Stereomutation and apicophilicity of diastereomeric spirophosphoranes (10-P-5)

Article abstract of DOI:10.1016/S0022-328X(01)01440-1

The apicophilicity for a series of monodentate groups were determined by deducing activation parameters of equilibration between diastereomeric spirophosphoranes bearing a Martin ligand and a modified Martin ligand. The equilibration was shown to be an in

Full text of DOI:10.1016/S0022-328X(01)01440-1

Journal of Organometallic Chemistry 643–644 (2002) 441–452
www.elsevier.com/locate/jorganchem
Stereomutation and apicophilicity of diastereomeric
spirophosphoranes (10-P-5)
Masaaki Nakamoto, Satoshi Kojima, Shiro Matsukawa, Yohsuke Yamamoto,
Kin-ya Akiba *
Department of Chemistry, Graduate School of Science, Hiroshima Uni6ersity, 1-3-1 Kagamiyama, Higashi-Hiroshima 739-8526, Japan
Received 27 July 2001; accepted 14 November 2001
Abstract
The apicophilicity for a series of monodentate groups were determined by deducing activation parameters of equilibration
between diastereomeric spirophosphoranes bearing a Martin ligand and a modified Martin ligand. The equilibration was shown
to be an intramolecular process and the order of apicophilicity of the groups turned out to be OMe:H\COMe:SMe\
NMe2\Me\n-Bu on the basis of activation enthalpy. The involvement of p-conjugation type interactions was revealed by X-ray
structures of symmetric analogs bearing COMe (acceptor) and NMe2 (donor) groups. © 2002 Published by Elsevier Science B.V.
Keywords: Spirophosphorane; Pentacoordinate; Apicophilicity; Hypervalent; NMR; X-ray analysis
results is that electronegative substituents prefer the
apical positions, while p donative or bulky ligands
prefer the equatorial positions. However, discrepancies
between series of apicophilicity provided by several
groups (deduced by different methods) have frequently
been pointed out, among the experimental ones in
particular. One of the reasons for this comes from the
fact that the experimental apicophilicity scales have
been established on the free energy of activation for
pseudorotation of phosphoranes using dynamic NMR
techniques, which is of kinetic and not of thermody-
namic nature. Thus, when interconversion processes
involve monocyclic and acyclic structures, what is actu-
ally being compared is the relative difference in energy
between the most stable TBP ground state and the
higher energy square pyramidal (SP) transition state
along the reaction coordinate. According to definition,
however, the ideal apicophilic scale should be deduced
from the comparison of two isomeric TBP states in
which the substituents in question can occupy an apical
position in one state and an equatorial position in the
other. Unfortunately, this ideal situation is attainable
only in theoretical calculations, which allow the co-exis-
tence of such isomers which may differ greatly in
energy, and in a very limited number of experimental
exceptions [9d,9e]. Furthermore, concerning the
stereoelectronic features of SP states, other than the
1. Introduction
Indications that pentacoordinate phosphorus inter-
mediates are involved in biological transformations in-
volving the phosphoryl group have stimulated studies
on model phosphorane compounds [1,2]. In order to
understand the mechanism of reactions involving these
hypervalent species, there are two aspects that must be
considered, which are characteristic of pentacoordinate
compounds in general and are not very familiar in the
chemistry of first row elements. They are pseudorota-
tion, a kinetic aspect [3], and apicophilicity, a thermo-
dynamic aspect [4]. The barrier of pseudorotation is
usually relatively low, causing rapid stereomutations in
contrast to tetracoordinate phosphorus species, which
are ordinarily stereochemically rigid [2,5]. As for api-
cophilicity, the relative preference for substituents to
occupy the apical positions as opposed to the equato-
rial positions in trigonal bipyramidal (TBP) structures,
a number of experimental results [2,6] and theoretical
calculations [7] have indicated that multiple factors are
influential. A general propensity deduced from these
* Corresponding author. Present address: Advanced Research Cen-
ter for Science and Engineering, Waseda University, 3-4-1 Ohkubo,
Shinjuku-ku, Tokyo 169-8555, Japan. Tel./fax: +81-3-5286-3165.
E-mail address: akibaky@mn.waseda.ac.jp (K.-y. Akiba).
0022-328X/02/$ - see front matter © 2002 Published by Elsevier Science B.V.
PII: S0022-328X(01)01440-1

442
M. Nakamoto et al. / Journal of Organometallic Chemistry 643–644 (2002) 441–452
theoretical study by Schleyer and co-workers [7d], there
are no reliable studies that actually link apicophilicity
(by definition) and the pseudorotation barrier. In their
analysis, it is shown that even for a simple one-step
pseudorotation process, a complete linear correlation is
not seen between the two. For spirophosphoranes with
small ring sizes (five membered or less), there have been
indications that the overall transition state (the transi-
tion state for the rate-determining pseudorotation pro-
cess) involves either a TBP configuration in which one
of the bidentates spans between two of the equatorial
positions or an SP configuration close to this TBP, and
that the energy surface in the vicinity of this configura-
tion is rather flat [2]. This implies that although the true
transition state might not be of TBP geometry, it will
be very close to a TBP in terms of energy. Thus a
spirophosphorane bearing two five-membered rings and
a monodentate ligand can be used as a suitable model
to determine apicophilicity of the monodentate ligand
because the ground state is TBP and the transition state
can also be regarded as TBP in which only the equato-
rial monodentate atom and a bidentate five-membered
ring atom have changed places. Although such
spirophosphoranes bearing five-membered rings have
been examined for pseudorotation in the past, they
were only a small set of compounds [6a,6b]. Further-
more, for this class of compounds, the analysis of
activation enthalpy, which is a more universal parame-
ter than the temperature-dependent activation free en-
ergy, has been overlooked. We have undertaken a
systematic examination of apicophilicity using
spirophosphoranes bearing monodentate substituents
with a wide range of electronegativities by utilizing
modified Martin spirophosphoranes [8,9], in which one
of the trifluoromethyl groups has been replaced with a
methyl group (1, 4–9) [10]. Herein we describe our
results [11].
gel, hexane/ether=5/1), and each diastereomer was
obtained as a crystalline solid stable to water.
1-Hydrospirophosphorane 1 was treated with several
electrophiles (acetyl chloride, elemental sulfur then
methyl iodide, methyl iodide and n-butyl bromide) in
the presence of base (DBU or NaH) to afford 1-substi-
tuted spirophosphoranes in good yields via phospho-
ranide 2. The electrophilic reaction proceeded with
essentially full retention of configuration [10]. Other
1-substituted spirophosphoranes with electronegative
groups (X=OMe, NMe2) were obtained by nucle-
ophilic substitutions of MeOH and Me2NH on 1-
chlorospirophosphorane 3, which was generated from
phosphoranide
2
with SO2Cl2. Using single
diastereomers of 1-exo, the resulting spirophosphoranes
were obtained as diastereomeric mixtures in the case of
4 and 7. This is probably due to the rapid stereomuta-
tion of 1-chlorospirophosphorane 3. Phosphoranide 2
and 1-chlorospirophosphorane 3 were also found to be
sensitive to moisture. Spirophosphoranes 4–9 were sta-
ble to moisture and heat. Owing to their relatively rigid
stereochemistry, the diastereomeric pairs could be sepa-
rated by chromatography, except for X=Me in which
the pair had identical Rf values.
The relative stereochemistries of the spirophospho-
1
ranes were assigned by difference H-NMR NOE ex-
periments. The series of diastereomers in which
irradiation of the bidentate CH3 resulted in intensity
enhancement at the monodentate bound to phosphorus
and the ortho proton of the same bidentate could be
assigned to the configuration designated exo, while the
diastereomers in which irradiation of the bidentate CH3
resulted in intensity enhancement at the two protons
ortho to phosphorus in the bidentate ligands were desig-
nated endo. Thus the relative stereochemistries could be
assigned as shown in Fig. 1.
X-ray analysis of 1-exo verified our assignment of
relative stereochemistry and showed nearly ideal TBP
structure as expected (Fig. 2). The hydrogen atom
attached to phosphorus was located on the difference
Fourier map and was found to occupy one of the
equatorial positions. It comes into notice that the more
electron-withdrawing group elongates the apical bond
2. Results and discussion
2.1. Synthesis and determination of stereochemistry
The preparation of 1-substituted spirophosphoranes
is outlined in Scheme 1. 1-Hydrospirophosphorane 1
was prepared in 60% yield with a diastereomeric ratio
of 1-exo:1-endo as ca. 2:1 by the successive treatment of
diethylaminodichlorophosphine with lithium 1,1,1,3,3,
,
(PꢀO(1)=1.766 A) compared to the other one
,
(PꢀO(2)=1.709 A). The methyl group was found to be
facing in the direction of the proton attached to
phosphorus.
Stereomutation between the diastereomeric pairs oc-
curred without any accompanying decomposition upon
heating them in solution. In principle, no change in the
chemical shifts or the shape of signals could be ob-
served in NMR spectra upon changing solvents or
temperature. This implies that the characteristic penta-
coordinate TBP structures were maintained for all the
compounds in solution.
3-hexafluoro-2-(2-lithiophenyl)-2-propoxide
(Martin
ligand) followed by lithium 1,1,1-trifluoro-2-(2-lithio-
phenyl)-2-propoxide (modified Martin ligand; from 2-
(2-bromophenyl)-1,1,1-trifluoro-2-propanol) and then
acid. The deactivated phosphine, diethylaminodichloro-
phosphine, was required to control the successive intro-
duction of the two ligands. The diastereomeric mixture
could be readily separated by chromatography (silica

