How to reduce the greenhouse gas emissions and air pollution caused by light and heavy duty vehicles with battery-electric, fuel cell-electric and catenary trucks

The reduction of greenhouse gas emissions is one of the greatest global challenges through 2050. Besides greenhouse gas emissions, air pollution, such as nitrogen oxide and particulate matter emissions, has gained increasing attention in agglomerated areas with transport vehicles being one of the main sources thereof. Alternative fuels that fulfill the greenhouse gas reduction goals also offer the possibility of solving the challenge of rising urban pollution. This work focuses on the electric drive option for heavy and light duty vehicle freight transport. In this study, fuel cell-electric vehicles, battery-electric vehicles and overhead catenary line trucks were investigated, taking a closer look at their potential to reduce greenhouse gas emissions and air pollution and also considering the investment and operating costs of the required infrastructure. This work was conducted using a bottom-up transport model for the federal state of North Rhine-Westphalia in Germany. Two scenarios for reducing these emissions were analyzed at a spatial level. In the first of these, selected federal highways with the highest traffic volume were equipped with overhead catenary lines for the operation of diesel-hybrid overhead trucks on them. For the second spatial scenario, the representative urban area of the city of Cologne was investigated in terms of air pollution, shifting articulated trucks to diesel-hybrid overhead trucks and rigid trucks, trailer trucks and light duty vehicles to battery-electric or fuel cell-electric drives. For the economic analysis, the building up of a hydrogen infrastructure in the cases of articulated trucks and all heavy duty vehicles were also taken into account. The results showed that diesel-hybrid overhead trucks are only a cost-efficient solution for highways with high traffic volume, whereas battery overhead trucks have a high uncertainty in terms of costs and technical feasibility. In general, the broad range of costs for battery overhead trucks makes them competitive with fuel cell-electric trucks. Articulated trucks have the highest potential to be operated as overhead trucks. However, the results indicated that air pollution is only partially reduced by switching conventional articulated trucks to electric drive models. The overall results show that a comprehensive approach such as fuel cell-electric drives for all trucks would most likely be more beneficial.


Introduction
The reduction of greenhouse gas (GHG) emissions, such as carbon dioxide (CO 2 ), as well as harmful emissions, such as particulate matter (PM) and nitrogen oxides (NO x ), are major challenges of contemporary society. While the Intergovernmental Panel on Climate Change IPCC (2013) has identified GHG emissions as the main drivers of the climate change, harmful emissions also affect human health, leading to higher risk of mortality and respiratory morbidity . In Europe in 2016, road transport produced 29% of NO x , 8% of PM with a diameter below 10 µm (PM10) and 10% of PM with a diameter below 2.5 µm (PM2.5) (EEA, 2018). In the case of GHG emissions, in 2017 19.7% of total EU-28 emissions came from road transport . Furthermore, in 2017 heavy duty vehicles (HDVs) 1 and light duty vehicles (LDVs) 2 produced 5.2% and 2.4% of total EU-28 GHG emissions . In total, HDV and LDV transport was responsible for 38.2% of road transport in EU-28 GHG emissions in 2017. To reduce harmful emissions, the European Commission set limit values for harmful emissions with the Council Directive 1999/30/EC in 1999(EC, 1999. In the case of GHG emissions, in 2015, the European Commission set forth a binding policy framework to reduce these in the EU by 30% of 1990 levels by 2030 in order to reach the climate goals laid out at the Paris Climate Conference (EC, 2014). Kluschke et al. (2019) have stated that in Europe's road freight transport sector, heavy freight trucks produce the largest share of CO 2 emissions. For the most populated state in Germany, which is North Rhine-Westphalia (NRW), the State Office for Nature, Environment and Consumer Protection North Rhine-Westphalia (LANUV, 2015) has developed a detailed spatial model to calculate GHGs and other harmful emissions. They showed that road transport produces 12% of CO 2 emissions, 30% of NO x emissions and 38% of PM10 emissions in the state of NRW in Germany (LANUV, 2015) 3 . Breuer et al. (2020) investigated harmful emissions in NRW in 2018 and found that about 10% of NO x in urban emission hotspots were produced by LDVs and up to 28% emanated from HDVs. Furthermore, they showed that in an emission hotspot, LDVs produce 30% and HDVs 20% of exhaust PM10 emissions. For road, tire and brake PM10 emissions in an exemplary emissions hotspot, 5-10% were produced by LDVs and 20-25% by HDVs . In summary, Breuer et al. (2020) noted that diesel vehicles are responsible for 70% of PM10 emissions and 92% of NO x emissions in an exemplary emission hotspot, whereas diesel vehicles had a share of 56% of the total distance driven.
This work contributes an in-depth analysis of the potential impact of electric drive options for LDVs and HDVs on GHG emissions and harmful emissions, followed by an economic assessment of different promising scenarios. For this purpose, the existing traffic model developed by Breuer et al. (2020) was adopted. Breuer et al. (2020) took NRW as a suitable model region for Germany, given its high impact of 20% on Germany's CO 2 emissions (IT.NRW, 2017) and by the fact that 14 cities in the state are being sued for exceeding harmful emission limits.
This paper is structured in the following manner. First, an overview of potential alternative fuel pathways for short-distance and longdistance freight transport is given. Second, GHG emissions, harmful emissions and the costs of alternative fuels are discussed. Third, the methodological framework for the simulation of future scenarios using electric drive options are explained. These include a best-case scenario in economic terms for GHG emissions reductions on federal highways, a best case scenario for urban pollution reduction for the selected emission hotspot of Cologne and further cases for the economic evaluation. Fourth, the present traffic volume of HDVs and LDVs, present emissions and future emissions are shown in spatial resolution and discussed in detail by applying the best case scenarios for GHG emissions and urban pollution reduction. Finally, the CO 2 -specific costs of the different cost cases, as well as their reduction potential, are assessed.  Fig. 2 illustrate selected potential fuel options for two different types of goods transport. Fig. 1 shows possible future fuel pathways for goods transport in urban areas, whereas Fig. 2 shows fuel pathways for goods transport with long-haul trucks. The preferred cradle-to-wheel routes for goods transport in a narrow district, which are compressed bio gas (CBG), fuel cell-electric trucks (FCETs) and battery-electric vehicles (BEVs), can be easily realized with a fleet of vehicles, although the fueling infrastructure must be changed (see Fig. 1). This change is limited to a single filling station at the hub. Because of the high chain efficiency and low eneergy density of batteries, BEVs have high potential for use in short distance transport and low-to-medium weight goods. Alternative options include hydrogen in FCETs, which is extremely important for buses and HDVs in general, and CBG to be combusted in ICEs. CBGs have the advantage of lower infrastructure costs and the possible retrofitting of existing vehicles. Fig. 2 shows selected options for future goods transport with longhaul trucks. CBG, hydrogen and dimethyl ether (DME) from Power-to-Fuel (P2F) are good candidates that use internal combustion engines. An electric drive system for long-haul trucks can be based on a catenary system or hydrogen for fuel cells.

