ArticlePDF Available

Performance Metrics Required of Next-Generation Batteries to Make a Practical Electric Semi Truck

Authors:
  • And Battery Aero

Abstract and Figures

There is enormous interest around the development of an electric semi truck, and in order to understand the practical limitations imposed by the battery pack, a rigorous and thorough analysis considering battery metrics as well as vehicle design parameters is required. In this study, we find that an electric semi truck, such as the envisioned Tesla Semi, would be limited to a driving range well under 600 miles, a small payload capacity, and a prohibitively high cost. Enabling a longer driving range of 600 miles needed by the current standards of the trucking industry might require a transition to beyond Li-ion batteries.
Content may be subject to copyright.
Performance Metrics Required of Next-
Generation Batteries to Make a Practical
Electric Semi Truck
There is enormous interest around the development of an
electric semi truck, and in order to understand the
practical limitations imposed by the battery pack, a
rigorous and thorough analysis considering battery metrics as
well as vehicle design parameters is required. In this study, we
nd that an electric semi truck, such as the envisioned Tesla
Semi, would be limited to a driving range well under 600 miles, a
small payload capacity, and a prohibitively high cost. Enabling a
longer driving range of 600 miles needed by the current
standards of the trucking industry might require a transition to
beyond Li-ion batteries.
In a recent TED talk,
1
Elon Musk discussed his vision for the
transition to sustainable transportation, a sector which currently
accounts for nearly one-third of the total greenhouse gas
emissions.
2
A major discussion point following this talk has been
the potential for Tesla Semi to revolutionize the trucking
industry,
35
which is the source of about one-fourth of the
emissions of the transportation sector while accounting for less
than one-tenth of the total on-road vehicles.
6,7
This has led to
enormous speculation, both among media and nancial markets,
on the cost and performance of the vehicle largely based on back-
of-the-envelope estimates. Given the huge variability and
limitations of such estimates, we feel compelled to carry out a
rigorous study that accounts for the large uncertainties in crucial
factors, such as the specic energy of the battery pack, vehicle
weight, drag, and rolling resistance, and how the factors inuence
each other. On the basis of this analysis, we suggest a potential
trajectory forward and targets for the required cost and specic
energy of the battery pack to enable mass penetration in this
sector; the implications of this analysis are summarized in Figure
1.
Vehicles like semi trucks fall under the Class 8category of
the EPA standards, and the statistics and estimates for such
vehicles can be found in a study by the National Academy.
8
The
main value proposition of the trucking industry hinges on
transporting a payload over a specied distance at a certain cost.
On the basis of this aspect, we rst address the limits of the range
and payload capacity of heavy-duty trucks with current Li-ion
battery technology with estimates for the cost of the pack. The
uncertainty in each of these parameters is quantied based on the
extensive data previously reported.
8,9
We then proceed to
examine how each of these limitations would be alleviated with
future and beyond Li-ion batteries.
The average annual distance traveled for a Class 8 vehicle is
about 75 000 miles,
8
which translates to about 300600 miles
traveled per day assuming a daily driving time of 816 h
throughout the year. The average payload carried by such
vehicles for commodities from dierent industries is about
14 500 kg (16 US ton)
8
and can be as high as 20 000 kg (22 tons).
The fully electric Tesla Semi would have to meet these
performance requirements at a reasonable cost of operation in
order to be a practical alternative to existing vehicles.
We begin our analysis with the estimation of the required
battery pack energy based on the standard dynamic vehicle
model, represented by the system of eqs 12
ρ
η
ηη
= · · · + · ·· + · ···
· −·
⎜⎟
⎜⎟
E C Av C W gv t W gvZ n
Wva D
v
1
2/
1
2
1
rms f
t
Pd
3
rr T T bw
bw
bw brk (1)
where the pack energy, (EP) depends on the energy utilized to
overcome aerodynamic drag forces, frictional forces, the road
gradient, and inertial forces. A signicant fraction of the energy
used to overcome inertial forces is recovered via regenerative
braking when the vehicle is decelerating. The important
parameters are the coecient of drag (Cd), average velocity
(v) and root-mean-square of the velocity (vrms), the coecient of
rolling resistance (Crr), the gross on-road vehicle weight (GVW)
represented by (WT) (includes the payload and battery pack),
the road gradient Z, and the total time taken for a xed driving
range determined from
D
v
. The road gradient term is accounted
Received: May 20, 2017
Accepted: May 31, 2017
Published: June 27, 2017
Figure 1. Summary of a comparison between current and beyond Li-
ion batteries for electrifying semi trucks.
