Solution NMR study of methyltin(IV)-cupferronates: Analysis of a unique dismutation reaction and preorganization process

Article abstract of DOI:10.1021/om050279y

The solution nuclear magnetic resonence of methyltin(IV)-cuferronates and a unique dismutation reaction and preorganization process was analyzed. The mechanism of the methyl-migrational process was proposed on the basis of solvent induced conformational equilibrium. Kinetic experiments in the NMR was carried out to prove the deduction concerning the reaction mechanism. It was observed that stronger lewis base solvents induced dismutation at markedly faster rates than methanol.

Full text of DOI:10.1021/om050279y

3784
Organometallics 2005, 24, 3784-3791
Solution NMR Study of Methyltin(IV)-Cupferronates:
Analysis of a Unique Dismutation Reaction and
Preorganization Process
Ga´bor Ta´rka´nyi* and Andrea Dea´k
Institute of Structural Chemistry, Chemical Research Center of the Hungarian Academy of
Sciences, Pusztaszeri u´t 59-67, H-1025 Budapest, Hungary
Received April 12, 2005
Multinuclear multidimensional 1H, 13C, and 119Sn NMR data indicated that both the cyclic
tetrameric [Me3Sn(ON(NO)Ph)]4 (1t) and dimeric [Me2Sn(ON(NO)Ph)2]2 (2d) supramolecules
are predominantly monomeric [Me3Sn(ON(NO)Ph)]1 (1m) and [Me2Sn(ON(NO)Ph)2]1 (2m) in
noncoordinating solvents (CDCl3, CD2Cl2) at room temperature. This has been proven by
the concerted interpretation of (1) Me-Sn-Me angles calculated from J-coupling data; (2)
tin coordination numbers deduced from 119Sn chemical shift data; and (3) molecular
hydrodynamic radii calculated from the measurement of translational diffusion constants.
Both 1m and 2m monomers underwent a cis-trans conformational exchange that is fast on
the chemical shift time scales. In noncoordinating solvents the dominance of the cis-TBP
and cis-Oh conformations was characteristic for 1m and 2m, respectively. In coordinating
solvents (DMSO, pyridine, methanol), the trans-conformers dominated and the formation
of specific solute-solvent 1m* and 2m* coordination complexes was confirmed. The solvent-
induced cis-trans conformational exchange was held responsible for the demethylation
1
reaction of 1m, which yielded 2m and the byproduct Me4Sn. H NMR kinetic experiments
have proven that the dismutation process followed a second-order 2 A f B + C type kinetics.
In Lewis base solvents, the dismutation product 2m was stabilized by the formation of a
heptacoordinated solute-solvent complex 2m*, which then underwent no further dismutation.
The stability constant of the pyridine complex of 2m was determined from the analysis of
NMR titration data. The mechanism of the methyl-migrational process was proposed on
the basis of solvent-induced 1m (cis) h 1m* (trans) conformational equilibrium. Low-
temperature 1H and 119Sn NMR experiments in noncoordinating solvents revealed the first
example of a 1m h 1t preorganization process leading to the re-formation of a solvated 20-
membered macrocycle (1t).
Introduction
received scant attention in the literature, and the
spectroscopic studies devoted to this area are frag-
mented.6-8 This led us to investigate the so far unchar-
acterized nature of the dismutation reaction of 1t by
solution NMR spectroscopy. We carried out kinetic
experiments in the NMR spectrometer to prove our
previous deduction concerning the reaction mecha-
nism: the appearance of the byproduct Me4Sn:
We have previously reported the preparation, X-ray
structural determination, and basic NMR characteriza-
tion of two supramolecular methyltin(IV)-N-nitroso-N-
phenylhydroxylaminato (-cupferronato, -ON(NO)Ph)
complexes: [Me3Sn(ON(NO)Ph)]4 (1t) and [Me2Sn(ON-
(NO)Ph)2]2 (2d) (see Chart 1).1,2 It was observed that the
crystallization of the cyclic tetramer 1t from methanol
yielded dimeric 2d. Additionally we have also seen that
the reaction of both 1t and 2d with N-donor ligands (such
as pyridine and 4,4′-bipyridine) gave identical prod-
ucts.2,3 We have deduced that the process behind the 1t
f 2d transformation is a methyl-migrational dismuta-
tion reaction.3 Similar dismutations in organometallic
chemistry may provide alternative synthetic routes, but
they are also potential sources of sample instability,
preventing the preparation of devised structures via
recrystallization.4,5 Such methyl-migrational processes
1t f 2d + 2 Me4Sn
We also aimed to clarify whether compounds 1t and
2d retain their self-assembled structures in the liquid
phase at room temperature or constitute predominantly
monomeric solvates. However, in contrast to the solid
state the evidence for the Sn-O bond formation in
(4) Gielen, M.; De Clercq, M.; De Poorter, B. J. Organomet. Chem.
1972, 34, 305-313.
(5) Bancroft, G. M.; Davies, B. W.; Payne, N. C.; Sham, T. K. J.
Chem. Soc., Dalton 1975, 973-978.
(1) Dea´k, A.; Haiduc, I.; Pa´rka´nyi, L.; Venter, M.; Ka´lma´n, A. Eur.
J. Inorg. Chem. 1999, 1593-1596.
(6) Eujen, R.; Jahn, N. J. Fluor. Chem. 1995, 71, 75-79.
(7) Craig, P. J.; Rapsomanikis, S. Inorg. Chim. Acta 1983, 80, L19-
L21.
