RETUZINUI‘G LIBRARY #3759351" ‘7'
M

 

 

22¢ per day per itu;

Place in bcuk return ta rt:.;e
charge from circa1;51:n recards

 

 

 

 

MULTINUCLEAR NMR STUDY ON THE
HIGHLY BRIDGED ARYL SUBSTITUTED
COATES CATION
AND
SIMILAR STUDIES ON THE

CLASSICAL ARYL SUBSTITUTED
2-ADAMANTYL AND 7-NORBORNYL CATIONS

By

Thomas P. Clausen

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Chemistry

1980

ABSTRACT
MULTINUCLEAR NMR STUDY ON THE
HIGHLY BRIDGED ARYL SUBSTITUTED
COATES CATION
AND
SIMILAR STUDIES ON THE

CLASSICAL ARYL SUBSTITUTED
2-ADAMANTYL AND 7-NORBORNYL CATIONS

By

Thomas P. Clausen

Following the work of Botto and Chambers, three series of aryl
'substituted carbocations were prepared and their NMR spectra taken.

The first series studied was the 9-ary1-9-pentacyclo[A.3.0.0?’40?’80§’Z)
nonyl cation; the only system that all in the field agree to be bridged.
The results from the study of it are twofold:

1) The NMR of aryl substituted bridged ions is in sharp
contrast to those of classical ions. The degree of
contrast is difficult to predict but qualitative trends
are readily apparent.

2) The perturbation of placing an aryl substituent on a
bridged system tends to stabilize the "classical" struc-
ture much more than the bridged structure. Thus weakly
bridged systems may easily become classical by the place-
ment of even very electron demanding aryl groups.

The second system studied was the 2-aryl-2-adamantyl cation.

It was found that the perturbation of placing an aryl group on the weakly-

bridged 2-adamanty1 cation produces a classical cation. Hence a

Thomas P. Clausen

lower limit for the amount of bridging that is detectable by this probe
has been established.

The third system studied was the 7-aryl-7—norbornyl cation with
the hope of inferring some unusual behavior for the parent cation. No
such behavior was found, however, and hence, if the geometry of the
7-norborny1 cation is deformed, then the energy difference between it
and its classical structure must be too small to measure by the probe

used.

-TABLE OF CONTENTS

Page
INTRODUCTION 0 O O 0 O O O O O O O O O O O O O O O O O O O O O O 1
Review of the Coates cation. . . . . . . . . . . . . . . 14
Review of the 2-adamanty1 cation . . . . . . . . . . . . 18
Review of the 7-norborny1 cation . . . . . . . . . . . . 26
RESULTS. 0 O 0 O O O O O O O O O O O O O O O O O O O O O O O O I 31
DISCUSSION 0 O O .0 O O O O O O O O O O O O O O O O O O O O O O O 41
Coates system. . . . . . . . . . . . . . . . . . . . . . 41
Z-Adamantyl SYStem o o o o o o o o o o o o o o o o o o o 49
7-N0rb0rny1 SYStem o o o o o o o o o o o o o o o o o o o 53
EXPERIMENTAL O O O O O O O O O O O O O O O O O O O I O O O I I O 56
Instrumentation. . . . . . . . . . . . . . . . . . . . . 56
Carbocation precursor formation. . . . . . . . . . . . . 56
Carbocation formation. . . . . . . . . . . . . . . . . . S7
Quench study on (83) . . . . . . . . . . . . . . . . . . 59

Measurement of the dependence of the 13C NMR of (27a) on
acid concentration . . . . . . . . . . . . . . . 59

Physical and spectral data of the carbocation

precursors . . . . . . . . . . . . . . . . . . . 60
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . 65

ii

LIST OF TABLES

TABLE PAGE
1. NMR chemical shifts of (68) and (69) . . . . . .I. . . . . 25
2. 1H NMR chemical shifts downfield from ms of (26). . . . . 32
3. 13C NMR chemical shifts downfield from TMS of (26) . . . . 33
4. 1H NMR chemical shifts downfield from ms of (27). . . . . 34
5. 13C NMR chemical shifts downfield from TMS of (27) . . . . 35
6. Dependence of 13C NMR chemical shifts of (27a) with re-

spect to acid concentration. . . . . . . . . . . . . . . . 37
7. 1H NMR chemical shifts downfield from TMS of (28) and

(83) . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
8. 13C NMR chemical shifts downfield from TMS of (28) and

(83) . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

9. Comparison of the 1H NMR of the ortho hydrogens between
(26), (27) and (28). . . . . . . . . . . . . . . . . . . . 45

iii

FIGURE

10.

ll.

12.

13.

14.

LIST OF FIGURES

Correlation of the chemical shift of (15) with:’+. .
Correlation of the chemical shift of (16) with ffil .
Correlation of the chemical shift of (17) with 0+.

Plot of the chemical shifts of the cationic center of
(15) VS. (19). o o o o o o o o e o o o o o o o 0

Plot of the chemical shifts of the cationic Center of
(19) V8. (22). o o o o o o o o o o o o o o o o o 0

Plot of the chemical shifts of the cationic center of
(25) vs. (20). . . . . . . . . . . . . . . . .

Calculated geometry of the 7-norbornyl cation. . . .

13C NMR chemical shifts of the cationic carbon of (26)
VS. (19) o o o ., o o o o o o o o o o o o o o o o o o o

13C NMR chemical shifts of the cationic carbon of (84)
VS. (19) o o o o o o o o o o o o o o o o o o o o o o o

A qualitative energy diagram for the bridge flipping
of (45). . . . . . . . . . . . . . . . . . . . . . .

A modification of figure 10 stressing the higher energy
of aryl substituted bridged cations compared to classi-

cal cations. . . . . . . . . . . . . . .

Molecular orbital diagram showing the interaction of
aryl groups with a bridged system. . . .

13C NMR chemical shifts of the cationic carbon of (27)
vs. (19) . . . . . . . . . . . . . . .

1 H NMR chemical shift of the nonequivalent geminal pro-
tons in (27) plotted against each other. .

iv

Page

10

11

3O

42

43

46

47

48

SO

51

LIST OF FIGURES--continued
FIGURE Page

13C NMR chemical shifts of the cationic carbon of (28)
vs. (19). . . . . . . . . . . . . . . . . . . . . . . . 54

15.

16. Bond polarization of the endo hydrogens of (28) by
through space interaction with the cationic center. . . 55

.INTRODUCTION

In 1939, Wilson1 made the suggestion that the proposed intermed-
iates (3) and (4) in the solvolysis of camphene hydrochloride (1) might
exist as a resonance hybrid (5) rather than separate entities. In the
early 1960's, structures of other cations showing similar delocaliza-
tion to that of (S) were proposed, the most famous being the 2-norbornyl
(6)2 and the cyclopropylcarbinyl (7)3 cations. Soon, practically every
carbocation known was reevaluated in terms of delocalized sigma bonds
and was given the designation of a "nonclassical cation". It was not
until the late 1970's, however, when the term "nonclassical cation"

was defined.48

cu
on
(I) . (2)

(3) (4)

 

 

(6) . m

In the early 1960's, when H.C. Brown5 and others challenged the
issue of nonclassical cations, an immense amount of extremely valuable
work was undertaken to prove or disprove the existence of these ions.
Possibly no other area of organic chemistry has had the time and scrutiny
of so many chemists as this question concerning the carbocation
geometry. Although even after a quarter of a century, many of the early
questions raised are still the objects of enthusiastic debate (the 2-
norbornyl and cyclopropylcarbinyl cationic structures for instance),
there is no doubt that these questions have inspired a great deal of
advancement in organic chemistry.

At present there can be no doubt as to the possibility of nonclas-
sical cations existing. The threercenter, two-electron bonds foUnd in
many boranes strongly indicate that carbocations (which are isoelec-
tronic with the boranes) may also enjoy the same type of bonding. Recen-
tly, structures (8) - (11) have gained wide approval over their classical
counter-parts.6'9 Still, however, the approval of structures (8) - (11)
is more than balanced by the rejection of a great many proposed nonclassi-

cal structures4b such as (12) - (14).

 

+
9H3

(I 2) (I 3) (I 4)

There remains, however, a large number of carbocations whose

 

structures have yet to be established as classical or nonclassical. Of
the many methods used to prove the structure of carbocations, NMR seems
to have been the most used. The main disadvantage of NMR studies has
been the lack of suitable model compounds. Thanks to the efforts of

Olah and others,10

we now have a workable supply of NMR data on proven

classical cations. The limited number of cations known to be bridged,

however, offers us little in the way of what is "normal" for the NMR of
. nonclassical cations. The major theme of this thesis is to enhance our
knowledge of what is normal for classical and nonclassical cations.

The method used to examine cations in this project was to correlate
chemical shifts of variousaC-aryl cations with those of known classical
cations as a function of electron demand. Olah has shown that a plot of
the chemical shift of the cationic center of cations (15) - (l7) corre-
lates in a somewhat linear fashion wither+ (see figures 1-3).11 From the
large standard deviations, however, it became clear that the effect of
different aryl groups on the chemical shifts was not just a function of
electron demand, and that a new set of<rC+ values might be established for
NMR studies in super acid media.12

Upon reexamination of Olah's results, it was found that excellent
correlations could be obtained by plotting the chemical shift of one
13

cationic center with that of the cationic center in another system.

An example of such a correlation is given in figure 4. In all cases, a

139-. 1. con-elation or the chemical shift of (15) with 0’

200 - +

 

300

 

 

-O.8 -O.4 O +0.4 +5.8

Irina-o 2. Correlation at tho clinical shift of (16) with r”

V fiv“V“V‘V“VV‘V

250

5 ”CO

300

 

 

l l l
—O,8

+Oc4 40.8
0.?

200

8 ”(3*

250

Figure 3. Correlation of the chemical shift of (17) with (1+

 

 

 

 

-O.8 -O.l. (l +0.4 +0.8

AI133 of (19)

330

3.10

290

270

259

231

 

Eigggg L. 219; gr thf c?egigg% shifts of the cationic
center of 15 vs. 1

 

 

22% 240 26) 2‘0 300

S 13c of (15)

Ar Ar, . Ar
Al.\ /~lk/ k)...

('5o—g) ' (Bo—d) - (We—e)

a) 4-OCH3 e) 4-F
b) 4-CH3 f) 4-Br
c) phenyl g) 3-F
d) 4-CF3

.straight line with a negligible standard deviation was obtained with
Lclose to a unit slope. It appears that the factors leading to the
deviations of the correlation of chemical shifts versus electron demand
could be cancelled in this manner.

