Shanghai Plans to Provide Up to 20% Subsidy to Buyers of More Fuel Efficient Vehicles
Pre-Orders in Japan for 2010 Prius to Top 80,000 By Debut on Monday

GM Quantifies CO2 and Fuel Consumption Reductions Via E-REVs And PHEVs, As Compared To “Conventional” Hybrids

by Jack Rosebro

Energy sources, paths, storage media, and propulsion systems available or in development. “FCEV” refers to all fuel cell vehicles, including E-REVs and fuel cell hybrids. Adapted from Tate et al. (2009). Click to enlarge.

General Motors has released a white paper that evaluates the CO2 reduction potential of extended-range electric vehicles (E-REVs) as well as plug-in hybrids (PHEVs), in combination with multiple vehicle charging scenarios, as compared to conventional hybrids. The paper was presented by authors Ed Tate and Peter Savagian at last month’s SAE 2009 World Congress in Detroit.

In the paper, the GM team broke down CO2 and fuel consumption reduction potentials into several categories:

  • The introduction of E-REVs and PHEVs to existing grids, displacing petroleum use;
  • The influence of charging availability and consumer behavior on CO2 reduction;
  • Changes in power grid mix as well as vehicle stock; and
  • The impact of selective and voluntary consumer behaviors on CO2 reduction.

Using 2008 estimates from the US Energy Information Administration (EIA) as well as the Organization of Petroleum Exporting Countries (OPEC), Tate and Savagian note that global primary energy use is projected to rise by 50% over the next 22 years, with worldwide vehicle population increasing by 60% over the same period of time. Per capita vehicle ownership is expected to rise by one-fourth during this time—from 12% to 15%—with private vehicle ownership functioning as “both a means to, and a dividend of, economic development.

Taking note of pending vehicular greenhouse gas (GHG) regulations such as California’s Assembly Bill 1493 (Pavley), the authors observe that GHG emissions from conventional vehicle powertrains are already regulated to some extent by fuel economy regulations in many countries, and that GHG regulations would function as additional de facto fuel economy regulations, as “the use of fuel and the production of CO2 are simply different measures of the amount of combustion” of fuel. To simplify compliance, they argue, “one regulation is sufficient to control both measures.”

CO2 Displacement Through Vehicle Electrification

“Electrification” is defined in the paper as referring to the relative increase in the electrical content and magnitude of a vehicle’s motive power as compared to the amount of energy that is derived from petroleum-based energy sources.

In a previous paper (The Electrification Of The Automobile: From Conventional Hybrid to Plug-In Hybrids To Extended-Range Electric Vehicles, SAE 2008-01-0458), Tate, Savagian, and Michael Harpster (also of GM) had evaluated the potential for petroleum displacement as well as emissions reductions, using vehicles similar to those in the current study.

However, that work had considered scenarios involving only one battery pack charge per vehicle per day. The current paper expands this work to several scenarios that involve multiple charge opportunities per driving day, and advocates a well-to-wheel analysis of vehicle-based CO2 emissions, rather than just tailpipe emissions, to inform several questions:

  • What kind of plug-in vehicles will enable CO2 reductions most effectively?
  • What is the most effective charging infrastructure for such vehicles?
  • What are the best test procedures and measures of merit for plug- in vehicles?

The study compared petroleum and CO2 displacement opportunities of three types of vehicles: PHEVs, E-REVs, and conventional full hybrids, on the assumption that they would be mid-sized vehicles equivalent to a 2009 Chevrolet Malibu sedan and operated in a variety of regional driving patterns using a variety of regional electrical power source profiles. Charge-sustaining fuel economy for all vehicles is assumed to be 36 miles per US gallon. The vehicles are defined as follows:

  • Hybrid Electric Vehicle (HEV): A 2.4L four-cylinder gasoline engine and a 1.8kWh, 30kW battery pack provide motive power through GM’s two- mode FWD hybrid system (earlier post). However, all motive power, including that from the battery pack, ultimately comes from the combustion of liquid fuel.