M. Nakamoto et al. / Journal of Organometallic Chemistry 643–644 (2002) 441–452
443
Scheme 1.
2.2. Kinetic measurements
sponds to the most polar phosphorane, and was
thought most likely to be affected by donor solvents if
there were any affect at all. As for the latter, PꢀH
phosphoranes are generally known to be in equilibrium
with ring opened tautomers, and thus stereomutation
had the possibility to proceed via tautomeric phos-
phinites and base is expected to assist such tautomeriza-
tion. The interconversion followed exact first-order
kinetics and differences in concentrations (0.02–0.1 M)
The equilibration processes of the diastereomers were
monitored by 19F-NMR signals of the pair of com-
pounds (Scheme 2). The measurements were performed
in aromatic hydrocarbons (toluene, p-xylene or t-butyl-
toluene) as solvent. In addition, pyridine was also used
as solvent in order to investigate solvent effects in the
case of X=OMe (4) and H (1). The former corre-

444
M. Nakamoto et al. / Journal of Organometallic Chemistry 643–644 (2002) 441–452
Fig. 1. Determination of relative stereochemistry by NOE measure-
ments.
Fig. 3. Eyring plots of 1/T vs. ln(k/T) for the stereomutation of the
diastereomers. Open circles correspond to the endo to exo process
and filled squares to the exo to endo process. Solid lines are for
measurements in aromatic hydrocarbon, whereas dashed lines are for
those in pyridine.
activation parameters, especially in the entropy value,
since transient zwitterionic phosphonium ions would be
involved. Thus, the stereomutation process involving
PꢀO bond dissociation–recombination is unlikely.
Other experimental facts were supportive of this inter-
pretation. Less than 5% deuteration of 1-exo could be
observed upon dissolution of this compound in CD3OD
even after the elapse of 24 h. During this time, however,
more than 10% had undergone stereomutation to 1-
endo. On the other hand, upon addition of four equiva-
lents of pyridine to the above solution, H–D exchange
was complete within 30 min, without any apparent
change in diastereomeric ratio [31P-NMR in CD3OD
for 1-exo-D: −50.1 (JPꢀD=110 Hz), for 1-endo-D:
−48.7 (JPꢀD=107 Hz)]. Furthermore, only a small
isotope effect (kH/kD=1.1) was observed for the
stereomutation at 50 °C. These results clearly show that
there is no relation between stereomutation and H–D
exchange or PꢀO bond dissociation.
Fig. 2. ORTEP drawing of 1-exo (X=H) with thermal ellipsoids at
30% probability level.
had no effect on the rate constants for all the phospho-
ranes examined. The activation parameters of the pro-
cess were estimated from the Eyring plot of the rates
for at least four different temperatures (Fig. 3 and
Table 1).
In the case of 1, there was practically no difference in
the values of the activation parameters and the activa-
tion entropy was small regardless of the solvent used. If
dissociation and/or association were the rate-determin-
ing step, a change in the reaction profile would have
been expected. This would result in changes in the
Scheme 2.

M. Nakamoto et al. / Journal of Organometallic Chemistry 643–644 (2002) 441–452
445
Table 1
Activation parameters for stereomutation of 1-substituted spirophosphoranes
X
Solvent
exo to endo
endo to exo
DH‡ a (kcal
DS‡ a (eu)
DG‡298 (kcal
DH‡ a (kcal
DS‡ a (eu)
DG‡298 (kcal mol−1
)
mol−1
)
mol−1
)
mol−1
)
OMe
H
(4) p-Xylene
Pyridine
(1) Toluene
Pyridine
24.290.5
25.190.6
24.290.5
24.490.8
29.490.3
30.890.6
31.890.6
33.590.6
33.890.9
−3.391.6
0.391.7
−4.991.4
−3.192.5
−4.590.9
1.491.5
−3.091.4
−4.091.3
−8.791.9
25.2
25.2
25.6
25.3
30.8
30.3
33.2
34.7
36.4
24.090.5
24.990.6
24.490.5
25.190.7
30.090.4
29.990.6
32.490.6
33.590.6
33.590.9
−2.691.6
−0.691.7
−5.091.4
−2.492.3
−2.391.0
1.691.7
−2.691.4
−4.091.3
−8.891.9
24.7
24.8
25.8
25.8
30.7
29.4
32.7
34.7
36.1
COMe (5) p-Xylene
SMe (6) p-Xylene
NMe2 (7) 4-Butyltoluene
Me
n-Bu
(8) 4-Butyltoluene
(9) 4-Butyltoluene
a Accuracy is shown as standard deviation.
For phosphorane 4, activation parameters were simi-
lar in both p-xylene and pyridine just as for 1. Thus, we
can conclude that nucleophilic assistance seen for stibo-
ranes [11b] is not involved here, and stereomutation is
straightforward. For other phosphoranes, the values of
activation entropy were near zero (less than five) except
for X=n-Bu (9). These kinetic features indicate that
the stereomutation of the whole series of compounds
should be of intramolecular nature without bond disso-
ciation which would gives rise to zwitterionic phospho-
nium ions. Thus, we arrive at an order of activation
enthalpies of OMe:HBCOMe:SMeBNMe2B
Me:n-Bu (see Table 1).
All stereomutation pathways of spirophosphoranes,
1, 4–9, by Berry pseudorotation can be topologically
depicted by the diagram shown in Fig. 4 [12]. With our
compounds, the ground state spirophosphoranes are 21
or 12 (the former is exo and the latter is endo). We can
deduce from this diagram that all multi-step pseudoro-
tation processes involve a high energy phosphorane
with a diequatorial five-membered ring (species with
index n5 or 5n, which are depicted with filled squares in
Fig. 4). Among these configurations, 35, 53 and 45, 54
are expected to be least favorable since in addition to
having ring strain resulting from the diequatorial biden-
tate, they have unfavorable carbon atoms of the biden-
tate in the apical position. Thus, these isomers can be
counted out and the lowest possible energy pseudorota-
tion pathways become those given in Scheme 3. For
cyclic phosphoranes (five-membered ring), as stated
above, high energy configurations such as 15, 51 and
25, 52 can be regarded as the transition state (or at least
energetically close) [2,7e]. Therefore the value of DG‡
for the pseudorotation between the ground state phos-
phoranes (equatorial monodentate) and the transition
state (apical monodentate) could be regarded as directly
related to the relative apicophilicity of the monodentate
attached to the phosphorus. The relative apicophilicity
scale can be deduced from the results given in Table 1.
On the basis of the values of DH‡ of these compounds,
we can conclude that the relative apicophilicity of
monodentate ligand is in the order of OMe:H\
COMe:SMe\NMe2\Me:n-Bu.
One can see that this apicophilicity scale does not
follow the electronegativity scale (Table 2). This is most
obvious for the COMe and NMe2 groups. For the
COMe group it should be noted that it is more api-
cophilic than alkyl groups (Me or Bu) by 4 kcal mol−1
and less apicophilic than the OMe group having similar
electronegativity by 6 kcal mol−1. Thus, we can see
that the identity of the element directly attached to the
phosphorus atom has a large contribution on api-
cophilicity. However, stabilization of the electron-rich
apical bond by conjugation with CꢁO p* seems also to
be of significance. This notion is supported by the
X-ray structure of 5% (Fig. 5), in which the Me of 5 has
Fig. 4. Restricted Desargus–Levi diagram. The double digits corre-
spond to diastereomers and the single digits to the pivot substituent
of the pseudorotation process.

