Kinetic study and determination of the enthalpies of activation of the dehydrogenation of titanium- and zirconium-doped NaAlH4 and Na3AlH6

Article abstract of DOI:10.1021/jp034543h

The rates of the dehydrogenation of the sodium alanates NaAlH₄ and Na₃AlH₆ doped with 2 mol % Ti or Zr have been measured over the temperature range 363-423 K. NaAlH₄ and Na₃AlH₆ undergo dehydrogenation at equal rates upon direct doping with titanium. However, Na₃AlH₆ arising from the dehydrogenation of Ti-doped NaAlH₄ undergoes dehydrogenation at much slower rates. Rate constants were determined from the slopes of the dehydrogenation profiles. On the basis of Eyring theory, the enthalpies of activation, ΔH^?, for the dehydrogenation reactions were determined to be ～100 kJ·mol^-1 for both Ti-doped NaAlH₄ and Na₃AlH₆ and ～135 kJ·mol^-1 for both Zr-doped NaAlH₄ and Na₃AlH₆. These results suggest that the dehydrogenation reaction pathways are highly sensitive to the nature and distribution of the dopant but not to differences in the Al-H bonding interactions in [AlH₄]^- and [AlH₆]^3- complex anions. Furthermore, we conclude that the kinetics are probably influenced by processes such as nucleation and growth and/or range atomic transport phenomenon.

Full text of DOI:10.1021/jp034543h

J. Phys. Chem. A 2003, 107, 7671-7674
7671
ARTICLES
Kinetic Study and Determination of the Enthalpies of Activation of the Dehydrogenation of
Titanium- and Zirconium-Doped NaAlH4 and Na3AlH6
Tetsu Kiyobayashi,† Sesha S. Srinivasan,‡ Dalin Sun,‡ and Craig M. Jensen*,‡
National Institute of AdVanced Science and Technology, Midorigaoka 1-8-31, Ikeda, Osaka 563-8577, Japan,
and Department of Chemistry, UniVersity of Hawaii, Honolulu, Hawaii 96822
ReceiVed: March 3, 2003; In Final Form: July 2, 2003
The rates of the dehydrogenation of the sodium alanates NaAlH4 and Na3AlH6 doped with 2 mol % Ti or Zr
have been measured over the temperature range 363-423 K. NaAlH4 and Na3AlH6 undergo dehydrogenation
at equal rates upon direct doping with titanium. However, Na3AlH6 arising from the dehydrogenation of
Ti-doped NaAlH4 undergoes dehydrogenation at much slower rates. Rate constants were determined from
the slopes of the dehydrogenation profiles. On the basis of Eyring theory, the enthalpies of activation, ∆Hq,
for the dehydrogenation reactions were determined to be ∼100 kJ‚mol-1 for both Ti-doped NaAlH4 and
Na3AlH6 and ∼135 kJ‚mol-1 for both Zr-doped NaAlH4 and Na3AlH6. These results suggest that the
dehydrogenation reaction pathways are highly sensitive to the nature and distribution of the dopant but not
to differences in the Al-H bonding interactions in [AlH4]- and [AlH6]3- complex anions. Furthermore, we
conclude that the kinetics are probably influenced by processes such as nucleation and growth and/or range
atomic transport phenomenon.
Introduction
on the dehydrogenation of NaAlH4 and Na3AlH6. We have
carried out systematic kinetic measurements over the temper-
ature range 363-423 K and interpreted the resulting dehydro-
genation profiles. This information has provided us with insights
into the dehydrogenation mechanisms of NaAlH4 and Na3AlH6.
