Snapshot 1 shows that at low temperature (e.g., ), the system is Type II. This is due to the fact that both -hexane and methylcyclopentane are only partially miscible in aniline. The dotted portions of the raffinate and extract curves are outside the triangular diagram and obviously unphysical (i.e. mole fractions outside the range [0,1]).

If temperature is increased, methylcyclohexane becomes more soluble than -hexane in aniline. At , the critical solution temperature for methycyclohexane in aniline, the system is at the borderline of Type II and Type I (see Snapshot 3). If you further increase the temperature (see Snapshot 4), the system is clearly of Type I, with aniline being more selective for methylcyclopentane than for hexane.

Type I systems have a plait point (indicated by the red dot in Snapshots 3 and 4). Now the raffinate and extract portions of the binodal curve meet inside the triangular diagram at the plait point. Type II systems do not exhibit a plait point. You can superimpose the experimental data obtained from [1] on the same ternary plot for equal to and .

Finally, the pressure has little effect on the LLE diagram. Indeed, liquid-phase activity coefficients, which are predicted using the NRTL model in the present calculation, have little or no pressure dependence.