Direct detection of dimethylstannylene and tetramethyldistannene in solution and the gas phase by laser flash photolysis of 1,1- dimethylstannacyclopent-3-enes

Article abstract of DOI:10.1021/ja052675d

The photochemistry of 1,1-dimethyl- and 1,1,3,4-tetramethylstannacyclopent- 3-ene (4a and 4b, respectively) has been studied in the gas phase and in hexane solution by steady-state and 193-nm laser flash photolysis methods. Photolysis of the two compounds results in the formation of 1,3-butadiene (from 4a) and 2,3-dimethyl-1,3-butadiene (from 4b) as the major products, suggesting that cycloreversion to yield dimethylstannylene (SnMe₂) is the main photodecomposition pathway of these molecules. Indeed, the stannylene has been trapped as the Sn-H insertion product upon photolysis of 4a in hexane containing trimethylstannane. Flash photolysis of 4a in the gas phase affords a transient absorbing in the 450-520-nm range that is assigned to SnMe₂ by comparison of its spectrum and reactivity to those previously reported from other precursors. Flash photolysis of 4b in hexane solution affords results consistent with the initial formation of SnMe₂ (λ _max ≈ 500 nm), which decays over ～10 μs to form tetramethyldistannene (5b; λ_max ≈ 470 nm). The distannene decays over the next ca. 50 μs to form at least two other longer-lived species, which are assigned to higher SnMe₂ oligomers. Time-dependent DFT calculations support the spectral assignments for SnMe₂ and Sn₂Me₄, and calculations examining the variation in bond dissociation energy with substituent (H, Me, and Ph) in disilenes, digermenes, and distannenes rule out the possibility that dimerization of SnMe₂ proceeds reversibly. Addition of methanol leads to reversible reaction with SnMe₂ to form a transient absorbing at λ_max ≈ 360 nm, which is assigned to the Lewis acid-base complex between SnMe ₂ and the alcohol.

Full text of DOI:10.1021/ja052675d

Published on Web 11/16/2005
Direct Detection of Dimethylstannylene and
Tetramethyldistannene in Solution and the Gas Phase by
Laser Flash Photolysis of 1,1-Dimethylstannacyclopent-3-enes
†
,‡
§
,§
Rosa Becerra, Peter P. Gaspar,* Cameron R. Harrington, William J. Leigh,*
§
,#
‡
Ignacio Vargas-Baca, Robin Walsh,* and Dong Zhou
Contribution from the Instituto de Quimica-Fisica ‘Rocasolano’, C.S.I.C., C/Serrano 119, 28006
Madrid, Spain, Department of Chemistry, Washington UniVersity, St. Louis, Missouri 63130,
Department of Chemistry, McMaster UniVersity, 1280 Main Street West, Hamilton, ON L8S 4M1
Canada, and School of Chemistry, UniVersity of Reading, Whiteknights, Reading, RG6 6AD U.K
Received April 25, 2005; E-mail: gaspar@wuchem.wustl.edu; leigh@mcmaster.ca; r.walsh@reading.ac.uk
Abstract: The photochemistry of 1,1-dimethyl- and 1,1,3,4-tetramethylstannacyclopent-3-ene (4a and 4b,
respectively) has been studied in the gas phase and in hexane solution by steady-state and 193-nm laser
flash photolysis methods. Photolysis of the two compounds results in the formation of 1,3-butadiene (from
4a) and 2,3-dimethyl-1,3-butadiene (from 4b) as the major products, suggesting that cycloreversion to
yield dimethylstannylene (SnMe ) is the main photodecomposition pathway of these molecules. Indeed,
2
the stannylene has been trapped as the Sn-H insertion product upon photolysis of 4a in hexane containing
trimethylstannane. Flash photolysis of 4a in the gas phase affords a transient absorbing in the 450-520-
nm range that is assigned to SnMe
from other precursors. Flash photolysis of 4b in hexane solution affords results consistent with the initial
formation of SnMe
(λmax ≈ 500 nm), which decays over ∼10 µs to form tetramethyldistannene (5b; λmax
470 nm). The distannene decays over the next ca. 50 µs to form at least two other longer-lived species,
which are assigned to higher SnMe oligomers. Time-dependent DFT calculations support the spectral
assignments for SnMe and Sn Me , and calculations examining the variation in bond dissociation energy
with substituent (H, Me, and Ph) in disilenes, digermenes, and distannenes rule out the possibility that
dimerization of SnMe proceeds reversibly. Addition of methanol leads to reversible reaction with SnMe
to form a transient absorbing at λmax ≈ 360 nm, which is assigned to the Lewis acid-base complex between
SnMe and the alcohol.
2
by comparison of its spectrum and reactivity to those previously reported
2
≈
2
2
2
4
2
2
2
electronic structure.11 It has been detected directly in the gas
phase only relatively recently, allowing study of its reactivity
toward various reagents such as alkenes, dienes, alkynes,
alcohols, alkyl halides, silicon and germanium hydrides, HCl,
and SO2. To our knowledge, however, SnMe2 has never been
detected directly in solution by time-resolved spectroscopic
methods.
For such studies to be successful, one needs a stable precursor
that undergoes photodecomposition to yield the transient of
interest cleanly and efficiently. The latter is crucial, because
the lowest-energy absorption band in the electronic spectra of
divalent group 14 compounds is characteristically weak, making
these species particularly difficult to detect reliably when they
are generated from precursors that are prone to other competing
photoreactions or whose photolysis proceeds in low quantum
Introduction
There has been considerable interest over the past 3 decades
in the chemistry of dialkyl and diaryl Sn(II) compounds
1
-5
(
“stannylenes”).
Several stable derivatives have been re-
1
2
3,4
ported, but remarkably little is known about the chemistry of
simpler derivatives that are transient species at room temper-
ature. Dimethylstannylene (SnMe2) has received the greatest
attention; some aspects of its chemistry in solution have been
6
-10
reported,
as have its infrared spectrum in a matrix at low
temperatures and ab initio calculations of its geometry and
†
Instituto de Quimica-Fisica ‘Rocasolano’.
Washington University.
McMaster University.
University of Reading.
‡
§
#
(
(
(
(
(
1) Neumann, W. P. Chem. ReV. 1991, 91, 311.
2) Driess, M.; Grutzmacher, H. Angew. Chem., Int. Ed. Engl. 1996, 35, 828.
3) Weidenbruch, M. Eur. J. Inorg. Chem. 1999, 1999, 373.
4) Tokitoh, N.; Okazaki, R. Coord. Chem. ReV. 2000, 210, 251.
5) Boganov, S. E.; Egorov, M. P.; Faustov, V. I.; Nefedov, O. M. In The
Chemistry of Organic Germanium, Tin and Lead Compounds; Rappoport,
Z., Ed.; John Wiley and Sons: New York, 2002; Vol. 2, pp 749-839.
6) Neumann, W. P.; Schwarz, A. Angew. Chem., Int. Ed. Engl. 1975, 14, 812.
7) Grugel, C.; Neumann, W. P.; Seifert, P. Tetrahedron Lett. 1977, 25, 2205.
8) Gross, L. W.; Moser, R.; Neumann, W. P.; Scherping, K. H. Tetrahedron
Lett. 1982, 23, 635.
yield. This has been demonstrated in some of our recent work
13,14
on transient germylenes in solution,
and the same is expected
(11) Bleckmann, P.; Maly, H.; Minkwitz, R.; Neumann, W. P.; Watta, B.;
Olbrich, G. Tetrahedron Lett. 1982, 23, 4655.
(12) Becerra, R.; Boganov, S. E.; Egorov, M. P.; Faustov, V. I.; Krylova, I. V.;
Nefedov, O. M.; Walsh, R. J. Am. Chem. Soc. 2002, 124, 7555.
(13) (a) Leigh, W. J.; Harrington, C. R.; Vargas-Baca, I. J. Am. Chem. Soc.
2004, 126, 16105. (b) Leigh, W. J.; Harrington, C. R. J. Am. Chem. Soc.
2005, 127, 5084.
(
(
(
(
9) Scherping, K. H.; Neumann, W. P. Organometallics 1982, 1, 1017.
(
10) Watta, B.; Neumann, W. P.; Sauer, J. Organometallics 1985, 4, 1954.
10.1021/ja052675d CCC: $30.25 © 2005 American Chemical Society
J. AM. CHEM. SOC. 2005, 127, 17469-17478
9
17469

A R T I C L E S
Becerra et al.
to be true of other group 14 carbene analogues such as silylenes
and stannylenes.
Results
Photochemical Product Studies. Gas chromatographic analy-
sis of the volatile products from preparative 193-nm photolysis
of 4a in the gas phase (0.2 Torr 4a, 100 Torr N2 as buffer gas)
indicates them to consist of 1,3-butadiene (ca. 80%), ethane (ca.
