Transportation Research Part D 70 (2019) 35–48

Contents lists available at ScienceDirect

Transportation Research Part D

journal homepage: www.elsevier.com/locate/trd

Fuel cell electric vehicles: An option to decarbonize heavy-duty

T
transport? Results from a Swiss case-study
Emir Çabukoglu, Gil Georges , Lukas Küng, Giacomo Pareschi,
⁎

Konstantinos Boulouchos
Aerothermochemistry and Combustion Systems Laboratory, ETH Zurich, Sonneggstrasse 3, 8092 Zurich, Switzerland
Swiss Competence Center for Energy Research on Efficient Technologies and Systems for Mobility, Zurich, Switzerland

ABSTRACT

CO2 emissions of road freight transport may seem secondary to passenger cars, but electrification could eliminate direct emissions of cars. For heavy-
duty trucks, it is unclear if substituting Diesel is even an option. We developed a data-driven approach to explore this issue: it estimates feasibility
considering the daily operation patterns of every vehicle in the fleet. This paper presents results for fuel cell propulsion systems. If every Swiss truck
drove on hydrogen produced exclusively by electrolysis, full decarbonisation would draw over 8 TWh of renewable electricity (13% of the national
consumption). That corresponds to roughly 60 km2 of photovoltaic panels with 1.5 GW peak power. We found that current fuel-cell technology
almost completely realized that potential, provided vehicles could refuel during the day. The autonomy range was generally better than with battery
electric systems without significant weight increase (relative to the original vehicle). Refuelling could take over half an hour, requiring a dense
energy infrastructure, able to refuel hundreds of vehicles in parallel to avoid congestion (i.e. vehicles waiting). The reduction of direct emissions was
easily overcompensated by indirect emissions of generation: the Swiss consumer mix lead to virtually no overall reduction, while natural gas
powerplants lead to a significant CO2 increase. We concluded that hydrogen is technically a very attractive decarbonisation agent for heavy-duty
vehicles, but significant investments may be required to ensure that (a) hydrogen production is truly renewable and (b) vehicles have adequate
access to additional energy during the day.

1. Introduction

1.1. Why do we consider fuel cell trucks?

Heavy-duty freight transport is an important CO2 emitter and its share in emissions grows worldwide (Calvo Ambel, 2015). The
potential solution for this problem is the electrification of heavy-duty vehicles (Eurelectric, 2017).
In Çabukoglu et al. (2018), we showed that although battery electric trucks can be an option for the future, there is still a need for
development of new technologies, i.e. having better batteries and battery swapping technology. Until that point is reached, it is not
possible to convert most of today’s conventional freight trucks into battery electric trucks.

Abbreviations: BEV, Battery Electric Vehicle; FCEV, Fuel Cell Electric Vehicle; PFCEV, Plug-in Fuel Cell Electric Vehicle; HEV, Hybrid Electric
Vehicle; PHEV, Plug-in Hybrid Electric Vehicle; NGV, Natural Gas Vehicles; ENTSO-E, European Network of Transmission System Operators for
Electricity; GTE, Goods Transport Survey; ICE, Internal Combustion Engine; LNG, Liquified Natural Gas; LSVA, Distance-specific Heavy-duty Vehicle
Tax; MPW, Maximum Permissible Weight; SOC, State of Charge; SMR, Steam Methane Reforming; pkm, Passenger-kilometer; vkm, Vehicle-kilo-
meter; tkm, Tonne-kilometer; Mt, Megaton; WHVC, World Harmonized Vehicle Cycle; PV, Photovoltaics; kph, Kilometers per hour; mph, Miles per
hour
⁎
Corresponding author at: Aerothermochemistry and Combustion Systems Laboratory, ETH Zurich, Sonneggstrasse 3, 8092 Zurich, Switzerland.
E-mail address: [email protected] (G. Georges).

https://doi.org/10.1016/j.trd.2019.03.004

Available online 27 March 2019

1361-9209/ © 2019 Elsevier Ltd. All rights reserved.
E. Çabukoglu, et al. Transportation Research Part D 70 (2019) 35–48

Another option on the way to decarbonization can be the fuel cell trucks since these would also avoid tail-pipe emissions and help
reducing fossil fuel use in freight transport. An important advantage of these vehicles compared to battery electric vehicles is that the
former ones have a longer range than the latter (Thomas, 2009; Morrison et al., 2018) and their fueling times are comparable to their
conventional counterparts (Talebian et al., 2018).
While fuel cell technology is commonly used in forklifts (Mayyas et al., 2016) and buses (Staffell et al., 2018) and the first
examples of fuel cell cars are already in the market (Toyota Mirai, 2018; Honda Clarity, 2018; Hyundai ix35, 2018); fuel cell trucks
are still uncommon. To our knowledge, there are currently only two fuel cell trucks developed.
The first one is the Coop ESORO truck (Factsheet Coop Esoro Truck, 2016) in Switzerland. The second one is Toyota’s hydrogen
fuel cell truck which is currently operating in the Port of Los Angeles (Toyota’s hydrogen fuel cell trucks are now moving goods
around the Port of LA, 2017). Besides, there are trucks which are announced to be available in the coming years. VDL announced that
a new fuel cell truck will be ready this year (VDL ready to go full force, 2018). Scania currently also develops a fuel cell refuse truck
which will be delivered in the end of 2019 (Scania delivers fuel cell refuse truck, 2018). Nikola Motors announced that they start with
the production of Nikola One in 2021 (Nikola One, 2017). Kenworth also develops its own fuel cell truck T680 based on an existing
model (Zero-Emission Kenworth T680 Equipped with Hydrogen Fuel Cell on Display at Consumer Electronics Show, 2018).

1.2. Related work

In the literature, there are various studies focusing on the conversion of conventional vehicles into fuel cell electric vehicles (see
Section 1.2.1) and on the decarbonization potential of fuel cell vehicles (see Section 1.2.2).

