Static and dynamic stereochemistry of tris(9-triptycyl)stannane derivatives

Article abstract of DOI:10.1246/bcsj.78.487

Static and dynamic stereochemistry of several tris(9-triptycyl)stannane derivatives Tp₃SnX (X = H, halogens, and alkyl groups) were studied. X-ray crystallography of the bromo and methyl compounds showed that the Sn atom was fundamentally tetrahedral and that the three Tp groups meshed with each other like bevel gears and formed a near-C₃ chiral conformation. The rate constants for enantiomerization of the chiral conformation in solution could be obtained by the lineshape analysis of the aromatic proton signals, even when a direct probe for enantiomerization such as a benzyl group was absent. The energy barrier to enantiomerization decreased as the substituent X became bulkier, suggesting that the ground state is more destabilized than the transition state upon introduction of a bulkier group.

Full text of DOI:10.1246/bcsj.78.487

Ó 2005 The Chemical Society of Japan
Bull. Chem. Soc. Jpn., 78, 487–497 (2005)
487
Static and Dynamic Stereochemistry of Tris(9-triptycyl)stannane
Derivatives#
ꢀ
Ryo Mikami, Yosuke Uchiyama, and Mao Minoura
Gaku Yamamoto, Shozo Ohta, Megumi Kaneko, Kaoru Mouri, Miki Ohkuma,
Department of Chemistry, School of Science, Kitasato University, Kitasato, Sagamihara, Kanagawa 228-8555
Received October 26, 2004; E-mail: gyama@kitasato-u.ac.jp
Static and dynamic stereochemistry of several tris(9-triptycyl)stannane derivatives Tp3SnX (X ¼ H, halogens, and
alkyl groups) were studied. X-ray crystallography of the bromo and methyl compounds showed that the Sn atom was
fundamentally tetrahedral and that the three Tp groups meshed with each other like bevel gears and formed a near-
C3 chiral conformation. The rate constants for enantiomerization of the chiral conformation in solution could be obtained
by the lineshape analysis of the aromatic proton signals, even when a direct probe for enantiomerization such as a benzyl
group was absent. The energy barrier to enantiomerization decreased as the substituent X became bulkier, suggesting
that the ground state is more destabilized than the transition state upon introduction of a bulkier group.
Compounds in which two 9-triptycyl (Tp) groups are attach-
ed to a central atom X, Tp2X, where X is CH2, NH, O, SiH2,
and so on, show very interesting stereochemical behavior.1
The two Tp groups are tightly meshed like bevel gears, and
correlated disrotation of the two Tp–X bonds takes place with
a very low energy barrier, while ‘‘slippage’’ of the gear, i.e.,
conrotation of the two Tp–X bonds, has a very high energy
barrier. Therefore, when a benzene ring of each Tp group is
suitably labeled, the phase relationship between the labeled
benzene rings remains intact unless the gear slippage occurs;
thus ‘‘phase’’ isomers are separately isolated if the energy bar-
rier to the gear slippage is sufficiently high. Typical examples
are given in Chart 1. The energy barriers to gearing could not
be obtained experimentally, but were estimated by molecular
two Tp groups, i.e., gear slippage, and thus the energy barrier
to any stereomutation involving the Tp–X rotations would be
significantly high.
Molecular model considerations show that Tp3X com-
pounds are highly congested, especially when the central atom
is of a low-period element such as C or N. Actually the only
compounds of this type so far reported are chlorotris(9-tripty-
cyl)germane (1) by Mislow and co-workers1c,2 and chloro-
tris(9-triptycyl)stannane (3):3 the latter was later shown to be
impure.2 Mislow and co-workers studied the stereochemistry
of 1.2 They revealed by X-ray crystallography that 1 adopted
gear-meshed chiral conformation. They resolved 1 into enan-
tiomers using liquid chromatography with a chiral column at
low temperatures, and found that racemization took place
ꢂ
mechanics calculations to be less than 4 kJ molꢁ1
.
1a The barri-
in methanol at ꢁ29 C with k ¼ 5:33 ꢃ 10ꢁ5 sꢁ1 and ꢀGz ¼
ers to gear slippage were obtained in terms of the isomeriza-
tion barriers of the phase isomers: those for compounds given
in Chart 1 were found to be 134.7, 163.6, and 175.7 kJ molꢁ1
for X ¼ CH2, NH, and O, respectively.1a The barrier heights
are naturally highly dependent on the nature of the central
atom X, the main factor being the Tp–X bond length. When
X ¼ SiH2, phase isomers could no more be isolated and the
NMR lineshape analysis gave the activation energy of 80
kJ molꢁ1 for the gear slippage.1a
It therefore seemed very intriguing to study the stereochem-
ical behavior in compounds where three Tp groups are attach-
ed to a central atom X, i.e., Tp3X.1c In these compounds, rota-
tions around the Tp–X bonds necessarily involve disrotation of
79:4 kJ molꢁ1. They did not however examine the mechanism
of the racemization, nor did they study the lineshape analysis
of the NMR spectra, although they found that compound 1
showed temperature-dependent NMR spectra.
We have recently developed a simple synthetic route to bro-
motris(9-triptycyl)stannane (2) and have succeeded in obtain-
ing various derivatives of the general formula Tp3SnX from
2, among which are compounds 3–7 (Chart 2). We therefore
decided to systematically study static and dynamic stereo-
chemistry of these compounds, by X-ray crystallography to re-
veal the molecular structures in the crystalline state and by dy-
namic NMR spectroscopy to reveal the conformational inter-
1: M = Ge, X = Cl
2: M = Sn, X = Br
3: M = Sn, X = Cl
Cl
Cl
X
X
Cl
4: M = Sn, X = F
M-X
3
5: M = Sn, X = CH3
6: M = Sn, X = CH2C6H5
7: M = Sn, X = H
Cl
Chart 1.
Chart 2.
Published on the web March 11, 2005; DOI 10.1246/bcsj.78.487

488 Bull. Chem. Soc. Jpn., 78, No. 3 (2005)
Stereochemistry of Tris(9-triptycyl)stannanes
conversion in solution. In this article, the synthesis, X-ray
crystallography, and dynamic NMR studies of a series of hal-
ostannanes 2–4 and alkyl- and hydridostannanes 5–7 are re-
ported and the stereochemical behavior of these compounds
is discussed. Some of the preliminary results have been pub-
lished.4
methylstannane).
The three Tp–Sn bonds are twisted from a regular staggered
conformation to similar extents. The three benzene rings in a
Tp group have dihedral angles of ca. ꢄ30, ꢅ90, and ꢄ150ꢂ
against the Sn–X bond axis (i.e., þ30, ꢁ90, and þ150ꢂ in
the P conformer5 and ꢁ30, þ90, and ꢁ150ꢂ in the M conform-
Results and Discussion
(a)
Syntheses.
