The Constitution of Cupric Chloride in Aqueous Solution.
[Received by the Editor, March 4, 1947; issued separately, April, 1948.]
Vapour pressure and spectrophotometric measurements have been made on solutions of cupric chloride. It is found that this salt is incompletely dissociated, forming the intermediate ion, CuCl+, to an appreciable extent. An approximate estimate suggests that at a concentration of 1 m, about 25% of the copper is present as CuCl+, a percentage which becomes about twice as great at a concentration of 5m.
The use of spectrophotometric measurements in elucidating the constitution of cobalt chloride in concentrated aqueous solution has been described recently by Robinson and Brown (1947), who concluded that the blue colour developed on the addition of concentrated solutions of other chlorides was due to the formation of the undissociated molecule of cobalt chloride. This transition from rose to blue solutions calls to mind the less spectacular change from blue to yellow exhibited by cupric chloride solutions on the addition of concentrated solutions of other chlorides and affords a problem which can also be attacked by spectrophotometric measurements.
Determination of Thermodynamic Quantities.
Isopiestic vapour pressure measurements have been made on cupric chloride solutions up to 2·8 m. by Robinson and Stokes (1940) and on cupric nitrate solutions up to 2 4 m. by Robinson, Wilson and Ayling (1942). These measurements have now been extended up to the saturated solutions, using calcium chloride or sulphuric acid as reference electrolytes. Table I gives the molalities of pairs of solutions
which were found to have equal vapour pressures. From these experimental results, the osmotic and activity coefficients have been evaluated and are recorded in Table II. A plot of the activity coefficients against [ unclear: ] m (Fig. 1) now merits consideration. On this graph the data for magnesium chloride have been plotted to illustrate the behaviour of a fully dissociated salt with a heavily hydrated cation. The data for cobalt chloride and cobalt nitrate have also been inserted; both of these salts belong to this class of salt, although at the highest concentrations cobalt chloride shows a slight departure from the expected curve, a behaviour which is associated with the development of blue colour and which has been ascribed to the formation of small amounts of the undissociated molecule. The lowest curve is that of zinc chloride; it is clearly most anomalous and its behaviour has been interpreted by Stokes (1947) to indicate the almost complete formation in moderately concentrated solution of the complex salt, Zn (ZnCl4). Of the two salts now studied, cupric nitrate falls into the category of hydrated fully dissociated salts, but cupric chloride exhibits anomalies similar to those of cobalt chloride and zinc chloride. In particular, attention may be drawn to the position of the cupric chloride curve below that of the nitrate.
|Cupric Chloride.||Cupric Nitrate.|
In the earlier paper it was shown that the formation of cobalt chloride molecules proceeded to so slight an extent that the vapour pressures of solutions containing equimolecular amounts of cobalt and calcium chloride could be calculated with fair accuracy from the vapour pressures of solutions of the constituent salts. The same was true for solutions containing two mols of lithium chloride to one mol of cobalt chloride. That this is not true for cupric chloride has already been observed by Robinson and Stokes (1946), who found that a solution of 1 m. K2CuCl4 had a vapour pressure lowering 8·8% less than the calculated. This significant departure from the calculated value is substantiated by further measurements, recorded in Table III, which show marked anomalies up to very high concentrations. This abnormal behaviour may be due to formation of the intermediate ion, CuCl+, the undissociated molecule, CuCl2——, or complex ions such as CuCl3 or CuCl4——. To investigate these possibilities, some light absorption experiments have been made.
Fig. 1.—The Activity Coefficients of Cupric Chloride and Nitrate, compared with those of other Salts of the same Valency Type.
Investigation of the absorption of light by cupric chloride solutions, using the Coleman, No. 11, spectrophotometer, showed that the blue solutions gave a strong absorption in the “far red.” This absorption
is typical of blue copper compounds and is frequently used in light filters to remove undesirable infra-red rays. On the addition of chloride ions, however, another absorption band appeared in the “violet” with a maximum about 375 mμ. This figure is only approximate, since it falls outside the accurate range of the instrument. Nevertheless, it was felt that by the additional use of the Coleman PC 6 filter with a maximum transmittance at 355–400 mμ, significant readings could be made. Furthermore, it was shown that Beer's Law relating intensity of light absorption to concentration was satisfactorily true at this wave length as is shown by the following values with a solution of 16·9 m lithium chloride and varying concentrations (mols per litre) of cupric chloride. D, the optical density, is defined by:—
D = log I/Io = k c,
where k is a constant.