14 and Chambers13

Botto continued this line of reasoning and found
similar linear relationships with the chemical shifts of compounds (18) -
(21). Compounds (15) - (17) also showed linear relationships with com—
pounds (18) - (21) and in all cases it was surprising to find slopes very
near unity. It appeared then, that some measure for detecting "normal"

cations was available. The prime question was whether this approach would

yield sufficiently different results for "unusual" cations.

Ar AV
6%” 4%”
(I8) (I 9) (20) (2 n

The first attempt to find unusual behavior in cations with increasing

electron demand (via NMR correlations of chemical shifts) was done by
Chambers13 with the 1-aryl-l-pheny1-l-ethyl cation, (ZZa—k). ~When plot-
ting the chemical shift of the cationic center in (22) with the cationic
center of any of the compounds (15) - (21), two straight lines were
obtained which intersected when the aryl group was phenyl (figure 5).
The rationale for this behavior rests on the fact that only one aromatic
ring may be conjugated with the cationic center at any given time due to
steric effects. Thus it is reasonable for the aryl group to be conju-
gated with the cationic center for those cases where it is less electron
demanding than phenyl (cases 22a-f). When the phenyl is more electron
donating than the aryl group, the aryl group would be twisted out of con-
jugation (cases 22h-k). It was thus established that by examining the
. chemical shifts of a series of aryl substituted cations, changes in geome-
try may be detected.

EJL~iE:/4£\f

l
(Sths

(22040

a) p-OCH g) phenyl

b) 3,4-(8h3)2 h) m-F

c) p-CH3 i) m-Cl

d) p-F j) p-CF

e) p-Cl k) 3,5- CF3)2
f) p—Br

Botto14

found breaks in the plots of the chemical shifts of
compounds (23) - (25) when plotted against any of the classical compounds

(15) - (21) (see figure 6 for a superb example). These breaks were inter-

5'13C+ of (19)

30(

220

26

24'

F1

 

9 Plot of the
center of

13+
5

10

 

he 1 shi ts of the cationi
vs (22

 

l l l

on c:

‘w‘ .

C of (22)

Figure 6.14 Plots of the chemical shifts of the cationic center or

(25) vs. (20).

11

F

 

 

'5 130* of (25)
3-3 a N
0 a
CS . —f .1 . . .
b \ ' I ' I
cl) 0
(I)
O
(“I
18 ..
N
N I.
\J‘I
O
‘a
CC;
+
8. N U 0
A ‘03 ' I
£3 (7)
V I
04
(‘J
8: h
151
(N
m -
33 ~ (I) 13
l l
2;; (T)
.n
04-” 040
N N //
E? ” <3

12

 

(23)

preted in the case of (22), as an abrupt change in their geometries. With
the highly stabilizing aryl groups, it was assumed that the systems need
no further stabilization from other portions of the molecule, while with
the electron withdrawing aryl groups, the systems would begin seeking
further stabilization elsewhere. The change in the resultant electronic
state would likely leave the aryl group in a less optimum position to
stabilize the cation by conjugation, thus requiring the system to seek
out still further stabilization. This feedback effect would continue until
g a new optimized geometry resulted. Thus the abrupt change in slope for
these plots seems reasonable.

The work in this dissertation is a continuation of the work done
by Botto and Chambers. The three Series studied are the aryl substituted
9-pentacyclo[4.3.0.0.2’40.3’80.5’7Jnony1 (26), the 2-adamanty1 (27) and

the 7—norborny1 (28) cations.

Ar
F. +

(26) (27) (28)

 

13

These systems were chosen for the following reasons:

1)

2)

3)

Although Botto's work suggests that the systems (23)-(25) are
nonclassical, the arguments proposed have not been accepted by all

in the field. Chemical shifts of cations known to be bridged are
required in order to extrapolate into the more questionable systems
such as the 2—norbornyl cation (6). The only cation that is accepted
as being bridged by all in the fieldls is the 9-pentacyclo
[4.3.0.Og’40?’80§’ZJnonyl cation (10)8 which would thus seem a
logical start for the characterization of nonclassical cations.

The 2-adamanty1 cation (29) has been postulated to possess a

barely detectable amount of bridging.l6’17

It is hoped that by corre-
lating the chemical shift of the cationic carbons of the 2-aryl-2—
adamantyl cations (27) with the chemical shifts of established
classical cations, a qualitative measure of how sensitive this probe

is can be determined. Also, any abnormal deviations observed would

lend support to the proposed bridged nature of (29).

Iii

(29)

No studies on the 7-norborny1 cation (30) concerning its struc-

ture in stable ion conditions have appeared in the literature.

14

It is hoped that the 7—aryl-7-norbornyl cations (28) will be stable
enough to be observed under super acid conditions. If such is the
case, perhaps some direct evidence concerning the electronic struc-

ture of the 7-norbornvl cation can be obtained.

.9

(30)

9-pentacycloE4.3.0.02’409’8OS’71nony1 cation. (10)

Perhaps the cation most accepted as being nonclassical is the
9-pentacyclo[4.3.0.02’409’80§’Z]nony1 cation, better known as the Coates
cation.15 Other cations have been made which resemble the Coates cation.
I Most notable are the 3—bicylo [3.1.0.] hexyl (31)18 and the endo-8—

tricycLoE3.2.1.0g’éJocty1 (32)19 cations. Although both of these

systems have been proposed to be nonclassical,l8’19 such proposals have
met resistance.20
+ +
N Or N
(31)

C)Y

(32)

 

The Coates cation on the other hand, offers a geometry which is

not subject to the classical interpretations imposed on (31?0

15

1 13

H NMR and . C NMR Show a system possessing a three fold symmetry.88’b

Both its
Rapid equilibration does not explain this symmetry because such equilibra-
tion would cause all NMR signals to coalesce.

Deuterium labeling studies in the solvolysis of 9-pentacyclol4.3.
0.0g’40?’80§’Z}nonyl-p-nitrobenzoates, (33) and (35), are only consistent

83 These

with the formation of the nonclassical intermediate (10).
studies further substantiate that only one cyclopropane ring may bridge
with the cationic center at any given time (notably the ring anti to

the nitrobenzoate). This observation is precedented in that other systems

are known that allow participation with only one functionality at 3

21,22 Zla-c

time. The most noteworthy examples are found in the solvolysis
of 7-chloronortricyc1ane (37) and 7—chloronorbornadiene (41).22’21C
In both of these systems, the relative rate of solvolysis is about the

same with two functionalities present as is with only one functionality,

but these rates are much larger than those systems with no functionality.

H N32 H OH

is.

“E in

<33) ' (34)

D

D ON 82 OH

e

(35) (36)

\

l
23'

16

H; ;C) H; SC) H Cl

(38) (39) (37)
Relative Rate 10
of Solvolysis (30-40)10 1 1010
(39) (40) (41)
Relative Rate 14
of Solvolysis 1 1011 10

A nonclassical structure for (10) is also supported by comparison
of the relative rate of solvolysis of its 9-substituted nitrobenzoate
(42), with that of the corresponding 7-substituted norbornane (43) as
a function of electron demand.8d (42) was found to be as high as 1014
times as reactive as (43). This large rate enhancement of (42) is best
explained by invoking bridging in the transition state of the solvolysis.

d8d that these rate differences were much less when

It was further note
the comparison was made with 7—substituted norbornenes (44) (where
participation is significant). Although no "break" was found in the

Hammetthplot for (42), thef’+ value of -2.05 is so low in magnitude

when compared to —5.27 for (43), that some charge delocalization seems

17

unavoidable.

R (3N5 R OPNB

 

(42) (43) (44)

The stability of the trishomocyclopropylium ion in (10) was
estimated by obtaining the activation energy required to effect a bridge

flip as shown in the following scheme.8b

Scheme I

 

(46) (46)

With R = H, the activation energy was sufficiently high so as to
be unmeasurable (an experimental value of.AC was not cited). With R =
methyl, however,¢§G_8°C = 13.0 Kcal/mole in contrast to the unsubstituted
cation; when R = phenyl, the bridge flipping occurs rapidly even at
-75°C (AG is less than 9.6 Kcal/mole). These results are in agreement
with those of the 7-substituted norbornadienyl cation, (46).23 Here, with
R again being hydrogen, methyl and phenyl, the AG values are greater than.

19.6 Kcal/mole (0°C), equal to 12.4 Kcal/mole (-l4°C) and less than 7.6

18

Kcal/mole (-100°C) respectively.

Finally, JorgensonBe has used MINDO/3 calculations to substantiate,
on theoretical grounds, the presence of the trishomocyclopropylium ion
in (10). His calculations substantiate the presence of C3v symmetry in
(10), which is consistent with its NMR spectrum. Also, his calculations
confirmed the activation energies found for the bridge flipping of (45).

With R = hydrogen, a calculated AC of 22.1 Kcal/mole was obtained, while

for R methyl, AG = 8.8 Kcal/mole. No calculation was done, however,

calc

for R phenyl.

2-adamanty1 cation (29)

 

The studies of the 2-adamantyl cations can be subdivided into two
major areas of investigation:
1) Rearrangements of the 2-adamantyl cations.
2) The possibility of the 2-adamantyl cations existing as
bridged species.

Rearrangements of the 2-adamanty1 cation

 

Unfortunately, observations of the 2-adamantyl cation (29) under
stable ion conditions have not been achieved due to decomposition. 24’25
Nonetheless, theory concerning its rearrangements agrees with a vast
amount of indirect evidence that requires its existence as a fleeting
intermediate during solvolysis reactions. 26,27
Perhaps one of the most striking properties of the 2-adamanty1
cation is the lack of 1,2 hydride shifts which are common to most cationic

28’29 This is expected and is readily €XP131“9d by the orthogo-

species.
nality between the migrating hydrogen and the ”empty p orbital" on the

cationic center. Indeed, the equilibration of the 2-adamantyl cation

(29) to its more stable isomer, the l-adamantyl cation,2‘9 has been shown

19

to involve intermolecular hydride shifts,28’3O or in some 30 2-adamantyl
cases,29’31 a series of Wagner-Meerwein shifts. 2—Methy1adamantane

(47) is an example of the latter case and rearranges cleanly to l-methyl—

29’31 A mechanism is given

14

adamantane (52) under Lewis acid conditions.
in scheme II and is consistent with labeling studies in which 2- C-2-
methyladamantane (47) is treated with a Lewis acid. The 14C label

was found on the C-1 position bearing the methyl as predicted by the mech-

ansim.

