Solid-state and high-resolution liquid 119Sn NMR spectroscopy of some monomeric, two-coordinate low-valent tin compounds: Very large chemical shift anisotropies

Article abstract of DOI:10.1021/ic000514z

High-resolution liquid- and solid-state ¹¹⁹Sn NMR spectroscopy was used to study the bonding environment in the series of monomeric, two-coordinate Sn(II) compounds of formula Sn(X)C₆H₃-2,6-Trip₂ (X = Cl, Cr(η

Full text of DOI:10.1021/ic000514z

5450
Inorg. Chem. 2000, 39, 5450-5453
Solid-State and High-Resolution Liquid 119Sn NMR Spectroscopy of Some Monomeric,
Two-Coordinate Low-Valent Tin Compounds: Very Large Chemical Shift Anisotropies
Barrett E. Eichler, Brian L. Phillips, Philip P. Power,* and Matthew P. Augustine*
Departments of Chemistry and Materials Science, One Shields Avenue, University of California,
Davis, California 95616
ReceiVed May 12, 2000
High-resolution liquid- and solid-state 119Sn NMR spectroscopy was used to study the bonding environment in
the series of monomeric, two-coordinate Sn(II) compounds of formula Sn(X)C6H3-2,6-Trip2 (X ) Cl, Cr(η5-
C5H5)(CO)3, t-Bu, Sn(Me)2C6H3-2,6-Trip2; Trip ) C6H2-2,4,6-i-Pr3). The trends in the principal components of
the chemical shift tensor extracted from the solid-state NMR data were consistent with the structures determined
by X-ray crystallography. Furthermore, the spectra for the first three compounds displayed the largest 119Sn NMR
chemical shift anisotropies (up to 3798 ppm) of any tin compound for which data are currently available. Relaxation
time based calculations for the dimetallic compound 2,6-Trip2H3C6S¨n-Sn(Me)2C6H3-2,6-Trip2 suggests that the
chemical shift anisotropy for the two-coordinate tin center may be as much as ca. 7098 ppm, which is as broad
as the 1 MHz bandwidth of the NMR spectrometer.
Introduction
SnSnR2 (R ) CH(SiMe3)2), which consist of weakly dimerized
SnR2 units, clearly demonstrate large chemical shift anisotropy,
as evidenced by the principal values δ11 ) 1600, δ22 ) 400,
and δ33 ) 100 ( 20 ppm, which is an order of magnitude greater
than that seen in Sn(IV) species. These values suggest strong
deshielding along the axis in the SnC2 ligand plane and
perpendicular to the Sn-Sn bond (i.e., δ11) and much greater
shielding along the axis perpendicular to the SnC2 coordination
plane at tin (i.e., δ33). We were therefore anxious to record the
solid-state spectra of some monomeric two-coordinate Sn(II)
species with the expectation that such a study would reveal large
anisotropies in their chemical shift tensors which would help
elucidate the nature of the bonding in these complexes. The
results of these studies are now described.
The use of 119Sn NMR spectroscopy has proven to be an
invaluable structural probe in all areas of tin chemistry.1 A large
body of data is now available2 which demonstrates that 119Sn
NMR studies provide accurate, relatively easily obtainable
information on the bonding to the tin atom. In many cases the
most important spectroscopic parameter available from these
studies is the value of the chemical shift (δ). While much of
the chemical shift data concern isotropic 119Sn shift values
measured in solution, solid-state 119Sn NMR measurements
usually afford more detailed information involving the δ11, δ22,
and δ33 components of the nuclear shielding tensor, information
that can provide considerably more insight into the bonding at
tin.3 Monomeric, two-coordinate tin(II) compounds (e.g., SnR2;
R ) alkyl,4 aryl,5 amide,6 or alkoxide7 groups) whose bonding
is often more complex than that in tin(IV) compounds owing
to the presence of a formally empty, nonbonding p-orbital and
a stereochemically active lone pair (in an orbital that is mostly
5s in character) at the tin centers, are particularly amenable to
solid-state 119Sn NMR investigation. Unfortunately, there are
no such studies available for this important compound class.
This may be due to the fact that monomeric compounds of this
kind are not numerous and that they usually require careful
handling owing to their extreme air and moisture sensitivity.
