Studies in aryltin chemistry Part XIV. Crystal and molecular structures of tris(o-methoxyphenyl)tin(IV) chloride, fluoride and oxide

Article abstract of DOI:10.1016/s0020-1693(99)00342-4

Full crystal structures are reported for the tris(o-methoxyphenyl)tin compounds, (o-CH₃OC₆H₄)₃SnX [X = Cl (A), F (B)] and [(o-CH₃OC₆H₄)₃Sn]₂O (C), together

Full text of DOI:10.1016/s0020-1693(99)00342-4

www.elsevier.nl/locate/ica
Inorganica Chimica Acta 294 (1999) 224–231
Studies in aryltin chemistry
Part XIV. Crystal and molecular structures of
tris(o-methoxyphenyl)tin(IV) chloride, fluoride and oxideꢀ
Ivor Wharf *, Anne-Marie Lebuis, Gillian A. Roper
Department of Chemistry, McGill Uni6ersity, Montreal, Que´bec, Canada H3A 2K6
Received 28 June 1999; accepted 26 July 1999
Abstract
Full crystal structures are reported for the tris(o-methoxyphenyl)tin compounds, (o-CH3OC6H4)3SnX [X=Cl (A), F (B)] and
[(o-CH3OC6H4)3Sn]2O (C), together with both solid-state and solution (CDCl3) 119Sn NMR data for B and C as well as
(Mes)3SnF and [(o-Tol)3Sn]2O. Compound B like A is monomeric with a similar but not identical arrangement of aryl groups
around the tin while NMR data indicate that B has almost the same geometry in solution as in the solid state. The anomalously
lower frequency l(119Sn) value for B is consistent with the more ionic character of the SnꢀF bond compared with other SnꢀX
(X=Cl, Br, I) bonds. Compound C is the first (Ar3Sn)2O with an SnꢀOꢀSn bond angle between 140 and 180°. The comparison
2
of solid-state with solution J(119Sn,119Sn) values for C and [(o-Tol)3Sn]2O suggests that the solid state favours structures having
larger SnꢀOꢀSn angles. Reacting C with water gives (o-CH3OC6H4)3SnOH only in solution, identified by its 13C and 119Sn NMR
spectra. © 1999 Elsevier Science S.A. All rights reserved.
Keywords: Crystal structures; Solid-state/solution 119Sn NMR; Tris(o-methoxyphenyl)tin compounds
1. Introduction
monomeric (Ar=2,4,6-Me3C6H2; X=F or OH [7]).
Previously we have used o-anisyl (o-methoxyphenyl) as
a sterically demanding aryl group to prepare the
monomeric (o-MeOC6H4)3SnNCS for comparison with
polymeric Ph3SnNCS [8–10] and (p-Tol)3SnNCS [8].
We now report an analogous study of (o-
MeOC6H4)3SnX (X=F or OH) for comparison with
previous mesityl (2,4,6-trimethylphenyl)- and phenyltin
analogues.
The (o-MeOC6H4)3SnX structures (X=I [11] or
NCS [8]) reported earlier show intramolecular d(Sn···O)
values much less than van der Waals which are sugges-
tive of increased coordination at tin. In addition the
shortest Sn···O interaction is approximately trans to the
SnꢀX bond and the resulting geometry around tin
resembles, to some extent, that found in the five-
The structures of triaryltin compounds, Ar3SnX
(X=Cl, Br, I) that have been reported to date, are
almost without exception monomeric having a quasi-
tetrahedral geometry at tin [2–4], regardless of the aryl
groups present. Only when at least one of the ring
substituents itself is a potential donor, does polymer
formation occur resulting in a trans-C3SnZY trigonal-
bipyramidal geometry at the tin. One example is [p-
MeS(O2)C6H4]3SnCl where the para-methylsulfonyl-
phenyl group acts as a bridge giving an intermolecular
organotin–sulfone adduct with the weak SnꢀO interac-
tion stabilised in the solid state by polymer formation
[2]. In contrast, structures of triaryltin fluorides and
hydroxides are dependent on the steric requirements of
the aryl groups around the central tin atom, being
either polymeric (Ar=Ph; X=F [5] or OH [6]) or
coordinate
(2-carbomethoxy-1,4-cyclohexadien-1-yl)-
dimethyltin halides (X=F, Cl, Br, I) [12]. However, in
this series of compounds, a detailed comparison of the
structure of the fluoride with those of the other halides
provided evidence for an intermolecular Sn···F bond,
ꢀ For Part XIII; see Ref. [1].
* Corresponding author.
E-mail address: wharf@chemistry.mcgill.ca (I. Wharf)
0020-1693/99/$ - see front matter © 1999 Elsevier Science S.A. All rights reserved.
PII: S0020-1693(99)00342-4

I. Wharf et al. / Inorganica Chimica Acta 294 (1999) 224–231
225
,
which although borderline to van der Waals (3.64 A)
refluxed for 24 h and then evaporated under reduced
pressure until only water was left. The white solid
obtained on stirring the cooled mixture with methanol
was filtered, washed with cold ethanol, and then recrys-
tallised from CDCl3–ethanol (1:4) to give the oxide.
Yield: 45%; m.p. 143°C. IR (mull), was(SnOSn): 853
cm−1. Anal. Calc. for C42H42O7Sn2: C, 56.29; H, 4.72.
Found: C, 56.56; H, 4.72%. Crystals for X-ray analysis
were obtained by slow cooling of a hot DMF solution
of C.
and thus stretching the concept of a ‘bonding interac-
tion’, was considered sufficient for the fluoride to be
taken as a dimer with six-coordinate tin atoms. Thus,
to make a similar comparison in this work, the crystal
structures of both (o-MeOC6H4)3SnF and (o-
MeOC6H4)3SnCl were determined.
