Articles

https://doi.org/10.1038/s41560-022-01065-y

Rapid battery cost declines accelerate the

prospects of all-electric interregional container
shipping
Jessica Kersey 1
, Natalie D. Popovich2 and Amol A. Phadke 2 ✉

International maritime shipping—powered by heavy fuel oil—is a major contributor to global CO2, SO2, and NOx emissions. The
direct electrification of maritime vessels has been underexplored as a low-emission option despite its considerable efficiency
advantage over electrofuels. Past studies on ship electrification have relied on outdated assumptions on battery cost, energy
density values and available on-board space. We show that at battery prices of US$100 kWh−1 the electrification of intrare-
gional trade routes of less than 1,500 km is economical, with minimal impact to ship carrying capacity. Including the environ-
mental costs increases the economical range to 5,000 km. If batteries achieve a US$50 kWh−1 price point, the economical range
nearly doubles. We describe a pathway for the battery electrification of containerships within this decade that electrifies over
40% of global containership traffic, reduces CO2 emissions by 14% for US-based vessels, and mitigates the health impacts of
air pollution on coastal communities.

T
ransporting 11 billion tonnes annually, the maritime shipping Maersk, the largest shipping company by volume, is already pilot-
industry handles nearly 90% of global trade by mass1,2. The ing battery hybridization on a containership operating between
industry’s meteoric growth has been underpinned by access East Asia and West Africa13. A fully electric 80 m containership,
to cheap, energy-dense heavy fuel oil (HFO). The shipping industry the Yara Birkeland, is expected to begin autonomous operation in
consumes 3.5 million barrels of low-grade HFO annually, produces Norway in the early 2020s. Similar battery-electric vessel projects
2.5% of total anthropogenic carbon dioxide equivalent (CO2e) emis- are underway in Japan, Sweden and Denmark14,15. However, system-
sions in 20182,3, and engenders enormous damages from marine atic analysis of the adoption potential for battery-electric container-
eutrophication and ecotoxicity, air pollution, and climate change ships has yet to be conducted. With the exception of these initial
impacts4. By 2050, maritime shipping emissions are projected to pilot projects, battery-electric propulsion has been underexplored
contribute as much as 17% of global CO2e emissions5,6. The indus- as a potential low-emissions alternative in the marine shipping sec-
try’s outsized contribution to criteria air pollutants—12% and 13% tor despite: its considerable emissions reduction potential; recent
of global annual anthropogenic SO2 and NOx emissions, respec- decline in battery costs; improvements in battery energy densities;
tively—caused an estimated 403,300 premature deaths from lung increasing availability of low-cost, renewably generated electricity;
cancer and cardiovascular disease in 20203,7. and its substantial efficiency advantage over e-fuels such as green
Mounting political pressure has prompted the International hydrogen and ammonia.
Maritime Organization (IMO) to take regulatory action to reduce Using the best-available battery costs and energy densities, we
GHG emissions consistent with the Paris Agreement. Actions examine the technical outlook, economic feasibility and environ-
include resolution MEPC.302(72), which aims to reduce annual mental impact of battery-electric containerships. We define two sce-
CO2e emissions by 50% by 2050 from 2008 levels8, and recom- narios: first, a baseline scenario using today’s best-available battery
mended amendments to the International Convention for the costs, HFO costs, battery energy densities and renewable energy
Prevention of Pollution from Ships (MARPOL)—whose members prices; and, second, a near-future scenario that tests the impacts
cover 99.4% of world shipping tonnage—to prohibit using or car- of projected 2030 improvements in these variables. By contrast to
rying HFO in Arctic waters after 20249,10. In concert, IMO’s 2020 most previous studies, we treat the volume repurposed to house
emissions standards reduced the allowable marine fuel sulfur con- the battery energy storage (BES) system as an opportunity cost
tent from 3.5% to 0.5% by mass11. instead of a fixed technical constraint. We specify eight container-
Faced with this tightening regulatory landscape, the marine ship size classes and model their energy needs, their CO2, NOx and
shipping industry is racing to identify commercially deployable SO2 emissions, and total cost of propulsion (TCP) across 13 major
zero-emission alternatives to HFO at a pace sufficient to sub- world trade routes—creating 104 unique scenarios of ship size and
stantially curb the sector’s emissions and avert catastrophic cli- route length that can be compared with almost any containership
mate change. Optimistic outlooks for zero-emissions alternatives operating today. We focus on battery-electric containerships and
for marine applications suggest that electrofuels (e-fuels) would briefly explore the implications of our results for electrifying other
increase the total cost of ownership for bulks carriers by 200–600% ship types. Our results suggest that over 40% of global containership
relative to HFO12. Such analysis prompts additional research into traffic could be electrified cost-effectively with current technology,
which existing propulsion technologies could achieve parity with reducing CO2 emissions by 14% for US-based vessels, and mitigat-
HFO in the near-future, particularly battery-electric propulsion. ing the health impacts of air pollution on coastal communities.

Energy and Resources Group, University of California, Berkeley, CA, USA. 2Energy Analysis & Environmental Impacts Division, Energy Technologies Area,
1

Lawrence Berkeley National Laboratory, Berkeley, CA, USA. ✉e-mail: [email protected]

664 Nature Energy | VOL 7 | July 2022 | 664–674 | www.nature.com/natureenergy

Nature Energy Articles
The search for low-emissions pathways for maritime small neo-Panamax ship, we estimate that a 5 GWh battery with
shipping lithium iron phosphate (LFP) chemistry, with a specific energy of
In the short term, most ship operators have turned to energy effi- 260 Wh kg−1 (ref. 34), will weigh 20,000 t and increase the draught by
ciency measures such as slow steaming (deliberately reducing a ship’s 1 m—a small fraction of the ship’s total height and well within the
cruising speed to reduce fuel consumption), route optimization and bounds of the vessel’s Scantling (maximum) draught. For voyages
hull fouling management to meet IMO mandates16. However, the longer than 5,000 km, the increase in draught exceeds the vessel’s
10–15% emissions reductions achievable through these measures Scantling draught.
are not sufficient to comply with forthcoming IMO efficiency regu- The distribution of additional weight also impacts the hydrody-
lations17,18. Hybrid battery technology has been explored a viable namics, aerodynamics, stability and energy consumption of a ves-
short-term solution to reduce—but not eliminate—emissions from sel35. Internal combustion engine (ICE) vessels use a ballast system
fossil-fuel energy sources. One study suggests a best-case scenario whereby water tanks charge and discharge depending on the cargo
for hybrid systems is only 14% reduction in emissions for dry bulk load to distribute weight and counteract buoyancy. Case studies
carriers (comprising 2% of global fleet emissions)19, not substantially of fully electric or hybrid propulsion systems suggest that ballast
better than the existing energy efficiency measures. Small modular systems can be partially or fully replaced by BES systems without
nuclear reactors, which have been used in military and submarine substantial impacts to symmetry (trim) and balance by distributing
applications for decades20, are a viable alternative, but are unlikely battery components throughout existing void, mechanical and bal-
to achieve wide-spread deployment in commercial vessels given the last spaces35. Furthermore, BES systems do not need to be arranged
regulatory challenges surrounding nuclear proliferation, safety and around a central drive shaft and can be more flexibly configured
waste disposal. Marine gas oil, liquefied petroleum gas, liquefied nat- within the vessel’s interior12,36. The volume of an onboard BES sys-
ural gas, methanol and their bio-derivations have received substan- tem depends on the ship’s power requirements, cruising speed,
tial attention as medium- to long-term options, but recent research voyage length, electrical efficiency and battery energy density.
has questioned the potential of these fuels to reach cost parity and Containership energy consumption can be approximated with the
considerably reduce lifecycle GHG emissions21–23. Not all transport Admiralty Law, a version of the propeller law that is widely used
modes are viable candidates for immediate and direct electrification; in first-order estimations of ship power requirements and fuel con-
commercial jet planes cannot reasonably be electrified until battery sumption37,38. Although a bottom-up approach to estimating energy
pack specific energy increases to three to ten times their current requirements would incorporate additional terms, our objective is
values24. It is within this context that propulsion technologies gen- to capture the relative changes in energy requirements between the
erated with renewable power have received the most attention. For two propulsion methods. Assuming an identical vessel and opera-
example, blue hydrogen (hydrogen produced from natural gas with tional profile, the energy needs of ICE and battery-electric ships dif-
carbon capture and storage) is expected to reduce GHG emissions fer only by the engine efficiencies and mass, which directly changes
by only 20% compared with burning natural gas25. Although renew- the vessel draught.
ably produced ammonia and hydrogen provide operational emis-
sions reductions, the inefficiency of the production process relative PSMCR × tvoyage V3average
eICE = × (1)
to HFO makes them unlikely to become sufficiently cost-competitive ηICE V3max
to displace fossil fuels26,27. By contrast, direct electrification is typi-
cally five times more efficient than e-fuels in the transportation sec- Equation (1) describes the energy needs of a ship with a
tor, exclusive of losses from e-fuel transport and storage27. low-speed, two-stroke marine ICE fed by IMO-compliant low-sulfur
By contrast to other modes where battery weight dramatically HFO, where PSMCR is the maximum continuous power rating (where
reduces payload capacity or range, such as light-duty vehicles and SCMR is the specified maximum continuous rating), Vaverage is the
planes, the sheer size of containerships means that the additional average cruising speed, Vmax is the maximum design speed, tvoyage
weight from the battery can potentially be offset with a smaller is the time to traverse the route and ηICE is the ICE tank-to-wake
percentage forfeiture of cargo. Past work has suggested that bat- efficiency.
tery electrification of marine vessels is unfavourable given the
2
low energy density of batteries relative to hydrocarbon fuels28–31. PSMCR × tvoyage T3 V3average
However, their assumptions about battery energy density and cost ebattery = × 2loaded × (2)
ηinverter × ηmotor Treference V3max
are outdated, differing in some cases by one to two orders of mag- 3