446
M. Nakamoto et al. / Journal of Organometallic Chemistry 643–644 (2002) 441–452
Scheme 3.
Table 2
Relative apicophilicity of monodentate ligands
Relative apicophilicity
OMe
:
H
\
COMe
:
SMe
\
NMe2
\
Me
:
n-Bu
DH‡ a
24.1
2.68
3.7
24.3
(2.10)
2.28
0.0
29.7
2.69
2.85
0.3
30.0
2.45
2.8
32.1
2.40
3.0
33.5
2.27
2.3
33.7
(2.28)
2.3
Electronegativity b
Group electronegativity c
d
|
0.3
0.3
0.17
−0.10
−0.10
I
a Average DH‡ of the values of exo to endo and endo to exo.
b Ref. [13].
c Ref. [14].
d Ref. [15].
been replaced by a CF3 group. As can be seen, the
carbonyl group is practically coplanar with the atoms
constituting the equatorial plane. Or in other words,
the supposedly present p-orbitals of the carbonyl group
are practically parallel with the apical bond. The possi-
bility of such a phenomenon has been hinted at in
theoretical calculations [7h].
The difference in apicophilicity between the OMe and
NMe2 groups is about 8 kcal mol−1. This implies the
presence of a large p-type interaction between the equa-
torial nitrogen lone pair and s* orbitals of the two
other equatorial PꢀC bonds, and is in good agreement
with assumptions from theoretical calculations [7d].
X-ray analysis of 7% (Fig. 6), the symmetric analog of 7
(CF3 in the place of Me) is supportive in this case also.
Thus, the nitrogen atom is practically flat (sp2), and the
NꢀC bonds are nearly coplanar with the apical bond
(implying that the lone pair on nitrogen is essentially in
the equatorial plane). Although OMe is also capable of
Fig. 5. ORTEP drawing of 5% (X=COMe) viewed along the direction
of the axis of the PꢀC(carbonyl) bond with thermal ellipsoids at 30%
probability level. Hydrogen atoms other than for the acetyl group
have been omitted for clarity.

M. Nakamoto et al. / Journal of Organometallic Chemistry 643–644 (2002) 441–452
447
group is expected to be more susceptible to steric
hindrance. For the remainder of the compound, the
moieties exerting the largest steric hindrance towards
the monodentate are the exo CF3 groups. However,
contrary to expectations from this sterical point of
view, the more hindered NMe2 compound has the
methyl groups closer to the CF3 groups. This observa-
tion supports the rationalization of the X-ray structures
based upon electronic effects as mentioned above
(Tables 3 and 4).
Hence, p-acceptor ligand COMe and p-donor ligand
NMe2 have higher equatorial site preference than ex-
pected from the electronegativity factor. Although a
good correlation between inductive substituent constant
|I and relative apicophilicity scales has been demon-
strated by Cavell and Buono, etc. [6], the correlation is
not strong in our case although activation enthalpy is
considered. Thus, we can conclude that the element
factor is of primary importance while p-type interac-
tions also contribute largely to relative apicophilicity.
The large negative entropy value for 9 (X=n-Bu)
compared with 8 (X=Me) can be accounted for by
considering the fact that, generally, the apical position
(which in this case should correspond to the transition
Fig. 6. ORTEP drawing of 7% (X=NMe2) viewed along the direction of
the axis of the PꢀN bond with thermal ellipsoids at 30% probability
level. Hydrogen atoms have been omitted for clarity.
such interactions, having lone pair electrons, because of
the highly electronegative nature of the oxygen atom,
this factor is expected to be of little significance. As for
the effect of sterics in compounds 5% and 7%, since the
P–N bond is shorter than the P–C(carbonyl) bond and
a NMe2 group is larger than a MeCO group, the former
Table 3
Crystallographic data for 1, 5%, and 7%
Compound
1 (X=H)
5% (X=COMe)
7% (X=NMe2)
Empirical formula
Formula weight
Crystal system
Space group
Crystal size (mm)
Color
C18H11F9O2P
462.25
Orthorhombic
Pbca
0.50×0.40×0.40
Colorless
Prism
C20H11F12O3P
558.26
Monoclinic
Cc
0.50×0.40×0.20
Colorless
Plate
C20H14F12NO2P
559.29
Orthorhombic
Pbca
0.55×0.50×0.30
Colorless
Prism
Habit
,
a (A)
19.501(4)
20.759(5)
9.152(2)
90
90
90
16.7000(4)
6.9890(2)
17.9830(4)
90
102.456(2)
90
7.5340(1)
21.6040(6)
26.7650(7)
90
90
90
,
b (A)
,
c (A)
h (°)
i (°)
 (°)
3
,
V (A )
3704(1)
8
2049.50(9)
4
4356.4(2)
8
Z
[
min
12.4
12.6
12.6
[
27.5
1.657
0.249
1848
28.0
1.809
0.266
1112
27.9
1.705
0.248
2240
max
Dcalc (g cm−3
)
Absorption coefficient (mm−1
F(000)
)
,
Radiation; u (A)
Mo–Ka, 0.7107
298
+h, 9k, 9l
3942
3413
3413
Mo–Ka, 0.7107
130
+h, +k, 9l
2513
2357
2357
Mo–Ka, 0.7107
130
+h, +k, +l
5582
4962
4962
Temperature (K)
Data collected
Total reflections collected
Reflections unique
Reflections used
Number of parameters refined
R, Rw
Goodness-of-fit (obs.)
271
325
325
0.0436, 0.1394
1.028
0.0429, 0.1287
1.019
0.0556, 0.2159
1.033
Maximum shift in final cycle
0.0027
0.20
0.0006
0.24
0.0013
0.19
−3
,
Final diff. map, max (e A
)