Viewpoint
http://pubs.acs.org/journal/aelccp
© 2017 American Chemical Society 1669 DOI: 10.1021/acsenergylett.7b00432
ACS Energy Lett. 2017, 2, 16691673
for using the expression (Z= r/100), where r is the percentage
road grade and (tf) is the fraction of time the vehicle spends at a
road grade of r%. We assume a road grade r of 1% and (tf) of 15%.
The other xed parameters are ρ, the density of air (1.2 kg/m3), g
the acceleration due to gravity (9.8 m/s2), Athe frontal area of
the vehicle (7.2 m2), athe mean acceleration or deceleration of
the vehicle (0.112 m/s2). ηbw is the battery-to-wheels eciency of
~85% which includes the battery discharge eciency of 95% and
a drivetrain eciency of 90%. ηbrk accounts for the eciency of
the brakes and is assumed to be 97%.
Class 8 vehicles currently have an average coecient of drag
(Cd) of 0.63,
8
with a projected value of 0.45
8
for future vehicles.
The velocity prole data is obtained from the NREL DriveCAT
database for heavy-duty trucks, and the mean and the root-mean-
square values of driving velocity are based on a set of CARB
HHDDT drive cycles namely the Cruise and Composite
segments. The mean rolling resistance (Crr) of truck tires is
estimated to be 0.0063
8
with projections of as low as 0.0045
8
for
future vehicles. The current average GVW is 27 000 kg (30
tons),
8
but for our analysis we assume a xed GVW of 36 000 kg
(40 tons), which is the legal limit for Class 8 vehicles stipulated
by the FHWA.
10
The pack weight, WPis given by, =W
E
Sf
P
P
Pburden
, determined on
the basis of a distribution of values for the specic energy SPat the
cell-level and a xed value of 0.48 for the packing burden factor.
11
fburden represents the weight for the thermal management
systems, module hardware, battery jackets, and other noncell
inactive materials used to assemble a practical battery pack. The
specic energy (SP) is considered with a mean value of 243 Wh/
kg, which is equivalent to that of a Panasonic NCR18650B cell,
and a maximum value of 300 Wh/kg for future Li-ion systems.
Another distribution of specic energies is considered for the
evaluation of beyond Li-ion systems, such as Li-metal coupled
with an advanced cathode, LiS, and Liair, with a mean specic
energy of 500Wh/kg at the cell level and a maximum value of 700
Wh/kg. It is worth pointing out that these are highly optimistic
estimates, and current designs of these batteries provide specic
energy far below the numbers used.
12
The maximum payload
capacity for the vehicle (WL) is given by
=− +WW WW()
LT PV (2)
which is the weight that remains of the GVW after accounting for
the weight of the pack (WP) and the empty vehicle weight (WV)
with their respective distributions. The empty vehicle weight
without the battery pack (WV) is considered to be in the range of
60008000 kg,
8
based on existing estimates for diesel-powered
vehicles without the weight of the engine.
Finally, the cost of the pack (CostP), given by =Cost E
PCost
P
kWh ,
is determined using the pack energy distribution obtained from
eq 1 and a distribution of values for the cost per kilowatt-hour of
the battery pack (CostkWh). The cost of the batteries based on
prior work
9
is assumed to have a mean value of $190/kWh for
Tesla battery packs and a maximum value of $350/kWh with a
projected future value of $150/kWh. For beyond Li-ion batteries,
we assume a mean cost of $120/kWh with a minimum value of
$80/kWh.
Each of these parameters are cast as truncated Gaussian
distributions (truncated within the limits of future projections
and known maximum or minimum values), and a summary of
this is listed in Table 1. Using the dynamic vehicle model and the
pack cost determination equations, we perform a standard
Monte Carlo simulation of all the possible outcomes based on
the distributions of each variable. The simulations are performed
for a driving range of 300, 600, and 900 miles. The resultant
required pack energy can be seen in Figure 2(a), and for each
driving range value, these uncertainties are propagated
throughout the system of equations. In Figure 2(bd), we can
compare the parameters of pack weight, cost per pack, and the
payload capacity between current Li-ion and beyond Li-ion (B-
Li) batteries.
The resulting distribution for the energy consumption per unit
distance, shown in Figure 2(a), is in the range of 2.2 and 2.9
kWh/mile and has a mean value of 2.6 kWh/mile or 65 Wh/ton-
mile. This is an extremely ecient system in comparison to
existing vehicles which have a mean estimated energy
consumption of 250 Wh/ton-mile or 155 ton-miles per gallon
of diesel.
8
The mean values of the required pack size or pack
energy (kWh) scale linearly with the distance and a massive 3100
kWh pack is needed for a range of 900 miles, nearly 30 times
the size of the battery packs in the Tesla Model S car. The values
of the required energy consumption per mile and required pack
size do not change with beyond Li-ion systems because they
depend mainly on the gross vehicle weight and not the weight of
the pack. Lower driving ranges of 300 and 600 miles require
battery packs of 1000 and 2000 kWh respectively.