nitude from today’s best-available figures of 210 Wh kg−1 specific

energy32 and US$100–134 kWh−1 (ref. 33). Furthermore, these stud- Equation (2) describes the energy needs of an equivalent
ies assumed that the maximum battery capacity is limited by the battery-electric ship, which includes a correction for increased
existing onboard space dedicated to mechanical propulsion systems draught due to battery system weight, where Tloaded is the draught
and fuel storage, so their findings suggest that battery-electric ships when loaded with the battery energy system, Treference is the typical
would require several recharges to traverse even short routes. operating draught, and ηmotor and ηinverter are motor and inverter effi-
ciencies, respectively.
Technical feasibility of battery-electric container shipping Nickel manganese cobalt oxide, LFP, nickel cobalt aluminium
The key technical constraint for battery-electric container ship- and lithium titanate oxide are commercially available lithium-ion
ping is the volume of the battery system and electric motor relative chemistries with the requisite cycle life, specific power, charge rates
to the volume occupied by a vessel’s existing engines, fuel stor- and operating temperatures to support container shipping applica-
age and mechanical space. The extra weight of the BES system is, tions39,40. The choice of battery chemistry depends on specific oper-
however, non-trivial in determining a vessel’s power requirements. ational characteristics. Vessels with shorter, more frequent voyages,
Operationally, containerships can increase their carrying capacity lower power requirements, and charging time constraints would
by increasing draught (that is, the vertical distance between the favour the high charge rates and long lifecycles of LFP batteries41,42.
waterline and the keel) on the basis of the Archimedes principle. A For ships with longer ranges and less frequent battery cycling, the
higher draught increases the hull resistance, and thus more power is relatively low cycle life and high energy density of nickel manganese
required to achieve the same speed. On voyages less than 5,000 km, cobalt oxide batteries may be more suitable. Given that electrifica-
we find that the necessary increase in power is less than 10% of the tion will probably be limited to small, short-range vessels until bat-
original power requirements. For example, for a 5,000 km range tery costs are further reduced, we model the use of LFP batteries.

Nature Energy | VOL 7 | July 2022 | 664–674 | www.nature.com/natureenergy 665

Articles Nature Energy

a b

i
ai

ai
ha

ha
ai

ai
gh

gh
gh

gh
ng

ng
ai

ai
ai

ai
an

an
gh

gh
an

an
ha

ha
gh

gh
Sh

Sh
an

an
Sh

Sh
–S

–S
an

an
e–

e–
Sh

Sh
k–

k–
ne

ne
Sh

Sh
or

or
n–

n–
r

r
ur

ur
n–

n–
Yo

Yo
ap

ap
ba

ba
bo

bo
sa

sa
ew

ew
ng

ng
ur

ur
el

el
Bu

Bu
Si

Si
M

M
N

N
D

D
Ultra-large co
0 0 ntainer vess
el (∼18,000
TEU)
Ultra
-larg
e co
ntain
Volume converted to battery (percent of total TEU capacity)

Volume converted to battery (percent of total TEU capacity)

er v
–10 esse –10
l (∼1
8,00 Sm
0 TE
U) all
fee
de
r (∼
–20 –20 1,0
00
TE
U)
Sm

–30 –30
all
fee
de
r(
∼1

–40 –40
,00
0T
EU
)

–50 –50

–60 –60

–70 –70

–80 Energy density = 470 Wh l

–1
Baseline –80 Energy density = 1,200 Wh l
–1
Near future
10,000

12,000

14,000

16,000

18,000

20,000

22,000

10,000

12,000

14,000

16,000

18,000

20,000

22,000
2,000

4,000

6,000

8,000

2,000

4,000

6,000

8,000
0

Voyage length (km) Voyage length (km)

Fig. 1 | Carrying capacity forfeited to onboard battery system as percentage of total TEU by voyage length. We model the volume of the ICE ship’s
combined engine and mechanical space, assuming a battery packing fraction of 0.76 and an 80% depth of discharge. The line thicknesses denote
increasing vessel carrying capacity. A small feeder, with a TEU capacity of around 1,000, is the smallest vessel modelled, whereas the ultra-large container
vessel, with a TEU capacity of around 18,000, is the largest. a, The baseline scenario results, with a battery energy density of 470 Wh l−1. In this scenario,
the battery volume is less than that of the existing ICE mechanical space at voyage lengths less than 1,300–2,000 km. The impacts of the battery system
volume on TEU forfeiture decreases as ship capacity increases, reflecting innovations in ultra-large containership design that optimize carrying capacity
and energy consumption better than feeder ships. b, The results with a battery energy density of 1,200 Wh l−1. In this near-future scenario, the net change
in carrying capacity is positive for voyages of up to 2,000–5,000 km, depending on ship type.

We find that minimal carrying capacity must be repurposed to capacity to charge within the available port time is less than 300 MW
house the battery system for most ship size classes and along short for all ship classes on voyages less than 10,000 km. We estimate that
to medium-length routes. For a small neo-Panamax containership, a 220 MW charger could charge a 7,650 TEU small neo-Panamax
representing an average containership in the global fleet, the vol- containership in 24 h. For longer voyages requiring larger battery
ume required by the battery system is less than the volume currently capacities, offshore charging infrastructure could be strategically
dedicated to the ICE and fuel tanks for routes under 3,000 km. For located in global shipping chokepoints such as the Strait of Hormuz,
the longest modelled route of 20,000 km for this ship class, the bat- the Panama Canal and the Strait of Malacca, where ships regularly
tery would occupy 2,500 twenty-foot equivalent unit (TEU) slots or queue for days awaiting passage.
32% of the ship’s carrying capacity. Supplementary Table 1 provides A number of contact-based options are already commercially
the baseline values used for each ship class. Figure 1 shows the per- available for the shore-to-ship interface, including manual and auto-
centage of ship carrying capacity forfeited to the BES system for the mated plugs from ABB, Cavotec, Mobimar, Zinus and Stemmann–
eight modelled ship classes across routes from 0 to 22,000 km, with Technik, with non-contact inductive charging solutions currently
current and near-future battery energy densities. We find that as car- under development46. Charging stations can be deployed at port
rying capacity increases, the percentage of total carrying capacity terminals or offshore to allow ships to charge while queuing for
volume occupied by batteries decreases because larger ships typically berth allocation.
have lower energy requirements per unit of carrying capacity43,44. The optimized and high-throughput nature of port operations
Megawatt-scale charging infrastructure will be required to meet (average berth utilization rates typically exceed 50%) support high
the large energy requirements of battery-electric containerships (for charging infrastructure utilization and associated cost reductions45.
example, 6,500 MWh for a small neo-Panamax containership over Adapting methods used for trucks40 and trains47 we estimate the
a 5,000 km route) without disrupting normal port operation. The levelized cost of a 300 MW charging station interconnected at the
average queuing time plus berthing time in a port is 31 h for con- transmission level to be US$0.03 kWh−1 at 50% utilization, inclusive
tainerships of 1,000–3,000 TEUs and 97 h for the largest contain- of hardware, installation, grid interconnection, and annual opera-
ership size classes of 10,000–20,000 TEUs45. The requisite charger tions and maintenance costs across the system lifetime48.