The interconversion (Ar3Sn)2O?Ar3SnOH is facile
when the Ar3Sn moiety can readily achieve a five-coor-
dinate trans-C3SnZY trigonal-bipyramidal geometry at
tin [1,13] and thus form a polymer structure like that of
Ph3SnOH [6]. This does not occur however when the
aryl groups provide more steric hindrance at tin as is
found with Ar=o-Tol, where there is no sign of hy-
droxide formation [1]. Here the effect of using Ar=o-
Anis (o-Anis=o-MeOC6H4) as a sterically demanding
aryl group on the above interconversion will be consid-
ered and the structure of the corresponding oxide is
reported for comparison with other (Ar3Sn)2O (Ar=Ph
[14] or o-Tol [15]) structures.
2.2. Crystal structure determinations
2.2.1. Data collection
For each compound a suitable crystal was selected
from those available and X-ray data were collected
(T=293(2) K) on a Rigaku AFC6S diffractometer.
Cell measurements were by standard methods. For each
compound, three standard reflections, measured every
250–150 reflections, showed 0.8–1.6% intensity decay
over the data collection period for A–C, respectively.
Full crystal parameters and data collection information
are listed in Table 1.
2. Experimental
All experimental details, including drying of solvents,
synthetic procedures, microanalyses and spectroscopic
methods (IR and both solution (CDCl3) and solid-state
NMR (119Sn, 13C)) were as given earlier [16–18]. N,N-
Dimethylformamide (DMF) was dried by refluxing
over CaH2 while potassium fluoride was heated in
vacuo until free-running before use. Tris(o-anisyl)tin
iodide (I) and chloride (A), both recrystallised from
ethanol, were previous preparations [19] as was tris(o-
tolyl)tin oxide [1]; (Mes)3SnOH and (Mes)3SnF (Mes=
2,4,6-Me3C6H2) were obtained using the literature
procedure [7].
2.2.2. Structure solution and refinement
The structures were solved by direct methods with
SHELXS-96 [20] and difference maps using SHELXL-96
[21]. For B, the absolute structure was confirmed by the
Flack parameter value of 0.036(14) [22]. All non-hydro-
gen atoms were anisotropic with H-atoms isotropic and
calculated at idealised positions using a riding model,
,
d(CꢀH) 0.93–0.98 A; for each methyl hydrogen, Uiso
was set at 50% greater than Uiso for its adjacent carbon
atom, while all other Uiso(H) values were at 1.20×
Uiso(C). Neutral atom scattering factors were from the
International Tables for Crystallography [23]. Further
solution and refinement details are in Table 1. Selected
geometric parameters and dihedral angles are reported
in Tables 2 and 3, respectively. ORTEP [24] views of the
molecules of B and C are shown in Figs. 1 and 2,
respectively.
2.1. Synthesis
2.1.1. Fluorotris(o-methoxyphenyl)tin(IV) (B)
A yellow solution of I (2.56 g) in DMF (20 ml) was
added under N2 to a slurry of potassium fluoride (3.0 g)
in DMF (25 ml). The colourless solution formed was
stirred at room temperature for 48 h and then evapo-
rated to dryness under vacuum. The residue was stirred
with water, filtered, washed with ether and dried in
vacuo; more product was obtained by adding water to
the filtrate. Recrystallisation of the crude product from
ethanol gave colourless crystals. Yield: 46%; m.p. 142–
143°C. IR (mull), w(SnꢀF): 537 cm−1. Anal. Calc. for
C21H21FO3Sn: C, 54.94; H, 4.61. Found: C, 55.25; H,
4.86%.
3. Results and discussion
3.1. Tris(o-methoxyphenyl)tin fluoride and chloride
Attempting to prepare (o-Anis)3SnF using the proce-
dure given for trimesityltin fluoride [7] i.e. by reacting
tris(o-anisyl)tin iodide with KF in aqueous acetone
always gave the crude oxideꢀhydroxide mixture (vide
infra), presumably obtained from OH− formed by reac-
tion of F− with the water present. Therefore, (o-
Anis)3SnF was prepared using dry DMF, an aprotic
medium where any trace OH− that might be formed
would also be a weaker nucleophile than the fluoride
2.1.2. v-Oxo-bis[tris(o-methoxyphenyl)tin(IV)] (C)
A mixture of I (1.40 g) in acetone (50 ml) and
potassium hydroxide (5.3 g) in water (5 ml) was

226
I. Wharf et al. / Inorganica Chimica Acta 294 (1999) 224–231
Table 1
Crystal parameters, and data collection and structure refinement information
A
B
C
Formula
C21H21ClO3Sn
475.56
monoclinic
P21/n
C
21H21FO3Sn
C42H42O7Sn2
896.21
triclinic
Formula mass
Crystal system
Space group
Cell constants
459.11
monoclinic
P21
(
P1
,
a (A)
8.412(2)
27.755(5)
8.987(10)
8.332(3)
7.9309(13)
15.562(3)
12.124(6)
14.811(8)
12.000(3)
91.44(4)
107.64(3)
106.48(4)
1954.7(15)
2; 1.523
,
b (A)
,
c (A)
h (°)
i (°)
k (°)
97.18(4)
104.44(2)
3
,
V (A )
2082(2)
4; 1.517
995.8(4)
2; 1.531
Z; Dcalc (g cm−1
)
Crystal size (mm)
0.92×0.25×0.05
1.376
0.46×0.11×0.05
10.462
0.47×0.20×0.20
1.326
v (mm−1
)
,
Radiation (u, A)
Transmission range
Data collection method
qmin; qmax (°)
Mo Ka1 (0.70930)
0.84–0.96 (psi-scans)
ꢀ scans
Cu Ka1 (1.54056)
0.05–0.19 (psi-scans)
ꢀ/2q scans
2.9; 60
Mo Ka1 (0.71069)
0.69–1.00 (psi-scans)
ꢀ/2q scans
2.7; 24.0
2.4; 26.0
h, k, l ranges
−10“10, 0“34,
−11“11
952
−9“+9, −9“9,
−18“18
460
0“14, −18“18,
−14“14
900
F(000)
No. reflections measured
No. unique reflections
No. observed reflections [I]2|(I)]
Rint
8231
4092
2897
0.079
6314
2949
2691
0.061
6460
6133
4144
0.033
No. of parameters
R1 observed (total)
wR2 observed (total)
Goodness-of-fit (S)
239
239
467
a
0.073 (0.117)
0.190 (0.215)
1.134
0.041 (0.047)
0.090 (0.094)
1.066
0.047 (0.092)
0.103 (0.119)
0.982
−3
,
Dzmax (Dzmin) (e A
)
1.349 (−1.505)
0.603 (−0.637)
0.727 (−0.845)
a R1=S(ꢀꢀFoꢀ−ꢀFcꢀꢀ)/SꢀFoꢀ. wR2=S[w(Fo2−Fc2)2/S(wFo2)2]0.5. S=S[w(Fo2−Fc2)2/(no. of reflections−no. of parameters)]0.5. Function min-
imised, S[w(Fo2−Fc2].