Nonetheless, studies8 on the related compounds such as R2-
Experimental Section
9
The synthesis of Sn(Cl)C6H3-2,6-Trip2 and {Sn(Cl)C6H3-2,6-
10
Mes2}2 (Mes ) C6H2-2,4,6-(CH3)3) have been previously reported.
The compounds Sn(X)C6H3-2,6-Trip2 (X ) Cr(η5-C5H5)(CO)3, t-Bu,
or Sn(Me)2C6H3-2,6-Trip2) were synthesized by the reaction of Sn-
(Cl)C6H3-2,6-Trip2 with Na[Cr(η5-C5H5)(CO)3], t-BuLi, or MeLi,
respectively. Full details of the synthesis, structural characterization,
and reactions of these three compounds will be provided separately.11
Solid-state 119Sn NMR spectra were recorded under conditions of magic
angle spinning at 9.4 T using a Chemagnetics Infinity NMR spectrom-
eter and probehead configured for 7.5 mm (o.d.) rotors. A combination
of fast relaxation and restricted sample quantities typically limited data
accumulation to between one and four rotational echoes. The absolute
value of the Fourier transform of the first rotational echo provided pure
phase powder patterns that were used to determine δnn values.12 Isotropic
liquid shifts were obtained using both this Chemagnetics solid-state
(1) Kennedy, J. D.; McFarlane, W. In Multinuclear NMR; Mason, J., Ed.;
Plenum: New York, 1987; Chapter 4.
(2) Wrackmeyer, B. Annu. Rep. NMR Spectrosc. 1999, 38, 203.
(3) Duncan, T. M. A Compilation of Chemical Shift Anisotropies; Farragut
Press: Chicago, 1990.
(4) Davidson, P. J.; Lappert, M. F. J. Chem. Soc. Chem. Commun. 1973,
317.
(5) Weidenbruch, M.; Schaeflke, J.; Scha¨fer, A.; Peters, K.; von Schnering,
H. G.; Marsmann, H. Angew. Chem., Int. Ed. Engl. 1994, 33, 1846.
(6) Harris, D. H.; Lappert, M. F. J. Chem. Soc. Chem. Commun. 1974,
895.
(7) Cetinkaya, B.; Gu¨mru¨kcu¨, I.; Lappert, M. F.; Atwood, J. L.; Rogers,
R. D.; Zaworotko, M. J. Am. Chem. Soc. 1980, 102, 2088.
(8) Zilm, K.; Lawless, G. A.; Merrill, R. M.; Millar, J. M.; Webb, G. G.
J. Am. Chem. Soc. 1987, 109, 7236.
(9) Pu, L.; Senge, M. O.; Olmstead, M. M.; Power, P. P. J. Am. Chem.
Soc. 1998, 120, 12682.
(10) Simons, R. S.; Pu, L.; Olmstead, M. M.; Power, P. P. Organometallics
1997, 16, 1920.
(11) (a) Eichler, B. E.; Power, P. P. Inorg. Chem. 2000, 39, 5444. (b)
Eichler, B. E.; Power, P. P. Unpublished work.
(12) Chingas, G.; Frydman, L.; Barrall, G.; Hartwood, J. S. in NMR
Spectroscopy: History, Theory and Applications; Blumlich, B.; Kuhn,
W., Eds.; VCH: Weinheim, 1992; pp 373-393.
10.1021/ic000514z CCC: $19.00 © 2000 American Chemical Society
Published on Web 11/05/2000

Liquid 119Sn NMR Spectroscopy of Sn(II) Compounds
Inorganic Chemistry, Vol. 39, No. 24, 2000 5451
Since the HOMO-LUMO gap in group 14 carbene analogues
increases with more electronegative substituents, a smaller
contribution to the paramagnetic shielding is expected to produce
a more upfield resonance (since paramagnetic effects augment
the applied field). This consideration has been used to account
for the higher field shifts for tin(II) amides and alkoxides.2,14,15
However, the observed isotropic chemical shift is not only a
sum of paramagnetic and diamagnetic terms, but also a function
of directionally dependent components.