One of our groups has published several studies of the
chemistry of GeH2 in the gas phase, generated by 193-nm
1
5-21
photolysis of 3,4-dimethylgermacyclopent-3-ene (1a),
as
well as two studies of GeMe2, similarly obtained by photolysis
1
0%), ethylene (ca. 5%), and acetylene (ca. 5%) (eq 1). This is
16,22
of 1,1-dimethylgermacyclopent-3-ene (2).
Following this and
consistent with the extrusion of SnMe2 as the major excited-
state reaction pathway.
other even earlier leads,23 others of us have recently shown that
compounds of this general type are also very useful as precursors
to substituted germylenes such as dimethyl-, diphenyl-, and
dimesitylgermylene (GeMe2, GePh2, and GeMes2, respectively)
for study by time-resolved methods in solution.13 In each case,
both the germylene and its (subsequently formed) digermene
dimer can be detected and characterized by laser flash photolysis
methods, using the substituted 1-germacyclopent-3-ene deriva-
tives 1b and 3a,b in conjunction with either 193- or 248-nm
excitation, depending on the absorption characteristics of the
particular compound. A similar methodology has proven suc-
cessful for the photochemical generation of reactive silylenes
in solution.2
Photolysis of a deoxygenated 0.08 M solution of 4a in hexane
containing 0.17 M trimethylstannane (Me SnH) was carried out
3
with 254-nm light; a similar solution containing only Me SnH
3
(0.17 M) was photolyzed in parallel for comparison. GC/MS
analysis of the photolyzates after ca. 10% conversion of 4a
4,25
allowed the detection of hexamethyldistannane (Sn Me ) and a
2
6
compound tentatively identified as pentamethyldistannane
HSn2Me5) on the basis of mass spectral evidence (see Sup-
(
porting Information) as the only volatile tin-containing products;
the latter was formed in roughly 30% of the yield of Sn2Me6 in
the photolysis of the 4a/HSnMe3 mixture, but in only trace
amounts in the photolysis of HSnMe3 alone. The results, which
are summarized in eq 2 as the product GC peak areas relative
to internal standard in the two experiments, suggest that the
Sn2Me6 formed is derived mainly from photolysis of HSnMe3,
whereas HSn2Me5 (the expected product of reaction of SnMe2
with HSnMe3) is derived from photolysis of 4a.
In this paper, we report the results of a study of the
photochemistry of the potential SnMe2 precursors 4a and 4b in
the gas phase and in hexane solution by steady-state and laser
flash photolysis methods. In the gas phase, 4a affords a transient
species with absorption characteristics similar to those previously
obtained for SnMe2 from other precursors,12 and which exhibits
similar reactivity toward stannylene scavengers. Photolysis of
4
b in solution, where a much wider range of monitoring
wavelengths are accessible, reveals a much richer transient
behavior and provides kinetic details of the initial steps in the
oligomerization of SnMe2, which is known from earlier work
Photolysis with a Zn resonance lamp (214 nm) of a deoxy-
genated 0.02 M solution of 4b in cyclohexane-d12 containing
9
,10
to be the main fate of the species under such conditions.
0
.5 M methanol (MeOH), with periodic monitoring of the
1
photolyzate by (600 MHz) H NMR spectroscopy, resulted in
the formation of 2,3-dimethyl-1,3-butadiene (DMB) and a
colorless precipitate that deposited on the walls of the NMR
tube. A collection of very weak peaks was also present in the
0
0
.5-0.8 ppm range of the spectrum (including a singlet at δ
.53 ppm), consistent with (SnMe2)n oligomers, but no new
1
0
signals were detectable in the δ 3.5-5.5 ppm region, where
those due to Sn-H and Sn-OMe protons would be expected.26
The conversion was ca. 12% after 2.5 h of photolysis and
produced an 80-90% yield of DMB relative to consumed
starting material. Virtually indistinguishable results were ob-
tained upon photolysis for the same period of time of 4b in
cyclohexane-d12 containing no added alcohol. It should be noted
that the yields of DMB produced in the two experiments were
identical within experimental error, indicating that the presence
of the alcohol in concentrations as high as 0.5 M has no effect
on the efficiency of the primary photoreaction of 4b.
(
14) Harrington, C. R.; Leigh, W. J.; Chan, B. K.; Gaspar, P. P.; Zhou, D. Can.
J. Chem. 2005, 83, 1324.
(
15) Becerra, R.; Walsh, R. Phys. Chem. Chem. Phys. 1999, 1, 5301.
16) Becerra, R.; Egorov, M. P.; Krylova, I. V.; Nefedov, O. M.; Walsh, R.
Chem. Phys. Lett. 2002, 351, 47.
(
(
(
(
(
17) Becerra, R.; Boganov, S. E.; Egorov, M. P.; Faustov, V. I.; Promyslov, V.
M.; Nefedov, O. M.; Walsh, R. Phys. Chem. Chem. Phys. 2002, 4, 5079.
18) Becerra, R.; Boganov, S. E.; Egorov, M. P.; Nefedov, O. M.; Walsh, R.
Chem. Phys. Lett. 1996, 260, 433.
19) Becerra, R.; Boganov, S. E.; Egorov, M. P.; Faustov, V. I.; Nefedov, O.
M.; Walsh, R. Phys. Chem. Chem. Phys. 2001, 3, 184.
20) Becerra, R.; Boganov, S. E.; Egorov, M. P.; Faustov, V. I.; Nefedov, O.
M.; Walsh, R. J. Am. Chem. Soc. 1998, 120, 12657.
(
21) Becerra, R.; Walsh, R. Phys. Chem. Chem. Phys. 2002, 4, 6001.
22) Becerra, R.; Boganov, S. E.; Egorov, M. P.; Lee, V. Ya.; Nefedov, O. M.;
Walsh, R. Chem. Phys. Lett. 1996, 250, 111.
(
(
23) Bobbitt, K. L.; Lei, D.; Maloney, V. M.; Parker, B. S.; Raible, J. M.; Gaspar,
P. P. In Frontiers of Organogermanium, -Tin and -Lead Chemistry;
Lukevics, E., Ignatovich, L., Eds.; Latvian Institute of Organic Synthesis:
Riga, Latvia, 1993; pp 41-53.
(25) Jiang, P.; Gaspar, P. P. J. Am. Chem. Soc. 2001, 123, 8622.
(26) Schumann, H.; Schumann, I. In Gmelin Handbook of Inorganic Chemistry;
Kr u¨ erke, U., Ed.; Springer-Verlag: Berlin, 1989; Part 17 (Sn; Organotin
Compounds), pp 78-79.
(
24) Steinmetz, M. G.; Yu, C. Organometallics 1992, 11, 2686.
17470 J. AM. CHEM. SOC.
9
VOL. 127, NO. 49, 2005

Direct Detection of SnMe
2
and Sn Me
2 4
A R T I C L E S
Figure 2. Plots of kdecay vs SO2 pressure, from 193-nm laser flash photolysis
of 4a (O) and SnMe4 (0)12 in the gas phase.
Figure 1. Transient absorption spectra obtained from 193-nm laser flash
photolysis of 4a (O, b) and SnMe4 (0) in the gas phase. The open symbols
are the transient absorbance (∆A) values obtained at the peak of the
excitation pulse. The solid points and curve represent the residual ∆A values
at the end of the time window monitored (50-100 µs).
Table 1. Absolute Rate Constants for Reaction of the Transient
Produced upon 193-nm Laser Flash Photolysis of 4a in the Gas
Phase at 298 Ka
k/109 M- s-
1
1
Laser Flash Photolysis of 4a in the Gas Phase at 193 nm.
For investigation of the transient spectrum, sample mixtures
were made up containing ca. 30 mTorr of 4a in 10 Torr of SF6.
Laser (193-nm) photolysis gives a transient species with strong
absorptions in the accessible monitoring range of 457.9-514.5
nm. The spectrum in this range has its strongest absorption at
substrate
this work
ref 12
SO2
HCl
MeOH
17.1 ( 0.9
0.40 ( 0.04
1.25 ( 0.07
2.32 ( 0.12
20.2 ( 0.8
0.49 ( 0.020
1.57 ( 0.04
4.72 ( 0.07
2-butyne
a
5
14.5 nm and matches those obtained previously from other
Rate constants converted from molecular units.
12
precursors to SnMe2 (e.g., SnMe4). The absorptions decay to
final steady, nonbaseline values. The intensities of these end
absorptions vary with wavelength, apparently peaking at ca. 480
nm. Whether the end absorptions are due to a stable final product
(i.e., stable for hours) or a much more slowly decaying transient
cannot be determined from these results, but the ratio of end
absorption to initial absorption (Ifinal/Iinitial) appears to be
independent of the initial transient concentration over a range
of a factor of 4. Figure 1 shows a comparison of transient spectra
1
2
obtained from 193-nm flash photolysis of 4a and SnMe4,
normalized at 514.5 nm (100%).
