"
AM
.
i
AV
.
2.MN
.
SOLLTE
ha
.
PO
.
1
i
THE
Muhammad
AN
WA
MARKO
UN
.
.
27
.
.
2
*
.
!
14.
2
AY
.
56
1
.
"
d
,
w
IY
2
UNCLASSIFIED
ORNL
..
: .
.
.
w
>
/
.
-
-
:

I

re
'I
C
14
TOF2

..
Y
.
11
.0
...AN
S
.
11
6
il
7
14
I
.
ri"
$2
.
.
.
.
W
:
11
.
H
.
1
17
.
.
.
M
.
ANNA
KI
CON
%
1
'
'
12. S.
.
:
"..
"
P
.
7
WWW
1.
Kul

R
M
A
4
-
.
LEGAL NOTICE
This report was prepared as an account of
Government sponsored work. Neither the
United States, nor the Commission, nor any
person acting on behalf of the Commission:
A. Makes any warranty or representa-
tion, expressed or implied, with respect to
the accuracy, completeness, or usefulness
of the information contained in this report,
or that the use of any information, appa-
ratus, method, or process disclosed in this
report may not infringe privately owned
rights; or
B. Assumes any liabilities with respect
to the use of, or for damages resulting from
the use of any information, apparatus,
method, or process disclosed in this report.
As used in the above, "person acting on
behalf of the Commission” includes any em-
ployee or contractor of the Commission, or
employee of such contractor, to the extent
that such employee or contractor of the
Commission, or employee of such contractor
prepares, disseminates, or provides access
to, any information pursuant to his employ-
ment or contract with the Commission, or
his employment with such contractor.
.
.
$
:
1
.
1
..
.
!
wi.
.
.
11.
!!
!
JH
.
4
.
"
I
HT
!
'.
JO
11
"Y
"
...
'
IN
H
'
"..
NI
..
.
1
PA
+
SUN
K.
.
"
.
'.'.
.
.
THUY
YA KWANI
2. S
t
YEN
**
ephen

**
WH
YLL
.
*
WYN
th
NA
,
M
.
TA
VU.
WWL
7
2
ULTUR
4
W
120
LICY
.
T."
A
X
.
W3
4
.
le
T
ivi
LE
2N
TOUT
Sping
ir A
.
1
>
:
..
.
,
24
S.
-
-
.
:
DTIE
MICROCARD
ISSUANCE
DATE
"
1.
,
ir.
'
N
A
.
;'
0
.
I
'
'Y
YK
:.
MY
sehing
1
.
.
-
-
-
.
.
.
8/ 18
3
?
4
A
.
'
-
1
:14
-
-
TU
-.-
-
.
2
.
...
-
..
-
<!
1
.
LA1
1
:
..
-
...
.-
1964
.
11WRIT
Yli.
S
'
N
'
::
:
" Hi
SM " 111
.
t-
..
!!
MILAN
IN"
:
"
IT.
" !
H
:
H
1
1

LEGAL NOTICE
TM. mopar wo womar mouw Qwonam rot wort, Matthew Vatted
watu, wer the countelen, nor may prakty na babae Comics
A. No my numaty or reputation, pro origu,
wrot towe mou-
rusy, claw, or who
fou o lovnathes connaio report, the
of my later methan, apariwa, we , w mocow duchowed to the rol way we went

2
.
Oenit 14
1. Arun way labisuhan wu roopact to the . or how to mana row Arou the
of my lor machen,
d, meer dachudde rent.
As well e whenu, "pom i
milo C
"
playev nomurecher of the C alestan, a w ayen w met rector,
Auch plague
u to, wapya wont wanne waru,
wermentus, o mondo *c W, may Internatin pinta a
n
oth the Covena, ur Neubayamu wa mkdcontentor.
CONF-599 _MASTER
A Simple Quasi Reference Electrode ... Applications in Controlled-
Potential Polarography and Voltametry and in Chronopotentiometry
7
D. J. Fisher, W. L. Belew, M. T. Kelley
F
Oak Ridge National Laboratory
Oak Ridge, Tennessee 37831
Section 5
ABSTRACT
AMA.
The quasi reference electrode (Q.R.E.) is a bright platinum wire or
other inert conductor immersed directly in the solution; it may have a
very small area. It is used instead of a true reference electrode but
must be used with potentiostavic or other circuits that prevent the
ein
.
2
drawing of cell current through the Q.R.E. When used in appropriate
situations, the results obtained by use of a Q.R.E. are like those
obtained with a true reference electrode except for the lack of
thermodynamic meaning of the Q.R.E. potential; hence, this simple
electrode is called a quasi reference electrode. A number of success-
ful applications of the Q.R.E. are described for controlled-potential
.
polarography and voltammetry and for chronopotentiometry in aqueous,
organic, and molten salt solvents. For example, the form of polaro-
S
IEEE
graphic waves is conventional except for the lack of meaning of the
observed E12 The Q.R.E. 18 convenient to use, simple to construct,
rugged, small, does not introduce impurities, can be used in corrosive
solutions, will operate at high temperatures, and has a very low
resistance. Tre Q.R.E. is used instead of a true reference electrode
un
i
when the quantity of interest is the change in emf between the working

Facsimile Price $_e l e
Microfilm Price $_L252
Available from the
Office of Technical Services
Department of Commerce
Washington 25, D. C.
i
Hy
electrode and the electrode used as reference or the change in current
at a constant or scanned applied cell voltage as a function of concen-
tration or some other variable. The predominant consideration leading
to the choice of a simple rather than a true reference electrode is the
relative convenience of use of these electrodes. The Q.R.E., like cther
simple electrodes, is not meant to compete with true reference elec-
trodes if they are necessary or available and equally convenient to use.
INTRODUCTION
A true, reversible, reference electrode has a stable, reproducible,
and calculable potential (within a few mV) with respect to that of a
standard hydrogen electrode. The potential is determined by a reversi.
ble, charge-transfer process involving an equilibrium between a metal
and its ions, or an equilibrium in conjuction with a reversible process
reference electrode must have an exchange current that is large compared
to the current that must pass through the electrode when it is used (1).
A chemically inert metal electrode immersed in a solution of un-
specified composition has an unpredictable and possibly unstable
potential; it can easily become polarized by the imposition for
potential measurement of the passage of any substantial current density.
Its potential 18 affected by oxidation-reduction couples, pH changes,
etc. which can vary unpredictably and at times its potential 18
affected by the nature and amount of impurities which are present
.
toimeno
rminations mediterz ro** His
}
or contaminants which are acquired; particularly, if net current is
.
S
-
drawn through the electrode, its potential will probably approach that
.
mity bittersoet
of the reaction having the highest exchange current, or, it may adopt
F
a mixed potential. To these sources of uncertainty of magnitude and
of lack of reproducibility from solution to solution and of variability
in a given solution of the potential of the electrode must be added
another:
strain-induced potential. If a metal is mechanically
strained by compression or bending, a potential is generated at the
strain site (2). This strain-induced potential can be of the order
of hundreds of millivolts and will be variable in magnitude, tending
to disappear after the deformation occurs if some subsequent annealing
takes place. Strain-induced potential has been observed to be a source
of difficulty when unannealed platinum gauze working electrodes are
used in controlled-potential coulometry. It has been observed in
potentiometric measurements to be a source of spurious, erratic,
irreproducible potential at the metal-glass-solution junction if a
strained glass-metal electrode seal is immersed in the solution phase (3).
