Note: Descriptions are shown in the official language in which they were submitted.
B
The present invention relates to a device and a method
for monitoring a component in a fluid mixture and more
particularly to such a device which utilizes a solid electrolyte.
~ork on the properties of solid electrolytes has been
going on in many countries since the last century. Haber and
St. Tolloczko were the first to study quantitatively the
chemical changes of solid electrolytes and to find that they
obeyed Faraday's law. Haber and others built high temperature
c~ncentration cells using solid electrolytes which measured the
e.m.f. produced by reactions between gases such as Co and 2 on
one side of the cell keeping a constant 02 concentration on
the other side. This enabled the calculation of thermodynamic
data for the gas at different temperatures. It was proposed to
use this method for power generation1 fuel cells, and for
gas analysîs.
The validity of the Nernst equation:-
e.m.f. = RT ln C2 ............... ~1)
NF C
where:
R - Universal gas constant
T - Absolute temperature
F - Faraday constant ,~
N - Number of electrons transferred
C~ - Concentration of a component on one side
of electrolyte.
C2 - Concentration of same component on the
other side of electrolyte.
(The e.m.f. is proportional to the chemical potential
of the component under equilibrium conditions which
in turn is related to its concentration or its partial
pressure in gases).
jl/i -2-
. . :: . ,......... -
. . : . ~ , : . ~ . .:, ~ - . , :
, , . ~ .. . . . . . ~
: - , .. : :- ' . : -. .
: :. -
was the first demonstrated for solid electrolytes by Katayama
who used amalgam concentration cells of the form (Hg + Pb)
Cl ¦Solid electrolyte¦ (Hg + Pb) C2
lead Bromide
By choosing an electrolyte system with the following
features:
i) does not form deposits on the electrodes,
ii) has the same basic reaction on either electrode,
iii) has nearly pure ionic conductivity, and
iv) has a conducting ion which is related to the
component of interest,
cells of the types mentioned can be utilized in a variety of
modcs:-
a) as concentration cells - if the concentration of a
component on one side is known (reference) the output e.m.f.
will be related to the concentration of that component on the
other side, concentration meter, thermodynamic data, etc.
b) as fuel cells for electric power generation, and
c) as pumps - if an electronic current is passed through
the electrolyte with the appropriate component having concentration
Cl and C2 on either side a transfer of part of the component from
one side to the other is effected, the extent of which is
primarily determined by the amount of electric current passed.
For each of these modes, the cell design, the electrolyte
material, and the electrode material is a matter of consideration.
Extensive use was made of solid electrolytes in the
construction of galvanic cells to gather thermodynamic data and
in the construction of fuel cells. Disc shaped solid
electrolytes made of solid solutions such as CaO in ~rO2 and
having oxygen ion vacancies in cells of the type A, A(O) /
solid electrylyte / B, B(O)
86~
where,
A(0) - metal oxide of metal A,
B(0) - metal oxide of metal B,
have been used to determine the molar free energy of formation
for a variety of oxides, sulfides and tellurides at elevates
temperatures. Their work re-generated interesting the use
of mixed crystals as solid electrolytes.
Now Nernst had observed, at the turn of the century, the
evolution of oxygen at the anode whilst passing a ~.C. current
in his "glow bar" element, which he usually made of mixed
crystal solid solutions such as 0.85 ZrO2 0.15 Y2 03 . Wagner
and Schotky related, thermo dynamically, the ionic defect
concentrations of a compound (ionic or solid solution), and
the deviation of the composition from exact stoichiometry,
to the activity of the component in the surroundings. Wagner
derived the expression:-
e.m.f = -1 1 ~"02 (tion) d~02 ............. (2)
~F .
_ 11 1 0 2
where,
tion - sum of ionic transference numbers of
the electrolyte.
F - Faraday constant.
02 - Chemical potential of oxygen at cathode.
~1'02 - Chemical potential of oxygen at anode.
for the e.m.f. produced in a galvanic oxygen concentration cell
involving a mixed conduction solid electrolyte.
Wagner a]so explained the electric conductivity of the
Nernst "glow bar" as oxygen ion conduction resulting from large
jl/~,~ -4-
- .
