The electrical double layer on thallium amalgam electrodes

JOURSAL

THE

OP ELECTRO_.S.zXLk-TIC-ALCHEMISTR17

ELECTRICAL

DOUBLE

LXYER

ON THALLIUM

AM_4LGX&I

ELECTRODES

Tyco

Laboratorces

(Rcccivcd

IMC.,

No\-ember

Bear

gth,

HzIl.

IF-altlrtrm

53.

_‘LInss. (c‘.S._-I

.)

~364)

1XTROL)UCTIOS In connection with a study of hydrogen overvoltage on thallium amalgam+, we have made some measurements of double-layer capacity on thallium amalgam electrodes in 0.1 II1 HC104 at ~5~. In addition, a criticalsurvey was made of the literatureconcerningthe electricaldoublelayer on thallium amalgams.Inthis paper, our experimental results are presented and compared with the literature values. In i-iew of the extensive literature data available on capacity, and the difficulty-of measuringaccuratezero-chargepotentials,a more detailed study- of the double layer on thallium amalgams does not seem appropriate at present. Measurements of the double-layer capacity of thallium amalgams have been made by DELAIL~Y _.XD KLEISERJI_.S~, and by BOGUSL_~~SKII -*SD DXMIASKIX~. Both sets of measurements appear to be consistent, butneitherinvestigator measured the double-layercapacityin 0.1 MHClO+themediumwhichwe choseforour overvoltage measurements_ Furthermore, one peculiarity exists in the measurements of capacity on 40% thallium amalgams in aqueous sodium fluoride solutionss. A potentialindependent frequency dispersion effect was observed, which is difficultto explain by any of the current theories of the electricaldoublelayer. We therefore thought it advisable to make some additional measurements of capacity, which we have reportedhere. From the data in the literature on the electrocapillary curves of thallium amalgam.+-10, together with capacity data"3, it is possible to obtain some information aboutthesurfacecomposition of the amalgams Theinfluence of surface composition on the rate of hydrogen evolution is relevant to the more general question of how alloy composition affects the rate of electrode reactions. ESPERINENTX The amalgams were prepared from triple-distilled mercury (Doe and Ingalls) pure thallium (American Smelting and Refining Co.) by weighing out and 99-999",/ therequiredquantitiesofmercuryandthalliumandcombiningthemunderanatmosphereof.argon.Theamalgamwasallowedtostandovernighttobecomehomogeneous, andthenfilteredinto the dropping-electrode reservoir_whereitwas stored.Thalliurn J_ EZectroanaZ.

Chem.,

9 (1965)

149-162

amalganllj; oxidize readily on exposure to air, but the material in the reservoir was maintained free of oxide by keepin, * it under an atmosphere of argon The method of measuring capacity was essentially that of GR_wAXEI~~~~ and has been described in detail presiousI_v f3_ Zero-charge patentials were measured b_v the streaming-electrode method 1+-L@_ Provided the electrode is ideally polarized, the potential at which no current flows between a rapidly dropping electrode and a reference electrode corresponds to the condition where there is no charging current, and hence no excess surface charge at the metal-salution interface; this is the zerncharge potential, or eIectrocapiIlar~- masimum. However, the cond.&ions %x- ideal polartiaation were nut perfectly satisfied by a thallium amalgam electrode. At the potential where RO current was observed, the capacity showed a slight frequency dispersion and the polarization resistance increed, indicating that thallium was dissolving at an observable rate. Furthermore, the measured zero-current potential was different, depending on whether it was approached from the negative side or the positive side. If the zero-current point XY+.S approached consistently from the negative side, and great care was taken to l&event any dissolution of amalgam. in contact with the solution, reproducible values were obtained_ On the other hand, if the dropping electrode, or the pool of amalgam at the bottom “of the cell, was allowed to become even 0.1 V more positive than the z~ro~cu~~~tpotential, tlrallium dissolved in the electrolyte, and the zero-current pot&ntials observed were &-reproducible_ The reproducible values couId be restored after long cathodic polarization, We have given the zero-current potentials measured in the absence of any thallium ion in the solution as zero-charge potentials, but in fact the true zero-&a-e potential is probably somewhat more positive, as will be discussed later. Extreme care was taken to elim.inate impurities in the solution. Sealed watertight juints and Teflon stopcocks were used to avoid ay organic material. The cell was washed with a chromic-sulphuric acid cleaning mixture and rinsed with tripledistilled conductivity water. Solutions were made from reagent-grade perchloric acid (J. T_ Bake&and conductivity water. The solution was pre-electrolyzed overnight at x mA@rn~, using an a-Gary mercury pool as cathode_ The mercury pool was discarded before any measurements were made_ Ail measurement-3 were made in o. IOO LM per&lo& acid solution at 25.0”. The electrolyte was saturated with hydrogen througbout the experiment.

