Mechanistic studies on the reactions of trimethylsilanide and trimethylstannylide ions with haloarenes in hexamethylphosphoramide

Article abstract of DOI:10.1016/S0022-328X(02)01572-3

The large positive ρ values obtained from Me3M^- ions reacting with appropriately-substituted chloroarenes (4.1+/-0.1, and 2.94+/-0.02 for M=Sn and Si respectively) can be accounted for the presence of a strong negative charge in the transition state of the substitution reaction. These ρ values for Me3Sn^- and Me3Si^- nucleophilic attack on ArX provide evidence for the overall larger selectivity of the former, compared with the more reactive, and less selective Me3Si^- ions. When 1-allyloxy-2-halobenzenes (X=Cl, I) are allowed to react with Me3Sn^- ions in HMPA, an ipso substitution product 6 is obtained, in addition to stannylated products on the allylic moiety. This reaction proceeds by a HME process, as opposed to the same reaction carried out in liquid ammonia, where (2,3-dihydro-benzofuran-3-ylmetyl)-trimetylstannane (4) is furnished in high yield through an S_RN1 mechanism. Reaction of Me3Si^- ions with 1-allyloxy-2-iodobenzene in HMPA, affords only the ipso-substituted product, namely 2-(allyloxyphenyl)-trimethylsilane (11), through the intermediacy of a hypervalent silicon species.

Full text of DOI:10.1016/S0022-328X(02)01572-3

Journal of Organometallic Chemistry 656 (2002) 108Á115
/
www.elsevier.com/locate/jorganchem
Mechanistic studies on the reactions of trimethylsilanide and
trimethylstannylide ions with haloarenes in
hexamethylphosphoramide
Al Postigo, Santiago E. Vaillard, Roberto A. Rossi*
Departamento de Qu´ımica Orga´nica, Facultad de Ciencias Qu´ımicas, Instituto de Investigaciones en Fisico Quimica de Co´rdoba (INFIQC), Universidad
Nacional de Co´rdoba, Ciudad Universitaria, 5000 Co´rdoba, Argentina
Received 14 March 2002; received in revised form 14 March 2002; accepted 13 May 2002
Abstract
The large positive r values obtained from Me3Mꢀ ions reacting with appropriately-substituted chloroarenes (4.19
0.1, and
/
2.949
/
0.02 for MꢁSn and Si, respectively) can be accounted for the presence of a strong negative charge in the transition state of
/
the substitution reaction. These r values for Me3Snꢀ and Me3Siꢀ nucleophilic attack on ArX provide evidence for the overall
larger selectivity of the former, compared with the more reactive, and less selective Me3Siꢀ ions. When 1-allyloxy-2-halobenzenes
(Xꢁ
Cl, I) are allowed to react with Me3Snꢀ ions in HMPA, an ipso substitution product 6 is obtained, in addition to stannylated
/
products on the allylic moiety. This reaction proceeds by a HME process, as opposed to the same reaction carried out in liquid
ammonia, where (2,3-dihydro-benzofuran-3-ylmethyl)-trimethylstannane (4) is furnished in high yield through an SRN1 mechanism.
Reaction of Me3Siꢀ ions with 1-allyloxy-2-iodobenzene in HMPA, affords only the ipso-substituted product, namely 2-(allyloxy-
phenyl)-trimethylsilane (11), through the intermediacy of a hypervalent silicon species. # 2002 Elsevier Science B.V. All rights
reserved.
Keywords: Trimethylsilanide ions; Trimethylstannylide ions; SRN1 reaction, HME reaction; Hypervalent silicon
1. Introduction
substitution. From these results it was concluded that
the reactions should proceed, at least partially, by a
radical mechanism [2].
We have recently described photostimulated reactions
of chloroarenes with triorganylstannyl ions by the SRN1
mechanism in liquid ammonia. These reactions afford
moderate-to-excellent yields of the nucleophilic substi-
tution products. Bromo and iodoarenes react by an
HME reaction [3,4]. When the aromatic substrate bears
two leaving groups, such as in p- and m-dichloroben-
Trimethylstannide ions (Me3Snꢀ) have been shown
to react with PhX (Cl, Br, and I) in tetraglyme as solvent
to produce the substitution product PhSnMe3, and
variable amounts of reduction product benzene. Me-
chanistic studies in this system revealed that the reaction
occurs by a Halogen Metal Exchange (HME) process in
a solvent cage [1]. When THF is used as solvent,
Me3Snꢀ ions react with o-, m-, and p-bromotoluenes
to afford the expected trimethylstannyl-substituted pro-
ducts [2].
zenes, disubstitution products are obtained in 88Á90%
/
yield [5]. The monosubstitution products are not inter-
On the other hand, when p-chloro and p-fluoroto-
luenes are allowed to react with Me3Snꢀ ions in THF as
solvent, the yields of cine substitution products are
enhanced. Radical traps favor cine substitution pro-
ducts, and Li metal enhances the yields of ipso
mediates in these reactions.