In the absence of unusual averaging effects, the liquid-state
shift is equal to the isotropic chemical shift, the average δiso
)
(δ11 + δ22 + δ33)/3 of the three principal components δ11, δ22,
and δ33 of the chemical shift tensor. Here δ11 and δ33 label the
downfield and upfield edges of the observed NMR spectrum in
the solid state whereas δ22 represents the frequency that
corresponds to a singularity in the powder pattern where the
spectral amplitude tends to infinity. Physically, these δnn values
describe the shape of an ellipsoid in three dimensions in the
principal axis system of the chemical shift tensor. This shape
is related to the topology of the electronic wave function at the
site of the nucleus and can therefore lead to details about
chemical bonding. The size of and difference between δnn values
for a particular site is a strong function of the symmetry and
structure of the bonding environment. This structural dependence
is illustrated by a comparison of shift tensor values for
tetraphenyl tin, SnPh4,16 and the already mentioned8 R2SnSnR2
(R ) CH(SiMe3)2) dimer (which is dissociated to SnR2
monomers in dilute solution) which have the values δiso ) -117
ppm and +700 ppm, respectively. The relaxed tetrahedral
environment at tin in SnPh4 yields an axially symmetric powder
pattern with δ11 ) -90 ppm and δ22 ) δ33 ) -130 ppm
whereas the less symmetric environment in R2SnSnR2 generates
δ11 ) +1600 ppm, δ22 ) +400 ppm, and δ33 ) +100 ppm,
thereby affording an asymmetric powder pattern with record
anisotropy ø ) 3(δ11 - δiso)/2 ) 1350 ppm.
Chemical Shift Trends in the Sn(X)C6H3-2,6-Trip2 Com-
pounds (X ) Cl, Cr(η5-C5H5)(CO)3, and t-Bu) and Related
Species. The 119Sn solid-state NMR spectra shown in Figure
2a-c correspond to the mononuclear compounds in Figure 1
with X ) Cl, Cr(η5-C5H5)(CO)3, and t-Bu, respectively. The
solid lines represent the actual experimental data while the
dashed lines represent fits used to extract δnn values. The δnn
values for these spectra and the spectra (not shown) for 2,6-
Trip2H3C6S¨n-Sn(Me)2C6H3-2,6-Trip2 and {Sn(µ-Cl)C6H3-2,6-
Mes2}2 are included in Table 2 along with a calculation of the
isotropic chemical shift and the Ar-Sn-X bond angle θ, CAr-Sn
bond length rAr, and Sn-X bond length rx values obtained from
X-ray data. It is important to note that the anisotropic chemical
shift tensor values for the X ) Sn(Me)2C6H3-2,6-Trip2 com-
pound represent only the Sn(III) site with δliq ) 257.4 ppm in
Table 1. All attempts at obtaining solid-state data for the Sn(I)
site failed, presumably, as discussed later, because the anisotropy
ø is too large to measure by pulse-Fourier transform techniques
at this field strength.
Figure 1. Model of the Sn(II) metal center structure in the compounds
used in this study. The bond lengths rAr between Sn and Ar ) 2,6-
Trip2C6H3 and rx between Sn and X ) Cl, Cr(η5-C5H5)(CO)3, t-Bu,
and Sn(Me)2(C6H3-2,6-Trip2) and Ar-Sn-X bond angle θ determined
by X-ray crystallography are summarized in Table 2.
Table 1. Summary of 119Sn Liquid State Chemical Shifts of
2,6-Trip2H3C6SnX (Trip ) C6H2-2,4,6-i-Pr3, X ) Cl,
Cr(η5-C5H5)(CO)3, t-Bu, Sn(Me)2C6H3-2,6-Trip2)
X
δliq(ppm)
reference
Cl
793.4
2,297.9
1,904.4
2856.9
257.4a
625.2
10
Cr(η5-C5H5)(CO)3
t-Bu
11b
11a
11a
a
Sn(Me)2C6H3-2,6-Trip2
{Sn(µ-Cl)C6H3-2,6-Mes2}2
9
a Chemical shift of the tetravalent tin atom.
system and a Varian Inova NMR spectrometer operating at 9.4 T. Solid-
state isotropic chemical shifts for samples giving several rotational
echoes were obtained from the magic angle spinning centerband.