The weight of the pack required for each of the driving range
values considered is shown in Figure 2(b). For 600 and 900
miles, we observe a mean pack weight of about 16 000 kg and
24 500 kg (18 and 27 tons), respectively. We observe an
approximate one-third reduction in the pack weight after a
transition from the most energy dense Li-ion battery with over
260 Wh/kg to a potental beyond Li-ion system with a specic
energy of 350 Wh/kg. In terms of mean values, a range of 600
miles would require a battery pack that would weigh twice as
much as the weight of the empty vehicle, but with batteries based
on Li metal, LiS, or Liair, it would weigh only as much as the
empty vehicle weight. The reduction in pack weight with beyond
Li-ion systems would have a signicant inuence on the energy
consumption of the vehicle, since the GVW of the vehicle would
reduce drastically with the same payload. The inuence of GVW
on the energy consumption can be seen by examining the
relation shown in eq 1.
The cost for the above-estimated pack size is shown in Figure
2(c), and we observe larger bounds around the cost of the battery
packs for a longer driving range. The cost estimates use the
uncertainties associated with the pack size and the cost per unit of
energy ($/kWh). For current Li-ion systems, the cost of the pack
is in the range of $160 000$210 000 for a driving range of 300
miles, and with beyond Li-ion systems, the value could be lower
than $100 000. For 600 miles, current Li-ion could cost as much
Table 1. Bounds and Mean Values of Variables
parameter [bounds], mean
Cd[0.45, 0.7], 0.63
Crr [0.0045, 0.01], 0.0063
WT(1000 kg) [27, 36], 32
WV(1000 kg) [7, 10], 8
v(m/s); (mph) [16, 21], 19; [36, 47], 43
vrms (m/s); (mph) [19, 24], 22; [43, 54], 49
CostkWh Li-ion ($/kWh) [150, 300], 190
CostkWh beyond Li-ion ($/kWh) [80, 200], 120
SPLi-ion (Wh/kg) [220, 300], 243
SPbeyond Li-ion (Wh/kg) [350, 700], 500
ACS Energy Letters Viewpoint
DOI: 10.1021/acsenergylett.7b00432
ACS Energy Lett. 2017, 2, 16691673
1670
as $400 000, but with a vehicle redesign to a low Cdof 0.45, a Crr
of less than 0.0045, the pack would cost $320 000 at $190/kWh,
which is the current estimate for the price of a Tesla battery pack.
Thelongestrangeconsideredof900mileswouldbe
commercially impractical, costing over $450 000. With a beyond
Li-ionbattery pack placed in a well-designedvehicle with
optimal values of design parameters mentioned before, a 600-
mile capable battery pack would cost $250 000, which is around
25% higher than the mean value of a current Li-ion battery pack
capable of a short 300 mile range. It should be noted that each of
these values is the cost of the battery pack alone, and the entire
vehicle would include several other costs. For comparison, an
equivalent diesel-powered vehicle would cost only $120 000, but
a true comparison should include the operating costs of the
vehicle and not only the initial costs in order to account for the
dierence in the price of electricity and fossil fuels as well as the
signicantly higher eciency of electric drivetrains.
The payload capacity of these vehicles, as stated before, is an
important parameter for the trucking industry, and in a fully
electric vehicle, the payload capacity would be reduced
signicantly because the battery pack weight forms a signicant
fraction of the GVW. The estimates for the maximum payload
capacity can be seen in Figure 2(d). For 600 miles, the vehicle
would house about 11 000 kg (12 tons) of payload, which is
about three-fourths of the current average payload carried by
Class 8 trucks, 14 500 kg (16 tons) as mentioned before. For the
same range, we could have a maximum payload of about 13 600
kg (15 tons) if the vehicle is designed with the lowest coecient
of drag, rolling resistance, and vehicle weight and the battery with
the highest possible specic energy with current Li-ion systems.
Another important observation is that a 600-mile capable battery
pack would weigh over 16 000 kg (18 tons), which is much more
than the available payload capacity of 12 tons. The weight of the
battery pack in comparison to the payload carried provides the
point for an interesting discussion, if the battery pack is much
heavier than the payload, then it implies that a greater fraction of
the energy consumed to move the vehicle is spent on moving the
battery pack rather than the payload. Only at a shorter range of
under 600 miles would the vehicle be practical considering the
average required payload capacity of over 16 tons.
