The following summarizes what I've learned in general about NiMH and Li-Ion
batteries as a result of:
* a full year of investigations of candidate batteries for Prius (and
soon Escape) conversions
* conversations with vendors
* responses to questions from journalists

While many people are excited about announcements of new batteries and new
battery technologies, this summary restricts itself to existing products.

First, here's a roundup of what Toyota has said about batteries.
As the interest in PHEVs has grown over the recent months, most of their
other objections have been fading away, and they seem to be left mainly
with continuing questions about battery performance and costs. These quotes
are culled from a larger roundup of automaker comments found at
http://autos.groups.yahoo.com/group/calcars-news/message/80 (which was
posted before the IEEE and LA Times Magazine stories came out).

TOYOTA STATEMENTS ON BATTERIES IN RECENT NEWS STORIES ABOUT PRIUS+:

Mr. Hermance of Toyota said that batteries today were not durable enough to
handle the wide range of charging up and charging down that a plug-in
hybrid would need, calling that the most damaging thing you can do to a
battery .... (NY TIMES)
Toyota spokeswoman Cindy Knight: "We think there need to be breakthroughs
in battery technology to make it commercially viable." (WIRED.COM)
"So you have a higher up-front cost, a heavier vehicle that gets less
fuel economy with less performance, and the prospect of replacing the
battery during [the car's] life," he says....Toyota's Hermance
insists that, barring a spectacular breakthrough in battery
chemistry, the cost of nickel-metal hydride batteries will remain around
$1100/kWh for the foreseeable future. He concedes that the Prius's
nickel-metal hydride battery packs have become significantly cheaper since
Toyota began producing the car for the Japanese market in late
1997--power densities have gone up, allowing the car to get the
same acceleration with a smaller battery pack. But energy density hasn't
really improved, so energy storage remains as expensive as ever. (IEEE
SPECTRUM)
Hermance says that while the PHEV concept has merit, it won't work with the
current generation of lithium-ion batteries, which, while powerful, are
both too expensive and temperamental for use in mass-production cars.
Depending on their chemistry, lithium-ion batteries tend to get really hot
-- thermal runaway, it's called—and, as the military well knows, to ignite.
"The betting line of developers is that a lithium-ion battery of sufficient
cost, durability and safety is three to five years away."... (LA TIMES)

MY GENERAL SUMMARY AND VIEWS:

Having read the LA Times article, I got the impression that Dan Neil is
very impressed with both the technology and the vehicle, even though he
believes Toyota's line that the batteries aren't ready yet. Actually, I'm
sure Toyota is right in this sense: specific cells that meet an auto
manufacturer's needs for size, performance, safety, price, reliability,
and lifetime have not yet been lifetime tested in deep discharge
vehicle applications, qualified as an automotive component, and placed
in high volume production. NiMH technology that can meet the challenge is
here now (see my remarks below about costs), but a battery manufacturer and
an auto company would have to simultaneously commit to a several-year
process of final design, proof and qualification, tooling, and production
ramp-up.

To quickly create a PHEV option or upgrade to existing hybrid vehicle
designs,
this several-year-long battery productization process would have to
occur in
parallel with, rather than before, the several-year-long PHEV power train
design and production ramp-up. Also, it would require an auto
manufacturer's
commitment to such a power train to fund that battery productization.
Toyota's statements are therefore overstated (as ours may be in
the opposite direction), as they imply that years of further battery
research and development are necessary before the final design/production
qualifying phase.

Li-ion chemistry, though less mature, has a weight advantage over NiMH.
Though in very high volume production for small sized applications like
laptop computers, its use in vehicles is new -- and there are safety issues
due to possible thermal runaway that must be addressed in battery pack
design
-- requiring extra qualification efforts. However, the Electric Power
Research Institute (EPRI), which funded the DaimlerChrysler PHEV Sprinter
van low volume production effort, states categorically that they have
Li-Ion
batteries that are fully ready for mass produced vehicles.

Some of these comments may be subject to reconsideration (or reinforcement)
as soon as we hear more details from EPRI -- I understand they will be
reporting new information and conclusions about both types of batteries in
the near future, correlated with their development work on the
DaimlerChrysler Sprinter project.

One other point not to be dismissed. Toyota has claimed that the added
weight of PHEV batteries would decrease overall mileage. We found that this
would be true, though only of city, not highway mileage. However, we found
that Toyota's existing hybrid battery is sufficiently inefficient that if
the PHEV battery were to have, say, half the internal resistance, the
decreased battery losses would make up for the added weight, giving the
vehicle at least equal mileage to that of the lighter simple hybrid.