Scheme 11
+ “3
IV

(47)

 

(49)

CH3

 

IV

(52) (51) (50)
Although 2-adamantyl cations do not undergo 1,2-hydride shifts,

32

1,3—hydride shifts appear to be facile. Such shifts are well documented

in cyclohexane derivatives.33 Only one system, however, has clearly demon-
strated 1,3-hydride shifts in the 2-adamantyl series in preference to

. . 34

Intermolecular hydride shifts.

Further rearrangements of 2-adamantyl cations will be discussed

with regard to possible nonclassical intermediates.

20

Nonclassical behavior of the 2-Adamanty1 cation

The 2-adamantyl cation possesses a rigid geometry that may show
some observable bridging. Such bridging (53), if it exists, would be
unsymmetrical and would thUs be expected to favor the less strained

contributing structure (29):
~ 4’

lhl- IOTA”

(53) (g9) ,(54)
Major Minor

The major studies which attempt to determine whether the 2-ada-

mantyl cation is classical or not involve solvolysis of 4—substituted-

protoadamantanes, (55) and (56).16.l7

H} . )4
,9 ,6

(55) (56)

Exo ‘ Endo

In solvolysis reactions, (55) was found to be about 104 times

16

more reactive than its endo isomer (56). The high reactivity of the

exo isomer is indicative of anchimeric assistance. Further insight into-

the solvolysis of 4-substituted-protoadamantanes was found by labeling

16

the 4-position with deuterium. With endo-4—substituted-4-d-proto-

21

adamantane (59) solvolysis gives the 2-Substituted adamantane (60),

with deuterium completely scrambled in the l and 2 positions. This
suggests a rapid degenerate 4-protoadamantyl---) 2—protoadamanty1 cation
rearrangement followed by a slower capture by solvent (see scheme III).
The exo isomer, however, yields only l-d-2-substituted adamantane (S8)
implying that the above degenerate rearrangement is slow compared to
solvent capture of the cation. These results are best interpreted by
anchimeric assistance in the transition state in going from (S7) to (58),
but say little of the nature of the intermediate involved.

Evidence for bridging in the intermediates formed from the solvol—
ysis of 2-adamantyl or 4-protoadamanty1 systems exists, but it is not con-
clusive. Solvolysis of optically active 2-adamantyl derivatives (61) and
(62), for instance, have yielded as high as 84% retention of configura-

tion in the products.35

RETENTION
0T5

CH3

(62)

 

 

 

 

 

P __+ f ; __, OH
,9 ' ,9 [9
T50 ,0 D D
123.1:
,9 z;
”D Slow H D
,9

23

The amount of retention as well as the product distribution in
the solvolysis of 2-adamanty1 derivatives may be made more pronounced by
adding substituents that would either stabilize the minor resonance form
(54) or destabilize the major resonance form (29) thereby increasing the
amount of bridging. The results of studies on such systems are shown

on the following page and are in accord with a bridged intermediate.l7a’26'27

17b it was found that adding successive methyl

In a related study,
groups adjacent to the 2-position of 2-adamantyl tosylates produces
an additive rather than a multiplicative effect towards the rate of sol-
volysis. Again, the data are given on the following page and are in
accord with a bridged system.

Although, as mentioned earlier, the 2-adamantyl cation has not
been observed under stable ion conditions, the tetramethyl-Z-adamantyl

36,37

cation (68) is stable enough to be analyzed by NMR. Presumably

this is a consequence of the inability of (68) to undergo the intermol-
ecular hydride shifts that are facile in the 2-adamantyl cation. The

13Cbnfllspectrum of (68) is striking when compared to that of the similar

30 cation (69) (see table 1).36 The higher field shift of the cationic

carbon in (68), and the lower field shifts for carbons l and 3 relative
to those of (69), suggest extensive charge dispersal from C—2 to C-1 and
C—3. These discrepancies cannot be reconciled by invoking a rapid equili—

bration of (68) with its 4-protoadamanty1 isomer (70) if "reasonable”

36,37 1

chemical shifts for (70) are assumed. Furthermore, the H NMR of

(68) shows the H-2 absorption at 5.10 ppm, a value over 8 ppm upfield

37

from what is expected from model compounds. These results strongly favor

a set of bridged structures (71) that are in rapid equilibration, which

allows for the observed symmetry of (68). (see scheme 1V).36’37

24

  
 

(OR 9R
R t
+ "’ *
1’ R

 

 

 

(63) (64) (65) (66)
SUBSTRATE %6L} %65 %66
H 99.6 .4 o
. CH3 73 27 0
'CN 6 2 O 38
RI
0T5
(. ——> SOLVOLYSIS
“ R2 PRODUCTS
(67)
SUBSTRATE RELATIVE RATE
RT = R2” H l
Rle, R2=CH3 14—21

R,= R2= CH3 38—70

25

 

 

 

 

 

 

 

 

(70)
Table 1 NMR chemical shifts (PPM from TMS) of (68) and (69).
Cation] CL?) '62 54,3310 C57 TCG
‘36 l38.6 92.3 5).o 26.2 46.7
(68)
'H am 2.53 l.68
(69, 'f’c 68.3 329.1 63.? 9.89....
H 2.68 2.08

 

 

 

 

 

 

 

 

 

Scheme IV

 

26

All of the evidence presented strongly suggests a bridged struc-
ture for 2D 2-adamanty1 cations. However, it is prudent to reemphasize
that if such bridging is in fact present, it is unsymmetrical, and hence,
probably weak. For instance, the excellent correlation of 2-adamantyl
tosylate with regard to the Foote-Schleyer equation38 is consistent with
little or no bridging of the 2-adamantyl cation during solvolysis.

7-norbornyl cation (30)

 

As a result of the instability of the 7-norborny1 cation, studies
associated with it and its precursors have been meager. In addition to
the scant amount of data concerning it, interpretations of such data have
been widely divergent, leading to three structures, (30),38 (72)39’40-42

and (73),“3 proposed for this elusive cation. The arguments which support

these structures will be briefly discussed below.

   

(30) ' (72) (73)

The first representation of the 7—norbornyl cation is that shown

by (30); i.e., a classical cation. Correlation with the Foote-Schleyer

1c

. 38 obs _ _ , ca
equation (log kre — 7.00, log krel

1 = ’7-04) is the only direct evi-

dence suggesting such a structure. However, from the scattering of data
points about the Foote-Schleyer equation, a rate enhancement by a factor.
of ten could be easily undetected implying the possibility of a small

amount of anchimeric assistance in the solvolysis of 7-norborny1 tosylates.

Furthermore, as stated earlier, the absence of any anchimeric assistance

27

in the transition state of such a solvolysis implies little about the
nature of the intermediate.

Perhaps the strongest case for (30) is the lack of sufficient
evidence to prove the existence of the more controversial structures,
(72) and (73). Nonetheless, evidence does exist for (72) and (73) and
is given below along with some classical interpretations.

Structure (72) was first postulated by Winstein and co—workers39
in 1958. Winstein's conclusions were based primarily on the products

obtained from the acetolySis of 7-norbornyl brosylate (74) and its isomer,
2—bicyclo[8.2.0]heptyl brosylate (75). The mechanism proposed invoked

the intermediate (72) to account for the absence of any of the endo isomer
of (76); i.e., (79). Similar arguments have been forwarded for the exclu-
sive formation of exo-2-norborny1 products in the solvolysis of endo and

44 Since this work, however, Brown45 has

exo-2-norbornyl derivatives.
refuted the need to impose nonclassical intermediates on the basis of

the exo/endo ratio of product distributions. To be explicit, regardless

of whether or not the intermediate is bridged, the exo face is the less
hindered and hence should capture the solvent preferentially.

Later, Gassman and others found additional evidence supporting
Winstein's proposed intermediate.l‘0'42 Gassman found that the acetolysis
of exo, exo-2,3-dideuterio-anti-tosyloxynorbornane (80) and the syn
isomer, yields the corresponding acetate (82) with about 90% retention.
From this, it is clear that the free 7-norborny1 cation (30) plays,
at best, a minor role in the solvolysis. Two possibilities were considered
or the intermediate, one being a nonclassical intermediate such as

(dz-72) and the other being a tight ion pair (81) that rapidly collapses

to (82).

28

 

CBS

(75). (72) . (74)
l H OAc
(77) (78)

0.46 ,

(76)
I‘llélé
f1

(79)

The presence of about 10% inversion in the product is most readily
explained by conversion of either (dz-72) or (81) to the free classical
cation (31) followed by solvent capture. An SN2 type mechanism is un-

likely because the addition of acetate ion has no observable effect on

41

the percent retention of configuration found in the product. The pos-

sibility of sulfur—oxygen bond breakage instead of carbon-oxygen bond clea-

18

vage of (80) to explain the amount of retention was excluded by an O-

labeling experiment.46

29

H 0T5 . ' (81)

(8°) (32)

 

(dz-72)

Although structure (73) could also be used to explain Gassman's
observations, it was not considered. The only mention of (73) was in a

paper by Dewar43

describing a MINDO/Z calculation to obtain an optimum
geometry for the 7-norbornyl cation. His calculations assumed a plane
of symmetry passing through atom 7 and bisecting the 2,3 and 5,6 bonds
(thereby excluding the possibility of (72)). The results of his calcu-
lations clearly favor an unsymmetrical cation as diagrammed on the
following page.

It is evident that very little can be conclusively stated
concerning the exact geometry of the 7-norbornyl cation. Perhaps Brown's

arguments should be applied here and the classical structure (30)

accepted until more convincing evidence for (72) and (73) can be forwarded.

30

Figure 7 Calculated geometry of the 7-norbornyl cation

H

.7 l76.9°

 

 

RESULTS

The details for the generation and observation of the carbocations
are given in the experimental section. The results of these observations
are briefly stated below.

The 1H NMR chemical shifts of thecx-aryl Coates cations (26a—e)
are presented in tables 2 and 3 respectively. The spectra were unaffected
from -100°C to about -500C at which time decomposition occurred. The
chemical shifts of the 9-pheny1 Coates cation (26b) have been reported

8b and are included in the tables.

earlier
Several unsuccessful attempts were made to generate the 9-(p-anisyl)
Coates cation (26f) and the 9-(3,5-dichloro-4-N,N-dimethylammoniophenyl)
Coates cation (26g). In the case of the electron rich cation (26f), the
complexity of the spectra observed suggested decomposition. For (26g),

however, rearrangement to two species predominated even at -100°C. No

attempts were made to elucidate the structures of the rearranged products.