Emissions reduction potential of alternative fuels
The primary goal of the energy transition in the context of transport is to reduce greenhouse gas emissions. While methane (CH 4 ) and nitrous oxide (N 2 O) have large global warming potentials with values of 28 and 265 (IPCC, 2014), direct CO 2 emissions with a share of 99% on total CO 2 -equivalents are the main driver in road transport in Germany (LANUV, 2015;UBA, 2020). Therefore, at first glance, the CO 2 emissions for various drive train technologies for LDVs and HDVs are key to transitioning from a fossil base to sustainable primary energy. Fig. 3 shows the specific CO 2 emissions for BEVs, fuel-cell electric vehicles (FCEVs) and internal combustion engine (ICE) vehicles fueled by the EU's electricity mix, with hydrogen from natural gas and fossil diesel, respectively. The production chain emissions are highest for BEVs, i.e., 136-150 g CO 2,eq. /MJ f , followed by hydrogen-fueled FCEVs based on natural gas with 126-141 g CO 2,eq. /MJ f (Edwards et al., 2014a). The combustion of fossil diesel in an internal combustion engine leads to 84.6-87.8 g CO 2,eq. /MJ f (Edwards et al., 2014a). It is also notable that the German electricity mix improved its carbon footprint from 154 g CO 2,eq. /MJ f in 2010 to 111 g CO 2,eq. /MJ f in 2019 (Icha and Kuhs, 2020). Considering the CO 2 emission factors, it is important to take into account that only the emissions factor of the German electricity mix is from 2019, whereas all other emission factors are based on the work of Edwards et al. (2014a), which was published in 2014. This is the reason for the large gap between the EU electricity mix and German electricity mix in terms of well-to-wheel emissions, as is shown in Fig. 3. The bar tank-to-wheel w/o credits in Fig. 3 shows the CO 2 emissions according to Europe's renewable energy directive (RED II). In this case, the CO 2 emission factor of BEVs and FCEVs drops to zero as a result of zero exhaust emissions, whereas conventional ICE fuels drop somewhat and the CO 2 emissions factor for power-to-fuel (P2F) fuels even increases from near zero (for well-to-wheel) to approximately the same as for conventional fuels. The reason for this is that, in the case of tank-towheel emissions, the upstream chain of P2F production, where the CO 2 is used as a resource for production, is not considered.
Additionally, the comparison of well-to-wheel emissions for the different options in g CO 2 /MJ f is somewhat misleading, because the total consumption is not reflected. It is important to note that the values in g CO 2 /MJ f must be converted into g CO 2 /km by applying the different conversion efficiencies (tank-to-wheel efficiencies), i.e., about 81% for BEVs, 55% for FCEV drive systems and 42% for ICE-powered HDVs (see Fig. 3).
The mechanical energy of the wheels is assumed to be 480 MJ/km for an HDV. For fossil-based chains, the order changes to 1060-1335 g CO 2 / km for FCEVs, 970-1050 g CO 2 /km for ICEs and 765-945 g CO 2 /km for BEVs. Finally, it must be stated that a strong CO 2 reduction can only be achieved by near total electricity generation based on renewables. Such well-to-wheel chains are hydrogen-fueled FCEVs (and trucks such as FCETs) and BEVs, both of which are based on renewable electricity, as well as ICEs fueled with biogas or electro fuels from renewable hydrogen and carbon dioxide. Carbon dioxide should ideally be originated by biomass, direct air capture or the capture of unavoidable process gases from industry. Fossil-based power plants are excluded as point sources.
2 Trucks with a gross vehicle weight < 3.5 t. 3 The cited paper provides the total emissions for 2016 and road transport emissions for 2013, which have been compared to gain the percentage shares.
As is outlined above, Europe's renewable energy directive (RED II) does not give any credits to synthetic fuels based on renewable fuels. Without credits for renewable liquid fuels, these count as fossil fuels and will not provide any incentives for the transport sector. Germany's National Hydrogen Strategy (BMWi, 2020) would like to change this situation to foster synthetic fuels in addition to hydrogen as the backbone of a future energy system.
The specific costs of renewable fuels play a key role in the economic profiles of LDVs and HDVs. Fig. 4 shows the specific CO 2 emissions and costs for different fuels based on Edwards et al. (2014b) andSchemme et al. (2020). Fossil-based fuels can be found in the range of 10-20 €/GJ f , bio fuels between 20 and 30 €/GJ f , hydrogen (H 2 ) at 45 €/GJ f for gaseous storage and 60 €/GJ f for liquid storage, whereas electro fuels are the most costly with 53-100 €/GJ f , depending on the type of fuel. It is clear that certain biofuels have serious CO 2 emissions in the range of 8-63 g CO 2,eq. /MJ f (see Fig. 4). Using waste such as liquid manure, black liquor or municipal waste is advantageous, because carbon capture and utilization chains avoid CO 2 emissions and lead to credits in between 45 and 75 g CO 2,eq. /MJ f . Progress for hydrogen and hydrogen-based fuels can only be made if the costs of hydrogen can be reduced from 4.6 €/kg to about 3 €/kg by shifting the hydrogen production to countries with excellent renewable energy sources and transport the hydrogen to  Europe. Finally, if near-zero CO 2 emissions are envisaged, the costs increase in the order of biofuels, hydrogen and electro fuels. Biofuels are not generally preferable, because a number of production chains emit a significant amount of CO 2 (see Fig. 4). If municipal waste, liquid manure, dry manure, black liquor, straw or woody residues are used as a feedstock, the CO 2 reduction is feasible and advantageous.
A further aspect for the selection of fuel and drive train technologies is the reduction of PM and NO x emissions. Liquid and gaseous electro fuels can reduce the CO 2 emissions in all mobile applications, but the limited pollutants NO X and PM will only partially be reduced. In case of biodiesel, fatty acid methyl esters (FAME) lead to increasing NO x emissions and strongly decreasing PM emissions, while hydrotreated vegetable oils (HVO) lead to decreasing NO x and PM emissions (Verbeek et al., 2014). The utilization of Gas to Liquid (GTL) diesel results in similar reduction as the utilization of HVO (Verbeek et al., 2014). According to Unglert et al. (2020), HVO and GTL diesel are very similar. However, Verbeek et al. (2014) state that, caused by limited data, the reduction of emissions in EURO VI vehicles is uncertain. The great advantage of these fuels is the reduction of local emissions from already existing vehicles with low emission standards due to the Drop-In possibility. In natural gas vehicles, the rate of emissions reduction depends on the engine technology used. Dual-Fuel concepts with a small amount of conventional diesel have most likely emissions in the same range as conventional EURO VI diesel engines (Otten et al., 2017;Vermeulen, 2019;Matzer et al., 2019). New high efficiency EURO VI gas engines in stoichiometric operation offer slightly lower emissions in comparison to conventional diesel engines (Matzer et al., 2019;Verbeek et al., 2014). A detailed extensive analysis of both biofuels and electro fuels regarding their emission behavior is performed in the review of Peters et al. (2021). Ongoing projects are investigating the measuring of emissions of different engine technologies that apply such fuels (C3-Mobility 2021; Virtual Institute 2020). The advantage is the existence of the corresponding infrastructure such as transport trucks and fueling stations. Due to the lower cradle-to-tank efficiency, more renewable energy must be introduced into the system.
Hydrogen as a fuel can also significantly reduce the CO 2 emissions for all mobile applications and completely reduce limited pollutants. Both, FCETs and the already above mentioned overhead catenary trucks, are zero-emission technologies and should be preferred for the elimination of local emissions. For trucks, articulated trucks have with Fig. 3. Specific CO 2 emissions for BEVs, FCEVs and ICEs fueled by renewable electricity, renewable hydrogen and electro fuels compared to BEVs using the EU electricity mix and today's German electricity mix, with hydrogen from natural gas and fossil diesel based on Edwards et al. (2014a) and Icha and Kuhs (2020). ( ) completely renewable power trains for well-to-wheel analysis. (*) tank-to-wheel analysis according to RED II without credits for synthetic fuels based on renewable hydrogen. 100,000 km/a the highest yearly mileages, while rigid and trailer trucks drive only 24,000 km/a (Bäumer et al., 2017). Articulated trucks have also a high share of distance driven on federal highways. According to the Federal Highway Research Institute (Bäumer et al., 2017), rigid trucks drive 32% of their overall distances on federal highways, trailer trucks 68% and articulated trucks 63%. Articulated trucks therefore have the highest affinity for operating as overhead catenary trucks. The question arises as to which technology, i.e. FCETs or overhead catenary trucks, offer the higher reduction potential in the case of GHG, NO x and PM emissions and at what costs. Previous studies such as Kühnel et al. (2018), Mareev and Sauer (2018), Gerbert et al. (2018), Mulholland et al. (2018), Plötz et al. (2019) and Wietschel et al. (2017) have researched overhead lines for heavy duty vehicles, but did not analyze their effect on local emissions. Only Gerbert et al. (2018) calculated CO 2specific costs for catenary trucks. A cost comparison, as shown in Fig. 4, is not possible for a ramp up of catenary trucks, because costs and CO 2 emissions strongly depend on the length of sections equipped with overhead lines and the local traffic. A spatial analysis is mandatory. As discussed above, FCETs and battery-/overhead-electric trucks are the only zero exhaust emission technologies and therefore comparable options. For this reason, we focus on electric drive systems and thus select BEVs and FCETs as options for short distance transport and the overhead catenary system and FCET as options for long-distance transport. This study provides information for deciding which technology should be considered in future mobility.

