Impact assessment of crude oil mix, electricity generation mix, and vehicle technology on road freight emission reduction in China

To achieve net zero emissions, the global transportation sector needs to reduce emissions by 90% from 2020 to 2050, and road freight has a significant potential to reduce emissions. In this context, emission reduction paths should be explored for road freight over the fuel life cycle. Based on panel data from 2015 to 2020 in China, China’s version of the GREET model was established to evaluate the impact of crude oil mix, electricity mix, and vehicle technology on China’s reduction in road freight emissions. The results show that the import share of China’s crude oil has increased from 2015 to 2020, resulting in an increase in the greenhouse gas (GHG) emission intensity of ICETs in the well-to-tank (WTT) stage by 7.3% in 2020 compared with 2015. Second, the share of China’s coal-fired electricity in the electricity mix decreased from 2015 to 2020, reducing the GHG emission intensity of battery electric trucks (BETs), by approximately 6.5% in 2020 compared to 2015. Third, different vehicle classes and types of BETs and fuel cell electric trucks (FCETs) have different emission reduction effects, and their potentials for energy-saving and emission reduction at various stages of the fuel life cycle are different. In addition, in a comparative study of vehicle technology, the results show that (1) for medium-duty trucks (MDTs) and heavy-duty trucks (HDTs), FCETs have lower GHG emission intensity than BETs, and replacing diesel-ICETs can significantly reduce GHG emissions from road freight; (2) for light-duty trucks (LDTs), BETs and FCETs have the highest GHG emission reduction potential; thus, improving technologies such as electricity generation, hydrogen fuel production, hydrogen fuel storage, and transportation will help to improve the emission reduction capabilities of BETs and FCETs. Therefore, policymakers should develop emission standards for road freight based on vehicle class, type, and technology.


Introduction
In 2020, CO 2 emissions of the global transport sector were 7.2 Gt. The International Energy Agency (IEA) has combined World Energy Model (WEM) and Energy Technology Perspectives (ETP) to simulate the net zero emission roadmap of the global transport sector. The results showed that the global transport sector should reduce emissions by 90% from 2020 to 2050. China's transport produced 950 Mt of CO 2 emissions in 2020, accounting for about 13% of global transport emissions and 9% of China's total energy-related emissions (IEA 2021a). Road freight is considered to be the main cause of the global greenhouse effect (Teixeira et al. 2021). By the end of 2021, there were 12.3196 million road business vehicles in China,including 11.7326 million trucks, accounting for approximately 95% of the total road business vehicles (Ministry of Transport of the People's Republic of China 2022). Trucks contribute the majority of EC and emissions (IEA 2021b). Therefore, controlling the EC and GHG emissions of trucks is the key to reducing road freight emissions.
As of the end of 2021, vehicle data released by the Ministry of Transport of the People's Republic of China showed that diesel-ICET, BETs, and FCETs account for approximately 92%, 2%, and 0.5% of business trucks, respectively. Diesel is the main fuel used by road freight vehicles. In contrast to diesel-ICET, the emissions of BETs and FCETs are entirely derived from upstream fuel extraction, production, and distribution processes (He et al. 2020). The most common GHGs are carbon dioxide, methane, and nitrous oxide; and more than 80% of these GHGs are generated in the process of energy production and consumption (El Hannach et al. 2019). Therefore, road freight emission reduction should be explored based on the fuel life cycle.
LCA has been widely used to assess EC and GHG emissions from road transport to evaluate the emission reduction potential of advanced vehicle technologies, particularly in urban road transport (Ou et al. 2009;Petrauskienė et al. 2020;Wang et al. 2020). The emission reduction effects of various vehicle technologies have attracted considerable attention owing to the application of advanced vehicle technologies in freight trucks. Bicer and Dincer (2017) showed that FCEVs have lower GHG emissions than BEVs owing to the impact of electricity mix. Booto et al. (2021) in their study concluded that the emission reduction effects of BETs are better than those of FCETs compared with low-sulfur diesel trucks conforming to Euro VI standards. Plötz (2022) noted that although electricity and hydrogen are two key energy carriers for realizing a low-carbon future, the application of BEV technology in freight should be the current focus until the hydrogen technology has sufficiently matured. Therefore, this study aimed to explore the determinants of road freight emission reduction in China by comparing the emission reduction effects of diesel-ICET, BETs, and FCETs throughout the fuel life cycle.

Literature review
Extensive studies have been conducted on energy-saving and emission reduction of vehicles. We have investigated the impact of crude oil mix, electricity mix, and vehicle technology on GHG emissions.

