Reversibility in free-radical reactions of aryltellurides with tributylstannyl, tributylgermyl and tris(trimethylsilyl)silyl radicals

Article abstract of DOI:10.1016/S0022-328X(97)00491-9

¹H, ¹³C, ²⁹Si, ⁷⁷Se, ¹¹⁹Sn and ¹²⁵Te NMR spectroscopies reveal that methyl, primary and secondary alkyl radicals, generated through the reaction of aryltelluroalkanes (4-9) with tributyltin hydride, tributylgermanium hydride or tris(trimethylsilyl)silane) under standard radical conditions (benzene, AIBN) are capable of displacing tributylstannyl, tributylgermyl and tris(trimethylsilyl)silyl radicals from aryltellurotributylstannanes (1, 2), aryltellurotributylgermanes (10, 11) and aryltellurotris(trimethylsilyl)silanes (13, 14) respectively. These observations are in agreement with high-level ab initio molecular orbital studies. Calculations using a (valence) double-ζ pseudopotential basis set supplemented with polarization functions and with the inclusion of electron correlation (MP2/DZP) predict energy barriers for the displacement of stannyl (SnH₃), germyl (GeH₃) and trisilylsilyl ((H₃Si)₃)Si) radicals by methyl, ethyl and iso-propyl radicals to lie between 22 and 39 kJ mol^-1, with reverse barriers of between 12 and 40 kJ mol^-1. Consequently, the use of aryltellurides as alkyl radical precursors together with (standard) chain-carrying reagents such as tributyltin hydride, tributylgermanium hydride and tris(trimethylsilyl)silane may be complicated with equilibria which may result in diminished reaction yields.

Full text of DOI:10.1016/S0022-328X(97)00491-9

Ž
.
Journal of Organometallic Chemistry 552 1998 145–157
Reversibility in free-radical reactions of aryltellurides with
ž
/
tributylstannyl, tributylgermyl and tris trimethylsilyl silyl radicals
)
Carl H. Schiesser , Melissa A. Skidmore
School of Chemistry, The UniÕersity of Melbourne, ParkÕille, Melbourne, Victoria 3052, Australia
Received 25 June 1997
Abstract
1H, 13C, 29Si, 77Se, 119Sn and 125Te NMR spectroscopies reveal that methyl, primary and secondary alkyl radicals, generated through
Ž
.
Ž
.
.
the reaction of aryltelluroalkanes 4–9 with tributyltin hydride, tributylgermanium hydride or tris trimethylsilyl silane under standard
Ž
.
Ž
.
radical conditions benzene, AIBN are capable of displacing tributylstannyl, tributylgermyl and tris trimethylsilyl silyl radicals from
Ž
.
Ž
.
Ž
.
Ž
.
aryltellurotributylstannanes 1, 2 , aryltellurotributylgermanes 10, 11 and aryltellurotris trimethylsilyl silanes 13, 14 respectively.
Ž
.
These observations are in agreement with high-level ab initio molecular orbital studies. Calculations using a valence double-z
Ž
.
pseudopotential basis set supplemented with polarization functions and with the inclusion of electron correlation MP2rDZP predict
Ž
.
Ž
.
ŽŽ
. .
.
energy barriers for the displacement of stannyl SnH3 , germyl GeH3 and trisilylsilyl H3Si Si radicals by methyl, ethyl and
3
iso-propyl radicals to lie between 22 and 39 kJ moly1, with reverse barriers of between 12 and 40 kJ moly1. Consequently, the use of
Ž
.
aryltellurides as alkyl radical precursors together with standard chain-carrying reagents such as tributyltin hydride, tributylgermanium
Ž
.
hydride and tris trimethylsilyl silane may be complicated with equilibria which may result in diminished reaction yields. q 1998 Elsevier
Science S.A.
Keywords: Alkyl radicals; Molecular orbital studies; Energy barriers; Aryltellurides
w x
constants for hydrogen transfer 13–20 to alkyl radicals
1. Introduction
6
ca. 10 My1 sy1 258 ; Eq. 4 provide for reagents
Ž
Ž
.
.
Free-radical chemistry currently enjoys unprece-
dented levels of popularity among synthetic chemists.
This is undoubtedly due to current levels of understand-
ing of the factors which govern the regio- and stereo-
chemistry of free-radical processes which did not exist
some 15–20 yrs ago and the ready availability of
reagents and precursors tailor-made for free-radical re-
w
x
superior to their silicon 21–27 and germanium
w
x
Ž
.
18,28,29 counterparts; only tris trimethylsilyl silane ri-
w
x
vals trialkyltin hydrides in its synthetic utility 26,27 . A
knowledge of rate constants is crucial to the successful
design of synthetic procedures involving these reagents.
Stannane chain-carrying reagents are useful because a
knowledge of the rate constants associated with hydro-
gen transfer and halogen abstraction allow, through
control of substrate concentration, necessary selectivity
w
x
actions 1–11 . Indeed, the discovery that trialkyltin
hydrides and alkyl halides react in a radical chain
manner to afford the corresponding trialkyltin halide
w x
criteria to be met 1 . Today, many classes of free-radi-
Ž
.
and alkane Eq. 1 marks a major turning point in the
development of radical techniques with synthetic appli-
cability. Tributyltin hydride and to a lesser extent,
triphenyltin hydride have been the reagents of choice
cal precursor have been developed for use with chain-
carrying reagents such as tributyltin hydride. More com-
monly used examples include dithio- and thionocarbon-
w
x
Ž
.
ates, 30–32 pyridine-2-thioneoxycarbonyl PTOC es-
w x
1 . Their ready availability and favourable rate con-
stants for attack of the corresponding tin-centred radi-
w
x
w
x
w x
ters, 30,33 imidate esters 34 and carbamates 35 as
w
x
well as aryl sulfides, selenides and tellurides 12 .
The ability to provide quality data relating to the
intimate details of reaction mechanisms is of critical
importance to their understanding. It is no accident that
the increase in the level of understanding of intramolec-
3
9
Ž
cals at the halogen atom in alkyl halides ca. 10 –10
My1 sy1 258 ; Eq. 3 , 12 coupled with useful rate
Ž
.
. w x
)
Corresponding author.
0022-328Xr98r$19.00 q 1998 Elsevier Science S.A. All rights reserved.

