4722
NOTES
X small drop of the dye mixture (dissolved in the solvent system used) wag placed a t the apex of the strips and after the ascending solvent fronts had coalesced a t the apex, a potential was applied across the strips by means of platinum electrodes immersed in the electrode vessels. Figure I was prepared from a tracing of paper strips and illustrates the separation of an electropositive dye (crystal violet) and an electronegative dye (Oil Red 0). In this separation a mixture of nitromethane (9 vol.) and glacial acetic acid (1 vol). was used. Similar separations have been effected in absolute ethyl alcohol, methanol and in pyridine-glacial acetic acid mixtures.b Included in Fig. 1 is an ascending chromatogram6 in the same nitromethane-glacial acetic acid solvent where i t will be noted no separation was achieved.
1701.
74
CATHODE 1 W
z+
0 0 W
a
d
YY I>o + > W r
0
EOSIN
!ETHYLENE VIOLET
CATHODE O I L RED 0
ALIZARINE BLUE
I NODE
Fig. ?.--Separation of dyc mixture in ethyl alcohol; 1000 volts 60 iriicraoinps, 50 iriin.
was apparent over the short distances traversed.
I t is perhaps significant that separations were achieved only with ions containing strongly polar groups (the dyes). The migration of the biological substances mentioned may be a passive electroendosmotic effect; however it is possible that these less polar materials could be separated in experiments of longer duration. The separation of the dye mixtures on the filter paper establishes the feasibility of electrophoretic separations in non-aqueous solvent systems. I t is possible that other solvent systems and experimental conditions may provide extension of this technique to a wider range 01 substances. ASCEND1NG CHROMATOGRAM Fig. 1.-Separation of oil red 0 (vertical lines) and crystal violet (horizontal lines) in nitromethane-glacial acetic acid mixture.
ANODE
In Fig. 2 the separation of a mixture of four dyes in absolute ethyl alcohol is illustrated. Cholesterol, palmitic acid, 17-hydroxy-1l-dehydrocorticosterone and testosterone were found to migrate toward the anode in the ethyl alcohol or nitromethane-glacial acetic acid systems; however, no separation of these mixtures of compounds (5) The most highly purified commerdally nvalhble solveatr were employed; however, no effort WM made to remove tr.ce.0 of E d B t u r e with which the solvent8 were in equilibrium, since the cell d d s R did w t completely exclude atmcupherfe moisture. ( 8 ) R~J. William. r a d R ,Kirby, Scirlrr, 147, 181 ( l 0 W .
ARMYMEDICAL SERVICE GRADUATE SCHOOL \\;ALTER REEDARMYMEDICAL CENTER ~I'ASHINGTON 12, D. C.
The Dissociation Energies of the First and Second Bond in H2S and Some Comments on a Recent Paper by Franklin and Lumpkin B Y A. H. SEHOS' RECEIVED APRIL21, 1952
Franklin and Lumpkin2 have recently derived a value of 38.4( * 5 ) kcal./mole for AHf(SH) from relevant thermochemical data and their measurements of appearance potentials of carbdnium ions (1) Post-doctoral fellow, 1851-1952. (2) J. I,. Frpnklin and €T, B. Lumpkin, Tnrm Jovrr&r-,7 4 # loa8
(lQW.
NOTES
Sept. 20, 1952
4723
from various mercaptans. Their value agrees strengths in HzO. I t seems, therefore, that their fairly well with the value of 32( *4) kcal /mole arbitrary choice of the lowest value for D ( S S ) for AHr(SH) determined in these l a b ~ r a t o r i e s , ~is the only reason for their disagreeing with Porter’s especially if, as pointed out by them, the heats value for D2. Surprisingly, however, in spite of of formation of radicals obtained by the electron their method of calculation they suggest that on general chemical grounds the highest value for impact method tend to be rather high. Since D ( S - 3 ) appears to be the correct one. D(II-SII) DI = Aflr(I-I) 5 AfIfr(SH) - Allfr(IIS) ( 1 ) Thus if D ( S S ) is, indeed, 102.5 kcal./mole then and the values for 1IZf(H) = 53 kcal./gram atom the sum of the two dissociation energies in H2S and for AHf(H2S) = -4.S kcal./mole are well is 175 kcal./mole, and since a value of -90 kcal./ e ~ t a b l i s h e d ~it” , ~is now possible to calculate in mole for D1 is supported by both electron impact terms of the above values for AHf(SH)the dissocia- and pyrolytic methods, Dz will be -83 kcal./mole, tion energy, D(H-SH), of the first bond in HzS which is in reasonable agreement with Porter’s at 95( *5) or 89( *4) kcal./mole, and Ramsay’s values obtained spectroscopically. -4 (denoted by 01) respectively. final settlement of this point can only be