Crude oil mix and GHG emissions
Crude oil mix has a substantial impact on reducing diesel-ICET emissions in countries that are highly dependent on crude oil imports. For countries with a single crude oil source, the quality of crude oil and the efficiency of vehicle fuel production have a major effect on the reduction of emissions from diesel-ICETs. For example, Masnadi et al. (2018) analyzed the quality of crude oil from 146 oil fields in 20 different countries and found that the emissions at the well-to-refinery stage vary widely, ranging from 1.5 to 46.9 g CO 2 eq/MJ. Rahman et al. (2015) compared GHG emissions of crude oil from five different sources in North America during their life cycle. They found that the difference in the GHG emissions of diesel oil extracted from two different regions during the WTT stage was 5.62 g CO 2 eq/MJ. Sheng et al. (2021) assessed the GHG emissions of ICEVs in New Zealand and Australia, and they also considered the crude oil production pathway. However, they used oil production efficiency data rather than crude oil mix analysis (Greene et al. 2020).
Although numerous previous studies have shown that crude oil production pathways affect the EC and GHG emissions of conventional energy vehicles (Da et al. 2022;El Hannach et al. 2019;Gan et al. 2021;, there are few studies on the specific analysis of the crude oil mix. In this study, the EC and GHG emissions of diesel-ICET in the WTT stage were evaluated by considering China's crude oil mix scenario.

Electricity mix and GHG emissions
For countries that use fossil fuels as the main source of electricity generation, the electricity generation structure seriously affects the emission reduction effects of BETs. For countries that use renewable energy as the main source of electricity generation, BETs and FCETs can provide significant emission reduction effects, but they also cause other environmental problems and there are cost-effectiveness problems. For example, the GHG emission intensity of electricity generation in the US FRCC region is approximately 55 g CO 2 eq/MJ higher than that in the WECC region because the share of fossil fuel generation in the FRCC region is 40% higher (Liu et al. 2021). In Turkey, the share of fossil fuel electricity generation decreased by approximately 10% in 2018 compared with 2014, and BETs reduced GHG emissions by 37 g CO 2 eq/km (Ugurlu 2022). Australia (83%) has a much higher share of fossil fuel electricity generation than New Zealand (17%), resulting in GHG emissions from BEVs that are nearly 50 g CO 2 eq/km higher (Sheng et al. 2021). Choi and Song (2018) found that in South Korea, where fossil fuels account for approximately 73% of electricity generation, the average GHG emissions of BEVs are 90-110 g CO 2 eq/km higher than those of ICEVs. However, BEVs may perform worse than ICEVs in terms of other environmental issues, including acidification, particulate matter formation, and toxicity, according to Bauer et al. (2015). Winkler et al. (2022) showed that although the emission reduction effect of FCEV is better than that of BEV in Germany, where renewable energy provides 47% of electricity generation, the cost of FCEV is also relatively higher.
As coal provides approximately 70% of the electricity generation in China (2015), the GHG emissions of BEVs in the fuel production stage account for 84% of the vehicle and fuel cycles (Da et al. 2022). The share of coal electricity generation declined in 2021 but remained above 60%. According to IEA forecasts, electricity generated from coal in China will be 39.57% in 2030, which remains higher than that of renewable sources, resulting in high global warming potential . In addition, the GHG emission intensity of electricity in different regions also varies significantly owing to the different electricity generation structures, thermal electricity generation technologies, and electricity transmission efficiencies. In regions with a high proportion of fossil fuel electricity generation, the carbon emissions of BEVs are higher (Tang et al. 2022). Under a scenario where the share of fossil fuel electricity generation in Hong Kong falls from 73 to 15% and renewables increase from 1 to 85% between 2019 and 2050, BETs will reduce emissions by 80% over the fuel life cycle compared with diesel-ICET .
Researchers generally believe that BEVs can mitigate climate change in the context of non-fossil fuel electricity generation (Choma and Ugaya 2017). The application of renewable energy in electricity generation can promote the emission reduction effect of BEVs; however, an electricity mix scenario with the greatest environmental advantage is still uncertain (Souza et al. 2018), and the impact of biomass, nuclear, and other electricity generation in the electricity mix is ignored. Therefore, in this study, the emission reduction effect of BETs was compared under the scenario of China's electricity mix from 2015 to 2020, and an electricity structure adjustment strategy conducive to emission reduction was explored.