At λ ) 501.7 nm, transient decays were studied using
precursor pressures between 11 and 44 mTorr (in SF6, 10 Torr).
The decays fit acceptably to first-order kinetics over a minimum
of 70% consumption of the transient with essentially no variation
of the decay constant [(5.5 ( 1.5) × 10 s ] with precursor
pressure over this range. Even at the highest precursor pressure,
there was no evidence of a second-order component to the decay.
Kinetic studies were undertaken with a variety of substrates
Figure 3. Transient absorption spectra, recorded by 193-nm flash photolysis
of 4b in deoxygenated hexane solution at 25 °C, 0.12-0.220 (b), 1.8-2.0
(O), and 20-80 (0) µs after the laser pulse.
4
-1
observed with 2-butyne, but the reaction seemed to be signifi-
cantly slower than observed previously. The absolute values
for the rate constants obtained from these experiments, using
only the lower decay constant values, are collected in Table 1.
Laser Flash Photolysis of 4b in Solution. Solution-phase
laser flash photolysis experiments with 4b were carried out with
the pulses from an ArF excimer laser (193 nm, ca. 25 ns, ca.
(MeOH, HCl, SO2, and 2-butyne) using both 501.7 and 514.5
nm as monitoring wavelengths, which were chosen to minimize
the effects of end absorption. However, all of the substrates
seem to quench the initial photochemical yield of the transient
-
1
(SnMe2) partially. This enhances the distorting effect of end
50 mJ), using rapidly flowing (ca. 5 mL min ) deoxygenated
-
5
absorption, making kinetic analysis more difficult, and also has
the effect of limiting the maximum values obtainable for decay
constants. This might be because, at high substrate pressures,
the decay trace is complicated by formation of the other
absorbing species (viz, the end absorption), which affects the
kinetics. An example of data obtained with SO2 as the substrate
is shown in Figure 2, along with the corresponding data obtained
previously using SnMe4 as the precursor.12 As can be seen, the
data points obtained here are in reasonable agreement with those
obtained previously, although at higher partial pressures of SO2,
they diminish slightly from the earlier values. This was also
true for reactions with HCl and MeOH. The effect was not
solutions of 4b (ca. 3 × 10 M) in dry hexane. Transient
absorptions were monitored over the 240-650-nm wavelength
range through several time windows after the laser pulse. At
least three different transient species were observed to form and
decay over a time scale of ca. 40 ms, which was the longest
time scale monitored in our experiments. Figure 3 shows
transient absorption spectra recorded 120-220 ns, 1.8-2.0 µs,
and 20-80 µs after excitation, and Figure 4 shows representative
transient decay/growth profiles recorded at monitoring wave-
lengths of 530, 450, and 260 nm. As the two figures show, laser
photolysis causes the prompt formation of a species absorbing
with an apparent λmax of ∼500 nm, the bulk of which decays
J. AM. CHEM. SOC.
9
VOL. 127, NO. 49, 2005 17471

A R T I C L E S
Becerra et al.
Figure 4. Transient decay/growth profiles, recorded by 193-nm flash photolysis of 4b in deoxygenated hexane solution at 25 °C at monitoring wavelengths
of (a) 530 and 450 nm and (b) 260 nm. The inset in (b) shows a trace recorded at 260 nm over a full scale of 40 ms.
wavelength edges of the two absorption bands over longer time
scales show strong similarities to one another, which might be
interpreted as the result of equilibrium between the two species.
Second, the fact that the apparent growth of the (470-nm) signal
assigned to 5b is much more rapid than the decay of the signal
assigned to SnMe2 does not represent an inconsistency in its
assignment; the growth of the second-formed transient appears
to be much more rapid than it actually is because the onset of
the decay of this species occurs quite early after it is initially
formed. Finally, the lifetime measured for the 260-nm transient
absorptions (τ ≈10 ms) should be considered a very crude lower
limit, as the longest lifetime our system is capable of measuring
accurately in the configuration used for these experiments is
ca. 3 ms.
Figure 5. Transient absorption spectra, recorded by 193-nm flash photolysis
-
5
of a ca. 3 × 10 M solution of 4b in dry deoxygenated hexane containing
3
mM MeOH at 23 °C, recorded 0.2-0.3 and 1.0-1.2 µs after the laser
pulse. The weak absorption band centered at ca. 460 nm is real and is due
to Sn2Me4 (5b). The inset shows a typical transient decay trace, recorded
at a monitoring wavelength of 360 nm.
over ca. 10 µs. A decay trace recorded at 540 nm over a factor
of 10 longer time scale fits well to second-order kinetics (see
Supporting Information), yielding a second-order decay constant
7
-1
of k/ꢀ540 ) (1.4 ( 0.1) × 10 cm s , where ꢀ540 is the molar
Addition of MeOH (0.2-3.0 mM) to the solution resulted in
significant reductions in the intensities of the signals due to
SnMe2 and 5b, but had little effect on the decay or growth times
of either species. In the case of the SnMe2 absorption (monitored
at 530 nm), a ca. 4-fold drop in intensity was observed at the
highest concentration of alcohol employed; at none of the
concentrations examined could a fast initial decay of the signal
be resolved. Similarly, the absorptions due to 5b (monitored at
extinction coefficient at 540 nm. A second transient species (λmax
≈
470 nm) is observed to be formed after the laser pulse,
reaching a maximum in concentration ca. 1 µs after the pulse
and then decaying with complex kinetics over the next 20-30
µs (see Figure 4 and Supporting Information). All of this is
accompanied by the slower growth of a stronger transient signal
centered at 260 nm, which decays with complex kinetics and
an apparent lifetime of ca. 10 ms. The growth of these
absorptions occurs over a time scale similar to that of the decay
of the longer-wavelength transients, suggesting that they are
produced by reaction of the latter species. Furthermore, the
complexity of their temporal behavior suggests that they are
due to more than one long-lived species.
We assign the first-formed transient to SnMe2 (λmax ≈ 500
nm); the second-formed transient to its dimerization product,
tetramethyldistannene (5b; λmax ≈ 470 nm); and the more slowly
formed species observed in the 240-300-nm range to one or
more SnMe2 oligomers, formed by reaction of 5b (eq 3). Several
comments regarding the time evolution of the transient signals
due to these species are warranted. First, there is substantial
spectral overlap between the 500- and 470-nm species, which
makes quantitative interpretation of the transient decays very
difficult. Indeed, decays recorded even at the long- and short-
4
40 nm) suffered a ca. 3-fold drop in maximum intensity over
the 0.2-3.0 mM concentration range, but there was no change
in the growth time of the signal and only small changes in the
decay time. These reductions in signal strength are much larger
than can be explained by screening of the excitation light by
the alcohol, which has a molar extinction coefficient of ꢀ193 )
3
-1
-1
27
3
9 ( 1 dm mol cm at 193 nm in hydrocarbon solvents.
Single-exponential decay analysis of a transient decay profile
recorded at 440 nm in the presence of 3 mM MeOH afforded
a reasonable fit and yielded a pseudo-first-order decay rate
4
-1
constant of kdecay ) (8.4 ( 0.7) × 10 s . The signal at 260
nm developed a fast growth component whose contribution
increased at the expense of the slow component with increasing
methanol concentration, until at 3 mM MeOH, the latter was
(27) Kerst, C.; Byloos, M.; Leigh, W. J. Can. J. Chem. 1997, 75, 975.
17472 J. AM. CHEM. SOC.
9
VOL. 127, NO. 49, 2005

Direct Detection of SnMe
2
and Sn Me
2 4
A R T I C L E S
no longer apparent in the transient profile; although the overall
decay rate appeared to increase slightly, there were only small
reductions in the maximum strength of the absorption over the
were included in the calculations for the Sn-containing species.
Electronic spectral calculations were then carried out on the
optimized geometries of SnMe2, 5b, 7-9, SiMe2, and 11b using
3
0,31
0.2-3.0 mM range in MeOH concentration.
the SAOP exchange-correlation potential,
with the same
These changes were accompanied by the appearance of a new
triple-ú double-polarization basis sets and relativistic adjustments
as were employed for the geometry optimizations; the latter were
used in the calculations of the spectra of the silicon-containing
species as well. We previously showed this method to afford
excitation energies for 10a-c and the corresponding germylenes
that agree with experimental values to within 0.12 eV or less.1
The relevant computed structural parameters for each of the
molecules studied are listed in the Supporting Information.
