5342 Table I Isomer Atom 0(1,2)
B3 B4 B8 B9 m(1,7) B2 B4 B5 B9 p(1,12) B2
QA.~
1.211 1.250 1.300 1,300 1.197 1.251 1.252 1.301 1.251
ZQAB
dd)
7 - 6 Obsda
0.123 0.128 0.131 0.136 0.123 0.127 0.131 0.132 0.127
187.8 188.8 190.0 190.0 187.6 188.8 188.8 190.0 188.8
15.2 13.9 9.9 2.7 17.3 13.6 11.0 7.0 15.2
Calcdb 16.7 12.5 7.4 6.8 18.2 12.5 11.9 7.2 12.5
observed some rather extreme cases of shifts of tautomeric equilibria between ground and excited states. If we examine the equilibrium diagram for the possible equilibria in the first two protonations of a dibasic acid, we arrive at the following scheme in which BH+ and B'H+ are two tautomeric first conjugate acids of the base B which differ only in the position of attachment of the p r ~ t o n . The ~ equilibrium between BH+
See ref 7. The chemical shifts are expressed in parts per million from boron trifluoride etherate. * These values are calculated using the equation of the least-squares line given in Figure 1: 8cnlcd = ,J( d l - 12.4 - 118.9(Q,ka ZQAB).
BH+
Q
Kr
B'H+
+
will probably dominate the chemical shifts of other boron compounds. Acknowledgment. We wish to thank the Office of Naval Research for support of this research. F. P. Boer The Dow Chemical Company, Eastern Research Laboratories Wayland, Massachusetts 01778 R. A. Hegstrom, M. D. Newton, J. A. Potenza, W. N. Lipscomb Department of Chemistry, Harcard Unicersity Cambridge, Massachusetts 02138 Receiaed August 29, 1966
Excited State pK's.
IV. Tautomeric Equilibria
Sir: I t has been shown' that the equilibrium constants for acid-base reactions in electronically excited states can be obtained, if only approximately, from the electronic absorption spectra of the conjugate acid-base pair by use of the Forster cycle. In principle, the
p-Dimethylaminoazobenzene p-Dimethylaminoazobenzene N-oxide P-Dimethylaminoazoxybenzene 6-Dimethylaminoazoxybenzene N-oxide 2.3-Aminonaohthol
and B'H+ is defined by the equilibrium constant Ky = [BH+]/[B'H+] = K*/K1= Ks/Kd. We have recently observed a number of such equilibria in which the equilibrium constants undergo tremendous changes upon excitation. Thus, the equilibrium between the azonium and ammonium forms of dimethylaminoazobenzene (DMAB) first conjugate acid, in which both forms are present in measurable quantities (KT = 7.15), shifts, upon excitation, in the direction of the pure azonium form to the complete Simiexclusion of the ammonium form (KT* = larly, in the first conjugate acid of the N-oxide of DMAB, the equilibrium shifts from a pure N-hydroxyDMAB cation (KT = lo6) in the ground state to a pure in the singlet excited azonium cation (KT* state. In dimethylaminoazoxybenzene a similar change from an ammonium to an azonium form occurs (KT= lo8: KT* = lo-*'). In the amino id-oxide of the latter compound, the change in KT is still quite large, but the equilibrium position is not reversed; the Nhydroxyammonium salt is stable in ground and excited states. However KT* is quite small, and it seems
4.42' -9.64e 5.271 7.84e -4.43" 10.76" -5.28, -4.91b 9.84, 4.110 4.11e 4.37f 4.37e -4.650 -5 c 5.6, 2.620 -15.5@ -0.41 -18.5, -8.020 -8.5gd 11.2~
1.22, 3.710 6.7~6.0,
3.71e 0.3e
3.89f 2.8i
3.89k -8.410 -8.2e 8.0C
-15.59e 10.le 2.68'
- 0 . 8 5 h -26.3' -9.02f 5.7f -7.6f 20f
1.39e -12, -1.8a 5.2f
-2f 6.4,
Calculated from the Hammett equation using parameters from H. H. Jaff6, Chem. Rec., 53, 191 (1953). * Presumed equal t o the pK Calculated of p-phenylazo-N,N,N-trimethylanilinium ion. 2 s c Calculated from the Hammett equation using parameters from ref 2c. from the Hammett equation assuming u ~ ~= uN(CHa)s. ( ~ ~e Calculated ~ ) ~from the Forster cycle. f Calculated from the relationships i Experimental. 1 j Constant within the ecruation scheme. 0 Experimental, Ellerhorst and Jaff6, unpublished work. h Experimental. ApK assumed-for equilibria 2 and 3. Assumed ApK = 0.