Vehicle technology and GHG emissions
The majority of research largely concurs that BEV and FCEV technologies can significantly reduce emissions from urban road transportation (Peng et al. 2018;Shen et al. 2019), but the emission reduction effect of road freight is controversial. For example, in urban road transport emission reduction, Souza et al. (2018) evaluated the emission reduction potential of electric vehicles in urban road transport from a life cycle perspective (including vehicle and battery recycling processes) and concluded that BEVs have the lowest impact on the environment. Sheng et al. (2021) compared vehicle technologies such as ICEVs, FCEVs, BEVs, and plug-in hybrid electric vehicles (PHEVs) in Class 3 light vehicles in New Zealand and Australia. They found that BEV technology was the best way to reduce emissions in transportation in the short term, and FCEV technology was the only way to reduce emissions next to BEV technology in the long term.
In a study on road freight emission reduction,  found that the emission reduction rates of BETs and FCETs in Hong Kong were 68% and 48%, respectively, compared with diesel-ICETs. Lao et al. (2021) verified the effectiveness of FCET emission reduction in the Beijing-Tianjin-Hebei-Shandong (BTHS) region through a case study. Ren et al. (2022) showed that the emission reduction potential of FCETs depends on the dominant role of renewable energy generation and the development of hydrogen storage and transportation technologies. Studies have also shown that BEV technology can effectively reduce GHG emissions from passenger and freight transportation (Ou et al. 2010).
In addition, many researchers have investigated the impact of alternative fuels on transportation emission reduction (Hao et al. 2010;Mahbub et al. 2017). For example, a study by Jhang et al. (2020) showed that adding 10% ethanol to gasoline could reduce 20.8 g CO 2 eq/km, and the gasoline-bioethanol blend is one of the best alternative fuels. Da et al. (2020) showed, in their study, that employing natural gas vehicles (NGVs) instead of diesel vehicles in China may have a significant negative impact on climate because of the high methane emissions of heavy-duty NGVs. Song et al. (2017) found that LNG-HDTs have a higher life-cycle EC than diesel vehicles, but their GHG emissions are still lower than those of diesel vehicles.
Researchers have previously focused on vehicle emission reduction, and most have assumed that advanced technology vehicles and diesel trucks have the same payloads. However, according to data on qualified vehicles released by the Ministry of Industry and Information Technology of China, most BETs and FCETs in China have higher payloads than diesel vehicles. Previous studies have explored methods for allocating emissions to road freight (Kellner 2022;Kellner and Schneiderbauer 2019). Therefore, in this study, the actual payload was considered to compare the GHG emission intensity of each unit of cargo transported by various vehicles.

Research scope and functional units
This study aims to compare the EC and GHG emission intensity of diesel-ICEV, BEV, and FCEV technologies applied to trucks to evaluate the impacts of China's crude oil mix, electricity mix, and vehicle technology on the reduction of road freight emissions. WTW analysis was applied to account for the EC and GHG emissions in the life cycle of vehicle fuel. The analytical framework is shown in Fig. 1. The WTW process is divided into WTT and TTW stages. The WTT stage is the process of feedstock extraction, transportation to the central plant, fuel production, and transportation to refueling stations, including feedstock and fuel processes. The TTW stage converts fuel into power during vehicle operation. This study assessed the EC and GHG emission intensity of vehicles completing one unit of cargo transportation during the fuel life cycle, in units of MJ/tonne•km and g CO 2 eq/tonne•km, respectively. The source of crude oil affects the energy loss and GHG emissions of diesel-ICETs in the WTT stage, the electricity mix affects the energy loss and GHG emissions of BETs in the WTT stage, and vehicle technology affects the energy use and GHG emissions in the TTW stage. Figure 1 shows the EC and GHG emissions of diesel-ICET, BETs, and FCETs in the WTT stage of diesel fuel production, electricity generation, and hydrogen production pathways, respectively.

Diesel fuel production pathway
Crude oil in the Chinese market is a mixture of crude oils from countries worldwide. China uses crude oil to produce diesel fuel. Diesel is produced through the processing of mixed crude oil by refineries (Masnadi et al. 2018), and the finished oil is transported and distributed to refueling stations. For example, 74% of China's crude oil in 2020 was imported, and only 26% was produced domestically (Table 1) . In 2021, crude oil in the Chinese market came from more than 40 countries/regions (International Energy Network 2022). Therefore, China's crude oil mix affects the energy loss and GHG emissions of the ICETs during the WTT stage.