Methodological framework
This paper builds on the analysis results of an emission model for the federal state of NRW in Germany. In previous work, Breuer et al. (2020) applied multivariate linear regression to statistical data such as local population, as well as gross domestic product, and combined it with existing traffic counts to approximate the traffic volume in every street section. Subsequently, the traffic volume of different vehicle classes was combined with the German fleet mix in 2018 based on Handbook of Emission Factors for Road Transport (HBEFA) 3.3 as well as with the corresponding emission factors from HBEFA 3.3 and the EMEP/EEA Air Pollutant Emission Inventory Guidebook 2016 (EEA, 2016). Based on Heldstab et al., 2003, it was assumed that 100% of PM10 exhaust emissions are also PM2.5 emissions. Breuer et al. (2020) used the developed approach to calculate CO 2 , NO x , HC, CO, PM 2.5 and PM 10 emissions from exhaust, as well as PM 2.5 and PM 10 emissions from tire, brake and road surface abrasion and the fuel consumption for eight vehicle classes in each street section. A detailed description of the methodology can be found in Breuer et al. (2020). The present work focuses on CO 2 , NO X and particulate matter emissions PM 2.5 . Furthermore, the fleet mix, total distances driven by different vehicles, the fuel mix and the emission factors in the model of Breuer et al. (2020), originally for the year 2018, have been adapted to the year 2017 in this study. The share of emission classes for the different vehicles in the applied fleet mix HB33 from HBEFA 3.3 is shown in the electronic supplementary information in the figures A 1 and A 2. Vehicle classes are described in table A 1 in the electronic supplementary information. The illustration of the spatial distribution of emissions and mileages in this work was carried out using the line sources from the street network, but also using urban areas as observation level. Urban areas in this study were defined by urban morphological zones from the European Environment Agency (EEA, 2014) with an area > 10 km 2 . The ranges in the legend of the spatial maps were produced using Jenks Natural Breaks Classification in the geographic information system application QGIS. The developed model approach from Breuer et al. (2020) was extended to calculate different future scenarios using the drive concepts diesel-hybrid catenary trucks, battery catenary trucks, fuel-cell electric trucks and battery-electric vehicles. Exhaust emissions of the concepts were assumed to be zero, with diesel-hybrid catenary trucks being the exception. In the following section the calculated spatial future scenarios, the different economic cases as well as assumptions will be explained.

Scenarios for the reduction of GHG emissions and harmful emissions
Following the analysis of present emissions in 2017, two potential future scenarios were applied to the model to reduce GHG emissions and harmful emissions. Fig. 5 shows an overview of both the calculated emission scenarios and the implemented technologies. The first scenario, Diesel-Hybrid Cat Ramp-Up, focuses on long-distance transport on federal highways and the resulting greenhouse gas emissions. In this scenario, articulated trucks on selected federal highways in NRW were shifted from diesel combustion engines to diesel-hybrid overhead catenary trucks. The second scenario, Urban Pollution Reduction, focuses on short distance transport in urban areas and the resulting harmful emissions. In the second scenario, LDVs in the urban area of Cologne were shifted from diesel combustion engines to FCETs and BEVs, and rigid trucks and trailer trucks were shifted from diesel combustion to FCETs on federal highways and to BEVs on federal roads, state roads, district roads and other streets. Furthermore, articulated trucks were shifted to diesel-hybrid overhead catenary trucks on federal highways in the second scenario. 4 For the first scenario Diesel-Hybrid Cat Ramp-Up, federal highways with the highest traffic volume for articulated trucks were equipped with overhead catenary lines. Wietschel et al. (2017) researched federal highways in Germany and developed contiguous corridors with the highest traffic volume of articulated trucks. Criteria besides the traffic volume were the mesh topology and the spatial distribution (Wietschel et al., 2017). As a result of this methodology, they classified these corridors into the groups 1000, 2000, 3000, 4000 and 8000. These groups are shown in Table 1. The group names describe the rounded length of the federal highway corridor/corridors. For a ramp up of overhead catenary trucks, Wietschel et al. (2017) recommend equipping these corridors starting with group 1000. From group 1000 upward to group 8000, it was sought, as already mentioned above, to include in each group the sections with the highest articulated truck traffic volume. It was also considered that the sections in each group are connected if possible. However, this rule does not apply for group 8000, which also includes single separate sections. Wietschel et al. (2017) also calculated the investment and operating costs for overhead catenary lines. They distinguished between two vehicle technologies and two cases: diesel-hybrid overhead catenary trucks and battery overhead catenary trucks, as well as a starting phase with the electric operation of trucks on federal highways in group 4000 and a final expansion with electric operation on all federal highways in Germany. For our study, the federal highways included in the case of diesel-hybrid overhead catenary trucks and the starting phase (i.e., federal highways included in group 4000), were acquired from the work of Wietschel et al. (2017). The other cases were investigated during the economic evaluation (see 2.2 Cases for the economic evaluation). The federal highway sections of group 4000 inside the model area NRW are listed in Table 2.
The scenario Urban Pollution Reduction focuses on harmful emissions in the urban area of the exemplary emission hotspot of Cologne and is an optimistic best case scenario. In this scenario, LDVs were fully shifted to FCETs and BEVs, rigid trucks and trailer trucks were shifted to FCETs on federal highways and to BEVs on federal roads, state roads, district roads and other streets. Moreover, articulated trucks were shifted to dieselhybrid overhead catenary trucks, which operate in electric mode on federal highways and in diesel mode on all other streets. In contrast to scenario Diesel-Hybrid Cat Ramp-Up, articulated trucks in Urban Pollution Reduction operate not only on selected federal highways of group 4000 in electric mode, but also on every federal highway section within the county of Cologne. Furthermore, particulate matter emissions from the abrasion of the overhead catenary lines were neglected. For both scenarios, a 100% replacement of LDVs and HDVs was assumed. In addition, BEVs and FCETs were considered with zero exhaust emissions and PM abrasion emissions of ICEVs, BEVs and FCETs were assumed to be the same.

