Time-resolved spectroscopic studies of the photochemistry of some diphenylgermylene (Ph2Ge:) precursors

1324

Time-resolved spectroscopic studies of the

photochemistry of some diphenylgermylene

(Ph₂Ge:) precursors¹

Cameron R. Harrington, William J. Leigh, Bryan K. Chan, Peter P. Gaspar, and

Dong Zhou

Abstract: The photochemistry of diphenylbis(trimethylsilyl)germane (2a) and 1,4-dihydro-5-methyl-1,2,3,4,9,9-

hexaphenyl-1,4-germanonaphthalene (11) has been studied in solution by steady-state and laser flash photolysis meth-

ods with a view to detecting the transient germylene derivative diphenylgermylene (Ph₂Ge), which has previously been

shown to be the major product of photolysis of 2a and a closely related derivative of 11. Steady-state trapping experi-

ments confirm the formation of Ph₂Ge as the major germanium containing primary product in both cases; with 2a, the

results indicate that other transient species are also formed in minor yields, including phenyl(trimethylsilyl)germylene

(Ph(TMS)Ge, ca. 6%) and diphenyl(trimethylsilyl)germyl radicals (Ph₂(TMS)Ge, ≥15%). Laser flash photolysis of 2a in

deoxygenated hexane solution yields a complex mixture of overlapping transient absorptions, which is shown to be

comprised of Ph₂Ge, tetraphenyldigermene (15) and its oligomerization products, and another species with spectral

characteristics similar to the Ph₂(TMS)Ge radical. The latter has been independently generated by hydrogen abstraction

from diphenyl(trimethylsilyl)germane by tert-butoxyl radicals. Compound 11 extrudes Ph₂Ge more cleanly and effi-

ciently upon photolysis in solution, yet laser flash photolysis affords excited triplet and triplet-derived species as the

only detectable transient products; interpretation of the results for this compound is made difficult by its slow thermal

decomposition to 5-methyl-1,2,3,4-tetraphenylnaphthalene. It is concluded that in spite of the fact that both 2a and 11

afford Ph₂Ge in high yield upon photolysis, they are poor precursors for study of the species in solution by time-

resolved UV–vis methods, owing to the formation of other, more strongly absorbing transient products than Ph₂Ge,

whose lowest energy absorption is characteristically weak.

Key words: germylene, germyl radical, flash photolysis, disilylgermane, photochemistry. 1338

Résumé : Afin d’essayer de détecter le dérivé germylène transitoire, diméthylgermylène (Ph₂Ge), qui avait été reconnu

antérieurement comme le produit principal de la photolyse du produit 2a et de son dérivé apparenté 11, on a étudié la

photochimie du diphénylbis(triméthylsilyl)germane (2a) et du 1,4-dihydro-5-méthyl-1,2,3,4,9,9-hexaphényl-1,4-

germanonaphtalène (11) en solution, par les méthodes de l’état stationnaire et de la photolyse éclair au laser. Les expé-

riences de piégeage de l’état stationnaire confirment dans les deux cas la formation du Ph₂Ge comme produit primaire

principal contenant du germanium; avec le produit 2a, les résultats indiquent que d’autres espèces transitoires se for-

ment aussi, avec des rendements mineurs, dont les radicaux phényl(triméthylsilyl)germylène (Ph(TMS)Ge; rendement

d’environ 6 %) et diphényl(triméthylsilyl)germyle (Ph₂(TMS)Ge; rendement supérieur à 15 %). La photolyse éclair au

laser du composé 2a en solution d’hexane désoxygéné mène à la formation d’un mélange complexe d’absorptions tran-

sitoires qui se superposent; il a été démontré qu’elles comprennent le Ph₂Ge, le tétraphényldigermène (15) et ses pro-

duits d’oligomérisation ainsi qu’une autre espèce de caractéristiques spectrales similaires au radical Ph₂(TMS)Ge. Ce

dernier a été généré de manière indépendante par abstraction d’hydrogène par des radicaux tert-butoxyles à partir du

diphényl(triméthylsilyl)germane. Par photolyse en solution, le composé 11 expulse le Ph₂Ge d’une façon beaucoup plus

propre et plus efficace, toutefois la photolyse éclair au laser ne fournit comme produits transitoires détectables que le

triplet excité et diverses espèces qui en dérivent; l’interprétation des résultats relatifs à ce composé est rendue difficile

par sa décomposition thermique lente en 5-méthyl-1,2,3,4-tétraphénylnaphtalène. On en conclut que, même si la photo-

lyse des composés 2a et 11 mène dans les deux cas à la formation de Ph₂Ge avec un rendement élevé, ils sont de pau-

vres précurseurs pour l’étude de ces espèces en solution par des méthodes UV–vis résolues dans le temps en raison de

la formation d’autres produits transitoires qui absorbent plus fortement que le Ph₂Ge dont la plus faible énergie

d’absorption est normalement faible.

Received 9 March 2005. Published on the NRC Research Press Web site at http://canjchem.nrc.ca on 22 October 2005.

C.R. Harrington, W.J. Leigh,²B.K. Chan, P.P. Gaspar, and D. Zhou. Department of Chemistry, McMaster University,

1280 Main Street West, Hamilton, ON L8S 4M1, Canada and Department of Chemistry, Washington University, St. Louis,

MO 63130, USA.

¹This article is part of a Special Issue dedicated to organic reaction mechanisms.

²Corresponding author: (e-mail: leigh@mcmaster.ca).

Can. J. Chem. 83: 1324–1338 (2005)

doi: 10.1139/V05-148

Harrington et al.

1325

Mots clés : germylène, radical germyle, photolyse éclair, disilylgermane, photochimie.

[Traduit par la Rédaction] Harrington et al.

Introduction

matrices at 77 K (1, 16, 17) — match the one reported more

recently in solution using 5 as the precursor.

Since the report nearly 20 years ago by Ando et al. (1) of

the UV spectra of several simple, photochemically generated

germylene derivatives in hydrocarbon matrixes at low tem-

perature, there have been a number of studies published on

the direct detection and study of reactive diarylgermylenes

(2–10) and dialkylgermylenes (6, 11–15) in solution by laser

flash photolysis techniques. Most of these studies employed

Group 14 catenates as photochemical precursors to these

“heavy carbene” analogues, such as 1–3 in the case of

diphenylgermylene (Ph₂Ge). While steady-state trapping ex-

periments showed that the germylene is the major transient

photoproduct in each case, the results of laser photolysis ex-

periments have been highly variable. For example, the UV–

vis absorption max of Ph₂Ge has been reported to occur at

various positions ranging from 445 to 630 nm, bracketing

the value of λ_max= 466 nm originally reported by Ando et

al. (1) in a 3-methylpentane matrix at 77 K using benzo-

germanorbornadiene derivative 4 and disilylgermane 2a as

germylene precursors. Thanks to the recent identification of

a precursor that yields this reactive species quantitatively

and in remarkably high quantum efficiency (5, eq. [1]), the

UV spectrum of Ph₂Ge in hydrocarbon solvents at room

temperature has now been conclusively established (9), as

have quantitative details of its reactivity toward dimerization

(9) and a number of representative metallylene scavengers

such as amines, carboxylic acids, halocarbons, Group 14 hy-

drides, dienes, alkenes, and alkynes (10). In hexane solution

One of the compounds employed by Ando et al. for the

photogeneration of Ph₂Ge — diphenylbis(trimethylsilyl)ger-

mane (2a) — has also been studied in fluid solution at room

temperature by flash photolysis methods (2, 6). Photolysis of

the compound in cyclohexane solution in the presence of

2,3-dimethyl-1,3-butadiene (DMB) was found to afford

hexamethyldisilane (Si₂Me₆), trimethylphenylsilane (PhSiMe₃),

and the diene trapping product 5 as the only detectable prod-

ucts (eq. [2]) (6), while laser flash photolysis of the com-

pound afforded a transient that decayed over several hundred

microseconds and exhibited λ_max= 445 nm (2), blue-shifted

somewhat relative to the spectrum recorded for this com-

pound in the earlier matrix experiments (1, 16). Absolute

rate constants were determined for reaction of the species

with various germylene scavengers, among them oxygen

(1.02 × 10⁸(mol/L)^–1s^–1), ethyldimethylsilane (k = 1.01 ×

10⁴(mol/L)^–1s^–1), 1-bromo-2-methylpropane (k = 6.36 ×

10⁵(mol/L)^–1s^–1), and 2,3-dimethyl-1,3-butadiene (2.75 ×

10⁴(mol/L)^–1s^–1) (2). Like the UV–vis spectrum, the reac-

tivity of the transient toward the various germylene scaven-

gers examined is in poor agreement with that reported much

later for Ph₂Ge using 5 as the precursor (9). It is noteworthy

that photolysis of the dimesityl analogue of 2a (6) also af-

fords the corresponding germylene (in this case dimesityl-

germylene, Mes₂Ge) quite cleanly, as shown by the results

of steady-state trapping experiments, matrix photolysis ex-

periments (1, 16), and laser flash photolysis experiments in

solution (7). In this case, the transient spectrum observed

(and assigned to Mes₂Ge, λ_max~ 550 nm) in solution (7)

matches that obtained in matrix experiments (1, 16) or in so-

lution from other precursors (7, 9) quite closely.

at room temperature, diphenylgermylene exhibits λ_max

=

500 nm (⑀ = 1650 390 (mol/L)^–1cm^–1) and dimerizes at

close to the diffusion-controlled rate to yield the transient

1,1,2,2-tetraphenyldigermene (Ph₂Ge=GePh₂, λ_max= 440 nm,

τ ~ 40 µs) (9). None of the UV spectra previously assigned

to Ph₂Ge — including that reported by Ando et al. in 3-MP

hν

[2]

Ph₂Ge(SiMe₃)₂

+

GePh₂

Hexane

2a

hν

5 (61%)

0.043 mol/L

[1]

+

[Ph₂Ge:]

GePh₂

Hexane

+ Si₂Me₆(68%)

λ_max= 500 nm

5

+ PhSiM e₃(27%)

τ ~ 2 µs

Mes₂Ge(SiMe₃)₂

Ph₂Ge MMe₃

Ph₃Ge GePh₃

6

MMe₃

(Mes = 2,4,6-trimethylphenyl)

1

2

a. M = Si

b. M = Ge

Recently, we reported a study of the photochemistry of

1,1,1-trimethyl-2,2,2-triphenylsilagermane (7) in solution by

steady-state and laser flash photolysis methods (18). This

compound also yields Ph₂Ge as the major germanium

containing photoproduct in solution, but it also affords

significant yields of the conjugated germene derived from

1,3-trimethylsilyl migration (8) and free radicals owing to

triplet-derived Ge—Si bond homolysis (eq. [3]), as revealed

by the results of steady-state trapping experiments. Only 8

GePh₂

Ph

Ph₂Ge GePh₂

Ph

Ph₂Ge (GePh₂)n

Ph

3

(n = 1-3)

4

1326

Can. J. Chem. Vol. 83, 2005

and triphenylgermyl radicals could be detected in laser flash

photolysis experiments with the molecule, presumably be-

cause the absorption spectra of these transients are signifi-

cantly more intense than that of Ph₂Ge, whose lowest energy

(n,p) electronic transition is characteristically weak. Further-

more, the lowest energy absorption band in the UV–vis

spectrum of 8 (λ_max~ 475 nm) occurs at a similar position

to the much weaker absorption band in the spectrum of the

germylene (λ_max= 500 nm) under similar conditions. We

speculated that an analogous mixture of products might well

be expected from photolysis of 2a under similar conditions,

and that the transient spectroscopic behavior reported earlier

for this compound might thus be better explained in terms of

the corresponding germene (9) and (or) the diphenyl(tri-

methylsilyl)germyl radical (10) (18).

