Note: Descriptions are shown in the official language in which they were submitted.
0~ 8 S ~ ~
Background o~ the Invention
1. Field of the Invention:--This invention relates to the
art of providing membranes for use in chlor-alkali elec~
trolysis membrane cells~ and in particular, to a method o~
pre-treating said membranes, before their insertion into a
cell, t~o improve the quality of said membranes
2. Description of the Prior Art:--The use of membrane-type
- electrolysis cells for the electrolysis of brine, producing
chlorine, hydrogen, and sodium hydroxide, is well known, as
for example, from U S Patent No 2J967J807. I~ is well
known that such membrane-type cells can be made by using a
sheet or film, approximately 0 1 to 0.25 or 0.5 millimeters
(about 4 to 10 or 20 mils) thick, of a copolymer of tetra-
fluoroethylene and sulfonylfluoride perfluorovinyl ether.
Suitable copolymer material is disclosed in U S. Patent
No. 3,282,875. Such cells offer an attractive alternative
to the customary diaphragm-type cells, using a diaphragm made
of asbestos or the like, because of the health hazards posed
by the manufacture and use of asbestos It is known that
such membranes have a tendency, when put into service, to
swell, thereby crea~ing water domains through which
hydroxide ions are transported much more readily than
sodium ions, owing to the Grotthus mechanism. I am not
aware that anyone has hitherto proposed any method or
-2-
S~4
practiceJ by means of which such swelling of the membrane
may be reduced, with corresponding favorable e~fects upon
the effective selectivity of the membranes and upon the
current efficiency o~ the chlor-alkali membrane cells ln
which they are used.
Summary of the Invention
By subjecting them to a heat treatment at 100 to
275 Centigrade for a time of several hours to four minutes,
membranes for use in chlor-alkali cells are given improved
properties: they contain less water, have improved selec-
tivity, exhibit higher current efficiency and lower power
consumption per unlt of product, and afford a product having
a lower salt content. It is also desirable, during ~he heat
i treatment, to apply a pressure of up to 9.76 kilograms per
square centimeter (10 tons per square foot).
Description of the Preferred Embodiments
The present invention is practiced upon membranes
for use in membrane-type chlor-alkali cells for the elec-
trolysis of brine to produce alkali-metal hydroxide,
chlorine3 and hydrogen. In particular, it is practiced on
membranes that are made of a copolymer of tetrafluoro-
ethylene and sulfonated perfluorovinyl ether, such as a
copolymer of tetrafluoroethylene and sulfonylfluoride per-
fluorovlnyl ethers. Such material is commercially available
7 ~ 56 ~
for use in such cells in the form of membranes having a
thickness ordinarily on the order of 0.10 to 0.25 milli-
meters ~4 to 10 mils) and having an equivalent weight
number on the order of 1000 to 1500. To improve ~he
strength of the membrane, some of the membranes are provided
with reinforcement of polytetrafluoroethylPne or the like;
others are not. Such membranes are useful ln their un-
treated condition~ but by the practice of the present in-
vention, their performance can be considerably improved.
In the practice of the invention, a membrane to
be treated is preferably placed between a pair of sligh~ly
larger thin sheets of polytetrafluoroethylene, to insure
against having the membrane adhere to anything with which
it is in contact during the thermal treatment. A conven-
ient way of practicing the invention is to insert the sand-
wich thus prepared into a hydraulic press having a pair of
electrically heated flat plates, and while exerting some
; pressure upon the sandwich, to bring it to a desired tem-
perature and hold it at such temperature for an appropriate
period of time. Satisfactory results have been ob~ained
without the exertion of any pressure, but in most instances
it is desirable to use a small pressure, such as o.976 to
4.88 kilograms per square centimeter (l to 5 ton8 per
Y8~;~;4
square foot). Pressuresas grea~ as 9.76 kilograms per
square centimeter (10 tons per square foot) can be used.
The duration of the heat treatment depends upon
the temperature. At a high temperature, such as 275 Centi-
grade, a short time such as four to five minutes is suf-
ficient, whereas at a low temperature such as 100 Centi-
grade, a time of several hours may be required. Preferably,
the temperature used is between 175 and 225 Centigrade, and
at that temperature, a time of five to twelve minutes is
satisfactory. Pre~erred results are obtained with the use
of a temperature of 200 Centigrade for seven minutes.
After the thermal treatment, the membrane i5 -.
allowed to cool to room temperature. Rapid cooling (one min-
or less) is acceptable, but a slower cooling rate (about
fifteen minutes) i~ preferred.
