Dehydrogenation properties of doped LiAlH4 compacts for hydrogen generator applications

Dehydrogenation properties of doped LiAlH4 compacts for hydrogen generator applications

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 9 ( 2 0 1 4 ) 1 6 3 6 2 e1 6 3 7 1 Available online at www.sciencedirect.co...

1MB Sizes 1 Downloads 11 Views

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 9 ( 2 0 1 4 ) 1 6 3 6 2 e1 6 3 7 1

Available online at www.sciencedirect.com

ScienceDirect journal homepage: www.elsevier.com/locate/he

Dehydrogenation properties of doped LiAlH4 compacts for hydrogen generator applications €ntzsch b,* Jie Fu a, Marcus Tegel b, Bernd Kieback a,b, Lars Ro €t Dresden, Institute for Materials Science, Helmholtzstraße 7, 01069 Dresden, Germany Technische Universita Fraunhofer Institute for Manufacturing Technology and Advanced Materials (IFAM), Branch Lab Dresden, Winterbergstraße 28, 01277 Dresden, Germany

a

b

article info

abstract

Article history:

Lithium aluminum hydride (LiAlH4) is an attractive hydrogen source for fuel cell systems

Received 5 May 2014

due to its high hydrogen storage capacity and the moderate dehydrogenation conditions. In

Received in revised form

this contribution, TiCl3- and ZrCl4-doped LiAlH4 powders are prepared and pelletized under

1 August 2014

different compaction pressures in a uniaxial press. At constant 80  C and a hydrogen

Accepted 10 August 2014

partial pressure of 0.1 MPa, the maximal hydrogen release of suchlike LiAlH4 compacts

Available online 2 September 2014

amounts to 6.64 wt.%-H2 (gravimetric capacity) and 53.88 g-H2 l1 (volumetric capacity).

Keywords:

under variation of the compaction pressure, temperature and hydrogen partial pressure.

Hydrogen storage material

Furthermore, the volume change of doped LiAlH4 compacts during dehydrogenation as

Lithium aluminum hydride

well as their short-term storability are investigated (shelf life).

Transition metal chloride doping

Copyright © 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights

The hydrogen release properties of the doped LiAlH4 compacts are studied systematically

reserved.

Hydrogen generation Compaction Shelf life

Introduction Hydrogen is the preferred fuel for most currently developed fuel cells. However, the efficient, compact and safe hydrogen storage is often a problem for fuel cell systems. Among the intensively investigated hydrogen storage materials, LiAlH4 is an attractive candidate to serve as a source of hydrogen in fuel cell systems because of its high theoretical hydrogen capacity (volumetric: 96.7 g-H2 l1; gravimetric: 10.6 wt.%-H2). Nearly 75% of its hydrogen capacity can be released at temperatures below 100  C after doping [1e4]. Although the reversible dehydrogenation of LiAlH4 must be carried out with the help of organic solvents [1,5,6], pristine LiAlH4 can serve as singleuse hydrogen storage material for fuel cell systems.

The dehydrogenation of pure LiAlH4 is generally accepted as a three-step process since the early 1970s [7]:

3 LiAlH4 / Li3AlH6 þ 3H2 þ 2 Al (5.31 wt.%-H2 of LiAlH4) (1.1)

Li3AlH6 / 3 LiH þ 3/2H2 þ Al (2.66 wt.%-H2 of LiAlH4)

(1.2)

3 LiH þ 3 Al / 3 LiAl þ 3/2H2 (2.66 wt.%-H2 of LiAlH4)

(1.3)

Most proton exchange membrane (PEM) fuel cells operate at near ambient pressure (0.1 MPa) and at temperatures between 30  C and 80  C [8]. To improve the energy efficiency of a

* Corresponding author. Tel.: þ49 351 2537 411; fax: þ49 351 2537 399. € ntzsch). E-mail address: [email protected] (L. Ro http://dx.doi.org/10.1016/j.ijhydene.2014.08.023 0360-3199/Copyright © 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.

16363

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 9 ( 2 0 1 4 ) 1 6 3 6 2 e1 6 3 7 1

fuel cell system with a metal hydride as hydrogen source, the heat required to trigger the dehydrogenation of LiAlH4 shall be taken from the exhaust heat of the fuel cell. Thus, it is reasonable to control the dehydrogenation temperature of LiAlH4 in the temperature range mentioned above. Since the temperature level to trigger Reaction 1.3 is significantly higher than 200  C, the study of the dehydrogenation properties of compacted LiAlH4 in this paper only refers to Reactions 1.1 and 1.2. However, if only Reactions 1.1 and 1.2 are taken into account, LiAlH4 can theoretically still release up to 7.97 wt.%H2, which is a considerable value for present hydrogen storage technologies. This study deals with the dehydrogenation properties of doped and compacted LiAlH4 for hydrogen generator applications in the PEM fuel cell systems. LiAlH4 is first doped with highly effective catalytic precursor, TiCl3 and ZrCl4, to improve its dehydrogenation kinetics [2,9e11]. The subsequent compaction of the doped LiAlH4 is performed aiming at a significantly higher volumetric hydrogen capacity compared to generally studied lithium alanate powder. The dependence of the dehydrogenation properties of doped LiAlH4 pellets on the compaction pressure during pelletization, on temperature and hydrogen partial pressure are systematically studied by thermal analysis. Furthermore, the volume change of compacted LiAlH4 during dehydrogenation will be reported and discussed. Moreover, the compacted LiAlH4 has been intensively tested in view of short-term storability from which its shelf life can be evaluated.