Results and discussion

The photochemistry of Ph₂Ge(SiMe₃)₂(2a)

The products of photolysis of 2a in the presence of DMB

(see eq. [2]) were originally identified, and their yields es-

tablished, from a preparative-scale photolysis experiment

carried to 64% conversion (6). Because secondary photolysis

of 5 can be expected to lead to its preferential loss from the

reaction mixture over such extended photolysis times, we

began the present study with an examination of the product

mixture at much lower conversions. Thus, a deoxygenated

0.05 mol/L solution of 2a in cyclohexane-d₁₂containing

0.1 mol/L of the diene was irradiated with 254 nm light,

1

monitoring the reaction periodically by (600 MHz) H NMR

spectroscopy over the 0%–16% conversion range. The NMR

spectra of the photolysate showed discrete evidence for the

formation of 5 and hexamethyldisilane (Si₂Me₆) in a ca. 1:1

ratio as the only detectable products. GC analysis of a simi-

lar photolysis in deoxygenated hexane solution (0.01 mol/L

2a, 0.03 mol/L DMB) also showed 5 to be the major product

of the reaction, but allowed the additional detection of

PhSiMe₃in minor amounts (≤6%), along with another minor

product that eluted shortly before 2a. The latter compound,

which is formed in similar yield to PhSiMe₃, was tentatively

identified as 3,4-dimethyl-1-phenyl-1-trimethylsilylgerma-

cyclopent-3-ene (12) on the basis of GC–MS evidence. The

results of these experiments are summarized in eq. [4].

Ph₂Ge:

Ph₃Ge

hν

[3]

Ph₃GeSiMe₃

+

PhSiMe₃

+

Me₃Si

+

Hexane

7

GePh₂

+

H

SiMe₃

8

Ge(Ph)SiMe₃

Ph₂GeSiMe₃

H

SiMe₃

hν

[4] Ph₂Ge(SiMe₃)₂

+

GePh₂

9

10

Hexane

2a

5 (89%)

0.03 mol/L

In the present paper, we report the results of a reinvestiga-

tion of the photochemistry of 2a in solution by steady-state

and laser flash photolysis methods, with a view to providing

a secure identification of the transient species observed in

the earlier work with this compound, and to see whether dis-

crete evidence for the formation of free Ph₂Ge might be ob-

tained. To aid in the interpretation of the results, germyl

radical 10 has been generated by an independent route and

its UV–vis spectrum and lifetime recorded under similar

conditions to those employed for laser photolysis experi-

ments with 2a. We have also synthesized an analogue (11) of

the benzogermanonorbornadiene derivative originally studied

by Ando et al. (4), and have studied its photochemistry using

similar methods. This compound differs from 4 only in the

presence of a methyl group situated on the benzannelated

ring, which was included so as to facilitate the determination

of its purity and the identification of products from photo-

chemical and (or) thermal decomposition, as compounds of

this type are known to exhibit pronounced thermal lability

under quite mild conditions (19–22).

+ Si₂Me₆

(0.94:1 relative to 5)

Ph

Ge

SiMe₃

+ PhSiMe₃

+

(1:1 relative to 12)

12 (6%)

Interestingly, photolysis of a 0.05 mol/L solution of 2a in

cyclohexane-d₁₂containing methanol (MeOH, 0.5 mol/L) af-

forded Si₂Me₆and the Ph₂Ge insertion product 13 as the ma-

jor products, but the latter was formed in only ca. 60% yield

relative to consumed 2a and in significantly lower yield than

the coproduct of Ph₂Ge extrusion, Si₂Me₆. The 0–0.5 ppm

region of the spectra also contained singlets attributable to

three additional minor products, whose resonances inte-

grated to 10:13:7; the total intensity of the three singlets ow-

ing to these products was ca. 30% relative to that of the

singlet owing to Si₂Me₆. The most minor of these products

was identified by ¹H NMR spectroscopy and GC as

disilyldigermane 14, by spiking the mixture with an authen-

tic sample of the compound; the other two were not identi-

fied. The results of these experiments are summarized in

eq. [5].

GePh₂

Ph

Taken together, these results indicate that while extrusion

of Ph₂Ge is the dominant photoreaction of 2a, accounting

for ca. 60% of the photochemistry, other minor processes oc-

cur as well. One of these is assigned to the alternate

germylene extrusion pathway, yielding Ph(TMS)Ge and

Ph

11

CH₃

Harrington et al.

1327

Fig. 1. Transient absorption spectra from laser flash photolysis

hν

H

[5]

MeOH

Ph₂Ge(SiMe₃)₂

+

of a flowed, deoxygenated solution of 2a (1.3 × 10^–4mol/L) in

anhydrous hexane containing (a) 0 mmol/L of AcOH and

(b) 3 mmol/L AcOH, recorded under identical conditions, 0.64–

2.56 µs (᭺), 24–39 µs (ٗ), and 324–325 µs (᭝) after the pulse.

Transient decay traces recorded at a monitoring wavelength of

440 nm are shown as inserts.

Ph₂GeOMe

C₆D₁₂

2a

0.5 mol/L

13 (60%)

Me₃Si

other

products

Ph₂GeGePh₂

SiMe₃

+

31%

14 (8%)

0.025

a

0.020

+ Si₂Me₆(1.17:1 relative to 13)

0.06

0.015

0.010

0.005

PhSiMe₃in about 6% chemical yield. Another is Si—Ge

bond homolysis to yield the trimethylsilyl and di-

phenyl(trimethylsilyl)germyl radicals, as evidenced by the

formation of minor amounts of digermane 14 in the

photolysis in the presence of MeOH. This process consti-

tutes at least 15% of the reactive excited-state decay pro-

cesses of the molecule, based on the observed yield of 14

and the assumption that 50% of the Ph₂(TMS)Ge radicals

produced will decay by recombination with trimethylsilyl

radical, regenerating 2a. Interestingly, 14 is not formed in

the diene scavenging experiment, and the complete absence

of additional SiMe₃resonances in the NMR spectrum of this

photolysis mixture indicates that the reason is not that the

Ph₂(TMS)Ge radical is efficiently scavenged by the diene. A

more likely mechanism by which the diene suppresses the

formation of free radicals is triplet quenching, since it is

known that the formation of radicals in the photolysis of

aryldisilanes, arylgermasilanes, and aryldigermanes derives

largely from the triplet state, and their formation can be

quenched to at least some extent by added diene (18, 23,

24).

Laser flash photolysis experiments with 2a employed

flowed, deoxygenated 1.3 × 10^–4mol/L solutions of the

compound in anhydrous hexane, and used the pulses from a

KrF excimer laser (248 nm, ~25 ns, ~100 mJ) for excitation.

As reported previously (2), laser photolysis of the compound

affords strong transient absorptions throughout the 300–

600 nm region of the spectrum, which decay over several

hundred microseconds with complex kinetics. Figure 1

shows transient spectra recorded at selected time intervals

over a ca. 350 µs time window after excitation in the ab-

sence and presence of 3 mmol/L acetic acid (AcOH). The

spectra in the absence of acid were collected with a view to

reproducing the originally reported conditions as closely as

possible, while those in the presence of acid were intended

to reveal those transient absorptions that remain behind

when Ph₂Ge and its oligomerization products are removed

from the mixture by trapping with AcOH, a potent

germylene scavenger (9, 10). The main (gross) effects of the

carboxylic acid are a marked reduction in the intensities of

the signals observed throughout the spectrum, and simplifi-

cation of the complex transient absorptions that arise below

ca. 340 nm. In the absence of AcOH (Fig. 1a), the spectrum

exhibits a band at λ_max= 300 nm during the earliest time

0.04

0.000

0

100

200

Time (µs)

300

0.02

0.00

300

400

500

600

Wavelength (nm)

0.015

0.010

0.005

0.000

b

0.06

0.04

0.02

0.00

0

100

200

Time (µs)

300

400

500

600

Wavelength (nm)

sient decay profile at this wavelength, which decays with

multiexponential kinetics in the absence of acid (see inset,

Fig. 1a). In the presence of 3 mmol/L AcOH the long wave-

length absorption decays with clean second-order kinetics

(see inset, Fig. 1b) and a rate coefficient of k_decay/ε = (7.4

0.2) × 10⁵cm^–1s^–1(r²= 0.987). The decay kinetics at

310 nm (not shown) are similar to those at 440 nm, suggest-

ing that the two absorption bands in the spectrum of Fig. 1b

belong to the same (as yet unidentified) transient species. It

can thus be concluded that the early (0.64–2.56 µs) spectrum

of Fig. 1a is the sum of the spectra of at least two transient

species of which all but one is eliminated in the presence of

3 mmol/L AcOH.

AcOH is known to react with Ph₂Ge irreversibly, with a

rate constant of 3.9 × 10⁹(mol/L)^–1s^–1(9). Thus, a concen-

tration of 3 mmol/L is sufficient to reduce the lifetime of

Ph₂Ge to <100 ns and more or less completely suppress its

dimerization to tetraphenyldigermene (15), the main mode

of decay of the germylene in the absence of scavengers (see

eq. [6]) (9). In the absence of scavengers, the concentration

window monitored, which evolves into a band with λ_max

~

310 nm after ca. 25 µs; only the latter is evident in the re-

gion below 340 nm in the presence of the carboxylic acid

(Fig. 1b). Addition of the acid also reduces the intensity of

the long wavelength absorption band centered at λ_max

~

445 nm, and markedly reduces the complexity of the tran-

1328

Can. J. Chem. Vol. 83, 2005

Fig. 2. (a) Transient absorption spectra from laser flash photo-

lysis of a flowed, deoxygenated solution of 2a (1.3 × 10^–4mol/L)

in anhydrous hexane, 30–80 ns (᭺) and 1.0–1.2 µs (ٗ) after the

pulse. The dashed line is the difference between the early and

late spectrum. (b) Transient absorption profiles recorded at

440 nm (᭺) and 510 nm (ٗ).

of 15 reaches a maximum in the first ca. 1 µs after the exci-

tation pulse, and then decays over several tens of microsec-

onds (9). We thus attribute that portion of the transient

spectrum of Fig. 1a that is eliminated in the presence of

AcOH to 15 (λ_max= 300, 440 nm (9)), formed by

dimerization of Ph₂Ge, and its oligomerization product(s).