Results and Discussion
To better understand bonding and electronic structure in the
series of compounds of formula Sn(X)C6H3-2,6-Trip2 illustrated
schematically in Figure 1 where X ) Cl, t-Bu, Cr(η5-C5H5)-
(CO)3, and Sn(Me)2(C6H3-2,6-Trip2), 119Sn NMR data in both
the liquid phase and the solid state were recorded. It is well-
known that the isotropic liquid state 119Sn chemical shift δliq is
sensitive to both valence and electronic structure.1,2 For example,
Sn(IV) compounds tend to be more shielded and resonate further
upfield (-1100 ppm < δliq < +200 ppm) than deshielded Sn-
(II) compounds (+200 ppm < δliq < +5000 ppm).2 A summary
of δliq in ppm for the compounds in Figure 1 is shown in Table
1, along with additional data for the related dinuclear tin com-
plex {Sn(µ-Cl)C6H3-2,6-Mes2}2,10 which is associated through
chloride bridging. The shifts for all of the compounds shown
in Figure 1 are greater than 200 ppm which indicates that the
tin is in the oxidation state +2 (note that the formal oxidation
states of the tin atoms in 2,6-Trip2H3C6S¨n-Sn(Me)2C6H3-2,6-
Trip2 are Sn(I) and Sn(III) although they are divalent and
tetravalent). However, there seems to be no apparent trend in
the chemical shifts of the series with X ) Cl, Cr(η5-C5H5)-
(CO)3, t-Bu, and Sn(Me)2(C6H3-2,6-Trip2). For example, the
electronegativity of the atoms bound to tin increases in the
order: Cr(1.6), Sn(1.96), C(2.55), Cl(3.16), suggesting that
electron density at the tin nucleus should be depleted in the
same order and thereby generate a chemical shift trend opposite
to what is observed. Other explanations for the order of δliq for
these compounds based on ligand electron withdrawing and
donating characteristics as well as their π-back-bonding capacity
also lead to predictions inconsistent with the observed δliq trend.
One plausible explanation is that the paramagnetic shielding, a
reflection of the mixing of ground and excited states, is increased
rather than decreased by more electronegative substituents.13
Although the trend in δliq values in Table 1 for the compounds
in the series X ) Cl, Cr(η5-C5H5)(CO)3, and t-Bu is not obvious,
the variation of each of the δnn components of the chemical
(13) Albright, T. A.; Burdett, J. K.; Whangbo, M. H. Orbital Interactions
in Chemistry; Wiley: New York, 1985; p 127.
(14) Braunschweig, H.; Chorley, R. W.; Hitchcock, P. B.; Lappert, M. F.
J. Chem. Soc., Chem. Commun. 1992, 1311.
(15) Wrackmeyer, B. UnkonVentionelle Wechselswirkungen in der Chemie
metallischer Elemente; Krebs, B., Ed.; VCH: Weinheim, 1992; pp
111-124.
(16) Komoroski, R. A.; Parker, R. G.; Mazany, A. M. J. Magn. Res. 1987,
73, 389.

5452 Inorganic Chemistry, Vol. 39, No. 24, 2000
Eichler et al.
