Linear relationships exist between the logarithms of equilibrium quotients, [SH+]/[S][H+], of diverse bases as they vary with acid concentration in moderately concentrated mineral acids. For purposes of formulating these linear free energy relationships in standard form, the equilibrium quotient for protonation of a hypothetical aromatic primary amine of pK, zero has been chosen as horizontal coordinate; this is given by (-Ho -log [H+]). Log ([SH+]/[S])+ HO is plotted against (Ho + log [H+]). 'The slope, 4, of the h e a r plot is a parameter which characterizes the response of the equilibrium quotient to changing acid concentration. The intercept represents the thermodynamic pK, of the base. This constitutes a general method for estimating the pK, of any base which undergoes protonation in moderately concentrated mineral acid, with reference to the single acidity function, Ho. For bases of diverse type, pK's estimated by this method are in good agreement with those estimated by the acidity function method. Use of the new correlation procedure a s a check on the validity of several acidity functions and its application to equilibria not involving proton gain or loss are also discussed.

ABSTR.-\CT Linear relationships exist between log k~ + I I o (for reactions of weakly basic substrates) or log k+ (for reactions of strongly basic substrates) and (Ho + log [Hf]). These are linear free energy relationships. For weakly basic sttbstrates, the correlations obtained are better than in plots of (log k+ + Ho) versus log an20 or of log k+ versus -Ha. The slopes in plots of log k+ or (log k+ + Ho), as appropriate, against (Ho + log [H+]) are taken a s parameter, 4, which characterizes the response of the reaction rate to changing rninera! acid col~centration. Values of 4 for reactions of strongly basic substrates reflect only relationships between protonated s~~b s t r a t e and transition state, and Inay be related to reaction mechanism. 4 values for reactions of weakly basic substrates reflect both equilibri~l~n protonation of the substrate and transformation of protonated substrate to transition state, and are therefore less directly related to ~nechanisin. However, the 4 values for the two steps are additive and that for the latter step can be obtained by subtraction if the overall 4 value and that for equilibrium protonation are known.The rates and states of equilibrium of reactions in moderately concentrated mineral acid solutions generally vary with the acid concentration. The extent and even the direction of variation differ from reaction to reaction. The description, quantification, and interpretation of these phenomena have occupied the attention of inany ivorlters, but understanding renlains illcomplete.In the accoinpanying paper (I) it is shown that linear free energy relatioilships (1.f.e.r.) exist between the logarithills of equilibrium quotients, [SI-If]/ [S] [Hf], and (or) secondorder rate coefficients for diverse reactions as these quantities vary lvith acid concentration. These relationships as they pertain to equilibria are considered in that paper. They are now considered as they pertain t o rate phenomena.T h a t 1.f.e.r. concerning kinetic data exist is illustrated by Fig. 1 of the accoinpanying paper and by Figs. 1 and 2 of this one. In the present Fig. 1, the logarithm of the secondorder rate coefficient (k+/[I-If], alternatively syinbolized kz) for the enolization of acetophenone (data of Zucker and I-Tainmett (2)) in perchloric acid solutions is plotted against log kz for the depolyinerization of trioxane (interpolated from data of Bell, Bascombe, and McCoubrey (3)) in perchloric acid solutions. Each point represents measurements a t a different acid concentration, points for lower acid concentrations being to~vard the left of the plot. In Fig. 2, log kz for ethyl acetate hydrolysis (interpolated froin data of Bell, Dowding, and Noble (4)) is plotted against log k+ for hydrolysis of N,N1-diphenylformamidine (data of DeFVolfe ( 5 ) ) , both in hydrochloric acid solutions. The latter substrate is strongly basic, essentially fully protonated a t all the acid concentrations studied, and one therefore plots log kJ, rather than log kz.For purposes of formulating 1.f.e.r. concerning e...

A five-step sequence has been established for the room temperature reaction of sulfur with benzylamine. Benzylamine polysulfides and benzylidenimine polysulfides occur as interconverting intermediates in this system. The final products are benzylammonium polysulfides, ammonia, and N-benzylidene benzylamine.

No abstract

The crystal and molecular structure of benzylidenimine trisulfide has been determined. Crystals are orthorhombic with a = 16.610(2), b = 9.984(6), c = 4.386(4), Z = 2 and space group P2,2,2. 773 unique reflections of positive measured intensity, recorded with a Syntex P i Automatic Diffractometer, were refined by full-matrix least-square methods to a conventional R value of 0.023. The molecule has a crystallographic two-fold axis and S-S bond lengths of 2.051(1) A. These are intermediate in length between the long inner and short outer S-S bonds of the chemically analogous tetrasulfide. The S-N and N-C bond lengths are 1.661(3) and 1.269(3), respectively. The crystal packing is determined by S-S van der Waals interaction between adjacent molecules along the c axis.