If we are 20 years into the EV revolution, why isn't US fuel consumption dropping more?
Part of the answer is where the fuel goes, and part is how we have been designing batteries. This essay makes one counterintuitive argument: cool the cell well enough and low-cost, well-known chemistry can carry the power, and a small pack built that way displaces more fuel for every kilogram than a large pack built for energy density. In the vehicles people actually drive, a fuel reserve is what lets the pack stay small. The short version is on the homepage. Every SampoVolt figure below is modeled, not test data. The highway figures are public statistics.
The gallons did not leave
US highway fuel climbed for decades, then leveled off. It did not settle into a long decline. From 2007 through 2024, gasoline plus special fuels, mostly diesel, stayed in a band from 159 to 186 billion gallons a year. The latest year, 2024, is 171.8 billion. The low year is 2020. The high year is 2022. The Federal Highway Administration publishes the series as Table MF-27.
The mix changed. Pickups, vans, and sport utility vehicles used 75.9 billion gallons in 2019. Cars used 48.6. Many crossovers that count as car SUVs are still inside the car figure, so the utility-vehicle share is at least that large. The Oak Ridge Transportation Energy Data Book, Edition 40, Tables 4.1 and 4.2, is the source. Through 2008 the split is the FHWA series. FHWA stopped publishing it after that. Later years are the Oak Ridge continuation.
Those are the vehicles people buy. In model year 2021, the EPA Automotive Trends Report, Table 3.2, puts truck SUVs at 44.7 percent of light-vehicle production, pickups at 16.1 percent, and car SUVs at 11.4 percent.
A kilogram of battery only saves fuel when it is working
A battery saves fuel on the miles it drives on electricity. Most days are short. A pack sized for the rare long day carries most of its capacity as ballast: paid for, hauled every mile, and rarely emptied. Each kilowatt-hour past the daily drive saves less fuel than the one before it.
A small pack that empties and refills nearly every day puts every kilogram to work. Over its life, each kilogram moves the vehicle many more miles on electricity, so it displaces more fuel. That is the whole idea.
For a truck or SUV, a pure battery pack has to be large. You haul that weight every day and rarely use all of it. The weight calls for heavier, more durable parts, and the price moves past most buyers. Building packs that large for so many vehicles also strains the supply chain and often requires more expensive chemistries.
Let hydrocarbons do what they do best
Hydrocarbon fuel is a remarkable store of energy. A tank holds a lot of it, costs little to carry, keeps for months, and refills in minutes. A tank is the right reserve: the long trip, the cold morning, the trailer, the day without a plug.
What an engine does badly is follow your right foot. So split the work. In a series hybrid the engine does not turn the wheels. It is a generator. The motor turns the wheels, and the battery feeds the motor. Everyday miles come off the battery. The tank covers the rest, without a second, rarely used battery riding along.
The tank is one part of why this works. It lets the battery be sized for power and the daily miles instead of the longest trip. Without it, the battery has to be the reserve as well, and energy density is the right target; that is the pure battery car. The other two parts are the cooling and the chemistry, below.
"A pack shaped for power and cycle life displaces more fuel per kilogram than a pack shaped for energy density."
The catch: small packs run hot
Here is why this is not already on the road. With no engine on the wheels, the battery alone has to deliver the motor's full peak. A right-sized pack lives at a high C-rate: a lot of power from not very many kilowatt-hours. High rates make heat.
The conventional pack is built around energy density: large flat prismatic cells, cooled through their broad faces. Heat has to cross the whole cell to get out, and those cells run out of thermal room near 2.7C. To deliver the same peak, the pack grows until the rate falls back inside the limits. It grows not because it needs the energy, but because it needs somewhere to put the heat. More battery. More mass. Less fuel saved per kilogram. Most hybrids end up back at a price few buyers can meet.
Cooling is the bridge.
Cool the edge
SampoVolt builds stacked ribbon electrodes. The current collector, the metal foil inside every electrode, is designed as the way out for both electricity and heat, and on edge it runs the short way out instead of across the cell. Cooling plates are shared between neighboring layers, so no electrode sits far from one. Packed appropriately, the stack is a low-resistance heat exchanger. Same LFP/graphite cell. Same coolant. Only the cooling path changes. We design from the pack in, not from a coin cell out.
Cool is only half of it. Even is the other half. A pack is only as good as its weakest cell, and a cell is only as good as its hottest electrode: one hot spot ages first and drags the whole pack with it. Keeping the spread small inside each cell and between electrodes is how the pack is designed to age at the same rate everywhere. That is a design goal, not a measured cycle life.
At a 5C charge, the modeled peak is 34.9 °C on the edge-cooled cell and 47.4 °C on the typical face-cooled cell. The spread inside the cell is 4.7 K against 16.6 K. The limits we design to are a 5 K spread and a 45 °C peak. Inside those limits the edge-cooled ribbon holds 5.15C indefinitely. On a 150 kW SUV, that is the 29 kWh pack. The face-cooled prismatic holds to 2.74C.
