Substituent effects on the reactions of diarylgermylenes and tetraaryldigermenes with acetic acid and other lewis bases in hydrocarbon solution

Article abstract of DOI:10.1021/om0609362

A series of three 1,1-diaryl-3,4-dimethylgermacyclopent-3-ene derivatives bearing polar substituents in the para-positions of the 1,1-diphenyl rings have been synthesized, and their photochemistry has been studied by steady state and laser flash photolysis methods. Photolysis in hydrocarbon solvents affords the corresponding diarylgermylenes in chemical and quantum yields similar to those of the parent (1,1-diphenyl) compound, as shown by the results of trapping experiments with methanol (MeOH) and acetic acid (HOAc). The germylenes exhibit UV/vis absorption spectra (λ_max -480-500 nm) and kinetic behavior (τ ≈ 2 μs) similar to that of the parent, diphenylgermylene (GePh₂), in hexane solution. Their decay results in the formation of new transient products assigned to the corresponding tetraaryldigermenes (Ge₂Ar₄), whose absorption spectra (λ_max = 440 nm) and lifetimes are also similar to those exhibited by the parent, Ge₂Ph₄. Both the germylenes and the corresponding digermenes are quenched rapidly by added HOAc and Et₂NH, with rate constants that increase with increasing electron-withdrawing power of the aryl substituents, consistent with reaction mechanisms that involve initial nucleophilic attack at germanium in all cases. The Lewis acid-base complexation of the three germylenes with THF and ethyl acetate (EtOAc) has also been studied and supports a two-step mechanism for the reaction with HOAc that involves initial, rate-determining nucleophilic attack at germanium followed by rapid proton transfer. A similar mechanism is proposed for reaction of this substrate with Ge₂Ph₄ on the basis of the measured Hammett ρ-value. Tetramesityldigermene reacts with HOAc in toluene or THF solution to afford the 1,2-addition product, 1-acetoxy-1,1,2,2-tetramesityldigermane, while no reaction could be detected with Et₂NH.

Full text of DOI:10.1021/om0609362

Organometallics 2007, 26, 1339-1348
1339
Substituent Effects on the Reactions of Diarylgermylenes and
Tetraaryldigermenes with Acetic Acid and Other Lewis Bases in
Hydrocarbon Solution
Lawrence A. Huck and William J. Leigh*
Department of Chemistry, McMaster UniVersity, Hamilton, Ontario L8S 4M1, Canada
ReceiVed October 11, 2006
A series of three 1,1-diaryl-3,4-dimethylgermacyclopent-3-ene derivatives bearing polar substituents
in the para-positions of the 1,1-diphenyl rings have been synthesized, and their photochemistry has been
studied by steady state and laser flash photolysis methods. Photolysis in hydrocarbon solvents affords
the corresponding diarylgermylenes in chemical and quantum yields similar to those of the parent (1,1-
diphenyl) compound, as shown by the results of trapping experiments with methanol (MeOH) and acetic
acid (HOAc). The germylenes exhibit UV/vis absorption spectra (λmax ) 480-500 nm) and kinetic behavior
(τ ≈ 2 µs) similar to that of the parent, diphenylgermylene (GePh2), in hexane solution. Their decay
results in the formation of new transient products assigned to the corresponding tetraaryldigermenes
(Ge2Ar4), whose absorption spectra (λmax ) 440 nm) and lifetimes are also similar to those exhibited by
the parent, Ge2Ph4. Both the germylenes and the corresponding digermenes are quenched rapidly by
added HOAc and Et2NH, with rate constants that increase with increasing electron-withdrawing power
of the aryl substituents, consistent with reaction mechanisms that involve initial nucleophilic attack at
germanium in all cases. The Lewis acid-base complexation of the three germylenes with THF and ethyl
acetate (EtOAc) has also been studied and supports a two-step mechanism for the reaction with HOAc
that involves initial, rate-determining nucleophilic attack at germanium followed by rapid proton transfer.
A similar mechanism is proposed for reaction of this substrate with Ge2Ph4 on the basis of the measured
Hammett F-value. Tetramesityldigermene reacts with HOAc in toluene or THF solution to afford the
1,2-addition product, 1-acetoxy-1,1,2,2-tetramesityldigermane, while no reaction could be detected with
Et2NH.
of singlet carbenes. For example, they all undergo facile
insertion into the O-H bonds of alcohols and carboxylic acids,
form Lewis acid-base complexes (“ylides”) with ethers, amines,
and other Lewis bases, and undergo (2+1)-cycloaddition
reactions with alkynes, alkenes, and dienes.1,6,7 It is generally
recognized that one of the main variables throughout the series,
all else being equal, is a systematic variation in overall reaction
thermodynamics, with reaction exothermicities decreasing down
the group.8-10 One particularly good example of this is the
(2+1)-cycloaddition with alkenes. While cyclopropanes are a
common structural motif in organic chemistry, most of the
known silacyclopropane derivatives enjoy only marginal stabil-
ity, and only two germacyclopropane derivatives have ever been
reported,11 to our knowledge; (2+1)-cycloreversion to generate
Introduction
The chemistry of germylenes has been of great interest for
many years, and a great deal of work has been done toward
defining the scope and mechanisms of many of their charac-
teristic reactions.1 All known examples of these typically highly
reactive, electrophilic species possess singlet ground states,2
which calculations indicate to be more than 20 kcal mol-1 below
the lowest triplet state in the parent molecule (GeH2) and simple
substituted derivatives.3 With only one recent exception,4 the
same is true of their organosilicon counterparts, silylenes,5 whose
chemistry has been even more extensively studied.6
Not surprisingly, the reactivity of transient silylenes and
germylenes shares many common qualitative features with those
(4) Sekiguchi, A.; Tanaka, T.; Ichinohe, M.; Akiyama, K.; Tero-Kuboto,
S. J. Am. Chem. Soc. 2003, 125, 4962.
(5) Apeloig, Y.; Pauncz, R.; Karni, M.; West, R.; Steiner, W.; Chapman,
D. Organometallics 2003, 22, 3250.
(6) For reviews published since 1990 see refs 1c-h and: (a) Weiden-
bruch, M. Coord. Chem. ReV. 1994, 130, 275. (b) Gaspar, P. P.; West, R.
In The Chemistry of Organic Silicon Compounds, Vol. 2; Rappoport, Z.,
Apeloig, Y., Eds.; John Wiley and Sons: New York, 1998; pp 2463-2568.
(c) Belzner, J.; Ihmels, H. AdV. Organomet. Chem. 1999, 43, 1. (d) Hill, N.
J.; West, R. J. Organomet. Chem. 2004, 689, 4165. (e) Ottosson, H.; Steel,
P. G. Chem. Eur. J. 2006, 12, 1576.
(7) Jones, M., Jr.; Moss, R. A. In ReactiVe Intermediate Chemistry; Moss,
R. A., Platz, M. S., Jones, M., Jr., Eds.; John Wiley & Sons: New York,
2004; pp 273-373.
(8) Horner, D. A.; Grev, R. S.; Schaefer, H. F., III. J. Am. Chem. Soc.
1992, 114, 2093.
(9) Sakai, S. Int. J. Quantum Chem. 1998, 70, 291.
(10) (a) Su, M.-D.; Chu, S.-Y. J. Am. Chem. Soc. 1999, 121, 11478. (b)
Su, M.-D. Chem. Eur. J. 2004, 10, 6073.
* Corresponding author. E-mail: leigh@mcmaster.ca.
(1) For reviews see: (a) Neumann, W. P. Chem. ReV. 1991, 91, 311. (b)
Rivie`re, P.; Rivie`re-Baudet, M.; Satge´, J. In ComprehensiVe Organometallic
Chemistry II; Davies, A. G., Ed.; Pergamon: New York, 1995; pp 137-
216. (c) Driess, M.; Gru¨tzmacher, H. Angew. Chem., Int. Ed. Engl. 1996,
35, 828. (d) Weidenbruch, M. Eur. J. Inorg. Chem. 1999, 1999, 373. (e)
Tokitoh, N.; Okazaki, R. Coord. Chem. ReV. 2000, 210, 251. (f) Weiden-
bruch, M. J. Organomet. Chem. 2002, 646, 39. (g) Tokitoh, N.; Ando, W.
