Foiled conjugation in α-oximino carbocations

Article abstract of DOI:10.1021/jo960073u

The 4-CHNOCH₃ group is a cation-stabilizing group when placed in the para-position of a cumyl cation. The effect of this group on cumyl cations when flanked by adjacent methyl groups has now been determined. Solvolysis rates of 3,5-(CH₃)₂-4-(CHNOCH₃)cumyl trifluoroacetates are somewhat slower than that of 3,5-dimethylcumyl trifluoroacetate. This is attributed to steric inhibition of the cation-stabilizing resonance effect of the p-oximino group. In a 1-adamantyl system, where an α-oximino group has been placed directly adjacent to a developing cationic center, solvolysis rates relative to 1-adamantyl mesylate are slowed by a factor of 10⁸. This is attributed a cation-destabilizing inductive effect where geometric constraints prevent stabilizing orbital overlap of the cationic center with the adjacent α-oximino group. This cation-destabilizing effect fades in the homoadamantyl and the bicyclo[3.3.1]nonyl systems, where rate-retarding effects are 1.6 × 10⁴ and 1.5 × 10², respectively. The behavior of geometrically constrained α-oximino cations parallels that of analogously constrained allylic cations. Computational studies at the HF/6-31G* level indicate that twisting the α-oximino group out of planarity with a tertiary cationic center into a perpendicular arrangement decreases stabilization by 21 kcal/mol. These studies suggest that conjugative interactions, and not ground state destabilization, are the most important factors in controlling rates of formation of α-oximino cations from mesylates and trifluoroacetates.

Full text of DOI:10.1021/jo960073u

3482
J . Org. Chem. 1996, 61, 3482-3489
F oiled Con ju ga tion in r-Oxim in o Ca r boca tion s
Xavier Creary* and Ziqi J iang
Department of Chemistry and Biochemistry, University of Notre Dame, Notre Dame, Indiana 46556
Received J anuary 11, 1996X
The 4-CHNOCH3 group is a cation-stabilizing group when placed in the para-position of a cumyl
cation. The effect of this group on cumyl cations when flanked by adjacent methyl groups has now
been determined. Solvolysis rates of 3,5-(CH3)2-4-(CHNOCH3)cumyl trifluoroacetates are some-
what slower than that of 3,5-dimethylcumyl trifluoroacetate. This is attributed to steric inhibi-
tion of the cation-stabilizing resonance effect of the p-oximino group. In a 1-adamantyl system,
where an R-oximino group has been placed directly adjacent to a developing cationic center, solvoly-
sis rates relative to 1-adamantyl mesylate are slowed by a factor of 108. This is attributed a cation-
destabilizing inductive effect where geometric constraints prevent stabilizing orbital overlap
of the cationic center with the adjacent R-oximino group. This cation-destabilizing effect
fades in the homoadamantyl and the bicyclo[3.3.1]nonyl systems, where rate-retarding effects
are 1.6 × 104 and 1.5 × 102, respectively. The behavior of geometrically constrained R-oximino
cations parallels that of analogously constrained allylic cations. Computational studies at the
HF/6-31G* level indicate that twisting the R-oximino group out of planarity with a tertiary ca-
tionic center into a perpendicular arrangement decreases stabilization by 21 kcal/mol. These
studies suggest that conjugative interactions, and not ground state destabilization, are the most
important factors in controlling rates of formation of R-oximino cations from mesylates and
trifluoroacetates.
Carbocations 1 substituted with formal electron-
withdrawing groups can often be generated with surpris-
ing ease.1 This is illustrated by the R-oximino cations of
general structure 2, which form quite readily under
solvolytic conditions.2,3 Rate comparisons suggest that
the R-CHNOCH3 group is substantially more cation
stabilizing than R-H and even more stabilizing than
R-CH3. This cation-stabilizing effect by the inductively
electron-withdrawing R-CHNOCH3 group was attributed
to a conjugative effect as represented by 2a . This effect
was supported by ab initio molecular orbital calculations,
which reveal extensive charge delocalization in cations
such as 2. In order to better understand the nature of
R-oximino stabilization of cations, we have now studied
the cations 3-6, where geometric constraints prevent
complete overlap of the cationic center with the conjugat-
ing CdN bond. Reported here are the results of these
studies.
tuted cumyl cation intermediate. By way of contrast, the
para-substituted analog 7c actually solvolyzes 1.35 times
faster than the unsubstituted cumyl trifluoroacetate 7a .
Resu lts a n d Discu ssion
One of the classic methods for evaluating the electronic
characteristics of a substituent is measurement of the
Hammett-Brown σ+ constant.4 The m-CHNOCH3 group
(σ+ ) 0.15) slows the solvolysis rate of trifluoroacetate
7b by a factor of 5 relative to the unsubstituted system
7a .2 This is indicative of a moderate electron-withdraw-
ing electronic effect which destabilizes the meta-substi-
X Abstract published in Advance ACS Abstracts, April 15, 1996.
(1) Creary, X. Chem. Rev. 1991, 91, 1625.
