EVs may make sense for places like California where there is a problem with air pollution due to weather inversions and smog and there is a wealthy sector of the population that can afford EVs. However, for the rest of us, they are not a great choice, even with generous subsidies. Range anxiety is still a concern although that problem is likely to go away in the future, likely in a few years, as more powerful batteries and new battery types such as solid-state batteries with longer ranges and faster charging times go on the market. The upfront cost is a big factor for most people and will continue to limit sales.
Mark P. Mills,
a physicist and engineer, who has long done research and writing for the
conservative think tank Manhattan Institute regarding the real-world challenges
of the energy transition, published a paper in July 2023, Electric Vehicles
for Everyone? The Impossible Dream, details the real challenges of EV
ramp-up, especially resource intensity and consumer uptake. This blog post is
mostly a summary and review of Mills’ paper.
Mills makes
two key points in the executive summary: 1) No one knows how much, if at
all, CO2 emissions will decline as EV use rises – it is a rough estimate. 2)
No one knows when or whether EVs will reach economic parity with the cars that
most people drive – EV cost is tied to critical minerals costs, which
fluctuate according to supply and demand.
He points out that
one-third of EVs sold are hybrids and that two-thirds of EVs sold in the U.S. in
2022 were Teslas, which are basically luxury cars for carbon-conscious rich
people, I mean who can afford a Tesla? Mills calls out the “ICE
prohibitionists.” Much like the anti-fossil fuel crowd, these advocates are
profoundly unrealistic. The vast majority of primary energy production is
fossil fuels, and the vast majority of mobile vehicles are powered by ICE
engines. While it is fine to advance more non-fossil fuel energy and more EVs,
problems emerge when there are attempts to mandate that advancement and put unrealistic
time limits on the mandates. He points out that for the average household personal
mobility is the biggest cost after a mortgage. “A car is the single most
expensive product that 98% of consumers ever purchase.”
ICE Emissions vs. EV Emissions
While it is
easy and straightforward to determine the emissions of ICE vehicles, this is
not so for EVs. After manufacturing emissions are determined ICE emissions are
mainly operational due to the burning of gasoline or diesel. EV emissions are
all related to manufacturing but their vastly higher resource requirements in the
form of critical minerals require much more work to determine the
manufacturing emissions. Battery minerals are what powers EVs and these
battery minerals are often rare and difficult to obtain and process. One might
also consider that the upfront emissions related to materials acquisition and
manufacture of EVs are much higher than the upfront emissions of ICE vehicles.
Mills points
out that EV material acquisition emissions are all rough estimates, but I think
these can be put into general ranges for each mineral so that the estimates are
not as “unknowable” as he seems to suggest. Mills writes:
“While there are dozens of variations, a typical EV battery weighs about 1,000 pounds and contains about 30 pounds of lithium, 60 pounds of cobalt, 130 pounds of nickel, 190 pounds of graphite, 90 pounds of copper, and about 400 pounds of steel, aluminum, and various plastic components.”
“These five elements total ~100,000 pounds of ore to
fabricate one EV battery. To properly account for all the earth moved, there’s
also the overburden, the materials first dug up to get to the ore; depending on
ore type and location, it averages three to seven tons of overburden removed to
access each ton of ore, thus ~500,000 pounds total. The exact number varies for
different batteries and mines. Note that this doesn’t include large quantities
of chemicals to process and refine the ores, or the mining/refining for the
other 400 pounds of battery minerals used (e.g., steel, aluminum).”
Most of the materials for ICE vehicles are iron, steel,
and plastic. These are mostly produced domestically in the U.S. and there is
much better transparency about the emissions associated with their mining and fabrication
than there is with many of the EV minerals produced and processed around the
globe with non-transparent China leading the pack by a vast margin.
Mills gives
the IEA’s life-cycle emissions for EVs vs. ICE vehicles shown below. He points
out the reliance on assumptions evident in the black error bars which
acknowledge that there is a chance that lifetime EV emissions could be
as high as ICE emissions in some cases. Those error bars are huge and that
lends credence to the possibility that EV emissions could be much higher than
predicted and shows a level of uncertainty that should be considered to be unacceptable.