(
)
146
C.H. Schiesser, M.A. SkidmorerJournal of Organometallic Chemistry 552 1998 145–157
Ž
.
ular homolytic addition ring-closure chemistry pro-
vided by many workers over the last two decades has
lead to the recent rapid increase in the use of that
methodology in synthesis. The fine detail able to be
provided by quality quantum computational data are
unrivalled; in no other way can the geometry and
electronic structure of transition states and unstable
intermediates along reaction pathways be understood.
phenyltellurotributylstannane, phenyltellurotributylger-
Ž
.
mane and phenyltellurotris trimethylsilyl silane respec-
tively, results which are consistent with ab initio molec-
ular orbital calculations.
2. Results and discussion
2.1. Reactions inÕolÕing phenyltellurotributylstannane
and 4-fluorophenyl tellurotributylstannane
init.
(
)
RyXqBu3SnH ™ RyHqBu3SnX
1
Ž .
Ž .
Ž
Phenyltellurotributylstannane
1
and 4-fluoro-
Bu3SnHqinit ™ Bu3Sn8
2
Ž .
.
Ž .
phenyl tellurotributylstannane 2 were prepared by re-
Ž
. w x
action of phenyltellurocyclohexane 3: ArsPh 39 or
Bu3Sn8qRX ™ Bu3SnXqR8
3
Ž .
Ž
.
Ž
. w x
4-fluorophenyl tellurocyclohexane 3: Ars4-FPh 40
R8qBu3SnH ™ Bu3Sn8qRH
4
Ž .
Ž
.
with tributyltin hydride in d6-benzene AIBN initiator
Ž
.
Ž
.
Scheme 1 . The tellurides 1, 2 proved to be light
1
D
Ea
s
26.3 kJ.moly
|
sensitive and generally unstable, attempted removal of
the solvent resulted in rapid decomposition. In benzene,
under nitrogen and in the dark, they appear to have an
indefinite lifetime. Consequently, 1 and 2 were charac-
terized by 1H, 13C, 119Sn and 125Te NMR spectro-
scopies. Specifically, 1 exhibits a singlet at dy1.3 ppm
8CH3 qH3SnTeH
H3Sn8qCH3TeH
D
Eas30.1 kJ.moly1
5
Ž .
Our recent interest in the development of new ho-
molytic substitution methods and modified stannanes
for use in free-radical synthesis has been aided by the
computer modelling of various homolytic substitution
reactions through the use of ab initio molecular orbital
theory. Of importance to the work described in this
paper are the results of high-level ab initio investiga-
tions into the attack of silyl, germyl and stannyl radicals
in the 119Sn NMR spectrum with satellite coupling to
Te J 119Sn–125Te s2693 Hz , while the 125Te NMR
spectrum of 1 revealed a singlet at d y209.4 ppm with
the expected satellite coupling to 117Sn and 119Sn
125
w
Ž
.
x
w
Ž
.
J
117Sn–125Te s 2578 Hz;
J
119Sn–125Te s 2692
Ž .
Hz . Similarly, 2 exhibits analogous signals: 119Sn NMR
x
119Sn–125Te s2656 Hz ; 125Te NMR d
Ž . x
w
x
w
at the halogen atom in halomethanes 36 and the
chalcogen atom in the analogous sulfides, selenides and
d 0.4
J
y213.9 J 125Te–117Sn s2526 Hz, J 125Te–119Sn s
w
Ž
.
Ž
.
w
x
x
tellurides 37,38 . These studies predict that, in accor-
2656 Hz . These spectroscopic data leave no doubt that
w
x
Ž
.
dance with expectation, 12 stannyl radicals react at the
halogen or chalcogen atom in halo- and
chalcogenomethanes in the following reactivity order:
I G Te )) Br G Se )) Cl G S. Somewhat surpris-
ingly, reactions at tellurium were calculated to be re-
the tellurostannanes 1, 2 had indeed been prepared in
situ.
Ž .
.
When one equivalent of 4-fluorotelluroanisole 4
w
x
41
Ž125
.
Ž
Te NMR, d 342.8 , tributyltin hydride 5 mol%
Ž
.
and AIBN 5 mol% were introduced into the NMR
solution of 1 and the colourless solution heated at 808C
for 2 h, 119Sn and 125Te NMR spectroscopies revealed
Ž
.
versible. Calculations involving a valence double-z
Ž
.
pseudopotential DZP basis set and electron correlation
Ž
.
Ž
.
Ž .
w x
MP2, QCISD predict that stannyl radical H3Sn Ø
undergoes reversible free-radical attack at the tellurium
the presence of telluroanisole 5 41 and 2, in addition
Ž
.
to starting material Fig. 1 by comparison with authen-
tic samples. Importantly, when 1 was heated at 808C in
benzene either in the absence of any reagents or in the
Ž
.
atom in methanetellurol MeTeH with expulsion of
Ž
.
methyl radical Eq. 5 . Our MP2rDZP calculations
provide energy barriers of 28.8 and 40.3 kJ moly1 for
the displacement of H3Sn Ø by methyl radical and the
re v e rs e re a c tio n re s p e c tiv e ly , w h ile
QCISDrDZPrrMP2rDZP calculations give values of
26.3 and 30.1 kJ moly1 for the same two reactions
Ž .
4
w x
41 and trib-
presence of 4-fluorotelluroanisole
Ž
.
Ž
.
utyltin hydride 5 mol% with no initiator AIBN , no
reaction was observed by 119Sn and 125Te NMR spectro-
scopies after 2 h at 808C.
We postulate that the formation of 2 and 5 involves
homolytic substitution by methyl radical, generated by
w
x
37,38 . Intrigued by this apparently contra-intuitive re-
sult and given the increased use of aryltellurides as
free-radical precursors, we began to explore the reaction
of methyl and other radicals with phenyltellurotributyl-
Ž
.
stannane and other group IV -containing tellurides. We
now report that methyl, primary- and secondary- alkyl
radicals are capable of displacing tributylstannyl, trib-
Ž
.
utylgermyl and tris trimethylsilyl silyl radicals from
Scheme 1.

(
)
C.H. Schiesser, M.A. SkidmorerJournal of Organometallic Chemistry 552 1998 145–157
147
Fig. 1. 125Te and 119Sn NMR spectra d6-benzene of the reaction of 1 with 4 at 808 in the presence of 5% Bu3SnH and 5% AIBN.
Ž
.
Ž .
reaction of 4-fluorotelluroanisole 4 with tributylstan-
cal from organo-tellurides assumes some importance
given the increased usage of tellurides as radical precur-
sors in synthesis. In order to shed further light on this
nyl radical, at the tellurium atom in 1 with expulsion of
further chain-carrying tributylstannyl radical Scheme
Ž
.
2 .
When the reaction was repeated using 2 and tel-
luroanisole 5 as starting materials, 119Sn and 125Te
NMR techniques revealed the presence of 1 and 4, as
expected on the basis of the mechanism depicted in
Scheme 2.
Ž .
The data presented so far are in agreement with our
Ž
.
previous ab initio calculations Eq. 5 and verify that
methyl radical is capable of displacing stannyl radical
from tellurium in organo-tellurides. Both experimen-
tally-derived and computational data raise the question
of the generality of these radical processes. Whether or
not alkyl radicals in general are also capable of displac-
ing chain-carrying species such as tributylstannyl radi-
Scheme 2.

(
)
148
C.H. Schiesser, M.A. SkidmorerJournal of Organometallic Chemistry 552 1998 145–157
question, we set about examining reactions analogous to
Ž .
and 4-fluorophenyltelluroethane 7 , or 2- phenyltel-
2.2. Reactions inÕolÕing phenyltellurotributylgermane
and 4-fluorophenyl tellurotributylgermane
w
x
(
)
those described above using phenyltelluroethane 6 42
Ž .
Ž
.
Ž .
w
x
Ž
.
luro butane 8 43 and 2- 4-fluorophenyltelluro butane
Ž .
Despite their widespread use in free-radical synthe-
ses, trialkylstannanes often deliver hydrogen atom to
alkyl radicals with rate constants too high for acceptable
9 instead of 4 and 5.
Tellurides 4–9 were prepared from the correspond-
ing alkyl bromide or iodide and either diphenyl ditel-
luride 44 or di 4-fluorophenyl ditelluride 44 accord-
ing to standard procedures. 40 When tellurides 6 or 8
Ž
.
w x
synthetic outcomes 1 . On occasion, despite the inher-
w
x
Ž
.
x
w x
ent expense, trialkylgermanes have replaced their tin
counterparts; their lower rate constants for the delivery
of hydrogen often lead to increased reaction yields,
especially when the primary bond-forming reaction of
w
Ž
.
were reacted with 4-fluorophenyl tellurotributylstan-
Ž .
nane 2 in the manner described previously, once again,
125Te and 119Sn NMR spectroscopies revealed the for-
interest e.g., intramolecular homolytic addition is slow
Ž
.
Ž .
1
w
x
mation of phenyltellurotributylstannane
and tel-
18,28,29 .
lurides 7 or 9 respectively. Typical 125Te NMR spectra
are displayed in Fig. 2. Similar reactions using 7 or 9
with 1 resulted in the formation of 2 as well as 6 or 8
We therefore turned our attention to the reactions of
the germanium analogues of 1 and 2 with methyl,
primary and secondary alkyl radicals with the intention
of determining whether or not tributylgermyl radicals
are also capable of reacting reversibly with alkyltel-
Ž
.
respectively Scheme 3 . These results clearly indicate
that primary and secondary alkyl radicals are also capa-
ble of displacing tributylstannyl radicals from tellurium.
Ž
.
Ž
lurides. Phenyltellurotributylgermane 10 and 4-fluo-
125
Ž . Ž . Ž .
Te NMR spectra d6-benzene of the reaction of 2 with 6 left ; and 1 with 9 right at 808 in the presence of 5% Bu3SnH and 5%
Fig. 2.
AIBN.

