i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 4 ( 2 0 0 9 ) 6 2 6 5 – 6 2 7 0

Avai lab le a t www.sc iencedi rec t .com

j ourna l homepage : www.e lsev ier . com/ loca te /he

Comments on solid state hydrogen storage systems designfor fuel cell vehicles

David Wengera,*, Wolfgang Polifkec, Eberhard Schmidt-Ihna, Tarek Abdel-Basetb,Steffen Mausa

aDaimler AG, Kirchheim/Teck-Nabern, GermanybChrysler LLC, Auburn Hills, MI, USAcTU Munchen, Lehrstuhl fur Thermodynamik, Garching, Germany

a r t i c l e i n f o

Article history:

Received 15 January 2008

Received in revised form

24 February 2009

Accepted 19 May 2009

Available online 3 July 2009

Keywords:

Fuel cell vehicle

Hydrogen storage

Metal hydride

Requirements

* Corresponding author at: Present address: Wfax: þ49 731142354.

E-mail address: david.wenger@wenger-en0360-3199/$ – see front matter ª 2009 Interndoi:10.1016/j.ijhydene.2009.05.072

a b s t r a c t

In recent years, significant research and development efforts were spent on hydrogen

storage technologies with the goal of realizing a breakthrough for fuel cell vehicle appli-

cations. This article scrutinizes design targets and material screening criteria for solid state

hydrogen storage. Adopting an automotive engineering point of view, four important, but

often neglected, issues are discussed: 1) volumetric storage capacity, 2) heat transfer for

desorption, 3) recharging at low temperatures and 4) cold start of the vehicle. The article

shall help to understand the requirements and support the research community when

screening new materials.

ª 2009 International Association for Hydrogen Energy. Published by Elsevier Ltd. All rights

reserved.

1. Introduction Nevertheless considerable challenges in terms of technology,

Hydrogen powered fuel cell vehicles represent one option to

face the challenges of increasing individual mobility demands

and decreasing oil reserves. While the protection of the

environment and climate is the main reason for the devel-

opment of this technology in Europe and Japan [1,2], the

reduction of dependence on oil imports from politically

unstable regions is the main goal in the US [3]. Furthermore,

with such technologies, the car manufacturers also expected

to develop competitive advantages by demonstrating

sustainability, innovation and technology leadership. Signifi-

cant progress has been achieved during the last 15 years.

enger Engineering GmbH

gineering.de (D. Wengerational Association for H

infrastructure development and cost reduction have to be

overcome before fuel cell vehicles will evolve from demon-

stration objects to a real business case.

Adopting a vehicle engineering point of view, the present

paper discusses four important requirements on solid state

hydrogen storage systems for fuel cell vehicle applications:

volumetric vs. gravimetric storage capacity, refueling time at

low temperatures, heat transfer for desorption and cold start

capability. It will be seen that adopting an automotive engi-

neering perspective leads to non-trivial, sometimes surprising

conclusions regarding the formulation of design targets for

solid state hydrogen storage system development.

, Schillerstraße 18, D-89077 Ulm, Germany. Tel.: þ49 163-8641621;

).ydrogen Energy. Published by Elsevier Ltd. All rights reserved.

Table 1 – Requirements of a fuel cell vehicle on hydrogenstorage systems [13,14].

Range 600 km

Usable hydrogen mass 6–10 kg

Hydrogen delivery rate 0–2 g/s

Hydrogen delivery pressure approx. 0.8 MPa

Operating temperature �40 to 85�C

Refueling time �3 min (98% capacity)

Cold start Like diesel engine

Power [kW]

0 50 100 150 200 250 300 350 400

Compact FC Sedan FC SUV FCCompact Gasoline Sedan GasolineCompact Diesel Sedan Diesel

SUV GasolineSUV Diesel

Sedan H2 ICE

6

5

4

3

2

1

0

Con

sum

ptio

n [M

J/km

]Fig. 2 – Specific energy consumption of different vehicles

[4,5].

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 4 ( 2 0 0 9 ) 6 2 6 5 – 6 2 7 06266

2. Requirements

Table 1 lists some important requirements on hydrogen

storage systems imposed by a typical 100 kW fuel cell vehicle.

