Note: Descriptions are shown in the official language in which they were submitted.
~20~
-1- 24-NT-04575
METHO'D FOR CONTROLLED HYDROGEN CHARGING OF METALS
_
FIELD OF THE INVENTION
This invention relates to metal hydrides and
their formation. It is particularly concerned with the
effec'ts of hydrogen upon certain metals and means for
ascertaining same.
BACKGROUND OF THE INVENTION
Metals which are exposed to gaseous hydrogen
or aqueous corrosive media can take up hydrogen from
such environments. The amount of hydrogen that can
dissolve in typical engineering metals such as iron,
aluminum and copper is limited by its solubility in
such'metals, and typically is only in the parts per
million range at room temperatures.
However, with many significant transition
metals such as zirconium, titanium, hafnium and their
alloys, hydrogen can form solid hydride phases in their
metal matrix. The presence of hydrogen within these
metals, or the hydrides formed thereby, can cause
embrittlement therein, and under some conditions
result in their failure.
Investigation of the effect of hydrogen
upon the mechanical and physical properties of metals
requires a reproducible method of charging controlled
amounts of hydrogen into metal being studied. Also,
in systems wherein hydrogen is stored in solid form,
there is a need for efficient means of charging the
hydrogen into a hbst metal.
~'
, ~,
,:
`` 120007~
24-NT-04575
--2--
The most direct method for charging hydrogen into
a metal that forms hydrides is to simply expose the metal
to a dry hydrogen environment in an enclosed chamber at a
temperature of about 200 to 500 . See Transition Metal
Hydrides, ed. by E.L. Muetterties, Marcell Debber, Inc.,
New York, 1971, page 14. This method of hydriding is
extremely sensitive to impurities in the environment,
particularly water vapor, and requires a high vacuum
before hydrogen gas can be introduced into the system.
Also, the surface of the metal must have all oxide removed,
which is especially difficult with active metals such
as zirconium or titanium.
Moreover, metal hydrides initially form at the
surface of the metal and migrate gradually into the
interior thereof. Volume expansion due to hydride form-
ation within the metal can produce spalling and complicate
hydrogen charging whereby the rate of the charging is
unpredictable. Accordingly, this procedure is only
practical for producing metal hydrides in metal particle
form, such as metal chips, and is not suitable for charging
controlled amounts of hydrogen or for forming metal
hydrides with a metal body of specific geometry.
Laboratory methods for charging controlled amounts
of hydrogen to metals require delicate equipment setups,
note Wienstein and Holtz, Transaction of ASM, Vol. 57,
1964, page 254. The metal samples must be heated to high
temperatures, for example, zirconium and its alloys is
heated to over 850C where it undergoes a phase trans-
formation which increases its capacity for dissolving
greater quantities of hydrogen. Controlled amounts of
hydrogen gas are applied to the heated samples whereupon
they are cooled to retain all hydrogen dissolved therein
while hot.
Although such high temperature techniques can
produce controlled quantities of hydrogen within a metal
matrix, the microstructure and in turn the physical and
mechanical properties of the metal sample will be altered
... .
''
, . ~ . , . . - .,
:,
, ,",
-` 12000~
24-NT-04575
--3--
by the high temperatures of the treatment.
Another method of charging hydrogen into a metal
is to corrode the metal in a corrosive environment, such as
aqueous lithium hydroxide (LioH) with zirconium and its
alloys at temperatures of about 300 to 400. See
"Corrosion and Hydrogen Pickup of Zircaloy in Concentrated
Lithium Hydroxide Solutions" by Kass, Corrosion, Vol. 25,
No. 1, January 1969. The corrosion reaction generates
hydrogen which in part enters the metal. However, the
reproducibility of this technique is uncertain because
of the variable susceptibility of metals to corrosion,
and the corrosion attack can deteriorate the surface
of the metal.
BRIED DESCRIPTION OF TEE DRAWING
The drawing comprises a flow sheet diagram
b--- illustrating the basic steps for the practice of the
method of this invention.