In ReactiVe Intermediate Chemistry; Moss, R. A., Platz, M. S., Jones, M.
Jr., Eds.; John Wiley & Sons: New York, 2004; pp 651-715. (h) Kira,
M.; Ishida, S.; Iwamoto, T. Chem. Rec. 2004, 4, 243.
(2) Ko¨cher, J.; Lehnig, M. Organometallics 1984, 3, 937.
(3) (a) 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; pp 749-
839 and references cited therein. (b) Szabados, A.; Hargittai, M. J. Phys.
Chem. A 2003, 107, 4314. (c) Ola´h, J.; De Proft, F.; Veszpre´mi, T.;
Geerlings, P. J. Phys. Chem. A 2004, 108, 490.
10.1021/om0609362 CCC: $37.00 © 2007 American Chemical Society
Publication on Web 02/07/2007

1340 Organometallics, Vol. 26, No. 6, 2007
Huck and Leigh
an alkene and the corresponding metallylene is the prevalent
mode of decomposition of sila- and germacyclopropane deriva-
tives in the absence of secondary trapping reactions.8,12-14
In this paper, we report the synthesis of three new diarylger-
mylene precursors, the characterization of their photochemistry
by steady state and laser flash photolysis techniques, and
absolute rate and/or equilibrium constants for the reactions of
the corresponding diarylgermylenes with tetrahydrofuran (THF),
N,N-diethylamine (Et2NH), ethyl acetate (EtOAc), and acetic
acid (HOAc). The compounds in question, the 1,1-diaryl-3,4-
dimethylgermacyclopent-3-ene derivatives 1b-d (eq 1), were
designed to enable the study of the effects of polar substituents
on various aspects of the reactivity of diphenylgermylene
(GePh2, 2a) in solution. The substituents (para-methyl, -fluoro,
and -trifluoromethyl, respectively) were chosen to provide a
reasonable span in electron-donating/-withdrawing ability through-
out the series, while minimizing the likelihood of any specific
effects on either the photochemical yield of the desired
diarylgermylenes (2b-d) or their kinetic behavior due to
complexation of the strongly electrophilic germylenes with their
precursors. Since the base reaction of GePh2 in solution is
dimerization to form tetraphenyldigermene (3a),16,22,26 the effects
of substituents on the kinetics and mechanisms of selected
reactions of the latter compound, in this case those with HOAc
and Et2NH, can also be studied.
The direct detection and study of transient organogermylenes
in solution by laser flash photolysis methods has been of
considerable interest. In solution, the direct detection of the
dimethyl (GeMe215-18), methylphenyl (GeMePh15b,c,19), diphenyl
(GePh216,20-23), and dimesityl (GeMes2;16,23,24 Mes ) 2,4,6-
trimethylphenyl) derivatives by time-resolved UV/vis spectro-
photometry has been widely reported, using variety of oligoger-
mane, silyl- and disilylgermane, benzo(7-germanorbornadiene),
and germacyclopent-3-ene derivatives as germylene precursors.
These compounds are all well established to undergo photo-
chemical germylene extrusion in moderate to high quantum and/
or chemical yields. However, most are known to afford other
transient photoproducts as well, which has made conclusive
identification of the weakly absorbing germylene of interest
problematic.3a Of these, germacyclopent-3-enes appear to be
uniquely suited as precursors for the study of transient ger-
mylenes in solution by time-resolved UV/vis methods, as
germylene extrusion characteristically proceeds in high quantum
yield and essentially quantitative chemical yield from these
compounds, and without the competing formation of other
(detectable) transient photoproducts in the primary photochemi-
cal event. Compounds of this type have recently allowed the
definitive identification of, and detailed kinetic studies to be
carried out for, all four of the “simple” transient germylene
derivatives listed above,16,17,19b,21,23 affording results that cor-
relate quite well with those of earlier studies of the parent and
dimethyl derivatives in the gas phase.25
The first objective was to define the effects of polar
substituents on the absolute rate and/or equilibrium constants
for the reactions of GePh2 with representative oxygen- and
nitrogen-based nucleophilic substrates such as THF and Et2-
NH. These are simple, well-known Lewis acid-base complex-
ation reactions,27-32 which proceed at rates close to the diffusion
limit in hexane solution;21,23 thus, the substituent effect on the
forward rate constant can be anticipated to be small, if
observable at all. Both reactions are reversible on the time scale
of our experiments because further unimolecular rearrangement
of the complexes to tetravalent product is thermodynamically
and(or) kinetically unfavorable, according to early studies of
the chemistry of hydridogermylamines33 and recent theoretical
calculations.34,35 We thought it would be particularly interesting
to examine the effect of substituents on the equilibrium constant
for the reaction of GePh2 with THF, as it is of the right
magnitude to be measurable under the conditions of our
experiments and might be expected to be somewhat more
(11) Ando, W.; Ohgaki, H.; Kabe, Y. Angew. Chem., Int. Ed. Engl. 1994,
33, 659.
(12) Seyferth, D.; Annarelli, D. C. J. Am. Chem. Soc. 1975, 97, 7162.
(13) (a) Driver, T. G.; Woerpel, K. A. J. Am. Chem. Soc. 2003, 125,
10659. (b) Jiang, P.; Trieber, D. I.; Gaspar, P. P. Organometallics 2003,
22, 2233.
(14) Birukov, A. A.; Faustov, V. I.; Egorov, M. P.; Nefedov, O. M. Russ.
Chem. Bull., Int. Ed. 2005, 54, 2003.
(15) (a) Tomoda, S.; Shimoda, M.; Takeuchi, Y.; Kajii, Y.; Obi, K.;
Tanaka, I.; Honda, K. J. Chem. Soc., Chem. Commun. 1988, 1988, 910.
(b) Wakasa, M.; Yoneda, I.; Mochida, K. J. Organomet. Chem. 1989, 366,
C1. (c) Mochida, K.; Yoneda, I.; Wakasa, M. J. Organomet. Chem. 1990,
399, 53. (d) Mochida, K.; Kanno, N.; Kato, R.; Kotani, M.; Yamauchi, S.;
Wakasa, M.; Hayashi, H. J. Organomet. Chem. 1991, 415, 191. (e) Mochida,
K.; Kikkawa, H.; Nakadaira, Y. J. Organomet. Chem. 1991, 412, 9. (f)
Mochida, K.; Tokura, S. Bull. Chem. Soc. Jpn. 1992, 65, 1642. (g) Mochida,
K.; Ginyama, H.; Takahashi, M.; Kira, M. J. Organomet. Chem. 1998, 553,
163.
(16) Leigh, W. J.; Harrington, C. R.; Vargas-Baca, I. J. Am. Chem. Soc.
2004, 126, 16105.
(17) Leigh, W. J.; Lollmahomed, F.; Harrington, C. R. Organometallics
2006, 25, 2055.
(18) Kaletina, M. V.; Plyusnin, V. F.; Grivin, V. P.; Korolev, V. V.;
Leshina, T. V. J. Phys. Chem. A 2006, 110, 13341.
(19) (a) Mochida, K.; Kimijima, K.; Chiba, H.; Wakasa, M.; Hayashi,
H. Organometallics 1994, 13, 404. (b) Leigh, W. J.; Dumbrava, I. G.;
Lollmahomed, F. Can. J. Chem. 2006, 84, 934.
(26) Massol, M.; Satge´, J.; Rivie`re, P.; Barrau, J. J. Organomet. Chem.
1970, 22, 599.
