The nature of the electronic factor governing diastereofacial selectivity in remotely substituted (X) 2-adamantyl cations: 5-X versus 4-X substitution

Article abstract of DOI:10.1002/poc.1247

A limited series of 4^eq-substituted (X) 2-methyleneadamantanes (6, Y=CH₂, X=F, Cl, Br, I, and SnMe3) has been synthesized and diastereoselectivities for their hydrochlorination (HCl/CH₂Cl ₂) have been determined. Diastereoselectivities for the fluorination (DAST/CH₂Cl₂) of secondary alcohol mixtures, obtained from the hydride reduction of the precursor ketones (6,Y=O) to the alkenes, have also been measured. A comparison of this selectivity data for nucleophilic trapping of 4^eq-substituted (X) 2-adamantyl cations (4, R=H and Me) with the corresponding information for 5-substituted (X) 2-adamantyl cations (1, R=H and Me) has revealed important distinctions between the two series. In particular, whereas extended hyperconjugative effects appear to be the predominant electronic effect governing facial selectivity in the 5,2-series, electrostatic influences prevail in the 4,2-disposition. Copyright

Full text of DOI:10.1002/poc.1247

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2007; 20: 791–798
Published online 17 August 2007 in Wiley InterScience
(www.interscience.wiley.com) DOI: 10.1002/poc.1247
The nature of the electronic factor governing
diastereofacial selectivity in remotely substituted (X)
2-adamantyl cations: 5-X versus 4-X substitution
*
William Adcock and Neil A. Trout
School of Chemistry, Physics, and Earth Sciences, Flinders University, Adelaide, SA 5001, Australia
Received 1 May 2007; revised 20 June 2007; accepted 22 June 2007
eq
—
—
—
ABSTRACT: A limited series of 4 -substituted (X) 2-methyleneadamantanes (6, Y CH , X F, Cl, Br, I, and
—
2
SnMe3) has been synthesized and diastereoselectivities for their hydrochlorination (HCl/CH2Cl2) have been
determined. Diastereoselectivities for the fluorination (DAST/CH2Cl2) of secondary alcohol mixtures, obtained
—
from the hydride reduction of the precursor ketones (6,Y O) to the alkenes, have also been measured. A comparison
eq
—
of this selectivity data for nucleophilic trapping of 4 -substituted (X) 2-adamantyl cations (4, R H and Me) with the
—
—
—
corresponding information for 5-substituted (X) 2-adamantyl cations (1, R H and Me) has revealed important
—
distinctions between the two series. In particular, whereas extended hyperconjugative effects appear to be the
predominant electronic effect governing facial selectivity in the 5,2-series, electrostatic influences prevail in the
4,2-disposition. Copyright # 2007 John Wiley & Sons, Ltd.
Supplementary electronic material for this paper is available in Wiley InterScience at http://www.mrw.interscience.
wiley.com/suppmat/0894-3230/suppmat/
KEYWORDS: adamantane; diastereofacial selectivity; substituent effects; hyperconjugation; coulombic interactions
INTRODUCTION
The p-facial selectivity has been ascribed to pre-
dominantly differential hyperconjugative effects induced
by the substituent at C5 determining the relative stability
of rapidly equilibrating pyramidalized anti(or E) and
syn(or Z) epimeric ions (2 and 3, respectively) prior to
rapid capture.2,3 However, the situation is not completely
defined as there is evidence that electrostatic effects may
also be a component of the electronic influence.5
The precise nature and significance of long-range
electronic effects governing diastereofacial selectivity
of additions to trigonal carbon centers remain a subject of
continuing inquiry and debate.1 Studies of model systems
in which the electronic factor is segregated from
complicating steric and conformational effects have
provided crucial insight into the problem. The most
widely deployed substrates of this kind are 5-substituted
(X) 2-adamantyl derivatives.2–4 Among these substrates
the most conspicuous examples of diastereoselectivity are
those reactions which are mediated by the formation of
More recently, extensive studies in the gas phase of
—
several of the tertiary ions (1, R Me) have provided
—
insight into the intrinsic factors governing face selection
in these systems.6 These results coupled with quantum
chemical calculations highlight that the aforementioned
—
—
—
model of diastereoselectivity for the tertiary ions (1, R
—
secondary (R H) and tertiary (R Me) 5-substituted
—
—
(X) adamant-2-yl cations (1).2,3
Me) is far from being unequivocal. In particular, the
apparent occurrence of many of these ions as a single
pyramidalized structure anti(2) and syn(3) for s-electron-
donating and -withdrawing substituents, respectively, in
the gas phase suggests that the intrinsic diastereoselec-
tivity of these cations is not influenced by an equilibrium
population of the two syn/anti invertomers. The observed
facial selectivity is a direct consequence of the different
space available to the incoming nucleophile on the anti
and syn faces of the pyramidalized cations (2 and 3,
respectively). Furthermore, a comparison of the gas
phase results with the solution data suggests that in the
latter medium diastereofacial selectivity of many of the
*Correspondence to: W. Adcock, School of Chemistry, Physics, and
Earth Sciences, Flinders University, GPO Box 2100, Adelaide, SA
5001, Australia.