and the other two methoxy groups (1) and (3), cis to
the SnꢀX bond with phenyl ring (2) almost coplanar
with the C(21)ꢀSnꢀX unit [Table 3(a,b)]. However,
while detailed structures of A and I are almost identi-
cal, the SnC3X unit in B is more distorted from trigonal
symmetry and the trans O(2)···SnꢀX angle distinctly
smaller, when compared with A or I [Table 2(a,b)]. This
change may be due to the increased interaction between
X and O(1) and O(3) in B which will occur with the
flattening of the SnC3 unit on going from A to B.
Detailed comparison of ortho-substituted Ar3SnX
structures with those of analogous Ph3SnX would be
unwieldy but the results of a simpler procedure are
shown in Table 5. The change in angle sums, Sqax=
SÚCSnX (°) and Sqeq=SÚCSnC (°) [12] is equivalent
to that used earlier to assess changes in the central tin
atom environment in various triaryltin pseudohalide
structures [8,18]. Even in cases where averaged data are
given, errors are less than 92 in Sqax and Sqeq, so the
changes observed, although small, are real.
ions present. While the formation of polymeric R3SnF
(R=Ph, o-Tol, alkyl) in aqueous alcohol media [25,26]
must in large part be due to their very low solubilities,
the factors governing the preparation of monomeric
Ar3SnF are clearly more subtle. However, once iso-
lated, (o-Anis)3SnF can be obtained unchanged from
protic media such as water or ethanol showing the
SnꢀF bond, once formed, is stable to protic solvolysis.
IR data [w(SnꢀF)] suggest that (o-Anis)3SnF is
monomeric like (Mes)3SnF (536 cm−1) [7]. This conclu-
sion is further supported by the very similar solid-state
and solution NMR results for each compound (Table
4(a)). However, unlike the mesityl case, the 119Sn NMR
spectrum of solid (o-Anis)3SnF consists of one doublet
caused by coupling to a single fluorine, indicating only
one molecule per asymmetric unit.
X-ray analysis of (o-Anis)3SnF (B) confirms that it is
indeed monomeric with the shortest intermolecular
,
Sn···F distance being 6.865 A. The arrangement of
o-anisyl groups around tin (Fig. 1) is the same as that
found for the isothiocyanate [8], the chloride (A) and
the iodide (I) [11]. This motif has one CH3O(2) trans
All Ph3SnX show almost the same distortion from
inner tetrahedral geometry with Sqeq\Sqax. However,

I. Wharf et al. / Inorganica Chimica Acta 294 (1999) 224–231
Table 2 (continued)
227
Table 2
,
Selected geometric parameters (A, °)
C(11)ꢀSnꢀC(21)
C(21)ꢀSnꢀC(31)
C(11)ꢀSnꢀC(31)
109.9(3)
118.6(3)
117.3(3)
C(41)ꢀSnꢀC(51)
C(51)ꢀSnꢀC(61)
C(41)ꢀSnꢀC(61)
110.7(3)
118.0(3)
110.2(3)
(a) Chlorotris(o-methoxyphenyl)tin(IV) (A)
SnꢀCl
2.371(5)
2.129(12)
1.361(14)
1.38(2)
SnꢀC(11)
SnꢀC(31)
O(1)ꢀC(17)
O(2)ꢀC(27)
O(3)ꢀC(37)
Sn···O(2)
Cl···O(1)
2.129(11)
2.105(13)
1.438(14)
1.42(2)
SnꢀC(21)
O(1)ꢀC(16)
O(2)ꢀC(26)
O(3)ꢀC(36)
Sn···O(1)
Sn···O(3)
Cl···O(2)
Sn(1)ꢀC(11)ꢀC(12) 119.2(6)
Sn(1)ꢀC(11)ꢀC(16) 121.8(6)
Sn(1)ꢀC(21)ꢀC(22) 120.3(6)
