View Article Online / Journal Homepage / Table of Contents for this issue

THE BEHAVIOUR OF THE LEAD DIOXIDE ELECTRODE

PART 6.-THE POTENTIALS OF THE LEAD DIOXIDE/LEAD SULPHATE
ELECTRODE

BY W. H. BECK,K. P. SINGHAND W. F. K. WYNNE-JONES

Physical Chemistry Laboratories, King’s College, Newcastle-upon-Tyne, 1
Published on 01 January 1959. Downloaded by University of Sheffield on 06/07/2014 12:28:37.

Received 3rd June, 1958

The e.m.f. of the cell (Pt)H2/H2S04(m)/PbSO4, PbOz(Pt) has been measured over the
temperature range 555°C and up to a sulphuric acid concentration of 8 m. The results
obtained agree satisfactorily with the vapour pressure data on the activities of water and
sulphuric acid and with the calorimetric data on the heat contents of the acid and water, if
the cell reaction is assumed to be that represented by the equation,
H2 + H2SO4 + PbO2 = 2H20 + PbS04.
In an earlier paper 1 we discussed the behaviour of the lead accumulator and
concluded that it is in accordance with Gladstone and Tribe’s “ double sulphation ”
theory of the cell reaction, and that this theory alone fits the data. This conclusion
was based on an analysis of the available e.m.f. data including those of Hamer 2 for
the cell
(Pt) H2/H2S04 (m)/PbS04, PbOZ(Pt). (1)
As pointed out by Shankman and Gordon,3 however, there is a small but definite
discrepancy between the values of the activities of water and sulphuric acid deduced
from their vapour pressure measurements and those calculated from the e.m.f. data
of Hamer 2 for cell I and also for the cell 4
(Pt) H2/H2S04 (m)/Hg2S04, Hg.
The discrepancy which corresponds to a maximum of about 2mV in the e.m.f. was
confirmed by the later isopiestic measurements of Stokes.5 Various suggestions
have been made to account for the deviation between the two sets of measurements
including a recent suggestion6 that the reaction between the lead dioxide and
sulphuric acid at the higher concentrations with evolution of oxygen may prevent
strictly reversible behaviour of cell I. The results of our experiments in which we
have measured the e.m.f. of cell I over a range of concentrations from 0.1 to about
8 m and at temperatures from 555°C show that the explanation is simpler, namely
that Hamer’s e.m.f. data are in error.

EXPERIMENTAL
MATERIm
WATER.-Laboratory-distilled water was twice distilled in a Pyrex apparatus, the first
time from a trace of alkalinepotassium permanganate,the second from a trace of phosphoric
acid.
SULPHURIC AcID.-Sulphuric acid solutions were made from B.D.H. A.R. sulphuric acid
of specific gravity 1.84. The acid was tested for sulphide, sulphite and chloride and was
found to be free from these impurities.
LEAD NITRATE.-Lead nitrate used in the preparation of lead sulphate and lead dioxide
was precipitated twice from a hot concentrated solution (60°C) of B.D.H. A.R. lead nitrate
by adding A.R. nitric acid to the solution. The precipitate was sucked dry in a sintered-
glass funnel and then transferred to a crystallizing basin which was covered by a watch
glass and left in an electric oven at 110°C for several hours.
33 1
 View Article Online

332 LEAD DIOXIDE ELECTRODE

LEAD S U L P H A T E . - ~ ~ ~ml
of 0-5 M lead nitrate solution and 500 ml of 0.5 pvf sulphuric
acid solution were added drop by drop from two dropping funnels to 2 1. of nearly boiling,
continuously stirred 0.1 M sulphuric acid solution. The precipitate of lead sulphate was
washed with water and digested in 2 M sulphuric acid solution for 24 h on a water-bath.
After digestion the lead sulphate was repeatedly washed with water and then stored in a
glass-stoppered flask containing 0.2 % sulphuric acid solution.
LEAD DIOXIDE.-A solution containing 100 g lead nitrate and 400 ml concentrated
Published on 01 January 1959. Downloaded by University of Sheffield on 06/07/2014 12:28:37.