The development of high-performance hydrogen carriers is
becoming increasingly important with the approach of practical,
hydrogen-powered vehicles and electronic devices. Over the past
several decades, metal hydrides have been utilized in limited
applications as hydrogen storage materials. However, a “holy
grail” material having both a high hydrogen capacity (g5 wt
% available hydrogen)1 and rapid hydrogen cycling kinetics at
low temperatures has not been attained despite the considerable
research efforts that have been concentrated on MgH2 and a
variety of interstitial hydrides.2 A number of complex hydrides
have considerably high hydrogen capacities. However, until
recently, they were precluded from consideration as potential
hydrogen storage materials due to their slow dehydrogenation
kinetics and poor reversibility. This thinking was changed by
the finding that charging NaAlH4 with a few mole percent of
selected transition metal dopants significantly enhances the
kinetics of the dehydrogenation process and renders it reversible
in the solid state under moderate conditions.3,4 This breakthrough
has been followed by a great deal of effort to develop alanates
as practical hydrogen storage materials.5-23 However, the
mechanism by which the dopants enhance the dehydrogenation
of NaAlH4 remains an enigma. An understanding of the
mechanistic role of the dopants would aid the development of
alanates that are suitable for practical applications.
Experimental Section
Materials. Titanium n-butoxide, Ti(OBun)4, and zirconium
n-propoxide, Zr(OPrn)4, were obtained from Aldrich Chemical
Inc. and used as received. Sodium aluminum hydride, NaAlH4,
was obtained from Albemarle Corp. and recrystallized from
tetrahydrofuran prior to use. NaAlH4 was doped with 2 mol %
Ti(OBun)4 of Zr(OPrn)4 according to the published procedure.5,14
Kinetic Measurements. The dehydrogenation studies were
carried out using an automated Sieverts’ type apparatus (LESCA
Co), which allowed for the accurate volumetric determination
of the amounts of hydrogen evolved. Rapid heating of the
sample to the desired temperature was accomplished by im-
mersing the sample reactor into a silicon oil bath (accuracy of
(1 K) preheated to a given temperature. Dehydrogenation was
performed against a back pressure of 0.1 MPa in a fixed volume.
Theory of Data Analysis. The dehydrogenation of NaAlH4
is composed of a two-step process seen in eqs 1 and 2.
k1
2
NaAlH4 98 1Na3AlH6 + Al + H2
(1)
(2)
3
3
The present study was conducted to obtain quantitative
information about the effect of titanium and zirconium dopants
k2
3
Na3AlH6
98 3NaH + Al + H2
2
* Corresponding author. Phone: 1-808-956-2769. Fax: 1-808-956-5908.
E-mail: jensen@gold.chem.hawaii.edu.
Assuming these reactions to be first-order and independent, the
differential form of the rate equation is
† National Institute of Advanced Science and Technology.
‡ University of Hawaii.
10.1021/jp034543h CCC: $25.00 © 2003 American Chemical Society
Published on Web 09/03/2003

7672 J. Phys. Chem. A, Vol. 107, No. 39, 2003
dNA(t)
Kiyobayashi et al.
) -kiNA(t)
The solution of eq 3 gives
-kit ) ln
(A ) NaAlH4, Na3AlH6; i ) 1, 2)
dt
(3)
NA(t)
(4)
0
(
)
NA
where ki is the reaction rate constant and NA(t) and NA0 are the
moles of the substance A present at times t and t ) 0,
respectively. It is appropriate to substitute NH (t) for NA(t) in
Figure 1. Dehydrogenation of Ti-doped NaAlH4 at 423 K. (A)
Experimental. (B) Theoretical: total (eq 10). (C) Theoretical: contribu-
tion from the dehydrogenation of NaAlH4 (first term on the right side
of eq 10). (D) Theoretical: contribution from Na3AlH6 (second term
on the right side of eq 10).
2
eq 4 because moles of hydrogen, NH (t), rather than NA(t) were
2
directly determined in our experiments. In terms of the stoich-
iometric factors of reactions in eqs 1 and 2, we obtain the
relations between NH (t) and the rate constant ki as
∆Gq
RT
2
kBT
ki ) κ
exp -
(11)
(
)
h
NH (t)
2
-k1t ) ln 1 -
(5)
(6)
0
(
)
N
where κ is the transmission coeffcient, kB is the Boltzmann
constant, h is Planck’s constant, T is the absolute temperature,
and R is the gas constant. Assuming that κ is independent of
the temperature, substituting ∆Gq ) ∆Hq - T∆Sq, where ∆Hq
and ∆Sq are respectively the enthalpy and entropy of activation,
into eq 11 and rearranging and taking the logarithm gives
NaAlH4
for the dehydrogenation of NaAlH4 and
2NH (t)
2
-k2t ) ln 1 -
0
(
)
3N
Na3AlH6
∆Hq 1
R T
ki
for that of Na3AlH6.
