Author Topic: Making your own hydrides, easily  (Read 2121 times)

0 Members and 1 Guest are viewing this topic.


  • Guest
Making your own hydrides, easily
« on: June 13, 2004, 06:23:00 PM »
Here is an article that I found that might be of interest to some....

Hydriding reactions induced by ball milling in group IV and V transition

Department of Physics, Dalhousie University, Halifax, Nova Scotia, Canada B3H 3J5

The early transition metals are known to absorb substantial
quantities of hydrogen [1, 2]. These metal hydrides
have a number of important commercial applications
such as hydrogen storage materials and electrodes
for batteries [2]. Particularly, in recent years there has
been interest in titanium hydride for these purposes because
of its high storage capacity [1]. However, such
applications have been limited because of the difficulty
in preparing metal hydrides. Typically hydrogenation
is performed by exposing a metal sample to H2 gas at
elevated temperature and/or pressure for several hours.
This is followed by slow cooling in order to maintain
maximum hydrogen content. The observation of the
preparation of nitrides by the ball milling of metals under
a nitrogen containing atmosphere [3, 4], suggest
the possibility that metal hydrides may be produced by
milling under an atmosphere of hydrogen. This process
has recently been reported for titanium [5–8] as well as
Mg and Zr [7, 8]. In the present work we report on investigations
of hydrogen absorption during ball milling
of all of the group IV and V early transition metals (Ti,
V, Zr, Nb, Hf and Ta). For these experiments we have
developed a hydrogen reservoir system which allows
us to maintain the pressure of the hydrogen at approximately
one atmosphere throughout the milling process.
We are also able to continuously monitor the amount
of hydrogen that is absorbed and demonstrate here that
the hydrogen absorption rates during the ball milling are
much higher than those obtained using other methods.
The starting material for all experiments reported
here was¡325 mesh elemental powder with a purity of
2N (Ti), 2N5 (V), 2N (Zr), 2N8 (Nb), 2N6 (Hf) and 3N
(Ta). Milling was carried out using a Spex Model 8000
ball mill. A custom made hardened steel vial was used.
This had a chamber that was approximately ellipsoidal
in shape in order to avoid problems with the powder becoming
caught in the corners of the more conventional
cylindrical shaped vial. In all cases a 1.0 g sample and
two 11.1 mm diameter hardened steel balls with a total
weight of 11.34 g were used. During milling the vial
was connected to a hydrogen gas reservoir system by
means of a Norton Masterflex 6104-24 tygon tube. The
total volume of the hydrogen systemwas approximately
8£103 cm3. For a nominal starting pressure of 1 atm,
the absorption of 2 hydrogen atoms per metal atom for
a 1 g sample corresponds to a decrease in the reservoir
pressure of between about 6% (for Ti) and about 1.5%
(for Ta), thus maintaining an approximately constant
hydrogen pressure on the sample during the milling process.
The pressure was measured by an Omega Model
PX303-015ASV strain gauge type pressure transducer.
The change in pressure in the system is related to the
number of hydrogen atoms absorbed per metal atom,
H=M as
2V A1P
where V is the volume of the reservoir system (in cm3),
A is the atomic weight of the metal, 1P is the pressure
change in the system (in atmospheres), R is the standard
volume of gas (in cm3) and m is the mass of the metal
sample (in g). In some cases the temperature of the
vial was monitored during milling by a type K thermocouple.
The structure of samples was determined using
room temperature X-ray diffraction measurements. A
Siemens D500 scanning diffractometer and CuK® radiation
was used.
The number of hydrogen atoms per metal atom
(H=M) absorbed as a function of milling time is illustrated
in Fig. 1 for each of the samples studied here.
In all cases the hydrogen absorption curves are characterized
by an initial period during which there is little
or no hydrogen absorption. This is followed by a period
of rapid hydrogen absorption and saturation for longer
times. In order to quantify these results we have defined
three quantities as given in Table I. These are the maximum
hydrogen absorption rate, (@H=@M)max, defined
as the maximum slope of a line tangent to the absorption
curve, the time for the onset of hydrogen absorption, t0,
defined as the H=M D0 intercept of the line tangent to
(@H=@M)max, and the saturation value H=M, given by
the asymptotic value of H=M for long milling times.
In general it is seen that those metals which have large
TABLE I Saturation values of H=M, maximum hydrogen absorption
rates (@H=@M)max, and time before absorption begins, t0, for the elements
studied here. The mean crystallite size as determined by X-ray
diffraction peak widths is given by hr i
M (H=M)sat (min¡1) t0(min) hr i (nm)