Sn(1)ꢀC(21)ꢀC(26) 121.0(6)
Sn(1)ꢀC(31)ꢀC(32) 120.6(6)
Sn(1)ꢀC(31)ꢀC(36) 121.3(6)
C(16)ꢀO(11)ꢀC(17) 120.1(8)
C(26)ꢀO(21)ꢀC(27) 119.9(7)
C(36)ꢀO(31)ꢀC(37) 120.3(8)
O(11)ꢀC(16)ꢀC(11) 114.8(7)
O(11)ꢀC(16)ꢀC(15) 124.3(8)
O(21)ꢀC(26)ꢀC(21) 115.5(7)
O(21)ꢀC(26)ꢀC(25) 123.5(8)
O(31)ꢀC(36)ꢀC(31) 115.3(7)
O(31)ꢀC(36)ꢀC(33) 122.5(9)
O(11)···Sn(1)ꢀO(1) 145.8(2)
O(21)···Sn(1)ꢀO(1) 102.3(2)
O(31)···Sn(1)ꢀO(1) 149.7(2)
Sn(2)ꢀC(41)ꢀC(42) 123.5(6)
Sn(2)ꢀC(41)ꢀC(46) 118.0(6)
Sn(2)ꢀC(51)ꢀC(52) 123.0(6)
Sn(2)ꢀC(51)ꢀC(56) 119.2(6)
Sn(2)ꢀC(61)ꢀC(62) 119.7(5)
Sn(2)ꢀC(61)ꢀC(66) 120.7(5)
C(46)ꢀO(41)ꢀC(47) 119.2(7)
C(56)ꢀO(51)ꢀC(57) 120.2(7)
C(66)ꢀO(61)ꢀC(67) 117.9(6)
O(41)ꢀC(46)ꢀC(41) 115.1(7)
O(41)ꢀC(46)ꢀC(45) 123.9(7)
O(51)ꢀC(56)ꢀC(51) 115.5(7)
O(51)ꢀC(56)ꢀC(55) 121.6(8)
O(61)ꢀC(66)ꢀC(61) 115.3(6)
O(61)ꢀC(66)ꢀC(65) 123.6(7)
1.36(2)
1.44(2)
3.090(9)
3.061(11)
5.180(10)
2.974(10)
3.534(10)
3.676(10)
Cl···O(3)
C(11)ꢀSnꢀCl
C(31)ꢀSnꢀCl
110.3(3)
106.6(4)
112.0(5)
118.7(8)
124.5(9)
121.8(11)
118.4(10)
118.3(13)
123.7(11)
125.2(12)
124(2)
C(21)ꢀSnꢀCl
100.9(3)
113.5(4)
112.8(5)
119.7(8)
116.8(9)
120.4(10)
119.3(12)
116.3(10)
113.4(11)
113.8(11)
79.5(2)
C(11)ꢀSnꢀC(21)
C(11)ꢀSnꢀC(31)
SnꢀC(11)ꢀC(16)
SnꢀC(21)ꢀC(26)
SnꢀC(31)ꢀC(36)
C(26)ꢀO(2)ꢀC(27)
O(1)ꢀC(16)ꢀC(11)
O(2)ꢀC(26)ꢀC(21)
O(3)ꢀC(36)ꢀC(31)
O(1)···SnꢀCl
C(21)ꢀSnꢀC(31)
SnꢀC(11)ꢀC(12)
SnꢀC(21)ꢀC(22)
SnꢀC(31)ꢀC(32)
C(16)ꢀO(1)ꢀC(17)
C(36)ꢀO(3)ꢀC(37)
O(1)ꢀC(16)ꢀC(15)
O(2)ꢀC(26)ꢀC(25)
O(3)ꢀC(36)ꢀC(35)
O(2)···SnꢀCl
O(41)···Sn(2)ꢀO(1)
O(51)···Sn(2)ꢀO(1)
O(61)···Sn(2)ꢀO(1) 147.9(20
78.4(2)
82.8(2)
Sn(1)ꢀOꢀSn(2)
167.0(4)
151.3(2)
O(3)···SnꢀCl
84.2(2)
(b) Fluorotris(o-methoxyphenyl)tin(IV) (B)
SnꢀF
1.972(5)
2.120(9)
1.365(11)
1.370(13)
1.382(11)
3.088(8)
3.108(7)
4.806(10)
SnꢀC(11)
SnꢀC(31)
O(1)ꢀC(17)
O(2)ꢀC(27)
O(3)ꢀC(37)
Sn···O(2)
F···O(1)
2.112(8)
2.129(8)
1.432(14)
1.423(11)
1.426(12)
3.026(8)
3.456(10)
3.333(10)
SnꢀC(21)
O(1)ꢀC(16)
O(2)ꢀC(26)
O(3)ꢀC(36)
Sn···O(1)
Sn···O(3)
F···O(2)
the general trend is for the C3Sn pyramid to steepen as
Cl is replaced by Br and then I, presumably due to
o-H···X interactions. Such intramolecular ortho-effects
are of much greater import in (Mes)3SnX where the
geometry around tin reflects the balance of repulsions
between proximal o-CH3 groups and interactions be-
tween distal o-CH3 groups with the axial X atom. The
result is for the C3Sn pyramid to be flatter than in the
Ph3SnX case and an increase in d(SnꢀI) [31]. At this
time X-ray data for (Mes)3SnX (X=Cl, Br) are not
available to show whether the small change in geometry
around tin in going from the fluoride to the iodide is
gradual over the series, or abrupt as F is replaced by
Cl. This abrupt change however occurs in the ortho-an-
isyl series. Thus, A and I have effectively the same
environment at tin which resembles that for Ph3SnX.
However, for B, Sqeq and Sqax values differ signifi-
cantly from those for A or I, resembling more those
found for (Mes)3SnX.
