Li-ion rechargeable batteries that last longer, re-charge more rapidly

July 4, 2016, Paul Scherrer Institut (PSI)
Haphazardly arranged graphite flakes in a conventional anode (above left and center): lithium ions attempting to dock or return to the cathode are forced to take detours (above right). But if the graphite is subjected to a rotating magnetic field (below), the flakes in the suspension align themselves vertically in parallel formation. They keep this orientation after they have been dried (below centre). The ions have shorter paths (below right). Credit: Juliette Billaud, Florian Bouville, Tommaso Magrini/Paul Scherrer Institute, ETH Zurich

Materials researchers at the Swiss Paul Scherrer Institute PSI in Villigen and the ETH Zurich have developed a very simple and cost-effective procedure for significantly enhancing the performance of conventional Li-ion rechargeable batteries. The procedure is scalable in size, so the use of rechargeable batteries will be optimized in all areas of application-whether in wristwatches, smartphones, laptops or cars. Battery storage capacity will be significantly extended, and charging times reduced. The researchers reported on their results in the latest issue of the research journal Nature Energy.

It's not necessary to re-invent the rechargeable in order to improve its performance. As Claire Villevieille, head of the battery materials research group at the Paul Scherrer Institute PSI says: "In the context of this competitive field, most researchers concentrate on the development of new materials." In cooperation with colleagues at the ETH in Zurich, Villevieille and co-researcher Juliette Billaud took a different approach: "We checked existing components with a view to fully exploiting their potential." Simply by optimizing the graphite anode - or negative electrode - on a conventional Li-ion battery, researchers were able to boost . "Under laboratory conditions, we were able to enhance storage capacity by a factor of up to 3. Owing to their complex construction, commercial batteries will not be able to fully replicate these results. But performance will definitely be enhanced, perhaps by as much as 30 - 50 percent: further experiments should yield more accurate prognoses."

Researchers point out that in terms of industrial implementation, improving existing components has the great advantage of requiring less developmental input than a new battery design using new materials. As Villevieille says: "We already have everything we need. If a manufacturer were willing to take on production, enhanced batteries could be ready for the market within one or two years." The procedure is simple, cost-effective and scalable for use on in all areas of application, from wristwatch to smartphone, from laptop to car. And it has the additional bonus of being transferable to other anode-cathode batteries such as those based on sodium.

Arranging the flakes

In this case, changing the way anodes work was the key to success. Anodes are made from graphite, i.e. carbon, arranged in tiny, densely packed flakes, comparable in appearance to dark grey cornflakes haphazardly compressed, as in a granola bar. When a Li-ion battery is charging, lithium ions pass from the cathode, or positive metal oxide electrode, through an electrolyte fluid to the anode, where they are stored in the graphite bar. When the battery is in use and thus discharging, the lithium ions pass back to the cathode but are forced to take many detours through the densely packed mass of graphite flakes, compromising battery performance.

These detours are largely avoidable if the flakes are arranged vertically during the anode production process so that they are massed parallel to one another, pointing from the electrode plane in the direction of the cathode. Adapting a method already used in the production of synthetic composite materials, this alignment was achieved by André Studart and a team of research experts in the field of material nanostructuration at the ETH Zurich. The method involves coating the graphite flakes with nanoparticles of iron oxide sensitive to a magnetic field and suspending them in ethanol. The suspended and already magnetized flakes are subsequently subjected to a magnetic field of 100 millitesla-about the strength of a fridge magnet. André Studart explains that "by rotating the magnet during this process, the platelets not only align vertically but in parallel formation to one another, like books on a shelf. As a result, they are perfectly ordered, reducing the diffusion distances covered by the lithium ions to a minimum."

Juliette Billaud, co-first author of the new study, and Claire Villevieille, head of the battery materials research group at the Paul Scherrer Institute. Credit: Markus Fischer/Paul Scherrer Institute
Shorter paths for the ions

Microscopic images show that if the magnet remains turned on during the ensuing drying process, the platelets keep their new orientation even when removed from the ethanol suspension. Instead of their formerly haphazard arrangement, the flakes in the compressed graphite bar are now parallel, enabling the lithium ions to flow much more easily and quickly, whilst also increasing by allowing more ions to dock during the charging process. Claire Villevieille emphasizes that the "chemical composition of batteries remains the same". The remaining iron oxide nanoparticles are negligible in quantity and do not influence battery function. "All we did was optimise the anode structure."

Explore further: Charging ahead with magnesium batteries

More information: Magnetically aligned graphite electrodes for high rate performance Li-ion batteries, J. Billaud, F. Bouville, T. Magrini, C. Villevieille, A.R. Studart, Nature Energy 4. July 2016 (online) DOI: 10.1038/nenergy.2016.97

Related Stories

Inexpensive material boosts battery capacity

October 23, 2013

Battery-powered cars offer many environmental benefits, but a car with a full tank of gasoline can travel further. By improving the energy capacity of lithium-ion batteries, a new electrode made from iron oxide nanoparticles ...