448
M. Nakamoto et al. / Journal of Organometallic Chemistry 643–644 (2002) 441–452
Table 4
Selected structural parameters for 1, 5%, and 7%
1 (X=H)
5% (X=COMe)
7% (X=NMe2)
Bond lengths
P(1)ꢀO(1)
P(1)ꢀO(2)
P(1)ꢀC(1)
P(1)ꢀC(10)
P(1)ꢀH(12)
1.710(2)
1.765(2)
1.814(3)
1.810(3)
1.286(1)
P(1)ꢀO(1)
P(1)ꢀO(2)
P(1)ꢀC(1)
P(1)ꢀC(10)
P(1)ꢀC(19)
1.761(2)
1.747(2)
1.805(3)
1.811(3)
1.870(3)
P(1)ꢀO(1)
P(1)ꢀO(2)
P(1)ꢀC(1)
P(1)ꢀC(10)
P(1)ꢀN(1)
1.739(1)
1.740(1)
1.831(2)
1.832(2)
1.658(2)
Bond angles
O(1)ꢀP(1)ꢀO(2)
O(1)ꢀP(1)ꢀC(1)
O(2)ꢀP(1)ꢀC(10)
C(1)ꢀP(1)ꢀC(10)
C(1)ꢀP(1)ꢀH(12)
C(10)ꢀP(1)ꢀH(12)
179.8(1)
89.0(1)
87.5(1)
128.7(1)
117.5(1)
113.8(1)
O(1)ꢀP(1)ꢀO(2)
O(1)ꢀP(1)ꢀC(1)
O(2)ꢀP(1)ꢀC(10)
C(1)ꢀP(1)ꢀC(10)
C(1)ꢀP(1)ꢀC(19)
C(10)ꢀP(1)ꢀC(19)
P(1)ꢀC(19)ꢀC(20)
P(1)ꢀC(19)ꢀO(3)
O(3)ꢀC(19)ꢀC(20)
178.06(10)
88.25(10)
88.2(1)
127.2(1)
114.0(1)
118.7(1)
116.8(2)
118.2(2)
125.0(3)
O(1)ꢀP(1)ꢀO(2)
O(1)ꢀP(1)ꢀC(1)
O(2)ꢀP(1)ꢀC(10)
C(1)ꢀP(1)ꢀC(10)
C(1)ꢀP(1)ꢀN(1)
C(10)ꢀP(1)ꢀN(1)
P(1)ꢀN(1)ꢀC(19)
P(1)ꢀN(1)ꢀC(20)
C(19)ꢀN(1)ꢀC(20)
172.39(6)
87.57(7)
87.29(6)
129.80(6)
115.03(7)
115.15(7)
124.0(1)
122.4(1)
112.2(2)
Torsion angles
O(1)ꢀP(1)ꢀC(19)ꢀC(20)
O(1)ꢀP(1)ꢀC(19)ꢀO(3)
O(2)ꢀP(1)ꢀC(19)ꢀC(20)
O(2)ꢀP(1)ꢀC(19)ꢀO(3)
Average torsion angle a
70.9(2)
−108.2(2)
−107.9(2)
73.0(2)
C(19)ꢀN(1)ꢀP(1)ꢀC(1)
C(19)ꢀN(1)ꢀP(1)ꢀC(10)
C(20)ꢀN(1)ꢀP(1)ꢀC(1)
C(20)ꢀN(1)ꢀP(1)ꢀC(10)
Average torsion angle a
110.5(1)
−70.8(2)
−54.8(2)
123.8(1)
62.8
17.8
a Average torsion angle from the equatorial plane.
state) is more sterically demanding than the equatorial
position, and thus requires higher order among the
substituents. The fact that activation enthalpy is essen-
tially the same for the two substituents indicates that
the electronic influence of the two substituents to api-
cophilicity is practically the same, and thus the ob-
served large difference in rates or free energies of
activation can be attributed almost completely to the
difference in relative sterical demands between the two.
This comparison presents an example where the separa-
tion of free energy of activation into enthalpy and
entropy terms is effective in deducing the relative steric
and electronic contributions to apicophilicity.
observed pseudorotation rates, hence steric factors
should not be ignored.
4. Experimental
4.1. General methods
All reactions involving air-sensitive chemicals were
carried out under N2. Ether and THF were freshly
distilled from sodium–benzophenone and all other sol-
vents were distilled from calcium hydride under an inert
atmosphere. Merck silica gel was used for column
chromatography and Merck silica gel 60 GF254 for
TLC. Melting points were measured with a Yanagi-
moto micro melting point apparatus and are uncor-
rected. 1H-NMR (400 MHz), 19F-NMR (376 MHz) and
31P-NMR (162MHz) spectra were recorded on a JEOL
EX-400 spectrometer. Chemical shifts (l) are reported
as parts per million downfield from internal Me4Si for
1H, from external trichlorofluoromethane for 19F and
from external 85% H3PO4 for 31P. Elemental analysis
was performed on a Perkin–Elmer 2400 CHN elemen-
tal analyzer.
3. Conclusion
In conclusion, through the determination of the pseu-
dorotational barriers for spirophosphoranes configura-
tionally stable at room temperature, we have
established yet another new system of pseudorotation
that we believe to be highly reflective of the energy
differences between pseudorotamers of TBP character.
Thus, apicophilicity derived from activation enthalpy is
OMe:H\COMe:SMe\NMe2\Me:n-Bu. Also
found was the importance of p-conjugation factors of
the COMe and NMe2 groups, which was supported by
X-ray structural analysis. In the case of the Me and
n-Bu groups, the use of activation enthalpy revealed
that the electronic contribution could be exaggerated in
4.2. 3-Methyl-3,3%,3%-tris(trifluoromethyl)-1u5-
1,1%(3H,3%H)-spirobi(2,1-benzoxaphosphole) (1)
To a solution of 1,1,1,3,3,3-hexafluoro-2-phenyl-2-
propanol (5.8 g, 24 mmol) in dry ether (20 ml) was