A short, hard pull is a different question from a steady charge. Even at a 7C demand, the temperature spread inside the cell stays near 2.5 K after 10 seconds, about half the 5 K limit. It rarely has to hold that alone. The controller starts the engine before the climb or the hard pull, and tow mode keeps the engine on, so the 70 kW generator and the battery share the load and the pack settles at about 4.6C.
That changes the generator too. The motor takes the launch, the grade, and the pass. The generator only has to supply steady energy, so it can be smaller and held near one efficient point. It makes electricity. It does not need a transmission to the wheels.
Half the battery, more fuel per kilogram
For an SUV at a 150 kW peak, the edge-cooled pack is 29.2 kWh and 249 kg. The face-cooled pack for the same peak is 55.6 kWh and 436 kg. About half the battery. The same comparison is about 50 kWh against 94 kWh on a pickup, and about 29 kWh against 56 kWh on a van.
That is a practical 60 to 70 mile series hybrid. At that sweet spot, about 33 lifetime gallons of fuel are displaced per kilogram of battery, against about 19 for the larger face-cooled pack. Over a year, about 2.2 gallons per kilogram against about 1.3. Past about 70 miles, even the edge-cooled pack starts to give some back: it grows faster than the miles it adds.
At a cell price of $55 to $95 per kWh, the cells are $1,609 to $2,779, against $3,057 to $5,280. About half the cell cost. Those are cells only, not the pack or the vehicle.
The pitch is thermal control that lets a small pack take harsh, high-power cycling day after day and still last. Improved cycle life is the design goal from holding the cell evenly. It is not a measured result, and this is not a measured charge time or a measured cycle life. No engine is selected, and no vehicle has been measured.
A smaller pack is also a supply-chain fact. Half the cells per vehicle means half the LFP, graphite, copper, and aluminum to buy for that vehicle, so the same cell supply reaches more vehicles. The aim is more packs in the vehicles that burn the fuel, and more fuel displaced. That is a design aim, not a counted fleet.
Low-cost materials are the point
Energy density is the score most battery programs chase. It is the right score for a pure battery car that has to hold a long trip inside the pack. It is the wrong score for this hybrid.
A series-hybrid pack is sized by power and heat. The chemistry can stay LFP and graphite, the low-cost pair the industry already knows how to make. Denser powder, nickel-rich cathodes, and cells built as power specialties are how you chase energy, or a rate the cooling could not support. Here the cooling supports the rate.
The second note runs the powder grades. In that model, moving from an older mainstream LFP to the densest pilot grade changes the hybrid pack by about 4 kg. Changing the cooling path changes it by 140 to 160 kg. Those figures are modeled, and they are a different cut from the 249 kg SUV pack above. The conclusion lines up: the heat path sets the size, and the pack does not need the densest powder to do this job.
What we are not claiming
Public highway statistics are not SampoVolt test data. Temperatures, pack sizes, cell cost, and gallons displaced per kilogram are model results. The comparison holds the cell and the coolant fixed and changes the cooling path. It does not say a vehicle has been built, a powder lot has been qualified, or a cycle life has been measured.
The figures are drawn again at the end of this page: the highway series, the shift from cars to trucks, the 2024 gallons, the fast-charge temperatures, the burst, the pack sizes, and the gallons per kilogram.
SampoVolt builds on the rolled-ribbon electrode invented by Thomas D. Kaun and developed at InvenTek (US Patent 7,195,840, now expired), with research support from Argonne National Laboratory. Not affiliated with the inventor or Rolled-Ribbon Battery Company. The patent is US 7,195,840.
Sources
- U.S. EPA, Automotive Trends Report, EPA-420-R-22-029, December 2022, Table 3.2. Model year 2021 light-vehicle production shares: truck SUVs 44.7 percent, pickups 16.1 percent, car SUVs 11.4 percent. https://www.epa.gov/system/files/documents/2022-12/420r22029.pdf
- Federal Highway Administration, Highway Statistics, Table MF-27. Gasoline, including gasohol, and special fuels, mostly diesel. The 2024 table is the latest year on the chart, 171.8 billion gallons. https://www.fhwa.dot.gov/policyinformation/statistics/2024/mf27.cfm
- Oak Ridge National Laboratory, Transportation Energy Data Book, Edition 40, Tables 4.1 and 4.2, June 2022 update. In 2019 the estimate is 75.9 billion gallons for pickups, vans, and SUVs, against 48.6 for cars. https://tedb.ornl.gov/wp-content/uploads/2022/03/TEDB_Ed_40.pdf
- Federal Highway Administration, Highway Statistics 2024, Table VM-1. Light-duty vehicles are 124.8 billion of the 171.8 billion gallon total. https://www.fhwa.dot.gov/policyinformation/statistics/2024/vm1.cfm
- Thomas D. Kaun, US Patent 7,195,840, rolled-ribbon cell structure. Now expired. https://image-ppubs.uspto.gov/dirsearch-public/print/downloadPdf/7195840