Comparison of the calculated structure of 9 to the reported
transient absorption centered at 360 nm, which appeared to be
formed within the duration of the laser pulse at the lowest MeOH
concentration employed (0.2 mM). The maximum absorbance
of the species increased with increasing MeOH concentration,
and the signal decayed with clean second-order kinetics and a
3a
6
-1
decay constant of k/ꢀ360 ) (7.2 ( 0.3) × 10 cm s , which did
not vary significantly with MeOH concentration over the 1-3
mM range. Figure 5 shows a transient absorption spectrum
recorded 1-1.2 µs after the pulse in the presence of 3 mM
MeOH, along with a transient decay trace recorded at 360 nm.
We assign the transient to the Lewis acid-base complex of
SnMe2 and MeOH (6; eq 4).
28
X-ray structure of the compound (see Supporting Information)
revealed reasonable agreement between experiment and theory,
although not quite as good as we found earlier for the
1
3a
homologous germylene derivative.
The calculated Sn-C
distance and C-Sn-C angle in 9 are 2.25 Å and 85.7°,
respectively, which should be compared to 2.22 Å and 86.7° in
2
8
the experimental structure. The calculated structures of the
other MR2 and the M2R4 species (M ) Si, Ge, Sn; R ) H, Me,
Ph) show the expected changes with increasing atomic number
of the group 14 elements [i.e., increasing M-R distances and
decreasing R-M-R angles in both species, as well as increasing
M-M bond distances and pyramidalization angles in the (trans-
bent) dimetallenes]. The structures of the parent and methylated
compounds generally compare quite favorably with the results
DFT Calculations. Time-dependent density functional theory
TD-DFT) calculations were carried out to predict the electronic
(
spectra of SnMe2, tetramethyldistannene (5b), and the three-
and four-membered cyclic SnMe2 oligomers 7 and 8, as well
as that of the stable stannylene 9 in order to assess the reliability
of the method employed through comparison to the known X-ray
3
2
33-38
of previous DFT calculations for the MH2, M2H4,
28
crystal structure and UV absorption spectrum of the compound.
3
9,40
37,38
MMe2,
and M2Me4
species. They also agree reasonably
Spectral calculations for SiMe2 and Si2Me4 (11b) were also
carried out at the same level of theory, to allow comparisons
with experimental spectra for the complete series of dimethyl-
metallylenes and tetramethyldimetallenes. To address the ques-
tion of possible reversibility in the dimerization of SnMe2 (vide
supra), we also calculated bond dissociation energies for the
series of distannenes, digermenes, and disilenes 5a-c, 10a-c,
and 11a-c, respectively (eq 5). Frequency calculations were
carried out on the H- and methyl-substituted derivatives, to
obtain the zero-point energy corrections, internal energies, and
entropies.
well with ab initio results, particularly for the Ge- and
Sn-containing species.
4
1-52
Geometry optimizations for the dimetallenes were carried out
under Ci symmetry in all cases; however, 5b was also optimized
without symmetry constraints after rotating the SndSn bond in
the Ci-symmetric (trans-bent) structure to a non-zero value. This
afforded a nonsymmetric trans-bent structure with a twist angle
of 11.78° about the SndSn bond, but with otherwise almost
identical structural features to the Ci-symmetric conformer. The
(
30) Gritsenko, O. V.; Schipper, P. R. T.; Baerends, E. J. Chem. Phys. Lett.
1999, 302, 199.
(
31) Schipper, P. R. T.; Gritsenko, O. V.; van Gisbergen, S. J. A.; Baerends, E.
J. J. Chem. Phys. 2004, 112, 1344.
(
(
(
(
(
32) Su, M.-D. J. Phys. Chem. A 2002, 106, 9563.
33) Jacobsen, H.; Ziegler, T. J. Am. Chem. Soc. 1994, 116, 3667.
34) Kapp, J.; Remko, M.; Schleyer, P. v. R. Inorg. Chem. 1997, 36, 4241.
35) Chen, W.-C.; Su, M.-D.; Chu, S.-Y. Organometallics 2001, 20, 564.
36) Mosey, N. J.; Baines, K. M.; Woo, T. K. J. Am. Chem. Soc. 2002, 124,
13306.
(37) Malcolm, N. O. J.; Gillespie, R. J.; Popelier, P. L. A. J. Chem. Soc., Dalton
Trans. 2002, 3333.
38) Su, M.-D. J. Phys. Chem. A 2004, 108, 823.
39) Su, M.-D.; Chu, S.-Y. J. Phys. Chem. A 1999, 103, 11011.
40) Su, M.-D. Chem. Eur. J. 2004, 10, 6073.
41) Fjeldberg, T.; Haaland, A.; Schilling, B. E. R.; Lappert, M. F.; Thorne, A.
J. J. Chem. Soc., Dalton Trans. 1986, 1551.
(
(
(
(
(
42) Goldberg, D. E.; Hitchcock, P. B.; Lappert, M. F.; Thomas, K. M.; Thorne,
A. J.; Fjeldberg, T.; Haaland, A.; Schilling, B. E. R. J. Chem. Soc., Dalton
Trans. 1986, 2387.
Geometry optimizations employed the exchange and correla-
29
tion functionals of Perdew and Wang (PW91) and uncontracted
Slater-type orbitals of triple-ú double-polarization (TZ2P) quality
as basis functions, as we used previously for calculation of the
structures and electronic spectra of the corresponding ger-
(
(
43) Liang, C.; Allen, L. C. J. Am. Chem. Soc. 1990, 112, 1039.
44) Grev, R. S.; Schaefer, H. F., III; Baines, K. M. J. Am. Chem. Soc. 1990,
112, 9458.
45) Trinquier, G. J. Am. Chem. Soc. 1990, 112, 2130.
46) Windus, T. L.; Gordon, M. S. J. Am. Chem. Soc. 1992, 114, 9559.
(
(
mylenes (GeH2, GeMe2, and GePh2), the germanium analogue
of stannylene 9, and digermenes 10a-c.13a Relativistic effects
(47) Karni, M.; Apeloig, Y. J. Am. Chem. Soc. 1990, 112, 8589.
(48) Boatz, J. A.; Gordon, M. S.; Sita, L. R. J. Phys. Chem. 1990, 94, 5488.
(
49) Heaven, M. W.; Metha, G. F.; Buntine, M. A. J. Phys. Chem. A 2001,
105, 1185.
(
28) Kira, M.; Yauchibara, R.; Hirano, R.; Kabuto, C.; Sakurai, H. J. Am. Chem.
Soc. 1991, 113, 7785.
(50) Hajgato, B.; Takahashi, M.; Kira, M.; Veszpremi, T. Chem. Eur. J. 2002,
8, 2126.
(51) Szabados, A.; Hargittai, M. J. Phys. Chem. A 2003, 107, 4314.
(52) Becerra, R.; Cannady, J. P.; Walsh, R. J. Phys. Chem. A 2004, 108, 3987.
(29) Perdew, J. P.; Chevary, J. A.; Vosko, S. H.; Jackson, K. A.; Pederson, M.
R.; Singh, D. J.; Fiolhais, C. Phys. ReV. B 1992, 46, 6671.
J. AM. CHEM. SOC.
9
VOL. 127, NO. 49, 2005 17473

A R T I C L E S
Becerra et al.
Table 3. Calculated Total Bonding Energy, Enthalpy, and Free
Table 2. Calculated and Experimental UV-Vis Absorption
Maxima for Stannylene 9; SnMe2; and SnMe2 Oligomers 5b, 7,
and 8a
Energy Changes for Dissociation of Dimetallenes R2MdMR2 to the
Corresponding Metallylenesa
λmax (nm) [E (eV)]
M
R
)
H
R
)
Me
R ) Ph
∆E (kcal mol-
1 b
)
compound
calcd
exptl
Si
Ge
Sn
70.4
53.4
33.9
60.2
43.7
29.0
51.1
35.0
24.1
9
505 [2.46], 346 [3.58], 484 [2.56], 370 [3.35],
81 [4.49], 271 [4.57]
502 [2.47], 239 [5.17]
2
285 [4.35], 247 [5.02]b
500 [2.50]
470 [2.64]
SnMe2
∆H° (kcal mol-
1 c
)
Me2SndSnMe2 (5b) Ci: 445 [2.79]
unsym: 447 [2.78]
298 [4.16]
308 (sh) [4.02], 249
Si
Ge
Sn
66.3
49.6
30.9
56.0
39.9
25.8
-
-
-
cyclo-(SnMe2)3 (7)
cyclo-(SnMe2)4 (8)
[
4.97], 239 [5.19]
S° (cal mol-1
-1 c
∆
K
)
39.2
37.6
32.6
Si
Ge
Sn
34.6
34.9
31.5
-
-
-
a
sh ) shoulder. b Reference 28.
total bonding energies in the two structures differed by only
-1 c
∆
G° (kcal mol
)
-
1
0
.2 kcal mol , in keeping with a rather flat potential energy
Si
Ge
Sn
56.0
39.2
21.5
44.3
28.7
16.1
-
-
-
33
surface for twisting about the SndSn bond. Electronic spectral
calculations were also carried out for this conformer and
revealed no significant differences in either the position or the
intensity of the longest-wavelength absorption band compared
to those for the Ci-symmetric structure. The calculated Sn-Sn
bond distance in 5b matches almost precisely the experimental
value (rSn-Sn ) 2.768 Å) reported for the stable tetraalkyldis-
tannene Sn2(CH(SiMe3)2)4.42
a
Values for M ) Sn include relativistic corrections. b At 0 K; not
corrected for zero-point energy contributions. c Corrected for zero-point
energy contributions.