Bromotris(9-triptycyl)stannane (2) was ob-
tained by treatment of a quarter molar amount of SnBr4 with
9-triptycyllithium prepared from 9-bromotriptycene and butyl-
lithium. A similar reaction of the 9-triptycyllithium with SnCl4
gave a ca. 7:1 mixture of chlorotris(9-triptycyl)stannane (3)
and the bromo compound 2, and the isolation of 3 in a pure
state was quite difficult as noticed by Mislow et al.2 However,
pure 3 was obtained by lithiation of 2 with t-butyllithium to
give tris(9-triptycyl)stannyllithium (8), followed by treatment
of 8 with SO2Cl2. Fluorotris(9-triptycyl)stannane (4) was ob-
tained by reaction of 2 with KF in the presence of 18-crown-
6. Methyltris(9-triptycyl)stannane (5) and benzyltris(9-tripty-
cyl)stannane (6) were synthesized by alkylation of 8 with
methyl iodide and benzyl bromide, respectively. Tris(9-tripty-
cyl)stannane (7) was prepared by reducing 2 with LiAlH4/
AlCl3.
(b)
X-ray Crystallography. Structural analysis by X-ray dif-
fraction was performed for single crystals of compounds 2 and
5. The perspective drawings of the molecular structures of
these compounds are shown in Fig. 1. Representative bond
lengths and angles are compiled in Table 1. In either com-
pound, the molecule adopts a chiral conformation with the
Sn–X bond as an approximate C3 axis, where three Tp groups
tightly intermesh with one another.
Table 1 shows that the Sn atom is tetrahedral in all the com-
pounds, though significantly distorted. The C–Sn–C angles
(av. 115ꢂ) are larger and the x–Sn–C angles (av. 103ꢂ; x de-
notes the central atom in group X) are smaller than the tetra-
hedral value of 109.5ꢂ, suggesting large steric repulsion among
ꢁ
the Tp groups. The Sn–C bond lengths are 2.29–2.33 A, signif-
ꢁ
icantly larger than the normal value (e.g., 2.144 A for tetra-
Fig. 1. The perspective drawings of X-ray molecular struc-
tures of (a) compound 2 and (b) compound 5, each in the P
conformation.
ꢂ
ꢂ
ꢁ
Table 1. Representative Bond Lengths (A), Bond Angles ( ), and Dihedral Angles ( ) Obtained by X-ray Crystallography
and MM2 Calculationsa)
C–Sn–C
x–Sn–C–C0
Compd
X
Sn–x
b)
Sn–C
x–Sn–C
b)
A
B
C
2
Br
2.528(1)
12
13
23
average
114.6(3)
1
2
3
2.318(9)
2.289(9)
2.328(9)
2.312
107.5(2)
98.1(2)
102.3(3)
102.6
30.7
38.9
24.8
31.5
31.7
ꢁ90:7
ꢁ80:0
ꢁ95:0
ꢁ88:6
ꢁ89.4
152.0
161.0
146.9
153.3
152.0
115.0(3)
116.4(3)
115.3
2.528
115.0
2.223
103.0
5
7
CH3
2.173(5)
12
13
23
average
113.7(2)
112.6(2)
116.0(2)
114.1
1
2
3
2.318(5)
2.291(5)
2.329(5)
2.313
108.0(2)
101.2(2)
103.6(2)
104.3
30.7
38.9
28.6
32.7
29.7
ꢁ88:8
ꢁ79:0
ꢁ91:7
ꢁ86:5
ꢁ89.5
150.3
159.7
148.7
152.9
150.5
2.161
1.688
112.0
2.241
104.8
H
114.9
2.184
102.3
27.8
ꢁ89.7
148.5
a) x denotes the central atom in group X. C denotes the bridgehead carbon C9 in a 9-triptycyl group, and C0 the carbon
bonded to C9. See text for A, B, and C. Values in italics are obtained by MM2 calculations. b) 1, 2, or 3 denotes an
arbitrary numbering of the three Tp groups.

G. Yamamoto et al.
Bull. Chem. Soc. Jpn., 78, No. 3 (2005)
489
~
A3
B1
C3 B3
B4
C4
A2
(d)
B2
B2
C2
A4
C1
A1
SnH
~
B4
C4
C3
A3A2
B3
C2
(c)
B1 C1
A4
A1
~
A3A2
B1
A4
C3
C4
B3
(b)
(a)
δ
B2
C2
C1
B4
A1
~
B4
C4
A2
A3
C3
B3
A4
C2
B2
C1
B1
A1
9.5
9.0
8.5
8.0
7.5
7.0
6.5
6.0
5.5
Fig. 2. The aromatic proton spectra of (a) compound 2, (b) compound 4, (c) compound 5, and (d) compound 7, in CDCl3 at 22 ꢂC
with the signal assignments. Singlets at ꢀ 5.3–5.4 are due to 10-H, and additional peaks at ꢀ ca. 7.25 are due to CHCl3.
er): the sites of these rings are hereafter referred to as A, B, and
C, respectively (Table 1).
Molecular Mechanics Studies. Molecular mechanics cal-
culations were performed using the MM2 program.6 A set of
parameters for tetrahedral tin by Horner and Newcomb7 were
A
B
C
C
incorporated and some still lacking parameters were added.
Representative lengths and angles obtained by the calculations
for compounds 2, 5, and 7 are included in Table 1.
B
X
A
C
The optimized structures for these compounds were of C3
symmetry and similar to those obtained by X-ray crystallogra-
phy mentioned above. Although the calculated bond lengths
are significantly deviated from the observed ones probably
due to the still poor Sn parameters, the bond angles and espe-
cially the dihedral angles around the Tp–Sn bonds are excel-
lently reproduced. These results, together with the fact that
these dihedral angles are almost independent of the substituent
X, will indicate that the molecular structures and conforma-
tions are mainly governed by non-bonded interactions among
the Tp groups.
B
A
X
Tp
Tp
X
A
Sn
1
B
1
2
6
Sn
6
5
A
1
2
3
Tp
Tp
B
5
6
2
3
C
C
5
4
10
4
Scheme 1.
NMR Analysis. At 22 ꢂC, all the compounds examined ex-
cept for the benzyl derivative 6 show twelve sharp signals for
By assuming that the molecules in solution adopt a C3 con-
formation similar to those found in the crystalline state or cal-
culated by MM2, namely that three benzene rings of a Tp
group, A, B, and C, have dihedral angles of ca. ꢄ30, ꢅ90,
and ꢄ150ꢂ, respectively, with respect to the Sn–X bond (See
Scheme 1 for the P conformer), twelve aromatic proton signals
of the Tp moiety were unambiguously assigned on the basis of
NOE and decoupling experiments (Table 2). The procedure
will be described below for the methyl derivative 5 as a typical
example.
1
the Tp aromatic protons in the H NMR spectra and 18 sharp
signals for the Tp aromatic carbons in the 13C NMR spectra,
indicating that the three Tp groups are equivalent, while the
three benzene rings in a Tp group are mutually nonequivalent.
1
The H NMR spectra in the aromatic region of several of the
compounds are shown in Fig. 2 together with the signal assign-
ments mentioned below. Compound 6 gives a broadened spec-
ꢂ
trum at 22 C, but shows the same spectral features as men-
tioned above at low temperatures. These features indicate that
the molecule in any compound adopts a conformation with C3
symmetry and the conformational change is frozen on the
NMR timescale.