carried out in a glovebox (MBraun) under argon atmosphere to prevent unwanted oxidation (<2 ppm O2, <3 ppm H2O).

Doping and pelletization process Pre-milled LiAlH4, Et2O and 2 mol% dopant (TiCl3 or ZrCl4) were added into a flask under argon atmosphere. The mixture was stirred for 10 min by a magnetic stirring bar with a speed of 1000 rpm. Then the mixture was evacuated for 90 min to remove the Et2O completely. The residual powder is the doped LiAlH4 powder. The whole procedure was performed at 0  C using an ice-water mixture. The pelletization of TiCl3/ZrCl4-doped LiAlH4 powder was carried out on Atlas Manual Hydraulic Press inside the argon glovebox (<2 ppm O2, <3 ppm H2O). Several pellets (13 mm diameter; 2.1e3.4 mm height) consisting of about 250 mg of doped LiAlH4 powder were prepared. Three different compaction pressures (15 MPa, 60 MPa, 180 MPa) were chosen in the pellets preparation.

Phase analysis X-ray diffraction (XRD) was performed on a Bruker D8 Advance diffractometer in BraggeBrentano geometry using CueKa1 radiation at a tube voltage of 40 kV and a tube current of 40 mA. The scanning range of the diffraction angle (2q) was from 10 to 100 . The powder samples for XRD test were prepared in an argon glovebox and covered with a Kapton® foil to prevent any unwanted oxidation during subsequent analysis.

Experimental

Dehydrogenation kinetics

Raw materials and pre-milling

Thermogravimetric analysis (TGA) to measure the dehydrogenation kinetics of doped LiAlH4 pellets was performed in a magnetic suspension balance (Rubo-therm MSB) with a precision of 10 mg at a hydrogen partial pressure (purity: 99.9999%) of 0.1 MPa. The applied heating rate amounts to 1 K/min.

Lithium aluminum hydride powder (LiAlH4, 99% purity) and zirconium (IV) chloride (ZrCl4, 98% purity) were purchased from Alfa-Aesar. Titanium (III) chloride (TiCl3, 98.5% purity) was purchased from SigmaeAldrich. Diethyl ether (Et2O, 99.9% purity) was purchased from VWR. Prior to doping, asreceived LiAlH4 powder was pre-milled for 30 min in a Fritsch P6 planetary ball mill using steel bowls and steel balls with a ball-to-powder weight ratio of 20:1 and a rotation speed of 300 rpm. The powder size of the pre-milled LiAlH4 and the ZrCl4 is about several micrometers. The size of TiCl3 powders is about several ten micrometers. The powder preparation was

Short-term storability Freshly doped LiAlH4 pellets are kept in argon atmosphere for 8 days to test their stability during storage. The experiments were carried out under two different temperatures to mimic different storage environment: (1) at room temperature: 20e24  C; (2) at normal refrigerator temperature: 2e6  C. The

Table 1 e Dehydrogenation properties of TiCl3 and ZrCl4-doped LiAlH4 powders. Dopant

Atmosphere Gas

5 5 2 4 2 2 2 2

wt.% TiCl3 wt.% ZrCl4 mol% TiCl3 mol% ZrCl4 mol% TiCl3 mol% TiCl3 mol% TiCl3 mol% ZrCl4

Ar

Dehydrogenation

Heating rate (K/min)

Max. temp. ( C)

Ref.

Pressure (Pa)

H2 released (wt.%)

Onset temp. ( C)

Not mentioned

~0.3 ~1.1 ~1.5 ~6 ~2.0 5.0 6.7 2.0

Not mentioned

2

327

[11]

>100 ~100 ~50 <80 <80

3 7 1.3 e 1

450 150 250 240 80

[9] [10] [2] [1] This work

Ar 105 Not mentioned Vacuum Not mentioned Not mentioned H2 105

16364

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 9 ( 2 0 1 4 ) 1 6 3 6 2 e1 6 3 7 1

Fig. 1 e X-ray diffractograms of pre-milled (undoped) and freshly doped LiAlH4 powders.

dehydrogenation of the doped LiAlH4 pellets was monitored every second day based on the samples' weight loss using an electronic balance (Sartorius, precision: 0.1 mg) to evaluate the shelf life of the doped samples.