Cases for the economic evaluation
Alongside the evaluation in the case of emissions, an economic evaluation with respect to costs in EUR per saved tones of CO 2 was performed in this work. The catenary line cases are based on Wietschel et al. (2017) and the FCET scenarios on investment and operating costs from Kramer et al. (2018). Table 3 provides an overview of the calculated cases. The first scenario is the already explained Diesel-Hybrid Cat Ramp-Up. In this scenario, diesel-hybrid overhead articulated catenary trucks are operated on selected highways, which are in group 1000-4000 from Wietschel et al. (2017). 90% of these highways are equipped with catenary lines. The resulting 10% of length are operated using the battery. Beyond these selected federal highways, diesel-hybrid trucks operate in diesel mode in the secondary street network and on the non-equipped federal highways. In the Diesel-Hybrid Cat case, dieselhybrid overhead articulated catenary trucks are operated on all federal highways in NRW and once again, 90% are equipped with catenary lines. In this scenario, diesel-hybrid overhead articulated catenary trucks operate in diesel mode within the secondary street network. For the Battery Cat scenario, the vehicle technology was changed to overhead catenary articulated trucks without diesel engines. In that scenario, these overhead catenary articulated trucks are operated on all federal highways in NRW. On every other street, they operate in battery mode. The scenario was subdivided into a minimal and a maximal cost case. For the minimal case, 30% of federal highways were equipped with

Table 1
Federal highway corridors by Wietschel et al. (2017) and federal highways in Germany with their absolute length, share of the total federal highway length and share of the total articulated trucks mileages.  Germany for all HDVs, as well as 100% equipment on federal highways using overhead catenary lines. Even for a conservative maximal scenario, this electrification share seemed somewhat high, and so it was adjusted in this work. The determination of the adjusted value is described in the next section. If the federal highways next to all 509 federal highway access points of the federal highways in NRW are equipped with 3 km (Wietschel et al., 2017) of overhead catenary lines, this would yield 68% electrification, ensuring that overhead catenary line articulated trucks, which have a range of 100-200 km (Wietschel et al., 2017), have fully charged batteries when they leave the federal highways for the secondary street network. Wietschel et al. (2017) stated that every point in the street network in Germany is within the range of 50 km from a federal highway and, in consequence, battery trucks can reach these points and even return to the federal highway access point. This hypothesis was checked for this study. Fig. 6 shows a 1 km 2 grid with the number of federal highway access points within the range of 50 km (a), 40 km (b), 30 km (c) and 20 km (d). Fig. 6a clearly illustrates that every point in NRW is reachable within a 50 km radius from the federal highway access points. Areas without a reachable highway access point appear with a radius of 40 km and increased when the radius was decreased to 30 or 20 km. However, federal highway access points that are outside of NRW were not considered in this analysis and may supply these white areas that are on the state borders. Based on the performed investigation, 68% of the federal highways were equipped with catenary lines in the maximum case of the scenario Battery Cat. The additional 15-21 bn. € for an overhead catenary line system for Germany by Kramer et al. (2018) were adjusted for articulated trucks in NRW using the work of Breuer et al.  2017) and vehicle energy consumption from HBEFA 4.1, the share of road transport energy consumption in NRW relating to Germany was calculated to be 20.57%. The share of energy consumption of articulated trucks on total road transport energy consumption inside NRW in 2017 was calculated to 13.28%, using the model based on Breuer et al. (2020), while the share of freight transport energy consumption was calculated to 28.23%. This leads to cost shares of 2.77% for the case Fuel Cell AT and 5.81% for the case Fuel Cell HDV. Calculated investment costs for the cases Fuel Cell AT and Fuel Cell HDV were 0.53 bn. € to 1.05 bn. € and 1.01 bn. € to 2.21 bn. € for infrastructure as well as 1.97 bn. € to 2.41 bn. € and 4.12 bn. € to 5.05 bn. € for production. Caused by these cost intervals, the scenarios Fuel Cell AT and Fuel Cell HDV were subdivided into a minimum cost case and a maximum cost case, respectively. The operating costs in both scenarios were Table 3 Analyzed cases in the economic evaluation, including a brief description, as well as the investment and operating costs used.