Mills also
points out that the IEA uses a “40 kWh battery pack, which is half the size
of the batteries in most popular EVs” and that the trend in the future is likely
to be bigger batteries with more range since range parity with ICE vehicles has
yet to be achieved. However, that may not be the case, since newer battery
types may end up being lighter and less mineral-intense. He also claims the IAE
analysis ignores that greater use of aluminum in EV frames reduces the weight.
Aluminum mining and processing are emissions intense, so he is claiming that
there are missing emissions here.
He gives
another graph from a 2022 study that compares emissions on the renewables-heavy
EU grid from Volkswagen diesel and EV models that shows that the EV emissions
are higher than the diesel emissions till the odometer reaches about 75,000
miles, again showing that EV emissions are front-end loaded. The emissions
would be a little higher on a global average grid. He shows a similar graph for
Volvo models. The global average for emissions parity seems to be about 55,000
miles at a glance but this could be off either way.
Mills next
gives the known unknowns, the uncertainties of EV emission determinations. He
gives ten of these which I will examine: 1) size of the battery
pack – he notes that most past analyses have been based on smaller and
lighter batteries but with the need for range parity a true equivalent comparison
with ICE vehicles will require using heavier batteries that use more minerals
in those analyses. 2) location of the mines – he notes
that 80-90% of the mines are located outside of the U.S. and that for each
mineral the emissions can vary quite a bit by where they are mined, especially
for copper (by up to 2 times) and nickel (by up to 3 times). 3) location
of minerals processing and refining facilities – Mills suggests that
some previous calculations may not consider that China processes 50-90% of the
world’s energy minerals. I would add that much of that processing would likely
be powered by coal. I am unsure whether or not previous analyses take this into
account, and he is vague on this point. 4) location of the
battery and EV assembly factory – he points out that assembly plants in
Norway with 90% hydroelectric power would have far less emissions than assembly
plants in China where two-thirds of all power is provided by coal. He also
notes that half of all EVs produced globally in 2022 were made in China. 5)
battery chemistry – here he notes that even with about a dozen different
battery chemistries their resource intensities are more or less the same. He
also notes that lithium-iron-phosphate (LPO) batteries, popular in China, do
not use cobalt and nickel but do have about 20% less energy density than those
that do so those advantages are offset by the loss in energy density. He doesn’t
mention some other new battery chemistries possible in the future that may have
the ability to reduce resource intensity without giving up energy density. 6)
material for the rest of the vehicle – here he asserts that EVs use far
more copper in the electric motor and wiring and far more aluminum in the
frames than do ICE vehicles and that many past analyses do not take this into
account. 7) emissions from EV power electronics – he notes here
that EVs use about 200% more electronics for power management and manufacturing
these electronics is more energy and emissions intense. He notes a study suggesting
that accounting for this is equivalent to driving an ICE vehicle about 3000
miles. 8) battery life span – this assumes the fact that fast charging
shortens battery life relative to slow charging. Another factor to consider
that I will mention is that the hype over V2G – using EVs to power the grid at
high energy demand times – also degrades the battery and shortens battery life. Some studies should assume
two batteries per vehicle. This is not hyperbole. My Toyota Prius required a
new battery, in this case though, a refurbished one so no additional resource
intensity. As the life of an EV battery is currently 8-10 years, some EV owners
will require two EV batteries for the life of the car. If we consider that some
ICE vehicles are still on the road after 30 years, even three batteries are not
out of the question. I do believe battery life will increase in the future so this will not be a concern at some point. 9) total miles –
here he gives some data that on average EVs are driven half the miles per year
as ICE vehicles. That means that it takes longer in time for an EV to achieve
emissions parity with ICE vehicles. That would make an EV's emissions per mile
higher (before parity). 10 Ice fuel efficiency – here he gives
the interesting point that ICE fuel economy is improving and is expected to
continue to improve so that predictions for 2030 are for a 30-50% improvement
in efficiency, which translates to equivalent lower emissions per mile. I plan
to write a post about ICE improvements and research in the near future. He
gives a graph for this, but it starts in 2018. If he had started in 2022/2023 which
he could and should have, the predicted improvement would be 10-30%. However,
in terms of the vehicles on the road, averages, and penetration of better
mileages in those averages, it is perhaps not a bad prediction.