Trimethylsilanide ions (Me3Siꢀ) cannot be prepared
in liquid ammonia due to reaction of this ion with the
solvent, to afford silylamines [6]. We have recently
reported the enhanced reactivity of Me3Siꢀ ions in
HMPA toward PhX (Cl, Br, I) to afford very good
yields of PhSiMe3 [7]. On the other hand, PhF reacts
with Me3Siꢀ ions to afford 76% of o- and p-trimethyl-
* Corresponding author. Tel.: ꢂ
433-4127/3030
E-mail address: rossi@dqo.fcq.unc.edu.ar (R.A. Rossi).
/
54-351-433-4170; fax: ꢂ54-351-
/
0022-328X/02/$ - see front matter # 2002 Elsevier Science B.V. All rights reserved.
PII: S 0 0 2 2 - 3 2 8 X ( 0 2 ) 0 1 5 7 2 - 3

A. Postigo et al. / Journal of Organometallic Chemistry 656 (2002) 108Á
/115
109
silylfluorobenzenes, and 14% of PhSiMe3 according to
Eq. (1).
solvents and given the enhanced reactivity of Me3Siꢀ
anion in HMPA toward haloarenes, we embarked on
providing further mechanistic evidence for the reactions
of the Me3Siꢀ and Me3Snꢀ ions toward haloarenes in
HMPA.
ð1Þ
Pyridine yields a quantitative yield of 4-trimethylsila-
nylpyridine [7] and N-methyl indole affords N-methyl-
(2-trimethylsilanyl)indole in 78% of isolated yield [8].
Benzene itself reacts in the dark with Me3Siꢀ ions to
afford 45% yield of PhSiMe3. Experiments aimed at
elucidating the mechanism of these reactions where a H
atom is substituted by the TMS group were carried out.
Both PhF and benzene show pronounced deuterium
2. Results and discussion
The reactivity order in competition reactions of
Me3Siꢀ ions with PhX in HMPA was found to be PhI
(3.5):PhBr (1.9):PhCl (1):PhF (0.08) [7]. These results
show that the span in reactivity between the four
halogens toward Me3Siꢀ ions is, from PhF to PhI,
only about 44 [7]. It has been suggested that these
nucleophilic substitutions take place through a silicon
hypervalent intermediate. The span in reactivity in
competition experiments of the four halobenzenes
toward acetone enolate anion in liquid ammonia in
kinetic isotope effects (3.29
/
0.1 and 1.790.1, respec-
/
tively) [7].
It has been proposed that the nucleophilic attack of
Me3Siꢀ ions on aromatic moieties take place through
the intermediacy of hypervalent silicon species, which
evolve to products through loss of an appropriate
leaving group (X in the case of haloarenes, and H in
the case of non-activated aromatic hydrocarbons or
heteroaromatic compounds without formal leaving
groups).
S
RN1 reactions is over 100 000 [9] and Kuivila et al.
found that the relative reactivity of PhX versus Me3Snꢀ
ions in tetraglyme was ca. PhI (106):PhBr (105):PhCl (1)
[1]. These reactions proceed by an in-cage HME
mechanism. This notorious span in reactivity is expected
in pure HME processes.
PhX (Xꢁ
I, Br) react with Me3Snꢀ ions in HMPA
/
In contrast, competition experiments, carried out in
the dark, among PhX toward Me3Snꢀ ions in HMPA
throw a relative reactivity pattern of PhI (77):PhBr
through a HME process. ArCl substrates undergo a
slow HME reaction with Me3Snꢀ ions. These reactions
are slightly accelerated upon photostimulation, indicat-
ing that an ET process competes with a polar mechan-
ism. However, when p-dichlorobenzene is allowed to
react with Me3Snꢀ ions in HMPA, both p-chlorotri-
methylstannylbenzene and p-bis-trimethylstannylben-
zene are formed [8], as opposed to when the same
reaction is carried out in liquid ammonia, where the
monosubstitution product is not an intermediate.
PhF conforms to a special case in the sense that the
intervention of a hypervalent tin species is postulated in
order to account for products such as p-fluorophenyl-
trimethylstannane. The ET process also operates under
irradiation to yield the straightforward substitution
product PhSnMe3 together with p-fluorophenyltri-
methylstannane (Eq. (2)). The fact that only the p-
isomer is formed, as opposed to the same reaction
carried out with Me3Siꢀ ions which renders both the o-
and p- TMS-substituted fluorobenzenes, suggests that
the Me3Snꢀ ions are less reactive, and more selective
than Me3Siꢀ ions.
Table 1
Competition experiments of PhX (XꢁF, Cl, Br, and I) for Me3Snꢀ
and Me3Siꢀ ions in HMPA
b
Substrates (mM)
Anion
(mM)
Products (relative %)
a
p-CH3C6H4Cl
(4338); PhI (482)
Me3Snꢀ
(482)
p-CH3C6H4SnMe3 (1.5%);
PhSnMe3 (98.7%)
p-CH3C6H4Cl (964); Me3Snꢀ
p-CH3C6H4SnMe3 (5.1%);
PhSnMe3 (95.1%)
p-CH3C6H4SnMe3 (88.7%); p-
PhBr (482)
(482)
p-CH3C6H4Cl (482); Me3Snꢀ
c
PhF (4338)
p-CH3C6H4Cl (85);
PhI (85)
(482)
Me3Siꢀ
(49)
FC6H4SnMe3 (11.2%)
p-CH3C6H4SiMe3 (22.5%);
d
PhSiMe3 (77.8%)
p-CH3C6H4SiMe3 (32.1%);
p-CH3C6H4Cl (85);
PhBr (85)
Me3Siꢀ
(49)
d
PhSiMe3 (67.9%)
p-CH3C6H4SiMe3 (92.6%);
p-CH3C6H4Cl (85);
PhF (85)
Me3Siꢀ
(49)
d
PhSiMe3 (7.4%)
Reaction conditions: dark, 25 8C, Me3Snꢀ 3 h, Me3Siꢀ (0.25 h).
