Article
Organometallics, Vol. 29, No. 17, 2010 3865
to four equivalents of sodium in liquid ammonia. Water and
other highly active hydrogen-donating impurities can be
most likely excluded as hydrogen source, since they are
expected to react rapidly with sodium in liquid ammonia
before the addition of the stannanes.
additional spectra in Figure S6 in the Supporting Informa-
tion), i.e., two additional signals featured when compared to
those resulting from experiments with dichlorodiphenylstan-
nane (occasionally, a minor signal at 27 ppm was also pre-
sent, probably as a result of a reaction with traces of oxygen).
The two main signals (with respect to the integrated inten-
sities) were linked to the binuclear species tetrabutyldistannide,
After exposure of dideuterodiphenylstannane to four equiva-
lents of sodium in liquid ammonia for 60 min, the 119Sn NMR
spectrum showed, as expected, on one hand the signal at -116
ppm of tetraphenyldistannide at the measurement temperature
of 220 K (see above) and, on the other hand, two signals in the
hydride region (Figure 3c). A small signal, which appeared as a
singlet at -192.9 ppm in the proton-broadband-decoupled
spectrum, split into a doublet in the proton-coupled spectrum
(Figure 3b; a series of additional spectra recorded at reaction
times between 30 min and 10 h are displayed in Figures S4 and S5
of the Supporting Information), which is indicative of the above-
mentioned hydrodiphenylstannide. Further, a pronounced three-
line feature with 1J(Sn,D) = 25 Hz at -196.8 ppm (center peak)
is indicative of the presence of deuterodiphenylstannide,
Ph2SnD- (NB: the three signals in the proton-coupled spectra
are poorly resolved due to additional couplings with the phenyl
protons).
2-
(Bu4Sn)2 (-161 ppm at 200 K), and hydrodibutylstannide,
Bu2SnH- (-228 ppm at 200 K). Accordingly, the latter signal
splits into a doublet in proton-coupled 119Sn NMR spectra
(1J(119Sn-1H) = 96 Hz at 220 K), but the resolution is relatively
low due to the coupling with the methylene protons of the butyl
groups. The resolution decreased even more upon very long
periods of measurement (>7 h), which is probably a result of the
limits in temperature control: the chemical shift strongly depends
on the temperature (Δδ = 0.5 ppm K-1; cf. Table 1; compare
also to the fluctuations of chemical shifts of phenylstannides
during NMR measurements, as displayed in Figures S4 and S5
in the Supporting Information).
The additional signals are presumably due to butyl group
migration (see also below, reactions with bromoethane section).
The signal at -136 ppm (at 200 K) obviously represents tri-
butylstannide, SnBu3-, since the reaction mixture of chlorotri-
butylstannane and two equivalents of sodium in liquid ammo-
nia resulted in a single peak at the same chemical shift.
The intensity of the Ph2SnH- signal observed after one hour
is by far too strong to be due to residual hydrogen atoms in the
1
starting compound Ph2SnD2 (as evident from analysis of H
and 119Sn NMR spectra of Ph2SnD2 in organic solvents).
Subsequently, the ratio between hydrodiphenylstannide and
deuterodiphenylstannide increased steadily during a period of
more than one week, indicating that further H-D exchange
occurred very slowly (∼25% exchange after 12 h at 220 K and
45% after one week, monitored directly in the NMR test tube).
It appears, therefore, that at least two processes contribute
to the H-D exchange: one largely advancing within one hour
or less, probably along the reaction path from Ph2SnD2 to
Ph2SnD-, and another one lasting for days or weeks. Since
Ph2SnD2 itself is essentially insoluble in liquid ammonia (it
dissolves only upon treatment with sodium), it seems unlikely
that the faster of the two processes is due to a H-D exchange of
Ph2SnD2 with the solvent (ammonia). Besides, it is worth noting
that the percentage of tetraphenyldistannide compared to the
sum of the two monostannides remained constant over time,
within experimental precision.
Since the hydrogen atoms of hydrodiphenylstannide ex-
tracted from dichlorodiphenylstannane or dihydrodiphenyl-
stannane at least partially stem from different sources, hydro-
diphenylstannide is formed by different processes. This might be
due to the higher reactivity of sodium to Sn-Cl than Sn-H
groups at initial stages of the reaction. Dichlorodiphenyl-
stannane may initially lose both chlorine atoms and quickly
react with hydrogen atoms from the solvent (liquid NH3) to
form the hydrodiphenylstannide. Dihydrodiphenylstannane
initially loses preferentially only one hydrogen atom, while the
other remains bound to the tin atom for a long time. The
exchange of the remaining hydrogen (deuterium) atom with
hydrogen atoms from the solvent is a second reaction step
proceeding at a different time scale and with complex reaction
order. The initial reactions that lead to the distannide and the
hydrostannide (or deuterostannide) are completed within less
than 60 min, whereas the second step takes days.
The signal at -212 ppm showed pronounced broadening in
the proton-coupled 119Sn NMR spectra, at least partially due to
nonresolved couplings to protons of the butyl groups. Yet the
coupling of 119Sn nuclei to protons of Sn-H bonds is signifi-
cantly stronger, and therefore, the line broadening increases
additionally in species with nonresolved Sn-H bonds. The full-
width at half-maximum (fwhm) of the peak at -212 ppm (-208
ppm at 220 K) extended in the proton-coupled spectra by 140%
(170% at 220 K) compared to the decoupled spectra (cf.
Table 1). This is even more than the signal of the hydrostannide
at -228 ppm (-218 ppm at 220 K), which was broadened by
about 80% (60% at 220 K). For comparison, the fwhm of the
two peaks at -136 and -161 ppm (-137 and -142 ppm at
220 K), which represent species without Sn-H bonds, was only
little influenced when proton-coupled and proton-decoupled
spectra are compared; the broadening amounted to only 15-
20% at 220 K and even less at 200 K. Thus, the signal at -212
ppm may represent, for instance, a mononuclear dihydrostan-
nide or, since the signal intensity was not sufficiently high to
allow detection of tin satellites, a binuclear hydrostannide.
Reaction of the Stannide Intermediates with Bromoethane.
As mentioned above, it has been postulated that the inter-
mediate diorganostannide dianion can be trapped by reac-
tion with organohalides (Scheme 1).20-23 Accordingly, we
transferred solutions with the in-situ-prepared stannides in
the final state into a large excess of precooled bromoethane.
In the case of the intermediates resulting from conversion of
dichlorodiphenylstannane, only diethyldiphenylstannane,
Et2Ph2Sn, was found after the reaction (Figure 4a, for
chemical shifts see Experimental Section).
The stannides resulting from conversion of dichlorodibutyl-
stannane with bromoethane yielded two additional products
compared to the analogous conversion with dichlorodiphenyl-
stannane. Note that in the former case also two additional
reaction intermediates were detected (see above). Besides the
expected main product, dibutyldiethylstannane, Bu2Et2Sn, also
tributylethylstannane, Bu3EtSn, and butyltriethylstannane,
BuEt3Sn, were found by 119Sn NMR analysis (Figure 4b; for
chemical shifts see Experimental Section), in line with the alkyl
Reaction of Dichlorodibutylstannane with Na in Liquid
Ammonia. 119Sn NMR spectra of dichlorodibutylstannane
exposed to sodium in liquid ammonia were more complex
than those of the diphenylstannanes. In the case of dichloro-
dibutylstannane, four major signals emerged (Figure 2c,