On the Triple Role of Fluoride Ions in Palladium-Catalyzed Stille Reactions

Article abstract of DOI:10.1002/chem.201503309

The mechanism of Stille reactions (cross-coupling of ArX with Ar′SnnBu₃) performed in the presence of fluoride ions is established. A triple role for fluoride ions is identified from kinetic data on the rate of the reactions of trans-[ArPdBr(PPh₃)₂] (Ar=Ph, p-(CN)C₆H₄) with Ar′SnBu₃ (Ar′=2-thiophenyl) in the presence of fluoride ions. Fluoride ions promote the rate-determining transmetallation by formation of trans-[ArPdF(PPh₃)₂], which reacts with Ar′SnBu₃ (Ar′=Ph, 2-thiophenyl) at room temperature, in contrast to trans-[ArPdBr(PPh₃)₂], which is unreactive. However, the concentration ratio [F^-]/[Ar′SnBu₃] must not be too high, because of the formation of unreactive anionic stannate [Ar′Sn(F)Bu₃]^-. This rationalises the two kinetically antagonistic roles exerted by the fluoride ions that are observed experimentally, and is found to be in agreement with the kinetic law. In addition, fluoride ions promote reductive elimination from trans-[ArPdAr′(PPh₃)₂] generated in the transmetallation step.

Full text of DOI:10.1002/chem.201503309

DOI: 10.1002/chem.201503309
Full Paper
&
CÀC Coupling |Hot Paper|
On the Triple Role of Fluoride Ions in Palladium-Catalyzed Stille
Reactions
Marius HervØ,[a] Guillaume Lef›vre,[b] Emily A. Mitchell,[c] Bert U. W. Maes,*[c] and
Anny Jutand*[a]
Abstract: The mechanism of Stille reactions (cross-coupling
of ArX with Ar’SnnBu3) performed in the presence of fluoride
ions is established. A triple role for fluoride ions is identified
from kinetic data on the rate of the reactions of trans-
[ArPdBr(PPh3)2] (Ar=Ph, p-(CN)C6H4) with Ar’SnBu3 (Ar’=2-
thiophenyl) in the presence of fluoride ions. Fluoride ions
promote the rate-determining transmetallation by formation
of trans-[ArPdF(PPh3)2], which reacts with Ar’SnBu3 (Ar’=Ph,
2-thiophenyl) at room temperature, in contrast to trans-
[ArPdBr(PPh3)2], which is unreactive. However, the concentra-
tion ratio [FÀ]/[Ar’SnBu3] must not be too high, because of
the formation of unreactive anionic stannate [Ar’Sn(F)Bu3]À.
This rationalises the two kinetically antagonistic roles exert-
ed by the fluoride ions that are observed experimentally,
and is found to be in agreement with the kinetic law. In ad-
dition, fluoride ions promote reductive elimination from
trans-[ArPdAr’(PPh3)2] generated in the transmetallation step.