(2) Dea´k, A.; Venter, M.; Ka´lma´n, A.; Pa´rka´nyi, L.; Radics, L.;
Haiduc, I. Eur. J. Inorg. Chem. 2000, 127-132.
(3) Dea´k, A.; Radics, L.; Ka´lma´n, A.; Pa´rka´nyi, L.; Haiduc, I. Eur.
J. Inorg. Chem. 2001, 2849-2856.
(8) Komura, M.; Tanaka, T.; Okawara, R. Inorg. Chim. Acta 1968,
2, 321-324.
10.1021/om050279y CCC: $30.25 © 2005 American Chemical Society
Publication on Web 06/23/2005

Study of Methyltin(IV)-Cupferronates
Organometallics, Vol. 24, No. 15, 2005 3785
Chart 1
Figure 1. 1H NMR kinetic study in methanol-d4 at 25 °C.
Spectra were acquired shortly after the dissolution of the
1t crystals in 0.75 mL of solvent (0.46 M for 1m). The
relative time spacings are indicated.
Results and Discussion
Investigation of the Dismutation Kinetics. Figure
1. shows the time-dependent 1H NMR spectra of 1t
shortly after dissolution of the crystals in methanol. The
dismutation reaction proceeded spontaneously at room
temperature, yielding a mixture of Me4Sn and solvated
2d. The reaction participants were identified both by
standard additions and by various spectroscopic means
(see later). The integral ratio of the newly formed methyl
moieties remained Me2Sn:Me4Sn ) 1:2 at each data
point up to 50% conversion. This observation has
verified our previous assumption that the methanol
solvation of 1t is accompanied by a mutual methyl group
migration between two central tin atoms, preventing the
preparation of 1t via recrystallization from said solvent.
Little is known about the exact mechanism of this
dismutation process. The concentration change of dis-
solved 1t recorded by the appropriate Me3Sn integrals
as a function of time revealed second-order kinetics. The
simplified kinetic equation that describes the system in
agreement with the measured stoichiometry of the
methyl protons is
solution often remains encoded in averaged NMR pa-
rameters due to the various dynamic (e.g., internal
motion, chemical exchange) processes at room temper-
ature.9 We expected from our previous experience that
the interaction of organotin(IV) compounds with coor-
dinating solvent molecules can interfere with the net-
work of solute-solute association equilibria. When some
of these processes induce specific solute-solute interac-
tions, this leads to ligand exchange reactions. Despite
years of spectroscopic studies in organotin chemistry,
the tendency of organotin(IV) compounds to form sol-
vated self-associated complexes still makes the charac-
terization of their solution structures far from straight-
forward.
We report a summary of solution NMR results after
dissolving the structurally well-defined crystals 1t and
2d in various solvents. Kinetic, temperature-dependent,
and tin-based titration measurements10 not yet pub-
lished for the above family of compounds are presented.
Diffusion ordered spectroscopy (DOSY) is utilized for the
characterization of the self-diffusion behavior of meth-
yltin(IV)-cupferronate solvates. Being inversely propor-
tional to the hydrodynamic radius, the diffusion con-
stant is an important molecular property in the
identification of dissolved supramolecular organo-
metallic species.11-13
2 A f B + C
(1)
In other words when the dissolved species 1t and 2d are
formally expressed in terms of the corresponding mono-
mers, 2 [1m] f [2m] + [Me4Sn] holds for the measured
NMR integral ratios. Accordingly the 1/[A] vs t repre-
sentation in Figure 2 provided a linear fit, and the rate
constant k (L‚mol-1‚s-1) could be determined from the
slope (at [A]0 ) 0.46 mol/L). The numeric results of the
kinetics measured in five different solvents using the
same initial concentrations are compared in Table 1. It
was found that stronger Lewis base solvents (C5D5N,
DMSO) induced dismutation at markedly faster rates
than methanol, while the process was two magnitudes
slower in acetone-d6 and nearly infinitely slow in a
nonpolarizing solvent, CDCl3. In the latter, radical
photodegradation of the cupferronato moiety accompa-
nied by the appearance of subsequent degradation
products prevented us from measuring the reaction
rates for more than 2 days.
(9) Pons, M.; Millet, O. Prog. Nucl. Magn. Reson. 2001, 38, 267-
324.
(10) Tsagatakis, J. K.; Chaniotakis, N. A.; Jurkschat, K.; Damoun,
S.; Geerlings, P.; Bouhdid, A.; Gielen, M.; Verbruggen, I.; Biesemans,
M.; Martins, J. C.; Willem, R. Helv. Chim. Acta 1999, 82, 531-542.
(11) Valentini, M.; Pregosin, P. S.; Ruegger, H. Organometallics
2000, 19, 2551-2555.
(12) Zuccaccia, C.; Bellachioma, G.; Cardaci, G.; Macchioni, A.
Organometallics 2000, 19, 4663-4665.
(13) Ribot, F.; Escax, V.; Martins, J. C.; Biesemans, M.; Ghys, L.;
Verbruggen, I.; Willem, R. Chem. Eur. J. 2004, 10, 1747-1751.
Solvent Effect and Oligomerization Equilibria.