Results and discussion TiCl3-and ZrCl4-doped LiAlH4 powders Undoped LiAlH4 powder releases a negligible amount of hydrogen at 80  C and 0.1 MPa hydrogen partial pressure (less than 0.2 wt.% within 580 min). In contrast, TiCl3-doped LiAlH4 and ZrCl4-doped LiAlH4 exhibit a considerable amount of hydrogen release under the same conditions. The mass loss of ZrCl4-doped LiAlH4 powder nearly reaches 2 wt.% after 580 min. Moreover, the TiCl3-doped LiAlH4 powder releases about 6.7 wt.% of hydrogen after 580 min Table 1 lists the dehydrogenation properties of TiCl3 and ZrCl4 doped LiAlH4 reported in recent publications. Compared with this work, the tests reported previously are often carried out either in vacuum or in argon atmosphere, which will benefit the dehydrogenation but is much different from practical applications. In this work, TiCl3-doped LiAlH4 shows an evidently higher amount of hydrogen release than previously reported even at

a lower temperature. In contrast, the hydrogen release of ZrCl4-doped LiAlH4 is not high enough at constant 80  C. However, its dehydrogenation properties will be greatly improved after the pelletizing process (cf. Section 3.2). Fig. 1 shows the X-ray diffractograms of pre-milled (undoped) and freshly doped LiAlH4 powders. Pre-milled LiAlH4 contains the main phases of LiAlH4 and the impurity phase of LiCl (Fig. 1a). Weak peaks from TiCl3 and ZrCl4, which are the peaks of as-received dopants with highest intensity, are found respectively in the TiCl3-doped LiAlH4 sample (Fig. 1b) and ZrCl4-doped LiAlH4 sample (Fig. 1c). This indicates that these chlorides have not completely reacted with LiAlH4 after doping. However, a partial reaction between LiAlH4 and the dopants cannot be excluded since weak Al peaks are observed in the TiCl3-doped LiAlH4 sample. Furthermore, some unidentified weak peaks between 12 and 15 (marked by question mark) have been detected in ZrCl4-doped LiAlH4 samples, thus, indicating the reaction between LiAlH4 and ZrCl4. This result is consistent with the assumption that TM chlorides act as the catalytic precursor during the dehydrogenation of LiAlH4 [2,11e15]. Although this assumption is widely accepted, the real catalyst has never been observed by XRD in the present study. It has to be noted that the atomic working mechanism of the catalyst is still ambiguous and controversial. In this study, the formation of metallic phases (Ti/Zr) or intermetallic phases (TixAly/ZrxAly), as catalyst, during the doping process are assumed. However, these phases are never detected by XRD in present study. Possible reasons are the formation of either amorphous phases or the formation of phases with nanoscale size at a small concentration, which is below the detection limits of the XRD [16]. It should be noticed that peaks from Li3AlH6 are not observed in both doped LiAlH4 samples, demonstrating the non-negligible decomposition of LiAlH4 has not been triggered during the doping process.

The influence of compaction pressure on the dehydrogenation of doped LiAlH4 To obtain a higher hydrogen volumetric capacity, the loose LiAlH4 powders are compacted into LiAlH4 pellets when used in the PEM fuel cell system. Fig. 2 displays exemplary photographic images of ZrCl4-doped LiAlH4 compacts before and after dehydrogenation. Evidently, the pellets do not disintegrate during dehydrogenation. Fig. 3 shows the dehydrogenation characteristics of 2 mol% ZrCl4-doped LiAlH4 pellets at

Fig. 2 e 2 mol% ZrCl4-doped LiAlH4 pellets compacted at: (a) 15 MPa (before dehydrogenation), (b) 15 MPa (after dehydrogenation), (c) 180 MPa (before dehydrogenation) and (d) 180 MPa (after dehydrogenation).

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 9 ( 2 0 1 4 ) 1 6 3 6 2 e1 6 3 7 1

Fig. 3 e Dehydrogenation characteristics of 2 mol% ZrCl4doped LiAlH4 pellets compacted under different pressures (0 MPa represents the loose powder).

16365

Fig. 4 e Dehydrogenation characteristics of 2 mol% TiCl3doped LiAlH4 pellets compacted at different pressures (0 MPa represents the loose powder).