(2) Creary, X.; Wang, Y.-X.; J iang, Z. J . Am. Chem. Soc. 1995, 117,
3044.
(3) For examples of R-oximino cations under nonsolvolytic condi-
tions, see: (a) Shatzmiller, S.; Lidor, R.; Shalom, E.; Bahar, E.; J .
Chem. Soc., Chem. Commun. 1984, 795. (b) Shatzmiller, S.; Shalom,
E.; Bahar, E. J . Chem. Soc., Chem. Commun. 1984, 1522. (c) Hansen,
J . F.; Strong, S. A. J . Heterocycl. Chem. 1977, 14, 1289. (d) Hansen, J .
F.; Kim, Y. I.; McCrotty, S. E.; Strong, S. A.; Zimmer, D. E. Ibid. 1980,
17, 475.
We have proposed that a cation-stabilizing resonance
interaction offsets the destabilizing inductive effect. To
further support this contention, the 3,5-dimethyl-substi-
tuted systems 9 and 10 have been prepared and solvolysis
rates have been compared to the “unsubstituted” analog
8. Rates of ethanolysis (Table 1) of these oxime deriva-
tives are depressed relative to 8. This is attributed to
inhibition of the potential cation-stabilizing conjugative
(4) Brown, H. C.; Okamoto, Y. J . Am. Chem. Soc. 1958, 80, 4979.
S0022-3263(96)00073-4 CCC: $12.00 © 1996 American Chemical Society

Foiled Conjugation in R-Oximino Carbocations
J . Org. Chem., Vol. 61, No. 10, 1996 3483
Ta ble 1. Solvolysis Ra tes of Su bstr a tes in Va r iou s
interaction by the o-dimethyl substituents; i.e., the
cationic intermediate prefers the unconjugated conforma-
tion as in 11.
Solven ts a t 25 °C
substrate
solvent
EtOH
EtOH
EtOH
ka (s-1
)
8
9
10
16
9.82 × 10-4
4.25 × 10-4
3.56 × 10-4
3.16 × 10-9 b
CF3CH2OH
2.92 × 10-6 (80.0 °C)
2.14 × 10-5 (100.0 °C)
2.99 × 10-1
17
18
19
20
21
25
26
32
CF3CH2OH
CF3CH2OH
CF3CH2OH
CH3OH
3.46 × 10-3
5.66 × 101 c
1.82 × 10-5
CH3OH
2.71 × 10-3
CF3CH2OH
CF3CH2OH
CF3CH2OH
5.58 × 10-5 d
7.91 × 10-6 d
3.74 × 10-3
a
Maximum standard deviations in duplicate runs were (1.5%
for rates determined by the 1H NMR method of ref 23. See the
b
Experimental Section for the kinetic method. Extrapolated from
data at higher temperature. c Estimated from the rate of the
corresponding trifluoroacetate (2.01 × 10-4 s-1) and a mesylate/
d
trifluoroacetate rate ratio of 2.81 × 105. See ref 8. Reference
23.
The effect of rotating a conjugating group out of
conjugation with a cationic center was first investigated
in the classic study of Martin5 and Schleyer.6 In the
tosylate 12, the inductive effect of the double bond results
in a retardation of solvolytic rate relative to the model
substrate 13. This approach has been greatly extended
by Takeuchi, who carried out solvolytic studies on a series
of substrates of general structure 14.7 Depending on ring
size, interaction of the developing cationic center of 15
with the double bond led to varying degrees of rate
enhancement or rate retardation.
oxime group, when allowed to interact with a cationic
center in a conjugative fashion, is even more cation
stabilizing than an R-CH3 group. Hence, studies on 22
and 25 show that cations 24 and 27 form under sol-
volytic conditions even faster than the tert-butyl and
2-methylnorbornyl cations, respectively. Cations 24 and
27 are highly delocalized intermediates, with charge
dispersed onto nitrogen and oxygen atoms. The cation-
destabilizing effect of the electron-withdrawing oxime
group is completely offset in these two R-oximino car-
bocations.
The effect of the oxime group on the solvolysis of
mesylate 18 is also to retard the rate relative to that of
the model substrate 19.8 However, the rate-retarding
effect of 10-4.2 is not as large as in mesylate 16. These
data suggest that the cationic center of 5 is beginning to
exhibit a small amount of conjugation with the CdN
We have used a similar approach in the study of
R-oximino cations. The mesylates 16, 18, and 20 were
prepared and solvolyzed in various solvents. In all cases
the simple substitution products were the only products
formed. Rate data are given in Table 1, along with data
for the model systems 17, 19, and 21 lacking the oxime
group. Comparison of the rate of 16 with that of
1-adamantyl mesylate, 17, shows that the oxime group
slows the rate by a factor of 108. This enormous rate
retardation is attributed to an inductive effect of the
oximine group which is directly attached to the develop-
ing cationic center of 4 but cannot interact in a con-
jugative fashion with that orthogonal cationic center. By
way of contrast, our previous study2 has shown that the
(5) Ree, B. R.; Martin, J . C. J . Am. Chem. Soc. 1970, 92, 1660.