666 Nature Energy | VOL 7 | July 2022 | 664–674 | www.nature.com/natureenergy

Nature Energy Articles
a b
$500 $500

$450 $38 $427 $450

$296
Total cost of propulsion (USD km–1)

Total cost of propulsion (USD km–1)

$400 $400

$350 $350

$300 $300

$250 $250

$200 $200 $170

$21
$17
$150 $150 $34 $132
$11 $93 $68 –$3
$100 $82 $100

$50 $50 $32 $1

$0 $0

ty

&M

ry

re

on

l
FO

&M

on

ta
es
ta
es

ac
ci

tte

tu

To
rb
To
rb

ag
ag

tr i

O
H

sp
O

c
Ba

ca
ca

tr u
ec

am
m

of
da

El

as
of

rg

d
ca

st
r
st

n
f
n

in

ti o

co
ti o

co

st

lu
lu

Lo

al
in
al

ol
ol

ci
rg
ci

rp
rp

So
So

ha

Ai
Ai

C
c d
$500 $38 $473 $500
$296
$450 $450
Total cost of propulsion (USD km–1)

Total cost of propulsion (USD km–1)

$400 $400
$350 $350
$300 $300
$250 $250
$200 $200
$128 $11 $139
$150 $150 $21 $125
$17
$100 –$5 $25 $87
$100 $34
$50 $50 $33 $1

$0 $0
FO

&M

on

al
es

ty

&M

ry

on

al
t

es
To
rb

ac

ur
ag

ci

t
H

tte
O

To
rb
ag
ca

ct
tri

sp
m

Ba

ca
ru
ec

m
da

of

st

da

of
El

rg
st

fra
n

ca

st
tio

co

n
in

tio

co
lu

st
al

lu
ol

Lo

al
in
ci

ol
rp

ci
rg
So

rp

So
ha
Ai

Ai
C

Fig. 2 | TCP including air pollution of a typical small neo-Panamax vessel. A neo-Panamax vessel of 7,650 TEU is modelled over a 1,565 km voyage.
a, The TCP of an ICE ship in the baseline scenario. b, The TCP of the battery-electric equivalent in the baseline scenario. c,d, The TCP of ICE (c) and
battery-electric (d) vessels in the near-future scenario. Coloured bars (red for ICE, teal for battery-electric) show non-environmental costs. Grey bars
and dashed lines capture environmental damages attributed to NOx, SO2 and CO2. Not accounting for environmental damages, in the baseline scenario,
the cost of the battery system and charging infrastructure outweigh the economic benefits of fuel switching, leading to a battery-electric TCP that is
US$39 km−1 higher than the ICE TCP. The baseline scenario assumes a battery cost of US$100 kWh−1, a battery volumetric energy density of 470 Wh l−1,
charging station utilization of 50%, wholesale electricity price of US$0.035 kWh−1, and a HFO cost of US$0.048 kWh−1 (equivalent to US$538 t−1); in the
near-future scenario, HFO costs of US$840 t−1 (representing a US$100 per tonne tax on CO2e), battery costs of US$50 kWh−1, battery energy density of
1,200 Wh l−1, and a charging infrastructure utilization rate of 70% lead to a battery-electric TCP that is US$52 km−1 lower than the ICE TCP. Accounting for
environmental damages increases the TCP advantage of the battery-electric ship dramatically.

Cost parity with HFO cost of the battery energy system separately, we omit the capital cost
We test the economic feasibility of a battery-electric containership of the vessel, given that propulsion systems constitute only a small
against that of a slow-speed, two-stroke ICE ship fuelled by very portion of ship newbuild costs and the cost advantage of electric
low sulfur fuel oil (VLSFO)—0.5% sulfur content—by calculat- motors relative to marine ICEs.
ing its TCP per kilometre by voyage length. For both ship types, In the baseline scenario, the TCP of a battery-electric ship
we calculate fuel, operations and maintenance costs, as well as the is lower than that of the incumbent ICE vessel only for ship
environmental costs of NOx, SO2 and CO2 emissions from direct classes larger than 8,000 TEUs over voyages of less than 1,000 km
combustion or grid electricity. For battery-electric vessels, we (refs. 5,40,47,49,50). Over longer voyages, the additional cost of the bat-
include the costs of an original and replacement battery set, the tery system, increased power requirements and charging infrastruc-
opportunity cost of forfeiting TEUs to the battery system and the ture outweighs the savings from fuel switching and the efficiency
levelized cost of charging equipment. As we account for the extra gains of direct electrification. However, if the environmental costs of

Nature Energy | VOL 7 | July 2022 | 664–674 | www.nature.com/natureenergy 667

Articles Nature Energy

ne
e
n

or
Ko

ur
ap
an

bo
g

g
on
us

el
in

–M
–H
–B

–S
ai

ai

ai

ai
gh

gh

gh

gh
an

an

an

an
Sh

Sh

Sh

Sh
$500

it y
lin

ns
se

de
ba

gy
el:

er
ss

en
ve

ry
ic

tte
ctr

ba
ele
$400

st
in

co
ry

ts
e
ur

ry
en
tte
ut

tte
f

em
Ba
ar

ba
e

ov
:n

in
el

pr

ns
ss

Im

tio
ve

uc
c
tri

ed
Total cost of propulsion (USD km–1)

ec

R
el
ry
$300 tte
Ba

ICE vessel: near future

$200

Increase in HFO prices

ICE vessel: baseline

$100

Vessel capital cost

$0
0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000 9,000 10,000
Voyage length (km)

Fig. 3 | TCP of ICE and battery-electric small neo-Panamax containerships in baseline and near-future scenarios excluding environmental costs. Red
and teal lines indicate the TCP of an ICE or battery-electric vessel, respectively. Dashed lines represent the near-future scenario. Stars indicate the point
at which battery-electric vessels achieve parity with ICE vessels for both baseline and near-future scenarios. In the baseline scenario, the TCP of the
battery-electric vessel is less than that of the ICE vessel at distances less than 1,000 km. In the near-future scenario, increases in HFO cost equivalent to
US$0.027 kWh−1 enable cost parity across ranges up to 3,300 km. Without increases in HFO prices, the range increases to 2,000 km in the near-future
scenario. Improvements in battery energy density produce small improvements in battery-electric vessel TCP by decreasing the volume forfeited from
the vessel’s carrying capacity to house the battery system. A capital cost of US$64 km−1 is depicted as a grey band to contextualize the magnitude of
the operating expenses79. The vertical dashed lines provide example routes and show that vessels traversing shorter, intraregional routes are prime for
electrification even in the baseline scenario.