In summary, a chemical.ly inert, metal electrode immersed in a solution
of unspecified composition can not be expected to act like a true
reference electrode, particularly if current is drawn through the
electrode.
There are a number of recognized occasions in the application of
?
electrochemical techniques where the use of a true reference electrode
24
is not necessary because the thermodynamic significance of the magnitude
of the total cell electromotive force (emf) is irrelevant. In these
cases, it is possible for reasons of convenience to use, instead,
simple electrodes that have a stable potential. For example, in many
experiments, the quantity of interest, rather than the absolute emf
between the electrodes, is the change in emf between the working
electrode and the reference electrode or the change in current at a
constant or scanned applied cell voltage as a function of some variable
such as the concentration of a species. In other cases, where a true
reference electrode is not available, a number of useful measurements
are nevertheless possible by use of a simple electrode as reference.
The predominant consideration involved in the choice of a true or of
a simple reference electrode is the convenience or the necessity of
using a simple electrode. The use of a simple reference electrode
18 not meant to compete with that of a true reference electrode where
the use of a true reference electrode is equally satisfactory and
convenient or where the actual magnitude of the cell emf is needed.
On the other hand, changes in cell emf due to concentration changes
sensed or caused during a titration or a polarogram or a chronopotentio-
gram are independent of the pocential of the reference electrode; there-
fore, the most essential characteristic required of any reference
electrode is that its potential have a constant value for the
duration of the experiment.
The use of simple "shielded" reference electrodes for potentiometric
titrations is well knom. The buret electrode of Willard and Boldyreff
is typical (4). To prepare this electrode, the orifice of the buret
is t'educed by drawing out the tip of the buret and a bright platinum
wire is sealed into the tip below the stopcock so that it makes contact
with the titrant. During a titration the buret tin 18 immersed in the
solution. This wire may be used as a simple reference electrode for
oxidation-reduction, neutralization, and precipitation titrations.
Another example of a simple reference electrode for oxidation-reduction
titrations has been reported by Muller (5). It consists of a platinum
wire sealed into a capillary containing a very small amount of the
original solution. The indicating electrode is another platinum wire
wound around the outside of the capillary tube. Simple electrodes such
as these have not been used for polarography because the usual polaro.
graphs draw the cell current through the reference electrode and thus
require a non-polarizable reference electrode.
In molten salt systems for uses in which the thermodynamic signi.
ficance of the total cell voltage is irrelevant so that stability of
potential and convenience of use are the principal requirements for
the reference electrode, a platinum or other chemically inert metal
electrode having a very large surface area relative to that of the
working electrode has been frequently utilized in lieu of a true reference
electrode. For example, Black and De Vries in performing voltammetry
with two platinum electrodes in molten salts report: "The platinum anode
appeared to behave as a nonpolarized electrode during the recording
of the polarographic waves in the molten salt. All potentials in
this work have been reported with reference to the p'atinum anode
and as read directly from the polarographic Instrument." (6). An
anode of large area was used.
EXPERIMENTAL
It is the purpose of this paper to describe a simple electrode, the
quasi reference electrode (Q.R.E.), and some applications in controlled-
potential. polarography and voltammetry and in chronopotentiometry.
The Q.R.E. originated in 1958 because of the development of a dc
polarograph at Oak Ridge National Laboratory that performs electrolysis
at controlled-potential by means of a potentiostatic polarizing circuit (7).
Three electrodes are contained in the polarographic cell. The cell
current is drawn through the counter and working (D.M.E.) electrodes
but does not flow through the reference electrode. Only a very small
current, of the order of 10 - amp., is drawn by the potentiostat
through the reference electrode. For this reason, an asbestos-fibre-tip
S.C.E. of the type used with pH meters is usually used as the reference
electrode. The counter electrode 18 a piece of platinum wire or
graphite rod immersed in the solution, without isolation.
The following considerations led to the origination of the Q.R.E.
If the cell current (or a substantial measuring current) must be passed
through the electrode used as reference, net electrolysis must occur at
this electrode. In order for net electrolysis to occur, the reaction
must be driven by a potential difference between the electrode surface
and the bound (Helm:oltz) layer in the solution adjacent to the surface.
If the electrode is a true reference electrode, the exchange current is
high, species for electrolysis are included with the electrode, and the
potential difference does not significantly exceed the reversible
potential of the electrode. On the other hand, if the electrode is an
unstrained, chemically inert, metal electrode immersed in an electrolyte
of unspecified composition, a portion of the total voltage applied to
the cell must be used to drive a reaction at this metal electrode.
The potential difference that is required depends upon what is avail-
able in the solution to react to allow the passage of cell and/or
measuring current through this electrode. For example, it may be
necessary to decompose some of the solvent itself or a redox couple
may be available in the solution. The potentiel difference that is
required at the electrode may be substantial if the available
reaction is irreversible; if the current density at the electrode 18
high, several reactions may have to be driven in order to allow the
passage of this current.
Thus, the potential assumed by the electrode
during the passage of current is unpredictable and variable. In
summary, an unstrained, chemically inert, metal electrode immersed in
an electrolyte of unspecified composition could not be expected to act
like a true reference electrode if substantial current were drawn through
the electrode. However, for example, in polarography, with a given solvent
and supporting electrolyte, in a cell of usual size, the overall composi-
tion of the solution does not change significantly as the result of
recording one or several polarograms or with various concentrations
of the electroactive species whose concentration is being determined.
Also, the oxygen concentration is kept very low by means of an inert
gas. Moreover, in controlled-potential polarography, no cell current
and only an extremely small ineasuring current is passed through the
reference electrode.
The thought occurred that this polarographic
situation is the converse of that just described at the beginning of
this paragraph. Simple curiosity led to trying a second platinum wire
of small area, immersed in the solution without isolation, in lieu of
a reference electrode. It immediately became apparent that there might
be practical applications of this simple electrode. Preliminary work
with the Q.R.E. in dc polarography has been reported (8); polarograms
of good form were obtained that had very reproducible wave heights.
The Q.R.E. has not been described in any detail prior to the present
paper; a number of successful applications of the Q.R.E. have been
made since 1958 in aqueous, organic, and mollten salt solutions.
The Q.R.E. consists of a bright platinum wire (or other inert
conductor) i romersed directly in the solution without isolation and is
used with potentiostatic or other circuits not drawing substantial
current through the Q.R.E. The Q.R.E. need not have a large surface
area to prevent significant polarization. Other inert conducting
materials such as graphite can be used instead of a noble metal. A
very small pool of mercury can be used as a Q.R.E. except at high
temperatures. The use of a Q.R.E. will produce results like those
obtained with a true reference electrode except for the lack of
thermodynamic meaning of the observed voltage; hence, the electrode 18
called a quasi reference electrode.