From a user’s perspective, vehicle range is perhaps the most

obvious design objective. The data in Fig. 1 indicate clearly

that the range of current fuel cell vehicles is unsatisfactory,

due to the difficulties of onboard hydrogen storage.

The mass of fuel required for a certain range can be derived

directly from the specific energy consumption of the vehicle,

see Fig. 2. Most fuel cell vehicles have a specific consumption

in the order of 1–1.2 MJ/km, which corresponds to roughly 1 kg

of hydrogen per 100 km.

The weight and volume of different vehicular hydrogen

storage systems are shown in Fig. 3. Apparently, high pressure

systems have by far the best gravimetric capacity, a factor of

crucial importance in vehicle engineering. Hence high pres-

sure storage was used in more than 90% of the fuel cell vehi-

cles introduced in the last years [6]. Although 70 MPa high

pressure storage can be regarded as the state-of-the-art in

hydrogen storage, the volumetric capacity of this technology

is still significantly below design targets stated, e.g. by DoE

[11]. This shortcoming, combined with restrictions in tank

design, is the main reason for the low range of current

hydrogen-fueled vehicles.

Solid state hydrogen storage is a technology that has

demonstrated significantly higher volumetric capacity, see

1400

1200

1000

800

600

400

200

0

Ran

ge [

km]

Power [kW]

Compact FC Sedan FC SUV FCCompact Gasoline Sedan GasolineCompact Diesel Sedan Diesel

SUV GasolineSUV Diesel

0 50 100 150 200 250 300 350 400

Fig. 1 – Range as a function of installed power in current

fuel cell-, gasoline- and diesel-powered vehicles [4,5].

again Fig. 3. The identification of storage materials with higher

capacity is an active area of ongoing research. It is the objec-

tive of the present paper to identify proper design objectives,

which should guide the further development of solid state

hydrogen storage materials suitable for fuel cell vehicle

applications.

2.1. Gravimetric vs. volumetric storage capacity

A disadvantage of metal hydride storage systems is in general

the high weight due to the low gravimetric storage capacity of

thermodynamically suitable materials. Recently, many

classes of materials were found to be technically reversible at

moderate pressures, e.g. alanates [15], amide/hydride-

systems [16] and borohydrides [17] (see Table 2). Also, kineti-

cally stabilized materials that are supposed to be regenerated

5

4

3

2

1

0

Grav. Storage Density [kgH2 / 100kg System]

Vol

. Sto

rage

Den

sity

[kgH

2 / 1

00l S

yste

m]

0 1 2 3 4 5 6 7

DoE Goal2010

35 MPa70 MPaLiquid HydrogenMetal Hydride

Fig. 3 – Volumetric and gravimetric storage density of

different automotive hydrogen storage systems

[7–9,10–12].

Table 2 – Properties of hydrogen storage materials investigated previously. A more exhaustive overview on the state-of-the-art is given in Ref. [30].

No. Reaction cmax [wt-%] Drh [kJ/mol] Source

(1) LiBH4þ 1/2MgH2 / LiHþ 1/2MgB2þ 2H2 11.46 41 [22]

(2) LiNH2þ LiH / Li2NHþH2 6.47 45 [16]

(3) Na2LiAlH6 / 2NaHþ LiHþAlþ 3/2H2 3.49 56 [23]

(4) Mg(NH2)2þ 2LiH / Li2MgN2H2þ 2H2 5.54 44 [24]

(5) 1/3Na3AlH6þ 2/3Al / NaHþAlþ 1/2H2 1.85 47 [15]

(6) NaAlH4 / 1/3Na3AlH6þ 2/3AlþH2 3.7 37 [15]

(7) LaNi5H6 / LaNi5þ 3H2 1.38 30 [25]

(8) Ti0.98Zr0.02V0.43Fe0.1Cr0.05Mn1.5H3 / Ti0.98Zr0.02V0.43Fe0.1Cr0.05Mn1.5þ 3/2H2 1.8 22–29 [26]

(9) Ti0.98Zr0.02V0.43Fe0.1Cr0.05Mn1.5þ 3/2H2 / Ti0.98Zr0.02V0.43Fe0.1Cr0.05Mn1.5H3 1.8 22–29 [27]

