Organogermanium reactive intermediates. The direct detection and characterization of transient germylenes and digermenes in solution

Article abstract of DOI:10.1021/ja046308y

Diphenylgermylene (Ph₂Ge) and its Ge=Ge doubly bonded dimer, tetraphenyldigermene (6a), have been characterized directly in solution for the first time by laser flash photolysis methods. The germylene is formed via (formal) cheletropic photocycloreversion of 3,4-dimethyl-1,1- diphenylgermacyclopent-3-ene (4a), which is shown to proceed in high chemical (>95%) and quantum yield (Φ = 0.62) by steady-state trapping experiments with methanol, acetic acid, isoprene, and triethylsilane. Flash photolysis of 4a in dry deoxygenated hexane at 23°C leads to the prompt formation of a transient assigned to Ph₂Ge (∈_max = 500 nm; ∈_max = 1650 M^-1 cm^-1), which decays with second-order kinetics (τ ≈ 3 μs), with the concomitant growth of a second transient species that is assigned to digermene 6a (τ ≈ 40 μs; λ_max = 440 nm). Analogous results are obtained from 1,1-dimesityl- and 1,1-dimethyl-3,4-dimethylgermacyclopent-3-ene (4b and 4c, respectively), which afford Mes₂Ge (τ = 20 μs; λ_max = 560 nm) and Me₂Ge (τ ≈ 2 μs; λs; λ_max = 480 nm), respectively, as well as the corresponding digermenes, tetramesityl- (6b; λ_max = 410 nm) and tetramethyldigermene (6c; λ_max = 370 nm). The results for the mesityl compound are compared to the analogous ones from laser flash photolysis of the known Mes₂Ge/6b precursor, hexamesitylcyclotrigermane. The spectra of the three germylenes and two of the digermenes are in excellent agreement with calculated spectra, derived from time-dependent DFT calculations. Absolute rate constants for dimerization of Ph₂Ge and Mes₂Ge and for their reaction with n-butylamine and acetic acid in hexane at 23°C are also reported.

Full text of DOI:10.1021/ja046308y

Published on Web 11/18/2004
Organogermanium Reactive Intermediates. The Direct
Detection and Characterization of Transient Germylenes and
Digermenes in Solution
William J. Leigh,* Cameron R. Harrington, and Ignacio Vargas-Baca
Contribution from the Department of Chemistry, McMaster UniVersity,
1280 Main Street West, Hamilton, ON Canada L8S 4M1
Received June 22, 2004; E-mail: leigh@mcmaster.ca
Abstract: Diphenylgermylene (Ph2Ge) and its GedGe doubly bonded dimer, tetraphenyldigermene (6a),
have been characterized directly in solution for the first time by laser flash photolysis methods. The germylene
is formed via (formal) cheletropic photocycloreversion of 3,4-dimethyl-1,1-diphenylgermacyclopent-3-ene
(4a), which is shown to proceed in high chemical (>95%) and quantum yield (Φ ) 0.62) by steady-state
trapping experiments with methanol, acetic acid, isoprene, and triethylsilane. Flash photolysis of 4a in dry
deoxygenated hexane at 23 °C leads to the prompt formation of a transient assigned to Ph2Ge (λmax ) 500
nm; ꢀmax ) 1650 M-1 cm-1), which decays with second-order kinetics (τ ≈ 3 µs), with the concomitant
growth of a second transient species that is assigned to digermene 6a (τ ≈ 40 µs; λmax ) 440 nm). Analogous
results are obtained from 1,1-dimesityl- and 1,1-dimethyl-3,4-dimethylgermacyclopent-3-ene (4b and 4c,
respectively), which afford Mes2Ge (τ ≈ 20 µs; λmax ) 560 nm) and Me2Ge (τ ≈ 2 µs; λmax ) 480 nm),
respectively, as well as the corresponding digermenes, tetramesityl- (6b; λmax ) 410 nm) and tetrameth-
yldigermene (6c; λmax ) 370 nm). The results for the mesityl compound are compared to the analogous
ones from laser flash photolysis of the known Mes2Ge/6b precursor, hexamesitylcyclotrigermane. The
spectra of the three germylenes and two of the digermenes are in excellent agreement with calculated
spectra, derived from time-dependent DFT calculations. Absolute rate constants for dimerization of Ph2Ge
and Mes2Ge and for their reaction with n-butylamine and acetic acid in hexane at 23 °C are also reported.
substituted (Ph2Ge) derivatives,20-23 using photochemical meth-
ods to generate the germylene of interest and time-resolved
Introduction
The chemistry of germylenes, the germanium analogues of
carbenes, has been of considerable interest over the past thirty
years. Several stable derivatives have been prepared and
characterized, and their chemistry has been extensively studied.1-4
Numerous time-resolved spectroscopic studies of the reactivities
of the simplest representatives in the gas phase or in solution
have also been reported, including studies of the parent molecule
(H2Ge)5-15 and the dimethyl- (Me2Ge)15-19 and diphenyl-
UV-vis spectrophotometry to monitor it and determine rate
constants for its reaction(s). Unfortunately, however, much of
the reported work on Me2Ge and Ph2Ge in solution is character-
ized by a distinct lack of consistency in the spectroscopic
properties and absolute reactivities of these transient species.24
It has been suggested that this problem may originate in the
photochemistry of the arylated oligogermane or disilylgermane
precursors that have usually been used for their generation;
compounds of this type are prone to competing photoreactions
such as M-M′ bond homolysis and [1,3]-sigmatropic rear-
rangements, which yield transient germanium-centered radicals
and conjugated germene derivatives, respectively, in addition
to the desired germylenes.24-27 These products are usually
formed in minor yields, but they absorb strongly in the mid-
(1) Neumann, W. P. Chem. ReV. 1991, 91, 311.
(2) Barrau, J.; Rima, G. Coord. Chem. ReV. 1998, 180, 593.
(3) Weidenbruch, M. Eur. J. Inorg. Chem. 1999, 1999, 373.
(4) Tokitoh, N.; Okazaki, R. Coord. Chem. ReV. 2000, 210, 251.
(5) Becerra, R.; Boganov, S. E.; Egorov, M. P.; Nefedov, O. M.; Walsh, R.
Chem. Phys. Lett. 1996, 260, 433.
(6) Becerra, R.; Boganov, S. E.; Egorov, M. P.; Faustov, V. I.; Nefedov, O.
M.; Walsh, R. J. Am. Chem. Soc. 1998, 120, 12657.
(7) Becerra, R.; Walsh, R. Phys. Chem. Chem. Phys. 1999, 1, 5301.
(8) Alexander, U. N.; Trout, N. A.; King, K. D.; Lawrance, W. D. Chem. Phys.
Lett. 1999, 299, 291.
(9) Alexander, U. N.; King, K. D.; Lawrance, W. D. Chem. Phys. Lett. 2000,
319, 529.
(10) Becerra, R.; Boganov, S. E.; Egorov, M. P.; Faustov, V. I.; Nefedov, O.
M.; Walsh, R. Can. J. Chem. 2000, 78, 1428.
(11) Becerra, R.; Boganov, S. E.; Egorov, M. P.; Faustov, V. I.; Nefedov, O.
M.; Walsh, R. Phys. Chem. Chem. Phys. 2001, 3, 184.
(12) Becerra, R.; Walsh, R. J. Organomet. Chem. 2001, 636, 49.
(13) 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.
(14) Alexander, U. N.; King, K. D.; Lawrance, W. D. Phys. Chem. Chem. Phys.
2003, 5, 1557.
(15) Becerra, R.; Egorov, M. P.; Krylova, I. V.; Nefedov, O. M.; Walsh, R.
Chem. Phys. Lett. 2002, 351, 47.
(16) Bobbitt, K. L.; Maloney, V. M.; Gaspar, P. P. Organometallics 1991, 10,
2772.
(17) Mochida, K.; Kanno, N.; Kato, R.; Kotani, M.; Yamauchi, S.; Wakasa,
M.; Hayashi, H. J. Organomet. Chem. 1991, 415, 191.
(18) Mochida, K.; Tokura, S. Bull. Chem. Soc. Jpn. 1992, 65, 1642.
(19) Becerra, R.; Boganov, S. E.; Egorov, M. P.; Lee, V. Y.; Nefedov, O. M.;
Walsh, R. Chem. Phys. Lett. 1996, 250, 111.
(20) Konieczny, S.; Jacobs, S. J.; Braddock Wilking, J. K.; Gaspar, P. P. J.
Organomet. Chem. 1988, 341, C17.
(21) Wakasa, M.; Yoneda, I.; Mochida, K. J. Organomet. Chem. 1989, 366,
C1.
(22) Mochida, K.; Yoneda, I.; Wakasa, M. J. Organomet. Chem. 1990, 399,
53.
(23) Mochida, K.; Wakasa, M.; Hayash, H. Phosphorus, Sulfur Silicon Relat.
Elem. 1999, 150-151, 237.
(24) Boganov, S. E.; Egorov, M. P.; Faustov, V. I.; Nefedov, O. M. In The
chemistry of organic germanium, tin and lead compounds - Vol. 2;
Rappoport, Z., Ed.; John Wiley and Sons: New York, 2002; p 749.
9
10.1021/ja046308y CCC: $27.50 © 2004 American Chemical Society
J. AM. CHEM. SOC. 2004, 126, 16105-16116
16105