1.2.1. Conversion of conventional vehicles into fuel cell electric vehicles

Kast et al. (2017) design fuel cell powertrains for various truck classes, simulate vehicles in different cycles and analyze the
technical feasibility (if vehicles can complete their duties using the new powertrain) of fuel cell trucks. In subsequent studies (Kast
et al., 2017; Gangloff John et al., 2017), they focus on the hydrogen tank and determine which vehicles have enough space for their
hydrogen tanks so that they can reach their destinations. All these studies use mobility data from surveys done with truck and fleet
operators and focus on a part of the truck fleet.
Other studies choose a specific vehicle class and focus on the substitution of the vehicles in this class with their fuel cell electric
counterparts. Lewis et al. (2017) use real world GPS data to size the fuel cell and battery for fuel cell delivery vans accordingly. Lee
and Hirota (2017) design a fuel cell garbage truck and estimate the hydrogen consumption of the new vehicle. Melo et al. (2014)
analyze a current urban bus fleet, design fuel cell powertrains for the vehicles using real world driving cycles and analyze the
technical feasibility of this conversion. Ribau et al. (2014) estimate the optimal fuel cell powertrain design for a specific bus and
compares it with its conventional counterpart from cost and efficiency perspectives. To our knowledge, there is no study investigating
the substitution of a whole national fleet with fuel cell vehicles.

1.2.2. CO2 reduction potential of a fuel cell vehicle fleet

There is a lack of literature about the CO2 reduction potential through a fuel cell truck fleet. To our knowledge, there is only one
study (Talebian et al., 2018) looking into electrification of road freight transport (using BEVs and FCEVs) and determining what share
of the truck fleet should be converted into electric trucks to achieve the CO2 reduction targets of British Columbia.
On the other hand, there are numerous studies estimating the CO2 reduction potential of fuel cell technology for light-duty
vehicles.
Thomas (2009) compares fuel cell vehicles with battery electric ones from the perspective of societal benefits for the next
100 years. He considers various properties including CO2 emissions, energy efficiency, costs and fueling infrastructure. Offer et al.
(2011) compare operational emissions of BEVs, FCEVs and PFCEVs for the UK. Sandy Thomas (2012) estimates the maximum
potential of BEVs, PHEVs and FCEVs to cut greenhouse gas (GHG) emissions in the U.S. In another study, Sandy Thomas (2009)
compares various alternative vehicle technologies including BEVs, HEVs, PHEVs, FCEVs, PFCEVs and NGVs considering their CO2
reduction potentials. Kromer and Heywood (2009) estimates the GHG reduction potential of electric propulsion systems (BEVs, HEVs,
PHEVs and FCEVs) for the light-duty vehicle fleet in the U.S. in 2030. Kim and Moon (2008) assess the CO2 mitigation potential of
fuel cell vehicles for Korea in various scenarios for the future considering different policy options. Nakata (2003) models the energy
consumption of fuel cell vehicles and estimate the CO2 reduction potential of these vehicles in Japan for different scenarios.
To our knowledge, there is not any study analyzing the CO2 reduction potential of a whole truck fleet using fuel cell powertrains.

1.3. Scope and goals

In this study, we explore the maximum penetration depth of fuel cell powertrain technology for the heavy-duty fleet and to which
extent it can help us to decarbonize the freight transport. We do this first by analyzing the fleet and determining which vehicles can
be converted into fuel cell vehicles depending on the available technology and fueling infrastructure. Then we analyze how the
demand for the infrastructure side will look like if we use fuel cell powertrains for heavy-duty vehicles in the future. To our
knowledge, this is the only study doing these analyses for the entire national heavy-duty fleet. An important reason for that is that
Switzerland is one of the three countries around the world, which follows every single vehicle in its heavy-duty fleet throughout the
year. As done in Çabukoglu et al. (2018), we look only into the technological potential. There are many aspects which can define if a
technology will be used or contribute to decarbonization in the future such as costs and market dynamics. This study analyses if fuel

36
E. Çabukoglu, et al. Transportation Research Part D 70 (2019) 35–48

cell trucks can technically fulfill the role conventional trucks have today since this is the prerequisite for any future adoption. This
study focuses only on fuel cell electric trucks. We will explore the alternatives such as hybrid and natural gas trucks in a subsequent
study.
This paper is structured as follows: Section 2 explains how we convert the mobility profiles of vehicles into energy demand
profiles, how we design the fuel cell powertrains for heavy-duty vehicles and how we determine which vehicles can be converted into
fuel cell electric vehicles. Section 3 displays our findings and how the results depend on the assumptions used.

2. Methodology

Our methodology consists of three parts. The first part is the derivation of the daily usage profiles of each heavy-duty vehicle in
the fleet. Afterwards these usage profiles are translated into mechanical energy. The last part is the design of the fuel-cell powertrain,
calculating the final energy demand and checking if the FCEV can complete all its missions throughout the year.
Since the first two steps are exactly the same as in the previous study (Çabukoglu et al., 2018), we summarize those in Sections 2.1
and 2.2. The last step is explained in detail in Section 2.3.

2.1. Generation of usage profiles

The basis of this part is the LSVA database of Switzerland which records the distance each distance drives for every day of the
year. LSVA (Heavy vehicle charges, 2018) is the performance-related heavy vehicle charge of Switzerland which contains vehicles
with a maximum permissible weight (MPW) of more than 3.5 t and depends on the kilometers driven. Because of this distance
dependence, Switzerland tracks each vehicle in its heavy-duty fleet and is thus one of the few countries (others being Belgium and
New Zealand) having complete coverage of the fleet.
LSVA is paid based on the MPWs of the vehicles and thus the actual payloads carried by heavy-duty vehicles are not considered in
the database. This information exists in the goods transport survey of Switzerland (Gütertransporterhebung (GTE), 2018) for a subset
of the fleet. We combine these two databases using the common variables of them, namely maximum permissible weight, curb
weight, body type and whether they have a trailer is attached. We cluster vehicles with similar payload profiles using decision trees
and in these groups we assign payloads from GTE to each record in LSVA. Fig. 1 shows an example of the variaty of valuable
information available in the LSVA database, In this figure, the distributions of daily mobility profiles of vehicles are shown depending
on the vehicle type and number of axles.

2.2. From usage profiles to useful energy demand

Knowing the mobility and load profile of the heavy-duty vehicles, it is possible to calculate the energy demand of these vehicles.
To calculate the energy needed on the wheel, the longitudinal force equation (Eq. (1)) is integrated through a representative heavy-
duty driving cycle, namely World Harmonized Vehicle Cycle (WHVC) (Silberholz et al., 2014).
1 dv (t )
Fprop (v (t )) = · air · cD· Af ·v 2 (t ) + m ·g · cr + m ·
2 dt (1)

where:

Fig. 1. The distribution of the distance Swiss heavy-duty vehicles drive on an average day by type of vehicle and number of axles, as reported by the
Swiss road tax system “LSVA”.