A key conclusion from this analysis is that, with current Li-ion
batteries, we would have no meaningful payload capacity if we
need a driving range of 900 miles since the battery pack and the
vehicle weight together would account for nearly the entire GVW
limit of 36 000 kg (40 tons). The payload capacity would increase
signicantly with a transition to beyond Li-ion systems, which
show a mean payload capacity of 23 500, 20 000, and 16 300 kg
(26, 22, and 18 tons) for 300, 600, and 900 miles respectively due
to the much higher specic energy with a mean value of 500 Wh/
kg equivalent to an advanced Li-ion or Li-S battery. The current
Figure 2. Estimated distributions of the required pack size, weight of the battery pack, cost of the pack, and maximum payload capacity for a fully
electric Class 8 truck with driving ranges of 300, 600, and 900 miles. (a) The estimated pack size (in kWh) for various driving ranges, which is
primarily determined by the gross vehicle weight. (b) The required pack weight for various driving ranges which is primarily determined by the
specic energy, and we see that Li-ion based battery packs (mean cell level specic energy of 243 Wh/kg) are signicantly heavier than potential
beyond Li-ion systems (mean cell level specic energy of 500 Wh/kg). (c) The pack cost for various driving ranges, which is determined by the
mean cost of the battery packs. Potential beyond Li-ion systems benet from requiring a signicantly smaller battery pack and higher energy in
each cell. The cost of the battery pack is a major bottleneck. (d) The lower pack weight of beyond Li-ion systems directly results in an increased
payload capacity, and the system would be practical for a long driving range of 600 miles. For reference, the current average payload carried by
Class 8 trucks is 16 US Ton, and the legal limit for the GVW is 40 US Ton. We observe that the low specic energy of current Li-ion batteries
renders a vehicle with a low payload capacity of under 10 US Ton for a driving range of 600 miles and higher.
The payload capacity of these vehicles
is an important parameter for the
trucking industry, and in a fully electric
vehicle, the payload capacity would be
reduced signicantly because the bat-
tery pack weight forms a signicant
fraction of the gross vehicle weight.
ACS Energy Letters Viewpoint
DOI: 10.1021/acsenergylett.7b00432
ACS Energy Lett. 2017, 2, 16691673
1671
required driving range close to 600 miles would have feasible
payload capacity only with much higher specic energy, and
current Li-ion batteries are clearly not suitable for longer driving
range.
The results of the optimistic scenario considered are shown in
Figure 3. With the high battery-to-wheels eciency and no road
gradient, the energy consumption is reduced to a range of 1.6 and
2.2 kWh/mile with a mean value of 1.9 kWh/mile or 47.5 Wh/
ton-mile. The pack required, shown in Figure 3(a) is now
reduced to 700, 1400, and 2000 kWh for 300, 600, and 900 miles
respectively. The pack weight, in Figure 3(b) is consequently
lower but still remains over 12 tons for a driving range of 600
miles or greater with current Li-ion batteries. The pack cost, in
Figure 3(c) also remains at very high values, where the pack
required for 600 and 900 miles costs over $250 000 and $350 000
respectively. In comparison to the payload capacity of the earlier
scenario shown in Figure 2(d), the payload capacity of the
optimistic scenario in Figure 3(d) is about 5 to 10 tons higher,
depending on the driving range. The zero road grade assumption
for the optimistic scenario reduces a signicant amount of the
energy consumption, but it is important to quantify this
assumption which translates to ignoring an energy consumption
increase of 1.6r kWh for every mile traveled at a road grade of r
%, for a GVW of 36 000 kg (40 tons). The 1.6r kWh per mile
quantity is derived from the road gradient term in eq 1.
Our analysis suggests that there are clear limitations in the cost
and payload capacity for fully electric Class 8 trucks like the Tesla
Semi, although the lower energy consumption of electric
drivetrains remains a compelling motivating factor. With all of
the parameters considered, as we attempt to design heavy-duty
vehicles with a longer range the limitations of current Li-ion
batteries are evidently magnied. Current Li-ion batteries would
not be technically feasible solutions because of their lower
specic energy values, and the longer driving range and higher
payload capacity required by the trucking industry would be met
only by beyond Li-ion solutions to the battery pack. Although
there exists a large uncertainty in the cost of the battery pack due
to the increased Li-ion production by the Tesla Gigafactory, the
initial investment cost for the battery pack would be the most
signicant limiting factor when compared against the cost of
existing diesel-powered vehicles. The targets needed for a driving
range of 600 miles and to carry a payload of over 10 tons are a
specic energy well in excess of 400 Wh/kg at the cell level
costing less than $100/kWh along with a vehicle designed with a
Cdof 0.45, a Crr of under 0.005, and an empty vehicle weight of
under 7000 kg. We end with a word of caution that autonomous
driving could potentially play a crucial role in changing the
landscape of the trucking industry, because a drastic change from
the current known driving patterns could have signicant impact
on the energy and power requirements of the vehicle; an analysis
of these eects is well beyond the scope of the present study.