(10) TiMn1.5V0.45Fe0.1þ 3/2H2 / TiMn1.5V0.45Fe0.1H3 1.5 28 [28]

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 4 ( 2 0 0 9 ) 6 2 6 5 – 6 2 7 0 6267

off-board were investigated intensively [18]. In terms of

gravimetric storage capacity, these materials are superior to

the low temperature metal hydrides. However, the increase in

gravimetric density was mainly achieved by using lighter

elements in combination with Ti-containing precursors. It can

be concluded that for a given hydrogen mass (derived from

a desired range) the material volume and therefore also the

necessary mass for the encasing pressure vessel may increase

and lead to a storage system that is heavier and larger than

current low temperature hydride systems.

Let us point out the consequence for systems engineering

in a short parameter study. For a given internal vessel volume,

the stored hydrogen mass is calculated as a function of typical

material properties like density, porosity and storage capacity.

As the required vessel and heat exchanger properties also

depend on the material properties, this approach implies

a certain inaccuracy, but nevertheless it provides a powerful

method with which to screen the potential of materials for

systems applications. The input values and resulting

hydrogen masses are given in Tables 3 and 4.

The results are somehow surprising: the volumetric

storage capacity of the system decreases despite the

increasing gravimetric capacity of the material. Even the net

gravimetric system capacity may decrease if the necessary

vessel becomes significantly larger and thus heavier. In

consequence this means that it is not enough to increase the

gravimetric storage capacity, the improvement also has to be

larger than the corresponding decrease in volumetric capacity

in order to realize an overall advantage (Fig. 3). It is therefore

suggested to place more emphasis on volumetric storage

density when screening new materials.

2.2. Transfer of heat of desorption

PEM Fuel Cells have an operating temperature of about

70–80�C, which implies that heat for desorption is available at

Table 3 – Input and resulting values of the parameterstudy (see Table 2 for the materials considered).

(8) (5)þ (6) (2) (1)

Typical material density [g/cm3] 6.2 1.25 1.26 0.66

Typical porosity [%] 50 50 50 50

Typical storage capacity [wt-%] 1.8 5.55 6.5 11.5

Stored hydrogen mass [10�2 g/cm3] 5.6 3.4 4.1 3.8

that temperature or lower. The necessary heat flow rate into

the hydrogen storage medium can be calculated from the

required hydrogen flow rate and the specific reaction enthalpy

of the storage material. A 100 kW fuel cell with an efficiency of

about 40% will require 2 g of hydrogen per second from the

storage system at full load. A typical low temperature metal

hydride with a reaction enthalpy of 20–30 kJ/mol will therefore

require a heat transfer rate of 20–30 kW at full power.

Fig. 4 illustrates the fraction of total fuel cell waste heat

required by the hydrogen storage medium versus fuel cell

power. The fraction is particularly high at low power, where

the fuel cell exhibits higher efficiency, and reaches

a maximum value of almost 50%! The transfer of such a large

amount of heat requires a large heat transfer coefficient

(difficult to realize technically), a large heat transfer area (high

weight) or a large temperature difference between the heat

transfer fluid and the material itself (difficult to realize due to

thermodynamic constraints).

The optimal solution lies in finding a material that desorbs

hydrogen at the lowest temperature possible with high pres-

sure equipment and sealing technologies to allow for a high

temperature difference. Selecting a storage material with the

appropriate temperature and pressure working range requires

an understanding of the operating envelope defined by the

vehicle operating conditions.

Current 70 MPa systems have a specified minimum

working temperature limit of�40�C. A typical fuel cell vehicle

requires a minimum input pressure of about 0.8 MPa in order

to overcome the pressure drop in pipes and valves. Thus, the

minimum condition for a thermodynamically ideal material is

set to an equilibrium pressure of 0.8 MPa at �40�C.

Conversely, the upper limit is given by the maximum

temperature and working pressure of the vessel. The

maximum temperature of the system is set at 80�C as defined

by the continuous coolant temperature. The pressure can be

arbitrarily chosen and must at least match the equilibrium

Table 4 – Input and resulting values of the parameterstudy (see Table 2 for the materials considered).