from largest to smallest as X ) Cr(η5-C5H5)(CO)3, > t-Bu, >
Cl, can be rationalized in terms of the variation of the Ar-
Sn-X bond angle θ. As θ decreases from 110.13° to 99.68° in
Table 2, the hybrid orbital containing the lone pair gains more
s character. This gain in s character at tin translates into
increased shielding owing to the increased electron density at
the 119Sn nucleus. Thus the most shielded δ33 value of -165.1
ppm is observed for the most electronegative substituent X )
Cl since this compound has the narrowest interligand angle at
tin. The narrowing of the angle is apparently caused by the
tendency of the more electronegative ligands to attract the most
p-character into the orbitals from the central element to which
they are bound, thereby allowing the s-character of the lone
pair to increase. Finally, the trend in δ22 values can be explained
from the X-ray data and the structure in Figure 1. Since the
direction of the δnn components must be orthogonal to each
other, δ22 must lie in the Ar-Sn-X plane and at an angle of
90° with respect to both δ11 and δ33. In other words, the δ22
component lies in the coordination plane of tin at a right angle
to the δ33 component that is directed through the lone pair. From
Table 2 it can be seen that both δ22 and rx decrease in the order
X ) Cr(η5-C5H5)(CO)3, > Cl, > t-Bu which supports the
suggestion that the direction of δ22 was at an angle of 90° to
the 33 direction and close to the Sn-X bond direction in Figure
1. The correlation of δ22 with bond length is possibly easier to
rationalize than the trends for δ11 and δ33. Here the increased
bond length indicates that electron density will be depleted from
around the metal center thus deshielding the 119Sn nucleus and
causing the X ) Cr(η5-C5H5)(CO)3 compound to resonate farther
downfield than the X ) t-Bu and Cl compounds. It is important
to note that differing energy gaps corresponding to directionally
dependent HOMO and LUMO’s as obtained from ab initio
calculations, such as those performed on a family of singlet
carbene molecules where similar trends are observed,17 would
provide further insight into the observed trends in the δnn
components by carefully tracking both the paramagnetic and
diamagnetic contributions to δnn.
Figure 2. Examples of solid-state 119Sn NMR powder patterns observed
in this study. The solid lines represent experimental data for the X )
Cl, Cr(η5-C5H5)(CO)3, and t-Bu compounds in a, b, and c, respectively.
The dashed lines indicate the fits used to extract the δ11, δ22, and δ33
shift tensor components.
shift tensor for these compounds can be directly related to
structure. Consider both the downfield δ11 and upfield δ33
components of the shift tensor. These matrix elements cor-
respond to the minimum and maximum values of the chemical
shielding or the minimum and maximum electron density along
orthogonal directions in the principal axis system of the chemical
shift tensor. With regard to the structure shown in Figure 1,
which has both an empty p-orbital and lone electron pair, it is
most likely that the 11 direction lies along the symmetry axis
of the empty p-orbital whereas the 33 direction involves the
hybrid orbital containing the lone electron pair. To be absolutely
certain of the orientation of the principal axis system of the
shift tensor in the molecular frame, either single crystal
measurements or investigation of the relationship between a
known dipolar coupling and the principal axis system of the
shift tensor would be required. However, in the absence of such
measurements this assignment of the 11 and 33 directions
appears to be consistent with back-bonding trends and X-ray
structural data. The structure in Figure 1 indicates essentially
zero electron density along the 11 direction. Therefore, one
would expect δ11 for these compounds to tend toward the bare
nucleus value. The finite but large value for δ11, coupled with
its variation in the order smallest to largest in the sequence X
) Cl, < Cr(η5-C5H5)(CO)3, < t-Bu can be explained by back-
bonding. The filled nonbonding orbitals in the Cl ligand more
efficiently overlap the p-orbital on tin than either the d-orbitals
on the chromium in the Cr(η5-C5H5)(CO)3 moiety or any
projection of C-C σ bond electron density from the t-Bu ligand.
Of these three X ligands, Cl back-donates the most electron
density to the tin, thus shielding the 119Sn nucleus and causing
the δ11 value to appear farther upfield. The same argument holds
for the Cr(η5-C5H5)(CO)3 ligand in comparison to the t-Bu
ligand since back-bonding orbital overlap is likely to be greater
for the former moiety.2,14,15 The trend in δ33 values in order
Comparison of Liquid- and Solid-State Isotropic Shifts.
It is interesting to compare δliq for each compound in Table 1
with δiso in Table 2 calculated from the δnn values and the magic
angle centerband isotropic shift δmas in Table 2. The difference
between δiso and δmas gives an estimate for the uncertainty of
the powder pattern fitting routine due to linebroadening and
noise in the spectral data. It is obvious from comparison that
δliq is not equal to δiso. Closer inspection of the data reveals
that there is a ca. 5% difference between δliq and δiso values for
both the X ) Cr(η5-C5H5)(CO)3 and t-Bu compounds. In terms
of the data sets shown in Figure 2, this 5% difference translates
into a change in δnn values of less than four points. More
bothersome, however, are the 17% and 35% difference between
δliq and δiso for the Sn(II) sites in the {Sn(µ-Cl)C6H3-2,6-Mes2}2
and X ) Cl compounds respectively and the 19% difference
for the Sn(III) site in the X ) Sn(Me)2(C6H3-2,6-Trip2)
compound. Barring incomplete averaging effects, chemical
exchange, or unusual solvent effects, δliq should always equal
δiso. The 17% and 19% difference recognized in the dinuclear
tin compounds could be due to any or all of these possibilities.