nitric acid in a total volume of 3 1. was electrolyzed at a temperature of 91-93°C and at an

anodic c.d. of 1.5 A/cm2 for 24; h, or, in one preparation, at 0.25 A/cm2 for 43 h. The
anode was a platinum gauze of estimated area 3 cm2 except at the lower c.d. at which a
platinum foil 8 cm2 in area was used ; the cathode consisted of a platinum wire and was
surrounded by a porous cup. Throughout electrolysis the solution was stirred by an
electrically driven glass stirrer.
At both current densities lead dioxide deposited in the form of a black, non-adherent
powder. This powder was collected, thoroughly washed with water and digested in 3 M
sulphuric acid for one week at 98-100°C. Digestion was carried out in a 250ml flask
containing about 150 ml of the acid solution and immersed up to its neck in a constant
level water bath kept within 2°C of boiling. A water-cooledrefluxing column connected to
the flask through a ground-glass joint prevented loss of water during digestion. A small
beaker placed over the top of the column prevented particles of dust from entering the flask,
while at the same time allowing the gas slowly generated during digestion, and presumably
oxygen formed by reaction between lead dioxide and sulphuric acid, to escape. At no
stage before, during or after digestion was the oxide ground or disturbed mechanically in
any other way. Earlier experiments had indicated that grinding the oxide resulted in
poorer reproducibility and constancy of the potentials of lead dioxide, lead sulphate
electrodes. After digestionall preparations of lead dioxidewere kept under 0.4 % sulphuric
acid in glass-stoppered conical flasks.
SULPHURIC ACID soLmIom-Solutions more concentrated than 0.3 m were made by
weighing out the calculated amounts of A.R. acid and water. Approximately 0.3 m and
more dilute solutions were made in a similar manner from a 6.095 m sulphuric acid solution
and water. All solutions were stored in Pyrex bottles fitted with ground-glass stoppers.
The molality of each solution was determined with an estimated error of f 0.05 % by titra-
tion against standard sodium hydroxide solutions using bromocresol purple as indicator.
APPARATUS
ELECTRICAL EQUIPMENT.-The e.m.f.'s were measured on a Tinsley vernier potentio-
meter reading directly to 0.01 mV using a Cambridge spot galvanometer of resistance 450
ohms and sensitivity 250 mm/pA as null detector. The whole apparatus was placed on a
metal shield earthed to the high-pressure water mains.
THERMOSTATS.-The water thermostats were maintained at temperatures of 5, 10,20,25,
35 and 45°C within the limits f0.01'C and at 55°Cwithin the limits & O.O4"C, by means of
+
mercury alcohol regulators. To obtain the temperatures of 5 and 10°C ethylene glycol
from a Frigidaire refrigerator was circulated through a coil of copper tubing immersed in
the water-bath. All temperatures were determined with an N.P.L. calibrated mercury-in-
glass thermometer which could be read to 0.005".
For some of the measurementsat 25°C an air thermostat was also used. The temperature
of the air oscillated within the limits i0.3"Cbut fairly rapidly so that the temperature of a
sealed tube of water kept in the thermostat and having a heat capacity less than that of any
of the cells used in the experiments varied by i 0.05"C only. The inner surface of the
walls of the thermostat was lined with zinc which together with the water thermostat tanks
was earthed to the high-pressure water mains.
CELL vEssELs.-Cell vessels were made from Pyrex or Monax glass and were of two types.
The first (I) was essentially an H-type cell vessel the two halves of which were connected
through a tap which was used to prevent hydrogen coming into contact with lead dioxide.
The second type of vessel (11) was designed to permit up to six oxide electrodes to be
tested at the same time. For this purpose the tube for the oxide electrode in vessel I was
replaced by six similar tubes connected to a central chamber (fig. 1). To prevent mixing of
solids from different oxide electrodes, the lead sulphate, lead dioxide pastes were placed in
the separate electrode vessels B (fig. 1). These were made from B14 thermometer pockets
and had a small pin-hole about 1 mm in diameter and about 1 cm above the level of the
paste to allow contact between the solution in the electrode vessel and the rest of the
solution in the cell.
 View Article Online

W . H . B E C K , K . P . S I N G H A N D W . F . K. W Y N N E - J O N E S 333
HYDROGEN ELECTRODES.-EleCtrOdeS consisted of a platinum foil, 1 cm x 2 cm. welded
to a platinum wire which was sealed into a B14 Monax thermometer pocket. They were
covered with a deposit of platinum black according to the instructions of Popoff, Kunz and
Snow 7 omitting the gold flash. Electrodes differing by more than 0.05 mV were given a
fresh coating of platinum black. Hydrogen was obtained from a cylinder and was passed
through a tube containing active copper at 400-600°C. Before entering a cell vessel, the
gas was saturated with the vapour of the test solution, by passage through four bubblers
Published on 01 January 1959. Downloaded by University of Sheffield on 06/07/2014 12:28:37.

similar to those described by Bichowsky and Storch.8

FIG.^^ .-Cell vessel 11.