The biggest
emissions from EVs by a very wide margin are the mining and processing of minerals.
He rightly points out that recycling of EV minerals will remain irrelevant and a
tiny percentage of total minerals for quite a long time. The main reason is
cost. The IEA estimates recycling may account for just 1-2% of EV minerals by 2030. Mining
and processing EV minerals also have many negative environmental impacts and
human rights issues, often amidst a significant lack of transparency that would
not be tolerated in the U.S. and other developed countries. While Congolese cobalt
mining, so-called dangerous artisanal mining by kids is a huge human rights concern,
the issue of water use by South American lithium brine operations is a minor environmental
issue as the area is extremely dry unliving desert and the water extracted from
the ground is twice as salty as seawater so not consumable.
Next, he
considers the fact that upstream EV emissions are actually rising. This is because mineral ore grades are lowering as high-grade ores are used up
so lower-grade ores are being mined more. This results in higher emissions
per ton of ore mined. This is particularly true for copper and nickel. Lower-grade ores simply mean more rock is mined to get the same amount of minerals.
This equals higher emissions. While making mining less energy intensive is
being pursued, even the IEA acknowledges that minerals production is likely to
get more energy intensive. Mills says “likely” is an understatement, and that
minerals production will definitely get more energy intensive. He gives data
that copper, nickel, and lithium production are all getting more energy-intensive, which also means more emissions intense. He also considers trends in
electrifying mining processes and technological trends in new battery chemistries and other battery breakthroughs. Here he points out that while such
improvements and breakthroughs are possible, they are not at all likely in a time
frame within a decade or two, while EV adoption is being pushed to be accelerated
more and more. The reality is that even with considerable incentives, the high
costs of EVs are slowing adoption and potentially affecting the economics of
auto manufacturers that are losing money on EVs even while they are investing
heavily. The recent autoworker strikes of the Big 3 in the U.S. suggest to me
that the Big 3 may be considering the potential of losses due to overfocus on
EVs in the near term in their negotiations. Indeed, I just heard a radio news segment that auto companies were citing the costly shift to EVs as one reason they are resisting UAW demands. The shift to EVs is also affecting autoworkers in other ways. More of them are losing their jobs since fewer workers are required to build EVs compared to ICE vehicles. Some of those workers can work at new battery plants but it seems that fewer workers will be needed overall.
Next, he
considers more uncertainties with EV emissions such as emissions of charging. These
depend on what powers the grid where they are charged and can vary dramatically.
In places like Wyoming or West Virginia where over 90% of the grid is powered
by coal, EV charging emissions are much higher than say Germany or California. When an
EV is charged is also a consideration since solar drops off at night and fossil
fuels provide the and in other more power in high-demand times. Other
considerations include real-world testing versus “sticker” mileage
calculations. He points to one study that showed that sticker mileage is more
accurate for ICE vehicles than for EVs, that actual EV mileage is about
12.5% lower than sticker, while ICE mileage is about 4% lower than sticker.
Car &
Driver’s Dave Vanderwerp describes why EVs are tested the way they are. They do
pre-test charging and standing-start acceleration in a standard way. They do a
75mph range test in a standard way on the same road to compare ranges for different
EVs. Vanderwerp notes that some EVs come in a full 20% below sticker range but
most German manufactured models come in at range or slightly above. Tesla has
been accused of vastly overstating range. He notes: “A range discrepancy
between EVs from different companies might not be as extreme as the numbers
would suggest. "400 miles of stated range for a Tesla and 300 miles for a
Porsche is pretty much the same number at real highway speeds.” One might ask the question - if Volkswagon paid a heavy price in the tens of billions for cheating on emissions tests, then why do companies like Tesla get to essentially cheat on range tests without consequence? A lower range does mean higher emissions per mile and even if those emissions are less than ICE emissions, it still gives a deceptive emissions determination.
Another
consideration is that EV mileage drops in lower temperatures – that EV mileage
drops by about 30% at 20deg F, while ICE mileage only drops by about 5% at that
temperature. In addition, an EV does not produce much waste heat while an ICE
vehicle scavenges waste heat from the engine to heat the interior of the car,
although he doesn’t point out that the opposite is likely true for air conditioning,
although heating is more energy intensive than A/C. In any case, EVs are significantly
more energy and emissions-intensive in colder climates in general.