(
)
C.H. Schiesser, M.A. SkidmorerJournal of Organometallic Chemistry 552 1998 145–157
149
Scheme 3.
.
Ž
.
Ž
rophenyl tellurotributylgermane 11 were prepared in
identical fashion to their tin counterparts, using tributyl-
germanium hydride and the appropriate aryltellurocy-
associated with silanes in general. Tris trimethyl-
.
Ž
.
silyl silane TTMSS was developed specifically to
overcome the poor chain propagation properties of
silanes in general and as a reagent to replace trialkyl-
stannanes which have undesirable toxicity and chro-
m atographic properties 26,27,45 . Indeed,
tris trimethylsilyl silane has been reported to behave
much more like a stannane than a silane in free-radical
Ž .
clohexane 3 . As was observed for the tellurostannanes
1, 2 , the germanium analogues 10, 11 proved to be
Ž
.
Ž
.
w
x
light-sensitive and unstable; their preparation in situ was
confirmed by careful H, 13C and 125Te NMR monitor-
1
Ž
.
Ž
.
ing of the reaction mixtures. Specifically, telluride 10
exhibits a singlet in the 125Te NMR spectrum at d
y78.7, while 11 displays the analogous singlet at d
w
x
chemistry 46 .
It therefore came as no surprise that alkyl radicals
Ž
.
Ž
y82.8. When phenyltellurotributylgermane 10 was
reacted with 5, 7 or 9 and 5 mol% tributylgermanium
hydride under identical conditions to those described for
proved to be capable of displacing tris trimethyl-
.
Ž
.
silyl silyl radicals from alkyltellurides 13, 14 . Once
again, due to inherent stability problems, reagents were
prepared in situ; phenyltellurotris trimethylsilyl silane
13 and 4-fluorophenyl tellurotris trimethylsilyl silane
the reaction of the tin analogue 1 , 125Te NMR spec-
Ž .
Ž
Ž
.
.
Ž
.
Ž
.
.
Ž
.
troscopy revealed the formation of 4-fluorophenyl tel-
Ž
.
Ž
lurotributylgermane 11 as well as 6, 8 or 10 respec-
tively. A typical 125Te NMR spectrum is shown in Fig.
3. In addition, similar reactions using 11 as well as 6, 8
or 10 as starting materials resulted in the formation of
10 together with 5, 7 or 9 respectively.
Once again, these data confirm that methyl, primary,
and secondary alkyl radicals are capable of displacing
tributylgermyl radicals from tellurogermanes 10, 11 .
These results strongly suggest that tributylgermyl radi-
cals are able to react reversibly with alkyltellurides.
14 were prepared by the reaction TTMSS with the
Ž .
appropriate aryltellurocyclohexane 3 . Tellurosilanes
13, 14 were characterized by H, 13C, 29Si and 125Te
1
Ž
.
NMR spectroscopy. Specifically, 13 exhibits a singlet at
d y192.9 with satellite coupling to 29Si J
29
w
Ž125
.
Te– Si
s135 Hz in the 125Te NMR spectrum and two singlets
x
at d y10.6 and y94.4, the latter displaying satellite
coupling to Te J 29 Si–125Te s135 Hz , in the 29Si
125
Ž
.
w
Ž
.
x
NMR spectrum, while 14 exhibits a singlets at d y195.9
Ž125
.
Ž29
.
Te , d y10.6 and y92.6 Si .
Ž
.
Ž
.
When phenyltellurotris trimethylsilyl silane 13 was
reacted with 5, 7 or 9 and 5 mol% TTMSS under
identical conditions to those described for the reaction
(
2.3. Reactions inÕolÕing phenyltellurotris trimethyl-
silyl silane and 4-fluorophenyl tellurotris trimethyl-
silyl silane
)
(
)
(
of the tin analogue 1 , 125Te NMR spectroscopy re-
)
Ž .
Ž
.
Ž
vealed the formation of 4-fluorophenyl tellurotris tri-
.
Ž
.
Trialkylsilanes are generally unsuitable reagents for
use in free-radical syntheses due to their general inabil-
ity to sustain chain reactions because of slow rates of
transfer of hydrogen atom to alkyl radicals 26,27 . Our
inability to prepare phenyltellurotriphenylsilane 12
methylsilyl silane 14 as well as 6, 8 or 10 respec-
tively. A typical 125Te NMR spectrum is shown in Fig.
3. Similarly, reactions using 14 together with 6, 8 or 10
as starting materials resulted in the formation of 13
together with 5, 7 or 9 respectively.
w
x
Ž
.
through the reaction of triphenylsilane with phenyltel-
lurocyclohexane demonstrates the inherent difficulties
Once again, these data confirm that methyl, primary,
and secondary alkyl radicals are capable of displacing

(
)
150
C.H. Schiesser, M.A. SkidmorerJournal of Organometallic Chemistry 552 1998 145–157
125
Ž
.
Ž
.
Fig. 3. Te NMR spectra d6-benzene of the reaction of 10 with 4 at 808 in the presence of 5% Bu3GeH and 5% AIBN left ; and 13 with 4 at
Ž
.
808 in the presence of 5% TTMSS and 5% AIBN right .
Ž
.
Ž
tris trimethylsilyl silyl radicals from tellurosilanes 13,
cally, 15 displays a singlet at d y23.3 in its 77Se NMR
spectrum, while 16 exhibits the analogous signal at d
y32.3. In addition, 119Sn NMR spectroscopy revealed
singlets at d 57.8 and 59.6 for 15 and 16 respectively.
.
14 . These results strongly suggest that, as was demon-
strated for tributylstannyl and germyl radicals tris tri-
methylsilyl silyl radicals are also able to react reversibly
with alkyltellurides. These results are also of signifi-
cance in inorganic chemistry, especially given the re-
Ž
.
Ž
.
Ž
.
When selenoanisole 17 , tributyltin hydride 5 mol%
Ž
.
and AIBN 5 mol% were introduced into a benzene
solution of 16 and the resultant solution heated at 808C
for 2 h, both 77Se and 119Sn NMR spectroscopy re-
Ž
.
x
cent importance of tris trimethylsilyl selenides and tel-
lurides in inorganic synthesis 47–50 .
w
Ž
.
vealed the presence of the selenostannane 15 in addi-
tion to starting materials 16, 17 by comparison with
Ž
.
2.4. Reactions inÕolÕing phenylselenotributylstannane
and 4-fluorophenyl selenotributylstannane
(
)
authentic samples. Importantly, no 4-fluorotelluroani-
Ž
.
sole 18 was observed. It seems likely that selenostan-
Ž
.
Given the widespread use of alkylselenides as free-
radical precursors, we chose to examine whether or not
tributylstannyl radicals react reversibly with alkyl se-
nane 15 arises by homolytic substitution of tributyl-
stannyl radicals generated from tributyltin hydride
Ž
.
AIBN at the selenium atom in 17 with expulsion of
w
x
lenides. To that end, phenylselenotributylstannane 51
methyl radical. These methyl radicals, in turn abstract
hydrogen atom from the stannane in preference to at-
tacking the selenium atom in 16.
Similar results are obtained when 18 is reacted with
15, namely 16 is observed in addition to starting mate-
rial. From these data we conclude that under standard
Ž
Ž
.
Ž
.
w
x
15 and 4-fluorophenyl tellurotributylstannane 51
.
16 were prepared by reaction of phenylselenocyclo-
w
x
Ž
.
hexane 51 and 4-fluorophenyl selenocyclohexane with
tributyltin hydride under standard conditions. Despite
being air-stable solids, 51 we chose to prepare solu-
tions in identical fashion to those prepared for the
w
x
Ž
radical reaction conditions i.e., tributyltin hydride,
Ž
.
.
tellurium analogues 1, 2 . These solutions were charac-
AIBN , methyl radicals are not capable of displacing
terized by 77Se and 119Sn NMR spectroscopies. Specifi-
tributylstannyl from tributylstannylselenides, strongly

