September 27, 2009
Silicon Nanotubes For 10 Times Better Batteries

A report in MIT's Technology Review bodes well for the future of electric cars. This could be a game changer.

In an advance that could help electric vehicles run longer between charges, researchers have shown that silicon nanotube electrodes can store 10 times more charge than the conventional graphite electrodes used in lithium-ion batteries.

Better anodes can absorb more lithium and so hold more charge.

Researchers at Stanford University and Hanyang University in Ansan, Korea, are developing the nanotube electrodes in collaboration with LG Chem, a Korean company that makes lithium-ion batteries, including those used in the Chevy Volt. When such a battery is charged, lithium ions move from the cathode to the anode. The new battery electrodes, described online in the journal Nano Letters, are anodes and can store much more energy than conventional graphite electrodes because they absorb much more lithium when the battery is charged.

We need a path of migration away from fossil-fuel powered cars. There's not enough cheap oil left for easy access and rising Asian demand combined with the coming of Peak Oil look set to drive up gasoline prices far higher than they are today. The costs of substitutes will determine how high the price of gasoline will go. Cheap high capacity batteries for long range electric cars would enable most driving to be transitioned to electric power. The incremental increase in electric demand could then come from nukes, wind, solar, and other non-fossil fuels energy sources.

Share |      Randall Parker, 2009 September 27 12:41 PM  Energy Batteries


Comments
Hong said at September 27, 2009 2:33 PM:

That's the question. Will the cost of producing these batteries be more damaging to the environment than fossil fueled powered engines?

AbuZuzu said at September 27, 2009 5:10 PM:

The nano-tubes make a great battery cathode. There needs to be a comparable breakthrough on the anode side to realize this 'breakthrough"

jay said at September 27, 2009 9:24 PM:


Cui reported this about 18 months ago. Thereafter, the Saudi's got involved funding his lab at Standford to
the tune of something like $5M to $10M.

This article doesn't mention the Saudis at all, so I'm not sure what happened there, but good ridence if so.

Brett_McS said at September 27, 2009 10:27 PM:

High power batteries will make a massive difference to everything, not just cars.

However, the electric power infrastructure is in a pretty diabolical state at the moment with greenie activism blocking near/medium term capacity increases. A significant fleet of electric cars will result in major blackouts, as things stand.

ScottN said at September 28, 2009 7:01 AM:

I disagree with Brett_Mcs.

Assuming a smarter grid, a fleet of electric cars will charge at night (when power rates are low) and potentially feed power back to the grid during the day (when power rates are high).

Less chance of blackouts, not more.

Dan C said at September 28, 2009 9:43 AM:

While I still am a proponent of migrating to fully electric vehicles, I had looked into what this would mean to the need for additional electrical generation capacity. I had performed a (relatively) simple equation to determine what additional generation capacity would be required to transition just 5% of the nation's automobile fleet to all electric - and the answer was approximately a doubling of the current generation capacity. I will say that this analysis did not take into consideration the fact that electric engines are more efficient than gas engines, although this fact will unlikely change the conclusion - that a significant shift from gas powered cars to electric will take a massive investment in generation (and transmission). Another fact that I didn't take into account, and just read about, is the load on the electric grid will be large for "fast charging" batteries. The peak generation required to service these technologies will be large. See this article for more information: http://www.lowtechmagazine.com/2009/03/fast-charging-electric-cars-off-peak-grid.html

It really looks like the transition to electric vehicles will necessarily be a slow one - a breakthrough on battery tech or not. We have to develop our clean generation sources at the same time.

Nick G said at September 28, 2009 11:16 AM:

Dan, two thoughts.

First, I agree that we'll go for a long time with plug-in hybrids like the Chevy Volt, which will eliminate the need for fast charging.

OTOH, I think your estimate of the need for changes to the grid to accommodate full EVs is too high. Just because some people drive long distances occasionally doesn't change the average power needed, which would only be about 9 KWHs per vehicle per day. If the grid penalizes people for charging at peak periods, that will move charging to the low periods in the middle of the night. Also, 90% of vehicles have off-road parking, so the infrastructure needed isn't quite as large as your estimate (or isn't needed as soon).