In l?ig_ x is presented a suullm ary of the available data on zero-charge potent&&s for t~um amalgams: The tiost precise measurements are those made by FRWMFCIX _hND G~RQD~~-~KAYA",who’measu.red the electrocapillary curves using a Lippman capillary electrometer, . a-d deterroin ed the point where the interfacial tension be-_ tweei the ar@lgam_ and N Na&C3~ solution_was a maximum, A similar set of measurem&+s eras-made .iri NaCN solutions by DELA~&Y AND.&E&ZRMAN~, using the le$s-

p2&~~~d~qkime m&h&d for d&&king the interkial tension_ These authors were una@lk to c&taih a k&r maximum, and determined the iero-charge point from the &3t

df.fh$cat&o_$i~%@ancl% bf &he electr&al$llary

cu.z~e with respect to the curve for

ELECTRICXL

DOUBLE

pure mercury. amalgams carried

LAYER

OX THALLILrM

The potent&s

than

those

XiSALGABI

ELECTRODES

*5=

so obtained are more than 0.1 V more positive in dilute

of FRUMKS

AND

GORODETZKA~A.

FRUMKIS ASD CIR~‘~~14 measured the potential at which no current was by a rapid dropping electrode, adjusting the potential by changing the con-

centration of thallium ion in the electrolyte_ These results agreed very closely with electrocapillary maximum measurements of FRUJIKIK AXD GORODETZKAYX*.

MOLE

Fig. I. Comparison BOGV’SLAVSKI1 Icxsrx~; x .

iiN

FRIJMKIX

FRACTION

of zero-charge

BOGUSLAWKI~

l-l

potentials

DAXASIiI~J; rl AXI) CIRX-ES’;.

the

DEL_~H_AY

measured AWD

by

various

~iLEISERL3I_Alu=;

workers: -+,

FRUMKIN

0.

thus work;

q

AZVD GORODETZ-

AXD DAMASKIS~ measured

the differential capacity of a senes and Identified the ~n~urn in the capacity curve with

of amalgams in 0.01 It;’ XaF, the zero-charge point. Recause they observed a frequency-dependence of the capacity at these potentials, they felt that there was some question as to whether the minimum

in capacity really corresponded to the zero-charge potential. Presumably a reversible dissolution of thallium was taking place, contributing a pseudo-capacity, which distorted the shape of the capacity curve, and shifted the miniruum to more negative potentials. The values obtained by this method are as much as 0.1 V more negative than those of FRUMKIS AND GORODETZKAYA~. Our measurements were made by measuring the potential at which no current flowed between a reference electrode and a rapidly dropping electrode. Our experimental arrangement was essentially the same as that of FRUMKIN _~XD CIRVES~*, except that the concentration of Tl+ in the electrolyte was kept as small as possible by never allowing the dropping electrode or pool of amalgam to become more positive than the zero-charge potential_ Our values agree most closely with those obtained by BOGUSLAVSKXI AND DAMASKIK_ In the concentrated amalgams, the results obtained by all workers are the same

within

0.02

V,

which

can

be

considered

adequate

agreement_

In more dilute

amalgams, however, the discrepaucy increases to as much as 0.2 V, which indicates some gross systematic difference in the results of the various measurement techniques. J.

Etectroanal.

Cfrem., g (rg65) rgg-162

,

x53

J

s.

RLXLER

The difficulty arises prin~arily from the dissolution of thahium. If this reaction can proceed at a finite rate, the electrode is no longer Ideally polarized, and the equations derived under that assumptions* are rm Zonger appIicabIe_ MOHXLXER~~ has treated the case where one of the components of the electrode undergoes a resersible charge eschange reaction to give one of the ~~mpoll~nt~ of the electrolyte. At the electrocapillary maximum, he finds that the charge on the electrode is not zero, but rather that the following relation is satisfied:

where y is interfacidi tension, Q is electrode potentia1, ~8 is chemical potential, Q is rhe excess electronic charge on the surface of the electrode, ~TP is the surface excess of TIT on the electrolyte side of the interface, and F is the faraday constant. A similar conclusion was reached by FRUMKIX AXD GORODETZUY_~~_ This means that under reversible conditions, the maximum of the eIectro5apillaz-y curve is not the same as the zerwcharge point. The true zero-charge point cannot be rigorously measured under conditions where there is a reversible reaction taking pIace_