37
E. Çabukoglu, et al. Transportation Research Part D 70 (2019) 35–48

• g is the standard acceleration due to gravity

• c isisthe
the ambient air density
•
air
vehicle’s aerodynamic drag coefficient
•
D
A is the vehicle’s frontal area
• c is the tire rolling resistance coefficient
f

• m is the total mass of the vehicle, consisting of the tractor (curb weight), the trailer (curb weight or 0 if there is no trailer attached)
r

and the payload carried — see Eq. (2)

m = mtowing vehicle + mtrailer + mpayload (2)

g and air are ambient properties which marginally vary with the location of the vehicle.

P (t )
WHVC total
dt 1
etotal = = · ( (Fprop (v (t )))·Fprop (v (t )) + Paux ) dt
WHVC
v (t ) dt d WHVC
(3)
1
where is the Heaviside step function ; Paux is the total auxiliary mechanical power demand of the various non-propulsive sub-
systems, ranging from air conditioning to power steering.

2.3. From useful to end energy demand

First we calculate the demand of diesel for the conventional vehicles. Then, we redesign each vehicle as explained in Section 2.3.1
and check if the new powertrain can allow the vehicle finish all its missions throughout the year.
To calculate the diesel demand for conventional vehicles, we assume a constant efficiency of 40% (see Table 1). Assessing the
hydrogen demand for fuel cell vehicles, we calculated the fuel cell vehicle efficiency for each vehicle by simulating the fuel cell
vehicle through the driving cycle (see Section 2.3.2) and will show in Section 3.5 how an assumption of a constant efficiency would
affect the results.

2.3.1. Fuel cell powertrain design

This section describes how we replace the conventional powertrains of trucks with fuel cell electric powertrains. The steps we
follow are as follows:

1. The Diesel engine and the gear-box are removed according to Table 1.
2. An electric motor with the same nominal power as the Diesel engine is added. Power electronics are also added accordingly.
3. Battery and fuel cell are added according to the constraints (gradeability and acceleration) described in the paragraph below.
4. The Diesel tanks of conventional vehicles are removed and hydrogen tanks are added into the same volume. In Section 3.5.3 we
analyze how the results would change if we could put also hydrogen tanks behind the driver cabin like in Nikola One truck (Nikola
One, 2017). For the volumes of fuel tanks, we use the values in Table 2.

There are two constraints defining the minimum size of the battery and fuel cell in a FCEV. The first one is the gradeability
constraint, which means that a heavy-duty vehicle should be able to drive on a slope with a specific speed using only its fuel cell. The
second constraint is that a heavy-duty vehicle should be able to accelerate from 0 kph to a specific speed in a specific time interval.
This can be done by using both the battery and the fuel cell. So, while the first constraint sets a lower limit for the fuel cell power, the
second one sets a lower limit for the sum of the fuel cell and battery power (Bernard et al., 2009).
The capability constraints used in this study are as follows:

• Gradeability: For heavy-duty vehicle,s the constraint chosen was driving on a slope of 5% with a speed of 40 kph (Jaggi, ESORO),
but it is seen that vehicles under 14 t cannot finish the driving cycle in a charge-sustaining mode ( SOC = 0 ), meaning that their
fuel cells are not big enough. So for these vehicles, a different constraint is chosen (Driving on a slope of 6% with a speed of
80.5 kph (50 mph ) (Duran, 2016)).2
• Acceleration capability: Accelerating from 0 kph to 85 kph in 36 s (Rasantes Sparschwein, 2014)

We chose the intersection of gradeability and acceleration lines for the design since this is the point that gives the minimum fuel
cell size. The reason for this choice is that large fuel cells (above 150 kW ) for mobile applications are still not commonplace and
should be researched/developed further (Kühn et al., 2017). On the other hand, keeping the fuel cell power the same and increasing
the battery power would result in a heavier vehicle without resulting in any efficiency gains. Thus the corner point shown in a circle
in Fig. 2 is chosen as the design point.3

1
The Heaviside step function (x ) = 1 if x > 0; (x ) = 0 otherwise.
2
These vehicles only correspond to 2% of the transport performance (tkm) in Switzerland and thus do not significantly affect the results.
3
Hydrogen compression also requires an additional electricity of 2.7 kWh/kg H2 (Linde, 2014)

38
E. Çabukoglu, et al. Transportation Research Part D 70 (2019) 35–48

Table 1
Technical properties of the components (or assemblies) used in powertrain design and energy demand computations.
Quantity Component Value Source

Weight-to-power ratio [kg/kW] ICE + gear-box 2.17 Plotkin and Singh (2009)
Electric Motor 0.80 Plotkin and Singh (2009)
Power Electronics 0.10 Plotkin and Singh (2009)
Fuel Cell 1.67 Plotkin and Singh (2009)
Battery 1.00 Bauer et al. (2015)

Volumetric mass density [kg/l] Hydrogen tank 0.250 Gangloff John et al. (2017)
Hydrogen (350 bar) 0.022 Gangloff John et al. (2017)
Diesel 0.832 Adebayo et al. (2013)

Conv. powertrain efficiency [–] ICE + gear-box 0.40 Çabukoglu et al. (2018)

Fuel cell vehicle powertrain efficiency [–] Battery 0.95 Drobnik and Jain (2013)
Inverter 0.96 Drobnik and Jain (2013)
Motor + Transmission 0.90 Drobnik and Jain (2013)
Recuperation 0.50

Hydrogen production technology efficiency [–] Electrolysis 0.63 Schildhauer (2017)

Compressor 0.92 Schildhauer (2017)

Inputs for steam methane reforming [MJ/kg H2 ] Electricity 2.00 Yazdanie et al. (2014)
Natural gas 165.00 Yazdanie et al. (2014)

Emission factors [g CO2 /kWh] Diesel 272.00 Röthlisberger (2014)

Swiss consumer mix 138.50 Messmer and Frischknecht (2016)
CCGT 325.00 Willnow (2013)
ENTSO-E 2010 (European mix) 343.80 E-Control (2011)
ENTSO-E 2015 (European mix) 413.60 E-Control (2016)

Table 2
Largest available fuel tank capacity by vehicle category, as currently available in the Swiss market.
Note that the values correspond to the product-portfolio of the market-leader in Switzerland at the
time of writing (Mercedes-Benz LKW Konfigurator - Actros, 2017).
Vehicle type MPW [kg] Fuel tank volume [l]

Rigid 3500–18,000 200

18,000–25,000 1260
25,000–26,000 1060
26,000–40,000 1260

Articulated 3500–18,000 200

18,000–23,300 1300
23,300–25,000 730
25,000–26,000 490
26,000–40,000 820

2.3.2. Fuel cell powertrain efficiency calculation

To estimate the powertrain efficiency, two operation strategies are examined (see Fig. 3). In the first approach, fuel cell always
runs on the average power demand of the vehicle in the driving cycle. In the second approach, fuel cell dynamically follows the power
demand of the vehicle. The fuel cell power never goes below 5% of the rated fuel cell power because of lifetime issues occurring below
this rate – this constraint is also mentioned by Jaggi (ESORO) in a private talk – and the upper limit (up to which limit the vehicle
follows the driving cycle) is determined so that the battery is used in charge-sustaining mode through the driving cycle. The driving
cycle simulations using these approaches are shown in Fig. 3. The component efficiencies used in this calculation are also shown in
Table 1.
We show the effects of calculating the efficiency using one of the two methods or assuming a constant efficiency (45% in our case)
on our results in Section 3.5.1.