F···O(3)
C(11)ꢀSnꢀF
C(31)ꢀSnꢀF
104.9(3)
106.5(3)
107.9(3)
121.0(6)
121.6(7)
120.9(6)
117.4(13)
118.4(8)
124.7(14)
125.0(10)
122.4(9)
147.2(3)
C(21)ꢀSnꢀF
98.4(3)
124.7(3)
112.2(3)
119.8(8)
119.0(7)
120.7(6)
118.1(9)
115.0(13)
114.2(10)
115.6(7)
83.0(3)
C(11)ꢀSnꢀC(21)
C(11)ꢀSnꢀC(31)
SnꢀC(11)ꢀC(16)
SnꢀC(21)ꢀC(26)
SnꢀC(31)ꢀC(36)
C(26)ꢀO(2)ꢀC(27)
O(1)ꢀC(16)ꢀC(11)
O(2)ꢀC(26)ꢀC(21)
O(3)ꢀC(36)ꢀC(31)
O(1)···SnꢀF
C(21)ꢀSnꢀC(31)
SnꢀC(11)ꢀC(12)
SnꢀC(21)ꢀC(22)
SnꢀC(31)ꢀC(32)
C(16)ꢀO(1)ꢀC(17)
C(36)ꢀO(3)ꢀC(37)
O(1)ꢀC(16)ꢀC(15)
O(2)ꢀC(26)ꢀC(25)
O(3)ꢀC(36)ꢀC(35)
O(2)···SnꢀF
O(3)···SnꢀF
78.5(2)
(c) Oxobis[tris(o-methoxyphenyl)tin(IV)] (C)
Sn(1)ꢀO(1)
1.940(5)
2.136(7)
2.133(7)
2.128(7)
1.344(10)
1.418(10)
1.368(9)
1.427(10)
1.374(10)
1.414(10)
3.127(7)
3.119(7)
3.133(6)
3.260(8)
3.100(11)
Sn(2)ꢀO(1)
1.915(5)
2.143(8)
2.111(8)
2.141(7)
1.372(9)
1.414(9)
1.375(9)
1.432(10)
1.364(9)
1.435(9)
3.052(6)
3.057(7)
3.117(5)
3.398(9)
3.096(9)
Sn(1)ꢀC(11)
Sn(1)ꢀC(21)
Sn(1)ꢀC(31)
O(11)ꢀC(16)
O(11)ꢀC(17)
O(21)ꢀC(26)
O(21)ꢀC(27)
O(31)ꢀC(36)
O(31)ꢀC(37)
Sn(1)···O(11)
Sn(1)···O(21)
Sn(1)···O(31)
O(1)···O(41)
O(31)···O(11)
Sn(2)ꢀC(41)
Sn(2)ꢀC(51)
Sn(2)ꢀC(61)
O(41)ꢀC(46)
O(41)ꢀC(47)
O(51)ꢀC(56)
O(51)ꢀC(57)
O(61)ꢀC(66)
O(61)ꢀC(67)
Sn(2)···O(41)
Sn(2)···O(51)
Sn(2)···O(61)
O(1)···O(51)
O(31)···O(21)
119Sn NMR data for (o-Anis)3SnX, when compared
with those for other Ar3SnX (Ar=Ph or Mes) (Table
5), also indicate that the effects of ortho-substituents
are not the same in the ortho-anisyl and mesityl series.
All three Ar3SnX series show the shift to higher fre-
quency of l(119) as X=I is replaced by more elec-
tronegative Br and then Cl, with (Mes)3SnF continuing
this trend. Plotting l(119Sn) data for Ph3SnX or (o-
Anis)3SnX versus (Mes)3SnX, gives l(Ph)=0.525
l(Mes)+1.13 (r=0.998), and l(o-Anis)=0.603
l(Mes)−3.85 (r=0.998), respectively. Inserting the
119Sn chemical shift for (Mes)3SnF then predicts
l(119Sn) values of −33.8 ppm for the hypothetical
Ph3SnF monomer in solution but more significantly,
C(11)ꢀSn(1)ꢀO(1)
C(21)ꢀSn(1)ꢀO(1)
C(31)ꢀSn(1)ꢀO(1)
99.7(3)
102.4(3)
105.9(3)
C(41)ꢀSn(2)ꢀO(1)
C(51)ꢀSn(2)ꢀO(1)
C(61)ꢀSn(2)ꢀO(1)
104.5(3)
112.3(3)
99.9(3)

228
I. Wharf et al. / Inorganica Chimica Acta 294 (1999) 224–231
Table 3
Selected dihedral angles (°) a
(a) Chlorotris(o-methoxyphenyl)tin(IV) (A)
(A) ClꢀSnꢀC(11)
(B) Phenyl ring (1)
54.9(4)
ClꢀSnꢀC(21)
Phenyl ring (2)
8.9(5)
ClꢀSnꢀC(31)
Phenyl ring (3)
68.0(5)
(b) Fluorotris(o-methoxyphenyl)tin(IV) (B)
(A) FꢀSnꢀC(11)
(B) Phenyl ring (1)
66.5(3)
FꢀSnꢀC(21)
Phenyl ring (2)
8.4(4)
FꢀSnꢀC(31)
Phenyl ring (3)
58.6(3)
(c) Oxobis[tris(o-methoxyphenyl)tin(IV)] (C)
(A) O(1)ꢀSn(1)ꢀC(11)
(B) Phenyl ring (1)
9.4(3)
O(1)ꢀSn(1)ꢀC(21) O(1)ꢀSn(1)ꢀC(31)
Phenyl ring (2)
82.5(3)
Phenyl ring (3)
19.6(5)
(A) O(1)ꢀSn(2)ꢀC(41)
(B) Phenyl ring (4)
59.1(3)
O(1)ꢀSn(2)ꢀC(51) O(1)ꢀSn(2)ꢀC(61)
Phenyl ring (5)
59.2(3)
Phenyl ring (6)
3.7(3)
Fig. 2. ORTEP view of the molecule of C showing the numbering
scheme adopted. Ellipsoids are drawn at 40% probability level.
a Angle between planes (A) and (B).