LEAD DIOXIDE, LEAD SULPHATE ELEcTRoDEs.-Oxide electrodes were made by vigorously

shaking approximately equal amounts of lead dioxide and lead sulphate with a few ml of
test solution in a glass-stoppered tube. The solution was renewed five or six times at
intervals of 20 to 30 min before the paste of lead dioxide and lead sulphate was transferred
to a cell or electrode vessel. Electrical contact with the paste was made with a platinum
electrode (A in fig. 1) which consisted of a piece of platinum foil 0.5 x 0-5 cm welded to
platinum wire sealed into a soda-glass tube carrying a B14 standard glass joint. Before use
these platinum electrodes were cleaned by immersion in boiling concentratednitric acid, and
thoroughly rinsed with tap and laboratory-distilledwater and finally with test solution.
ASSEMBLYOF cELLS.-The practice adopted by Hamer'2 of filling the cell vessels in vucuo
was not followed,though care was taken to reduce exposureto the atmosphereto a minimum.
All ground joints and the tap on the hydrogen outlet limb of the cell vessels were well-
greased. Since the tap on the cross-limb of the cells was always kept closed after the cells
had been set up, a strip in the centre of the barrel and key had to be left ungreased to provide
an electrically conducting path between the two halves of the cell. A hard-grade high-
vacuum grease was used for this purpose except for some measurementsat 25°C for which a
soft-grade high-vacuum grease was employed.

PRESSURE CORRECTION
All e.m.f.'s were corrected to a standard hydrogen pressure of 1 atm, using the vapour-
pressure data for aqueous sulphuric-acid solutions of Bronsted,9 Grollman and Frazer 10
and Collins 11 for this purpose. No correction was applied for the depth of the hydrogen
inlet jet below the surface of the solution in the hydrogen-electrode compartment of the
cells. The maximum value of this depth was 4 cm and the correction for this according to
the equation given by Hills and Ives 12 is 1-2mm of mercury in the hydrogen pressure, and
therefore only 0.02 mV in the e.m.f.
 View Article Online

334 LEAD DIOXIDE ELECTRODE

RESULTS
REPRODUCIBILITY AND CONSTANCY OF ELECTRODE POTENTIALS
Measurements were first carried out to test the reproducibility and constancy of the
potentials of the oxide electrodes. For this purpose several electrodes were made from
each of five batches of lead dioxide including the one made at the lower c.d. The potentials
of these electrodes with respect to a hydrogen electrode dipping in the same solution were
Published on 01 January 1959. Downloaded by University of Sheffield on 06/07/2014 12:28:37.

then determined on a number of days over a period of up to 30 days, at one or more

concentrations in the range 0.1 to 8-272m. For each determination at least six readings of
the p.d. between the electrode under test and the hydrogen electrode were taken at approxi-
mately half-hourly intervals. These agreed to within f 0.1 mV except at 6.095 and 8.272 m
where the potentials of the electrodes tested rose continuously during the first day of test.
Between each determination of their potential, the oxide electrodes were stored in closed
tubes containing test solution at 25°C. The measurements were made with cell vessels of
type I1 in the air thermostat at 25°C.

5 10 15 days 5 10 15 days

-0.1
L I I I I 1

+O.I

-0.1

1
0 IS 20 25
1
30 dcys
I
FIG. 2.-Variation of e.m.f. with time.

The constancy of the potentials is illustrated in fig. 2, where the deviation of the potential
on a particular day from its mean value over the whole period of test has been plotted
against the time for three electrodes tested respectively at 0.1, 6.095 and 8.272 m. The
result for the lowest concentration does not really indicate a tendency for the potential to
decrease with time. When all the results are taken into account no significant trend is
discernible, the values of the potential rising and falling within the limits & 0.12 mV except
at 8.272 m, where as fig. 2 shows the limits after the first day were & 0.2 mV.