These two dinuclear complexes also have two bulky aromatic
ligands and are nearly twice the size of the other compounds,
thus forcing the tumbling rate in solution to be slower. The
decreased tumbling rate in the liquid could lead to incomplete
averaging of the shift tensor components and yield a skewed
(17) Wiberg, K. B.; Hammer, J. D.; Keith, T. A.; Zilm, K. W. J. Phys.
Chem. A 1999, 103, 21.

Liquid 119Sn NMR Spectroscopy of Sn(II) Compounds
Inorganic Chemistry, Vol. 39, No. 24, 2000 5453
Table 2. Summary of 119Sn Solid-State NMR Data and X-ray Parameters in ArSnX Compounds (Ar ) C6H3-2,6-Trip2; X ) Cl,
Cr(η5-C5H5)(CO)3, t-Bu, Sn(Me)2C6H3-2,6-Trip2)
X
δ
11 (ppm)
δ
22 (ppm)
δ
33 (ppm)
δ
iso (ppm)
δ
mas (ppm)
ø
θ (deg)
rx (Å)
rAr (Å)
Cl
3021
4426
4494
781
826
1322
697
111
866
-165
814
697
58
-259
1227
2187
1962
317
1176
-
1829
263
-
2691
3358
3798
-
99.68
110.13
101.79
119.3
-
2.409
2.847
2.228
2.891
-
2.180
2.214
2.211
2.201
Cr(η5-C5H5)(CO)3
t-Bu
a
Sn(Me)2C6H3-2,6-Trip2
{Sn(µ-Cl)C6H3-2,6-Mes2}
†
1643
750
-
-
a Chemical shift tensor information for only the Sn(III) site.
value for δliq. The other possibility for this discrepancy,
applicable also to the Sn(Cl)C6H3-2,6-Trip2 compound, is
chemical exchange. As noted8 by Zilm and co-workers, the 119Sn
NMR spectrum of R2SnSnR2 (R ) CH(SiMe3)2) in liquid phase
displays effects of chemical exchange (between the dimer R2-
SnSnR2 and the monomer SnR2), and the true isotropic shift
can only be obtained from solid-state spectra. Examination of
the X-ray crystal structures of the two dinuclear tin complexes
studied here, 2,6-Trip2H3C6S¨n-Sn(Me)2C6H3-2,6-Trip2 and
{Sn(µ-Cl)C6H3-2,6-Mes2}2), and the monomeric X ) Cl
compound, suggests that only in the case of {Sn(µ-Cl)C6H3-
2,6-Mes2}2, where a monomer-dimer equilibrium is possible,
could such a process account for the discrepancy between the
solid- and liquid-phase spectra.
Purcell technique gives T2 ) 3.8 ms for the Sn(III) site at δliq
) +257.4 ppm and T2 ) 38 µs for the Sn(I) site at δliq
)
+2856.9 ppm. An estimate of the size of the chemical shift
anisotropy for the Sn(I) site can be made by assuming that T2
is governed by this anisotropy, a reasonable assumption for
large, heavy molecules far from the extreme narrowing limit.18
This relaxation rate in terms of the 119Sn gyromagnetic ratio γ,
magnetic field B0, correlation time τc, shift tensor elements δnn,
and the isotropic shift δiso is given by
2
2γ2B0 τ
1
T2
)
c((δ11 - δiso)2 + (δ22 - δiso)2 +
15
(δ33 - δiso)2) × 10-12 (1)
where the 10-12 is a conversion factor between ppm and absolute
shielding. Taking δiso ≈ δliq from Table 1, T2 ) 3.8 ms, and
the δnn values from Table 2 for the Sn(III) site gives τc ) 261
ns from eq 1. Using this τc value along with T2 ) 38 µs and
δliq ) +2856.9 ppm for the Sn(II) site, the approximation of
an axially symmetric shift tensor δ22 ) δ33, and the definition
of δiso ) (δ11 + δ22 + δ33)/3 allows calculation of δ11 and δ33
from eq 1 as +7589 ppm and +491 ppm, respectively. Although
an axially symmetric tensor was assumed, these values are most
likely accurate to within (1000 ppm. Comparison of the
anisotropy for these estimated components ø ) 7098 ppm with
ø ) 3798 ppm for the X ) C(CH3)3 compound, the widest
powder pattern observed in this study, indicates that the Sn(I)
pattern will be about two times wider or about 1.0 MHz, well
beyond both the excitation bandwidth and detection limits of
the pulsed NMR spectrometer. These limitations were seen by
Exploration of 2,6-Trip2H3C6S2n-Sn(Me)2C6H3-2,6-Trip2.