The Q.R.E. is to be used only in circuits that do not draw any
substantial current through the electrode. This requirement is met
in many potentiostatic circuits, for example, for polarography or
for voltammetry. This requirement is also met by using a voltage
follower, a positive amplification, or a potentiometric mode operational
amplifier system to prevent drawing load current through the electrode.
In the chronopotentiometer (9), a voltage follower was used.
For the potential to be stable although meaningless, the electrode
must be free of strain-induced potential. If the data to be taken is
the change in cell voltage resulting from changes in a controlled
variable, the potential of the Q.R.E. must be stable for at least
the duration of the experiment. Noble metal wires can be annealed by
heat treatment in order to remove strain-induced potentials. However,
as will be shown by means of some dc polarographic data, if the
objective is to measure wave heights of regular polarograms, the
10
relatively slow drift in observed E, 12 value due to the use of an
unannealed wire for the Q.R.E. has no adverse effect on the reproduci.
bility of the measured wave heights.
The potential of the Q.R.E. is more stable if the solution 18
kept well-sparged. This is presumably due to the prevention of
significant variations in oxygen concentration.
RESULTS
Applications of the Q.R.E. in Controlled-Potential DC Polarography
Conventional polarographs draw the cell current through the
reference electrode and as a consequence only reference electrodes
that are not polarized by the passage of current may be used. Also,
to minimize IR losses, the resistance of the electrode including that
of the salt bridge and Junction must be low. The requirements of
non-polarization and of low resistance demand a rather large and a
reversible reference electrode and this complicates the construction
of small cells. Small cells are necessary for microcoulometry or
where only small amounts of solution are available. There are many
reference electrodes that are suitable for use with conventional
polarographs in most aqueous solutions, but in molten salt or organic
solutions, a suitable reference electrode is not always available.
The Oak Ridge National Laboratory (ORNL) Q-1988 series models
of controlled-potential dc polarographs contain a potentiostatic
polsrizing circuit and so do not draw the cell current through the
reference electrode (10). With these polarographs, the Q.R.E. has been
used in aqueous, organic, and molten salt solutions. The relative
placement of the three electrodes in the cell is not critical except
in solutions having a very high specific resistance and the latter
restriction applies whether a true or a quasi reference electrode 18
used. The ORNL model Q-1988A polarograph is now also produced and
sold by commercial corporations.
Aqueous Solutions, Cell of Conventional (~20 ml) Size
The effect of the physical history of the platinum wire used for
a Q.R.E. upon the utility of the polarograms for concentration
measurement was investigated. The test solution was 10-4 M ca
in
0.1 M KC1. Polarograms were obtained with a Q.R.E. consisting of a
new, bare, 18 gauge (0.040 inch 0.D.), platinum wire. About one.
half inch of the wire was immersed in the solution. Other polarograms
were obtained after the wire had been bent and twisted and artr»r the
wire had been heated to redness and quencined in concentrated nitric
acid. Although the observed E, is changed as a result of these physical
treatments, the shapes and heights of the waves were not affected by
the physical history of the wire. Thus, the utility of a Q.R.E. for
concentration measurement is not dependent on the physical history of
the inert material used for the Q.R.E.
Mechanical strain in the material of the Q.R.E. results in an
unstable, strain potential at the site of the strain. This strain
12
potential is included in series with the initial and scan voltages that
are applied to the cell. If the end that is to be immersed is hammered
just before the Q.R.E. is put into the cell, the strain potential 18
several hundred millivolts and decays in several minutes to tens of
millivolts. By recording the current at the D.M.E. äuring the first
few minutes after hammering and immersing the Q.R.E., at a constant
initial potential, as the major part of the strain induced voltage
annea.ls out at room temperature, a polarogram can be obtained. For
example, regular polarograms recorded with a 10-4 M ca** in 0.1 M
KCl solution by means of linear scanning and a S.C.E. and by means of
the decay of the strain potential that was induced by hammering the
tip of a platinum wire Q.R.E. had the same height. Annealing with
heat prior to the use of a Q.R.E. will reduce any strain-induced
potentials that may be present. ence, more reproducible, although
still thermodynamically meaningless, observed E, 12 values will be
obtained for replicate polarograms made with an annealed Q.R.E.
With the ORNL model Q-1988-ES controlled-potential and derivative
dc polarograph (8), regular polarograms were made with a solution
of 30°* M ca** in 0.1 M kci in an unthermostatted cell using in
sucression various electrodes as a reference electrode. Observations
are shown in Table I.
13
Table 1
Behavior of Noble Metals as a Quasi Reference
Electrode in Aqueous Solution
Range of 14(100)
Electrode Used
as Reference
No. of
Polarograms
Range of
Observed E, 12
mV
ave. la
Beckman S.C.E.
o
1.1
1.3
Unannealed Pt Wire Q.R.E. 5
Annealed Pt Wire Q.R.E. 4
N o
Unannealed Au Wire Q.R.E. 4
Small Hg Pool Q.R.E.
10-4 M ca
in 0.1 M KC1
ORNL model Q-1988-ES controlled.
potential dc polarograph
The same value of wave height, 17, is obtained with all of the electrodes
that were used as a reference electrode. The observed value of E, 12
is strictly reproducible and has thermodynamic significance only when
a true reference electrode, for example, a S.C.E., 18 used. When an
unannealed platinum or gold wire is used as a Q.R.E., the range of
observed values of E 12 18 very large, but the range of the observed
values of 1a 18 satisfactorily small and comparable to that obtained
with the S.C.E. Polarograms obtained with an unstrained noble metal
Q.R.E. (e.g., an annealed platinum wire or a small mercury pool) do
not shift about as much on the voltage axis; the reproducibility of
wave heights obtained with an unstrained Q.R.E. also is satisfactory.
The slow shifting of the replicate polarograms on the voltage axis when
strain-induced potentials are present in the Q.R.E. does not adversely
affect the measurement of heights of regular polarograms.
For a simple, reversible, diffusion-controlled process, at 25°C,
the peak width of a first derivative dc polarogram in millivolts at
half peak height is 90.6/n. Determinations of n for cadmium by this
measurement from derivative polarograms made respectively with a
S.C.E. and a Q.R.E. are in good agreement.
The following polarograms made with a platinum wire as Q.R.E.
are other examples of tests of the performance of a Q.R.E. 1.1 unthe: -
mostatted aquecus solutions. The values of range of La(100) for
four polarograms of 10" M N1(NH ) ** in 0.2 M NH, C1 and 1 M MH, OH was
ave, s
.
-
.
-
1.6%. The values of (E 12; S.C.E. - observed E, /2, Q.R.F.) for waves
due to cut, pb**, ri*, in*** and ca** simultaneously present (each
~10-4 M) in 1 M KC1 were 168 + 6 mV.
The Q.R.E. has also been tested in aqueous solutions with the
ORNL model Q-1988-FES controlled-potential and derivative dc polarograph
(10). This polarograph was designed for rapid regular and first and
second derivative dc polarography; it can also be used at customary, slow
.