- ~ : ~ - ;'
6v~3
concentration of mobile oxygen vacancies in the lattice. Hund who studied
the density and x-ray diffractions of such mixed oxide solid electrolytes,
found that a lower valent cation substituted in the lattice result in
oxide vacancies providing thus a path for diffusing oxygen ions. Weininger
and Zeemany were the first to demonstrate quantitatively that the oxygen
ion is the carrier in solid electrolytes such as 0.85 Zr 02 0.15 Y203 by
measuring the evolved oxygen and correlating it to the current passed
through the electrolyte. Kingery and co-workers employed the stable
isotope 018 and mass spectrometer analysis to determine the oxygen ion
mohility in the cubic flourite-structure phase of the solid solution
0.85 ZrO2 0.15 CaO and found it to be near unity.
Interest in oxides such as zirconia, which early in the century was
directed to its possible uses as a refractory material, had shifted by
the fifties to its characteristic as a solid electrolyte and workers
from many countries contributed to this knowledge. This enabled workers
such as Peters and Mobuis to improve the design of the Haber basic gas
concentration cell using mixed oxide solid electrolyte discs - ThO2, La203,
ZrOz and Y203 and to use it to investigate the equilibria.
Co + 1/2 02 ~ Co2
and C + Co2 ~ 2Co at difEerent temperatures -
1000 to 1600 K -. Peters and Mobuis disclose in German Democratic
Republic Patent No. 21,673, granted August 7, 1961, a practical design
of a gas analyse based on such a cell which operates at high temperatures
and uses either a gas or a sealed metal/metal oxide mixture as a reference.
Ruka et al US Patent No. 3,400,054, issued September 3, 1968, also
describes a cell construction capable of being used as a fuel cell, as
an 2 from gas separator or as an 2 partial pressure sensor.
~ -5-
; ~
:; . , :, ,~
:: , ~ . : :: :
36~
As operational experience was gathered from the use of
such devices, many practical problems appeared and many
improvements have been proposed. The most serious of these
problems are:
a) Fragility and fracture due mainly to the use of
large size ceramic units with low thermal conductivity, to
the presence of temperature gradients across the ceramic,
and to bad thermal matching of materials;
b) large errors due to active sensor areas being within
a temperature gradient, to different temperatures at the
sample and reference sides, to mounting the temperature sensor
at a position different from the active sensor area, and to
leakage across seals;
c) errors due to sample and reference components not
reaching the equilibrium temperature;
d) complex designs leading to high manufacturing cost,
and difficulty in manufacturing and servicing;
e) non-versatile design leading to specialized sensors.
In one particular aspect the present invention provides
a concentration cell having an ion-conductive partition,
first ion-supplying means arranged to apply ions to one side
of said partition for conducting ions to the other side of
said partition, second ion-supplying means arranged to apply
ions to said other side of said partition for conducting
ions to said one side of said partition, and regulating
means for causing said first ion supplying means to supply
ions to said one side of said partition in such concentration
as to maintain the net conduction of ions across said partition
at a predetermined fixed value, wherein said first ion-
supplying means is a chamber having said one side of said
J~ 6-
' . . ;
~86~
partition providing at least a portion of the inner surface
of the wall of said chamber and also includes a chemical
means for producing a concentration of gas inside said chamber
in proportion to the temperature of said chemical means;
and heating means for heating the chemical means, said
regulating means being responsive to the voltage across said
partition and connected to said heating means for causing
said heating means to vary the temperature of said chemical
means such as to maintain the voltage across said partition
at a fixed va]ue, said heating means arranged to keep the
temperatures of ~he opposite sides of said partition substantially
equal to each other.
In another particular aspect the present invention
provides a method of measuring ion concentration which includes
providing an ionic conductor, applying a known concentration
of gas to a first portion of a surface of said conductor for
conduction through said conductor of gas ions to a second
portion of said surface, applying an unknown concentration
of gas to said second portion, and varying sald known con-
centration such as to malntain a fixed rate of conduction ofgas ions from one said portion to the other by exposing said
one portion to oxygen gas evolving from a metal/metal oxide
reference means, and varying the temperature of said refer-
ence means such as to maintain said fixed rate, while
maintaining said first portion and said second portion at
substantially the same temperature.
A preferred embodiment of the invention provides a device
comprising a small tube or cylinder having positioned across
its center a small thin solid electrolyte disc having ionic
~.,
-6a-
'. .- '-'~
98
conduction appropriate to the component to be monitored.
Symmetrical compartments on either side of the disc act as
reference and sample chambers. A heater wire wound in grooves
cut on the outside of the tube provide the disc with even
heat resulting in equal temperature on either side of the disc.
Metallic particles define an electronic conducting area and
hold a temperature sensing element in close contact with the
disc on each side.
~'
j 1 / ~J `1 ~ ` - 7-
- , .
8~ 8
The small si~e of the device and its symmetrical thermal
design give it great resistance to thermal shock resulting
in its ability to respond very quickly to changes in its
required set temperature. This enables the device to be
operated in a constant e.m.f. mode when a suitable reference
is sealed in the reference chamber and a fixed e.m.f. operating
point selected. The operating temperature is adjusted by an
automatic electronic control system which alters it until the
concentration of the reference component in relation to the
sample component produces an output across the disc which is
equal to the selected e.m.f.. The temperature of the disc is
then related to the concentration of the sample component.