NOx, SO2 and CO2 are considered, the cost-effective range increases an average vessel in the global fleet across a 1,565 km voyage from
to 5,000 km across all size classes given the high emissions rates of Hong Kong to Shanghai. Figure 3 depicts the relationship between
HFO relative to the emissions intensity of the US grid. the TCP and voyage length for a small neo-Panamax vessel. The
Under the near-future scenario, the TCP of battery-electric results show improvements in TCP and gains in achievable range by
shipping is lower than that of the incumbent ICE ship at ranges improving charging infrastructure utilization, battery pack cost, and
around 3,000 km for all ship classes. Including environmental costs, battery energy density from baseline to near-future values. Figure
this range expands to 6,500 km for smaller-capacity ships and up 4 displays the difference in TCP between ICE and battery-electric
to 12,000 km for the largest ship classes. However, although these vessels for all vessel size classes across all modelled voyage lengths,
longer ranges are cost-effective, the weight of the batteries drives exclusive of environmental costs.
vessel draught beyond safe operating parameters and thus they are The primary constraint for cost parity of battery-electric ships
unlikely to be candidates for full electrification without substantial with ICE ships over longer ranges is the battery cost. Battery prices
changes in ship design. The fact that bulk carriers such as iron ore need to reach US$20 kWh−1 for a 10,000 km range battery-electric
carriers have much higher weight and draught limits than contain- ship capable of crossing the Atlantic or Pacific Ocean to be
erships points to the possibility of accommodating the additional cost-effective without recharging. Current commercial lithium bat-
weight and draught by changing the ship design. tery technologies, and emerging technologies such as solid-state
Figure 2 presents the TCP analysis in the baseline and near-future batteries, are not projected to decline to this extent given the cost of
scenarios for a 7,650 TEU small neo-Panamax vessel, representing the materials used in these batteries51. However, battery technolo-

668 Nature Energy | VOL 7 | July 2022 | 664–674 | www.nature.com/natureenergy

Nature Energy Articles
a b
Small feeder

Large feeder

Classic panamax and wide-beam

Small neo-Panamax

Large neo-Panamax

Maxi neo-Panamax

Neo post-Panamax

Ultra-large container vessel

10,000
11,000
12,000
13,000
14,000
15,000
16,000
17,000
18,000
19,000
20,000
21,000
22,000

10,000
11,000
12,000
13,000
14,000
15,000
16,000
17,000
18,000
19,000
20,000
21,000
22,000
1,000
2,000
3,000
4,000
5,000
6,000
7,000
8,000
9,000

1,000
2,000
3,000
4,000
5,000
6,000
7,000
8,000
9,000
0

0
Voyage length (km) Voyage length (km)

$0

00
40

20

00

80

60

40

20

$2
1,

1,

1,

–$

–$

–$

–$
–$

–$

–$
Difference in TCP between ICE and battery electric vessel (USD km–1)

Fig. 4 | Difference in TCP between battery-electric and ICE vessels for all eight size classes for voyages up to 22,000 km. a,b, The baseline (a) and
near-future (b) scenarios. TCP excludes environmental costs. A positive value indicates that the TCP of the battery-electric ship is lower than that of
the ICE equivalent, whereas a negative value represents a lower ICE TCP. The TCP difference is larger in magnitude for larger ship classes, indicating
the difficulty of cost-effectively electrifying large containerships over intercontinental routes, but also the potential economic benefit of phasing in
battery-electric vessels over short to medium intraregional routes.

gies designed for long duration storage applications from low-cost primarily constrained by weight rather than volume41. Energy den-
materials are under development. Iron–air batteries, for example, sity by weight is therefore the critical technical parameter for the
offer comparable energy density at a fraction of the cost of current batteries that would power these ships. At the same time, some bulk
lithium-ion batteries and may offer pathways for cost-competitive carriers and oil tankers are designed to carry up to 400,000 t—more
long-range shipping52. than twice the weight of the largest containerships59.
For a 5,000 km range dry bulk carrier, we estimate that the battery
Deployment potential of battery-electric shipping system will constitute 5–6% of the ship weight with current battery
An estimated 42.3 trillion TEUs (40% of global trade) traversed technology and 3–4% with projected increases in energy density by
intraregional routes in 201953. However, this proportion is prob- 203028,41,60. Factors such as the extent to which ships operate at their
ably an underestimate owing to recent trends in containership weight limit, opportunity cost of foregone weight carrying capacity,
logistics and the regionalization of trade54, including an 1,100% and the cost of modest increases to weight carrying capacity of the
increase in average containership capacity between 1968 and ships will determine the impact of battery weight on the economics
201555. The sector’s trend towards containership gigantism has of these ship types.
promoted a hub-and-spoke model of trade, whereby high-capacity
mega-containerships transport goods over long distances from one Emissions reduction potential of battery electrification
hub to another54. From the destination hub, a host of smaller feeder Battery-electric container shipping would eliminate all direct com-
ships transport the containers to their final destinations in smaller bustion emissions and considerably ameliorate localized air pol-
regional ports. Nearly all of these feeder ships traverse short routes lution and related health impacts in communities near ports and
that could be electrified, which would increase battery-electric con- global trade lanes61. However, lifecycle emissions reductions depend
tainership adoption well beyond the potential suggested by intrare- on the pollution intensity of the electricity source as well as trans-
gional trade figures. Figure 5 depicts the ten best-connected ports mission, distribution and charging losses. We compare the CO2,
in the world, all of which are intraregional routes less than 5,000 km NOx and SO2 emissions intensities of a small neo-Panamax contain-
in length2. Moreover, feeder ships are older on average than their ership with a slow-speed diesel engine running on HFO or VLSFO
larger-capacity counterparts, and many are reaching the end of to a battery-electric vessel across a range of realistic well-to-wake
their useful service lives56. The 2020 IMO regulation limiting sul- emissions intensities (Fig. 6). The input tank-to-wake emissions
fur content will probably lead to the premature scrapping of these factors (g kWh−1) include downstream losses attributable to trans-
fuel-inefficient ships, creating an opportunity for battery-electric mission, power conversion, shore-side storage and electric motor
models to enter the fleet57. losses. Battery-electric vessels would also eliminate direct emissions
Although containerships, with their standardized cargo and of black carbon, which is a particular concern for the sizeable per-
volume dependency, are useful for understanding the technoeco- centage of vessels operating in Arctic waters given its demonstrated
nomics of battery-electric shipping, they represent only 23% of role in reducing snow albedo and accelerating ice melt62.
total maritime shipping emissions58. Achieving larger emissions Reductions in carbon emissions and air pollutants are highly
reductions will require electrifying additional ship types, including dependent on the generation matrix of the grid where the vessel
oil tankers, bulk carriers, general cargo ships and cruise liners. Of is charged. Assuming an average grid carbon intensity of 535 g
those, bulk carriers and oil tankers seem to have the largest emis- CO2 kWh−1 (inclusive of transmission, conversion and motor inef-
sion footprint. Unlike containerships, some of these ship types are ficiency losses), a battery-electric containership charged in a US

Nature Energy | VOL 7 | July 2022 | 664–674 | www.nature.com/natureenergy 669

Articles Nature Energy

Busan

Shanghai

<500 km Hong Kong

500–1,000 km
1,000–2,000 km
4,000–5,000 km

Hamburg
Felixstowe Rotterdam

Antwerp

Port
Kelang

Singapore

Fig. 5 | Top-ten bilateral maritime trading partners by shipping connectivity in 2019. UNCTAD’s liner shipping bilateral connectivity index quantifies the
extent to which ports in two countries are connected by maritime trade. The index is based on trade indicators, which include the minimum number of
trans-shipments required to get from country A to B, the number of common direct third-country connections between the country pair, the number of
direct connections, the level of competition of shipping services connecting the country pair, and the size of the largest ship connecting the country pair92.
Connectivity is strongest over short, intraregional routes of less than 5,000 km.