Introduction
R’SnR3, and iii) reductive elimination of ArR’ from [ArPdR’Ln]
(n=1 or 2).[1] Stille reactions are known to be more efficient in
the presence of additives such as copper,[3] chloride[4] and fluo-
ride anions,[5] or copper with fluorides,[6] owing to a synergetic
effect. The beneficial role of CuI in Stille reactions was rational-
ised as a complexation of one ligand L of [ArPdXL2] to CuI.[3c]
The ensuing unsaturated [ArPdXL] complex would be thus
more reactive in the transmetallation step with CH2=CHÀ
The Stille reaction is a cross-coupling reaction between aryl
halides and organostannane derivatives [Eq. (1)].[1] The popu-
larity of this reaction is based on the mild reaction conditions
required to create carbon–carbon bonds. Its high functional
group compatibility makes it a suitable tool to couple highly
functionalised subunits in the synthesis of complex natural
products.[1] In addition, organostannanes can be easily synthes-
ised in a variety of ways and stored without special precau-
tions.[2]
SnBu3, allowing for coordination of a h2-C=C bond to Pd.[3c,1e]
A
transfer of the R’ group of R’SnBu3 to CuI to form a more reac-
tive {R’Cu} species in the transmetallation was also disclo-
sed.[3c,1f] We reported a dually beneficial role of chloride ions in
the Stille reaction when using the precatalyst [PdCl2(PPh3)2],[4]
namely i) in situ halide metathesis in [ArPdX(PPh3)2] (X=I, Br)
leading to [ArPdCl(PPh3)2], which is more reactive in the trans-
metallation step with organostannane derivatives, and ii) stabi-
lisation of [Pd0(PPh3)2] (formed by reduction of [PdCl2(PPh3)2]
by R’SnBu3) with formation of stable anionic [Pd0Cl(PPh3)2]À,
which prevents the Pd0 decomposition that occurs in the ab-
sence of ClÀ. A high catalyst loading is maintained in the pres-
ence of ClÀ and the catalytic reactions become faster.[4] The ac-
celerating effect of fluoride ions in Stille reactions[5] is proposed
to be due to the formation of anionic [R’SnFR3]À, which is as-
sumed to be more reactive in the transmetallation step with
[ArPdXLn] than the neutral R’SnR3.[5b,e,f] However, a triple role of
fluoride ions in the Suzuki–Miyaura[7] and Hiyama reactions[8]
was recently discovered. It is established that the complexes
that react with the nucleophiles [Ar’B(OH)2 or Ar’Si(OMe)3] are
not trans-[ArPdXL2] (L=PPh3; unreactive in the transmetalla-
tion step) but trans-[ArPdFL2] generated by reaction of FÀ with
trans-[ArPdXL2].[7,8] However, fluoride ions exert a decelerating
The postulated mechanism involves at least three steps:
i) oxidative addition of a [Pd0Ln] complex with ArX to generate
[ArPdXLn] (n=1 or 2), ii) transmetallation of [ArPdXLn] by
[a] M. HervØ, Dr. A. Jutand
Ecole Normale SupØrieure-PSL Research University, DØpartement de Chimie,
Sorbonne UniversitØs-UPMC Univ Paris 06, CNRS UMR 8640 PASTEUR
24 Rue Lhomond, 75231 Paris Cedex 5 (France)
E-mail: Anny.Jutand@ens.fr
[b] Dr. G. Lef›vre
CEA-Saclay, DSM-IRAMIS-NIMBE-LCMCE, UMR 3685
91191 Gif Sur Yvette, Cedex (France)
[c] Dr. E. A. Mitchell, Prof. B. U. W. Maes
University of Antwerp
Groenenborgerlaan 171, 2020 Antwerp (Belgium)
E-mail: bert.maes@uantwerpen.be
role at high concentrations by formation of unreactive anionic
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201503309.
À
[Ar’BFn(OH)3Àn
]
(n=1–3)[7] and [Ar’SiF(OMe)3]À[8] respectively,
Chem. Eur. J. 2015, 21, 18401 – 18406
18401
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
leading to two kinetically antagonistic roles of fluoride ions in
the transmetallation, the rate of which is controlled by the
concentration ratio [FÀ]/[Ar’B(OH)2] or [FÀ]/[Ar’Si(OMe)3], re-
spectively.[7,8] In addition, in both cross-coupling reactions, fluo-
ride ions catalyze the reductive elimination in the intermediate
trans-[ArPdAr’L2] to form ArAr’ and [Pd0L2].[7,8] We report herein
that fluoride ions fulfil similar roles in the Stille reaction, which
is in contrast with the current mechanistic view.
see the Supporting Information, Figure S1a). But [Pd0(PPh3)3]
was formed in 97% yield together with 3H (96% yield) upon
addition of FÀ (a=2.5 equiv vs. 1H) to a solution of 1H (2 mm),
2-ThSnBu3 (b=7.5 equiv vs. 1H) and PPh3 (4 mm; see the Sup-
porting Information, Figure S1b,c). These observations confirm
that the fluoride ions considerably accelerate the reaction in
[Eq (2)], a key step in Stille reactions, at room temperature.
