Electric Cars Don't Use Fossil Fuel, but what's the Environmental Impact and Life Cycle of the Batteries?

It's really good to know that electric vehicles have less emissions overall, but when are they going to come out with and electric vehicle that is affordable. Even though Tesla is coming out with the "S", it is still out of reach of most consumers @ $50k.

I think there might be a smart car within reach, but that is just simply too small for an American highway with all the CarbonPig SUVs out there...

Great article - thanks,

Kevin

Thanks for posting this, Matt. I was pleased to see the information about lithium sources and mining.

For the chart you showed, Tesla assumed that the electricity to power its car was made by natural gas generators, reasonable for the USA on average but not for the area I live in. About 90% of the power in my local grid is hydro power in an average water year, which means that the efficiency of the plug-in Tesla would be even better for me than for the average American. However, Tesla probably did not include the energy cost of making the vehicle battery in its efficiency figures. Most assessments don’t. That’s a problem, it turns out, because the energy cost of making a lithium battery is significant.

I’m more interested in plug-in hybrids than pure electrics. Pure electrics will be fine for the commute but won’t be suitable as an only car for the next few years. They don’t go far enough on a single charge to make a highway trip.

I’ve read that a lithium battery takes 1,500-2,000 MJ of primary energy per 1 kW battery capacity to make, including the energy cost of extracting and refining the lithium. Looking at the 16 kWh battery now being made for the Volt, I’d be able to drive a 2010 Civic about 30,000 km or 19,000 miles before the energy penalty of making the lithium battery was absorbed. Interestingly, that battery is almost exactly the same size as the one being put into the all-electric smart car. A recent study at Carnegie Mellon University showed that the lifetime greenhouse gas emissions from a plug-in hybrid are about 40% less than a gasoline-powered car if the electricity comes from typical US sources, or almost 80% less if the electricity grid has a mostly renewable power source, like mine.

I can see I should learn how to put hyperlinks into my replies here, so let me try. The passage "a lithium battery takes 1,500-2,000 MJ of primary energy per 1 kW battery capacity to make, including the energy cost of extracting and refining the lithium" was based on a 2008 paper from Carnegie Mellon . And, of course, the “per 1 kW battery capacity” should have been “per 1 kWh”, an embarrassing slip. I’ve looked at an independent study from Japan where the energy input also came close to 1500 MJ per kWh for making an EV battery. The Japanese authors went on to calculate the CO2 emissions at 75 kg per kWh, which, for making the 53 kWh battery in a Tesla would be close to 4 metric tonnes, or 1.2 metric tonnes for the 16 kWh battery now being made for the Volt. Interestingly, the Japanese paper then looked at the implications of recycling the batteries at end-of-life, calculating that recycling added less than 3% to the lifetime footprint for CO2. I don’t think you could find an EV battery for $5K. The current cost is $1,000 to $1,200 per kWh according to a study by the Boston Consulting Group (subscription required) reported in Automotive News. That’s at least $16,000 for a Volt’s battery, and it implies the 16.5 kWh battery in the trial release of 1,000 electric smart cars costs about the same as an entire smart Passion cabriolet with gasoline engine. It’s revealing to me that Daimler has never announced a retail price for the electric smart, which is available only for lease and only in selected markets. Xavier Mosquet, Boston Consulting’s global automotive practice leader, said the high cost of batteries will prevent EV’s from penetrating the mainstream market for at least the next few years. I'd like to think he's wrong but I doubt he is.

Bringing the discussion back down to earth, what do all these numbers mean to a person concerned with minimising their ecological footprint and their personal contribution to global warming?

The simple answer is that the emissions associated with making a Li-ion battery are significant in the short term but not over the lifetime of a vehicle. For the 16 kWh battery in the Volt, the mileage threshold seems to be around 30,000 km compared to driving a regular Honda Civic; once past that point the energy penalty from making the battery has been absorbed and driving the EV contributes less to global warming - at least 40% less but as much as 80+% less, depending on the proportion of renewables making up the power grid used for battery charging.

But, we'll have to pay for the privilege of being green. How much depends on the government subsidy that will undoubtedly become a hot topic of discussion once the Volt is released. It's presently $7,500 for a Volt or equivalent, and I don't think that's enough for it to attract the number of buyers it should. Of course, GM could sell each one at a spectacular loss, but their major shareholder might object.

I think the analysis is fundamentally flawed but I agree with the broad conclusion that Li-ion powered EVs are kinder to our environment even when considering the manufacture of the cells as well as the operation of the vehicle.