80  C at a H2 back-pressure of 0.1 MPa. The dehydrogenation curves of the powder sample and the pellet compacted at 15 MPa nearly overlap with each other. As the compaction pressure increases, the dehydrogenation kinetics of ZrCl4doped LiAlH4 pellets is significantly improved. The most plausible explanation is that the contact of ZrCl4 and LiAlH4 particles is promoted during compaction. Table 2 lists the relation between the compaction pressure and the properties of ZrCl4-doped LiAlH4 pellets. It can be seen that the density of the pellets increases as the compaction pressure increases. Consequently, the porosity of the compacted pellets greatly decreases from 34.09% to 4.54% when the compaction pressure is increased from 15 MPa to 180 MPa. After 580 min of dehydrogenation, both the hydrogen gravimetric capacity and the hydrogen volumetric capacity of ZrCl4-doped LiAlH4 pellets increase with increasing compaction pressure. As a result, ZrCl4-doped LiAlH4 pellets with higher compaction pressure show evidently better comprehensive dehydrogenation properties with regard to both hydrogen release amount and dehydrogenation kinetics. Fig. 4 exhibits the dehydrogenation characteristics of 2 mol % TiCl3-doped LiAlH4 pellets at 80  C with a H2 back-pressure of 0.1 MPa. It can be seen that the effect of compaction pressure on the dehydrogenation kinetics of TiCl3-doped LiAlH4 pellets is not as pronounced as in the case of ZrCl4-doped LiAlH4 pellets. The respective curves nearly overlap with each other. Interestingly, the powder sample shows the best

dehydrogenation kinetics. All three compacted pellets nearly have the same dehydrogenation kinetics during the first 100 min. However, the dehydrogenation of pellets compacted at 60 MPa and at 180 MPa slightly slows down after 100 min. Similar to the ZrCl4-doped LiAlH4 pellets, when the compaction pressure is increased from 15 MPa to 180 MPa, the density of the TiCl3-doped LiAlH4 pellets increase while the porosity of the pellets decreases (see Table 2). In general, the hydrogen volumetric capacity of the doped LiAlH4 pellets is greatly enhanced. It should be noticed that the maximum hydrogen release of TiCl3-doped LiAlH4 pellets is 53.88 g l1 (compacted at 180 MPa), which is much higher than the volumetric capacity of recently developed high-pressure composite cylinders (700 bar, about 39 g-H2 l1). It is close to the volumetric capacity of liquid hydrogen storage (71 g-H2 l1) [17].

The influence of temperature on the dehydrogenation of doped LiAlH4 compacts To take advantage of the exhaust heat of PEM fuel cell systems to trigger dehydrogenation, the dehydrogenation temperature of doped LiAlH4 pellets should be some K lower than the operation temperature of the PEM fuel cell (80  C). Figs. 5 and 6 respectively show the dehydrogenation characteristics of ZrCl4- and TiCl3-doped LiAlH4 pellets at different temperature at a H2 back-pressure of 0.1 MPa. According to Fig. 5, ZrCl4-

Table 2 e Properties of doped LiAlH4 pellets compacted at different pressure.

ZrCl4-doped LiAlH4 TiCl3-doped LiAlH4

Mass (mg)

Compaction pressure (MPa)

Density (g cm3)

Porosity (%)

Amount of H2 release after 580 min (g l1)

Amount of H2 release after 580 min (wt.%)

252.7 250.1 249.8 247.9 258.7 250.4

15 60 180 15 60 180

0.604 0.725 0.875 0.566 0.696 0.858

34.09 20.97 4.54 38.28 24.09 6.47

11.90 18.53 31.96 37.59 44.86 53.88

1.97 2.56 3.71 6.64 6.33 6.28

16366

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 9 ( 2 0 1 4 ) 1 6 3 6 2 e1 6 3 7 1

dehydrogenation kinetics in practical application. Thus, 80  C is necessary for the dehydrogenation of ZrCl4-doped LiAlH4 pellets while 60  C is sufficient for TiCl3-doped LiAlH4 pellets. The temperature dependence of the rate constant (k) is described by the Arrhenius equation (the experimental data is fit by the first-order model): k ¼ A exp (Ea/RT), where A is the preexponential (frequency) factor, Ea is the activation energy, T is absolute temperature and R is the gas constant. In accordance to Fig. 7(a and b), Ea of ZrCl4- and TiCl3-doped LiAlH4 are calculated as 83.5 kJ/mol and 63.6 kJ/mol, respectively. The preexponential factor A of ZrCl4- and TiCl3-doped LiAlH4 are calculated as 1.33  109 and 4.04  107, respectively. The results fall within the range of Ea and A values reported before [12,18]. Evidently, the catalytic effect of ZrCl4 on reducing Ea of LiAlH4 dehydrogenation is weaker than that of TiCl3. Fig. 5 e Dehydrogenation characteristics of ZrCl4-doped LiAlH4 pellets (compacted at 180 MPa) at different temperatures.