Since the reaction rates were influenced by the coordi-

3786 Organometallics, Vol. 24, No. 15, 2005
Ta´rka´nyi and Dea´k
Table 1. Dependence of the Rate of Reaction k on the Choice of the Solvent for the Dismutation of 1t/m
(0.46 M for 1m)
CDCl3
acetone
methanol
pyridine
DMSO
k (L‚mol-1‚s-1
)
3.38 × 10-7 ( 3.7 × 10-9 2.97 × 10-5 ( 4.5 × 10-8 6.98 × 10-5 ( 1.3 × 10-7 9.03 × 10-5 ( 5.4 × 10-8
1
Table 2. Chemical Shift δ(119Sn) and J-Coupling [2J(119Sn-1H), J(119Sn-13C)] Data of
Methyltin(IV)-Cupferronates 1t, 2d, and 3m Dissolved in Various Solvents (30 mM for the monomers)a
solvent
NMR parameter
compound
CDCl3
acetone
methanol
pyridine
DMSO
δSn119 (ppm)
1m/t
2m/d
3m
1m/t
2m/d
3m
1m/t
2m/d
3m
+114.0
-159.1
-478.6
54.8
+100.0
-166.9
-478.0
58.0
+58.1
-316.6
-479.3
64.6
108.4
124.9
481
+32.4
-367.7
-474.1
63.2
112.0
124.2
482
+19.1
-361.4
-474.6
61.2
115.2
123.4
509
2
| J1H-119Sn| (Hz)
76.8
79.3
125.3
405
660
1178
125.9
425
679
1186
1
| J13C-119Sn| (Hz)
1029
1187
1119
1174
1118
1163
a Indexes refer to the possible equilibrium between monomeric/oligomeric states.
nating strength of the solvent, we attempted to find the
spectral differences between measurements in coordi-
nating and noncoordinating solvents that could ratio-
nalize this correlation. Because the stoichiometry of the
dismutation reaction is such that it allows the inter-
pretation of the mechanism on the basis of intramo-
lecular rearrangements within the tetrameric species
1t, it raised the question whether compounds 1t and 2d
exist as stable symmetric self-associates or are predomi-
nantly monomeric (1m and 2m) in solution. Answering
this requires some circumspection due to the presence
of [1t] ≡ [1m]4 h 4 [1m] and/or [2d] ≡ [2m]2 h 2 [2m]
equilibria in solution. Initially, it is unspecified whether
these equilibria are fast on the chemical shift time scales
(1H, 13C, 119Sn) or are shifted toward either species:
monomeric or oligomeric. In both extreme cases one set
of resonances are detected due to the symmetric nature
of the supramolecules. However when the equilibria are
shifted by concentration and temperature changes or
by the presence of complexing ligands (e.g., coordinating
solvents), the dynamic nature of the various solute-
solute and solute-solvent complexation processes can
be fully characterized. These experimental results will
be presented in the following.
Figure 2. 1/[A] vs t representation of the second-order
dismutation kinetics of dissolved 1t in various solvents (25
°C, [A]0 ) 0.46 mol/L for 1m).
tion of organotin coordination geometries without prior
knowledge about the complexation dynamics. Neverthe-
less, on the basis of the literature data for other five-
membered organotin-chelates15 (such as maltolates,16
tropolonates17-19) it can be said that the 119Sn shift value
of 2d in CDCl3 contradicts with a heptacoordinated state
of the tin that was characteristic for its crystal structure.
The CN ) 6 for compound 2d and CN ) 7 for compound
3m can be rationalized according to analogy with
tropolonates provided that both are monomeric in
solution. The 119Sn chemical shift for dissolved 1t is
acceptable for the predominance of a pentacoordinated
cis-trigonal bipyramid (TBP) conformation of a monomer
(1m) (see Chart 1), but the possibility for the formation
of a symmetrical pentacoordinated macrocycle (1t) in
equilibrium with a monomeric monodendate distorted
tetrahedron (Th) is also a plausible explanation at this
point.
Analysis of the 119Sn Chemical Shifts. We extend
here our structural study of methyltin(IV)-cupferronates
by the synthesis and characterization of a third com-
pound in the row: the monomethyltin(IV) derivative,
[MeSn(ON(NO)Ph)3]1 (3m) (Chart 1). The room-temper-
ature spectroscopic properties of compounds 1t, 2d, and
3m dissolved in various solvents are collected in Table
2. The 119Sn chemical shifts for CDCl3-solvated 1t
(+114.0 ppm) and 2d (-159.1 ppm) lie at an intermedi-
ate region for tin coordination numbers CN ) 4 or 5
and CN ) 5 or 6, respectively.14 Given that the bidentate
cupferronato anion has the tendency to show not only
five-membered chelating but also bridging- and bimetal-
lic bridging-chelating coordination patterns, in principle
we face the problem that the dynamic equilibrium of
all four coordination types (including the monodentate)
may add its contribution to the observed 119Sn chemical
shift in solution. Therefore 119Sn chemical shift data
alone are insufficient for the unambiguous determina-
(15) Otera, J.; Hinoishi, T.; Kawabe, Y.; Okawara, R. Chem. Lett.
1981 273-274.
(16) Bhattacharya, S.; Seth, N.; Gupta, V. D.; No¨th, H.; Polborn,
K.; Thomann, M.; Schwenk, H. Chem. Ber. 1994, 192, 1895-1900.
(17) Otera, J. J. Organomet. Chem. 1981, 221, 57-61.
(18) Otera, J.; Kusaba, A.; Hinoishi, T.; Kawasaki, Y. J. Organomet.
Chem. 1982, 228, 223-228.
(19) Camacho-Camacho, C.; Contreras, R.; No¨th, H.; Bechmann, M.;
Sebald, S.; Milius, W.; Wrackmeyer, B. Magn. Reson. Chem. 2002, 40,
31-40.
(14) Na´dvorn´ık, M.; Holee`ek, J.; Handl´ıø, K. J. Organomet. Chem.
1984, 275, 43-51.