ln
) -
+ constant
(12)
( )
T
When the reactions in eqs 1 and 2 are treated as consecutive
first-order reactions, the simultaneous differential equations
Results and Discussion
dNNaAlH (t)
4
The overall dehydrogenation profile of NaAlH4 doped with
2 mol % Ti(OBun)4 (Ti-doped NaAlH4) is seen in Figure 1,
) -k1NNaAlH
4
dt
The solid line represents the hydrogen evolution NH (t) from
2
dNNa AlH (t)
3
6
1
3
Ti-doped NaAlH4 at 423 K. The profile is composed of two
portions: a rapid reaction (0 e t/s e 1000) followed by a slow
reaction (1000 e t/s). Obviously, the first reaction corresponds
to the dehydrogenation of NaAlH4 (eq 1) and the second to the
dehydrogenation of Na3AlH6 (eq 2). The broken line in the
figure shows the close simulation of the profile that was obtained
by using eq 10 and values of k1 ) 3.5 × 10-3 s-1 and k2 ) 2.3
× 10-5 s-1. These rate constants confirm earlier observations
that the first dehydrogenation reaction in the stepwise dehy-
drogenation of Ti-doped NaAlH4 is much faster than the
second.5,7,11,15,20,22 In the case of NaAlH4 doped with 2 mol %
Zr(OPrn)4 (Zr-doped NaAlH4), the second step is so much slower
than the first one that the rate constant, k2, could not be
determined within a practical time frame. A priori, these results
suggest that both the titanium and zirconium dopants provide a
much more pronounced intrinsic kinetic enhancement to the first
dehydrogenation reaction. However, further experiments, de-
scribed below, have shown that this is not the case. Although
the experimental profile is closely simulated by eq 10 at 423
K, it deviates significantly from the model at temperatures lower
than 383 K. This is apparently due to kinetic complications
imposed processes not considered in eq 10 (i.e., nucleation and
growth or atomic diffusion) that are likely to limit this solid
state transformation at the lower temperatures. To clarify the
kinetic constraints on the dehydrogenation process, we decided
to determine the rate constants k1 and k2 of eqs 1 and 2
individually through independent measurement of the dehydro-
genation kinetics of the Ti- and Zr-doped NaAlH4 and Na3-
AlH6.
) -k2NNa AlH + k1NNaAlH
4
(7)
(8)
3
6
dt
yield the solution
NNaAlH (t)
4
) exp(-k1t)
0
NNaAlH
4
and
N
(t)
k1
Na3AlH6
1
)
{exp(-k2t) - exp(-k1t)} (9)
0
3 k1 - k2
NNaAlH
4
0
by applying the initial condition NNa AlH ) 0. For the same
reason as mentioned above, NH (t) is substituted for NNaAlH (t)
3
6
2
4
and NNa AlH (t) in eqs 8 and 9. Thus the relation between NH (t)
3
6
2
and the rate constants k1 and k2 is obtained,
NH (t)
k2
2
0 ) 1 -
{1 - exp(-k1t)} +
{
}
2(k1 - k2)
N
NaAlH4
k1
{1 - exp(-k2t)} (10)
{
}
2(k1 - k2)
The first term on the right-hand side of eq 10 is the contribution
from the dehydrogenation of NaAlH4, and the second term is
that of Na3AlH6.
According to the Eyring theory,24 the relation between the
rate constant ki and the Gibbs free energy of activation ∆Gq is
given by
Na3AlH6 was prepared through dehydrogenation of NaAlH4
in a 350 mL stainless steel reactor at 480-490 K for 3 h

Dehydrogenation of Ti- and Zr-Doped NaAlH4
J. Phys. Chem. A, Vol. 107, No. 39, 2003 7673
Figure 3. Eyring plots of rate constants for the dehydrogenation of
(A) Ti-doped NaAlH4 9; (B) Ti-doped Na3AlH6 0; (C) Zr-doped
NaAlH4 b; and (D) Zr-doped Na3AlH6 O.