Recommended for you

Coffee-based colloids for direct solar absorption

March 22, 2019

Solar energy is one of the most promising resources to help reduce fossil fuel consumption and mitigate greenhouse gas emissions to power a sustainable future. Devices presently in use to convert solar energy into thermal ...

EPA adviser is promoting harmful ideas, scientists say

March 22, 2019

The Trump administration's reliance on industry-funded environmental specialists is again coming under fire, this time by researchers who say that Louis Anthony "Tony" Cox Jr., who leads a key Environmental Protection Agency ...

17 comments

Adjust slider to filter visible comments by rank

Display comments: newest first

ahaveland
4.6 / 5 (7) Jul 04, 2016
Getting 3x improvement in capacity and longevity into production, and for no additional cost would be a real game changer.
Grid and home storage would accelerate and the world would go electric even sooner and really shake up the commodity markets.
Today is not a good time to be in the fossil fuel industry.
kochevnik
1 / 5 (1) Jul 04, 2016
Nickel-iron battery still best for solar energy storage, then lithium or lead-acid depending upon current demand

Petrol has far superior energy density. Something that could make sugars or hydrocarbons from atmosphere would have superior energy density capability over any battery. Perhaps make carbon nanotubes from atmosphere using solar power? Burn them for energy or keep them for industry
fmfbrestel
4.4 / 5 (7) Jul 04, 2016
Getting 3x improvement in capacity and longevity into production, and for no additional cost would be a real game changer.
... snip ...

Yeah, it would be great if they could wave their magic wands to do that, but if you keep reading....
Owing to their complex construction, commercial batteries will not be able to fully replicate these results. But performance will definitely be enhanced, perhaps by as much as 30 - 50 percent: further experiments should yield more accurate prognosis.


No such thing as a free lunch. 30-50% at best, and there will definitely be some additional costs. Magnets are not free, extra processing steps to attach the iron oxides to the graphite isn't free either. if those steps increase the cost by even 10%, the $/Watt improvement might only be 20%.

Still good, and definitely still worth doing, but far from an overnight game changer.
fmfbrestel
5 / 5 (5) Jul 04, 2016
Where this might see good application however is in fields that don't care about cost, only pure performance -- such as spaceflight. a 3x improvement in energy density when it costs $1000 per pound to put something into orbit is nothing to sneeze at, and very well might justify the production costs associated with the full laboratory setup.

Other "cost as no object" applications for military projects might also be interested. solar powered drones that never have to land because they have awesome batteries will not care about the cost.
Da Schneib
4.5 / 5 (8) Jul 04, 2016
I'll be interested to see what happens when a process engineer gets ahold of the technique. If the limit in a lab is 3x, I bet a good process team could get 2x out of it in an industrial production environment.
Eikka
5 / 5 (5) Jul 05, 2016
I'll be interested to see what happens when a process engineer gets ahold of the technique. If the limit in a lab is 3x, I bet a good process team could get 2x out of it in an industrial production environment.


I think the question is about repeatability. In a process that needs to be done in mass at high speed, the results will vary, so some battery cells might well come out excellent and others not so much.

The actual capacity of cells varies over a range as it is, and they have to be painstakingly paired into battery systems to get maximum performance. Simply throwing a bunch of batteries together creates a balancing problem where you have to run the pack according to the weakest cells and that drags down the capacity of the whole, reducing the potential gain.

So, for more money you could have specially picked cells that perform well, and for less money you can have a grab-bag of different cells that together perform poorly but cheaply.
Lord_jag
5 / 5 (7) Jul 07, 2016
In a process that needs to be done in mass at high speed

Why?

If the process takes an hour to complete, have multiple processes doing the same thing side by side.

Canning soda isn't very fast compared to how fast we drink it. There are bottling plants all over the world in almost every city with multiple machines bottling at max speed 24/7 to meet demand.

You don't think all cars come from one assembly line do you?

So once the process is completed, let the machine take the time it takes to do the job right, then duplicate the machine as many times as it takes to produce the volume required.
Lord_jag
5 / 5 (7) Jul 07, 2016
The actual capacity of cells varies over a range as it is, and they have to be painstakingly paired into battery systems to get maximum performance.


Oh so much wrong here.

You make a false assumption that the testing of the capacity/performance of the battery could not be easily automated. You also falsely assume that cells cannot be individually managed by a insignificantly small charge controller - the kind already installed in every lithium battery pack.

Consider what happens with CPU chips. Your thinking would have every one of them painstakingly characterized so that they don't burn out when installed where in fact an automated system tests every CPU and rates them according to how much ram isn't corrupt and the maximum stable clock speed.