Study of Methyltin(IV)-Cupferronates
Organometallics, Vol. 24, No. 15, 2005 3787
2
1
Table 3. Calculated Me-Sn-Me Bond Angles (deg), Calculated from J(119Sn-1H)a and J(119Sn-13C)b
J-Coupling Data
Me-Sn-Me
compound
CDCl3
acetone
methanol
pyridine
DMSO
calcd from
1m/t
2m/d
1m/t
2m/d
109.4
127.0
110.6
134.4
111.0
130.0
112.4
136.2
115.3
179.5
117.7
168.9
114.3
187.5
117.7
177.3
112.9
195.0
120.3
177.2
2J(119Sn-1H)
calcd from
1J(119Sn-13C)
a Refers to ref 23 in text. b Refers to ref 24 in text.
Table 4. Diffusion Constants (D) and the
Both 1t and 2d experienced significantly large upfield
119Sn shifts when dissolved in coordinating solvents,
suggesting higher coordination numbers around the
central tin atoms. This observation is in agreement with
the faster dismutation rates of 1t found in Lewis base
solvents, but no further dismutation of 2d was experi-
enced. Since the 119Sn shifts of 3m were unaffected by
the coordinating strength of the solvent, the hypothesis
about its predominantly monomeric state in solution
seems reasonable.
Approximate Spherical Hydrodynamic Radii (r) of
Methyltin(IV)-Cupferronates in CDCl3 (25 °C, 50
mM for monomers)
compound
D (× 10-10 m2/s)
r (Å)
1m
2m
3m
12.08 ( 0.07
9.44 ( 0.04
8.00 ( 0.05
3.3 ( 0.1
4.3 ( 0.2
5.0 ( 0.2
trans-Oh conformation has been suggested earlier for the
bis(kojato)dimethyltin(IV) complex in DMSO although
without consideration of a solvent coordination model.25
Analysis of the J Coupling Constants. Additional
structural characterization of the dissolved compounds
1t, 2d, and 3m was possible from the analysis of their
relevant one- 1J(119Sn-13C) (1J) and two-bond 2J(119Sn-
1H) (2J) scalar coupling constants.20 The experimental
values determined in various solvents are collected in
Table 3. Lockhart et al. have determined from solid-
1
2
The | J| and | J| values for 3m were completely un-
affected by the coordinating strength of the solvent, and
we assumed that its central tin atom is in a conforma-
tionally stable heptacoordinated state.26
Self-Diffusion Behavior and Concentration De-
pendence. The monomeric nature of solvated methyl-
tin(IV)-cupferronates was further clarified by diffusion
ordered spectroscopy (DOSY).27,28 The apparent diffu-
sion constants were determined by recording the proton
2D-DOSY spectrum for the artificial mixture of all three
compounds 1t, 2d, and 3m (Figure 3). Using the Stokes-
Einstein equation29
1
2
state NMR21 that the | J| and | J| absolute values of
these coupling constants can be used to estimate the
Me-Sn-Me bond angles with reasonably good ac-
curacy.22,23 Their theory was extended to solution NMR
as well with a notice that depending on the nature of
the chelating ligands the Me-Sn-Me angles may differ
1
2
significantly whether determined from the | J| or | J|
values.24 For solvated 2d we experienced a minor but
similar discrepancy, as the bond angles became 134.6°
kBT
D )
(3)
6πηr
1
2
(| J|) vs 127.0° (| J|) in CDCl3 (see Table 3). If 2d were
predominantly monomeric (2m) in solution, a possible
explanation for this would be a cis-octahedral (cis-Oh)
conformation around the central tin. Our in vacuo PM3
semiempirical calculations reinforced this, as the equi-
librium geometry for the monomeric cis conformer was
ca. 17 kJ/mol lower in potential energy than the corre-
sponding trans-octahedral (trans-Oh) one (Chart 1). For
the effective hydrodynamic radii (r) were calculated and
are listed in Table 4 (kB is the Boltzmann constant, T
is absolute temperature, η is viscosity). These provided
structural information about the molecular size of the
solvates in a spherical approximation. Increasing the
number of bulky Ph(NO)NO- substituents resulted in
a stepwise increase in the hydrodynamic radii. Both 1t
and 2d were bulky assemblies in the solid state, fitting
into a sphere with at least 8-10 Å radius. The smaller
experimental r values in Table 4, however, reinforced
our previous hypothesis that 1t and 2d fail to form
“stable” self-associated supramolecular complexes in
solution at room temperature: they are predominantly
monomeric solvates (1m and 2m). Combining these
results with the geometrical data available in Table 3,
the Me-Sn-Me angles suggest that 1m is dominantly
in cis-trigonal bipyramidal (cis-TBP) conformation,
whereas 2m is in a cis-Oh state in dilute solutions of
noncoordinating solvents (see Chart 1).
2
compound 1t the approximate bond angle 113° (| J|) in
CDCl3 contradicted the trans-trigonal bipyramydal (trans-
TBP) geometry necessary for the macrocyclic organiza-
tion and provided further evidence for its monomeric
nature.
1
2
The | J| and | J| coupling constants of dissolved 1t and
2d increased with the coordinating strength of the
solvents and therefore yielded larger Me-Sn-Me bond
angle values (see Table 4). In DMSO the corresponding
Me-Sn-Me bond angles reached the characteristic
solid-state values ∼120° (1t) and ∼180° (2d), for which
solute-solvent complexation of the presumably mono-
2
meric forms 1m and 2m is envisaged. Based on | J| the
Despite our efforts in observing variations in 1H, 13C,
and 119Sn chemical shifts, any room-temperature con-
(20) Howard, W. F.; Crecely, R. W.; Nelson, W. H. Inorg. Chem.