Figure 2. Comparison of the dehydrogenation profiles at 383 K of
(A) Ti-doped NaAlH4; (B) Ti-doped Na3AlH6; (C) Zr-doped NaAlH4;
and (D) Zr-doped Na3AlH6.
TABLE 2: Enthalpies of Activation, ∆Hq, of Ti- and
TABLE 1: Dehydrogenation Rate Constants (k/s-1) of Ti-
Zr-Doped NaAlH4 and Na3AlH6 (in kJ per mol)
and Zr-Doped NaAlH4 and Na3AlH6
Ti-doped
Zr-doped
Ti-doped
Zr-doped
NaAlH4
Na3AlH6
100 ( 7
99 ( 13
134 ( 16
135 ( 17
T/K
NaAlH4
Na3AlH6
NaAlH4
Na3AlH6
363
373
383
393
403
413
423
2.3 × 10-5a
2.0 × 10-4a
2.4 × 10-4a
4.8 × 10-4
1.1 × 10-3a
2.4 × 10-3a
4.8 × 10-3
1.3 × 10-6
6.6 × 10-6
3.2 × 10-5
8.6 × 10-5
2.1 × 10-4
4.7 × 10-4
8.7 × 10-4
7.7 × 10-5
2.6 × 10-4
5.0 × 10-4a
1.1 × 10-3
2.6 × 10-3a
3.6 × 10-3
NaAlH4) ≈ ∆Hq (Ti doped Na3AlH6) ≈ 100 kJ‚mol-1 and ∆Hq
(Zr-doped NaAlH4) ≈ ∆Hq (Zr doped Na3AlH6) ≈ 135
4.5 × 10-6
1.4 × 10-6a
2.9 × 10-5a
1.1 × 10-4a
2.8 × 10-4a
kJ‚mol-1
.
Conclusions
a Average of two determinations.
We have observed that both NaAlH4 and Na3AlH6 undergo
dehydrogenation at nearly equal rates directly following me-
chanical doping with titanium. This finding was quite unex-
pected, as the Na3AlH6 arising from the initial dehydrogenation
of Ti-doped NaAlH4 undergoes dehydrogenation at much slower
rates. Thus Ti doping does not innately enhance the dehydro-
genation NaAlH4 to a greater extent than Na3AlH6. Instead it
seems plausible that the dehydrogenation of doped NaAlH4
yields Na3AlH6 in which the Ti dopant is not properly distributed
for the maximum enhancement of the dehydrogenation of the
hexahydride. Our finding that Ti-doped NaAlH4 and Na3AlH6
undergo dehydrogenation at the same rates clearly indicates that
the dehydrogenation kinetics are insensitive to variation in the
Al-H bonding interactions and are probably influenced by
processes such as nucleation and growth or the diffusion of
atoms and molecules. These processes may be less important
in determining the dehydrogenation kinetics of Zr-doped
NaAlH4 and Na3AlH6, as, unlike the Ti-doped materials, directly
Zr-doped Na3AlH6 undergoes dehydrogenation at significantly
different rates than Zr-doped NaAlH4.
At high temperatures (ca. 423 K) the two-step dehydroge-
nation of Ti-doped NaAlH4 to NaH is well represented by a
consecutive, first-order reaction kinetics given by eq 7. However,
at temperatures lower than 383 K, the dehydrogenation kinetics
deviate significantly from this model. This observation provides
further confirmation that processes other than the breaking of
Al-H bonds in the complex anions play significant roles in
determining the rates of dehydrogenation of these materials at
lower temperatures.
according to the procedure of Ashby.25 Analysis of the product
by X-ray diffraction confirmed it to be a mixture of Na3AlH6
and Al with no traces of NaAlH4. Measurement of the amount
of hydrogen released during heating also verified that the
reaction seen in eq 1 proceeds to completion.