port generates approximately 0.78 g CO2 km−1 (ref. 63). This is a 16% A direct electrification pathway can leverage higher efficiency
reduction from HFO and VLSFO, which produce approximately compared with e-fuels as well as future cost reductions and improve-
0.93 and 0.91 CO2 km−1, respectively. Battery electrification yields an ments in battery technology driven by wide-scale battery deployment
86% reduction over VLSFO in per-kilometre SO2 emissions in the in road transport and stationary storage66. Strategic adjustments to
US but only a 4% reduction in China64. NOx emissions are reduced container shipping logistics could provide a partial solution to the
approximately 83% and 42% over VLSFO for vessels charged at US range challenges facing battery-electric vessels and facilitate the
and Chinese ports, respectively. These findings point to the need electrification of long-distance transoceanic routes. Major mari-
to couple charging infrastructure with collocated renewable energy time chokepoints—such as the Suez Canal, Strait of Gibraltar, Strait
generation to fully capitalize on the emissions reduction potential of of Malacca and Cape of Good Hope—present an opportunity for
battery electrification65. long-range vessels to recharge offshore while queuing for passage.
Breaking the longest voyages into segments could facilitate elec-
Discussion trification of a much larger percentage of global maritime trade.
We show that battery-electric ships powered by renewable electric- Offshore charging in ports and along shipping trade routes could
ity offer a near-term pathway to cut shipping emissions over intra- facilitate collocation of charging stations with renewable generation
regional and inland routes. At battery prices of US$100 kWh−1, the sources, eliminate direct emissions and alleviate range constraints.
TCP of a battery-electric containership is lower than that of an ICE Two-thirds of global ship traffic occurs within 370 km of the shore,
equivalent over routes of less than 1,000 km—without considering where wind potential is highest67,68. Furthermore, the cost of off-
the costs of environmental and health damages. With policy sup- shore wind is expected to decline 37–49% by 2050, beating 2015
port to internalize the environmental costs of HFO and near-future predictions69 by 50%.
battery prices of US$50 kWh−1, routes upwards of 5,000 km can be Electrification provides several benefits over e-fuel alternatives
electrified cost-effectively. Future research should consider how in addition to global availability and cost-competitiveness. For the
opportunities for intermediate recharging affect the overall eco- same power rating, the capital cost and volume of electric motors
nomics of battery electrification. If vessels were able to recharge at are typically smaller than the capital cost and volume of ICEs29,70.
distinct points en route, the battery cost, forfeited TEUs and addi- Hence, retrofitting or hybridizing existing ships with electric
tional energy requirements from battery weight will each decrease, drivetrains during propulsion system overhauls is technically and
potentially making longer-range trips economically feasible. economically viable and could accelerate the electrification of the

670 Nature Energy | VOL 7 | July 2022 | 664–674 | www.nature.com/natureenergy

Nature Energy Articles
a b

ia

a
ab

re
ko
Ar

na
na

h
i
ud

ut

hi

S
hi

K
So
Sa

U
C
20,000

U
C
1.6

1.4 18,000

1.2
Carbon intensity (g CO2 km–1)

SO2 intensity (g SO2 km–1)

16,000

1.0

14,000
0.8

6,000
0.6

4,000
0.4

0.2 2,000

0 0
1,000 800 600 400 200 0 3.0 2.5 2.0 1.5 1.0 0.5 0
Grid-to-wake carbon intensity of electricity (g CO2 kWh–1) Grid-to-wake SO2 intensity of electricity (g SO2 kWh–1)

c
na
hi

S
U
C

24,000
HFO
VLSFO
Battery electric
22,000

20,000
NOx intensity (g NOx km–1)

18,000

16,000

8,000

6,000

4,000

2,000

0
3.0 2.5 2.0 1.5 1.0 0.5 0
Grid-to-wake NOx intensity of electricity (g NOx kWh–1)

Fig. 6 | Comparison of CO2, SO2 and NOx emissions intensities of a battery-electric, HFO and VLSFO small neo-Panamax containerships charged in different
countries. The x-axes describe the well-to-wake intensities of the electric grid supplying the battery-electric vessel. The y-axes represent emissions intensities
of a 7,650 TEU small neo-Panamax ship. The red and orange lines represent HFO- and VLSFO-fuelled containerships, respectively. Blue wedges represent
emissions from a battery-electric ship, which vary by grid emissions intensity. a, We show that CO2 emissions reductions depend on grid carbon intensity. A
battery-electric vessel charged on Saudi Arabia’s carbon-intense grid yields a 46% increase in CO2 emissions over an HFO-fuelled vessel. The cleaner US and
UK grids yield 14–16% and 51–52% reductions in CO2 emissions, respectively, over HFO and VSLFO. b, We demonstrate that a battery-electric vessel charged
in the US would yield 86% and 97% SO2 emissions reductions over VLSFO and HFO, respectively. A battery-electric vessel charged with China’s coal-reliant
grid would yield a 4% SO2 reduction over VLSFO and 77% over HFO. c, For NOx, a battery-electric vessel charged in the US yields 83% and 96% reductions
over VLSFO and HFO, respectively. A vessel charged in China would yield NOx emissions reductions of 42% over VLSFO and 88% over HFO. Emissions
reductions improve rapidly with penetration of renewable energy. A ship charged with 100% renewable energy would eliminate downstream emissions.

global fleet. One advantage of having dual-fuel capabilities is that Our analysis suggests that rapidly improving battery technology
these battery-electric ships could serve as large emergency back-up may enable direct electrification to play a key role in decarbonizing
power plants during increasingly common extreme events leading the shipping industry. Although direct electrification has become
to power supply disruptions. For example, battery-electric ships a technically feasible and cost-effective pathway for zero-emission
modelled in this paper will have 5–10 GWh of storage capacity. In shipping, several challenges need to be addressed for commer-
comparison, the generation deficit that caused the 2020 California cial deployment. The operating costs of battery-electric ships are
blackouts, leaving more than 800,000 customers without power much lower than those of conventional ships, but their upfront
during an extreme heatwave, was less71 than 5 GWh. costs will be much higher primarily due to the cost of the batteries.