In Table I are listed the results of our capacity measurements on three amalgams in 0 700 M NC104 at 2.fj”.The measurements shown were made at 5 kc, but ex-

2

r_m5

-

2 1-5

--x.20 --IzTO -

IA30

*

I T-9 IS.0 1S.S

“I_.+

21.8 21.6

2x.7 23.8 (&Lo) = ag.6

-Qo.go -

~0.6 20”9

o.so

-0,“io

-aGo

38-F 42?_4b’=

--o@ --o-5.5 52-g

-0-p

_--o-45 -0.40 --0:35

24-3 27.0 30-3

=

c5J-31'*c 34-I=

_ 66.8~

ELECTRICAL

DOUBLE

LAYER

OS

THALLIUM

AXkLGAM

153

ELECTRODES

cept where noted (potentials more positive than the zero-charge point) the values were independent of frequency from o.=j-xo kc. The polarization resistance was always less than 0.1 Q cm”. The results quoted in the table represent averages of 2-4 measurements of capacity, but the data were not othervvise smoothed. The value obtained for 31.2% amalgam at -o.So V was discarded because of a fault_v measurement of balance time, and the value listed in the table at this point was interpolated from a smooth curve through the other values. These results are plotted in Fig. 2, together with a capacity curve for pure mercury in the same potential region, for comparison. The zero-charge points shown in Fig z are those of FRUMWS ASII GOROIIETZEXYA~.

-”

--Lz

--1 I

POTENTIAL

Fig chzge

-ID

vs. REV

-Q9

-08

-07

H2 ELECTRODE.

-06

-05

O.lN

HCIO,

-04

2. Differential capacity of thallium amalgam electrodes: 0, experimental point (ref. 4). (a), 40-s% Tl, (b). 31.2% Tl; (c), IO.I~/~ Tl; (d), pure Hg.

data;

0,

zero-

the capacity of thallium amalgams and 307(~ amalgams agree with ours within o-5 .uF/crn” at potentials more negative than -0-55 V in the case of the IO?/, amalV in the case of the 30% amalgam. At potentials more positive than g=-G and -02 these values, DELAEAY AXD KLEINERMA~T’S values are higher than ours, presumably because CN- is more strongly adsorbed at the electrode surface than is ClOa-. In Fig_ 3, our capacity me&urements on 40-5 “/o thallium amalgam in 0.1 M HC104 are compared xvith the measurements made by BOGUSLAVSKII AND DAM_~sKIN~ -on 40.0°i0 amalgam in 0.1 M KCL Within the experimental error, our data show no DELAHXY

AND KLEINERMAK~

also measured

in 0.2 &I NaCN at 30”. Their results for 10%

J. EbctroanaL Chenz., g (1965) 149-162

.$&&tidence 06 frequency from 0.5-x0 kc. At those points where a detectable differ&nce -is &served between measurements at different frequencies, both the 0.5 kc .+id the 16 kg :tiegsuremerxts xqere lower than the more accurate 5 kc measurements. These d..i$x~~~&‘,6a~ almost certainly be attributed to stray capacitances and inducta&sin-the app&&u&.~& contrast, BOGUSLAT~‘SKXI ASD DAMASRIX’S me~urem~nts at okg‘.kc ar&‘higher than those at r.o kc by 1-r-5 @?/cm”, greater by a factor of five than the.sc&&.of-.tfieir da& abatit a smooth curve_

-1.5 ._.I

..-tzr

.-.-is

-G?

--1J

-LO

-as

$.OTESWiL\i~-vi‘F&Ln;

-?.a

-0.7

ELECTf?UDE.