M. Nakamoto et al. / Journal of Organometallic Chemistry 643–644 (2002) 441–452
449
added n-BuLi in hexane (1.6 M, 32 ml, 51 mmol) at
−78 °C under N2, followed by tetramethylethylenedi-
amine (TMEDA, 0.28 g, 2.4 mmol). The solution was
then allowed to warm to room temperature (r.t.). After
stirring for 10 h, the dilithio salt of the alcohol which
had precipitated was dissolved with THF (20 ml) to
give a clear solution. The solution was added dropwise
via cannula to diethylaminodichlorophosphine (3.55 g,
20 mmol) in dry ether (10 ml) at −78 °C under N2. In
a separate flask, to a solution of 2-(2-bromophenyl)-
1,1,1-trifluoro-2-propanol (3.55 g, 20 mmol) in dry
ether (10 ml) was added n-BuLi in hexane (1.6 M, 28
ml, 44 mmol) at −78 °C under N2. The mixture was
subsequently allowed to warm to r.t., and stirred for 14
h. The suspension formed gave a clear yellow solution
upon addition of THF (10 ml). This solution was added
dropwise to the former mixture at −78 °C under N2,
and the resulting solution was allowed to warm to r.t.
After stirring for 4 h, the mixture was quenched with 2
M HCl, and 6 M HCl was added until the solution
became acidic. The aqueous layer was extracted with
ether (50 ml×4), the combined organic layer was
washed with H2O (10 ml) and brine (20 ml), followed
by drying over MgSO4 and filtration. The organic
solvent was then evaporated in vacuo. Purification of
the residue by column chromatography (silica gel, hex-
ane–CH2Cl2=2/1) gave hydrophosphorane 1 as a
diastereomeric mixture. Combined yield, 5.42 g (57%,
diastereomeric ratio=60: 40 (exo:endo)).
and brine, dried over MgSO4 and filtered. Then, the
organic solvent was evaporated in vacuo. The residue
was purified by chromatography (silica gel, hexane) to
give the product along with recovered material 1 (30%).
Combined yield, 216 mg (44%, diastereomeric ratio=
65:35 (exo:endo)).
1
4-exo: m.p. 128–130 °C; H-NMR (CDCl3) l 1.82 (s,
3H), 3.55 (d, JPꢀH=14.6 Hz, 3H), 7.49–7.74 (m, 6H),
8.29–8.43 (m, 2H); 19F-NMR (CDCl3) l −79.9 (s, 3F),
−75.2 (q, J=9.8 Hz, 3F), −75.0 (q, J=9.8 Hz, 3F);
31P-NMR (CDCl3)
l
−16.1; Anal. Calc. for
C19H14F9O3P: C, 46.36; H, 2.87. Found: C, 46.42; H,
2.88%.
1
4-endo: m.p. 125–127 °C; H-NMR (CDCl3) l 1.82
(s, 3H), 3.60 (d, JPꢀH=14.6 Hz, 3H), 7.52–7.74 (m,
6H), 8.31–8.42 (m, 2H); 19F-NMR (CDCl3) l −79.4
(s, 3F), −75.1 (s, 6F); 31P-NMR (CDCl3) l −18.1;
Anal. Calc. for C19H14F9O3P: C, 46.36; H, 2.87. Found:
C, 46.51; H, 2.85%.
4.4. 1-(1-Oxoethyl)-3-methyl-3,3%,3%-tris(trifluoro-
methyl)-1u5-1,1%(3H,3%H)-spirobi(2,1-benzoxaphosphole)
(5)
To solution of 1 (415 mg, 0.90 mmol, 66:34) in ether
(5 ml) was added DBU (0.17 ml, 1.2 equivalents) at r.t.
After stirring for 2 h, the solution was allowed to cool
to 0 °C and acetyl chloride (0.1 ml, 1.3 equivalents) was
added, then the mixture was stirred for 10 h at r.t. The
resulting solution was quenched with sat. NH4Cl and
followed by extraction with ether (10 ml×3). The
collected organic layer was washed with brine (5 ml),
dried over MgSO4, filtered, and concentrated in vacuo.
The crude product 5 was obtained along with starting
reactant 1 (about 25% recovered). This diastereomer
was separated and purified by column chromatography
and TLC (silica gel, hexane–CH2Cl2=2/1). Both were
recrystallized from MeCN. Combined yield, 268 mg
(58%, diastereomeric ratio=60:40 (exo:endo)).
1
1-exo: m.p. 110–111 °C; H-NMR (CDCl3) l 1.77 (s,
3H), 7.56–7.58 (m, 2H), 7.61–7.76 (m, 4H), 8.23–8.34
(m, 2H), 8.35 (d, JPꢀH=712 Hz, 1H); 19F-NMR
(CDCl3 l −79.6 (s, 3F), −76.4 (q, J=9.3 Hz, 3F),
−75.2 (q, J=9.3 Hz, 3F); 31P-NMR (CDCl3) l
−49.4; Anal. Calc. for C18H11F9O2P: C, 46.77; H, 2.62.
Found: C, 46.47; H, 2.58%.
1
1-endo: m.p. 127–128 °C; H-NMR (CDCl3) l 1.85
(s, 3H), 7.54–7.60 (m, 2H), 7.64–7.77 (m, 4H), 8.21
(dd, J=7.3 Hz, J=11.7 Hz, 1H), 8.29 (dd, J=7.3 Hz,
J=10.7 Hz, 1H), 8.36 (d, JPꢀH=725 Hz, 1H); 19F-
NMR (CDCl3) l −80.8 (s, 3F), −76.3 (q, J=9.3 Hz,
3F), −75.0 (q, J=9.3 Hz, 3F); 31P-NMR (CDCl3) l
−47.8; Anal. Calc. for C18H11F9O2P: C, 46.77; H, 2.62.
Found: C, 46.78; H, 2.42%.
1
5-exo: m.p. 147–148 °C; H-NMR (CDCl3) l 1.67 (s,
3H), 2.21 (d, JPꢀH=6.3 Hz, 3H), 7.44–7.78 (m, 6H),
8.23–8.44 (m, 2H); 19F-NMR (CDCl3).l −79.6 (s, 3F),
−75.5 (q, J=9.8 Hz, 3F), −74.8 (q, J=9.8 Hz, 3F);
31P-NMR (CDCl3).l −37.3; Anal. Calc. for
C20H14F9O3P: C, 47.64; H, 2.80. Found: C, 47.49; H,
2.89%.
4.3. 1-Methoxy-3-methyl-3,3%,3%-tris(trifluoromethyl)-
1u5-1,1%(3H,3%H)-spirobi(2,1-benzoxaphosphole) (4)
1
5-endo: m.p. 145–148 °C; H-NMR (CDCl3) l 1.85
To a solution of 1 (470 mg, 1.02 mmol, 66:34) in dry
ether (5 ml) was added DBU (0.22 ml, 1.4 mmol) at r.t.
After stirring for 1 h, sulfuryl chloride (0.17 ml, 2.0
mmol) was added at 0 °C, followed by addition of any
excess amount of methanol (3 ml). The solution was
quenched with H2O and extracted with ether (10 ml×
3). The combined organic layer was washed with H2O
(s, 3H), 2.27 (d, JPꢀH=5.9 Hz, 3H), 7.54–7.76 (m, 6H),
8.32–8.44 (m, 2H); 19F-NMR (CDCl3) l −80.6 (s, 3F),
−75.8 (q, J=9.8 Hz, 3F), −74.8 (q, J=9.8 Hz, 3F);
31P-NMR (CDCl3)
l
−39.1; Anal. Calc. for
C20H14F9O3P: C, 47.64; H, 2.80. Found: C, 47.61; H,
2.81%.