Irradiation of the 10-H signal of 5 at ꢀ 5.36 in CDCl3 en-
hanced the intensity of three aromatic signals at ꢀ 7.25, 7.33,
and 7.47, which would naturally be assigned to either of A4,

490 Bull. Chem. Soc. Jpn., 78, No. 3 (2005)
Table 2. 1H NMR Chemical Shifts in ꢀa)
Stereochemistry of Tris(9-triptycyl)stannanes
Compd
X
2
3
4
5
6b)
7
Br
Cl
F
CH3
CH2Ph
H
A1
A2
A3
A4
9.577
7.102
7.088
7.451
9.414
7.100
7.088
7.465
9.076c)
7.079
7.094
7.487
8.276
7.035
7.073
7.471
8.211
6.698
7.010
7.470
8.670
7.025
7.086
7.484
B1
B2
B3
B4
8.268
5.729
6.663
7.260
8.151
5.718
6.659
7.268
7.582
5.700
6.640
7.274
7.878
5.622
6.618
7.249
7.938
5.702
6.687
7.312
7.146
5.612
6.591
7.241
C1
C2
C3
C4
7.490
5.822
6.774
7.355
7.524
5.888
6.796
7.380
7.650
6.152
6.863
7.451
7.719
5.890
6.740
7.325
7.760
5.890
6.777
7.364
7.914
6.197
6.851
7.415
10
X
5.338
5.352
5.400
5.356
5.422
5.405
2.177
(43.3)
4.322, 4.553d)
(49.6)
8.857
(1614)
a) Obtained at 22 ꢂC unless otherwise stated. Interproton coupling constants are not shown.
Values in parentheses are 119Sn–H coupling constants in Hz. b) Obtained at ꢁ29 ꢂC. 2JH{H
for the methylene protons is 15.6 Hz. c) JH{F ¼ 3:6 Hz. d) Methylene protons.
B4, or C4. Decoupling experiments easily revealed three sets
ly dependent on the stereochemical relationship between the
relevant nuclei, 3JC{Sn is larger than 2JC{Sn in 9-trimethylstann-
yl-9,10-dihydro-9,10-ethanoanthracene (34.1 and 5.9 Hz, re-
spectively).10 The assignments to A, B, or C were made on
the basis of the signal broadening: carbons (as well as hydro-
gens) on ring B gave the broadest signal and those on ring C
gave the sharpest one at an appropriate temperature range, as
discussed later in detail. The assignments from the signal
broadening did not contradict with those from the CH-COSY
studies for the tertiary carbons.
of the four adjacent protons. Irradiation of the methyl signal
at ꢀ 2.18 enhanced the aromatic signal at ꢀ 8.28, and this signal
was assigned to A1 because this was the only hydrogen close
to the methyl group (the distance between A1 and the methyl
ꢁ
carbon is 2.405, 2.466, or 2.500 A by X-ray crystallography
ꢁ
and 2.431 A by MM2). Irradiation of A1 enhanced the methyl
signal and the signal at ꢀ 7.88, so this aromatic signal was as-
signed to B1, because the B1 hydrogen on a flanking Tp group
is located close to A1 (the distance between A1 and B1 is
ꢁ
2.339 A by MM2). In this way, the twelve aromatic signals
of 5 were completely assigned with no contradiction as given
in Table 2.
In the fluorine compound 4, many carbons showed detecta-
ble couplings with 19F: the bridgehead carbons (9-C), C5 and
C6 carbons, and the ring-A carbons except for A4 (Table 3).
Most of them will be ascribed mainly to a through-bond mech-
anism, but the large coupling of the A1 carbon (J ¼ 21:0 Hz)
will be due mainly to a through-space mechanism. As for the
3JC{F couplings with the A6, B6, and C6 carbons, a Karplus-
type stereochemical relationship seems to hold (J of 2.7, ꢆ0,
and 1.7 Hz for the dihedral angles of 30, 90, and 150ꢂ, respec-
tively).
Similarly, for compounds 6 and 7, irradiation of the pro-
ton(s) in the X group (CH2 in 6 or Sn–H in 7) afforded the
NOE enhancement of the signal at the lowest field in the aro-
matic region, which was assigned to A1 for each compound.
For the fluoro compound 4, the lowest field signal at ꢀ 9.08
showed an additional splitting (J ¼ 3:6 Hz) assignable to a
through-space coupling with the 19F nucleus,8,9 indicating
the proximity of this proton and the fluorine, and therefore this
signal was assigned to A1. For compounds 2 and 3, the lowest-
field signal was reasonably assigned to A1, although no exper-
imental confirmation was available. The assigned chemical
shifts are compiled in Table 2.
In the 13C NMR spectra, all the 18 aromatic carbons of the
Tp moieties were separately observed and could mostly be as-
signed (Table 3). The assignments were made as follows. The
CH-COSY experiments easily enabled the assignments of the
tertiary carbon signals. For the quaternary carbons, those
showing large C–Sn couplings (J ¼ 34{53 Hz) were assigned
to those at position 5 (A5, B5, or C5) and those with smaller
couplings (J < 14 Hz) to those at position 6 (A6, B6, or
C6), on the basis of the report that, while coupling constants
between Sn and aromatic carbons in benzylstannanes are high-
It has been documented that tetra-coordinate halostannanes
adopt a penta-coordinate trigonal-bipyramidal geometry in ba-
sic solvents such as pyridine, dimethyl sulfoxide (DMSO), and
hexamethylphosphoric triamide (HMPA) by coordination of a
solvent molecule to the tin atom.11,12 The coordination is ex-
emplified by a high-field shift of the 119Sn signal, a low-field
shift of the carbon signal directly bound to Sn, and the increase
in the 1JSn{C value upon changing the solvent from CDCl3 to a
basic solvent. For example, bromotriphenylstannane showed
1
ꢀSn, ꢀC, and JSn{C (Hz) of ꢁ59:8, 136.9, and 595.0, respec-
tively, in CDCl3, and ꢁ228:4, 143.84, and 793.0, respectively,
in DMSO-d6.11 In order to study whether such an effect oper-
ates in the Tp3SnX compounds, NMR spectra were measured
in DMSO-d6, but no significant solvent effects were observed.
For example, in the bromo compound 2, ꢀSn were ꢁ43:6 and