Electricity generation pathway
The electricity generation pathway in China involves a combination of electricity generation using coal, natural gas, water, solar energy, and other primary energy sources. In 2020, 61% of China's electricity was generated from coal. Despite a downward trend from 2015 to 2020, the share of coal electricity generation is still more than 60% (see Table 2). Studies have suggested that BEVs produce almost no GHG emissions while driving. However, owing to the use of fossil fuels, which produce a significant amount of GHGs, the electricity generation process has a high emission reduction potential (El Hannach et al. 2019;He et al. 2020). Therefore, China's electricity mix affects the energy loss and GHG emissions of the BETs during the WTT stage.

Hydrogen production pathway
Currently, China's FCETs use gaseous hydrogen as fuel. The production pathways are shown in Fig. 2. Gaseous hydrogen is produced in central plants, piped to the bulk terminal, and transported to refueling stations by truck (Argonne National Laboratory 2021). The methods of hydrogen production in China include coal gasification, natural gas reforming, industrial by-products, and electrolysis of water, which account for 62%, 19%, 18%, and 1% of the total production, respectively (China EV100 2020). The main mode of transportation of gaseous hydrogen is via the tube trailer. The source and method of hydrogen production and transportation distance of the tube trailer affect the energy loss and GHG emissions of FCETs in the WTT stage.

TTW analysis of trucks
Trucks are the main transport vehicles for road freight in China and are the major contributors to energy use and emissions (IEA 2021b). In this study, we compared the energy-saving and emission-reduction effects of BETs and FCETs relative to diesel-ICET. Commercial trucks that meet China's safety technical standards were selected to evaluate the EC and GHG emissions of trucks with four classes of GVWR (see Table 3 in Sect. 3.4.1 for vehicle parameters). Fuel economy and vehicle payload are key parameters that affect the energy use and GHG emission intensity in the TTW stage. Fuel economy is affected by traffic conditions, driving habits, and many other factors . In this study, the fuel economy of vehicles under integrated operating conditions was used, and the fuel economy of other vehicle technologies was presented as the ratio of the baseline diesel vehicles.

EC and GHG emission intensity per unit of freight turnover
This study used the GREET model developed by Argonne National Laboratory (Argonne National Laboratory 2021). The model was updated in terms of feedstock source, fuel  CH 4 , and N 2 O over the fuel life cycle and converting them into GWP (i.e., 1, 30, and 265, respectively), in units of g CO 2 eq/tonne•km. The total EC and GHG emissions from road freight are composed of two components: fuel production (i.e., well-to-tank) and vehicle operation (i.e., tank-towheel). The vehicle life cycle of trucks was not considered in this study.

Crude oil mix
Based on the usability and availability of the data, sources of crude oil in China were obtained through regional statistics (excluding China). For example, crude oil from North America in China's market is the total supply of oil from Canada, Mexico, and the United States. The scenarios for China's crude oil mix from 2015 to 2020 are presented in Table 1. Owing to the differences in petroleum product attributes and oil recovery technologies in oilfields worldwide, the energy efficiency and GHG emission intensity of crude oil recovery processes vary significantly (Masnadi et al. 2018). The average values of field product attribute, such as API gravity and S-content averages, for each region were used (Table in the Appendix).
It is assumed that imported crude oil is shipped from each region to China, and the shipping distance is the average shipping distance from the major ports in each region to the Dalian Port of China, with data obtained from China Shipping Online (Table in the Appendix).

Electricity mix
China primarily uses coal, natural gas, water, wind, and solar energy for electricity generation. The share of the primary energy electricity generation in China's electricity mix from 2015 to 2020 is presented in Table 2. Different electricity mix scenarios affected the EC and GHG emissions of BETs in the WTT stage. In the scenario of electricity mix in 2020, the WTT energy efficiency is only 41.4% (Table in the Appendix) owing to the share of coal electricity generation.

Vehicle technology
For road freight, trucks are classified as lightduty trucks (1.8 t < GVWR ≤ 4.5 t), medium-duty trucks (4.5 t < GVWR ≤ 12 t), and heavy-duty trucks (GVWR > 12 t) (Ministry of Transport of the People's Republic of China 2019). In this study, 12 different freight trucks that met China's VI standard were selected. The key parameters of these vehicles are listed in Table 3. These vehicles are divided into four GVWR, three vehicle types, and three vehicle technologies. The aim was to compare the EC and GHG emissions of these vehicles during the WTW stage. The fuel economy of diesel vehicles was used as the baseline for comparison, and that of other vehicle technologies was presented as the relative ratio of the baseline. Actual payload data for various vehicles were applied.