44-360 nm,65,66 respectively). In the case of 9, although the
calculated oscillator strength for the first transition [Sn(n,p); f
0.004] is comparable to the experimental value of 0.006, the
intensity of the second band is significantly overestimated. This
transition corresponds to promotion of a Si-C(σ) bonding
electron to the Sn(p) LUMO. This same discrepancy was evident
in our calculations of the spectra of the germanium analogue
of this molecule and was essentially independent of the size
of the basis set and the potential model used for the calculation
of excitations.
Table 3 summarizes the results of the bond dissociation
energy (BDE) calculations for the three series of hydrido-,
methyl-, and phenyl-substituted dimetallenes, in the form of total
energies (at 0 K) for all nine compounds and zpe-corrected
standard enthalpies and free energies for the hydrido- and
methyl-substituted derivatives (at 298 K). The entropy changes
required for calculation of the latter quantities are also included
in the table. The smallest values of ∆S° were obtained for the
distannenes, consistent with the weaker MdM bonds and lower
energies for the rocking modes that disappear upon dissociation,
compared to the disilenes and digermenes. To help assess the
accuracy of the calculation for 5b, we also calculated the Sn-
Sn BDE in hexamethyldistannane. The value obtained, ∆E )
3
)
The calculated Sn-Sn bond distances in 7 and 8 are within
the ranges reported experimentally for those stable cyclotristan-
53,54
55-58
nanes
and cyclotetrastannanes
that have been character-
ized crystallographically, and are close to the values calculated
for the parent compounds by ab initio methods.59 The calculated
structure of 8 is puckered with a fold angle of 39.6°, a larger
deviation from planarity than those in reported experimental
13a
structures, in which the fold angles range from 0° to ca. 30°
depending on substituent.5
5-58
Previous ab initio calculations
on cyclotetrastannane (cyclo-Sn4H8) predicted the molecule to
5
9
be planar.
The results of the electronic spectral calculations for the
stannylenes and (SnMe2)n oligomers are summarized in Table
2
along with the corresponding experimental values of Kira and
28
co-workers for 9 and those of the present work for SnMe2
and Sn2Me4 (5b). There is excellent agreement between the
predicted and experimental spectra of the stannylenes in both
cases, as well as with previous calculations of the absorption
1
2
maximum of SnMe2 using ab initio and TD-DFT methods.
Similarly, the predicted values of the lowest-energy absorption
bands in the UV-vis spectra of SiMe2 [λmax ) 457 nm (2.71
eV)] and tetramethyldisilene [11b; λmax ) 365 nm (3.39 eV)]
-1
-1
6
8.8 kcal mol , corresponds to ∆H° ≈ 66 kcal mol , assuming
reasonable values for the internal energy and zpe corrections.
This can be compared to the best current experimental estimate
(
see Supporting Information) fall within or just outside the
6
0-64
ranges in experimental values (λmax ) 453-470 nm
and
-1 67
of ∆H° ) 68.5 ( 2.0 kcal mol .
(
(
(
(
(
(
53) Cardin, C. J.; Cardin, D. J.; Constantine, S. P.; Drew, M. G. B.; Rashid,
Discussion
H.; Convery, M. A.; Fenske, D. J. Chem. Soc., Dalton Trans. 1998, 2749.
54) Masamune, S.; Sita, L. R.; Williams, D. J. J. Am. Chem. Soc. 1983, 105,
Most aspects of the photochemistry of 4a and 4b, including
the transient spectroscopic behavior, bear strong qualitative
similarities to those observed previously for the germacyclo-
6
30.
55) Lappert, M. F.; Leung, W. P.; Raston, C. L.; Thorne, A. J.; Skelton, B.
W.; White, A. H. J. Organomet. Chem. 1982, 233, C28.
56) Puff, H.; Bach, C.; Reuter, H.; Schuh, W. J. Organomet. Chem. 1984, 277,
1
7.
57) Puff, H.; Bach, C.; Schuh, W.; Zimmer, R. J. Organomet. Chem. 1986,
(62) Baggott, J. E.; Blitz, M. A.; Frey, H. M.; Walsh, R. J. Am. Chem. Soc.
1990, 112, 8337.
3
12, 313.
58) Cardin, C. J.; Cardin, D. J.; Convery, M. A.; Devereux, M. M.; Kelly, N.
B. J. Organomet. Chem. 1991, 414, C9.
(63) Baggott, J. E.; Blitz, M. A.; Frey, H. M.; Lightfoot, P. D.; Walsh, R. J.
Chem. Soc., Faraday Trans. 2 1988, 84, 515.
(
59) Rubio, J.; Illas, F. J. Mol. Struct. (THEOCHEM) 1984, 110, 131.
60) Michalczyk, M. J.; Fink, M. J.; De Young, D. J.; Carlson, C. W.; Welsh,
K. M.; West, R.; Michl, J. Silicon, Germanium, Tin Lead Compds. 1986,
(64) Levin, G.; Das, P. K.; Bilgrien, C.; Lee, C. L. Organometallics 1989, 8,
1206.
(
(65) Yamaji, M.; Hamanishi, K.; Takahashi, T.; Shizuka, H. J. Photochem.
Photobiol. A: Chem. 1994, 81, 1.
9
, 75.
(61) Baggott, J. E.; Blitz, M. A.; Frey, H. M.; Lightfoot, P. D.; Walsh, R. Int.
(66) Sekiguchi, A.; Maruki, I.; Ebata, K.; Kabuto, C.; Sakurai, H. J. Chem.
Soc., Chem. Commun. 1991, 341.
J. Chem. Kinet. 1992, 24, 127.
17474 J. AM. CHEM. SOC.
9
VOL. 127, NO. 49, 2005

Direct Detection of SnMe
2
and Sn Me
2 4
A R T I C L E S
pentene derivatives 1-313a,15-23 and are consistent with formal
cheletropic cycloreversion to yield SnMe2 and the corresponding
conjugated diene as the primary photochemical reaction of these
molecules. In the gas phase, flash photolysis of 4a affords a
transient product with UV-vis spectrum and reactivity similar
to those assigned previously to SnMe2 in experiments with other
photochemical precursors to the molecule.12 Likewise, laser
photolysis of 4b in solution leads to behavior that is strikingly
similar to that observed from the related germanium compounds
corresponding dimetallenes (5b, 10b, 11b) predict a similar trend
in absorption maxima for these compounds, with predicted
values of λmax ) 445, 374, and 365 nm for the distannene,
digermene, and disilene, respectively. The agreement between
the predicted and experimental solution-phase spectra of
1
3a,73,74
Ge2Me4 (10b; λmax ) 370 nm)
and Si2Me4 (11b; λmax )
65
360 nm) is again outstanding and affords a considerable degree
of confidence in the ability of the theoretical method to predict
the position of the lowest-energy absorption band in Sn2Me4
(5b) with reasonable accuracy. We can thus be fairly certain
that the spectrum that grows in over the first ca. 1 µs after pulsed
laser photolysis of 4b is due to the distannene (λmax ≈ 470 nm),
shifted to the blue of the stannylene absorption band but
overlapped partially with it.
1
a and 3a,b, which was attributed to the initial formation of
the corresponding germylene derivatives (GeMe2, GePh2, and
GeMes2, respectively) followed by the products of dimerization
(
the corresponding digermenes) and then their subsequent
13a
oligomerization products. In all four cases (viz, 1a, 3a,b, and
b), this process is characterized by the sequential formation
4
Interestingly, the temporal behaviors of the transient signals
of species that absorb at successively shorter wavelengths than
the initially formed transient and exhibit successively longer
lifetimes, as would be expected. We thus assign the initially
formed species from laser photolysis of 4b to SnMe2 (λmax ≈
observed at the long- and short-wavelength edges (e.g., 530-
5
40 and 430-450 nm, respectively) of the broad transient
absorptions observed upon flash photolysis of 4b in solution
differ considerably from one another only in the first ca. 2 µs
after the laser pulse and then follow qualitatively similar time
dependences thereafter. These similarities presumably result
simply from the substantial overlap in the spectra of the two
species. Nevertheless, we considered in some detail the alterna-
tive possibility, that this behavior results from reVersible
interconversion of the stannylene and distannene over the time
scale of their eventual removal from the system via oligomer-
ization.