G. Yamamoto et al.
Bull. Chem. Soc. Jpn., 78, No. 3 (2005)
491
Table 3. 13C Chemical Shifts (ꢀ) and C–Sn/C–F Coupling Constants (Hz) for Compounds 2–7a)
2 (X ¼ Br)
3 (X ¼ Cl)
4 (X ¼ F)
5 (X ¼ CH3)
6 (X ¼ CH2Ph)b)
7 (X ¼ H)
Carbon
ꢀ
J
ꢀ
J
ꢀ
J
ꢀ
J
ꢀ
J
ꢀ
J
A1
A2
A3
A4
A5
A6
129.06 26
128.94 25
128.82 —, 21.0
125.14 —, 1.7
125.63 —, 1.0
123.55 9
147.10 45, 0.5
147.36 13, 2.7
128.32 24
129.05
c)
124.99
c)
147.76d)
147.20d)
128.62 51
123.99
125.25
123.70
—
—
9
124.27
125.33
—
—
123.88
125.18
123.91
—
—
8
124.68
125.60
123.83
—
—
8
123.69 10
147.59 41
147.70 14
147.69 44, 41
147.81 14
148.55 38
148.66
147.53 35, 34
145.31
—
3
B1
B2
B3
B4
B5
B6
128.89 37
128.94 33
128.91 29
129.56 29
129.61
c)
124.85
c)
146.85d)
145.52d)
129.18 25
123.06
125.34
122.92
—
—
9
123.05
125.28
122.94
—
—
9
123.00
124.99
123.02
—
—
9
122.67
124.88
122.74
—
—
7
122.76
124.61
122.76
—
—
8
146.55 47, 44
144.68
146.70 46
144.55
147.01 46
144.49
147.24 37
145.64
147.35 39, 38
145.68
7
7
7
4
7
C1
C2
C3
C4
C5
C6
129.82 45
129.52 47
128.54 48
131.24 27
131.06
c)
125.19
c)
146.04d)
145.45d)
130.27 30
122.98
125.89
—
—
123.08
125.94
—
—
123.66
126.05
—
—
123.08
125.38
122.90
—
—
7
123.83
125.72
123.37
—
—
9
123.27 13
146.39 53, 50
143.87 10
123.39 13
146.55 53, 51
143.95 11
123.85 13
146.83 51, 0.7
144.36 11, 1.7
146.40 37
145.41
146.40 41, 40
148.24
4
8
9
10
ꢁ
i
79.86 403, 385 80.25 416, 398 76.92 437, 418, 8.1 73.54 389, 372 75.65
54.91
35.28
356, 341 68.52 429, 410
55.37 4
55.12
7
55.18
8
55.27
6
55.51 5
12.04 226
—
234, 223
139.23
132.01
127.90
125.31
31, 29
40
—
o
m
p
6
ꢂ
a) Measured at 22 C unless otherwise stated. JC{Sn values are given in plain type and JC{F values in italic type. Where two JC{Sn
values are cited, the former is for 119Sn and the latter is for 117Sn. Where one JC{Sn value is cited, the difference due to the isotopes
is not observed. ‘‘—’’ means that JC{Sn is not detected. b) Chemical shifts were measured at ꢁ29 ꢂC, while J values were measured at
22 ꢂC where cited. J values for the Tp aromatic carbons were not obtained. c) Appeared at ꢀ 122.4–123.6 and could not be assigned.
d) Tentatively assigned.
ꢁ34:8 in CDCl3 and DMSO-d6, respectively, and ꢀC for the
bridgehead carbon (C9) of the Tp moiety were 79.86 and
79.62, respectively. These data indicate that no significant co-
ordination of the solvent molecule takes place in DMSO. Such
absence of coordination will be ascribed to the steric hindrance
of coordination by the three bulky Tp groups.
Dynamic NMR Studies. When the temperature was ele-
vated, the aromatic signals of the Tp moiety showed broaden-
ing and coalescence in all compounds, although the fast-ex-
change-limit spectra were not obtained even at the highest
temperature examined (ca. 140 ꢂC). 1H NMR spectra of the
methyl compound 5 in 1,1,2,2-tetrachloroethane-d2 (TCE) at
several temperatures are shown in Fig. 3 as a typical example.
These lineshape changes will reasonably be ascribed to the
conversion of one C3 conformer to its enantiomeric conformer
by rotation of the Tp–Sn bonds, a process hereafter referred to
as enantiomerization (Scheme 2).
Although lineshape analysis of the aromatic proton signals
should afford the rate constants for the enantiomerization, it
did not seem so straightforward because the lineshape changes
are deeply affected by the mechanistic details of the site ex-
change. So the discussion will start with compound 6 that car-
ries a benzyl group as the direct probe for the enantiomeriza-
tion. The methylene protons of the benzyl group in 6 will be
diastereotopic and thus anisochronous if the enantiomerization
is frozen on the NMR timescale, affording an AB-quartet sig-
nal. As the enantiomerization becomes faster, the signal will
broaden, coalesce, and finally afford a sharp single peak. This
was actually observed in variable temperature NMR experi-
ments. The temperature dependence of the methylene signals
is shown in Fig. 4. The spectra are somewhat complicated
by the presence of the Sn-coupled signals, but the lineshapes
were excellently reproduced by simulations using the DNMR5
program.13 The best-fit theoretical spectra are also shown in
Fig. 4. The least-squares analysis of the Eyring plot gave the
kinetic parameters shown in Table 4. It should be noticed that
these data were obtained without a knowledge of the mecha-
nistic details of the enantiomerization.
Now the discussion goes back to the lineshape analysis of
the aromatic proton signals. When the enantiomerization is
slow, the three benzene rings of the Tp moiety are nonequiva-
lent, and they become indistinguishable as the enantiomeriza-
tion becomes fast on the NMR timescale. The lineshape
changes are described in terms of three-site exchange, e.g.,
among A1, B1, and C1, and thus three independent rate con-
stants are fundamentally necessary: those for the A1 ! B1,
A1 ! C1, and B1 ! C1 processes. The reverse processes
(e.g., B1 ! A1) naturally have the same rate constants as

492 Bull. Chem. Soc. Jpn., 78, No. 3 (2005)
Stereochemistry of Tris(9-triptycyl)stannanes
Fig. 3. The aromatic proton spectra of compound 5 in 1,1,2,2-tetrachloroethane-d2 at various temperatures.
mode all the benzene rings eclipse the Sn–X or Sn–Tp bond
during the rotation. Therefore, sites A and B exchange by
mode i, sites B and C exchange by mode ii, but direct exchange
between A and C never occurs by either mode.
In the enantiomerization of a molecule of the present com-
pounds, each of the three Tp–Sn bonds rotates either by mode i
or mode ii, and therefore there are four possible combinations
of modes: (a) i, i, ii; (b) i, ii, ii; (c) i, i, i; and (d) ii, ii, ii. In
Mechanism (c), all Tp–Sn bonds rotate by mode i, and thus site
C should never move to the other sites. This means that the
signals of the protons at this site should remain sharp irrespec-
tive of the temperature, which contradicts the observed NMR
lineshapes. Therefore, Mechanism (c) is rejected. Similarly
Mechanism (d) is also rejected, because site A should never
move to the other sites in this mechanism.
In Mechanism (a), the exchange between site A and site B
(mode i) takes place twice as often as the exchange between
site B and site C (mode ii), and the direct exchange between
site A and site C never occurs. The enantiomerization rate is
determined by the lifetime of the most frequently exchanging
site, i.e., site B in this case, and the rate constant for the enan-
tiomerization, ke, is given by the sum of the rate constants
kðB ! AÞ and kðB ! CÞ, where kðB ! AÞ ¼ 2kðB ! CÞ as
discussed above. Thus ke ¼ 3kðB ! CÞ ¼ 3kðB ! AÞ=2.
Therefore, a simple relationship holds among the rate con-
stants for the site exchange in terms of the rate constant for
the enantiomerization, ke, as given in Scheme 4(a). If Mecha-
nism (b) operates, the rate constants for the site exchange will
be as shown in Scheme 4(b).
X
X
P
M
Scheme 2.
the respective forward processes because the three sites have
the same populations. As explained in the next paragraph, a
simple relationship holds among these three rate constants.
