New Carbon Nanotube Electrode Material for Li-ion Batteries Tackles the Power Performance Gap Between Electrochemical Capacitors and Batteries
20 June 2010
Researchers at MIT have developed a new carbon nanotube electrode material for a Li-ion battery based on redox reactions of functional groups on the surfaces of the nanotubes. The electrode, which is several micrometers thick, can store lithium up to a reversible gravimetric capacity of ~200 mAh g-1electrode while also delivering 100 kW kgelectrode-1 of power and providing lifetimes in excess of thousands of cycles, both of which are comparable to electrochemical capacitor electrodes.
A paper on the work, led by Associate Professor of Mechanical Engineering and Materials Science and Engineering Yang Shao-Horn, in collaboration with Bayer Chair Professor of Chemical Engineering Paula Hammond, was published online in the journal Nature Nanotechnology 20 June. The lead authors are chemical engineering student Seung Woo Lee PhD ’10 and postdoctoral researcher Naoaki Yabuuchi.
Layer-by-layer techniques are used to assemble an electrode that consists of additive-free, densely packed and functionalized multiwalled carbon nanotubes. A device using the nanotube electrode as the positive electrode and lithium titanium oxide as a negative electrode had a gravimetric energy ~5 times higher than conventional electrochemical capacitors and power delivery ~10 times higher than conventional lithium-ion batteries.
—Lee et al.
The performance can be attributed to good conduction of ions and electrons in the electrode, and efficient lithium storage on the surface of the nanotubes, the researchers said.
While such electrodes might initially find applications in small portable devices, with further research they might also lead to improved batteries for larger applications, such as in vehicles, the team suggests.
The layer-by-layer fabrication method involves alternately dipping a base material in solutions containing carbon nanotubes that have been treated with simple organic compounds that give them either a positive or negative net charge. When these layers are alternated on a surface, they bond tightly together because of the complementary charges, making a stable and durable film.
The carbon nanotubes self-assemble into a tightly bound structure that is porous at the nanometer scale. In addition, the carbon nanotubes have many oxygen groups on their surfaces, which can store a large number of lithium ions; this enables carbon nanotubes for the first time to serve as the positive electrode in lithium batteries, instead of just the negative electrode.
This electrostatic self-assembly process is important, Hammond says, because ordinarily carbon nanotubes on a surface tend to clump together in bundles, leaving fewer exposed surfaces to undergo reactions. By incorporating organic molecules on the nanotubes, they assemble in a way that “has a high degree of porosity while having a great number of nanotubes present,” she says.
The electrodes the team produced had thicknesses up to a few microns, and the improvements in energy delivery only were seen at high-power output levels. In future work, the team aims to produce thicker electrodes and extend the improved performance to low-power outputs as well, they say.
In its present form, the material might have applications for small, portable electronic devices, says Shao-Horn, but if the reported high-power capability were demonstrated in a much thicker form—with thicknesses of hundreds of microns—it might eventually be suitable for other applications such as hybrid cars.
While the electrode material was produced by alternately dipping a substrate into two different solutions—a relatively time-consuming process—Hammond suggests that the process could be modified by instead spraying the alternate layers onto a moving ribbon of material, a technique now being developed in her lab.
This could eventually open the possibility of a continuous manufacturing process that could be scaled up to high volumes for commercial production, and could also be used to produce thicker electrodes with a greater power capacity.
Funding for the work was provided by the Dupont-MIT Alliance; the US Office of Naval Research; and the MRSEC Program of the National Science Foundation.
Seung Woo Lee, Naoaki Yabuuchi, Betar M. Gallant, Shuo Chen, Byeong-Su Kim, Paula T. Hammond, & Yang Shao-Horn (2010) High-power lithium batteries from functionalized carbon nanotube electrodes. Nature Nanotechnology doi: 10.1038/nnano.2010.116
I am confused: Is this a battery or a supercapacitor. The headline indicates battery, but the numbers seem to say supercapacitor.
I'm having trouble getting the actual numbers here. The 200mAh/g and 100kW/kg numbers are for the electrode alone, I wonder what the cell's numbers would look like.
They later talk about:
" A device using the nanotube electrode as the positive electrode and lithium titanium oxide as a negative electrode had a gravimetric energy ~5 times higher than conventional electrochemical capacitors and power delivery ~10 times higher than conventional lithium-ion batteries. "
Considering that "conventional electrochemical capacitors" on the market max out around 6Wh/kg, I guess this is then around 30Wh/kg and since the better lithium ion batteries are about 500W/kg then this cell would have 5kW/kg?
Sounds like a supercapacitor and a battery made babies and most of the DNA came from the supercap to me.
Posted by: DaveD | 21 June 2010 at 05:58 AM
I am not aware of any lithium battery which gets anywhere near 500wh/kg - you would need to be using lithium sulphur or something.
I'm not sure what the carbon nanotubes are bringing to the party, as lithium titanate batteries have great power density anyway, far more than other lithium chemistries, and in a car would make capacitors pointless.
Toshiba with the SCiB have hit 100wh/kg.