Fig. 6 e Dehydrogenation characteristics of TiCl3-doped LiAlH4 pellets (compacted at 15 MPa) at different temperatures.

doped LiAlH4 pellets release 3.71 wt.% of hydrogen at 80  C within 580 min while the sample with the same composition release only 0.76 wt.% of hydrogen at 60  C within 580 min (see Table 3). In contrast, even at 60  C, the TiCl3-doped LiAlH4 pellet can release more than 5.5 wt.% of H2 within 580 min (see Fig. 6 and Table 4). The results demonstrate there is a trade-off between the proper temperature and the enough

The influence of H2 partial pressure on the dehydrogenation of doped LiAlH4 compacts Theoretically, the decomposition of LiAlH4 should be obstructed by high H2 back-pressure. Wang et al. has successfully used high H2 pressure (97.5 bar) to prevent the dehydrogenation of LiAlH4 during ball milling [5]. However, there is no report about the effect of H2 back-pressure on the dehydrogenation of doped LiAlH4 within the operation pressure range of PEM fuel cells. To mimic the practical application of LiAlH4 pellets as hydrogen sources in PEM fuel cell system, Figs. 8 and 9 and Tables 5 and 6 depict the dehydrogenation characteristics of doped LiAlH4 pellets at 80  C at different H2 back-pressure (within the operation pressure range of PEM fuel cells: 0.1e0.4 MPa). From the results of Fig. 8, the dehydrogenation curves of ZrCl4-doped LiAlH4 pellets at 0.2 MPa and 0.4 MPa almost overlap. When the dehydrogenation pressure is reduced to 0.1 MPa, the dehydrogenation kinetics of the tested pellet is improved. The result suggests that within the simulated operation pressure range, the lower H2 back-pressure leads to the improved dehydrogenation kinetics of ZrCl4-doped LiAlH4 pellets. However, the improvement of the dehydrogenation kinetics at lower H2 backpressure is not significant. Similar results have been observed for TiCl3-doped LiAlH4 pellets (Fig. 9). The dehydrogenation curves of TiCl3-doped LiAlH4 pellets at 0.1 MPa and 0.2 MPa almost overlap, which is a little faster than that of the pellet desorbing hydrogen at 0.4 MPa.

Volume variation of LiAlH4 compacts during dehydrogenation Fig. 10 displays a simple design of a standard cylindrical tank for hydrogen storage [19,20]. Considering the volume variation

Table 3 e Properties of ZrCl4-doped LiAlH4 pellets (compacted at 180 MPa) dehydrogenated at different temperature.

ZrCl4-doped LiAlH4

Mass (mg)

Dehydrogenation temperature ( C)

Density (g cm3)

Porosity (%)

Amount of H2 release after 580 min (g l1)

Amount of H2 release after 580 min (wt.%)

249.0 248.6 249.8

60 70 80

0.873 0.871 0.875

4.85 5.00 4.54

6.67 11.84 31.96

0.76 1.36 3.71

16367

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 9 ( 2 0 1 4 ) 1 6 3 6 2 e1 6 3 7 1

Table 4 e Properties of TiCl3-doped LiAlH4 pellets (compacted at 15 MPa) dehydrogenated at different temperature.

TiCl3-doped LiAlH4

Mass (mg)

Dehydrogenation temperature ( C)

Density (g cm3)

Porosity (%)

Amount of H2 release after 580 min (g l1)

Amount of H2 release after 580 min (wt.%)

247.3 250.9 247.9

60 70 80

0.556 0.556 0.566

39.34 39.36 38.28

32.76 34.76 37.59

5.89 6.25 6.64

Fig. 7 e The plot of ln (k) versus the reciprocal temperature for (a) ZrCl4-doped LiAlH4 pellets (compacted at 180 MPa), (b) TiCl3-doped LiAlH4 pellets (compacted at 15 MPa).

of the LiAlH4 pellets after dehydrogenation, the size of the cylindrical tank should be designed referring to the data in Tables 7 and 8. According to Table 7, after dehydrogenation, the volume of all ZrCl4-doped LiAlH4 pellets has increased in both the radial direction and the axial direction, the increment is from 6.3% to 12.0%. On the contrary, the volume of all TiCl3doped LiAlH4 pellets decreases in both directions after dehydrogenation (see Table 8). Before suggesting explanations for the contradiction of volume change between ZrCl4- and TiCl3doped LiAlH4 pellets, it should be known that the volume change of doped LiAlH4 pellets can be ascribed to three reasons: (1) The phase transformation (LiAlH4 / LiH þ Al þ 3/ 2H2[) of the pellets leads to a decreased volume. LiAlH4 has a

monoclinic unit cell with space group of P121/c. After dehydrogenation, both LiH and Al have face-centered cubic structures (Fm-3m), which are densely packed arrangements of atoms; (2) During the dehydrogenation process, the particles inside the pellets will stretch, distort or break to make a path for hydrogen gas to escape, simultaneously inducing volume swelling, or even breakage of pellet samples. According to the investigations of Beattie [21], doped LiAlH4 samples decompose at a lower temperate and dehydrogenate more slowly than the undoped LiAlH4 samples. A benign outgassing of the doped LiAlH4 samples leads to less morphological change than undoped LiAlH4 samples. In other words, the benign outgassing may cause less volume swelling than the drastic

Fig. 8 e Dehydrogenation characteristics of ZrCl4-doped LiAlH4 pellets (compacted at 180 MPa) at different H2 backpressure.

Fig. 9 e Dehydrogenation characteristics of TiCl3-doped LiAlH4 pellets (compacted at 15 MPa) at different H2 backpressure.