Next, he
discusses parity, including cost parity, operating costs parity, convenience
parity, and parity in terms of supply chains and environmental impacts. EVs
still cost significantly more than ICE vehicles. However, they are cheaper to
operate, cheaper to fuel, and require far less regular maintenance including no
oil changes. Future costs of EVs are dependent on battery minerals costs which
in turn are dependent on costs of minerals and supply chains from foreign
countries. While minerals costs had been dropping, they have risen or stagnated
in the last few years for a number of reasons, some temporary and some not. The
graph below simply shows that the global mining industry is not well ready for
a deep acceleration of energy transition buildout which includes EVs.
Mills thinks that mineral mining and processing monopolies,
cartels, and protectionist rules will stay the same or increase in the future,
even as countries like the U.S. seek to develop domestic supplies. He shows some
predictions that copper and nickel prices are set to rise from the mid-2020s to
the 2040s.
He also
considers the need for massive new charging infrastructure in an accelerated EV
scenario. When I had a plug-in hybrid EV, I had a Level 2 charger installed at
home. I did some of the work myself which lowered the cost. It was great, to be
honest. But if tens of thousands of people in a small area each have one
installed it would strain the local grid. At current EV ranges there is a need
for significantly more available chargers per area than for gas stations. These
factors require the utility to upgrade equipment. Public chargers have added
costs, especially fast chargers.
As I have
argued in the past, if and when EVs are adopted at levels exceeding ICE cars
then all the money lost in gasoline taxes that funds many things would have to
be recouped somehow. For a few years now states have been enacting EV ownership
taxes, usually at a few hundred dollars per year. Massive EV expansion will
require grid upgrades and grid transmission expansion. This is in addition to
those needs for wind and solar expansion so grid reliability could also be
affected.
He mentions that
EV owners report a higher incidence of problems than ICE owners in the first
year of ownership, often electrical issues. I can say having owned a plug-in hybrid
EV for over a year I experienced zero issues. He also considers battery
waste disposal costs which could end up being bolted on to the purchase price.
While pro-green
enthusiasts may say that EVs make us more energy-independent they rarely
concede that they make us far more mining and processing-dependent. Minerals
production and processing are even more concentrated than oil production. The
graph below shows just how dependent we are on China for EV minerals and mineral
processing. The graph may be a few years old since now the U.S. exports more
LNG than Qatar which is not as depicted.
He notes: “The U.S. today is dependent on imports for 100% of some 17 critical minerals, and, for 28 others, net imports account for more than half of existing domestic demand. Assembling batteries (or solar hardware) here creates underlying dependencies equivalent to assembling conventional automobiles domestically but importing all the key parts and all the fuel.” We should learn from the mistake of EU dependence on Russian oil & gas and not get overly dependent on Chinese minerals and processing. While it is true that we are working on domestic and friendlier country supplies and processing capacity, that is likely to take a long time and not likely to compete with Chinese prices kept low by government subsidies.
While I am
annoyed by guys with big trucks “rolling coal” in front of me while I have to
breathe it – this happened to me just yesterday – I am also bullish on efficiency
improvements to ICE vehicles that could render EV benefits less beneficial. I
also like electric motors, their efficiency, the ability to charge up at home
at night, to get fantastic mileage, a quiet ride, no oil changes, etc. I don’t
have the data, but I tend to disagree with Mills' suggestions about EV reliability.
I think they are probably much more reliable than he suggests. Though just a hybrid, buying a
Prius on the first day of 2006 turned out to be one of the best financial
decisions I ever made for a number of reasons. However, we do have to be
realistic. While I don’t agree with all of Mills’ assessments it was a great paper,
and I would guess based on all of his points that on that first graph, the real-world
emissions advantage of EVs is up on that error bar – my guess is that the
reality is that an EV could be as much 75% as emissions intense as an ICE vehicle
and that conclusion added to the other issues and inconveniences of EVs
suggests that currently a lower mileage ICE vehicle, a hybrid, or a plug-in
hybrid are much better choices. I realize that this is almost twice as much as
other studies, but Mills’ work strongly suggests that we are significantly underestimating
the emissions intensity of EVs. That may not be true in the future, but I think
it is now. In any case, I can no longer afford anything but a used ICE vehicle
anyway and that is true for many. It is also true that EV subsidies have gone
vastly disproportionately to wealthy people since one has to be wealthy to
afford a Tesla.