For simplicity, the molecules are assumed to exist in confor-
mations with the dihedral angles of ꢄ30, ꢅ90, and ꢄ150ꢂ be-
tween the Sn–X bond and the respective benzene ring plane as
mentioned above. Small deviations of the dihedral angles from
these values do not affect the discussion below. The enantio-
merization then requires simultaneous rotation of the three
Tp–Sn bonds by 60ꢂ in either direction. Two modes of rotation
are possible for one Tp–Sn bond as shown in Scheme 3: The
indicated Tp group of the left conformer (P) in Scheme 3 ro-
tates either clockwise (mode i) or anticlockwise (mode ii). In
mode i, the gray benzene ring originally at site A moves to site
B by 60ꢂ rotation, the blank one at site B moves to site A,
while the dark one at site C does not change its site. In this
mode, any benzene ring does not eclipse the Sn–X or Sn–Tp
bond during the rotation. In mode ii, the blank benzene ring
at site B moves to site C, the dark one at site C moves to site
B, but the gray one at site A does not change its site. In this
Qualitatively, Mechanism (a) predicts that a signal due to a
proton on the ring B (Signal B) will be the broadest, Signal C
the sharpest, and Signal A intermediate among a set of three

G. Yamamoto et al.
Bull. Chem. Soc. Jpn., 78, No. 3 (2005)
493
Temp / °C
k / s−1
(300)
55
49
43
37
31
25
18
200
134
88
56
36
(21)
10
δ
4.7
4.6
4.5
4.4
4.3
4.2
Fig. 4. The observed methylene proton signals of compound 6 in CDCl3 at various temperatures (left), and the calculated spectra
with best-fit rate constants (right). Rate constants in parentheses mean that the spectra were calculated based on the data given in
Table 6.
Table 4. Free Energies of Activation and Rate Constants for the Enantiomerization at 350 K in TCE
and DMSO
TCE
DMSO
ꢀGz (350 K)
kJ molꢁ1
ꢀGz (350 K)
kJ molꢁ1
k (350 K)
sꢁ1
k (350 K)
sꢁ1
Compd
X
2
3
4
5
6
Br
Cl
F
CH3
CH2Ph
70.5
71.3
76.8
220
170
25
67.9
69.2
78.1
75.4
550
340
16
77.2
22
41
65.9a)
64.3a,b)
81.6
1:1 ꢃ 103a)
40a,b)
4.9
7
1
H
Cl
84.1
2.1
a) Solvent: CDCl3. b) At 300 K.
A
A
A
X
B
0
0
Sn
ke/3
2ke/3
ke/3
Tp
Tp
0
0
2ke/3
ke/3
X
A
i
2ke/3
2ke/3
(b)
C
B
M
M
B
C
B
C
Sn
ke/3
Tp
Tp
X
A
(a)
C
ii
B
Sn
Scheme 4.
Tp
Tp
P
C
signals when the enantiomerization takes place at a moderate
rate on the NMR timescale, while Signal A will be the sharpest
when Mechanism (b) operates. The spectra in Fig. 3, for exam-
Scheme 3.

494 Bull. Chem. Soc. Jpn., 78, No. 3 (2005)
Stereochemistry of Tris(9-triptycyl)stannanes
A1
B1
C1
Fig. 5. The observed peri-proton signals of compound 6 in CDCl3 at various temperatures (left), and the calculated spectra with
best-fit rate constants (right).
ple, seem to agree better with Mechanism (a).
lacking the potential diastereotopic groups in the X moiety
of Tp3SnX, namely, 2, 3, 4, 5, and 7, as well as to the germane
reported by Mislow and co-workers,2 Tp3GeCl (1). Tempera-
On the basis of the discussion above, the temperature-de-
pendent lineshapes of the aromatic proton signals of compound
6 were analyzed using the DNMR3 program.14 Calculations
assuming Mechanism (a) excellently reproduced the experi-
mental spectra. The observed and calculated spectra of the peri
protons (A1, B1, and C1) are shown in Fig. 5. The kinetic pa-
rameters obtained are given in Table 3; these values are in ex-
cellent agreements with those obtained from the analysis of the
methylene proton signals. This shows that the relationship in
Scheme 4(a) and thus the enantiomerization mechanism men-
tioned above are valid. Although the possibility of small con-
tamination of Mechanism (b) can not be ruled out, it can be
safely said that the extent of contamination is too small to
affect the lineshape.
Preference for Mechanism (a) will be understood as follows.
Two of the three Tp–Sn bonds would disrotate in a bevel gear
fashion, and thus one bond rotates in mode i and the other in
mode ii. Mode i would have a lower energy barrier than mode
ii, because the former mode does not require the eclipsing of
the benzene rings with the Sn–X or Sn–Tp bonds during the
rotation. Therefore, the third Tp–Sn bond would prefer mode i.
Qualitatively, the signal due to a nucleus (1H or 13C) in site
C is the sharpest, that in site B is the broadest, and that in site
A is intermediate, as typically shown in spectra at 68–88 ꢂC in
Fig. 3 and those at 25–37 ꢂC in Fig. 5. This feature enabled the
assignments of the quaternary carbons as mentioned before
(Table 3).
1
ture dependent H NMR spectra were obtained for these com-
pounds in TCE, except for compound 6 where chloroform-d
was used, and also in DMSO-d6. The lineshapes of the aromat-
ic signals were analyzed as mentioned above, and the Eyring
plots of the obtained rate constants afforded the activation pa-
rameters. The data are shown in Table 6 in the Experimental
Section, where the temperature ranges and the signals used
for the analysis are also included. The ꢀGz and k values at
350 K calculated from the parameters given in Table 6 are
compiled in Table 4.
Table 4 reveals several aspects worth consideration. For one
thing, the energy barrier to enantiomerization decreases on go-
ing from 7 to 5 to 6 in the order of increasing steric demand of
the group X. This indicates that the transition state for enantio-
merization is destabilized to a lesser extent than the ground
state upon replacing X by a bulkier group. A similar tendency
was also observed in the halogen compounds 2, 3, and 4: the
bulkier the halogen, the lower the barrier. However, 5 has a
significantly higher barrier than 3 and 4, though the effective
ꢁ 15
van der Waals radius of CH3 (1.80 A ) is similar to those
15
ꢁ
of Cl and Br (1.73 and 1.86 A, respectively ). This may be un-
derstood by taking into account the ease of the Sn–halogen
bond elongation. A C–X bond (X ¼ Cl or Br) has a smaller
stretching force constant than a C–C bond as shown by the
MM2 parameters, and so it will be reasonable to assume that
a Sn–X bond (X ¼ Cl or Br) has a smaller stretching force
constant than a Sn–C bond. When the Sn–X bond is elongated
at the transition state for enantiomerization, the Tp–Sn–Tp an-
Since the methodology for determining the enantiomeriza-
tion rate constants from the lineshape analysis of the aromatic
signals was established, this was applied to the compounds

G. Yamamoto et al.
Bull. Chem. Soc. Jpn., 78, No. 3 (2005)
495
J ¼ 7:6, 0.9 Hz, B2), 5.888 (3H, td, J ¼ 7:6, 1.1 Hz, C2), 6.659
(3H, t, J ¼ 7:3 Hz, B3), 6.796 (3H, t, J ¼ 7:3 Hz, C3), 7.088
(3H, m, A3), 7.100 (3H, m, A2), 7.268 (3H, d, J ¼ 7:4 Hz, B4),
7.380 (3H, dd, J ¼ 7:2, 0.9 Hz, C4), 7.465 (3H, m, A4), 7.524
(3H, d, J ¼ 7:6 Hz, C1), 8.151 (3H, d, J ¼ 7:6 Hz, B1), 9.412
(3H, m, A1). 119Sn NMR (CDCl3) ꢀ ꢁ1:2.
gles can more easily open up, and thus the destabilization of
the transition state will be smaller than in the cases where
the bond elongation is not easily expected, resulting in the low-
ering of the barrier.