16368

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 9 ( 2 0 1 4 ) 1 6 3 6 2 e1 6 3 7 1

Table 5 e Properties of ZrCl4-doped LiAlH4 pellets (compacted at 180 MPa) dehydrogenated at different H2 back-pressure. Mass (mg)

Dehydrogenation pressure (H2) (MPa)

Density (g cm3)

Porosity (%)

Amount of H2 release after 580 min (g l1)

Amount of H2 release after 580 min (wt.%)

249.8 246.5 245.8

0.1 0.2 0.4

0.875 0.864 0.861

4.54 5.80 6.07

31.96 28.69 27.23

3.71 3.32 3.16

ZrCl4- doped LiAlH4

Table 6 e Properties of TiCl3-doped LiAlH4 pellets (compacted at 15 MPa) dehydrogenated at different H2 back-pressure. Mass (mg)

Dehydrogenation pressure (H2) (MPa)

Density (g cm3)

Porosity (%)

Amount of H2 release after 580 min (g l1)

Amount of H2 release after 580 min (wt.%)

247.9 246.7 243.3

0.1 0.2 0.4

0.566 0.563 0.556

38.28 38.58 39.41

37.59 37.23 34.71

6.64 6.61 6.25

TiCl3-doped LiAlH4

of ZrCl4-doped LiAlH4 pellets, factor (2) is the dominant factor of the volume change since more than 50% of hydrogen storage capacity is released in a drastic manner only at higher temperatures 240  C. In connection with Figs. 3, 5 and 8, it is easy to find that the pellet desorbs less hydrogen within 580 min has a larger volume variation after complete dehydrogenation. In contrast, more than 80% of the hydrogen storage capacity of TiCl3-doped LiAlH4 pellets has already been released in a soft manner in the first step (cf. Figs. 4, 6 and 9). Therefore, factor (1) appears to become dominant for the volume change of TiCl3-doped LiAlH4 pellets. Since the volume of TiCl3- and ZrCl4-doped LiAlH4 changes in opposite directions, co-doping might be a possible method to obtain volume-stable LiAlH4 pellets.

Fig. 10 e Simple design of a cylindrical hydrogen tank with LiAlH4 compacts.

outgassing; (3) During dehydrogenation, the formation of hydrogen bubbles inside pellets also might cause the volume swelling of the pellets. The swelling continues until the hydrogen bubbles contact with open pores. All of the three factors influence the doped LiAlH4 pellets in different extent and finally lead to the swelling or shrinkage of different samples. Referring to the doped LiAlH4 pellets in Tables 7 and 8, the data are collected after two steps of dehydrogenation: Firstly, the doped LiAlH4 pellets experience a benign dehydrogenation process at constant 80  C for 580 min (cf. Figs. 3e6, Figs. 8 and 9); Secondly, the residual hydrogen storage capacity is rapidly released when the temperature increases to 240  C. In the case

Short-term storability of doped LiAlH4 compacts Freshly doped LiAlH4 compacts are kept in argon atmosphere for 8 days to study their stability during the storage (shelf live). The errors of weight loss are within ±0.2 wt.%. Fig. 11(a) presents the mass loss of the ZrCl4-doped LiAlH4 compacts at room temperature. The pellet compacted at 180 MPa releases 1.43 wt.% of hydrogen within 8 days while the pellet compacted at 15 MPa releases only 0.44 wt.% of hydrogen in the same duration. The results reveal that the dehydrogenation kinetics of compacts at room temperature

Table 7 e The volume variation of 2 mol% ZrCl4-doped LiAlH4 pellets. Preparation conditions

Dehydrogenation conditions

Variation after dehydrogenation

Mass (mg)

Compaction pressure (MPa)

Temperature ( C)

H2 pressure (bar)

Diameter (%)

Thickness (%)

Volume (%)

252.7 250.1 249.0 248.6 249.8 246.5 245.8

15 60 180 180 180 180 180

80 80 60 70 80 80 80

1 1 1 1 1 2 4

þ2.7 þ3.1 þ3.5 þ2.7 þ1.9 þ2.3 þ2.7

þ4.8 þ1.9 þ4.7 þ4.7 þ2.3 þ2.3 þ2.3

þ10.5 þ8.3 þ12.0 þ10.4 þ6.3 þ7.1 þ7.9

16369

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 9 ( 2 0 1 4 ) 1 6 3 6 2 e1 6 3 7 1

Table 8 e The volume variation of 2 mol% TiCl3-doped LiAlH4 pellets. Preparation conditions

Dehydrogenation conditions

Variation after dehydrogenation

Mass (mg)

Compaction pressure (MPa)

Temperature ( C)

H2 pressure (bar)

Diameter (%)

Thickness (%)

Volume (%)