Typical value of . (8) (5)þ (6) (2) (1)

Material density [g/cm3] 6.2 1.25 1.26 0.66

Porosity [%] 50 50 50 50

Gravimetric capacity [wt-%] 1.8 5.55 6.5 11.5

Volumetric capacity [10�2 g/cm3] 5.6 3.4 4.1 3.8

50

40

30

20

10

0Hea

t of

Des

orpt

ion

/ Tot

al W

aste

Hea

t [%

]

Fuel Cell Power [kW]0 20 40 60 80 100

Δrh = 30 kJ/mol

Δrh = 25 kJ/mol

Δrh = 20 kJ/mol

Fig. 4 – Fraction of fuel cell waste heat needed for

desorption of hydrogen from metal hydride.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 4 ( 2 0 0 9 ) 6 2 6 5 – 6 2 7 06268

pressure at that temperature. As 70 MPa-systems are avail-

able, even metal hydride systems with this working pressure

would be conceivable. Although the advantages of high pres-

sure metal hydride tanks have been known for many years

[19], such storage systems have only been presented recently

[20,7]. These concepts are very interesting from an engi-

neering point of view. However, apart from high cost of such

systems the issue remains how to safely avoid the over-

heating during fast filling [21].

Fig. 5 shows various well-investigated hydrides relevant for

research and application. The figure highlights that only the

low temperature metal hydrides are compatible with the

thermodynamic constraints imposed by the fuel cell system.

Identifying a material that has higher storage capacity at these

conditions must remain a focus of further research activities.

2.3. Refueling time at low temperatures

One important requirement is the refueling time of the

storage system. Although some authors consider a refueling

Inverse Temperature [10-3/K]

1.5 2 2.5 3 3.5 4

Equ

ilibr

ium

Pre

ssur

e [M

Pa]

101

100

10-1

10-2

(2)

(3)

(4)

(1)

(10)

(9)

(8)(7)

(6)

(5)

T=85°C T=-40°C

Fig. 5 – Overview of different hydrides relevant for research

and application. The legend is given in Table 2.

time of 10 min acceptable, the comparison with gasoline

vehicles indicates that 3 min or less must be achieved under

all ambient temperatures. Current 70 MPa systems with 4 kg

of hydrogen fulfill this requirement [14] and set the bench-

mark for metal hydride systems. Thermodynamically,

absorption is possible with all the materials depicted in Fig. 5

and a charging pressure of 5 MPa. In practice, most of the

‘‘new’’ materials react much slower than desired, especially

when the temperature of the hydride is low. Fig. 6 shows

absorption experiments on a sample of 25 g of sodium alanate

with a Ti-precursor prepared at Forschungszentrum Karls-

ruhe. Obviously, the absorption time strongly depends on

temperature and pressure. Although all experiments reached

a final capacity of more than 4 wt-% (not shown here), the

reaction was very slow at temperatures below 80�C. The

reason is that for the reaction (Eq. (1)), not only an activation

barrier has to be overcome, but also a solid–solid reaction

must take place. This in turn requires metal diffusion which is

slow at low temperatures [29]. This metal to metal interaction

may hinder the fast reaction kinetics required to satisfy the

fast fill requirements under vehicle conditions with the

material classes (1)–(6) in Fig. 5. In this case, off-board loading

with additional heat supply would be necessary.

NaHþAlþ 32H2/

13

Na3AlH6 þ23

AlþH2/NaAlH4 (1)

2.4. Cold start

One major drawback of the first generation of fuel cell vehicles

was their limitation to temperatures above 0�C. Second

generation vehicles already allow for starting at temperatures

down to �20�C. With a high pressure tank hydrogen is easily

supplied even at such conditions, whereas a metal hydride

tank has fundamental difficulties to provide sufficient

hydrogen.

There are several aspects to consider when starting a fuel

cell system at temperatures below 0�C, but one rule always

holds true: the fuel cell must heat up as fast as possible to an

2

1.5

1

0.5

0

Hyd

roge

n C

once

ntra

tion

[w

t-%

]

0 60 120 180

Time [s]

60°C / 2.5 MPa 80°C / 2.5 MPa 100°C / 2.5 MPa40°C / 5 MPa100°C / 5 MPa 120°C / 5 MPa

80°C / 5 MPa60°C / 5 MPa

Fig. 6 – Absorption experiments with NaAlH4.

