The oldest data involving this system, determined by Klemenc and Low (1) in 1930, appear to be low and are imprecise because of the method used for saturation of water. There is, however, some evidence that the method allows the separation of the solubilities of the two crystallographic forms of l,4-dichlorobenzene although the experimental values give no distinct temperature for the conversion from the a-form to the b-form. The solubility measured by Gross and Saylor (2) at 308.2 K is probably too low because of the short tine periods involved in the equilibrations. For the same reason, the value determined by Andrews and Keefer (3) must be rejected as doubtful. Booth and Everson (4) have applied a residue-volume method of Vaughn and Nutting (5) and found the solubility of 1,4-dichlorobenzene in water to be less than 0.5 g(l)/dm³(2). Their method is rapid but insensitive giving only rough values for substances having low solubilities.

More reliable data have been produced by Wauchope and Getzen (6) who reported errors to be in the range of 2 percent. These data are supported by the solubility values obtained by Vesala (7) as part of a study of the transfer free energies of certain nonelectrolytes from H₂O to D₂O. The recent data of Aquan-Yuen et al. (8) of the solubilities of 1,4-dichlorobenzene in aqueous electrolyte solutions produce a value for the solubility in water that agrees well with the above values. Banerjee et al. (9) report a solubility that is somewhat different from the former ones. The reason for the differences may be attributed to the radiochemical method of analysis used. For instance, the radiochemical purity of the substrate and the quenching of the samples in the scintillation analysis remain open to question. Their value can thus be regarded only as a slight support to the other data - even in cases of a full agreement.

The recommended solubility values are calculated on the basis of the data given by Wauchope and Getzen (6). However, instead of their smoothed equation, the calculations are done using an equation of simpler form. This, in turn, suggests that the magnitude of the reported errors is too optimistic. (In the work of Wauchope and Getzen, the meaning of the term "av.% dev.obsd. smoothed solubility" is somewhat unclear.) The values calculated from the simpler equation and the experimental data allow a fairly good estimate of precision. A standard deviation for a single value established in this fashion is of the order of 4.0 mg(l)/kg(2) or 0.027 mmol(l)/kg(2). The existence of the two crystallographic forms of the solute in the range of temperatures where the solubilities were measured is probably one reason for the deviations. However, this brings about no greater effect in the lower temperature range which suggests that the value for 298.2 K, for instance, is quite reliable. The simplified equation for concentration, g(l)/kg(2), in terms of Absolute temperature, T, is as follows:

The observed values from the seven relevant data are shown in Figure 1 together with the calculated behavior (shown as a solid line) from the simpler equation discussed above. The densitites of pure water and the saturated solutions were assumed to be equal for the determination of the reported molarity values rom the calculated g(1)/kg(2) values.

The recommended g(1)/kg solubility values for solid 1,4-dichlorobenzene inwater calculated from values obtained from equation [1] together with corresponding molarity and mole fraction values are listed in Table 1.

Experimental Data:   (Notes on the Nomenclature)