Shashank Sripad
Venkatasubramanian Viswanathan*
Figure 3. The gure shows the estimated distributions for the optimistic scenario, with no road gradient, a high battery-to-wheels eciency, and a
relaxed oversizing fraction. With the much lower energy consumption, the required pack size, shown in part (a), is more than 25% lower than the
estimates in Figure 2(a). (b) The pack weight is over 12 tons for a driving range of over 600 miles but would be reduced to half the value with a
battery pack based on beyond Li-ion batteries. (c) For the optimistic scenario, the pack would still cost over $200 000 for a driving range of 600
miles or higher. We also observe that the cost of a 900-mile capable beyond Li-ion pack would cost as much as a current Li-ion pack capable of 600
miles, similar to what is seen in Figure 2(c). (d) The lower pack weight of the optimistic scenario provides an additional payload capacity of up to
10 tons compared to Figure 2(d). At the same time it is important to point out that these optimistic parameters considered could be achieved for
specic use-cases with a at road, a highly ecient powertrain, and use-cases where only a fraction of the rated driving range is utilized implying a
low depth-of-discharge.
With all of the parameters considered,
as we attempt to design heavy-duty
vehicles with a longer range the
limitations of current Li-ion batteries
are evidently magnied.
ACS Energy Letters Viewpoint
DOI: 10.1021/acsenergylett.7b00432
ACS Energy Lett. 2017, 2, 16691673
1672
Department of Mechanical Engineering, Carnegie Mellon
University, Pittsburgh, Pennsylvania 15213, United States
AUTHOR INFORMATION
Corresponding Author
*E-mail: venkvis@cmu.edu.
ORCID
Venkatasubramanian Viswanathan: 0000-0003-1060-5495
Notes
Views expressed in this Viewpoint are those of the authors and
not necessarily the views of the ACS.
The authors declare no competing nancial interest.
ACKNOWLEDGMENTS
The authors acknowledge helpful discussions with Steve LeVine,
writer and journalist. The authors also thank Godfrey Sikha and
co-workers at Tesla, Inc. for their helpful comments and
suggestions.
ABBREVIATIONS
TED Technology, Entertainment, and Design con-
ferences
EPA Environmental Protection Agency
GVW Gross vehicle weight
NREL National Renewable Energy Laboratory
DriveCAT Drive Cycle Analysis Tool
CARB HHDDT California Air Resources Board - Heavy Heavy
Duty Diesel Truck
FHWA Federal Highway Administration
REFERENCES
(1) Musk, E. The Future Were Building and Boring. www.ted.com/
talks/elon_musk_the_future_we_re_building_and_boring, 2017.
(2) Greene, D. L.; Plotkin, S. Reducing greenhouse gas emission from US
transportation; Technical Report; Center for Climate and Energy
Solutions: Arlington, VA, 2011.
(3) Banker, S. The Tesla Semi Is a Bad Idea. www.forbes.com/sites/
stevebanker/2017/04/14/the-tesla-semi-a-bad-idea/#9740cfb62961
(accessed May 10, 2017), 2017.
(4) Jenkins, A. Will Anybody Actually Use Teslas Electric Semi Truck?
www.fortune.com/2017/04/14/tesla-semi-truck/ (accessed May 10,
2017), 2017.
(5) Hull, D. Musk Lays out Timeline for Tesla Semi Truck, Electric
Pickup. https://www.bloomberg.com/news/articles/2017-04-13/
musk-sketches-out-timeline-for-tesla-electric-semi-truck-pickup (ac-
cessed May 10, 2017), 2017.
(6) Sen, B.; Ercan, T.; Tatari, O. Does a battery-electric truck make a
difference?-Life cycle emissions, costs, and externality analysis of
alternative fuel-powered Class 8 heavy-duty trucks in the United States.
J. Cleaner Prod. 2017,141, 110121.
(7) Anair, D.; Mahmassani, A.; Cooke, D. Engines for Change;
Technical Report; Union of Concerned Scientists: Cambridge, MA,
2015.
(8) National Research Council. Technologies and approaches to reducing
the fuel consumption of medium-and heavy-duty vehicles; Technical Report,
Natl. Academies Press: Washington, DC, 2010.
(9) Sakti, A.; Azevedo, I. M.; Fuchs, E. R.; Michalek, J. J.; Gallagher, K.
G.; Whitacre, J. F. Consistency and robustness of forecasting for
emerging technologies: The case of Li-ion batteries for electric vehicles.
Energy Policy 2017,106, 415426.
(10) Fixing Americas Surface Act (FAST Act) Truck Size and Weight
Provisions; Report, Federal Highway Administration, U.S. Department
of Transportation, 2016.