Figure 2 compares the dehydrogenation profiles of Ti- and
Zr-doped NaAlH4 with those of Ti- and Zr-doped Na3AlH6 at
T ) 383 K. Determination of the slopes of the dehydrogenation
profiles (indicated by the broken lines in Figure 2) during the
acceleratory period gives the rate constants26 k1 and k2 in terms
of eq 4: k1 ) 2.4 × 10-4 s-1 and k2 ) 2.6 × 10-4 s-1 for
Ti-doped hydrides; and k1 ) 3.2 × 10-5 s-1 and k2 ) 4.5 ×
10-6 s-1 for Zr-doped hydrides. Thus the dehydrogenation of
Na3AlH6 that is directly doped with Ti or Zr undergoes
dehydrogenation at much faster rates than those observed for
the hexahydride arising from the dehydrogenation of doped
NaAlH4. Remarkably, the rates observed for directly Ti-doped
Na3AlH6 match those of Ti-doped NaAlH4. However, hexahy-
dride that is directly doped with Zr undergoes significantly
slower dehydrogenation than Zr-doped NaAlH4.
Kinetic measurements were carried out over temperature
ranges of 363-423 K for Ti- and Zr-doped NaAlH4, 373-423
K for Ti-doped Na3AlH6, and 383-423 K for Zr-doped Na3-
AlH6. To check the reproducibility and the reliability of the
data, the measurements were repeated twice for some temper-
atures using different batches of hydride. The dehydrogenation
rates of Ti-doped NaAlH4 and Na3AlH6 were consistently found
to be much higher than those of the Zr-doped hydrides. The
obtained rate constants are summarized in Table 1. The
activation enthalpies, ∆Hq, of dehydrogenation were determined
from the Eyring plots (ln(k/T) vs 1/T) shown in Figure 3. The
values derived for ∆Hq are summarized in Table 2. Although
the standard deviations are large, it can be concluded that the
activation enthalpies of dehydrogenation of Ti-doped NaAlH4
and Na3AlH6 are lower than those of the Zr-doped hydrides.
For both the Ti- and Zr-doped hydrides, the activation enthalpies
of NaAlH4 and Na3AlH6 are comparable: i.e., ∆Hq (Ti-doped
We have found excellent agreement in the values determined
for the activation enthalpies of the dehydrogenation of NaAlH4
and Na3AlH6 with the same dopant (Ti doped ≈ 100 kJ per
mol and Zr doped ≈ 135 kJ per mol). This suggests that the
dehydrogenation mechanisms of Ti- and Zr-doped NaAlH4 and
Na3AlH6 all involve similar reaction pathways in which the
dominant energetic consideration is the positioning of dopants
rather than the strength of Al-H bonding interactions. In situ
X-ray diffraction8 and scanning electron microscopy17 studies
have shown that the phases arising from the dehydrogenation
of Ti-doped NaAlH4 are composed of large crystallites with

7674 J. Phys. Chem. A, Vol. 107, No. 39, 2003
Kiyobayashi et al.
(5) Jensen, C. M.; Zidan, R.; Mariels, N.; Hee, A. G.; Hagen, C. Int.
J. Hydrogen Energy 1999, 23, 461.
(6) Zidan, R. A.; Takara, S.; Hee, A. G.; Jensen, C. M. J. Alloys Compd.
1999, 285, 119.
(7) Bogdanovic, B.; Brand, R. A.; Marjanovic, A.; Schwickardi, M.;
To¨lle, J. J. Alloys Compd. 2000, 302, 36-58.
(8) Gross, K. J.; Guthrie, S.; Takara, S.; Thomas, G. J. Alloys Compd.
2000, 297, 270.
(9) Zaluska, A.; Zaluski, L.; Stro¨m-Olsen, J. O. J. Alloys Compd. 2000,
298, 125.
(10) Balema, V. P.; Dennis, K. W.; Pecharsky, V. K. Chem. Commun.