NICKEL-METAL HYDRIDE
NiMH may be more attractive than Li-Ion to automakers at this point, since
they are much more proven and are already in use in mass produced hybrid
vehicles, with 10 year, 150,000 mile warranties. Neither do they have the
safety concerns that must be addressed to use Li-Ion. Though the long
lifetimes of hybrid NiMH batteries are partially due to their restricted
state-of-charge (SOC) range, the RAV4EV fleet has proven that
1990's-technology NiMH cells can survive for 120,000+ miles in
deep-discharge EV use.

The high (Toyota said $1100/kWh) price of NiMH hybrid batteries is largely
due to their super high power handling requirements (the Prius's battery is
capable of amazing 30C (30 times the Ah rating) discharge and 20C charge
rates). An NiMH PHEV battery pack would need power-handling capabilities
1/2 to 1/4 that of a pure hybrid's battery, merely because it would be 4 to
7 times as large while handling only the same maximum power (though it
should do so more efficiently to make up for the PHEV's added weight).
Thus, a PHEV battery would be significantly less expensive per kWh.

At the other extreme, 9Ah, 1.2V, 11W-hr D cells (with insufficient power
handling capabilities even for PHEVs) sell for as little as $4.50 in
moderate quantities of e.g. 10,000 cells or 110 kWh (vs. 50,000+ kWh
needed for just 10,000 vehicles), yielding $410/kWh in this expensive,
small packaging and low volume purchase.

Given all this, my guess is that an automobile manufacturer would pay no
more than $600/kWh for PHEV NiMH cells capable of 150,000 miles' lifetime,
in production quantities. At $600/kWh, a 5.3 kWh pack for a PRIUS+ capable
of 40 miles' EV-assist range, would cost Toyota $3200. Remember, this is
in place of the $1400 they already claim to spend (at $1100/kWh) on the
current hybrid battery, yielding a difference of only $1800.

LITHIUM-ION:
As for battery costs, the best data I have to go on is that high quality
high-discharge-rate 2Ah (3.7V, 7.5W-hr) 18650 cells (used in laptop
computers) go for around $3 in moderate quantities of e.g. 10,000 cells
or 75 kWh (as opposed to 50,000+ kWh needed for just 10,000 vehicles).
This amounts to around $400/kWh. While packaging, electronics, and safety
systems will add to a pack's cost, automotive volume purchasing and larger
cell designs could easily be expected to cut costs a similar amount,
leaving
an auto manufacturer with a cost of around $400/kWh in the near term --
presumably much less in the long run, as Li-Ion raw material costs are low
and there is no single patent holder for the chemistry, unlike for NiMH.

At this near-term price, a 5.3 kWh Li-Ion PRIUS+ battery pack capable of a
40 mile EV-assist range would cost Toyota $2100. Subtract $1400 for the
existing hybrid battery, and you get an increment of only $700!

Actually, as of now, Li-Ion cycle life in an EV is unknown due to limited
real-world experience. Some manufacturers are claiming 2000 deep discharge
cycles for their cells in the laboratory. This translates to around 6 years
of full discharge every day, or maybe 80,000 miles.

Additionally two things are known to maximize Li-Ion cycle life: holding
maximum charge voltage to slightly below the normal 4.2V (thereby
sacrificing some of the cells' capacity) and keeping maximum cell
temperatures well below their normal 60 deg C rating. Both can be easily
provided for in a PHEV, the latter e.g. via liquid cooling (already
available
in a Prius, as its electronics is liquid cooled).

One more helpful factor: battery performance normally decreases slowly, and
end-of-life is normally specified as when capacity is reduced to 80%.
However, a PHEV, unlike a pure electric vehicle, could easily perform well
with e.g. only 60% of its battery's original capacity. Using this criterion
could as much as double the battery's lifetime.

Finally, the only way to discover and verify traction battery lifetimes is
to build and drive a fleet of vehicles, as all the bench testing in the
world will not accurately emulate real-world driving. Then, because Li-Ion
technology is improving so rapidly, by the time the batteries are
approaching
their end of life, new ones may already be expected to last dramatically
longer. NiMH cells built in the late 1990s for the RAV4EV fleet were
expected
to last for just 75,000 miles, but are still going strong after over
120,000
miles. Will the same thing happen for Li-Ion? No one knows, but, due to
national security and global warming issues, we don't have the decades it
would take to find out the hard way; we must find a shortcut.

Even if Toyota budgeted to replace every single PRIUS+'s battery during the
vehicle's warranty, and the price went down by only 33% during that time,
and servicing cost that 33%, their battery cost increment over that for a
standard Prius would still be no more than $2800 during the first year of
production.

+++++++++++++++++++++++++++++++++++++++++
Ron Gremban, rgremban@...
California Cars Initiative, a nonprofit organization:
http://www.CalCars.org
Moderator & Technical Lead
http://www.priusplus.org
PRIUS+ PHEV Conversion Group: http://groups.yahoo.com/group/priusplus
Newsletter: http://groups.yahoo.com/group/calcars-news
+++++++++++++++++++++++++++++++++++++++++