-ck§

O.itd

-a5 &iCtu,

ELECTRICAL

DOUBLE

LAYER

OX THALLIU&f

AMALG_4XI

ELECTRODES

155

cussion that the values measured by various methods disagree because of the reversible dissolution of thallium from the amalgam. MOHILNER~~ has shown that it is possible to calculate the difference of potential across the diffuse double layer, even when one of the components can be reversibly transferred across the interface_ He has shown that this potential drop is zero not when the surface charge is zero, but rather when the condition of eq. (I) is satisfied. Unfortunately, under the conditions of our overvoltage experiments, the criteria for reversibility- are far from satisfied_ In the absence of any thallium ion in the electrolyte, the thallium amalgam electrode approximates an ideally polarized electrode at potentials more negative than the region of the zero-charge potential. Thus IMOHILXER’S equations cannot be used to calculate the potential drop across the diffuse double layer at the potentials where hydrogen e\-olution is appreciable_ On the other hand, the equations for an ideally polarized electrode without specific adsorption i break down in the region of the zero-charge potential, thereby makmg the calculated values of the potential drop across the diffuse double layer somewhat uncertain. Since the experimental conditions of FRUME~IX -*ND GORODETZKAYA~. and approximate most closely the conditions where the potential FRUIkIKIh- AND cIR\‘ES14, drop across the diffuse double layer is zero, we have chosen their values for the zerocharge potentia1, and used the theoretical equations for the ideally polarized electrode to calculate properties of the double layer. Although such a procedure is far from rigorous, the agreement of the integrated capacity curve with the experimental electrocapillarcurve (discussed later) shows that the errors introduced in the overpotential correction1 are at most a few millivolts, approximately the same order of magnitude as the experimental errors in the overvoltage measurements.

I -G!

I

-u

F+TEQTlAL

I -ICI vs..

1 -as Rr7t.

I -aa

H2

ELE-E.

1 -a7

1

1

-0.E

-a5

O.IN

HCIO,

Fig. 4. Integral capacity of the Helmholtz polynomjal functions given in Table 2_ 0.

-

-double layer_ The curves y(., Tl; I), 31_2*/~ Tl; +

40-5

J- ELzc~voana2_

were calculated from ro.r% Tl: A, Hg. Chem.,

g (1965)

the

x49-162

The integral capacity of the Hehnholtz layer, calculated using the zero-charge potentials given in Table I, accordin g to the iteration method described previously~a are shown in Fig. 4. For ease in subsequent computations, these results were fitted by the least-squares nlethod to a quadratic ~~lynornia~: Ko = co + cr $25,+ cz #bo”

(2)

where $a is the potentiaf with respect tu the zero-charge pamt. The curves shown on Fig_ 4 were calculated from eqn. (2) usin,= the coefficients listed in Table 2. Note that the ~&ES for the amalgams are quite similar, but very different from those for mercury. At corresponding values of surface charge, the capacity of the thallium amalgams ss essentially the same, but nearly twice the value for pure mercur>-. This is in. agreement with our observations on indium amalgams in perchloric acidla. Figure 5 shows the curves af the potential drop across the diffuse double layer, $2, as a function 01 overpotentiai T. for mercury and the three amalgams, Note that the contribution of the diffuse double laver is an appreciable fraction of the total patential drop, approximately xoo mV at an overpotential of 1-0 ‘f’. The differences

ELECTRICAL DOUBLE LAYER ON THALLIUAI AMALGAJLELECTRODES

I57

between mercury and the amalgams, however, are only about IO mV m the potential range around 1.0 V. This effect results from the cancellation of two factors; although the zero-charge potential of the amalgams is more negative (larger value of 7). the capacity is larger, and hence the curve 42 rises more steeply. Fortuitously, the curves for mercury and the amalgams nearly intersect in the region between 0-S and r-3 V.

One of the factors which can influence the rate of the hydrogen evolution reaction at the surface of a thallium amalgam is the concentration of thallium and mercury at the surface. This concentration need not be the same as in the bulk and the ratio of surface to bulk concentration will in general be a function both of the bulk concentration and the potential. FRUAIKIX XXD GORODETZKXPA” have made measurements of the interfacial tension between thallium amalgams and normal Na2S04. Since at negative potentials, the specific adsorption of both SOq2- and ClOa- should be negligiblel2.1*, the interfacial tension of the thallium amalgams should be essentially the same whether the electrolyte is 0.1 M HClOd or 1.0 A; Na2SOa. The most sensitive way to see any differences in the structure of the electrical double layer is to look at the differential FRUMKIX AKD GORODETZKAE-A’S data are accurate enough to capacity. Fortunately, permit t\vo successive differentiations to obtain the capacity

c =

3”y 0.1

-

a$’