2000, 1665.
sizes greater than 100 nm. These findings indicate that processes
must be operating in which the constituent metal atoms are
transported over a long range. Interestingly, the values of ∆Hq
that we have determined in this study are close to the value
reported for the self-diffusion activation energy of metallic
aluminum of 142 kJ‚mol-1 27
This suggests that the rate-
.
determining step in the dehydrogenation of the doped alanates
at lower temperatures may also be a long-range atomic transport
phenomenon.
Studies currently in progress of the rehydrogenation kinetics
should reinforce our conclusions and perhaps provide further
insight. Additionally, microscopic studies of the dehydrogenation
reactions based on the knowledge obtained from our macro-
scopic observations should elucidate the details of the reaction
mechanism(s) and function of dopants.
(11) Jensen, C. M.; Gross, K. J. Appl. Phys. A 2001, 72, 213.
(12) Bogdanovic, B.; Schwickardi, M. Appl. Phys. A 2001, 72, 221.
(13) Chen, J.; Kuriyama, N.; Xu, Q.; Takeshita, H. T.; Sakai, T. J. Phys.
Chem. B 2001, 105, 11214.
(14) Jensen, C. M.; Zidan, R. A. U.S. Patent 6,471,935, 2002.
(15) Gross, K. J.; Thomas, G. T.; Jensen, C. M. J. Alloys Compd. 2002,
330-332, 691.
(16) Gross, K.; Sandrock, G.; Thomas, G.; Jensen, C.; Meeker, D.;
Takara, S. J. Alloys Compd. 2002, 330-332, 696.
(17) Thomas, G. J.; Gross, K.; Yang, N. Y. C.; Jensen, C. M. J. Alloys
Compd. 2002, 330-332, 702.
(18) Sun, D.; Kiyobayashi, T.; Takeshita, H. T.; Kuriyama, N.; Jensen,
C. M. J. Alloys Compd. 2002, 337, L8.
Acknowledgment. The study was administered through the
New Energy and Industrial Technology Development Organiza-
tion (NEDO) as a part of the International Clean Energy
Network using Hydrogen Conversion (WE-NET) program with
funding from the Ministry of Economy, Trade and Industry of
Japan.
(19) Meisner, G. P.; Tibbetts, G. G.; Pinkerton, F. E.; Olk, C. H.; Balogh,
M. P J. Alloys Compd. 2002, 337, 254.
(20) Sandrock, G.; Gross, K.; Thomas, G. J. Alloys Compd. 2002, 339,
299.
(21) Fichtner, M.; Fuhr, O. J. Alloys Compd. 2002, 345, 286.
(22) Bogdanovic, B.; Sandrock, G. MRS Bull. 2002, 27, 712.
(23) Balogh, M. P.; Tibbetts, G. G.; Pinkerton, F. E.; Meisner, G. P.;
Olk, C. H. J. Alloys Compd. 2003, 350, 136.
(24) Glasstone, S.; Laidler, K. J.; Eyring, H. The Theory of Rate
Processes; McGraw-Hill: New York, 1941; p 11.
(25) Ashby, E. C.; Kobertz, P. Inorg. Chem. 1966, 5, 1615.
(26) Young, D. A. The International Encyclopedia of Physical Chemistry
and Chemical Physics Topic 21: Solid and Surface Kinetics Vol 1:
Dehydrogenation of Solids; Pergamon Press: Oxford, 1966.
(27) Borg, R. J.; Dienes, G. J. The Physical Chemistry of Solids;
Academic Press: San Diego, 1992; p 466.
References and Notes
(1) Gravimetric hydrogen density targets of 5.0, 6.0, and 5 wt % have
been set by the Japanese government, WE-NET project (http://www.
enaa.or.jp/WE-NET), the International Energy Agency (http://hydpark-
.ca.sandia.gov/iea.html), and the U.S. Department of Energy (http://
www.eere.energy.gov/hydrogenandfuelcells/mission), respectively.
(2) Bowman, R. C.; Fultz, B. MRS Bull. 2002, 27, 688.
(3) Bogdanovic, B.; Schwickardi, M. J. Alloys Compd. 1997, 253-
254, 1.
(4) Bogdanovic, B.; Schwickardi, M. International Patent WO97/03919,
1997.