247.9 258.7 250.4 247.3 250.9 246.7 243.3

15 60 180 15 15 15 15

80 80 80 60 70 80 80

1 1 1 1 1 2 4

1.9 2.3 2.3 2.3 1.9 1.5 1.5

1.5 3.6 2.3 1.5 2.9 1.5 1.5

5.3 6.3 6.7 6.0 6.6 4.5 4.5

Fig. 11 e Dehydrogenation of ZrCl4-doped LiAlH4 compacts at (a) room temperature and (b) refrigerator temperature.

becomes slower when the compaction pressure decrease. For comparison, the mass loss of the ZrCl4-doped LiAlH4 compacts stored at refrigerator temperature is less than 0.1 wt.% and can be neglected, therefore (Fig. 11(b)). Fig. 12(a) presents the mass loss of the TiCl3-doped LiAlH4 compacts at room temperature. Unlike the ZrCl4-doped LiAlH4 compacts, the performances of three samples are nearly the same. After 8 days, about 4.5 wt.% of hydrogen are released from the three TiCl3-doped LiAlH4 compacts. For comparison, the three TiCl3-doped LiAlH4 compacts stored at refrigerator temperature release less than 0.5 wt.% of hydrogen after 8 days (Fig. 12(b)). Consequently, storing ZrCl4- and TiCl3-

doped LiAlH4 compacts at refrigerator temperature leads to much less hydrogen release than that of the compacts stored at room temperature during the short-term storage process. In addition, these trends of dehydrogenation behaviors of the doped LiAlH4 compacts are similar when they are placed at 80  C and at room temperature. In other words, a sample shows the best dehydrogenation kinetics at 80  C will also has the fastest dehydrogenation kinetics at room temperature. Thus, the dehydrogenation behavior of ZrCl4- and TiCl3-doped LiAlH4 compacts at room temperature can be estimated when their performance at higher temperature is observed.

Fig. 12 e Dehydrogenation of TiCl3-doped LiAlH4 compacts at (a) room temperature and (b) refrigerator temperature.

16370

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 9 ( 2 0 1 4 ) 1 6 3 6 2 e1 6 3 7 1

Although the performances of ZrCl4- and TiCl3-doped LiAlH4 compacts discussed in this paper is attractive, there are further problems to be solved in view of the practical application as hydrogen source for PEM fuel cell systems. For example, an advanced heat transport in LiAlH4 compacts would promote the dehydrogenation kinetics. Furthermore, the gas transport properties of LiAlH4 compacts needs to be improved to reduce the pellets swelling during dehydrogenation. Thus, the heat conductivity and gas permeability of LiAlH4 compacts should be tested thoroughly before, during and after dehydrogenation in order to deduce the optimum dimensions and operation conditions of doped LiAlH4 compacts. Possible methods of improving the heat and gas transport of doped LiAlH4 compacts include coating it with metal, mixing it with secondary materials with high thermal conductivity (e.g. graphite) or filling it in a matrix configuration (such as metal foams or other cellular structures) [22e25].

Conclusions 2 mol% TiCl3-and ZrCl4-doped LiAlH4 powders were prepared and pelletized at different compaction pressures. The relation between the compaction pressure, the dehydrogenation conditions and the performance of doped LiAlH4 compacts can be concluded as follows: a) As the compaction pressure increases, the density of the doped LiAlH4 compacts increases while the porosity of the compacts decreases. Consequently, the hydrogen volumetric capacity of the doped LiAlH4 compacts is greatly enhanced. b) As the compaction pressure increases, the dehydrogenation kinetics of the ZrCl4-doped LiAlH4 compacts improves. In contrast, the compaction pressure does not show evident influence on the dehydrogenation kinetics of TiCl3doped LiAlH4 compacts. c) The dehydrogenation kinetics of the ZrCl4- and TiCl3doped LiAlH4 compact samples are greatly improved when the temperature increases. d) The effect of H2 back-pressure (in the range of 0.1e0.4 MPa) on the dehydrogenation kinetics of the doped LiAlH4 compacts is not significant. e) After dehydrogenation, the volume swelling of ZrCl4-doped LiAlH4 compacts and the volume shrinkage of TiCl3-doped LiAlH4 compacts has been observed. f) Storing ZrCl4- and TiCl3-doped LiAlH4 compacts at refrigerator temperature leads to much less hydrogen release than that of the compacts stored at room temperature during a short-term storage period (8 days).

Acknowledgments Jie Fu Thanks the China Scholarship Council for financial support. Further, support from the Fraunhofer Attract program is gratefully acknowledged. The authors thank M.