(11) Eroglu, D.; Ha, S.; Gallagher, K. G. Fraction of the theoretical
specific energy achieved on pack level for hypothetical battery
chemistries. J. Power Sources 2014,267,1419.
(12) Sapunkov, O.; Pande, V.; Khetan, A.; Choomwattana, C.;
Viswanathan, V. Quantifying the promise of beyond Li-ion batteries.
Transl. Mater. Res. 2015,2, 045002.
NOTE ADDED AFTER ASAP PUBLICATION
This paper was published ASAP on June 27, 2017 with an error in
equation 1. The corrected paper was reposted on July 3, 2017.
We end with a word of caution that
autonomous driving could potentially
play a crucial role in changing the
landscape of the trucking industry.
ACS Energy Letters Viewpoint
DOI: 10.1021/acsenergylett.7b00432
ACS Energy Lett. 2017, 2, 16691673
1673
... For example, in [Mareev, 2018], the simulation model determines the aging of the battery and thus the life span based on the vehicle energy requirements. While in other works, like [Sripad & Viswanathan, 2017] and [Link et al., 2021], the energy required to overcome the driving resistances is calculated with a single simplified equation using an overall vehicle efficiency, which represents the efficiency from the battery to the wheel. ...
... However, some works in the literature instead simulate the relevant roads using the GPS coordinates of the particular road and possibly supplemented with road elevation information. Subsequently, constant speed values are assumed based on the maximum speed allowed on the given road, as in [Mareev, 2018], [Earl et al., 2018] and [Zhao et al., 2018], or cycle-specific average speed values based on driving cycle databases are used, as in [Sripad & Viswanathan, 2017] and [Link, et al., 2021]. Similarly, constant acceleration and deceleration values are also assumed, representing the vehicle's acceleration or braking to keep with the set speed value. ...
Thesis
Electrification has emerged as the optimal approach for achieving higher energy efficiency and reduced emissions in the realm of road vehicles. While electrification is gaining ground in the passenger car and light-duty vehicle sectors, it exhibits a lower degree of maturity in the heavy-duty truck domain. One of the key factors contributing to this situation is the substantial initial investment costs resulting from higher energy consumption and more demanding applications, leading to markedly different economic viability conditions for electric heavy-duty trucks. Conversely, electric vehicles offer the advantage of significantly lower energy consumption and reduced maintenance expenses. Consequently, the economic feasibility of employing electric heavy-duty trucks hinges on whether the higher initial investment costs can be offset by lower operating expenditures, necessitating a comprehensive analysis of operational profiles and working conditions to ascertain the truck's realistic energy consumption for a given application. This analysis enables the prediction of whether electrifying the specific heavy-duty truck represents a viable option. Furthermore, it is crucial to underscore other favorable aspects of electric trucks, such as emissions and noise reduction, as well as enhanced efficiency. To address this matter, this thesis presents a simulation-based approach to analyze and evaluate the electrification potential of a special-purpose heavy-duty truck employed in a specific context. The study investigates the competitiveness of an electric prototype truck, developed within a research project, in comparison to conventional diesel catering lift trucks utilized at airports. Real individual usage profiles form the basis for the assessment. Initially, the relevant operating profiles and work conditions pertaining to the catering lift trucks at Frankfurt Airport are determined. Subsequently, a simulation model is employed to calculate the energy consumption for complete work cycles, encompassing driving and operation of the lifting system, utilizing recorded operational data. Based on the simulation results, an efficiency analysis is conducted for both the driving and lifting systems, as well as the overall vehicle. This is followed by a total cost of ownership analysis, which considers all costs associated with the acquisition, operation, and disposal of the trucks, aiming to determine the economic potential and cost differentials throughout their service lifespan. The life cycle environmental impact of the considered catering lift trucks is also evaluated, with a focus on the aspects that distinguish the environmental balance between the two truck technologies. Additionally, noise emissions from both trucks are measured and assessed during driving and lifting system operation. The findings of this study demonstrate that the electric truck exhibits significant advantages in terms of efficiency improvement and consumption reduction compared to the conventional diesel truck. The total cost of ownership analysis reveals that while the electric truck entails substantial incremental acquisition costs, it compensates for this with notable benefits in operational expenses and residual value as compared to the diesel truck. However, the economic operation and profitability of the electric truck are heavily contingent on the acquisition costs and battery prices. The comparison of the environmental impact between the two trucks indicates that, even when utilizing the current German electricity mix, the electric truck demonstrates a superior environmental footprint in comparison to the diesel truck. This advantage is further enhanced when exclusively utilizing electricity from renewable energy sources. Lastly, the acoustic measurements comparison demonstrates that the electric truck generates lower noise levels than the diesel truck, both during driving and lifting system operation.