450
M. Nakamoto et al. / Journal of Organometallic Chemistry 643–644 (2002) 441–452
4.7. 1-(Dimethylamino)-3-methyl-3,3%,3%-tris(trifluoro-
4.5. 1-(1-Oxoethyl)-3,3,3%,3%-tetrakis(trifluoromethyl)-
1u5-1,1%(3H,3%H)-spirobi(2,1-benzoxaphosphole) (5%)
methyl)-1u5-1,1%(3H,3%H)-spirobi(2,1-benzoxaphosphole)
(7)
To solution of (ML)2PH (ML=the Martin ligand)
(104 mg, 0.201 mmol) in ether (2 ml) was added DBU
(0.04 ml, 1.2 equivalents) at r.t. After stirring for 10
min, the solution was allowed to cool to 0 °C and acetyl
chloride (0.02 ml, 1.3 equivalents) was added, and then
the mixture was stirred for 5 h at r.t. The resulting
solution was quenched with H2O and then extracted
with ether (25 ml×2). The collected organic layer was
washed with brine (30 ml), dried over MgSO4, filtered,
and concentrated in vacuo. The crude product was
separated by TLC (silica gel, hexane–CH2Cl2=2/1) to
afford 103 mg of 5% (92%) as a white solid. Recrystal-
lization from MeCN yielded colorless crystals of 5%.
To a solution of 1 (290 mg, 0.63 mmol, 55:45) in
THF (3 ml) was added t-BuLi in pentane (1.7 M, 0.42
ml, 1.1 equivalents) at −80 °C under N2 and the
resulting solution was allowed to warm to 0 °C. SO2Cl2
(0.08 ml, 1.5 equivalents) was then added and the
reaction mixture was stirred for 0.5 h at r.t. Dimethy-
lamine (b.p. 7 °C) which was generated from aqueous
solution of dimethylamine (40% w/w) and NaOH, and
passed through a KOH column was then bubbled into
the solution at 0 °C. The solution was quenched with 1
M HCl and extracted with ether (50 ml×3). The
resulting organic layer was dried over MgSO4, filtered
and concentrated in vacuo. The diastereomers were
separated and purified by TLC (silica gel, hexane–
CH2Cl2=5/1) to give compound 7 along with hydroxy
phosphorane (X=OH) in 38% and starting reactant 1
in 5%. The obtained product 7 was recrystallized from
MeCN to yield colorless crystals. Combined yield, 103
mg (47%, diastereomeric ratio=95:5 (exo:endo)).
1
5%: m.p. 164–166 °C; H-NMR (CDCl3) l 2.26 (d,
J
PꢀH=6.0 Hz, 3H), 7.70–7.75 (m, 6H), 8.43 (dd, J=
11.4, 6.5 Hz, 2H); 19F-NMR (CDCl3) l −74.9 (q,
J=9.2 Hz, 6F), −75.7 (q, J=9.2 Hz, 6F); 31P-NMR
(CDCl3) l −37.1; Anal. Calc. for C20H11F12O3P: C,
43.03; H, 1.99. Found: C, 42.95; H, 1.96%.
1
7-exo: m.p. 140–141 °C; H-NMR (CDCl3) l 1.77 (s,
3H), 2.65 (d, JPꢀH=11 Hz, 6H), 7.52–7.77 (m, 6H),
8.2–8.5 (m, 2H); 19F-NMR (CDCl3) l −79.5 (s, 3F),
−75.4 (q, J=8.8 Hz, 3F), −75.2 (q, J=8.8 Hz, 3F);
4.6. 3-Methyl-1-methylthio-3,3%,3%-tris(trifluoromethyl)-
1u5-1,1%(3H,3%H)-spirobi(2,1-benzoxaphosphole) (6)
31P-NMR (CDCl3)
l
−29.5; Anal. Calc. for
To a solution of 1 (456 mg, 0.98 mmol, 24:76) in
THF (15 ml) was added DBU (0.18 ml, 1.2 mmol) at
r.t. After stirring for 2 h, elemental sulfur (44 mg, 1.4
mmol) in THF was added and the mixture was stirred
for 16 h to give an orange suspension. Methyl iodide
(0.3 ml, 4.8 mmol) was then added. After stirring for 1
h, the solution was quenched with H2O, followed by
extraction with ether (10 ml×3). The organic layer was
dried over MgSO4, filtered and concentrated in vacuo
to give a crude product (diastereomeric ratio=27:73).
Diastereomeric separation and purification were
achieved by TLC (silica gel, hexane–ether=5/1) and
both were recrystallized from MeCN. Combined yield,
431 mg (86%, diastereomeric ratio=24:76 (exo:endo)).
C20H17F9NO2P: C, 47.54; H, 3.39; N, 2.77. Found: C,
47.37; H, 3.06; N, 2.70%.
7-endo: m.p. 142–143 °C; H-NMR (CDCl3) l 1.72
1
(s, 3H), 2.62 (d, JPꢀH=12 Hz, 6H), 7.50–7.75 (m, 6H),
8.2–8.4 (m, 2H); 19F-NMR (CDCl3) l −80.3 (s, 3F),
−75.4 (q, J=9.8 Hz, 3F), −75.2 (q, J=9.8 Hz, 3F);
31P-NMR (CDCl3)
l
−28.3; Anal. Calc. for
C20H17F9NO2P: C, 47.54; H, 2.77. Found: C, 47.45; H,
2.78%.
4.8. 1-(Dimethylamino)-3,3,3%,3%-tetrakis(trifluoro-
methyl)-1u5-1,1%(3H,3%H)-spirobi(2,1-benzoxaphosphole)
(7%)
To a solution of (ML)2PH (ML=the Martin ligand)
(110 mg, 0.213 mmol) in THF (2.5 ml) was added PhLi
in cyclohexane–ether (1.0 M, 0.23 ml, 1.1 equivalents)
at −78 °C under Ar and the resulting solution was
allowed to warm to 0 °C. SO2Cl2 (0.03 ml, 1.8 equiva-
lents) was then added and the reaction mixture was
stirred for 20 min at 0 C. The volatile materials were
removed in vacuo, and the residue was dissolved into
THF (2 ml) after purging Ar. An excess amount of
dimethylamine (2.0 M solution in THF, 3 ml) was
added to the solution at r.t., then the solution was
stirred overnight. The solution was quenched with 1 M
HCl and extracted with ether (30 ml×2). The resulting
organic layer was dried over MgSO4, filtered and con-
1
6-exo: m.p. 136–137 °C; H-NMR (CDCl3) l 1.82 (s,
3H), 2.21 (d, JPꢀH=18.1 Hz, 3H), 7.50–7.71 (m, 6H),
8.20–8.37 (m, 2H); 19F-NMR (CDCl3) l −79.2 (s, 3F),
−74.4 (q, J=8.6 Hz, 3F), −74.3 (q, J=8.6 Hz, 3F);
31P-NMR (CDCl3)
l
−24.1; Anal. Calc. for
C19H14F9O2PS: C, 44.89; H, 2.77. Found: C, 44.95; H,
2.55%.
1
6-endo: m.p. 123–124 °C; H-NMR (CDCl3) l 1.82
(s, 3H), 2.24 (d, JPꢀH=18.6 Hz, 3H), 7.48–7.71 (m,
6H), 8.28–8.35 (m, 2H); 19F-NMR (CDCl3) l −78.9
(s, 3F), −74.0 (s, 6F); 31P-NMR (CDCl3) l −24.8;
Anal. Calc. for C19H14F9O2PS: C, 44.89; H, 2.77.
Found: C, 44.76; H, 2.60%.

