A survey of reported compositions along the solubility curve and compositions of coexisting phases in equilibrium for the system nitromethane + methanol + water is given in Table 1.
Solubility curve
The system nitromethane + methanol + water forms a miscibility gap of Type I, Treybal.4 Only one binary pair of components, nitromethane + water, is partially miscible. The data for this system were previously compiled and critically evaluated.5 The recommended value for the solubility of nitromethane in water at 293.2 K is x1" = 0.0345; and at 313.2 K, x1" = 0.0426. The solubility of nitromethane in water reported in Zhuravlev and Maslovskaya1 (0.090 mass fraction; 0.028 mole fraction) and at 313.2 K in Maslovskaya2 (0.100 mass fraction; 0.032 mole fraction) are rejected because they are inconsistent with recommended value from Sazonov et al.5
Solubility data obtained by the titration method for the ternary system nitromethane + methanol + water were reported in three references. Data reported at 293.2 K by Zhuravlev and Maslovskaya,1 and by Tolstova and Vasil'eva,3 are in agreement. Data of Maslovskaya,2 which were reported at 313.2 K, are less than those in Zhuravlev and Maslovskaya1 and Tolstova and Vasil'eva.3 Thus, with increasing temperature the mutual solubility of the components increases, Fig. 2 . The data for the nitromethane + methanol + water system are consistent with the data for the nitromethane + ethanol + water system measured and reported by Maslovskaya.2 Experimental errors were not reported, but are estimated by the evaluator to be ± 0.002 mass fractions. Solubility data1,3 at 293.2 K (water-rich and nitromethane-rich branches together) are described by an algebraic equation in the form:
x2 = a + b·x3 + c· ln(x3) + d· (ln(x3))2 + e· (x3)1/2.
The proposed equation gives a smaller standard deviation than any other tested equations. The equation parameters (a, b, c, d, e) were calculated by the least squares method and presented in Table 2. Binodal curves in this system were calculated and compared separately for solubility data reported in Zhuravlev and Maslovskaya1 and Tolstova and Vasil'eva3 as well as for the combined data.
Liquid-liquid equilibrium (LLE) data
LLE data for the ternary system nitromethane + methanol + water were reported in Maslovskaya,2 and Tolstova and Vasil'eva.3 Composition of each phase, Maslovskaya,2 was calculated from the index of refraction-composition curve of saturated mixtures prepared at the time of the saturation curve measurements. Compositions of phases in equilibrium reported in Tolstova and Vasil'eva3 were measured by gas chromatography but detailed information about experimental procedures is not reported. The tie lines cover the whole range of the miscibility gap. Data reported at 293.2 K by Maslovskaya2 and Tolstova and Vasil'eva3 are in agreement. The direction of the tie lines changes with concentration, but is similar in all reported papers. Plait points were reported in both cases: at 293.2 K is the values are x1 = 0.297, x2 = 0.286, Zhuravlev and Maslovskaya,1 and x1 = 0.361, x2 = 0.272, Tolstova and Vasil'eva.3
Experimental errors were not reported, but are estimated by the evaluator to be ± 0.005 mass fractions. The experimental liquid-liquid equilibrium data at 293.2 K, Maslovskaya,2 and Tolstova and Vasil'eva,3 are used for construction of the smoothed curve for distribution of methanol between water-rich and nitromethane-rich phases, Fig. 3 . These data are described by an algebraic equation
x2" = -4.3771(x2')2 + 2.2409x2' - 0.0042.
Calculated and experimental distribution data, Refs 2 and 3, are presented in Fig.2 for comparison.
Recommended or tentative data
Solubility and liquid-liquid equilibrium data measured by different methods are in agreement with one another. Experimental data on solubility1-3 at 293 K and 313 K are presented in Fig. 2. The ternary solubility data1-3 were used to obtain equation parameters of the binodal curve.
In view of the inaccurate results, obtained in Zhuravlev et al.,1 Maslovskaya,2 and Tolstova and Vasil'eva,3 for the binary nitromethane + water system, the evaluator considers the ternary system data to be tentative and concludes that new studies on the ternary system would be of value.