Arrhenius parameters for the head-to-tail dimerizations of 1,1-diphenylsilene and 1,1-diphenylgermene in solution

Article abstract of DOI:10.1021/om0103409

Arrhenius parameters for the head-to-tail dimerization of 1,1-diphenylsilene and 1,1-diphenylgermene have been determined in hexane solution by nanosecond laser flash photolysis. Both compounds exhibit near-zero activation barriers over the range 0-60°C a

Full text of DOI:10.1021/om0103409

Organometallics 2001, 20, 4537-4541
4537
Ar r h en iu s P a r a m eter s for th e Hea d -to-Ta il
Dim er iza tion s of 1,1-Dip h en ylsilen e a n d
1
,1-Dip h en ylger m en e in Solu tion
Tracy L. Morkin and William J . Leigh*
Department of Chemistry, McMaster University, 1280 Main Street West,
Hamilton, Ontario, Canada L8S 4M1
Received April 25, 2001
Arrhenius parameters for the head-to-tail dimerization of 1,1-diphenylsilene and 1,1-
diphenylgermene have been determined in hexane solution by nanosecond laser flash
photolysis. Both compounds exhibit near-zero activation barriers over the range 0-60 °C
9
10
-1 -1
and preexponential factors of ca. 10 and 10
M
s , respectively. While the results do not
allow one to distinguish conclusively between concerted and stepwise mechanisms for the
reaction, they provide an indication of what the kinetic behavior of the putative reaction
intermediates must be if dimerization proceeds nonconcertedly.
-5
In tr od u ction
dition, or head-to-head ene-dimerization.1 Theoretical
calculations suggest that the regiochemistry of the
reaction is dictated by the electronic nature of the
substituents at carbon and silicon and their effect on
The chemistry of silicon-carbon and germanium-
carbon double bonds has been the subject of considerable
1
-6
interest over the past ∼30 years.
As might be
δ+
δ-
16,17
the polarity of the SidC bond.
Similar factors
appear to control the regiochemistry of dimerization of
expected, the chemistry of silenes and germenes share
many common features; for example, they both exhibit
potent reactivity as electrophiles, as dienophiles, and
toward dimerization, and most of their reactions are
highly regioselective. Much of the recent mechanistic
work on their rich reactivity has been devoted to the
study of nucleophilic additions, and the mechanisms of
these reactions are now reasonably well understood, at
22,23
germenes.
Head-to-tail dimerization is most common, as this is
the preferred regiochemistry for relatively polar, elec-
trophilic silenes such as 1,1-dimethylsilene (2a ). The
kinetics of dimerization of 2a have been studied in the
gas phase by two groups, using the gas-phase thermoly-
sis or photolysis of 1,1-dimethylsilacyclobutane (1a ) to
7
-13
least for silenes.
Relatively little experimental work
24,25
generate the silene (eq 1).
Gusel’nikov and co-
has been done on the kinetics and mechanism of the
dimerization of silenes and germenes, despite the
considerable interest that theoreticians have shown in
-15
3
workers determined a value of kdim ) 5.9 × 10
cm
-1
s
(corresponding to a solution-phase value of only
6
-1 -1
3
.55 × 10 M
s ) and found the rate constant to be
1
4-22
these reactions.
temperature independent over the 25-300 °C range
3
-1
26
Silenes are known to dimerize in three ways: head-
to-tail [2+2]-cycloaddition, head-to-head [2+2]-cycload-
(E ) 0 ( 1 kcal/mol; log(A/cm s ) ) -14.2). Potz-
a
inger and co-workers later reexamined the dimerization
of 2a using laser flash photolysis techniques and
-
11
3
-1
(
1) Gusel’nikov, L. E.; Nametkin, N. S.; Vdovin, V. M. Acc. Chem.
Res. 1975, 8, 18.