.
scan-rates.
Ir. these tests, the D.M.E. had a hammer-controlled drop time
.
.
- i
.
of 0.5 second. Average-current polarograms were recorded by means of
. ne----m
the dual, parallel-T, low-pass filter. The area of the Q.R.E. is not
ina.
m
criticul because of the very low amount of current drawn through the
Q.R.E. by the potentiostat. At scan rates of 0.3 and 3 volts/min.,
a
EAR
H
respectively, regular and first derivative polarograms of 10-4 M cd
IN
ii: 0.1 M KCl with a Q.R.E. consisting of a cone of platinum approximately
177
0.003 inches high with a base 0.003 inches in diameter had the same
forms as those obtained with an S.C.E. The following tests were made
with a Q.R.E. consisting of a piece of bare, bright, annealed, 20
gauge (0.032 inch diameter) platinum wire that had about one-half inch
of its length immersed in the solutions. Polarograms were recorded
for 5 x 10-4 M VO," in 0.1 M HNO3. Fourteen regular polarograms
recorded without sparging between polarograms had a range in observed
E, 1 of 55 mv. Eight regular polarograms recorded with sparging
between each polarogram had a range in observed E, 12 of 5 mV. The
16
relative instability of the Q.R.E. potential in the first set 1.8
presumed to be due to leakage of oxygen into the solution; the top
of the polarographic cell was not well sealed. Subsequent test
solutions were kept well sparged. Five reguler polarograms were
recorded at a scan rate of 0.3 volt/min. for 5 x 10-4 M N1** in
0.2 M KNO. The range in observed E 1/2 was 10 mv. Four first deri.
vative polarograms recorded at a scan rate of 3 volts/min. had a
range in observed Ebeak of 15 mV and, a range in peak height of 1.5%.
Five regular polarograms of 10 M ca in 0.1 M KCl recorded at a
scan rate of 0.3 volt/min. had the same value of wave height (all
superimposed within a pen width) and the range of observed EV2 was
5 mv. Regular and first derivative polarograms were recorded with
10-4 M cain 0.1 M KC1. At this concentration, and presumably, at
lower concentrations, shifts in residual current values and in Q.R.E,
potential require that the heights of regular polarograms be measured
by extrapolation of the initial linear portion of the waves rather
than by measurement relative to corresponding polarograms of the
supporting electrolyte alone. The regular-wave heights at 10-4 and 10-5
M had a tenfold ratio even though the heights were measured by extrapola-
tion. The first derivative polarograms were recorded at a scan rate of
3 volts/min.; 20-4 M derivative polarograms were more reproducibie than
the corresponding sets of regular polarograms. Slow drifts in Q.R.E.
potential would not significantly affect rapid scan rates during the
17
recording of a derivative polarogram.
The reproducibility of derivarivo
polarograms of the supporting electrolyte at high sensitivity is better
than that of regular polarograms of the supporting electrolyte. The
error introduced by extrapolation 18 less for derivative polarograms
than for regular polarograms. At low concentrations (105 M or less),
at fast scan rates, first derivative polarograms are more satisfactory
than regular polarograms for concentration measurement.
The above observations may be briefly summarized as follows. The
-
.
form of a polarographic wave made with a Q.R.E. is convencional in all
.
.
.
.
respects except for the lack of thermodynamic significance of the
.
.
.
.
observed 51/2
Aqueous Solutions, Cell of Small (~0.3 mi) Size
A small cell can be constructed more easily with a Q.R.E. than
with a true reference electrode. A small, single-compartment cell was
assembled that contained 0.3 ml of solution. The electrodes were a
Sargent 2 to 5 sec. D.M.E. and two bright platinum wires, one of which
:
-
*
-
.
.
.
-
-
.
-
-
-
-
served as a counter electrode and the other as a Q.R.E. The Q.R.E.
was e. iength of 18 gauge, bare wire; about one-half inch was immersed in
the solution. Because of the small sample volume, the wave heights of
successive polarograms of 10-4 M ca** in 0.1 M KCl decreased at the rate
of 1.5% per polarogram, which is in agreement with the calculated
depletion due to electrolysis. Even with this small sample volume, for
.
--
<
-
.
7
18
a solution having the usual order of magnitude of specific resistance the
relative placement of the three electrodes is not critical. Maximum
current polarograms were made by means of the peak follower. A plot of
log i / (12 - 1) vs E for one of these polarograms is shown in Figure 1.
.
Figure 1. Test of Polarographic Wave Obtained with Platinum Wire
Quasi Reference Electrode
2
*
-
UNCLASSIFIED
ORNL-LR-DWG. 45383

-Reciprocal of Slope
= 0.034
(Theoretical for n = 2
is 0.030 )
CONDITIONS:
Controlled - Potential and
Derivative Polarograph
10-4 M Cd" in 0.1 M KCI
Peak Follower
Electronic Scan
Cell
Anode, pt Wire
Controlled Electrode, D.M.E.
Quasi-reference Electrode,
Pt Wire
Sample Volume, 0.3 ml
0.014
0.68
0.80
0.72 0.76
-EOME, volts
14 Sept. 1959
DJF
TEST OF POLAROGRAPHIC WAVE OBTAINED
WITH PLATINUM WIRE QUASI-REFERENCE ELECTRODE
20
The observed relationship is linear and the reciprocal of the slope
of the line is 0.031 which is in good agreement with the theoretical
value for n equal to 2 of 0.030. With the quadruple, parallel-T, low-
pass filter and the derivative computer, first derivative average-
current de polarograms were recorded. The width of a derivative wave
at half peak height was the same with a Q.R.E. as that obtained with
a S.C.E. in a larger volume of solution. From these tests, it is
seen that polarographic waves obtained with a Q.R.E. fit the usual
equations for the reduction process. Also, the ratio was computed of
the wave height obtained with the parallel-T filter to that obtained
with the peak follower. The ratio of 0.82 is in good agreement with
values of 0.81 and 0.813 that were obtained with a S.C.E. (8) and
with values reported by Taylor et al. (11) from oscillographic
measurements of millimolar cadmium which were 0.810 and 0.805 for the
Organic Solvents
Many true reference electrodes have been described that each
function properly in one or more of various organic sol.vents. The
attainment of a true reference electrode in an organic solvent is
sometimes made difficult by the uncertain magnitudes and by the
variation of junction potentials at interfaces between different
solvents where thermodynamically irreversible processes occur or by the
leakage of water from an aqueous salt bridge into an organic system.
If the significance of the actual magnitude of the cell voltage need not
be interpreted, the use of a Q.R.E. may be more convenient, for example,
for concentration measurement by controlled-potential polarography.
The following example demonstrates that the Q.R.E. can be used
satisfactorily in organic solvents. Controlled-potential polarograms
were recorded of a solution of 10-3 M ca** in n-propanol with 0.05 M
lithium perchlorate as supporting electrolyte. A Sargent 2 to 5 sec.