(27) Rivie`re, P.; Satge´, J.; Castel, A. C. R. Acad. Sci. Paris 1975, 281,
835.
(28) Rivie`re, P.; Castel, A.; Satge´, J. J. Organomet. Chem. 1982, 232,
123.
(29) Barrau, J.; Bouchaut, M.; Lavayssiere, H.; Dousse, G.; Satge´, J. J.
Organomet. Chem. 1983, 243, 281.
(30) Ando, W.; Itoh, H.; Tsumuraya, T. Organometallics 1989, 8, 2759.
(31) Zemlyansky, N. N.; Borisova, I. V.; Kuznetsova, M. G.; Khrustalev,
V. N.; Ustynyuk, Y. A.; Nechaev, M. S.; Lunin, V. V.; Barrau, J.; Rima,
G. Organometallics 2003, 22, 1675.
(20) (a) Konieczny, S.; Jacobs, S. J.; Braddock Wilking, J. K.; Gaspar,
P. P. J. Organomet. Chem. 1988, 341, C17. (b) Mochida, K.; Wakasa, M.;
Nakadaira, Y.; Sakaguchi, Y.; Hayashi, H. Organometallics 1988, 7, 1869.
(c) Bobbitt, K. L.; Maloney, V. M.; Gaspar, P. P. Organometallics 1991,
10, 2772. (d) Mochida, K.; Wakasa, M.; Hayashi, H. Phosphorus, Sulfur
Silicon Rel. Elem. 1999, 150-151, 237.
(21) Leigh, W. J.; Harrington, C. R. J. Am. Chem. Soc. 2005, 127, 5084.
(22) Harrington, C. R.; Leigh, W. J.; Chan, B. K.; Gaspar, P. P.; Zhou,
D. Can. J. Chem. 2005, 83, 1324.
(23) Leigh, W. J.; Lollmahomed, F.; Harrington, C. R.; McDonald, J.
M. Organometallics 2006, 25, 5424.
(24) Toltl, N. P.; Leigh, W. J.; Kollegger, G. M.; Stibbs, W. G.; Baines,
K. M. Organometallics 1996, 15, 3732.
(32) Ola´h, J.; De Proft, F.; Veszpre´mi, T.; Geerlings, P. J. Phys. Chem.
A 2005, 109, 1608.
(33) Rivie`re, P.; Rivie`re-Baudet, M.; Couret, C.; Satge´, J. Synth. React.
Inorg. Met.-Org. Chem. 1974, 4, 295.
(25) Boganov, S. E.; Egorov, M. P.; Faustov, V. I.; Krylova, I. V.;
Nefedov, O. M.; Becerra, R.; Walsh, R. Russ. Chem. Bull., Int. Ed. 2005,
54, 483, and references therein.
(34) Su, M.-D.; Chu, S.-Y. J. Phys. Chem. A 1999, 103, 11011.
(35) Heaven, M. W.; Metha, G. F.; Buntine, M. A. J. Phys. Chem. A
2001, 105, 1185.

Reactions of Diarylgermylenes and Tetraaryldigermenes
Organometallics, Vol. 26, No. 6, 2007 1341
sensitive to polar substituents than the forward rate constant
for the process.
of concentration versus time plots for the starting materials and
products (eq 3). The reactions were monitored by H NMR
1
The second objective was to examine the effects on the O-H
insertion reaction with HOAc. This reaction is closely related
to that with alcohols, which was first discovered by Lappert
and co-workers36 and has since been shown to proceed via initial
Lewis acid-base complexation followed by rate-determining
H-migration,23,30,37 the second step most likely proceeding
catalytically.23,37 Both steps in the sequence are reversible, as
has been known since the work of Satge´ and co-workers on the
chemistry of hydridoalkoxygermanes.26 The primary question
to be answered in the case of the analogous reaction with
carboxylic acids, which in contrast to the reaction with alcohols
proceeds irreversibly,16 is whether the mechanism is similar to
that of the reaction with alcohols (with the carboxylic acid
playing the role of the base in the initial step) or involves
protonation of the germylene in the initial (or perhaps only
kinetically significant) step. Presumably, the two mechanisms
can be expected to lead to quite different responses of the
absolute rate constant to aryl substituent. Our study of this
reaction was further expanded to include an examination of the
Lewis acid-base complexation of 2a-d with ethyl acetate
(EtOAc), to provide additional relevant mechanistic information
on the reaction.
The effects of ring substituents on the rate constants for
quenching of Ge2Ph4 by HOAc and Et2NH are also reported.
These processes occur at measurable rates under the conditions
of our experiments,16 although the products are unknown and
cannot be easily determined using 1a-d as precursors because
the digermene is produced by germylene dimerization; anything
that one might add to react with it reacts with the germylene
faster and quenches the formation of the digermene altogether.
We have thus carried out product studies of the reactions of
the two substrates with tetramesityldigermene (3e), a well-
known, relatively stable tetraaryldigermene that can be synthe-
sized in good yields by low-temperature photolysis of hexa-
mesitylcyclotrigermane (4; eq 2),38-40 as a model for the
evidently rapid reactions of the less hindered tetraaryldigermenes
with these substrates.
spectroscopy as a function of photolysis time over the 0-25%
conversion range, above which the spectra became increasingly
complex owing (presumably) to secondary photolysis of the
primary products. None of the three methoxygermanes survived
either column or gas chromatography of the reaction mixtures
without substantial decomposition, so they were identified by
1
comparison of the H NMR spectra of the crude photolyzates
to that of methoxydiphenylgermane (5a), with additional support
from high-resolution mass spectrometry. Quantum yields for
the formation of 5b-d were determined using the formation of
5a from 1a as a secondary actinometer (Φ ) 0.55 ( 0.0716)
and were found to be the same within experimental error as
that for the parent compound in each case.
Similar results were obtained from photolyses of the three
compounds in hexane containing HOAc (0.4 M), which led to
the formation of DMB and the corresponding acetoxygermanes
6b-d (eq 3) as the major products. The latter showed similar
sensitivities to attempted column or gas chromatographic
isolation as 5a-d, so their identification was also made
spectroscopically using the crude reaction mixtures. However,
1
considerably higher conversions were possible before the H
NMR spectra began to show signs of secondary photolytic
decomposition of the primary products, compared to the
experiments with MeOH as the scavenger. This allowed
additional confirmation of the product structures to be obtained
by 13C NMR and infrared spectroscopy.
Product studies were also carried out on the reaction of
tetramesityldigermene (3e) with HOAc. The digermene was
prepared by photolysis of a 1.6 mM solution of 4 in anhydrous
toluene at ca. -50 °C (eq 2), following the procedure of Baines
and co-workers.40 Addition of HOAc (90 mM) as a cooled
solution in toluene to the resulting solution of 3e at ca. -50
°C, followed by warming to room temperature, resulted in
disappearance of the intense yellow color characteristic of the
digermene over a period of ca. 10 min. The mixture, after
evaporation of solvent and excess HOAc, contained residual 4,
1-hydroxy-1,1,2,2-tetramesityldigermene (7f), and 1-acetoxy-
1,1,2,2-tetramesityldigermane (7e; eq 4), which was isolated and
Results
Compounds 1b-d were synthesized by reaction of 1,1-
dichloro-3,4-dimethylgermacyclopent-3-ene with the corre-
sponding substituted phenylmagnesium bromides16 and were
obtained in analytically pure form after column chromatography
and several recrystallizations from hexanes.
Steady state photolysis of deoxygenated 0.02 M solutions of
1b-d in cyclohexane-d12 containing methanol (0.2 M) afforded
equal amounts of 2,3-dimethylbutadiene (DMB) and the cor-
responding diarylmethoxygermanes (5b-d) in estimated chemi-
cal yields of 77-88%, as determined from the relative slopes
1
identified on the basis of its H and 13C NMR, infrared, and
mass spectra. The yield of 7f depended on the rigor with which
the solvent was dried prior to use and did not change during
column chromatography of the crude reaction mixture. The
compound thus appears to be formed by reaction of 3e with
traces of water present in the solvent;41 lower yields of 7e
(36) Lappert, M. F.; Miles, S. J.; Atwood, J. L.; Zaworotko, M. J.; Carty,
A. J. J. Organomet. Chem. 1981, 212, C4.
(37) Klein, B.; Neumann, W. P.; Weisbeck, M. P.; Wienken, S. J.