2) Raabe, G.; Michl, J . In The Chemistry of Organic Silicon
reported a value of kdim ) (3.3 ( 0.8) × 10
cm s at
25
∼
298 K. This corresponds to a solution phase rate
(
1
0
-1 -1
Compounds; Patai, S., Rappoport, Z., Eds.; J ohn Wiley & Sons: New
York, 1989; pp 1015-1142.
constant of 2 × 10
M
s
and is more reasonable
(
(
3) Brook, A. G.; Brook, M. A. Adv. Organomet. Chem. 1996, 39, 71.
4) Baines, K. M.; Stibbs, W. G. Adv. Organomet. Chem. 1996, 39,
(16) Apeloig, Y.; Karni, M. J . Am. Chem. Soc. 1984, 106, 6676.
(17) Morokuma, K. In Organosilicon and Bioorganosilicon Chem-
istry: Structure, Bonding, Reactivity and Synthetic Application; Saku-
rai, H., Ed.; Ellis Horwood Publishers: Chichester, 1985; p 37.
(18) Seidl, E. T.; Grev, R. S.; Schaefer, H. F., III. J . Am. Chem. Soc.
1992, 114, 3643.
2
75.
5) Muller, T.; Ziche, W.; Auner, N. In The Chemistry of Organic
(
Silicon Compounds; Rappoport, Z., Apeloig, Y., Eds.; J ohn Wiley &
Sons Ltd.: New York, 1998; pp 857-1062.
(
(
6) Barrau, J .; Escudi e´ , J .; Satg e´ , J . Chem. Rev. 1990, 90, 283.
7) Kira, M.; Maruyama, T.; Sakurai, H. J . Am. Chem. Soc. 1991,
(19) Bernardi, F.; Bottoni, A.; Olivucci, M.; Robb, M. A.; Venturini,
A. J . Am. Chem. Soc. 1993, 115, 3322.
1
13, 3986.
(20) Bernardi, F.; Bottoni, A.; Olivucci, M.; Venturini, A.; Robb, M.
A. J . Chem. Soc., Faraday Trans. 1994, 90, 1617.
(21) Venturini, A.; Bernardi, F.; Olivucci, M.; Robb, M. A.; Rossi, A.
J . Am. Chem. Soc. 1998, 120, 1912.
(
(
(
8) Wiberg, N. J . Organomet. Chem. 1984, 273, 141.
9) Leigh, W. J . Pure Appl. Chem. 1999, 71, 453.
10) Leigh, W. J .; Boukherroub, R.; Kerst, C. J . Am. Chem. Soc.
1
998, 120, 9504.
11) Leigh, W. J .; Kerst, C.; Boukherroub, R.; Morkin, T. L.; J enkins,
S.; Sung, K.; Tidwell, T. T. J . Am. Chem. Soc. 1999, 121, 4744.
12) Leigh, W. J .; Toltl, N. P.; Apodeca, P.; Castruita, M.; Pannell,
K. H. Organometallics 2000, 19, 3232.
(22) Kudin, K. N.; Margrave, J . L.; Khabashesku, V. N. J . Phys.
Chem. A 1998, 102, 744.
(23) Barton, T. J .; Hoekman, S. K. J . Am. Chem. Soc. 1980, 102, 2,
1584.
(24) Gusel’nikov, L. E.; Flowers, M. C. J . Chem. Soc., Chem.
Commun. 1967, 864.
(25) Brix, T.; Bastian, E.; Potzinger, P. N. J . Photochem. Photobiol.
A: Chem. 1989, 49, 287.
(
(
(13) Toltl, N. P.; Leigh, W. J . J . Am. Chem. Soc. 1998, 120, 1172.
(14) Ahlrichs, R.; Heinzmann, R. J . Am. Chem. Soc. 1977, 99, 7452.
(15) Morokuma, K.; Kato, S.; Kitaura, K.; Obara, S.; Ohta, K.;
Hanamura, S. In New Horizons of Quantum Chemistry; Lowdin, P.-
O., Pullman, R., Eds.; D. Reidel: Dordrecht, 1983; pp 221-241.