Energy efficiency during the WTT stage
The WTT energy efficiency is the ratio of 1 MJ of energy used by the vehicles to the total energy input, as a percentage. As shown in Formula (1), the lower the energy loss at the WTT stage, the higher its energy efficiency.
El f uel is the energy loss in fuel production and transportation when a vehicle uses 1 MJ of energy; El feedstock is the energy loss in feedstock production and transportation when a vehicle uses 1 MJ of energy. The updated GREET model was used to evaluate El f uel and El feedstock .

Energy use during the TTW stage
TTW energy use is calculated by the fuel consumption per unit of cargo transport, as shown in Formula (2), and the unit of energy use is MJ/tonne•km.
LHV F i is the calorific value of a specific fuel, and a low calorific value is used in this study to maintain consistency with the GREET model. M Payload V t is the payload of a particular vehicle (related to GVWR, vehicle type, and vehicle technology). The fuel economy of diesel-ICETs is measured by diesel-equivalent MPG, whereas the fuel economy of other vehicle technologies is based on the relative changes in diesel trucks, using data provided by the GREET model.

EC and GHG emission intensity during the WTW stage
The WTW EC is the sum of the primary EC per unit of cargo transport. It is the total sum of TTW energy use and WTT energy loss, as shown in Eq. (3), and the unit is MJ/ tonne•km.
The GHG emission intensity at the WTW stage is the total GHG emission at the vehicle operation, fuel production, and feedstock production stages. It is the GHG emission per unit of cargo transport, in units of g CO 2 eq/ tonne•km. According to the GWP of the three greenhouse gases over the last 100 years, the various greenhouse gases were converted into carbon dioxide equivalents, as shown in Formula (4).
(1) Ef = 1 1 + El feedstock + El f uel × 100% (2) (3) E WTW = E TTW + E TTW * El feedstock + El f uel GWP i is the GWP of i greenhouse gases; r i (V t ) is the GHG emission rate of i greenhouse gases by a particular vehicle (using GREET model data).

Impact of crude oil mix
The energy efficiency and GHG emissions of diesel fuel in the WTT stage under different crude oil mix scenarios are shown in Fig. 3. This figure shows that with an upward trend in the share of China's crude oil imports from 2015 to 2020, energy efficiency at the WTT stage decreased and GHG emissions increased. This can be attributed to the following points: (1) the long shipping distance of crude oil imports increases energy loss and GHG emissions in the process of crude oil transportation, and (2) China's crude oil originates from more than 40 countries/regions, which have significant differences in oil field productivity, recovery technology, and combustion efficiency. For example, Saudi Arabia's oil field is characterized by high reservoir production, low water production, and low combustion efficiency; the Orinoco field belt in South America has a high GHG intensity of 31.9 g CO 2 eq/MJ since steam flooding is used to enhance oil recovery (Masnadi et al. 2018). Figure 3 shows that under China's crude oil mix scenario in 2020, the production efficiency of diesel fuel is about 83.5%, and the indirect GHG emission of diesel-ICETs is about 17.4 g CO 2 eq/MJ. From 2015 to 2020, the import share of North America in China's crude oil showed an increasing trend. However, diesel production efficiency with North American crude oil is about 63% and GHG emission intensity in the WTT stage is about 46.8 g CO 2 eq/MJ. Therefore, increasing crude oil imports from North America is not conducive to reducing China's road freight emissions. The share of crude oil from the Middle East in China's market is the highest and showed an increasing trend. The diesel production efficiency with crude oil from the Middle East is about 84.5%, second only to the Asia-Pacific region and Africa. The GHG emission intensity in the WTT stage is about 16.0 g CO 2 eq/MJ. Therefore, increasing crude oil imports from the Middle East will help reduce the emissions of China's road freight. The share of crude oil from other Asia-Pacific regions (excluding China) in China's market has an increasing trend, with an energy efficiency of 87.1% and GHG intensity of 15.2 g CO 2 eq/MJ in the WTT stage. Increasing crude oil imports from the Asia-Pacific region will help reduce emissions in China's road freight. This study assumes that all crude oil imports are shipped, which means the EC and GHG emission intensity during fuel production may be overestimated for regions using pipeline and rail transport.