Organomet. Chem. 1993, 446, 149.
(38) Masamune, S.; Hanzawa, Y.; Williams, D. J. J. Am. Chem. Soc.
1982, 104, 6136.
relative to 7f were obtained when the reaction was carried out
in THF solution.
(39) Tsumuraya, T.; Sato, S.; Ando, W. Organometallics 1988, 7, 2015.
(40) Samuel, M. S.; Jennings, M. C.; Baines, K. M. J. Organomet. Chem.
2001, 636, 130.
(41) Baines, K. M.; Cooke, J. A.; Dixon, C. E.; Liu, H. W.; Netherton,
M. R. Organometallics 1994, 13, 631.

1342 Organometallics, Vol. 26, No. 6, 2007
Huck and Leigh
Figure 1. Transient UV/vis absorption spectra from laser flash photolysis of dry, deoxygenated solutions of 1d (0.003 M) at 25 °C, (a) in
hexane 30-100 ns (O) and 3.50-3.52 µs (b) after the laser pulse and (b) in hexane containing 4.3 mM THF 115-140 ns (O) and 3.49-
3.54 µs (b) after the laser pulse. The insets show transient absorption profiles recorded at selected wavelengths within the ranges monitored.
No evidence of reaction could be discerned upon addition of
Et2NH to similarly prepared THF solutions of 3e. The same
was true of experiments using n-butylamine as the potential co-
reactant.
maxima of the germylenes appear to be blue-shifted from that
of the unsubstituted derivative (2a, λmax ) 500 nm), with the
largest differences being observed for the para-fluoro (2c, λmax
) 480 nm) and para-trifluoromethyl derivatives (2d, λmax
)
Laser Flash Photolysis Studies. Laser flash photolysis
studies were carried out on flowed, deoxygenated solutions of
1a-d (ca. 0.003 M) in anhydrous hexane using a pulsed KrF
excimer laser (∼25 ns, ∼100 mJ, 248 nm) for excitation. As
with the parent compound (1a),16 laser photolysis of 1b-d
resulted in the formation of two transient species, one that is
formed with the laser pulse (λmax ≈ 300 nm and 480-500 nm;
490 nm); that for 2b appears at λmax ) 495 nm and is not
significantly different from that of 2a. While the spectra were
recorded on reasonably short time scales (3.6 µs full scale) and
those for the germylenes were taken over a very narrow time
window at the peak of the laser pulse, it is difficult to ensure
that there is no distortion in the spectra due to small contribu-
tions from the corresponding digermenes, whose formation is
fastest when the germylene concentration is highest. In an
attempt to check for such contributions, a spectrum of 2c was
recorded in hexane containing 2 mM HOAc, where the lifetime
of the germylene is reduced to ca. 100 ns and formation of the
digermene is quenched almost entirely. The spectrum was found
to be essentially identical to that obtained in the absence of
added acid. The possibility of a spectral shift due to complex-
ation of 2c with the precursor was ruled out on the basis of a
spectrum recorded with a 30 µM solution of 1c in hexane using
an ArF laser (193 nm) for excitation, which again exhibited
λmax ) 480 nm.
τ ≈ 2 µs) and one that is considerably longer-lived (λmax
≈
440 nm; τ ≈ 20 µs) and is formed over a similar time scale to
the decay of the first-formed species. The former are assigned
to the corresponding diarylgermylenes (2b-d), while the
second-formed species are assigned to the corresponding
tetraaryldigermenes (3b-d), formed by dimerization of 2b-d.
Figure 1a shows transient absorption spectra from flash pho-
tolysis of 1d under these conditions, recorded at selected time
intervals after the laser pulse, along with representative growth/
decay traces recorded at monitoring wavelengths of 500 and
440 nm.
The germylene decays, after scaled subtraction of traces
recorded on the same time scale at 440 nm in order to remove
the underlying growth/decay component due to the correspond-
ing digermene absorptions, fit acceptably to second-order decay
kinetics (eq 5) and afforded rate coefficients of kdim/ꢀ500 ) (8
( 2) × 106 cm s-1 for 2b and (2.8 ( 0.8) × 107 cm s-1 for 2c
and 2d. A value of kdim/ꢀ500 ) (9 ( 2) × 106 cm s-1 was
obtained for 2a, which is in acceptable agreement with the value
of kdim/ꢀ530 ) (1.1 ( 0.9) × 107 cm s-1 reported by us earlier
from (similarly corrected) decays recorded at 530 nm,16 given
the difference in extinction coefficients at the two monitoring
wavelengths. The scaling factors used for subtraction of the
underlying digermene absorptions were determined from tran-
sient spectra recorded for hexane solutions of 1b-d containing
4-15 mM THF, as described below. The results suggest that
the absolute rate constants for dimerization of 2c,d are slightly
larger than those for 2a,b, barring large variations in extinction
coefficient throughout the series. A value of kdim ) (1.2 ( 0.2)
× 1010 M-1 s-1 for 2a (hexane; 23 °C) was reported in our
earlier study of the parent diarylgermylene,16 which is within a
factor of 2 of the diffusional limit in hexane at these temper-
atures.
Addition of small amounts of tetrahydrofuran (THF) to the
solutions of 1b-d led to distinct changes in the form of the
transient signals at 500 nm, causing them to develop a distinct
initial rapid decay component superimposed on the growth/decay
profile of the corresponding digermene (3b-d); the latter was
in all cases essentially unaltered in the presence of low
concentrations of THF compared to its form in pure hexane
solution. The initial decay became unresolvable from the laser
pulse in the presence of ∼2 mM THF, as we found previously
for 2a.23 Accompanying this behavior was the appearance of
new transient absorptions centered at λmax ≈ 350 nm in all three
cases, which exhibited a resolvable growth that quickened as
successively higher concentrations of THF were added; the
growth became unresolvable from the laser pulse at roughly
the same THF concentrations as did the germylene decays.
Figure 1b shows transient spectra and representative transient
growth/decay profiles recorded by laser photolysis of the
solution of 1d in the presence of 4.3 mM THF. Transient
absorption spectra recorded for 1b and 1c in the presence of
15.0 and 7.3 mM THF, respectively, are shown in the Supporting
Information along with the spectra in neat hexane solution.
Relative extinction coefficients at 500 and 440 nm (ꢀ500/ꢀ440
)
of 0.150, 0.110, and 0.135 for 3b, 3c, and 3d, respectively, were
determined from the spectra of the digermenes obtained in these
experiments. The behavior exhibited by the three derivatives
in the presence of THF is consistent with reversible scavenging
∆At ) ∆A0/(1 + 2kdim(∆A0/ꢀl)t)
(5)
While the UV/vis spectra of 3b-d are all essentially identical
to that of tetraphenyldigermene (Ge2Ph4, 3a), the absorption

Reactions of Diarylgermylenes and Tetraaryldigermenes
Organometallics, Vol. 26, No. 6, 2007 1343
series of germylenes, if at all. On the other hand, Keq varies
quite significantly, increasing in size as the electron-withdrawing
power of the aryl substituents increases, from a low of Keq
≈
5000 M-1 in the case of 2b to a value in excess of 25 000 M-1
in the case of 2d.
kdecay ) k-Q + kQ[Q]
∆A0/∆Ares ) 1 + Keq[Q]
(7)
(8)
(9)
∆At ) ∆Ares + (∆A0 - ∆Ares) exp(-kdecayt)
Addition of HOAc and Et2NH to hexane solutions of 1b-d
had very different effects on the germylene and digermene decay
profiles; the results were again quite similar to those found in
our previous study of 1a in the presence of these reagents.16
Addition of small amounts of these substrates resulted in marked
decreases in the peak intensities of the digermene signals and
concomitant acceleration of both their initial growth and
subsequent decay, all in proportion to the concentration of added
reagent; the decays fit acceptably to pseudo-first-order kinetics
in the presence of the two substrates over the concentration
ranges in which they could be detected. Decays recorded at 500
nm, where 2b-d are the dominant absorbers, also clearly
accelerated in the presence of the two substrates and followed
good first-order kinetics once the concentrations were high
enough to reduce the peak intensity of the digermene absorptions
(at 440 nm) to less than ca. 25% of their values in the absence
of the scavenger.