  • Plug-In Hybrid Electric Vehicle (PHEV): A 2.4L four-cylinder gasoline engine and a 2 to 16 kWh (2, 4, 8 and 16), 53kW battery pack provide motive power through GM’s two-mode FWD hybrid system. Motive power comes from the battery pack when available and when power and speed requirements are below 53kW and/or 56 MPH. Once the battery pack is depleted to a given state of charge (SOC), the vehicle operates as a conventional hybrid vehicle. The range in battery pack capacity allows for evaluation of the effect of battery pack size on liquid fuel consumption.

  • Extended-Range Electric Vehicle (E-REV): A 1.4L four-cylinder gasoline engine and a 2 to 16 kWh (2, 4, 8 and 16), 120kW battery pack provide motive power through GM’s E-REV propulsion system. When the battery pack is depleted to a given state of charge (SOC), the vehicle operates as a conventional hybrid vehicle. There are no speed or power constraints that force the engine to run; fuel is burned only after the battery pack is depleted.

Study Methodology

The researchers began by comparing behavioral models of the three vehicles with trip information from the 2001 National Household Travel Survey (NHTS) of US travel behavior, conducted by the Department of Transportation (DOT). The behavioral vehicle models profile energy consumption at the fuel tank, the battery pack, and the charging station, by defining:

  • Consumption of electrical energy in charge-depleting mode;
  • Fuel energy used, if any, during charge-depleting mode;
  • Consumption of fuel in charge-sustaining mode;
  • Battery pack capacity;
  • Charger power rate;
  • Charger efficiency;
  • Restrictions on charging time; and
  • Restrictions on charging opportunities.

The models assume that plug-in vehicles start each day at full charge.

NHTS data includes trip distances, travel times, travel speeds, trip origin locations, and trip destination locations for approximately 50,000 vehicles over more than 130,000 trips, and evaluates vehicles which move as well as vehicle that remain parked during the travel day. Data was collected at a rate of once per minute per vehicle. NHTS data shows that most cars are left at home during the weekday; the vehicles that are used to commute to work are obvious candidates for recharging after their commute.

Three charging scenarios were evaluated:

  • Charging exclusively at home, between the hours of 9 PM and 9 AM;
  • Charging at home, as above, plus unrestricted charging at work; and
  • Unrestricted charging at home and at work, as well as other opportunities.

Each charging scenario was broken down into three subsets, reflecting charging rates of 1.1, 3.3, and 6.6kW, (at potentials of 110, 220, and 400V, respectively) and assuming 90% efficiency for all chargers.

Data on primary power used to fuel electrical grids was taken from current and projected status of the EPA’s Emissions and Generation Resource Integrated Database (eGRID). In general, the study uses a 2005 EPA estimate of California average electric generation emissions (0.32 kg of CO2 per kWh); however, the EPA estimates that such emissions vary by region from 0.23 to 0.89 kg of CO2 per kWh.


Within any given driving profile, reductions in fuel consumption and CO2 production are primarily influenced by battery size, powertrain architecture, and charging scenario. If a relatively modest battery pack capacity is employed and combined with a restrictive charging scenario, PHEV and E-REV fuel consumption and CO2 reduction potential (as compared to a conventional hybrid) are roughly equal to one another.

An aggressive battery pack size coupled with an aggressive charging scenario can, however, reduce either plug-in vehicle architecture’s fuel consumption by as much as half. As battery size and charging opportunities increase, the rate of reduction of fuel consumption and CO2 production begins to favor an E-REV architecture.

Tate and Savagian also looked at the role of the electrical power grid in terms of projected peak power draw from a given population of vehicles as well as the total projected energy use by those vehicles while charging. Using the median charging rate of 3.3kW, for example, the peak expected power consumption of an 8kWh battery pack is 813 watts, which the authors equate to “a 50-inch plasma television set or a few high-wattage light bulbs.

“Your hot water heater will probably use more energy than an E-REV that travels 10K miles per year.”
—Pete Savagian

Peak grid loading is primarily affected by both charging locations and charging time restrictions. For example, the peak grid loading of a 4kWh E-REV that is charged only at home with a 1.1kW charger will be similar to peak grid loading if the same vehicle is charged with a 3.3kW charger at both home and work, even though the total amount of charge to the battery pack will be very different.