(26) Gusel’nikov, L. E.; Konobeyevsky, K. S.; Vdovin, V. M.; Namet-
kin, N. S. Dokl. Akad. Nauk. SSSR (Engl. transl.) 1977, 235, 791.
1
0.1021/om0103409 CCC: $20.00 © 2001 American Chemical Society
Publication on Web 09/26/2001

4
538 Organometallics, Vol. 20, No. 22, 2001
Morkin and Leigh
considering the very high reactivity that 2a exhibits
It is also consistent with a more
recent determination of the rate constant for head-to-
the stabilized germene 9b dimerizes in head-to-head
toward nucleophiles.1
0,27
fashion, the same as its silicon counterpart 9a (eq 5).
34,35
Except for one example from our group (vide infra),36
no kinetic studies of germene dimerization have yet
been reported.
tail dimerization of 1-methylsilene (kdim ) (2.4 ( 0.5) ×
-11
3
-1
28
1
0
cm s at 300 K). While there is indirect support
from other studies for the original conclusion that the
2
9
rate of dimerization of 2a is temperature independent,
the problem has never been reexamined by direct
methods.
Head-to-head regiochemistry is favored for silenes of
relatively low SidC bond polarity and electrophilicity.
For example, the 1,1-bis(trimethylsilyl)silene 5, formed
in the photolysis of acylpolysilane 4, dimerizes to give
two products: the 1,2-disilacyclobutane 6 and the ene-
dimer 7 (eq 2).30 The two products were proposed to
result either from concerted [2+2]-cycloaddition and
ene-reaction, respectively, or via a stepwise pathway
involving the common 1,4-biradical intermediate 8 (eq
High-level theoretical studies of the mechanism of the
head-to-tail dimerization of the parent silene (H2Sid
CH2, 11) have led to conflicting conclusions. Schaefer
and co-workers proposed a concerted (π2s+π2s) mech-
anism on the basis of calculations carried out at the
CCSD (DZ+d) level of theory, in which reaction proceeds
via a C2h-symmetric transition state that is 3.8 kcal/
3
). The kinetics of dimerization of 5 in cyclohexane
18
mol higher in energy than the reactants. Two later
solution have been studied by Conlin and co-workers,
studies by Bernardi and co-workers concluded in favor
7
-1
who reported a rate constant of kdim ) 1.3 × 10 M
19,21
of stepwise mechanisms for the reaction.
In the first
-
1
s
at 293 K and the Arrhenius parameters Ea ) 0.2 (
study, CAS-SCF calculations, with geometries optimized
at the DZ+d or 3-21G* levels of theory, predicted the
reaction to proceed via the initial, effectively irreversible
formation of the trans-1,3-disilabutane-1,4-diyl biradical
-
1
-1
31
0
.1 kcal/mol and log(A/M
s ) ) 7 ( 1. These values
are consistent with either concerted or stepwise dimer-
ization mechanisms, provided that the formation of the
biradical intermediate (8) by the latter pathway is
reversible.
1
2 (eq 6). The biradical then undergoes bond rotation
and coupling to yield the head-to-tail dimer (13) through
1
9
an activation barrier of 2.7 kcal/mol. In the second,
carried out at the CASPT2/6-311G*//CAS-SCF/6-31G*
level of theory, the reaction was predicted to proceed
via the initial formation of the trans-2,3-disilabutane-
1
,4-diyl biradical 14, of energy ∼5.3 kcal/mol lower than
the reactants. Collapse of the biradical to 13 by 1,2-
sigmatropic shift (eq 10) then proceeds through an
2
1
activation barrier of ca. 4.2 kcal/mol, via a transition
state similar to that originally proposed by Schaefer and
co-workers for the concerted dimerization pathway.18
Only one theoretical study of germene dimerization has
2
2
been reported, in which a concerted mechanism analo-
1
8
gous to that of Seidl et al. was proposed.