(
)
C.H. Schiesser, M.A. SkidmorerJournal of Organometallic Chemistry 552 1998 145–157
151
Ž
Ž
.
w
wS
x
suggesting that tributylstannyl radicals react irreversibly
with alkyl selenides under these conditions.
eroatoms in this study exponents d z s0.284, 58
Ž
.
w
x
Ž
.
Swi
x
Ž
.
d z
s0.230, 58 d z s0.200, 59 d z
s
Gwe
x
Ž
.
Sn w
x
Ž
.
x
.
0.60, 55–57 d z Se s0.38, 58 d z Te s0.30 58
while the double-z all-electron basis sets of Dunning
2.5. Ab initio molecular orbital calculations
w
x
60 with an additional set of polarisation functions
Ž
Ž
.
Ž
.
.
exponents d z C s0.75 and p z H s1.00 were used
In order to shed further light on the ability of alkyl
radicals to displace trialkylstannyl radicals from tel-
for C, H, N and O. We refer to this basis set as DZP
w
x
Ž
.
throughout this work 37,38,61 .
lurostannanes 1, 2 , trialkylgermyl radicals from 10 and
Ž
.
Calculations were performed on a Sun Sparcserver
10r512, DEC AlphaStation 400 4r233, DEC Al-
phaServer 8400, Cray Y-MP4Er364 or Cray J916 com-
puter.
Extensive searching of the potential energy surfaces
for the reactions depicted in Scheme 4 located hyperva-
lent structures 19–25 at both SCFrDZP and
MP2rDZP levels of theory. Vibrational frequency anal-
ysis revealed that these structures correspond to transi-
11, as well as the tris trimethylsilyl silyl moiety from
Ž
.
tellurosilanes 13 and 14 , we chose to examine the
potential energy surfaces for the attack of ethyl and
iso-propyl radicals at the tellurium atom in stannyltel-
Ž
.
Ž
.
lurol H3SnTeH , germyltellurol H3GeTeH and silyl-
Ž
.
tellurol H3SiTeH , as well as attack of methyl radical
ŽŽ
Ž
.
Ž
.
.
.
.
at tris trisilyl silyltellurol H3Si SiTeH with expul-
3
Ž
sion of the appropriate group IV -centred leaving radi-
cal Scheme 4 . Comparison of the data generated in
this manner with those obtained in our previous study
Ž
.
Ž
.
tion states for the displacement of the group IV -centred
Ž . Ž .
w
x
radical Y by alkyl radical R from tellurium. At the
involving methyl radical, 37,38 as well as the experi-
mental data should serve to provide significant insight
into the nature of the reactions in question.
Ž
.
SCFrMP2 level, transition states 19–25 are predicted
to be of CS symmetry, while at the higher level
Ž
.
MP2rDZP , significant deviation from the collinear
All ab initio molecular orbital calculations were car-
w
x
arrangement of attacking and leaving radicals found at
the SCF level is predicted; all transition states adopt C1
ried out using the Gaussian 92 and 94 programs 52,53 .
Geometry optimisations were performed using standard
gradient techniques at the SCF and MP2 levels of
theory using RHF and UHF methods for closed and
Ž
.
symmetry at MP2rDZP. Dihedral angles R–Te–H–Y
of between 1508 and 1678 are predicted and are consis-
w
x
tent with our previously published data for the similar
open shell systems, respectively 54 . Further single-
point QCISD calculations were performed on each of
the MP2 optimised structures. When correlated methods
were used calculations were performed using the frozen
core approximation. Whenever geometry optimisations
were performed, vibrational frequencies were calculated
to determine the nature of located stationary points.
Calculations were performed on all reactants, products
and transition states to obtain barriers and energies of
reaction.
w
x
reactions involving methyl radical 37,38 . These devia-
tions from collinearity have been discussed in terms of a
favourable interaction of the lone-pair electrons on
chalcogen with the three-centre, three-electron ‘bond’ in
w
x
transition states of this sort 62 .
MP2rDZP—generated transition structures 19–25
are displayed in Fig. 4, while the calculated energy
Ž
.
a
Ž
barriers for the forward and reverse reactions D E1 ,
D E2a in Scheme 4 are listed in Table 1. Calculated
.
Ž
.
energies of all structures in this study can be found in
Table 2.
The valence double-z pseudopotential basis sets of
Wadt and Hay 55–57 supplemented with a single set
of d-type polarisation functions were used for the het-
w
x
Inspection of Fig. 4 reveals a close similarity be-
tween the structures generated in this study to those in
our previous study. For example, the C–Te separation
˚
Ž
.
in transition state 23, at 2.437 A MP2 is slightly
˚
longer that the similar distance in 24, at 2.411 A at the
same level of theory and somewhat shorter than the
C–Te distance in the analogous transition state involv-
˚
Ž
Ž .
.
ing methyl radical MeTe H SnH3 at 2.457 A
Ž
.
w
x
MP2rDZP 37,38 . A corresponding lengthening of
the Te–Sn distance is predicted, with MP2rDZP calcu-
˚
Ž
lated distances of 2.859 A when the attacking radical is
˚
˚
Ž
w
x
.
methyl, 37,38 2.886 A ethyl and 2.906 A iso-pro-
pyl . Similar trends are predicted for the silyl and
.
germyl containing transition states and are consistent
with the increasing degree of ‘lateness’ expected in
progressing from methyl to ethyl and iso-propyl radi-
cals as attacking species. It is interesting to compare
Ž
.
these data with those for transition state 25 in which
Scheme 4.

(
)
152
C.H. Schiesser, M.A. SkidmorerJournal of Organometallic Chemistry 552 1998 145–157
Ž
. Ž
.
Fig. 4. MP2rDZP generated structures of transition states 19–25 . SCFrDZP data in parentheses .
Ž
.
the tris trisilyl silyl radical is involved as the leaving
In other related work, we have established that the
inclusion of electron correlation is necessary in order to
reliably model homolytic substitution reactions at a
species. The MP2rDZP calculated C–Te distance, at
˚
2.478 A, is significantly longer than the corresponding
w
x
separation in the similar transition state involving SiH3
variety of main-group elements 12,37,38,59,61 . It is
therefore not surprising that the SCF level of theory also
models the reactions in this study poorly. Interestingly,
the QCISDrDZPrrMP2rDZP single-point data also
diverge somewhat from the MP2-generated data. We
believe that these differences are most likely to be due
˚
Ž
w
Ž .
.
MeTe H SiH3 which is calculated to be 2.393 A
x
37,38 . Indeed, this distance resembles the C–Te dis-
tance in the reaction involving S nH
3
˚
.
Ž
Ž .
.Ž
MeTe H SnH3 2.489 A more closely. This result is
interesting because, as previously mentioned, TTMSS
often behaves more like a stannane than a silane, 46 an
observation which has been extensively exploited in
w
x
to the differences between the MP2rDZP and
1
QCISDrDZP potential energy surfaces.
Unfortu-
w x
free-radical chemistry 1 .
The calculated energy barriers presented in Table 1
also reveal some interesting trends. Firstly, the
SCFrDZP generated data provide energy barriers which
are significantly higher for the forward reactions in all
cases than the corresponding MP2rDZP generated data.
1 For example, Me2Cl is predicted to be a transition state at both
Ž
.
MP2rDZP and QCISDrDZP fully optimized levels of theory. The
˚
former calculations provides a C–Cl distance of 2.062 A, while at the
higher level this separation is predicted to be 2.117 A. C.H. Schiesser
˚
and L.M. Wild, unpublished.