Extrapolation of the kinetic data for 1 in Table 6 gives k ¼
1:7 ꢃ 10ꢁ5 sꢁ1 at ꢁ29 ꢂC for the enantiomerization. This val-
ue is compared with k ¼ 2:67 ꢃ 10ꢁ5 sꢁ1 calculated from the
racemization rate in methanol reported by Mislow and co-
workers. If the difference in the solvent used and in the tem-
perature range of the measurements is allowed for, these val-
ues are considered to agree very well. This agreement also will
support the validity of finding the racemization rate constants
from the lineshape analysis of the aromatic proton signals.
Fluorotris(9-triptycyl)stannane (4). To a solution of 200 mg
(0.208 mmol) of 2 in 20 mL of benzene was added 24 mg (0.416
mmol) of KF and 11 mg (0.042 mmol) of 18-crown-6, and the
mixture was heated under reflux for 24 h. GPC of the residual
mass followed by recrystallization from chloroform–hexane af-
forded 156 mg (0.174 mmol, 84%) of 4, mp 407–409 ꢂC (dec).
Found: C, 80.41; H, 4.40%. Calcd for C60H39FSn: C, 80.28; H,
4.38%. 1H NMR (CDCl3) ꢀ 5.400 (3H, s, 10-H), 5.700 (3H, td,
J ¼ 7:6, 1.3 Hz, B2), 6.152 (3H, td, J ¼ 7:6, 1.2 Hz, C2), 6.640
(3H, td, J ¼ 7:5, 0.8 Hz, B3), 6.863 (3H, t, J ¼ 7:3 Hz, C3),
7.03–7.13 (6H, m, A2 and A3), 7.274 (3H, dd, J ¼ 7:2, 1.1 Hz,
B4), 7.451 (3H, dd, J ¼ 7:3, 1.1 Hz, C4), 7.487 (3H, m, A4),
7.582 (3H, d, J ¼ 7:6 Hz, B1), 7.650 (3H, d, J ¼ 7:4 Hz, C1),
9.076 (3H, m, A1). The A1 signal appeared as a doublet
(J ¼ 3:6 Hz) upon irradiation of the A2/A3 signal at ꢀ 7.08.
119Sn NMR (CDCl3) ꢀ 30.4 (1JSn{F ¼ 2564 Hz). 19F NMR
Experimental
General.
Melting points are not corrected. 1H, 13C, and
19F NMR spectra were obtained on a Bruker ARX-300 spectrom-
eter operating at 300.1 MHz for 1H, 75.4 MHz for 13C, and 282.4
MHz for 19F, respectively. Chemical shifts are referenced with
internal tetramethylsilane (ꢀH ¼ 0:00), CDCl3 (ꢀC ¼ 77:00), and
C6F6 (ꢀF ¼ ꢁ163:0). The 13C NMR spectral data are compiled in
Table 3. In variable-temperature experiments, temperatures were
calibrated using a methanol sample or an ethylene glycol sample
and are reliable to ꢄ1 ꢂC. 119Sn NMR spectra were obtained on a
JEOL JNM-EX270 at 100.50 MHz or on a JEOL JNM-EX400 at
148.95 MHz, and chemical shifts are referenced with external
Me4Sn (ꢀ ¼ 0:00). Preparative gel permeation chromatography
(GPC) was performed on an LC-908 Liquid Chromatograph
(Japan Analytical Industry Co., Ltd.) using a series of JAIGEL
1H and 2H columns and chloroform as the eluent.
1
1
119
117
(CDCl3) ꢀ ꢁ172:5 ( J
¼ 2565 Hz, J
¼ 2451 Hz).
Sn{F
Sn{F
Methyltris(9-triptycyl)stannane (5). To a stirred solution of
200 mg (0.209 mmol) of bromotris(9-triptycyl)stannane (2) and
0.15 mL (0.84 mmol) of hexamethylphosphoric triamide (HMPA)
in 30 mL of THF at ꢁ78 ꢂC was added 0.52 mL (0.84 mmol) of t-
butyllithium in pentane (1.60 mol Lꢁ1) and the mixture was stirred
for 2 h at this temperature. To the solution was added 0.91 mL
(14.6 mmol) of methyl iodide and the mixture was allowed to
warm up to room temperature. After quenching with water, the
solvent was removed by evaporation at reduced pressure. Column
chromatography on silica gel with dichloromethane as the eluent
followed by GPC gave 120 mg (0.134 mmol, 64%) of 5, mp
386 ꢂC (dec). Found: C, 81.93; H, 4.97%. Calcd for C61H42Sn:
Bromotris(9-triptycyl)stannane (2). To a suspension of 3.33
g (10.0 mmol) of 9-bromotriptycene in 60 mL of diethyl ether was
added 6.7 mL (10.0 mmol) of butyllithium in hexane (1.58
mol Lꢁ1), and the mixture was stirred at room temperature for 4
h. The supernatant solvent was decanted off, and the residue
was washed several times with diethyl ether, and then was suspend-
ed in 120 mL of diethyl ether. To the stirred suspension was added
0.33 mL (2.5 mmol) of SnBr4 and the mixture was stirred for 32 h
at room temperature. After quenching with water, the solids were
collected by filtration to give 1.85 g (1.93 mmol, 77%) of 2, mp
370 ꢂC (dec). Found: C, 75.35; H, 3.89%. Calcd for C60H39BrSn:
C, 75.18; H, 4.10%. 1H NMR (CDCl3) ꢀ 5.338 (3H, s, 10-H),
5.729 (3H, t, J ¼ 7:3 Hz, B2), 5.822 (3H, td, J ¼ 7:6, 1.1 Hz,
C2), 6.663 (3H, t, J ¼ 7:3 Hz, B3), 6.774 (3H, t, J ¼ 7:3 Hz,
C3), 7.088 (3H, m, A3), 7.102 (3H, m, A2), 7.260 (3H, d, J ¼
7:0 Hz, B4), 7.355 (3H, d, J ¼ 7:1 Hz, C4), 7.451 (3H, m, A4),
7.490 (3H, d, J ¼ 7:6 Hz, C1), 8.268 (3H, d, J ¼ 7:6 Hz, B1),
9.577 (3H, m, A1). 119Sn NMR (CDCl3) ꢀ ꢁ43:6.
1
2
C, 81.98; H, 4.74%. H NMR (CDCl3) ꢀ 2.177 (3H, s, JSn{H
¼
43:3 Hz, CH3), 5.356 (3H, s, 10-H), 5.622 (3H, td, J ¼ 7:6, 1.2
Hz, B2), 5.890 (3H, td, J ¼ 7:6, 1.1 Hz, C2), 6.618 (3H, t, J ¼
7:1 Hz, B3), 6.740 (3H, t, J ¼ 7:0 Hz, C3), 7.073 (3H, m, A3),
7.035 (3H, m, A2), 7.249 (3H, td, J ¼ 7:1, 1.0 Hz, B4), 7.325
(3H, td, J ¼ 7:2, 1.0 Hz), 7.471 (3H, dd, J ¼ 6:8, 1.8 Hz, A4),
7.719 (3H, d, J ¼ 7:6 Hz, C1), 7.878 (3H, d, J ¼ 7:6 Hz, B1),
8.276 (3H, d, J ¼ 7:0 Hz, A1). 119Sn NMR (CDCl3) ꢀ ꢁ59:4.