D.M.E. was used. The cell resistance from S.C.E. to D.M.E. was 9600
ohms. Polarographic measurements obtained with an unannealed 18 gauge
platinum wire Q.R.E. and with a S.C.E. reference electrode are shown
in Table II.
22
Table 2
Behavior of Unannealed Platinum Wire as a Quasi Reference
Electrode in an Organic Solvent
Range of
Range of la(100)
Electrode Used
as Reference
No. of
Polarograms
Observed E1/2
ave. 1.
mV
Beckman S.C.E.
2.0
Unannealed Pt Wire Q.R.E.
4
14
0.9
10-3 M ca"
in n-propanol with 0.05 M Licio,
ORNI, model Q-1988-ES controlled-potential
dc polarograph
Although the range of observed E, 19 values was greater with the unannealed
Q.R.E. than with the S.C.E., the range of wave heights measured with the
Q.R.E. was less than that measurer with the S.C.E.
The same average
value of wave height was obtained with the Q.R.E. and the S.C.E.
In
this medium, as an example of an organic solvent, the use of the
unannealed Q.R.E. is satisfactory for the controlled-potential polaro-
graphic determination of concentration.
Organic Solvent Extracts
It has been shown that controlled-potential dc polarography is
useful for determining the concentration of metals present as complexes
in organic solvent extracts. For example, a very satisfactory sensitive
method was developed (12) for the determination of uranium extracted
with 0.1 M tri-n-octylphosphine oxide in cyclohexane from an aqueous
phase that is 2 N in nitric acid. An equal volume of 0.2 M liti.ium
perchlorate in ethanol is added to the separated extract in order to
provide supporting electrolyte. The specific resistance of the
resulting solution is about 1400 ohm-cm.
In solutions having a
specific resistance of this order or less, at usual cell current
magnitudes, the relative placement of the three cell electrodes used
in controlled-potential polarography is not critical. For example,
with the reference electrode tip located as usual outside the IR
bulk
regions (7) but at a distance of 10 drop radii or more from the
...
.
surface of the D.M.E. drops, for a specific resistance of 1400 ohm-cm,
a drop radius of 0.035 cm., and a cell current of 1 ma, the 1Rnner
error is less than 5 mv. For the highest uranium concentration that
is of interest in this method, 10-4 M, the diffusion current value
was about 0.7 ua. However, with some organic solvents or organic
solvent extracts, the specific resistance is higher and so the
product of cell current and inner solution resistance may be signifi.
cant and require appropriate positioning of the tip of the potentiostat
reference electrode.
With a D.M.E., a substantial fraction of the IR loss in the
solution is located in the high current-density region within several
drop radii of the surface of the mercury drop. With a properly
functioning potentiostat, the location of the potential difference
between the controlled electrode (D.M.E. in polarography) and the
reference electrode that is equal to the command voltage set into
the potentiostat (sum of initial and scan voltages in dc polarography)
is the equipotential surface that passes through the tip of the
potentiostat reference electrode. According to Kortum and Bockris (13),
in terms of Stern's theory, the voltage that drives the reaction is
that between the controlled electrode and the Helmholtz or bound
layer that is about one ionic radius away from the electrode. The
diffuse Gouy double layer extends from the Helmholtz layer into the
bulk of the solution and there is an additional exponential drop of
25
potential across the Gouy layer. Under ideal conditions, with a very small
reference electrode prube placed as close as possible to the electrode
surface, the total of these two voltages, the potential difference
between the controlled electrode and the solution, can be measured with-
out including additional voltage due to iR in the solution. In the
case where the controlled electrodc is a D.M.E., according to Laitinen
in Kolthoff et al. (14), the solution resistance may be regarded as
consisting of the sum of an inner and a bulk : esistance. The inner
resistance is located in the region from the double layer to about
ten drop radii and the bulk resistance of interest is in the current
paths in the remainder of the solution. The average value of the
inner resistance has been calculated by Ilkovič (15, 15) to be:
Renner = der where r is the maximum radius of the D.M.E. drop and o
is the specific resistance of the solution. If the tip of the
potentiostat reference electrode is located out of the iRzt regions
in the solution, on the opposite side of the D.M.E. from the counter
electrode (7), but is more than 10 r from the surface of the drop, the D.M.E
voltage relative to the solution will be less than the intended value by the
product of the average value of the cell current and the average value of the
inner resistance. The existence of this uncorrected iR voltage at the
Interface of the D.M.E. and the solution was pointed out by Kelley, Jones,
and Fisher (7) who also correctly stated that the magnitude of this
error is insignificant for usual polarographic conditions. However,
26
as can be predicted by the above Ilkovič formula, under some polarographic
conditions, this error io not insignificant. Methods of correcting
derivative dc polarograms for this error have been developed by Schaap
and McKinney (17, 18).
It was decided to demonstrate whether or not, in very high specific
resistance solutions, controlled-potential polarographic electrolysis
could be free of significant i(Renner + Rub) error by locating the
tip of the potentiostat reference electrode within 0.1 r of the surface
of the D.M.E. The first demonstrations (12) were made with a Q.R.E.
and with a Smoler vertical orifice D.M.E. (19, 20, 21, 22) sharpened
to a point, having a drop time of 0.5 sec., and a mercury flow rate
of approximately 5 mg. / sec. The value of r at -0.5 V. vs. the S.C.E.
was 0.035 cm. The mercury drops from a Smoler D.M.E. are relatively
small and grow principally in a downward direction; this allows the
placement of a polarograph potentiostat reference electrode tip within
0.1 r of the upper surface of the growing mercury drop. About 10% of
Rino is located at a distance of 0.1 r from the upper surface of the
inner
drops from a Smoler D.M.E. A small Q.R.E. was used that was made from a
sharpened 3 mil diameter platinum wire sealed in glass so that the
immersed platinum was a cone having a height about equal to the
diameter of the wire.
This Q.R.E. was mounted on a micrometer screw
mechanism arranged so that the tip could be located at known distances
from the upper surface of the growing mercury drop of the Smoler D.M.E.
UN
7:
-
-
1
w
-
-
.
-
AT".
"
20F2
14
PI
ORNL
UNCLASSIFIED

.
W
,
LJ
.
*
14
:
L
.

En
..
..
...
.
... ...
-...
.
.
.
. wow ..-
.
-.
.



-
-
-
-
-
2
7
-
:
ic
:
+
t
:
-
"
.
14
.
.
-
-
i
11?
PO
'.i
.
E
V2:
.
"
-
23.
..sas
-
*
- two
Fy-
KR12
.
TV
*
TS
PSY
FLY
3.11
.
TAS
1
V
LL
Tr.
S
GYET*
111
AL
29,
YO
--
-
..
:
:
.
C
.
i
.
E
.
.
E
TV
.
.
"
.
.
.
-
-
4
-
11
.
-
.
-
.
4.
IM
...
!
...
T!
T
,
K
DTIE
MICROCARD
ISSUANCE
DATE
8/ 18
1961
"
!
"
.
.
ta
.
.