2.4. Evaluation

We analyzed the technical potential of fuel cell technology for the heavy-duty fleet. In reality, vehicles do not have unlimited time
to fuel since they have to complete their missions of the day and be also competitive with their conventional counterparts. Because of
that, we set a limit to the number of fuellings a vehicle can perform in a day and show our results depending on this limit. We
demonstrate the potential of fuel cell technology using the following indicators:

39
E. Çabukoglu, et al. Transportation Research Part D 70 (2019) 35–48

Fig. 2. The constraints considered in fuel cell electric powertrain design.

Fig. 3. Approaches to operate the fuel cell electric powertrain.

• the CO 2 emissions avoided by substituting the conventional vehicles with their fuel cell electric counterparts and using various
electricity sources (including CO2 -neutral electricity) or steam methane reforming for hydrogen production (Throughout this
study, when avoided CO2 emissions are mentioned, CO2 -neutral electricity is assumed, unless stated otherwise)
• the number of substitutable vehicles
• the amount of vkm and tkm driven by fuel cell electric vehicles
• the total hydrogen and electricity demand
• the number of fuellings on any given day

2.5. Hydrogen fuelling infrastructure requirements

Currently there are only a few hydrogen fuelling stations in Switzerland (Switzerland gets its first hydrogen filling station, 2016;
The First Public Hydrogen Fueling Station in Switzerland, 2016), but many of them will be needed in the future if the fuel cell
vehicles become commonplace.
To estimate the demand on the infrastructure, we modeled the fuelling stations converting the multi-agent, discrete event si-
mulation used in Çabukoglu et al. (2018) into a hydrogen fuelling station simulation, i.e. changing the fuelling rates. This simulation
goes through each day of the vehicles and simulates the congestion in the fuelling stations. Each vehicle drives at a constant average
speed until it reaches its destination for the day or its battery is empty. Each fuelling station can serve one vehicle at a time and
therefore some vehicles wait for their turn, while another vehicle gets fueled with a pre-defined fuelling rate.
In this study we varied two parameters and investigated how many stations we need in Switzerland so that vehicles never spend
more than 1 h per day to be fuelled. We set this constraint so that these vehicles can compete with their conventional counterparts.
Considering the breaks drivers take during the day, one hour is a reasonable time being available for fuelling.
We focused on the following parameters:

1. Fuelling rate: Here we assumed two different fuelling rates according to SAE J2601/2 (2014). Two rates used are the slow fuelling
option (30 g/s ) and the normal fuelling option (60 g/s ).
2. When vehicles go to fuelling stations. Vehicles can follow three different strategies:
(a) Vehicles get fueled when their tank is empty
(b) Vehicles get fueled when their tank is empty or when they start the day
(c) Vehicles get fueled when their tank is empty or when they end the day

40
E. Çabukoglu, et al. Transportation Research Part D 70 (2019) 35–48

3. Results and discussion

3.1. Maximum CO2 mitigation potential

To understand the technical limits of fuel cell powertrain technology with respect to CO2 mitigation, we analyzed the case of
substituting all vehicles with fuel cell electric vehicles. We did this by removing the limit of number of fuellings a vehicle can get daily.
Substituting all trucks in Switzerland with fuel cell electric ones causes an annual electricity and a hydrogen demand of 8.21 TWh
and 142, 769 t respectively. The daily hydrogen and electricity demands throughout the year are shown in Fig. 5. Here it is assumed
that hydrogen is produced on the day it is needed by the vehicles. If the electrolysers work 24 h per day (perfect distribution of the
daily demand), the peak electricity demand reaches 1.5 GW on the days of high freight transport activity.
Avoided CO2 emissions are shown in Fig. 4 as a function of the electricity mix. It is seen that less than one third of today’s
emissions can be avoided even using the rather clean electricity mix of Switzerland. Using electricity from the ENTSO-E mix or gas-
fired CCGT plants, current emissions caused by heavy-duty vehicles can even be doubled. Thus, to achieve a meaningful CO2 miti-
gation, renewable energy technologies should be used (no operational emissions). Since most of the hydro potential is already used
and wind energy potential is limited in Switzerland (Densing et al., 2014), the most appropriate solution from the potential point of
view is using photovoltaics (PV) to generate the required hydrogen for fuel cell vehicles. This would also eliminate the problem of
storing solar energy when it is not needed. If all the electricity needed to produce hydrogen is generated through PV, an area of
58.6 km2 – around two thirds of the area of Zurich – is needed for PV installations.

3.2. Substitution assuming no fuelling infrastructure on roads

Nowadays, hydrogen tanking stations are not commonplace in Switzerland. So, we analyzed the case assuming that tanking
stations are not available on the roads and each user fuels its vehicles in its home base when vehicles come back in the evening (one
fuelling is allowed per day). This assumption allows around 30% of the vehicles be converted into fuel cell vehicles, while the rest
need fuelling at least once during the day. The vehicles which can be converted into fuel cell vehicles (30% of the fleet) produce only
11% of the heavy-duty emissions since these vehicles are the ones which drive short distances. The distribution of the substitutable
vehicles according to vehicle classes is shown in Table 3. If the electrolysers work 24 h per day, the peak electricity demand for
electrolysers reach 165 MW in the days of high freight transport demand.
The substituted vehicles need around 15, 165 t of hydrogen and 0.87 TWh of electricity to produce this hydrogen. A PV area of
6.2 km2 is needed to generate this electricity. The fluctuations of the demand throughout the year are shown in Fig. 6.