Product yields based on starting PhX, corrected to 1:1 relative
molar concentration. The relative yields result from average of
a
b
duplicate experiments whose Do 52s.
c
Upon irradiation (3 h, 366 nm) of a mixture consisting of p-
CH3C6H4Cl (482 mM), PhF (482 mM), and Me3Snꢀ (482 mM) in
HMPA, product relative yields are the following: p-CH3C6H4SnMe3
(90.6%), PhSnMe3 (2.9%), p-FC6H4SnMe3 (6.5%).
ð2Þ
d
The same ratio of relative yields was also obtained, within
experimental errors, under the following different reaction conditions:
p-CH3C6H4Cl (255 mM), PhX (255 mM), and Me3Siꢀ (147 mM),
reaction time: 1 h.
Considering the mechanistic evidence amounted for
the reactivity of Me3Snꢀ ions toward PhX in different

110
A. Postigo et al. / Journal of Organometallic Chemistry 656 (2002) 108Á115
/
(19):PhCl (1):PhF (0.13) (Table 1). In this case, the span
in reactivity is ca. 590, which when compared with
Me3Siꢀ ions, denotes a larger selectivity for the reac-
tions of Me3Snꢀ with PhX (being Me3Siꢀ ions more
reactive, are hence rendered less selective).
p-Chlorobenzotrifluoride 1 reacts with Me3Snꢀ ions
in HMPA in the dark to furnish trimethyl-(4-trifluor-
omethyl-phenyl)stannane (2) in 85% yield (Eq. (3)). The
meta isomer is not observed under our detection limits
(B0.5%) [10].
/
Fig. 1. Hammett plots for the reactions of Me3Snꢀ (m) and Me3Siꢀ
(m) ions with ArX in HMPA.
ð3Þ
propenyloxy)halobenzenes (3) with these ions as radical
probes.
Linear free energy relationships (LFER) or Hammett
plots were constructed with Me3Snꢀ and Me3Siꢀ ions
upon reacting in the dark with appropriately-substituted
When o-(2-propenyloxy)chlorobenzene (3a) is sub-
jected to reaction with Me3Snꢀ ions in liquid ammonia,
(2,3-dihydro-benzofuran-3-yl-methyl)-trimethyl-stan-
nane (4) is obtained (87%) exclusively (Eq. (4)) (Table 3,
experiment 1).
chloroarenes ArCl (p-XPhCl, where Xꢁ/CH3O, CH3,
H, F, Cl, CF3).
The slopes of these plots (obtained from plotting the
log of kArCl/kPhCl vs. substituent constant s) yielded r
0.1 for the reactions of Me3Snꢀ ions, and
values of 4.19
/
rꢁ2.949
/
/
0.02 for the reactions of Me3Siꢀ ions (Table
ð4Þ
2), indicating, in these cases, that a negative charge in
the transition states of the substitution reactions of ArX
either with Me3Snꢀ or Me3Siꢀ ions is strongly devel-
oping (Fig. 1).
In order to determine the nature of the intermediates
involved in the reactions of haloarenes with Me3Snꢀ
and Me3Siꢀ ions, we examined the behavior of o-(2-
This result is explained by the intermediacy of the
SRN1 mechanism [11]. When 3a accepts one electron,
radical anion 3aꢀ is formed, which fragments into
+
radical 5. This radical intermediate is trapped by the
Table 2
Competition experiments of PhCl and p-RC6H4Cl toward Me3Mꢀ ions (MꢁSn, Si)
a
Experiment
p-RC6H4Cl
Me3Mꢀ
M
Relative yields (%)
log kArCl/kPhCl
R
PhMMe3
p-RC6H4MMe3
1
2
MeO
Me
F
Sn
Sn
Sn
Sn
Sn
Si
92.5
74.0
9.8
7.5
ꢀ1.09
ꢀ0.45
0.96
26.0
90.2
90.5
99.4
20.1
33.4
79.6
86.6
97.9
3
4
Cl
9.5
0.98
2.21
5
CF3
OMe
Me
F
0.60
79.9
66.6
20.4
13.4
2.1
6
ꢀ0.60
ꢀ0.30
0.59
7
Si
8
Si
9
10
Cl
CF3
Si
0.81
1.67
Si
a
In the reactions with Me3Snꢀ ions, the concentration of the nucleophile, PhCl and p-RC6H4Cl were 360 mM each in HMPA. The reactions were
allowed to proceed for 2 h at 40 8C. For the reactions with Me3Siꢀ ions, the concentration of the nucleophile was 490 mM, and PhCl and p-
RC6H4Cl were 980 mM in HMPA respectively. The reactions were allowed to proceed for 2 h at 40 8C. Yields were calculated based on the internal
standard method, using trimethylsilylbenzene or trimethylstannylbenzene as calibrated internal standards. Values of kArCl were calculated based on
disappearance of starting materials (PhCl, and ArCl), employing the following expression: kPhCl/kArClꢁln[(PhCl)o/(PhCl)t]/ln[(ArCl)o/(ArCl)t],
where (PhCl)o and (ArCl)o are the initial concentrations and (PhCl)t and (ArCl)t are the concentrations at time t. See: J.F. Bunnett, in: E.S. Lewis
(Ed.), Investigation of rates and mechanism of reactions, third ed., Part I, Wiley-Interscience, NY, 1974. The relative yields were the result of an
average of duplicate experiments with standard deviation less than 2s.