−44.0 ppm for (o-Anis)3SnF. The l(119Sn) values of
(o-Anis)3SnX (X=Cl, Br, I) and (o-Anis)4Sn are more
negative compared with the phenyltin data [19], per-
haps due to weak Sn···O interactions increasing shield-
ing at tin since there is little structural change on going
from a phenyl compound to its o-anisyl analogue.
However, l(119Sn) for the fluoride is at much lower
frequency than expected on this basis and this appears
to coincide with the abrupt flattening of the C3Sn
pyramid in going from (o-Anis)3SnCl to (o-Anis)3SnF.
A similar large shift to low frequency for l(119Sn) of
(Mes)3SnI compared with Ph3SnI was also earlier corre-
lated with the opening up of the CSnC angles around
tin [19]. For the trimesityltin system this is mainly due
to the steric crowding of the ortho-methyl groups. For
(o-Anis)3SnX, since flattening of the C3Sn pyramid as
X=Cl is replaced by F increases the steric strain in the
molecule, it is more likely due to the high greater ionic
character of the SnꢀF bond causing an increase in
s-character of the SnꢀC bonds with the accompanying
opening out of the CSnC angles. However, while such
correlations of trends in solution 119Sn data with solid-
state structural changes must be viewed as speculative,
the close similarity of solution and solid-state 119Sn
data for both (o-Anis)3SnF and (Mes)3SnF would sug-
gest the same structural features are responsible for the
NMR trends in both CDCl3 solution and the solid-
state. Solid-state 119Sn NMR studies of a wide range of
Ar3SnX (X=Cl, Br) are now in progress for compari-
son with data from earlier solution studies [16,19].
3.2. Tris(o-methoxyphenyl)tin oxide and hydroxide
All freshly prepared samples of ‘(o-Anis)3SnOH’ gave
two peaks in the 119Sn spectrum (CDCl3 solution) at
−83 and −98 ppm, the latter usually predominant.
Saturating one of these solutions with water eliminated
the lower frequency peak, thus allowing the peak at
−83 ppm to be assigned to the hydroxide [Table 4(b)].
Recrystallising the crude ‘(o-Anis)3SnOH’ from ethanol
or ethanol/CDCl3 gave the OH-free oxide (IR), con-
firmed by the single 119Sn resonance at −98 ppm with
satellites due to 119Sn–117Sn spin coupling, observed in
CDCl3 solution [Table 4(b)]. Indeed, the almost com-
plete loss of OH from ‘(o-Anis)3SnOH’ samples occurs
at room temperature simply by more rigorous drying in
vacuum or leaving them open to the atmosphere
overnight. This behaviour means, as found earlier for
(neophyl)3SnOH [34], that (o-Anis)3SnOH can not be
isolated and may only be identified in solution.
Fig. 1. ORTEP view of the molecule of B showing the numbering
scheme adopted. Ellipsoids are drawn at 40% probability level. The
same numbering scheme was also used for molecule A.

I. Wharf et al. / Inorganica Chimica Acta 294 (1999) 224–231
229
Table 4
Solution (CDCl3) and solid-state NMR data a for Ar3SnX
(a) X=F
Ar=
o-Anis
Mes
b
Solution
Solid
Solution
Solid
l(119Sn) ppm
−74.05
714.6
2286.5
12.7
−77.10
2240.0
−66.51
625.7
2285.5
9.9
−70.6, −82.2
1J(119Sn–13C) Hz
1J(119Sn–19F) Hz
2J(19F–13C) Hz
2300, 2256
(b)
X=1/2O
o-Anis
X=OH
c
Ar=
o-Tol
o-Anis
Mesd
Solution
Solid
Solution d
Solid
Solution
Solution
l(119Sn) ppm
−98.42
660.2
471.2
−93.26, −109.76
650.6
−77.20
617.1
576.1
−54.08
801.6
−83.12
679.3
−101.78
614.7
1J(119Sn–13C) Hz
2J(119Sn–119Sn) Hz
a [Sn]=0.2–0.4 M; Ref: (119Sn), ext. (CH3)4Sn; (13C), int. (CH3)4Si.
b Ref [27].
c CDCl3 solution of [(o-Anis)3Sn]2O saturated with water.
d Ref. [1].
2
IR data [was(SnOSn)] indicated that [(o-Anis)3Sn]2O
(C) in the solid state might be more like linear [(o-
Tol)3Sn]2O (839 cm−1) [1,15] rather than angular
(Ph3Sn)2O (776 cm−1) [13,14]. However, the solid-state
119Sn NMR spectrum of the o-anisyltin system shows
that there are in fact two distinct tin environments in
contrast to the single peak observed with symmetric
[(o-Tol)3Sn]2 [Table 4(b)]. Furthermore, from the corre-
lation of J(119Sn,119Sn) with molecular structure given
by Lockhart et al. [15], we can predict that C would be
the first distannoxane with an SnꢀOꢀSn angle between
140 and 180°.