The kinetics of the reaction presented in [Eq. (2)] were moni-
tored through the rate of formation of [Pd0(PPh3)3] (4) by chro-
noamperometry at a rotating disk electrode (RDE) polarised at
+0.10 V vs. SCE, on the oxidation wave of 4.[9a] The growth of
the oxidation current of 4 (proportional to its concentration at
time t) was recorded with time after addition of nBu4NF (a=6
vs. 1H) to a solution of 1H (2 mm), 2 (b=7.5 vs. 1H) and PPh3
(4 mm) in DMF at 258C (Figure 2a). The plot of lnx [x=(ilimÀit)/
ilim; ilim is the final oxidation current of 4 and it is oxidation cur-
rent of 4 at time t] versus time was linear (Figure 2b), attesting
to a first-order reaction for the
Results and Discussion
The model reactions starting from isolated trans-
[ArPdBr(PPh3)2] (1) and 2-thienyl-SnnBu3 (2-ThSnBu3, 2; b equiv
vs. 1) in the presence of nBu4NF (a equiv vs. 1) were per-
formed at room temperature in DMF [Eq. (2)]. Excess PPh3
(2 equiv vs. 1) was used to stabilise the palladium(0) formed in
[Eq. (2)] as [Pd0(PPh3)3] (4).
palladium complex: lnx=Àkobs
”
t. The observed rate constant
kobs, which characterised the rate
of formation of [Pd0(PPh3)3] (the
same rate as the formation of
3H) in [Eq (2)], was calculated
from the slope of the straight
line (Figure 2b): kobs =0.147 sÀ1
(DMF, 258C, a=6, b=7.5).
The reactions were monitored by cyclic voltammetry (CV)
A series of similar kinetic studies were performed with vary-
ing amounts of FÀ and 2-ThSnBu3 (2) relative to the PdII com-
plex 1H (a and b; see the Supporting Information, Figures S2–
S10). The plots of kobs versus a at constant b exhibited a maxi-
mum (Figure 3a). This indicates that the fluoride ions are in-
performed versus time.[9] Indeed, complex 4 was characterised
by its oxidation peak at Epox = +0.05 V vs. SCE.[10] The com-
[11]
plexes 1H and 1CN
were characterised by their reduction
peaks at Epred =À1.83 and À1.77 V, respectively, and the cross-
coupling products 3H and 3CN by their reduction peaks at Erped
=
À2.40 and À1.85 V, respectively, similar to those of the corre-
sponding authentic samples.[12] Interestingly, since the reduc-
tion or oxidation currents are always proportional to the con-
centration of the electroactive species, their evolution with
time can be easily observed.[9a]
It was first observed that trans-[{p-(CN)C6H4}PdBr(PPh3)2] (1CN
;
C0 =2 mm in DMF) did not react with 2-ThSnBu3 (2) at room
temperature, even in the presence of a large excess of 2 (b>
30 equiv vs. 1CN) and after a long reaction time (>3 h; reaction
performed the presence of 4 mm PPh3). Indeed, the reduction
peak current of 1CN (proportional to its concentration at any
time)[9a] did not decrease with time (Figure 1a). After addition
of nBu4NF (a=6 equiv vs. 1CN) to a solution of 1CN (2 mm), 2
(b=7.5 equiv vs. 1CN) and PPh3 (4 mm), the solution turned
yellow with concomitant formation of 4 (yellow complex) and
the cross-coupling product 3CN, as detected by their oxidation
(Figure 1b) and reduction peak (Figure 1c), respectively. The
yields of 4 (99%) and 3CN (93%) were determined by consider-
ing the growth of the oxidation peak current of 4 and the re-
duction peak current of 3CN after addition of a known amount
of authentic samples of 4 and 3CN, respectively.