... Nevertheless, for Na(110)/NaF(110) and Na(110)/NaF(111), weaker levels of sodiophilicity were able to reduce the critical temperatures related with void dissolution. This can also be seen in Figure S9, where values of the both Ω (11) and Ω (12) are plotted against approximate interface adhesion. A more in-depth discussion of the relationship (or lack thereof) between approximate adhesion and thermodynamic parameters can be found in the Supporting Information. ...
... Interfaces with weaker interactions, such as (b), present no large charge redistribution within the Na-slab. of the ϵ (11) V V parameter, with darker, more opaque lines corresponding to stronger interactions, and lighter, more transparent lines corresponding to weaker interactions. A similar plot for ϵ (12) V V is shown in Figure S10. It can be observed that there exists some relationship between charge density heterogeneity and thermodynamic parameters, despite it not being substantial. ...
Article
Full-text available
Techno-economic, humanitarian, and safety concerns limit the possible uses of conventional lithium-ion and lithium-metal batteries. Sodium-based batteries constitute a promising alternative to address these issues; however, due to the similarities...
Preprint
Next generation batteries based on lithium (Li) metal anodes have been plagued by the dendritic electrodeposition of Li metal on the anode during cycling, resulting in short circuit and capacity loss. Suppression of dendritic growth through the use of solid electrolytes has emerged as one of the most promising strategies for enabling the use of Li metal anodes. We perform a computational screening of over 12,000 inorganic solids based on their ability to suppress dendrite initiation in contact with Li metal anode. Properties for mechanically isotropic and anisotropic interfaces that can be used in stability criteria for determining the propensity of dendrite initiation are usually obtained from computationally expensive first-principles methods. In order to obtain a large dataset for screening, we use machine learning models to predict the mechanical properties of several new solid electrolytes. We train a convolutional neural network on the shear and bulk moduli purely on structural features of the material. We use AdaBoost, Lasso and Bayesian ridge regression to train the elastic constants, where the choice of the model depended on the size of the training data and the noise that it can handle. Our models give us direct interpretability by revealing the dominant structural features affecting the elastic constants. The stiffness is found to increase with a decrease in volume per atom, increase in minimum anion-anion separation, and increase in sublattice (all but Li) packing fraction. Cross-validation/test performance suggests our models generalize well. We predict over 20 mechanically anisotropic interfaces between Li metal and 6 solid electrolytes which can be used to suppress dendrite growth. Our screened candidates are generally soft and highly anisotropic, and present opportunities for simultaneously obtaining dendrite suppression and high ionic conductivity in solid electrolytes.
Article
Full-text available
There is a growing consensus that future specific energy improvements in Li–ion batteries may not ever be sufficient to allow mass market adoption of electric vehicles, as we approach the physical limits of storage capacity of current Li–ion batteries. Several 'beyond li-ion' (BLI) chemistries are being explored as possible high-energy-density alternatives to Li–ion batteries. In this article, we focus on analyzing three BLI battery systems: Li–air, Li–sulphur and Na–air. We present a comprehensive discussion of the fundamental material challenges associated with these chemistries and document the progress being made in translating next-generation battery systems from the lab to the market. We also carry out a critical examination of the hype surrounding emerging battery technologies. We report, for the first time, a hype chart for batteries akin to those popularized by Gartner, Inc. for emerging technologies. We expect this hype chart to give us better insights on the respective standings of the current BLI technologies.
Article
There are a large number of accounts about rapidly declining costs of batteries with potentially transformative effects, but these accounts often are not based on detailed design and technical information. Using a method ideally suited for that purpose, we find that when experts are free to assume any battery pack design, a majority of the cost estimates are consistent with the ranges reported in the literature, although the range is notably large. However, we also find that 55% of relevant experts’ component-level cost projections are inconsistent with their total pack-level projections, and 55% of relevant experts’ elicited cost projections are inconsistent with the cost projections generated by putting their design- and process-level assumptions into our process-based cost model (PBCM). These results suggest a need for better understanding of the technical assumptions driving popular consensus regarding future costs. Approaches focusing on technological details first, followed by non-aggregated and systemic cost estimates while keeping the experts aware of any discrepancies, should they arise, may result in more accurate forecasts.