A R T I C L E S
Leigh et al.
UV (400-500 nm), in exactly the same spectral region that the
germylenes of interest are expected to absorb. The proper
interpretation of the results of these studies was also hindered
by the nonavailability of quantitative kinetic data for the
reactions of transient germenes, the first of these studies having
been reported only relatively recently.28 Complexation of the
germylene with the precursor, solvent. or a photolysis co-product
may also be a contributing factor.24
1,1-dimesityl (4b; mesityl ) 2,4,6-trimethylphenyl) analogue.
These compounds were identified as potential precursors to
diphenyl- and dimesitylgermylene, respectively, based simply
on the fact that the 1,1-dihydro derivative (4d) and 1,1-
dimethylgermacyclopent-3-ene (5) are known to extrude the
corresponding germylenes upon photolysis in the gas phase with
193 nm light, in amounts sufficient for their detection by time-
resolved UV-vis spectroscopy.6,19,32 The dimesityl compound
4b serves as a useful control against which to judge the results
for 4a, as the UV spectrum and reactivity of dimesitylgermylene
(Mes2Ge) have been previously characterized in solution25 and
in low-temperature matrixes,29,30,33 with quite good agreement
between the various studies that have been reported. We have
also carried out a preliminary study of the photochemistry of
1,1,3,4-tetramethylgermacyclopent-3-ene (4c) in solution by far-
UV (193 nm) laser flash photolysis techniques,34 to compare
with previous gas- and solution-phase flash photolysis studies
of Me2Ge, and with the results obtained from 248 nm laser
photolysis of the arylated derivatives 4a and 4b.
For example, the UV spectrum of Ph2Ge, generated by
photolysis of 1-3 in a 3-methylpentane (3-MP) glass at 77 K,
has been reported to exhibit λmax ≈ 465 nm.21,22,29,30 On the
other hand, laser flash photolysis studies of 1 and 2 in
cyclohexane solution at room temperature have afforded tran-
sient UV spectra that are different from those observed in 3-MP
at 77 K and kinetic behavior that varies with the precursor.20-22
A species with λmax ) 445 nm and second-order lifetime (τ) of
ca. 270 µs was assigned to Ph2Ge in experiments using 1 as
precursor,20 while a transient with λmax ) 450 nm and (second-
order) τ ≈ 350 ns was later assigned to the same species on the
basis of similar experiments with 2.21,22 Absolute rate constants
for reaction of the species with hydridosilanes and 2,3-dimethyl-
1,3-butadiene (DMB) differed by about an order of magnitude
from the two precursors, and no evidence was found for reaction
with ethanol. Steady-state photolysis experiments verified that
Ph2Ge extrusion is the major photoreaction of both 1 and 2 but
also indicate that other competing processes occur as well.
Unfortunately, to our knowledge benzogermanorbornadiene 3,
the one precursor that is evidently not prone to the competing
photoreactions that 1 and 2 undergo,30,31 has not yet been studied
in fluid solution by time-resolved spectroscopic methods. The
analogous literature on Me2Ge is much more extensive than
that on Ph2Ge and is in a correspondingly greater state of
confusion.24
In this paper, we present the results of steady-state (lamp)
photolysis of these compounds in the presence of various
germylene trapping reagents, which show that this class of
precursor does indeed constitute a highly efficient, essentially
quantitative source of the corresponding germylene derivative.
The germylenes and their corresponding digermene dimers (6a-
c) have been detected and characterized by laser flash photolysis
methods, using time-dependent DFT calculations to support the
transient assignments. The assignment of tetraphenyldigermene
(6a) is further supported by product studies using high-intensity
laser excitation, which enables competitive trapping of the
digermene and the germylene from which it is formed. Absolute
rate constants for dimerization of the two diarylgermylenes and
for reaction with n-butylamine and acetic acid are also reported,
the latter to provide further support for the transient assignments.
The results reported here form the basis of a more detailed
kinetic study of the reactions of Ph2Ge and 6a with a broad
range of reactants, including amines, alcohols, carboxylic acids,
ketones, alkenes, alkynes, dienes, Group 14 hydrides, halocar-
bons, and oxygen, which will be reported separately.35
We have thus sought to develop an efficient photochemical
precursor to Ph2Ge that should not be prone to competing
photoreactions of the types characteristic of arylated oligoger-
manes or disilylgermanes and is more straightforward to prepare
and handle than molecules such as 3. Herein we report the first
results of our efforts, in the form of a steady-state and laser
flash photolysis study of the photochemistry of 3,4-dimethyl-
1,1-diphenylgermacyclopent-3-ene (4a) and the corresponding
(25) Toltl, N. P.; Leigh, W. J.; Kollegger, G. M.; Stibbs, W. G.; Baines, K. M.
Organometallics 1996, 15, 3732.
(26) Baines, K. M.; Cooke, J. A.; Dixon, C. E.; Liu, H.; Netherton, M. R.
Organometallics 1994, 13, 631.
Results and Discussion
(27) Leigh, W. J.; Toltl, N. P.; Apodeca, P.; Castruita, M.; Pannell, K. H.
Organometallics 2000, 19, 3232.
Compounds 4a-c were prepared in 80-90% yield by
reaction of the corresponding aryl or alkyl Grignard reagent
with 1,1-dichloro-3,4-dimethylgermacyclopent-3-ene (7; eq 1),
which was prepared in high yield via a two-step procedure
(28) Toltl, N. P.; Leigh, W. J. J. Am. Chem. Soc. 1998, 120, 1172.
(29) Ando, W.; Tsumuraya, T.; Sekiguchi, A. Chem. Lett. 1987, 317.
(30) Ando, W.; Itoh, H.; Tsumuraya, T. Organometallics 1989, 8, 2759.
(31) Kocher, J.; Lehnig, M.; Neumann, W. P. Organometallics 1988, 7, 1201.
(32) Note added in proof: We recently learned that preliminary details of the
photolysis of 4a in the presence of isoprene were reported several years
ago in a conference proceeding. See: Bobbitt, K. L.; Lei, D.; Maloney, V.
M.; Parker, B. S.; Raible, J. M.; Gaspar, P. P. In Frontiers of organoger-
manium, -tin and -lead chemistry; Lukevics, E., Ignatovich, L., Eds.; Latvian
Institute of Organic Synthesis: Riga, 1993; p 41. We thank Professor P.
P. Gaspar for bringing this to our attention and for providing a reprint of
the article.
(33) Ando, W.; Itoh, H.; Tsumuraya, T.; Yoshida, H. Organometallics 1988, 7,
1880.
(34) Kerst, C.; Byloos, M.; Leigh, W. J. Can. J. Chem. 1997, 75, 975.
(35) Leigh, W. J.; Harrington, C. R. Di- and Trivalent Organogermanium
Reactive Intermediates. Kinetics and Mechanisms of Some Reactions of
Diphenylgermylene and Tetraphenyldigermene in Solution. To be published.
9
16106 J. AM. CHEM. SOC. VOL. 126, NO. 49, 2004