TABLE1
no. of mean e.m.f. max. deviation
electrodes V mV
5 1.56728 + 0.06
- 0.04
0.47 17 13 1.60539 + 0-14
- 0.10
1.129 10 1.63179 + 0.11
- 0.09
2.217 13 1.65946 + 0.14
- 0.11
3.900 2 1.69443 i 0-02
6.095 2 1.73514 & 0.02
8.272 1 1-77099

The results of all the measurements are summarized in table 1. At each concentration
the number of electrodes tested is shown in column 2, the mean e.m.f., obtained by averaging
the values for the individual electrodes, in column 3, and the maximum deviation of these
values from the mean in column 4. These results show that the reproducibility was
better than f 0.15 mV. The measurements also showed that there was no significant
 View Article Online

W. H. BECK, K . P . SINGH AND W. F . K . WYNNE-JONES 335

difference between electrodes made from lead dioxide prepared at the higher c.d. and those
from oxide prepared at the lower c.d. ; between electrodesfreshly made up from the stocks of
lead dioxide and lead sulphate and those that had already been tested at other concentra-
tions ; or between such electrodes and others made by combining the pastes of several
electrodes previously tested at other concentrations.
DEPENDENCE OF e.m.f. ON ACID CONCENTRATION AND TEMPERATURE
Published on 01 January 1959. Downloaded by University of Sheffield on 06/07/2014 12:28:37.

In a second series of experimentsthe e.m.f. of cell (I) was measured at a series of tempera-
tures in the range 5-55°C at acid concentrations from 0-1 to 7.199 m. For these measure-
ments two stocks of lead dioxide, lead sulphate paste were used. The first was made by
combining all the electrode pastes previously made from one batch of lead dioxide for the
reproducibility tests just described. The second, which was used to provide a check at one
concentration,was made in the same way from a11 the electrode pastes made for these tests
from a second batch of lead dioxide. Both batches of lead dioxidehad been prepared at the
higher c.d. Cells were assembled using for the lead dioxideelectrodes,paste taken from one
or other of these two stocks. When the e.m.f. of a cell had been measured over the whole
temperature range as described below, the paste of lead dioxide and lead sulphate was re-
covered from the cell, washed with thrice distilled water and restored to the appropriate stock
which was kept under approximately0-1m sulphuricacid in a glass-stoppered tube. Further
samples of paste were then withdrawn for other cells, the pastes being alwaysrecovered from
each cell and restored to the stock. The cell vessels used were of type I.
TABLE2
e.m.f. (V)
temp.
OK 0.1000 m 0-1996 m 0.2917 m 0-4717m 1.129 m 2.217 m 3.900 m 4.973 m 6.095 m 7.199m

278.16 1.57410 1.59097 1.60031 1.61237 1.63738 1.66364 1.69856 1.71984 1.74035 1.75965
283.16 1.57173 1.58893 1.59833 1.61065 1.63606 1.66259 1.69743 1.71864 1.73891 1.75820
293.16 1.56730 1.58480 1.59450 1.60720 1-63336 1.66045 1.69524 1.71603 1.73640 1.75560
298.16 1.56509 1.58284 1.59266 1.60550 1.63195 1.65943 1.69420 1.71491 1.73516 1.75425
308.16 1,56086 1.57887 1.58896 1.60232 1.62948 1.65720 1.69212 1.71260 1.73285 1.75187
318.16 155683 1.57510 1.58544 1.59924 1.62701 1.65518 1.69013 1.71056 1.73073 1.74964
328.16 1-55284 1-57146 1.58197 1.59617 1.62455 1.65320 1.68826 1.70878 1.72856 1.74740