In this paper, we report the first determination of the
Arrhenius parameters for the head-to-tail dimerization
of a transient silene and its germanium homologue in
solution. 1,1-Diphenylsilene (16a ) and 1,1-diphenylger-
mene (16b), generated by photolysis of the correspond-
ing 1,1-diphenylmetallacyclobutanes 15a ,b, are both
known to dimerize to give the head-to-tail cycloadducts
As might be expected, the same bond polarity factors
appear to govern the regiochemistry of germene dimer-
ization. For example, 1,1-dimethylgermene (2b) dimer-
izes with head-to-tail regiochemistry to yield 1,1,3,3-
3
6-38
1
7a ,b (eq 8).
We have previously shown that the
dimerizations of these two compounds proceed with
1
0
-1 -1
tetramethyl-1,3-digermacyclobutane 3b (eq 4),3
2,33
absolute rate constants on the order of 10
M
s
in
while
hexane solution at room temperature, the fastest of any
(
27) Kerst, C.; Boukherroub, R.; Leigh, W. J . J . Photochem. Photo-
biol. A: Chem. 1997, 110, 243.
28) Vatsa, R. K.; Kumar, A.; Naik, P. D.; Upadhyaya, H. P.;
Pavanaja, U. B.; Saini, R. D.; Mittal, J . P.; Pola, J . Chem. Phys. Lett.
996, 255, 129.
29) Bastian, E.; Potzinger, P.; Ritter, A.; Schuchmann, H.-P.; von
Sonntag, C.; Weddle, G. Ber. Bunsen-Ges. Phys. Chem. 1980, 84, 56.
(33) Khabashesku, V. N.; Kudin, K. N.; Tamas, J .; Boganov, S. E.;
Margrave, J . L.; Nefedov, O. M. J . Am. Chem. Soc. 1998, 120, 5005.
(34) Bravo-Zhivotovskii, D.; Zharov, I.; Kapon, M.; Apeloig, Y. J .
Chem. Soc., Chem. Commun. 1995, 1625.
(35) Bravo-Zhivotovskii, D.; Braude, V.; Stanger, A.; Kapon, M.;
Apeloig, Y. Organometallics 1992, 11, 2326.
(
1
(
(
(
30) Baines, K. M.; Brook, A. G. Organometallics 1987, 6, 692.
31) Zhang, S.; Conlin, R. T.; McGarry, P. F.; Scaiano, J . C.
(36) Toltl, N. P.; Stradiotto, M. J .; Morkin, T. L.; Leigh, W. J .
Organometallics 1999, 18, 5643.
Organometallics 1992, 11, 2317.
32) Namavari, M.; Conlin, R. T. Organometallics 1992, 11, 3307.
(37) Auner, N.; Grobe, J . J . Organomet. Chem. 1980, 197, 147.
(38) J utzi, P.; Langer, P. J . Organomet. Chem. 1980, 202, 401.
(

Dimerizations of 1,1-Diphenylsilene and -germene
Organometallics, Vol. 20, No. 22, 2001 4539
F igu r e 1. Arrhenius plots for dimerization of 1,1-diphenylsilene (16a ) and 1,1-diphenylgermene (16b) in hexane. The
dashed lines and associated solid points represent the temperature dependence of the diffusional rate constant in hexane,
calculated using the modified Debye equation (kdiff ) 8RT/3000η) and published temperature-viscosity data.
Ta ble 1. Ra te Con sta n ts a n d Ar r h en iu s
P a r a m eter s for Dim er iza tion of 1,1-Dip h en ylsilen e
(
16a ) a n d 1,1-Dip h en ylger m en e (16b) in
a
Deoxygen a ted Hexa n e Solu tion
metallaene
Ph2SidCH2 (16a )
Ph2GedCH2 (16b)
kdim25°C/M-1
Ea/kcal mol
s
-1 b
(9 ( 6) × 10
9
(8 ( 3) × 10
+0.8 ( 0.4
10.4 ( 0.4
9
-
1
-1.0 ( 0.6
9.2 ( 0.4
log(A/M s-1
-
1
)
a
Errors are reported as (2σ, corresponding to confidence
b
intervals of 94-95%. Interpolated from the Arrhenius param-
eters.