Engineer-Poet said at September 28, 2009 12:28 PM:

I don't know where Dan made his mistake, but it was a doozy.  Using relatively generous allowances for vehicle efficiency (I'm seen figures which are some 25% lower than my estimate), it would only take about a 40% increase in net generation to replace all gasoline and all diesel used in the USA.  The increase in generation capacity would be much smaller, because off-peak charging would largely use existing generators for more of the day.

Nick G said at September 28, 2009 2:09 PM:

E-P,

I agree.

It's worth noting that the logical source of energy for those EVs would be wind power. EV charging can be scheduled, so it's a perfect match with the variability of wind. If we increase demand by 40% for EVs, and power them with wind, we've integrated a roughly 1/3 market share for wind absolutely painlessly.

Of course, wind isn't completely random, so EV charging (and, eventually, V2G) would actually enable integration of a much larger % market share - easily 50%.

Engineer-Poet said at October 1, 2009 6:30 AM:

I've been wondering about how well that would work with different generation mixes.  Wind works really well in combination with generation which uses relatively scarce or expensive energy and is easily turned up and down, like hydro or gas turbines.  But if the expansion into zero-carbon energy is base load like nuclear, there's no real savings from turning down the non-wind part.  EVs work very well to level generation variations on the order of hours, but poorly for variations on the order of days.

The implication is that another element is required to make wind fit well if most other generation is inherently base load.  CAES is a possibility, as it can level variations on the order of days or even weeks and can buffer nuclear as well as wind.

Nick G said at October 2, 2009 1:46 PM:

Wind works really well in combination with generation which uses relatively scarce or expensive energy and is easily turned up and down, like hydro or gas turbines.

Yes. I'd phrase it slightly differently: Wind works really well in combination with generation which uses relatively cheap equipment (like gas turbines), so that most of your expense is in the fuel. It works just as well whether the fuel is cheap and plentiful or scarce or expensive, because you don't use that much.

But if the expansion into zero-carbon energy is base load like nuclear, there's no real savings from turning down the non-wind part.

Yes.

EVs work very well to level generation variations on the order of hours, but poorly for variations on the order of days.

That depends on the size of the EV resource, and whether we take advantage of V2G. If we have 220M EVs, each with 100KWH of battery storage, that's 22 TWH's. That's enough to store the US's current average electrical generation for two full days, and enough to replace generation for 8 days of a 25% generation shortage (shortages larger than that are very, very unlikely in the scenario we're talking about here - there will always be some light and wind, and of course the nuclear will still be there). If these are Extended Range EVs, like the Volt - well, a fleet like that could store another 100KWH in it's gas tanks, for 16 days of backup.

The implication is that another element is required to make wind fit well if most other generation is inherently base load. CAES is a possibility, as it can level variations on the order of days or even weeks and can buffer nuclear as well as wind.

Sure, CAES would work. I kind've think that it won't be that big because of the problems of heat storage, or heat loss and the need for co-firing of something else.

In the medium term, we could just keep the nat gas and coal plants. That could work because the long outages we're concerned about would happen rarely, so this kind of backup would only need to handle maybe 5% of total KWHs. In the long-term, you could use biomass.

Engineer-Poet said at October 2, 2009 6:33 PM:
Wind works really well in combination with generation which uses relatively cheap equipment (like gas turbines)
True.  The economics of any generator which has a very low capacity factor depends on low capital cost and not nearly as much on energy cost.  Hold that thought, I'm coming back to it.
EVs work very well to level generation variations on the order of hours, but poorly for variations on the order of days.
That depends on the size of the EV resource, and whether we take advantage of V2G.
No, not for quite a while.  Until EVs have enough battery capacity to run the average mileage for several days without charging, they are going to need to be recharged more or less every day.  For instance, my 2004 calculation found that gasoline-powered vehicles would need on the order of 107 GW of electricity average.  It would take a nation-full of vehicles like the Chevy Volt to get up to a full 24 hours of usable storage, and several times as much before you have days of buffer.  That's not coming for a long time yet, and the capital expenditure will be huge.  Schemes like Project Better Place defeat range limitations and battery cost by moving most "storage" to the grid, so you'd be back to mere hours.
Sure, CAES would work. I kind've think that it won't be that big because of the problems of heat storage, or heat loss and the need for co-firing of something else.
I know there are studies of adiabatic CAES going on, and if you are mining caverns to store air you can easily mine a bigger one and fill it with crushed rock.  You feed hot compressed air into the center and the rock stores the heat for you as the air goes past it, returning most of it when you draw the air out again.  About the only thing cheaper than crushed rock is air, and the storage for CAES uses little else.