The long wavelength absorption band in the spectrum of 15

is almost exactly coincident with that of the longer lived

(nonquenchable) species that was erroneously identified as

owing to Ph₂Ge in the original work (2).

30-80 ns

a

1.0-1.2µs

0.03

0.02

0.01

0.00

x2

[6]

[Ph₂Ge]

Ph₂Ge GePh₂

15

AcOH

OAc

Ph₂Ge

H

300 350 400 450 500 550 600

Wavelength (nm)

Direct evidence for the formation of Ph₂Ge from 2a was

obtained by examination of the transient spectroscopic be-

havior over shorter timescales. Figure 2 shows transient

spectra recorded at selected time intervals within ca. 1 µs af-

ter excitation, along with representative transient absorption

profiles recorded at 510 and 440 nm. As can be seen from

the data, the long wavelength absorption band in the tran-

sient spectrum from 2a is centered at λ_max~ 465 nm during

the first ca. 100 ns after excitation, and appears to shift to

λ_max~ 445 nm (its position in the spectra of Fig. 1) during

the next ~900 ns. Inspection of the transient absorption pro-

file at 440 nm reveals a slight increase in signal strength at

this wavelength over this early time period, indicative of the

delayed formation of a species that is not present (or at least

present in lower concentration) immediately after the ca.

25 ns laser pulse. Thus, the apparent shift in the position of

the long wavelength transient absorption during the first mi-

crosecond after excitation can be understood as owing to the

superposition of three distinct transient absorptions that each

evolve differently over this timescale: one that decays almost

completely (465 nm < λ_max< 550 nm), one that grows in

(400 nm < λ_max< 460 nm), and one that is essentially

nondecaying (λ_max~ 445 nm, see Fig. 1b). This behavior is

b

0.03

0.02

0.01

0.00

510 nm

440 nm

0

1

2

3

Time (µs)

cant portion of the signal at 510 nm is quenched by the re-

agent, concomitant with a reduction in the yield of a tran-

sient absorbing at 440 nm. Nonlinear least-squares analysis

of the initial rapidly decaying portion of the 510 nm decay

profiles at several AcOH concentrations as a single expo-

nential decay (with the more slowly decaying residual ab-

sorption treated as nondecaying) afforded quite reasonable

fits; the resulting pseudo-first-order decay rate constants

(k_decay) are plotted in Fig. 3c vs. AcOH concentration. Linear

least-squares analysis of these data according to eq. [7],

where k₀is the effective pseudo-first-order decay rate con-

stant at [AcOH] = 0 and k_AcOHis the bimolecular rate con-

stant for scavenging by the carboxylic acid, affords a slope

consistent with the initial formation of Ph₂Ge (λ_max

=

500 nm (9)), which decays over 1 to 2 µs by dimerization to

yield 15 (λ_max= 440 nm (9)). The timescale over which

these changes occur correlates almost precisely with that

found in our previous study of these species, generated by

laser photolysis of 5 (9). The transient decay profiles shown

in Fig. 2 bear only qualitative similarities to those previously

obtained from 5 (9), however, owing to the presence of the

third transient species, which absorbs relative strongly over

the same range of the spectrum, and whose decay proceeds

over a much longer timescale.

Inspection of the transient absorption profiles over the

short timescale as AcOH is added to the solution provides

somewhat clearer evidence for the presence of Ph₂Ge and its

dimerization product in the transient mixture. Figures 3a and

3b show a series of decay profiles recorded at the two wave-

lengths in the presence of 0, 0.1, and 2.0 mmol/L of added

carboxylic acid, demonstrating quite clearly that a signifi-

of k_AcOH= (5.5

0.5) × 10⁹(mol/L)^–1s^–1, in excellent

agreement with the value reported by us earlier for the O-H

insertion reaction of Ph₂Ge with AcOH (k_AcOH= 3.9 ×

10⁹(mol/L)^–1s^–1(9)). Figure 3b also shows the result of

subtracting the decay trace at 440 nm in the presence of

2.0 mmol/L AcOH (where formation of 15 should be more

or less completely suppressed) from that recorded in the ab-

Harrington et al.

1329

Fig. 3. Transient decay traces recorded with a deoxygenated an-

hydrous hexane solution of 2a at (a) 510 nm and (b) 440 nm in

the presence of 0 mmol/L AcOH (᭺), 0.1 mmol/L AcOH (ٗ),

and 2.0 mmol/L AcOH (᭝). The grey trace in (b) is the differ-

ence between the 0 and 2.0 mmol/L traces. (c) Plot of the rate

constant for the initial rapid decay at 510 nm (k_decay) vs. [AcOH]

over the 0.3–2.0 mmol/L concentration range.

sence of scavenger. This is also in good agreement with the

growth–decay profile for 15 that is observed using 5 as the

germylene precursor (9).

[7]

k_decay= k₀+ k_Q[Q]

Addition of n-butyl amine (BuNH₂, 5.0 mmol/L) or

MeOH (50 mmol/L) to the hexane solutions of 2a had simi-

lar effects as AcOH on the overall transient behavior

(Fig. 1), eliminating the absorptions owing to Ph₂Ge and its

products and leaving behind the same long-lived transient

species that is observed in the presence of AcOH. Like the

carboxylic acid, both of these reagents are potent germylene

scavengers, reacting with Ph₂Ge at close to the diffusion-

controlled rate in both cases (9).³Addition of carbon tetra-

chloride (CCl₄), chloroform (CHCl₃), or oxygen (O₂) to hex-

ane solutions of 2a in hexane containing 50 mmol/L MeOH

caused an increase in the decay rate of the surviving long-

lived transient (monitored at 440 nm) in proportion to the

concentration of added reagent, and plots of k_decayvs. [Q]

0.02

a

0.01

0.00

according to eq. [7] were linear, with slopes of k_CC1

=

(7.3

0.7) × 10⁶mol/L^–1s^–1, k_CHC1= (1.2

0.1⁴) ×

10⁴mol/L^–1s^–1and k_O= (6.6 0.8) ×³10⁷mol/L^–1s^–1for

0

1

2

3

2

CCl₄, CHCl₃, and O₂, respectively. The k_Ovalue is in good

Time (µs)

agreement with that reported for this spec²ies in the original

study (k_O= 1.02 × 10⁸mol/L^–1s^–1) (2). It is interesting to

2

note that the transient exhibits similar (but two to three times

greater) reactivity towards O₂and CCl₄as Ph₂Ge does under

similar conditions (10), but exhibits ca. 20 times lower reac-

tivity toward CHCl₃(10).

b

0.03

0.02

0.01

0.00

Finally, we examined the flash photolysis behavior of 2a

in deoxygenated tetrahydrofuran (THF) solution, as Ph₂Ge

and other reactive germylenes are known to form Lewis

acid–base complexes with ethers, with substantially blue-

shifted absorption maxima compared to those of the free

species (16, 17); for example, generation of Ph₂Ge in a 2-

methyltetrahydrofuran glass at 77 K leads to a species ab-

sorbing at λ_max= 325 nm, which was assigned to the

germylene–ether complex (17). The transient spectra ob-

tained by flash photolysis of 2a in THF solution are similar

to those of Fig. 1b, but exhibit a weak shoulder on the long

wavelength edge of the rising 310 nm absorption, consistent

with the presence of the Ph₂Ge-complex. The spectrum of

the complex could be isolated by addition of 0.2 mol/L CCl₄

to the solution, which shortened the lifetime of the

310/445 nm transient absorptions to ca. 500 ns and left be-

0

1

2

3

Time (µs)

c

15

10

5

hind longer lived absorptions (τ ~ 2 µs), centered at λ_max

=

345 nm that decayed with second-order kinetics (see Fig. 4).

The spectrum is in reasonably good agreement with that as-

signed to the Ph₂Ge–THF Lewis acid–base complex (eq. [8])

in the earlier matrix experiments (17). From the lifetime of

the 310/445 nm species under these conditions, a rate con-

stant of ca. 1 × 10⁷mol/L^–1s^–1can be estimated for its reac-

tion with CCl₄, in good agreement with the value measured

in hexane solution (vide supra). The extended lifetime of the

Ph₂Ge–THF complex under these conditions presumably re-

sults from the fact that it is much less reactive toward CCl₄

than the free germylene, whose concentration is reduced by

orders of magnitude relative to that in hexane solution owing

to the solvent complexation equilibrium. The “stabilizing”

slope = (5.5 0.5) x 10⁹(mol/L)^–1s^–1

0

0.000

0.001

0.002

[AcOH] (mol/L)

³W.J. Leigh, C.R. Harrington, F. Lollmahomed, and J.J. McDonald. To be published.

1330

Can. J. Chem. Vol. 83, 2005

Fig. 4. Transient absorption spectra recorded by laser flash

photolysis of a deoxygenated 1.3 × 10^–4mol/L solution of 2a in

THF containing 0.2 mol/L CCl₄, 0.6–2.1 µs (᭺) and 81–85 µs

(ٗ) after the (248 nm) laser pulse. The inset shows a transient

decay profile recorded at 350 nm.

activity of the Si=C bond by roughly a factor of 10 relative

to a methyl substituent (35), which is similar to a phenyl

group in its effects on Si=C reactivity (36, 37). Finally, like

its silicon counterpart (16), germene 8 exhibits significantly

higher reactivity toward O₂and chlorocarbons than it does

toward alcohols or other nucleophiles, owing to the radical-

like nature of these reactions and the radical-stabilizing

effect of the cyclohexadienyl substituent attached to the ger-

menic carbon atom (18). These considerations suggest that

the reactivity of the 310/445 nm transient species of Fig. 1b

is not inconsistent with what might be expected for germene

9. The only piece of data that argues against this assignment

is the UV absorption maximum, which lies ca. 35 nm

(1800 cm^–1) to the blue of that exhibited by germene 8 (18).

In silenes, replacement of a methyl substituent at silicon

with a trimethylsilyl group produces a ca. 4000 cm^–1red

shift in the UV absorption maximum (35); this fact, coupled

with the facts that replacement of one of the phenyl groups

in 16 with a methyl group causes a ca. 2000 cm^–1blue shift

in the absorption maximum (23) and silenes and germenes

with the same substituents exhibit essentially identical UV–

vis absorption maxima (18, 38), leads to the expectation that

the absorption maximum of 9 should be quite similar to that

of 8 (λ_max~ 475 nm).