it impossible to completely eliminate the pseudo-first-
order component from the decays, even with the highest
intensities we can achieve with our laser. Thus, the
second-order decay rate constants (kdim) were calculated
by nonlinear least-squares fitting of absorbance vs time
data to eq 9, where l is the path length of the cell (3
mm), ꢀ is the extinction coefficient of 16a at 325 nm
-
1
-1 36
reaction that these two species undergo.36 The present
(ꢀ325 nm ) 8900 ( 1800 M
cm ), and k1 is the pseudo-
first-order component of the decay. The analyses gener-
study, combined with other evidence from previously
reported work, allows us to address the mechanism(s)
of this characteristic reaction of SidC and GedC double
bonds.
4
-1
ally afforded k1-values on the order of 10 s , while the
second-order coefficients () 2kdim∆ODo/lꢀ) were on the
6
-1
order of (2-3) × 10 s
.
The decay traces for germene 16b were more pre-
dominantly second-order, but plots of 1/(∆OD)t vs time
deviated significantly from linearity because of the
presence of a residual absorption that we assume is due
to the formation of a relatively long-lived, minor pho-
toproduct. Therefore, the decays were fit to eq 10
Resu lts a n d Discu ssion
Laser flash photolysis of continuously flowing, nitrogen-
saturated solutions of 15a and 15b allows the corre-
sponding 1,1-diphenylmetallaenes 16a ,b to be detected
directly, as transient absorptions centered at λmax ) 325
-
1
-1 36
(
ꢀ325 nm ) 11400 ( 2700 M cm ), to correct for the
residual absorption (∆OD∞), which did not exceed 10%
of the initial signal intensity in any case.
9
,36
nm.
In sodium-distilled hexane and using oven-dried
glassware, the transients decay with kinetics ranging
from mixed second-order and pseudo-first-order in the
case of 16a to pure second-order in the case of 16b.
Although relatively insensitive to the presence of
oxygen, silene 16a is so reactive toward water and
alcohols that under our experimental conditions we find
Second-order decay rate constants for 16a and 16b
were determined at several temperatures over the

4
0
540 Organometallics, Vol. 20, No. 22, 2001
Morkin and Leigh
-60 °C range and are shown in the form of Arrhenius
Sch em e 1
plots in Figure 1. For comparison, the temperature
dependence of the diffusional rate constant (calculated
using the modified Debye equation (kdiff ) 8RT/3000η)
3
9
and published temperature-viscosity data
(Ea )
-
1
-1
2
.4 ( 0.5; log(A/M
s
) ) 12.1 ( 0.4) is also plotted in
Figure 1. Table 1 lists the Arrhenius parameters
obtained from linear least-squares analysis of the two
sets of data, along with “best fit” values of kdim at
The near-zero activation energies for dimerization of
both 16a and 16b are similar to those reported previ-
26,29
ously for 2a in the gas phase
and, given the high
2
5 °C, calculated from the Arrhenius parameters. We
degree of exothermicity associated with the process, are
note that the errors in the Arrhenius parameters are
reported as (2σ, which represents confidence intervals
close to 95% in both cases.
46-48
9
consistent with either concerted
or stepwise mech-
anisms for the reaction. The preexponential factors are
q
associated with entropies of activation of ∆S ) -18 (
The similar absolute reactivities and Arrhenius pa-
rameters for dimerization of 16a and 16b are quite
2 and -13 ( 2 cal/mol‚K for 16a and 16b, respectively,
at 300 K.
interesting, considering the substantial differences in
While we cannot yet conclusively decide between
concerted and stepwise dimerization mechanisms in
either case, it is possible to define for the stepwise
mechanism what the kinetic behavior of the putative
reaction intermediates must be in order to be consistent
with the experimentally observed activation energies.