Transient Germylenes and Digermenes in Solution
A R T I C L E S
starting from germanium tetrachloride. The three compounds
exhibited spectroscopic and analytical properties similar to
previously reported data.36-38
(see Supporting Information), with the identity of 12 being
confirmed by spiking the photolyzed mixture with a small
amount of an authentic, independently prepared sample of the
compound. The relative yields of 12 and 9a in these experiments
varied depending on the laser intensity and alcohol concentra-
tion, with the yield of 12 relative to that of 9a being significantly
lower (or not detectable at all) in experiments carried out at
lower laser intensities or in the presence of higher concentrations
of MeOH. For example, only 9a could be detected upon
photolysis of 4a in the presence of 0.03 M MeOH, under
otherwise identical conditions as required to produce 12 and
9a in a ratio of ca. 1:4 from photolysis of 4a in the presence of
0.003 M MeOH. These results are consistent with competitive
trapping of Ph2Ge with MeOH and dimerization to form
tetraphenyldigermene (6a), which is subsequently trapped as
the expected methanol addition product (12; eq 3). The success
of the experiment depends on the use of a sufficiently high laser
flux to ensure that Ph2Ge is formed in relatively high concentra-
tions and sufficiently low methanol concentrations to ensure
that dimerization of the germylene competes effectively with
trapping by the alcohol.
Photolysis of argon-saturated cyclohexane-d12 solutions of 4a
(0.05 M) in the presence of 0.5 M triethylsilane, methanol, acetic
acid, or isoprene, with low-pressure mercury lamps (254 nm),
led to the formation of 2,3-dimethyl-1,3-butadiene (DMB) and
equal yields of the expected germylene-trapping products 8a,
9a, 10a, and 11a, respectively. The products, which were in
The quantum yield for formation of 9a from 4a was
determined to be ΦPh2Ge ) 0.62 ( 0.08, by merry-go-round
photolysis of deoxygenated, optically matched solutions of 4a
(0.037 M) and 1,1-diphenylsilacyclobutane (13; 0.043 M) in
methanolic (0.5 M) cyclohexane-d12 solution, with periodic
monitoring of the photolyzates by 1H NMR spectroscopy
between 0 and ca. 25% conversion of 4a. The photolysis of 13
under these conditions yields ethylene and methoxymethyl-
diphenylsilane (14; eq 4) with a quantum yield of Φ14 ) 0.21
( 0.02.40
each case formed in chemical yields exceeding 95% at conver-
sions below ca. 20%, were identified by H NMR and FTIR
1
spectroscopy and GC/MS analyses of the crude reaction
mixtures. The assignments were corroborated by comparisons
to authentic samples in three of the four cases (8a, 9a, 11a);
the acetoxygermane (10a) could not be successfully isolated.
Similarly, compounds 9b and 11b were obtained along with
DMB in quantitative yields upon photolysis of cyclohexane-
d12 solutions of 4b in the presence of methanol or isoprene,
under similar conditions to those used for the photolyses of 4a.
Finally, 214 nm photolysis of deoxygenated solutions of 4c (0.02
M) in cyclohexane-d12 containing methanol (0.035 M) or
isoprene (0.005 M) afforded DMB and roughly equal amounts
1
of 9c or 11c, respectively, according to H NMR, GC, and/or
GC/MS analysis of the crude product mixtures between ca. 3
and 10% conversion of the starting material.
Laser Flash Photolysis of 4a-c. The Direct Detection of
Germylenes and Digermenes. Laser flash photolysis of
continuously flowing, deoxygenated solutions of 4a (ca. 0.003
M) in dried hexane with the pulses from a KrF excimer laser
(248 nm, ca. 25 ns, ca. 100 mJ) leads to the formation of two
transient species. This is illustrated in Figure 1a, in the form of
transient absorption spectra recorded in point-by-point fashion
55-170 ns and 3.0-3.2 µs after the laser pulse. The first
transient is formed with the pulse and exhibits an absorption
maximum at λmax ) 500 nm. The signal due to this species is
superimposed on that of a second, much longer-lived transient,
which exhibits λmax ) 440 nm, grows in with approximate
second-order kinetics over ca. 2 µs, and then decays with mixed
first- and second-order kinetics and a lifetime τ ≈ 40 µs. The
absorptions due to this second transient overlap considerably
with those of the initially formed transient, so much so that the
kinetic form of the decay of the initially formed species cannot
be determined from the raw data recorded at 500 nm. However,
the decay profile of the initially formed transient could be
Preparative laser photolysis of a deoxygenated solution of
4a (0.003 M) in hexane containing MeOH (0.003 M), with the
focused pulses from a KrF excimer laser (248 nm; ca. 25 ns,
ca. 150 mJ) in a laser-drop apparatus,39 afforded a mixture of
DMB, 9a, and 1-methoxy-1,1,2,2-tetraphenyldigermane (12; eq
2). The conversion of 4a in this experiment was ca. 2%. The
(36) Mazerolles, P.; Manuel, G. Bull. Soc. Chim. Fr. 1965, 56, 327.
(37) Tsumuraya, T.; Kabe, Y.; Ando, W. J. Organomet. Chem. 1994, 482, 131.
(38) Schriewer, M.; Neumann, W. P. J. Am. Chem. Soc. 1983, 105, 887.
(39) Banks, J. T.; Garcia, H.; Miranda, M. A.; Perez-Prieto, J.; Scaiano, J. C. J.
Am. Chem. Soc. 1995, 117, 5049.
(40) Toltl, N. P.; Stradiotto, M. J.; Morkin, T. L.; Leigh, W. J. Organometallics
1999, 18, 5643.
products were identified by (600 MHz) 1H NMR spectroscopy
9
J. AM. CHEM. SOC. VOL. 126, NO. 49, 2004 16107

A R T I C L E S
Leigh et al.
one, partly because the 440 nm transient absorption contains
minor contributions from the 500 nm species; accordingly, it
afforded a decay profile that fit acceptably to mixed first- and
second-order kinetics with a lifetime τ ≈ 1.5 µs. Traces recorded
at 530 nm were ca. 40% weaker than those recorded at 500
nm, but fit reasonably well to second-order kinetics under
scrupulously dry, deoxygenated conditions with a (second-order)
lifetime of ca. 3 µs (vide infra). The reasonably good qualitative
agreement between the decay rate of the 500 nm species and
the apparent growth rate of the 440 nm species is consistent
with the latter being formed from the former via a second-order
kinetic process, allowing assignment of the initially formed
transient to diphenylgermylene (Ph2Ge) and the subsequently
formed one to its dimer, tetraphenyldigermene (6a). We note
that quantitative agreement between the germylene decay and
digermene growth kinetics can only be expected under circum-
stances where the digermene is stable over a much longer time
scale than that in which it is formed (and it is not; see Figure
1a, insert) and where other reactive pathways (e.g., reaction with
impurities, the precursor, the photolysis coproduct, or the
dimerization product as it is formed) do not compete with
dimerization. This will be discussed in more detail later in the
paper.
The transient spectra shown in Figure 1a both contain a
second, significantly more intense absorption band centered at
λmax ) 290-300 nm, a result of the fact that transient decay
profiles recorded at this wavelength consist of both a fast and
slow decay component, superimposed on a stable residual
absorption that is most likely associated with oligomerization
products of 6a. The decay characteristics of the signals in this
region of the spectrum are too complex for reliable assignments
to be made at this time.
Similarly, flash photolysis of a flowed, deoxygenated 0.004
M solution of 4b in dry hexane affords the known dimesitylger-
mylene (λmax ) 325, 560 nm),25,30 which decays over a similar
time scale as the growth of absorptions due to tetramesityldi-
germene (6b; λmax ) 410 nm),25,30,41 as illustrated in Figure
1b. Again, only qualitative agreement is observed between the
decay kinetics of the germylene absorption and the apparent
growth kinetics of the digermene, but both fit reasonably well
to second-order kinetics and exhibit decay and growth constants
that match each other within a factor of 2. The optical yield of
dimesitylgermylene from 4b is significantly higher than that
obtained from either hexamesitylcyclotrigermane (15) or bis-
(trimethylsilyl)dimesitylgermane (16), which have both been
studied previously by our group by flash photolysis methods.25
However, the yield of 6b by dimerization of Mes2Ge: is
significantly lower than that obtained from 15 under similar
conditions, and the digermene appears to be unstable, decaying
with a pseudo-first-order lifetime of ca. 2 ms (vide infra).
Figure 1. Transient UV absorption spectra from laser flash photolysis of
(a) a 0.003 M solution of 4a in dried, deoxygenated hexane solution at 23
°C, recorded 55-170 ns (O) and 3.0-3.2 µs (b) after the (248 nm) laser
pulse; (b) a 0.004 M solution of 4b under similar conditions, recorded 50-
400 ns (O) and 16.8-17.8 µs (b) after the (248 nm) laser pulse; and (c) a
5 × 10-5 M solution of 4c in dried, deoxygenated hexane solution at 23
°C, recorded 50-400 ns (O) and 7.8-8.6 µs (b) after the (193 nm) laser
pulse. The inserts show typical transient absorption profiles recorded at (a)
500 and 440 nm; (b) 550 and 410 nm; and (c) 480 and 370 nm. The third
transient decay profile shown in the insert in part a was calculated by
subtracting the 440 nm growth/decay profile (× 0.25) from the 500 nm
decay profile.
Finally, far-UV (193 nm) laser flash photolysis of deoxy-
genated hexane solutions of 4c (ca. 5 × 10-5 M) afforded similar
transient behavior to that described above for 4a and 4b. The
signals obtained in these experiments were substantially weaker,
extracted from the raw decays recorded at 500 nm by subtraction
of the profile recorded at 440 nm, scaled by the approximate
relative absorbances of the 440 nm species at 500 and 440 nm
(see Figure 1a, insert; ꢀ500/ꢀ440 ≈ 0.25). The procedure is a crude
(41) Masamune, S.; Hanzawa, Y.; Williams, D. J. J. Am. Chem. Soc. 1982,
104, 6136.
9
16108 J. AM. CHEM. SOC. VOL. 126, NO. 49, 2004