A. Postigo et al. / Journal of Organometallic Chemistry 656 (2002) 108Á
/115
111
Table 3
Reactions of o-(2-Propenyloxy)halobenzenes (3) with SnMe3ꢀ ions in HMPA
a
b
Experiment
Substrate (mM)
SnMe3ꢀ (mM)
Reaction conditions
Product(s) (%)
c
d
1
3a, (10)
10
hn, 90 min
Dark, 120 min
hn, 90 min
4 (87)
6 (6), 7 (23), 8 (ca. 4%)
e
2
3
4
5
6
3a, (313)
3a, (313)
3b, (415)
3b, (751)
3b, (128)
418
418
418
f
6 (16), 7 (29), 8 (10)
6 (38), 7 (35), 7-r (12), 8a (13)
Dark, 120 min
Dark, 120 min
Dark, 24 h
g
h
1352
256
11 (51)
11 (75), 7 (5)
g
i
a
HMPA (2Á
/
7 ml) under N2.
b
c
d
e
f
Determined by glc, unless otherwise indicated.
In 200 ml of liquid ammonia.
Isolated yield.
Substrate recovered in 69% yield.
Substrate recovered in 39% yield.
The nucleophile was Me3Si- ions.
Substrate recovered in 49% yield.
Substrate recovered in 24% yield.
g
h
i
double bond faster than coupling with the nucleophile
occurs, affording the rearranged radical 5-r (Eq. (5)).
Radical 5-r reacts with Me3Snꢀ ions to yield 4 (Eq. (6)).
There is precedent that rearrangement of radicals can
occur along the SRN1 mechanism [12].
12% yields, respectively. This is supporting evidence for
the absence of free radicals in the HMPA solution.
Product 8a (13%, vide infra) was not isolated from the
reaction mixture but identified by a GC-spike experi-
ment with an authentic sample synthesized by an
independent route (see Experimental). The low com-
bined yields of products 8 precluded their direct isola-
tion from the reaction mixture. Product 8b was
tentatively identified on the basis of its GCHRMS
data as being probably another positional isomer of 8a
(presumably trimethyl-(3-phenoxy-allyl)-stannane)).
The reduction products observed (i.e. 7 and 7-r) do
not arise by proton abstraction from the solvent; (cf.
reaction of PhI with Me3Snꢀ which yields 98% of
substitution product, where reduction product benzene
is not observed [8]). It is proposed, instead, that when 3
reacts with Me3Snꢀ ions by a HME reaction, ion pair 9
is formed, which collapses to product 6 (Eq. (8)). A Hꢂ
shift from the allyl group to the ring, renders anion 10-r,
which is trapped by Me3SnX to yield 8, or it is
protonated when the reaction is quenched to afford 7
(Eq. (9)).
ð5Þ
ð6Þ
On the contrary, when 3a is treated with Me3Snꢀ ions
in HMPA, (2-allyloxy-phenyl)-trimethyl-stannane (6) is
found in low yield in the reaction mixture (6%), along
with allyloxy-benzene (7) (23%), and traces of a rear-
ranged reduction product 7-r (Eq. (7)). Other substitu-
tion products resulting from stannylation on the allyl
moiety (8, ca. 4% yield) are observed. This reaction is
slightly accelerated by light (Table 3, experiments 2Á3).
/
ð8Þ
ð7Þ
ð9Þ
It is noteworthy that though the reaction of 3a with
Me3Snꢀ ions is accelerated by light, no cyclized product
is obtained, which is probably indicating that an in-cage
ET process is taking place to some extent in the
substitution reaction by Me3Snꢀ ions.
On the other hand, when 3b is treated in the dark with
Me3Snꢀ ions, 38% of 6 is obtained. The yield of 6 is not
enhanced upon photostimulation of the reaction mix-
ture. Reduction products 7 and 7-r are found in 34 and

112
A. Postigo et al. / Journal of Organometallic Chemistry 656 (2002) 108Á115
/
In the same fashion, when 3b is treated with Me3Siꢀ
ions in HMPA, (2-allyloxy-phenyl)-trimethylsilane (11)
is obtained according to Eq. (10).
species), depending on the solvent and on the nucleo-
phile.