Cl +
CH3 4 NH(CH3)2
+ Cl

1H NMR and 13C NMR spectra of the a-aryl adamantyl cations

The
(27a-i) are given in tables 4 and 5 respectively. Except for (27i), all of

these systems were stable up towards the boiling point of the solvent (+1OOC).

31

Amnmmv aow.~(
Annmav H~¢.n)n<

Anm h.wuhmvv «ow.n)
Ana m.wuhmvv mom.nlu<

Asmmmv mow.e(mo~.~)
Aux w.ouhmuv mmm.~Iu¢
Am©.5vl
ABV omc.hlm¢m.nlu<
Aux m.nahmwv mnw.nl

Aha n.eusuev oma.mmu<
euc.uu mo

.omHz

Nmo.m

owm.m

wmn.m

Am~.~v
oqm.m

naa.N

 

no «0

.nm monoummmu Eoum coxmu mum mammnuaoume a“ enemaszm

omo.~

nmq.m

mam.m

Ame.mv
mmm.m

oew.m

5 mu m No

wmm.q

omn.q

mHo.m

Aoo.ev
moo.¢

mmooINAm

movum.m

c o m

m UI mole

e o

m Ulmlm

mmoo

emeoummone

 

msouw Hmu<

Aoomv

Acemv

Aoemv

mAnomv

Amway

mcsomaoo

 

.Aemv mo Amze seem ammo humane amasseee

mzz ma N wanmw

32

.nm mocmumwou Boom coxmu mum wfimozucoume CH mumn552m

 

me.oe . eo.oe me.em Ha.me mmeoumflmeov-m.m AceNe
ee.oe . om.em He.em mm.oh . emeoumeoue Aeeme
oe.oe mm.mm He.em em.ee . emeoue-m Aeewv
AeH.oev ANm.AmV Amo.emv Ame.ewv
Ne.oe . om.em me.em wo.mw mmco exeemv
me.--mmo he.oe mm.em hm.wm om.eoH emcoummo-e Acemv
dale)». Nola i .mJIHIo. mm anew TC... 393

 

.Aemv mo Amze acne zeev humane sconceee mzz o m «Home

ma

33

Table 4 JTINMR chemical shifts (PPM from TMS) of (27)

 

 

Compound Aryl group C1,3 04,8, C5,7 Misc.
9,10a & C6
(27a) 4-CH30-06H4 4.220 2.700 2.288 CH3— 4.372
2.363 Ar-8.740(2H;d;j=9.5Hz)
-7.424(2H;d;j=9.5hz)
(27b) A-CH3-C6H4 4.438 2.811 2.311 CH3-2.842
2.469 Ar-8.698(2H;d;J=7.25Hz)
-7.838(2H;d;J=7.25hz)
(27c) 4—F-C6H4 4.459 2.888 2.333 Ar-7.680(2H;t;J=8.5hz)
2.546 -8.912-8.970(2H;m)
(27d) C6H5 4.571 2.939 2.355 Ar-8.811(2H;d;J=8.3Hz)
2.590 —8.668(lH;t;J=8.3Hz)
-8.022(2H;t;J=8.3Hz)
(27e) 3-Br-C6H4 4.576b 3.001 2.365 Ar—8.879(1H;s)
4.528 2.643 -8.748(1H;d;J=7.8Hz)

-8.646(1H;d;J=7.8Hz)
-7.9l4(1H;t;J=8.0Hz)

(27f) 4-CF3-C6H4 4.685 3.099 2.390 Ar-8.876(2H;d;J=8.3Hz)
2.726 -8.221(2H;d;J=8.3Hz)
(27g) 3,5-(CF3)2-C6H3 4.745 3.202 2.430 Ar-9.l32(lH;s)
2.827 8.988(2H;s)
(27h) 3,5-C12-4- 4.681 3.286 2.427 CH3-3.774(d;J=4.5Hz)
N(CH3)2-C6H2 2.917 Ar-8.773(2H;s)
(27i) 3,5-C12-4- 4.819 3.482 2.496 Ar-8.610(2H;s)
CN-C6H2 3.061 2.459

 

8All of these peaks appeared as doublet with a coupling constant varying
between 11.0 and 13.0 hz.

bThe unsymmetrical aryl group causes these two hydrogens to become non-
equivalent.

34

 

Nee (Misseoz
m.am m.sm s.em e.Ne e.os~ (a- Ho-m.m Aesmv
N m mmeo
s.om o.sm ~.em m.mm N.emm - A movnm.m Amsmv
m.om H.sm s.~m m.em c.mm~ emco-mmo-e Adamo
m.om m.em m.Hm H.em H.5AN emeouamum Amsmv
H.oM m.em m.se s.am m.ss~ memo AeeNV
o.om o.am e.me ~.Hm N.me~ ameoum-e Acsmv
s.e~-mmo m.m~ m.em m.me «.se m.oe~ emeoummoue Aesmo
c.8mummo m.m~ m.em m.me m.ae 8.4mm emeo-0mmo(e Assay
.emaz .mw same oH.s.m.eo maso mm. .mmcam Haws ecsowan

 

.ANNV mo Amze comm ammo mousse accesses mzz 02 m mamas

35

36

The latter system, however, rearranged and consequently, no 13

C NMR
'could be obtained for it even at -110°C. To determine if this rearrange-
ment was unique for the adamantyl system, the corresponding cyclopentyl
cation (19) was generated and was found to be a complex mixture of
products. No definitive spectra could be obtained for it.

The 2-(p-anisy1)adamanty1 cation showed a peculiarity in its
13C NMR. All peaks showed a dependence on the concentration of acid used.
This dependence is shown semiquatitatively in table 6.

Tables 7 and 8 show the 1H NMR and 13

C NMR results respectively
for the 7-aryl-7-norborny1 cations (28a-f). Olah47 has obtained chemical
shifts for several of these systems and his results are included in table
8. The peak assignments for the exo and endo hydrogens were deduced by

generating the 7-(p-fluorophenyl)norbornyl cation (28c) with deuterium

placed on the exo face of the 2 and 3 positions.

 

(exo, exo-2, 3—D2-2aé)
The 7-(3,5-dichloro-4-N,N—dimethylammoniophenyl)-7-norbornyl
cation (28g) could not be generated. All attempts to make it resulted
in the formation of a system showing less than the expected symmetry of
the classical cation. Furthermore, quenching the sample with sodium meth-
oxide dissolved in methanol (five fold excess) at -80°C yielded the start-
ing alcohol and not the expected methyl ether. To insure that the methyl

ether was not cleaved to the alcohol under the work—up conditions, the

o.mm
q.wo

~.Ho

.xmmm ecu mo mmosnmoun onu mo mmsmomn oumaaxoumem ma

w.om

m.om

H.mN

a.m~

H.5m

0.5m

n.0m

N.om

h.mo

o.mm

n.0m

N.oq

m.n¢

oa.s.m.eo

n.0m

w.~m

N.oq

w.¢¢

m.Ho

¢.omm

mm.eem

«.mqm

m.nm~

smash Hmeaamee No seem

 

mucmam>fiovm o

mmucmam>fisvm q

mucmam>wsvo N

uamHm>fisum H

 

new: wave
mo unseen mumafixouee<

 

.GOHumuuaouaoo mace ou uoumwmu sues Amnmv.uo Amze Baum Snag mumwsm Hmoaaono mzz omH mo mocomcoeoo c canoe

37

.chfiuwmoe once was oxm ecu cu accuoue oaoaxnume ecu cwwmmm ou meme ma unsouum 02o .
.o>Hum>wpom oueusevfivIMsN)OXe.oxm mnu mo muuowmm ecu Bouw monummcfi mums humane HmoHEoLo smacks

.Nm m.m new w.m coosuon wcwmum> acoumaoo wswaesoo m saws mumfinsom mm venomous mxmma omens mo HH<m

 

Aamm.muemmv osn.a
seem.muemcv Hms.H

 

 

 

 

Amvess.e-a< Asmm.euaucv mms.a Nmeomxmmovz
Aamm.muemcvsoA.mummo Aamm.nusscv smN.~ Hes.m (eumsoum.m exams
Aeneavoea.su N m mmeo
Ammmmvflsm.sua< mum.~ mem.~ seo.e - A movum.m Amway
Aamm.muemcsmavmem.mu a e m
Aamo.mnmmmmmmvecs.mua< one.~ eNm.N mas.m m on sous Ammmv
Aamo.munsesm~v~ma.mu m c
Aamm.husucmmm88ss.mua< cam.~ ~e~.~ cos.m m o Aemmv
Asmm.musmemm~vsss.su
“amm~v~mm.m-as ma~.~ ma~.~ som.m emeo-m-e excmmv
,Aamo.sunmemm~vose.su
hamm.munmeum~vmmm.mua<
esm.eummo moa.~ mos.~ cae.m emeo-0mmo-e hammv
.eaez cfiomamve.mawa~o axoxmve.m.m.~o s.Ho sebum Haa< casemEOo

 

.Ammv mam Ammo mo Amze some game menace assesses mzz me a assay

38

.mcfiu owomEoum one Ou wands no a me moonumo 0.81338 93 smfimmm 3 once we unsouum ozn

.NG NUCOHQme BORN flmxmu mum mfimwfiufimhmm CH mhwflgzm

 

m c.8N Nee (MAmmovz

m.ee( so o.m~ A.Ne m.moA (e- Ho-m.m eAmmV
A~.A~V Ae.Aev AH.mmNV m e N m
o.A~ «.se s.~m~ m o u A movlm.m aAemNV
AA.ANV Am.eev Am.AA~V e e m
e.A~ e.ee H.mA~ m on sous mAmmNV
Am.emv A~.eev As.me~v
~.A~ e.ee m.ec~ mmeo aAcmNV
e.AN A.ee m.AmN ammo-e-e Aemmv

AA.m~vmmo Am.e~v A~.mev Ae.~mmv emcoummoue mAsmNV

Am.oev Am.A~v AA.Aev As.s-V

o.oe-mmo o.A~ n.4e e.s- emeoummoue aAmmNV

.eaez c.m.m.~o eaao mm macaw Ashe acacmEco

 

 

.Ammv mam Ammv mo Amze some ammo mousse Assesses mzz omH m cases

39

40

crude product was analyzed by mass spectrometry and showed the molecular
ion peak for the alcohol. No peak was observed for the methyl ether. It
‘was thus determined that the species observed was the protonated alcohol
(83). Upon warming of the solution containing (83) to about -80°C, the
system slowly ionized to cations whose structures could not be elucidated.

