STUDIES I N ELECTROCAPILLARITY Part IT’. The Effect of Salts on the Electrocapillary Curves of Solutions containing Surface-active Substances BY J. A . V. BUTLER A S D A . WIGHTMAN
According to a theory of electrocapillary action which has been put forward by one of us,’ the lowering of the surface tension at the interface between mercury and the solution produced by the presence of many organic substances is related to the potential difference by an equation of the form Ay
=
Ayme-a(V-Vrn)?
where Vm is the potential difference at which the surface tension lowering (for the g’ven solution) has its maximum value A-ym, and A-y is the lowering at the potential difference V. The curve of A-y against V (which we shall call the adsorption curve) as determined by this equation is completely symmetrical about the potential difference V. The study of Gouy’s data? for numerous organic substances in normal sodium sulphate solution showed that while this equation was, in most cases, in excellent agreement with the right hand (negative) branches of the adsorption curves, these were often unsymmetrical and lower on the left (positive) side. In Parts I1 and I11 of this series of paper^,^ the elect’rocapillary behaviour of solutions containing two surface-active substances has been examined, and it has been shown that the adsorption of one substance is in general reduced by the presence of a second surface-active substance, in some cases to the extent of the complete exclusion of the first substance. Since the sulphate ion, like most inorganic anions, itself causes a lowering of the surface tension on the positive side of the electrocapillary curve, it appeared probable that the asymmetry of the adsorption curves in normal sodium sulphate might be due to the partial displacement of the adsorbed substances at the more positive potentials by adsorbed sulphate ions. We have therefore examined the effect of varying the nature and concentration of the salt on the adsorption curves of some typical surface-active substances. I t is well known that the electrocapillary curves of most inorganic salt solutions are practically parallel a t high negative potential differences, and may be made t o coincide in the region from - 1.2 to - 1.8 volts (referred to the normal calomel electrode) by small displacements parallel to the axis of the potential difference. These displacements have been ascribed to the liquid junction potential difference between the standard electrode and the solution under examination. F h e n taking the surface tension lowering caused by a substance as the difference between the surface tension of the salt J. A. V. Butler: R o c . Roy. SOC., 122A, 399 (1929). * A n n . Chim. Phys., 8,291;9, 7 j (1906). J. Phys. Chem., 34, 2297, 2841 (1930).
3294
J. A. V. BUTLER AND A. WIGHTMAN
solution and that of the solution containing the same concentration of the salt, together with the added substance, it is important to take into account any displacement due to this cause. KO difficulty arises with non-ionised solutes a t a moderate concentration, as the curves of the mixtures usually coincide with those of the primitive salt solutions at the extreme right (for solutions containing 4 mols per cent of alcohol a slight displacement is necessary to make them coincide at - 1.8 volts). But when the added substance is also a salt, as in the case of sodium salicylate, and has a concentration
FIG.I Surface tension lowerings of ethyl alcohol (4M) in lithium chloride solutions. Electrocapillary curves of lithium chloride in water in the upper part of the diagram.
which is comparable with, or greater than that of the inorganic salt, it is necessary to displace the curve of the mixture until it coincides on the extreme right with the curve of the primitive solution. In order to facilitate the comparison of the behaviour of salt solutions of different concentrations, the electrocapillary curves of the salt solutions have been displaced so as to coincide with that of normal potassium chloride at - 1.5 volts. In the diagrams given in this paper, the adsorption curves have also been given the same displacement. The displacements required are listed with the experimental data below. (I) Ethyl alcohol. I n Fig. I the upper part of the diagram shows the electrocapillary curves of solutions of lithium chloride 'n water, and the lower part , the surface tension lowerings produced by the presence of ethyl
STUDIES I N ELECTROCAPILLARITT
3295
alcohol to the extent of 4 mols per cent. I t can be seen that the surface tension of the primitive solution is raised by increasing the salt concentration at potential differences greater than -0.7 volts, while below -0.4 volts it is positively adsorbed to an increasing extent as the potential of the mercury becomes more positive. The behaviour of the alcohol is closely related to that of the salt. In the region in which the salt is negatively adsorbed the surface tension lowering of the alcohol increases with the salt concentration but in the region in which the salt is positively adsorbed the surface tension lowering of the alcohol falls off rapidly as the salt concentration is increased. The adsorption curves of alcohol, which are markedly asymmetrical at the higher concentrations of the salt, thus become broader as that is decreased and are very nearly symmetrical at the smallest salt concentration.