E-mail: William.adcock@flinders.edu.au
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 791–798

792
W. ADCOCK AND N. A. TROUT
—
tertiary ions (1, R Me) may arise in part from the
—
differential solvation of the two faces of a single
pyramidalized ion.
In the light of the aforementioned evolving picture of
diastereoselectivity for 1 it is of interest to consider how
face selection in 2-adamantyl cations will be affected by
moving the substituent from the 5- to the 4-position (4 and
5). le Noble and co-workers2
However, a significant feature of the aforementioned
results not emphasized previously is that the rela-
tive preference for syn addition in 1 and 4(1>4) is
diametrically opposite to that for the reduction of the
—
ketones (6 and 7, Y O; 6>7)! This strongly suggests
—
that other factors (possibly electrostatic effects)3,5
besides delocalized electrical effects must be influen-
cing diastereoselectivity in these systems as the trend is
inexplicable in terms of purely a TS hyperconjugative
model.1a Consequently, we were prompted to examine
further how face selection in 2-adamantyl cations will
be affected by having the substituent at the 4-(4) versus
the 5-position (1). We limited our investigation of
s-electron-withdrawing groups to the halogen series
since these particular substituents have similar electro-
static field constants (sF values)3 but vastly different
electronegativity parameters (sx values).10 Their value
lies in the fact that the pattern of induced diastereos-
electivities by the halogen subset reflects importantly
on the nature of the electronic effect governing
diastereoselectivity. If the selectivity order parallels
the sF values (F ꢀ Br ꢀ Cl ꢀ I) then through-space field
effects are dominant whilst an order that follows sx
values (F > Br > Cl > I) signifies dominant hypercon-
jugative effects (a function of the s-inductive pertur-
bation of C4 and C5 in 4,2- and 5,2-adamantyl
derivatives, respectively).3,5 To complete the substi-
tuent set we also included Sn(CH3)3, an electropositive
substituent with virtually no electrostatic field influ-
ence (sF ꢁ 0) but with a hyperconjugative effect on
diastereoselectivity in 1 opposite in direction to that of
the halogens.3b,6
pondered this question and perfunctorily dismissed
consideration of 5 since here the axial disposed group
will obviously affect stereochemistry in a steric way. In
the case of 4, the situation where the substituent is located
in the sterically unbiased equatorial position, they con-
cluded that a priori an answer cannot be unequivocally
provided since it is not self evident that the inductive
effect of a s-electron-withdrawing group on the donor
hyperconjugative ability of the C3—C4 bond with the
electron deficient center (C2) will be reduced, more or
less, than the extended (or double) hyperconjugative effect
of a 5-substituent (antiperiplanar electron-withdrawing
effect involving both the C3—C4 and C1—C9 donor
bonds with C2).3,7 Relevant to the above question are
solvolysis rate studies of 5- and 4eq-substituted (X)
2eq-adamantyl p-nitrobenzenesulfonates by Grob et al.8
which revealed that electron-withdrawing polar effects
are transmitted more effectively in the 5,2-disposition
compared to the 4,2. Consequently, a greater preference
8
—
—
for syn attack in 1 (R H) than 4 (R H) is observed.
—
—
Interestingly, Kaselj and le Noble9 previously explored
the effect on facial selectivity of moving the substituent
from the 5- to the 4-position on the the hydride reduction
of some appropriately substituted (X)-2-adamantanones
—
—
(6 and 7,Y O; X F and Br). It was noted that
—
—
despite the substituent being closer to the stereoinductive
center in 6 and, therefore, with the possibility it might
have a greater perturbation than in 7, the observed
expected preference for syn addition in 6 was unexpect-
edly only slightly greater than that for 7. Moreover,
it was also found that the facial selectivity for the
Herein, we report the results of our study which
was essentially twofold. First, we wished to deter-
mine diastereoselectivities for the nucleophilic trap-
—
—
ping of the secondary and tertiary ions of 4 (R
H
and CH3, respectively) in order to make comparisons
with the corresponding data for 1. Second, in order
to provide information regarding the structure of
the possible syn(or Z) and anti(or E) epimeric ions of
reduction of 4eq,9eq-dibromoadamantan-2-one (E/Z ¼ 86/
—
14) is not even twice that of a single bromine (6,Y O;
—
—
—
X
Br. E/Z ¼ 76/24). It was concluded largely by
default that the results probably highlight the importance
of extended hyperconjugation in the transition state
(TS) of 7.7
4 we have carried out full geometry optimizations
at the B3LYP/6-31Gꢂ level of theory where X
and Cl.
—
—
F
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 791–798
DOI: 10.1002/poc

FACIAL SELECTIVITY IN 2-ADAMANTYL CATIONS
793
EXPERIMENTAL
Table 1. Product distributions for the hydrochlorination
of 4eq- and 5-substituted (X)-2-methyleneadamantanes (6
—
and 7, Y CH ; respectively)
—
2
Synthesis of compounds and general
procedures for hydrochlorination and
fluorination
CH2Cl2
NO2CH3
6
7a
6
7a
—
—
—
—
All the alkenes (6,Y CH ) and ketones (6,Y O) are
2
—
X
%E
%Z
%E
%Z
%E
%Z
%E
%Z
known compounds (X F, Cl, Br, I, and Sn(CH ) ; see
3 3
—
Supplementary Material). Except for the tin ketone, all
were prepared essentially by literature procedures. Some
essential synthetic details together with general pro-
cedures (hydrochlorination, fluorination, reduction, and
methylation) are reported in the Supplementary Material.