The authors conclude that charging scenarios are at least as influential as battery pack size when considering the ability of a given plug-in vehicle to optimize reductions of fuel consumption and well-to-wheel CO2 emissions.

If publicly available data sets that provided summary data similar to the NHTS [data] with second-by-second velocity information were available,” they add, “detailed simulations could be performed instead of the behavioral models used here.

Several additional conclusions were drawn from an analysis of the data:

  1. Fuel consumption and CO2 production of both PHEVs and E-REVs are lower than that of equivalent full hybrids, with expected fuel savings of around 70% and CO2 reduction of around 40%, assuming a grid power mix similar to that of the current California grid, and assuming that the vehicle is charged daily and driven an average of 100 miles or less per day.

  2. Fuel consumption and CO2 production of E-REVs—particularly those with larger battery packs—are lower than that of equivalent plug-in hybrids. Assuming a battery pack size of 8kWh for both vehicle types, E-REV fuel consumption is 22% lower and CO2 output is 8% lower than that of a PHEV, with 3.3 kW charging twice a day and a California-type grid power mix.

  3. Although the impact of PHEVs and E-REVs on grid loading is expected to be minimal, peak loading can be minimized by making charging widely available at work. Assuming that drivers are also charging at home, this is expected to further reduce a given vehicle’s fuel consumption and CO2 production.

  4. As the electrical grid shifts to renewables, grid-based CO2 emissions related to vehicle operation per mile traveled will decline, further decreasing well-to-wheel CO2 emissions from both PHEVs and E- REVs. A consumer preference for green and renewable electricity programs would leverage carbon emissions even further.


  • Tate et al (2009) The CO2 Benefits of Electrification: E-REVs, PHEVs and Charging Scenarios. (SAE 2009-01-1311)

  • Tate et al. (2008) The Electrification of the Automobile: From Conventional Hybrid, to Plug-in Hybrids, to Extended-Range Electric Vehicles. (SAE 2008-01-0458).



Good study but GM may not be around long enough to see the first results.

If GM and Chrysler end up being majority own and run by UAW members, they will be in Chapter 7 bankruptcy within 3 years.

Government (USA and Canada) will lose up to $50B with those two.

On the other hand, GM and Chrystler closing down may be the very best thing for Ford's survival and a helping hand for Toyota, Honda, Hyundai and others making and selling cars in USA and Canada.

It would certainly not be the end of the world. There will be no shortage of very good cars.


Without actually having read the report, I noticed that solar powered cars are not included in the analysis. It assumes that all energy used to propel the car comes from either the grid as electricity, or from the fuel tank on board the car. However, much energy can be derived from leaving your car parked in the sun all day. This quite simple feature could be relatively cheaply incorporated into new cars if mass production techniques were employed. Basically, you just lay down thin film solar panels on your bodywork and then cover with some kind of UV resistant epoxy. Then hook them all up to the battery charging system. Pretty simple, and shouldn't cost more than an extra couple thousand dollars at most, and that cost will decrease every year.

- average sun shining down: 800 W per m2
- average daily illumination: 8 hours
- solar panel efficiency: 25%
- area of solar panels on car: 4 m2
- therefore, energy available per day: 6.4 kW-hr
- estimate of actual energy delivered to batteries due to other efficiency losses: 5 kW-hr
- mileage of a good electric car: 10 kW-hr per 100 km
- therefore, daily range from solar panels: 50 km
- bring this down due to inclement weather: 35 km

That's a very substantial figure, and would cover most peoples' commutes. Considering that this is all based on existing technology available now, and all that needs to happen is for some car maker to incorporate it into their manufacturing, which seems very likely within 10 years, then this is something that needs to be factored into the analysis.

Nick Lyons

I'm curious as to why they didn't include pure EVs in their analysis.


"If GM and Chrysler end up being majority own and run by UAW members, they will be in Chapter 7 bankruptcy within 3 years."