(6) Buss, V.; Gleiter, R.; Schleyer, P. v. R. J . Am. Chem. Soc. 1971,
93, 3927.
(7) Takeuchi, K.; Kitagawa, T.; Ohga, Y.; Toshida, M.; Akiyama, F.;
Tsugeno, A. J . Org. Chem. 1992, 57, 280.
(8) The rate of mesylate 19 was estimated from that of the
corresponding trifluoroacetate derivative. A mesylate/trifluoroacetate
rate ratio of 2.81 × 105 was measured for the 1-adamantyl system 17,
and this value was used to calculate the rate of 19.

3484 J . Org. Chem., Vol. 61, No. 10, 1996
Creary and J iang
ates in a parallel fashion in both allylic and R-oximino
constrained carbocations.
The rate data on mesylate 16 imply a substantial
interaction of the CdN bond with the adjacent developing
cationic center in 6. The bridgehead alkene 33 can be
considered as a model for the interaction of adjacent sp2-
hybridized centers in the bicyclo[3.3.1]nonyl system. This
strained alkene 33 can be isolated and is thermally stable
at room temperature.9 The interaction of the sp2-hybrid-
ized centers in 33 is therefore quite good despite the
bridgehead nature of this alkene. This is in contrast to
the alkene 34 which readily dimerizes at room temper-
ature10 and the alkene 35 which dimerizes when warmed
above 70 K.11 It is therefore reasonable to expect a
substantial interaction of the cationic center of 6 with
the adjacent unsaturated center.
bond. In the cation 4, the CdN bond is held rigidly
orthogonal to the cationic center, as depicted in 28.
However, the additional methylene group in 5 renders
this cation less rigid, and hence, the double bond in 5
can achieve partial overlap with the cationic center, as
depicted in 29.
Com p u ta tion a l Stu d ies. In order to gain further
insights into the nature of cations 4-6 and the ability of
the oxime group to interact with cationic centers, ab initio
molecular orbital computational studies12 were carried
out on these cations, as well as on the hydrocarbons 33-
35. These alkenes present a useful model for the
interaction of a bridgehead sp2-hybridized carbons with
an adjacent sp2-hybridized carbon. The HF/6-31G*-
optimized geometries of 33-35 are shown with all but
the olefinic hydrogens removed for clarity. These geom-
etries all indicate a significant interaction between the
two sp2-hybridized centers. In the case of alkene 33, the
C1C3H plane is only rotated 15° with respect to the
C2C8C9 plane (as shown in the Newman projection 33a ).
In other words, the p-bond in this alkene is distorted by
a relatively small amount. This computational result is
in line with the observed kinetic stability of this alkene.10
The alkene 34 is more twisted than 33, with the
analogous interplanar angle being 30° in 34. Finally, the
alkene 35 is the most twisted of the three, with a 49°
interplanar angle. However, the fact that the angle is
not 90°, and the CdC bond length (1.339 Å) is signifi-
cantly shorter than a normal C-C bond, suggests that
The solvolytic rate data for the bicyclic mesylate 20
indicate that the importance of conjugation has again
increased, but not to the extent that the inductive effect
of the oxime group is entirely negated. The rate-
retarding effect of the oxime group is now only a factor
of 149 and suggests that there is significant overlap of
the cationic center of 6 with the CdN bond as depicted
in 30. However, the overlap in cation 6 is still not
equivalent to that in the unconstrained cations 24 and
27, where conjugative stabilization is maximal (as de-
picted in 31).
(9) (a) Wiseman, J . R.; Pletcher, W. A. J . Am. Chem. Soc. 1970,
92, 956. (b) Marshall, J . A.; Faubl, H. J . Am. Chem. Soc. 1970, 92,
948.
(10) Martella, D. J .; J ones, M., J r.; Schleyer, P. v. R.; Maier, W. F.
J . Am. Chem. Soc. 1979, 101, 7634.
(11) (a) Conlin, R. T.; Miller, R. D.; Michl, J . J . Am. Chem. Soc. 1979,
101, 7637. (b) Michl, J .; Radziszewski, G. J .; Downing, J . W. Pure Appl.
Chem. 1983, 55, 315.
(12) Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.;
J ohnson, B. G.; Robb, M. A.; Cheeseman, J . R.; Keith, T.; Petersson,
G. A.; Montgomery, J . A.; Raghavachari, K.; Al-Laham, M. A.;
Zakrzewski, V. G.; Ortiz, J . V.; Foresman, J . B.; Peng,C. Y.; Ayala, P.
Y.; Chen, W.; Wong, M. W.; Andres, J . L.; Replogle, E. S.; Gomperts,
R.; Martin, R. L.; Fox, D. J .; Binkley, J . S.; Defrees, D. J .; Baker, J .;
Stewart, J . P.; Head-Gordon, M.; Gonzalez, C.; Pople, J . A. Gaussian
94, Revision B.3, Gaussian, Inc., Pittsburgh, PA, 1995.