The NMR spectra of (83) are included in tables 7 and 8.

 

DISCUSSION

Coates system (26)

 

Figure 8 is a plot of the 13C NMR chemical shifts of the cationic

13C NMR chemical shifts

13

centers of (26a-e) (taken from table 3) versus the
of the corresponding(x—arylcyclopentyl cations (19). The striking fea-
ture of this plot is the overall negative slope which is in sharp contrast
to the corresponding plots of classical cations (refer to figure 4). This
negative slope is best explained as a result of an increase in bridging
with the substitution of increasingly electron demanding aryl groups.
It has been observed for bridged systems that the apical carbon has drama-
tic high field chemical shifts. Compounds (8)6, (9)7, (10)8, and (11)9
I have the respective 13c+ chemical shifts of 22.5, -23.04, 29.55 and 1.5
ppm from TMS. These shifts are well over 200 ppm upfield from what would
be expected for the corresponding classical structures. Hence, it is ex-
pected, that by increasing the degree of bridging in (26) by the placement
of electron demanding aryl substituents on the bridged carbon, an
upfield shift of the cationic center would result, thus causing a negative
slope when plotted against that of a classical system.

Recently, Olahl'7 has reported the 13C NMR spectra of a variety of

13 +

7-ary1-7-norbornenyl cations (84). Figure 9 is a plot of the C chem-
ical shifts of (84) against those of (19). The similarity between figures
8 and 9 is readily apparent, indicating a general behavior for «baryl sub-
stituted bridged cations.

It is tempting to interpret the nonlinearity of figures 8 and 9 in

41’

110~
0 (260)
105.
Slope '-'-’ ‘- 1e55
Correlation coefficient = .993
100') O= work presented in this thesis
0:- work from reference 8d
95 ‘
90 ~
(3
v (26 b)
3
9‘, 85 "
a.
O
‘15
g\
‘- 80 .
Va
0 (26C)
75 ~
0 (26d)
70 ‘
65 .. o (269) 2
()0 I 1 V U T 1
255 260 265 270 275 280 '235

42

 

Figure 8 13c NMR chemical shifts of the cationic carbon of (26liversus

 

 

5 130+ of (19)

230

210

190

170

"+ of (82.)

130

513

710

43

 

Figge 213E NMR chemical shifts of the cationic carbon of (62) versus
. *

 

 

J 130’r of (19)

* Taken from reference 47.

0 l3" OCH3
- o p- CH3
A D-H
° p\"CF3
1 1 l l 1 Pals-(CF3)2
225 235 225 255 265 275 2;:

44

Ar

(84)

"48 or in terms of the "onset of €18881C81

terms of "saturation of bridging
behavior".48 However, there is no reason why such plots between classical
and nonclassical systems must be linear and thus the small deviations

are not sufficient to interpret with any confidence. Perhaps if more

data could be obtained to extend the plots, further interpretations could

be forwarded. Unfortunately, attempts to do so for (26) failed.

An unexpected result from table 3 is the almost negligible effect
of the aryl substituents on the 13C chemical shifts of the other bridging
carbons. Over the range of substituents studied, these sites showed a
steady downfield shift of less than 4 ppm with increasing electron demand.
Perhaps the chemical shift dependence on the degree of bridging for these
positions is cancelled by their dependence on the amount of charge devel-
oped. Such a cancellation appears less fortuitous, however, when compared

13C chemical shifts of

to Olah'sl'7 results on (84). In this system, the
the vinyl carbons vary only about 4 ppm but with an upfield trend with in-
creasing electron demand. Whether this is a general property of bridged
«earyl substituted cations cannot be decided until more data are obtained.
Nonetheless, this result does warn that little can be deduced concerning

the classical versus nonclassical nature of aryl substituted cations by

observing sites removed from the aryl group.

45

The chemical shifts of the aryl protons shown in table 4 deserve
some comment. Comparison of these chemical shifts to those of cations (27)
and (28) reveal that they are significantly upfield from what at first is
expected (see table 9). FUrthermore, this upfield trend is most pronounced
for the more electron demanding aryl substituents, which suggest some re-
lationship with the degree of bridging present. The most reasonable expla-
nation is that, as the system becomes more bridged, more charge is placed
on sites away from the aryl group. This would result in less than the
normal amount of charge delocalized onto the aromatic ring causing an up-

field shift of the aromatic protons.

 

 

 

 

Aryl gTOUp H !h of (26) Ho tho of (27) Hortho of (2“)
A—CFB-CéHA 7.903 8.876 8.964

The symmetry shown by the spectra of (26) implies a rapid bridge
flipping process described earlier in scheme 1.8b It was hoped that with
the strongly electron withdrawing aryl substituent, (26) would become
sufficiently bridged to enable the observation of its "frozen out" form
rather than its averaged spectrum. Unfortunately, the spectra obtained
were unchanged over the temperature range studied. Even at -lld’C, the
NMR spectra of the highly bridged (26e) gave no indication of freezing oUt.
In contrast to this observation, the NMR spectra of the methyl substituted

Coates cation (45; R=CH3) and the parent Coates cation (10) coalesce at

46

-5°C and over +10°C respectively8b (decomposition commences before complete
averaging is observed). Thus the activation energies for bridge flipping
are in the order: EH) ECH3))E Ar.
a a . a

The relative rates of solvolysis of substituted nitrobenzoates of
(42) are l:1.3(10)2:l.3(10)1 for R - H, CH3 and 3,5-(CF3)2C6H3 respective-
1y.8d Assuming similar ground state energies for the nitrobenzoates and
that the transition states resemble the intermediates formed, then the
Hammond postulate requires the ground state energies of (45) to decrease
in the order of R = H, Ar, CH3. The above results are represented in fig-

ure 10 as an energy diagram for the bridge flipping of (45; R-H, CH3, Ar).

196.24., 0 '0._-'..___-a.._ = =9: :e o. 311-: . h 0) ,o;e _.. 13 o

 

 

 

 

47

To accomodate the relative energies of the substituted Coates cat-
ions and the activation energies for bridge flipping, an important conse-
quence can be deduced; i.e., aryl groups stabilize nonclassical cations
far less than classical cations whereas alkyl groups stabilize nonclassi-
cal cations comparably (but still less) than classical cations.

If one assumes in the solvolysis of (42) that an appreciable amount
of classical character is developed in the transition state, then by the
previous statement, the transition state will be greatly stabilized by the
aryl substituent when compared to a totally bridged transition state. This
in turn will modify the energy diagram in figure 10 to make the energy level
of the aryl substituted bridged species much higher in energy as shown in
figure 11. The net effect may be that electron withdrawing aryl groups
may actually destabilize nonclassical cations while they stabilize classi-

cal cations.

Fiah : A “a; _ .2 09 .' ‘ 1- . 0 = =:= 9: 9: 27:1: =n- of a l

substitgted bridged cations eggpazed to classical cations.

 

R=Fl
R'Ar

 

R=CH3

an an

 

48

An explanation for the smaller Stabilizing influence of aryl groups
on bridged systems may be inferred from a molecular orbital diagram show-
ing the interaction between a cyclopropenium cation (a model for bridged
cations) and benzene (see figure 12). It is apparent that for benzene
to donate electrons gig conjugation with the cyclopropenium ion, a bonding
orbital of benzene must interact with an antibonding orbital of the cyclo-
propenium cation (shown by the arrow in figure 12). .This interaction must
be minimal and its stabilizing effect therefore must be small when compared
to the interaction of benzene with an "empty" nonbonding orbital of a

classical cation.

Fi_812 "0.“-2i - 0 .-:._ 0. .-.."'."‘ :90. 9° .9: 91: a _01 .‘ us 34018

W

 

 

O 9

In summary, the aryl substituted Coates system (26) as a model for
nonclassical cations shows an inverse relationship of the chemical shifts
of the cationic carbons when plotted against classical systems. Other
positions, however, are much less sensitive toward the aryl group due most
probably to a cancellation of the effects of charge and hybridization on
the 13CNMR chemical shifts. Whether this is fortuitous remains to be de-.

termined.

49

The inability to slow down the bridge flipping of the aryl substi-
tuted Coates cations sufficiently to observe the "frozen out" form is most
readily explained by the lowering of the energies of the classical inter-
mediates much more so than the bridged cations. This explanation is suppor-
ted both by theory (figure 12) and by experiment (figures 10 and 11).

The direct observation of the aryl substituted Coates system now
allows a more rigorous examination of other aryl substituted cations with
regard to their classical or nonclassical structures.

Adamantyl system (27)
13C+

Figure 13 is a plot of the chemical shifts of the adamantyl

system (27a-h) versus the 13C+ chemical shifts of the i-arylcyclopentyl
cations (19). That an excellent correlation is obtained is obvious. It
must be concluded, therefore, that, over the range of substituents studied,
the adamantyl system is classical. An attempt to extend the plot by in-
cluding the powerfully electron withdrawing group, 3,5-dichloro-4-cyano-
phenyl, gave onlyiilTINMR spectrum of (271). The interpretation of the
spectrum of (271) was complicated by the vast amount of rearranged isomers
present. Only after several attempts at generating (271) could a reliable

13c NMR

leflflifor it be obtained. Because of the time required to observe
spectra, no 130 chemical shifts could be obtained on unrearranged (271).
Indeed, this is the first report of a cation with such a powerfully elec-
tron withdrawing group attached to the cationic carbon, presumably due to
the resistance of 3° adamantyl cations to rearrange.

lHlflfliof (271), the nonequivalent methyl-

In order to make use of the
ene protons are plotted against each other in figure 14. Again, a well

defined correlation is obtained, suggesting that even the highly electron

deficient adamantyl system (271) is classical.

290

280

270

230

T.)
f.)
O

50

 

 

 

 

F e

e = 0(27h)

Sigielatizgi coefficient .= .999 0 (27g
o 271‘)
0 (276)
. 0 27b)
.. 0 (270)
/
230 27.0 250 260 270 280 29:) 1
5 130+ or (19)

Fi a , IHNH .9‘!) -

3o5~

3.4‘

1 . ‘*’
5 H of hlgh field methylene protonts

N
e
\D

2.8 1

2.71

 

2.6

2.3 2.4 2.5 5.6 2.7 218 2.9

51

9: 999:0. (£81392.