5
00 nm); the second one to its dimerization product, tetra-
methyldistannene (5b; λmax ≈ 470 nm); and the long-lived
species absorbing below 300 nm to higher SnMe2 oligomers
(eq 3).
The absorption spectrum of the stannylene in solution is
similar to that measured in the gas-phase experiments, although
because of instrumental limitations, we have been able to detect
only the short-wavelength rising edge of the gas-phase absorp-
tion band. Qualitative comparisons of the (1-2 µs) time scale
over which the initial decay of the stannylene and the growth
of its primary product occur to that observed in previous
experiments with GePh2 (for which a rate constant for dimer-
The possibility seems a reasonable one, as most of the stable
acyclic distannenes that are currently known are dissociative
42,75-80
in solution.
These compounds all contain sterically bulky
substituents, and most are tetraaryl-substituted; the individual
1
0
-1 -1
13a
ization of kdim ) 1.1 × 10
M
s
has been measured )
effects of steric and electronic factors on the SndSn bond
suggest that the dimerization of SnMe2 in solution proceeds at
a similar rate. This is consistent with the second-order decay
81
strengths in these compounds are not clear. We thus decided
to expand the scope of our DFT calculations on SnMe2 and 5b
to examine the predicted variations in bond dissociation energy
of the SndSn bond as a function of substituent, by bracketing
the value for the tetramethyl system with those for the parent
distannene (for which previous calculations have been re-
7
constant measured at 540 nm of k/ꢀ540 ) (1.4 ( 0.1) × 10 cm
-1
s ; if we assume that the extinction coefficient of the stannylene
3
-1
at its absorption maximum is on the order of 1000 dm mol
-
1
68,69
), a value of k ≈ 7 × 10 M s-1
9
-1
cm (cf. that for SiMe2
is obtained for the absolute rate constant for dimerization. To
our knowledge, the product of this reaction, tetramethyldistan-
nene (5b), has not been detected directly before under any set
of conditions.
33,34,44
ported
) and tetraphenyldistannene (5c), as a model for the
known tetraaryldistannenes that are dissociative in solution. To
provide a benchmark for the calculations on the tin systems,
we also carried out similar calculations for the corresponding
digermenes (10) and disilenes (11) (Table 3) and hexameth-
yldistannane, all using the same theoretical method.
Comparison of the experimental spectra of SnMe2 (λmax )
22,70
5
4
00 nm), GeMe2 (λmax ) 480 nm),
and SiMe2 (λmax ) 453-
65 nm)6
0,64
reveals a consistent trend toward decreasing
The trend of decreasing BDE with increasing atomic number
of the group 14 element, which is observed in the present
excitation energy with increasing size of the group 14 element,
28
similar to that exhibited by 9 (λmax ) 484 nm ) and its Ge and
Si analogues (λmax ) 450 nm71 and 440 nm, respectively).
The trend is reproduced reasonably faithfully by the TD-DFT
calculations, which predict lowest-energy absorption maxima
of 502, 463, and 457 nm for SnMe2, GeMe2,1 and SiMe2,
respectively. Calculations using the same method for the
72
(73) Mochida, K.; Kayamori, T.; Wakasa, M.; Hayashi, H.; Egorov, M. P.
Organometallics 2000, 19, 3379.
(
74) Taraban, M. B.; Volkova, O. S.; Plyusnin, V. F.; Ivanov, Y. V.; Leshina,
T. V.; Egorov, M. P.; Nefedov, O. M.; Kayamori, T.; Mochida, K. J.
Organomet. Chem. 2000, 601, 324.
3a
(
75) Goldberg, D. E.; Harris, D. H.; Lappert, M. F.; Thomas, K. M. J. Chem.
Soc., Chem. Commun. 1976, 261.
(
76) Klinkhammer, K. W.; Schwarz, W. Angew. Chem., Int. Ed. Engl. 1995,
34, 1334.
(
67) Martinho Simoes, J. A.; Liebman, J. F.; Slayden, S. W. In The Chemistry
of Organic, Germanium, Tin and Lead Compounds; Patai, S., Ed.; John
Wiley & Sons Ltd.: New York, 1995; Vol. 1, pp 245-266.
(77) Weidenbruch, M.; Kilian, H.; Peters, K.; von Schnering, H. G.; Marsmann,
H. Chem. Ber. 1995, 128, 983.
(78) Klinkhammer, K. W.; Fassler, T. F.; Grutzmacher, H. Angew. Chem., Int.
Ed. 1998, 37, 124.
(
68) Drahnak, T. J.; Michl, J.; West, R. J. Am. Chem. Soc. 1979, 101, 5427.
69) Raabe, G.; Vancik, H.; West, R.; Michl, J. J. Am. Chem. Soc. 1986, 108,
(
(79) Della Bona, M. A.; Cassani, M. C.; Keates, J. M.; Lawless, G. A.; Lappert,
M. F.; Sturmann, M.; Weidenbruch, M. J. Chem. Soc., Dalton Trans. 1998,
1187.
(80) Sturmann, M.; Saak, W.; Klinkhammer, K. W.; Weidenbruch, M. Z. Anorg.
Allg. Chem. 1999, 625, 1955.
(81) (a) Power, P. P. J. Chem. Soc., Dalton Trans. 1998, 2939. (b) Power, P. P.
Chem. ReV. 1999, 99, 3463.
6
71.
(
70) Mochida, K.; Tokura, S. Bull. Chem. Soc. Jpn. 1992, 65, 1642.
71) Kira, M.; Ishida, S.; Iwamoto, T.; Ichinohe, M.; Kabuto, C.; Ignatovich,
L.; Sakurai, H. Chem. Lett. 1999, 1999, 263.
(
(72) Kira, M.; Ishida, S.; Iwamoto, T.; Kabuto, C. J. Am. Chem. Soc. 1999,
1
21, 9722.
J. AM. CHEM. SOC.
9
VOL. 127, NO. 49, 2005 17475

A R T I C L E S
Becerra et al.
1
calculations for all three substituents, is well established from
earlier ab initio and DFT calculations for the parent dimetal-
leads to estimates of ca. 36, 20, and 11 kcal mol- for the
expected ∆G° values for dissociation of the three derivatives,
assuming reasonable values for the entropy changes. The
3
3,34,44
lenes.
Our calculated ∆E values (0 K, zpe-corrected) for
-
1
-1
disilene (11a; 65.6 kcal mol ), digermene (10a; 49.1 kcal
estimated ∆G° for 5c can be compared to the 0-2 kcal mol
value that would be required for the reported “dissociative”
tetraaryldistannenes. The reason these distannenes are dis-
-1
-1
mol ), and distannene (5a; 30.7 kcal mol ) are all higher than
values calculated previously with other methods,3
3,34,44
but the
75
present value for 11a comes closest to the experimental value
sociative, while 5b is not, must be ascribed to steric factors,
which appear to contribute as much as 8-10 kcal mol to the
dissociation energies of these highly hindered derivatives.
-
1
82
-1
of 63.3 kcal mol , reported by Ruscic and Berkowitz. The
agreement with experimental data is also excellent in the case
of the BDE of hexamethyldistannane (vide supra).
The question naturally arises from the solution studies as to
whether distannene 5b might also be formed in detectable
amounts in the gas phase. We believe that it is, based on the
rather close match between the gas-phase end absorptions and
the spectrum assigned to 5b in solution (Figure 1), as well as
the kinetic arguments presented in the following discussion.
Similar long-lived absorptions, as well as the formation of solid
precipitates (so-called “dust”), were evident in the experiments
with the four other SnMe2 precursors that were studied
The calculations for the methyl- and phenyl-substituted
systems reveal a successive reduction in BDE throughout the
three series of dimetallenes as the hydrogens in the parent
compounds are replaced with methyl and phenyl substituents.
This can be interpreted in terms of a substituent effect on the
stabilities of the corresponding metallylenes. What is known
of the differences in the reactivities of the parent metallylenes
compared to those of the dimethyl analogues for the silicon and
6
1-63,83
germanium analogues supports this view; both SiMe2
12
previously.