(
)
C.H. Schiesser, M.A. SkidmorerJournal of Organometallic Chemistry 552 1998 145–157
153
Table 1
a
a
SCFrDZP, MP2rDZP and QCISDrDZPrrMP2rDZP calculated energy barriersa for the forward D E1 and reverse D E2 reactions in this
Ž
.
Ž
.
Ž
.
study Scheme 4
Ž .
Radical R
TS
SCF
D E1a
MP2
D E1a
QCISDrMP2
D E1a
D E2a
D E2a
D E2a
92.9b
100.5
109.6
79.9b
86.6
36.5b
31.8
28.7
39.1b
33.4
29.8
41.1b
36.1
33.9
43.8
37.6b
38.9
37.2
30.7b
30.0
26.0
28.8b
26.8
22.4
29.9
19.1b
13.5
12.8
29.0b
23.2
20.2
40.3b
33.2
29.8
16.1
37.2b
46.0
49.2
28.4b
31.9
32.7
26.3b
27.8
27.6
28.0
9.6b
9.4
8.6
Ž .
MeTe H SiH3
MeP
EtP
i-PrP
MeP
EtP
i-PrP
MeP
EtP
i-PrP
MeP
w
x
19
w
20
x
Ž .
MeTe H GeH3
19.9b
13.7
10.7
30.1b
22.6
18.4
16.6
w
21
22
x
w
x
95.0
Ž .
MeTe H SnH3
74.4b
80.6
w
23
w
24
w
25
x
x
x
88.6
77.8
a Energies in kJ moly1
.
b
w
x
Taken from 37,38 .
Ž
.
Ž
.
nately, QCISDrDZP full optimizations on the struc-
tures in this study are tasks beyond our current re-
sources.
experimental NMR data. Once again, differences in
the MP2rDZP and QCISDrDZP generated potential
energy surfaces are most likely to be responsible for the
Ž
As expected from previous calculations, reactions of
alkyl radicals at silyltellurol H3SiTeH with expulsion
significant divergence between MP2 and QCISD single
Ž
.
.
point data.
of SiH3 are predicted to be endothermic. MP2rDZP
Inspection of Table 1 reveals that in all cases in this
a
Ž
.
calculations predict energy barriers D E1 , Scheme 4 of
study, MP2rDZP calculations predict that alkyl radical
37—39 kJ moly1, regardless of the attacking species,
attack at tellurium with expulsion of SnH3 is exother-
a
a
Ž
.
Ž
.
Ž .
Ž
.
Ž
.
with reverse barriers D E2 of 19.1 Me , 13.5 Et and
mic, with forward barriers D E1 of 28.8 Me , 26.8
12.8 i-Pr kJ moly1. The trends in D E2a are as ex-
Et and 22.4 i-Pr kJ moly1. The analogous reverse
Ž
.
Ž
.
Ž
.
a
Ž
Ž
.
Ž
.
Ž
.
pected on the basis of leaving radical stability i-Pr)Et
barriers D E2 are predicted to be 40.3 Me , 33.2 Et
)Me . Similar trends in D E1 and D E2a are observed
a
and 29.8 i-Pr kJ moly1. Similar considerations to
those applied to the analogous reactions involving
germyl radical leads to predicted equilibrium constants
.
Ž
.
at the QCISDrDZPrrMP2rDZP level of theory.
Analogous calculations involving germyltellurol
H3GeTeH reveal energy barriers for the forward and
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
KE of approximately 50 Me , 10 Et and 12 i-Pr .
reverse reactions which are much more similar in mag-
Clearly then, alkyl radicals in general are predicted to
be capable of displacing stannyl radicals from tellurium,
once again, in agreement with the previously presented
nitude than the previously presented silicon data.
a
Ž
.
MP2rDZP calculations predict barriers D E1 of 30.7,
30.0 and 26.0 kJ moly1 for the reactions of methyl,
ethyl and iso-propyl radicals respectively at tellurium
with expulsion of GeH3, while the reverse reactions are
predicted to proceed with barriers of 29.0, 23.2 and 20.2
kJ moly1 respectively. Differences of only 1.7 to 6.8 kJ
moly1 in the forward and reverse barriers would be
expected to manifest themselves as equilibria, with
Ž
.
experimental NMR data.
Ž
.
It is interesting to note that tris trisilyl silyl radicals
are calculated to react exothermically with methyltel-
lurol; MP2rDZP calculations predict D E2a to be 16.1
kJ moly1, with the barrier for methyl radical attack
a
D E1 predicted to be 29.9 kJ moly1. Similar data are
Ž
.
provided by the single-point QCISD data. We have
previously reported that the silyl group is a poor model
for the trimethylsilyl group in ab initio calculations
Ž
.
equilibrium constants KE of between approximately
0.6 and 0.1 at 808C, assuming similar entropy terms.
w x
involving substituted carbenium ions; 64 it is highly
These data, combined with the MP2rDZP calculated
y1
Ž
.
and the knowl-
Ž
.
activation energies 20–31 kJ mol
likely that Si SiH3 also performs poorly as a model
3
Ž
.
edge that many effective chain-carrying radical reac-
tions proceed with barriers of less than approximately
for Si SiMe3 in this study.
3
30 kJ moly1 12 suggest that reactions involving tri-
alkylgermyl radicals at alkyltellurides with the forma-
tion of alkyl radicals are indeed reversible and are
consistent with our NMR studies detailed above. Inter-
w
x
2.6. Equilibria in radical reactions
Because of the nature of many free-radical reactions
which rely on the establishment of chain mechanisms
Ž
.
Ž
.
estingly, the QCISDrMP2 single point calculations
e.g., Scheme 2 , it is unlikely that true equilibria can
predict these reactions to be essentially irreversible
ever be established because of competing chain-propa-
gating steps. Consequently, the NMR data presented in
y4
Ž
.
and are not consistent with the available
KE ;10

(
)
154
C.H. Schiesser, M.A. SkidmorerJournal of Organometallic Chemistry 552 1998 145–157
Table 2
SCFrDZP, MP2rDZP and QCISDrDZPa calculated energiesb of the reactants, products and transition states in this study
Structure
SCF
MP2
QCISDrMP2a
MeP
EtP
i-PrP
PSiH3
PGeH3
PSnH3
y39.57176c,d
y78.61706c
y117.66350c
y5.46984d
y5.34597d
y4.94363c,d
y20.31222
y48.04744d
y87.08781
y126.13002
y13.96715d
y13.83733d
y13.43211d
y28.80051
y53.50337d
y53.37852d
y52.97540d
y92.54590
y92.42140
y92.01846
y131.58919
y131.46464
y131.06187
y68.34303
y39.69727c,d
y78.88130c
y118.06814c
y5.55964d
y5.43346d
y5.02396c,d
y20.69008
y48.29930d
y87.48223
y126.66945
y14.17021d
y14.03700d
y13.62247d
y29.29736
y53.85230d
y53.72171d
y53.80792d
y93.03601
y92.90687
y92.49354
y132.22418
y132.09522
y131.68208
y68.98324
y39.71891c,d
y78.91699c
y118.11713c
y5.58357d
y5.45694d
y5.04697c,d
y20.77086
y48.33181d
y87.52711
y126.72578
y14.20770d
y14.07400d
y13.65911d
y29.38810
y53.91172d
y53.78117d
y53.36730d
y93.10713
y92.97881
y92.56548
y132.30611
y132.17866
y131.76572
y69.09634
Ž
.
3
PSi SiH3
MeTeH
EtTeH
i PrTeH
H3SiTeH
H3GeTeH
H3SnTeH
Ž
.
H3Si 3SiTeH
Ž .
MeTe H SiH3
MeTe H GeH3
MeTe H SnH3
Ž .
Ž .
Ž .
Žw x.
EtTe H SiH3 19
Ž . Žw x.
EtTe H GeH3 20
Ž . Žw x.
EtTe H SnH3 21
i
Ž .
Ž .
Žw x.
Žw x.
PrTe H SiH3 22
i
i
PrTe H GeH3 23
Ž . Žw x.
PrTe H SnH3 24
Ž . . Žw x.
Ž
MeTe H Si SiH3 25
3
aQCISDrDZPrMP2rDZP.
b
y1
Ž
.
.
Energies in Hartrees 1Eh s2626 kJmol
cw
x
61,63 .
dw
x
37,38 .
this paper serve only to establish that both the forward
and reverse reactions as presented in Schemes 2 and 3
are occurring; the determination of equilibrium constant
data from these NMR experiments is a highly unlikely
outcome.
tributyltin hydride; it is therefore not surprising that
alkyldestannylation at tellurium is competitive with hy-
drogen abstraction from tributyltin hydride under the
reaction conditions described.
Similar considerations when applied to the reaction
of methyl radical at the selenium atom in 15, with an
The calculated value of 22.4 kJ moly1 for D E1a in
the reaction of iso-propyl radical with stannyltellurol is
likely to represent an upper limit for the activation
energy for the reaction of iso-propylradical with aryltel-
approximate activation energy of 50 kJ moly1 37,38
w
x
y1
sy1 estimate of 9 leads to approxi-
w
x
and log ArM
mate relative reaction rates of 2=10y4 and 1.0 for
attack of methyl radical at 15 and tributyltin hydride
respectively under the described reaction conditions.
With this in mind, it is not surprising that methyl radical
failed to displace the tributylstannyl radical from the
arylselenotributyltin species 17, 18 ; abstraction of hy-
drogen atom from the tributylstannane present would be
expected to be the favoured process.
Ž
.
lurotributyltin 1, 2 . Given that a value of 9 represents
y1
a reasonable estimate for the log ArM sy1 term in
w
x
the Arrhenius expression for intermolecular homolytic
substitution at a heavy main-group heteroatom, 65
2
w
x
Ž
.
the rate constant for homolytic displacement of tributyl-
stannyl radical from 1 by iso-propyl radical can be
estimated to be greater than 4.8=105 My1 sy1 at
808C. The rate constant for hydrogen abstraction by
secondary alkyl radicals from tributyltin hydride is well
2.7. Conclusion
established to be 3.6=106 My1 sy1 at 808C. 19,65
w
x
29Si, 119Sn and 125Te NMR spectroscopy together
with ab initio molecular orbital calculations confirm that
under standard radical conditions, alkyl radicals are
capable of displacing tributylstannyl, tributylgermyl and
Ž
.
As the reactions in question Scheme 3 are carried out
in the presence of 5 mol% tributyltin hydride, the
iso-propyl radical can be estimated to react more than
approximately 3 times more effectively with 1 than with
Ž
.
tris trimethylsilyl silyl radicals from the tellurium atom
in alkyl tellurides.
2 Based on the reaction of 1-octyl radical with iodocyclohexane.
Consequently, reactions involving aryltellurides as
radical precursors together with chain carrying species
w
x
65