The kinetic behavior of mesylates 16, 18, and 20 and
trifluoroacetate 25 can be compared to that of the allylic
derivatives 12, 14a , and 14b (studied by Martin,5 Schley-
er,6 and Takeuchi7) and the allylic trifluoroacetate 32.
As in the case of R-oximino cations, the rate-retarding
effect of a perpendicular allylic double bond is maximal
and decreases as the double bond is brought into conju-
gation. Thus, a plot (Figure 1) of solvolytic rate data for
the oxime derivatives 16, 18, 20, and 25 vs analogous
data for the allylic systems 12, 14a , 14b, and 32 is linear.
This linear free energy relationship supports the idea
that geometric inhibition of resonance stabilization oper-

Foiled Conjugation in R-Oximino Carbocations
J . Org. Chem., Vol. 61, No. 10, 1996 3485
CdN bonds of the cations are always twisted out of
conjugation with the cationic center, and the amount of
twisting is always larger than the analogous alkene twist.
In cation 6 the C1C3N plane is rotated 31° with respect
there still remains a significant interaction between the
sp2-hybridized centers of 35.
to the C2C8C9 plane. This is indicative of a substantial
interaction of the cationic center C1 with the CdN
bond, but this interaction is not of the same magnitude
as in the alkene 33 (where the interplanar angle is
only 15°). A comparison of bond lengths of cation 6 with
those of the neutral oxime 36 and the planar cation 24
supports the idea of partial conjugation. The 1.445 Å
bond length of the C1C2 bond of cation 6 is between
that of the single bond (1.526 Å) of 36 and the conju-
gated C1C2 bond (1.401 Å) of the planar cation 24. The
CN and NO bonds of cation 6 are also intermediate
between those of the oxime 36 and cation 24. These
structural features are all in line with the idea of
repressed (but not completely eliminated) CdN conjuga-
tion in the cation 6.
The calculated geometry of cation 5 is in line with a
further decrease in the importance of conjugation of the
cationic center with the oxime functionality relative to
the less constrained cation 6. The cationic center of 5 is
rotated 60° with respect to the oxime group. The
substituted 1-adamantyl cation 4 is now completely
unconjugated with the oxime group as revealed by the
calculated geometry (90° rotation of the cationic center
relative to the CdN bond). The â-C-C bonds of the ring
are significantly lengthened (1.618 Å), which suggests
The HF/6-31G* geometries of the cations 4-6 are also
shown, and they parallel those of alkenes 33-35. The

3486 J . Org. Chem., Vol. 61, No. 10, 1996
Creary and J iang
from the completely aligned arrangement in 24 to the
perpendicular arrangement in 4 is therefore a 21.7 kcal/
mol decrease in stabilization.
that cation 4 receives stabilization mainly by the type of
interaction represented in 4a (or 4b). The unimportance
of CdN conjugation is manifested by the +C-C, CdN,
and N-O bond lengths which retain alternating single,
double, and single bond character, respectively.
The effect of twisting the cationic center out of conju-
gation with the CdN bond was also examined using the
isodesmic reactions 1 and 2 (Scheme 1).13 Results of
these calculations at the HF/6-31G* level are shown in
Table 2. While formation of the planar conjugated cation
24 is favored by 7.3 kcal/mol, the reaction becomes
increasingly unfavorable as the oxime group is twisted
out of conjugation with the cationic center. In the case
of the perpendicular R-oximino cation 4, the destabiliza-
tion relative to the 1-adamantyl cation amounts to 14.4
kcal/mol. The overall effect of rotating the oxime group
Finally, the effect of twisting the stabilizing oxime
group out of conjugation with the cationic center can be
evaluated theoretically in the cation 24. Energies of
geometrically constrained cations were calculated as a
function of the NCCC dihedral angle. Energies of these
cations were calculated by rotating the oximino group out
of conjugation with the cationic center by 10° increments.
These rotated cations were further constrained to prevent
closure to cyclized cations by fixing the NCC angle at
116.9° (the same as the optimized angle in the planar
cation 24). The results of these calculations are displayed
in Figure 2. The effect of this rotation is to destabilize
the cation by 20.9 kcal/mol as the dihedral angle is
increased to 90°. This value of 20.9 kcal/mol is consistent
with the net destabilization (21.7 kcal/mol) calculated
from the isodesmic reactions in Table 2.
Con clu sion s. The cation-stabilizing effect of a
4-CHNOCH3 group on a cumyl cation can be negated by
placement of methyl groups in the 3- and 5-positions on
the phenyl ring. This is attributed to steric inhibition of
(13) An alternative approach to evaluating this effect is the calcula-
tion of hydride affinities. Relative hydride affinities of cations 4, 5, 6,
and 24 are 0, 1.8, 6.6, and 15.2 kcal/mol, respectively, at the HF/6-
31G* level.
F igu r e 1. Plot of log krel for solvolysis of R-CNOCH3 systems
vs log krel for solvolysis of R-CdCH2 systems.