Ti 2.0 0.158 10 8
V 0.8 0.030 54 5
Zr 2.0 0.18 5 7
Nb 0.9 0.071 21 6
Hf 1.8 0.12 17 7
Ta 0.7 0.10 18 9
values of t0, also have small values of (@H=@M)max.
This relationship is seen for the metals studied here in
Fig. 3.
The structure of the samples milled to saturation have
been determined by X-ray diffraction techniques. A
typical pattern is illustrated in Fig. 2. In all cases the
patterns of the samples milled to saturation are consistent
with those of metal hydrides with the stoichiometry
as indicated in Table I and do not show the presence of
any elemental metal. All X-ray diffraction patterns of
milled samples show measurable line broadening indicative
of small crystallites. The average crystallite
size as determined by the Scherrer formula is given in
Table I.
Although the bulk diffusion rate of hydrogen in early
transition metals is high [5], diffusion rates during conventional
hydrogenation are limited by the presence of
surface oxides. The greatly increased diffusion rates observed
during ball milling has been attributed to several
factors [5, 7].
(i) the formation of a large amount of oxide-free
(ii) the rapid reduction in grain size and
(iii) the introduction of significant lattice defects.
In the first case, the chemisorption of H2 molecules
and their dissociation to form H atoms is promoted. In the second case, the reduced grain size means shorter
diffusion lengths are necessary. It is also relevant that
the embrittlement of the metals, which results from hydrogen
absorption, further increases the effectiveness
of the milling process at reducing the size of the grains.
Finally it is known [8] that the introduction of lattice defects,
such as dislocations, provides convenient routes
along which hydrogen can diffuse through the metal.
In the present study the initial period during which
there is little or no hydrogen absorption by the sample
corresponds to a period during which sample grains are
reduced in size and lattice defects are introduced. This
reduces the diffusion length necessary for hydrogen absorption
and introduces lattice defects such as dislocations
along which hydrogen can readily diffuse. This is
followed by a period during which hydrogen actively
diffuses into the sample. The rate at which hydrogen
diffusion occurs does not show any direct relationship
with literature values of the bulk hydrogen diffusion
coefficient [1]. However, Fig. 3 illustrates an inverse
relationship between the values of (@H=@M)max and
t0 measured in the present work. This suggests that the
maximum rate at which hydrogen diffuses into the sample
is limited by the rate at which fresh surfaces and lattice
defects are produced by milling. Presumably, the
effectiveness of the milling process at promoting hydrogen
absorption is determined to a large extent by
the degree to which the partially hydrogenated sample
becomes embrittled. The average grain size of the fully
hydrogenated samples produced in the present investigation
is in the range of 5–9 nm, as indicated in Table I.
This is somewhat smaller than that reported in earlier
work for TiH2 produced by ball milling [5]. This suggests
that the continuous supply of hydrogen gas used
in the present work may have been more effective at
promoting rapid hydrogen diffusion which resulted in
more substantial grain size reduction and faster saturation
times. Thus rapid hydrogen absorption during ball
milling appears to begin at some threshold grain size
and/or defect concentration and proceeds to saturation
as a result of continued particle size reduction.
From a commercial standpoint the present results are
important as they indicate that the diffusion of hydrogen
into the early transition metals during ball milling
is very rapid provided the hydrogen pressure is maintained
at a reasonable level throughout the milling process.
This method is, therefore, a simple and efficient
means of producing single phase hydrides from elemental
early transition metal precursors.

1. G. ALEFELD and J . VOLKL, “Hydrogen in metals” (Springer
Verlag, Berlin, 1978).
2. K. M. MACKAY, “Hydrogen compounds of the metallic elements”
(Spon, London, 1965).
3. A. CALKA, Appl. Phys. Lett. 59 (1991) 1568.
4. A. CALKA and J . S . WILLIAMS, Mater. Sci. Forum 88–90
(1992) 787.
5. D. A. SMALL, G. R. MACKAY and R. A. DUNLAP,
J. Alloys and Compounds, submitted.
6. H. ZHANG and E. H. KISI, J. Phys. Condens, Matter 9 (1997)
7. Y. CHEN and J . S . WILLIAMS, J. Alloys and Compounds 217
(1995) 181.
8. Idem., Mater. Sci. Forum 225–227 (1996) 881.


  • Guest
Almost rated as incomprehensible
« Reply #1 on: June 13, 2004, 07:20:00 PM »
Could you please edit the typography of the article for clarity?

It is very hard to read when you cut/paste a whole article without removing any trash characters, no tables edited, not dividing it into paragraphs - and last but not least posting it with hard line breaks at 60 columns???