The relative selectivity data listed in the various Tables
below are the average of determinations by several
methods (13C, 1H, and 19F NMR, VPC-MS, and VPC) and
are accurate to ꢃ3%.
F
7
6
7
13
93
94
93
87
10
17
22
34
90
83
78
66
6
15
15
21
94
85
85
79
0
3
17
26
100
97
83
74
Cl
Br
I
a Taken from References 3 and 5.
been rationalized in terms of the relative stability of
—
equilibrated solvated epimeric cations (2Ð3; R CH )
—
3
being governed predominantly by differential hypercon-
jugation.5 Further, the solvent effect has been ascribed
to a mechanistic difference: intimate ion-pairs in
CH2Cl2 (e ¼ 8.9) and free ions in NO2CH3 (e ¼ 37.5).5
Alternatively, if only the solvated pyramidalized Z-cation
Computational methods. The cation calculations
reported below were carried out at the B3LYP/6-31Gꢂ
level of theory utilizing the GAUSSIAN 98 program
package.11 Analytical frequency calculations were
performed on the minima and transition states of the
density functional theory (DFT) optimized cation
structures to determine zero-point vibrational energies
(ZPVE) and, as well, to ensure Nimag ¼ 0 and 1 for the
minima and transition states. NBO analyses were
executed using the GAUSSIAN 98 program on several
of the epimeric cations of 1 and 4. The NBO approach is
described in detail by Weinhold and co-workers12 and no
detailed account is necessary here. Suffice to state that it is
useful methodology for estimating quantitatively the
energy of hyperconjugative effects by treating the
delocalizing interactions by a standard second-order
perturbation approach to provide so-called E(2) energies.
—
—
(3, R CH ) exists then the observed selectivity will be a
3
function of the different physical space available to the
nucleophile in its approach to the asymmetric faces of the
distorted ion. In addition, differential solvation of the two
faces may also play a role.6
If we assume that a rapid equilibrium also exists
—
between the solvated E- and Z-cations of 4 (8Ð9, R
—
CH3, respectively) then the observed selectivity order
(F ꢀ Cl ꢀ Br ꢀ I) suggests that their relative stability is a
consequence of differential electrostatic field effects. This
seems reasonable if the decreased donor capacity of the
C3—C4 bond by the s-inductive effect in the E-cation is
essentially compensated by a concominant increased
donor effect from the C1—C9 bond as a result of
enhanced electron demand at the reaction center (C2)13
(see calculations below).The destabilizing electrostatic
interaction between the C—X dipole and the delocalized
positive charge will clearly be greater in the E-cation than
the Z-cation because Coulombic repulsion
RESULTS AND DISCUSSION
Stereoselectivity
—
Hydrochlorination of alkenes (6, Y CH ). Dias-
—
tereoselectivities (Z/E) for the hydrochlorination of 6
2
—
—
(Y CH , X Halogens) in CH Cl and NO CH are set
2
—
—
2
2
2
3
out in Table 1. The corresponding values for 7 are also
listed in order to facilitate comparison. Unfortunately, the
protolytic instability of Sn(CH3)3 precluded a result for
this group. An examination of the data reveals that the
selectivity order (F ꢀ Br ꢀ Cl ꢀ I) for 6 essentially
parallels the sF values of the substituents.3 Moreover,
except for F, the facial selectivity decreases on going
from CH2Cl2 to NO2CH3 as solvent. This contrasts with
the general picture for 7 where facial selectivity exhibits
an electronegativity order (F > Cl > Br > I) and the
aforementioned solvent change increases the selectivity
for all the halogens. The overall pattern of results for 7 has
is maximized in the former ion due to the delocalized
positive charge (including that in the CH3 group) being
towards the polar C—X bond. Within this framework the
observed solvent effect noted above is probably a
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 791–798
DOI: 10.1002/poc

794
W. ADCOCK AND N. A. TROUT
manifestation of the solvent encroaching on the space
between the C—X dipole and the reaction site. Hence, the
bulk dielectric of the medium impinges significantly on
the effective dielectric constant.14 An alternative expla-
nation in terms of the existence of a single solvated
Z-cation can also be advanced as above for 1.6
field-effect of the halogen substituents (F ꢀ Br ꢀ Cl ꢀ I).
Hence, it appears that differential electrostatic effects
appear to predominantly govern facial selectivity in the
—
—
secondary ions of 4 (R H) as noted above for the
—
corresponding tertiary species 4 (R CH ).
—
3
However, this interpretation is contingent on rapid
equilibration of the ions (EÐZ) prior to trapping by the
fluoride ion. This imperative requires that the pure
epimers yield identical product mixtures. Consequently,
in order to test this assumption we carried out
fluorinations of some of the pure epimeric alcohols
(10; E- and Z-fluoro and -bromo alcohols, see Table 3).
Most importantly, it can be seen that the product mixtures
from the pure alcohols within both series are not the same.