Scenario
Description Investment costs Operating cost

Diesel-Hybrid Cat Ramp-Up
Diesel-hybrid overhead articulated catenary trucks were operated on the selected highways inside Group 1000-4000 from Wietschel et al. (2017). 90% were equipped with catenary lines. assumed to be 5% of the investment costs based on Kramer et al. (2018). Table 4 provides an overview of the parameters for the economic evaluation. The interest rate of 4%, as well as the 40 years of depreciation for infrastructure and 20 years of depreciation for production plants were taken from Kramer et al. (2018), who assumed these values for the economic evaluation of the infrastructures and fuel production of battery-electric and fuel cell-electric vehicles. The working capital was assumed to be 10% of the investment costs. . 7a shows the traffic volume of HDVs in 2017 on a street level in millions of vehicles on federal highways, federal roads, state roads and district roads. The traffic on the remaining other streets was also calculated and considered during the simulation, but for reasons of clarity is not shown here. The highest traffic volume is on the federal highways A42, A2, A3, A4 and A1. The numbers reach between 5.8 and 8.8 million HDVs per year. Similarly, Fig. 7b shows the LDV traffic on federal highways, federal roads, state roads and district roads in the local state of North-Rhine Westphalia in 2017 indicated by numbers in millions of vehicles. As for HDVs, LDV traffic on the remaining others streets was also calculated and is considered in the following emission calculations, but for reasons of clarity is not shown in Fig. 7b. High numbers in the category 3.1-7.9 million LDVs per year were reached on federal highways around the city of Cologne, on the A3 between Cologne and Düsseldorf and the Ruhr area around Duisburg, Essen and Dortmund. These figures support the impression that LDVs play an important role in road transport and clearly show that HDV traffic is mostly focused on federal highways, whereas high LDV traffic occurs around and within urban areas.  Fig. 8a displays the simulated CO 2 emissions caused by road traffic on federal highways. These emissions are subdivided into caused by LDVs, HDVs (i.e., rigid trucks, trailer trucks, articulated trucks and buses) and remaining vehicles (i.e., passenger cars, motorcycles and other vehicles) for selected federal highways with the highest CO 2 emissions. The highest CO 2 emissions can be observed in the Ruhr area on the federal highways A3, A42 and A1, as well as around the city of Cologne on the highways A4 and A3, with values of 11-17 kt CO 2 per km. These highways also belong to the corridors defined by Wietschel et al. (2017). For the majority of the federal highways shown with high CO 2 emissions, the share of LDVs and HDVs is nearly 50% of the total CO 2 emissions (see Fig. 8). The share of LDVs itself is about 10-15% and plays a minor role in CO 2 emissions most likely due to the short distances traveled by retail traffic. Fig. 8b shows the NO x emissions in the state of NRW in 2017, indicated by t per km 2 and sketched for selected urban areas. NO x emissions were subdivided for HDVs without federal highways (i.e., HDVs on federal roads, state roads, district roads and other streets), HDVs on federal highways, as well as LDVs and remaining vehicles on all streets. Federal highways in urban areas are marked in blue in Fig. 8b-d. The NO x hotspots were identified in the Ruhr area around Duisburg and  Essen, with values of 15-18 t/km 2 in the northern part of Cologne and in the urban area of Wuppertal. The city of Wuppertal is embedded between two federal highways, namely the A46 and A1, and is located in a narrow valley. The share of road freight transport amounts 26-37% in urban areas, with a majority of this being HDV transport (18-27%). The share of HDVs driven on federal highways and other roads accounted for NO x emissions in urban areas of between 0 and 7% and 10-25%, depending on the individual location (see Fig. 8b). Fig. 8c shows the  PM10 emissions in NRW in 2017 indicated by t per km 2 and sketched for selected urban areas. The PM10 emissions were also subdivided for HDVs not on federal highways, HDVs on federal highways, LDVs and remaining vehicles. The hotspots can be identified again in the Ruhr area with values between 1 and 1.2 t/km 2 and in the urban area of Wuppertal. Further on, critical values of between 0.8 and 1 t/km 2 were found in the northern part and some areas in the south of Cologne, as well as in Aachen, Bonn and Krefeld (an urban area southwest of Duisburg). The share of road freight transport on PM10 emissions amounts to 26-39% in urban areas, with the majority of HDV transport amounting to 12-25%. The share of HDVs driven on and off of federal highways induced PM10 emissions in urban areas differs between 0-12% and 12-20%, depending on the individual location. Fig. 8c clearly shows that the effect of LDV transport on PM10 emissionsindicated by a share of 10-14% -is somewhat higher than that for NO x emissions (8-11%). The reason for this is that NO x emissions only occur due to fuel combustion, whereas PM emissions have different sources. PM emissions originate from fuel combustion and corresponding tail pipe emissions, due to rubber friction from wheels, as well as friction from asphalt and brake dust. The latter three sources depend on the number of vehicles on the road. Road transport by LDVs is indicated by a higher number of vehicles than HDV transport, whereas the fuel consumption and therefore the PM10 emissions from tail pipes are higher for HDVs than LDVs. Fig. 8d shows PM2.5 emissions in NRW in 2017 indicated by kg per km 2 and sketched for selected urban areas. As for PM10 emissions, the hot spots were identified in the Ruhr area, with values of 650-780 kg/km 2 and in the urban area of Wuppertal. As stated above, the city is embedded between two federal highways, i.e., the A46 and A1, and is located in a narrow valley. Further on, critical values of between 530 and 650 kg/km 2 were observed in the northern part and in some areas in the south of Cologne, as well as in Aachen, Bonn and Krefeld (see Fig. 8d). The share of road freight transport in PM2.5 emissions does not differ in all of the selected urban areas in relation to PM10 emissions. Fig. 9 shows the share of wear emissions (tires, brakes, surface) and exhaust emissions from HDVs, LDVs and remaining vehicles for the ten urban areas with the highest PM emissions in 2017. As was already observed in Fig. 8b-d, the share in each urban area seems to be fairly similar. Overall, the share of abrasion and exhaust emissions on PM10 emissions is about 75% and 25%, respectively. This relationship changes in the case of PM2.5 emissions. For these, about 65% is produced through abrasion and 35% via exhaust. Exhaust PM emissions are fully PM2.5 emissions (Heldstab et al., 2003), whereas particulate matter from abrasion is only partly PM2.5. HDV PM emissions are primarily Fig. 9. Share of wear emissions (tire, brake and surface) and exhaust emissions from HDVs, LDVs and remaining vehicles for the ten urban areas with the highest PM emissions in 2017. produced by abrasion. The reason for this is probably the high loads and resulting weights of HDVs. LDVs produce almost the same amount of PM10 abrasion and exhaut emissions, whereas for PM2.5 emissions, the share of exhaust emissions is larger. For the remaining vehicles, which mainly consist of passenger cars in the case of total distance driven, the share of PM emissions from abrasion is much bigger than from exhaust. This is a consequence of the share of gasoline vehicles on passenger cars. HDVs and LDVs in Germany are mainly powered by diesel, whereas for passenger cars, only about 56% are diesel-based. As was found in the work of Breuer et al. (2020), exhaust PM10 emissions from passenger cars powered by diesel in the current German fleet are four times higher than for passenger cars powered by gasoline.

Emissions from light duty and heavy duty vehicles
Which findings can be drawn from this model approach adapted to a populous region in Germany such as the federal state of NRW? -CO 2 emission hotspots are in the Ruhr area on the federal highways A3, A42 and A1, as well as around the city of Cologne on the highways A4 and A3. A change in the drive system technologies and a fuel switch could achieve a strong reduction in the limited pollutants in metropolitan areas and improve air quality, depending on technology and application.