reached Furthermore, the dissociation energy D (S-H) by an unambiguous determination of D(S-S) or of the second bond in H2S (denoted by Dz) can be AHf(s). calculated by the similar expression ;Vofe: Franklin and Lumpkin give D(HS-SH) = S0.4 kcal./mole, a value, which was presumably D(S-H) = Dz AH@) 4- AHf(H) - Allr(SH) (2) obtained by thein by using in their calculations Unfortunately, it is impossible a t present to decide AElf(HZS2)1rq = -3.6 kcal./mole instead of AHron one particular value for D2 since in equation ( 2 ) (H2S2)g,as required. Since the heat of vaporizaAHr(S) may have one of three values 53.5, 57 or tion of H2Szis probably -8 kcal /mole, it follows 66.5 kcal./mole5 depending on the choice of one of that D(HS-SH), based on their value for AIIf(SH), the three possible values of 76, 83 or 102.5 kcal./ should have actually been i 2 . 4 kcal./mole. On the mole for D ( S S ) in S Z . ~ Hence, D2 can be 67, other hand on the basis of our value for AHf(SH) 70.5 or 80 kcal./mole on the basis of Franklin and D(HS-SH) is 59.6 kcal./mole, a value which is Lumpkin’s value for AL\Hf(SH)or 73.5, 77 or 86.5 comparable with D(H0-OH) of 54 k ~ a l . / m o l e . ~ kcal./mole on the basis of our value for AHr(SH). (’0 \I. Srwarc, Chem R c r s , 47, 75 (1950) By adding equations (1) and (2) it follows that DIVISION OF CHEMISTRY the sum of the two S-H dissociation energies in SATIOVAL RESEARCH COUSCIL HzS can vary between 163 and 175 kcal./mole OTTA\VA, ONTA4RI0, CAVADA depending on whether D ( S S ) is taken as 76 or 102.5 kcal./mole. The uncertainty in choosing a particular value for D ( S S ) is due to the difficulty of interpreting unambiguously the spectrum of SZ. Quantum Mechanical Calculations of Orientation However, Gaydone after analyzing all relevant data in Aromatic Substitution’ is inclined to favor the highest value. From recent BY JOHN D. ROBERTS A N D ANDREW STREITWIESER. JR.* interpretations of the S-H spectrum Porter’ and RECEIVED MARCH11, 1952 Ramsay8 deduce values of S5 and 83 kcal./rnole, respectively, for Dz. These, in conjunction with Wheland3 has shown definitely that’ the molecuthe values given above for D 1 support also the lar orbital method makes possible a single unified highest value for D ( S S ) treatment of electrophilic, free-radical and nucleoFranklin and Lumpkin disagree with Porter’s philic substitution reactions of aromatic molecules. value for D2 and suggest instead a value of 67 kcal./ His approach was based on calculations of the enmole. Since the sum of their values for D1 and ergies of activated complexes of type I where z is a DS,respectively, 95 and 67 kcal./mole, is 162 kca1.l unit positive charge, an unpaired electron or a unit mole it appears that they tacitly assumed the lowest negative charge depending on whether the attackvalue for D ( S S ) . Indeed, an analysis of their ing reagent (R) is an electrophilic, radical or nucleoresults indicates that they used this value through- philic species. out their calculations. They argue that if Porter’s H H value of 85 kcal./mole for D2 is correct then D1 \ / must be 77 kcal./mole, i.e., D1would then be smaller ( X ) H y a RH than Dz, which is contrary to what one would expect by comparison with the corresponding bond H/ I‘ I
(3) This value has been calculated by using the data of our determination of D(CHrSH) by the “toluene-carrier” technique [A. H. Sehon and B. deB. Darwent, A. C. S. Meeting, Buffalo, March, 1952. A full account will be published shortly]. (4) (a) Selected Values of Chemical and Thermodynamic Properties, N. B. S., Washington, 1947; (b) K. K. Kelley, U.S.Dept of Interior, Bureau of Mines, Bulletin 406 (1937). (5) Since [ A B f ( S ) (AHkSdg D C S S ) ] / Z where AHf(Sdg = 80.80 IrcWmole. [W. H. Evans and D. D. Wagman, Natl., Bur. Standards, Report No. 1037, Washington, 1851.1 (6) A. G. Gaydon, “Dissociation Energlea and Spectra of Diotoraic Molecules,” Chapman and Hall, Ltd., London, 1947. (7) G. Porter, Discussion Faraday Soc., €4,80 (1860). ( 8 ) P,A, Rammay, ibid,, 80 (1960).
-
+
I
With various substituent groups (X) and reasonable assignments t o appropriate parameters, he found that the pattern of aromatic substitution could be well reproduced. I n the present work, the simple molecular orbital (1) Supported by the research proram of the U. 9. Atomic Energy Commirsion under C O B ~ XAT(aO-l)906. .~~ (2) U. 9. Atomic Energy Commission Post-Doctoral Fellow, 19611862. (a) 0. w. wm.ad, ~m JOVRNAL, ri, wo ( 1 8 ~ .