3.3. Considering development of fuelling infrastructure

We then analyzed the potential of FCEVs considering the development of fuelling infrastructure. This analysis showed that when
hydrogen fuelling stations become available on the roads, more vehicles can be converted into FCEVs (see Fig. 7a). Fig. 7b shows that
the heavy vehicles can be converted much easier into FCEVs compared to the vehicles lighter than 18 t . The reason for this is that the
light vehicles have much less space in them for hydrogen tanks (see Table 2) and need to get fueled more often than the heavy ones.
Fig. 7c shows than when electrolysis is used to produce hydrogen, then even with the clean Swiss consumer mix, not much CO2 can
be avoided. When the ENTSO-E mix or a gas-fired CCGT plant is used, the heavy-duty vehicle emissions are more or less doubled. In case
of using steam methane reforming to produce hydrogen, the emissions of FCEVs stay around a similar level as the conventional vehicles.

3.4. Congestion in fuelling stations

Up to this point in this study, fuelling was an instantaneous event happening without any constraints. To estimate the demand in
the fuelling stations, we used our discrete event simulation explained in Section 2.5. We defined different cases depending on the
fuelling rate and the fuelling strategy of the vehicles.

Fig. 4. Avoided CO2 emissions against CO2 intensity of electricity – case: whole heavy-duty fleet is converted to FCEV.

41
E. Çabukoglu, et al. Transportation Research Part D 70 (2019) 35–48

Fig. 5. Daily demand on fuel and electricity assuming that all vehicles are converted to fuel cell vehicles.

Table 3
Absolute numbers of the Swiss fleet by vehicle class and relative share of substitutable vehicles in each class assuming
no fuelling infrastructure on roads
Vehicle class Vehicles

in the fleet electrifiable through FCEVs

Rigid Trucks 42,434 35.0%

Articulated Trucks 10,385 7.2%

3500 kg⩽MPW < 12,000 kg 5582 36.3%

12,000 kg⩽MPW < 18,000 kg 6060 40.2%
18,000 kg⩽MPW < 26,000 kg 20,483 27.6%
26,000 kg⩽MPW < 32,000 kg 9748 24.8%
MPW ⩾32,000 kg 10,946 28.0%

All Trucks 52,819 29.5%

Fig. 6. Daily demand on fuel and electricity assuming that vehicles can only get fueled in the night (around 30% of the fleet is substituted).

As a reference, we allowed vehicles to get fuelled a maximum of three times per day. We chose this case so that we can analyse the
infrastructure demand when most of the vehicles are converted (around 90% of the vehicles are converted with three fuellings per day
– see Fig. 7a), but still vehicles can stay competitive with their conventional counterparts (more than three fuellings result in vehicles
spending hours for getting fuelled). The results of the analysis are shown in Fig. 8. The maximum time spent at a station by a vehicle –
sum of times spent for fuelling and waiting in the queue per day – is plotted against number of fuel dispensers available. As expected,
it is seen that when a fuelling rate of 60 g/s is chosen, much less fuel dispensers are needed compared to slower fuelling. The more
interesting finding is that if vehicles are forced to get fuelled in the evening before they go to their homebase, then the fuelling of the
fleet is distributed so well that this case results in a much smaller infrastructure demand than other strategies. When comparing the

42
E. Çabukoglu, et al. Transportation Research Part D 70 (2019) 35–48

Fig. 7. Allowing daily fueling.

other two strategies, it is seen that forcing the vehicle to get fuelled in the morning results in a more balanced fuelling demand –
meaning that less dispensers are needed – than the case in which vehicles get fuelled only when the tank is empty.
The horizontal line in Fig. 8 shows the 1 h limit (a vehicle spends at most 1 h in stations per day) and it is investigated how many
dispensers are needed to reach this limit. While this limit is reached very easily for the morning and evening fuelling cases with 60 g/s ,
slow fuelling in the evening and fast fuelling whenever the tank is empty result in a demand of around 450 dispensers. On the other
hand, in the two other cases (whenever and morning refuelling at 30 g/s ) we need more than 1.00 dispensers (currently there are 3424
tanking stations in Switzerland (Erdöl-Vereinigung, 2017)). These cases underline the impacts of the planning of refuellings and the
fuelling technology we have on the infrastructure demand we will encounter in the future.
Here we would also like to show the impacts of more extreme cases for fuelling rates and storage pressure. To show this, the
strategy of fuelling whenever tank is empty is chosen. Fig. 9 shows for this case how the demand for fuel dispensers change when
refuelling rate increases to 300 g/s (a theoretical value which is five times as fast as the current fast charging rate) or the storage
pressure reaches 700 bar . This storage pressure is currently available for some vehicles in the market (Gangloff John et al., 2017).
It is seen that while doubling the storage pressure slightly reduces the congestion in fuelling stations, the increase in the fuelling
rate has a much greater impact on the congestion. Using a fuelling rate of 300 g/s almost completely ends the congestion in the
stations and almost no vehicles wait for other vehicles being fuelled (Figs. 10 and 11).

3.5. Sensitivity analyses

To demonstrate that the uncertainties on the technical parameters do not affect our findings significantly, we analyze possible
values for various variables and show the impacts. Since our main indicator is the avoided CO2 emissions throughout this study, we
use this value to show the impacts.
Doing the sensitivity analyses, we investigated the impacts for all possible numbers of daily fuelling allowed and we saw that the
impacts of most of the properties are very small. Therefore we only show the extreme impacts that we encountered.

43
E. Çabukoglu, et al. Transportation Research Part D 70 (2019) 35–48

Fig. 8. Number of fuel dispensers needed for different fuelling strategies.

In the sensitivity analyses, we see that changing the weight-to-power ratios results in a change of less than 1% CO2 avoidance,
while changing the vehicle properties (such as improving aerodynamic or tyre properties) can have an impact of up to 5% . There are
three properties which we look more in detail in the following sections, namely the fuel cell powertrain operation strategy, the
recuperation efficiency and the fuel tank size.

3.5.1. Fuel cell powertrain operation strategies

To calculate powertrain efficiency of the fuel cell vehicles, two strategies are used as explained in Section 2.3.2. It is seen that the
first approach – where the fuel cell is always operated at the average power demand of the vehicle throughout the driving cycle –
results in a higher efficiency than the second one. The reason for this is that the fuel cell can be always operated close to the optimum
efficiency point in the first approach, while the fuel cell power fluctuates between 5% and almost 85% of the rated fuel cell power in
the second approach. This results in ”undesirable” operation points from the efficiency point of view.
To see the effect of the efficiency assumption on the results, we analyzed three cases:

1. First approach: Fuel cell operates at the average power demand of the vehicle throughout the driving cycle and battery does the
rest of the work.
2. Second approach: Fuel cell follows dynamically the power demand of the vehicle throughout the driving cycle. The lower limit is
5% of the rated power (Jaggi, ESRO) and the higher limit – up to which power rate it follows the power demand – is decided so that
the batteries are operated in charge-sustaining mode in the driving cycle.
3. Constant efficiency for all vehicles (45%): As a third option, an average efficiency is chosen as 45% since some powertrain effi-
ciencies lay above this limit while some others are below according to two strategies discussed above.