Scheme 1 defines the form of a stepwise mechanism
involving a single intermediate, analogous to the mech-
anisms that have been previously established for the
their reactivity toward nucleophiles.1
3,40
For example,
1
6a reacts at close to the diffusion-controlled rate with
methanol in solution, via a stepwise mechanism that is
overall first-order in alcohol. In contrast, germene 16b
reacts by a mechanism that is strictly second-order in
methanol, with the first-order pathway being too slow
to be detected. The similarities in the rates of dimer-
ization and methanol addition to 16a are not surprising,
considering what is known of the thermochemistries of
the two reactions in simpler derivatives. For example,
the heats of dimerization and hydration of 1,1-dimeth-
ylsilene (2a ), calculated from experimental heats of
4
9
addition of alcohols and other nucleophiles to transient
5
0
silenes in solution. Mechanisms of this type, when the
reaction is strongly exothermic, will exhibit a bell-
shaped Arrhenius behavior provided that a large enough
temperature range can be spanned experimentally.9
This behavior is caused by the fact that the rates of the
two reaction pathways available to the intermediates
reversion to reactants (k-I) and collapse to products
(kP)svary in different ways with temperature. The
activation energy is negative over the (high) tempera-
ture range where k-I > kP, positive over the (low)
temperature range where k-I < kP, and zero at the
temperature where k-I ≈ kP. If the dimerizations of both
16a and 16b are nonconcerted, then the near-zero
activation energies that are observed indicate that k-I
and kP must be of similar magnitudes to one another
over the 0-60 °C temperature range; however, the
intermediate from 16a must revert to reactants margin-
ally faster than collapse to products (since its Ea is
slightly negative), while the opposite must be true for
the intermediate from 16b.
4
1
formation for 2a (∆Hf° ) 9 kcal/mol ), 3a (∆Hf° )
4
2
-
51.6 kcal/mol ), trimethylsilanol (∆Hf° ) -119 kcal/
4
3
39
mol ), and water (∆Hf° ) -56 kcal/mol ), are ∆Hdim )
69.6 and ∆Hhyd ) -70.0 kcal/mol, respectively. These
-
compare reasonably well with high-level theoretical
estimates of the corresponding parameters for the
parent silene, H2SidCH2: ∆Hdim ) -77 to -83 kcal/
mol1
4,18-21
and ∆Hhyd ) -77 kcal/mol. A self-consistent
44
set of data for germene dimerization and hydration
are unfortunately not available. However, values of
2
2
45
∆
Hdim ) -75.3 kcal/mol and ∆Hhyd ) -61.7 kcal/mol
have been reported for head-to-tail dimerization and
hydration of the parent molecule (H2GedCH2) on the
basis of RHF/6-311G(d,p) and MP3/DZ ab initio calcula-
tions, respectively. These data indicate that the dimer-
ization of silenes and germenes of analogous structure
have comparable overall exothermicities, consistent
with the similarities in the absolute rate constants for
dimerization of 16a and 16b. Nucleophilic addition, on
the other hand, maintains a strong exothermicity in
silenes, but is significantly less exothermic in the case
of germenes. This too is reflected in the relative reac-
tivities of 16a and 16b toward methanol addition.
In our view, the most reasonable stepwise reaction
mechanism for dimerization of the 1,1-diphenyl-substi-
tuted metallaenes 16a and 16b is one that involves the
intermediacy of the 1,3-dimetalla-1,4-butanediyl biradi-
cals 18, and so further examination of the viability of
this mechanism requires that we have independent
information on how such biradicals behave. In principle,
this information should be available from studies of the
photochemistry of the corresponding 1,3-dimetallacy-
clobutanes, since the photolysis of these compounds is
(
39) CRC Handbook of Chemistry and Physics; CRC Press: Boca
Raton, 1995; pp 6-241.
40) Leigh, W. J .; Bradaric, C. J .; Sluggett, G. W. J . Am. Chem. Soc.
993, 115, 5332.
41) Brix, Th.; Arthur, N. L.; Potzinger, P. J . Phys. Chem. 1989, 93,
193.