Impact of electricity mix
Considering the scenario of China's electricity mix from 2015 to 2020, the energy loss and GHG emissions at the WTT stage of electric energy are shown in Fig. 4. Since 2015, the share of coal electricity generation in China has decreased, thus reducing indirect BET emissions and fossil fuel losses and improving the overall efficiency of electricity generation. In terms of the electricity mix, in 2020 when compared with 2019, the share of fossil fuels decreased by 1.31%, the share of biomass energy increased by 0.24%, and the share of renewable energy increased by 1.03%. However, in 2020, the energy production efficiency of electricity decreased by approximately 1.2%, and the GHG emissions of BETs increased by approximately 4 g CO 2 eq/MJ. It was found that the main reason for the decrease in energy production efficiency and increase in GHG emissions was the application of boiler technology for biomass electricity generation, which resulted in low-efficiency biomass electricity generation at 20%. If the technology of biomass electricity generation is completely converted from boiler to integrated gasification combined cycle (IGCC), the electricity generation efficiency using biomass can be increased to 45%. It should be noted that there are costs associated with technology conversion, which are not the focus of this study. The results show that under the existing electricity generation technology, increasing the share of biomass is not conducive to the emission reduction of BETs. However, if generation technology is improved, increasing the share of biomass electricity generation can promote BET emission reduction.

Impact of vehicle technology
Under the scenario of crude oil mix and electricity mix in 2020, the EC and GHG emission intensity of 12 freight trucks with China's VI standard were compared (see Table 3 in Sect. 3.4.1 for key vehicle parameters), as shown in Figs. 5 and 6, respectively. It is hypothesized that the hydrogen produced in China is produced by natural gas. The average transportation distance of hydrogen from the bulk terminal to the refueling station is assumed to be 200 km.

Energy consumption
In the case of GVWR, the development of MDTs and HDTs helps reduce the EC per unit of cargo transport. Figure 5 shows that 9-tonne, 18-tonne, and 31-tonne diesel-ICET reduce energy losses by 3.02, 3.55, and 3.84 MJ/tonne•km, respectively, saving 65%, 76%, and 83% energy compared to 4-tonne trucks. For BETs, the EC per unit of cargo transport for an 18-tonne truck was the lowest. For FCETs, the 18-tonne and 31-tonne trucks had the lowest EC per unit of cargo transport. From the perspective of the fuel life cycle, diesel-ICETs have the greatest energy-saving potential in the feedstock and TTW stages, BETs have the greatest energy-saving potential in the fuel stage of the WTT, and FCETs have the greatest energy-saving potential in the fuel stage of the WTT and TTW stages. Figure 5 demonstrates that, in the feedstock stage, the EC of diesel-ICETs is much higher than those of BETs and FCETs, mainly because of China's high dependence on crude oil imports. This suggests that optimizing China's crude oil mix will help reduce the energy loss in the feedstock stage (see Sect. 4.1 for details). In addition, improving the fuel economy and payload of diesel-ICETs can help reduce energy use during the TTW stage. The energy loss of BETs in the fuel stage is significantly higher than that of diesel-ICETs and FCETs, which is mainly influenced by China's electricity mix and electricity generation technology. Reducing the share of coal and improving technology can help reduce BET's primary EC (see Sect. 4.2 for details). Currently, FCETs rely primarily on fossil fuels to produce hydrogen. Fuel production, storage, and transportation technologies restrict the application of FCETs (Hao et al. 2018;Yang et al. 2020), but FCETs still have good application prospects in reducing the EC of MDT.

Greenhouse gas emissions
As shown in Fig. 6, from the GVWR of trucks, the emission reduction effects of different truck levels are different.  (1) 18-tonne BETs and 18-tonne and 31-tonne FCETs had the lowest GHG emission intensities of 47, 48, and 49 g CO 2 eq/tonne•km, respectively. Figure 6 shows that the GHG emission intensity of the 18-tonne BETs was 38 g CO 2 eq/km less than that of the 18-tonne ICETs. The average payload of an 18-tonne truck is approximately 10 tons, which implies that employing a fully loaded 18-tonne BET will reduce GHG emissions by 380 g CO 2 eq/km. (2) 4-tonne BETs and FCETs had the greatest GHG emission reduction potential, which was mainly reflected in the fuel stage of the WTT. Figure 6 shows that the magnitudes of the emission reduction for 4-tonne BETs and FCETs were 211 and 200 g CO 2 eq/tonne•km, respectively. In the fuel stage, the GHG emission intensities of BETs and FCETs were approximately 109 and 119 g CO 2 eq/tonne•km higher than that of diesel-ICETs, respectively. (3) The GHG emission intensity of 31-tonne BETs was higher than that of diesel-ICETs, as shown in Fig. 6. The magnitude of the emission reduction for 31-tonne BETs and FCETs was less significant at − 3 and 14 g CO 2 eq/tonne•km, respectively. (4) The GHG emissions of BETs and FCETs mainly originated from the fuel stage of the WTT, accounting for 93.8% and 93% of the total GHG emissions, respectively. Compared with FCETs, the fuel stage of BETs has a higher GHG emission intensity, which is mainly influenced by China's electricity generation sources and technologies (see Sect. 4.2). Hydrogen production primarily uses natural gas, which is not significantly affected by the source of electricity generation. (5) Under the impact of China's crude oil mix, the GHG emission intensity of diesel-ICETs during the feedstock stage is much higher than that of BETs and FCETs, as shown in Fig. 6.