Doing this, it is seen that choosing the first or second approach does have an impact of less than 5% percentage points on the
avoided CO2 emissions. Assuming a constant efficiency (45%) gives very similar results to the first approach (difference less than
2.7% ).

3.5.2. Recuperation efficiency

Recuperation efficiency – what share of the kinetic energy on the wheels can be recovered while braking (the rest is lost as heat) –
is a parameter which affects the results meaningfully. Our base assumption is the 50% and the efficiency is varied between 30% and
70% . The effect of this on the avoided CO2 avoidance was up to 6% percentage points.

Fig. 9. Number of fuel dispensers needed for different fuelling rates and storage pressures.

44
E. Çabukoglu, et al. Transportation Research Part D 70 (2019) 35–48

Fig. 10. Effect of operation strategies on the fuel cell electric powertrain efficiency.

Fig. 11. Effect of the assumed recuperation efficiency of FCEVs on the avoided CO2 emissions.

3.5.3. Fuel tank size

Generally two possible places are presented for the hydrogen tanks in fuel cell trucks (Gangloff John et al., 2017), namely the side
rail and the back of the cab. Our basic assumption is that we remove the diesel tanks of conventional vehicles (on the side rail) and
add the hydrogen tanks into the same volume. But to see the impact of back of cab design, we also analyze the case with the hydrogen
tanks being both in side rail and back of cab. In this case, since a vehicle can store more hydrogen, it needs to visit the hydrogen
fuelling stations less often. This makes the conversion into fuel cell vehicles much easier – meaning that more vehicles are sub-
stitutable – as it can be seen in Fig. 12.
Using current diesel tank volumes for hydrogen tanks only allows around 10% of CO2 avoidance for the case of no fuelling
infrastructure on the roads. This value is almost five times higher when twice as much space is used. Besides, if there is fuelling
infrastructure available, for a CO2 avoidance of 90% vehicles need up to four fuellings per day using their current tank volumes, while
they could finish all their missions only fuelling once per day doubling the volume of their hydrogen tanks.

4. Conclusions and outlook

We determined which share of the heavy-duty fleet in Switzerland can be converted to fuel cell vehicles depending on the
development of fuelling infrastructure. Fueling only in the home-base of freight vehicle operators – meaning no fuelling opportunity
during the day – allows around 30% of the vehicles being substituted with FCEVs. Having hydrogen fuelling stations around
Switzerland and allowing vehicles to fuel multiple times per day brings the conversion rate close to 100% . This, on the other hand,
results in a demand on the infrastructure side. If all heavy-duty vehicles are converted into FCEVs, more than 140, 000 t of hydrogen is
needed annually. Assuming that electrolysers are used to produce this hydrogen, annual electricity demand increases by more than
8 TWh in Switzerland (an increase of more than 13% of the national demand). If the electrolysers operate 24 h per day (perfect
distribution of the daily demand), the peak electricity demand reaches 1.5 GW in the days of high freight transport activity. To fuel
the vehicles, we will need hydrogen fuelling stations and the demand for these stations will heavily depend on the strategy of vehicles
to go to stations, i.e. whether they go to stations whenever the tank is empty or if they also visit stations in the beginning or at the end
of the day, and the fueling rate available. As an example; if trucks are get fuelled in the evening, before they go to their home-base,
with a fueling rate of 30 g/s, 450 dispensers are needed so that not a single truck spends more than 1 h per day in fuelling stations. On
the other hand, if the trucks only go to fuelling stations when their tank gets empty, we need more than 1.00 dispensers (currently
there are 3424 tanking stations in Switzerland (Erdöl-Vereinigung, 2017)).
If the required infrastructure is ready and the fleet is converted into FCEVs, whether any CO2 avoidance is achieved depends on

45
E. Çabukoglu, et al. Transportation Research Part D 70 (2019) 35–48

Fig. 12. Effect of fuel tank size on the avoided CO2 emissions.

the way hydrogen is produced. When steam methane reforming is used for hydrogen production, the CO2 emissions slightly increase
compared to the conventional fleet. On the other hand, in case of using electrolysers, the ENTSO-E mix or using gas-fired CCGT plants
doubles the emissions of the current fleet. The only solution for meaningful CO2 mitigation is using renewable energy technologies to
power electrolysers. In case of Switzerland, the PV is proposed as the solution since wind power potential is very limited and most
hydro potential is already used.
To conclude, substituting conventional trucks with their fuel cell electric counterparts will require new infrastructure for hy-
drogen production/fueling and increase the electricity consumption if electrolysis is used to produce hydrogen. Besides, CO2 miti-
gation is more difficult to achieve with the fuel cell electric vehicles than their battery electric counterparts because of the lower
efficiency of the powertrain and the extra step of fuel production, i.e. electrolysis or steam methane reforming. But assuming that the
new infrastructure is built and a renewable energy source is used to power the electrolysers, fuel cell trucks have a higher technical
potential than battery electric trucks because of their longer range.
In our previous study (Çabukoglu et al., 2018), we analyzed the potential of battery electric propulsion for heavy-duty vehicles in
Switzerland. This study does a similar analysis for fuel cell electric vehicles. In the previous study, it was shown that battery electric
powertrains can supply the required energy to finish daily missions for most of the vehicles only if the technologies such as battery
energy densities or battery swapping technology improve substantially, which is not expected in the near future. On the other hand,
in the fuel cell electric study we show that even with conservative assumptions, vehicles powered using fuel cell electric powertrains
can mostly finish their missions throughout the year.
There are also other options which can be interesting in the future to decarbonize freight transport sector. In a subsequent study,
we will focus on these technologies, namely hybrid powertrains and natural gas engines. This analysis will give a more complete
picture of possible technologies for heavy-duty vehicles in the future.