71
27
and could be located within 0.1 r of the upper surface. Because of the
very small area of this Q.R.E., to ensure that insignificant current
would be drawn through the Q.R.E. by the potentiostat input, a voltage
follower was fabricated from a Philbrick type P2 operational amplifier (23)
and connected in series with the Q.R.E. and the input of the potentiostat.
For this purpose, the use of the Q.R.E. is very convenient as compared
to the use of a Luggin-Haber probe and a true reference electrode.
The performance of the Q.R.E. in this application was very satis-
factory. The heights of replicate polarograms were very reproducible.
With a particular medium and for a fixed location of the Q.R.E, tip,
the range of observed E12 values for replicate polarograms was less
than 5 mV. These asſiects of the performance of the Q.R.E. are illustrated
in Figure 2 which consists of superimposed triplicate, first derivative,
average current, controlled-potential polarograms of millimolar ca
with 0.1 M tetra-n-hexylammonium iodide as supporting electrolyte in
2-octanol. The specific resistance of this solution was 22,000 ohm-cm.
and the Q.R.E. was within 0.1 r of the upper surface of the Smoler
D.M.E. drops for the polarograms reproduced in Figure 2.
1
E
28
Figure 2. Triplicate, Superimposed, Controlled-Potential Polarograms
Obtained with Quasi Reference Electrode
-,.
-
UNCLASSIFKD
ORNL-Owo. 83-TIMA

IDIPINDIMIT
7.2 wa / min.
ORNL controlled potential and derivative polarograph
Smoler vertical orifice D.M.E.
#Drop time, 1/2 second
Scan rate, 3.0 volts /minute
Sample :
4x10-3 M Cd in 2-octanol
0.1 M tetra-n-hexylammonium icdide
Specific resistance, 22,000 ohm-cm.
Reference to D.M.E. distance 0.4 drop radius
mm
*
!
llion
i
ie,
f
.
...
..
..
...
.,;
.
DNOU
UTODO0I
UUUUU
UUUUUUUUUUUUUU1101000
-0.1
-0.2 -0.3 -0.4 -0.5 -0.6
VOLTS vs QUASI REFERENCE ELECTRODE
-0.7
Although the waves in this medium, due to maxima, are not very well-formed,
the degree of reproducibility of the polarograms obtained with a R.R.E. is
satisfactory. Also, the data obtained with the Q.R.E. is satisfactory
for demonstrating the degree to which this placement of the reference
electrode tip eliminates iR errors. The observed values of Rinman
Inner
were somewhat higher than the values predicted from the above Ilkovič
equation. This Ilkovič equation is derived for a simplified model
consisting of two concentric, equipotential, spherical, surfaces having
radii of r and r' where r'>>r, and separated by a medium having a
specific resistance of p. Devay (24) has pointed out that the shielding
caused by the capillary itself prevents the current lines from the D.M.E.
from extending out radially in all directions and has derived a shielding
correction factor of 16/9 to account for the higher onserved values of
Rinner: An empirical correction factor of 3/2 was proposed by Kolthoff
et al. (14). This correction is also discussed by Schaap and McKinney
(17, 18). The data obtained is discussed in some detail elsewhere (12),
and will not be repeated here since the primary objective was to
investigate the degree to which error due to iRinner can be prevented
by appropriate location of the tip of the potentiostat reference elec-
trode above the drops from a Smoler D.M.E, in solutions having very
high specific resistances. It was found that the above Ilkovič
equation is a practical guide for estimating the magnitude of iRinnor
inner
errors for known p, 1 and r values. In solutions having specific
resistances up to about 20,000 ohm-cm., polarograms recorded with the
tip within 0.1 r had insignificant amounts of 1R error. Since supporting
electrolyte is desirable for the usual reasons, and ions are required
for transport, organic media probably can be found for most analyses of
interest that have specific resistances well below 20,000 ohi-cm.
Molten Armonium Bifluoride
In 1958, H. P. Raaen performed a cursory evaluation (25) of the
polarography of niobium in molten ammonium bifluoride at about 150°C
by means of an ORNL model Q-1988 controlied-potential dc polarograph (7)
and a platinum wire Q.R.E. This work was limited by the lack of a
satisfactory D.M.E., a satisfactory Teflon D.M.E. (26) having not
yet been developed.
Application of the Q.R.E. in Controlled-Potential AC Polarography
Satisfactory polarograms have been obtained by R. W. Stelzner (27)
with a platinim wire Q.R.E, using a controlled-potential ac polarograph.
The polarograms are reproducible with respect to the height of the
wave, although the observed half-wave potential is subject to small
and unpredictable deviation from the accepted value.
Applications of the Q.R.E. in Molten Salt Solvents
The selection of a true reference electrode and the establishment
of its reliability, reproducibility, and thermodynamic significance in
molten salt systems at high temperatures is difficult but is required
if certain thermodynamic data is to be gathered by means of potentiometric
measurements. Other useful thermodynamic data can be obtained by observing
only changes in cell voltage rather than its absolute value. There are
situations where it is not required to know the thermodynamic significance
of cell voltages; for example, concentration can be measured by voltammetry,
chronopotentiometry, or other electrochemical methods without knowing
absolute values of the cell voltage.
The determination of the concentra.
tions of impurities such as the corrosion products, iron, nickel, and
chromium, and of other species, directly in the molten state, in
corrosive molten fluoride systems that are compor.ents of reactor fuels
is of practical importance. At Oak Ridge ilational Laboratory, work has
been in progress since 1962 to develop instrumentation and methods for
this purpose; this work has been done with the Q.R.E.
Chronopotentiometry
Mamantov has used a Q.R.E. to investigate the chronopotentiometry
of uranium in molten fluoride melts (28). A precision chronopotentiometer
was used that was fabricated (9, 29) from operational amplifier systems.
A voltage follower system prevents any significant amount of current
from being drawn through the Q.R.E.
Chronopotentiometric studies of iron in molten LiF-BeF, (66-34 mole %)
eutectic by means of the precision chronopotentiometer (9, 29), an
indicator electrode made from pyrolytic graphite which was properly
oriented and insulated in boron nitride and a platinum Q.R.E. have
been reported by Manning and Mamantov (30).
32
Controlled-Potential DC Voltammetry
D. L. Manning has investigated the voltammetry of silver in molten
equimolar sodium nitrate - potassium nitrate eutectic over the tempera-
ture range of 246 to 326°c (31). The voltammograms were recorded by
means of an ORNI. model Q-1988 controlled-potential de polarograph (7),
a stationary platinum microelectrode, a 0.5-mm. diameter (24 gauge)
platinuni wire Q.R.E., and an isolated platinum counter electrode. The
parallel-T, RC filter was useful in this work. The redox reaction
properties of silver in this molten salt solvent are known from pre-
viously published work. Manning's objective was to find out whether
information obtained with the Q.R.E. would agree with the known redox
behavior of silver before applying the Q.R.E. to the study of other
moiten salt system electrochemistry. Manning found that with the Q.P.E.
the reduction of Ag" to Agº proceeds reversibly, the observed half-wave
potentials are reproducible and the limiting current is proportional
to the concentration of silver so that the current-voltage curves are
analytically useful. From the results the activation energy for the
current-limiting process is calculated to be approximately 6.4 kcal./mole,
the effective thickness of the diffusion layer is about 0.1 mm., and the
average n value is very nearly 1 and has no particuler trend with
temperature. The latter three values are in good agreement with
those obtained by others who did not use the Q.R.E. Reverse scan,
stripping curves were also recorded and the stripping curves did not
hesitate as they passed through the zero-current axis. Current-voltage
curves were recorded at rates from 25 to 100 mV/mi.n. and the limiting
current was essentially independent of scan rate; evidently, steady-
state conditions were achieved very rapidly in the melt. The scope
and results of these investigations show that the Q.R.E. performed
satisfactorily.