The trea~ed membrane is then inserted into a
chlor-alkali cell and used in the same manner as an un-
treated membrane~
The invention described above is illustrated by
the following specific examples.
Example 1
A 0.125-millimeter (five-mil) thick piece of
polytetrafluoroethylene-reinforced membrane materialJ made
~5~
.. , .. ~ , . ...
" ~ 7 8 ~ ~
of a copolymer o~ tetrafluoroethylene wlth sulfona~ed per-
fluorovinyl ether and having an equivalent weight number o
about 1100, was boiled briefly in a 1 Normal aqueous solu-
tion of hydrochloric acid and then removed. In this state,
it could have been inserted directly into a chlor-alkali
cell. The piece was wiped dry, sandwiched between two
sheets of polytetrafluoroethylene, and placed into a hy-
draulic press that had been pre-heated to 225 Centigrade.
A pressure of 6.83 kilograms per square centimeter (7 tons
per square foot) was then applied for a period o five
minutes, and the membrane was then allowed to cool in the
press after the pressure had been released. This took about
15 minutes. The membrane was removed from between the
sheets of polytetrafluoroethylene and inserted into a chlor-
alkali cell having dimensionally stable anodes and steel
cathodes. The cell was then operated at a cell current of
25 amperes. Saturated brine having a pH of 4 was fed to
the anode compartment at a rate of about 200 milliliters
per hour7 and 80 milliliters per hour of water were fe~ to
the cathode compartment, which produced an 18 weight percent
aqueous solution of sodium hydroxide. The cell operated
at ~.85 volts and with a current efficiency of 78 percent.
The energy consumption was 1~2 wat~-hours per mole of
sodium hydroxlde.
.
-6-
.
~7~ 5~ ~
For comparison, a similar membrane was inser~ed
into a similar chlor-alkali cell, immediately after having
been boiled brie~ly in hydrochloric acid. This chlor-alkali
cell was operated under substantially th~ same conditions,
exhibiting a cell voltage of 3.~5 volts, a curren~ effi-
ciency of 59 percent, and an energy consumption o~ 152
watt-hours per mole of sodium hydroxide. The thermal treat-
ment according to the invention increased the current efi-
ciency from 59 percent to 78 percent, and it lowered the
energy consumption from 152 to 132 watt-hours per mole.
Example 2
Example 1 was repeated, except that (1) the mem- -
brane was 0.178 millimeters (7 mils) thick and had an
equivalent weight number of 1200, and (2) the temperature
used in the thermal treatment was 250 Centigrade. Again,
for comparison, the results with an identical but untreated
membrane were observed. The ~reated membrane gave current
efficiency of 82 percent, a cell voltage of ~.7 volts, and
an energy consumption of 121 watt-hours per mole. When the
treated membrane was used for a period of ~ive months3 the
current efficiency remained at about 80 percent. The un-
treated membrane gave a current eficiency of 69 percent,
a cell voltage of ~.85 volts, and an energy consumption of
. ~ . . . ............................ . .
,
~ 85~4
149 watt-hours per mole.
Example 3
An unreinforced membrane of 1200 equivalent
weight number and having a thickness of 0.254 millimeters
(10 mils) was thermally treated at 250 Centigrade for five
minutes~ It was then inserted into a cell, as in Example 1,
and used to produce an aqueous solution containing 18 weight
percent of sodium hydroxide. The current efficiency was
82 percent, the cell voltage was 3.3 volts, and the energy
consumption was 108 watt-hours per mole. The sodium
chloride content of the product from the cell containing
the treated membrane was 200 milligrams per liter.
In comparison, when a substantially identical but
! untreated membrane was used, the current efficiency was
64 percent~ the cell voltage was 3.1 volts, and the energy
consumption was 130 watt-hours per mole. Moreover, the
hydroxide product contained 1.5 grams per liter of sodium
chloride.
- Example 4
Copolymerized tetrafluoroethylene and sulfonated
- perfluorinated vinyl ether of equivalent weight number 1350
was used to prepare a reinforced membrane 0.1016 millimeters
(~ mils) thick. The membrane was thermally treated a~ 225
.. ..
-8- ~
..
. . .. ~ . : .
.
~ ~ 7~ S ~ ~
Cen~igrade for five minutes. When lnserted lnto a cell,
the thermally treated membrane gave a cell voltage of 4.0,
a current efficiency of 90 percent, and an energy con-
sumption o~ 120 watt-hours per mole. In comparison, an
untreated membrane gave a cell voltage of 3.3, a current
efficiency of 68 percent, and an energy consumption of
130 watt-hours per mole.