16,22
and GeMe2
are substantially less reactive than the corre-
The gas-phase decays of SnMe2 were quite noisy, but they
fit reasonably well to single exponentials and showed no obvious
second-order kinetic component; the decay constants (kdec)
sponding parent molecules in the gas phase.8
4-86
Recent results
for GePh2 in solution indicate that it exhibits similar or slightly
lower reactivity than GeMe2 in all cases that have been studied
so far, although the comparisons that have been made were
necessarily confined to gas-phase rate constants for the simpler
germylene, because of a lack of reliable solution-phase values
for reactions with the same or similar substrates.1
4
-1
varied over the range (4-7) × 10 s , but did not vary
systematically with variations in the initial transient concentra-
tion over a factor of 4, as would be expected if dimerization
contributed significantly to the decay kinetics. We thus conclude
that a pseudo-first-order decay process is mainly responsible
for the decay of the stannylene in the gas-phase experiments.
This process can most likely be identified as surface reactions
with the solid particulates that cannot be removed by pumping
the reaction vessel, as the lack of dependence of the decay rate
constants on precursor pressure in the absence of added
scavengers rules out pseudo-first-order reaction with the precur-
sor (4a).
3b
Table 3 also contains the calculated values of ∆G° for
dissociation of the parent dimetallenes and tetramethyldimet-
-
1
allenes. The value of ∆G° )16.1 kcal mol for 5b is roughly
-1
1
0 kcal mol higher than the value required by the experimental
data if dimerization of SnMe2 is reversible. The latter was
estimated assuming a concentration of ca. 10 µM for the amount
of SnMe2 produced by the laser pulse (which is consistent with
the typical transient absorbances observed in our solution-phase
Nevertheless, the kinetic analyses can accommodate a
contribution of 5-15% from dimerization, which corresponds
roughly to the yield of 5b that is indicated by the relative
intensities of the initial and end absorptions in our experiments,
given that the calculations (see Supporting Information) predict
the extinction coefficient of 5b to be ca. 10 times greater than
that of SnMe2 at the respective absorption maxima. Similar
differences are predicted for germylenes and their digermene
dimers and have been corroborated experimentally in the case
3
-1
experiments and an extinction coefficient of ca. 1000 dm mol
-
1
cm ), with the assumption that roughly half of it remains after
the conversion to 5b comes to equilibrium. This leads to an
-
5
estimate of Keq ≈ 10 M for the equilibrium constant for
dissociation of the distannene and a value of ∆G° ≈ 5 kcal
-
1
mol , assuming that dimerization was fully reversible under
the conditions of our experiments. The comparison ignores
solvation effects, but such effects can be expected to be small
considering that, in hexane solution, the only solvation interac-
tion possible is that due to dispersion forces. Despite the
uncertainties in these estimates, the calculations would appear
to rule out reversibility.
1
3a
of the GeMes2/Ge2Mes4 system. The rate constant required
9
to give this amount of dimer is estimated to be krec ≈ 8 × 10
-
1
-1
M
s
(see Supporting Information), which is in good
9
-1 -1
agreement with the solution value (∼7 × 10 M s ) obtained
in this work and is about a factor of 20 less than the collision
rate. We have also verified by RRKM calculations that SnMe2
recombination in the gas phase should not be significantly
pressure-dependent under our experimental conditions (see
Supporting Information). We note, however, that the end
absorptions were also present in decays obtained in the presence
of stannylene scavengers, although we did not examine them
near their absorption maxima.
The zero-point energy contributions to the dissociation
-
1
energies are in the ranges of 3.2-4.8 kcal mol for the parent
-
1
dimetallenes and 2.5-4.1 kcal mol
for the tetramethyl
derivatives, with the largest values being obtained for the
disilenes and the smallest for the distannenes. Applying similar
corrections to the calculated ∆E values for the phenylated
-
1
systems affords estimates of ca. 48, 32, and 21 kcal mol for
the BDEs of 11c, 10c, and 5c, respectively, at 298 K. This then
(
82) Ruscic, B.; Berkowitz, J. J. Chem. Phys. 1991, 95, 2416.
83) Baggott, J. E.; Blitz, M. A.; Frey, H. M.; Lightfoot, P. D.; Walsh, R. Chem.
Phys. Lett. 1987, 135, 39.
Turning briefly to the products formed in solution over the
longer time scale, it is clear from the form of the growth/decay
profiles that their formation is linked to the decay of SnMe2
and 5b, and so they can be assigned to higher SnMe2 oligomers
with some degree of confidence. The decay kinetics of the
(
(84) Jasinski, J. M.; Becerra, R.; Walsh, R. Chem. ReV. 1995, 95, 1203.
(85) Becerra, R.; Walsh, R. Res. Chem. Kinet. 1995, 3, 263.
(86) Boganov, S. E.; Egorov, M. P.; Faustov, V. I.; Krylova, I. V.; Nefedov, O.
M.; Becerra, R.; Walsh, R. Russ. Chem. Bull., Int. Ed. 2005, 54, 483.
17476 J. AM. CHEM. SOC.
9
VOL. 127, NO. 49, 2005

Direct Detection of SnMe
2
and Sn Me
2 4
A R T I C L E S
signals recorded in the 240-300-nm monitoring range are
complex, and we tentatively interpret them as consisting of at
least two components, at least one of which exhibits a lifetime
in excess of 10 ms. The calculated UV absorption spectra of
the three- and four-membered cyclic oligomers of SnMe2 (7
and 8, respectively) indicate that these two compounds could
well contribute to the transient absorptions observed in this
spectral range. Unfortunately, it is not possible to be any more
specific than that in regard to the identities of these species;
the observed absorption maximum (λmax ) 265 nm, E ) 4.68
eV) is closest to that calculated for cyclotetrastannane 8, but
the level of agreement is rather poor. No UV spectral data for
compounds of these types have been reported, to our knowledge.
The remaining set of data to be discussed is that obtained
from flash photolysis of 4b in hexane containing methanol, one
Figure 6. Plot of the ratio of initial transient absorbances due to SnMe2
(
monitored at 530 nm) in the absence and presence of MeOH (∆A530,0 and
∆A530,Q, respectively) vs [MeOH]. The solid line represents the least-squares
fit of the data to eq 6.
of the few reactive scavengers that can be studied in solution
with 193-nm laser excitation.2
7,87-89
The gas-phase results
consistent with the effects noted above is fast, reVersible reaction
of SnMe2 with the alcohol to yield complex 6. We thus interpret
the stannylene transient absorptions measured in the presence
of MeOH as being due to residual free stannylene present in
rapid equilibrium with the complex. Under such conditions, the
signal intensities due to residual free stannylene are related to
the equilibrium constant for the reaction according to eq 6, where
indicate that SnMe2 reacts with MeOH with a rate constant of
9
-1 -1
ca. 1.2 × 10 M
s
(see Table 1 and ref 12). Interestingly,
we were unable to detect any acceleration of the decay of SnMe2
in solution in the presence of the alcohol; the only effects noted
were a marked reduction in the intensities of the signals due to
the stannylene and distannene over the 0.2-3 mM range in
alcohol concentration, with the effect on the stannylene signal
being the greater. Furthermore, there was no change in the
growth time of the distannene signal over this concentration
range either. Despite this, the product of the reaction can be
readily detected in the form of a new, long-lived transient
absorbing with λmax ≈ 360 nm, which decays with second-order
kinetics (Figure 3) and is assigned to the Lewis acid-base
complex of the stannylene with the alcohol (6; eq 3). The recent
theoretical calculations of Su predict this species to be the
primary product in the reaction of SnMe2 with MeOH, and
∆A0 and ∆AQ are the initial transient absorbances in the absence
and presence of the alcohol at concentration [Q], respectively;
Keq is the equilibrium constant for reaction; and the excitation
laser intensity is roughly constant throughout the experiment.
A plot of the data according to eq 6 is shown in Figure 6; the
-
1
slope is the equilibrium constant, Keq ) 1060 ( 170 M . This
-
1
corresponds to a free energy difference of ca. 2 kcal mol in
favor of the complex in hexane solution at 298 K, in reasonable
40
agreement with the recent calculations of Su, with appropriate
assumptions as to entropic factors. It should be noted that, with
-
1
predict an activation energy of at least 15 kcal mol for its
-
1
an equilibrium constant on the order of 1000 M , the forward
further reaction to yield the formal O-H insertion product,
methoxydimethylstannane.40 The magnitude of this barrier could
rate constant for the process would need to be less than ca. 5 ×
9
-1 -1
1
0 M
s
in order for us to be able to distinguish an initial
explain why the methoxystannane is in fact not formed in
detectable amounts in steady-state trapping experiments (vide
supra); the ultimate fate of SnMe2 in the latter experiments,
even in the presence of relatively high concentrations of alcohol,
is oligomerization. Few examples of stable alkoxydialkylhy-
fast decay (due to the approach to equilibrium) from noise under
the conditions of our experiments. This limit is defined mainly
by the strength of the transient signals obtained, which was
rather low in the present study, but is somewhat higher than
9
-1 -1
dridostannanes are known,2
6,90,91
the absolute rate constant (k ≈ 10 M s ) determined earlier
for reaction of SnMe2 with MeOH in the gas phase. As the
latter was determined at relatively low pressures and pressure
dependence has not been studied, it might well be significantly
lower than the value in solution.
but water and MeOH are
known to undergo oxidative addition to Lappert’s stannylene
{
Sn[CH(SiMe3)2]2} to afford the corresponding (stable) O-H
91
insertion products. The reaction was proposed to proceed via
initial formation of the corresponding stannylene-alcohol
(
water) complex, but if this complex can rearrange readily to
∆
A /∆A ) 1 + K [Q]
(6)
0
Q
eq
product, then the required H migration must either have a lower
barrier than that calculated by Su for the SnMe2-MeOH
complex (6)40 or proceed catalytically.