To further investigate the source of this solid/liquid chemical
shift discrepancy and to determine why 119Sn solid-state NMR
spectra for the divalent tin (formally Sn(I)) site in the 2,6-
Trip2H3C6S¨n-Sn(Me)2C6H3-2,6-Trip2 compound could not be
obtained, its solution 119Sn NMR spectra were studied in the
temperature range -80 °C to +25 °C. At room temperature
the 119Sn solution NMR spectrum contains two peaks at +2856.9
ppm and +257.4 ppm, each flanked by satellites due to a 119Sn-
119Sn J coupling value of 8332 Hz. It is likely that the 119Sn-
117Sn J coupling is also present, but this splitting is probably
within the broad line width of the satellites. No evidence of
peak separation ear-marking different products and no evidence
of new peaks were observed in this temperature range. The J
coupling also remained constant with temperature. The chemical
shifts δliq and line width did change with temperature. The shift
1
investigating the Sn(I) site in this compound using H-119Sn
δliq for the Sn(I) site shifted upfield while δliq for the tetravalent
cross polarization with the 119Sn rf carrier set to the isotropic
+2856.9 ppm shift. The resulting spectrum was a 50 kHz wide
Gaussian line, a width directly corresponding to the rf-excitation
bandwidth of the NMR spectrometer. This result supports the
existence of an extremely broad chemical shift powder pattern
for the Sn(I) site in the 2,6-Trip2H3C6S¨n-Sn(Me)2C6H3-2,6-
Trip2 compound.
(formally Sn(III)) site shifted downfield as the temperature was
decreased. Lowering the temperature also increased the line
width of the spectrum for each site reflecting the longer
rotational correlation time and shorter transverse relaxation time
T2 at lower temperature. This variable temperature data (together
with bonding considerations) suggest that chemical exchange
between different chemical compounds is unlikely to be the
source of the discrepancy between δliq and δiso values. The
change in δliq with temperature suggests that it is more likely
the large size of the molecules, coupled with the extremely large
shift anisotropies shown in Table 2, are the source of this
difference. Larger molecules translate into longer correlation
times and thus less-efficient motional averaging. For shift
anisotropies on the order of tens of ppm, it is likely that
decreased motion will not substantially change δiso from δliq.
On the other hand, where anisotropies on the order of thousands
of ppm are observed in the solid state, as they are here, decreased
motion could profoundly affect the difference between δliq and
δiso.
Conclusion
The δnn components of the shift tensor in the series of
compounds with X ) Cl, Cr(η5-C5H5)(CO)3, and t-Bu can be
explained on the basis of chemical structure rather than
electronegativity values of the substituents. It is this point which
underscores the strength of solid-state NMR over liquid-state
experiments where trends in chemical shift display no apparent
correlation with structural data.
Acknowledgment. We are grateful to the NSF for financial
support. M.P.A. thanks The Packard Foundation for support.
Finally, it is important to comment on the absence of solid-
state data in Table 2 for the Sn(I) site in the 2,6-Trip2H3C6-
S¨n-Sn(Me)2(C6H3-2,6-Trip2) compound. Measurement of the
room-temperature spin-spin relaxation time T2 using the Carr-
IC000514Z
(18) Poole, C. P.; Farach, H. A. Relaxation in Magnetic Resonance;
Academic Press: New York, 1971; p 75.