Acknowledgments

The authors gratefully acknowledge the financial support of the Swiss Federal Office of Energy (BFE) [SI/501311-01] and the
Swiss Competence Center for Energy Research in Efficient Technologies and Systems for Mobility (SCCER mobility), funded by the
Swiss Commission for Technology and Innovation (CTI). We thank the Federal Statistical Office (BFS), the Federal Roads Office
(ASTRA) and the Federal Customs Administration (EZV) for providing the different datasets. We thank Michael Hugentobler for
operationalizing the GTE dataset. We thank Jonas Bütikofer for the inItial cataloguing of the vehicle fleet. We thank Urs Cabalzar and
Patrick Stadelmann for sharing their knowledge of hydrogen infrastructure. We also thank Diego Jaggi and Stefan Camenzind of
ESORO for providing helpful insights about fuel cell powertrain design for heavy-duty vehicles.

Appendix A. Supplementary material

Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.trd.2019.03.004.

References

Adebayo, G., Ameen, O., Abass, L., 2013. Physico-chemical properties of biodiesel produced from Jatropha curcas oil and palm oil. J. Oil Palm Res. 25 (2), 159–164.
Bauer, C., Hofer, J., Althaus, H.-J., Duce, A.D., Simons, A., 2015. The environmental performance of current and future passenger vehicles: life cycle assessment based
on a novel scenario analysis framework. Appl. Energy 157, 871–883. <https://ac.els-cdn.com/S0306261915000252/1-s2.0-S0306261915000252-main.pdf?_
tid=18467c68-a754-11e7-8f3c-00000aacb35e&acdnat=1506936708_83ee08f141a3d63e40ac22334e878fbb>.
Bernard, J., Delprat, S., Büchi, F.N., Guerra, T.M., 2009. Fuel-cell hybrid powertrain: toward minimization of hydrogen consumption. IEEE Trans. Veh. Technol. 58 (7),
3168–3176.