SUMMARY OF THE INVENTION
. . . _
This invention comprises a procedure for the
substantially exclusive application of hydrogen gas or
its isotopes comprising tritium and deuterium, to a
selectively defined surface area of a hydridable metal
or article thereof, and in a controlled, determinable
quantity. The deposition of a thin layer or plating
of a non-hydriding metal such as copper only over the
hydriding metal surface portion of the selectively defined
area effectively confines the hydrogen gas, permeation to
such area while inhibiting the permeation of other gaseous
materials such as water vapor or oxygen. By measuring
given conditions of the procedure, the amount of hydrogen
gas penetrating the selectively defined area of the metal
can be determined and in turn controlled.
OBJECTS OF THE INVENTION
It is a primary object of this invention to
35~ provide a method for the charging of hydrogen or its
isotopes to metals wherein the quantity of hydrogen applied
is controllable
:i
- ~
.. ~ ~
,~,: ' "
`-` 1200~
24-NT-04575
_~ _
It is also an object of this invention to
provide a method for applying hydrogen gas selectively to
a defined area of a metal surface, and to the exclusion
of other gases of the immediate atmosphere.
It is another object of this invention to provide
means for investigating the effects of hydrogen upon metals,
and structures thereof.
It is another objective to provide a method for
producing transition metal hydride components of any
geometry with the least dimensional distortion.
It is a further object of this invention to
provide a method for the reproducible charging of hydrogen
gas in controlled amounts for evaluating any changes in the
chemical or physical properties of metals attributable to
hydrogen~
It is another object of this invention to provide
an efficient method of charging hydrogen into metals for
the purpose of storing the hydrogen.
DETAILED DESCRIPTION OF THE INVENTION
This invention comprises a unique method for
charging hydrogen, including its isotopes, to hydride
forming metals, such as, for example, zirconium, titanium,
hafnium and their alloys. The method includes the
application of a thin layer of a non-hydriding metal such
as copper over the portion of a metal surface to be
subjected to the hydriding action, and then subjecting such
portion to hydrogen gas at elevated temperatures.
In accordance with this invention, the surface
portion of the hydride forming metal, or of an article
formed of such metal, is thoroughly cleaned to render it
receptive to the application of a thin metal layer thereover,
such as by plating or other conventional techniques for
metal bonding.
For instance, the portion of the surface of the
hydride forming metal can be prepared by etching with a
hydrogen fluoride and nitric acid solution, e.g., about
2-5% HF and about 30-50% HNO3 with the balance water, or
~,
' :, . . ''
.
,
:, .- . ,. .: - . - ,
^` lZ0(~077
24-NT-04575
--5--
by other common metal pickling procedures.
A thin layer of a non-hydriding metal, such as
copper, is then applied over the cleansed surface portion
of the metal to be subjected to a hydriding treatment.
Conventional metal plating processes can be used for the
application of the thin metal layer, comprising for
example, electrolytic or electroless procedures, such
as disclosed in U.S. Patent No. 4,017,368 issued
April 22, 1977 to Wax et al and No. 4,093,756 issued
June 6, 1978 to Donaghy.
In accordance with a preferred embodiment of
this invention, the thin metal layer such as copper
applied to the hydriding metal surface, should be about
4 to about 20 microns in thickness, and typically about
10 to 15 microns.
Upon the application of such a thin layer of
non-hydriding metal over the surface portion of the
hydriding metal, the said composite is exposed to
hydrogen gas at elevated temperatures. For instance
the metal material, or article thereof, can be placed
within the enclosure of a furnace and subjected to
hydrogen gas. A preferred procedure for this invention
comprises applying an ambient temperature of about 200
to about 400C (392-752F) within an atmosphere of low
pressure hydrogen gas such as supplied with a continuous
flow of gas from a hydrogen source into an enclosing
chamber. The gas flow can be sustained by burning with
a torch arrangement at a gas exit.
In the practice of this invention, the thin
layer of copper or other non-hydriding metal overlying
the surface of the hydriding metal substrata, functions
as a hydrogen diffusion window which enables easy
diffusion of the hydrogen gas therethrough on into the
underlying metal substrata while at the same time
inhibiting diffusion passage therethrough of other gaseous
media such as oxygen or water vapor. In other words the
thin non-hydrizing metal layer performs as a filter that
-;-
.. . . .
;
~2~)007~
24-NT-04575
--6--
permits ready passage of hydrogen gas and retards passage
of oxygen and water vapor. Thus, a preferential charging
of hydrogen by selective diffusion can be attained within
the surface area of the hydriding metal provided with the
overlying non-hydriding metal layer.