90

80

70

60

50

40

30

20

10

0

Coo

lant

Tem

pera

ture

[°C

]

Time after vehicle start [min]0 20 40 60 80 100 120 140

Fig. 7 – Heating curves of a Mercedes-Benz F-Cell fuel cell

system during daily operation (own experiments).

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 4 ( 2 0 0 9 ) 6 2 6 5 – 6 2 7 0 6269

uncritical temperature, e.g. 50�C, to ensure good performance,

avoid degradation and remove liquid water. Therefore the

whole fuel cell waste heat must be used exclusively for that

purpose.

Fig. 7 shows the coolant temperature of a Mercedes-Benz

F-Cell vs. time during daily operation. The properties of the

fuel cell system are given elsewhere [9]. In the experiment,

both parameters were logged when the driver started the

vehicle. Most of the drive cycles are rather short. The

temperature curve strongly depends on the initial tempera-

ture of the system, the ambient temperature and the driving

cycle. As these experiments were performed in summer, the

lowest temperature recorded is about 10�C. However, the time

gap between vehicle start and availability of waste heat at the

desired temperature (e.g. 70 �C, depending on the design of the

heat exchanger of the metal hydride system) has to be

bridged. At relatively high initial temperatures this appears to

be feasible by taking advantage of the sensible heat of the

material itself and the support of the battery system, but at

low initial temperatures this might prove very challenging. It

must thus be concluded that the cold start is a crucial issue for

a combined automotive fuel cell/metal hydride system,

especially if one considers that at low power, where the fuel

cell has its best efficiency, almost half of the waste heat is

needed for desorption (Fig. 4).

3. Conclusion

Adopting an automotive engineering point of view, four

important aspects of solid state hydrogen storage systems for

fuel cell vehicle applications have been discussed. It was

shown that there are certain issues of this storage technology

which may hinder the widespread application in conventional

fuel cell vehicles. Overall system mass is strongly dependent

on both material gravimetric and volumetric capacity. Cold

start and fueling time also represent significant engineering

and thermodynamic challenges, which must be resolved.

Further research on storage materials and systems

engineering should take the findings discussed in this paper

into account in order to adequately explore the limitations

and capabilities of this storage technology.

Acknowledgements

The authors would like to thank the working group of Dr. M.

Fichtner at Forschungszentrum Karlsruhe, Germany, for

providing the material used for the experiment cited in this

paper.

r e f e r e n c e s

[1] Commission of the European Communities, editor. Anenergy policy for Europe. Brussels; 2007.

[2] Agency for Natural Resources and Energy, Ministry ofEconomy, Trade and Industry, editors. Energy in Japan 2006.Tokyo; 2006.

[3] U.S. Department of Energy, editor. National energy policy.Washington DC; 2001.

[4] U.S. Department of Energy, editor. Fuel cell vehicle worldsurvey 2003. Washington DC; 2004.

[5] Robert Bosch GmbH, editor. KraftfahrtechnischesTaschenbuch. 26th ed. Wiesbaden: Vieweg; 2007.

[6] L-B-Systemtechnik GmbH, www.h2cars.de; 2007.[7] Young RC. Advances in solid hydrogen storage systems. In:

14th annual U.S. hydrogen conference, NHA, WashingtonDC; 2003.

[8] Maus S. Modellierung und Simulation der Betankung vonFahrzeugbehaltern mit komprimiertem Wasserstoff,Fortschritt-Berichte VDI Reihe 3 Nr. 879. Dusseldorf: VDIVerlag; 2007.

[9] Wuchner E. Wasserstoff – Kraftstoff furBrennstoffzellenfahrzeuge. In: 20% Erneuerbare Energien –Moderne Speichertechnologien als Voraussetzung? Berlin:Forum fur Zukunftsenergien e.V; 2007.

[10] Mori D, Haraikawa N, Kobayashi N, Kubo H, Toh K, TsuzukiM, et al. High-pressure metal hydride tank for fuel cellvehicles, in: Materials Research Society Symposium, vol.884E; 2005.

[11] U.S. Department of Energy, editor. Hydrogen, fuel cells &infrastructure technologies program: multi-year research,development and demonstration plan. Washington DC; 2005.