The SnMe2-MeOH complex decays with clean second-order
6
kinetics and a rate coefficient of k/ꢀ360 ) (7.2 ( 0.3) × 10 cm
Because screening of the excitation light and excited-state
quenching effects can both be discounted as possible trivial
reasons for the drop in the signal strengths upon addition of
the alcohol, the only reasonable remaining explanation that is
-
1
s
in the presence of 3 mM alcohol, similar to the value
determined from the decay of the stannylene in the absence of
added MeOH. The extinction coefficients of the complex (Figure
) and the free stannylene (Figure 3) at their respective
absorption maxima appear to be quite similar to one another,
so it can be concluded that the presence of the alcohol at low
concentrations has little effect on the overall dimerization rate.
In the limit where dimerization proceeds only by combination
of free SnMe2 molecules, the overall rate constant is expected
to be reduced from its value in the absence of the alcohol by
5
(
(
(
87) Kerst, C.; Boukherroub, R.; Leigh, W. J. J. Photochem. Photobiol. A: Chem.
997, 110, 243.
88) Leigh, W. J.; Boukherroub, R.; Kerst, C. J. Am. Chem. Soc. 1998, 120,
1
9
504.
89) Leigh, W. J.; Kerst, C.; Boukherroub, R.; Morkin, T. L.; Jenkins, S.; Sung,
K.; Tidwell, T. T. J. Am. Chem. Soc. 1999, 121, 4744.
(
(
90) Hayashi, K.; Iyoda, J.; Shiihara, I. J. Organomet. Chem. 1967, 10, 81.
91) Schager, F.; Goddard, R.; Seevogel, K.; Porschke, K. R. Organometallics
1
998, 17, 1546.
J. AM. CHEM. SOC.
9
VOL. 127, NO. 49, 2005 17477

A R T I C L E S
Becerra et al.
the factor Keq[MeOH], or a factor of roughly 3 in the presence
of 3 mM MeOH. The observed reductions in the strength of
the signal due to 5b in the presence of the alcohol are consistent
with the idea that formation of the distannene does not involve
dimerization of the complex, but proceeds via reaction of free
stannylene in solution, either with itself or with the complex
with which it is in equilibrium.
to the Lewis acid-base complex Me2SnO(H)Me, in equilibrium
with free SnMe2 and MeOH. The equilibrium constant (Keq ≈
-1
1000 M ) is too low to allow measurement of the rate constant
for the association process, but the data allow a lower limit of
9
-1 -1
k > 5 × 10 M
s
to be established. This is consistent with
9
-1 -1
the value of k ≈ 1 × 10 M
s
measured in the gas phase at
low pressure. The complex appears to decay by redissociation
followed by SnMe2 dimerization.
In contrast to the substantial reactivity exhibited by SnMe2
with MeOH, our results indicate that tetramethyldistannene (5b)
is relatively unreactive toward the alcohol; from the pseudo-
first-order decay rate constant measured in the presence of 3
Sn Me has not previously been observed directly under any
2
4
set of conditions. SnMe has been detected previously only in
2
the gas phase, although instrumental limitations prevented the
7
-1 -1
mM MeOH, an upper limit of ca. 3 × 10 M
s
can be
absorption maximum of this species from being observed under
those conditions. The successful determination of the complete
UV-vis absorption spectra of these two species achieved in
this work strengthens our understanding of organotin(II) chem-
istry. Further work on these and related heavy carbene and
alkene systems is in progress in our laboratories.
estimated for the absolute rate constant for the reaction.
Summary and Conclusions
Strong evidence has been obtained for the formation of highly
reactive dimethylstannylene (SnMe2) from the 193-nm pho-
tolysis of rational stannacyclopentene precursors (4a) in the gas
phase and (4b) in solution. Steady-state photolysis in both cases
leads to formation of the corresponding dienes, and in solution,
SnMe2 has been trapped by reaction with Me3SnH to afford a
product that has been tentatively identified as Me3SnSnMe2H,
the product of Sn-H insertion.
Laser flash photolysis of 4b in solution produces a sequence
of transient species with partially overlapping broad-band
spectra, led by formation of strong absorptions at 500 nm,
followed by 470 nm, and eventually 260 nm. These are assigned
to SnMe2, Sn2Me4, and higher SnMe2 oligomers (possibly cyclo-
Sn3Me6 or cyclo-Sn4Me8), respectively, on the basis of both their
time evolutions and their calculated (TD-DFT) spectra. The
dimerization of SnMe2 is very rapid, proceeding at close to the
diffusion-controlled rate.
Experimental Section
The solvent hexanes (EMD Omnisolv) was purified by repeated
washing with concentrated sulfuric acid, followed by distilled water,
and preliminary drying over anhydrous sodium sulfate. It was then
refluxed for several days under nitrogen over sodium/potassium
amalgam and distilled. Deoxygenated samples of the solvent exhibited
an absorbance of ca. 0.2 at 193 nm in a 7 × 7 mm quartz cell. Methanol
(
2
Sigma-Aldrich) was distilled from Mg/I . The 1-stannacyclopent-3-
92
enes 4a and 4b were prepared as recently described, and 4a was further
purified before use by vacuum distillation. Spectroscopic data for these
compounds are listed in the Supporting Information.
All gases used in this work were deoxygenated thoroughly prior to
use. Commercial samples of reactive substrates used were obtained as
follows: 2-Butyne (GC purity > 99%) was from Cambrian Gases. HCl
(>99%) and MeOH (Gold label, >99%) were from Sigma-Aldrich.
The same long-wavelength (500 nm) transient absorption
band is observed in the gas-phase experiments, coincident with
earlier observations of SnMe2 generated from other precursors.
Long-lived absorptions are also observed in the 450-515-nm
wavelength region in the gas-phase experiments, following the
decay of the SnMe2 signal. The kinetic and spectroscopic
evidence indicates that these could very well be due to
Sn2Me4, formed in minor yield by dimerization of the primary
intermediate. Gas-phase trapping kinetics are broadly consistent
with earlier studies.
In solution, the spectral overlap of the bands from SnMe2
and Sn2Me4 makes it difficult to determine whether SnMe2
dimerizes irreversibly or reaches an equilibrium with Sn2Me4,
but the latter possibility can be ruled out on the basis of DFT
calculations of the bond dissociation energy of the distannene.
The calculations we have undertaken include comparisons of
geometries and transition energies with both the analogous
germanium and silicon systems, as well as those for the parent
2
SO (99.5%) was from BDH.
The instrumentation and procedures employed for gas- and solution-
phase laser flash photolysis experiments and details of the computational
methods employed are summarized in the Supporting Information.
Acknowledgment. We thank the following for financial
support: the Natural Sciences and Engineering Research Council
(NSERC) of Canada, the Canada Foundation for Innovation and
the Ontario Innovation Trust (W.J.L., C.R.H., and I.V.-B.), the
DGICYT (Spain) under Project BQU2002-03381 (R.B.), the
Royal Society for a Spain-U.K. Joint Project grant (No. 15636;
R.W.), and the United States National Science Foundation under
Grants CHE-9981715 and CHE-0316124 (P.P.G. and D.Z.).
Supporting Information Available: Details of the experi-
mental and computational procedures employed in this work,
computed geometric parameters and excitation energies, RRKM
calculations and kinetic derivations, spectroscopic data for 4a
and 4b, transient decay profiles recorded in hexane over
extended time scales, experimental data used for the estimation
of the equilibrium constant for methanol complexation, and
details of the steady-state trapping experiments. This material
is available free of charge via the Internet at http://pubs.acs.org.
(hydrogenated) and phenyl-substituted counterparts. The limited
experimental data available are in excellent agreement with these
calculations, providing a high level of confidence in comparisons
between analogous molecules based on elements in different
rows of the periodic table.
Addition of MeOH in solution leads to the formation of a
new transient species absorbing at 360 nm, which is assigned
JA052675D
(92) Zhou, D. Ph.D. Thesis, Washington University, St. Louis, MO, 2004.
17478 J. AM. CHEM. SOC.
9
VOL. 127, NO. 49, 2005