46
E. Çabukoglu, et al. Transportation Research Part D 70 (2019) 35–48

Çabukoglu, E., Georges, G., Küng, L., Pareschi, G., Boulouchos, K., 2018. Battery electric propulsion: an option for heavy-duty vehicles? Results from a Swiss case-
study. Transp. Res. Part C: Emerg. Technol. 88, 107–123.
Calvo Ambel, C., 2015. Too big to ignore – truck CO2 emissions in 2030, Technical Report September, Transport & Environment.
Densing, M., Hirschberg, S., Turton, H., 2014. Review of swiss electricity scenarios 2050. PSI Bericht(14).
Drobnik, J., Jain, P., 2013. Electric and hybrid vehicle power electronics efficiency, testing and reliability. Evs 27, 1–12.
Duran, A., 2016. ‘EVS29 International Battery, Hybrid and Fuel Cell Electric Vehicle Symposium Driving an Industry: Medium and.. Driving an Industry: Medium and
Heavy Duty Fuel Cell Electric Truck Component Sizing’, (June 2016).
E-Control, 2011. Gesamtaufbringung nach ENTSO-E 2010, Technical report. <https://www.e-control.at/documents/20903/-/-/8c1acae6-16b1-40e8-bf86-
96a3082e4e9f>.
E-Control, 2016. Gesamtaufbringung nach ENTSO-E 2015, Technical report. <https://www.e-control.at/documents/20903/26393/ENTSOE_2015_Gesamtjahr_2016-
04-04.pdf/e7e7eb5e-d947-45ce-9314-b8da5d13def8>.
Erdöl-Vereinigung, 2017. Jahresbericht 2016, Technical report. <https://www.erdoel.ch/de/treibstoffe-und-mobilitaet/publikationen/jahresbericht/jahresbericht-
2016>.
Eurelectric, 2017. Electrification of Heavy Duty Vehicles, Technical Report September. <https://www.eurelectric.org/media/2161/electrification_of_heavy_duty_
vehicles-2017-030-0588-01-e.pdf>.
Factsheet Coop Esoro Truck, 2016. <http://www.esoro.ch/deutsch/content/aktuelles/images/Factsheet_Lastwagen_E.pdf>.
Gangloff John J., J., Kast, J., Morrison, G., Marcinkoski, J., 2017. Design space assessment of hydrogen storage onboard medium and heavy duty fuel cell electric
trucks. J. Electrochem. Energy Convers. Storage 14 (2), 21004–21008. https://doi.org/10.1115/1.4036508.
Gütertransporterhebung (GTE), 2018. <https://www.bfs.admin.ch/bfs/de/home/statistiken/mobilitaet-verkehr/erhebungen/gte.html>.
Heavy vehicle charges, 2018. <https://www.ezv.admin.ch/ezv/en/home/information-companies/transport-travel-documents-road-taxes/heavy-vehicle-charges-
performance-related-and-lump-sum-.html>.
Honda Clarity, 2018. <https://automobiles.honda.com/clarity-fuel-cell>.
Hyundai ix35, 2018. URL: <http://www.hyundai.com/eu/en/Showroom/Eco/ix35-Fuel-Cell/PIP/index.html>.
Jaggi (ESORO), F., 2018. Personal Communication.
Kast, J., Morrison, G., Gangloff, J.J., Vijayagopal, R., Marcinkoski, J., 2017. Designing hydrogen fuel cell electric trucks in a diverse medium and heavy duty market.
Res. Transp. Econ. 1–10. https://doi.org/10.1016/j.retrec.2017.07.006.
Kast, J., Vijayagopal, R., Gangloff, J.J., Marcinkoski, J., 2017. Clean commercial transportation: medium and heavy duty fuel cell electric trucks. Int. J. Hydrogen
Energy 42 (7), 4508–4517. https://doi.org/10.1016/j.ijhydene.2016.12.129.
Kim, J., Moon, I., 2008. The role of hydrogen in the road transportation sector for a sustainable energy system: a case study of Korea. Int. J. Hydrogen Energy 33 (24),
7326–7337. https://doi.org/10.1016/j.ijhydene.2008.09.050.
Kromer, M.A., Heywood, J.B., 2009. A comparative assessment of electric propulsion systems in the 2030 us light-duty vehicle fleet. SAE Int. J. Engines 1 (1), 372–391.
Kühn, A., Thielmann, A., Sauer, A., Plötz, P., Moll, C., Stütz, S., Schellert, M., Rüdiger, D., 2017. Wissenschaftliche Beratung des BMVI zur Teilstudie ”Brennstoffzellen-
Lkw: kritische Forschungsbedarf und Marktpotential”. (August).
Lee, H., Hirota, T., 2017. A Study on Hybrid Power Plant System of Fuel Cell and Li-Ion Battery for Garbage Truck, pp. 0–5.
Lewis, M., Hearn, C., Feng, X., Hanlin, J., Levin, J., Ambrosio, J., Guggenheim, P., Walker, C., 2017. Design and modeling for hydrogen fuel cell conversion of parcel
delivery trucks. In: 2017 IEEE Transportation and Electrification Conference and Expo, ITEC 2017 (June), pp. 674–678.
Linde, 2014. Hydrogen technologies. The Ionic Compressor 90 MPa - IC90.
Mayyas, A., Wei, M., Chan, S.H., Lipman, T., 2016. Fuel cell forklift deployment in the USA. Fuel Cells: Data Facts Figures 334–342. <http://doi.wiley.com/10.1002/
9783527693924.ch33>.
Melo, P., Ribau, J., Silva, C., 2014. Urban bus fleet conversion to hybrid fuel cell optimal powertrains. Procedia - Soc. Behav. Sci. 111, 692–701. <http://linkinghub.
elsevier.com/retrieve/pii/S1877042814001049>.
Mercedes-Benz LKW Konfigurator - Actros, 2017. <https://toc.mercedes-benz.com/LKW_Konfigurator_TOC_de1/toc.dll?>.
Messmer, A., Frischknecht, R., 2016. Umweltbilanz Strommix Schweiz 2014. treeze Ltd, fair life cycle thinking.
Morrison, G., Stevens, J., Joseck, F., 2018. Relative economic competitiveness of light-duty battery electric and fuel cell electric vehicles. Transp. Res. Part C: Emerg.
Technol. 87 (June 2017), 183–196. <http://linkinghub.elsevier.com/retrieve/pii/S0968090X18300056>.
Nakata, T., 2003. Energy modeling on cleaner vehicles for reducing CO2emissions in Japan. J. Cleaner Prod. 11 (4), 389–396.
Nikola One, 2017. <https://nikolamotor.com/one>.
Offer, G.J., Contestabile, M., Howey, D.A., Clague, R., Brandon, N.P., 2011. Techno-economic and behavioural analysis of battery electric, hydrogen fuel cell and
hybrid vehicles in a future sustainable road transport system in the UK. Energy Policy 39 (4), 1939–1950. https://doi.org/10.1016/j.enpol.2011.01.006.
Plotkin, S., Singh, M., 2009. Multi-Path Transportation Futures Study: Vehicle Characterization and Scenario Analyses. Argonne National Lab.
Rasantes Sparschwein, 2014. Trucker (October), 11. <https://www.bus.man.eu/man/media/content_medien/etc/d38_special_folder/d38_doc/201410_MAN_
TGXD38_Trucker_Supertest.pdf>.
Ribau, J., Viegas, R., Angelino, A., Moutinho, A., Silva, C., 2014. A new offline optimization approach for designing a fuel cell hybrid bus. Transp. Res. Part C: Emerg.
Technol. 42, 14–27.
Röthlisberger, R., 2014. Faktenblatt CO2 -Emissionsfaktoren des Treibhausgasinventars der Schweiz. BAFU - Bundesamt für Umwelt - Abteilung Klima.
SAE J2601/2, 2014. <https://saemobilus.sae.org/content/j2601/2_201409>.
Sandy Thomas, C.E., 2009. Transportation options in a carbon-constrained world: hybrids, plug-in hybrids, biofuels, fuel cell electric vehicles, and battery electric
vehicles. Int. J. Hydrogen Energy 34 (23), 9279–9296. https://doi.org/10.1016/j.ijhydene.2009.09.058.
Sandy Thomas, C.E., 2012. How green are electric vehicles? Int. J. Hydrogen Energy 37 (7), 6053–6062. https://doi.org/10.1016/j.ijhydene.2011.12.118.
Scania delivers fuel cell refuse truck, 2018. <https://www.scania.com/group/en/scania-delivers-fuel-cell-refuse-truck/>.
Schildhauer, T., Witte, J., Vogel, F., Cabalzar, U., Brügger, M., Bach, C., Schürch, C.2017. Future mobility, Technical Report 1, Bundesamt für Energie (BFE). <https://
www.aramis.admin.ch/Default.aspx?DocumentID=3905&Load=true>.
Silberholz, G., Six, C., Fredriksson, J., 2014. GRPE-HDH Research Project. <https://wiki.unece.org/download/attachments/18546752/HDH-17-07e.pdf?api=
v2%0A>.
Staffell, I., Scamman, D., Abad, A.V., Balcombe, P., Dodds, P.E., Ekins, P., Shah, N., Ward, K.R., 2018. The role and status of hydrogen and fuel cells across the global
energy system.
Switzerland gets its first hydrogen filling station, 2016. <https://www.azom.com/article.aspx?ArticleID=13317>.
Talebian, H., Herrera, O.E., Tran, M., Mérida, W., 2018. Electrification of road freight transport: Policy implications in British Columbia. Energy Policy 115 (September
2017), 109–118. <http://linkinghub.elsevier.com/retrieve/pii/S0301421518300041>.
The First Public Hydrogen Fueling Station in Switzerland, 2016. <https://www.azom.com/article.aspx?ArticleID=13317>.
Thomas, C.E., 2009. Fuel cell and battery electric vehicles compared. Int. J. Hydrogen Energy 34 (15), 6005–6020. https://doi.org/10.1016/j.ijhydene.2009.06.003.
Toyota Mirai, 2018. <https://ssl.toyota.com/mirai/fcv.html>.
Toyota’s hydrogen fuel cell trucks are now moving goods around the Port of LA, 2017. <https://www.theverge.com/2017/10/12/16461412/toyota-hydrogen-fuel-
cell-truck-port-la>.

47
E. Çabukoglu, et al. Transportation Research Part D 70 (2019) 35–48

VDL ready to go full force, 2018. <http://www.nweurope.eu/projects/project-search/h2share-hydrogen-solutions-for-heavy-duty-transport/news/vdl-ready-to-go-

full-force/>.
Willnow, K., 2013. Energy efficiency technologies ANNEX III technical report for thermal power plants. WEC Knowl. Network (August), 1–30.
Yazdanie, M., Noembrini, F., Dossetto, L., Boulouchos, K., 2014. A comparative analysis of well-to-wheel primary energy demand and greenhouse gas emissions for the
operation of alternative and conventional vehicles in Switzerland, considering various energy carrier production pathways. J. Power Sources 249, 333–348.
https://doi.org/10.1016/j.jpowsour.2013.10.043.
Zero-Emission Kenworth T680 Equipped with Hydrogen Fuel Cell on Display at Consumer Electronics Show, 2018. <https://www.kenworth.com/news/news-
releases/2018/january/t680-zect/>.

48