Article
In valuing new active materials chemistries for advanced batteries, the theoretical specific energy is commonly used to motivate research and development. A packaging factor is then used to relate the theoretical specific energy to the pack-level specific energy. As this factor is typically assumed constant, higher theoretical specific energies are judged to result in higher pack-level specific energies. To test this implicit assumption, we calculated the fraction of the theoretical specific energy achieved on the pack level for hypothetical cell chemistries with various open-circuit voltages and theoretical specific energies using a peer-review bottom-up battery design model. The pack-level specific energy shows significant dependence on the open-circuit voltage and electrochemical impedance due to changes in the quantity of inactive materials required. At low-valued average open-circuit voltages, systems with dramatically different theoretical specific energies may result in battery packs similar in mass and volume. The fraction of the theoretical specific energy achieved on the pack level is higher for the lower theoretical specific energy systems mainly because the active materials mass dominates the pack mass. Finally, low-valued area-specific impedance is shown to be critical for chemistries of high theoretical specific energy and low open-circuit voltage to achieve higher pack-level specific energies.
Engines for Change; Technical Report; Union of Concerned Scientists: Cambridge, MA, 2015. (8) National Research Council. Technologies and approaches to reducing the fuel consumption of medium-and heavy-duty vehicles
  • J Cleaner Prod Anair
  • D Mahmassani
  • A Cooke
J. Cleaner Prod. 2017, 141, 110−121. (7) Anair, D.; Mahmassani, A.; Cooke, D. Engines for Change; Technical Report; Union of Concerned Scientists: Cambridge, MA, 2015. (8) National Research Council. Technologies and approaches to reducing the fuel consumption of medium-and heavy-duty vehicles; Technical Report, Natl. Academies Press: Washington, DC, 2010. (9) Sakti, A.; Azevedo, I. M.; Fuchs, E. R.; Michalek, J. J.; Gallagher, K.
) Sen, B.; Ercan, T.; Tatari, O. Does a battery-electric truck make a difference?-Life cycle emissions, costs, and externality analysis of alternative fuel-powered Class 8 heavy-duty trucks in the United States
  • E Musk
  • D L Greene
  • S Plotkin
Musk, E. The Future We're Building − and Boring. www.ted.com/ talks/elon_musk_the_future_we_re_building_and_boring, 2017. (2) Greene, D. L.; Plotkin, S. Reducing greenhouse gas emission from US transportation; Technical Report; Center for Climate and Energy Solutions: Arlington, VA, 2011. (3) Banker, S. The Tesla Semi Is a Bad Idea. www.forbes.com/sites/ stevebanker/2017/04/14/the-tesla-semi-a-bad-idea/#9740cfb62961 (accessed May 10, 2017), 2017. (4) Jenkins, A. Will Anybody Actually Use Tesla's Electric Semi Truck? www.fortune.com/2017/04/14/tesla-semi-truck/ (accessed May 10, 2017), 2017. (5) Hull, D. Musk Lays out Timeline for Tesla Semi Truck, Electric Pickup. https://www.bloomberg.com/news/articles/2017-04-13/ musk-sketches-out-timeline-for-tesla-electric-semi-truck-pickup (accessed May 10, 2017), 2017. (6) Sen, B.; Ercan, T.; Tatari, O. Does a battery-electric truck make a difference?-Life cycle emissions, costs, and externality analysis of alternative fuel-powered Class 8 heavy-duty trucks in the United States. J. Cleaner Prod. 2017, 141, 110−121. (7) Anair, D.; Mahmassani, A.; Cooke, D. Engines for Change;
National Research Council. Technologies and approaches to reducing the fuel consumption of medium-and heavy-duty vehicles
  • A Sakti
  • I M Azevedo
  • E R Fuchs
  • J J Michalek
  • K Gallagher
Technical Report; Union of Concerned Scientists: Cambridge, MA, 2015. (8) National Research Council. Technologies and approaches to reducing the fuel consumption of medium-and heavy-duty vehicles; Technical Report, Natl. Academies Press: Washington, DC, 2010. (9) Sakti, A.; Azevedo, I. M.; Fuchs, E. R.; Michalek, J. J.; Gallagher, K.
Consistency and robustness of forecasting for emerging technologies: The case of Li-ion batteries for electric vehicles
  • G Whitacre
  • J F Eroglu
  • D Ha
  • S Gallagher
  • K G Sapunkov
  • O Pande
  • V Khetan
  • A Choomwattana
  • C Viswanathan
G.; Whitacre, J. F. Consistency and robustness of forecasting for emerging technologies: The case of Li-ion batteries for electric vehicles. Energy Policy 2017, 106, 415−426. (10) Fixing America's Surface Act (FAST Act) Truck Size and Weight Provisions; Report, Federal Highway Administration, U.S. Department of Transportation, 2016. (11) Eroglu, D.; Ha, S.; Gallagher, K. G. Fraction of the theoretical specific energy achieved on pack level for hypothetical battery chemistries. J. Power Sources 2014, 267, 14−19. (12) Sapunkov, O.; Pande, V.; Khetan, A.; Choomwattana, C.; Viswanathan, V. Quantifying the promise of beyond Li-ion batteries. Transl. Mater. Res. 2015, 2, 045002.