0.010

0.005

0.000

0

20

40

60

80

Time (µs)

0.005

0.000

300

400

500

600

Wavelength (nm)

effect of ethers and other Lewis bases on reactive

germylenes is well-recognized (25–28).

SiPh₂

_

+

Ph₂Ge

O

[8]

Ph₂Ge +

O

H

SiMe₃

The two most reasonable candidates for the 310/445 nm

species of Fig. 1b are the diphenyl(trimethylsilyl)germyl

radical (10), which our steady-state results indicate is

formed upon photolysis of 2a in significant yield, and the 1-

germahexatriene derivative 9, which would be derived from

formal 1,3-trimethylsilyl migration into the ortho position of

one of the phenyl groups in the excited singlet state of 2a.

While we have no evidence for the formation of 9 from

steady-state photolysis experiments, we can nevertheless not

easily discount the possibility that it is formed in minor

yields and escaped detection in the product studies; trapping

of germenes (and silenes) of this general structure by dienes

and alcohols is known to afford more than a single product

(18, 23, 29). Furthermore, photorearrangements of this type

are common in arylgermasilanes (6, 18, 30) and aryldisi-

lanes (31–33) in which the reaction was initially discovered

(34). Germene 9 can be expected to be a much stronger ab-

sorber than Ph₂Ge, and exhibit similar reactivity to that actu-

ally observed for the 310/445 nm transient (18).

The rationale behind the latter statement is as follows.

The rate constants for reaction of AcOH and MeOH with

germene 8 (one of only two germenes of this type that have

been studied (18)) are on the order of 4 × 10⁵mol/L^–1s^–1or

less, three orders of magnitude lower than the homologous

silene derivative (16) (18). The reactivity of 9 with nucleo-

philes can be expected to be significantly lower than that of

8 because of the presence of the trimethylsilyl substituent at

Ge, based on analogies in transient silene chemistry;

trimethylsilyl substitution at silicon in silenes reduces the re-

16

To address the other possible transient assignment, germyl

radical 10, we have independently generated the radical by

hydrogen abstraction from the corresponding germyl hydride

(17) by tert-butoxyl radicals (eq. [9]), generated by 308 nm

laser flash photolysis of 5–30 mmol/L solutions of 17 in di-

tert-butylperoxide (BOOB) – isooctane (1:4) solution, using

a similar procedure to that used previously for the generation

of triphenylgermyl and other Group 14 radicals (39, 40).

Figure 5 shows the resulting transient absorption spectrum

of the radical, a representative transient absorption profile il-

lustrating its growth kinetics, and a plot of the pseudo-first-

order rate constant for growth of the radical absorption vs.

concentration of 17, the slope of which is the absolute rate

constant for hydrogen abstraction from 17 by tert-butoxyl

radical. The value obtained, k = (2.4 0.4) × 10⁸(mol/L)^–1s^–1,

is roughly twice higher than that for the analogous reaction of

triphenylgermane (Ph₃GeH) (39),⁴which is the expected result

on the basis of comparisons of the corresponding rate con-

stants for hydrogen abstraction from tris(trimethylsilyl)ger-

mane and triethylgermane (41), and analogies taken from

silicon radical chemistry (42). A possible alternate assign-

ment of the transient product to the diphenylgermyl radical,

which might be formed by trimethylsilyl transfer from 17

rather than the expected hydrogen abstraction to form 10,

can be ruled out on the basis of comparisons of the spectrum

to that reported for the diphenylgermyl radical (λ_max= 325,

405 nm (39b)).⁵The authentic spectrum of radical 10

⁴We also remeasured the rate constant for H abstraction from Ph₃GeH in the course of this work and obtained a value of k = (1.2 0.1) ×

10⁸(mol/L)^–1s^–1, in excellent agreement with the previously reported value of 9 × 10⁷(mol/L)^–1s^–1(39).

⁵We thank a referee for alerting us to this possible alternate assignment.

Harrington et al.

1331

Fig. 5. (a) Transient absorption spectrum of the di-

ilar to that of 8 with the same reagents, significantly more so

than it is to the reactivity of the triphenylgermyl radical. In

fact, a significantly lower reactivity toward halocarbons and

oxygen compared to triphenylgermyl is what one would ex-

pect of radical 10, as the stabilizing effect of trimethylsilyl

substitution on both germanium and (in particular) silicon

centered radicals is well-known (41–43).

phenyl(trimethylsilyl)germyl radical, recorded 3.7–3.9 µs after

308 nm laser photolysis of a 0.040 mol/L solution of

Ph₂Ge(H)SiMe₃(17) in BOOB–isooctane (1:4, v/v); the inset

shows a typical growth profile of the radical, recorded at a mon-

itoring wavelength of 420 nm. (b) Plot of k_growthat 420 nm vs.

the concentration of added 17. The solid line represents the lin-

ear least-squares fit of the data to eq. [6].

The photochemistry of 1,4-dihydro-5-methyl-1,2,3,4,9,9-

hexaphenyl-1,4-germanonaphthalene (11)

0.04

0.03

0.02

0.01

0.00

a

Compound 11 was synthesized in ca. 80% yield by cyclo-

addition of 4-methylbenzyne (19) to hexaphenylgermole (18;

see eq. [10]) in dichloromethane at –78 °C, by an analogous

procedure to that reported for the synthesis of 4 (16). The

reaction mixture also contained small amounts of the 4-

methylbenzyne dimerization products 20 and 6-methyl-

1,2,3,4-tetraphenylnaphthalene (21); small amounts of 18 are

very likely present as well, although the compound could

not be specifically detected in the NMR spectrum of the

mixture. The amount of 21 in the sample increased slowly

relative to 11 over a few hours, thus, the latter was isolated

from the reaction mixture by cold (ca. –15 °C) radial chro-

0.10

0.05

0.00

0

1

2

3

Time (µs)

1

matography, characterized on the basis of its H and ¹³C

300

400

500

600

NMR and mass spectra (see the Experimental section), and

stored in the dark at –78 °C. However, in spite of reasonable

efforts, a sample of 11 containing less than ca. 5% 21 could

not be obtained.

Wavelength (nm)

8

6

4

2

0

b

Ph

Ge

Ph

CH₂Cl₂

-78 °C

Ph

[10]

+

Ph

19

18

Ph

Ge

Ph

R

+

Ph

R'

20

Ph

a. R = H, R' = Me

b. R = Me, R' = H

11

Ph

0.00

0.01

0.02

0.03

0.04

Ph

[Ph₂(Me₃Si)GeH] (mol/L)

+

(Fig. 5a) agrees reasonably well with the spectrum of the

310/445 nm transient from photolysis of 2a (Fig. 1b), al-

though the match is not precise. We did not attempt to mea-

sure rate constants for reaction of authentic 10 with O₂or

with CCl₄for comparison to those for the 310/445 nm tran-

sient, the latter because the germane precursor is unstable in

the presence of the halocarbon and light.

Ph

21

To quantify the thermal lability of 11, a 0.055 mol/L solu-

tion of the compound was prepared in deoxygenated cyclo-

hexane-d₁₂containing 0.5 mol/L DMB and monitored over

time at 23 °C by NMR spectroscopy. The mixture contained

ca. 10% 21 at the beginning of the experiment, evidently

formed by thermal decomposition of 11 subsequent to its

initial purification. The NMR spectrum was measured at se-

lected time intervals over the following 20 h, which revealed

slow conversion of 11 to 21 and 5 (eq. [11]). A plot of

Ph

t-Bu

O

[9]

Me Si Ge H

+ t-BuOH

Me₃Si Ge

+

3

Ph

17

10

log([11]/(mol/L)) vs. time was reasonably linear (r²

=

Table 1 lists the UV absorption maxima and absolute rate

constants for reaction of triphenylgermyl radicals (germene

8) and the 310/445 nm transient from flash photolysis of 2a.

As mentioned previously, the reactivity of the species is sim-

0.9886; see Fig. 6) and linear least-squares analysis afforded

a first-order rate constant of (1.6 0.1) × 10^–5s^–1for the de-

composition of 11, which yields 21 and Ph₂Ge, trapped as

DMB adduct 5 at 23 °C. This corresponds to a half-life of

1332

Can. J. Chem. Vol. 83, 2005

Table 1. Comparison of UV–vis absorption spectra and absolute rate constants for the reaction of

the 310/445 nm transient from 2a, triphenylgermyl radicals (Ph₃Ge), and germene 8 with

halocarbons and O₂.

310/445 nm Transient

Ph₃Ge

8

λ_max(hexane solution, nm)

310, 445

7.3×10⁶

328, 405 (50)

5.8×10⁸(3)

475 (18)

2.9×10⁷(18)

k_CC1((mol/L)^–1s^–1)

4

k_CHC1((mol/L)^–1s^–1)

1.2×10⁴

6.6×10⁷

5.5×10^{7 6}

3.8×10⁹(3)

<3×10⁶(18)

9.6×10⁷(18)

3

k_O((mol/L)^–1s^–1)

2

ca. 12 h, or (put another way) a decomposition rate of ca.

6% per hour at 23 °C.

Fig. 6. The thermal decomposition of germanorbornadiene 11 in

cyclohexane-d₁₂at 23 °C. The solid line represents the linear

least-squares fit of the data, affording a first-order rate constant

k = (1.6 0.1) × 10^–5s^–1, r²= 0.9886).

Ph

Ge

∆ or hν

Ph

-1.0

-1.5

-2.0

[11]

GePh₂

DMB

Ph

5

11

Ph

+

Ph

21

Steady-state photolysis of 11 was carried out on argon

outgassed 0.05 mol/L solutions in cyclohexane-d₁₂contain-

ing 0.5 mol/L MeOH or DMB with 254 nm light, with the

1

0

500

1000

course of the photolyses periodically monitored by H NMR

spectroscopy; the samples of 11 each contained ca. 10% 21

at the beginning of the experiment. While irradiations and

NMR analyses were all carried out at room temperature, the

samples were otherwise stored at –78 °C to minimize ther-

mal decomposition. Irradiation to ca. 50% conversion of 11

showed 21 and the corresponding Ph₂Ge trapping products

(methoxygermane 13 or germacyclopentene 5) to be the only

detectable photolysis products. While no attempts were

made to determine the quantum yield for Ph₂Ge extrusion

from the compound, it was estimated to be of similar magni-

tude to that from 5 for which a value of Φ = 0.62 has been

reported (9). Görner et al. (44) reported a quantum yield of

Φ = 0.4 for the extrusion of Me₂Ge from the related com-

pound 22 in deoxygenated methylcyclohexane solution.