1985, 24, 2204-2208.
(25) Otera, J.; Kawasaki, Y.; Tanaka, T. Inorg. Chim. Acta 1967, 1,
(21) Manders, W. F.; Lockhart, T. P. J. Organomet. Chem. 1985,
294-296.
297, 143-147.
(26) Otera, J.; Hinoishi, T.; Okawara, R. J. Organomet. Chem. 1980,
(22) Lockhart, T. P.; Manders, W. F.; Schlemper, E. O. J. Am. Chem.
202, C93-C94.
Soc. 1985, 107, 7451-7453.
(27) Morris, K. F.; Johnson, C. S., Jr. J. Am. Chem. Soc. 1992, 114,
(23) Lockhart, T. P.; Manders, W. F. Inorg. Chem. 1986, 25, 892-
3139-3141.
895.
(28) Johnson, C. S., Jr. Prog. Nucl. Magn. Reson. 1999, 34, 203-
(24) Lockhart, T. P.; Davidson, F. Organometallics 1987, 6, 2471-
256.
2478.
(29) Edward, J. T. J. Chem. Educ. 1970, 47, 261-270.

3788 Organometallics, Vol. 24, No. 15, 2005
Ta´rka´nyi and Dea´k
1
Figure 5. Low-temperature H NMR measurements for
1m in CD2Cl2 at 130 mM concentration.
solvated 1t is rather close to its isotropic shift found
earlier in solid-state NMR experiments (-1.5 ppm, +30
°C).3 Accordingly, Figure 4 shows the first solution
evidence for a preorganization process leading to sol-
vated 1t containing a 20-membered carbon-free metal-
lamacrocycle. We note here that although Gielen et al.
have published the solid-state structure of a similar
macrocyclic tetramer with a 16-membered Sn4C4O8 ring,
no spectroscopic data were provided about its dynamics
and existence in the solution.30
Figure 3. 1H-DOSY-NMR measured for the mixture of
methyltin(IV)-cupferronates (CDCl3, 50 mM for the mono-
mers 1m, 2m, and 3m). The horizontal scale represents the
1H dimension (ppm) at the aliphatic region, whereas the
vertical scale shows the diffusion scale (m2/s).
Figure 5 shows a series of low-temperature proton
spectra for the 4 [1m] h [1t] equilibrium where the
population growth of the 1H NMR signal assigned to 1t
was followed until resharpening of the resonances below
the coalescence temperature (ca. -70 °C). The increase
2
in the | J| coupling constant seen on the 117Sn/119Sn
satellites (from 54.8 to 69.3 Hz) also verifies the 1m
f
1t transition as it corresponds to the 109° f 120° Me-
Sn-Me bond-angle variation. Since 1H resonances were
2
exchange broadened, the | J| coupling constant for
1
solvated 1t at -100 °C was determined from the H-
coupled 119Sn spectra. In contrast to 119Sn NMR, the 1H
chemical shift change between the 1m and 1t states was
rather small (ca. 40 Hz), which rationalizes the fact why
room-temperature concentration-dependent chemical
shift variations were not indicative of a low-population
1t (trans-TBP) species.
Figure 4. Temperature-dependent 119Sn NMR measure-
ments for 1m in CD2Cl2 at 130 mM concentration.
Complexation of 2m with Lewis Bases. When 2m
was likewise cooled in CDCl3/CD2Cl2 solvents, upfield
119Sn shifts were observed, suggesting a higher tin
coordination number at low temperatures. In contrast
to 1m, however, the 119Sn resonance of 2m was severely
broadened around -230 ppm at temperatures near the
freeze-out point of the solute (see Figure S1). Therefore,
resharpening of a second 119Sn resonance could not be
detected and identification of the solvated supramolecu-
lar species 2d was not possible. Nevertheless, a more
shielded 119Sn nucleus was in agreement with a cis-Oh
f trans-Oh conformational change, which is the pre-
requisite for the homodimerization process leading to
2d (Chart 1).
centration dependent experiment for 2m in CDCl3 has
failed to provide experimental detail for a strong
transient self-association behavior. For 1m some degree
1
of H line broadening effects and a less than 10 ppm
upfield 119Sn shift were detected in more concentrated
solutions (0.5 M or more) in CDCl3.
Observation of the Preorganization Process for
1t. Cooling 1m in CDCl3 provided clear evidence for the
reconstitution of the tetrameric species 1t. To extend the
temperature range toward lower values, we replaced
CDCl3 with CD2Cl2 (Figure 4). At -60 °C and below a
sharp resonance appeared significantly upfield to the
1m signal, indicating a change in the molecular geom-
etry. The upfield shift of the ca. +110 ppm resonance
can be rationalized by the cis-TBP f trans-TBP con-
formational change associated with the 1m f 1t transi-
tion. The 119Sn chemical shift (+14.3 ppm, -100 °C) of
By considering all spectroscopic data it seems that 2m
in N- or O-donor Lewis basic solvents is in fast equi-
(30) Gielen, M.; Khloufi, A. E.; Biesemans, M.; Kayser, F.; Willem,
R.; Mahieu, B.; Maes, D.; Lisgarten, J. N.; Wyns, L.; Moreira, A.;
Chattopadhay, T. K.; Palmer, R. A. Organometallics 1994, 13, 2849-
2854.

