Checking the shelf

Batteries have been the weak link of electric vehicles for well over a century, so any development is of great interest.  One bit of recent news was very exciting:  Altair Nanomaterials announced a new anode material for lithium-ion (Li-ion) batteries which triples their current capacity and drastically shortens the necessary charging time.  The implications for EV and HEV use is obvious:  more power and better regenerative braking from the same battery pack.

According to certain authorities, the average commuter travels 22 miles a day or less; this means that a car which can travel 22 miles on electric power alone can eliminate these people's fuel requirements for work travel, and cut total fuel needs by as much as 80%.  Longer all-electric range translates to less fuel use.

According to EPRI, a compact electric vehicle would require about 250 watt-hours per mile of range (it is not clear if this is measured at the charger input, the charger output, or between the batteries and the motor).  Others differ; AC Propulsion claims 205 Wh/mile for their tzero (presumably as output from the batteries), while Commuter Cars says a Tango would need about 180 Wh/mile.  For a slightly larger vehicle, the EPRI figure 250 Wh/mile seems to be reasonable for a BOTE analysis.

Range is the other figure.  30 miles is well over the average commute, and would certainly capture the 80% reduction in fuel requirements projected by analyses which find 22 miles is sufficient.  To obtain 30 miles range at 250 Wh/mile and 80% discharge, a battery would require a capacity of 30 * 0.250 / .8 = 9.375 KWh; call it 10 KWh even, for simplicity's sake.

The last element is power.  To meet consumer demands, a car will probably need at least 100 horsepower, perhaps 150 horsepower.  This means that the battery must be able to supply 75 to 112 kilowatts of power for acceleration.

For batteries, I like to look at batteryspace.com.  Their best Li-ion offering at the moment is a pack of 50 cells in the 18650 configuration (18 mm diameter by 65 mm long), which store 2000 milliamp-hours (2 amp-hours) at 3.6 volts nominal; for this they're asking roughly $5.00/cell.  The 50-pack is specified at 81 ounces, or roughly 45.6 grams/cell.  The specifications say that they are limited to a 2.5 C (5 amp) discharge rate.  Suppose that Altair's electrode technology can triple this to 7.5 C; at that rate, a 10 kWh battery would be able to supply 75 kW_peak, nearly as much as a typical NiMH battery.

For NiMH, the cost leader is a 10-pack of C cells, 4500 mAH at 1.2 volts nominal for $3.30/cell.  Assuming a 10 C discharge rate, a 10 kWh pack  would be able to supply 100 kW_peak.

I chose to assume two different configurations:  a commuter car with 75 kW (100 HP) of power, and a sport model with 112 kW (150 HP) of power, with 30 miles minimum all-electric range at 100% discharge. My calculations came out like this:

Salient points:

• Li-ion batteries are getting very close to NiMH in cost per unit energy.  (This is new in my experience.)
  
• The commuter configuration with the NiMH battery sits right at the "sweet spot"; it has neither excess power nor excess capacity for the range requirement.  This is partly due to the cells chosen; some NiMH cells have much higher discharge rates (up to 20 C) and could provide very high performance for similar weight and only slightly greater cost.
• Cars using Li-ion batteries are power-limited and require greater battery capacity to meet performance specifications.  This adds cost.
• The Li-ion batteries make up for this with a substantially greater all-electric range.
• The Li-ion batteries are also substantially lighter, by as much as an adult passenger's worth for the sport configuration.
• Either Li-ion car would probably be able to run entirely on electricity for a large majority of most user's driving.
  

What can we expect in the future?

• Cost of Li-ion cells will continue to fall.  If they follow the standard experience curve of 20% cost reduction for every 2x increase in cumulative production, an 8x production increase will see the cost of the commuter battery close to $3500.
• Cost of NiMH will also fall, but probably not as fast.
• Li-ion will probably be the cost leader for both energy/$ and power/$ in a few years.
  

• Cost of a Li-ion battery will approximate the cost of a gasoline drivetrain when it hits the $2000-3000 range.
• This requires about a 2.5-4x decrease in cost, depending on configuration.
• We can expect this at a 16x to 64x increase in cumulative production.
  
• Use of cells for traction batteries will consume far greater volume than portable electronic gear, and would increase production much more rapidly than the current trend.

¶ 3/02/2005 11:56:00 PM