GHG Life-Cycle Comparisons between Propane and Full-Electric Medium-Duty VehiclesGokul VishwanathanDirector, Research and SustainabilityPropane Education & Research CouncilNov. 3rd, 2021

This presentation is a condensed version of a white paper that was published last year

White paper with references can be downloaded from Propane Education & Research Council (PERC) website:

https://propane.com/research-development/emissions/decarbonization-of-md-hd-vehicles-with-propane/

What is Renewable Propane?

Source: NREL and Abhari, et al. Process for producing jet fuel and LPG from renewable sources. 7,846,323 B2, 2010

5 – 6 wt% Propane

6-7 wt% C3 + C4

• Isomerize• Crack

• Hydrogenate double bonds

• Deoxygenate

• Potential yield of propane ~ 5-7% (wt.%) depending on the end-product (Renewable Diesel (RD) or Sustainable Aviation Fuel (SAF))

Used cooking oil (e.g., soybean)

Fossil or green Fossil or green

Triglyceride

• Over 790M Gal/y (2.4M t/y) RD/SAF production in service →~68M gallons potential• Claimed growth of RD/SAF to over 6.3B Gal/y (19M t/y) → ~540M gallons potential• Highest capacity in Louisiana area > 2.5B Gal/y (7.8M t/y), California > 1.8B Gal/y (5.7M

t/y)• None on the east coast

Commercial HEFA/HVO Production of RD and SAF - USA

Company Location Existing Gal/y

Additional Gal/y

Technology

1 Renewable Energy Group Geismar, LA 90M 250M Expanding 2023

2 Diamond Green – Valero Norco, LA 290M 400M Expanding 2021 Ecofining

3 Marathon Dickinson, ND 180M HydroFlex

4 NextChem/ Saola Energy Garnett, KA 5M

5 BP Cherry Point Birch Bay, WA 42M

6 World Energy Paramount, CA 35M 230M Expanding 2023

7 Sinclair/HollyFrontier Sinclair, WY 150M 240M

8 CVR Wynnewood Wynnewood, OK 100M Under construction HydroFlex

9 Diamond Green – Valero Port Arthur, TX 470M Under construction 2024

10 Emerald Biofuels Port Arthur, TX 110M Under construction

11 Green Fuels Port of Columbia, LA 32M Planning 2025

12 Holly Cheyenne, Wy 90M Planning 2022

13 Marathon Martinez, CA 730M Converting 2023 HydroFlex

14 Next Renewables Columbia River, OR 750M Planned 2024

15 Phillips 66 Rodeo, CA 650M Planning 2024

16 Ryze Renewables Las Vegas, NV 100M Planning

17 Ryze Renewables Reno, NV 50M Planning

18 St Josheph Renewable Fuels Newton, IL 90M Planning

19 Bakersfield Renewable Fuels Bakersfield, CA 230M Converting 2022 HydroFlex

20 HollyFrontier Artesia, NM 110M Converting 2022

21 Grön Fuels Baton Rouge, LA 900M Planning 2030

Total 792M 5,532M

Biofuels Digest, Feb. 2021. https://www.biofuelsdigest.com/bdigest/2021/02/08/50-renewable-diesel-projects-and-the-technologies-behind-them/10/ , Biodiesel Magazine, Jan. 2021.

83.19

3326

20.526.5

37.5 38.543.5

27 28.5

0

10

20

30

40

50

60

70

80

90

Fossil Distilled corn oil NA sourced usedcooking oil

US sourced non-rendered used

cooking oil

SA sourced usedcooking oil

NA sourcedanimal fat

(tallow)

SA sourcedanimal fat

(tallow)

Asia pacificsourced animal

fat (tallow)

SanimaxMontreal animal

fat (tallow)

Sanimax USAanimal fat

(tallow)

Car

bo

n In

ten

sity

(gC

O2

eq/M

J)

Feedstock Source

Propane/LPG (Conventional and renewable) Carbon Intensities

Renewable Propane – CARB Provisional Carbon Intensities

• Carbon intensity can be 4X lower than conventional propane.• CI of production in LA is about 14 gCO2eq/MJ but by the time it reaches CA via

rail-car it is 25-30ish gCO2eq/MJ!