42) Gusel’nikov, L. E.; Potapov, V. K.; Volnina, E. A.; Orlov, V. Yu.;
Vdovin, V. M.; Nametkin, N. S. Dokl. Akad. Nauk. SSSR 1976, 229,
53.
43) Becerra, R.; Walsh, R. In The Chemistry of Organic Silicon
(
1
(
8
(46) Houk, K. N.; Rondan, N. G.; Mareda, J . Tetrahedron 1985, 41,
1555.
(47) Mayr, H.; Schneider, R.; Grabis, U. J . Am. Chem. Soc. 1990,
112, 4460.
(48) Patz, M.; Mayr, H.; Bartl, J .; Steenken, S. Angew. Chem., Int.
Ed. Engl. 1995, 34, 490.
(49) Bradaric, C. J .; Leigh, W. J . Can. J . Chem. 1997, 75, 1393.
(50) (a) Morkin, T. L.; Leigh, W. J . Acc. Chem. Res. 2001, 34, 129.
(b) Morkin, T. L.; Owens, T. R.; Leigh, W. J . In The Chemistry of
Organic Silicon Compounds; Rappoport, Z., Apeloig, Y., Eds.; J ohn
Wiley & Sons: New York, 2001; in press.
(
7
(
Compounds; Rappoport, Z., Apeloig, Y., Eds.; J ohn Wiley & Sons: New
York, 1998; pp 153-180.
(44) Nagase, S.; Kudo, T.; Ito, K. In Applied Quantum Chemistry;
Smith, V. H., J r., Schaefer, H. F., Morokuma, K., Eds.; D. Reidel:
Dordrecht, 1986; pp 249-267.
(45) Nagase, S.; Kudo, T. Organometallics 1984, 3, 324.

Dimerizations of 1,1-Diphenylsilene and -germene
Organometallics, Vol. 20, No. 22, 2001 4541
thought to proceed via initial scission of one of the ring
hexane solution. Both reactions are characterized by
near-zero Arrhenius activation energies and Arrhenius
3
6,51,52
C-M bonds.
information on both cleavage and coupling of 1,3-disila-
,4-butanediyl biradicals is the report by J ones and co-
The only example available that gives
9
10
-1 -1
preexponential factors on the order of 10 -10
M
s .
1
In neither case do the results allow one to distinguish
conclusively between concerted and stepwise mecha-
nisms for the reaction. However, they do provide an
indication of what the kinetic behavior of the putative
reaction intermediates must be in order for a noncon-
certed mechanism, perhaps involving the intermediacy
of the corresponding 1,3-dimetalla-1,4-butanediyl bi-
radical, to be consistent with the experimentally ob-
served activation energies. A stepwise mechanism in-
volving a different biradical intermediate, such as the
2,3-disila-1,4-butanediyl recently predicted by theory for
the head-to-tail dimerization of the parent silene,21
seems unlikely, but nevertheless cannot be rigorously
excluded as a possibility.
workers of the photolyses of cis- and trans-1,1,3,3-
tetraphenyl-2,4-dineopentyl-1,3-disilacyclobutane (19),
each of which results in the formation of the other
5
1
geometric isomer and the acyclic isomers 20 (eq 11).
These products are those expected from bond rotation/
recoupling and hydrogen migration within the 1,4-
biradical intermediate 21. Photolysis of cis- and trans-
1
9 in the presence of methanol resulted in the additional
formation of methoxysilane 23, the methanol-trapping
product of silene 22. Thus, it is at least clear that 1,4-
biradicals of this type undergo competitive ring closure
and â-scission to silene, as expected. Unfortunately, the
crucial question of the relative rates of these processes
cannot be answered quantitatively without careful
product analyses at low conversions, and data of this
type were not reported. The photolysis of disilacyclobu-
tane 17a has been shown to lead to the formation of
The head-to-tail dimerization of 1,1-diphenylsilene
(16a ) exhibits an Arrhenius activation energy of E )
a
-1.0 ( 0.4 kcal/mol and activation parameters of
q
q
∆H ) -1.6 ( 0.3 kcal/mol and ∆S ) -18 ( 2 cal/mol‚
K at 300 K. For this to be consistent with the stepwise
mechanism, the biradical intermediate would have to
revert to silene somewhat faster than it collapses to
product by bond rotation and coupling. In contrast, an
Arrhenius activation energy of Ea ) +0.8 ( 0.4 kcal/
3
6
1
6a , but in very low quantum yield (Φ16a ) 0.007).