It's hard to predict what will happen here. Contrary to popular belief, the UAW did not cause the demise of GM or Chrys. It was all caused by poor decisions made by upper managment. (including weak bargaining with the UAW)



Yes but your number are too rosy. 25% is elusive for affordable solar panel, the very best are about 23%. 800 whrs/m2 is for orthogonal orientation to the light, so elusive to get unless you have a tracking system. and 8hrs/day of this solar flux is rare. 0.1Kwhrs/Kms is also optimistic unless you have an Aptera type of sleek vehicle but then you won't stick 4 m2 of solar panel on it.

a more reasonable estimate is that covering your car roof of solar cell could give you 10-15 Kms/day, not bad but quite an expensive technologie for such a short distance that could be easily covered with a bike ...


Marc BC,

What you mention is true but it is such a corner case that would be such a small percentage of the customer base that it isn't worth talking about other than as a footnote lest the researcher be ridiculed for trying to push an unlikely scenario (for the main customer base) as the prime motivator for adopting the technology...

John Baldwin

GM many not know this but they have a subsidiary called Opel that makes a car called a Zafira CNG Turbo that runs on natural gas and also petrol with very low CO2/km, 500km range, good performance. So no need to waste approx 50% energy to convert natural gas into liquid or hydrogen or electricity. Across the world GM subsidiaries are making CNG cars, someone should point this out to GM folk in the US.

Even better, home refuelling devcices mean that the natural gas can be put in the car during the night, in future using surplus renewable energy.

Will S

Hybrid Electric Vehicle (HEV): A 2.4L four-cylinder gasoline engine
Plug-In Hybrid Electric Vehicle (PHEV): A 2.4L four-cylinder gasoline engine
Extended-Range Electric Vehicle (E-REV): A 1.4L four-cylinder gasoline engine

Since the Prius has been using a 1.5L engine, using a 2.4L for the HEV and PHEV skews the results so significantly that I don't even consider these results to be worth looking at. GM is clearly trying to build a case for how the Volt is so superior to the Prius that they won't dare use real figures.


The Big 3 thought there was no market for fuel efficient cars, but their lack of them drove them to bankruptcy. When gas prices shoot back up again (assuming the economy recovers), cars as Mark describes will indeed have a sufficient market. Watch the Aptera, Loremo, MiEV, and others pull market share away from the Declining 3.

Andrey Levin


You arguably right in your analysis. I would add that Ford (with tremendous debt) is nowhere to go but into Ch.11 reorganization, to shed off big chunk of the debt.

Interesting, that “collapse” of UK auto industry did not lead to reduction of vehicles produced in UK; it is stable over two decades at 1.7 million vehicles per year. The only difference is that these vehicles are produced mostly by Toyota, Honda, and Nissan (about 0.9 million), and most of remaining (Mini, Jaguar, Landrover, Vauxhall) is controlled by foreign auto companies. Engineering, manufacturing, assembling, etc. jobs are remaining into UK proper:

Probably the same will happen with indigenous American auto industry.



Here an idea:
Instead of putting the solar panels on the car where they weigh a ton and cost thousands to repair, why not have the solar panels feed into the grid? This means that you can also use other kinds of renewable energy such as wind and hydro, if you happen not to live in SoCal or Arizona.

If you want to get even more sophisticated, you can store solar energy in the form of ethanol or biodiesel. That way excess energy produced in the summer could be used in the winter. Solar receptors can be made locally in bio-factories that look suspiciously like fields.

Seriously, if you want some real-world numbers to "plug in" to your equations, look at the World Solar Challenge.
Experimental cars with huge solar panels and barely enough space for a driver can't make your numbers in the middle of the Australian Outback. Good luck getting a solar-powered car to even budge on a Chicago winter morning.


John even the old fuel cell systems of a few years back could beat a common cng fueled car as far as miles per unit of nat gas used. The current ones would blow em out of the water.


My numbers are correct.

The Tesla Roadster gets 9.1 km per kW-hr so I actually underestimated a bit (110 Wh/km)

Orthogonal solar irradiance is 1000 Watts. My 800 Watts may be a bit optimistic but not if it is summer and sunny. Average insolation for the whole world is 250 Watts but that includes night time, clouds, and polar latitudes, so what you'd expect on an average on a sunny day in typical situations is probably something like 500 - 800 Watts. Obviously in winter in Canada you aren't going to get these numbers.