Source: CARB

What is Renewable Dimethyl Ether (rDME)?

• DME vs. Propane

• Similar thermophysical properties

• Disparate chemical/combustion properties (high cetane vs. high octane)

• Ideal for blending

• Conventional and renewable propane can be blended with renewable DME to drastically reduce the fuel’s carbon intensity

Source: Oberon Fuels

Carbon intensity can be as low as -278 gCO2eq/MJ

What is Renewable Dimethyl Ether (rDME)?

ALL models are wrong, but some are useful

– British Statistician George E.P. Box

• Purpose: Evaluate ΔCO2eq emissions between Medium Duty (MD) Propane vehicle vs. MD Full-Electric vehicle (MDEV). To that intent, we are NOT looking at exact CO2eq emissions but an analysis to find the Δ

• CO2eq emissions from body, doors, chassis, tires and tire replacement, wheels and wheel replacements, final assembly, interior and exterior, lead-acid battery are assumed similar or within the noise between the two vehicles such that the Δ is negligible

• Credit for Li-ion battery second life (e.g. grid applications) is not assumed

• Recycling of components including Li-ion battery is not assumed

• CO2eq emissions from End-of-Life (EOL) are assumed similar or within the noise between the two vehicles such that the Δ is negligible

• Electricity produced in a state in assumed to be used though it is imported from other states (e.g. VT, CA etc.) i.e. net electricity Carbon Intensity (CI) could be production based or consumption based but production-based CI is assumed

• Truck is charged only within the state though it is driven for 200 miles

Purpose and High-Level Assumptions

Major AssumptionsParameter Value

Vehicle Class 6-7 Truck

State-level Electricity CI (lb/MWh) • 2019 EIA (Combustion) + 2019 production emissions. Adopted CARB

methodology for each state’s energy-mix

• Decarbonized electric grid scenario - 95% reduction from above CI

Propane WTW CI (gCO2e/MJ) CARB for PADD5 (83.19) and adopted CARB method for each state in a PADD,

Renewable Propane (45 for HVO – CARB Approved), Renewable DME (-278 –

CARB approved)

T&D Losses 5% for all states (Based on EIA National Average)

Miles Driven Per Day 200

Miles Driven Per Year 60,000

Life (miles) 300,000 (5 years)

MDEV Charging, power conversion and

battery roundtrip Losses

10%

Net vehicle energy Efficiency (kWh/mile)

and Battery Size (kWh)

2.08 and 520

Life of Li-ion Battery (cycles) • 1000

• 5000 (million-mile battery)

Petroleum Administration for Defense Districts (PADD).

Major AssumptionsParameter Value

Battery Manufacturing CO2eq Emissions

(kgCO2eq/kWh)

• 140 (ICCT Study)

• 61 (Decarbonized electric grid scenario)

Regenerative Braking Energy Savings 20%

Propane Vehicle MPG • 5.5 for propane and renewable propane

• 5.3 when blended with DME due to its 3% lower energy content

Engine Peak Power (kW) 260

Accounted for:

• GHG footprint of engine, after-treatment and transmission production

• GHG footprint of fluids for MDEV and Propane vehicle and their replacement

• GHG footprint of MDEV motor, inverter, controller, transmission and cooling system

• MDEV service interval – 40,000 miles. Propane vehicle service interval (oil - 5,000 miles, radiator coolant –15,000 miles and all others – 40,000 miles)

Boundary of LCA

Vehicle Production

Powertrain Production

Vehicle Usage

Fuel Production/Electricity Generation

Vehicle Maintenance

End of Life

Waste material recycling/disposal

Parts recycling/disposal

Parts recycling/disposal

Powertrain Maintenance

Adapted from Kawamoto et al. (2020). Estimation of CO2eq Emissions of Internal Combustion Engine Vehicle and Battery Electric Vehicle Using LCA, Sustainability, 2019

Individually Calculated

Not calculated assuming negligible Δ Out of scope

U.S. Average, Just as an Example

Some Studies only consider this

Well-to-Wheels Carbon Intensity Comparisons of energy carrier (gCO2eq/MJ)