Plots of kdecay versus [Q] according to eq 10, where Q is the
substrate and k0 is the hypothetical first-order decay rate constant
at [Q] ) 0, were linear for each of the germylenes and
digermenes, as illustrated in Figure 3 for the experiments with
HOAc as scavenger. The digermene decays fit acceptably to
pseudo-first-order kinetics in the presence of the two substrates
over the concentration ranges in which they could be detected.
Rate constants for reaction of GePh2 (2a) and Ge2Ph4 (3a) by
Figure 2. Corrected transient decay profiles for germylenes 2b
and 2c in hexane containing 0.50 and 0.36 mM THF, respectively,
at 25 °C. The solid lines are the nonlinear least-squares fits of the
data to eq 9.
of 2b-d by the ether to yield the corresponding 2-THF Lewis
acid-base complexes (7; eq 6),23 which exhibit their own
discrete absorptions centered at λmax ≈ 350 nm.
In addition to the behavior noted above, decay profiles for
germylenes 2b and 2c, after correction for the underlying
absorptions due to the corresponding digermenes, exhibited a
bimodal form in the presence of sub-millimolar concentrations
of THF. The decays consisted of a fast initial decay component
and a very slowly decaying residual absorption, similar to that
reported by us previously for 2a in the presence of THF.23 The
fast initial decay component is associated with the approach to
equilibrium with 7, for which the pseudo-first-order rate constant
is given by eq 7, where kQ and k-Q are the absolute rate constants
for the forward and reverse components of the process and Q
is the scavenger (THF). The slowly decaying residual absorption
is associated with that due to free germylene remaining after
equilibration is complete and undergoing slow dimerization; the
equilibrium constant (Keq ) kQ/k-Q) is approximately related
to the initial and residual transient absorbances (∆A0 and ∆Ares,
respectively) as defined by eq 8, where ∆A0 and ∆Ares are
obtained from analysis of ∆At versus time profiles according
to eq 9.23 Corrected decay traces for 2b and 2c in hexane
containing 0.50 and 0.36 mM THF, respectively, are shown in
Figure 2. Nonlinear least-squares fitting of the two decay profiles
afforded decay rate constants of kdecay ) (5.6 ( 0.3) × 106 and
(4.4 ( 0.3) × 106 s-1, respectively, leading to an estimate of
kTHF ≈ 1.1 × 1010 M-1 s-1 in both cases. Similarly, the ∆A0
and ∆Ares values obtained from the analyses afforded estimates
of Keq ≈ 5200 M-1 for 2b and Keq ≈ 20 000 M-1 for 2c.
Germylene 2d appeared to decay completely to the pre-pulse
HOAc were redetermined; the value obtained for 2a (kHOAc
)
(4.4 ( 0.6) × 109 M-1 s-1) matches that reported in our earlier
study,16 while the value for 3a (kHOAc ) (7.0 ( 0.4) × 107
M-1 s-1) is somewhat higher than our previously estimated
upper limit (kHOAc < 4 × 107 M-1 s-1), which was based on a
single decay rate constant obtained at the highest HOAc
concentration that was studied.16 Quenching by acetic acid-Od
(DOAc) was also studied for three of the germylenes (2a, 2b,
and 2c) and two of the digermenes (3b and 3d) and found to
proceed with the same rate constant as HOAc within experi-
mental error in each case. The absolute rate constants obtained
in these experiments are listed in Table 1, along with the
absorption maxima of the germylenes (2a-d) and their THF
complexes (7a-d).
kdecay ) k0 + kQ[Q]
(10)
As was also found for 2a,16 quenching of 2b-d by diethy-
lamine was accompanied by the formation of transient products
absorbing at λmax ≈ 320 nm, which can be assigned to the Ar2-
Ge-Et2NH Lewis acid-base complexes 8 (eq 11). The species
decayed with mixed-order kinetics over several tens of micro-
seconds. As was found for the THF complexes 7, there was no
discernible variation in the position of the absorption maxima
of these species throughout the series.
level in the presence of 0.35 mM THF (consistent with Keq
>
25 000 M-1), with a kdecay value corresponding to kTHF ≈ 1.2
× 1010 M-1 s-1. For comparison, a similar analysis of data
recorded for 2a in hexane containing 0.40 mM THF afforded
estimates of kTHF ≈ 9 × 109 M-1 s-1 and Keq ≈ 16 000 M-1
,
25-30% higher and lower, respectively, than the (more ac-
curate) values determined by us previously from data recorded
at several THF concentrations over the 0.2-0.8 mM range.23
We thus conclude that kTHF varies only slightly throughout the
To assist in the interpretation of the results for HOAc, we
also examined the behavior of GePh2 and the three substituted
derivatives in the presence of ethyl acetate (EtOAc). Steady-

1344 Organometallics, Vol. 26, No. 6, 2007
Huck and Leigh
Figure 3. Plots of the pseudo-first-order rate constants for decay of (a) diarylgermylenes 2b-d and (b) tetraaryldigermenes 3a-d in
hexane at 25 °C in the presence of HOAc.
Table 1. Quantum Yields for Formation of Diarylgermylenes (2a-d), UV/vis Absorption Maxima, and Absolute Rate or
Equilibrium Constants for Reaction of Diarylgermylenes (2a-d) and Tetraaryldigermenes (3a-d) with Acetic Acid (HOAc),
N,N-Diethylamine (Et2NH), and Ethyl Acetate (EtOAc) in Hexane Solution at 25 °C
GeAr2 (2)
Ge2Ar4 (3)
kHOAc
k
kHOAc
(107 M-1 s-1
k
Et2NH
Et2NH
precursor
Φ2
λ
max (nm)
(109 M-1 s-1
)
(109 M-1 s-1
)
KEtOAc (M-1
)
)
(109 M-1 s-1
)
a
1a
1b
1c
1d
0.55 ( 0.07
0.50 ( 0.09
0.47 ( 0.12
0.55 ( 0.12
500
495
480
490
3.9 ( 0.7 (4.4 ( 0.6) b
3.4 ( 0.6 (3.1 ( 0.8) b
4.1 ( 0.8
7.3 ( 0.9
6.9 ( 0.8
7.3 ( 0.4
9.4 ( 1.0
405 ( 7
129 ( 5
460 ( 80
900 ( 25
7.0 ( 0.4
2.4 ( 0.4
3.1 ( 0.3
4.9 ( 0.5
5.8 ( 0.6
4.0 ( 0.5 (4.5 ( 0.8) b
9.0 ( 1.0
6.2 ( 1.0 (7.2 ( 0.9) b
36 ( 6 (28 ( 3)b
a Quantum yields for the formation of methoxygermanes 5a-d from photolysis of 1a-d in C6D12 containing 0.2 M MeOH. The values for 1b,c employed
the photolysis of 1a as secondary actinometer. b Value obtained using acetic acid-Od as scavenger.