Time (min)

3.6 µs after the laser pulse. The results are consistent with

the formation of at least three transient species. The shortest

lived transient exhibits very strong absorptions centered at

λ_max= 360 nm, is formed with the laser pulse, and decays

with a lifetime of τ ~ 300 ns. The second species is assigned

to the weak shoulder on the 360 nm band in the 50–150 ns

spectrum. Its apparent absorption maximum is ca. 480 nm

(see the expanded spectra in Fig. 7b), which is clearly an un-

derestimate caused by the substantial overlap with the

360 nm band. Nevertheless, the absorptions in the 460–

540 nm region of the spectrum are at least five times more

intense than would be expected if Ph₂Ge was the only con-

tributor to the absorptions in this range, assuming the

germylene is produced at an initial concentration of ca.

10 µmol/L, which is the maximum value we would expect

under the conditions of our experiments. The third species is

significantly longer lived than the other two and exhibits an

apparent λ_max~ 460 nm, too far to the red to be assigned to

the expected germylene dimerization product (digermene

15). While it is perhaps relevant to point out that the spec-

trum of this species is similar to that reported in the previous

matrix photolysis of 4 (1), we have insufficient information

to make a reliable assignment of the species. At no monitor-

ing wavelength in the range studied was a growth in the

transient signal discernible, although this is to be expected

Me Me

Ge

Ph

22

Laser flash photolysis experiments were carried out with

flowing, argon outgassed solutions of freshly chromato-

graphed 11 (ca. 5 × 10^–5mol/L, containing ca. 5% 21) in

anhydrous hexane, again using the KrF excimer laser as ex-

citation source. Figure 7 shows a series of transient absorp-

tion spectra recorded at various time intervals over the first

⁶W.J. Leigh and N.P. Toltl. Unpublished results.

Harrington et al.

1333

Fig. 7. (a) Transient absorption spectra recorded by laser flash

photolysis of a deoxygenated 5 × 10^–5mol/L solution of 11 in

anhydrous hexane at 23 °C, 50–150 ns (᭺) and 3.2–3.5 µs (ٗ)

after the (248 nm) laser pulse; the inset shows a transient decay

profile recorded at 360 nm. (b) An expanded plot of the spectra

of (a) with a spectrum recorded 1.4–1.6 µs after the pulse in-

cluded; the vertical line marks the apparent maximum of the

long wavelength absorption band in the 50–150 ns spectrum.

65% of the absorptions decaying on a very fast timescale

and the remaining ~35% exhibiting a lifetime similar to that

observed in the absence of O₂. This behaviour suggests that

the transient absorptions produced upon flash photolysis of

11 are dominated by triplet or triplet-derived species.

Our results for 11 are similar to those reported for the

related compound 22 by Görner and co-workers (44), who

assigned the transients observed upon 248 nm laser flash

photolysis of the compound to the triplet state of 1,2,3,4-

tetraphenylnaphthalene, the coproduct of Me₂Ge photoex-

trusion from this compound. They were similarly unable to

obtain discrete evidence for the formation of a transient as-

signable to the germylene (44).

0.4

a

0.30

0.3

0.2

0.25

0.1

As was also reported in the earlier study of 22 (44), virtu-

ally identical transient spectroscopic behaviour was observed

upon laser flash photolysis of (static) 1 × 10^–5mol/L solu-

tions of the tetraphenylnaphthalene coproduct (21, in the

present case) in both deoxygenated and oxygen saturated

hexane solution, with the transient spectra and lifetimes at

various wavelengths coinciding quite closely to those ob-

tained using 11 under corresponding conditions. In this case,

saturation of the solution with O₂decreased the lifetime of

the 500 nm transient absorptions to ca. 50 ns from which

can be estimated a quenching rate constant of k_O~ 1.3 ×

10⁹(mol/L)^–1s^–1, in good agreement with that ob²tained in

experiments using 11 as the precursor.

0.0

0.20

0

1

2

3

Time (µs)

0.15

0.10

0.05

0.00

50-150 ns

3.2-3.5µs

300

400

500

600

Wavelength (nm)

Görner and co-workers (44) explained their results for 22,

which were much more extensive than the present ones for

11 and involved purer samples (22 is evidently significantly

more stable thermally than 11 (19)), in terms of a mecha-

nism in which the (1,2,3,4-tetraphenylnaphthalene) coproduct

of Me₂Ge extrusion is formed in the excited singlet state,

which then yields the triplet after intersystem crossing. The

alternative possibility that the naphthalene triplet is directly

produced along with (singlet) Me₂Ge from the triplet state of

the precursor was ruled out on the basis of CIDNP experi-

ments. Because our samples of 11 contained small amounts

of the corresponding naphthalene derivative (21), we cannot

ascertain to what extent the transient mixture produced in

our experiments with the Ph₂Ge precursor is contaminated

by cophotolysis of the impurity. Nevertheless, our results

verify that the transient spectroscopic behaviour observed in

flash photolysis experiments with compounds of this type is

dominated by triplet and (or) triplet-derived species (for one

reason or another), and that simple germylenes such as

Ph₂Ge or Me₂Ge are impossible to detect reliably (if at all)

in fluid solution from such precursors by time-resolved UV

spectroscopic methods.

b

0.03

0.02

0.01

0.00

0.06

0.05

0.04

0.03

0.02

0.01

0.00

0

1

2

3

Time (µs)

300

400

500

600

Wavelength (nm)

considering the substantial overlap with the strongly absorb-

ing species at 360 nm.

Similar experiments carried out on an O₂saturated solu-

tion of 11 verified that Ph₂Ge and its oligomerization prod-

ucts contribute in only a very minor way, if at all, to the

transient absorptions observed from flash photolysis of this

compound. Under these conditions the transient at 500 nm

exhibited a lifetime of ca. 75 ns, and decayed to a residual

absorption level roughly 4% of that immediately present af-

ter the pulse. A quenching rate constant of k_O~ 9 ×

10⁸(mol/L)^–1s^–1can be estimated from the lifetime²(using a

value of [O₂] = 0.015 mol/L for the concentration of O₂in

O₂saturated hexane), which is ca. 30 times higher than the

rate constant reported by us recently for the reaction of

Summary and conclusions

In spite of the indication from product studies that the di-

rect photolyses of disilylgermane 2a and benzogermanor-

bornadiene 11 afford Ph₂Ge in very high yields, they are

clearly unsuitable as precursors of the germylene for study

by time-resolved UV–vis spectroscopic methods in solution.

In the case of 2a, this is due to the competing formation of a

second transient product, which, although it is formed in rel-

atively low yield, absorbs strongly over the same spectral

range as Ph₂Ge and its dimerization product (digermene 15),

and is much longer lived than the relatively weakly absorb-

Ph₂Ge with O₂(k_O~ 2.7 × 10⁷(mol/L)^–1s^–1(10)). The

other major change noted upon saturation of the solution

with O₂was a change in the form of the decay of the

360 nm transient to biexponential kinetics, with roughly

2

1334

Can. J. Chem. Vol. 83, 2005

ing germylene. The germylene and its dimer can be success-

fully detected in laser flash photolysis experiments with this

compound, but their relatively weak absorptions are super-

imposed on a background of intense, relatively long-lived

absorptions owing to the minor product. This makes them

very easy to miss altogether, particularly when little or no

independent information on their spectra and reactivities ex-

ists, as was the case in the original study. The UV–vis ab-

sorption spectrum of the species and its reactivity toward

halocarbons and oxygen is broadly consistent with its

assignment to the diphenyl(trimethylsilyl)germyl radical,

which product studies show is formed in minor, but signifi-

cant, yields upon photolysis of 2a. Radical formation is sup-

pressed in the presence of high concentrations of diene,

suggesting that the Si—Ge bond homolysis process that is

responsible for its formation is derived from the triplet state

of the disilylgermane. The alternate assignment of the spe-

cies to germene 9 cannot be ruled out.

Other factors frustrate attempts to detect the germylene

and its dimer in laser flash photolysis experiments with the

benzogermanorbornadiene derivative 11, in spite of the fact

that product studies indicate its photodecomposition to pro-

ceed cleanly and in relatively high quantum yield. One of

these is the compound’s rather low thermal stability, which

makes it exceedingly difficult to purify to the level necessary

for studies of this type; the major product of thermal decom-

position (tetraphenylnaphthalene derivative 21) absorbs

strongly at the wavelength necessary to excite the germylene

precursor, and its relatively efficiently populated triplet state

absorbs very strongly over the same range as the germylene

does. Earlier studies with a related compound (44) indicate

that even when highly pure samples of the precursor can be

obtained, the only transients detectable by time-resolved UV

spectroscopic methods are derived from the excited triplet

state of the aromatic coproduct. This is thought to be due to

the fact that the photodecomposition reaction proceeds adia-

batically, ultimately yielding the coproduct in its excited

triplet state. In the present experiments with 11, these two

complicating factors combine to make the triplet state of

tetraphenylnaphthalene 21 the dominant transient detectable

upon laser flash photolysis of the compound in solution at

ambient temperatures.

vania), while GC–MS analyses were carried out using a

Micromass GCT instrument interfaced to an Agilent 6890

Series GC system with a J & W DB-XLB 30 m column.

Melting points were determined using a Mettler FP82 hot

stage mounted on a polarizing microscope and are uncor-

rected. Radial chromatographic separations were carried out

using a Chromatotron^®(Harrison Research, Inc., Palo Alto,

California) with silica gel 60 plates (4 mm thick) and with

hexane as the eluant. Column chromatography was per-

formed on a 30 mm × 600 mm column using silica gel

(~100 g, acid washed, 230–400 mesh) (Silicycle). Steady-

state photolyses were carried out in a Rayonet^®photochemi-

cal reactor (Southern New England Ultraviolet Co.,

Branford, Connecticut) equipped with a merry-go-round ap-

paratus and six RPR-2537 lamps (254 nm).

4-Methyl-2-nitroaniline (Sigma-Aldrich, St. Louis, Mis-

souri), chlorotrimethylsilane (Gelest, Morrisville, Pennsylva-

nia), tert-butyllithium (1.7 mol/L in pentane, Sigma-Aldrich),

dimethylmalonate (Sigma-Aldrich), and lead tetraacetate

(Sigma-Aldrich) were used as received from the suppliers.

2,3-Dimethyl-1,3-butadiene (Sigma-Aldrich) was purified

before use by bulb-to-bulb distillation under vacuum. Tetra-

hydrofuran and hexanes (Caledon reagent grade, Georgetown,

Ontario) were dried by passage through activated alumina

under nitrogen using a Solv-Tek solvent purification system

(Solv-Tek, Inc., Berryville, Virginia). Dichloromethane, meth-

anol, acetone and pentane (all Caledon reagent grade) were

used as received. Triethylamine (Sigma-Aldrich) was

refluxed over solid KOH and distilled before use.