Transient Germylenes and Digermenes in Solution
A R T I C L E S
Figure 2. Pseudo-first-order rate constants for the decay of Ph2Ge (O) and Mes2Ge (0) in the presence of (a) n-BuNH2 and (b) AcOH in dried hexane
solution at 23 °C, at concentrations at which the formation of 6a and 6b is more than 80% quenched.
Table 1. Absolute Rate Constants for Reaction of
Diarylgermylenes (Ph2Ge and Mes2Ge) and Tetraaryldigermenes
(6a and 6b) with n-BuNH2 and AcOH in Dried Hexane Solution at
however, due most likely to a combination of lower quantum
yield, weaker excitation pulse energy (roughly half that at 248
nm), and screening of the excitation light by the solvent (which
accounts for ca. 30% of the solution absorbance at 193 nm).
Transient spectra and typical decay/growth profiles are shown
in Figure 1c; they consist of a prompt, relatively weak absorption
centered at 480 nm, which decays with approximate second-
order kinetics (τ ≈ 2 µs) to form a second species that absorbs
at λmax ) 370 nm and then decays over the next several hundred
microseconds. The absorption spectrum of the initially formed
transient is similar to that reported previously for Me2Ge in the
gas-phase (from laser photolysis of 5b and pentamethyldiger-
mane (17))15,19 and in solution (from laser photolysis of
decamethylcyclopentagermane (18)).18 That of the second
transient matches the spectrum previously assigned to tetram-
ethyldigermene (6c) by Mochida and co-workers.17,42
23 °C
1
1
kQ/109 M-1 s-
kQ/109 M-1 s-
Q
Ph2Gea
Mes2Gea
6aa
6bb
n-BuNH2
AcOH
10.1 ( 0.6
3.9 ( 0.7
7.0 ( 0.30
1.6 ( 0.3
2.4 ( 0.3
<0.04
<7 × 10-7
<10-7
a Using 4a or 4b as precursor. b Using 15 as precursor.
acid-base complexes between diarylgermylenes and aliphatic
alcohols in low-temperature matrixes has been reported by Ando
and co-workers.30 On the other hand, the presence of residual
oxygen resulting from incomplete deoxygenation had only small
effects on the intensities and decay profiles of the signals due
to Ar2Ge and the corresponding digermenes.
Addition of n-butylamine (n-BuNH2) or acetic acid (AcOH)
to the solutions of 4a and 4b caused enhancements in the decay
rates of the signals assigned to Ph2Ge and Mes2Ge and
reductions in the maximum intensities of the signals due to 6a
and 6b, in proportion to the concentration of added reagent.
This is consistent with trapping of the germylenes by n-BuNH2
and AcOH, in competition with dimerization to the correspond-
ing digermenes. At concentrations above those required to
reduce the digermene signal intensities to less than ca. 20% of
their values in the absence of added reagent (ca. 0.1-0.5 mM),
the decay profiles of the germylene signals followed clean first-
order kinetics and plots of the pseudo-first-order decay rate
constants (kdecay) vs concentration were linear. Typical plots are
shown in Figure 2. The linear behavior is consistent with the
relationship of eq 5, where kQ is the second-order rate constant
for reaction of the germylene with the added reagent (Q) and
k0 is the (hypothetical) pseudo-first-order rate constant for
germylene decay in the absence of Q. The absolute rate constants
for reaction of Ph2Ge and Mes2Ge with these two reagents are
collected in Table 1.
The intensities of the signals observed in experiments with
4a and 4b depend critically on the concentration of water in
the solvent, with the strongest signals being observed with
samples of hexane that had been dried by distillation from
lithium aluminum hydride or sodium/potassium amalgam, using
oven-baked glassware and sample cells. These effects are much
more pronounced for 4a than for 4b and are indicative of
reaction of the germylenes with water or other hydroxylic
impurities, which results in reductions in the intensities of the
signals observed for both the germylene and the digermene. For
example, laser flash photolysis of 4a in a sample of untreated,
HPLC-grade hexane ([H2O] ≈ 1.5 mM) afforded signals due
to Ph2Ge and 6a that were roughly 55% and 75% weaker,
respectively, than those recorded in the dried solvent. Interest-
ingly, the bulk decay and growth kinetics of the two species
are approximately the same under the two sets of conditions,
suggestive of a rapid equilibrium between the free germylene
and its complex with water.35 Indeed, such a phenomenon was
suggested several years ago by Neumann and co-workers for
Me2Ge,43 and spectroscopic evidence for the formation of Lewis
kdecay ) k0 + kQ[Q]
(5)
In the case of the amine, acceleration of the decay of the
signal due to 6a and a change to first-order decay kinetics was
evident over the (low) concentration ranges where the species
could still be detected. A plot of kdecay vs [n-BuNH2] was linear
(see Figure 3) and afforded a quenching rate constant of kQ )
(2.4 ( 0.3) × 109 M-1 s-1. No such acceleration was evident
in the decay of 6b over a similar concentration range in amine.
(42) Mochida, K.; Kayamori, T.; Wakasa, M.; Hayashi, H.; Egorov, M. P.
Organometallics 2000, 19, 3379.
(43) Klein, B.; Neumann, W. P.; Weisbeck, M. P.; Wienken, S. J. Organomet.
Chem. 1993, 446, 149.
9
J. AM. CHEM. SOC. VOL. 126, NO. 49, 2004 16109

A R T I C L E S
Leigh et al.
the fits and absolute rate constants that are ca. 4 and 2 times
greater, respectively, than the value measured using the bulk
concentrations. However, we have no information on which to
base an opinion on the relative reactivities of the monomeric
and dimeric forms of AcOH toward O-H insertion by the
germylene.
The course of the reaction of n-BuNH2 with 6a has not yet
been ascertained but can most likely be identified as 1,2-addition
across the GedGe bond. The substantially greater reactivity of
6a toward the amine compared to the carboxylic acid is
suggestive of a mechanism involving initial nucleophilic attack
at germanium, analogous to the addition of nucleophiles to
disilenes.45,46 As expected, 6b is several orders of magnitude
less reactive than 6a, probably owing mainly to the greater
degree of steric protection of the GedGe bond that is afforded
by the mesityl substituents in the molecule.
Figure 3. Pseudo-first-order rate constants for the decay of digermene 6a
in the presence of n-BuNH2 in dried hexane solution at 23 °C, over the
concentration range at which its formation is less than 90% quenched. The
solid line represents the best (linear least squares) fit of the data to eq 5.
The extinction coefficient for Ph2Ge in hexane solution at
the absorption maximum was determined by benzophenone
actinometry.47,48 The method involves comparing the initial ∆A
value at 500 nm (∆A500,0) due to Ph2Ge to that recorded at 525
nm for a solution of benzophenone, whose irradiation yields
the triplet state (λmax ) 525 nm, ꢀ525nm ) 6250 ( 1250 M-1
cm-1) with unit quantum yield.49 The substrate and actinometer
solutions were optically matched at the laser wavelength (248
nm) to ensure equal light absorption, both solutions were flowed,
and the laser intensity was varied using neutral density filters.
As expected, plots of ∆Aλ,0 vs laser intensity at the two
monitoring wavelengths were linear for both compounds at low
to moderate laser intensities (see Supporting Information).48 The
extinction coefficient for Ph2Ge at its absorption maximum was
then calculated from the relative slopes of the plots and the
quantum yield for formation of Ph2Ge determined above.
Duplicate determinations yielded an average value of ꢀPh2Ge,500
A slight acceleration of the decay rates of 6a was evident in
the experiments with AcOH, allowing assignment of an upper
limit of ∼4 × 107 M-1 s-1 for the rate constant for its reaction
with AcOH. Upper limits for the rate constants for reaction of
these two reagents with 6b are on the order of 102 M-1 s-1 and
were established using 15 as the precursor.
These data are a subset of an extensive study of the kinetics
and mechanisms of the reactions of these species with a variety
of germylene and digermene scavengers, which has been carried
out in parallel with the present work and will be reported
separately.35 They are included here because they effectively
rule out a number of possible alternative transient assignments,
such as excited triplet states, free radicals, or radical ions. On
the other hand, both amines and carboxylic acids are known to
react with transient germylenes.43 Absolute rate constants have
not yet been reported in any case, however.30,33
In both cases, addition of n-BuNH2 also resulted in the
formation of a new transient species that decayed with mixed-
order kinetics and exhibited λmax e 325 nm. We tentatively
assign these species to the Lewis acid-base complexes between
Ar2Ge and the amine (19; eq 6), as complexes of this type have
been reported previously upon generation of these and other
germylenes in low-temperature glasses.30,33 Interestingly, steady-
) 1650 ( 390 M-1 cm-1
.
The maximum absorption intensities at 440 nm were also
measured during these experiments; plots of ∆A440,max vs laser
intensity also showed good linearity, and the average slopes
afforded an estimate of ꢀ6a,440 ) 5970 ( 1250 M-1 cm-1 for
6a. However, this value was calculated without consideration
for the fact that 6a decays over the time scale in which it is
formed and with the assumption that it is formed from Ph2Ge
in quantitative yield, which we do not believe to be true (vide
infra). It should thus be considered a lower limit, differing from
the true value by at least a factor of 2.
state photolysis of cyclohexane-d12 solutions of 4a containing
n-BuNH2 leads to the formation of oligomeric material and very
small amounts of nitrogen-containing products that are detect-
able only at higher conversions of the starting material. On the
other hand, the reaction with AcOH is quite clean, resulting in
the exclusive formation of the O-H insertion product (e.g., 10a
from Ph2Ge) with no detectable intermediates. It should be noted
that the carboxylic acid exists as a mixture of the monomeric
species and the hydrogen-bonded dimer in hexane even at sub-
millimolar bulk concentrations.44 Replotting the decay rate
constants of Figure 2b against either the monomeric AcOH
concentration or the sum of the monomer and dimer concentra-
tions (calculated using the monomer-dimer equilibrium constant
of Fujii and co-workers for acetic acid in hexane at 25 °C, K )
3200 M-1 44) led to significant improvements in the quality of
As mentioned above, Ph2Ge has previously been reported to
exhibit a UV absorption maximum at 445-450 nm in hydro-
carbon solvents at room temperature20-22,50 and at 465 nm in
hydrocarbon glasses at 77K,22,30 in contrast to the results
reported above from laser photolysis of 4a. Previous assignments
of the UV spectrum of 6a are similarly contradictory, with
(45) Sakurai, H. In The Chemistry of Organic Silicon Compounds; Rappoport,
Z., Apeloig, Y., Eds.; Wiley and Sons: Ltd: 1998; pp 827-855.
(46) Morkin, T. L.; Owens, T. R.; Leigh, W. J. In The chemistry of organic
silicon compounds, Volume 3; Rappoport, Z., Apeloig, Y., Eds.; John Wiley
and Sons: New York, 2001; pp 949-1026.
(47) Carmichael, I.; Hug, G. L. In CRC handbook of organic photochemistry,
Vol. I; Scaiano, J. C., Ed.; CRC Press: Boca Raton, FL, 1989; pp 369-
404.
(48) Wintgens, V.; Johnston, L. J.; Scaiano, J. C. J. Am. Chem. Soc. 1988, 110,
5117.
(49) Carmichael, I.; Helman, W. P.; Hug, G. L. J. Phys. Chem. Ref. Data 1987,
16, 239.
(50) Mochida, K.; Wakasa, M.; Nakadaira, Y.; Sakaguchi, Y.; Hayashi, H.
Organometallics 1988, 7, 1869.
(44) Fujii, Y.; Yamada, H.; Mizuta, M. J. Phys. Chem. 1988, 92, 6768.
9
16110 J. AM. CHEM. SOC. VOL. 126, NO. 49, 2004