FIG.2 Surface tension lowerings of ethyl alcohol (4M) in potassium bromide and iodide solutions.
The behaviour of alcohol in potassium bromide and iodide solutions is very similar. (See Fig. 2 . The electrocapillary curves of some of the salt solutions are shown in Fig. 3 ) . The bromide and iodide ions are much more surface-active than the chloride ion and when they are present the electrocapillary curve begins to fall at much higher values of the potential difference. As might be expected in these solutions the displacement of the alcohol by the surface active anion begins to take effect further to the right of the curves. The effect of ?; potassium bromide on the adsorption of alcohol is about the same as that of S / I O potassium iodide, and that of N/zo potassium bromide about the same as that of N/z lithium chloride. Thus it appears that solutions containing chloride, bromide and iodide ions, having concentrations in the ratios I O O : I O : I , have approximately the same effect on the behaviour of alcohol. The electrocapillary curves of solutions containing chloride, bromide and iodide ions, in these ratios, also approximate to each other. We may therefore conclude that the effect of these ions on the behaviour of alcohol is proportional to their effect on the electrocapillary curve in the primitive solutions, and that their surface activities have relativemagnitudes approximately in the ratios I : I O : I O O .
3296
J. A. V. BUTLER AND A. WIGHTMAN
(2). Phenol. The behaviour of phenol is very similar. Fig. 3 shows the surface tension lowerings of N / I O phenol in a series of solutions containing the halogen ions, and in the upper part of the diagram the corresponding electrocapillary curves of the salt solutions. The electrocapillary curves of potassium chloride solutions are nearly identical with those of lithium chloride solutions at the same concentrations, and the behaviour of phenol in the
000
300
TOO
FIG.3 Surface tension lowerings of phenol (SII O ) in potassium halide solutions. Electrocapillary curves of the salt solutions in wiiter in the upper part of the l a g r a m .
former is very similar to that of alcohol in the latter. The adsorption curve of phenol becomes more symmetrical as the concentration of potassium chloride is decreased. The effect of the bromide and iodide ions on the behaviour of phenol, is very similar in kind and ext,ent to their effect on alcohol. Fig. 4 shows the behaviour of phenol in sodium sulphate and ammonium nitrate solutions. The latter salt is of interest because measurements can be extended with its solutions further in the positive direction than wit'h any of the other salts employed. The electrocapillary curve is lowered by an increase of the concentration of this salt, at all potentials more positive than
STUDIES I N ELECTROCAPILLARITT
3297
-0.7, and the surface tension lowering of the phenol is reduced by an increase of the salt concentration over the same range. Sodium sulphate is notable in that an increase of its concentration from S/IO to N / I raises the adsorption curve of phenol through practically the whole of the region in which measurements can be obtained. (3) Sodzum salicylate. In Part I of this series of papers’ it was pointed out that strong electrolytes having surface-active positive ions (e.g. substituted ammonium salts) show a maximum adsorption when the potential
FIG.4 Surface tension lowerings of phenol (N/Io) in sodium sulphate and ammonium nitrate solutions. I . Sodium sulphate (N); (2) sodiumsulphate (N/1oj;(3)ammoniumnitrate (N); (4) ammonium nitrate (N/Io).
difference has a high negative value (-1.3 to -1.j volts), whereas the maximum for non-ionised substances falls within the range -0.4 to -0.7 volts. I t was to be expected on theoretical grounds that the maximum of the adsorption curves of strong electrolytes having surface-active anions would occur at, a similar interval towards positive values. The adsorption curves of a number of salts of organic acids were measured in X/I sodium sulphate and it was found that their maxima occurred between -0.13 and -0.30 volts, which are not nearly so positive as was expected. It seemed probable that this anomaly was due to the reduction of the adsorption of the organic anions at the extreme left of the curves owing to the adsorption of sulphate ions, which cause a considerable lowering of the surface tension in this region. J. Phys. Chem.,
34, 2286 (1930).
3298
J. A . V. BCTLER AND
A.
WIGHTMAS
We have therefore determined the surface tension lowerings of S / I O sodium salicylate in solutions of sodium sulphate, ammonium nitrate, potassium chloride and potassium bromide, of various concentrations. The curves obtained are shown in Figs. 5 and 6. I t is evident that as the concentration of the inorganic salt is increased, the maximum of the adsorption curve of the salicylate moves towards more positive values of the potential difference, and a t the smallest salt concentrations is beyond the most positive values for which measurements can be obtained. Thus, when the salt concentration is sufficiently reduced we realise the behaviour which was expected on theoretical grounds.