Foiled Conjugation in R-Oximino Carbocations
J . Org. Chem., Vol. 61, No. 10, 1996 3487
Sch em e 1
Ta ble 2. Ca lcu la ted Va lu es (HF /6-31G*) of ∆E (k ca l/m ol)
for th e Isod esm ic Rea ction s 1 a n d 2
system
adamantyl
homoadamantyl
bicyclo[3.3.1]nonyl
tert-butyl
R-oximino cation
∆E (kcal/mol)
4
5
6
14.4
11.9
4.3
24
-7.3
the cation-stabilizing resonance effect of the oximino
group. The oximino group now becomes a weak cation-
destabilizing group. The R-oximino group can also
become a relatively strong cation-destabilizing group
when geometric constraints prevent interaction with an
adjacent cationic center. Hence, placement of an oximino
group adjacent to the cationic center in a 1-adamantyl
system slows the rate of cation formation by a factor of
108. This cation-destabilizing effect fades as geometric
constraints are lessened in the homoadamantyl and the
bicyclo[3.3.1]nonyl systems. The behavior of these geo-
metrically constrained R-oximino cations parallels that
of analogously constrained allylic cations. Computational
studies at the HF/6-31G* level indicate that twisting the
R-oximino group out of planarity with a tertiary cationic
center into a perpendicular arrangement decreases sta-
bilization by 21 kcal/mol.
These studies also address the possibility of ground
state destabilization of the substrate as the origin of the
rapid rates of solvolyses of R-oximino systems. Kirmse
and Houk14 have pointed out that there can be a
destabilizing geminal interaction between an electron-
withdrawing R-substituent and a leaving group such as
triflate or mesylate. This can raise the ground state
energy of the substrate, thereby leading to greater than
expected solvolysis rates. Our present findings (that
rates are so highly dependent on geometry of the incipi-
ent cation in a fashion that parallels allyl systems)
suggest that ground state destabilization is not the most
important factor in the formation of R-oximino cations.
The major rate-controlling factor is R-oximino conjuga-
tion.
F igu r e 2. Calculated energies (HF/6-31G*) as the cationic
center is rotated out of conjugation with R-CHNOCH3.
solution after concentration of the solvent used in the work-
up: 1H NMR of trifluoroacetate 8 (Et2O) δ 6.96 (br s, 2 H),
6.89 (br s, 1 H), 2.27, (s, 6 H), 1.82 (s, 6 H); 13C NMR of
trifluoroacetate 8 (Et2O) δ 143.67, 138.13, 129.57, 122.28,
87.53, 27.97, 21.02.
P r ep a r a tion of Tr iflu or oa ceta tes 9 a n d 10. These
trifluoroacetates were prepared by the sequence shown below.
A solution of 6.14 g of 2,6-dimethyl-4-bromobenzaldehyde16 and
Exp er im en ta l Section
P r ep a r a tion of Tr iflu or oa ceta te 8. This trifluoroacetate
was prepared by reaction of 3,5-dimethylcumyl alcohol in ether
with trifluoroacetic anhydride and 2,6-lutidine using the
previously described procedure.15 This trifluoroacetate was
stored in ether solution since removal of the ether solvent led
to rapid decomposition. NMR spectra were recorded in ether
(14) a) Kirmse, W.; Goer, B. J . Am. Chem. Soc. 1990, 112, 4556. (b)
Wu, Y.-D.; Kirmse, W.; Houk, K. Ibid. 1990, 112, 4557.
(15) Creary, X.; Casingal, V. P.; Leahy, C. E. J . Am. Chem. Soc. 1993,
115, 1734.

3488 J . Org. Chem., Vol. 61, No. 10, 1996
Creary and J iang
4.59 g of trimethylorthoformate in 60 mL of methanol was
stirred, and 270 mg of p-toluenesulfonic acid was added. The
mixture was stirred at room temperature for 20 h and 93 mg
of sodium methoxide was then added. The excess methanol
was removed using a rotary evaporator and the residue taken
up into ether and washed with a small amount of water. The
ether extract was then washed with saturated NaCl solution
and dried over MgSO4. The solvent was removed using a
rotary evaporator, and the residue was distilled to give 7.19 g
(96% yield) of 2,6-dimethyl-4-bromobenzaldehyde dimethyl
temperature for 10 min and then taken up into ether. The
mixture was then washed with cold water, dilute HCl solution,
and saturated NaCl solution. The organic extract was then
dried over MgSO4, and the solvent was removed using a rotary
evaporator.
The crude sulfinate ester was dissolved in 6 mL of CH2Cl2
and then cooled to 0 °C. m-Chloroperbenzoic acid (100 mg of
85%) was then added, and the mixture was then stirred at
room temperature for 1.5 h. The mixture was then taken up
into ether, and the ether solution was washed with a combined
solution of NaI-Na2S2O3-KOH. The ether solution was then
dried over MgSO4, and the solvents were removed using a
rotary evaporator leaving 102 mg (80% yield) of mesylate 16:
mp 57-59 °C; 1H NMR (CDCl3) δ 3.884 (s, 3 H), 3.726 (br s, 1
H), 3.226 (s, 3 H), 2.45 and 2.39 (AB quartet, J ) 11.7 Hz, 4
H), 2.28 (m, 2 H), 1.77 (br s, 6 H); 13C NMR (CDCl3) δ 160.08,
87.46, 61.59, 44.92, 40.78, 36.14, 34.69, 31.78, 30.37. Anal.