Manning has investigated (32) the voltammetry of Iron in molten
LiF-NaF-KF (46.5 - 11.5 - 42 mole %) eutectic over the temperature
range of 470 to 545°C by means of the ORNL model Q-1988 controlled-
potential de polarograph (I), an isolated platinum counter electrode,
a platinum micro electrode, and a 1 mm. diameter (18 gauge) platinum
wire Q.R.E. The Q.R.E. performed satisfactorily in these investigations.
Manning and Mainantov have used an ORNL model Q-1988 controlled-
potential dc polarograph (7), an isolated counter electrode, one of
several micro electrodes, and a p.latinum Q.R.E. for voltammetry of
zirconium and uranium in molten LiF-NaF-KF (46.5 - 1.1.5 - 42 mole %) (28).
The performance of the Q.R.E. was satisfactory for this work.
Manning and Memantov have reported (30) the use for voltammetry
of an ORNL model Q-1988 controlled-potential de polarograph (7) that
was modified by means of design information furnished by H. C. Jones
to add current ranges up to 5 ma., electronic scanning (33) rates from
0.05 to 20 volts/min. and voltage dividers to drive an XY recorder.
The other apparatus included an indicator electrode made from pyrolytic
graphite which was properly oriented and insulated in boron nitride, ar.
1solated platinum counter electrode, and a platinum Q.R.E. Results
were reported (30) for rapid scan voltammetric studies of iron in
molten LiF-NaF-KF (46.5 - 11.5 - 42 mole %) eutectic and in molten
LiF-BeF, 166 - 34 mole %) eutectic. At scan rates above about 1 volt/min.,
at 500°C, the reduction of Fe(II) to the metal in LiF-NaF-KF was diffusion
controlled since a plot of 1 peak V8. (scan r
vs. (scan rate) 1/2 was linear. From the
slope of this line, the diffusion coefficient of Fe(II) in molten
LiF-NaF-KF was found to be 1 x 10-6 cm /sec. In molten LiF-BeF,
eutectic, at a scan rate of 5 volt/min., the peak current was propor-
tional to the molality of Fe(II) in the range of 2 to 5 millimolal. The
diffusion coefficient of Fe(II) in this medium was found to be 5 x 10-6
cm /sec. The relative average deviation in ineer for the anodic
peak
stripping voltammetry of iron was 3.5%. The stripping curve revealed
no hesitation as it passed through zero current which is evidence that
the reaction at the pyrolytic graphite electrode proceeds reversibly.
Generally speaking, the current-voltage curves were well defined,
especially at faster scan rates where the transport process is more
nearly diffusion controlled. The Q.R.E. performed satisfactorily in
this investigation.
Manning has reported results on the rapid scan voltammetry and
anodic stripping voltammetry of nickel in molten LiF-NaF-KF (46.5 - 11.5 - 42
mole %) eutectic (34). An ORNL model Q-1988 controlled-potential de
35
polarograph (7) (modified as just described to add fast electonic scan
rates and higher current ranges), an isolated platinum counter electrode,
a pyrolytic graphite indicator electrode sheathed in boron nitride, and
a platinum Q.R.E. were used. To purify the eutectic by pre-electrolysis,
an ORNL model Q-2005 controlled-potential coulometric titrator (35, 36)
was used with a platinum Q.R.E. Rapid scanning (3 v/min.) anodic
stripping voltammetry was the more sensitive method. The reproducibility
of the area under replicate anodic stripping curves is about 5%. As
little as 1 ppm of nickel can be detected and the areas are a useful
measure of the concentration of nickel. The observed E, 10 for the
reduction of N1(II) to Ni(0) 18 about -0.2 volt vs. the platinum Q.R.E.
The voltammograms are well-defined and have a peak at faster (up to 10
V/min.) scan rates indicating that the transport process is essentially
diffusion controlled at faster scan rates. The peak current at faster
scan rates is directly proportional to the concentration of nickel.
From data consisting of observed peak current vs. scan rate, the
diffusion coefficient of nickel was calculated at 500, 570, and 600°C.
The activation energy of the current-limiting process was also obtained.
The performance of the Q.R.E. Was satisfactory for this work.
Results accomplished by means of controlled-potential voltammetry
with an ORNL model Q-1988 controlled-potential dc polarograph (7) using
three molten fluoride solvents (LiF-NaF-KF, 46.5 - 11.5 - 42 mole %;
LiF-BeFg, 66 - 34 mole %; anú LiF.-BeF, -ZrF, 65.6 - 30.7 - 3.7 mole %)
36
have been reported by Mamantov, Manning, and Dale (37). Well-defined and
useful voltammograms were obtained, particularly at rapid scan rates that
are available with the niodified polarograph. Ions studied include
iron(II), nickel(II), zirconium(IV), and uranium(IV). Rapid scan
anodic stripping voltammetry was found to be more satisfactory for
concentration measurement than direct vol.tammetry. Diffusion coeffl-
cients and other parameters were measured satisfactorily by the
latter technique. The performance of a platinum Q.R.E. that was
used for this work was satisfactory.
DISCUSSION
The Q.R.E. is a bright platinim wire or other inert conductor
immersed without isolation directly in the solution and used with
potentiostatic or other circuits that prevent the drawing of cell
current through the Q.R.E. It may have a very small surface area.
The most essential requirement of a simple electrode that is to be
used instead of a true reference electrode is that it have a constant
potential for the duration of the experiment. However, a slow drift
in Q.R.E. potential does not adversely affect the measurement of
the heights of regular polarograms, or, at fast scan rates, of
derivative polarograms.
That the Q.R.E. is a satisfactory simple
electrode is shown by its successful application in controlled-
potential polarography and voltammetry and in chronopotentiometry iri
aqueous, organic, and molten salt solutions that have an unspecified
but relatively constant overall composition. It must be remembered
that the potential of an unstrained Q.R.E., though stable, is
thermodynamically meaningless in that its value in various solutions
cannot be calculated.
The Q.R.E. has been shown to have the 'following advantages.
It is convenient to use, cheap, simple to construct, and rugged
and can be physically small.
It can be used in aqueous and in
non-aqueous media including organic solvents and molten salts for
which true reference electrodes are difficult to obtain. It can be
used, for example, in corrosive electrolyte solutions since it is
constructed from a relatively inert conductor. It does not
introduce impurities such as k*, c1", Hg++, Ag+, or water.