4. Experimental
ð10Þ
4.1. General methods
Gas chromatographic analyses were performed on a
The reaction pathway for the formation of 11 is not
likely to be through silicon-analog intermediates such as
9 or 10-r, as no reduction products are observed in this
case (either 7 or 7-r) from the reaction mixture. Instead,
the exclusive formation of 11 suggests that the inter-
mediate in this reaction is a silicon hypervalent species
(Eq. (11)), as it has been proposed for the reactions of
PhX with Me3Siꢀ ions [7]. At longer reaction times (24
h) the yield of 11 increases up to 75%, and a small
amount of 7 (5%) is found.
HewlettÁ
ionization detector, a HewlettÁ
integrator, an one of the following columns: HP1 5 mꢃ
0.17 mm column.
/Packard 5890 Series II instrument with a flame
/Packard 3396 Series III
/
0.17 mm column, and a DB-1, 30 mꢃ
/
1H-NMR (200.13 MHz) and 13C-NMR (50.32 MHz)
were conducted on a Bruker AC 200 spectrometer in
deuterochloroform as solvent or other solvent as
indicated. Coupling constants (J) are given in Hz units.
GCMS analyses were carried out on a Shimadzu GCMS
QP 5050 spectrometer equipped with a DB-5, 30 mꢃ
/
0.18 mm ID column. IR spectra were conducted on a
JASCO instrument, and referenced to polystyrene. The
absorptions are given in wavenumbers (cmꢀ1). High
resolution mass spectra were conducted at the McMas-
ter Regional Center for Mass Spectrometry, at McMas-
ter University, Canada.
ð11Þ
4.2. Materials
3. Conclusions
PhX (I, Br, Cl, F), p-chlorotoluene, p-dichloroben-
zene, p-chloroanisole, p-chlorobenzotrifluoride, p-
chlorofluorobenzene, trimethylsilylbenzene, trimethyl-
tinchloride, p-bis-trimethylsilylbenzene, hexamethyldisi-
lane, sodium methoxide, p-dinitrobenzene, di-tert-
butylnitroxide, N-chlorosuccinimide, phenol, allyl bro-
mide, potassium carbonate, dimethylformamide, hex-
ane, decane, and eicosane were commercially available
and used as received from the supplier. HMPA was
distilled twice under reduced pressure and kept under
nitrogen.
The large positive r values obtained from Me3Mꢀ
ions reacting with PhX (4.19
/
0.1, and 2.9490.02 for
/
MꢁSn and Si, respectively) can be accounted for the
/
presence of a strong negative charge in the transition
state of the substitution reaction. As far as we are
concerned, the influence of substituents on nucleophilic
substitution rates of ArX substrates by stannyl or silyl
anions has not been reported. r Values for Me3Snꢀ and
Me3Siꢀ nucleophilic attack on ArX substrates provide
evidence for the overall larger selectivity of the former,
compared with the more reactive, and less selective
Me3Siꢀ ions.
4.3. Reactions of ArX with Me3Snꢀ ions in HMPA
When 1-allyloxy-2-halobenzenes (XꢁCl, I) are al-
/
lowed to react with Me3Snꢀ ions in HMPA, an ipso
substitution product 6, in addition to stannylated
products on the allylic moiety, are obtained, indicating
that the reaction occurs by an HME process, as opposed
to the same reaction carried out in liquid ammonia,
where the ring closure product (2,3-dihydro-benzofuran-
3-ylmethyl)-trimethylstannane (4) is furnished in high
yield through an SRN1 mechanism. On the contrary, 1-
allyloxy-2-iodobenzene reacts with Me3Siꢀ ions to
afford only the straightforward substitution product,
through a hypervalent silicon intermediate, ruling out an
ET and HME processes. It is interesting to note that 1-
allyloxy-2-halobenzenes react by three different mechan-
isms (i.e. HME, ET, and through hypervalent metallic
Into a 15 ml tube previously flame-dried, Na metal
(1.8 mmol) and HMPA (2.5 ml) were introduced and the
tube sealed with a rubber septum. The suspension was
stirred with a micro stir bar and was de-oxygenated and
blanketed with a nitrogen atmosphere thrice, and left
under N2. The stirred suspension turned slowly deep
blue, and a deoxygenated solution of Me3SnCl (ca. 0.93
mmol) in HMPA (1 ml) was slowly introduced by
syringe and the solution left stirring for additional 20
min. The mixture turned colorless, with some Na
suspended. The tube was immersed in an ultrasound
bath, and left under sonication until all Na was
dissolved and reacted with Me3SnCl (from 2 to 3 h).
At this point, the solution was deep orange. The deep

A. Postigo et al. / Journal of Organometallic Chemistry 656 (2002) 108Á
/115
113
orange solution obtained was indicative of the forma-
tion of Me3Snꢀ ions. At this time, for the dark arm of
reactions, the tube was covered with aluminum foil to
protect it from laboratory light, removed from the
ultrasound bath, and the neat ArX (ca. 0.84 mmol)
was slowly introduced by syringe through the rubber
septum. The dark reactions were left to progress under
stirring, whereas the photochemical counterparts were
irradiated in a photochemical reactor employing two
medium pressure (400 W) water-cooled Hg lamps
emitting maximally at 366 nm. The mixtures were then
cooled down in an ice bath and quenched slowly with
water and extracted three times into pentane. The
pentane layers were washed twice with doubly-distilled
water. The organic layers were gathered and dried over
sodium sulfate, filtered, evaporated and chromato-
graphed (for isolation purposes, the number of moles
of the reactants was scaled up by a factor of 10, and the
solvent volume, HMPA, was 7.5 ml) over a 2 mm
thickness silica-gel plate using radial chromatography
(solvent of elution was hexane). The isolated compounds
were characterized by standard spectroscopic techniques
(1H/13C-NMR, and MS). For GC quantification, the
internal standard method was used, employing tri-
methylsilylbenzene, trimethylstannylbenzene, eicosane
or n-decane as internal standards.