Calcd for C12H19NO4S: C, 52.72; H, 7.01. Found: C, 53.00;
H, 7.07.
1
acetal: bp 98-100 °C (3 mm); H NMR (CDCl3) δ 7.146 (s, 2
H), 5.430 (s, 1 H), 3.390 (s, 6 H), 2.414 (s, 6 H); 13C NMR
(CDCl3) δ 139.23, 133.30, 131.63, 121.86, 104.33, 55.27, 20.13.
Conversion of 2,6-dimethyl-4-bromobenzaldehyde dimethyl
acetal to 2,6-dimethyl-4-(2-hydroxy-2-propyl)benzaldehyde was
completely analogous to the previously described procedure:
17
1H NMR (CDCl3) δ 10.565 (s, 1 H), 7.204 (s, 2 H), 2.614 (s,
6 H), 2.20 (br s, 1 H), 1.577 (s, 6 H); 13C NMR (CDCl3) δ 193.12,
154.21, 141.41, 130.85, 125.83, 72.33, 31.54, 20.78.
Conversion of 2,6-dimethyl-4-(2-hydroxy-2-propyl)benzal-
dehyde to the O-methyl oxime derivative was completely
analogous to the previously described procedure.2 The O-
methyl oxime consisted of an 88:12 mixture of anti and syn
isomers. These were separated by silica gel chromatography
(elution with increasing amounts of ether in hexanes). The
anti-oxime eluted first followed by the syn-oxime. anti-
Oxime: 1H NMR (CDCl3) δ 8.336 (s, 1 H), 7.154 (s, 2 H), 3.966
(s, 3 H), 2.407 (s, 6 H), 1.92 (br s, 1 H), 1.547 (s, 6 H); 13C
NMR (CDCl3) δ 149.56, 147.88, 137.49, 127.71, 124.54, 72.26,
61.85, 31.62, 21.44. Anal. Calcd for C13H19NO2: C, 70.54; H,
8.66. Found: C, 70.60; H, 8.49.
P r ep a r a tion of 1-Ad a m a n tyl Mesyla te, 17. Mesylate 17
was prepared by oxidation of the sulfinate ester of 1-adaman-
tanol using a procedure completely analogous to the prepara-
tion of mesylate 16. Thus, reaction of 370 mg of 1-adaman-
tanol with 359 mg of CH3SOCl and 492 mg of Et3N gave 547
mg of the crude sulfinate ester. Oxidation of this ester with
570 mg of 85% m-chloroperbenzoic acid gave 549 mg of a
mixture consisting of 87% 1-adamantyl mesylate, 17, and 13%
1-adamantanol. Attempted chromatographic purification on
silica gel led to decomposition of 17. Recrystallization fron
hexanes did not improve the purity: 1H NMR of 17 (CDCl3) δ
2.995 (s, 3 H), 2.238 (br s, 9 H), 1.667 (br s, 6 H); 13C NMR
(CDCl3) δ 91.85, 42.96, 41.03, 35.61, 31.52. The mixture
containing 87% of mesylate 17 was used for kinetic studies.
P r ep a r a tion of Mesyla te 18. 1-Hydroxyhomoadamantan-
2-one O-methyl oxime was prepared by reaction of 1-hydroxy-
homoadamantan-2-one19 with methoxylamine hydrochloride in
pyridine using a procedure completely analogous to the
preparation of 1-hydroxy-2-adamantanone O-methyl oxime
described above. This oxime derivative was converted to
mesylate 18, mp 138-139 °C (90% yield), using a procedure
completely analogous to the preparation of mesylate 16 (by
initial conversion to the methyl sulfinate ester followed by
m-chloroperbenzoic acid oxidation to the sulfonate ester): 1H
NMR (CDCl3) δ 3.907, (s, 3 H), 3.209 (s, 3 H), 2.649 (d, J )
4.2 Hz, 2 H), 2.40 (d, J ) 13.8 Hz, 2 H), 2.33 (m, 2 H), 2.19 (m,
1 H), 2.13 (br s, 2 H), 1.90 (m, 2 H), 1.67 (m, 1 H), 1.58 (m, 1
H), 1.510 (d, J ) 13.8 Hz, 2 H); 13C NMR (CDCl3) δ 160.64,
92.29, 61.67, 42.99, 40.78, 36.10, 34.31, 32.81, 27.78, 27.52.
Anal. Calcd for C13H21NO4S: C, 54.33; H, 7.37. Found: C,
54.32; H, 7.53.