Scenario Diesel-Hybrid Cat Ramp-Up: Reduction of GHG emissions on selected federal highways
In the first scenario, the replacement of ICE-driven articulated trucks by diesel-hybrid overhead catenary trucks was analyzed. Fig. 10 shows the tank-to-wheel CO 2 emissions of road transport on selected federal highways in NRW for (a) the base case of 2017 and (b) the simulated case with catenary lines for articulated trucks. Caused by only the investigation of tank-to-wheel emissions, CO 2 emissions from electricity generation were not considered at this point. In terms of well-to-wheel emissions, CO 2 emissions of articulated trucks operated with catenary lines would only be close to zero if the electricity generation is 100% from renewables (see Fig. 3). The share of articulated truck CO 2 emissions on total CO 2 emissions on those federal highways in the base case without catenary lines amounts to 10-25% (see Fig. 10a). The highest CO 2 emissions were calculated to 12,000-17,000 t/km per year on parts of the A3 and 8,000-10,000 t/km per year on the main parts of the A2, with some hotspots reaching 10,000 to 12,000 t/km per year. As Fig. 10b shows, the replacement of ICE-driven articulated trucks by diesel-hybrid overhead catenary trucks leads to a significant CO 2 reduction. Other low CO 2 emission technologies such as FCEVs would in this case lead to the same result, because the CO 2 exhaust emissions of overhead catenary line trucks and FCEVs are both zero on the federal highways. CO 2 emissions on the A2 were reduced to 5,000-8,000 t/km, and only a small section of the A3 remains at 12,000-17,000 t/km after the implementation of diesel-hybrid overhead catenary trucks according to the Diesel-Hybrid Cat Ramp-Up scenario. On the A3, large sections showed CO 2 emissions in the range 8,000-12,000 t/km. Fig. 11 shows the present share of CO 2 emissions from articulated trucks on selected federal highways referring to: (a) total CO 2 emissions in NRW from road transport; and (b) total CO 2 emissions from road transport on federal highways in NRW. It is clear that the deployment of diesel-hybrid overhead catenary systems on these selected federal highways would reduce total road transport CO 2 emissions in NRW by 3.7% and total road transport CO 2 emissions on federal highways by 9.6%. The shares of articulated trucks on selected federal highways are below 1%, with each relating to the CO 2 emissions of the complete traffic on all roads (see Fig. 11a) and between 0.5% and 2.4% each relating to the CO 2 emissions on federal highways in NRW (see Fig. 11b).   Table 2. indicated by kg per km. Fig. 12a and c were simulated for the base case assuming today's combustion engines und fuel mix. Fig. 12b and d consider the case of using FCETs and BEVs for LDV freight transport and for truck transport, as well as diesel-hybrid overhead catenary trucks on federal highways (see Fig. 5).

Scenario urban pollution reduction: Reduction of harmful emissions in urban areas
The highest emissions in the base case (a) and (c) can be observed on the ring of federal highways surrounding the city of Cologne. These amount to 36-47 kg NO x /km on the A3 federal highway and 24-36 kg NO x /km on the A1 and A4. The PM2.5 emissions amount to 1.2-1.7 kg PM2.5/km on the federal highway A3 and parts of the A4, and 0.7-1.2 kg PM2.5/km on the A1 and parts of the A4. In the case of replacing ICEdriven freight transport with HDVs and LDVs by diesel-hybrid overhead catenary trucks, FCETs and BEVs, limited emissions were partly reduced (see Fig. 12b and d). NO x emissions were reduced to 24-36 kg NO x /km on the northern part of the federal highway A3 and 10-24 kg NO x /km on all other federal highways. PM2.5 emissions were reduced to 0.7-1.2 kg PM2.5/km on the northern part of the federal highway A3 and 0.4-0.7 kg PM2.5/km on all other federal highways. Fig. 13 shows the NO x and PM2.5 emissions in the urban area of the emission hotspot of Cologne in 2017, as indicated by kg per km 2 and subdivided for articulated trucks, rigid and trailer trucks, light duty vehiles and remaining vehicles. The categories articulated trucks as well as rigid and trailer trucks are further subdivided in vehicles on federal highways and on roads off of federal highways. The left columns show the emissions in 2017 and the right columns those of the scenario Urban Pollution Reduction.
Road transport in the urban area of Cologne accounted for approximately 12,000 kg/km 2 NO x and 500 kg/km 2 PM2.5 emissions in 2017 (see Fig. 13). A share of 1/3 was caused by freight transport on roads, i. e., nearly 4,000 kg/km 2 NO x and 180 kg/km 2 PM2.5 emissions. Freight transport by LDVs was responsible for 1,500 kg/ km 2 NO x and 80 kg/ km 2 PM2.5 emissions. The corresponding emissions for HDV transport amounted to 2,500 kg/km 2 of NO x and 100 kg/km 2 of PM2.5, whereas the relationship between articulated trucks and other trucks (i.e., rigid trucks and trailer trucks) was calculated to 40:60 for NO x and 50:50 for PM2.5. Articulated trucks in Cologne were primarily driven on federal highways, whereas rigid trucks and trailer trucks were similarly used on federal highways and all other roads. As is shown in Fig. 8b-d, this is most likely not the case for all urban areas and makes Cologne a best case scenario for the application of diesel-hybrid overhead catenary Fig. 11. Present share of tank-to-wheel CO 2 emissions from articulated trucks on selected federal highways referring to: (a) total CO 2 emissions in NRW from road transport and (b) total CO 2 emissions from road transport on federal highways in NRW.
articulated trucks. Applying the scenario Urban Pollution Reduction reduces the NO x and PM2.5 emissions by 29% and 17%, respectively (see Fig. 13). The reason for the slightly lower reduction of PM2.5 emissions is the large share of abrasion emissions in PM (see Fig. 9). The diesel-hybrid overhead catenary articulated trucks on highways reduce articulated truck NO x emissions in Cologne by 81% and PM2.5 emissions by 24%. For both, NO x and PM2.5, the share of emissions in the base case, which is produced by articulated trucks not driven on federal highways is, with 1.27% for NO x and 1.19% for PM2.5, small (see Fig. 13, left columns, articulated trucks w/o federal highways). For PM2.5, this 1.19% consists of 34% exhaust emissions and 66% abrasion emissions. Table 5 shows the calculated CO 2 emissions from all vehicle classes for the selected federal highway sections of the scenario Diesel-Hybrid Cat Ramp-Up, for all federal highways and for all streets in NRW for 2017. These values are important to calculate the CO 2 -specific costs of the different economic cases. Fig. 14a presents an overview of the CO 2 -specific costs of the five different investigated scenarios. For the first case, Diesel-Hybrid Cat Ramp-Up, the distance for operating the diesel-hybrid overhead catenary trucks according to Wietschel et al. (2017) is 717 km (see Table 5). Therefore, 645.3 km (90%) must be equipped with overhead catenary lines. This distance, with investment costs of 1.7 m. €/km and annual operating costs of 30,556 €/km. lead to investment costs of 1.1 bn. € and operating costs of 19.7 m. €/a. The CO 2 reduction potential is 1.38 m. t/a (see Table 5, with the CO 2 emissions of articulated trucks on selected federal highways A1-A45). Assuming a depletion period of 40 years, an interest rate of 4% and a working capital share of 10%, the CO 2 -specific costs were determined to be 57.7 €/t of CO 2 . For operating diesel-hybrid overhead catenary trucks on 100% federal highways in NRW in the scenario Diesel-Hybrid Cat, the investment costs and annual operating  Wietschel et al. (2017) lead to CO 2specific costs of about 40.92 €/t CO 2 to equip 30% of federal highways in NRW with catenary lines in the minimal cost case. For the maximal cost case, the mentioned investment and operating costs to equip 68% of federal highways in NRW with catenary lines, and additional infrastructure costs of 1.93 bn. € lead to 114.94 €/t CO 2 (see Fig. 14 Breuer et al. (2020) used aggregated emission factors for vehicle classes, which led to equal energy consumption per kilometer of all HDVs (i.e., rigid trucks, trailer trucks and articulated trucks). Fig. 14b shows the reduction of CO 2 emissions in NRW, as well as NO x and PM2.5 emissions in the urban area of Cologne for different cases. The case of Diesel-Hybrid Cat Ramp-Up leads to a CO 2 reduction of 3.84% and NO x and PM2.5 reductions in Cologne of 0.25% and 0.09%. The harmful pollution reduction in this case is limited due to the limited length of federal highways with overhead catenary lines (see Fig. 10). The case Diesel-Hybrid Cat results in a CO 2 reduction of 8.62%, whereas cases where articulated trucks operate in electric mode on every street (i. e., Battery Cat and Fuel Cell AT) have a higher reduction of 13.55%.