Study of Methyltin(IV)-Cupferronates
Organometallics, Vol. 24, No. 15, 2005 3789
Chart 2
librium between the monomeric hexacoordinated cis-Oh
and heptacoordinated trans-methyl state according to
the model for the pyridine complex in Chart 2. In
connection to the previously described structural modi-
fication of 2d by pyridine,3 here we were interested in
the determination of the Ka room-temperature equilib-
rium constant (eq 4) for the above process.
[complex]
[1m][Py]
Figure 6. Stack plot of 13 1H-119Sn-HMQC experiments
recorded for 2m at various pyridine concentrations in a
series of titration experiments (c2m ) 50 mM, CDCl3, 25
°C). F1 and F2 dimensions represent the 119Sn and 1H
chemical shifts, respectively. The pyridine total concentra-
tions (cpy) corresponding to the individual titration points
are indicated Figure 7.
Ka )
(4)
By the titration of 2m with pyridine in chloroform we
aimed to unravel the origin of the large N-donor-induced
119Sn chemical shifts with the separation of the specific-
ity and the strength of the solute-solvent complexation
interaction. Specificity is defined by the 119Sn chemical
shift difference between the complexed and uncom-
plexed species (∆δ∞ ) |δcomplexed - δfree|). When ∆δ∞ is
several tens of ppms large, it provides evidence for the
change in the tin coordination number and/or confor-
mational change. The magnitude of the association
constant (Ka) on the other hand is related to the strength
(or stability) of the interaction. As it was pointed out in
a previous study,31 the two parameters (∆δ∞ and Ka)
responsible for the magnitude of the solvent-induced
chemical shifts (|δobs - δfree|) are independent and their
effect is separable. Using the gradient-enhanced version
of indirect detection 1H-119Sn-HMQC experiments32-36
we recorded the 119Sn chemical shifts for 2m after each
pyridine addition (Figure 6). By knowing the total
concentration of the dissolved species (c2m for 2m and
cpy for pyridine:) the Ka association constant was deter-
mined from the δobs vs cpy plot by a well-described
nonlinear fitting procedure (Figure 7) using eq 5.
Figure 7. Plot of δobs as a function of the pyridine total
concentration (cpy) at constant organotin concentration
(c2m). The 119Sn chemical shift values stem from Figure 6
(CDCl3, 25 °C).
chemical shift for 2m. We found that a relatively small
Ka association constant (Ka ) 2.78 ( 0.02 L/mol) and a
large shift difference (∆δ∞ ) +215.3 ppm) were char-
acteristic of this system at room temperature. This
allows us to conclude that the above complexation
interaction is rather a highly specific than a highly
stable one: despite the small stability of the complex
large |δobs - δfree| solvent shifts are induced at room
temperature due to remarkable changes in the tin
geometry. It is worth noting that although the 1H shift
variations were not significant in the titrations (Figure
δobs
)
-1
-1
(δcomplex - δfree)(c2m + cpy + Ka
-
(c2m + cpy + Ka ) )
2 - 4c2mcpy
x
+
2c2m
δfree (5)
where δobs are the measured 119Sn shifts in the presence
of pyridine, whereas δfree is the uncomplexed 119Sn
(31) Ta´rka´nyi, G. J. Chromatogr. A 2002, 961, 257-276.
(32) Kayser, F.; Biesemans, M.; Boualam, M.; Tiekink, E. R. T.;
Khloufi, A. E.; Meunier-Piret, J.; Bouhdid, A.; Jurkschat, K.; Gielen,
M.; Willem, R. Organometallics 1994, 13, 1098-1113.
(33) Willem, R.; Bouhdid, A.; Kayser, F.; Delmotte, A.; Gielen, M.;
Martins, J. C.; Biesemans, M.; Mahieu, B.; Tiekink, E. R. T. Organo-
metallics 1996, 15, 1920-1929.
(34) Willem, R.; Bouhdid, A.; Meddour, A.; Camacho-Camacho, C.;
Mercier, F.; Gielen, M.; Biesemans, M.; Ribot, F.; Sanchez, C.; Tiekink,
E. R. T. Organometallics 1997, 16, 4377-4385.
(35) Ribot, F.; Sanchez, C.; Meddour, A.; Gielen, M.; Tiekink, E. R.
T.; Biesemans, M.; Willem, R. J. Organomet. Chem. 1998, 552, 177-
186.
(36) Martins, J. C.; Biesemans, M.; Willem, R. Prog. Nucl. Magn.
Reson. 2000, 36, 271-322.

3790 Organometallics, Vol. 24, No. 15, 2005
Ta´rka´nyi and Dea´k
1
6), H-119Sn-HMQC experiments revealed large 119Sn
It is out of the scope of this paper to further describe
each contribution that determines the relative rate of
the dismutation for the various coordinating solvents
(CD3OD, DMSO-d6, and C5D5N). The fine balance
between Lewis basicity, steric, and byproduct (Me4Sn)
solvation factors in methylated solvents (e.g., DMSO)
is probably responsible for the fastest reaction rates
found in DMSO (Table 1).
shift changes, which makes this technique useful for the
characterization of weak transient complexes of orga-
notins in dilute solutions. The values δcomplex (-374.4
ppm) and ∆δ∞ (+215.3 ppm) determined by nonlinear
fitting are also important structural parameters proving
a CN ) 6 to 7 coordination change around the central
tin. Similar titration experiments for diorganotins have
not yet been reported in the literature.