[12] TUV Rheinland e.V. In: Project Coordinator Motor Vehiclesand Road Transport, editor. Alternative energy sources forroad transport – hydrogen drive test. Verlag TUV Rheinland,Koln; 1990.

[13] Barral K, Perrin J. StorHy onboard storage targets. In: IPHEinternational conference on hydrogen storage. Lucca, Italy:IPHE; 2005.

[14] Hyundai Motor Company. Compressed hydrogen fuelingstation & 70 MPa fueling process specification – final draftversion 5.0. In: DaimlerChrysler AG, General MotorsCorporation, Ford Motor Company, Nissan Motor Company,Honda Motor Company, Toyota Motor Company,Volkswagen AG, editors. Detroit: NextEnergy, www.nextenergy.org/industryservices/70MPa_Specification_Docs.asp; 2007.

[15] Bogdanovic B, Schwickardi M. Ti-doped alkali metalaluminium hydrides as potential novel reversible hydrogenstorage materials. Journal of Alloys and Compounds 1997;253–254:1–9.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 4 ( 2 0 0 9 ) 6 2 6 5 – 6 2 7 06270

[16] Chen P, Xiong Z, Luo J, Lin J, Tan KL. Interaction of hydrogenwith metal nitrides and imides. Nature 2002;420:302–4.

[17] Zuttel A, Wenger P, Rentsch S, Sudan P, Mauron P,Emmenegger C. LiBH4 a new hydrogen storage material.Journal of Power Sources 2003;118:1–7.

[18] Graetz J, Reilly J. Kinetically stabilized hydrogen storagematerials. Scripta Materialia 2007;56:835–9.

[19] Bernauer O. Wasserstoffspeicher, patent DE 3514500-C1;1986.

[20] Mori D, Haraikawa N, Kobayashi N, Shinozawa T,Matsunaga T, Kubo H, et al. High-pressure metal hydridetank for fuel cell vehicles. In: IPHE international conferenceon hydrogen storage. Lucca, Italy: IPHE; 2005.

[21] Toh K, Kubo H, Isogai Y, Mori D, Hirose K, Kobayashi N.Thermal analysis of high-pressure metal hydride tank forautomotive application. In: Materials Research SocietySymposium, vol. 927. San Francisco, CA: MRS; 2006.

[22] Vajo JJ, Skeith SL, Mertens F. Reversible storage of hydrogenin destabilized LiBH4. Journal of Physical Chemistry B 2005;109(9):3719–22.

[23] Fossdal A, Brinks HW, Fonnelop JE, Hauback BC. Pressure–composition isotherms and thermodynamic properties ofTiF3-enhanced Na(2)LiA1H(6). Journal of Alloys andCompounds 2005;397(1–2):135–9.

[24] Xiong ZT, Hu JJ, Wu GT, Chen P, Luo WF, Gross K, et al.Thermodynamic and kinetic investigations of the hydrogenstorage in the Li-Mg-N-H system. Journal of Alloys andCompounds 2005;398(1–2):235–9.

[25] Mendelsohn M, Gruen D, Austin D. LaNi5�xAlx is a versatilealloy system for metal hydride applications. Nature 1977;269:45–77.

[26] Bernauer O, Topler J, Noreus D, Hempelmann R, Richter D.Fundamentals and properties of some Ti/Mn based lavesphase hydrides. International Journal of Hydrogen Energy1989;14(3):187–200.

[27] Fischer T. Simulation von Metallhydridspeichern,Dissertation, Universitat Dortmund, AbteilungChemietechnik; 1987.

[28] Roehl F. Entwicklung, experimentelle Erprobung undSimulation thermochemischer Wasserstoffkompressorenauf der Basis von Metallhydriden, Dissertation, TU Hamburg-Harburg, Arbeitsbereich VerfahrenstechnischerApparatebau; 2001.

[29] Bogdanovic B, Eberle U, Felderhoff M, Schuth F. Complexaluminium hydrides. Scripta Materialia 2007;56:813–6.

[30] Sakintuna B, Lamari-Darkrim F, Hirscher M. Metal hydridematerials for solid hydrogen storage: a review. InternationalJournal of Hydrogen Energy 2007;32:1121–40.