Transient Germylenes and Digermenes in Solution
A R T I C L E S
values of λmax varying between 320 nm20 and 450 nm23 in
hydrocarbon solution at room temperature. Similarly, there is
poor consensus as to the position of the UV spectral maximum
of Me2Ge, with reported values varying between 320 and 490
nm.24 The spectra reported here for the three germylenes are
similar to those reported previously for the corresponding
silylene derivatives Me2Si (λmax ) 465 nm in cyclohexane at
∼20 °C51,52), Ph2Si (λmax ) 495 nm in 3-methylpentane at 77
K53), and Mes2Si (λmax ) 580 nm in cyclohexane at 25 °C54);
analogous similarities exist between the spectra of 6b and 6c
and those of tetramesityldisilene (λmax ) 420 nm in cyclohexane
at 25 °C54a) and tetramethyldisilene (λmax ) 360 nm in
cyclohexane at 22 °C54b), respectively. A reinvestigation of the
time-resolved spectroscopy of 1 and a similar study of an
analogue of 3 is in progress; the preliminary results of this work
are in fact fully compatible with the present results for 4a.55
Nevertheless, we decided to compare the germylene and
digermene spectra obtained in the present work with the results
of time-dependent density functional theory (TD-DFT) calcula-
tions, to establish the validity of our transient assignments
further.
larly, the calculated structural parameters for digermene (Ge2H4)
and 6c agree well with the results of previous calculations for
the two molecules.63-68 In the cases of 6a and 6b, the calculated
bond distances are generally in good agreement with crystal-
lographic data reported for tetrakis(2,6-diethylphenyl)digermene
(22),69 but the GedGe distances are overestimated by as much
as 0.1 Å. More serious deviations from the crystallographic
structure are observed in the bond and torsional angles about
the germanium atoms, and the calculations predict a much
greater degree of pyramidalization at germanium than is
observed in the crystal structure of 22.
Although the dimensions of the structures calculated here for
6a and 6b fall within the experimentally observed range for 22
and are in agreement with theoretical studies of smaller
molecules,70-72 the results for all digermenes must be taken with
caution. The accurate prediction of the structure of compounds
containing double bonds between heavy Group 14 elements is
notoriously difficult; the potential surfaces are fairly flat73
resulting in great variability of the GedGe bond distance
(2.213-2.454 Å), the pyramidalization angle (∼0-45°), and
the twist angle (<10°).3,69,74,75
TD-DFT Calculations of the Electronic Spectra of Ger-
mylenes and Digermenes. The choice of basis set and density
functional used in the calculations was made by first carrying
out a series of geometry optimizations for the stable germylene
20 and comparing the results to the reported single-crystal X-ray
structure of the compound.56 Four functionals were investigated,
Preliminary TD-DFT calculations were then carried out for
20 to identify the best basis set and exchange-correlation model
for the prediction of electronic transitions in these species. The
best balance between accuracy (as defined by the degree of
agreement with the experimental UV spectrum56) and computa-
tion time was obtained using the statistical average of different
model potentials for occupied KS orbitals (SAOP)76,77 with
triple-ú double-polarization basis sets and the inclusion of
relativistic effects (ZORA formalism). The basis sets were
reduced to triple-ú single-polarization to expedite the calcula-
tions for the digermenes.
The calculated lowest energy transitions in the UV-visible
spectra of the three germylenes and the corresponding di-
germenes studied in this work are collected in Table 2, along
with the experimental absorption maxima determined by laser
flash photolysis of 4a-c. Calculated and experimental λmax
values for 20 are also included in the table. There is excellent
using uncontracted Slater-type orbitals (STOs) of triple-ú quality
as basis functions (see Supporting Information). The exchange
and correlation functionals of Perdew and Wang (PW91) were
found to afford the best agreement with the reported crystal-
lographic data for 20 and were thus used in subsequent geometry
optimizations for the other species investigated. The calculated
R-Ge distances and R-Ge-R bond angles (R ) H or C) for
H2Ge, Me2Ge, and Ph2Ge (see Supporting Information) and
vibrational frequencies for H2Ge agreed well with the results
of previous calculations at other levels of theory1,57-61 and with
experimental structural data for H2Ge62 and the stable diarylger-
mylene bis[2,6-bis(1-naphthyl)phenyl]germylene (21).59 Simi-
(57) BelBruno, J. J. Heteroatom. Chem. 1998, 9, 195.
(58) Su, M.-D.; Chu, S.-Y. J. Am. Chem. Soc. 1999, 121, 11478.
(59) Wegner, G. L.; Berger, R. J. F.; Schier, A.; Schmidbaur, H. Organometallics
2001, 20, 418.
(60) Szabados, A.; Hargittai, M. J. Phys. Chem. A 2003, 107, 4314.
(61) Lemierre, V.; Chrostowska, A.; Dargelos, A.; Baylere, P.; Leigh, W. J.;
Harrington, C. R. Appl. Organomet. Chem. 2004, 18, 000.
(62) Karolczak, J.; Harper, W. W.; Grev, R. S.; Clouthier, D. J. J. Chem. Phys.
1995, 103, 2839.
(63) Liang, C.; Allen, L. C. J. Am. Chem. Soc. 1990, 112, 1039.
(64) Grev, R. S.; Schaefer, H. F., III Organometallics 1992, 11, 3489.
(65) Chen, W.-C.; Su, M.-D.; Chu, S.-Y. Organometallics 2001, 20, 564.
(66) Malcolm, N. O. J.; Gillespie, R. J.; Popelier, P. L. A. J. Chem. Soc., Dalton
Trans. 2002, 3333.
(67) Mosey, N. J.; Baines, K. M.; Woo, T. K. J. Am. Chem. Soc. 2002, 124,
13306.
(68) Wang, X.; Andrews, L.; Kushto, G. P. J. Phys. Chem. A 2002, 106, 5809.
(69) Snow, J. T.; Murakami, S.; Masamune, S.; Williams, D. J. Tetrahedron
Lett. 1984, 25, 4191.
(70) Trinquier, G. J. Am. Chem. Soc. 1990, 112, 2130.
(71) Grev, R. S.; Schaefer, H. F., III; Baines, K. M. J. Am. Chem. Soc. 1990,
112, 9458.
(51) Levin, G.; Das, P. K.; Lee, C. L. Organometallics 1988, 7, 1231.
(52) Levin, G.; Das, P. K.; Bilgrien, C.; Lee, C. L. Organometallics 1989, 8,
1206.
(53) Michalczyk, M. J.; Fink, M. J.; De Young, D. J.; Carlson, C. W.; Welsh,
K. M.; West, R.; Michl, J. Silicon, Germanium, Tin and Lead Compounds
1986, 9, 750.
(72) Windus, T. L.; Gordon, M. S. J. Am. Chem. Soc. 1992, 114, 9559.
(73) Jacobsen, H.; Ziegler, T. J. Am. Chem. Soc. 1994, 116, 3667.
(74) Batcheller, S. A.; Tsumuraya, T.; Tempkin, O.; Davis, W. M.; Masamune,
S. J. Am. Chem. Soc. 1990, 112, 9394.
(54) (a) Conlin, R. T.; Netto-Ferreira, J. C.; Zhang, S.; Scaiano, J. C.
Organometallics 1990, 9, 1332. (b) Yamaji, M.; Hamanishi, K.; Takahashi,
T.; Shizuka, H. J. Photochem. Photobiol., A 1994, 81, 1.
(55) Leigh, W. J.; Chan, B. R.; Harrington, C. R.; Gaspar, P. P. Time-resolved
spectroscopic studies of the photochemistry of some diphenylgermylene
precursors. To be submitted.
(75) Schafer, A.; Saak, W.; Weidenbruch, M.; Marsmann, H.; Henkel, G. Chem.
Ber./Recl. 1997, 130, 1733.
(76) Gritsenko, O. V.; Schipper, P. R. T.; Baerends, E. J. Chem. Phys. Lett.
1999, 302, 199.
(56) Kira, M.; Ishida, S.; Iwamoto, T.; Ichinohe, M.; Kabuto, C.; Ignatovich,
L.; Sakurai, H. Chem. Lett. 1999, 263.
(77) Schipper, P. R. T.; Gritsenko, O. V.; van Gisbergen, S. J. A.; Baerends, E.
J. J. Chem. Phys. 2004, 112, 1344.
9
J. AM. CHEM. SOC. VOL. 126, NO. 49, 2004 16111