The preferred embodiment provides a simple design which
is conductive to low cost and easy manufacturing and results in
a very small and rugged sensor that can survive harsh environments,
large temperature fluctuations, thermal cycling and vibration,
while offering at the same time a more reliable and accurate
measurement. Another feature of the design i9 its versatility,
for example the reference and sample chambers are interchangeable.
Also, the ease of manufacturing a dlsc shaped electrolyte
allows a wide choice of electrolytes to suit the application.
The feature of the embodiment, which is possible due to its
fast thermal response time, is its operation in a constant
e.m.f. mode when a suitabie reference is sealed in the
reference compartment, and a fixed e.m.f. set or operating
point selected. The operating temperature is adjusted by an
automatic electronic control system which alters it until the
concentration of the reference component in relation to the
~ 8-
... . ..
::.: . . - -: .
,::. ~ . :. :. -
. .
36~8
sample component produce an output across the disc which is
equal to the selected e.m.f. set point. The temperature of the
disc is then related to the concentration of the sample com-
ponent.
These and other features and advantages of the present
invention will become apparent from the following description
of an embodiment thereof given by way of example when taken in
conjunction with the accompanying drawings, in which:-
Figure 1 shows a section through the centre of a device
showing :its symmetry;
Figure 2 shows a way of making the disc and compartmentsin one piece;
Figure 3 shows a cross-sectional end view of the device
shown in Figure l;
Figure 4 shows a block diagram of the electronic control
system needed to operate the device in a constant e.m.f. mode;
Figure 5 shows a graph of the temperature distribution of
the centre of the sensor along its length;
Figure 6 shows a graph of the thermal response of the
sensor to a change in set temperature;
Figure 7 shows a graph of temperature difference across
the disc due to variations in sample flow rate;
Figure 8 shows a graph of e.m.f. v temperature for a set
of constant e.m.f. lines for a Pd/PdO sealed-in reference; and
Figure 9 shows a graph of sensor temperature for a constant
e.m.f. corresponding to 10% 02 set point.
The basic construction of the device is shown in Figure 1.
A small and thin solid electrolyte disc 1, having a high ionic
conductivity appropriate to the measurement required and a low
electronic conductivity, separates two chambers one being a
reference chamber 2 and the other a sample chamber 3, the
j 1/ _9_
8~8
chambers being interchangeable due to the symmetry of the design.
A specific area 4 on each side of the disc is coated with a
layer of porous electronically conducting powder, this can
be applied by sputtering or by using a commercial paste. The
choice of powder depends on whether an equilibrium state is
desired, then a metal such as platinum would be suitable since
platinum acts as a catalyst to speed up the operation, or a
non equilibrium state is desired where a metal such as silver
would be suitable.
The coating is used also to embed a thermocouple of
very fine wire 5 into the centre of each face of the disc. In
most cases it is found that the temperature of each face of
the disc is nearly identical in which case one thermocouple
suffices, and only an electrical conductor wire is needed on
the opposite face. Each thermocouple provides an electric
signal related to the temperature of the associated disc face.
The e m.f. across the disc can be measured using wires of the
same material from each thermocouple. The chambers are defined
by a thin walled ceramic cylinder manufactured out of an
appropriate material having a zero porosity, thls could be of
the same composition aæ the disc or a ceramic with matching
thermal characteristics. The chambers and disc can be made in
one part 6. Figure 2 if they have the same composition? or
they can be made in separate cylindrical sections 7 as shown in
Figure 1. When made in sections they can be joined to the
disc with a metallic based, a glass based, or a ceramic based
gas tight seal 8. The end sections 9 and 10 are also joined
to the chambers and sealed gas tight. Means for admitting 11
and removing 12 a fluid into the chamber 2 are provided by pipes
., . ~ . .- . :
: . : . : :, :,........ :
sealed gas tight to the end section 9. The choice of pipe
material is dependant on the intended operating temperature
range and on the degree ofthe corrosion potential of the
fluid. The thin electric wires are brought outside the chamber,
through holes 13 which are sealed gas tight, and then fused
to thicker wires 14 of the same composition which provide the
electrical terminations through the end section 9. ;;
A heater wire 15 is tightly wound round the ceramic
cylinder in grooves cut on its outside and covered with a
ceramic cement 16. The end connections of the heater winding
are also brought out through the end section 9.
A cylinder of insulating material such as an appropriate
ceramic 17 is cemented with a refractory cement to the end
sections and the space between it and the heater can be filled
with a high temperature insulation fibrous material 18. A thin
stainless steel outer cover 19 cemented to the end sections
acts as an overall sheath. Reference material is sealed in the
reference chamber 2 obviating the need for the inlet and outlet
pipes.