In all further work, therefore, measurements were made at 375 mμ, a wave length taken to be characteristic of the yellow compound. In order to take advantage of the most sensitive density range, it was found that a copper concentration of 0.0005 mols per litre was optimal and all further measurements were made at this concentration.
It was realised early in the work that blank values were extremely sensitive to contamination, probably by iron, which produces a very intense yellow colour at high chloride concentration, or by organic colouring matter from cork, etc. Rigid precautions against contamination had to be taken and glass-stoppered flasks and weighing bottles and chemically pure reagents were used throughout the work. Blanks without added cupric chloride were prepared with the same care as were the test solutions.
Temperature was found to have negligible effect on the formation of the yellow compound. Measurements were made at a temperature in the vicinity of 23°.
Following the method developed for cobalt chloride, a series of measurements was made at a cupric concentration of 0 0005IN, with increasing quantities of lithium chloride. In the absence of lithium chloride, the optical density was zero and this increased to a flat maximum of 1–04 at 23m lithium chloride, a concentration which was realised by raising the temperature slightly. The fraction of copper present in the form of the yellow compound was then calculated by the equation:
α = D/1·04,
D being the optical density at intermediate chloride concentrations.
To determine the effect of chloride concentration on the formation of the yellow compound, the hydrochloric acid-perchloric acid system, of approximately constant water activity, was used as described by Robinson and Brown (1947). Table IV gives the optical density, D, and the fractional degree, α, of formation of the yellow compound in eight mixtures of hydrochloric acid and perchloric acid. The water activities, a [ unclear: ] , were calculated from the data of
Åkerlöf and Teare (1937) for hydrochloric acid and of Robinson and Baker (1947) for perchloric acid.
|mHCl||mHClO4||D||α||a [ unclear: ]||—logK1||—logK2||—logK3||—logK4|
|α = K1 (1-α) [Cl]|
|α = K2 (1-α) [Cl]2|
|α = K3 (1-α) [Cl]3|
|α = K1 (1-α) [Cl]1|
Four possible equilibria are represented by the following equations:—
Cu++ + Cl− ⇌ CuCl+
Cu++ + 2Cl− ⇌ CuCl2
Cu++ + 3Cl− ⇌ CuCl3−
Cu++ + 4Cl− ⇌ CuCl4− -
corresponding to each of which there are the equilibrium constants, K1, K2, K3 and K4. On substituting values of the chloride concentrations and the measured values of α, it was found that only K1 was to any degree constant. It follows, therefore, that the formation of the ion, CuCl +, must be the predominant factor in the production of the yellow colour. With increasing chloride concentration, K1 increases from 0·13 to 0·25 whilst K2 decreases from 0·17 to 0·024. Thus the values of K1 are not so satisfactorily constant as were the values obtained for cobalt chloride. This change in the value of K1 over a range of chloride concentration may well mean that, whilst formation of the CuCl+ ion is the most important effect, it is accompanied by some formation of the undissociated molecule, CuCl2.
K1 corresponds to approximately 25% formation of the intermediate ion, CuCl +, at a concentration of 1 m cupric chloride, free from other salts, a figure which increases to approximately 50% at 5 m. This should be contrasted with the formation of the undissociated molecule, CoCl2 in cobalt chloride solutions, which occurs to the extent of only 4% at 1 m and 15% at 5m.
The author wishes to thank Dr. R. A. Robinson for his interest in this work.
Åkerlof, G., and Teare, J. W., 1937. J. Amer. Chem. Soc., 59, 1855.
Robinson, R. A., and Baker, O. J., 1946. Trans. Roy. Soc. New Zealand, 76, 250.
Robinson, R. A., and Brown, J. B., 1947. Ibid., 77, 1.
Robinson, R. A, and Stokes, R. H., 1940. Trans. Faraday Soc., 36, 1137.
Robinson, R. A., and Stokes, R. H., 1946. Ibid., 41, 752.
Robinson, R. A., Wilson. J. M., and Ayling, H. S., 1942. J. Amer. Chem. Soc., 64, 1469.
Stokes, R. H., 1947. Trans. Faraday Soc., 43 (in press).