Moreover, any oxide of zirconium, etc. occurring
at the interface of the overlying non-hydriding metal layer
and the underlying substrata of hydriding metal will
rapidly dissolve or become permeable whereby hydrogen
gas can penetrate through such as intermediate oxide body.
Further in accordance with this invention, the
amount of hydrogen gas charging attained within the given
area of the hydrogen diffusion window of the non-hydriding
metal layer, can be determined and thereby controlled
by measuring and manipulating certain conditions of the
charging operation. Specifically, each of the following
conditions of the operation should be measured or
determined: (l) the thickness of the layer of non-
hydriding metal overlying the hydriding metal surface;
(2) the ambient temperature of the hydrogen gas application;
and ~3) the time period of the hydrogen application.
Having ascertained these factors, the flux of
hydrogen gas passing through the layer of non-hydriding
metal providing the hydrogen window can be calculated
when expressed as:
_ H ~CH
J - -D d
where DCHu is the diffusivity of hydrogen in copper at test
temperature, ~CH is the hydrogen concentration difference
between copper outer and inner surface (or zirconium outer
surface), d is the plating thickness.
Accordingly, the amount of hydrogen gas applied
in the charging process can be controlled by manipulating
the thickness of the layer of non-hydriding metal, or the
temperature of the hydrogen gas application, or the duration
of the hydrogen gas~application.
~,,
'
.
: , ~
120007~
24-NT-04575
--7--
Typically, it is most convenient to fix the
thickness of the layer of non-hydriding metal overlying
the hydriding metal surface and the hydrogen charging
temperature~ and simply vary the charging time to control
the amount of hydrogen gas charged into the hydriding metal.
Charging temperatures at the lower end of the
preferred range consisting of about 200 to about 500C
are more feasible if the quantity of hydrogen gas to be
applied is low, such as less than about 100 parts per
million. The hydrogen gas flux is limited by diffusion
through the layer of non-hydriding metal, and the
mobility of hydrogen in hydriding metals such as
zirconium alloys is high at temperatures above about
200C. Thus, the hydrogen gas and in turn hydrides will
disperse within the matrix of the hydriding metal rather
than concentrating at its surface. This technique
accordingly provides for the formation of metallic hydride
in any complicated geometry with no significant physical
distortion of the specimen undergoing evaluation. The
charging technique of this invention also constitutes
an economical method for producing metal hydrides.
Moreover, if hydrogen gas charging is desired
only for certain or limited surface area of the hydriding
metal, or article thereof, or if hydrogen charging is
unneeded or detrimental in given surface areas thereof,
a layer of the non-hydriding metal can be applied only
onto a selected portion or portions of the surface of
such hydriding metal or the article, with the remaining
surface portions free of any overlying metal layers.
Then, the hydrogen gas is applied by first passing it
through water whereby it picks up water vapor before
contacting the metal. The water vapor thus entrained in
the continuous supply of hydrogen gas is sufficient to
passivate the exposed surface of the hydriding metal
devoid of the layer of non-hydriding metal, whereby
hydrogen gas is precluded from entering such exposed
surfaces. Thus, as a practical matter, the hydrogen gas
1200077
24-NT-04575
--8--
only enters the hydriding metal surface through the
window provided by the layer of non-hydriding metal.
XAMPLES
Test specimens were r4 1" long, half rings of
8 x 8 zirconium alloy namely Zircaloy-2 of a composition
disclosed in U.S. Patent No. 2,772,964.
The test specimens included samples of copper
plated over the inside diameter of the Zircaloy-2 tubing
half rings.
The tubing test specimens were placed in a
pyrex glass tubing installed in a Lindberg tube furnace.
The temperature of the furnace was measured with a
Chromel-Alumel thermocouple, and was recorded with a strip
chart recorder. A wet hydrogen gas mixture was produced
by passing 1 atmosphere hydrogen gas through a bubbler
containing distilled water at room temperature (22C).
The hydrogen flow rate was controlled at about 330 cc/min
by the flowmeter. The H2/H2O ratio was calculated from
the electrochemical potential outputs of the oxygen meter.