.' ‘9 -' o
in olotted at: 9' 9.: .9 ther.

Slope - 1.16
Correlation coefficient = .995

0 ODS

0(27h)

0 (279)

(27 f)

/

o (27e)

(27d)
(276)

0 (27b)

T r

51” of low field methylene proton

3.0

9
(271)

52

If the parent 2-adamanty1 cation (29) is indeed bridged, then it
‘must be assumed that our probe is too coarse to detect the presence of
bridging in ions (27a-i). It may be argued that even the powerfully elec-
tron withdrawing aryl groUps can still stabilize cations by delocalization
of charge into the aromatic ring. This in turn may cause an unsubstituted
nonclassical cation to become classical when substituted with an aryl group.
Thus it is likely for a system such as the 2-adamantyl cation (29), which
is at best weakly bridged, to become classical upon substitution with an
aryl group.

Perhaps the most significant result in this study is the effect of
acid concentration on the 130 NMR of the 2—(p-anisy1)-2-adamanty1 cation
(27a) (table 6). .Workers in the past have reported difficulties in reprodu-

12 This is the first direct

cing NMR spectra of methoxy substituted cations.
.observation of and-(p-anisyl)cation which adequately explains the irreproe
ducible spectra. The most feasible interpretation of the NMR dependence
on acid concentration is that a rapid equilibration shown in scheme V

is in progress.

° b on of 6 e e e 0 acid

 

 

(27a)

53

As the acid concentration is increased, the equilibrium is shifted
from the electron rich cation (27a) to the electron deficient cation (86).
The net effect of downfield shifts in the NMR spectra of (27a) as the acid
concentration is increased 18 in accord with such a mechanism.

In conclusion, the study presented on the aryl substituted adamantyl
cations (27a-i) warns of two pitfalls. The first is that the probe of in-
creasing electron demand can be used to detect only highly bridged cations.
What degree of bridging is detectable is still uncertain,but it must be
maintained that this probe is rather coarse and should not be used to prove
that any given system is classical. The second pitfall is that any data
concerning the p-anisyl group under strongly acidic media should not be
weighted heavily Unless precautions are taken to insure no O-protonation.

7-norbornyl system (28)

13C+ chemical shifts of the 7-norbornyl

Figure 15 is a plot of the
system (28) against the corresponding 13C+ chemical shifts of (19). As
in the adamantyl system (27), an excellent correlation is obtained indica-
ting that at best, the 7-norbornyl cation (30) is too weakly bridged to be
detected by this probe.All attempts to extend the plot beyond the bistri-
flouromethylphenyl substituent fOund in (28f) failed due to decomposition
of the cation.

It is interesting to note that the endo hydrogens of (28) are at

1H NMR than the exo hydrogens. That this observation

lower field in the
becomes more pronounced as the aryl group becomes more electron demanding

is consistent with a through space bond polarization of the endo hydrogens

by the cationic center. This is shown in figure 16.

54

Figure (5136 NMR chemical shifts of he cationic carbon of 8 ve

 

 

290 .
Slope = 1.03
Correlation coefficient = .999 6 CP3)2
280 . /
‘3 41"(31:25
270 i
A Q 2’
00
360 .. /
s /°4-F
.,
9° °4’CH3
w250 “
O = work presented in this thesis
21,0 . A = work from reference 1,7
236 7 6 4"OCH3
220 T t v I V 1
230 21.0 250 13 + 260 270 280 ' 290
J C of (19)

U8

1

 

55

_Eigure 16 Bond pglarization of the ggdg hydrogens of (28) by through space

interaction with the cationic center:

 

 

. EXPERIMENTAL

All melting points were taken on a Thomas Hoover capillary melting
point apparatus and are uncorrected.

Mass spectra were obtained using a Finnigan 4000 mass spectrometer
with an INCOS data system. In all cases, the correct molecular ion peak
was obtained.

Infrared spectra are uncorrected and were taken on a Perkin
Elmer sodium chloride spectrometer, model 137.

1H NMR spectra of the carbocation precursors were done on a Varian
T-60 spectrometer using deuterochloroform (CDC13) as the solvent and tetra-
methylsilane (TMS) as an internal standard.

13C NMR spectra of the carbocation precursors were taken on a
Varian CFT-20 spectrometer using deuteriochloroform as the solvent and inter-
nal standard (79.00 ppm relative to TMS). All spectra were decoupled.

All NMR spectra of the carbocations (26) and (28) were taken on a
Bruker WM—250 spectrometer equipped with a variable temperature controller.
13C NMR of the adamantyl cations (27) were obtained on a Varian CFT-20
spectrometer equipped with a Varian V—6040 N—M-R variable temperature con-
troller. Peaks were referenced to internal dideuteriomethylene chloride
(CD2C12) for all spectra (54.5 ppm for 13C NMR and 5.374 ppm for 1H NMR
from TMS) with the exception of the 13C NMR of the adamantyl cations (27)
which were referenced against external d6-acetone (204.5 ppm from TMS;

corrected).

Carbocation precursors

 

All of the carbocation precursors are alcohols obtained by react-

57

ing the corresponding ketone with 1.1 equivalents of the appropriate Grig-
nard reagent. 7—Norbornanone49 and 9-pentacyclo[4.3.0.02’409’80792]nona-

“011850

were made by literature procedures. 2-Adamantanone was purchased
from Aldrich and was Used as is. Yields of the Grignard reactions ranged
from 40-85%. A description of the procedure used is given below.

In a flame-dried lOO—mL, round-bottomed, three—necked flask equipped
with a stir bar, condenser, addition funnel and a nitrogen bubbler, was
placed 20 mL of dry ether and 0.23 grams of finely cut magnesium ribbon.
Stirring was started and 0.011 moles of the appropriate aryl bromide was
added. Occasional heating and/or a crystal of iodine was needed to start
the reaction. When the reaction.was under way, more ether was added (20-50
mL) and the mixture was stirred until the reaction stopped (one half hour
after refluxing ends).

0.01 Moles of the corresponding ketone, dissolved in about 20 mL of
dry ether was added dropwise and the mixture then stirred for an additional
hour. The reaction was quenched with a saturated aqueous ammonium chlo-
ride solution and the layers separated. The ether layer was washed with a
saturated sodium chloride solution, dried (sodium sulfate) and the ether
removed. The resulting crystalline compound was recrystallized from pentane
repeatedly until a constant melting point was obtained.

The important physical and spectral data for the precursors are
given at the end of this section.

Carbocation formation

 

All carbocations were generated in a 10mm NMR tube. Three proced—

ures were used:
1) For the 13C NMR samples of the adamantyl systems (27), about

0.10 g of precursor was placed in the nmr tube which was equipped with

58

a 2mm insert filled with d6-acetone. Approximately 2 mL of sulfuryl flouride
chloride was condensed in the tube and the mixture was mixed by bubbling

a slow steady stream of argon through it until all the precursor was dis-
solved. The flow of argonmwas continued while the tube‘was cooled in a dry
ice-acetone bath (-78°C) followed by the rapid addition of about 0.5 mL of
100% magic acid (obtained from Aldrich Chemical Co.). The argon flow was
continued until complete mixing was accomplished. Spectra were taken
immediately afterwards.

2) All carbocations, except for those of the adamantyl series
intended for 13C NMR and the highly unstable cations (see procedure 3), were
prepared similarly to that shown in procedure 1 with the following excep-
tions. 7

a) The 10mm NMR tube was not equipped with a 2mm insert.

b) The precursor was initially dissolved in about 0.5 mL of
dz-methylene chloride, followed by the introduction of
the sulfuryl fluoride chloride.

c) In the case of the Coates system (26), flourosulfonic
acid was preferred over magic acid.8d

3) This procedure was Used for the highly unstable and difficult—
to-make cations ((26f), (26g), (27i) and (28g)). In a 10mm NMR tube
was placed 0.5 mL of 100% magic acid followed by about 2 mL of sulfuryl
flouride chloride. The contents were mixed by bubbling argon and cooled
to -l30°C (liquid N2-pentane bath). A slow introduction of the precursor
(0.10 g) dissolved in 0.5 mL of d2-methy1ene chloride was accomplished -
while the mixture was mixed with the argon flow. The temperature was
slowly raised until a homogeneous solution resulted and then the tempera-
ture was again lowered to -130°C. Spectra were recorded immediately after

carbocation formation.

59

Quench study on (83).

 

About 15 mL of sulfuryl fluoride chloride was condensed in an
apparatus described by Hart.51 To this solution was added 1.0 mL of 100%
magic acid and the apparatus was cooled to ~80°C (dry ice—acetone bath).
The contents were mixed by the passage of a steady stream of argon through
the mixture. 0.20 g of the precursor, dissolved in 0.5 mL of methylene
chloride, was added slowly and the mixture was mixed for about 15 minutes.

The above acid solution was rapidly added to a rapidly stirred
solution of 75 grams of sodium methoxide dissolved in 1.0 liter of methanol
held at -80°C. Extraction of the resultant solution with ether, followed
by drying (NaZSOA) and removal of the ether yielded a colorless oil. NMR
and mass spectral analysis confirmed that the oil was the starting alcohol.
No evidence could be found for the presence of the methyl ether.

13C NMR of (27a) on acid concentration.

Measurement of the dependence of the
(27a) was generated as described by procedure 1 using about 0.25 mL

of 100% magic acid (about 1 equivalent) and the 13C NMR of this sample was

recorded. More acid was measured into the NMR tube and the NMR was re-

corded again. This procedure was repeated until no change was observed in

the NMR spectra. The results are recorded in table 6.

60

Physical and spectral data of the carbocation precursors

9-aryl-9-pentacyclo[4.3.0.0g’402’BOS’21nonanols

1

4‘N(CH3)2"3,5"C12‘C6H23 mp 156-157°C; HNMR (CDC13) 7.16 (2H, 3),

2.76 (6H, s), 2.73 (2H, m), 2.23 (5H, m), 1.63 (2H, m); 130 NMR
(c0013) 145.3, 141.1, 134.8, 128.7, 105.8, 50.0, 42.1, 39.3,
36.6, 35.7, 35.0; ir (nujol)/u 2.90, 6.45, 11.32; m/e = 321,

323, 325.