method should be applicable to any type of equilibrium constants. Of particular interest may be tautomeric equilibria which, when occurring in polar media, may be expressed as a combination of several acidbase reactions, and which may be established fast enough to occur in excited singlet states. In connection with our studies of the basic properties of amino-substituted azo and azoxy compounds, we have ( I ) CJ., e.g. (a) H. H. Jaffe, D. L. Beveridge, and H. L. Jones, J . Am. Chem. SOC.,86, 2932 (1964); (b) H. H. Jaffe and H. L. Jones, J . Org. Chem., 30, 964 (1965). (2) (a) S. J. Yeh and H. H. Jaffk, J . Am. Chem. SOC.,81, 3283 (1959); (b) M. Isaks and H. H. Jaffe, ibid., 86, 2209 (1964); (c) C. S . Hahn and H. H . JaffG, ibid., 84, 949 (1962).
Journul of the American Chemical Society
88:22
feasible that the hydroxyazonium form may be observable. Finally, a similar analysis of the aminonaphthols4 suggests that KT changes are much smaller, and no reversal of stable forms occurs; i.e., the neutral form rather than the zwitterion is the stable form in both ground and excited states. The data here presented are subject to a great deal of uncertainty. First, the Forster cycie is used throughout, and is known to yield only approximate (3) Note that throughout basicities of bases are expressed as p& of the conjugate acid. (4) D. W. Ellis and L. B. Rogers, Spectrochim. Acta, 20, 1709 (1964). ( 5 ) (a) E. L. Wehry and L. B. Rogers, ibid., 21, 1976(1965); (b) J . C. Haycook, S . F. Mason, and B. E. Smith, J. Chem. Soc., 4897 (1963).
Nouember 20, 1966
5343
Second, many pK's are estimated on the basis of the Hammett equation or on the basis of crude approximation of substituent effects, as explained in the footnotes to Table I. However, the effects observed are so large that uncertainties of even a power of ten in KT appear insignificant. Thus, we can conclude that, at least occasionally, we may expect to see tautomeric equilibria shift completely from one side to the other upon electronic excitation. This fact may be of considerable importance in considerations of the mechanism of photochemical reactions, because the reacting species may be a tautomer not occurring in the ground state. (6)
i
U.S. Public Health Service Fellow, 1963-1965.
(7) National Aeronautics and Space Administration Trainee, 1 9 6 4 1966.