where C is the differential capacity in ,,~IF/cm2, */ is the inter-facial tension in erg/cm”, and 4 is the potential with respect to a reference electrodel”. This differentiation was done in two ways. First, the simple second-differences between the interfacial tension values were used. These results are plotted as crosses on Fig. 6, and show considerable scatter. The other method used was to fit a least-squares polynomial to the data and to differentiate in analytically_ The best fit was obtained with a Sth-order polyno2nial, and the values of its derivative are shown are circles on Fig. 6. The line on Fig. 6 represents our experimental measurements of the differential capacity (Table I). Except for a slight increase at potentials more negative than -1.0 V, the values obtained from the polynomial fit to FRUMKIS AXD GORODETZKAYA’S data agree with our experimental capacity within the experimental error. On the other hand, when our capacity data is integrated to obtain the interfacial tension, the results shown in Fig. 7 are obtained. The interfacial tension was obtained by integrating the capacity twice, fixing the position of the integrated curve to coincide with FRUMEIN AND GORODETZKAE-_a’_~measurement at the electrocapillary maximum_ In Fig. 7, their experimental data are shown as circles, and the result of inte,grating our capacity data is shown as a line. Even though the capacity is essentially the same in both cases (Fig. 6), our integrated interfacial tention is about I o/o higher than FRUMKTS AND GORODETZK-4~~‘s in the negative potential region. This is due in part to their slightly higher capacity values (which make the curve bend more steeply) but is primarily due to a shift in the zero-charge potential: The integrated curve can be made to agree within o.z~/~ with the data of ‘FRUMKIN AND GORODETZKAYA if the electrocapillary ma~~urn point is taken to be /_ ElectroanaE. Chem..

g (1965)

149-162

J_ X’*‘.BUTLER

-r-3

--tz

--IA

POTENTIAL

-,a

-as

VSL RRc

-ai9

-07

f-f2 EL~~X-RODE.

-0s

QIN

-05

--04

HCIO,

Fig_- 6.. Gompsrison of drfferentkl capacity of EOOj TI amalgam calcufated by differentiating the eiectrocapiflary cume xvith experimental differential capacity: -, differentiaI capacity (this work) ; X , 2nd difference of experimental pomts; Q. zsx3 derivative sth-order polynomial+.

approximateiy 15 mV more positive in 0.1 M HClC3, than in I AT Nak33~. One ex,planation for this is that the incipient dissolution of thahiurn at potentials near the ekctrocapillary maximum introduces uncertainties of this order of ma@tude. As we saw in our discussion of the zero-charge potential, this is quite a reasonable expkmation. The other possibibty is that the sulkte ion is slightly more strongly ad_-sorbed than the perchlorate ion at positive potentials. This would cause a shift in the electracapillary maximurk, but would also imply that the capacity in NazSOG should be higher than in WClO~ at the positive end of the curve. As c.an be seen from Fig. fjl thisk not the case. The question of specific adsorption of anions on thalhum amalgams but has-been briefly touched by FRUMIUN4*14 and by BOGUSLAVSKXI AND DAXASKIN~, detaiIe+ measurements and calculations have been carried out only for mercuryls. Being thus convinced that only minor differences exist between the directly measured ‘interfaciaf tension and the integrated interfacial tension, we can use the _con_c&tratio& depcndenco of interfacial tension to obtain the surface excess of thafltirun. The for@ of-&e Gibbs adso&&ion-isotherxr which applies to this case isal.sa

-where y-is th@ mter&ciat teusi_on, ar~is-tbeacf5vity of thaWurn in the arrialgam, and Z%I is#~e snrf~cebxoek cjf _thallinm, with respect to mercury. The -activity of -tk.IJium_ _. -1. -Eksti+x&~~. XZhmn.; g _fxg65) ~4g-x6?

._

_-

-_

ELECTRXCAL

-8-2

-‘.I

DOUBLE

-Lo

POTENTWL

LAYER

-0s

vs,

--QB

REV

ON

-07

THALLXUM

-0.6

H,EUXTRQDE.O.IM

_A~XALG_&f

ELECTRODES

=59

--03

-0.5

HCK?,

Fig. r_ Comparison of experimental interfacia1 tensian measurements with values for ro”ja Tl amafgam obtamed by integrating the capacity curve. The twu sets of values were made to coincide at the electracapillary maximum: -, integrated capacity curve (this work); 0, experimental interfaclal tensions.