Here are the rest of the specs including power densities:
Posted by: Davemart | 21 June 2010 at 06:42 AM
In formula 1, the A123 cells (lithium-iron-phosphate) used in the MacLaren were reportedly giving out (and taking in) 20 kW/kg last year.
Of course they were probably pulverised in a relatively short time by that kind of abuse.
Posted by: clett | 21 June 2010 at 06:46 AM
That was 500W/kg, not Wh/kg. I was speaking of the power density. I believe there are a few approaching that range for power density though most are closer to 300-350W/kg. I should have been more specific that I was speaking of power when I mixed in both power density and energy density in the same paragraph. Sorry :-)
On the energy density front, Lithium (depending on which chemistry obviously) ranges between about 100Wh/kg and as high as ~260Wh/kg (the new panasonic batteries announced for early next year).
Posted by: DaveD | 21 June 2010 at 06:52 AM
Wow, I wondered how much the A123 packs could handle! Good info, thanks.
I would love to see how they survived that abuse and whether or not they lasted multiple races before having to be replaced.
Of course, they have at least 8 engines per car, per season at a cost of nearly $250,000 per engine so what do they care if they replaced a few kg of batteries between races LOL
Posted by: DaveD | 21 June 2010 at 06:54 AM
Yeah, I thought you were talking about energy density, which is why I changed your 500w/kg to wh/kg.
On power density I would agree that 500w/kg is readily attainable - the Toshiba I referenced does a lot better than that. Check out the specs.
I will have a look at the Panasonic - life can often get confusing between energy density at the cell level and at the module and battery system level.
I believe although I am not certain that the Toshiba I referenced refers to the module level.
They are working on getting it up to 150wh/kg, which should allow 200 mile range EVs, as my understanding is that the batteries on the Leaf for instance, have around 88wh/kg at the module level.
Posted by: Davemart | 21 June 2010 at 07:01 AM
Yeah, I was being lazy, I should have provided the reference on the Panasonic batteries. They announce fiscal year 2012 which "ends in March 2012"...so I guess it begins in March of 2011. They have already delivered some cells to Tesla for testing, but I can't find details beyond that.
Here is the link, and I re-did the math and it is 266Wh/kg at the cell level:
Posted by: DaveD | 21 June 2010 at 07:16 AM
I ran some numbers on what kind of energy they could get back from an F1 at it's most aggressive corners. If I recall, one of the corners,at Manza I believe it was, has braking from ~200mph down to about 75mph in 1.2 seconds.
Assuming about 700kg including car, driver and some fuel, that is about 1.6 MEGA Watts of available regen braking power!
Of course you wouldn't do all of that with regen braking, and that would give you about 550Wh of energy (after aerodynamic and frictional losses) and the KERS systems are limited to about 111Wh (400kJ) so they couldn't be pumping more than about 333kW of power into the battery pack.
If they are taking 20kW/kg then it would be about 16.65 kg, of the 30kg total for the KERS system, would have been A123 batteries.
Those numbers are at least rational so I could believe they are in the right ball park. Very impressive for the batteries to handle all that. As you said though...I wonder if it destroys them?
Posted by: DaveD | 21 June 2010 at 07:55 AM
Thanks for the Panasonic links.
The Toshiba would handle the regen without blinking, as it can charge to 95% in 5 minutes at a charge rate of 240kw.
It is good for 6,000 cycles down to 80% at that charge rate.
Posted by: Davemart | 21 June 2010 at 08:29 AM
The F1 rules at the time were for 60 kW max in and out of the battery, so they could have done that with just 3 kg of batteries!
PS KERS is coming back for next year's F1 season, but the top teams couldn't get everyone to agree to a higher power / energy limit in the regulations, so it will still be limited to 60 kW and 400 kJ.
Posted by: clett | 21 June 2010 at 09:25 AM
What a shame they didn't up the power and kJ they could store! They could have doubled both and cut back a little bit on the fuel allowed to see what these things could really do.
It would also be fun to see if it let them cut back on the wear and tear of the traditional brakes a bit to up the regen.
Frankly, with the way Hamilton and Vettel try to tear up their brakes, I can't believe that they haven't had more brake failures this year with the extra weight of the fuel loads.
Posted by: DaveD | 21 June 2010 at 10:11 AM
@ Davemart: "Toshiba with the SCiB have hit 100wh/kg": not according to the link to the Toshiba PDF you provided. 4AH at 12V for a 1KG pack indicates an energy density of 48wh/kg?
Posted by: Chris O | 21 June 2010 at 02:00 PM
Check bottom right for their comment on their 20AH battery.
Other info from Toshiba indicates that this reaches around 100Wh/kg in a module, I assume:
'As for EVs, we will begin to ship samples with the nominal capacity of 20Ah and the energy density of 100Wh/kg this fall. The energy density is sufficient for a small EV, but in order to drive a larger EV, the energy density needs to be raised to around 150Wh/kg. We are already working on this, and we are putting further efforts into research for EV batteries (Fig.6).'
Posted by: Davemart | 21 June 2010 at 02:13 PM
Still looks like A123 has the leading technology for now. The Japanese will be pushing hard to beat that and this will further accelerate EV battery improvement. All in all a very good environment for progress.
Posted by: Reel$$ | 03 July 2010 at 07:12 AM