Nature Energy | VOL 7 | July 2022 | 664–674 | www.nature.com/natureenergy 671

Articles Nature Energy

Innovative financing and business models are required to address two-stroke ICE fuelled by HFO with an onboard scrubber system for compliance
higher upfront costs. Transmission-connected charging stations with IMO sulfur emissions regulations. Cost drivers for the traditional ship
include HFO costs, which vary by scenario as described above, and operations and
with capacities of hundreds of megawats—similar to large scale grid maintenance costs, including periodic repairs, regular maintenance, and operation
connected storage facilities—will have to be built to support ship of a scrubber system to comply with recent IMO sulfur emissions standards
charging. Given that environmental damages from conventional (estimated at US$5 MWh−1), and excluding other shipping operational expenses
ships are an order of magnitude higher than the propulsion costs of such as labour, insurance, and port charges76. These expenses are developed from
industry benchmarks and academic research77,78.
these ships, policies such as financial incentives for demonstrations
The battery-electric TCP model accounts for the cost of electricity, TEU
and regulations will play a critical role in supporting the transition forfeiture, additional capital costs of the original and replacement of BES systems,
to zero-emissions shipping. operations and maintenance, and the levelized cost of charging infrastructure.
Battery costs are defined as the uniform annual payment for upfront battery capital
Methods costs plus replacement costs over the service lifetime of the ship (25 years)79.
Modelling approach. The energetic requirements of an ICE containership and its LFP batteries are assumed to need replacement after 5,000 cycles or 20 years,
battery-electric analogue depend on the ship size and voyage distance. Past studies whichever comes first39,40. Battery decommissioning costs are neglected based on
have approached this analysis by studying specific real-life ships along their usual the assumption that batteries will have a second life application80. Battery capital
routes28–30. Although this approach has strengths in terms of data availability, an costs are assumed to be additional to ship newbuild capital costs, allowing us to
important limitation is the generalizability of the results to similar ships of different neglect the inclusion of newbuild costs for both battery-electric and ICE ship
route length and carrying capacity. To improve modelling sensitivity to ship size types. Given the relatively low cost of marine engines compared to the total ship
class and route length, we specify eight containership size classes and model newbuild capital cost, this assumption is reasonable and conservative. This study
their energy needs, emissions and economics across 13 major world trade routes, assumes the case of a newbuild only and does not consider retrofit costs, although
ranging from a 911 km voyage from Shanghai, China, to Busan, South Korea, to a the economics of battery electrification through vessel retrofits are an important
20,476 km inter-Atlantic voyage from Shanghai, China, to Santos, Brazil. area of future research.
We define two technoeconomic scenarios. The baseline scenario considers We assume the operations and maintenance cost of the battery-electric
the state of technology in the near future with a volumetric battery energy density vessel is 50% of the ICE equivalent, commensurate with savings on electric
of 470 Wh l−1, battery cost of US$100 kWh−1, HFO cost of US$0.048 kWh−1, vehicles and exclusive of the operating expenses of an onboard scrubber81. An
charging infrastructure utilization of 50% equivalent to US$0.029 kWh−1, and economic penalty or credit, characterized as TEU forfeiture, is included in the
electricity price of US$0.035 kWh−1. The near-future scenario assumes a battery TCP analysis to account for carrying capacity gained or lost based on the volume
cost of US$50 kWh−1, volumetric energy density of 1,200 Wh l−1 and charging requirements of the battery system relative to the ICE ship baseline. The volume
infrastructure utilization of 70%, or a US$0.021 kWh−1 levelized cost of charging differential quantified in TEUs is multiplied by the freight rate for the trade lane,
infrastructure divided by half to account for the inequality in global trade flows that results in
underutilization in carrying capacity for at least one leg of a roundtrip voyage56,82.
Modelling containership technical parameters. The rated energy of the Supplementary Table 2 summarizes the data inputs to the TCP model.
battery must be large enough to supply power for the entirety of each one-way We adapt previous research on electric trains40 and trucking47 to estimate
voyage, assuming the ship can be charged at both its port of origin and its the levelized cost of megawatt-scale charging infrastructure and electricity costs.
destination. Each marine engine is designed with a SMCR, which describes We use an electricity cost of US$0.035 kWh−1 in line with historical real-time
its maximum power output during continuous operation. The average power prices published by the California Independent System Operator (CAISO) for
output is lower than the SMCR, as the engine seldom runs at its maximum 2017 through 201983. This price is inclusive of the cost of generation, compliance
power output even while cruising. The engine load factor describes the ratio with California’s renewable portfolio standards, applicable CAISO fees for a
of the ship’s average power output during normal operations to its SMCR and direct-access customer, demand charges, and applicable delivery charges. The
can be estimated as the cubed ratio of the ship’s average speed to its maximum charging infrastructure cost includes hardware costs, grid connection fees,
design speed. The marine engine manufacturer MAN Diesel Turbo publishes operations and maintenance expenses, and the cost of installation. Supplementary
SMCR and maximum design speed values for containerships based on Holtrop Fig. 1 provides a summary of the components that comprise the total levelized
and Mennen’s power prediction calculation method72. Additional energy charging infrastructure costs.
requirements during manoeuvreing and hoteling, as well as energy savings Where environmental costs are presented, we assume a marginal cost of
through slow steaming practices, are neglected. The voyage time varies NOx and SO2 of US$13,000 t−1 and US$24,000 t−1, respectively84, and a social cost
depending on the route length and the average speed. The average speed is fixed of carbon of US$43 t−1 in line with the US Environmental Protection Agency’s
at 80% of the design speed, which equates to 37 km h−1 (20 knots) for any ship regulatory guidelines85. Notably, this value is about one-third that considered
7,650 TEUs or larger. Auxiliary engine power needs are assumed to be 22% of sufficient to remain below a 1.5 degree Celsius86.
propulsion engine power per port emission inventory best practices73. We use
the Admiralty Law load factor to account for resistance created by additional Environmental impacts. To quantify the potential environmental impacts of
displacement from the weight of the BES system for the battery-electric vessels battery-electric container shipping, we use published tank-to-wake CO2, NOx
per equation (2)38. Design draught, maximum draught, vessel length and vessel and SO2 emissions factors for a slow-speed, two-stroke ICE ship, as described
breadth are taken from MAN Diesel Turbo and used to convert from battery in Supplementary Table 3. Emissions intensities are converted to per-kilometre
weight to change in draught based on the Archimedes principle, which states that intensities by multiplying by the energy consumption in kilowatt hours of a
the weight of the displaced water is equal to the weight of the ship72. We assume battery-electric small neo-Panamax containership87,88.
an ICE tank-to-wake efficiency of 50% and electric motor and inverter efficiencies To estimate the emissions of a battery-electric vessel, we calculate a
of 95% each28. Batteries yield an 80% efficiency improvement compared to their tank-to-wake emissions intensity across a range of real-life grid emission factors
ICE counterparts, which translates to a 30% decrease in total energy needs for the sourced from multiple countries63. To convert from grid intensities to tank-to-wake
battery-electric ship. emissions intensities, we apply 5% transmission and distribution losses, 10% AC/
Daily HFO fuel consumption is derived from an empirical study of DC power conversion losses89, 5% DC/AC conversion losses28 and 5% electric
containership fuel consumption74. We assume a containership carries enough motor efficiency losses28. Calculated tank-to-wake carbon emissions for each
fuel for a day’s voyage; in reality, this figure is probably higher, because ships often country are presented in Supplementary Table 4. We exclude emissions from
carry fuel for several days after bunkering. The mass and volume of BES and battery production owing to the wide variation in estimates, which depend
propulsion systems are the total energetic needs of the battery system (including on where primary materials are extracted and potential end-of-life recycling
efficiency gains from electric propulsion) multiplied by the assumed volumetric opportunities90. To ensure a direct comparison with alternative fuels, we use
or specific energy of the battery, depending on the scenario, with a 0.76 battery tank-to-wake emissions factors rather than well-to-wake emissions factors91.
packing fraction and 80% depth of discharge. We calculate TEU forfeiture by
converting BES system volume in excess of the existing mechanical and fuel storage Data availability
space to standard 2.6 m × 2.4 m × 6.1 m TEUs. The net change in weight used to All data and assumptions necessary to replicate this study’s analysis are included in
correct the power estimates for the battery-electric vessels is the weight of the this published article and its Supplementary information.
battery system and electric propulsion system (assumed to weigh 50% that of the
ICE propulsion system), less the weight of the fuel storage and ICE engine, less
the weight of TEUs forfeited to the battery energy system assuming an average
Code availability
The source code and data underlying the figures presented in this manuscript is
loaded TEU weight of 28.2 t (ref. 75). ICE system weights and volumes are based on
available at https://doi.org/10.5281/zenodo.6594089.
correlations developed by29.

TCP analysis. We quantify economic feasibility through a TCP framework,

Received: 27 August 2021; Accepted: 1 June 2022;
whereby a battery-electric containership is compared to a reference ship with a Published online: 18 July 2022