1,1,2,3,4,5-Hexaphenylgermole was prepared by the proce-

dure of Ferman et al. (45).

Bis(trimethylsilyl)diphenylgermane (2a)

To a flame-dried 150 mL two-neck round-bottom flask

equipped with a rubber septum, magnetic stirring bar, and

nitrogen inlet was added anhydr. THF (30 mL) via a syringe.

The solution was cooled to –40 °C (MeCN–CO₂bath) and

5.2 g (22.8 mmol) of diphenylgermane was added by syringe

dropwise through the septum. A solution of tert-butyllithium

(1.7 mol/L in pentane, 32 mL, 54.4 mmol) was then added

dropwise using a syringe in several portions over 10 min.

The initial brilliant yellow reaction mixture because of the

diphenylgermyl anion transformed to a dark red solution af-

ter the half-way point of the addition, signifying the pres-

ence of the diphenylgermyl dianion (Ph₂GeLi₂) (46). The

solution was stirred at –40 °C for 25 min and then neat

chlorotrimethylsilane (9 mL, 7.7 g, 71.3 mmol) was added

dropwise via a syringe, causing a slight increase in tempera-

ture and the mixture to turn yellow and cloudy. The cooling

bath was removed and the reaction mixture was stirred for

24 h, during which time a white precipitate formed. Satu-

rated aqueous ammonium chloride was added slowly until

the precipitate dissolved, diethyl ether (25 mL) was added,

the layers were separated, and the aqueous portion was ex-

tracted with additional ether (3 × 25 mL). The combined

ether extracts were washed with saturated brine (25 mL) and

water (25 mL), dried over anhydrous magnesium sulphate,

filtered, and the solvent was removed on the rotary evapora-

tor. The resulting yellow oil was purified by vacuum distilla-

tion (6.7 g, 18.1 mmol, 79%, 110 °C (0.02 mmHg,

1 mmHg = 133.322 4 Pa)) and recrystallized at –40 °C sev-

Experimental

¹H, ¹³C, and ²⁹Si NMR spectra were recorded on Bruker

AV 200 or AV 600 spectrometers in deuteriochloroform

1

solution unless otherwise noted; H and ¹³C NMR spectra

were referenced to the residual solvent proton and ¹³C sig-

nals, respectively. High-resolution electron impact and atmo-

spheric pressure chemical ionization (APCI) mass spectra

and exact masses were determined on a Micromass TofSpec

2E (MALDI-TOF mass spectrometer) performed on a

Micromass Ultima ESI/APCI triple quadrupole mass spec-

trometer. IR spectra were recorded on a BioRad FTS-40 FT

IR spectrometer and are reported in wavenumbers (cm^–1).

Gas chromatographic analyses were carried out using a

Hewlett-Packard 5890 gas chromatograph equipped with a

conventional heated injector, a flame ionization detector, a

Hewlett-Packard 3396A integrator, and a SPB-5 capillary

column (25 m, 0.25 µm, Supelco, Inc., Bellefonte, Pennsyl-

Harrington et al.

1335

eral times from acetone–methanol (1:1, v/v) to yield colour-

removal of the pentane under vacuum afforded a colourless

liquid. The product was further purified by bulb-to-bulb dis-

tillation (twice) to yield a colourless solid (0.55 g,

1.83 mmol, 35%, mp 36.0–37.2 °C), which was identified as

17 on the basis of the following spectral data: IR (solid):

3068 (m), 2955 (s), 2915 (s), 1998 (s, Ge-H), 1483 (m),

less needles with a melting point of approximately –20 °C.

1

The compound was identified as 2a by comparison of its H

and ¹³C NMR and mass spectra to the previously reported

1

data (6). The purity was estimated to be >99.5% by GC. H

NMR (CDCl₃) δ: 0.20 (s, 18H), 7.21–7.24 (m, 6H), 7.34–

7.39 (m, 4H). ¹³C NMR (CDCl₃) δ: 0.33, 127.5, 128.1,

135.8, 139.4. MS (EI) m/z: 376 (1, ⁷⁶Ge), 374 (5, ⁷⁴Ge), 372

(3, ⁷²Ge), 370 (2, ⁷⁰Ge), 359 (20), 357 (15), 355 (10), 301

(10), 299 (7), 297 (5), 226 (36), 224 (75), 222 (68), 220

(65), 197 (67), 151 (40), 149 (32), 147 (27), 136 (33), 135

(100), 107 (23), 73 (65), 45 (25). Exact mass calcd. for

C₁₈H₂₈GeSi₂: 374.0941; found: 374.0950.

1

1430 (m), 1246 (s), 1088 (m), 976 (w), 836 (s). H NMR

(C₆D₆) δ: 0.18 (s, 9H), 5.08 (s, 1H), 7.12–7.18 (m, 6H),

7.55–7.57 (m, 4H). ¹³C NMR (C₆D₆) δ: –0.63, 128.5, 128.6,

135.8, 137.4. ²⁹Si NMR (C₆D₆) δ: –10.1. MS (EI) m/z: 299.7

(2), 229.9 (13), 228.9 (31), 227.9 (86), 227.0 (100), 226.0

(86), 225.0 (70), 224.1 (56), 223.1 (44), 198.1 (6), 197.1

(25), 195.2 (6), 154.1 (5) 153.1 (15), 152.1 (14), 151.1 (58),

150.1 (15), 149.1 (45), 148.1 (4), 147.1 (32), 136.2 (8),

135.2 (50).

1,1,2,2-Tetraphenyl-1,2-bis(trimethylsilyl)digermane (14)

To a flame-dried 25 mL one-neck round-bottom flask

equipped with a septum and magnetic stirring bar, under ni-

trogen, was added 1,1,2,2-tetraphenyldigermane (0.51 g,

1.12 mmol) (47) in anhydr. THF (4 mL) via a syringe. The

solution was cooled to –40 °C (MeCN–CO₂bath) and tert-

butyllithium (1.4 mL, 2.38 mmol) was added dropwise via a

syringe in several portions over 10 min. The dark yellow re-

action mixture was stirred for 40 min at –40 °C and

chlorotrimethylsilane (0.4 mL, 3 mmol) was added dropwise

via a syringe over 2 min, causing the mixture to turn colour-

less. The reaction mixture was warmed to ambient tempera-

ture and stirred for 1 h after which saturated aqueous

ammonium chloride (2 mL) and distilled water (4 mL) were

added slowly until the solution clarified. The work up con-

sisted of dilution with diethyl ether (20 mL) and extraction

of the aqueous layer with more ether (2 × 20 mL). The com-

bined organic layers were washed with saturated brine

(20 mL) and distilled water (20 mL), dried over magnesium

sulphate, filtered, and the solvent was removed on the rotary

evaporator. The resulting colourless solid (>95% pure by

NMR) was identified as 14 (mp 212–214 °C) on the basis of

the following spectroscopic data: IR (solid): 3066 (s), 2956

(m), 1580 (w), 1481 (m), 1429 (m), 1245 (s), 1046 (s), 833

1-Amino-5-methylbenzotriazole was prepared by a modi-

fication of the procedure of Campbell and Rees (48), starting

from 4-methyl-2-nitronaniline and dimethyl malonate. The

original synthetic work did not include NMR or MS data for

the compound or its synthetic precursors, so these are listed

in the following.

Dimethyl mesoxalate 4-methyl-2-nitrophenylhydrazone

Bronze-coloured crystals, mp 158–160 °C. IR (neat):

2991 (m), 1745 (s), 1725 (s), 1450 (w), 1368 (m), 1174 (m),

1

944 (w). H NMR (CDCl₃) δ: 2.35 (s, 3H), 3.87 (s, 3H),

3.93 (s, 3H), 7.24 (d, J = 8.5 Hz, 1H), 7.91 (d, J = 8.5 Hz,

1H), 7.97 (s, 1H), 14.08 (s, 1H). ¹³C NMR (CDCl₃) δ: 20.3,

52.5, 52.6, 117.1, 124.1, 127.2, 133.4, 134.4, 136.2, 136.9,

161.8, 163.2. MS (EI) m/z: 295 (2), 141 (5), 89 (6), 87 (20),

72 (100), 71 (99), 57 (85). Exact mass calcd. for

C₁₂H₁₃N₃O₆: 295.0804; found: 295.0802.

Dimethyl mesoxalate 2-amino-4-methylphenylhydrazone

Orange granules, mp 128 to 129 °C. IR (neat): 3464 (m),

3311 (w), 2990 (w), 1745 (s), 1725 (s), 1680 (m), 1651 (m),

1530 (s), 1440 (m), 1360 (s), 1240 (s), 1200 (m), 800 (m).

¹H NMR (CDCl₃) δ: 2.22 (s, 3H), 3.80 (s, 3H), 3.86 (s, 3H),

5.0 (bs, 2H), 6.52 (d, J = 8.4 Hz, 1H), 6.54 (s, 1H), 6.90 (d,

J = 8.4 Hz, 1H), 13.29 (s, 1H). ¹³C NMR (CDCl₃) δ: 20.9,

51.8, 51.9, 112.8, 118.5, 119.0, 119.2, 124.4, 136.1, 137.0,

163.4, 164.4. MS (EI) m/z 265 (2), 263 (3), 205 (8), 177 (5),

159 (5), 146 (5), 133 (5), 120 (10), 119 (12), 91 (50), 85

(80), 84 (100), 71 (60), 49 (85). Exact mass calcd. for

C₁₂H₁₅N₃O₄: 265.1063; found: 265.1067.

1

(s). H NMR (C₆D₆) δ: 0.14 (s, 18H), 7.11 (t, J = 7.2 Hz,

4H), 7.15 (t, J = 7.2 Hz, 8H) 7.64 (d, J = 7.8 Hz, 8H). ¹³C

NMR (C₆D₆) δ: 0.24, 128.3, 128.4, 136.5, 139.8. ²⁹Si NMR

(C₆D₆) δ: –6.1. MS (EI) m/z: 602 (1), 601 (1), 600 (2), 599

(1), 598 (1.5), 587 (4), 585 (5), 583 (4), 529 (4), 527 (5),

525 (4), 451 (60), 450 (98), 448 (100), 446 (50), 301 (69),

299 (55), 297 (43), 151 (56), 149 (45), 147 (34). Exact mass

calcd. for C₃₀H₃₈Ge₂Si₂: 602.0936; found: 602.0947.