The hydrogen gas was finally burnt at a torch at the line
outlet. Before testing was started, the test line was
continuously flushed with nitrogen gas for 2 hours to
remove any residual oxygen~
Two series of tests were carried out. In the first
test, tubing samples were exposed to a wet hydrogen
environment having H2/H20 ~ 33 + 5 at 355 + 8C for 72
hours. In the second test, the H2/H2O ratio was ~ 18 + 5,
temperature 400 C, and test duration 360 hours. The
difference in H2/H2O in the above tests is likely to be
due to the change in the test temperature.
Post-test evaluation included mainly metallographic
examination. Neutrographic examination was also carried out.
RES`ULTS ~ND DISCUS~SION
1) 355 C - 72 Hour Te~t
-35 The wet hydrogen environment contained water vapor
at about 0.45 psig partial pressure (H2H2O f~ 33), which
- ~2000~7~
24-NT-04575
_g _
significantly exceeded the reported minimum value for
inhibiting Zr-2 hydriding, 0.002 psig. This was confirmed
from metallographic examination of the autoclaved sample,
showing no accelerated hydriding. A vapor deposited
nickel film on an autoclaved test sample, which presumably
would accelerate surface hydrogen dissociation, was
shown to have no significant effect in hydrogen pickup of
the zirconium alloy sample.
Three samples with copper plated on pickled
surface were found to form hydride platelets. The
hydrided samples all had a thin massive hydride layer of
up to about lO m at the interface between the Zircaloy and
the copper layer. The e~uivalent hydrogen content
estimated from the density of uniform hydride platelets
is about 300 ppm for all of the three samples. The
interface oxide in the sample with copper plated on an
autoclaved tubing was found to effectively prevent hydrogen
from entering the cladding. The protection by this
interface oxide is, however, believ0d to be only temporary
as will be discussed below.
The autoclave treatment of a sample with copper
plated on a pickled tubing resulted in some interesting
features. It can be seen that the copper surface is shining
and free of any tarnish. A continuous interface layer,
presumably an oxide, was formed by the autoclave treatment
and the copper barrier was still visually intact. Hydrogen
pickup test of this sample showed that it also resisted
hydriding attack similar to the sample with copper plated
directly on an autoclaved tubing.
II) 400C - 360 Hour Test
The post test surface appearance of the ten samples
tested was investigated. In this test, the partial pressure
of water vapor, about 0.8 psig (H2/H2O 18) is slightly
higher than the value used in the 355C - 72 hour test. The
difference in H2/H2O ratio was not expected to result in
any significant influence in the hydriding of Zircaloy-2.
~.
, :
()0077
24-NT-04575
--10--
Micrographic examination again showed that in this passive
environment all Zircaloy samples without a copper plating,
i.e., autoclaved, pickled, oxidized~ and pickled plus a
50A Platinum film, were resistant to hydriding attack.
Samples with copper plating on either pickled,
oxidized, or autoclaved tubing were heavily hydrided.
Some of the hydrides, exhibit cellular structure. On
the sample with copper on pickled tubing, an oxide-like
layer was found beneath the copper barrier, suggesting
that oxidant has probably diffused across the copper
barrier or from specimen edges to form a zirconium oxide
film as similar to the autoclave treatment discussed
previously. Presence of the interface oxide during
testing, however, did not prevent hydrogen from entering
the Zircaloy. In fact, the preformed autoclaved or
oxidized oxide films beneath the copper barrier all remain
visually intact while hydriding of the Zircaloys beneath
the oxide films occurred. The interface oxide however
seems to retard hydrogen entry to the Zircaloys, where
the density of hydrides decreases with increasing oxide
thickness.
The difference in the results of the copper on
autoclaved samples tested at 355C and 400 C suggests
that the interface oxide provided transient protection to
the Zircaloy. Lacking in oxygen supply from external
sources, the ZrO2 oxide will gradually become sub-
stoichiometric due to diffusion of oxygen into the
Zircaloy. The sub-stoichiometric ZrO2 oxide has been
proposed to be permeable to hydrogen, leading to hydriding
of Zircaloys. The incubation time for Zircaloy hydriding
to occur would therefore decrease with increasing
temperature and with decreasing interface oxide thickness.
Small changes in the value of H2/H2O in the test environ-
ment is not likely to have effect in Zircaloy-2 hydriding
since it is a diffusion-controlled process.