Previously,7b we noted this situation for the E- and
Z-bromo alcohols of 11 ((E)-Br and (Z)-Br, Table 3)
which agreed with the observations of le Noble and
co-workers15 for some other reactions mediated by
secondary 2-adamantyl cations when the 5-substituent
(X) is a good s-electron acceptor. In contrast, when the
—
—
—
Fluorination of alcohols (10, Y OH, Z H and
—
—
—
Y H, Z OH). Product distributions for the fluorina-
tions (DAST/CH2Cl2) of mixtures of secondary alcohols
—
—
—
—
—
—
(10, Y OH, Z H and Y H, Z OH; E/Zꢀ70/30, see
—
—
—
—
Supplementary Material) are set out in Table 2. It should
be noted that fluorination of the 4,2 tin-alcohol
—
5-substituent (X) is a s-electron donor (X Si(CH ) ) the
3 3
—
epimeric alcohols give identical product mixtures.3,7
pertinent aspect of the data in Table 3 is the reverse
A
—
selectivity result (E > Z) for fluorination of 10 (Y OH,
—
H; X F). We believe this highlights the significant
—
—
—
—
—
—
—
—
—
mixture 10 (Y OH, Z H and Y H, Z OH; X
—
Z
—
—
—
—
influence of extended hyperconjugation (coupling of the
n-orbital of F with the electron deficiency at C2 via
the C3—C4 s-bond; nF—sC—C—sþ) on the stability of 8
Sn(CH3)3) did not yield any tin-fluorides. The only
product detected was 2,4-dehydroadamantane (see
Supplementary Material). The corresponding results for
the previously reported fluoride product mixtures from
similar fluorinations of 5,2-secondary alcohol mixtures
(E/Zꢀ60/40) are also listed in Table 2 in order to facilitate
comparison. The latter clearly display an electronegativ-
ity order of facial selectivity in accord with the dominance
of extended hyperconjugation controlling the relative
stability of the rapidly equilibrating ions (EÐZ).7b By
—
—
(R H, X F). In this ion the preferred stereoelectronic
—
—
requirement (antiperiplanarity of the participating orbi-
tals)16 for through-bond transmission is met but not in the
corresponding Z-cation. Noteworthy, is that this reson-
ance phenomenon does not appear to influence the
relative stability of the less electron demanding tertiary
—
cations (8Ð9, R CH ; Table 1). It is worth noting that
—
3
this delocalized interaction involving fluorine as a
substituent has been invoked by Nelsen et al.17a and
Cieplak17b to explain the order of formal redox potentials
for oxidation and VIP parameters for 4eq-halogenated-
biadamantylidene and the unusual selectivity order for the
contrast, but similar to the results above for the tertiary
—
ion (4, R CH ), the selectivity pattern for trapping of
—
the secondary ion (4, R H) parallels the electrostatic
3
—
—
Table 2. Product distributions for the fluorinationa of mix-
tures (E and Z isomers) of 4eq- and 5-substituted (X)
—
—
hydride reduction of 6 (Y O, X Halogens) utilizing
— —
LiHAl(OC(CH3)3, respectively. However, it should be
—
—
—
—
Adamantan-2-ols (10 and 11; Y OH, Z H and Y H,
—
—
—
Z
OH, respectively)
—
10b
11c
Table 3. Product distributions for the fluorinationa of some
pure epimers of 4eq- and 5-substituted (X) adamantan-2-ols
X
%E
%Z
%E
%Z
—
—
—
—
—
(10 and 11; Y OH, Z H and Y H, Z OH, respectively)
—
—
—
F
34
37
36
35
66
63
64
65
0d
6
15
94
85
10b
11
Cl
Br
I
19
35
81
65
X
%E
%Z
%E
%Z
Sn(CH3)3
0d
>98e
Tracee
(E)-F
(Z)-F
(E)-Br
(Z)-Br
54
10
36
17
46
90
64
83
12b
2b
88b
98b
73c
90c
a DAST/CH2Cl2.
27c
10c
b See Table 1 (NaBH4) in the Supplementary Material for composition of the
—
—
—
—
secondary alcohol mixtures of 10 (Y OH, Z H; Y H, Z OH).
—
—
—
—
c Taken from Reference 5. Composition of alcohol mixtures (E/Z ꢁ 60/40).
d No substitution product detected only propellane formation. See
Supplementary Material.
a DAST/CH2Cl2.
b This study. Alcohols available from a previous study (Reference 6a).
c Taken from Reference 7b.
e Fragmentation predominant.