! Example 5
Example ~ was repeated, except that the thermal
treatment was conduc~ed at 200 Centigrade for four minutes
and at a pressure of 2.928 kilograms per square cen~imeter ;
(three tons per square foot). When tested in a chlor-alkali
cell, the resulting membrane gave a current efficiency of
82 percent, the same as in Example 3, in which the un~reated
- membrane had a current efficiency of 64 percent.
Example 6
Example ~ was repeated, except that the thermal
; treatment was conducted in an oven at 200 Centigrade for
~hirty minutes without any pressure. When tested in a
chlor-alkali cell, the mem~rane so treated gave a current
efficiency of 78 p~rcent.
" ~Z .
Example 3 was repeated, except that the thermal
.
.
_9
.
.
56~
treatment was conducted in an oven at 110 Centigrade or
a period of ~our hours. When tes~ed in a chlor~alkali cell,
the membrane gave a current efficiency of 77 percent.
In addition, tests were conducted wi~h respect to
membranes of 1100 and 1200 equivalent weight num~er to
demonstrate that a thermal treatment in accordance with the
present invention yields a membrane which has a decreased
tendency to absorb water (in comparison with an untreated
membrane). Treated and corresponding untreated ~embranes
were brought to equilibrium in an atmosphere having 50
percent xelative humidity, and the water contents of the
membranes were then determined by the base catalyzed,
extrapolated Karl Fischqr me~hod. For membranes of 1100
equivalent weight number, the wa~er contents were 8.72
weigh~ percent for the untreated and 7.50 weight percent
for the treated, a difference of 14 percent. For membranes
of 1200 equivalent weight number, the water contents were
7.45 weight percent for the untreated and 6.55 weight percent
for the treated, a difference of 12 percent.
While there have been shown and described herein
certain embodiments of the invention, it is intended that
- - there be covered as well any change or modification therein
which may be made withou~ departing from the spiri~ and ;
scope of the invention. -~
Membrane~ treated as herein taught may find other
uses in which greater selectivity is wanted.
-10- , . ... .
.
. . . .
. . . . ..
~ 1~785ti~
SUPPLEMENTARY DISC~OSURE
_ . _ _ . _ . .
In the application as originally iled, a method ~or
obtaining a selective membrane for use in chlor-alkali cells
has already been disclosed. Said membrane prior to use is
subjected to a particular thermal treatment.
Now, it has been found that the thermal treatment
may be conducted at a temperature of 175 to 225 Centigrade,
for a time of three hours to one half hour. Preferred results
are obtained with the use of a temperature of 200Centigrade
for two hours.
Further, after the thermal treatment, the membrane
is allowed to cool to room temperature. Rapid cooling (one
minute or less) is acceptable, but a slower cooling ra-te ~at
least 15 minutes, and up to several hours, preferably about
2 or 3 hours) is preferred.
As a result of such thermal treatment, the membranes
exhibit improved properties: they have improved selectivity,
give higher current efficiencies and lower power consumption
per unit of product obtained, and afford a product having a
lower salt content.
The underlying reason for these changes is that as
a result of the heat treatment, there occurs a morphological
transition in the membrane material. This can be seen clearly
from X-ray diffraction data upon membranes in their untreated
and treated states. Untreated, the membrane is characterized
by two lacttice constants, one at 5.7 Angstrom units and one
at 34 Angstrom units. The former is attributable to the lateral
spacing of the polymer chains. The latter is related to the spa-
cing of the sulfonic acid groups. Treated, the membrane exhibitslattice constants of 5.7, 27, and 140 Angstrom units. The X-ray
diffraction data demonstrate that the spacing between sulfonic acid
~ 'r
groups has heen diminished~ as is evidenced b~ the decrease in
lattice constant from 34 to 27 Anys-trom units. Those skilled in
the art of ion-exchange membranes know ~hat closer spacing o
sulfonic acid groups means better membrane selectivity.
Moreover, the appearance of an overstructure with a
spacing of 140 Angstrom units indicates that, af~er treatment,
there is a more regular ordering of the resin. Those skilled
in the art will again appreciate that the more regular ordering
can be expected to improve the selectivity of the membrane and
its other mechanical and transport properties. Indeed, the
treated membrane, as compared to one untreated, was 25% higher in
tensile strength and 50% lower in permeability for gases.
The new reacting conditions of preparing the membrane
will now be further understood by means of the following non-res-
trictive example.
Example 8
Example 6 of the original application was repeated,
except that the treatment at 200 Centigrade was or two hours.
The curr~nt efficiency was 81%.
- 12 -
. . - . .