3,5-(CF3)2-C6H3: mp 79-80°C;]TlNMR (00613) 7.77.(2h, s), 7.65 (1H, s),

3.18 (1H, s), 2.55-2.95 (2H, m), 2.50-1.83 (4H, m), 1.77-1.37
(2H, m); 13C‘NMR(CDC13) 145.3, 133.7, 132.0, 128.1, 121.2, 105.8,
50.2, 39.5, 36.5, 35.6, 34.8; it (nujol)/L 2.94, 6.12, 11.13;

m/e = 346.
1

4-CF3-C6H4: mp 73-74°C; IINMR (CDC13) 7.43 (4H, s), 2.83 (3H, br s),

2.48-2.17 (4H, m), 1.70-1.43 (2H, m); 13CDflfli(CDCl3) 146.7,
128.2, 124.9, 124.8, 106.2, 49.9, 39.2, 36.4, 35.2, 34.9; if

(nujol)/4 2.50, 11.00, 11.90, 12.50; m/e = 278.
1

3'F‘C6H4‘ mp 69-69.5°C; H NMR (CDC13) 7.50-6.65 (4H, m), 3.14-2.70

C6H52

(23, m), 2.66 (1H, s), 2.48-1.93 (4H, m), 1.81-1.47 (2H, m);

13C)nn((0001 1.68.5, 156.3, 145.6, 145.2, 129.6, 129.1, 123.3,

3)
115.4, 114.5, 114.3, 113.5, 106.2, 49.9, 39.1, 36.4, 35.1, 35.0;
ir (nujol)92.93, 6.30, 11.70, 12.55; m/e = 228.

mp 83-84°C;]TINMR (CDC13) 7.6-7.0 (5H, m), 2.85 (2H, br s),
2.65 (1H, s), 2.95-1.92 (4H, m), 1.75-1.42 (2H, m); 13c NMR
(0001 ) 142.9, 127.9, 127.7, 127.1, 106.8, 49.1, 39.0, 36.5,

35.0; ir (nujol)%L 2.95, 13.12, 14.43; m/e = 210.

61

4-CH3-C6H4: mp 77-78°C; lHlUfl!(CDCl3) 7.0 (4H, 9, J=8Hz), 2.93-2.66
(2H, m), 2.52 (1H, s), 2.36-2.28 (7H, m), 1.77-1.50 (2H, m);
13C‘NMR(CDC13) 140.0, 136.4, 128.4, 127.4, 106.4, 49.6, 38.8,
36.5, 34.9, 20.8; ir (nuJoI)22.2.90, 12.2; m/e = 224.

4-0CH3-C6H4: mp 66-68°c; 1HNMR(CDC13) 7.50 (2H, d, J=6Hz), 6.63
(2H, , J=6Hz), 3.65 (3H, s), 2.78 (3H, br s), 2.33-l.87 (4H,
m), 1.67—1.37 (2H, m); 13ernn((CDc13) 158.5, 135.3, 128.7, 113.2,
106.4, 55.0, 49.9, 39.0, 36.6, 35.0; ir (nujol)22 2.89, 2.96,

6.21, 12.00; m/e = 240.

2-Aryl-2-adamatanols

 

4-CN-3,5-C12-C6H2: mp 199-201°c:1'

IINMR (CDC13) 7.45 (2H, s), 2.44-
2.28 (4H, m), 1.82-1.75 (11H, m); 13crnfl((c0c13) 154.7, 140.2,
217.4, 114.9, 77.0, 38.8, 37.1, 36.2, 34.2, 28.6, 28.1; ir (nujol)
.222.88, 4.50, 6.31, 6.16, 11.45, 12.25; m/e = 321, 323, 325.

4-N(Ch3)2-3,5-012-C6H2: mp 125-127°C;]TINMR (CDC13) 7.20 (2H, s),
2.80 (6H, s), 2.40-1.60 (15H, m); 13crefli(00013) 145.1, 143.6,
135.3, 126.5, 75.0, 42.0, 37.4, 35.5, 34.6, 32.7, 27.2, 26.6;
ir (nujol)/4 3.00, 6.87, 11.44, 12.56; m/e = 339, 341, 343.

3,5-(CF3)2-C6H3: mp 89-89.5°C;]71NMR (CDCL3) 7.79 (2H, s), 7.60
(1H, s), 2.47 (2H, br s), 2.27 (1H, br s), 1.80-1.67 (12H, m);
13crnnz<cnc13) 151.2, 136.6, 134.9, 130.0, 129.9, 125.0, 120.5,
79.2, 41.1, 39.5, 38.4, 36.4, 31.1, 30.4; ir (nujol)/4 2.93,

6.82, 7.28, 11.08; m/e = 364.

62

4-(CF3)-C6H4: mp 86—87°C; HHNMR (CDC13) 7.45 (4H, s), 2.42 (2H, br s),
2.22 (1H, br s), 1.73-1.68 (12H, m); 13Chflfll(CDCl3) 149.2, 125.8,
125.3, 125.1, 75.1, 37.3, 35.4, 34.4, 32.5, 27.2, 26.6; ir (nujol)
,/42.94, 6.12, 11.93; m/e - 296.

3-Br-C6H4: mp 74-76°c; 1H10fl{(CDCl3) 7.67 (1H, s), 7.48-7.40 (3H, m),
2.58-2.37 (48, m), 1.91-1.54 (9H, m); 13crnnt(00013) 147.9,
141.5, 130.2, 128.9, 124.1, 122.9, 75.7, 37.5, 35.6, 34.7, 32.9,
27.3, 26.8, ir (nujol)22 3.06, 3.52, 6.31, 12.80, 14.40; m/e =
306, 308.

C6H5: mp 79-80°C;]TlNMR (00013) 7.48-7.00 (5H, m), 2.44 (2H, br s),
2.27 (18, br s), 1.65 (12H, br s); 13cnnn((00013) 145.2, 128.5,
127.0, 125.3, 75.4, 37.5, 35.5, 34.7, 32.8, 27.3, 26.8; ir (nujol)
,222.96, 6.20, 12.96, 14.30; m/e - 228.

4-F-C6H4: mp 79-80°C; 18inn((c0013) 7.50 (28 dd, J=9Hz, 5.5H2), 7.041
(2H t, J=9Hz), 2.52 (1H, 63), 1.51-1.90 (12H, m); 13cnnnz(00013)

167.1, 155.5, 141.3, 141.1, 127.4, 127.0, 115.7, 114.7, 75.1,
37.6, 35.8, 34.7, 32.8, 27.4, 26.8; ir (nujo1)2t 2.95, 3.50,
6.26, 12.0; m/e = 246.

4-CH3-C6H4: mp 77-78.5°C: lurnnl(CDCI3) 7.30 (28, d, J=8Hz), 7.04
(2H, d, J=8Hz), 2.46 (2H, br s), 2.26 (4H, br s), 1.95 (28, m),
1.66 (9H, br s), 1.47 (2H, s);13C NMR (CDC13) 144.1, 138.4,
130.4, 126.9, 77.0, 39 3, 37.3, 36.5, 34.6, 29.1, 28.6, 22.5;

ir (nujol)/U. 2.86, 2.91, 6.62, 12.25; m/e = 242.

63

4-OCH3-C6H4: mp 96-98°C;]TINMR (CD013) 7.25 (28, d, J=9Hz), 6.70
(2H, d, J-8Hz), 3.70 (3H, s), 2.45 (2H, br s), 2.28 (1H, br s),

1.68-1.50 (12H, m); m/e - 258.

7-Aryl-7-norbornanols

 

4-8-(083)2-3,5-012-c682:' mp (oi1);]iINMR (00013) 7.15 (2H, s),
2.83 (68, s), 2.25 (48, m), 1.93 (18, s), 1.42 (6H, m);13crnn(
(00013) 145.5, 140.7, 135.2, 127.9, 87.0, 42.1, 28.3, 27.2; ir
(neat)24 3.0, 3.49, 3.55, 6.90, 12.55, 13.65; m/e = 399, 401, 403.

3,5-(CF3)2-C6H3: mp 65-66°c;]iiuhR (CD013) 7.70 (2H, s), 7.60 (1H, s),
2.33 (2H, br s), 2.22-1.88 (2H, m), 1.62-1.26 (7H, m);13CNMR
(00013) 145.0, 132.0, 130.2, 127.5, 121.4, 87.4, 42.4, 28.3,
27.2; it (nujollx4 3.03, 3.48, 3.53, 6.89, 11.10, 11.81; m/e =
324.

4-(CF3)-C6H4: mp 79-80°c;]i(NMR (00013) 7.61 (4H, 8), 2.407 (2H, 8),
2.155 (2H, d, J=7.25Hz), 1.80 (18, s), 1.48-1.24 (68, m); 13c NMR
(00013) 146.4, 127.6, 125.5, 87.5, 42.1, 28.3, 27.2; it (nujol)
,213.09, 3.48, 3.55, 6.18, 6.90, 11.90; m/e = 256.

. l
C6H5: mp 46-47 C;

IINMR (00013) 7.35-7.00 (5H, m), 2.32 (28, br s),
2.28-1.83 (38, m), 1.53-1.17 (68, m);13CNMR (CDC13) 142.5,
128.4, 127.4, 127.0, 87.7, 41.9, 28.4, 27.3; ir (nn3o1)22 3.05,
3.47, 3.54, 6.86, 13.05, 14.36; m/e - 188.

1
exo,exo-2,3-d2-P-F-C6H4: mp (oil); liNMR (CDCl 7.42-7.13 (2H, m),

3)
6.85 (2H, t, J=8.5Hz), 2.35-1.60 (58, m), 1.60-.97 (4H, m),
13crnn((CD013) 132.6, 129.06, 128.64, 115.88, 114.82, 87.29,
42.22, 28.44, 27.29; ir (neat)22 3.00, 3.45, 3.51, 4.60, 6.25,

6.67, 11.95, 12.31; m/e = 208.

64

1

P-CH30-C6H4: mp 50-51°c; HNMR (00013) 7.30 (2H, d, J=8Hz), 6.75
(28, d, J=8Hz), 3.73 (38, s), 2.28-1.17 (118, m);13CNMR (0001.3)
1.58.9, 128.3, 127.7, 113.9, 87.5, 55.2, 42.2, 28.6, 27.5; it

(nujol)/U. 2.95, 3.46, 3.52, 6.21, 6.62, 12.02; m/e = 218.