A R T I C L E S
Leigh et al.
Figure 4. Transient decay/growth profiles recorded by laser flash photolysis of a 5 × 10-5 M solution of 15 in dried, deoxygenated hexane solution at 23
°C recorded at (a) 550 nm (Mes2Ge) and (b) 410 nm (6b). The solid lines represent nonlinear least-squares fits to the second-order kinetic eqs 8 and 9,
respectively. The arrow in part b indicates the initial ∆A value due to 6b formed during the laser pulse.
Table 2. Calculated (TZ2P-PW91-SAOP-ZORA) and Experimental UV/vis Absorption Maxima for Dialkyl- and Diarylgermylenes and Their
Digermene Dimers
R2Ge
R2GedGeR2
R
calcd [nm (eV)]
exptl [nm (eV)]
calcd [nm (eV)]
exptl [nm (eV)]
[C(SiMe3)2CH2-] (20)
462 (2.68)
493 (2.52)
463 (2.68)
514 (2.41)
586 (2.19)
450 (2.76)a
514 (2.41)b
480 (2.58)
500 (2.48)
560 (2.23)
H
347 (3.57)
374 (3.32)
460 (2.70)
467 (2.66)
Me
Ph
Mes
370 (3.35)
440 (2.82)
410 (3.03)
a Reference 56. b Reference 79.
agreement in the position of the long-wavelength absorption
maxima for most compounds. The largest deviation was obtained
for 6b; while the 0.37 eV difference is not unusual in TD-DFT
spectroscopic calculations, the method failed to place the
absorption maximum at shorter wavelengths than that of 6a.
The difficulties commonly encountered in the optimization of
the structures of digermenes are presumably behind the limited
accuracy of the calculated electronic structure and spectroscopic
transitions in this case.
Absolute Rate Constants for Dimerization of Ph2Ge and
Mes2Ge. As mentioned earlier, only qualitative agreement is
observed between the decay kinetics of the three germylenes
and the growth kinetics of the corresponding digermenes, with
the decay of the germylene appearing to lag behind the
formation of the digermene by a factor of ca. 2 in each case.
Furthermore, the yield of 6b via dimerization of Mes2Ge appears
to be significantly lower from 4b than that observed previously
using 15 as the precursor, although the quantum yield of Mes2-
Ge appears to be higher.25 We thus decided to reinvestigate the
flash photolysis behavior of 15 in more detail, to ensure that a
meaningful comparison of results from the two precursors would
be possible. Photolysis of this compound affords Mes2Ge and
6b in equal initial amounts (eq 7), a feature which ultimately
allows the dimerization kinetics of Mes2Ge to be fully charac-
terized, given a value of ꢀ6b,410 ) 20 000 for the extinction
coefficient of 6b at its long-wavelength absorption maximum.78
It also allows the yield of 6b via dimerization of Mes2Ge to be
determined with reasonable precision directly from the growth
profile of the digermene.
using 15 as the precursor are shown in Figure 4. The two sets
of data exhibit excellent (nonlinear least squares) fits to second-
order kinetics (equations 8 and 9), affording rate coefficients
() 2kdim[Mes2Ge]0) of (5.0 ( 0.6) × 104 s-1 (Figure 4a) and
(6.1 ( 0.2) × 104 s-1 (Figure 4b). The yield of 6b formed via
dimerization of Mes2Ge from this precursor is given by twice
the ratio of the best-fit values of (∆A6b,∞ - ∆A6b,0) and ∆A6b,0
;
the results of 10 separate measurements afford an average value
of 93 ( 3%. Similar averaging of the initial ∆A0 values for the
germylene and digermene absorptions (∆AMes2Ge,0 and ∆A6b,0
,
respectively, recorded under conditions of constant laser inten-
sity) leads to a value of ꢀ6b,410/ꢀMes2Ge,550 ) 13.9 ( 0.1 for the
ratio of the extinction coefficients due to 6b and Mes2Ge at
their respective approximate absorption maxima. This yields a
value of ꢀMes2Ge,550 ) 1440 ( 400 for the extinction coefficient
of the germylene absorption at 550 nm, assuming an error of
ca. 20% in the measured ꢀ410 value for 6b,78 from which can
be calculated an estimate of kdim ) (5.4 ( 1.5) × 109 M-1 s-1
for the absolute rate constant for dimerization of Mes2Ge in
hexane solution at 23 °C. The average kdim value obtained from
the digermene data is ca. 15% larger than this, but the difference
is well within the estimated error limits of the two determina-
tions. The value of kdim reported here for Mes2Ge is quite similar
to that reported by Conlin and co-workers for Mes2Si under
similar conditions (kdim ) 8.5 × 109 M-1 s-1 in cyclohexane at
25 °C).54
The results obtained with this compound in the present work
are in excellent agreement with those reported previously;25
examples of typical decay/growth profiles for Mes2Ge and 6b
∆AAr2Ge,t ) ∆AAr2Ge,0/(1 + 2kdim[Ar2Ge]0t)
(8)
∆A6,t ) ∆A6,0 + ∆A6,∞{1 - 1/(1 + 2kdim[Ar2Ge]0t)} (9)
(78) Baines, K. M. Private communication.
9
16112 J. AM. CHEM. SOC. VOL. 126, NO. 49, 2004

Transient Germylenes and Digermenes in Solution
A R T I C L E S
of 6b formed during the laser pulse (∆A6b,0 in eq 9), to facilitate
comparison of the digermene growth profiles obtained from the
two precursors. The data of Figure 5a and b have also been
normalized against the initial ∆AMes2Ge,0 values for the germylene
transient absorptions at 550 nm for the two precursors, and then
the plots have been scaled according to the predicted (100%)
theoretical yield of digermene, to give an indication of the
difference in the apparent yields of 6b (as defined by the
maximum ∆A6b value in the growth profile) via germylene
dimerization from 15 and 4b. To help visualize the effect that
the decay of 6b has on its apparent yield in experiments with
4b as precursor, kinetic simulations were carried out to model
the time evolution of the digermene absorption, using the value
of [Mes2Ge]0 defined by the germylene decay traces in Figure
5a and b and the extinction coefficients and kdim value
determined above, and working in the decay kinetics of the
digermene signals determined over longer time scales from the
two precursors. The simulated growth curve with no decay of
the digermene included reproduced the data of Figure 5a quite
closely, although (as it should) ∆A6b,∞ levels off at a value
corresponding to 100% yield, rather than the ca. 93% value
exhibited by the actual data. The results of the simulation for
4b, with the decay kinetics of the digermene included, are shown
as the dashed curve in Figure 5b. They suggest that, with 4b as
precursor, there are one or more reactions of Mes2Ge that
compete with the formation of 6b via dimerization, hence
reducing the observed yield. The behavior exhibited using 15
as precursor allows us to rule out reaction with 6b or solvent
impurities, leaving reaction with something specific to the
precursor, or its photochemistry as the most likely source of
the competing process(es). Possible candidates are reaction of
Mes2Ge with 4b, with a residual impurity from its synthesis,
or with the photolysis coproduct, 2,3-dimethyl-1,3-butadiene
(DMB). The reaction with 4b can be ruled out on the basis of
an experiment in which 4c (as a model for the expected reactive
site in 4b, but transparent at the excitation wavelength) was
added to solutions of 15; the addition of up to 7.4 mM 4c had
no discernible effect on either the decay kinetics of Mes2Ge or
the yield or growth kinetics of 6b, allowing an upper limit of k
< 5 × 105 M-1 s-1 to be estimated for the rate constant for
reaction of Mes2Ge with precursor 4b. Reaction of Mes2Ge with
the diene coproduct (DMB) also seems an unlikely possibility,
but cannot be completely ruled out.35 We are also unable to
rule out the remaining possibility of impurity quenching; our
sample of 4b contains ca. 3% of an unknown impurity that we
have been unable to remove. Given the nominal concentration
of 4b employed in our flash photolysis experiments (0.004 M),
the impurity would be present at a concentration of ca. 10-4
M, roughly 10 times greater than the concentration of DMB
produced in the laser pulse. Simulations indicate that, at this
concentration, it would need to react with Mes2Ge with a rate
constant of ca. 3 × 107 M-1 s-1 to reproduce the experimental
growth/decay curve for 6b with reasonable precision.
Figure 5. Transient decay/growth profiles recorded by laser flash photolysis
of dried, deoxygenated hexane solutions of (a) 15 (7.0 × 10-5 M), (b) 4b
(0.0039 M), and (c) 4a (0.0030 M) at the monitoring wavelengths indicated.
All three solutions were optically matched at the excitation wavelength (248
nm). The 410 nm growth trace in part a is corrected for the initial ∆A
value (∆A0) due to 6b formed during the laser pulse (e.g., see Figure 4b).
The plots of parts a and b were normalized against the ∆A0 values of the
germylene absorptions at 550 nm, to accentuate the differences in the
dimerization yield of 6b (right-hand axes) from the two precursors. The
dashed gray line in part b is the simulated growth/decay curve for 6b,
assuming that Mes2Ge undergoes exclusive dimerization and employing
the extinction coefficients and rate constants given in the text for
dimerization of Mes2Ge and the subsequent decay of 6b.
Figure 5c shows decay and growth traces for Ph2Ge and 6a,
recorded over a 4-fold shorter time scale than the data for the
mesityl system and at wavelengths to the red and blue of the
absorption maxima due to the germylene and digermene,
respectively, to minimize overlap in the decay/growth profiles.
The decay trace shown for Ph2Ge at 530 nm is expanded in
Figure 6; it fits acceptably to second-order kinetics (equation
Figure 5 shows transient decay/growth profiles recorded by
flash photolysis of optically matched (at 248 nm) solutions of
15 (7 × 10-5 M), 4b (0.0039 M), and 4a (0.0030 M) in dry,
deoxygenated hexane under conditions of approximately con-
stant laser intensity; the ∆A6b data for 15 have been adjusted
by subtracting the initial ∆A-value at 410 nm due to the portion
9
J. AM. CHEM. SOC. VOL. 126, NO. 49, 2004 16113