amalgams is knixvn, hating been cdctiated by LEWIS AND I~ANDALL~~ from the E.M.F. measurements of RKWLRDS AND DANIEL-S**. T&t surface excess can be obtained from the concentration dependence of inte&atiaJ. tensi_on in Tao complementary ways. The first is to make use of eqn- (4) in the form given ; to plot y a5 a function of log a&, and measure the slope directly. Such

in its

3x30

J_ S_

BUTLER

a plot is shown ix. Fig 3. Note, however, that the slope at small ralues of the activity is very xaall, and uTLfessan expanded scale is used, difficult to measure accurately. The second method is to use eqn. (4) in the form

and -to measure the slope of a plot of y as a function of OTI_Such a plot is shown in Fig. 9. Note that the slope of the curve in Fig. 9 is greatest in the region where the slope of fhe’ curve in Fig. 8 is smallest. Thus the two methods complement each other, the fkst being more accurate in the high-concentration region, and the second being more accurate in the low-concentration region.

The values of Z’r, obtained by the two metbuds are shown on Fig- IO, for sev~~+al values of.poteni5a.l in the region where hydrogen overvoItage measurementsi have been made, Note that the values of .Pa are negative; the surface concentration of thal.hmn_ is lower than the bnlk concentration if the concentration of mercury is assum& to be uniform out to the interface,. The ratio of_rTl to the bnlk concentration is of interest, since it is relatively iudepetident of- concentxation,- ks can be seen from Fig” IX. If the variation of ton+.&ration with the surface -were the same for aU the amalgams, then . _distane&rom -T’v_+& woul$he independ& of concehtrakon. The decrease ok .Z%&TI as the con-cer~t&ic&3xxxxases +ow~ that in the ember-c~nceu~a~on ama.@*he surface ‘ck3&ukat&n is.probabIy closer to the bulk concentration than in the dilute amalgams. @houg& . the is known, tile aftsol& vaIue of the surface . - &&ace excess-Fm i5~qqti~&oae.canriof be’ ob+i.ned without prior knowledge of theshape of the conc&tition. ]e~~il&-G; a -d_+ectiopm pi?rpen&cdar_ to- $he suiface. _A cfose-packed mono&q@+: $f 3iA diameter -$Ixeres g&es a-sur$ace concetitka~on of 21. - 10-10~ mole~/ccm~

ELECTRICAL

DOUBLE

LAYER

OK

THALLIUM

AMALGAM

x61

ELECTRODES

and I’T~ in a 40% thallium amalgam is -8 . ro--~O moles/cmz. Thus, if all the thallium were removed from the first atomrc layer, this would almost precisely account for the observed value of I’TI. Of course, a concentration profile which is unifo?rn up to one atomic layer from the surface, and zero for that single layer is absurd, but if one assumes that the surface deficiency arises primarily in the first four or five atomic layers, one finds that the surface concentration is one-half to three-fourths of the bulk concentration, depending on the concentration profile chosen. Thus it is rather unlikely that the observed similarity in hydrogen overvoltage

Fig.

IO. surfact?

Fj =

O.SO_

excess

of

thallium

with

rfqx%%

to

KIXXCU~:

t.

q =

K.20;

X,

?'j=

Fig. 1~. Surfaca tfscess of thallium compared to the bulk concn. of thaIhum_ The refatwe of thesurface CORCILis greater at the lower concns. f , q = I .zo ; x.~=I.00;O,~=090:*,~= OAO.

X.00;

a,

depletion

162

f.

s_ BU-TLER

betr;veen mercury and the thallium amalgams1 is due entirely to a depletion of thallium at the surface. Although the surface concentration of thallium is certainly louver than the bu& concentration, the ratio of surface to bulk concentration is probably greater than 0.5, and is not very different for the different amalgams.

. This work was supported by the U.S. Office of Naval Research, Materials Sciences Division, Contract No. NOnr-3+5(oo). ARPA Order No. 302-6~. The author thanks Mrs. EVELYK A. BARRON-APPS for her competent assistance tvith the experimental work, Dr. A. C. MAKRIXXESfor his penetrating criticism and helpful suggestions, and Mrs. MABY L. M~EHAN for assisting with the experiments and calculations. SUM.XIRY

Measurements of double-layer capacity and zero-charge potential have been made on so, 30, and 40% thallium amalgams. These resuhs have been compared with existing literature data. The integrated capacity curves agree with literature values for experimental electrocapillary curves. The surface concentration of thallium, at potentials where overvoltage measurements have been made, is one-half to threefourths of the bulk concentration. In contrast to results reported in the literature, the double:layer capacity of 40% thallium amalgam shows no dependence on frequency, within experimental error in the range from 0.5-10 kc.

REZFERENCES