MIT students develop liquid fuel for electric cars | Hacker News Reader

MIT students develop liquid fuel for electric cars

i.autoblog.com

153 pointsjamesjyu26 comments

The highest energy density listed for the electrolytes on the wikipedia page for flow batteries is 75 watt hours/kg, that is about 0.25Mj/kg. Gasoline is 42Mj/kg.

The original MIT release claims a 10 fold improvement in energy densities over existing flow batteries, but that is still more than an order of magnitude worse than the incumbent fuel.

Pardon my ignorance, but are the flow batteries you mention the same type used in existing electric vehicles? It can't be assumed that the energy density of flow batteries will ever match as gasoline, but people might be willing to accept the difference if it means more infrastructure adoption and less time spent recharging vehicles.

Good point.

I've not seen flow batteries in existing vehicles. Electric vehicles use NiMH batteries (0.25 MJ/kg), lithium batteries (0.4 to 0.7 MJ/kg) or lead acid (0.14 MJ/kg fork trucks, but they need mass anyway).

Guessing at the numbers MIT is avoiding naming in that press release, they might get to something three times better than the current lithium batteries, but you also have to add in the mass of the pumps, and the reaction part of the battery.

It could "win", but it won't be a game changer.

Putting aside vehicles though, flow batteries have some interesting properties for utility scale energy storage. The rate at which you can store or retrieve energy is independent from the amount of energy you can store. The first is by the size of your reactor, the second is by the volume of your storage tanks. It is possible if they are using common enough elements for their fluid that they could be a good answer for peak load shifting or production storing.

Here's the numbers:

...we estimate that optimized SSFC systems using established lithium intercalation compounds could have energy densities of 300–500 Wh L−1 (specific energy 130–250 Wh kg−1), which would satisfy metrics considered necessary for widespread adoption of all-electric vehicles.[4] Further improvements would be possible by ‘dropping in’ higher-energy-density or lower-cost storage compounds in the SSFC platform as they are developed.

http://onlinelibrary.wiley.com/doi/10.1002/aenm.201100152/fu...

(linked from http://web.mit.edu/newsoffice/2011/flow-batteries-0606.html)

So 0.47-0.90 MJ/kg. Based on theoretical calculations -- they haven't actually engineered this.

Isn't that about the level where it gets interesting? My car can go something over 400 miles on a tank of gas right now. If their new batteries can get 40 miles on a "tank", and be refueled as easily as my car can now, that seems right on the threshold of being a very viable product.

Also, gas tanks could be a lot bigger without ruining cars. Things would be slightly less efficient, but if the solution is just being reused then pumping thirty gallons at once might not be the end of the world (as opposed to the ten that currently fit in my tank). If the solution weighs as much as gasoline, then that's an extra 120 lbs when the tank is full, which is a lot, but it's <5% extra.

Not to mention if the tank is full of a fluid that is not particularly dangerous, you might be able to integrate it into the structure of the car, similar to motorcycles with oil-in-frame design.

> If their new batteries can get 40 miles on a "tank", and be refueled as easily as my car can now, that seems right on the threshold of being a very viable product.

Not really. I go 300 miles between fillups and would like to go more. 10x as many visits to the "energy station" is a huge obstacle.

Not to mention that I often go on trips that are longer than 40 miles. Losing 5 minutes to fillups every 3-40 miles is unworkable. And, I often go places where the gas station density is low, so I'm going to have to fill up more often and there are places that I just can't go on a 3-40 mile tank.

As a commute-only vehicle, 30-40 miles can make sense, but few of us can afford significant money for such an "extra" car.

Single people can't, but commute-only vehicles are common for families. The minivan takes care of major trips, and the commuter car is used for the spouse with the longer commute.

Does the calculation of efficiency take into account the efficiency of conversion? I know that the actual potential-to-kinetic energy conversion rate of gas-powered cars is about 20% ( http://en.wikipedia.org/wiki/Engine_efficiency#Gasoline_.28p... ). Wikipedia claims an electric car is 3x more efficient than that ( http://en.wikipedia.org/wiki/Electric_vehicle#Energy_efficie... ).