The procedure followed was to measure the e.m.f. of a cell first in the water thermostat at
25°C and then in the tanks at 5, 10,20,25,35,45,55 and 25°C in that order. Three days
were required to cover the complete temperature range, the cell being kept closed by night in
the tank at 25°C. In the first run at 25°C hydrogen was bubbled copiously for nearly 2 h
and thereafter at a lower rate. The e.m.f. was by
then constant to within i 0.05 m't' except at T~~~~ 3 . - co- ~ ~
6.095 m and 7.199 my where it continued to rise EFFICIENT op e.m.f. Am HEAT
slowly on the first day but became constant to within AT 2 5 0 ~ .
f 0-05 mV by the next morning. When the cell
was transferred to the tank at 5"C, a fairly rapid __ X lo3
--dE
Hzso4 dT
-AH
stream of hydrogen for 1 h was sufficient to bring m (kcall
03
the e.m.f. to a value constant to within & 0.05 mV
+
and h in the remaining runs. The three deter- o'looO 0.4320 78-156
minations of the e.m.f at 25°C agreed well with each 0'1996 0-3967 78-490
other except at the highest concentration 7.199m 0.2917 0.3721 78.604
where the third and highest value exceeded the first 0'4717 0-3290 78.604
and lowest by 0-3 mV. One cell only was measured 1*129 0.2570 78-834
at each concentration except at 1-129m where a 2'217 0.2122 79.485
second cell made from the second stock of paste 3.900 0.2104 81 -065
was measured as a check. 4.973 0.2314 82.309
The results of the measurements are given in 6'095 0.2417 83.385
table 2 where the values for 25°C are those found on 7'199 0-2512 84.397
the second day. Results have been given for only
one of the two cells measured at 1.129m. The e.m.f. of the second cell differed from
those given by not more than k 0-1mV.
Temperature coefficients were obtained by fitting quadratic equations to the e.m.f.,
temperature data at each concentration and differentiating. These equations reproduced
 View Article Online

336 L E A D D I O X I D E ELECTRODE

the experimental e.m.f. to within f.0.05 mV. The temperature coefficients obtained from
them are shown in table 3, and the values listed for the heat of reaction in the same table were
calculated using the equation

DISCUSSION
Published on 01 January 1959. Downloaded by University of Sheffield on 06/07/2014 12:28:37.

THE DEPENDENCE OF e.m.f. ON ACID CONCENTRATION

If the reaction
Hz + H2SO4 + PbOz = PbS04 + 2H20 (2)
occurs reversibly in cell I, then the e.m.f. of this cell should vary with acid concentra-
tion according to the equation
+
E = E" (RT/2F)ln(4m3r,t3/a2w), (3)
where yf and a, are the stoichiometric mean activity coefficient of sulphuric acid,
and the activity of water respectively at molality rn. The value of E" calculated
from the observed e.m.f. by means of this equation should therefore be constant.
When the results of the measurements at 25°C in both the air and water thermostats
are tested in this way, they yield the values shown in the last two columns of table 4.
The values quoted in columns 4 and 5 of the table for the activity coefficients of
acid and the activities of water respectively have been interpolated from the data
given by Stokes.13 The same test has been applied to the measurements of Hamer 2
on the same cell at 25°C and the results are shown in fig. 3 where the values of E"
taken from the last column of table 4 have been included for comparison.

1.6872 c I I

0 n

0
0 0
0
0
0
1.6842 -
0 (R TI2 F) I n (4m38$3?w)
I-I -
1 -

It is clear from table 4 and fig. 3 that the results of the present investigation
do not confirm the small discrepancy between the values of the activities of water
and sulphuric acid calculated from Hamer's e.m.f.'s and those calculated from
vapour-pressure measurements. On the contrary they show that the variation of
the e.m.f. with acid concentration is in strict accordance with the assumption that
the reaction represented by eqn. (2) occurs reversibly over the whole concentration
range examined, if the values obtained from vapour pressure measurements are
taken for the activities of acid and water.
 View Article Online

W. H . B E C K , K . P . SINGH AND W. F. K . WYNNE-JONES 337

TABLE
4.
A = (RT/2F) In (4rn3y*3/uw2) = in V at 25°C.
0.02957 1oglo(4rn~3y~3/a,2)
e.m.f. (V) EOW
H2S04 air water air water
m thermostat thermostat '
f =W A(n thermostat thermostat
Published on 01 January 1959. Downloaded by University of Sheffield on 06/07/2014 12:28:37.

01OOO 156509 0.2655 0.9963 - 0.12191 1.68700

-1085 1.56728 ,2584 -9961 - -11981 1.68709
-1996 1.58284 -2092 -9928 - *lo436 1.68720
*2917 1-59266 -1845 -9898 - -09451 1.68717
-4717 1.60539 1-60548 - 1 5 ~ ~9829 - -08170 1.68709 1-68718
1.129 1,63179 1.63195 *1293 -9561 - ,05520 1.68699 1-68715
2,217 1.65946 1.65943 -1294 -9003 - *02761 1.68707 1.68704
3.900 1.69443 1.69420 -1667 -7874 + -00735 1.68708 1~68685
4.973 1.71494 -2065 -7057 + -02775 1.68719
6.095 1.73514 1-73516 -2619 -6186 + -04816 1-68698 1.68700
7-199 1.75425 -3295 -5366 + -06709 1.68716
8,272 1.77099 -4071 -4635 4 -08433 1.68666