The X-ray diffraction study of C confirms these
predictions, the molecule (Fig. 2) having two quite
different geometries around each tin and a bent
SnꢀOꢀSn skeleton (167°). The architecture of o-anisyl
Table 5
Comparison of tin atom environments with solution (CDCl3) NMR results for various Ar3SnX complexes
Ar3SnX
Ar
Geometric parameters a
NMR data b
d(SnꢀX) (A)
Sqax
l(119Sn) (ppm)
1J(119Sn–13C) (Hz)
c
c
,
X
Sqeq
(Sqeq−Sqax)
Ph
Cl d
Br e
I f
2.360
2.495
2.704
314
316
319
341
340
337
27
24
18
−44.81
−60.01
−113.38
615.7
596.3
570.9
o-Anis
F
1.972
2.371
310
318
345
338
35
20
−74.05
−56.68
−74.31
−135.65
714.6
684.8
666.2
635.8
Cl
Br
I g
2.713
1.961
318
303
338
349
20
46
Mes
F h
Cl
Br
I i
−66.51
−84.39
−120.98
−217.10
625.7
596.1
579.8
554.0
2.750
308
346
38
a Averaged values when two or more data sets are available.
b This work or Ref. [19].
c Sqax=SÚCSnX (°), Sqeq=SÚCSnC(°).
d Ref. [28,29].
e Ref. [30].
f Ref. [31,32].
g Ref. [11].
h Ref. [7].
i Ref. [31].

230
I. Wharf et al. / Inorganica Chimica Acta 294 (1999) 224–231
Table 6
Correlation of 119Sn NMR data with structural parameters for various (Ar3Sn)2O complexes
(a) 119Sn NMR data
Ar
l(119Sn) (ppm)
2J(119Sn–119Sn) (Hz)
CDCl3
CDCl3
Solid
Solid
421 b
Ph
−83.47 a
−80.5
−98.42
−75.2 b
−93.26
−54.08
410.8 a
o-Anis
o-Tol
471.2
650.6
801.6
−109.76
−77.20 a
576.1 a
(b) Structural parameters
Ar
c
c
,
Sn
d(SnꢀO) (A)
ÚSnOSn (°)
137.3
Sqax
Sqeq
(Sqeq−Sqax)
Ph d
(1)
(2)
(1)
(2)
1.957
1.953
1.940
1.915
1.922
315
320
308
317
323
340
336
346
337
333
25
16
38
20
10
o-Anis
o-Tol e
167.0
180.0
a Ref. [1].
b Ref. [33].
c Sqax=SÚCSnO (°), Sqeq=SÚCSnC (°).
d Ref. [14].
e Ref. [15].
groups around Sn(2) is almost the same as that found
for A and B, having one CH3O(6) trans and the other
two methoxy groups (4) and (5) cis to the Sn(2)ꢀO
bond [Table 2(c)] with phenyl ring (6) almost coplanar
with the C(61)ꢀSnꢀO(1) unit [Table 3(c)]. In contrast,
for Sn(1) all three methoxy groups are trans to the
Sn(1)ꢀO bond, presumably to lessen interatomic repul-
sions between the two (o-Anis)3Sn units.
[(o-Anis)3Sn]2O, Sqax and Sqeq values are comparable
with those found in the corresponding triaryltin halides
(Table 5), but for the tris(o-tolyl)tin oxide, Sqax=323
and Sqeq=333 are unexpectedly close to the ‘ideal’
tetrahedral value, Sqax=Sqeq=328°. In contrast, the
geometries around the two atoms tin in (o-Tol)3SnNCS
(Sqax=314°, Sqeq=342°; Sqax=311°, Sqeq=344°) are
similar to that found in (o-Anis)3SnNCS (Sqax=315°,
Sqeq=333°) [8] which itself is between the tin ge-
ometries determined for the corresponding chloride A
and fluoride B (Table 5).
The factors which determine the amplitude of the
MꢀOꢀM angle in (R3M)2O (M=C–Pb) are as yet
unresolved. These include the relative ground state
(linear or bent) energies of the molecule, the conse-
quences of intramolecular interactions between the two
end R3M units, as well as crystal packing effects in the
solid state. Several years ago Glidewell [35] argued that
electronic rather than steric factors would predominate,
and could account for the change in geometry from
angular (Ph3C)2O [36] to linear (Ph3Si)2O [37] which has
the same symmetric structure as [(o-Tol)3Sn]2O [15].
However, in contradiction, both (Ph3Ge)2O [38] and
(Ph3Sn)2O [14] have bent MꢀOꢀM skeletons, leading to
the conclusion that in these systems a large amplitude
bending vibration can occur, meaning that it is charac-
teristic of (R3M)2O that the MꢀOꢀM angle may be
markedly changed with the expenditure of very little
energy. In this case, crystal packing effects then can
stabilise one molecular shape over the another. Our
Correlation of solid-state 119Sn NMR data with
structural parameters (Table 6) for the three (Ar3Sn)2O
(Ar=Ph, o-Anis, o-Tol) gives ÚSnOSn=0.114
2J(119Sn,119Sn)+89.93; r=0.995. Inserting CDCl3 solu-
tion 2J(119Sn,119Sn) values in the equation then indi-
cates SnꢀOꢀSn angles, in solution, of 137, 144, and
156° for the given (Ar3Sn)2O species, Ar=Ph, o-Anis,
o-Tol, respectively. Thus, for [(o-Anis)3Sn]2O as was
found earlier for two hexabenzyldistannoxanes [15], it is
2
the change in the J value and not the chemical shift
which suggests a structural change occurs on going
from the solid state to the solution. However, for
[(o-Tol)3Sn]2O, both NMR parameters are affected
when the sample is dissolved with the SnꢀOꢀSn skele-
ton changing from linear to bent. In addition our data
do not show the previously reported considerable high
field shift of the tin resonance of [(o-Tol)3Sn]2O [15].