Note: They are compared on the same scale

CARB approved value = 82.92. Higher value here as charging losses are accounted for

Conventional Propane Today’s Grid Electricity

Simulated CasesCase Detail

Case I Comparison of conventional propane vehicle vs. MDEV

Case II Comparison of renewable propane vehicle vs. MDEV

Case III Comparison of propane/renewable DME blend (80%-20% by mass) vehicle vs. MDEV

Case IV Comparison of renewable propane/renewable DME blend (80%-20% by mass) vehicle vs. MDEV

• Renewable fuel and components production CIs assumed the same as status-quo even under decarbonized electric grid scenario. In reality, the CI of renewable fuels and production CIs will be lower due to cleaner electricity. Calculation of these are out-of-scope

• Propane vehicle fuel economy has been kept the same as status-quo even under decarbonized electric grid scenario. In reality, the fuel economy will improve significantly due to evolution of engine technologies (~25 years from now)

Case-I: ΔCO2eq for One Truck: Green - Propane is Better, Red – MDEV is better

Today, Propane is a cleaner solution for 38 states and DC

Numbers represent ΔLC CO2 emissions in US tons per MD vehicle

Case-II: ΔCO2eq for One Truck: Green – Renewable Propane is Better, Red – MDEV is better

Today, Renewable Propane is a cleaner solution for all states (and DC) but Vermont

Numbers represent ΔLC CO2 emissions in US tons per MD vehicle

Case-III: ΔCO2eq for One Truck: Green – Propane/R-DME is Better, Red – MDEV is better

Today, Propane/R-DME blend is a cleaner solution for all states (and DC) but Vermont

Numbers represent ΔLC CO2 emissions in US tons per MD vehicle

Extension of Study - Tool Demonstration

Conventional Propane MDVs & EVs - CO2eq Emissions

“Average” Grid Emissions28 states/territories propane MDVs are better24 states/territories EVs are better

“Marginal” Grid Emissions49 states/territories propane MDVs are better3 states/territories EVs are better (Hydroelectric dominated)

Policies are made considering “average” emissions

Conventional Propane MDVs & EVs - NOx Emissions

“Average” Grid Emissions44 states/territories propane ultra-low NOx MDVs are better8 states/territories EVs are better

“Marginal” Grid Emissions51 states/territories propane ultra-low NOx MDVs are better1 state EVs are better (Hydro-electric dominated)

• Mitigating CO2eq is extremely important but so are criteria pollutants and hence LCA of criteria pollutants should also be given equal emphasis. Currently, several propane and natural gas engines are being certified for CA 0.05 & 0.02 g/hp-hr optional/voluntary NOx standards.

• kWh/mi – Where is it measured (plug, battery, wheels)? Based on usable battery capacity or nameplate battery capacity? Altoona testing captures it well.

• Vampire loads become extremely important for MDVs/HDVs. A MA school bus study shows charging can lead to 1.4-2.4 kWh/mile depending on how the EV is charged. How does this impact V2G?

• “Average” power generation emissions are sometimes only considered in LCA. Some regulations are being enforced referencing only “average” power generation emissions (e.g., CA Port At-Berth Regulation). Upstream/feedstock emissions are often not considered and more so for renewable energy sources such as solar, wind etc. (should land use be considered here?)

• Should we compare 1 MD/HD ICEV vs. 1 EV or is it 1.X EVs to 1 ICEV? gCO2eq/ton-mile?

• PERC is in the process of releasing another report on forklifts LCA. This has been shared with CARB as CA is contemplating a rule making for banning/retiring internal combustion engine forklifts by 2025 (large fleets) and 2030 (small fleets). Currently, CA employs >300,000 ICEV forklifts.

Closing Thoughts

Gokul Vishwanathan, PhD

Director, Research and Sustainability

t 202-452-8975 / c 202-843-5128

https://www.linkedin.com/in/gokul-vishwanathan-1906301b/

1140 Connecticut Ave. NW

Suite 1075

Washington, DC 20036

Contact Information

E-mail: gokul.vishwanathan@propane.com