So theoretically, energy density of a battery tech only needs to hit one-third the efficiency of gasoline to match mileage. So the goal is 14Mj/kg, and a 10x increase would be 2.5Mj/kg -- so still lower, but getting much closer.

Based on your data and a sibling comment of yours, it would mean we would simply need fuel tanks that hold 5x more fuel. That doesn't seem unreasonable as the production cost of a fully electric car can be less as the engines are MUCH simpler (esp. when ordered in mass).

I think weight (and sloshing) would impact handling and economy in a severely negative if tanks were 5x as large. 400 pounds of liquid would be pretty bad.

Original article: http://web.mit.edu/newsoffice/2011/flow-batteries-0606.html

"The company has already raised more than $16 million in venture capital and federal research financing." - Awesome! Much better details in this.

Original paper: http://onlinelibrary.wiley.com/doi/10.1002/aenm.201100152/ab... (free access until 27 June 2011)

This looks like a major breakthrough. Love the nickname - "Cambridge crude". As with all flow batteries, power density is low. But you could possibly compensate for that in vehicles with ultracapacitors which could also recapture energy from braking.

Worth noting that the liquid is rechargeable/reusable: http://en.wikipedia.org/wiki/Flow_batteries

According to the Wikipedia page, the main issue with these batteries is low energy density. If this group has figured it out, they may be on to something huge.

I agree, if a development like this were to happen for electric vehicles, adoption of them would skyrocket. The one thought I have is how much will this cost to produce initially per liter, as well as over time when it reaches a critical mass?

As I've learned when such promising technology is announced, I'll wait for system integration prototypes to roll out.

Often in these sorts of articles, you can tell what the big problems are by the fact that they're not even mentioned. The big unmentioned problems here seem to be:

a) Cost

b) Cost

c) Safety, and

d) Cost

I'm not sure what materials they're making these things from, but would it be fair to guess that it's something exotic and hard to fabricate?

From the article: "The tech supposedly makes the batteries up to ten times more efficient than their traditional counterparts, and even more importantly, the new tech is cheaper to produce."

Cheaper is meaningless in this context. For example, if the traditional counterparts cost $1,000/mile then $100/mile is still cheaper but still far from being pratical. We have no idea what it's cheaper than and by how much it's cheaper by.

Of course, my numbers are completely made. It's just meant as an example.

I'm pretty sure it is saying 10x better and cheaper than a standard lithium ion for the same battery size. If they are not saying that then the sentence makes no sense.

Cost is a very complex thing, are you considering externalities?

In other words, we're paying a pretty damn high price for "gas" right now (including externalities like pollution, wars, and oil spills). Alternatives have not taken off due to chicken/egg problem of adoption<->economies of scale.

Also gas is not safe. It's highly combustible and toxic to humans: http://medicmagic.net/living-near-a-gas-station-increase-can...

When a major change is made to how something is done, there really needs to be a very significant improvement that makes the cost of changing worthwhile.

I think eventually this will happen with electric vehicles, but no one knows for sure. Which is probably why we shouldn't subsidize solutions that are not compelling enough. The current electric vehicles aren't even as good as what they are attempting to replace and wouldn't be available at all without the massive subsidies.

If we do subsidize things in an effort to "jumpstart" the change, we might actually switch to a system that is not compelling enough and might get persuaded to jumpstart the next generation, which might have been compelling enough on its own. And be stuck with the cost of the subsidies or paying the higher cost directly.

Of course, such changes become so monolithic to begin with when government picks a standard and encourages everyone to adopt it. It may drive down costs in the short run, but it reduces progress in the long run.

For example, the adoption of the NTSC tv format and enforcing it using the FCC. We got color. Then we got closed captioning. And that was probably all the innovation we got over 50 years. But look at how quickly video has evolved on personal computers by comparison.

"liquid fuel for electric cars" -- what if we used refined petroleum as a secondary energy source? we probably wouldn't even need a large electric motor any more.

Here's a much better article with diagrams and more details: http://www.greencarcongress.com/2011/05/ssfc-20110526.html

This really will be a major step forward if it pans out. Even if it doesn't work so great for cars it could be amazing for grid storage.

Reply on news.ycombinator.com