(
)
C.H. Schiesser, M.A. SkidmorerJournal of Organometallic Chemistry 552 1998 145–157
155
1
colourless solution was examined by H, 13C, 119Sn and
125Te NMR spectroscopy which revealed the absence of
starting materials and the in situ formation of 1 together
such as tributyltin hydride, tributylgermium hydride and
Ž
.
tris trimethylsilyl silane may involve reversible ho-
molytic substitution at tellurium leading to reduced
reactions yields.
with cyclohexane as the only products of reaction.
1
Ž
.
Ž
H NMR: d 0.85 t, 9H, Js7.5 Hz ; 1.09–1.14 m,
.
Ž
.
Ž
.
6H ; 1.20–1.31 m, 6H ; 1.41 s, cyclohexane ; 1.44–
Ž
.
Ž
.
Ž
3. Experimental
1.55 m, 6H ; 6.85 t, 2H, Js7.5 Hz ; 7.00 t, 1H,
Js7.1 Hz ; 7.85 dd, 2H, Js7.2, 1.2 Hz . 13C NMR:
.
J
Ž
.
13C-119Sn s296.6 Hz,
13C–117Sn s
Ž
.
w
Ž
.
Ž
J
.
Tris trimethylsilyl silane was purchased from Fluka,
tellurium powder was purchased from Aldrich and trib-
d
13.5
283.0 Hz ; 13.9; 27.3 J 13C–119Sn s56.6 Hz ; 27.3
x
w
Ž
.
x
3
13C–119Sn s21.9 Hz ; 103.5;
Ž
.
w
Ž
.
x
J
x
utylgermanium chloride was a gift. Tributylgerma-
cyclohexane ; 29.9
J
125.2; 129.0; 142.2. 119Sn NMR: d y1.3
119Sn–
Ž
w
nium hydride and tributyltin hydride were prepared
Sn– C s299.1 Hz ; 125Te
1
125
13
Ž119
w
.
w
x
.
following literature procedures 66,67 . Compounds 3–8,
17, 18 were prepared according to previously published
Te s2693 Hz,
J
Ž
NMR:
d
y209.4
J
125 Te–119 Sn s 2693 Hz,
.
125Te–117Sn s2578 Hz .
w
x
Ž
.
x
procedures 40,42,43,68 .
J
w
Ž
.
x
Ž .
4-Fluorophenyl telluro ethane 7 was prepared fol-
Ž .
lowing Standard Protocol E in our previously pub-
lished procedure 40 using bis 4-fluorophenyl ditel-
(
)
( )
3.2. 4-fluorophenyl tellurotributylstannane 2
w
x
Ž
.
Ž
luride and iodoethane. Kugelrohr distillation 458Cr0.08
¨
The title compound was prepared according to the
general procedure using 4-fluorophenyl tellurocyc-
mm afforded 7 as a yellow oil 70% . 1H NMR
.
Ž
.
Ž
.
Ž
Ž
Ž
.
Ž
Ž
CDCl3: d 1.37 t, 3H, Js7.5 Hz ; 2.45 q, 2H,
lohexane.
13
.
.
Ž .
.
1
Js7.5 Hz ; 6.59 m, 2H ; 7.42 m . C NMR C6 D6 :
Ž
.
Ž
H NMR: d 0.87 t, 9H, Js7.5 Hz ; 1.09–1.14 m,
19
Ž
.
w
Ž13
.
d 0.6; 17.4; 105.5 JC – F s3.5 Hz ; 116.6 J C– F
.
Ž
.
Ž
.
6H ; 1.20–1.31 m, 6H ; 1.41 s, cyclohexane ; 1.44–
19
x
w
Ž13
. x
1.55 m, 6H ; 6.58 m, 2H ; 7.27 m, 2H . 13C NMR: d
s 20.8 Hz ; 141.3
J
x
C– F s 7.1 Hz 163.2
Ž
w
.
Ž
.
Ž
.
J
C– F s246.8 Hz . 125Te NMR: d 551.2. MS
19
w
Ž13
.
13.8 J 13C–119Sn s296.6 Hz, J 13C–117Sn s283.0
Ž
.
Ž
.
q
Ž
.
Ž
.
Ž
.
13C–119Sn s63.7
mre 254 30%, M ; 225 29% ; 130 25% ; 95
x
Ž
Ž
.
w
Ž
J
.
Hz ; 27.3 cyclohexane ; 27.4
Ž
.
Hz ; 30.0 J 13C–119Sn s21.9 Hz ; 105.6 J C– F
Ž13
x
.
100% . HRMS: C8 H9 FTe requires 253.9758; found
253.9763.
19
x
w
.
x
w
19
x
w
Ž13
.
s 4.1 Hz ; 116.5
J
C– F s 15.8 Hz ; 144.0
wŽ
.
x
Ž .
9
Ž13
.
x
w
Ž13
C– F s7.6 Hz ; 164.5 J C– F s247.8 Hz .
.
x
2- 4-Fluorophenyl telluro butane was prepared
following Standard Protocol E in our previously pub-
lished procedure 40 using bis 4-fluorophenyl ditel-
luride and 2-iodobutane. Kugelrohr distillation
19
19
w
J
Sn NMR d 0.4 J 119Sn–125Te s2656 Hz . 125Te
Ž .
119
w
Ž
.
x
w
x
Ž
¨
.
NMR d y213.9 J 125Te–119Sn s2656 Hz, J 125Te–
w
Ž
.
Ž
117
.
x
Sn s2526 Hz .
1
Ž
.
Ž
.
508Cr0.08 mm afforded 9 as a yellow oil 70% . H
Ž
.
Ž
.
NMR CDCl3 : d 0.97 t, 3H, Js7.1 Hz ; 1.57–1.74
(
)
3.3. Phenyltellurotributylgermane 10
Ž
Ž
.
.
Ž
.
Ž
.
m, 5H ; 3.34 q, 1H, Js7.1 Hz ; 6.91 m, 2H ; 7.75
13
Ž
m, 2H . C NMR C6 D : d 13.9; 23.9; 26.3; 32.4;
w
Ž13
.
6 x
w
Ž13
.
19
19
The title compound was prepared according to the
general procedure using tributylgermanium hydride and
AIBN 15 mol% with heating at 808C for 4 h.
105.5 J C– F s3.6 Hz ; 116.4 J C– F s20.4
19
19
x.
w
Ž13
.
x
w
Ž13
.
Hz ; 142.6 J C– F s7.7 Hz ; 162.9 J C– F
125
Ž
.
x
Ž
s247.8 Hz . Te NMR: d 645.8. MS mre 282 3%,
1
Ž
.
Ž
Mq ; 225 16% ; 130 12% , 95 68% ; 91 100% .
HRMS: C10 H13FTe requires 280.0064; found 280.0060.
H NMR: d 0.80 t, 9H, Js7.5 Hz ; 0.98–1.04 m,
.
Ž
.
Ž
.
Ž
.
Ž
.
.
Ž
.
Ž
.
6H ; 1.18–1.35 m, 6H ; 1.41 s, cyclohexane ; 1.35–
Ž
.
Ž
.
Ž
1.40 m, 6H ; 6.78 t, 2H, Js7.5 Hz ; 6.94 t, 1H,
Js7.5 Hz ; 7.85 dd, 2H, Js7.2, 1.2 Hz . 13C NMR:
.
Ž
.
3.1. General procedure for the in situ preparation of
Ž
.
d 13.9; 16.6; 26.5; 27.3 cyclohexane ; 27.7; 114.0;
(
)
(
)
selenostannanes 1, 2 , selenogermanes 10, 11 and
126.9; 128.8; 142.1. 125Te NMR d y78.8.
(
)
selenosilanes 13, 14 . Preparation of phenyltel-
lurotributylstannane 1
( )
(
)
( )
3.4. 4-fluorophenyl tellurotributylgermane 11
Ž
.
Ž
Tributyltin hydride 94 ml, 0.35 mmol and AIBN 5
.
mol% was added to a solution of phenyltellurocyclo-
The title compound was prepared according to the
Ž
. Ž
.
Ž
Ž .
general procedure using 4-fluorophenyl tellurocyc-
hexane 3, ArsPh 100 mg, 0.35 mmol in C6 D6 0.5
.
Ž .
lohexane, AIBN 15 mol% and tributylgermanium hy-
1
ml in an NMR tube. The solution was heated at 808C
Ž
dride with heating at 808C for 4 h. H NMR: d 0.89 t,
for 2 h while shielded from background light. The
.
Ž
.
Ž
9H, Js7.5 Hz ; 1.03–1.09 m, 6H ; 1.20–1.35 m,
.
Ž
.
Ž
.
6H ; 1.41 s, cyclohexane ; 1.39–1.45 m, 6H ; 6.59
m, 2H ; 7.53 m, 2H . 13C NMR: d 13.3; 17.1; 26.3;
3 We thank Professor D. Dakternieks for providing a sample of
tributylgermanium chloride.
Ž
.
Ž
.
19
Ž
.
w
Ž13
.
27.3 cyclohexane ; 28.5; 96.2; 116.3
J
C– F s