672 Nature Energy | VOL 7 | July 2022 | 664–674 | www.nature.com/natureenergy

Nature Energy Articles
References 32. Edelstein, S. Report: EV battery costs hit another low in 2021, but they might
1. IPCC Climate Change 2021: The Physical Science Basis. Contribution of rise in 2022. Creen Car Reports (1 December 2021); https://www.
Working Group I to the Sixth Assessment Report of the Intergovernmental Panel greencarreports.com/news/1134307_report-ev-battery-costs-might-rise-in-
on Climate Change (Cambridge Univ. Press, 2021). 2022#:~:text=Lithium%2Dion%20battery%20pack%20prices,average%20
2. 50 Years of Review of Maritime Transport, 1968–2018: Reflecting on the pack%20price%20for%202022
Past, Exploring the Future (United Nations Conference on Trade and 33. Henze, V. Battery pack prices cited below $100/kWh for the first time in
Development, 2018). 2020, while market average sits at $137/kWh. BloombergNEF (16 December
3. Fourth IMO Greenhouse Gas Study (IMO, 2020). 2020); https://about.bnef.com/blog/battery-pack-prices-cited-below-
4. Ytreberg, E., Åström, S. & Fridell, E. Valuating environmental impacts 100-kwh-for-the-first-time-in-2020-while-market-average-sits-at-137-kwh/
from ship emissions—the marine perspective. J. Environ. Manag. 282, 34. Kane, M. VW-related Guoxuan high-tech launches record-setting 210 Wh/kg
111958 (2021). LFP battery cells. Inside EVs (24 January 2021); https://insideevs.com/
5. Comer, B, Olmer, N., Mao, X., Roy, B. & Rutherford, D. Prevalence of Heavy news/481770/guoxuan-210-whkg-lfp-battery-cells/
Fuel Oil and Black Carbon in Arctic Shipping, 2015 to 2025 (International 35. Dedes, E. K, Hudson, D. A. & Turnock, S. R. Technical Feasibility of Hybrid
Council on Clean Transportation, 2017). Propulsion Systems to Reduce Exhaust Emissions of Bulk Carriers (The Royal
6. Cames, M., Graichen, J, Siemons, A. & Cook, V. Emission Reduction Targets Institution of Naval Architects, 2012).
for International Aviation and Shipping (European Parliament, 2015). 36. Bolvashenkov, I., Herzog, H.-G. & Rubinraut, A. Possible ways to improve the
7. Sofiev, M. et al. Cleaner fuels for ships provide public health benefits with efficiency and competitiveness of modern ships with electric propulsion
climate tradeoffs. Nat. Commun. 9, 406 (2018). systems. In IEEE Vehicle Power and Propulsion Conference 1–9 (IEEE, 2014).
8. Initial IMO Strategy on Reduction of GHG Emissions from Ships (IMO, 2018). 37. Moreno-Gutiérrez, J. et al. Methodologies for estimating shipping emissions
9. Sub-Committee on Pollution Prevention and Response (PPR 7) (IMO, 2020); and energy consumption: a comparative analysis of current methods. Energy
http://www.imo.org/en/MediaCentre/MeetingSummaries/PPR/Pages/PPR- 86, 603–616 (2015).
7th-Session.aspx 38. Brown, I. N. & Aldridge, M. F. Power models and average ship parameter
10. Harrison, J. Making the Law of the Sea: A Study in the Development of effects on marine emissions inventories. J. Air Waste Manag. Assoc. 69,
International Law (Cambridge Univ. Press, 2011). 852–763 (2019).
11. MARPOL Annex VI (US Environmental Protection Agency, 2020); 39. Chen, L., Tong, Y. & Dong, Z. Li-Ion Battery performance degradation
https://www.epa.gov/enforcement/marpol-annex-vi-and-act-prevent- modeling for the optimal design and energy management of electrified
pollution-ships-apps#marpol propulsion systems. Energies 13, 1629 (2020).
12. Stolz, B., Held, M., Georges, G. & Boulouchos, K. Techno-economic analysis 40. Popovich, N. D., Phadke, A. A., Rajagopal, D. & Tasar, E. Economic,
of renewable fuels for ships carrying bulk cargo in Europe. Nat. Energy 7, environmental, and grid resilience benefits of converting diesel trains to
203–212 (2022). battery-electric. Nat. Energy 6, 1017–1025 (2021).
13. Maersk to pilot a battery system to improve power production. Maersk 41. EMSA Marine Battery Study: Electrical Energy Storage for Ships
(6 November 2019); https://www.maersk.com/news/articles/2019/11/06/ (European Maritime Safety Agency, 2019).
maersk-to-pilot-a-battery-system-to-improve-power-production 42. Assessment of Battery Technology for Rail Propulsion Application
14. Hockenos, P. Europe takes first steps in electrifying world’s shipping fleets. (US Department of Transportation Federal Railroad Administration, 2017).
YaleEnvironment360 (22 February 2018); https://e360.yale.edu/features/ 43. BatPaC: Battery Manufacturing Cost Estimation (Argonne National
europe-takes-first-steps-in-electrifying-worlds-shipping-fleets Laboratory, 2022); https://www.anl.gov/partnerships/batpac-battery-
15. Brook-Jones, C. All-electric E-pusher type M vessel from Kotug. Electric & manufacturing-cost-estimation
Hybrid Marine Energy International (17 February 2022); https://www. 44. Phadke, A., Aditya, K. & Abhyankar, N. Why Regional and Long-Haul
electrichybridmarinetechnology.com/news/ports-and-harbours/all-electric- Trucks are Primed for Electrification Now (Lawrence Berkeley National
e-pusher-type-m-vessel-from-kotug.html Laboratory, 2021).
16. Cullinane, K. & Cullinane, S. Atmospheric emissions from shipping: the need 45. Park, N. K. & Suh, S. C. Tendency toward mega containerships and the
for regulation and approaches to compliance. Transp. Rev. 33, 377–401 (2013). constraints of container terminals. J. Mar. Sci. Eng. 7, 131 (2019).
17. Cariou, P. Is slow steaming a sustainable means of reducing CO2 emissions 46. Karimi, S., Zadeh, M. & Suul, J. A. Shore charging for plug-in
from container shipping?. Transp. Res. D 16, 260–264 (2011). battery-powered ships: power system architecture, infrastructure, and control.
18. Ammar, N. R. Energy- and cost-efficiency analysis of greenhouse gas IEEE 8, 47–61 (2020).
emission reduction using slow steaming of ships: case study RO–RO cargo 47. Phadke, A., McCall, M. & Rajagopal, D. Reforming electricity rates to enable
vessel. Ships Offshore Struct. 13, 868–876 (2018). economically competitive electric trucking. Environ. Res. Lett. 14, 124047 (2019).
19. Dedes, E. K., Hudson, D. A. & Turnock, S. R. Assessing the potential of 48. Berdichevsky, G. & Yushin, G. The Future of Energy Storage: Towards a Perfect
hybrid energy technology to reduce exhaust emissions from global shipping. Battery with Global Scale (Sila, 2020).
Energy Policy 40, 204–218 (2012). 49. Henze, V. Battery Pack Prices Cited Below $100/kWh for the First Time
20. Mallouppas, G. & Yfantis, E. A. Decarbonization in shipping industry: a in 2020, While Market Average Sits at $137/kWh. BloombergNEF
review of research, technology development, and innovation proposals. (16 December 2020).
J. Mar. Sci. Eng. 2021, 415 (2021). 50. Placke, T., Kloepsch, R., Simon, D. & Winter, M. Lithium ion, lithium metal,
21. Bengtsson, S., Andersson, K. & Fridell, E. A comparative life cycle assessment and alternative rechargeable battery technologies: the odyssey for high energy
of marine fuels: liquefied natural gas and three other fossil fuels. Proc. Inst. density. J. Solid State Electrochem. 21, 1939–1964 (2017).
Mech. Eng. M 225, 97–110 (2011). 51. Schmuch, R., Wagner, R., Hörpel, G., Placke, T. & Winter, M. Performance
22. El-Houjeiri, H., Monfort, J.-C., Bouchard, J. & Przesmitzki, S. Life cycle and cost of materials for lithium-based rechargeable automotive batteries.
assessment of greenhouse gas emissions from marine fuels: a case study of Nat. Energy 3, 267–278 (2018).
Saudi crude oil versus natural gas in different global regions. J. Ind. Ecol. 23, 52. Kian Tan, W., Kawamura, G., Muto, H. & Matsuda, A. in Sustainable
374–388 (2018). Materials for Next Generation Energy Devices Ch. 3, 59–83 (Elsevier, 2021)
23. Navigating to a Renewable Future: Solutions for Decarbonising Shipping 53. UNCTAD’s Review of Maritime Transport 2020: Highlights and Figures on Asia
(International Renewable Energy Agency, 2019). and the Pacific (UNCTAD, 2020); https://unctad.org/press-material/unctads-
24. Schäfer, A. W. et al. Technological, economic and environmental prospects of review-maritime-transport-2020-highlights-and-figures-asia-and-pacific
all-electric aircraft. Nat. Energy 4, 160–166 (2018). 54. Haramlambides, H. E. Gigantism in container shipping, port and global
25. Howarth, R. W. & Jacobson, M. Z. How green is blue hydrogen? Energy Sci. logistics: a time-lapse into the future. Marit. Econ. Logist. 21, 1–60 (2019).
Eng. 9, 1676–1687 (2021). 55. About the Industry: Container Ship Design (World Shipping Council, 2020);
26. Tan, W., Bhavnagri, K. & Chatterton, R. Hydrogen: the economics of http://www.worldshipping.org/about-the-industry/liner-ships/container-
powering ships. BloombergNEF (27 March 2020). ship-design
27. Ueckerdt, F. et al. Potential and risks of hydrogen-based e-fuels in climate 56. Review of Maritime Transport (United Nations Conference on Trade and
change mitigation. Nat. Clim. Change 11, 384–393 (2021). Development, 2019).
28. Comer, B. Transitioning Away from Heavy Fuel oil in Arctic Shipping 57. Shipping Market Review (Danish Ship Finance, 2021).
(The International Council on Clean Transportation, 2019). 58. Olmer, N., Comer, B., Roy, B., Mao, X. & Rutherford, D. Greenhouse Gas
29. Minnehan, J. J. & Pratt, J. W. Practical Application Limits of Fuel Cells and Emissions from Global Shipping, 2013–2015 (International Council on Clean
Batteries for Zero Emission Vessels (Sandia National Laboratories, 2017). Transportation, 2017).
30. Zero-emission vessels 2030: how do we get there? Lloyd’s Register Insights 59. Stopford, M. Maritime Economics 3rd edn (Routledge, 2009).
(5 February 2018). 60. Janek, J. & Zeier, W. G. A solid future for battery development. Nat. Energy 1,
31. Wu, P. & Bucknall, R. On the design of plug-in hybrid fuel cell and lithium 16141 (2016).
battery propulsion systems for coastal ships. In 13th International Marine 61. McArthur, D. P. & Osland, L. Ships in a city harbour: an economic valuation
Design Conference (Univ. College London, 2018). of atmospheric emissiosn. Transportation Res. D 21, 47–52 (2013).