A R T I C L E S
Leigh et al.
tetramesityldigermene (6b) with an absolute rate constant of
ca. 6 × 109 M-1 s-1. This is distinctly slower than the
dimerization of Ph2Ge, as might be expected, but the value
indicates that steric effects on the rate constant for dimerization
due to the mesityl substituents are relatively small. The
somewhat lower reactivity of Mes2Ge compared to Ph2Ge is
also reflected in the absolute rate constants for reaction of the
two species with n-butylamine and acetic acid, both of which
are potent germylene scavengers, reacting with absolute rate
constants that are within a factor of 20 of the diffusion controlled
rate in solution.
Analogous results to those for 4a are also observed upon 193
nm laser flash photolysis of 1,1,3,4-tetramethylgermacyclopent-
3-ene (4c) in hexane solution, affording a definitive UV/vis
spectrum of Me2Ge (λmax ) 480 nm) in solution at ambient
temperatures for the first time. The spectrum agrees well with
that reported for Me2Ge in the gas phase, but with only one of
the several condensed phase spectra that have been assigned to
this species previously. The transient dimerizes to yield the
previously characterized tetramethyldigermene (6c, λmax ) 370
nm), which then decays over several hundred microseconds.
More complete characterization of Me2Ge and 6c must await
the development of a more efficient photoprecursor that absorbs
at longer wavelengths in the UV, so as to avoid the problems
associated with far-UV excitation in solution. Work of this type
is in progress in our laboratory.
Figure 6. Transient decay trace recorded by laser flash photolysis of a
0.003 M solution of 4a in dry, deoxygenated hexane at 23 °C. The solid
line represents the best fit of the data to eq 8.
8), affording a rate coefficient ()2kdim[Ph2Ge]0) of (3.1 ( 0.1)
× 105 s-1. This leads to a value of kdim ) (1.1 ( 0.2) × 1010
M-1s-1, using a value of ꢀPh2Ge,530 ) 990 M-1cm-1 (a factor of
0.60 of the ꢀPh2Ge,500 value determined earlier) for the calculation
of [Ph2Ge]0 from the initial ∆A-value at 530 nm. Comparison
of the growth/decay data for 6a to the results of kinetic
simulations is not warranted in this case because of the lack of
reliable extinction coefficient data for the digermene, as well
as the fact that Ph2Ge exhibits significantly higher reactivity
than Mes2Ge toward dienes,35 and may also react with 6a as it
is formed. Nevertheless, it is clear from the data obtained for
Mes2Ge/6b that the maximum concentration of 6a achieved from
dimerization of Ph2Ge (as measured by the maximum in the
growth/decay curve) must be at least a factor of 2 lower than
half the initial germylene concentration. This leads to the
conclusion that the true extinction coefficient of 6a at its
absorption maximum must be at least 2 times greater than the
value of ca. 6000 M-1 cm-1 estimated earlier by benzophenone
actinometry.
The UV/vis spectra of all three germylenes studied in this
work, as well as those of digermenes 6a and 6c and the known
stable germylene 20, agree very closely with the results of time-
dependent DFT calculations on the molecules, and also with
the reported spectra of the corresponding silicon analogues.
Close attention has been paid to the identification of the most
appropriate basis sets used in the geometry optimizations and
spectral calculations, not only to maximize the reliability of the
results but also to ensure that the methods employed will be
readily extendable to other di- and trivalent Group 14 reactive
intermediates that are under investigation in our laboratory.
The present work demonstrates that the successful detection
of reactive germylenes in solution depends on several key
factors. The most obvious of these is the photochemistry of the
precursor(s) employed for generation of the species of interest,
which for optimum results should be highly efficient and, more
importantly, devoid of other competing photoreactions that
produce (even minor amounts of) other reactive intermediates
such as germanium-centered radicals or germenes. Such species,
particularly arylated derivatives, absorb much more strongly in
the UV/visible spectrum than do germylenes, whose long
wavelength absorption bands are due to an orbitally forbidden
(n,p) electronic transition and hence possess extinction coef-
ficients on the order of only 103. Another, perhaps less obvious
factor is the experimental conditions employed for the experi-
ment; in particular, the use of scrupulously dry solvents and
glassware is critical for the successful observation of reactive
germylenes in solution. This is because scavenging of the species
by water is rapid and reVersible, which reduces the concentration
of free germylene present in solution.35 At the concentrations
of water present in commercial, HPLC-grade hydrocarbon
solvents, this results in much weaker transient signals than those
observed under anhydrous conditions, for both the germylene
and its digermene dimer, but has little effect on the apparent
Summary and Conclusions
1,1-Diphenylgermylene (Ph2Ge) and its dimer, tetraphenyl-
digermene (6a), have been detected and characterized in solution
for the first time by laser flash photolysis methods. The
germylene exhibits a weak long-wavelength UV/vis absorption
at λmax ) 500 nm (ꢀmax ) 1650 ( 390 M-1 cm-1) and dimerizes
with an absolute rate constant of ca. 1 × 1010 M-1 s-1 in hexane
solution at ambient temperatures to yield 6a. The latter is a
significantly stronger absorber, exhibiting λmax ) 440 nm (ꢀmax
> 12 000 M-1 cm-1) and decaying with mixed first- and second-
order kinetics and τ ≈ 40 µs. Trapping experiments with
methanol, acetic acid, isoprene, and triethylsilane afford the
expected products of reaction of the germylene with the trapping
reagent in essentially quantitative yield using conventional UV
lamps as the excitation source. The product of 1,2-addition of
MeOH to 6a has been detected as well upon high-intensity laser
photolysis of 4a, under conditions where the germylene is
produced in sufficiently high concentrations to allow dimeriza-
tion of the species to compete with trapping by the alcohol.
Analogous results are obtained upon laser flash photolysis
of the 1,1-dimesityl analogue 4b, which yields the previously
characterized Mes2Ge. The latter dimerizes to yield the known
9
16114 J. AM. CHEM. SOC. VOL. 126, NO. 49, 2004