M. Nakamoto et al. / Journal of Organometallic Chemistry 643–644 (2002) 441–452
451
centrated in vacuo. The residue was purified by TLC
(silica gel, hexane) to give 44.3 mg of compound 7%
(37%). The obtained product 7% was recrystallized from
MeCN to yield colorless crystals.
yielding 9-endo (colorless oil 63%). The same procedure
was carried out for 1-exo to prepare the other
diastereomer 9-exo (63%). Combined yield, 83 mg
(63%, diastereomeric ratio= B2:\98 (exo:endo)).
1
1
7%: m.p. 136–137 °C; H-NMR (CDCl3) l 2.65 (d,
9-exo: H-NMR (CDCl3) l 0.82 (t, J=7.3 Hz, 3H),
J
PꢀH=11.1 Hz, 6H), 7.63–7.71 (m, 6H), 8.39–8.44 (m,
1.28 (m, 2H), 1.57–1.83 (m, 2H), 1.76 (s, 3H), 2.21 (dt,
2H); 19F-NMR (CDCl3) l −75.4 (br s, 12F); 31P-NMR
(CDCl3) l −25.1; Anal. Calc. for C20H14F12NO2P: C,
42.95; H, 2.52; N, 2.50. Found: C, 42.90; H, 2.50; N,
2.40%.
JPꢀH=15 Hz, J=7.3 Hz, 2H), 7.51–7.72 (m, 6H),
8.34–8.43 (m, 2H); 19F-NMR (CDCl3) l −79.67 (s,
3F), −75.4 (m, 6F); 31P-NMR (CDCl3) l −22.2;
HRMS (FAB+) Calc. for C22H20F9O2P: 519.1141.
Found: 519.1125.
1
4.9. 1,3-Dimethyl-3,3%,3%-tris(trifluoromethyl)-
9-endo: H-NMR (CDCl3) l 0.81 (t, J=7.3 Hz, 3H),
1u5-1,1%(3H,3%H)-spirobi(2,1-benzoxaphosphole) (8)
1.26 (m, 2H), 1.68–1.84 (m, 2H), 1.73 (s, 3H), 2.22
(dddd, JPꢀH=12 Hz, 2J=12 Hz, 3J=12 Hz, 3J=4 Hz,
2
3
To a solution of 1 (48 mg, 0.1 mmol, 20:80) in THF
(3 ml) was added DBU (1.8-diazabicycle[5.4.0]undec-7-
ene, 0.025 ml, 0.17 mmol) at r.t. After stirring for 2 h,
excess methyl iodide (0.2 ml, ca. 30 mmol) was added.
The solution was quenched with H2O, followed by
extraction with ether (10 ml×3). The resulting organic
layer was dried over MgSO4 and filtered. The organic
solvent was evaporated in vacuo. The residue was
recrystallized from MeCN to give 8. The diastereomers
could not be separated by TLC. However, they could
be separately prepared from their corresponding
diastereomerically pure 1. Combined yield, 31 mg (62%,
diastereomeric ratio=19:81 (exo:endo)).
1H), 2.41 (dddd, JPꢀH=17 Hz, J=12 Hz, J=12 Hz,
3J=5 Hz, 1H), 7.44–7.80 (m, 6H), 8.26–8.45 (m, 2H);
19F-NMR (CDCl3) l −80.1 (s, 3F), −75.4 (q, J=9.7
Hz, 3F), −75.0 (q, J=9.7 Hz, 3F); 31P-NMR (CDCl3)
l −22.6; Anal. Calc. for C22H20F9O2P: C, 50.98; H,
3.89. Found: C, 50.77; H, 3.90%.
4.11. Kinetic measurements
Freshly distilled toluene, p-xylene or 4-butyltoluene
was used as solvent and the diastereomerically enriched
samples were sealed in NMR tubes under N2. Kinetic
measurements of pseudorotation processes were carried
out with a JEOL EX-400 spectrometer by monitoring
the diastereomeric ratio from the integration of 19F-
NMR signals. The range of temperature of the rate
measurements were 40–70 °C for 1, 40–70 °C for 4,
90–125 °C for 5, 85–110 °C for 6, 130–170 °C for 7,
150–180 °C for 8, and 160–185 °C for 9.
1
8-exo: m.p. 116–117 °C; H-NMR (CDCl3) l 1.75 (s,
3H), 2.03 (d, JPꢀH=16.6 Hz, 3H), 7.50–7.58 (m, 2H),
7.60–7.71 (m, 4H), 8.33 (dd, J=7.8 Hz, J=11.1 Hz,
1H), 8.39 (m, 1H); 19F-NMR (CDCl3) l −79.8 (s, 3F),
−75.5 (q, J=9.8 Hz, 3F), −75.3 (q, J=9.8 Hz, 3F);
31P-NMR (CDCl3)
l
−26.9; Anal. Calc. for
C19H14F9O2P: C, 47.92; H, 2.96. Found: C, 48.04; H,
2.64%.
8-endo: m.p. 118–120 °C; H-NMR (CDCl3) l 1.75
(s, 3H), 2.11 (d, JPꢀH=16.7 Hz, 3H), 7.48–7.58 (m,
2H), 7.60–7.72 (m, 4H), 8.31 (dd, J=7.8 Hz, J=11.2
Hz, 1H), 8.40 (m, 1H); 19F-NMR (CDCl3) l −80.1 (s,
3F), −75.4 (q, J=9.8 Hz, 3F), −74.9 (q, J=9.8 Hz,
3F); 31P-NMR (CDCl3) l −26.1; Anal. Calc. for
C19H14F9O2P: C, 47.92; H, 2.96. Found: C, 47.94; H,
2.61%.
The data was analyzed based upon reversible first-or-
der kinetics using the equation of
1
ln[(xe−x0)/(xe−x)]=k−(K+1)t
where xe is the equilibrium ratio, x0 is the ratio ob-
served at t=0, x is the ratio observed at constant
intervals, k− is the rate constant (endo to exo), and K
is the equilibrium constant (K=(100−xe)/xe=k+/
k−). Least square analyses provided the rate constants
given in Table 2. The activation parameters were calcu-
lated from the Eyring plot of the rates at four different
temperatures or more.
4.10. 1-Butyl-3-methyl-3,3%,3%-tris(trifluoromethyl)-
1u5-1,1%(3H,3%H)-spirobi(2,1-benzoxaphosphole) (9)
4.12. X-ray structure determinations
To a solution of 1 (119 mg, 0.29 mmol, exo:endo=
B2:\98) in ether (3 ml) was added DBU (1.8-diaza-
bicycle[5.4.0]-undec-7-ene, 0.05 ml, 0.32 mmol) at r.t.
After stirring for 2 h, n-butyl bromide (0.06 ml, 0.58
mmol) was added. The solution was quenched with
H2O, followed by extraction with ether (10 ml×3). The
resulting organic layer was dried over MgSO4 and
filtered. The organic solvent was evaporated in vacuo.
PTLC (hexane–CH2Cl2=5/1, silica gel) was performed
For 1-exo, measurements were carried out on a Mac
Science MXC3 diffractometer with graphite-monochro-
,
mated Mo–Ka radiation (u=0.71073 A) for data col-
lection. Lattice parameters were determined by
least-square fitting of 31 reflections with 51B2qB60°.
Data were collected with the 2q/ꢀ scan mode. All data
were corrected for extinction. The structures were
solved by a direct method with the SIR-92 program [16].