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 791–798
DOI: 10.1002/poc

FACIAL SELECTIVITY IN 2-ADAMANTYL CATIONS
795
Table 4. Calculated energies a,b and zero–point vibrational energies (ZPVE)b of 4-substituted (Xeq) 2-adamantyl cations (4; Z
—
and E, R H and Me)
—
c
d
Systems
Eelec
ZPVE
Eo
Nimag
DEo
—
—
—
—
H
—
R
R
H, X
ꢄ389.820158
ꢄ389.8175889
ꢄ489.048056
ꢄ489.04467
ꢄ489.045711
ꢄ849.408076
ꢄ849.403993
ꢄ849.405003
ꢄ528.385582
ꢄ888.745629
0.23131
0.22082
ꢄ389.5934766
ꢄ389.5913833
ꢄ488.824558
ꢄ488.821788
ꢄ488.822239
ꢄ849.186131
ꢄ849.182641
ꢄ849.183116
ꢄ528.134073
ꢄ888.495733
0
0
1.31
0
1.74
1.46
0
2.19
1.89
—
1e
0
—
—
—
H, X H(TS)
—
—
Z(R H, X F)
0.223498
0.222885
0.223472
0.221945
0.221352
0.221887
0.251509
0.249896
—
—
ZÐE(TS)
—
1f
0
—
—
E(R H, X F)
—
—
—
Z(R H, X Cl)
0
—
—
ZÐE (TS)
—
1g
0
—
—
E(R H, X Cl)
—
—
—
Z(R Me, X F)
—
—
0
0
—
—
Z(R Me, X Cl)
—
—
—
a,b Calculations carried out at the B3LYP/6-31Gꢂ level of theory utilizing Gaussian 98. All the structures were fully optimized, and analytical frequency
calculations were performed on the minima and transition state to ensure Nimag ¼ 0 and 1, respectively.
b Energies given in hartrees.
c Eo ¼ Eelec þ ZPVE.
d Energy differentials given in kcal/mol.
e y ¼ ꢄ296.12 cmꢄ1
.
f y ¼ ꢄ256.4 cmꢄ1
.
g y ¼ ꢄ264.2 cmꢄ1
.
—
and 4 (R CH ). The favored invertomer of the tertiary
—
3
noted that the latter data are puzzling in the light of the
data from this study (see Table 1 and associated
comments in the Supplementary Material). Interestingly,
this intramolecular interaction (nX—sC—Cꢄsꢂ) appears
also to be significant in the neutral ground state as
reflected by various nmr parameters.18 In the extreme,
when electron demand is high, extended hyperconjuga-
tion can lead ultimately to fragmentation.19
—
ions is syn(or Z) (3 and 9; R CH ). In contrast, two
3
—
zero-order critical structures have been found for the
—
—
secondary ions (1 and 4, R H). The syn(or Z) ion is
—
favored in all cases (3; R H). Relevant aspects of their
—
geometries are displayed in Table 22 of the Supple-
mentary Material. A pertinent feature is that the
electron-deficient center (C2) is pyramidalized to varying
degrees, dependent on electron demand, with the C2—H
and C2—CH3 bonds bent towards and away from the
substituent in the E and Z ions, respectively. A further
distortion is the concomitant tilting of the C1—C2—C3
bridge in the same direction as the C2—H and C2—CH3
bonds (not listed in Table 22 of the Supplementary
Material). These distortions allow for better overlap of
the Cþ orbital with the C1—C9 (and C3—C4) and
C1—C8 (and C3—C10) bonds in the E and Z conformers,
respectively. The significant lengthening of these particu-
lar bonds relative to the other C—C bonds on the opposite
Theoretical calculations. The B3LYP/6-31Gꢂ-computed
—
—
H
—
critical structures of the ions (1,R H and 4,R
—
—
and CH ; X F and Cl) are given in Table 4. The
—
calculation of the other ions (1, R H, X F; R CH ,
3
—
—
—
—
—
—
3
—
X
F and Cl) of interest in this study were previously
—
reported in other investigations.6,20 To facilitate com-
—
parisons the corresponding results for 1 (R H) are listed
—
in Table 5. After ZPVE corrections, only a single
zero-order critical structure has been identified on the
potential energy surfaces (PES) of the tertiary ions of 16
Table 5. Calculated energiesa,b and zero-point vibrational energies (ZPVE)b of 5-substituted (X)-2-adamantyl cations (Z and E,
—
R
H)
—
c
d
Systems
—
Eelec
ZPVE
Eo
Nimag
DEo
—
Z(R H, X F)
ꢄ489.0501095
ꢄ489.0429465
ꢄ489.0436883
ꢄ849.4105028
ꢄ849.4052490
ꢄ849.4066456
0.22268
0.22247
0.22268
0.22160
0.21921
0.22144
ꢄ488.8318851
ꢄ488.8249259
ꢄ488.8254648
ꢄ849.1933338
ꢄ849.1906003
ꢄ849.1896364
0
0
—
—
EÐZ (TS)
—
1e
0
4.37
4.03
0
1.72
2.32
—
—
E(R H, X F)
—
—
—
Z(R H, X Cl)
0
—
—
EÐZ (TS)
1f
0
—
—
E(R H, X Cl)
—
—
a,b Calculations carried out at the B3LYP/6-31Gꢂ level of theory. All the structures were fully optimized, and analytical frequency calculations were performed
on the minima and transition state to ensure Nimag ¼ 0 and 1, respectively.
b Energies given in hartrees.
c Eo ¼ Eelec þ ZPVE.
d Energy differentials given in kcal/mol.
e y ¼ ꢄ286.47 cmꢄ1
.
.
f y ¼ ꢄ230.61 cmꢄ1
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 791–798
DOI: 10.1002/poc

796
W. ADCOCK AND N. A. TROUT
face in each invertomer is clearly another manifestation of
hyperconjugation. Also noticeable is the lower symmetry
of the 4-X ions (4) compared to the 5-X species (1; Cs).