Proposed Mechanism of the Dismutation Reac-
tion. Because the solution structures of 1m and 2m are
now well characterized, we arrive at setting up a
proposed mechanism for the dismutation process. The
spectroscopic differences for 1m between measurements
in coordinating and noncoordinating solvents (Tables 2
and 3) can be interpreted on the basis of a 1m (cis-TBP)
f 1m* (trans-TBP) conformational change rather than
according to a 1m (cis-TBP) f 1t (trans-TBP) preorga-
nization model. The solute-solute interactions between
the monomeric 1m species, which only dominate at low
temperatures in noncoordinating solvents, are replaced
by solute-solvent interactions at room temperature in
the presence of N- or O-donors in the solvent bulk. As
a result, the three methyl carbons of 1m are pushed by
the O- or N-donor solvent (S) into a single plane, forcing
the cupferronato anion to coordinate (from the opposite
direction) rather as a monodendate ligand (1m* in Chart
1). Similar solvent effects were described for other
trimethyltin derivatives with monodendate ligands.37
The solvent-perturbed conformational equilibrium for
Conclusions
We have proven that the supramolecular structures
1t and 2d are predominantly monomeric (1m and 2m) in
solution at room temperature. Both monomeric struc-
tures underwent a cis-trans conformational exchange
that is fast on the chemical shift time scales. In
noncoordinating solvents (CDCl3, CD2Cl2) the domi-
nance of the cis-TBP and cis-Oh conformations was
characteristic for 1m and 2m, respectively. Coordinating
solvents (DMSO, pyridine, and methanol) increased the
population of the trans-states, increasing the coordina-
tion number of the central tins. Solvent-induced con-
formational change was held responsible for a unique
demethylation process in which 1m was converted into
2m, yielding the byproduct Me4Sn. Such transformations
may explain why no trimethyltin(IV) derivatives of five-
membered bidendate anions (cupferronates, kojates,
tropolonates) were ever reported as being crystallized
from methanol or from other coordinating solvents.
1H NMR kinetic experiments carried out in the
spectrometer have proven that the dismutation process
followed a second-order 2 A f B + C type kinetics. The
rate of the reaction increased with the coordinating
strength of the solvent, but the kinetic model has not
changed. It seems that the dismutation of 1t in solution
is initiated by intermolecular 1m + 1m* collisions rather
than by intramolecular rearrangements within the
supramolecule 1t. The product 2m was stabilized by the
formation of specific heptacoordinated solute-solvent
complexes in Lewis base solvents 2m*, which underwent
no further dismutation. The underlying cis-Oh f trans-
Oh conformational equilibrium was verified by the
calculation of the Me-Sn-Me angles on the basis of
J-coupling data in the absence and in the presence of
O- or N-donor ligands. Additionally, the stability of the
solute-solvent complexes of 2m with pyridine and 4,4′-
bipyridyl was characterized by 119Sn NMR titration
experiments.
1
m is the most plausible explanation for the mechanism
of the dismutation reaction. Its role is most likely to
initiate a process leading to a nucleophilic substitution
in which the cupferronato anion is kept in a monoden-
date transition state whose lifetime is long enough for
the proper ligand exchanges to take place via bimolecu-
lar collisions. This is in agreement with the fact that
the dismutation reaction proceeded at slightly faster
rates in DMSO when repeated with lower [A]0 initial
concentrations of 1m (k ) 1.0671 × 10-4 ( 1.3 × 10-7
L‚mol-1‚s-1 at [A]0 ) 0.110 mol/l, and k ) 1.6538 × 10-4
( 3.7 × 10-7 L‚mol-1‚s-1 at [A]0 ) 0.030 mol/L, see Table
1). This finding has modified our understanding of the
dismutation kinetics, as we previously assumed that a
relatively high initial concentration of 1m is a prereq-
uisite of the transformation. Because titrations have
shown that 2m is stabilized by solvent complexation, we
summarize the role of solvent activation according to
the following:
Low-temperature 1H and 119Sn NMR experiments in
noncoordinating solvents revealed the first example of
a preorganization process toward a solvated 20-mem-
bered macrocyclic structure in 1t. The thermodynamic
properties of the preorganization process in comparison
with those of other triorganotin(IV) derivatives will be
published elsewhere.
Compound 3m prepared as a standard for diffusion
experiments showed neither solute-solute or solute-
solvent complexation behavior nor spontaneous dis-
mutation reaction in any solvents.
solvent
1m
y
z 1m*
(I)
1m + 1m* f 2m* + Me4Sn
(II)
solvent
2m* y
z 2m
(III)
Both solvent-coordinated complexes 1m* and 2m* are
abundant because of the omnipresent O- or N-donors
in the bulk solvent. Processes I and III are very fast
and reversible. The rate-determining step is then pro-
cess II (as confirmed by the kinetic data), and conse-
quently the order of the reaction remains 2.
Experimental Section
NMR Experiments. NMR spectra were recorded on a 400
MHz (for 1H) Varian INOVA spectrometer equipped with a
Varian 5 mm 1H-19F/{15N-31P} Z-gradient indirect detection
(37) Bolles, T. F.; Drago, R. S. J. Am. Chem. Soc. 1966, 88, 5730-
5734.

Study of Methyltin(IV)-Cupferronates
Organometallics, Vol. 24, No. 15, 2005 3791
probe and a direct detection Varian 15N-31P/{1H-19F} swit-
chable broadband probe (-100 to +150 °C). 1H and 13C
chemical shifts are referenced to the residual solvent signals,
whereas 119Sn shifts are given relative to the external Me4Sn
(0.00 ppm) signal. 99.95% perdeuterated solvents were pur-
chased from Cambridge Isotope Laboratories. Kinetic experi-
ments were carried out in 5 mm NMR tubes sealed at
constriction shortly after dissolution of the crystalline samples.
molar ratio of the solutes was followed by accurate 1H
integration of the pertinent resonances. 1H-119Sn-gHMQC
experiments were run to obtain 119Sn chemical shift data in
10.5 Hz digital resolution in the F1 dimension. Titration data
were plotted and fitted in Microcal Origin 6.1 using a least-
squares nonlinear fitting algorithm.