FIQ.5 Surface tension lowerings produced by sodium salicylate (N / IO) in potassium chloride
and potassium bromide solutions.
Experimental Data The experimental arrangements were similar to those of the previous papers of this series. Since it was possible that small differences of surface tension would be of importance, the vessel containing the solution at the tip of the capillary was water-jacketted and kept at a constant temperature (zoo). The salts used were the best available A.R. quality. The ethyl alcohol was dried over quicklime, and distilled. The phenol was also purified by distillation, and had the mp. 40.8'. The following tables give the surface tension lowerings produced by the presence of ethyl alcohol phenol and sodium salicylate in the inorganic salt
STUDIES I N ELECTROCAPILLARITY
3299
solutions. In determining these values, the electrocapillary curve of the mixture has been displaced (if required) parallel to the axis of the potential differences, so as to make it coincide with the curve of the primitive salt solution a t - 1.8 volts. The displacement required is given in the column D1. DPgives the d'splacements required to make the electrocapillary curves of the salt solutions coincide with the curve of normal potassium chloride at (and above) - 1.5 volts.
FIG.6 Surface tension lowerings produced by sodium salic late (N/Io) in ammonium nitrate and sodium sulphate g o h o n s .
The potential differences are given with reference to the normal calomel electrode. The surface tension lowerings are given in terms of Gouy's scale, Le. the maximum surface tension between mercury and normal sodium sulphate at zoo is taken as 1001.7, corresponding to a maximum value of 1000 for the surface tension between mercury and water, according to Gouy's measurements.' 1 The ratio of the maximum heights of the electrocapillary curve in normal sodium sulphate,and water ( I O O I . ~ : was ~ ~ )determined a t 18". Our measurements are a t zoo, but this difference of temperature cannot appreciably affect the ratio.
3. A . 1'. BUTLER AND A . WIGHTMAS
3300
The data from which the electrocapillary curves of the solutions of salts in water have been constructed are not given here, for they would occupy a considerable space, and the general features of the curves can be seen in the diagrams. We propose, however, to publish the data for these curves in another paper, and to give a theoretical discussion of them.
TABLE I Surface Tension Lowerings of Ethyl Alcohol (4 mols per cent) in various Salt Solutions Potential difference
zM
Iithium Chloride' rM o.gM o.ogh1
-0.0
0.5
0.0
-0.2
2.2
j.1
Potassium Bromide o.rN o.ojN
IS
Potassium Iodide 1s
0,lX
-
_
3.6
0.3 6.9
17.3
8.5
6.3
-0.4
18.8
21.7
25.8
31.0
10.0
28.1
(16.6)
- 0.6
41.0
40.3
39.0
34.7
25.4
38.8
12.5
13.7 28.3
-0.8
41.5 28.6
37.4 26. I
36.3 24.7
34.7
32.9 23.6
38.1
18.8
3c.2
2j.0
28.1
21.5
2,7.$
12.j
13.2
13.1
16.1
4.9
16.4
12.2
14.7
6.0
5.3
j.8
6.8
-
7 ti
4.7
.5.?
0.2
0. j
1.0
2.4
', 2
0,j
1 2
DI
so.01 so.01 so.01
+O.OI
~ 0 . 0.j1
Dz
-0.01
+0.07
tO.005
-1.0
-1.2 - 1.4 - 1.6
0.0
+O.OI
_.
0.0
+e.oj:
* In the lithium chloride solutions, which are t a h r n iron> another series of meaiurements, tht
concentrations are expressed as gram molecules lier they refer to gram-equivalents per litre.
IOOO
grams of solvent; in the other solution
T A ~ ~I1L E Surface Tension I,on-eriii,Ts of I'henol ( S ' I O ) in various Salt Solutions Potential difference
Potassium Chloride _ _ _ _
~
13
O.IA
o.o,5S
.
.
Fotnssiiiin Bromidp IN 0.1x 0.0jS
Potassium ~ ~ _Iodide . _ _ 1 3 !),I% CJ.O~S
.