This point is particularly exemplified by the NBO E(2)
energies of hyperconjugative interactions (C—C ! C2þ)
in 1 and 4. Note that the donor effects in the latter system
of the four flanking C—C bonds are all different whereas
in the former only those on the opposite face in each
invertomer are different.
interactions between the flanking C—C bonds and the
electron deficient center (C2). These parameters are set
out in Table 6. One of the more significant aspects of these
results is that the decreased donor capacity of the C3—C4
bond by the s-inductive effect in the E-cations of system 4
—
(R H) is accompanied by a concominant increased
—
donor effect from the C1—C9 bond as a result of
enhanced electron demand at the reaction center (C2). On
the other hand, as expected, the donor effects of the
C3—C4 and C1—C9 bonds of the corresponding ions of
Although the calculations are for isolated molecules in
the gas phase, the finding that only one structural
minimum exists on the PES for the tertiary ions raises the
possibility that the relative reactivity of the two faces of a
single solvated ion may determine the stereoselectivity of
these systems in solution (see above).6 This is not the case
for the secondary ions where two minima have been
located. Interestingly, Tomoda et al.21 have recently
presented an excellent linear correlation (r2 ¼ 0.97)
between the calculated (HF/6-31Gꢂ level of theory)
energy difference between Z- and E-ions of 1 and the
observed stereoselectivities (ln(syn/anti))7b in support of
the idea that (p-facial selectivity of capture of 1 is
determined by the relative stability of rapidly equilibrat-
ing pyramidalized syn(or Z) and anti(or E) epimeric ions
(2 and 3, respectively) prior to rapid capture.2,3 It should
be noted that unlike the higher level B3LYP/6-31Gꢂ
calculations in this and other studies,6,20 the lower level
HF/6-31Gꢂ calculations locate two minima on the PES of
—
system 1 (R H) are both reduced equally by extended
—
hyperconjugation. Thus, the calculations highlight an
important distinction between the substituent-induced
perturbations of hyperconjugative effects in the anti(or E)
epimeric ions of 1 and 4. The result clearly supports the
idea expressed above that this phenomenon is probably
responsible for differential electrostatic effects being the
apparent dominant factor governing facial selectivity in 4
—
(see subsection Hydrochlorination of Alkenes (6, Y
—
CH2) above). A further pertinent observation is the
considerable reduction in the energy of hyperconjugative
interactions on reducing electron demand (cf. correspond-
ing interactions of 20 and 3 ions (R H and CH ,
0
—
—
3
respectively).
Finally, by use of Eqn (1) we determined the relative
hydride affinities (kcal/mol) listed in Table 7. The results
exemplify the pronounced destabilizing effects of the
electron-withdrawing substituents on both ions (E and Z)
for each structure (1 and 4, R H):
the 30 ions (1, R CH ). Furthermore, the authors
—
—
—
—
3
assume incorrectly (see above) that the equilibration
0
HAdþ þ XAdH ! HAdH þ XAdþ
(1)
—
ZÐE involving 2 ions (1, R H) is complete for all
—
substituents.
For each epimeric pair, the effect is more pronounced for
the E-cation in accord with expectations based on field
inductive/hyperconjugative effects. However, the signifi-
cantly larger influence for Cl versus F on corresponding
ions for each structure is puzzling. Noteworthy though is
the fact that the energy difference (kcal/mol) between the
A useful way of quantitatively describing electron
delocalization interactions is by the energies of the
second-order perturbation analysis of the Fock matrix
elements in the NBO basis (E(2)).12 Consequently, we
carried out an NBO analysis of the secondary and tertiary
—
—
—
ions of 1 and 4 (Z and E, R H and Me; respectively) in
—
E- and Z-ions of 1 (X F, ꢄ3.98; X Cl, ꢄ2.38) and 4
—
—
order to obtain E(2) values for the hyperconjugative
— —
(X F, ꢄ1.47; X Cl, ꢄ1.92) roughly parallels the
—
—
Table 6. Selected NBO E(2) energies (kcal/mol)a,b of hyperconjugative interactions (C–C ! C2þ) in 1 and 4
System
C1–C8
C1–C9
C3–C4
C3–C10
—
—
—
1(Z; R H, X F)
29.03
1.17
28.07
1.17
20.04
19.48
26.27
0.92
27.85
0.92
0.59
25.90
0.74
25.97
1.85
2.04
1.05
33.73
0.90
34.47
2.46
1.92
0.59
25.90
0.74
25.97
1.85
2.04
0.97
17.98
0.77
17.07
2.21
2.26
29.03
1.17
28.07
1.17
20.04
19.48
28.07
1.87
27.20
1.57
—
—
—
—
1(E; R H, X F)
—
—
—
—
1(Z; R H, X Cl)
—
—
—
—
1(E; R H, X Cl)
—
—
—
—
1(Z; R Me, X F)
—
—
—
—
1(Z; R Me, X Cl)
—
—
—
—
4(Z; R H, X F)
—
—
—
—
4(E; R H, X F)
—
—
—
—
4(Z; R H, X Cl)
—
—
—
—
4(E; R H, X Cl)
—
—
—
—
4(Z; R Me, X F)
17.95
18.97
19.60
18.86
—
—
—
—
4(Z; R Me, X Cl)
—
a,b Only energies >0.5 kcal/mol are shown.