procedure: Into a 50 ml two-necked round-bottomed
flask equipped with a water-cooled condenser, and an
inlet for a thermometer, allyloxybenzene (11.2 mmol),
N-chlorosuccinimide (22.4 mmol), benzoyl peroxide
(0.95 mmol) and previously distilled CCl4 (25 ml) were
introduced. The reaction flask was heated in an oil bath
at 60 8C for 48 h. The reaction mixture was filtered,
then quenched with water, and extracted into CCl4. The
organic layers were dried over sodium sulfate, filtered,
evaporated. The reaction mixture contained two major
products chlorinated at position 1 and another at
position 3 of the allyl chain, that is, (3-chloropropeny-
loxy)-benzene, and (1-chloro-allyloxy)-benzene. The
mixture was chromatographed over silica-gel using
column chromatography. The elution solvent was pet-
roleum ether. Standard spectroscopic techniques were
used to characterize these compounds.
4.5. Syntheses of tin-derived products
4.5.1. (2,3-Dihydro-benzofuran-3-ylmethyl)-trimethyl-
stannane (4)
Into a three-necked, 500 ml round-bottomed flask
equipped with a cold finger condenser charged with
ethanol, a nitrogen inlet and a magnetic stirrer were
condensed 250 ml of ammonia previously dried with Na
metal under nitrogen. Trimethyltin chloride (1.15 mmol)
was then added and Na metal (2.65 mmol) in small
pieces was introduced, waiting for total de-coloration
The standardized procedure for the reactions of
trimethylsiliconide ions is given in reference [7].
Phenyltrimethyltin and trimethylphenylsilane were
isolated and their spectroscopic data compared with
those of authentic samples. Trimethyl-p-tolyl-stannane
[13], (4-chlorophenyl)-trimethylstannane [14], (4-fluoro-
phenyl)-trimethylstannane [13,15], 1,4-bis-trimethyl-
stannanyl-benzene (6) [13], (4-methoxy-phenyl)-tri-
between each addition. A lemonÁyellow solution is
/
obtained. Substrate 3a (1 mmol) was dissolved in 1 ml
of anhydrous ethyl ether and added to the solution. The
reaction mixture was irradiated for 120 min (366 nm),
and then quenched by adding ammonium nitrate in
excess. The ammonia was allowed to evaporate and
water (50 ml) was then added. The aqueous phase was
methylstannane
[13],
trimethyl-(4-trifluoromethyl-
phenyl)-stannane [13], were characterized by standard
spectroscopic techniques and the data matched well with
those reported in the literature. Analogously, trimethyl-
p-tolyl-silane [13,16,17], (4-methoxy-phenyl)-trimethyl-
silane [13,17], (4-chlorophenyl)-trimethylsilane [17,18],
(4-fluorophenyl)-trimethyl-silane [13,16,17], trimethyl-
(4-trifluoromethyl-phenyl)-silane [13,17], trimethyl-(3-
trifluoro-phenyl)-silane [16,17], 1,4-bis-trimethylsilan-
nyl-benzene, 1-allyoxy-2-chloro-benzene (3a) [19], 1-
allyoxy-2-iodo-benzene (3b) [19], were characterized by
standard techniques, and the spectroscopic data
matched well with those reported in the literature.
Allyloxy-benzene and 1-propenyloxy-benzene were
extracted with CH2Cl2 (3ꢃ50 ml) and the organic
/
layers were dried over anhydrous MgSO4. The solvent
was evaporated and product 4 was purified by silica-gel
column chromatography (isolated product yield: 87%)
using a mixture of CH2Cl2: petroleum ether (60Á
15:85. Product 4 was characterized by 1H-NMR and
13C-NMR, IR, EIHRMS, and GCÁ
MS data as follows.
1H-NMR (CDCl3) d (ppm): 0.06 (ss, 9H), 1.32
(double quintet, 2H); 3.76 (octet, 1H, Jꢁ8.4, 13.2
Hz); 4.04 (dd, 1H, Jꢁ8.4, 7.3 Hz); 4.65 (dd, 1H, Jꢁ
8.4, 8.76 Hz), 6.83 (cplx m, 2 H); 7.12 (cplx.m, 2 H). 13C-
NMR (CDCl3) d (ppm): ꢀ9.49, 17.19, 40.43, 79.16,
MS
/
80 8C),
/
/
/
/
characterized by 1H-NMR and GCÁ
compared with literature values [19].