P r ep a r a tion of 1-Hom oa d a m a n tyl Tr iflu or oa ceta te. A
solution of 100 mg of 1-hydroxyhomoadamantane20 and 110
mg of 2,6-lutidine in 8 mL of ether was cooled to 0 °C, and
177 mg of trifluoroacetic anhydride was added dropwise. The
mixture was stirred for 5 min at 0 °C, and water was then
added. The ether phase was washed with dilute HCl solution,
NaHCO3 solution, and saturated NaCl solution and then dried
over MgSO4. Solvent removal using a rotary evaporator left
155 mg of 1-homoadamantyl trifluoroacetate as a clear oil
which was used for kinetic studies without further purifica-
tion: 1H NMR (CDCl3) δ 2.34 (m, 4 H), 2.10 (m, 5 H), 1.88 (m,
2 H), 1.77 (m, 2 H), 1.55 (m, 4 H); 13C NMR (CDCl3) δ 155.98
(q, J ) 41 Hz), 114.51 (q, J ) 287 Hz), 92.76, 43.01, 37.13,
36.21, 35.14, 30.74, 28.96, 27.46; exact mass calcd for C13H17F3O2
262.1181, found 262.1135.
syn-Oxime: 1H NMR (CDCl3) δ 7.480 (s, 1 H), 7.163 (s, 2
H), 3.895 (s, 3 H), 2.257 (s, 6 H), 1.96 (br s, 1 H), 1.554 (s, 6
H); 13C NMR (CDCl3) δ 149.49, 147.17, 135.59, 129.57, 123.38,
72.28, 61.97, 31.64, 20.14; exact mass (FAB) calcd for C13H20
NO2 222.1497, found 222.1472.
-
Conversion of these oximes to the corresponding trifluoro-
acetates 9 and 10 was completely analogous to the previously
described procedure.2 These trifluoroacetates were stored in
ether solution. Removal of the ether solvent leads to eventual
decomposition at room temperature. CDCl3 solutions of 9 and
10 also decompose, and spectra were recorded by concentrating
the ether solutions used in the workup of 9 and 10. Trifluo-
roacetate 9: 1H NMR (Et2O) δ 8.31 (s, 1 H), 7.05 (s, 2 H), 3.90
(s, 3 H), 2.39 (s, 6 H), 1.83 (s, 6 H); 13C NMR (Et2O) δ 147.08,
143.74, 138.10, 129.52, 124.62, 87.17, 61.40, 27.72, 21.24.
Trifluoroacetate 10: 1H NMR (Et2O) δ 7.43 (s, 1 H), 7.02 (s, 2
H), 3.80 (s, 3 H), 2.21, (s, 6 H), 1.84 (s, 6 H); 13C NMR (Et2O)
δ 146.17, 136.18, 123.18, 116.45, 87.28, 61.30, 27.86, 19.74.
P r ep a r a tion of 1-Hyd r oxy-2-a d a m a n ta n on e O-Meth yl
Oxim e. A solution of 158 mg of 1-hydroxy-2-adamantanone18
in 6 mL of pyridine was stirred as 95 mg of methoxylamine
hydrochloride was added. After 20 h at room temperature,
the mixture was taken up into ether and the mixture was
washed with water. The organic extract was then washed with
dilute HCl solution and saturated NaCl solution and then
dried over MgSO4. The solvent was then removed using a
rotary evaporator leaving 157 mg (85% yield) of 1-hydroxy-2-
adamantanone O-methyl oxime, mp 47-48 °C, which solidified
on standing: 1H NMR (CDCl3) δ 3.851 (s, 3 H), 3.554, (br s, 1
H), 2.187 (br s, 2 H), 2.03 (d, J ) 11.7 Hz, 2 H), 1.90-1.65 (m,
8 H); 13C NMR (CDCl3) δ 164.47, 70.03, 61.55, 40.98, 36.82,
34.86, 30.41, 29.60. Anal. Calcd for C11H17NO2: C, 67.66; H,
8.78. Found: C, 67.80; H, 8.86.
P r ep a r a tion of Mesyla te 16. Mesylate 16 was prepared
by peracid oxidation of the sulfinate ester of 1-hydroxy-2-
adamantanone O-methyl oxime. A solution of 92 mg of
1-hydroxy-2-adamantanone O-methyl oxime and 95 mg of Et3N
in 4 mL of CH2Cl2 was cooled to -15 °C, and 70 mg of CH3-
SOCl was added. The mixture was then warmed to room
P r ep a r a tion of Mesyla te 20. 1-Hydroxybicyclo[3.3.1]-
nonan-2-one O-methyl oxime was prepared from 1-hydroxy-
bicyclo[3.3.1]nonan-2-one21 and methoxylamine hydrochloride
in pyridine using a procedure completely analogous to the
(19) Takeuchi, K.; Yoshida, M.; Nishida, M.; Kohama, A.; Kitagawa,
T. Synthesis 1991, 37.
(20) Nordlander, J . E.; J indal, S. P.; Schleyer, P. v. R.; Fort, R. C.,
J r.; Harper, J . J .; Nicholas, R. D. J . Am. Chem. Soc. 1966, 88, 4475.