In another embodiment, when the fluid to be tested is in
an open environment into which the sensor is inserted, the
fluid could diffuse into the test chamber through a filter disc
or a flame trap if necessary. The outer sheath 19 then to be
replaced by an extended sheath having an opening at the trap
or filter disc.
When the device has a sealed in reference, the device is
operated in a constant e.m.f. mode as outlined in the block
diagram shown in Figure 4. The desired set point or constant
e.m.f. is selected from voltage divider 22. An electronic
8~
comparator 23 compares the set e.m.f. with the output voltage
of the decive and produces a signal, when they differ, of
polarity indicating the sense of the difference which is fed
to the temperature controller 2~l. This either increases or
decreases the electrical energy reaching the heater 15 as needed
to decrease the difference. The device settles down to a
temperature at which the sealed-in reference produces a
concentration of the component related to the sample component
so that the output voltage of the device is identical to the
set point voltage.
The temperature of the device as read by the thermocouple
5 and after suitable conditioning 25 can be used to drive a
display 26. The most common use of solid electrolyte devices
at present is for the measurement of the partial pressure of
oxygen in a gas mixture, so this application will be chosen for
the following example although by proper choice of the materials
other components can be measured. For example, if it were
desired to measure chlorine, a chlorine ion conductive solid
electrolyte would be chosen and a reference of metal/metal
chloride would be used.
Example 1
This is an example of selecting an appropriate sealed
reference for oxygen measurement in the constant e.m.f. mode.
The reference is required to have an oxygen partial
pressure, which is a function of temperature, that is very
stable and repeatable. A system of metal/metal oxide has this
feature, and the second requirement would be to find partial
pressures corresponding to the range of 2 measurement and in
a temperature range that is within the working range of the
~ 12-
,:
,: :: : ;
,
,. . ;:: . ~ :
,
.-: . : . .~,
` ~rf~ ~6$8
solid electrolyte used. The range of oxygen is chosen to
be within 1-100% and one of the common oxygen type ion conductor
electrolytes that operate between 800 - 1300 K is used.
An oxygen ion conducting material was chosen for the disc
which was 5 mm diameter and 0.5 mm thick and an active area of
about 2 mm diameter. Pt/13% Rh 0.001 diameter wire was used
for the thermocouple which were cemented using Johnson and
Matt'ney N 758 Pt. paste. The disc was fused to the two
~ylindrical halves using thin copper foil and heated to over
1000C. The heater winding was 0.002 Pt wire covered with
sauereisen no. 8 cement, this was also used for all other seals,
but other suitable cements may be used. The temperature
distribution of the centre of the sensor along its length
with 20 W input power is shown in Figure 5. The measured
temperature difference between the faces of the disc was less
than 1C. The response of the sensor to a chang'e in set
temperature, in this case switching on 20 W from cold, is shown
in Figure 6 where a time constant of about 15 seconds is evident.
Repeated thermal cycling and shock showed no sign of fracture
in the sensor. By appropriately positioning the inlet pipe
the-effect of the fluld cooling the disc can be minimised as
shown in Figure 7.
Now for a metal/metal oxide system the oxygen partial
pressure C2 at a given temperature is given by:-
ln C2 = A + BT ................................... (3)
where A and B are constants dependant on the oxide system used.
Using this with equation (1), n = 4 for oxygen.
e.m.f. = A + BT ~ 4F ln Cl ............................ (4)
An oxide system that satisfies our requirement is aPd/Pd 0 system and using the data given by Fouletier J. Appl.
; -13-
Electrochem. Vol. 5, III (1975) in conjunction with equation
(4), a family of curves can be drawn each corresponding to a
constant e.m.f. and relating the partial pressure of oxygen
with the temperature Figure 8. ~y examining this we can select
a constant e.m.f. line which is suited for the o~ygen range
and temperature range of the sensor. Assume that the zero
e.m.f. line is selected, this gives us the partial pressure of
the oxygen at the sample side
log Cl = 10.3 - 1175~ ............................... (5)
Thls shows that the working temperature of the device
ls related to the sample concentration when it is operated in
a constant e.m.f. mode and the variation in temperature
corresponding to an oxygen sample range of 15 - 5~ 02 for
a zero e.m.f. is shown in Figure 9.
The sensor is connected in a circuit as shown in Figure
4 whlch controls the temperature of the sensor and hence the
refe~ence material in the reference chamber 2 until the
concentration of the reference component equals the concentration
in the sample component and hence no e.m.f. is generated across
the disc. The temperature of the disc is thus related to the
concentration of the sample component and the temperature
indication can be used as an indication of concentration.
il6.-~ -14-
, :. , -.
. - :
- , . .. : :