/
MS data and
/
109.59, 120.53, 123.84, 125.54, 133.47, 159.56. GCÁ
/
(m/z, %): 298 ([M]ꢂ, 2), 283 (isotopic cluster, 91), 281
(64), 279 (39), 253 (5), 249 (2), 225 (4), 223 (2), 221 (2),
185 (21), 165 (67), 163 (42), 161 (33), 151 (23), 133 (100),
131 (58), 115 (11), 105 (32), 91 (24), 77 (43), 65 (12), 51
4.4. Synthesis and characterization of substrates
4.4.1. (1-Chloro-allyloxy)-benzene
This compound was synthesized from allyloxybenzene
and N-chlorosuccinimide, according to the following
(8). EIHRMS: ([M]ꢂꢀ
/
15 for C11H15O120Sn): 283.0145.
Calc.: 283.0145. IR (NaCl pellet) n (cmꢀ1): 2974, 2896

114
A. Postigo et al. / Journal of Organometallic Chemistry 656 (2002) 108Á115
/
(s), 1610 (m), 1597 (s), 1481 (s), 1460 (s), 1224 (s), 970
(m), 748 (s), 526 (s).
1.82, 3.7, 17.2 Hz), 6.18 (double quintet, 1H, 17.2Hz, 5.1
Hz, Jꢁ
12.1), 7.01 (cplx.m, 2H), 7.41 (cplx.m, 2H). 13C-
NMR (in CDCl3) d (ppm): ꢀ0.92, 68.59, 110.47,
/
/
117.08, 120.53, 128.13, 130.61, 133.52, 135.02, 163.25.
GCMS (m/z, %): 206 ([M]ꢂ, 45), 191 (95), 175 (17), 163
(96), 151 (81), 150 (39), 149 (71), 135 (100), 15 (21), 91
(63), 73 (31), 59 (21), 41 (31).
4.5.2. (2-Allyloxy-phenyl)-trimethylstannane (6)
This compound [20] was isolated (38% yield) as
indicated above from reaction of 1-allyloxy-2-iodoben-
zene and Me3Snꢀ ions in HMPA, according to the
general procedure (vide supra), and characterized by 1H-
NMR and 13C-NMR, EIHRMS, and GCÁ
/
MS data as
follow: H-NMR (in CD3CN) d (ppm): 0.25 (ss, 9H),
4.54 (dt, 2H, Jꢁ1.82, 3.3, 8.4 Hz), 5.26 (dq, 1H, Jꢁ
1.82, 3.3, 10.6 Hz), 5.39 (dq, 1H, Jꢁ1.82, 3.6, 17.5 Hz),
6.09 (double quintet, 1 H, Jꢁ5.12, 10.6, 15.7 Hz), 6.93
(cplx.m, 2H), 7.35 (cplx.m, 2H). 13C-NMR (in CD3CN)
d (ppm): ꢀ8.74, 70.06, 112.04, 118.05, 122.69, 131.15,
131.50, 135.41, 138.04, 164.34. HRMS ([M]ꢂꢀ
15 for
C11H15O120Sn): 283.0098. Calc.: 283.0145. GCÁ
MS: 283
1
Acknowledgements
/
/
/
This work was supported in part by the Consejo de
Investigaciones de la Provincia de Co´rdoba (CONI-
/
COR), the Consejo Nacional de Investigaciones Cient´ı
-
/
ficas y Te´cnicas (CONICET), SECYT, Universidad
Nacional de Co´rdoba and FONCYT, Argentina.
S.E.V. gratefully acknowledges receipt of a fellowship
from CONICET.
/
/
(79), 281 (58), 279 (35), 268 (0.5), 253 (4), 242 (9), 240
(6), 219 (17), 212 (6), 133 (100), 131 (8), 120 (2), 105 (4),
91 (32), 69 (7).
References
4.5.3. Trimethyl-(1-phenoxy-allyl)-stannane (8a)
This compound was synthesized from (1-chloro-
allyloxy)-benzene (vide supra) and Me3Snꢀ in HMPA,
in 50% yield (determined by GC), according to the
following general procedure: To an HMPA solution of
sodium trimethyltin (4.7 mmol in 2 ml of solvent)
prepared as described above, (1-chloro-allyloxy)-ben-
zene (2.4 mmol dissolved in 1 ml of HMPA) was slowly
introduced by syringe, and the mixture was left stirring
to react for 48 h at room temperature (r.t.) in the dark.
The reaction was worked up as usual, and the crude was
chromatographed over silca-gel column chromatogra-
phy, using CH2Cl2:petroleum ether (5:95) as eluants
[1] K.R. Wursthorn, H.G. Kuivila, G.F. Smith, J. Am. Chem. Soc.
100 (1978) 2779.
[2] (a) J.P. Quintard, S. Hauvette-Frey, M. Pereyre, J. Organomet.
Chem. 112 (1976) C11;
(b) J.P. Quintard, S. Hauvette-Frey, M. Pereyre, J. Organomet.
Chem. 159 (1978) 147.
[3] C.C. Yammal, J.C. Podesta´, R.A. Rossi, J. Org. Chem. 57 (1992)
5720.
[4] C.C. Yammal, J.C. Podesta´, R.A. Rossi, J. Organomet. Chem.
509 (1996) 1.
[5] (a) E.F. Co´rsico, R.A. Rossi, Synlett (2000) 227;
(b) E.F. Co´rsico, R.A. Rossi, Synlett (2000) 230.
[6] D. Wittenberg, H. Gilman, Quaterly Rev. (1960) 116.