There are no junction potentiaïs, which may be variable, at
interfaces of different solvents. It will operate at high tempera-
tures.
The Q.R.E. has a very low resistance.
The Q.R.E., like other simple electrodes, is not meant to
compete with true reference electrodes if they are necessary or
available and equally convenient to use. There are, however, a
number of recognized occasions in the application of electrochemical
techniques where the use of a true reference electrode is not
required because the thermodynamic significance of the magnitude
of potential relative to that of a standard hydrogen electrode is
irrelevant; the quantity of interest in many techniques is the
38
change of emf between the working electrode and the electrode used as
a reference or the change in current at a constant or scanned Applied
cell voltage as a function of concentration or some other variable.
In other cases, a true reference electrode is not available. The
predominant consideration leading to the choice of a simple electrode,
e.8., the Q.R.E., rather than of a true reference electrode is the
relative convenience of use of these electrodes. When used in
appropriate situations, the results obtained by use of a Q.R.E. are
like those obtained with a true reference electrode except for the
lack of thermodynamic meaning of the Q.R.E. potential; hence, the
electrode is called a quasi reference electrode. This paper summarizes
a number of successful applications of the Q.R.E. which have been
made since 1958 in aqueous, organic, and molten salt solutions.
REFERENCES CITED
1. Ives, David J. G., Janz, George J., "Reference Electrodes .,
Theory and Practice," Academic Press, New York, N. Y., 1st. Ed.,
1961, pp. 17-21, 578-579, and 591-592.
2. Nobe, K., Baum, E., Seyer, W. F., J. Electrochem. Soc. (1961)
108, 97.
3. Garrett, A. B., Hogge, E., Heiks, R., Science (1940) 92, 18.
4. Willard, H. H., Boldyreff, A. W., J. Am. Chem. Soc. (1929) 51,
471.
5. Müller, E., 2. Physik. Chem. (Leipzig) (1928) 135, 102.
6. Black. E. D., DeVries, T., Anal. Chem. (1955) 27, 905.
7. Kelley, M. T., Jones, H. C., Fisher, D. J., Anal. Chem.
(1959) 31, 1475.
8. Kelley, M. T., Fisher, D. J., Cooke, W. D., Jones, H. C.,
"Controlled-Potential and Derivative Polarography," pp. 158-182 in:
Longmuir, I. S., Ed., "Advances in Polarography," Pergamon Press,
Oxford, Ist. Eż., 1960, Vol. i.
9. Maddox, W. L., Fisher, D. J., "Precision Chronopotentiometer,"
Oak Ridge National Laboratory Rept. ORNL-3060, Office of Technical
Services, Department of Commerce, Washington 25, D. C., issued 1961,
p. 4.
10. Fisher, D. J., Belew, W. L., Kelley, M. T., "A Controlled-
Potential and Derivative DC Polarograph for Rapid Regular and First
40
and Second Derivative DC Polarography: Circuits, Theory, and Performance
Characteristics," (submitted for publication).
11. Taylor, J. K., Smith, R. E., Cooter, I. L., J. Res. Natl. Bur.
Stand. (1949) 42, 387.
12. Belew, W. L., "Application of Controlled-Potential and Deriva-
Extracts and Compensation of Polarographic Cell Resistance in High
Specific Resistance Solutions," M. S. Thesis, Department of Chemistry,
The University of Tennessee, Knoxville, Tennessee, December 1963.
13. Kortum, G., Bockris, J. O'M., "Textbook of Electrochemistry,"
Elsevier Publ. Co., New York, N. Y., Vol. II, 1951, pp. 364-365.
14. Kolthoff, I. M., Marshall, J. C., Gupta, S. L., J. Electroanal.
Chem. (1962) 3, 209.
15. Ilkovič, D. Coll. Czechoslovak Chem. Communs. (1932) 4, 480.
16. Iliovič, D., Coll. Czechoslovak Chem. Communs. (1936) 8, 13.
17. Schaap, W. B., McKinney, P. S., Anal. Chem. (1964) 36, 29.
18. Schaap, W. B., McKinney, P. S., "Resistance Compensation in
Polarography ... Application to High-Resistance Nonaqueous Systems and
to High Current-Density Aqueous Systems," (submitted for publication
to Anal. Chem.).
19. Smoler, I., Coll. Czechoslovak Chem. Comuns. (1954) 19, 238.
20. Kalvoda, R., Smoler, I., Zavodskaya Lab. (1961) 27, 549.
21. Kuta, J. and Smoler, I., "The Instantaneous Currents (i-t Curves)
41
on Single Drops," pp. 43-63 in: Zuman, P., Kolthoff, I. M., Ed.,
"Progress 1. Polarography," Interscience Publishers, New York, X. .,
1962, Vol. 1.
22. Smoler, I., J. Electroanal. Chem. (1963) 6, 465.
23. George A. Philbrick Researches, Inc., Boston 16, Mass.,
"GAP/R Technical Bulletin P2".
24. Devay, J., Acta Chim. Acad. Sci. Hung. (1963) 35, 255,
25. Raaen, H, P., "Polarography of Niobium in Ammonium Bifluoride,"
Oak Ridge National Laboratory Rept. ORNL-2662, Office of Technical
Services, U. S. Department of Commerce, Washington 25, D. C., issued
1959, pp. 3.2-32.
26. Raaen, H. P., Anal. Chem. (1962) 34, 1714.
27. Stelzner, R. W., "AC Polarography," Oak Ridge National
Laboratory Rept. ORNL-3060, Office of Technical Services, Department
of Commerce, Washington 25, D. C., issued 1961, pp. 8-9.
28. Manning, D. L., Mamantov, G., J. Electroanal. Chem. (1963) 6,
328.
29. Maddox, W. L., Fisher, D. J., "Modification to the Precision
Chronopotentiometer," Oak Ridge National Laboratory Rept. ORNL-3537,
Office of Technical Services, U. S. Department of Commerce, Washington
25, D. C., issued 1964, p. 2.
30. Manning, D. L., Mamantov, G., J. Electroanal. Chem. (1964) 7,
102.
42
31. Manning, D. L., T-Lanta (1963) 10, 255.
32. Manning, D. L., J. Electroana 1. Chem. (1963) 6, 227.
33. Kelley, M. T., f'i sher, D. J., Jones, H. C., Anal. Chem.
(1960) 32, 1262.
34. Manning, D. L., "Voltammetry of Nickel in Molten Lithium
Fluoride - Potassium Fluoride - Sodium Fluoride," J. Electroanal.
Chem. (in press).
35. Kelley, M. T., Jones, . C., Fisher, D. J., Anal. Chem.
(1959) 31, 488 and 956.
36. Kelley, M. T., Jones, H. C., Fisher, D. J., Talanta (1960) 6,
185.
37. Mamantov, G., Manning, D. L., Dale, J. M., "Voltammetry in
Molten Fluorides," Third International Congress of Polarography,
Southampton, England, July 19-25, 1964,
work performed at Oak Ridge National Laboratory operated by Union Carbide
Nuclear Company for the U. S. Atomic Energy Commission.

END
#
.com
..