q
q
mol (∆H ) +0.2 ( 0.4 kcal/mol; ∆S ) -13 ( 2 cal/
mol‚K at 300 K) is observed for the head-to-tail dimer-
ization of 1,1-diphenylgermene (16b). This is consistent
with a mechanism involving the initial reversible for-
mation of 1,3-digerma-1,4-butanediyl biradical 18b,
which undergoes coupling in preference to cleavage.
Further work in this area will be directed at delineat-
ing the chemistry of 1,3-dimetalla-1,4-butanediyl bi-
radicals, in an effort to define the mechanism(s) of the
head-to-tail dimerization of transient silenes and
germenes more precisely.
The results of similar studies of the photochemistry
of 17b seem more conclusive than in the silicon case,
since photolysis of this compound in methanolic hexane
Exp er im en ta l Section
solution leads to the formation of methoxygermane 24
with a quantum yield of Φ24 ) 0.11.36 This value
Hexane (BDH Omnisolv) was distilled from sodium under
an atmosphere of dry nitrogen. 1,1-Diphenylsilacyclobutane
corresponds to the quantum yield for formation of 16b
from the excited state of the digermacyclobutane, as the
experiment was carried out under conditions of quan-
titative trapping of the germene by the alcohol. Indeed,
(
15a ) and 1,1-diphenylgermacyclobutane (15b) were synthe-
13,53
sized according to published methods.
flash photolysis experiments employed the pulses (248 nm;
5-20 ns; 120-140 mJ ) from a Lambda Physik Compex 100
excimer laser filled with F /Kr/He mixtures and a computer-
Nanosecond laser
1
1
1
1
6b can be readily detected by laser flash photolysis of
7b, in contrast to the behavior of the silicon analogue
7a under identical conditions.36 These results suggest
2
5
4
controlled detection system as described previously. Solutions
were prepared at concentrations such that the absorbance at
the excitation wavelength was ca. 0.7 (∼0.003 M) and were
flowed continuously through a 3 × 7 mm Suprasil flow cell
connected to a calibrated 100 mL reservoir. Solution temper-
atures were measured with a Teflon-coated copper/constantan
thermocouple that was inserted directly into the flow cell.
that biradical 18b undergoes â-scission with reasonable
efficiency relative to coupling to regenerate 17b. We do
not know what the relative rates of biradical â-scission
(k-I) and coupling (kP) are; nevertheless, the possibility
that k-I is similar to but smaller than kP, as the slightly
positive Ea for dimerization of 16b demands, is certainly
a viable one.
Ack n ow led gm en t. We wish to thank the Natural
Sciences and Engineering Research Council of Canada
for financial support and for a post-graduate scholarship
to T.L.M.
Su m m a r y a n d Con clu sion s
1
,1-Diphenylsilene and 1,1-diphenylgermene exhibit
OM0103409
similar reactivities toward head-to-tail dimerization in
(
51) J ung, I. N.; Pae, D. H.; Yoo, B. R.; Lee, M. E.; J ones, P. R.
Organometallics 1989, 8, 2017.
52) Yoo, B. R.; Lee, M. E.; J ung, I. N. Organometallics 1992, 11,
626.
(53) Leigh, W. J .; Bradaric, C. J .; Kerst, C.; Banisch, J . H. Orga-
nometallics 1996, 15, 2246.
(54) Leigh, W. J .; Workentin, M. S.; Andrew, D. J . Photochem.
Photobiol. A: Chem. 1991, 57, 97.
(
1