McKinsey & Company Thinks EV Emissions
Intensity Can Be Steeply Reduced During the Next 5-10 Years
McKinsey & Company released a study examining EV decarbonization efforts in February 2023. Regarding the variations in EV emissions, they noted: “Emission levels from EV battery production depend on a variety of factors, including design choices, vehicle type, range, and freight requirements, as well as production and sourcing locations. The energy sources used to produce various battery components are one of the biggest factors explaining the wide variation in the carbon footprint of different OEMs (original equipment manufacturers).” The graph below compares EVs and ICE emissions by material and type.
The authors believe
that steep reductions in carbon emissions are possible in the next 5-10 years. They
cite battery chemistry, production technology, the selection of raw material suppliers, and transportation routes as key areas where emissions can be
reduced. What powers the mining, manufacturing, and delivery processes is also a
factor. The graph below shows variances by region and production sector within
each region.
They note that
China makes 70% of global batteries and has the most emissions-intensive
manufacturing. Citing incentives like the EU's carbon border adjustment mechanism
and the U.S. Inflation Reduction Act they think it is feasible to drastically reduce
EV emissions by 2030. I am somewhat skeptical knowing all the challenges such
as supply chain issues, regulatory and permitting issues, and public opposition
issues rampant with renewables acceleration. They are bullish on
electrification in the near term. I don’t disagree with their assessment, but I
do think they are too optimistic with their timeframe. I think it will take
another 5 or more likely 10 years. I do think that range issues, charging
issues, emissions issues, battery waste disposal issues, and safety issues will
all be solved in time. I also think that the rush to ban ICE vehicles and to
make other mandates is both unnecessary and bound to fail for a number of
reasons.
References:
EPA is
ignoring the glaring problem with dirty electric vehicles. Benjamin Zycher. The
Hill. EPA is ignoring the glaring problem
with dirty electric vehicles (msn.com)
Electric
Vehicles for Everyone? The Impossible Dream. Mark P. Mills. Manhattan
Institute. July 2023. Electric Vehicles for Everyone? The
Impossible Dream | Manhattan Institute (manhattan-institute.org)
The
Hidden Carbon Footprint Of Electric Vehicles. Quina Baterna. Slash Gear. August
19, 2023. The Hidden Carbon Footprint Of
Electric Vehicles (msn.com)
The
race to decarbonize electric-vehicle batteries. McKinsey & Company. February
23, 2023. The race to decarbonize
electric-vehicle batteries | McKinsey
Electric
Vehicle Myths. U.S. EPA. Electric Vehicle Myths | US EPA
The
Hazards of Electric Car Batteries and Their Recycling. Taotianchen Wan and
Yikai Wang. 2022 IOP Conf. Ser.: Earth Environ. Sci. 1011 012026. The Hazards of Electric Car Batteries
and Their Recycling - IOPscience
Is it
Ethical to Purchase a Lithium Battery Powered EV? Ronald Stein, P.E. June 13,
2022. Is it Ethical to Purchase a Lithium
Battery Powered EV? – The Heartland Institute
Electric
cars are made of pollution and human misery. Kathryn Porter. The Telegraph.
August 20, 2023. Electric cars are made of pollution
and human misery (msn.com)
EVs
Fall Short of EPA Estimates by a Much Larger Margin Than Gas Cars in Our
Real-World Highway Testing. Caleb Miller. Car and Driver. August 20, 2023. EVs Fall Short of EPA Estimates by a
Much Larger Margin Than Gas Cars in Our Real-World Highway Testing (msn.com)
Why We
Test EVs the Way We Do. Car and Driver. Dave Vanderwerp. October 3, 2022. Why We Test EVs the Way We Do
(caranddriver.com)
No comments:
Post a Comment