Computational Study. Molecular modeling and geometry
optimizations have been performed on a Silicon Graphics
Octane computer at the PM3 semiempirical level, employing
the MOPAC package in InsightII.
1
The tube was then inserted in the spectrometer, and the H
NMR spectra were monitored by setting a 600 s preacquisition
delay in an arrayed overnight experiment.
Preparation of Methyltris(N-nitroso-N-phenylhydrox-
ylaminato)tin(IV) (3m). The ammonium salt NH4[PhN(O)-
NO] (0.70 g, 4.5 mmol) dissolved in 15 mL of distilled water
was added to a solution of MeSnCl3 (0.36 g, 1.5 mmol) in 5
mL of distilled water. Precipitation of the complex occurred
immediately. After stirring for 15 min at room temperature
the precipitate was filtered off and washed with cold distilled
water and dried in air. The complex is hygroscopic and, after
few weeks of standing, converted into monohydrated material
according to 1H NMR (CDCl3, 25 °C). Yield: 0.73 g, 89%. Mp:
79-80 °C. FT-IR (KBr): ν 3432 cm-1 (O-H) (w b), 3105-2921
(w b), 1591 (O-H) (w), 1486 (Ph) (m), 1463 (Ph) (ms), 1347
(N-N) (ms), 1295 (vs), 1224 (NdO) (ms), 1184 (NdO) (ms),
1165 (ms), 1066 (ms), 1022 (m), 937 (ONNO) (s), 926 (ms sh),
756 (Ph) (s), 694 (Ph) (s), 683 (OSnO) (s sh), 617 (w), 600 (sh),
573 (w b), 505 (w), 401(Sn-O) (vs). 1H NMR (399.89 MHz,
Room-temperature 119Sn chemical shift data were obtained
1
from H-119Sn-gHMQC experiments (for example see Figure
S2), while one-dimensional 119Sn was measured at low tem-
peratures. In variable-temperature studies a 20 min precon-
ditioning time was applied before each acquisition. 128 scans
were collected for the 1H-decoupled one-dimensional 119Sn
spectra with 5 s repetition time. VT gas was cooled by liquid
nitrogen. 2J(119Sn-1H) coupling constants were determined
directly from 1H NMR spectra (Figure S3). Using the method
1
proposed by Kupcˇe and Wrackmeyer et al.38-41 the J(119Sn-
13C) coupling constants were determined from the 119Sn
satellites seen in F1 traces of the gradient-enhanced 1H-13C-
HSQC experiments (2.7 Hz digital resolution) (see Figure S4).
The DOSY experiment was carried out in a 5 mm Shigemi
tube (CDCl3 matched) at 25 °C to minimize convection effects.
A Performa I. gradient amplifier was used. The gradient
strength was calibrated by using 5 w/w% sucrose in D2O at
25 °C (D ) 5.22 × 10-10 m2/s). The bipolar pulse-pair
stimulated echo (Dbppste) sequence was used for acquiring
diffusion data with 50 ms diffusion delay, 16 squared incre-
ments for gradient levels, and 16 transients. The Varian DOSY
package was used for the processing. Chloroform viscosity, η
) 5.42 mPa, was used for the calculation of the hydrodynamic
radii at 25 °C.
2
CDCl3): δ 1.12 (s, J(117Sn/119Sn-1H) ) 120.1/125.3 Hz, 3H,
CH3); 1.55 (s, 2H, H2O); 7.43-7.50 (m, 9H, CH); 7.92-8.00 (m,
6H, CH). 13C NMR (100.56 MHz, CDCl3): δ 7.3 (s, 1J(117Sn/
119Sn-13C) ) 1123/1178 Hz, CH3); 119.8 (s, CH); 129.2 (s, CH);
130.5 (s, CH); 139.6 (s, C). 119Sn NMR (149.08 MHz, CDCl3):
δ -478.6 ppm. Anal. Calcd for anhydrous 3m C19H18,N6,O6,Sn
(545.10): C, 41.9; H, 3.3; N, 15.4. Found: C, 41.9; H, 3.5; N,
15.4.
Titrations were carried out in a 5 mm NMR tube by adding
volumes of pyridine solution with a Gilson 50 µL analytical
pipet keeping the total organotin concentration constant. The
Acknowledgment. The authors would like to thank
Dr. Pe´ter Sa´ndor (Varian Deutschland Gmbh) for NMR
application support. We thank the Hungarian Research
Fund (OTKA) Grant T047321 for supporting the work.
(38) Kupcˇe, Eh.; Wrackmeyer, B.; Granger, P. J. Magn. Reson. 1969,
100, 401-405.
(39) Wrackmeyer, B.; Kupcˇe, Eh.; Ku¨mmerlen, J. Magn. Reson. Chem.
1992, 30, 403-407.
Supporting Information Available: Four figures show-
ing details of NMR experiments. This material is available free
of charge via the Internet at http://pubs.acs.org
(40) Wrackmeyer, B.; Kupcˇe, Eh.; Kehr, G.; Sebald, A. Magn. Reson.
Chem. 1992, 30, 964-968.
(41) Wrackmeyer, B.; Horchler von Locquenghien, K.; Kupcˇe, Eh.;
Sebald, A. Magn. Reson. Chem. 1993, 31, 45-50.
OM050279Y