Benzyltris(9-triptycyl)stannane (6). To a stirred solution of
1.00 g (1.04 mmol) of bromotris(9-triptycyl)stannane (2) and 0.55
mL (3.13 mmol) of HMPA in 150 mL of THF at ꢁ78 ꢂC was add-
ed 2.11 mL (3.13 mmol) of t-butyllithium in pentane (1.48
mol Lꢁ1) and the mixture was stirred for 2 h at this temperature.
To the solution was added 0.62 mL (5.12 mmol) of benzyl bro-
mide and the mixture was allowed to slowly warm up to room
temperature. After quenching with water, the solvent was removed
by evaporation at reduced pressure. Column chromatography on
silica gel with dichloromethane as the eluent followed by GPC
Chlorotris(9-triptycyl)stannane (3). To a stirred solution of
200 mg (0.208 mmol) of 2 and 1.1 mL (6.3 mmol) of hexamethyl-
ꢂ
phosphoric triamide in 20 mL of dry tetrahydrofuran at ꢁ78 C
under Ar was added 0.45 mL (0.63 mmol) of t-butyllithium in
pentane (1.40 mol Lꢁ1), and the mixture was stirred at this temper-
ature for 2 h. To this solution was added 0.10 mL (1.46 mmol) of
sulfuryl chloride, and the solution was left to warm up to room
temperature. After addition of 10 mL of water, the mixture was
extracted with diethyl ether. The extracts were dried over MgSO4
and evaporated. The residue was recrystallized from CHCl3 to
give 153 mg (1.70 mmol, 80%) of 3, mp 360–370 ꢂC (dec).
Found: C, 78.50; H, 4.33%. Calcd for C60H39ClSn: C, 78.84; H,
4.30%. 1H NMR (CDCl3) ꢀ 5.352 (3H, s, 10-H), 5.718 (3H, td,
ꢂ
gave 296 mg (0.305 mmol, 29%) of 6, mp 295 C (dec). Found:
C, 83.30; H, 4.88%. Calcd for C67H46Sn: C, 82.98; H, 4.78%.
1H NMR (CDCl3, ꢁ29 ꢂC) ꢀ 4.322 (1H, d, J ¼ 15:6 Hz, 2JSn{H
¼
2
49:6 Hz), 4.553 (1H, d, J ¼ 15:6 Hz, JSn{H ¼ 49:6 Hz), 5.422
(3H, s, 10-H), 5.702 (3H, td, J ¼ 7:6, 1.2 Hz, B2), 5.890 (3H,
td, J ¼ 7:6, 1.2 Hz, C2), 6.687 (3H, t, J ¼ 7:0 Hz, B3), 6.698
(3H, td, J ¼ 7:6, 1.2 Hz, A2), 6.777 (3H, t, J ¼ 7:4 Hz, C3),
6.87–6.97 (3H, m, m- and p-H), 7.010 (3H, t, J ¼ 7:4 Hz, A3),
7.312 (3H, dd, J ¼ 7:2, 1.1 Hz, B4), 7.36 (2H, m, o-H), 7.364

496 Bull. Chem. Soc. Jpn., 78, No. 3 (2005)
Stereochemistry of Tris(9-triptycyl)stannanes
(3H, dd, J ¼ 7:3, 1.2 Hz, C4), 7.470 (3H, dd, J ¼ 7:3, 1.2 Hz,
A4), 7.760 (3H, d, J ¼ 7:6 Hz, C1), 7.938 (3H, d, J ¼ 7:6 Hz,
B1), 8.221 (3H, d, J ¼ 7:7 Hz, A1). 119Sn NMR (CDCl3) ꢀ ꢁ69:4.
Tris(9-triptycyl)stannane (7). To a stirred solution of 200 mg
(0.208 mmol) of 2 in 20 mL of dry THF at 0 ꢂC was added 85 mg
(0.63 mg) of AlCl3 and then 10 mg (0.26 mmol) of LiAlH4 and the
mixture was stirred for 24 h at room temperature. The mixture was
quenched with water and extracted with diethyl ether. The extracts
were washed with water, dried over MgSO4, and evaporated.
Recrystallization of the residue from chloroform gave 115 mg
(0.131 mmol, 63%) of 7, mp 300–320 ꢂC (dec). Found: C,
81.96; H, 4.83%. Calcd for C60H40Sn: C, 81.92; H, 4.58%.
1H NMR (CDCl3) ꢀ 5.405 (3H, s, 10-H), 5.612 (3H, td, J ¼ 7:6,
1.3 Hz, B2), 6.197 (3H, td, J ¼ 7:6, 1.3 Hz, C2), 6.591 (3H, td,
J ¼ 7:5, 0.9 Hz, B3), 6.851 (3H, td, J ¼ 7:4, 0.7 Hz, C3), 7.025
(3H, td, J ¼ 7:4, 1.5 Hz, A2), 7.086 (3H, td, J ¼ 7:3, 1.1 Hz,
A3), 7.146 (3H, d, J ¼ 7:6 Hz, B1), 7.241 (3H, dd, J ¼ 7:2, 1.2
Hz, B4), 7.415 (3H, dd, J ¼ 7:3, 1.2 Hz, C4), 7.484 (3H, dd,
J ¼ 7:1, 1.4 Hz, A4), 7.914 (3H, d, J ¼ 7:5 Hz, C1), 8.670 (3H,
258285 and 226291 for compounds 2 and 5, respectively. Copies
of the data can be obtained free of charge via http//www.ccdc.
cam.ac.uk/conts/retrieving.html (or from the Cambridge Crystal-
lographic Data Centre, 12, Union Road, Cambridge, CB2 1EZ,
UK; Fax: +44 1223 336033; e-mail: ordeposit@ccdc.cam.ac.uk).
Lineshape Analysis. Total lineshape analysis was performed
by visual matching of experimental spectra with theoretical spec-
tra computed on an NEC PC9821Xs personal computer equipped
with a Mutoh PP-210 plotter. For the analysis of aromatic proton
signals, the DNMR3K program, a modified version of the
DNMR3 program14 converted for use on personal computers by
Dr. Hiroshi Kihara, was used, while the DNMR5 program13 was
used for the simulation of the methylene proton signals of com-
pound 4. Temperature dependences of chemical shift differences
and T2 values were properly taken into account. The enthalpies
and entropies of activation obtained by the least-squares analysis
of the Eyring plots of the rate constants are compiled in
Table 6.
Table 5. Crystal Data and Parameters for Data Collection,
Structure Determination, and Refinement
1
d, J ¼ 7:3 Hz, A1), 8.857 (1H, s, JSn{H ¼ 1614, 1543 Hz,
Sn–H). 119Sn NMR (CDCl3) ꢀ ꢁ246:2.