1. Li-ion batteries are not "primary" cells. Primary cells are the little AA, C, and D cells that are not rechargeable. Even though they use Lithium they are completely different than a Li-ion battery. Therefore I think the manufacturing analysis is meaningless.

2. Tesla uses Sanyo cells for the Roadster, not Panasonic. Maybe they plan to use Panasonic for the Model S which is not yet in production and won't be until at least 2012.

3. 180,000 miles is quite optimistic. The useful vehicle life for a Li-ion battery depends on many factors. "Li-ion" is really a family of chemistries each of which uses different anode materials. Each chemistry has a different cycle life and they vary widely. Also cycle life can be substantially improved by not using 100% of the cell capacity on each cycle - that's why the Volt battery is 16kW-h when it uses only 8kW-h per cycle. Other factors include temperature, charging algorithms, and calendar life.

4. The analysis ignores the fact that Li-ion batteries have the potential of being re-marketed for stationary energy storage applications when their useful life in a vehicle has ended. That's because they still have around 75% of their energy storage capacity remaining.

Definitely crucial to look critically at the environmental impacts of electric cars. In the end, if EVs are powered 100-percent by renewable energy, you have a zero-pollution vehicle in terms of actual auto emissions. Hard to think of anything that could clear the air in urban centers around the world quicker than that. Also, if the li-ion batteries and the car itself are produced by power generated by renewables, poof, there goes the carbon footprint -- or most of it (I suppose you still have to factor in transportation of the raw materials, battery, etc. by ship, truck, train, etc.)

While I'm very definitely concerned that lithium be mined properly and those who mine it be treated fairly, it seems clear to me that nothing can reduce air pollution more quickly in the world than the EV + renewable energy equation.

A very interesting piece of research: well done, this is an area I have been looking at in some detail in the past year and getting any sort of figures has been very difficult. Very few battery manufacturers release their carbon footprint for manufacturing lithium based batteries, making it very difficult to get accurate figures. You are to be congratulated.

You may be interested to hear that I did manage to get some information from two battery manufacturers - one who gave me the information verbally on the strict understanding that the information was kept to myself, the other from French manufacturer, SAFT.

Both battery manufacturers came up with similar figures for their carbon footprint - between 20-23kg for every 1kWh of battery storage. In neither case was I able to see the calculations used for finding these figures. However, I was told that around 14-15kg is associated with the mining process and transportation of the lithium. The remainder is associated with the battery manufacture.

My gut instinct is that those figures are too low, but I have yet to clarify this. My gut instinct suggests to me that the real figure would be somewhere in the region of 75kg/kWh, but I have no figures to back this up.

Likewise, my gut instinct is that your figures are too high. I am wondering whether the Carnegie Mellon Economic Input-Output Life Cycle Assessment is flawed when looking at lithium-based batteries? At present, there is a huge disparity between the mining cost for lithium and the sale price for lithium. There is also a huge disparity between the production cost of a lithium-based battery and the sale price for a lithium-based battery. What is the correct price to ascribe to a lithium-based battery? The production cost? The cost of the raw materials and the production cost as separate figures? The trade price? The retail price? There are huge disparities between each set of figures. If I run the tool based on the estimated manufactured price for lithium-ion batteries, then the figures the Carnegie Mellon Economic Input-Output Life Cycle Assessment tool come up with are around 90kg/kWh.

Incidentally, there is a wide disparity into the size of different battery packs used by different electric cars. The Reva NXR Intercity I test drove last year in Frankfurt had 12kWh of lithium-ion batteries. The Reva dc-drive I use has 9.6kWh of lead acid batteries. My Mitsubishi iMiEV has 16kWh of lithium-ion batteries. I believe the Tesla has 53kWh of lithium-ion batteries, whilst the Mini-e (and, I believe, the Nissan LEAF) has 35kWh of lithium-ion batteries. Making the statement in your article that "roughly 13.5 Metric Tonnes of CO2e are produced when manufacturing an electric car battery" does not take into account the different sizes of battery packs.

However, this is not to detract from the work you have done. It is good to see that other people are looking at this and working hard to get an accurate set of figures so that the consumer can see whether or not electric cars make environmental sense. It is interesting to note that whatever figures you choose - yours, mine, or completely made up from thin air - electric cars appear to make good environmental sense as the manufacturing impact of the batteries is always less than the ongoing impact of using fuel in an internal combustion engine.

Great article, but I think you're missing some vital context here, which that current estimates are that our personal individual annual carbon footprints should be at the most around two tonnes, declining to one tonne over the next 100 years.

So, by your calculation, buying an EV blows around 6 years' carbon budget – just on the battery (never mind the carbon footprint for the rest of the vehicle).