Dimethyl (5-methylbenzotriazol-1-yl)iminomalonate

Yellow powder, mp 115 to 116 °C. IR (neat): 2990 (w),

1745 (s), 1625 (m), 1430 (w), 1300 (m), 1240 (s), 1160 (m),

Diphenyltrimethylsilylgermane (17)

To a 50 mL Schlenk tube connected to a vacuum line and

equipped with a magnetic stirring bar was added diphenyl-

germane (1.18 g, 5.16 mmol) and freshly distilled THF

(5 mL). The tube was cooled to –40 °C and tert-butyllithium

in hexane (1.58 mol/L, 3.26 mL, 5.16 mmol)) was added

over 5 min. The resulting yellow mixture was allowed to

warm to –20 °C and then stirred for 20 min after which

chlorotrimethylsilane (1 mL, freshly distilled from calcium

hydride) was added via a syringe, causing the immediate

formation of a colourless precipitate. The mixture was

warmed to room temperature and stirred for 1 h after which

the solvent and excess chlorosilane were removed under

high vacuum. The residue was extracted with dry pentane;

1

1030 (s), 812 (s). H NMR (CDCl₃) δ: 2.51 (s, 3H), 3.99 (s,

3H), 4.02 (s, 3H), 5.1 (bs, 2H), 7.44 (d, J = 8.4 Hz, 1H),

7.73 (d, J = 8.4 Hz, 1H), 7.79 (s, 1H). ¹³C NMR (CDCl₃) δ:

21.4, 53.5, 53.8, 110.2, 119.4, 130.2, 132.0, 136.2, 140.0,

145.3, 160.8, 162.5. MS (EI) m/z: 276 (2), 205 (4), 177 (6),

120 (5), 119 (7), 91 (15), 88 (28), 86 (95), 84 (100), 51 (70),

49 (99), 47 (55). Exact mass calcd. for C₁₂H₁₂N₄O₄:

276.0859; found: 276.0864.

1-Amino-5-methylbenzotriazole

Colourless needles, mp 91 to 92 °C (lit. value (48) mp 87

to 88 °C). IR (neat): 3295 (m), 3180 (m), 2990 (w), 1745

1336

Can. J. Chem. Vol. 83, 2005

(m), 1725 (m), 1235 (m), 1150 (m), 1110 (m), 958 (s), 810

a relative yield of 1:1.08. No other peaks were evident in the

0–0.3 ppm region of the spectrum, although it should be

noted that the signal owing to PhSiMe₃(δ 0.233) is almost

coincident with that owing to 2a (δ 0.228). In a second ex-

periment, a deoxygenated hexane solution containing 2a

(0.01 mol/L) and DMB (0.03 mol/L) was photolyzed to ca.

60% conversion with periodic monitoring by GC. The prod-

uct mixture contained (in order of increasing elution time)

PhSiMe₃(2.3%), 12 (2.5%), 5 (41%), and 2a (50%); Si₂Me₆

elutes with the solvent under these conditions. The products

were identified by GC–MS analysis; the product assigned to

12 exhibited the following mass spectrum: MS (EI) m/z: 308

(2, ⁷⁶Ge), 306 (8, ⁷⁴Ge), 304 (5, ⁷²Ge), 302 (3, ⁷⁰Ge), 291

(2), 289 (1), 287 (1), 226 (6), 224 (42), 222 (28), 220 (19),

156 (39), 154 (25), 152 (17), 151 (17), 149 (12), 147 (10),

136 (19), 135 (100), 73 (40), 45 (5).

Steady-state photolysis of a deoxygenated solution (ca.

0.75 mL) of 2a (0.05 mol/L) in cyclohexane-d₁₂containing

MeOH (0.5 mol/L) was carried out in a similar manner to

that previously described for the DMB trapping experiment.

The NMR spectra of the photolysate showed the growth of

peaks assignable to Si₂Me₆(δ 0.049), 13 (δ: 3.54 (s, 3H),

6.12 (s, 1H), 7.2–7.32 (m, 6H), 7.51–7.53 (m, 4H)), and 14

(δ: 0.027 (s, 18H), 7.15–7.19 (m, 12H), 7.36–7.40 (m, 8H))

on the basis of comparisons with the spectra of authentic

samples, together with two additional Me₃Si containing spe-

cies (δ 0.057 and 0.063), which were not identified; the rela-

tive intensities of the Si₂Me₆singlet, the MeO singlet owing

to 13 (δ 3.54), and the –SiMe₃singlets owing to 14 (δ 0.027)

and the two unidentified components were 1.00 : 0.139

0.013 : 0.114 0.020 : 0.103 0.027 : 0.121 0.006. This

gives average relative yields of 1.0 : 0.84 0.07 : 0.11

0.02 : 0.20 0.05 : 0.24 0.02 for the five products, assum-

ing that the singlets for the two unidentified components

represent a single trimethylsilyl group each.

1

(s). H NMR (CDCl₃) δ: 2.46 (s, 3H), 5.73 (s, 2H), 7.28 (d,

J = 8.4 Hz, 1H), 7.47 (d, J = 8.4 Hz, 1H), 7.66 (s, 1H). ¹³C

NMR (CDCl₃) δ: 21.4, 109.3, 118.4, 129.9, 130.9, 134.1,

145.0. MS (EI) m/z: 148 (4), 133 (4), 120 (25), 119 (35), 92

(20), 91 (90), 86 (75), 84 (100), 77 (15), 65 (22), 51 (17), 50

(75). Exact mass calcd. for C₇H₈N₄: 148.0749; found: 148.0750.

1,4-Dihydro-5-methyl-1,2,3,4,9,9-hexaphenyl-1,4-

germanonaphthalene (11)

1,4-Dihydro-5-methyl-1,2,3,4,9,9-hexaphenyl-1,4-germanon-

aphthalene (11) was prepared using the method of Ando et

al. (17) for the synthesis of 4. To a flame-dried 100 mL two-

neck round-bottom flask equipped with an addition funnel

and magnetic stir bar under N₂was added anhydrous

dichloromethane (30 mL) containing 1,1,2,3,4,5-hexaphenyl-

germole (2.00 g, 3.42 mmol) and 1-amino-5-methylbenzo-

triazole (0.53 g, 3.59 mmol). The resulting yellowish

solution was cooled to –78 °C (i-PrOH–CO₂bath) and a so-

lution of lead tetraacetate (1.67 g, 3.77 mmol) in anhydrous

dichloromethane (15 mL) added dropwise with stirring over

50 min. The resulting brown-tinged solution was allowed to

warm gradually to room temperature over 1.5 h and then

quenched with saturated aqueous sodium carbonate (20 mL).

The organic layer was separated, dried with magnesium sul-

fate, filtered, and the solvent was removed on the rotary

evaporator to yield a yellow oil. The oil was purified by

radial chromatography in the freezer at –15 °C using

precooled dichloromethane–pentane (1:3, v/v) as the eluant,

which yielded three easily separable fractions. The first two

fractions were identified as a 1:1 mixture of 3(3′)-dimethyl-

biphenylene and 3(4′)-dimethylbiphenylene (20a and 20b,

respectively, 0.05 g, 0.28 mmol) (49) and 6-methyl-1,2,3,4-

tetraphenylnaphthalene (21) (0.35 g, 0.78 mmol) by compar-

ison of their ¹H NMR spectra to the literature data. The third

fraction (1.8 g, 2.67 mmol, 78%) consisted of a pale yellow

powder that was identified as 11 on the basis of the follow-

Steady-state photolyses (254 nm) of 11 were carried out

in similar fashion, on argon outgassed 0.046 mol/L solutions

of the compound in cyclohexane-d₁₂containing methanol

(0.5 mol/L) and DMB (0.5 mol/L). The photolyses were

taken to ca. 60% conversion and were monitored periodi-

1

ing spectroscopic data: H NMR (CDCl₃) δ: 2.34 (s, 3H),

6.62 (bs, 4H), 6.80 (d, J = 7.2 Hz, 2H), 6.90–6.96 (m, 7H),

7.07–7.13 (m, 12H), 7.19 (d, J = 7.2 Hz, 1H), 7.39 (d, J =

7.2 Hz, 2H), 7.39 (t, J = 7.2 Hz, 2H), 7.44 (t, J = 7.2 Hz,

1H), 7.75 (d, J = 7.2 Hz, 2H). ¹³C NMR (CDCl₃) δ: 21.5,

61.1, 64.4, 121.8, 122.9, 124.9, 125.5, 125.7, 127.1, 127.6,

128.0, 128.3, 129.7, 129.9, 130.0, 130.1, 131.5, 133.7, 135.0,

135.8, 136.1, 137.1, 138.0, 138.1, 138.3, 138.4, 145.1, 145.8,

146.2, 147.9. MS (APCI) m/z: 677.2 (20, ⁷⁶Ge, M + 1), 675.2

(100, ⁷⁴Ge, M + 1), 673.1 (60, ⁷²Ge, M + 1), 671.2 (45,

⁷⁰Ge, M + 1).

1

cally by H NMR spectroscopy. In both cases, the formation

of the diphenylgermylene adduct (5 and 13, respectively)

was accompanied by an equal yield of 21; no other products

could be detected in yields greater than ca. 3%. The identi-

ties of the products were verified by comparison of the spec-

tra of the crude photolysis mixtures with those of authentic

samples.

Thermolysis of benzogermanorbornadiene 11

The thermolysis of 11 was carried out with argon out-

gassed 0.046 mol/L solutions of the compound in cyclohexane-

d₁₂containing DMB (0.5 mol/L). Thermal decomposition of

Steady-state photolysis experiments

Steady-state photolysis of a deoxygenated solution (ca.

0.75 mL) of 2a (0.05 mol/L) in cyclohexane-d₁₂containing

DMB (0.1 mol/L) was carried out in a quartz NMR tube,

1

11 was monitored by 200 MHz H NMR spectroscopy, mea-

1

suring the relative intensities of the δ 2.28 singlet owing to

11 and the residual proton peak owing to the solvent, every

20 min over a 24 h period. The loss of 11 was accompanied

monitoring the reaction periodically by H NMR spectros-

copy over the 0%–16% conversion range. The NMR spectra

of the photolysate showed the growth of peaks assignable to

hexamethyldisilane (Si₂Me₆, δ 0.049) and 5 (δ: 1.78 (s, 6H)

1.97 (s, 4H), 7.20–7.24 (m, 6H), 7.40–7.43 (m, 4H)) on the

basis of comparisons with the spectra of authentic samples;

the relative intensities of the Si₂Me₆singlet and the δ

1.78 ppm singlet owing to 5 averaged to 1.0:3.24, indicating

by the formation of 21 (δ_CH2.31) and 5.

3

Laser flash photolysis experiments

Laser flash photolysis experiments with 2a, 11, and 21

employed the pulses from a Lambda Physik Compex 120

Harrington et al.