(
)
156
C.H. Schiesser, M.A. SkidmorerJournal of Organometallic Chemistry 552 1998 145–157
19
20.6 Hz ; 143.8 J C– F s7.5Hz ; 163.5 J 13C–
w
w
w
w
x
Ž
.
13 D.J. Carlsson, K.U. Ingold, J. Am. Chem. Soc. 90 1968 1055.
x
w
Ž13
.
x
w Ž
F s248.3 Hz . 125Te NMR d y82.8.
19
x
Ž
.
14 D.J. Carlsson, K.U. Ingold, J. Am. Chem. Soc. 90 1968 7047.
.
x
x
Ž
.
15 M. Newcomb, Tetrahedron 49 1993 1151.
x
16 D.V. Avila, K.U. Ingold, J. Lusztyk, W.R. Dolbier Jr., H.-Q.
Ž
.
(
)
( )
Pan, M. Muir, J. Am. Chem. Soc. 116 1994 99.
3.5. Phenyltellurotris trimethylsilyl silane 13
w
w
x
17 S.J. Garden, D.V. Avila, A.L.J. Beckwith, V.W. Bowry, K.U.
Ž
.
Ingold, J. Lusztyk, J. Org. Chem. 61 1996 805.
The title compound was prepared according to the
general procedure using tris trimethylsilyl silane.
x
18 L.J. Johnston, J. Lusztyk, D.D.M. Wayner, A.N. Abeywick-
Ž
.
rema, A.L.J. Beckwith, J.C. Scaiano, K.U. Ingold, J. Am.
1
Ž
.
Ž
.
Ž
.
Chem. Soc. 107 1985 4594.
H NMR: d 0.26 s, 27H ; 1.41 cyclohexane ; 6.81
w
w
w
x
19 C. Chatgilialoglu, K.U. Ingold, J.C. Scaiano, J. Am. Chem. Soc.
Ž
.
Ž
.
Ž
t, 2H, Js7.5 Hz ; 7.00 t, 1H, Js7.5 Hz ; 7.82 dd,
13
Ž .
103 1981 7739.
29
.
w
Ž13
.
2H, Js6.7, 1.5 Hz . C NMR: d 1.54 J C– Si
x
20 O.M. Musa, J.H. Horner, H. Shahin, M. Newcomb, J. Am.
x
Ž
.
s53.6 Hz ; 27.3 cyclohexane ; 104.4; 127.7; 129.0;
142.3. 29Si NMR d y94.4; 10.6. 125Te NMR d
y192.9.
Ž
.
Chem. Soc. 118 1996 3862.
x
Ž
.
21 J. Lusztyk, B. Maillard, K.U. Ingold, J. Org. Chem. 51 1986
2457.
w
w
w
x
Ž
.
22 M. Newcomb, S.U. Park, J. Am. Chem. Soc. 108 1986 4132.
23 C. Chatgilialoglu, Acc. Chem. Res. 25 1992 188.
24 C. Chatgilialoglu, A. Guerrini, M. Lucarini, J. Org. Chem. 57
x
Ž
.
(
)
(
)
( )
3.6. 4-fluorophenyl tellurotris trimethylsilyl silane 14
x
Ž
.
1992 3405.
w
w
w
w
w
w
x
25 C. Chatgilialoglu, C. Ferreri, M. Lucarini, J. Org. Chem. 58
The title compound was prepared according to the
general procedure using 4-fluorophenyl tellurocyc-
lohexane and tris trimethylsilyl silane.
Ž
.
1993 249.
Ž
.
x
26 C. Chatgilialoglu, D. Griller, M. Lesage, J. Org. Chem. 53
Ž
.
Ž
.
1988 3641.
1
Ž
.
.
Ž
.
29
H NMR: d 0.19 s, 27H ; 1.41 cyclohexane ; 6.87
x
27 M. Ballestri, C. Chatgilialoglu, K.B. Clark, D. Griller, B. Giese,
13
Ž13
m, 2H ; 7.73 m, 2H . C NMR: d 2.01 J C– Si
.
B. Kopping, J. Org. Chem. 56 1991 678.
Ž .
Ž
.
Ž
w
19
Ž13
.
x
28 J. Lusztyk, B. Maillard, D.A. Lindsay, K.U. Ingold, J. Am.
x
Ž
.
w
s47.1 Hz ; 27.3 cyclohexane ; 105.6 J C– F s
Ž
.
Chem. Soc. 105 1983 3578.
19
x
w
Ž13
4.1 Hz ; 116.6 J C– F s20.4 Hz ; 143.0 J 13C–
.
x
w Ž
x
29 J. Lusztyk, B. Maillard, S. Deycard, D.A. Lindsay, K.U. Ingold,
C– F s247.9 Hz . 29Si
Ž13
. x
19
19
.
x
w
F s7.6 Hz ; 165.0
J
Ž
.
J. Org. Chem. 52 1987 3509.
NMR d y92.6 J 29 Si–125Te s135 Hz ; 10.6. 125Te
NMR d y195.9.
x
Ž
.
w
Ž
.
x
30 D.H.R. Barton, Tetrahedron 48 1992 2529, references cited
therein.
w
w
x
x
Ž
.
31 M.J. Robins, J.S. Wilson, J. Am. Chem. Soc. 103 1981 932.
32 M.J. Robins, J.S. Wilson, F. Hansske, J. Am. Chem. Soc. 105
Ž
.
1981 4059.
w
x
33 D.H.R. Barton, D. Crich, W.B. Motherwell, Tetrahedron 41
Acknowledgements
Ž
.
1985 3901.
w
w
w
x
Ž
Ž
.
.
34 J.L. Esker, M. Newcomb, J. Org. Chem. 58 1993 4933.
35 J.L. Esker, M. Newcomb, J. Org. Chem. 59 1994 2779.
36 C.H. Schiesser, B.A. Smart, T.-A. Tran, Tetrahedron 51 1995
We thank the Australian Research Council for finan-
cial support and the award of an Australian Postgradu-
ate Award to MAS. The support of the Ormond Super-
computer Facility, a joint venture between the Univer-
sity of Melbourne and the Royal Melbourne Institute of
Technology is gratefully acknowledged.
x
x
Ž
.
.
3327.
w
w
x
x
Ž
.
37 C.H. Schiesser, B.A. Smart, Tetrahedron 51 1995 6051.
38 C.H. Schiesser, B.A. Smart, T.-A. Tran, Tetrahedron 51 1995
Ž
10651.
w
x
39 H. Duddeck, P. Wagner, A. Biallass, Mag. Res. Chem. 29
Ž
.
1991 248.
w
w
x
x
Ž
.
40 M.A. Lucas, C.H. Schiesser, J. Org. Chem. 61 1996 5754.
41 D.H. O’Brien, N. Dereu, C.-K. Huang, K.J. Irgolic, F.F. Knapp
References
Ž
.
Jr., Organometallics 2 1983 305.
w
w
x
42 M. Baiwir, G. Llabres, L. Christiaens, A. Luxen, J. Piette, Mag.
w x
1
B. Giese, Radicals in Organic Synthesis: Formation of Carbon–
Carbon Bonds, Pergamon, Oxford 1986 .
Ž
.
Res. Chem. 24 1986 362.
Ž
.
x
43 N.P. Luthra, J.D. Odom, The Chemistry of Organic Selenium
w x
Ž
.
2
W.P. Neumann, Synthesis 1987 p. 665.
Ž
.
and Tellurium Compounds, S. Patai, Z., Rappoport, Eds. , Vol.
w x
Ž
Ž
.
.
3
D.P. Curran, Synthesis 1988 p. 417.
D.P. Curran, Synthesis 1988 p. 489.
C.P. Jasperse, D.P. Curran, T.L. Fevig, Chem. Rev. 91 1991
Ž
.
1, Chap. 6, Wiley, New York 1986 .
w x
4
w
w
w
x
Ž
.
44 W.S. Haller, K.J. Irgolic, J. Organomet. Chem. 38 1972 97.
45 D. Crich, S. Sun, J. Org. Chem. 61 1996 7200.
46 C. Chatgilialoglu, D. Griller, M. Lesage, J. Org. Chem. 54
w x
Ž
.
5
x
Ž
.
1237.
x
w x
6
U. Gerigk, M. Gerlach, W.P. Neumann, R. Vieler, V. Weintritt,
Ž
.
1989 2492.
Ž
.
Synthesis 1990 p. 448.
w
w
w
x
47 P.J. Bonasia, D.E. Gindelberger, B.O. Dabbousi, J. Arnold, J.
w x
Ž
.
7
J. Light, R. Breslow, Tetrahedron Lett. 31 1990 2957.
F. Ferkous, D. Messadi, B. De Jeso, M. Degueil-Castaing, B.
Maillard, J. Organometal. Chem. 420 1991 315.
W.P. Neumann, M. Peterseim, React. Polym. 20 1993 189.
Ž
.
Am. Chem. Soc. 114 1992 5209.
w x
8
x
Ž
.
48 D.E. Gindelberger, J. Arnold, J. Am. Chem. Soc. 114 1992
Ž
.
6242.
w x
Ž
.
.
9
x
x
49 P.J. Bonasia, V. Christou, J. Arnold, J. Am. Chem. Soc. 115
w
w
w
Ž
10 D.P. Curran, S. Hadida, J. Am. Chem. Soc. 118 1996 2531.
11 D. Nanni, D.P. Curran, Tetrahedron Asym. 7 1996 2417.
12 C.H. Schiesser, L.M. Wild, Tetrahedron 52 1996 13265, refer-
Ž
.
1993 6777.
x
x
Ž
.
w
w
x
x
Ž
.
Ž
50 A.L. Seligson, J. Arnold, J. Am. Chem. Soc. 115 1993 8214.
51 C. Tamborski, F.E. Ford, E.J. Soloski, J. Org. Chem. 28 1963
Ž
.
.
ences cited therein.
237.