Discussion
The GREET model was applied to evaluate the EC and GHG emissions during the fuel life cycle. Subsequently, the results were compared with the relevant research literature, as shown in Table 4.
(1) The results of this study for vehicle diesel, hydrogen, and electric energy in the WTT stage are consistent with those in the relevant literature. Masnadi et al. (2018) used the Oil Production Greenhouse Gas Emissions Estimator (OPGEE) to evaluate a crude oil mix scenario in China in 2015. The average GHG emis-sions in the WTT feedstock stage were 8.4 g CO 2 eq/ MJ, which is comparable to 8.1 g CO 2 eq/MJ found in this study (Table of the Appendix). Further, the average GHG emission for global crude oil in 2018 was approximately 8.7 g CO 2 eq/MJ (Ankathi et al. 2022), and for Chinese crude oil, it was 9.1 g CO 2 eq/MJ (Table of the Appendix), which is slightly higher than the global average. Compared with 2015, GHG emissions from diesel fuel in the WTT stage increased significantly in 2020 (by 1.4 g CO 2 eq/MJ), which is mainly attributed to China's increasing dependence on crude oil imports. Our results estimated the production efficiency of hydrogen for vehicles in 2020 to be approximately 61% (Table of the Appendix). Lao et al. (2021) also employed the GREET model to estimate the production efficiency of hydrogen and obtained 60% in 2017, which is similar to the results of this study. Li and Yang (2020) used statistical analysis to evaluate the electricity carbon intensity of 31 Chinese provinces in 2018. They found that carbon dioxide emissions from coal electricity generation in 2018 ranged from 8 to 215 g CO 2 eq/MJ, with a median of 174 g CO 2 eq/MJ. Under the electricity mix scenario in 2018, the results in this work show that the total GHG emissions at the WTT stage are approximately 194 g CO 2 eq/MJ, whereas the carbon dioxide emissions at the electricity generation stage are approximately 181 g CO 2 eq/MJ (Table of the  Appendix). The GHG emissions during coal combustion were approximately 174.9 g CO 2 eq/MJ, which is consistent with the reference value. Li and Yang (2020) estimated that the electricity generation efficiency in the WTT stage was 42%, which supports the results obtained in this study. Winkler et al. (2022) reported that WTT GHG emissions from electricity in Germany were 145 g CO 2 eq/MJ in 2019. In this study, WTT GHG emissions in China were 177 g CO 2 eq/MJ in 2019. The GHG emissions were 32 g CO 2 eq/MJ higher than those of German BETs because only 19% of Germany's electricity comes from fossil fuels compared to 65% in China. From 2015 to 2020, GHG emissions in the feedstock stage of electricity generation remained unchanged (12 g CO 2 eq/MJ), and the GHG emissions in the fuel stage generally showed a downward trend (Table of the Appendix). Compared to 2015, estimated GHG emissions from the fuel stage of electricity generation in 2020 decreased by approximately 12 g CO 2 eq/MJ. (2) In contrast to other previous studies, this study has evaluated the EC and GHG emissions per unit of cargo transport of different vehicles by vehicle type, fuel economy, and payload and this study has allocated them to goods transportation. Lao et al. (2021) estimated that the amount of emission reduction should be 63 and 79 kg CO 2 eq/100 km for the 14-24-tonne FCET (fuel economy at 28.2 L/100 km) and 29-31-tonne FCET (fuel economy at 38.1 L/100 km) in 2017, respectively. This study has assessed that the amount of emission reduction is 37 and 14 g CO 2 eq/tonne•km for 18-tonne FCETs (fuel economy at 23.9 L/100 km) and 31-tonne FCETs (fuel economy at 35.9 L/100 km), respectively. Owing to the payload impact, the emission reduction per cargo of the 31-tonne FCET was much lower than that of the 18-tonne FCETs. When the emission reduction per unit of cargo in this study was converted to the emission reduction per vehicle, according to the payload of each vehicle type, the amount of emission reduction for 18-tonne and 31-tonne FCETs becomes 44 and 56 kg CO 2 eq/100 km, respectively. Compared with 2017, the fuel economy of ICETs in 2021 was significantly improved, resulting in a decrease in the relative emission reduction effect of FCETs. In addi-tion, the GHG emissions per unit of cargo are separately assessed by different vehicles in units of g CO 2 eq/tonne•km, which helps allocate GHG emissions to transported goods.