Solar panels of 25% efficiency exist, actually they have experimental ones up to 40% and they are improving significantly every year. I don't understand why they can make them for my backpacking solar panel but they can't somehow put this on a car. Price will obviously come down, just like computer CPU power has.

Solar panels do not weigh a ton. I'd estimate the weight of solar panels would be about 15 pounds, and then maybe 20 pounds of epoxy to seal them up. And with regenerative braking you get a lot of that energy back.

Just like if you went back in time 20 years and told people what your laptop could do they'd laugh at you. Well, technology advances, and the solar energy for powering electric cars 50 km and beyond is right there shining down on us, it will inevitably happen, probably sooner than later.



The wikipedia page on the Roadster has this to say:
"n August, 2008, Tesla Motors reported on testing with the new, single-speed gearbox and upgraded electronics of Powertrain 1.5 which yielded an EPA range of 244 mi (393 km) and an EPA combined cycle, plug-to-wheel efficiency of 28kW·h/100mi (174 W·h/km, 3.57 mi/kW·h)."

Your estimate is very optimistic. It doesn't account for shadows cast by nearby buildings and trees.

Where I live (The Netherlands) an ideally placed solar panel can deliver 1 kWh per Wp (watt peak) per year. Placed horizontally, that figure drops to 800 Wh/Wp/yr. Factoring in the shadows I would estimate 600 Wh/Wp/yr a realistic maximum for a car roof mounted solar panel. (I forget about the dirt quickly accumulating on a horizontal surface, causing even more degradation).

I also fail to see how you can cram 4 m2 of solar panels on the roof alone. 2 m2 is the limit I guess. There is also the practical issues and costs related to covering a curved roof with flat monocrystalline solar cells.

Assume 20% realistic efficiency for commercial solar cells. That is 400 Wp that can be mounted on the roof.

400 * 600 = 240 kWh per year, enough for ~4 km/day.

In very sunny regions, that figure could double. The rest of the world will be somewhere in between.


Here's a hypothetical: what happens if zero point energy is allowed to develop? Eliminating the cost of traditional consumables and providing ubiquitous energy to the entire human population.

Would this retire all energy industries overnight? Would that be bad? Would the puppet masters tolerate the loss of their puppets?

I also fail to see how you can cram 4 m2 of solar panels on the roof alone. 2 m2 is the limit I guess.
There is also the hood and the trunk, as well as the windshield and rear window when you aren't driving. On an SUV you could find more than 4 m.
There is also the practical issues and costs related to covering a curved roof with flat monocrystalline solar cells.
Well, my backpacking solar panel is fully flexible and submersible in salt water, somehow they manage to do it!
EPA combined cycle, plug-to-wheel efficiency of 174 W·h/km
That's plug-to-wheel efficiency, and in my above calculations I already factored in an efficiency drop in charging the batteries of about 20%, which is probably too high. From the Wikipedia page it says it "uses 135 W·h/km (4.60 mi/kW·h) battery-to-wheel, and has an efficiency of 92% on average.[3][4]" This is 13.5 kW-hr / 100 km, which is admittedly more than my 10 kW-hr / 100 km, but certainly much less than your 174 figure.
Your estimate is very optimistic. It doesn't account for shadows cast by nearby buildings and trees.
That's why I dropped the final number from 50 to 35 km. You'd have people trying to park their cars in the sun.
Assume 20% realistic efficiency for commercial solar cells. That is 400 Wp that can be mounted on the roof. 400 * 600 = 240 kWh per year, enough for ~4 km/day.
There is an aftermarket company that makes bolt on solar panels for the roof of the Prius and they claim it adds an extra 20 miles a day. Whether you believe this figure for an average day or not is another question. And that's just from the roof. Imagine what advances in technology will bring!
Will S

Venturi Astrolab is working on such a car, though the collector space may not be quite what you had in mind.

Venturi Astrolab Solar Car

Will S

The link didn't come through;

The comments to this entry are closed.