[7] A. Postigo, R.A. Rossi, Org. Lett. 8 (2001) 1197.
[8] A. Postigo, S.E. Vaillard, R.A. Rossi, unpublished results.
[9] J.F. Bunnett, Acc. Chem. Res. 11 (1978) 413.
[10] It is known that 2,3-dehydrobenzotrifluoride, upon reaction with
PhLi, affords 46% of the respective meta-substituted isomer. See
R.W. Hoffmann, Dehydrobenzene and Cycloalkanes, Verlag
(39% isolated yield), and characterized by 1H-NMR,
1
MS, and EIHRMS data as follow: H-NMR (in
GCÁ
/
CDCl3) d (ppm): 0.26 (s, 9H), 3.96 (dt, 1H), 5.45 (tq, 2
H), 5.35 (cplx. m, 1H), 6.80 (cplx. m, 2H), 6.90 (cplx.m,
1H), 7.28 (cplx.m, 2H). HRMS ([M]ꢂꢀ
/
15 for
MS
Chemie. GmbH. Academic Press, 11 (1967) 106Á107.
/
C11H15O120Sn): 283.0160. Calc.: 283.0145. GCÁ
/
[11] (a) For reviews, see: (a) R.A. Rossi, R.H. de Rossi, Aromatic
Substitution by the SRN1 Mechanism; ACS Monograph 178;
Washington, D.C., 1983;
(m/z, %): 298 ([M]ꢂ., 5), 283 (25), 281 (20), 279 (15),
253 (2), 213 (3), 166 (15), 165 (100), 163 (86), 161 (51),
150 (5), 133 (23), 117 (4), 115 (3), 105 (28), 94 (6), 77
(18), 51 (15).
(b) R.A. Rossi, A.B. Pierini, A.N. Santiago, Aromatic Substitu-
tion by the SRN1 Reaction. In: L.A. Paquette. and R. Bittman
(Eds.), Organic Reactions, Wiley, New York, 1999; Vol. 54, pp. 1-
271;
(c) R.K. Norris, in: B.M. Trost (Ed.), Comprehensive Organic
Synthesis, 4 (1991) 451;
(d) R.A. Rossi, A.B Pierini, A.B. Pene´nory, in: S. Patai, Z.
˜ ˜
4.5.4. Compound 8b
/
HRMS ([M]ꢂꢀ15 for C11H15O117Sn): 283.0145.
Calc.: 283.0166.
Rappoport (Eds.), The Chemistry of Functional Group, Wiley,
Chichester, Suppl. D2, Ch. 24, 1995, p. 1395.
[12] (a) G.F. Meijs, A.L.J. Beckwith, J. Am. Chem. Soc. 108 (1986)
5890;
4.5.5. (2-Allyloxy-phenyl)-trimethyl-silane (11)
The compound [21] was isolated as indicated above,
(b) A.L.J. Beckwith, G.F. Meijs, J. Org. Chem. 52 (1987) 1922;
(c) A.L.J. Beckwith, S.M. Palacios, J. Phys. Org. Chem. 4 (1991)
404.
and characterized by 1H-NMR and 13C-NMR, and
1
MS data as follow: H-NMR (in CD3COCD3) d
GCÁ
/
[13] S.M. Moerlein, J. Organomet. Chem. 319 (1987) 29.
[14] C. Eaborn, H.L. Hornfeld, D.R.M. Walton, J. Organomet.
Chem. 10 (1967) 529.
(ppm): 0.35 (ss, 9 H), 4.68 (dt, 2H, Jꢁ
/1.46, 3.3, 5.5 Hz),
5.33 (dq, 1H, Jꢁ1.46, 3.3, 12.1 Hz), 5.51 (dq, 1H, Jꢁ
/
/

A. Postigo et al. / Journal of Organometallic Chemistry 656 (2002) 108Á
/115
115
[15] A. Smith, W. Adcock, W. Kitching, J. Am. Chem. Soc. 92 (1970)
6140.
[20] (a) Y.V. Krotikov, V.S. Zargorodnii, A.A. Petrov, Metallorg.
Khim. 54 (1992) 960;
[16] J. Nishimura, J. Furukawa, N. Kawabata, J. Organomet. Chem.
29 (1971) 237.
(b) Y.V. Krotikov, V.S. Zargorodnii, A.A. Petrov, Chem. Abstr.
118 (1993) 81060z.
[17] M.E. Freeburger, L. Spialter, J. Am. Chem. Soc. 93 (1971)
1894.
[21] (a) Y.V. Bogolepova, I.V. Vygoeskaya, G.A. Kliger, S.M. Loktev,
Nafte Khimiya 31 (1991) 805;
[18] W. Haubold, J. Herdtle, W. Gollinger, J. Organomet. Chem. 315
(1986) 1.
(b) Y.V. Bogolepova, I.V. Vygoeskaya, G.A. Kliger, S.M. Loktev,
Chem. Abstr. 116 (1992) 12827y;
[19] A.N. Abeywickrema, A.L.J. Beckwith, Tetrahedron Lett. (1986)
109.
(c) J.P. Pillot, J. Dunogues, J. Geval, Eur. Polym. J. 25 (1989)
285.