Eckardt, T. Richter, T. Himmelreich and V. Pacheco for experimental assistance.

references

[1] Liu X, McGrady GS, Langmi HW, Jensen CM. Facile cycling of Ti-doped LiAlH4 for high performance hydrogen storage. J Am Chem Soc 2009;131:5032e3. [2] Easton DS, Schneibel JH, Speakman SA. Factors affecting hydrogen release from LiAlH4. J Alloys Compds 2005;398:245e8. [3] Ismail M, Zhao Y, Yu XB, Nevirkovets IP, Dou SX. Significantly improved dehydrogenation of LiAlH4 catalysed with TiO2 nanopowder. Int J Hydrogen Energy 2011;36:8327e34. [4] Li P, Li Z, Zhai F, Wan Q, Li X, Qu X, et al. NiFe2O4 nanoparticles catalytic effects of improving LiAlH4 dehydrogenation properties. J Phys Chem C 2013;117:25917e25. [5] Wang J, Ebner AD, Ritter JA. Physiochemical pathway for cyclic dehydrogenation and rehydrogenation of LiAlH4. J Am Chem Soc 2006;128:5949e54. [6] Graetz J, Wegrzyn J, Reilly JJ. Regeneration of lithium aluminum hydride. J Am Chem Soc 2008;130:17790e4. [7] Dilts JA, Ashby EC. Thermal decomposition of complex metal hydrides. Inorg Chem 1972;11:1230e6. [8] Larminie J, Dicks A. Fuel cell systems explained. 2nd ed. London: Wiley; 2003. [9] Resan M, Hampton MD, Lomness JK, Slattery DK. Effects of various catalysts on hydrogen release and uptake characteristics of LiAlH4. Int J Hydrogen Energy 2005;30:1413e6. [10] Suttisawat Y, Rangsunvigit P, Kitiyanan B, Muangsin N, Kulprathipanja S. Catalytic effect of Zr and Hf on hydrogen desorption/absorption of NaAlH4 and LiAlH4. Int J Hydrogen Energy 2007;32:1277e85. [11] Kojima Y, Kawai Y, Matsumoto M, Haga T. Hydrogen release of catalyzed lithium aluminum hydride by a mechanochemical reaction. J Alloys Compds 2008;462:275e8. [12] Chen J, Kuriyama N, Xu Q, Takeshita HT, Sakai T. Reversible hydrogen storage via titanium-catalyzed LiAlH4 and Li3AlH6. J Phys Chem B 2001;105:11214e20. [13] Balema VP, Balema L. Missing pieces of the puzzle or about some unresolved issues in solid state chemistry of alkali metal aluminohydrides. Phys Chem Chem Phys 2005;7:1310e4. [14] Resan M, Hampton MD, Lomness JK, Slattery DK. Effect of TixAly catalysts on hydrogen storage properties of LiAlH4 and NaAlH4. Int J Hydrogen Energy 2005;30:1417e21. [15] Shim JH, Lee GJ, Lee BJ, Oh YJ, Cho YW. Thermal decomposition of L3AlH6 with TiAl3 catalyst. Catal Today 2007;120:292e7. [16] Kim HS, Suhr DS, Kim GH, Kum DW. Analysis of x-ray diffraction patterns from mechanically alloyed AleTi powders. Metal Mater 1996;2:15e21. [17] Zu¨ttel A, Borgschulte A, Schlapbach L. Hydrogen as a future energy carrier. Weinheim: Wiley-VCH; 2008. [18] Varin RA, Zbroniec L. Fast and slow dehydrogenation of ball milled lithium alanate (LiAlH4) catalyzed with manganese chloride (MnCl2) as compared to nanometric nickel catalyst. J Alloys Compds 2011;S509:S736e9. [19] Ye J, Jiang L, Li Z, Liu X, Wang S, Li X. Numerical analysis of heat and mass transfer during absorption of hydrogen in metal hydride based hydrogen storage tanks. Int J Hydrogen Energy 2010;35:8216e24.

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 9 ( 2 0 1 4 ) 1 6 3 6 2 e1 6 3 7 1

[20] Askri F, Ben Salah M, Jemni A, Ben Nasrallah S. Optimization of hydrogen storage in metal-hydride tanks. Int J Hydrogen Energy 2009;34:897e905. [21] Beattie SD, McGrady GS. Hydrogen desorption studies of NaAlH4 and LiAlH4 by in situ heating in an ESEM. Int J Hydrogen Energy 2009;34:9151e6. [22] Chen Y, Sequeira CAC, Chen CP, Wang XH, Wang QD. Metal hydride beds and hydrogen supply tanks as minitype PEMFC hydrogen sources. Int J Hydrogen Energy 2003;28:329e33. € ntzsch L, Hu JJ, Weißga € rber T, Kieback B, [23] Pohlmann C, Ro Fichtner M. Tailored heat transfer characteristics of

16371

pelletized LiNH2eMgH2 and NaAlH4 hydrogen storage materials. J Power Sources 2012;205:173e9. [24] Chaise A, de Rango P, Marty Ph, Fruchart D, Miraglia S, Olive's R, et al. Enhancement of hydrogen sorption in magnesium hydride using expanded natural graphite. Int J Hydrogen Energy 2009;34:8589e96. [25] Mellouli S, Dhaou H, Askri F, Jemni A, Ben Nasrallah S. Hydrogen storage in metal hydride tanks equipped with metal foam heat exchanger. Int J Hydrogen Energy 2009;34:9393e401.