Molecular Mechanics Studies. Molecular mechanics calcula-
tions were performed employing the MM2 program6 on an NEC
EWS-4800/360 workstation. A set of parameters involving tetra-
hedral tin were incorporated according to Horner and Newcomb.7
Compound
2
5
Empirical formula
Formula weight
Crystal system
Space group
C60H39BrSn
958.59
tetragonal
P42=n
C61H42Sn 2CHCl3
ꢈ
1132.48
orthorhombic
Pbca
In addition, following parameters were newly added: Sn–Br:
ꢁ ꢁ1
3
ꢁ
Ke ¼ 2:0 mdyn A , lo ¼ 2:53 A; Br–Sn–C(sp ): Kb ¼ 0:43
ꢁ
a/A
27.7835 (2)
27.7835 (2)
13.2767 (2)
10248.6 (2)
8
1.268
0.7107
180.1
54.5
20.0474 (1)
21.7181 (2)
23.1934 (2)
10098.2 (1)
8
1.490
0.7107
130.2
54.5
mdyn radꢁ2
,
ꢂo ¼ 98:0ꢂ; Br–Sn–C(sp3)–C(sp2): V3 ¼ 0:066
ꢁ
b/A
kcal molꢁ1. These values were tentatively assigned based on the
existing parameters in the force field and those in Ref. 7.
ꢁ
c/A
V/A
ꢁ 3
X-ray Crystallography.
Crystals of compounds 2 and 5
Z
grown from dichloromethane–hexane were used. The crystal data
and the parameters for data collection, structure determination,
and refinement are summarized in Table 5. Diffraction data
were collected on a Rigaku AFC7R or a Rigaku/MSC Mercury
CCD diffractometer and calculations were performed using the
SHELXL97 program.16 The structures were solved by direct
methods followed by full-matrix least-squares refinement with
all non-hydrogen atoms anisotropic and all hydrogen atoms iso-
Dc/g cmꢁ3
ꢄ(Mo Kꢁ)/cmꢁ1
Temp/K
2’max/
No. of reflections measured
ꢂ
Total
Unique
No. of refinement variables 560
11605
5125
11278
7954
632
tropic. Reflection data with jIj > 2:0ꢃðIÞ were used. The function
Final R; Rw
GOF
0.0582; 0.0801 0.0474; 0.0983
1.877
2
minimized was ꢂwðjFoj ꢁ jFcjÞ where w ¼ ½ꢃ2ðFoÞꢇꢁ1
.
1.210
Crystallographic data have been deposited with Cambridge
Crystallographic Data Centre: Deposition number CCDC-
R ¼ ꢂjjFoj ꢁ jFcjj=ꢂjFoj, Rw on F2.
Table 6. Activation Parameters for the Enantiomerization
ꢀHz
kJ molꢁ1
ꢀSz
J molꢁ1 Kꢁ1
Temp rangeb)
Compd
X
Solventa)
Obsd signals
ꢂC
2
3
4
5
6
Br
TCE
DMSO
TCE
DMSO
TCE
DMSO
TCE
DMSO
CDCl3
68:1 ꢄ 2:4
62:0 ꢄ 0:9
66:3 ꢄ 1:5
67:7 ꢄ 0:7
67:5 ꢄ 1:3
76:8 ꢄ 0:8
75:3 ꢄ 0:9
72:6 ꢄ 0:8
54:5 ꢄ 1:9
54:7 ꢄ 0:8
82:4 ꢄ 1:9
76:3 ꢄ 2:3
ꢁ6:9 ꢄ 7:1
ꢁ16:8 ꢄ 2:7
ꢁ14:3 ꢄ 4:3
ꢁ4:5 ꢄ 2:2
ꢁ26:7 ꢄ 3:5
ꢁ3:5 ꢄ 2:1
ꢁ5:3 ꢄ 2:3
ꢁ8:0 ꢄ 2:2
ꢁ32:5 ꢄ 6:1
ꢁ31:7 ꢄ 2:6
2:6 ꢄ 4:7
48–87
32–75
47–90
A1, B1; B2, C2, B3, C3
A1, B1; B3, C3
A1, B1; B2, C2
A1, B1; B3, C3
A1, B1, C1; B2, C2: B3, C3
B2, C2
Cl
39–75
F
88–123
93–123
88–123
71–104
25–61
25–49
97–138
112–142
CH3
CH2Ph
A1, B1, C1
A1, B1; B3, C3
A1, B1, C1
CH2
A1, C1; B2, C2
A1, B1, C1
7
1
H
Cl
TCE
TCE
ꢁ22:3 ꢄ 5:7
a) TCE: CDCl2CDCl2; DMSO: (CD3)2SO. b) The temperature range where the lineshape was analyzed.

G. Yamamoto et al.
Bull. Chem. Soc. Jpn., 78, No. 3 (2005)
497
(1966).
6
399 (1983).
We thank Dr. Kei Goto and Mr. Norikiyo Nakagawa, the
University of Tokyo, for obtaining the 119Sn spectra. This
work was partially supported by a Grant-in-Aid for Scientific
Research (No. 16550043) from the Japan Society of the
Promotion of Science. The support of the Kitasato University
Research Grant for Young Researchers (to M. M.) is also
acknowledged.
N. L. Allinger and H. L. Flanagan, J. Comput. Chem., 4,
7
(1991).
8
J. H. Horner and M. Newcomb, Organometallics, 10, 1732
For the through-space mechanism in spin–spin coupling,
see: J. Hilton and L. H. Sutcliffe, Prog. NMR Spectrosc., 10, 27
(1975).
1
9
tives, see: G. Yamamoto and M. Oki, J. Org. Chem., 49, 1913
For through-space H–19F couplings in triptycene deriva-
References
¯
¯
¯
(1984); G. Yamamoto and M. Oki, Tetrahedron Lett., 26, 457
#
This paper is dedicated to Professor Michinori Oki on the
occasion of his 77th birthday.
a) H. Iwamura, J. Mol. Struct., 126, 401 (1985). b) H.
(1985).
10 W. Adcock, B. D. Gupta, W. Kitching, and D. Doddrell,
J. Organomet. Chem., 102, 297 (1975).
1
Iwamura and K. Mislow, Acc. Chem. Res., 21, 175 (1988). c)
K. Mislow, Chemtracts: Org. Chem., 2, 151 (1989).
ˇ
´
´
´ˇ
ˇ
11 J. Holecek, M. Nadvornık, K. Handlır, and A. Lycka,
2
a) J. M. Chance, J. H. Geiger, and K. Mislow, J. Am. Chem.
J. Organomet. Chem., 241, 177 (1983).
ˇ
ˇ
´ˇ
´
´
Soc., 111, 2326 (1989). b) J. M. Chance, J. H. Geiger, Y.
Okamoto, R. Aburatani, and K. Mislow, J. Am. Chem. Soc.,
112, 3540 (1990).
3 R. J. Ranson and R. M. G. Roberts, J. Organomet. Chem.,
107, 295 (1976).
12 J. Holecek, A. Lycka, K. Handlır, and M. Nadvornık,
Collect. Czech. Chem. Commun., 53, 571 (1988).
13 D. S. Stephenson and G. Binsch, QCPE Program No. 365;
QCMP Program No. 059.
14 D. A. Kleier and G. Binsch, QCPE Program No. 165.
15 G. Bott, L. D. Field, and S. Sternhell, J. Am. Chem. Soc.,
102, 5618 (1980).
4 G. Yamamoto, M. Kaneko, M. Ohkuma, and M. Minoura,
Chem. Lett., 32, 964 (2003).
5 For the usage of the M=P descriptors, see: R. S. Cahn, C.
Ingold, and V. Prelog, Angew. Chem., Int. Ed. Engl., 5, 385
16 G. M. Sheldrick, ‘‘Program for the Refinement of Crystal
Structures,’’ University of Gottingen, Germany (1997).
¨