452
M. Nakamoto et al. / Journal of Organometallic Chemistry 643–644 (2002) 441–452
(f) L.V. Griend, R.G. Cavell, Inorg. Chem. 22 (1983) 1817.
[7] (a) R. Hoffmann, J.M. Howell, E.L. Muetterties, J. Am. Chem.
Soc. 94 (1972) 3047;
The hydrogen atoms were included in the refinement on
,
geometrically calculated positions (CꢀH=1.0 A) riding
on their carrier atoms with isotropic thermal parame-
ters. For 5% and 7%, measurements were carried out on a
Mac Science DIP2030 imaging plate instrument
equipped with graphite-monochromated Mo–Ka radia-
(b) R.S. McDowell, A. Streitwieser Jr., J. Am. Chem. Soc. 107
(1985) 5849;
(c) J.A. Deiters, R.R. Holmes, J.M. Holmes, J. Am. Chem. Soc.
110 (1988) 7672;
(d) P. Wang, Y. Zhang, R. Glaser, A.E. Reed, P.v.R. Schleyer,
A. Streitwieser, J. Am. Chem. Soc. 113 (1991) 55;
(e) H. Wasada, K. Hirao, J. Am. Chem. Soc. 114 (1992) 16;
(f) G.R.J. Thatcher, A.S. Campbell, J. Org. Chem. 58 (1993)
2272;
(g) P. Wang, Y. Zhang, R. Glaser, A. Streitwieser, P.v.R.
Schleyer, J. Comput. Chem. 14 (1993) 522;
(h) G.R.J. Thatcher, E.S. Krol, D.R. Cameron, J. Chem. Soc.
Perkin Trans 2 (1994) 683;
(i) B.D. Wladkowski, M. Krauss, W.J. Stevens, J. Phys. Chem.
99 (1995) 4490.
,
tion (u=0.71073 A). Unit cell parameters were deter-
mined by autoindexing several images in each data set
separately with the DENZO program [17]. For each data
set, rotation images were collected in 3° increments with
a total rotation of 180° about ƒ. Data were processed
by using the SCALEPACK program [17]. The structures
were solved by a direct method and refined by full-ma-
trix least-squares methods with the TeXsan program
[16]. All non-hydrogen atoms were refined with an-
isotropic thermal parameters and were included in the
[8] (a) W.H. Stevenson III, S. Wilson, J.C. Martin, W.B. Farnham,
J. Am. Chem. Soc. 107 (1985) 6340;
,
refinement at calculated positions (CꢀH=1.0 A) riding
(b) I. Granoth, J.C. Martin, J. Am. Chem. Soc. 101 (1979) 4618
and 4623;
(c) E.F. Perozzi, R.S. Michalak, G.D. Figuly, W.H. Stevenson
III, D.B. Dess, M.R. Ross, J.C. Martin, J. Org. Chem. 46 (1981)
1049;
on their carrier atom with isotropic thermal parameters.
The structures were solved by a direct method with the
SIR-92 program.
(d) M.R. Ross, J.C. Martin, J. Am. Chem. Soc. 103 (1981) 1234;
(e) J.C. Martin, Science 221 (1983) 509;
5. Supplementary material
(f) S.K. Chopra, J.C. Martin, Heteroatom Chem. 2 (1991) 71;
(g) C.D. Moon, S.K. Chopra, J.C. Martin, in: E.N. Walsh, E.J.
Griffiths, R.W. Parry, L.D. Quin (Eds.), Phosphorus Chemistry,
Developments in American Science, ACS Symposium Series 486,
American Chemical Society: Washington, DC, 1992, pp. 128–
136.
Crystallographic data (excluding structure factors)
for the structures reported in this paper have been
deposited with the Cambridge Crystallographic Data
Center, CCDC nos. 168164 (5%), 168165 (7%), and
168166 (1-exo). Copies of this information may be
obtained free of charge from The Director, CCDC, 12
Union Road, Cambridge CB2 1EZ, UK (Fax: +44-
1223-336033; e-mail: deposit@ccdc.cam.ac.uk or www:
http://www.ccdc.cam.ac.uk).
[9] (a) S. Kojima, K. Kajiyama, K.-y. Akiba, Tetrahedron Lett. 35
(1994) 7037;
(b) S. Kojima, K. Kajiyama, K.-y. Akiba, Bull. Chem. Soc. Jpn.
68 (1995) 1785;
(c) K. Kajiyama, S. Kojima, K.-y. Akiba, Tetrahedron Lett. 37
(1996) 8409;
(d) S. Kojima, K. Kajiyama, M. Nakamoto, K.-y. Akiba, J. Am.
Chem. Soc. 118 (1996) 12866;
(e) K. Kajiyama, M. Yoshimune, M. Nakamoto, S. Matsukawa,
S. Kojima, Y. Yamamoto, K.-y. Akiba, Org. Lett. 4 (2001) 1873.
[10] Preliminary results have been reported in part in the following:
(a) S. Kojima, M. Nakamoto, K. Kajiyama, K.-y. Akiba, Tetra-
hedron Lett. 36 (1995) 2261; (b) S. Kojima, M. Nakamoto, K.
Yamazaki, K.-y. Akiba, Tetrahedron Lett. 38 (1997) 4107.
[11] (a) For the systematic examination of the pseudorotation of
pentacoordinate silicates using Martin ligands: [8a];
(b) For pentacoordinate stiboranes: S. Kojima, Y. Doi, M.
Okuda, K.-y. Akiba, Organometallics 14 (1995) 1928.
[12] K. Mislow, Acc. Chem. Res. 3 (1970) 321.
[13] (a) J.E. Huheey, J. Phys. Chem. 69 (1965) 3284;
(b) J.E. Huheey, J. Phys. Chem. 70 (1966) 2086.
[14] P.R. Wells, Prog. Phys. Org. Chem. 6 (1968) 111.
[15] M. Charton, Prog. Phys. Org. Chem. 13 (1981) 119.
[16] TeXsan: Single-Crystal Analysis Software, version 1.9; Molecu-
lar Structure Corporation: The Woodlands, TX 77381, USA,
1998. The program is available from Mac Science Co.
[17] Z. Otwinowsky, W. Minor, DENZO and SCALEPACK. in: C.W.
Carter, Jr., R.M. Sweet (Eds.), Processing of X-ray Diffraction
Data Collected in Oscillation Mode, Methods in Enzymology,
Vol. 276, Macromolecular Crystallography, Part A; Academic
Press: New York, 1997, pp. 307–326. The program is available
from Mac Science Co.
References
[1] (a) F.H. Westheimer, Acc. Chem. Res. 1 (1968) 70;
(b) G.R.J. Thatcher, R. Kluger, Adv. Phys. Org. Chem. 25
(1989) 99 and references cited therein.
[2] R.R. Holmes, Pentacoordinated Phosphorus—Structure and
Spectroscopy, Vols. I, II, ACS Monograph 175, 176, American
Chemical Society, Washington, DC, 1980.
[3] R.S. Berry, J. Chem. Phys. 32 (1960) 933.
[4] E.L. Muetterties, W. Mahler, R. Schmutzler, Inorg. Chem. 2
(1963) 613.
[5] J. Moc, K. Morokuma, J. Am. Chem. Soc. 117 (1995) 11790 and
references therein.
[6] (a) S. Trippett, Phosphorus Sulfur 1 (1976) 89;
(b) S. Trippett, Pure Appl. Chem. 40 (1970) 595;
(c) G. Buono, J.R. Llinas, J. Am. Chem. Soc. 103 (1981) 4532;
(d) M. Eisenhut, H.L. Mitchell, D.D. Traficante, R.J. Kaufman,
J.M. Deutch, G.M. Whitesides, J. Am. Chem. Soc. 96 (1974)
5385;
(e) C.G. Moreland, G.O. Doak, L.B. Littlefield, N.S. Walker,
J.W. Gilje, R.W. Braun, A.H. Cowley, J. Am. Chem. Soc. 98
(1976) 2161;