Inspection of tin atom environments [Sqax, Sqeq,
(Sqeq−Sqax)] in the three (Ar3Sn)2O (Ar=Ph, o-Anis,
o-Tol) [Table 6(b)] provides more evidence of the spe-
cial nature of the hexakis-(o-tolyl)distannoxane struc-
ture in the solid state. Thus, for both (Ph3Sn)2O and

I. Wharf et al. / Inorganica Chimica Acta 294 (1999) 224–231
231
[11] R.A. Howie, J.-N. Ross, J.L. Wardell, J.N. Low, Acta Crystal-
logr., Sect. C 50 (1994) 229.
[12] U. Kolb, M. Dra¨ger, B. Jousseaume, Organometallics 10 (1991)
2737.
[13] I. Wharf, H. Lamparski, R. Reeleder, Appl. Organomet. Chem.
11 (1997) 969.
[14] C. Glidewell, D.C. Liles, Acta Crystallogr., Sect. B 34 (1978)
1693.
results, together with those of Lockhart et al. [15],
indicate that (R3Sn)2O structures with larger SnꢀOꢀSn
angles are more favoured in the solid state. Thus,
crystal packing effects are significant in determining the
size of the SnꢀOꢀSn angle while both linear and bent
R3SnꢀOꢀSnR3 structures have very similar energies.
[15] T.P. Lockhart, H. Puff, W. Schluh, H. Reuter, T.N. Mitchell, J.
Organomet. Chem. 366 (1989) 61.
[16] I. Wharf, Inorg. Chim. Acta 159 (1989) 41.
[17] I. Wharf, A. Bastone, E.J. Bures, Can. J. Spectrosc. 41 (1996)
122.
Acknowledgements
The financial assistance of the ‘Fonds FCAR (pro-
gramme Soutien aux Equipes de Recherche)’ of the
[18] I. Wharf, R. Wojtowski, C. Bowes, A.-M. Lebuis, M.
Onyszchuk, Can. J. Chem. 76 (1998) 1827.
´
‘Gouvernement du Que´bec’ is most gratefully acknowl-
edged. We also thank Dr Ce´line Pearson for carrying
out the X-ray structure determination of compound A.
The authorities of Dawson College, Montreal, are also
thanked for providing release time, thus enabling this
research to be carried out.
[19] I. Wharf, M.G. Simard, J. Organomet. Chem. 532 (1997) 1.
[20] G.M. Sheldrick, SHELXS-96, Program for the solution of crystal
structures, University of Go¨ttingen, Germany, 1996.
[21] G.M. Sheldrick, SHELXL-96, Program for structure analysis,
University of Go¨ttingen, Germany, 1996.
[22] H.D. Flack, Acta Crystallogr., Sect. A 39 (1983) 876.
[23] International Tables for Crystallography, vol. C, Kluwer, Dor-
drecht, Holland, 1992, Tables 4.2.6.
[24] C.K. Johnson, ORTEPII, Rep. ORNL-5138, Oak Ridge National
Laboratory, TN, USA, 1976.
[25] E. Krause, R. Becker, Chem. Ber. 53 (1926) 173.
[26] D. Tudela, R. Fernandez, V.K. Belsky, V.E. Zavodnik, J. Chem.
Soc., Dalton Trans. (1996) 2123.
[27] H. Bai, R.K. Harris, H. Reuter, J. Organomet. Chem. 408 (1991)
167.
[28] J.S. Tse, F.L. Lee, E.J. Gabe, Acta Crystallogr., Sect. C 42 (1986)
1876.
[29] S.W. Ng, Acta Crystallogr., Sect. C 51 (1995) 2292.
[30] H. Preut, F. Huber, Acta Crystallogr., Sect. B 35 (1979) 2292.
[31] M.G. Simard, I. Wharf, Acta Crystallogr., Sect. C 50 (1994) 397.
[32] S.W. Ng, Acta Crystallogr., Sect. C 51 (1995) 629.
[33] R.K. Harris, A. Sebald, Magn. Res. Chem. 27 (1989) 81.
[34] T.P. Lockhart, J. Organomet. Chem. 287 (1989) 179.
[35] C. Glidewell, J. Organomet. Chem. 159 (1978) 23.
[36] C. Glidewell, D.C. Liles, Acta Crystallogr., Sect. B 34 (1978) 696.
[37] C. Glidewell, D.C. Liles, Acta Crystallogr., Sect. B 34 (1978) 124.
[38] C. Glidewell, D.C. Liles, Acta Crystallogr., Sect. B 34 (1978) 119.
References
[1] I. Wharf, Appl. Organomet. Chem. (1999) in press.
[2] I. Wharf, A.-M. Lebuis, H. Lamparski, Acta Crystallogr., Sect.
C 52 (1996) 2477 and Refs. therein.
[3] I. Wharf, A.-M. Lebuis, Acta Crystallogr., Sect. C 52 (1996)
3025.
[4] I. Wharf, M.G. Simard, J. Organomet. Chem. 532 (1974) 1.
[5] D. Tudela, E. Gutierrez-Puebla, A. Monge, J. Chem. Soc.,
Dalton Trans. (1992) 1069.
[6] C. Glidewell, D.C. Liles, Acta Crystallogr., Sect. B 34 (1978)
129.
[7] H. Reuter, H. Puff, J. Organomet. Chem. 379 (1989) 223.
[8] I. Wharf, M.G. Simard, Inorg. Chim. Acta 282 (1998) 30.
[9] A.M. Domingos, G.M. Sheldrick, J. Organomet. Chem. 67
(1974) 257.
[10] L.E. Khoo, X.-M. Chen, T.C. Mak, Acta Crystallogr., Sect. C
47 (1991) 2647.
.