Nature Energy | VOL 7 | July 2022 | 664–674 | www.nature.com/natureenergy 673

Articles Nature Energy
62. Zhang, Q., Wan, Z., Hemmings, B. & Abbasov, F. Reducing black carbon 86. Coninck, d. H. et al. Strengthening and Implementing the Global Response
emissions from Arctic shipping: solutions and policy implications. J. Clean. (IPCC, 2018).
Prod. 241, 118261 (2019). 87. Comer, B. & Osipova, L. Counting for Well-to-Wake Carbon Dioxide
63. Brown to Green: The G20 Transition to a Low-Carbon Economy Equivalent Emissions in Maritime Transportation Climate Policies
(Climate Transparency, 2018). (International Council on Clean Transport, 2021).
64. de Chalendar, J. A., Taggart, J. & Benson, S. M. Tracking emissions in the US 88. Lindstad, H., Eskeland, G. S., Psaraftis, H. N., Sandaas, I. & Strømmane, A.
electricity system. Proc. Natl Acad. Sci. USA 116, 25497–25502 (2017). H. Maritime shipping and emissions: a three-layered, damage-based
65. Ammonia: Zero-Carbon Fertiliser, Fuel and Energy Store Policy Briefing approach. Ocean Eng. 110, 94–101 (2015).
(The Royal Society, 2020). 89. How Much Electricity is Lost in Electricity Transmission and Distribution in the
66. Muratori, M. et al. The rise of electric vehicles—2020 status and future. United States? (US Energy Information Administration, 2021); https://www.
Prog. Energy 3, 022002 (2021). eia.gov/tools/faqs/faq.php?id=105&t=3
67. Adelaja, A., McKeown, C., Calnin, B. & Hailu, Y. Assessing offshore wind 90. Gaines, L., Sullivan, J. & Burnham, A. Life-Cycle Analysis for Lithium-Ion
potential. Energy Pol. 42, 191–200 (2012). Battery Production and Recycling (Argonne National Laboratory, 2011).
68. Weng, C. & Corbett, J. J. Geographical characterization of ship traffic and 91. Third IMO GHG Study 2014: Executive Summary and Final Report
emission. J. Transp. Res. Board https://doi.org/10.1177/0361198105190900113 (IMO, 2015).
(2005). 92. Fugazza, M. Bilateral maritime connectivity since 2006: a primer using new
69. Wiser, R. et al. Expert elicitation survey predicts 37% to 49% declines in wind liner shipping bilateral connectivity index (LSBCI) calculations. UNCTAD
energy costs by 2050. Nat. Energy 6, 555–565 (2021). Transport and Trade Facilitation Newsletter (4 December 2019).
70. Nguyen, H. P. et al. The electric propulsion system as a green solution for
management strategy of CO2 emission in ocean shipping: a comprehensive
review. Int. Trans. Electrical Energy Sys. 31, e12580 (2020).
Acknowledgements
We thank J. Zuboy for his assistance in preparing the manuscript. We also acknowledge
71. Root Cause Analysis: Mid-August 2020 Extreme Heat Wave (California
the insights from our counterparts at the US Department of Transportation Maritime
Independent System Operator, California Public Utilities Commission,
Administration. N.P. and A.P. received funding for this work from the Hewlett
California Energy Commission, 2021).
Foundation under grant no. 2019–9467.
72. Propulsion Trends in Container Vessels (MAN Diesel & Turbo, 2019).
73. U.S. Environmental Protection Agency: Current Methodologies in
Preparing Mobile Source Port-Related Emission Inventories–Final Report Author contributions
(US Environmental Protection Agency, 2009). J.K. developed the model, conducted the analysis, and wrote the draft manuscript. A.P.
74. Czermanski, E., Cirella, G. T., Oniszczuk-Jastrzabek, A., Pawlowska, B. & conceived the project, secured funding and reviewed to the manuscript. N.P. provided
Notteboom, T. An energy consumption approach to estimate air emission input data for charging infrastructure and wrote portions of the manuscript.
reductions in container shipping. Energies 14, 278 (2021).
75. Menon, H. TEU in Shipping—Everything you Wanted to Know
(Marine Insight, 2021).
Competing interests
The authors declare no competing interests.
76. Costs and Benefits of LNG as Ship Fuel for Container Vessels (MAN Diesel &
Turbo, 2012).
77. Ship Operating Costs Annual Review and Forecast 2017/2018 (Drewery, 2017). Additional information
78. Apostolidis, A., Merikas, A. & Merika, A. The modelling of maintenance cost: Supplementary information The online version contains supplementary material
the case of container-ships in dry-dock. J. Comput. Optim. Econ. Financ. 5, available at https://doi.org/10.1038/s41560-022-01065-y.
27–39 (2013).
79. Dinu, O. & Ilie, A. M. Maritime vessel obsolescence, life cycle cost and design Correspondence and requests for materials should be addressed to Amol A. Phadke.
service life. In IOP Conference Series: Materials Science and Engineering Vol. Peer review information Nature Energy thanks Stephen R Turnock and
95 (IOP, 2015). the other, anonymous, reviewer(s) for their contribution to the peer review
80. Ioakimidis, C. S., Murillo-Marrodán, A., Bagheri, A., Thomas, D. & of this work.
Genikomsakis, K. N. Life cycle assessment of a lithium iron phosphate (LFP) Reprints and permissions information is available at www.nature.com/reprints.
electric vehicle battery in second life application scenario. Sustainability 11,
2527 (2019). Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in
81. Harto, C. Electric vehicle ownership costs: today’s electric vehicles offer big published maps and institutional affiliations.
savings for consumers. Consumer Reports (October 8 2020). Open Access This article is licensed under a Creative Commons
82. Baik, J. S. The study on impacts of mega container ships on ports. Pan-Pac. J. Attribution 4.0 International License, which permits use, sharing, adap-
Supply Chain Manag. Appl. Pract. 1, 22–40 (2017). tation, distribution and reproduction in any medium or format, as long
83. CAISO Real-time Price (LCG Consulting, 2022); http://www.energyonline. as you give appropriate credit to the original author(s) and the source, provide a link to
com/Data/GenericData.aspx?DataId=19&CAISO___Real-time_Price the Creative Commons license, and indicate if changes were made. The images or other
84. Goodkind, A. L., Tessum, C. W., Coggins, J. S., Hill, J. D. & Marshall, J. D. third party material in this article are included in the article’s Creative Commons license,
Fine-scale damage estimates of particulate matter air pollution reveal unless indicated otherwise in a credit line to the material. If material is not included in
opportunities for location-specific mitigation of emissions. Proc. Natl Acad. the article’s Creative Commons license and your intended use is not permitted by statu-
Sci. USA 116, 8775–8780 (2019). tory regulation or exceeds the permitted use, you will need to obtain permission directly
85. The Social Cost of Carbon (US Environmental Protection Agency, 2021); from the copyright holder. To view a copy of this license, visit http://creativecommons.
https://19january2017snapshot.epa.gov/climatechange/social-cost-carbon_. org/licenses/by/4.0/.
html#main-content © The Author(s) 2022

674 Nature Energy | VOL 7 | July 2022 | 664–674 | www.nature.com/natureenergy