Oh this one starts making errors at 2.8 GHz. We'll call that an i7-930. That one still works all the way up to 3.2 GHz with no errors. That one will sell that as a i7-960.
gkam
1 / 5 (4) Jul 07, 2016
Lord Jag is correct. In the business, they are called "bin numbers". A "bin one" is Military-spec, a "perfect" device. Standard consumer quality devices were coded bin six in our system, and the other bin numbers represented different grades and results in the tests; for example, bin two was power supply failure - overcurrent.

We watched our yield numbers very carefully.
TheGhostofOtto1923
3 / 5 (4) Jul 07, 2016
Lord Jag is correct. In the business
Business? What business might that be? One of the ones you got kicked out of?

Is this like that 100yo chart you thought governed exhaust stack emissions opacity but was actually replaced 40 years ago by CFR regs?

You know, back when you were 'in' that business?
Da Schneib
4.2 / 5 (5) Jul 07, 2016
You make a false assumption that the testing of the capacity/performance of the battery could not be easily automated.
Precisely, @Lord_jag; this process is called "quality assurance." You sell different grades; people buy what they can afford. Your example with Intel processors is very apt.
Eikka
5 / 5 (1) Jul 09, 2016
Oh so much wrong here.

You make a false assumption that the testing of the capacity/performance of the battery could not be easily automated.


I'm not saying it can't be automated. I'm saying it's costly because it involves charging and discharging the cell a couple times to find out its parameters such as capacity and ESR. It's generally slow to do in a mass-production setting.

I've seen people who make EVs buy a bunch of prismatic A123 cells from the factory, and they do have to test and pair them up.

You also falsely assume that cells cannot be individually managed by a insignificantly small charge controller


When you have a hundred cells in series to get the high voltages needed for a modern electric car, you can't magically "manage" it when one cell runs empty before the others.

the kind already installed in every lithium battery pack.


Those are simply low voltage fuses that prevent recharge if a cell voltage drops below a set point.
Eikka
5 / 5 (1) Jul 09, 2016
Consider what happens with CPU chips.


A CPU is easily tested in seconds.

A battery cell isn't, because it behaves differently if you charge/discharge it at different rates. Is the capacity the same at 20C dis/charge rate as it is with 1C or 0.1C? Nope, it isn't, and there's no nice mathematical model that would relate one to the other across different battery chemistries, makes and models.

There's currently no method of reliably and consistently testing or "binning" battery cells except by discharging and recharging them over a number of hours, which is why the cheapest cells are not tested individually - a batch is tested by random pick and the buyer gets a datasheet that shows the likely capacity distribution (min/max/typ).

Eikka
5 / 5 (1) Jul 09, 2016
For example, take a typical lithium cell data sheet:

https://www.adafr...0mAh.pdf


Nominal Capacity 2200mAh ±2%
Standard charge current 0.2C
Standard discharge current 0.5C


When you put such cells in series and charge them, you have to take into account that the difference in voltage between the best and the worst cells can be 4% which is kinda important because lithium cells are really picky about overcharging and wear out faster, or can even catch fire

If the cut-off limit is 4.3 Volts, the worst case scenario is that one or few cells are pushed to 4.47 Volts and near destruction. That means you have to limit all the cells to 4.1 Volts to make sure none of them go over the limit

The problem for the manufacturer is that while they can indeed test the cells and bin them, most of them won't fit a neat bin like 2200 mAh +-1% and because the customer demands more cells, they have to relax the tolerances to +-2% or +-5% and so-on
Science1st
Jul 09, 2016
This comment has been removed by a moderator.
Eikka
2.7 / 5 (3) Jul 09, 2016
Of course you can individually monitor each cell's voltage to keep track of battery balance, but that means cost and complexity and more points of failure.

For applications like cellphones, the actual capacity of the battery doesn't really matter because there's only one cell in the battery. For systems like electric cars, you need a hundred cells in series because you need high voltage to deliver the high power. Monitoring individual cells becomes costly due to complexity - one reason why Tesla's batteres are expensive - and even if you do, what are you going to do about it? You still can't charge or discharge each cell individually without a ridiculously complex and expensive charge controller.

If you read about how the actual battery packs in EVs are made, they're typically arranged into modules with a number of cells under a single controller, such as 4x4 or 4x8 cells in series-parallel configuration to reduce the complexity of the charging system.
fmfbrestel
4.7 / 5 (3) Jul 09, 2016
For applications like cellphones, the actual capacity of the battery doesn't really matter because there's only one cell in the battery.

Not true. That one "cell" has many thin cells stacked together wired in parallel and most definitely contains a charge controller.

For systems like electric cars, you need a hundred cells in series because you need high voltage to deliver the high power. Monitoring individual cells becomes costly due to complexity - one reason why Tesla's batteres are expensive
Again, FALSE. -- Tesla's batteries are the cheapest Li-ion batteries on the market and its not even really close. A small part of the reason they are cheap is because each "cell" does NOT have a charge controller because they put many of these "cells" together in parallel to make mini-modules (or mega cells) and have one controller for each mini-module. The enabling production method being very strict binning of individual "true cells" .

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.