Conclusion
In this study, the GREET model was employed to investigate the impact of China's crude oil mix, electricity mix, and vehicle technology on EC and GHG emissions of road freight. From a fuel life cycle perspective, the contribution of this study is the assessment of indirect and direct environmental impacts per unit of cargo transport. This study updates the GREET model in terms of feedstock source, fuel pathway, and vehicle technology. The actual payloads of BETs and FCETs were used rather than the default values. Available studies have used the payloads of baseline diesel vehicles (Lao et al. 2021;Ren et al. 2022), but the current payloads Our results show that China's crude oil and electricity mixes have a significant impact on indirect EC and GHG emissions per unit of cargo transport. Approximately 92% of China's road freight uses diesel trucks, and China's dependence on crude oil imports has reached more than 70%. Therefore, the source of crude oil has become a key factor in reducing road freight emissions in China. This suggests that an increasing share of crude oil imports from North America will lead to an increase in indirect GHG emissions from road freight, and an increased share of China's crude oil imports from the Middle East and other Asia-Pacific regions will help reduce indirect GHG emissions from road freight. The use of BETs in road freight transport in China is relatively low (2%). The dominance of coal electricity generation in China's electricity generation structure and the increase in the share of biomass electricity generation are not conducive to road freight emission reduction. Policy implementation supporting electricity generation from renewable energy sources such as water, wind, and solar should be continued, and policy support to improve biomass electricity generation technology should be strengthened. Only 0.5% of the road freight in China uses FCETs. GHG emissions from hydrogen fuel production should be reduced by increasing investment in hydrogen production technology, storage, and transportation infrastructure.
From the perspective of the fuel life cycle, diesel-ICET has a greater potential for energy saving and emission reduction during the vehicle operation stage (TTW). Research and development of alternative energy sources and alternative fuel technologies should be supported to improve fuel efficiency. (Greenberg and Evans 2017;Kazancoglu et al. 2021). China enforced the sixth-stage emission restriction standard for heavy-duty diesel vehicles in 2019 to reduce the environmental pollution caused by trucks (Ministry of Ecology and Environment of the People's Republic of China 2018). This led to an improvement in the fuel economy of diesel-ICETs. BETs and FCETs have greater energy-saving potential in the WTT fuel production stage and vehicle operation stage (TTW) (see Fig. 5 and Table 9) and significant emission reduction potential in the WTT fuel production stage (see Fig. 6 and Table 10). In 2019, BEV and FCEV technologies were used in trucks. Refrigerated vehicles employing these vehicle technologies can reduce emissions by 58% and 55%, respectively (see Fig. 6). The application of these two advanced vehicle technologies to road freight will significantly reduce GHG emissions per unit of cargo. However, the large-scale application of BEV and FCEV technologies in freight will put great pressure on the energy sector to reduce emissions under the current electricity mix scenario, hydrogen storage, and transportation technology. Decision-makers should formulate application and promotion measures for specific vehicle classes, types, and technologies. BEV technology should be popularized in LDTs while minimizing their weight. The application of medium and heavy-duty FCETs in road freight should also be increased based on technological improvements in hydrogen fuel production, storage, and transportation.
Based on the results, policymakers should formulate transport emission reduction policies along the automotive energy supply chain from energy supply and delivery to energy use. If the government reasonably considers the "upstream" emissions of new energy vehicles in the formulation of policies and regulations, it will help China achieve a GHG emission peak soon (He et al. 2020). Although this study focuses on road freight emission reduction, the EC and GHG emissions involved in energy supply, production, and transportation are also significant for China's energy sector. The emission reduction of road freight is a global concern for environmental sustainability, and this research method can also be used to assess it in other countries. This study focuses on the impact of the crude oil mix and electricity mix on road freight emission reduction. With the increasing application of FCEV technology in road freight, hydrogen production materials and pathways will also become important factors affecting road freight emission reductions in China. Fuel, vehicle technology, and vehicle acquisition costs are also important factors that influence emission reduction (Lajunen and Lipman 2016;Orsi et al. 2016;Zeng et al. 2021), and the impact of these factors on road freight emission reduction will be considered in the future.      Table 9 Energy consumption (MJ/tonne•km) of various trucks during the WTW stage