state photolysis of a solution of 1a in C6D12 containing EtOAc
(0.3 M) resulted in the formation of DMB and germanium-
containing polymers and showed no evidence for the formation
of products that might be ascribable to trapping of 2a by the
Figure 4. (a) Transient UV/vis absorption spectra recorded by flash
ester. Flash photolysis of hexane solutions of 1a in the presence
photolysis of a deoxygenated 0.003 M solution of 1a in hexane
of 1.5-15 mM EtOAc afforded germylene absorptions whose
containing 0.03 M EtOAc at 25 °C, 100-160 ns (O) and 6.6-6.7
initial intensities decreased systematically with increasing
µs (b) after the laser pulse; the inset shows transient absorption
substrate concentration, but with no change in the form of the
transient decay profiles. No effect of any sort was observed on
profiles recorded at 350 and 440 nm. (b) Plots of (∆A0)0/(∆A0)Q
for quenching of the peak signal intensities due to 2a-d by EtOAc
in hexane at 25 °C and for 2a at 60 °C. The solid lines are the
least-squares fits of the data to eq 12.
the digermene absorptions at 440 nm. New transient absorptions
centered at λmax ) 350 nm were observed to appear, which
increased in initial intensity as that of the germylene signal (at
500 nm) decreased; at no concentration was an initial growth
in the signal resolvable from the laser pulse, however. These
new absorptions decayed over 10-20 µs, concomitantly with
the growth of the digermene absorptions at 440 nm; the latter
were slowed significantly in the presence of EtOAc compared
to the growth rates in the absence of the substrate. Only the
digermene and the 350 nm absorptions could be detected in
the presence of 30 mM EtOAc, as the transient spectra of Figure
4a illustrate. The behavior is consistent with reversible com-
plexation of 2a with EtOAc, under conditions where the
equilibrium constant is sufficiently small and the forward rate
constant sufficiently large that the approach to equilibrium
cannot be detected on the time scale of our experiments.23 The
signal detected for GePh2 at 500 nm in the presence of low
concentrations of EtOAc thus corresponds to free germylene
in equilibrium with the complex and is related to the equilibrium
constant by eq 12, where (∆A0)0 and (∆A0)Q are the initial signal
intensities in the absence and presence of Q (EtOAc, in this
case).23 Similar behavior was observed for each of the ger-
mylene/digermene systems studied in this work, and plots of
(∆A0)0/(∆A0)Q versus [EtOAc] were in each case linear, as
shown in Figure 4b. The species absorbing at 350 nm are
assigned to the Lewis acid-base complexes of germylenes
2a-d with the ester (9a-d; eq 13). The Keq values determined
in these experiments are collected with the other kinetic data
in Table 1.
(∆A0)0/(∆A0)Q ) 1 + Keq[Q]
(12)
Complexation of GePh2 (2a) with EtOAc and THF was also
studied at 60 °C, where the effects of added substrate on the
germylene signal intensities were found to be similar in form
but significantly less pronounced than at room temperature. The
(corrected) germylene decay in the absence of added substrate
was significantly faster at this temperature than at 25 °C, but
followed clean second-order kinetics with a decay coefficient

Reactions of Diarylgermylenes and Tetraaryldigermenes
Organometallics, Vol. 26, No. 6, 2007 1345
of kdim/ꢀ500 ) (2.2 ( 0.3) × 107 cm s-1. Addition of EtOAc
resulted in reductions in initial germylene signal strength in an
analogous manner to those observed at 25 °C, and a plot of
(∆A0)0/(∆A0)Q versus [EtOAc] according to eq 12 afforded a
value of Keq ) 89 ( 7 M-1 for complexation with EtOAc (see
Figure 4b). Decays recorded in the presence of THF were
bimodal and were analyzed as two single-exponential decays
in order to estimate the decay rate constant of the fast initial
component associated with the complexation reaction; the
resulting plot of kdecay versus [THF] according to eq 7 afforded
kQ ) (9 ( 2) × 109 M-1 s-1. The ∆Ares values required for
determination of Keq were estimated as the ∆A values at the
break-points in the bimodal decays, and the resulting plot of
∆A0/∆Ares versus [THF] according to eq 8 afforded a value of
Keq ) 3100 ( 200 M-1 (see Supporting Information).
Figure 5. Hammett plots of the absolute rate and equilibrium
constants for Lewis acid-base complexation of diarylgermylenes
2a-d with Et2NH (kQ; ∆), THF (Keq; ]), and EtOAc (Keq; 0),
and of the absolute rate constants for reaction of 2a-d with AcOH
(kQ; O) in hexane solution at 25 °C. The shaded area in the lower
portion of the graph represents the indeterminable range in log Keq,
where Keq is too large to be measured (i.e., Keq > 25 000 M-1);
the point for 2d lies in this range.
absence of germylene scavengers. Another potential explanation
for the spectral shifts is complexation of the germylene with
its precursor, which is present at a concentration of ca. 3 mM.
This too can be ruled out for 2c, since the spectrum recorded
by 193 nm flash photolysis of a 30 µM solution of 1c was
identical to that obtained with the 248 nm laser at the ca. 100-
fold higher precursor concentration. On the basis of the fact
that the magnitude of the spectral shift correlates roughly with
the inductiVe electron-withdrawing power of the aryl substituents
(as characterized by the σI values for fluoro and trifluorom-
ethyl43), we speculate that the effect may originate in an
inductive effect on the energy of the nonbonding molecular
orbital in GePh2.
The equilibrium constants for complexation of 2a-c with
EtOAc are ca. 50 times smaller than the corresponding ones
for complexation with THF, matching the trends in the solution
phase Brønsted and hydrogen-bond basicities of the two
compounds.44,45 The corresponding thermodynamic parameters
can be estimated from the measured equilibrium constants at
25 and 60 °C, which yields values of ∆H° ≈ -11 ( 2 kcal/
mol and ∆S° ≈ -24 ( 5 cal K-1 mol-1 for complexation with
THF and ∆H° ≈ -8.5 ( 0.6 kcal/mol and ∆S° ≈ -23 ( 2 cal
K-1 mol-1 for complexation with EtOAc, with the gas phase
at 298 K and 1 bar as standard reference state. It is interesting
to note that the Ph2Ge-THF and Ph2Ge-EtOAc complexes
exhibit very similar UV/vis absorption maxima, in keeping with
the similar thermodynamic stabilities of the two species.
Plots of the rate and equilibrium constants for the four
reactions of 2a-d studied in this work versus Hammett
substituent constants43 are collected together in Figure 5.
Excellent correlations are observed for the reactions with HOAc
and Et2NH, affording Hammett F-values of +0.19 ( 0.01 and
+0.10 ( 0.01, respectively, with the smaller value being
obtained for the faster of the two reactions. The largest F-values
are exhibited by the complexation equilibria with THF (F )
Discussion
The results demonstrate that polar substituents attached to
the phenyl rings of GePh2 have small but distinct effects on the
rate and equilibrium constants for reaction of the species with
nucleophilic substrates. The particular reactions studied here are
all very fast, but they all appear to be accelerated by remote
electron-withdrawing substituents attached to the phenyl rings.
Distinct effects are also observed on the lowest energy UV/vis
absorption maximum of the species, which is associated with
the n,p electronic transition in the molecule.1g,42 In contrast, the
same substituents have little effect on the UV/vis spectra of
the corresponding tetraaryldigermenes or on the course or
efficiency of the photochemistry of the 1,1-diarylgermacyclo-
pent-3-ene precursors. Again though, regular effects are ob-
served on the reactivity of the digermenes, particularly in the
case of reaction with HOAc, which is the intrinsically slower
of the two reactions that have been studied.
The apparent blue shift in λmax exhibited by the para-methyl-
substituted diarylgermylene (2b) relative to that of the unsub-
stituted compound (2a) is too small to be considered really
significant, but those for the para-CF3 (2d; λmax ) 490 nm)
and para-F (2c; λmax ) 480 nm) derivatives are well outside
the limits of error in our measurements. One potential problem
in pinpointing the absorption maxima of these compounds
precisely is the possibility that the absorption bands are distorted
due to small contributions from the corresponding digermenes,
as their formation is quite rapid during the first 50-100 ns after
the laser pulse, making it difficult to avoid bleed-through of
the product transient absorptions in the earliest time window
after the laser pulse that can be monitored in our experiments.