(21) Takeuchi, K.; Ikai, K.; Yoshida, M.; Tsugeno, A. Tetrahedron
1988, 44, 5681.
(16) Nakayama, T. A.; Khorana, H. G. J . Org. Chem. 1990, 55, 4953.
(17) Creary, X.; Wang, Y.-X. J . Org. Chem. 1992, 57, 4761.
(18) (a) Curran, W. V.; Angier, R. B. Chem. Commun. 1967, 563.
(b) Schleyer, P. v. R.; Buss, V. J . Am. Chem. Soc. 1969, 91, 5880. (c)
Martin, J . C.; Ree, B. R. Ibid. 1969, 91, 5882.

Foiled Conjugation in R-Oximino Carbocations
J . Org. Chem., Vol. 61, No. 10, 1996 3489
preparation of 1-hydroxy-2-adamantanone O-methyl oxime
described above. Conversion to the mesylate 20, mp 65-66
°C (80% yield), was completely analogous to the preparations
of 16 and 18 (via the sulfinate ester): 1H NMR (CDCl3) δ 3.901
(s, 3 H), 3.099 (s, 3 H), 2.86 (m, 1 H), 2.726 (m, 1 H), 2.58 (m,
1 H), 2.37 (m, 1 H), 2.15 (m, 1 H), 1.98 (m, 2 H), 1.84 (m, 1 H),
1.71 (m, 1 H), 1.50 (m, 3 H), 1.38 (m, 1 H); 13C NMR (CDCl3)
δ 159.26, 88.27, 61.84, 41.17, 39.82, 37.89, 31.20, 29.16, 24.63,
22.38, 20.24. Anal. Calcd for C11H19NO4S: C, 50.56; H, 7.33.
Found: C, 50.57; H, 7.36.
× 10-4 M 2,6-lutidine; 272 nm) and 21 (methanol containing
2.5 × 10-4 M 2,6-lutidine; 270 nm) were also monitored by
UV spectroscopy. Rates of solvolyses of 16, 18, and 32 as well
as 1-homoadamantyl trifluoroacetate (in trifluoroethanol con-
taining 0.05 M 2,6-lutidine) and 20 (in methanol containing
1
0.05 M 2,6-lutidine) were monitored by H NMR spectroscopy
using the recently described method based on measurement
of the chemical shift of the methyl groups of the buffering base
2,6-lutidine.23
Com p u t a t ion a l St u d ies. Ab initio molecular orbital
calculations were performed using either the Gaussian 92 or
Gaussian 94 series of programs.12 All structures were char-
acterized as true minima via frequency calculations which
showed no negative frequencies.
P r ep a r a tion of Mesyla te 21. This mesylate was prepared
as previously described.7
P r ep a r a tion of Tr iflu or oa ceta te 32. This trifluoroac-
etate was prepared (90% yield) by reaction of exo-2-vinyl-endo-
2-norborneol in ether with trifluoroacetic anhydride and 2,6-
lutidine using the previously described procedure.15 The neat
trifluoroacetate 32 decomposed on prolonged standing at room
temperature: 1H NMR (CDCl3) δ 6.064 (d of d, J ) 17.4, 10.9
Hz, 1 H), 5.253 (d, J ) 17.4 Hz, 1 H), 5.235 (d, J ) 10.9 Hz, 1
H), 2.70 (m, 1 H), 2.33 (m, 1 H), 2.13 (d of d of d, J ) 14.1, 4.8,
2.7 Hz, 1 H), 1.74-1.42 (m, 5 H), 1.40-1.26 (m, 2 H); 13C NMR
(CDCl3) δ 156.33 (q, J ) 41.4 Hz), 138.95, 115.84, 114.56 (q, J
) 286.8 Hz), 92.69, 45.87, 42.67, 37.10, 36.06, 28.55, 22.11l
exact mass calcd for C11H13F3O2 234.0868, found 234.0883.
Kin etics P r oced u r es. Rates of reaction of the substrates
described in this paper were determined by either UV spec-
troscopy or 1H NMR spectroscopy. Rates of solvolyses of
trifluoroacetates 8 (245 nm), 9 (275 nm), and 10 (272 nm) in
ethanol were monitored by UV spectroscopy at the wave-
lengths given using previously described methods.22 Rates of
solvolyses of mesylate 17 (in trifluoroethanol containing 2.5
Ack n ow led gm en t is made to the National Science
Foundation for support of this research. We also thank
Professor Olaf Wiest for helpful discussions.
Su p p or tin g In for m a tion Ava ila ble: A table of HF/6-
31G* energies of optimized structures (1 page). This material
is contained in libraries on microfiche, immediately follows this
article in the microfilm version of the journal, and can be
ordered from the ACS; see any current masthead page for
ordering information.
J O960073U
(22) Creary, X.; Hatoum, H.; Barton, A.; Aldridge, T. E. J . Org.
Chem. 1992, 57, 1887.
(23) Creary, X.; J iang, Z. J . Org. Chem. 1994, 59, 5106.