Early this year I was working on an idea for CAES using nuclear heat, but the standard CAES system didn't have enough of an efficiency advantage over supercritical CO2 turbines to make the output boost worthwhile (~50% for supercritical CO2 turbines, ~80% heat to electric for CAES).  Adiabatic CAES changes that game and may make it worthwhile.  I have no time to look at this right now, I'm picking apart Dittmar's bogus series on TOD at the moment and it's eating all my time for in-depth writing.

Randall Parker said at October 2, 2009 8:30 PM:

E-P makes a good point about the Chevy Volt and similar PHEVs. You really need pure EVs to effectively do V2G. We are years away from having enough EVs to do V2G on a serious scale. So wind has scaling problems in the mean time.

We also need dynamic pricing to create the incentives for the development of better ways to vary demand. A lot of the storage needs to be done after the electric power is used. For example, make extra ice or heat up molten salts in a house or commercial building when electric prices are low.

This reminds me: How much pure ice would a refrigerator/freezer need to make in order to be able to avoid/shift electric power usage for, say, 12 hours or 24 hours? What's the incremental cost for making refrigerator/freezer appliances that have big load shifting capability? Freezers already make ice. You'd think this would be a small incremental cost to give them this capability. I wonder what the pennies per kwh price gap needs to be to make them capable of doing load shifting in this manner.

Are ground sink heat pumps well or poorly equipped for demand shifting? I would expect they work more poorly the bigger the temperature shift they generate. But they would need to make bigger temperature shifts on their inputs in order to make storeable heat and cold.

I also wonder what other easy ways to shift electric power could be developed. Your thoughts?

Engineer-Poet said at October 3, 2009 7:59 AM:

Don't most refrigerators have only one evaporator for both the refrigerator and freezer sections?  It operates at only one temperature, so there is no loss from making ice for the refrigerator section.  (Making something cold enough to keep the freezer at 0°F for a while might increase energy requirements a bit).  I suspect that it would make more sense to use vacuum-insulated panels to cut the overall energy demand and rely on thermal mass of the contents than to fuss with lots of ice.

The Ice Bear product spec is vague on total EER, but says it can manage 30 ton-hours of cooling (360,000 BTU) while shifting 35 kWh of load.  That's roughly an EER of 10.  I suspect that a ground-source system would do better due to lower condenser temps and back pressure.

Nick G said at October 5, 2009 3:14 PM:

E-P said That depends on the size of the EV resource, and whether we take advantage of V2G. - No, not for quite a while.

We're actually in agreement. The point is, we won't need EVs to buffer wind's variation for quite some time: it will take a while for us to get to the 10-20% of KWH market share that the current grid can accomodate. We need to remember that the existing grid can do quite a bit more DSM than it does now: a lot of A/C (for instance) could go on existing DSM programs. DSM is underutilized because utilities prefer to build generation.

Randall said E-P makes a good point about the Chevy Volt and similar PHEVs. You really need pure EVs to effectively do V2G. We are years away from having enough EVs to do V2G on a serious scale. So wind has scaling problems in the mean time.

See my note to E-P: he and I were really talking about buffering on the scale of days. At the moment, we don't need EVs, and we don't need V2G. A relatively small number of Volts using dynamic charging can provide a significant amount of hourly buffering, which is all we'll need for quite a while.

Post a comment
Comments:
Name (not anon or anonymous):
Email Address:
URL:
Remember info?

                       
Go Read More Posts On FuturePundit
Site Traffic Info
The contents of this site are copyright