This can be minimized by working at lower laser intensities in
order to reduce the initial germylene concentration and slow
the dimerization reaction or by recording spectra at even shorter
time scales in the presence of an irreversible scavenger (such
as HOAc) at concentrations high enough to eliminate digermene
formation altogether. The spectrum of 2c was re-recorded under
both these sets of conditions, but showed no discernible change
from that recorded with higher laser intensities and in the
(43) Hansch, C.; Leo, A.; Taft, R. W. Chem. ReV. 1991, 91, 165.
(44) Arnett, E. M.; Mitchell, E. J.; Murty, T. S. S. R. J. Am. Chem. Soc.
1974, 96, 3875.
(45) Spencer, J. N.; Grushow, A.; Ganunis, T. F.; Allott, K. N.; Kneizys,
S. P.; Willis, H.; Puppala, S.; Salata, C. M.; Zafar, A. I.; Stein, B. J.; Hahn,
L. C. J. Solution Chem. 1989, 18, 471.
(42) Ando, W.; Tsumuraya, T.; Sekiguchi, A. Chem. Lett. 1987, 317.

1346 Organometallics, Vol. 26, No. 6, 2007
Huck and Leigh
+1.3 ( 0.2; three data points) and EtOAc (F ) +0.51 ( 0.18;
four data points). The substantial substituent effect on the
equilibrium constants for complexation with THF contrasts
sharply with the minimal variation in the forward rate constants
for the process, which results from the fact that formation of
the complex occurs at rates close to the diffusion limit. The
same limitation applies to the rate constants for complexation
with Et2NH, which is also reversible but has equilibrium
constants that are too large to be measured under the conditions
of our experiments.16 The germylene must clearly play the role
of an electrophile in these complexation reactions, and thus the
positive F-values that characterize them confirm the expectation
that this aspect of germylene reactivity should be enhanced by
the presence of electron-withdrawing substituents in the mol-
ecule.
reversible reactions are characterized in kinetic experiments of
this type by the detection of both the approach to equilibrium
(which provides a measure of the forward rate constant kQ) and
the residual quantity of limiting reactant that remains after the
approach to equilibrium is complete (which provides a measure
of the equilibrium constant Keq, given the ability to measure
the quantity of limiting reactant at the beginning of the approach
to equilibrium as well). The fact that only the latter could be
detected in our experiments means that kQ must be on the order
of 109 M-1 s-1 or greater, given the maximum time resolution
of our instrument and the magnitude of Keq that characterizes
the reaction.23 The behavior in the presence of EtOAc contrasts
that observed with Et2NH and other amines, which also react
reversibly with GePh2 in hexane solution; in these cases the
equilibrium constants are relatively large (Keq > ca. 2.5 × 104
M-1), and thus only the forward rate constants for complexation
can be measured.16,21 The intermediate (“ideal”) situation, where
both kQ and Keq can be measured, is obtained for the reactions
of 2a and 2b with THF, as the data of Figure 2 illustrate. The
same behavior has been reported previously for the reactions
of GePh2 with methanol, tert-butanol, terminal alkenes, and
isoprene.16,23
Hammett plots of the rate constants for quenching of
digermenes 3a-d by Et2NH and HOAc are shown in Figure 6.
As can be seen from the data, an excellent correlation is
observed for the reaction with HOAc, for which the rate
constants vary over the range (4-36) × 107 M-1 s-1 throughout
the series of compounds studied and lead to a F-value of +0.33
( 0.01. The rate constants for quenching by the amine, which
are within a factor of 10 of the diffusion-controlled rate in
hexane at 25 °C, vary by a factor of only ca. 2 throughout the
series of compounds and correlate relatively poorly with
σ-values (F ) +0.11 ( 0.07). As found for the reactions of
these substrates with 2a-d, the results indicate that the effects
of remote polar substituents are relatively small in both cases,
but are consistent with mechanisms in which the substrate plays
the role of nucleophile in the rate-determining step of the
quenching process in both cases.
The rapid reactions of these substrates with the precursors to
3a-d (i.e., germylenes 2a-d) preclude the formation of
digermene trapping products in tractable yields under the
conditions of steady state photolysis experiments with 1a-d.
Nevertheless, the most likely course of the reactions with HOAc
can be inferred from the identity of the product obtained from
reaction of tetramesityldigermene (3e) with this substrate: the
formal 1,2-addition product 6e (Vide supra). It is interesting to
note that the latter reaction is on the order of 9 orders of
magnitude slower than those of the less sterically hindered
derivatives 3a-d; a ballpark estimate of kHOAc ≈ 10-1 M-1
s-1 can be derived from the time required for disappearance of
the characteristic yellow color due to 3e in the presence of ca.
0.01 M HOAc in THF at room temperature, which is consistent
with the upper limit of ca. 102 M-1 s-1 that was estimated by
flash photolysis in our previous kinetic study of 3e in hexane
solution.16
The rate constants for reaction of 2a-d with HOAc exhibit
a sensitivity to substituents similar to the rate and equilibrium
constants for Lewis acid-base complexation with the ether,
amine, and ester, which is clearly consistent with a mechanism
in which the carboxylic acid plays the role of nucleophile in
the rate-determining step. A mechanism involving initial, rate-
determining complexation of the germylene with the carboxylic
acid, followed by rapid migration of the acidic proton to
germanium to complete the reaction (eq 14), is consistent with
the results. The lack of a significant isotope effect on the rate
constants for reaction of 2b and 2d with HOAc and DOAc also
supports this mechanism, though much less conclusively because
the rates are so fast that little effect would be expected even if
H-transfer were involved in the rate-controlling step of the
reaction. Thus, even though it is relatively small, the observed
substituent effect provides mechanistic information that cannot
be readily obtained in other ways. If the rate constant for the
second (proton migration) step is fast relative to reversion of
the complex to free reactants (i.e., k2 >> k-1), then the overall
rate constant for the reaction will be defined by the rate constant
for complexation, i.e., kHOAc ≈ k1. If we assume that the
equilibrium constant for complexation of the germylene with
EtOAc provides a reasonable model for that associated with
the initial step in the reaction with HOAc, then an estimate of
k-1 ≈ 1 × 107 s-1 is obtained for the rate constant for reversion
of the complex to the free reactants. This in turn sets a lower
limit of ca. 108 s-1 for k2, the rate constant for (intramolecular)
H-migration in the complex of GePh2 with HOAc. This can be
compared to our recent estimate of k2 ≈ 104 s-1 for the rate
constant of the analogous process in the zwitterionic intermediate
involved in the O-H insertion reaction of GePh2 with metha-
nol.23 In the latter case, intramolecular proton transfer is
sufficiently slow that the intermediate Lewis acid-base complex
can be detected directly by laser flash photolysis methods.
Complexation is the initial step in the reaction with both aliphatic
alcohols and carboxylic acids and represents the rate-determining
step for decay of the germylene in both cases. It is also the
rate-determining step for product formation in the reaction with
HOAc. In the case of alcohol insertions, where H-migration in
the intermediate complex is slowed by both lower acidity and
tighter geometric requirements in the transition state, the
H-migration step represents the rate-determining step for forma-
tion of the final product.
