We have adapted the anion chain length model of Cleaver and Sime to predict the distribution of anions in a polysulfide melt, Na2Sx (1 -< X -6), as a function of composition the mole fraction of total sulfur. This was then used to calculate the EMF of a Na/S celt from Xs = 0.75 to 0.56, where Xs is the mole fraction of sulfur. This range corresponds to the entire two-phase region. The predicted anion distribution is similar to that of Cleaver and Sime except for the prediction of the appearance of Na2S to high states of discharge; the calculated EMF provides a good fit to the literature values. Figure 1 illustrates how the EMF of the Na/S cell undergoes a transition from 2.076V when the sulfur electrode is in the two-phase, sulfur plus sodium polysulfide state, to 1.74V in the two-phase sodium polysulfide plus Na2S2(s) state. The former two-phase zone terminates when the mole fraction of total sulfur Xs = 0.729, the latter occurs at Xs = 0.571. Between these two zones, when 0.571 < Xs < 0.729, only one phase is present, a sodium polysulfide liquid, and the EMF decreases almost linearly with decreasing Xs.If at Xs = 0.714, a ratio corresponding to the stoichiometric formula Na2Ss, the sulfur electrode half-reaction is 4Na2S5 + 2Na § + 2e ~ 5Na2S4 then, as the cell discharges, and Xs decreases, the EMF would exhibit the behavior of a typical EMF titration curve. In fact, as the composition of the sulfur electrode passes through values of Xs corresponding to the stoichiometric compositions of Na2Ss, Na2S4, Na2S3, the EMF would be expected to pass through successive titration curves.The fact that the EMF follows a smooth transition from one two-phase region to the other suggests that the polysulfide melt may be comprised of several polysulfide species in equilibrium and that each of these make a contribution to the EMF. Thus, as the cell discharges one species may be in the majority and make the major contribution to the EMF, but this prodominance diminishes on further discharge, and another species of lower chain length becomes the majority species. If a sufficiently large number of species are present, then the EMF would be expected to behave in the experimentally observed manner.Cleaver and Sime (1) have developed a model that predicts the distribution of anion chain lengths in the polysulfide melt, and they indicate how it can be used to calculate the EMF of the Na/S cell as a function of the composition of the sulfur electrode. They assumed that the melt contains a distribution of polysulfides from Na~S2 upwards. An upper limit is not imposed on the chain length. In order to reduce the number of parameters required to fit the model to the data for sulfur activity, they assumed that the equilibrium constant for the reaction na2Sj-, + S --* na2Sj kj -XNazs(~-Das had some functional dependence on j, where S is dissolved sulfur, and as its activity in the polysulfide melt. The relationship between kj and j, that provided the best fit between the predictions of the model, and the experimental activity data i...