R. H. Ellerhorst,' H. H. JaRe', Albert L. Miller' Depanmem of Chemistry, Uniwrsity of Cincimati Cincinnati, Ohio 45221 Receiued October 6, 1966 Inner-Sphere Mechanisms for the Reduction of Cobalt(II1) Complexes by Iron(I1)'
Sir: Since the discovery2 of the bridged activated complex in the Co(NH3)&1*+-CrZ+ reaction, many electrontransfer reactions have been studied in order to determine whether inner- or outer-sphere mechanisms obtain.8 For reducing agents such as chromium(11)' and pentacyanocobaltate(II),6 the product criterion of mechanism is applicable.' For other reducing agents (iron(II), vanadium(II), eur~pium(II)),"~ however, the primary reaction products cannot, in general, be identified since the rates of substitution in the coordination spheres of the oxidation products are usually rapid compared to the rates of the oxidation-reduction reactions.lQ.11 In spite of the lability of iron(II1) complexes, we have found that the reductions of C O ( C ~ O ~ ) ~Co~-, (H .EDTA)Cl-, Co(en)z(OH2)Cl2+,Co(NHa)3(OHz)%Cl2+, Co(NH3)3(0H%)~Ns2+, and Co(NH&(0H#22O4+ by iron(I1) were sufficiently rapid so that the primary iron(111) products of these reactions could he identified.'% The identification was performed by determining the spectrum of the iron(II1) product and by observing its rate of formation and decay. These measurements were made with the rapid-flow apparatus described previously. l a
x t Figure 1. Transmittance us. time curves for the C O ( C , O ~ ) ~ ~ - - F ~ ~ + M, [Fez+] = 2.5 X reaction at 25". ICO(CO~)~'-I= 1.0 X 10-2 M, [HCIOJ = 0.92 M,ionic strengih = 1.0 M. Upper curve (36a) shows the disappearance of Co(C~O,),'-: wavelength, Mx) mp; abscissa scale, 500 msec per major division. Lower curve (36b) shows the formation and disappearance of the intermediate FeCOrc: wavelength, 310 m p ; abscissa scale, 2 sec per major division.
The Co(C20&'--Fe2+ ceeds in two stages. Co(GO&-
Reaction.
This reaction pro-
+ Fe'+ +FeGO,+ + Co*++ 2G04*-
+
FeC04+ 2Ht
e Fea+ + H G O .
(1)
(2)
(1964). (8) 1. H. Espennon, Inorg. Chem., 4, I21 (1965). (9) H. Diebler and H. Taube, ibid., 4. 1029 (1965). (IO) It must be noted, however, that FeCIa+ has been identified as the
By using different wavelengths, both the disappearance of CO(C~O,),~-and the formation and subsequent decay of FeC2O4+were observed. The disappearance of the cobalt(II1) complex was followed a t 600 mp, an absorption maximum of C O ( C ~ O ~ and ) ~ ~a- region where none of the other reactants or products absorb significantly. The rate constant for the oxidation0.1) X 10' M-' sec-' at reduction reaction is (3.3 25.0" and ionic strength 1.0 M." The formation and disappearance of FeC,O,+ was followed a t 310 mp. As shown in the lower curve of Figure 1, a rapid decrease in transmittance associated with the oxidationreduction step is followed by an increase in transmittance corresponding to the disappearance of the FeC204+formed in the first step. Because of the ex-
primary product of the CoC12+-Fer+ reaetion.11 (11) T. I. Conaechioli, G. H. Nancollas, and N. Sutin, J. Am. Chcm. Soc., 86, 1453 (1964). (12) This criterion appears to be applicable to the Co(NH*)aGOaH'+Fer+reaction. Calculations indicate that at low acid concentration the rate of this reactions becomes comparable to the rate of dissociation ofFeGO,+.lP Weare planningto stndy thisreactionin detail.
(13) 0. Dulz and N. Sutin, Inorg. Chem., 2,917 (1963). (14) The rate constant for reaction I is 1.15 x IOs M-I sec-I at 20' and zero ionic strength: I. Barrett and J. H. Baxendale, Trans. Faradoy Soc., 52. 210 (1956). The lower rate constant determined in the present work is consistent with the effect of ionic strength on the reaction rate.
(1) Research performed under the auspices of the U. S. Atomic Energy Commission. (2) H. Taube, H. Myers, and R. L.Rich, 3. Am. Chem. Soc., 15,4118
,.,,. ,1022,
(3) For a recent review of this subject, see N. Sutin, Ann. Reo. Phys. Chem., 17, 119 (1966). (4) H. Taube, Adwn. Inorg. Chem. Radiochem., 1, 1 (1959). ( 5 ) 1. P. Candlin, 1. Halpern, and S. Nakamura, 3. Am. Chsm. Soc.,
85, 2517 (1963).
(6) A. M. Zwickel and H. Taube, lbld., 83, 793 (1961). (7) I. P. Candlin, 1. Halpern, and D. L. Trimm, ibld., 86, 1019
*
Communicafionsto the m i f o r