Economic evaluation
Finally, the case Fuel Cell HDV, where all HDVs are operated in electric mode on every street, has a CO 2 reduction of 28.38%. In the case of harmful emissions in the urban area of Cologne, the change from dieselhybrid catenary articulated trucks to battery catenary articulated trucks leads to a higher reduction of 6.65% and 2.44% for NO x and PM2.5 emissions (see Fig. 14b, Diesel-Hybrid Cat and Battery Cat). The reduction of harmful emissions in Cologne for the case Fuel Cell AT is the same as the latter one. A higher reduction is only achieved by operating all HDVs in electric mode with fuel cells in the case of Fuel Cell HDV. This case results in a NO x reduction of 19.05% and a PM2.5 reduction of 6.66%.

Discussion
The conducted analysis of present CO 2 emissions and harmful emissions (NO x , PM10, PM2.5) clearly shows that freight transport by LDVs and HDVs accounts for 50% of CO 2 emissions on much frequented federal highways and 33% of NO x and 40% of particulate matter emissions in a representative urban area with high air pollution (see Fig. 8). Therefore, it is very important to introduce measures to reduce these values in accordance with the CO 2 reduction targets and to improve the air quality in urban areas. A fully electricity-based solution could use battery overhead catenary trucks or fuel cell-electric trucks for the mission profile of articulated trucks and a battery-driven or fuel celldriven E-drive for rigid trucks, trailer trucks and LDVs. CO 2 emissions could be completely avoided if the demanded electricity is fully derived from renewables (see Fig. 3). In this regard, the main disadvantage of BEVs is the restricted range. Diesel-hybrid overhead catenary trucks would be a partially electricity-based solution, only reducing emissions  Table 5 2017 tank-to-wheel CO 2 emissions in kt for the selected federal highway sections, all federal highways and all streets in NRW segmented for the different vehicle classes. Also displayed are the lengths of the federal highway sections that have been equipped with overhead catenary lines according to the scenario Diesel-Hybrid Cat Ramp-Up.
on federal highways.
As is shown in Fig. 7, trucks and articulated trucks (as part of HDVs) have been primarily operated on federal highways, whereas LDV traffic mostly occurs around urban areas. Bäumer et al. (2017) calculated that 37% of total LDV mileages, 32% of rigid truck mileages, 68% of trailer truck mileages and 63% of articulated truck mileages are driven on federal highways. Nevertheless, the contribution of CO 2 emissions from selected federal highways, which were equipped with a catenary system in the ramp-up phase of a diesel-hybrid overhead catenary line system scenario (Diesel-Hybrid Cat Ramp-Up), is only 3.9% against complete traffic CO 2 emissions (see Fig. 11a). However, in the scenario Diesel-Hybrid Cat Ramp-Up, only articulated trucks were assumed to switch to electric operation due to their having the highest affinity for this technology. Moreover, switching rigid trucks and trailer trucks to 100% electric operation would lead to an additional reduction of 1016 kt of CO 2 in the case of overhead catenary lines for federal highways in the group 4000 or 2295 kt of CO 2 for electric operation on all federal highways in NRW, which is, respectively, 2.8% and 6.4% of the total road transport CO 2 in 2017 (see Table 5).
During the calculations for this work, abrasion from the overhead catenary lines for articulated trucks was neglected. Pregger and Friedrich (2003) calculated the particulate matter abrasion emissions of the catenary lines of trains to be 0.08 g/km for PM10 and 0.012 g/km for PM2.5. Applying these emission factors to the driven distance from diesel-hybrid overhead articulated trucks in the scenario Urban Pollution Reduction in the urban area of Cologne for a rough estimation leads to additional PM10 emissions of 41 kg/km 2 and PM2.5 emissions of 6 kg/ km 2 , which were 5.5% and 1.2% of 2017 PM10 and PM2.5 emissions in Cologne. The additional emissions from the overhead catenary lines of the articulated trucks in the scenario Urban Pollution Reduction will reduce the achieved particulate matter reduction. Furthermore, the urban area of Cologne has a small share of emissions produced by HDVs on streets, that are not federal highways (see Fig. 8b-d). For the urban areas of cities like Essen or Duisburg, where the share of non-federal highway HDV traffic is higher, the potential reduction of harmful emissions through diesel-hybrid overhead catenary trucks would be much smaller because these would operate in diesel mode off of the federal highways and would have similar emissions as diesel trucks. Articulated trucks without diesel engines and larger batteries instead would solve this problem, but the investment costs per km would also be much greater for a battery-hybrid overhead catenary system, as reported by Wietschel et al. (2017). Another solution would be FCETs, which have zero exhaust emissions and no need for overhead catenary lines. The analysis of the urban area of Cologne showed that the use of overhead catenary line technology for articulated trucks, as performed in the scenario Urban Pollution Reduction in this work, has a larger impact in air pollution than using it for rigid and trailer trucks, because a majority of harmful rigid and trailer truck emissions are produced off of federal highways in the urban area of Cologne (68% of NO x and 65% of PM2.5 exhaust emissions, see Fig. 13).
The investigation of the different cases from an economic point of view showed that diesel-hybrid overhead articulated trucks are only a cost-efficient technology if used on highways with high traffic volume (see Fig. 14), whereas battery overhead catenary trucks have a high range of costs depending on the necessary expansion of the overhead catenary line system and the electricity infrastructure. An analysis conducted in this work showed that for the maximal cost case of batteryoverhead articulated trucks, every point in NRW is reachable within 50 km distance from highway access points. However, NRW is a highly populous state and this analysis would probably produce another result in a less populated area, especially against the background that an attractive overhead catenary line system should be an international  Table 4).
approach. Furthermore, the economic analysis showed that the minimum and maximum CO 2 reduction costs via fuel cell-electric trucks are fairly close. The CO 2 -specific costs of FCETs consist mostly of costs for the hydrogen production infrastructure. The hydrogen production infrastructure's CAPEX and OPEX account for up to 25% of the total yearly costs (see Fig. 14b), given the assumption that hydrogen infrastructure costs are linearly proportional to fuel consumption and the infrastructure is scalable without changing the CO 2 -specific costs. An expansion from the operation of articulated trucks to operating all trucks as FCETs would lead to an increase in CO 2 reduction by 28%. Gerbert et al. (2018) calculated, based on the results of Wietschel et al. (2017), CO 2 -equivalent costs of about 60 €/t CO2e to 300 €/t CO2e for equipping 4,000 km and 8,000 km of federal highways with overhead lines (see Table 1) and operate diesel-hybrid trucks on them. While the lower value fits with the calculated value for the scenario Diesel-Hybrid Cat Ramp-Up, the higher value for equipping 8,000 km with catenary lines is even higher than the in this work calculated value for all federal highways in NRW (see Fig. 14). However, they calculated well-to-wheel emissions and did not do a spatial analysis, what could have caused the difference. The calculation of tank-to-wheel emissions in this work was chosen to assess both technologies, catenary trucks and fuel-cell electric trucks, with a preferably neutral approach. Well-to-wheel emissions will strongly depend on the energy source (see Fig. 3 and Fig. 4). Using H 2 from natural gas for FCETs and electricity from the German or European electricity mix for catenary trucks would benefit the latter (see Fig. 3), while assuming imported hydrogen from renewable energy sources as proposed in Hydrogen Council (2020) would benefit FCETs.
In addition to the reduction in GHG emissions through the discussed technologies, there is also the option of alternative fuels such as DME, synthetic GTL diesel, HVO, FAME or compressed and liquid natural gas. However, as discussed already above, although these alternative fuels could completely reduce CO 2 emissions through the utilization of CO 2 during production, they reduce harmful emissions only partially or not at all. The reduction strongly depends on the emission concept of vehicle. Drop-In fuels offer especially the advantage of reducing local emissions of already existing vehicles with no or old emission reduction technologies. Hydrogen and alternative liquid or gaseous fuels also offer the possibility to be imported from regions with superior renewable energy resources.
All scenarios and cases in this study demonstrate the maximum emission reduction potential, because the fleet's development in this investigation was neglected. On the one hand, the future rising traffic volume would lead to higher emissions. On the other, improved efficiencies and higher emission standards would lead to lower emissions with conventional fuels and drive trains (Matthias et al., 2020). Higher future emissions standards will most likely have the higher impact on future emissions. Therefore, a modified version of the base scenario, including the fleet development, would probably narrow the gap between results of the base scenario and those of Diesel-Hybrid Cat Ramp-Up and Urban Pollution Reduction scenarios, as well as the economic cases, and lead to more precise values. The influence of international traffic in terms of switching the drive technology was also neglected in the present work (the driven distances from international vehicles were considered). On federal highways, the share of international traffic on the mileages of LDVs, articulated trucks, rigid trucks and trailer trucks was 21%, 41%, 21% and36% in 2014 Bäumer et al. (2017). Switching the drive technology of these vehicles is potentially an uncertain assumption.