—
—
—
—
—
Parent ions (X H, R H): 27.03, 1.02, 1.02, 27.03. (X H, R Me): 18.18, 2.54, 2.54, 18.18.
—
—
—
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 791–798
DOI: 10.1002/poc

FACIAL SELECTIVITY IN 2-ADAMANTYL CATIONS
797
Table 7. Relative hydride affinities (DE)a,b of Some E- and
tellano G, Andreu C, Boix C, Mello R, Asensio G. Org. Lett. 2000;
2: 831–834. DOI: 10.1021/ol000017m (see also references cited
therein); (c) Gonzalez-Nunez ME, Castellano G, Andreu C, Royo
J, Baguena M, Mello R, Asensio G. J. Am. Chem. Soc. 2001; 123:
7487–7491. DOI: 10.1021/ja003667u; (d) Mehta G, Singh SR,
Gagliardini V, Priyakumar UD, Sastry GN. Tetrahedron Lett. 2001;
42: 8527–8530. DOI: 10.1016/S0040-4039(01)01818-4 (see also
references cited therein); (e) Mehta G, Singh SR, Balanarayan P,
Gadre SR. Org. Lett. 2002; 4: 2297–2300. DOI: 10.1021/
ol026005d (see also references cited therein); (f) Yadav VK, Singh
L. J. Org. Chem. 2005; 70: 692–695. DOI: 10.1021/jo048878m
—
Z-Cations of 1 and 4(R H)
—
Structure
—
—
Epimer
DE (kcal/mol)
1(X F)
E
Z
E
Z
E
Z
E
Z
7.88
3.90
10.57
8.19
6.62
5.15
9.04
7.12
—
1(X F)
—
—
1(X Cl)
—
—
1(X Cl)
—
—
4(X F)
—
—
´
(see also references cited therein); (g) Gonzalez-Nun˜ez ME, Royo
´
4(X F)
—
—
4(X Cl)
—
—
J, Mello R, Baguena M, Ferrer JM, deArellano CR, Asenio G,
Prakash GKS. J. Org. Chem. 2005; 70: 7919–7924. DOI: 10.1021/
jo0509511 (see also references cited therein); (h) Wilmot N,
Marsella MJ. Org. Lett. 2006; 14: 3109–3112. DOI: 10.1021/
ol061224a (see also references cited therein); (i) Yadav VK, Gupta
A, Balamurugan R, Sriramurthy V, Kumar NV. J. Org. Chem.
2006; 71: 4178–4182. DOI: 10.1021/jo060291f (see also refer-
ences cited therein).
4(X Cl)
—
aEffects of substituents on the energy (kcal/mol)of the following isodesmic
reaction: HAdþ þ XAdH ! HAdH þ XAdþ.
bEnergies calculated at the B3LYP/6-31Gꢂ level of theory
observed stereoselectivities (see Table 2) given that the
—
2. Kaselj M, Chung W-S, le Noble WJ. Chem. Rev. 1999; 99:
1387–1414. DOI: 10.1021/cr980364y (see also references cited
therein).
3. Adcock W, Trout NA. Chem. Rev. 1999; 99: 1415–1436. DOI:
10.1021/cr980390v (see also references cited therein).
4. (a) Knoll W, Bobek MM, Kalchhauser H, Rosenberg MG, Brinker
UH. Org. Lett. 2003; 5: 2943–2946. DOI: 10.1021/ol0350700 (see
also references cited therein); (b) Kaneno D, Tomoda S. Org. Lett.
2003; 5: 2947–2949. DOI: 10.1021/ol035073c (see also references
cited therein).
equilibria (2Ð3 and 8Ð9, R H) are probably
—
incomplete (see Table 3). It should be noted that the
differences between F and Cl in 1 and 4 are probably
manifestations of extended hyperconjugation in the
E-cations (2 and 8, R H; sCX —sC—Cꢄsþ and
ꢂ
—
—
nF—sC—Cꢄsþ, respectively).
5. Adcock W, Cotton J, Trout NA. J. Org. Chem. 1994; 59:
1867–1876 J. Org. Chem. 1995; 60: 7074–7074.
6. (a) Filippi A, Trout NA, Brunelle P, Adcock W, Sorensen TS,
Speranza M. J. Am. Chem. Soc. 2001; 123: 6396–6403. DOI:
10.1021/ja010352l; (b) Filippi A, Trout NA, Brunelle P, Adcock
W, Sorensen TS, Speranza M. J. Org. Chem. 2004; 69: 5537–5546.
DOI: 10.1021/jo04948ll.
7. (a) Adcock W, Coope J, Shiner VJ, Jr, Trout NA. J. Org. Chem.
1990; 55: 1411–1412; (b) Adcock W, Trout NA. J. Org. Chem.
1991; 56: 3229–3238.
8. (a) Grob CA, Wittwer G, Rao KR. Helv. Chim. 1985; 68: 651–660;
(b) Grob CA, Wang G, Wang C. Tetrahedron Lett. 1987; 28:
1247–1250. DOI: 10.1016/S0040-4039(00)95337-1.
9. Kaselj M, le Noble WJ. J. Org. Chem. 1996; 61: 4157–4160. DOI:
10.1021/jo9601066.