The corollary to this conclusion is that the Lewis acid-base
complexation of GePh2 with HOAc and EtOAc both proceed
very rapidly, with rate constants that are within a factor of ∼10
of the diffusional limit in hexane solution. The behavior
exhibited by 2a-d in the presence of EtOAc is in fact consistent
with a forward rate constant of this magnitude.23 Ideally,
Rate constants for reaction of two of the substituted di-
germenes with acetic acid-Od have been determined and found

Reactions of Diarylgermylenes and Tetraaryldigermenes
Organometallics, Vol. 26, No. 6, 2007 1347
more slowly than by Et2NH,21 and no new transient absorptions
that might be assigned to such a species are evident in flash
photolysis experiments with 1a-d in the presence of the
secondary amine. The failure of 3e to react with either diethyl-
or n-butylamine, coupled with the observation of near diffusion-
controlled quenching of 3a-d by the secondary amine, indicates
that the influence of steric effects on this reaction is much greater
than is the case with HOAc addition. Further study of the
reaction will clearly be necessary in order to understand these
results more completely.
Summary and Conclusions
The effects of polar ring substituents on the rate constant for
the O-H insertion reaction of diphenylgermylene (GePh2, 2a)
with acetic acid (HOAc) in hexane solution have been deter-
mined and compared to those on the rate and(or) equilibrium
constants for Lewis acid-base complexation of the germylene
with THF, EtOAc, and Et2NH under similar conditions. Each
of these reactions proceed at rates within a factor of ∼5 of the
diffusion limit in hexane at 25 °C, and thus the rate constants
are relatively insensitive to remote substitution with polar
substituents on the phenyl rings. Nevertheless, small but discrete
effects are observed, and the rate constants for reaction of the
four germylenes with HOAc and Et2NH correlate with Hammett
substituent constants, affording F-values of +0.19 ( 0.01 and
+0.10 ( 0.01, respectively. The equilibrium constants for
complexation with THF and EtOAc in hexane are affected much
more strongly by aryl substituents, with the data affording
equilibrium F-values of +1.3 ( 0.2 and +0.5 ( 0.2, respec-
tively. The positive F-values are consistent with mechanisms
in which the germylene plays the role of electrophile in the
kinetically significant step in every case. On the basis of these
results, a mechanism for HOAc insertion involving initial, rate-
determining Lewis acid-base complexation of the germylene
at the carbonyl oxygen followed by fast proton transfer is
proposed.
Figure 6. Hammett plots of the absolute rate constants for
quenching of tetraaryldigermenes 3a-d with Et2NH (∆) and HOAc
(O) in hexane solution at 25 °C.
to be the same within experimental error as those for reaction
with the protiated carboxylic acid in both cases. For rate
constants on the order of 108 M-1 s-1, the lack of a kinetic
isotope effect is significant and rules out both a concerted
mechanism and a stepwise one initiated by protonation; the latter
is also ruled out by the positive Hammett F-value. A mechanism
involving rate-determining nucleophilic addition to form the
zwitterionic addition product, which proceeds to the final
product by fast H-migration (eq 15), is consistent with all of
the data. The mechanism of eq 15 is related to the one proposed
by Baines and co-workers for the reaction of 3e with aliphatic
aldehydes.46
Both HOAc and Et2NH also quench the lifetimes of the
corresponding tetraaryldigermenes (3a-d), which are formed
by dimerization of the diarylgermylenes and can be monitored
directly at low substrate concentrations. Quenching by HOAc
most likely proceeds via 1,2-addition, based on the identity of
the product formed from reaction of the carboxylic acid with
tetramesityldigermene (3e). The second-order rate constants for
quenching of 3a-d by HOAc again correlate with Hammett
substituent constants, leading to a reaction constant of F ) +0.33
( 0.07. This and the lack of a significant kinetic isotope effect
on the reaction are consistent with a mechanism involving
nucleophilic attack of the substrate at the GedGe bond in the
initial (rate-determining) step, to form a zwitterionic intermediate
that proceeds to the overall 1,2-addition product by fast proton
transfer from oxygen to germanium. The reaction is exquisitely
sensitive to steric effects, as shown by the ca. 109-fold lower
rate constant for reaction of HOAc with tetramesityldigermene
(3e) compared to that measured for the parent tetraaryldigermene
(Ge2Ph4; 3a). An even greater steric effect is evident in the
reaction of tetraaryldigermenes with primary and secondary
amines, which proceeds at close to the diffusion-controlled rate
in the case of 3a but to no detectable extent whatsoever in the
case of the tetramesityl derivative.
In view of the indication from our kinetic experiments of a
very rapid quenching interaction between the transient tetraaryl-
digermenes (3a-d) and Et2NH, we were somewhat surprised
to find no evidence for reaction of either Et2NH or n-butylamine
with tetramesityldigermene (3e), even after 2 days in THF
solution at 4 °C. We thus have no information on the likely
identity of the product(s) formed upon quenching of 3a-d by
the amine. To our knowledge there are no known examples of
the reaction, and the only potentially relevant example known
in disilene chemistry is that of ammonia with a (sterically
stabilized) 1,3-tetrasilabutadiene derivative.47 The small positive
F-value suggests that the first (and possibly only) step in the
reaction involves the amine acting as a nucleophile, perhaps to
form a zwitterionic species analogous to the proposed interme-
diate for HOAc addition (see eq 15). It remains unclear whether
the primary product of the reaction with 3a-d proceeds further
to form the 1,2-addition product that might be expected or
simply moderates oligomerization of the digermene as a result
of it being formed reversibly. In this regard, it should be noted
that 3a is quenched by the tertiary amine Et3N at least 100 times
Mechanistic studies of several other characteristic reactions
of transient diarylgermylenes in solution are in progress.
Experimental Section
(46) Samuel, M. S.; Baines, K. M. J. Am. Chem. Soc. 2003, 125, 12702.
(47) Boomgaarden, S.; Saak, W.; Weidenbruch, M.; Marsmann, H. Z.
Anorg. Allg. Chem. 2001, 627, 349.
Tetrahydrofuran (Caledon) and hexanes were dried by passage
through activated alumina under nitrogen using a Solv-Tek solvent

1348 Organometallics, Vol. 26, No. 6, 2007
Huck and Leigh
purification system. Glacial acetic acid (Caledon) and acetic acid-d
(Aldrich) were used as received from the suppliers. N,N-Diethy-
lamine (Aldrich) was refluxed over KOH and distilled. Ethyl acetate
(Caledon) was dried by passage through activated silica gel. 3,4-
Dimethyl-1,1-diphenylgermacyclopent-3-ene (1a)16 and hexamesi-
tylcyclotrigermane (4)48 were prepared as previously described.
Details of the synthesis of the other compounds reported in this
work are described in the Supporting Information.
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) mixtures, and a Luzchem Research
mLFP-111 laser flash photolysis system, modified as described
previously.16 Solutions were prepared at concentrations such that
the absorbance at the excitation wavelength was between ca. 0.7
and 0.9 and were pumped continuously through a vacuum oven-
dried, thermostated 7 × 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, using a Masterflex
77390 peristaltic pump fitted with Teflon tubing (Cole-Parmer
Instrument Co.). The glassware, sample cell, and transfer lines were
dried in a vacuum oven at 65-85 °C before use. Solution
temperatures were measured with a Teflon-coated copper/constantan
thermocouple inserted into the thermostated sample compartment
in close proximity to the sample cell. Reagents were added directly
to the reservoir by microliter syringe as aliquots of standard
solutions. 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. Rate constants
were calculated by linear least-squares analysis of decay rate-
concentration data (generally 4-7 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.
Acknowledgment. We thank the Natural Sciences and
Engineering Research Council of Canada for financial support,
Teck-Cominco Metals Ltd. for a generous gift of germanium
tetrachloride, and the McMaster Regional Centre for Mass
Spectrometry for high-resolution mass spectra. L.A.H. thanks
the Province of Ontario for an Ontario Graduate Scholarship.
Supporting Information Available: Details of the preparation
and characterization of compounds, representative transient decay
and growth/decay profiles, time-resolved UV/vis absorption spectra,
and details of kinetic analyses. This material is available free of
charge via the Internet at http://pubs.acs.org.
(48) Tsumuraya, T.; Kabe, Y.; Ando, W. J. Organomet. Chem. 1994,
482, 131.
OM0609362