e.m.f.
THE TEMPERATURE COEFFICIENT OF THE
The heat AH of the reaction represented by eqn. (2) at an acid molality m, is
related to the standard heat of reaction AH", by the equation
AH = AH" f 2L1- L2,
where L1 and L2 are respectively the relative partial molal heat contents of water and
acid in the sulphuric acid solution of molality rn. Values for the quantity (2L1 - L2)
can be interpolated from calorimetric data for L1 and L2, while AH can be calculated
from the e.m.f. and temperaturecoefficients of the e.m.f. of cell I. If thecalorimetric
and e.m.f. data are in agreement

should be a constant independent of the acid molality. The result of this test is
shown in table 5 in which the values calculated for AH at 25°C from eqn. (1) are
listed in column 2, and the values found for (2L1 - L2) by interpolation from the
data of Craig and Vinal14 in column 3. The values obtained for AH" are shown in
the last column of the table and are also displayed in fig. 4 where values calculated
for AH" in the same way but from the e.m.f. data of Hamer are shown for comparison.
I I I I I
-73.00
0
AH'(kcal)
e
0 0

-72'75

-7250
t-.
ez .o
0.5
1
O

-___L--I;
I I3
ma 1
2.0
-
0
e u o
1
FIG.4.-Standard heat of cell reaction.
0 present work ; 0 Hamer's measurements.

It is clear that the agreement between the calorimetric and the present e.m.f.
data is excellent. It seems distinctly poorer, however, for Hamer's measurements,
suggesting that the small divergence between Hamer's results and theory first noted
in the dependence of e.m.f. on acid concentration is also reflected in the temperature
coefficient of the e.m.f.
 View Article Online

338 LEAD D I O X I D E ELECTRODE

As our results at 25°C are in satisfactory agreement with the vapour-pressure

data on the activities of water and sulphuric acid and as the temperature coefficients
are also in good agreement with the calorimetric measurements, we regard it as
definitely established that the cell reaction of the lead accumulator is that postulated
in the reaction equation,
Pb + Pb02 + 2H2SO4 = 2PbS04 + 2H20.
Published on 01 January 1959. Downloaded by University of Sheffield on 06/07/2014 12:28:37.

TABLE
5.
H2SO4 -AH -W1--Lz) -A H 0
m (kcall kcal kcal
0.1OOo 78.156 5-472 72684
*1996 78-490 5.771 72.7 19
-2917 78.604 5-854 72.750
-4717 78.604 5.870 72-734
1.129 78.834 6-085 72-749
2-217 79.485 6.739 72-746
3.900 81-065 8-320 72.745
4.973 82.309 9-535 72.774
6-095 83.385 10.660 72.725
7-199 84-397 11-676 72-721

average 72-735

The PbO,/PbS04 electrode can therefore be regarded as behaving reversibly at

least over the temperature range 545°C and up to a sulphuric acid concentration of
8 m. The electrode requires care in preparation but we are unable to find any
difference in the methods used by ourselves and by Hamer from the published
accounts to explain the difference in our results.

1 Beck and Wynne-Jones, Trans. Faraaizy Sac., 1954, 50, 136.

2 Hamer, J. Amer. Chem. Sac., 1935, 57,9.
3 Shankman and Gordon, J. Amer. Chem. Sac., 1939, 61, 2370.
4 Harned and Hamer, J. Amer. Chem. Sac., 1935,57,27.
5 Stokes, J. Amer. Chem. Sac., 1947,69, 1291.
6 Hamer, Nat. Bur. Stand., circ. 524 (Electrochemical Constants, 1953), p. 184.
7 Popoff, Kunz and Snow, J. Physic. Chem., 1928,32,1056.
* Bichowsky and Storch, J. Amer. Chem. Sac., 1915,37,2696.
9 Bronsted, 2.physik. Chem., 1910, 68, 693.
10 Grollman and Frazer, J. Amer. Chem. Soc., 1925,47, 712.
11 Collins, J. Physic. Chem., 1933, 37, 1191.
12 Hills and Ives, Nature, 1949, 163,997.
13 Stokes, Trans. Faraday Sac., 1948,44,295.
14 Craig and Vinal, J. Res. Nat. Bur. Stand., 1940, 24, 482.