_
STUDIES IN ELECTROCAPILLARITT
TAI~LP; I1 (Continued) A m mon i urn N i t,rnt e 0.IN o.oj?i
Sod i urn SUIph n te
Potential diff erencc
IX
IN
0.05N
0.13
-
+OS4
$0.2
9.4 27.4
I
0.0
-0.2 - o,.$
64. I
--
42.2
I /*I
53. I
7 9.4
59.7 59.5
- 0.6
72.2
- 0.8 - 1.0
.;6.4 29.1
48. I
- 1.2 - 1.4
7 .4
2 .O
20.7
1.3
0.5
111 112
-
-
+o.o.jj
0.0
TABLE 111 Surfnce Tension Lowerings of Sodium S:dicyl:ite (N/Io) in vzrious Salt Solutions Potassium Chloride O.IN 0.0jh’ 0.02N
o ten tiul .fference
IN
-0.1
-
-
0.0
-
76.8 57.0
-0.2 - 0.4 - 0.6 - 0.8 - 1.0 - 1.2
49.1
58.9
60.j
jI.0
42.0
30.4 11.4 3.0
23.7 3.7
1)1 I:, 2
-
’otentinl .iff erence
Dz
16.5
19.2 32.j 37.3 2j.4
40.2 46.6 35.3 23.2 8.4 3.0
j2.3
2.9
9.4
0.0
1.3
(49.6) 5j.j 44.4
4.0
7.2
1*9
2 .o
0.0
30.1 15.6
-0.07
-0.01
-0.01
+O.II
fo.00;
+ o . o ~ fo.065
Ammonium Sitrnte O.IN o.ojN 0.02N
81.2 79.7 79.4
SG. j
74. I
72.0
95.8 77.7 j 5.6
106.6 100.6
84.3 79.2
62.3
j j.8
42.6
42 - 4
40.2
35.7
28.3
20.7
30.8
10.4
4-2
7.7
17.8 2.9
3.0 -
0.0
0.0
0.0
-0.01
j -0.01
0.0
+0.03
j
+O.O
- 0.05
jj
+
0.0s
IN
81.2 88.6 79.4 55.7 30.3 12.3 3.9 I
+o.or
98.0
42.5 30.3 13.3
+0.06
IN
o.otS
76.2 45.6 28.9 13.6 4.0
- 0 . 0 2 ;.
50.8
D1
-
+o.oj
51.8 52.9 56.3 5 2.3
- 1.2
-
_ I
78.7 53.6 32.3
i 3.0
0.2
to.1
-0.2 - 0.4 - 0.6 - 0.s - 1.0
94.6
--
3j.7 18.7 3.9 3.0 - 0.030
t0 . 2 0 .o
IO0
L
62.9 71.5
Potnsiiim I3romide O . I N 0.05 N 0.02X
IN
-0.01
-0.03
-0.10
+O.IO
+O.I7
Sodium Sulphate O.IN o.ojN O.WN
90. I 9j.3 84.6 64.2 38.9 18.3 5.6 0.8 -0.02
+o.oSj
’
99.9
107.1
99.9 78.8
104.0
51.8 28.8 11.9 3.2 0.0
-0.04 to.08
81.2 52.1
30.1 13.6 3.4 0.0 -0.05 j
+0.13j
3302
J. A. V. BUTLER AND A. WIGHTMAS
Summary The effect of varying the nature and the concentration of inorganic salts on the electrocapillary curves of solutions containing ethyl alcohol, phenol and sodium salicylate has been investigated. 2. In general the adsorption curves of ethyl alcohol and phenol become more symmetrical as the salt concentration is lowered. The surface-active bromide and iodide ions greatly reduce the adsorption of these solutes in the region in which the ions themselves are adsorbed. In solutions containing chloride, nitrate and sulphate ions, the surface tension lowering caused by the organic substances is increased by increasing the salt concentration in the region in which the salt is negatively adsorbed, and decreased in the region in which the salt is positively adsorbed. 3. The maximum of the adsorption curve of sodium salicylate moves towards positive potentisls as the concentration of the inorganic salt is decreased and its magnitude increases. I t has been shown that the maxima for strong electrolytes with surface-active cations occur between - 1.3 and -1.j volts, and for non-ionised substances in the region of -0.j volts. I t is now found that at very small concentrations of the inorganic salt, the maximum for a salt with a surface-active anion occurs at a potential more positive than +o.z. The behavior R-hich was expected on theoretical grounds has thus been established. We are indebted to the Department of Scientific and Industrial Research for a Studentship, granted to one of us (A. W.); which enabled him to take part in this work, and to Imperial Chemical Industries, Ltd., for a grant which defrayed the cost of part of the apparatus. I.
Chemistry Department, University of Edinburgh, June 18, 1931.