Conclusions
In this work, we presented an extensive analysis of the emission reduction potential of overhead catenary line trucks, battery-electric vehicles and fuel cell-electric trucks as technologies for light and heavy duty vehicles based on a detailed bottom-up approach. Emissions were investigated for the present case, for two future scenarios using battery-electric vehicles, diesel-hybrid catenary trucks and fuel cellelectric trucks to reduce CO 2 overall and harmful emissions in urban areas and, finally, for five different economic cases. For the two spatially investigated scenarios, CO 2 emissions were examined on selected federal highways in North Rhine-Westphalia, and the harmful pollutants NO x , PM10 and PM2.5 were researched for urban areas and for the selected emission hotspot Cologne. For the five economic cases, the investment and operational costs were calculated and combined with the model results for CO 2 emissions, as well as harmful emissions in Cologne. The results showed that light and heavy duty vehicle traffic produce up to 50% of present CO 2 emissions on federal highways and 30-40% of harmful air pollution in urban areas. The analysis of urban air pollution showed that diesel-hybrid overhead catenary technology could, but not necessarily would, effectively reduce air pollution, depending on the respective share of mileages of heavy duty vehicles on and off of federal highways. Furthermore, the investigation of urban area air pollution indicated that using fuel cell-electric trucks or battery-electric drivetrains for light duty vehicles, rigid trucks and trailer trucks, as well as diesel-hybrid overhead catenary trucks for articulated trucks in a selected urban area, leads to a reduction of NO x and PM2.5 emissions of 29% and 17%, respectively. Operating diesel-hybrid overhead catenary articulated trucks on 717 km of federal highways in North Rhine-Westphalia would lead to a CO 2 reduction of 3.84% at costs of 58 €/t CO 2 , whereas these articulated trucks on all federal highways in North Rhine-Westphalia would lead to a CO 2 reduction of 8.62% at costs of 94 €/t CO 2 . For the option of operating battery-hybrid overhead catenary articulated trucks on all federal highways in North Rhine-Westphalia and, accordingly, in battery mode on all remaining streets, leads to a CO 2 reduction of 13.55% at costs of 41-115 €/t CO 2 . Finally, operating articulated trucks as fuel cell-electric trucks on all streets would reduce CO 2 emissions for 13.55% at costs of 63-86 €/t CO 2 , whereas operating all trucks as fuel cell-electric trucks on all streets would reduce CO 2 emissions by 28% at the same costs.
The main conclusions of this work are as follows: • Operating diesel-hybrid overhead catenary articulated trucks is an effective technology for reducing GHG emissions on federal highways. Although it is cost-efficient for highways with high traffic volumes, costs will increase to a high level if this technology is used on all federal highways. The diesel engine offers a high flexibility in the case of driving on non-federal highway roads, but leads to air pollution in urban areas, depending on the city. Although articulated trucks have a high affinity for this technology, its application to rigid trucks and trailer trucks may not meet the requirements of the users. • Battery hybrid overhead catenary trucks are an effective technology for reducing GHG emissions and air pollution on all streets. This technology features either with high costs and high flexibility in terms of driving on non-highway roads or with low costs and low flexibility. The technical feasibly of the latter may not be a given, especially in areas with a low population density and resulting long drive distances in the secondary street network off of federal highways. • Fuel cell-electric trucks would solve the challenges of GHG emissions and air pollution in urban areas. These vehicles are more flexible and not dependent on charging sections on the highways. However, apart from low infrastructure costs compared to the overhead catenary line technology, the majority of costs for GHG reduction through fuel cell-electric trucks arise from the investment and operating costs of the hydrogen production infrastructure.
Finally, fuel cell technology is most likely the more flexible and overall cost-efficient technology. Fuel cell-electric trucks and passenger cars use partly the same infrastructure and can be used on all roads. Therefore, a combined approach of battery-electric vehicles, fuel cellelectric trucks and passenger cars seems most attractive. Furthermore, this study emphasize the necessity of replacing the existing diesel freight transport fleet with low GHG emissions and above all zero air pollution technologies.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.