REFERENCES

. T.P. Nevell, E. de Salas, and C.L. Wilson, J. Chem. Soc., 1188

 

(1939).

. S. Winstein and D. Trifan, J. Amer. Chem. Soc., 13, 1147, 1154

 

(1952).

 

. J.D. Roberts and R.H. Mazur, J. Amer. Chem. Soc., 13, 3542

(1951).

. H.C. Brown, "The Nonclassical Ion Problem", Plenum Press, New York

and London, 1977, p. 47.

b) Ibid., p. 4.

. H.C. Brown, "The Transition State", Chem. Soc. (London) Spec.

Pub1., 16, 140-158 (1962).

. a) H. Hogeveen and P.W. Kwant, Tetrahedron Letters, 1665 (1973).

b) H.T. Jonkman and W.C. Nieuwpoort, Tetrahedron Letters, 1671
(1973).

 

c) H. Hogeveen and P.W. Kwant, Tetrahedron Letters, 4351 (1974).

d) H. Hogeveen and P.W. Kwant, J. Amer. Chem. Soc., 25, 7315
(1973).

e) H. Hogeveen and P.W. Kwant, J. Amer. Chem. Soc., 96, 2208
(1974).

a) W.-D. Stohrer and R. Hoffmann, J. Amer. Chem. Soc., 94,
1661 (1972).

b) M.J.S. Dewar and R.C. Hadden, J. Amer. Chem. Soc., 95,
5836 (1973).

c) H. Kollmar, H.0. Smith and P.v.R. Schleyer, J. Amer. Chem.
Soc., 25, 5834 (1973).

d) S. Masamune, M. Sakai and H. Oms, J. Amer. Chem. Soc., 24
8956 (1972).

e) K.N. Honk, J. Amer. Chem. Soc., 24, 8958 (1972).

65

10.

11.

12.

13.

14.

15.

66

f) S. Masamune, M. Sakai, and H. Oms, J. Amer. Chem. Soc., 94,
8965 (1972). ”"

 

g) W.J. Hehre and P.v.R. Schleyer, J. Amer. Chem. Soc., 95,
5837 (1973).

 

h) S. Masamune, Pure App. Chem., 49, 861 (1974).

 

. a) R.M. Coates and J.L. Kirkpatrick, J. Amer. Chem. Soc., 92,

 

4883 (1970).

b) R.M. Coates and E. Fretz, J. Amer. Chem. Soc., 2], 2538 (1975).

 

c) R.M. Coates and E. Fretz, J. Amer. Chem. Soc., 92, 287
(1977).

 

d) H.C. Brown and M. Ravindranathan, J. Amer. Chem. Soc.,
22, 289 (1977).

 

e) W.L. Jorgensen, Tetrahedron Letters, 3033 (1976).

 

. a) H. Hart and M. Kuzuya, J. Amer. Chem. Soc., 95, 4096 (1973).

 

b) H. Hart and M. Kuzuya, Tetrahedron Letters, 4123 (1973).

 

c) H. Hart and M. Kuzuya, J. Amer. Chem. Soc., 94, 8958 (1972).

 

 

d) H. Hart and M. Kuzuya, J. Amer. Chem. Soc., 26, 6436 (1974).

e) H. Hart and R. Willer, Tetrahedron Letters, 4189 (1978).
"Carbonium Ions", Vol. I—IV, G.A. Olah and P.v.R. Schleyer, Eds. John
Wiley and Sons, New York, London and Sydney (especially Chapters

7 and 28). '

a) G.A. Olah, et. a1., J. Amer. Chem. Soc., 24, 2044 (1972).

b) G.A. Olah, J. Amer. Chem. Soc., 2;, 464 (1971).

a) D. Kelly, Private Communication.

b) H.C. Brown, Private Communication.

W.T. Chambers, Ph.D. Thesis, Michigan State University, (1975).

R. Botto, Ph.D. Thesis, Michigan State University, (1975).

H.C. Brown, "The Nonclassical Ion Problem", Plenum Press, New York
and London, 1977, p. 274.

67

16. D. Lenoir, R.B. Hall and P.v.R. Schleyer, J. Amer. Chem. Soc.,

l7.

18.

19.

20.

21.

22.

23.

9g

8)

b)

a)

b)

C)

d)

e)

f)

a)

b)

 

, 2138 (1974).

D. Lenoir, D.J. Raber and P.v.R. Schleyer, J. Amer. Chem.
Soc., 26, 2149 (1974).

 

P.v.R. Schleyer, et. a1., J. Amer. Chem. Soc., 26, 2157

 

S. Winstein and J. Sonnenberg, J. Amer. Chem. Soc., 8;,
3235 (1961)

 

S. Masamune and M. Sakai, Can. J. Chem., Q2, 855, 858 (1974).

 

S. Winstein, E.G. Friederich, R. Baker and Y. Lin, Tetrahedron,
Supp1., No. 8 (Part II), 621 (1966).

 

S. Winstein, Quart. Rev. Chem. Soc., 22, 141 (1969).

 

W.J. leNoble, B.L. Yates and A.W. Scaplehorn, J. Amer. Chem.
Soc., 82, 3751 (1967).

 

H. Kucynski, M. Walkowicz and C. Walkowicz, Rocz. Chem., 42,
1277 (1968).

 

H. Tanida, T. Tsuji and T. Irie, J. Amer. Chem. Soc., 82,
1953 (1967).

 

M.A. Battiste, C.L. Deyrup, R.B. Pincock and J. Haywood-Farmer,
J. Amer. Chem. Soc., 82, 1954 (1967). '

 

H.C. Brown, "Nonclassical Ion Problem", Plenum Press, New York
and London, 1977, p. 272.

a)

b)

C)
a)

b)

a)

b)

H.C. Richey, Jr., and N.C. Buckley, J. Amer. Chem. Soc., 8;,
3057 (1963).

 

H.C. Brown and E.N. Peters, J. Amer. Chem. Soc., 2], 1927
(1975).

 

P.v.R. Schleyer, J. Amer. Chem. Soc., 8Q, 1854 (1964).

S. Winstein, M. Shatavsky, C. Norton and R.B. Woodward, J,
Amer. Chem. Soc., 11, 4183 (1955).

 

S. Winstein and C. 0rdronneau, J. Amer. Chem. Soc., 82, 2084
(1960).

 

 

M.J.S. Dewar and W.W. Schoeller, Tetrahedron,’2], 4401 (1971).

R.K. Lustgarten, M. Brookhart and S. Winstein, J. Amer. Chem.
Soc., 24, 2347 (1972).

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

68 ~

G.A. Olah and G. Liang, J. Amer. Chem. Soc., 194 (1973).

,2.

 

P. Vogel, M. Saunders, W. Thielecke and P.v.R. Schleyer, Tetrahedron
Letters, 1429 (1971).

 

D. Lenoir, Chem. Ber., 226, 2366 (1973).

D. Farcasiu, J. Amer. Chem. Soc., 22, 5301 (1976).

 

P.v.R. Schleyer, et. a1., J. Amer. Chem. Soc., 22, 5246 (1970).

 

Z. Majerski, P.v.R. Schleyer and A.P. Wolf, Ibid. 22, 5731 (1970).

a) H.W. Geluk and J.L.M.A. Schlatmann, Chem. Commun., 426 (1967).

 

b) H.W. Geluk and J.L.M.A. Schlatmann, Tetrahedron, 23, 5361,
5369 (1968).

 

c) H.W. Geluk and J.L.M.A. Schlatmann, Recl. Trav. Chim. Pays.-
Bas., fig, 13 (1969).

 

d) M.A. McKervey, J.R. Alford, J.F. McCarrity and E.J.F. Rea,
Tetrahedron Lett., 5165 (1968).

 

e) H.W. Whitlock and M.W. Siefken, J. Amer. Chem. Soc., 29, 4929
(1968).

 

Z. Majerski, S.H. Liggero, P.v.R. Schleyer and A.P. Wolfe,
Chem. Comm., 1596 (1970).

 

E.I. Bagrii, E.I. Anosova and P.I. Sanin, Neftekhimdya, }2,
639 (1971).

 

R.C. Fort, Jr., "Adamantane", Marcel Dekker, Inc., New York, 1976,
Vol 5.

E. Boelema, J.H. Wieringa, H. Wynberg and J. Strating, Tetrahedron
Lett., 2377 (1973).

 

35. J.A. Bone, J.R. Pritt, and M.C. Whiting, J. Chem. Soc., Perkin II

36.

37.

38.

 

1447 (1975).

P.v.R. Schleyer, D. Lenoir, P. Mison, G. Liang, G.K. Kurya Prakash
and G.A. Olah, J. Amer. Chem. Soc., 1Q2, 683 (1980).

 

D. Lenoir, P. Mison, E. Hyson, P.v.R. Schleyer, M. Saunders,
P. Vogel and L.A. Telkowski, J. Amer. Chem. Soc., 2Q, 2157 (1974).

0.8. Foote, J. Amer. Chem. Soc., g9, 1855 (1964).

 

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

69

S. Winstein, F. Gadient, E.T. Stafford and P.E. Kleindinst, g,
Amer. Chem. Soc., QQ, 5395 (1958).

 

P.G. Gassman and J.M. Hornback, J. Amer. Chem. Soc., 82, 2487 (1967).

 

F. Miles, J. Amer. Chem. Soc., 82, 2488 (1967).

 

F.B. Miles, J. Amer. Chem. Soc., 29, 1265 (1968).

 

M.J.S. Dewar and W.W. Schoeller, Tetrahedron, 21, 4401 (1971).

 

A. Colter, R.C. Friedrich, N.J. Holness and S. Winstein, J. Amer.
Chem. Soc., 8], 378 (1965).

H.C. Brown, "The Nonclassical Ion Problem", Plenum Press, New York
and London, 1977, p. 94-98.

P.G. Gassman, J.M. Hornback and J.M. Pascone, Tetrahedron Letters,
1425-1428 (1971).

 

G.A. Olah, Private Communication. To be published soon.

D.G. Farnum and A.D. Wolf, J. Amer. Chem. Soc., 22, 5683 (1977).

 

P.G. Gassman and P.G. Pape, J. Organic Chemistry, 22, 160 (1964).

 

W.G. Dauben and R.J. Twieg, Tetrahedron Letters, 531 (1974).

 

H. Hart and M. Kuzuya, J. Amer. Chem. Soc., 29, 6436 (1974).