Transient Germylenes and Digermenes in Solution
A R T I C L E S
nitrogen just prior to assembling the sample-handling system at the
beginning of an experiment. Solution temperatures were measured with
a Teflon-coated copper/constantan thermocouple inserted directly into
the flow cell. Transient decay and growth rate constants were calculated
by nonlinear least-squares analysis of the absorbance-time profiles using
the Prism 3.0 software package (GraphPad Software, Inc.) and the
appropriate user-defined fitting equations, after importing the raw data
from the Luzchem mLFP software. Reagents were added directly to
the reservoir by microliter syringe as aliquots of standard solutions.
Rate constants were calculated by linear least-squares analysis of decay
rate-concentration data (5-7 points) that spanned as large a range in
transient decay rate as possible. Errors are quoted as twice the standard
deviation obtained from the least-squares analyses.
lifetimes of the species. The use of flow techniques is also
essential in methods requiring extensive signal averaging, as
those employed here do. This results from the fact that the main
fate of these species in the absence of scavengers is oligomer-
ization, which yields products that absorb in the same range as
the precursor and produce transient species of their own upon
excitation. Finally, as our parallel study of the reactivity of Ph2-
Ge shows in some detail,35 simple diarylgermylenes are
extremely potent electrophiles and react rapidly with even
relatively weak nucleophiles, often to form Lewis acid-base
complexes or other intermediates whose absorption spectra are
blue-shifted relative to those of the free germylenes. We believe
that factors of this type may well explain some of the
discrepancies between the UV spectra reported here for Ph2Ge
and Me2Ge and those reported previously, particularly those in
low-temperature matrixes where diffusion of the free, photo-
chemically generated germylene away from a weakly nucleo-
philic coproduct is inhibited.
The quantum yield for formation of methoxydiphenylgermane (9a)
was determined by merry-go-round photolysis of deoxygenated, opti-
cally matched solutions of 4a (0.037 M) and 1,1-diphenylsilacyclobu-
tane (13; 0.043 M) in methanolic (0.5 M) cyclohexane-d12 solution,
using a Rayonet photochemical reactor fitted with four RPR-2537 low-
pressure Hg lamps. The photolyses were monitored as a function of
time between 0 and ca. 25% conversion of 4a by 600 MHz NMR
spectroscopy, using the peak due to residual C6D11H as internal
integration standard. The quantum yield was calculated from the relative
slopes of concentration vs time plots for the formation of 9a from 4a
and 14 from 13 (see Supporting Information) and the reported quantum
yield for the latter reaction (Φ14 ) 0.21 ( 0.0240). The value reported
for the formation of 9a from 4a (Φ9a ) 0.62 ( 0.08) is the average of
triplicate determinations.
Future work will explore the versatility of the germacyclo-
pent-3-enyl system for the generation and study of other aryl-
and alkylgermylene and digermene derivatives by laser flash
photolysis methods.
Experimental Section
The procedures employed for the synthesis of 4a-c, authentic
samples of photoproducts, and semipreparative photolysis experiments,
as well as detailed spectroscopic data for the compounds reported in
this work, are described in the Supporting Information. Hexamesityl-
cyclotrigermane was prepared according to the literature procedure80
and recrystallized several times from hexanes.
Hexanes (EMD Omnisolv) was dried by refluxing for several days
under argon over sodium/potassium amalgam followed by distillation.
For 193 nm flash photolysis experiments, this procedure was preceded
by repeated washing of the solvent with concentrated sulfuric acid,
followed by distilled water and preliminary drying over anhydrous
sodium sulfate. Deoxygenated samples of hexane used for 193 nm work
exhibited an absorbance of no more than 0.3 at 193 nm in a 7 × 7
mm2 quartz cell. n-Butylamine (Sigma-Aldrich) was refluxed for 12 h
over potassium hydroxide and distilled. Glacial acetic acid was used
as received from Sigma-Aldrich.
Laser flash photolysis experiments employed the pulses from a
Lambda Physik Compex 120 excimer laser, filled with F2/Kr/Ne (248
nm; ∼25 ns; 100 ( 5 mJ) or F2/Ar/Ne (193 nm; 20-25 ns; ca. 50 mJ)
mixtures, and a Luzchem Research mLFP-111 laser flash photolysis
system. The latter has been modified to incorporate an external 150 W
high-pressure Xe lamp as monitoring source, powered by a Kratos LPS-
251HR power supply, which provides a dynamic range in time of 106
(1 µs to 1 s full scale). A Pyrex filter inserted between the sample cell
and monochromator protects the PMT from scattered light from the
laser when the monitoring wavelength is g330 nm; the excitation beam
is delivered at a 90° angle relative to the monitoring beam through a
5 mm diameter circular slit, which defines the path length over which
transients are produced. Solutions were prepared at concentrations such
that the absorbance at the excitation wavelength (248 or 193 nm) was
between ca. 0.7 and 0.9 and were flowed continuously through a
thermostated 7 × 7 mm2 Suprasil flow cell connected to a calibrated
100 mL reservoir, fitted with a glass frit to allow bubbling of nitrogen
or argon gas through the solution for at least 30 min prior to and then
throughout the duration of each experiment. The glassware, sample
cells, and transfer lines used for these experiments were stored in a 65
°C vacuum oven when not in use, and the oven was vented with dry
Determination of the extinction coefficient for the Ph2Ge absorption
at 500 nm was carried out by benzophenone actinometry, using the
so-called intensity variation method.47,48 The substrate and actinometer
solutions were optically matched at the laser wavelength (248 nm) to
ensure equal light absorption, and both were flowed through the same
sample cell for sequential measurement of ∆Aλ,0 (λ ) 500 and 525 nm
for Ph2Ge and the benzophenone triplet, respectively) values as a
function of laser intensity, which was controlled using neutral density
filters constructed from wire screens. Both solutions were deoxygenated
prior to the measurements with a stream of nitrogen and restored to
their original volumes at the end of the deoxygenation procedure with
fresh (deoxygenated) hexane. The resulting plots of ∆Aλ,0 vs laser
intensity, from whose slopes was calculated the extinction coefficient
of the Ph2Ge absorption using values of Φ ) 1.0 and ꢀ525nm ) 6250 (
1250 M-1 cm-1 for the benzophenone triplet49 and the quantum yield
for Ph2Ge formation determined above, are included in the Supporting
Information. The value obtained, ꢀ500-nm ) 1650 ( 390 M-1 cm-1, is
the average of duplicate determinations.
Computational Details. A. Molecular Structures. Calculations
were performed with the ADF 2003.01 density functional theory
package (SCM).81-83 Unless otherwise indicated, the calculation of
model geometries was gradient-corrected with the exchange and
correlation functionals of Perdew and Wang (PW91);84 all basis
functions were of triple-ú quality and were composed of uncontracted
Slater-type orbitals (STOs), including all core electrons and two
auxiliary basis set of STOs for polarization.
(81) te Velde, G.; Bickelhaupt, F. M.; van Gisbergen, S. J. A.; Fonseca Guerra,
C.; Baerends, E. J.; Snijders, J. G.; Ziegler, T. J. Comput. Chem. 2001, 22,
931.
(82) Fonseca Guerra, C.; Snijders, J. G.; te Velde, G.; Baerends, E. J. Theor.
Chem. Acc. 1998, 99, 391.
(83) Baerends, E. J.; Autschbach, J.; Berces, A.; Bo, C.; Boerrigter, P. M.;
Cavallo, L.; Chong, D. P.; Deng, L.; Dickson, R. M.; Ellis, D. E.; Fan, L.;
Fischer, T. H.; Fonseca Guerra, C.; van Gisbergen, S. J. A.; Groeneveld,
J. A.; Gritsenko, O. V.; Gruning, M.; Harris, F. E.; van den Hoek, P.;
Jacobsen, H.; van Kessel, G.; Kootstra, F.; van Lenthe, E.; Osinga, V. P.;
Patchkovskii, S.; Philipsen, P. H. T.; Post, D.; Pye, C. C.; Ravenek, W.;
Ros, P.; Schipper, P. R. T.; Schreckenbach, G.; Snijders, J. G.; Sola, M.;
Swart, M.; Swerhone, D.; te Velde, G.; Vernooijs, P.; Versluis, L.; Visser,
O.; van Wezenbeek, E.; Wiesenekker, G.; Wolff, S. K.; Woo, T. K.; Ziegler,
T. ADF2003.01. Amsterdam, SCM.
(79) Saito, K.; Obi, K. Chem. Phys. 1994, 187, 381.
(80) Ando, W.; Tsumuraya, T. J. Chem. Soc., Chem. Commun. 1987, 1514.
(84) Perdew, J. P. Phys. ReV. B 1992, 46, 6671.
9
J. AM. CHEM. SOC. VOL. 126, NO. 49, 2004 16115

A R T I C L E S
Leigh et al.
B. Spectroscopic Calculations. The method is based on the time-
dependent extension of density functional theory (TD-DFT) imple-
mented85-89 in the ADF package. The adiabatic local density ap-
proximation (ALDA) was used for the exchange-correlation kernel90,91
and the differentiated static LDA expression was used with the Vosko-
Wilk-Nusair parametrization.92 For the exchange-correlation potentials
in the zeroth-order KS equations, the excitation energies and oscillator
strengths were obtained using the Davidson iterative diagonalization
method.
Kinetic simulations were carried out using KinTekSim version 3.0.3
(KinTek Corp.).
Acknowledgment. We thank the Natural Sciences and
Engineering Research Council of Canada, the Canada Founda-
tion for Innovation, and the Ontario Innovation Trust for
financial support; the McMaster Regional Centre for Mass
Spectrometry for mass spectrometric analyses; and Professor
M. Kira for providing the original spectroscopic data for
germylene 20 and its silicon and tin congeners.
(85) Gross, E. K. U.; Kohn, W. AdV. Quantum Chem. 1990, 21, 255.
(86) Gross, E. K. U.; Ullrich, C. A.; Gossmann, U. J. Density functional theory
of time-dependent systems; Plenum: New York, 1995; p 149.
(87) Gross, E. K. U.; Dobson, J. F.; Petersilka, M. Density Functional Theory;
Springer: Heidelberg, 1996.
Supporting Information Available: Details of the synthesis
and characterization of compounds and photoproducts, quantum
yield and transient extinction coefficient data for Ph2Ge, and
computational details for the germylenes and digermenes studied
in this work. This material is available free of charge via the
Internet at http://pubs.acs.org.
(88) van Gisbergen, S. J. A.; Snijders, J. G.; Baerends, E. J. J. Chem. Phys.
1995, 103, 9347.
(89) van Gisbergen, S. J. A.; Snijders, J. G.; Baerends, E. J. Comput. Phys.
Commun. 1999, 118, 119.
(90) van Gisbergen, S. J. A.; Snijders, J. G.; Baerends, E. J. Phys. ReV. Lett.
1997, 78, 3097.
(91) van Gisbergen, S. J. A.; Snijders, J. G.; Baerends, E. J. J. Chem. Phys.
1998, 109, 10644.
(92) Vosko, S. H.; Wilk, L.; Nusair, M. Can. J. Phys. 1980, 58, 1200.
JA046308Y
9
16116 J. AM. CHEM. SOC. VOL. 126, NO. 49, 2004