12992
J. Am. Chem. Soc. 1998, 120, 12992-12993
Scheme 1
Reversal of the Stereochemistry of Kinetic
Protonation by Intramolecular Proton Delivery and a
Remarkable Dependence of Selectivity on Donor
Concentration1
Howard E. Zimmerman* and Alexey Ignatchenko
Chemistry Department, UniVersity of Wisconsin
Madison, Wisconsin, 53706
ReceiVed July 3, 1998
A very large number of reactions proceed by way of transient
enolic intermediates and related carbanion species. Nearly four
decades ago we reported2 that the carbon being protonated has a
transition state which is close to sp2 hybridized. Also it was noted
that as a consequence the preferred attack of the proton donor is
from the less hindered side of the delocalized species to afford
the less stable of two alternative stereoisomeric products. A long
series of our publications on the subject followed that initial
report.3 One relatively recent and typical example is given in
eq 1.3f
That the formation of the more stable of the two stereoisomeric
ketone products did not result from epimerization of an initially
formed endo-benzoyl ketone was evidenced by the lack of
reactivity of the endo-benzoyl ketone under conditions used for
the ketonization runs. Thus, with the highest concentration of
acetic acid used, less than 4% epimerization was observed.
A remarkable observation was that the stereoselectivity of
reaction of the endo-pyridyl enol was a function of the concentra-
tion of proton donor acetic acid employed, whereas the selectivity
of ketonization in the exo-pyridyl enol 4-exo was independent of
the donor concentration. This signifies that in the endo-pyridyl
ketonization, the formation of the two stereoisomers cannot have
the same kinetic order in (e.g.) acetic acid.
In ketonization of the endo-pyridyl enol, in 1.4 M acetic acid
the exo-benzoyl:endo-benzoyl product ratio was 97:3, whereas
in the more dilute 0.08 M acid a ratio of 1:3 resulted. We conclude
that more acetic acid molecules participate in the intramolecular
protonation than in the intermolecular process. We designate the
amount of the endo-benzoyl product 8, formed by the ordinary
intermolecular protonation, as P1 and the amount of exo-benzoyl
product 7, as Pn. Equations 2 give the extent of formation of the
In the present paper we report a unique example where
intramolecular proton delivery permits reversal of the ubiquitous
phenomenon and a remarkable dependence of the stereoselectivity
on proton donor concentration. The reactivity of the two enols,
4-Exo and 5-Endo, was the object of this study. The correspond-
ing silyl enol ethers4 were employed as precursors. Generation
of the enols with tetrabutylammonium fluoride led to the
anticipated stereochemistry in the case of 4-Exo where the less
hindered protonation with ammonium ion or acetic acid led to
the endo-benzoyl ketone 6.
In contrast, parallel ketonization of the endo-pyridyl counterpart
5-Endo, under conditions sufficiently acidic to permit prior partial
protonation of the pyridyl nitrogen, led to reversal of the reaction
stereochemistry. Thus, while ketonization of the endo-enol 5-Endo
with ammonium ion gave the endo-benzoyl-endo-pyridyl ketone
8 resulting from the normal, less hindered approach of the proton
donor, with 1.4 M acetic acid in THF, the stereoselectivity
favoring formation of the exo-benzoyl-endo-pyridyl ketone 7 was
97:3 (Scheme 1). Similar results were obtained in isopropyl
alcohol and DMSO solvents as well as with formic acid in THF.
dP1/dt ) k1[HA][Enol] and dPn/dt ) kn[HA]n[Enol] (2a,b)
two stereoisomers after total ketonization.5 Division of the two
equations gives, in logarithmic form, eq 3. Here we note the slope
log(Pn/P1) ) (n - 1) log[HA] + log(kn/k1)
(3)
of a plot of log(Pn/P1) versus the log of acid concentration [HA]
is (n - 1) and affords the difference in number of acid molecules
in the transition state for the intramolecular compared with the
ordinary intermolecular process.
(5) (a) Pyridine and acetic acid have been reported6 in relatively non-polar
solvents to exist in rapidly equilibrated7 clusters. Thus, eqs 2 and 3 assume
dissociation of the clusters prior to transition-state formation. However, if
clustering persists, eqs 2 and 3 are replaced by
(1) (a) This is paper 250. (b) For paper 249, see: Zimmerman, H. E. J.
Phys. Chem. A 1997, 102, 5616-5621.
dPm/dt ) km[HA]m[Enol] and dPn/dt ) kn[HA]n[Enol] (4a,b)
(2) (a) Zimmerman, H. E. J. Org. Chem. 1955, 20, 549-557; (b)
Zimmerman, H. E.; Nevins, T. E. J. Am. Chem. Soc. 1957, 79, 6559-6561.
(3) (a) For a review, see: Zimmerman, H. E. Acc. Chem. Res. 1987, 20,
263-268; A few references of the fourteen follow: (b) Zimmerman, H. E.;
Mariano, P. S. J. Am. Chem. Soc. 1968, 90, 6091-6096; (c) Zimmerman, H.
E. J. Am. Chem. Soc. 1957, 79, 6554-6558; (d) Zimmerman, H. E.; Chang,
W.-H. J. Am. Chem. Soc. 1959, 81, 3634-3643; (e) Zimmerman, H. E.;
Cutshall, T. W. J. Am. Chem. Soc. 1958, 80, 2893-2896; (f) Zimmerman, H.
E.; Linder, L. W. J. Org. Chem. 1985, 48, 1637-1646. (g) Interestingly ref
3a and d contain an algebraic predecessor to the computer molecular mechanics
analysis of acyclic conformational analysis of electrophilic attack on an sp2
hybridized carbon, a subject of considerable subsequent interest in studies
arriving at the same conclusions.
log(Pn/Pm) ) (n - m) log[HA] + log(kn/km)
(5)
and the slope again gives the difference in number of acid molecules in
the alternative transition states. We note that ketonization rates are slow
by comparison8 and thus cluster formation is not rate-limiting. However,
the relation between difference in reaction order and slope is independent
of this. (b) In eqs 2-5, as a consequence of the complexity of the media
with uncertain extents of hydrogen bonding and protonation, the constants
k1, kn, km may not be true rate constants but may incorporate equilibrium
constants. However, these constants are uninvolved in the slope obtained.
(6) Akiyama, Y.; Wakisaka, A.; Mizukami, F.; Sakaguchik, K. J. Chem.
Soc. Perkin Trans. 2 1998, 95-99.
(7) Golubev, N. S.; Smirnov, S. N.; Gindin, V. A.; Denisov, G. S.; Benedict,
H.; Limbach, H.-H. J. Am. Chem. Soc. 1994, 116, 12055-12056.
(8) Andraos, J.; Kresge, A. J.; Obraztsov, P. A. J. Phys. Org. Chem. 1992,
5, 322-326.
(4) Runs were made at 30 °C. All compounds were properly characterized
using elemental analysis, HRMS, NMR, IR, and X-Ray analysis in selected
cases. Kinetic details will be given in our full publication.
10.1021/ja9823262 CCC: $15.00 © 1998 American Chemical Society
Published on Web 12/02/1998