1337

excimer laser filled with F₂–Kr–Ne mixtures (248 nm,

~25 ns, 100 5 mJ) and a Luzchem Research mLFP-111

laser flash photolysis system, modified as previously de-

scribed (9). Solutions were prepared at concentrations such

that the absorbance at the excitation wavelength was be-

tween ca. 0.7 and 0.9, and were continuously flowed through

a vacuum oven-dried, thermostatted 7 mm × 7 mm Suprasil

flow cell connected to a calibrated 100 mL reservoir, fitted

with a glass frit to allow bubbling of argon gas through the

solution for at least 30 min prior to and then throughout the

duration of each experiment; the entire apparatus was dried

in a vacuum oven at 65–85 °C before use. Solution tempera-

tures were measured with a Teflon-coated copper/constantan

thermocouple directly inserted into the flow cell. Reagents

were added directly to the reservoir by microliter syringe as

aliquots of standard solutions. In experiments with those

scavengers that absorb at the laser wavelength, the concen-

trations employed were kept low enough to ensure that the

fraction of the excitation light absorbed by the scavenger

never exceeded ca. 10%, to avoid complications in the kinet-

ics owing to screening effects or the formation of potentially

reactive species from cophotolysis of the scavenger. For ex-

periments using oxygen as a scavenger, solutions of selected

O₂–N₂ratios were prepared by adding the appropriate vol-

ume of a nitrogen saturated solution of 1a in hexane to an

oxygen saturated hexane solution of the same concentration

in 1a.

Cominco Metals Ltd. for a generous gift of germanium

tetrachloride, the McMaster Regional Centre for Mass Spec-

trometry for mass spectra, J.C. Scaiano for the use of his la-

ser photolysis system, and M. Chretien for technical

assistance. PPG and DZ are grateful for financial support

from the United States National Science Foundation under

grants CHE-9981715 and CHE-0316124. BKC thanks the

Reactive Intermediate Student Exchange (RISE) and

NSERC for summer scholarships.

References

1. W. Ando, T. Tsumuraya, and A. Sekiguchi. Chem. Lett. 317

(1987).

2. S. Konieczny, S.J. Jacobs, J.K. Braddock Wilking, and P.P.

Gaspar. J. Organomet. Chem. 341, C17 (1988).

3. K. Mochida, M. Wakasa, Y. Nakadaira, Y. Sakaguchi, and H.

Hayashi. Organometallics, 7, 1869 (1988).

4. M. Wakasa, I. Yoneda, and K. Mochida. J. Organomet. Chem.

366, C1 (1989).

5. K. Mochida, I. Yoneda, and M. Wakasa. J. Organomet. Chem.

399, 53 (1990).

6. K.L. Bobbitt, V.M. Maloney, and P.P. Gaspar. Organometallics,

10, 2772 (1991).

7. N.P. Toltl, W.J. Leigh, G.M. Kollegger, W.G. Stibbs, and K.M.

Baines. Organometallics, 15, 3732 (1996).

8. K. Mochida, M. Wakasa, and H. Hayashi. Phosphorus Sulfur

Silicon Relat. Elem. 150–151, 237 (1999).

9. W.J. Leigh, C.R. Harrington, and I. Vargas-Baca. J. Am.

Chem. Soc. 126, 16105 (2004).

10. W.J. Leigh and C.R. Harrington. J. Am. Chem. Soc. 127, 5084

(2005).

11. K. Mochida, N. Kanno, R. Kato, M. Kotani, S. Yamauchi, M.

Wakasa, and H. Hayashi. J. Organomet. Chem. 415, 191 (1991).

12. K. Mochida and S. Tokura. Organometallics, 11, 2752 (1992).

13. K. Mochida and S. Tokura. Bull. Chem. Soc. Jpn. 65, 1642

(1992).

Flash photolysis experiments with compound 17 and tri-

phenylgermane were carried out using Professor J.C. Scaiano’s

system at the University of Ottawa and a Lumonics EX-530

excimer laser filled with Xe–HCl–Ne mixtures (308 nm,

~6 ns, 70

5 mJ) for excitation. Signals from the

monochromator/photomultiplier were captured by

a

Tektronix 2440 digitizer and transferred to a PowerMacintosh

computer programmed in the LabVIEW 4.1 environment

from National Instruments (Austin, Texas). Detailed descrip-

tions of similar laser systems have been provided elsewhere

(51). Argon outgassed solutions of 18 (5–30 mmol/L) in

BOOB–isooctane (1:4, v/v) were contained in 7 mm × 7 mm

(static) quartz cells. Experiments were carried out at room

temperature (22 °C). Growth rate constants for radical ab-

sorptions were determined by biexponential nonlinear least-

squares analysis of the growth–decay profiles, recorded at

420 nm in the case of 17 and 340 nm in the case of

triphenylgermane.

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., San Diego, California) and the appropriate

user-defined fitting equations, after importing the raw data

from the Luzchem mLFP software. Rate constants were cal-

culated by linear least-squares analysis of decay rate – con-

centration data (generally four to seven 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.

14. K. Mochida, K. Kimijima, and H. Chiba. Organometallics, 13,

404 (1994).

15. M.B. Taraban, O.S. Volkova, V.F. Plyusnin, Y.V. Ivanov, T.V.

Leshina, M.P. Egorov, O.M. Nefedov, T. Kayamori, and K.

Mochida. J. Organomet. Chem. 601, 324 (2000).

16. W. Ando, H. Itoh, T. Tsumuraya, and H. Yoshida. Organo-

metallics, 7, 1880 (1988).

17. W. Ando, H. Itoh, and T. Tsumuraya. Organometallics, 8, 2759

(1989).

18. W.J. Leigh, N.P. Toltl, P. Apodeca, M. Castruita, and K.H.

Pannell. Organometallics, 19, 3232 (2000).

19. W.P. Neumann and M. Schriewer. Tetrahedron Lett. 21, 3273

(1980).

20. J. Kocher, M. Lehnig, and W.P. Neumann. Organometallics, 7,

1201 (1988).

21. W.P. Neumann, E. Michels, and J. Kocher. Tetrahedron Lett.

28, 3783 (1987).

22. W. Ando and T. Tsumuraya. Organometallics, 8, 1467 (1989).

23. W.J. Leigh and G.W. Sluggett. J. Am. Chem. Soc. 115, 7531

(1993).

24. W.J. Leigh and G.W. Sluggett. Organometallics, 13, 269 (1994).

25. P. Riviere, J. Satge, and A. Castel. C.R. Acad. Sci. Paris, 281,

835 (1975).

26. J. Barrau, M. Bouchaut, H. Lavayssiere, G. Dousse, and J.

Satge. J. Organomet. Chem. 243, 281 (1983).

27. S. Mazieres, H. Gornitzka, and C. Couret. Polyhedron, 21,

2827 (2002).

Acknowledgements

We thank the Natural Sciences and Engineering Research

Council of Canada (NSERC) for financial support, Teck

1338

Can. J. Chem. Vol. 83, 2005

28. N.N. Zemlyansky, I.V. Borisova, M.G. Kuznetsova, V.N.

Khrustalev, Y.A. Ustynyuk, M.S. Nechaev, V.V. Lunin, J.

Barrau, and G. Rima. Organometallics, 22, 1675 (2003).

29. W.J. Leigh and G.W. Sluggett. J. Am. Chem. Soc. 116, 10468

(1994).

40. K.U. Ingold, J. Lusztyk, and J.C. Scaiano. J. Am. Chem. Soc.

106, 343 (1984).

41. C. Chatgilialoglu and M. Ballestri. Organometallics, 14, 5017

(1995).

42. C. Chatgilialoglu. Chem. Rev. 95, 1229 (1995).

43. S. Bernadoni, M. Lucarini, G.F. Pedulli, L. Valgimigli, V.

Gevorgyan, and C. Chatgilialoglu. J. Org. Chem. 62, 8009

(1997).

44. H. Görner, M. Lehnig, and M. Weisbeck. J. Photochem. Photo-

biol. A, 94, 157 (1996).

30. K. Mochida, H. Kikkawa, and Y. Nakadaira. J. Organomet.

Chem. 412, 9 (1991).

31. M. Ishikawa and M. Kumada. Adv. Organomet. Chem. 19, 51

(1981).

32. M. Kira and T. Miyazawa. In The chemistry of organic silicon

compounds. Edited by Z. Rappoport and Y. Apeloig. John

Wiley and Sons, New York. 1998. p. 1311.

33. M.G. Steinmetz. Chem. Rev. 95, 1527 (1995).

34. M. Ishikawa, T. Fuchikami, T. Sugaya, and M. Kumada. J.

Am. Chem. Soc. 97, 5923 (1975).

45. J. Ferman, J.P. Kakareka, W.T. Klooster, J.L. Mullin, J.

Quattrucci, J.S. Ricci, H.J. Tracy, W.J. Vining, and S. Wallace.

Inorg. Chem. 38, 2464 (1999).

46. A. Castel, P. Riviere, J. Satge, and H.Y. Ko. Organometallics,

9, 205 (1990).

35. W.J. Leigh, R. Boukherroub, and C. Kerst. J. Am. Chem. Soc.

47. A. Castel, P. Riviere, J. Satge, D. Desor, M. Ahbala, and C.

Abdenadher. Inorg. Chim. Acta, 212, 51 (1993).

48. C.D. Campbell and C.W. Rees. J. Chem. Soc. (C), 742 (1969).

49 T. Kitamura, M. Yamane, K. Inoue, M. Todaka, N. Fukatsu, Z.

Meng, and Y. Fujiwara. J. Am. Chem. Soc. 121, 11674 (1999).

50. C. Chatgilialoglu. In CRC handbook of photochemistry. Vol II.

Edited by J.C. Scaiano. CRC Press, Boca Raton, Florida. 1989.

p. 3.

120, 9504 (1998).

36. T.L. Morkin, T.R. Owens, and W.J. Leigh. In The chemistry of

organic silicon compounds. Vol. 3. Edited by Z. Rappoport and

Y. Apeloig. John Wiley and Sons, New York. 2001. p. 949.

37. T.L. Morkin and W.J. Leigh. Acc. Chem. Res. 34, 129 (2001).

38. N.P. Toltl and W.J. Leigh. J. Am. Chem. Soc. 120, 1172 (1998).

39. (a) C. Chatgilialoglu, K.U. Ingold, J. Lusztyk, A.S. Nazran,

and J.C. Scaiano. Organometallics 2, 1332 (1983); (b) K.B.

Clark and D. Griller. Organometallics, 10, 746 (1991).

51. J.C. Scaiano, M. Tanner, and D. Weir. J. Am. Chem. Soc. 107,

4396 (1985).

Article Doi

Time-resolved spectroscopic studies of the photochemistry of some diphenylgermylene (Ph2Ge:) precursors

DOI: 10.1139/v05-148

Source and publish data:

Authors:

Article abstract of DOI:10.1139/v05-148

Full text of DOI:10.1139/v05-148

Products guided by the article

R&D Labs maybe for 878631-66-4

Relevant to this article

Hot Product