Letters
J. Phys. Chem. B, Vol. 109, No. 9, 2005 3721
Consequently, the variation of equilibrium pressure with tem-
perature should display a lower enthalpy, i.e., a lower slope,
above ∼360 °C. The measured equilibrium pressure at 450 °C
is indeed lower than the pressure extrapolated from lower
21
temperatures. These data points may indicate a transition from
reaction 1 at temperatures below ∼360 °C to reaction 2 at higher
temperatures. However, additional data are necessary to clearly
resolve this transition. At higher capacities, above approximately
8.5 wt %, a second plateau would be expected for the isotherms
at 400 and 450 °C corresponding to hydrogenation/dehydroge-
nation equilibrium of Mg. Thus far, we have not detected a
distinct second plateau. However, the slope of the isotherm at
4
00 °C gives an equilibrium pressure of ∼19 bar at 9 wt %,
Figure 4. Van’t Hoff plots for destabilized LiBH
4
+ 1/
2
MgH
2
, pure
which is close to the expected pressure for MgH2/Mg. Although
two plateaus have not been observed in the isotherm data, the
temperature ramp desorption measurements (see Figure 1b)
show two desorption steps that may correspond to dehydroge-
nation of MgH2 followed by reaction of Mg with LiBH4 to form
MgB2. As mentioned above, dehydrogenation under vacuum
results in formation of Mg, not MgB2. Thus, it appears that a
finite hydrogen pressure, approximately 2 bar after the first
desorption step for the experiment shown in Figure 1b, is
necessary for the subsequent reaction of Mg with LiBH4 to yield
MgB2. Further work is underway to completely understand this
behavior.
Extrapolation of the equilibrium pressures at 315-400 °C to
lower temperatures gives a pressure of 1 bar at 225 °C. Direct
measurements at this temperature have not been possible because
the kinetics are currently too slow. Following work with
NaAlH4, 2-3 mol % TiCl3 was added as a catalyst because
initial experiments without added TiCl3 displayed poor perfor-
mance. However, the catalyst composition and system process-
ing, i.e., milling conditions and particle and crystallite sizes,
must still be optimized to obtain reasonable reaction rates at
low temperatures and to realize the potential of this system for
hydrogen storage. In addition, a temperature of 225 °C for an
equilibrium pressure of 1 bar is higher than desired for most
applications. Ideally, an equilibrium pressure of 1 bar would
occur at e150 °C. Thus, further thermodynamic destabilization
is also required.
LiBH
the absorption isotherms at 4 wt %. A linear fit to the data at 315-400
C indicates a dehydrogenation enthalpy of 40.5 kJ/(mol of H ) and an
equilibrium pressure of 1 bar at 225 °C. Curve b shows an estimate of
to LiH + B.15 Curve c
the behavior for dehydrogenation of LiBH
shows the equilibrium pressure for MgH /Mg from ref 20. Addition of
MgH increases the equilibrium pressure by approximately 10 times
while lowering the enthalpy by 25 kJ/(mol of H ) compared with pure
LiBH
4 2
, and MgH . Curve a shows equilibrium pressures obtained from
°
2
4
2
2
2
2
2
4
.
varied from 4.5 bar at 315 °C to 19 bar at 450 °C. Absorption
and desorption isotherms obtained at 400 °C display a hysteresis
of 2-3 bar. For all of these measurements, the kinetics were
slow and times up to 100 h were necessary to attain equilibrium.
Because of the slow kinetics, only a single point was obtained
at 315 °C.
A preliminary van’t Hoff plot (logarithm of the equilibrium
pressure versus the inverse of the absolute temperature) using
absorption equilibrium pressures at 4 wt % (see Figure 3) is
19
shown in Figure 4. From 315 to 400 °C the behavior is linear
with an enthalpy of 40.5 kJ/(mol of H2) and an entropy of 81.3
J/(K mol of H2). At 450 °C (1000/T ) 1.38) the equilibrium
pressure is lower than the pressure predicted on the basis of an
extrapolation of the linear behavior at lower temperatures (see
discussion below). Also shown in Figure 3 are equilibrium
pressures for MgH2/Mg obtained from the IEA/DOE/SNL
2
0
database and an estimate of the equilibrium pressure for
15
dehydrogenation of pure LiBH4 to LiH + B. The enthalpy
for the LiBH4/LiH + B system is estimated to be 67 kJ/(mol of
H2). Compared with pure LiBH4, the hydrogenation/dehydro-
1
In summary, we have shown that addition of /2MgH2 to
LiBH4 yields a destabilized, reversible hydrogen storage material
system with a capacity of approximately 8-10 wt %. The
hydrogenation/dehydrogenation enthalpy is reduced by 25 kJ/
1
genation enthalpy for the LiBH4 + /2MgH2 system is lower
by 25 kJ/(mol of H2) and at 400 °C the equilibrium pressure is
increased from approximately 1 to 12 bar. Alternatively,
extrapolating the linear behavior gives a temperature of 225 °C
for an equilibrium hydrogen pressure of 1 bar. Overall, the
(
mol of H2) compared with pure LiBH4, and the tempera-
ture for an equilibrium pressure of 1 bar is estimated to be 225
C.
°
1
equilibrium pressure indicates that addition of /2MgH2 signifi-
cantly destabilizes LiBH4 for hydrogen storage.
References and Notes
Interestingly, the equilibrium pressure behavior for the LiBH4
(
(
1) Grochala, W.; Edwards, P. P. Chem ReV. 2004, 104, 1283.
2) Bogdanovi c´ , B.; Schwickardi, M. J. Alloy Compd. 1997, 253-254,
+ 1
(
/2MgH2 system crosses the curve for MgH2/Mg at ∼360 °C
1000/T ) 1.57). At temperatures below 360 °C the equilibrium
1.
pressures are greater than those for pure MgH2. Thus, in addition
to LiBH4, the MgH2 is also destabilized. In this region the
combined LiBH4 + /2MgH2 system has equilibrium pressures
higher than either individual component. Above 360 °C the
equilibrium pressures, obtained from the isotherms at 4 wt %,
are below the equilibrium pressures for MgH2/Mg. Under these
conditions, the system reacts according to
(3) Sandrock, G.; Gross, K.; Thomas, G. J. Alloy Compd. 2002, 339,
(4) Jensen, C. M.; Gross, K. J. Appl Phys. A 2001, 72, 213.
2
99.
1
(
02.
5) Chen, P.; Xiong, Z.; Lou, J.; Lin, J.; Tan, K. L. Nature 2002, 420,
3
(6) Chen, P.; Xiong, Z.; Lou, J.; Lin, J.; Tan, K. L. J. Phys. Chem. B
2003, 107, 10967.
(7) Ichikawa, T.; Isobe, S.; Hanada, N.; Fujii, H. J. Alloy Compd. 2004,
3
65, 271.
(8) Luo, W. J. Alloy Compd. 2004, 381, 284.
9) Leng, H.; Ichikawa, T.; Hino, S.; Hanada, N.; Isobe, S.; Fujii, H.
J. Phys. Chem. B 2004, 108, 8763.
10) Orimo, S.; Nakamori, Y.; Kitahara, G.; Miwa, K.; Ohba, N.;
Noritake, T.; Towata, S. Appl. Phys. A 2004, 79, 1765.
1
1
3
LiBH + / Mg T LiH + / MgB + / H
2
(2)
4
2
2
2
2
(
Because hydrogenation of Mg is exothermic, the enthalpy for
reaction 2 should be less than the enthalpy for reaction 1.
(