(
)
C.H. Schiesser, M.A. SkidmorerJournal of Organometallic Chemistry 552 1998 145–157
157
w
w
x
52 M.J. Frisch, G.W. Trucks, M. Head-Gordon, P.M.W. Gill,
V. Jonas, K.F. Kohler, R. Stegmann, A. Veldkamp, G. Frenk-
¨
Ž .
ing, Chem. Phys. Lett. 208 1993 237.
59 B.A. Smart, C.H. Schiesser, J. Comput. Chem. 16 1995 1055.
M.W. Wong, J.B. Foresman, B.G. Johnson, H.B. Schlegel,
M.A. Robb, E.S. Replogle, R. Gomperts, J.L. Andres, K.
Raghavachari, J.S. Binkley, C. Gonzales, R.L. Martin, D.J. Fox,
D.J. Defrees, J. Baker, J.J.P. Stewart, J.A. Pople, Gaussian 92,
w
w
x
Ž
.
x
60 T.H. Dunning, P.J. Hay, Modern Theoretical Chemistry, Chap.
Ž
.
1. Plenum, New York 1976 pp. 1–28.
Ž
.
w
w
w
w
w
x
Revision F, Gaussian, Pittsburgh, PA 1992 .
61 C.H. Schiesser, M.L. Styles, L.M. Wild, J. Chem. Soc. Perkin
x
Ž .
Trans. 2 1996 2257.
62 J.E. Lyons, C.H. Schiesser, J. Organomet. Chem. 437 1992
53 M.J. Frisch, G.W. Trucks, H.B. Schlegel, P.M.W. Gill, B.G.
Johnson, M.A. Robb, J.R. Cheeseman, T. Keith, G.A. Petersson,
J.A. Montgomery, K. Raghavachari, M.A. Al-Laham, V.G.
Zakrzewski, J.V. Ortiz, J.B. Foresman, C.Y. Peng, P.Y. Ayala,
W. Chen, M.W. Wong, J.L. Andres, E.S. Replogle, R. Gom-
perts, R.L. Martin, D.J. Fox, J.S. Binkley, D.J. Defrees, J.
Baker, J.J.P. Stewart, M. Head-Gordon, C. Gonzalez, J.A. Pople,
x
Ž
.
165.
x
63 D. Dakternieks, D.J. Henry, C.H. Schiesser, J. Chem. Soc.
Ž
.
Perkin Trans. 2 1997 1665.
x
64 W. Adcock, C.I. Clark, C.H. Schiesser, J. Am. Chem. Soc. 118
Ž
.
1996 11541.
Ž
.
x
65 Landolt-Bornstein, Numerical Data and Functional Relation-
¨
Gaussian 94, Revision B.3, Gaussian, Pittsburgh, PA 1995 .
w
x
Ž .
ships in Science and Technology, IIr18, H. Fischer Ed. ,
54 W.J. Hehre, L. Radom, P.v.R. Schleyer, J.A. Pople, Ab Initio
Ž
.
Ž .
Springer-Verlag, Berlin 1994 .
Molecular Orbital Theory, Wiley, New York 1986 .
w
w
w
w
x
Ž
.
.
w
w
x Ž .
66 H.G. Kuivila, Synthesis, 1970 p. 499.
55 W.R. Wadt, P.J. Hay, J. Chem. Phys. 82 1985 284.
56 P.J. Hay, W.R. Wadt, J. Chem. Phys. 82 1985 270.
57 P.J. Hay, W.R. Wadt, J. Chem. Phys. 82 1985 299.
58 A. Hollwarth, M. Bohme, S. Dapprich, A.W. Ehlers, A. Gobbi,
x
Ž
Ž
x
67 K. Hayashi, J. Iyoda, I. Shiihara, J. Organomet. Chem. 10
x
.
Ž
.
1967 81.
x
w
x Ž .
68 K. Surowiec, T. Fuchigami, J. Org. Chem. 57 1992 5781.
¨
¨