10. See References [40–43] in Reference [3].
11. Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA,
Cheeseman JR, Zakrzewski VG, Montgomery JA, Jr, Stratmann
RE, Burant JC, Dapprich S, Millam JM, Daniels AD, Kudin KN,
Strain MC, Farkas O, Tomasi J, Barone V, Cossi M, Cammi R,
Mennucci B, Pomelli C, Adamo C, Clifford S, Ochterski J,
Petersson GA, Ayala PY, Cui Q, Morokuma K, Malick DK, Rabuck
AD, Raghavachari K, Foresman JB, Cioslowski J, Ortiz JV, Baboul
AG, Stefanov BB, Liu G, Liashenko A, Piskorz P, Komaromi I,
Gomperts R, Martin RL, Fox DJ, Keith T, Al-Laham MA, Peng
CY, Nanayakkara A, Challacombe M, Gill PMW, Johnson B, Chen
W, Wong MW, Andres JL, Gonzalez C, Head-Gordon M, Replogle
ES, Pople JA. Gaussian 98, Revision A.9; Gaussian: Pittsburgh,
PA, 1998.
CONCLUSIONS
The results of this study reveal a distinct difference in the
nature of the electronic factor of polar electron-withdrawing
groups governing facial selectivity in the nucleophilic
trapping of 5- versus 4eq-substituted (X)-2-adamantyl
cations (1 and 4, respectively). Whereas the selectivity
trend (Z/E) of the former ions for the halogen subset
parallels their electronegativies (F > Cl > Br > I), by
contrast, the corresponding values for the latter ions
mirror their polar field constants (sF; F ꢀ Br ꢀ Cl ꢀ I).3 A
significant feature of this result is that if transmission of
the polar inductive effect was predominantly via
successive polarization of the C—C bonds (s-inductive
effect),22 then one would have expected the selectivity for
4 to also have followed the electronegativity of the subs-
tituents. Clearly, differential electrostatic effects (through-
space) prevail here in determining the relative stability of
the equilibrating solvated ions (8Ð9) which govern the
facial selectivity of rapid nucleophilic trapping of 4.
12. (a) Reed AE, Curtis LA, Weinhold F. Chem. Rev. 1988; 88:
899–926; (b) Weinhold F. In Encyclopedia of Computational
Chemistry, vol. 3, Schleyer PvR (ed.). Wiley: New York, 1998;
1792;; (c) Glendening ED, Reed AE, Carpenter JE, Weinhold F.
NBO Version 3.1.
Acknowledgements
13. Kirmse W, Wonner A, Allen AD, Tidwell TT. J. Am. Chem. Soc.
1992; 114: 8828–8835 (see also references cited therein).
14. Kirkwood JG. J. Chem. Phys. 1934; 2: 351–361 Kirkwood JG,
Westheimer FH. Ibid. 1938; 6: 506–512, 513–517 Ibid. 1939; 7:
437; Ehrenson S. J. Phys. Chem. 1977; 81: 1520–1526.
15. Cheung CK, Tseng LT, Lin MH, le Noble WJ. J. Am. Chem. Soc.
1986; 108: 1598–1605 1987; 109: 7239–7239.
We are grateful to the Australian Research Council for
partial financial support of this work.
REFERENCES
16. (a) Hoffmann R, Imamura A, Hehre WJ. J. Am. Chem. Soc. 1968;
90: 1499–1509; (b) Hoffmann R. Acc. Chem. Res. 1971; 4: 1–9; (c)
Gleiter R, Stohrer W-D, Hoffmann R. Helv. Chim. Acta. 1972; 55:
893–906.
1. (a) For recent reviews see: le Noble WJ, Gung BW (eds). Thematic
issue on diastereoselection. Chem. Rev. 1999; 99: 1069–1480.
DOI: 10.1021/cr9803636; (b) Gonzalez-Nunez ME, Royo J, Cas-
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 791–798
DOI: 10.1002/poc

798
W. ADCOCK AND N. A. TROUT
17. (a) Nelsen SF, Klein SJ, TrieberII DA, Ismagilov RF, Powell DR.
J. Org. Chem. 1997; 62: 6539–6546. DOI: 10.1021/jo970252r; (b)
Cieplak AJ. Chem. Rev. 1999; 99: 1265–1336. DOI: 10.1021/
cr980381n.
18. Duddeck H, Islam MR. Tetrahedron 1981; 37: 1193–
1197.
20. Fraschetti C, Novara FR, Antonello Filippi A, Trout NA, Adcock
W, Sorensen TS, Speranza M. J. Org. Chem. 2007; 72: 4077–4083.
DOI: 10.1021/jo0702140.
21. Kaneno D, Tomoda S. Tetrahedron Lett. 2004; 45: 4559–4562.
DOI: 10.1016/j.tetlet.2004.04.036.
22. (a) Exner O. J. Phys. Org. Chem. 1999; 12: 265–274. DOI:
10.1002/poc.124; (b) Charton M. J. Phys. Org. Chem. 1999; 12:
275–282. DOI: 10.1002/poc.125.
19. Sasaki T, Eguchi S, Toru T. J. Org. Chem. 1971; 36: 3460–3461
(see also references cited therein).
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 791–798
DOI: 10.1002/poc