Figure 1. Cyclic voltammetry performed at a gold disk electrode (d=1 mm)
in DMF containing nBu4NBF4 (0.3m) as supporting electrolyte with a scan
rate of 0.5 VsÀ1 at 258C: a) Reduction of trans-[{p-(CN)C6H4}PdBr(PPh3)2] (1CN
;
2 mm) in the presence of PPh3 (4 mm). The same voltammogram is observed
in the presence of 2-ThSnBu3 (15 equiv vs. 1CN); b) oxidation of [Pd0(PPh3)3]
(4) formed upon addition of nBu4NF (12 mm, from a 1m mother solution in
THF) to a solution of 1CN (2 mm), 2-ThSnBu3 (15 mm) and PPh3 (4 mm) after
100 s; c) reduction of 3CN formed together with 4 under the experimental
conditions of Figure 1b.
In a similar way, no reaction of trans-[PhPdBr(PPh3)2] (1H;
2 mm in DMF) with 2-ThSnBu3 (b=10–40 equiv vs. 1H) was ob-
served at room temperature, in the presence of PPh3 (4 mm;
Chem. Eur. J. 2015, 21, 18401 – 18406
www.chemeurj.org
18402
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
The dependence of kobs versus the concentration of the fluo-
ride ions and 2-ThSnBu3 indicates that [Pd0(PPh3)3] is formed at
the same rate as 1H disappears. In other words, the reductive
elimination from [PhPd(2-Th)(PPh3)2] that generates [Pd0(PPh3)3]
(the reaction order for which in 2-ThSnBu3 and FÀ should be
zero) is faster than the transmetallation that generates
[PhPd(2-Th)(PPh3)2]. Consequently, kobs characterises the rate of
the rate-determining transmetallation.
It has been established that fluoride ions reversibly ex-
change the halide of trans-[ArPdX(PPh3)2] (1) to generate well-
characterised trans-[ArPdF(PPh3)2] [5; Eq. (4) in Scheme 1].[7a]
The complex trans-[{p-(CN)C6H4}PdF(PPh3)2] (5CN) has been syn-
thesised[7a] and submitted to reaction with 2-ThSnBu3 (2) in the
presence of PPh3 (2 equiv vs. 5CN). The reaction was monitored
Figure 2. Kinetics of the formation of [Pd0(PPh3)3] (4) in the reaction of trans-
[PhPdBr(PPh3)2] (1H; 2 mm) with 2-ThSnBu3 (7.5 equiv vs. 1H, 15 mm) in the
presence of FÀ (6 equiv vs. 1H, 12 mm) and PPh3 (2 equiv vs. 1H, 4 mm) in
DMF at 258C: a) Evolution of the oxidation current i of [Pd0(PPh3)3] (propor-
tional to its concentration) versus time; i was determined by chronoamper-
ometry performed at a rotating gold disk electrode (d=2 mm) polarised at
+0.1 V vs. SCE; b) plot of lnx versus time: x=(ilimÀit)/ilim ; ilim =final oxidation
current of [Pd0(PPh3)3]; it =oxidation current of [Pd0(PPh3)3] at time t, deter-
mined in Figure 2a.
by cyclic voltammetry. The reduction peak of 5CN (2 mm, Erped
=
À1.92 V; see the Supporting Information, Figure S13a) rapidly
disappeared after addition of 2 (b=7.5 equiv vs. 5CN) at 258C.
[Pd0(PPh3)3] (4) was formed in 45% yield after 3 min. The ex-
pected
intermediate
complex
trans-[{p-(CN)C6H4}Pd(2-
Th)(PPh3)2] (6CN,Th) was detected at Epred =À1.37 V (see the Sup-
porting Information, Figure S13b)[14] but with a low reduction
peak because of the formation of 4 by reductive elimination
(see the Supporting Information, Figure S13b). The products
3CN and 4 were generated in quantitative yields after 10 min
(see the Supporting Information, Figure S13c,d). Since no reac-
tion of trans-[{p-(CN)C6H4}PdBrL2] (1CN) with 2-ThSnBu3 was ob-
served at the same concentrations (see above), it is thus estab-
lished that the transmetallation of trans-p-(CN)C6H4-PdFL2 by 2-
ThSnBu3 [Eq. (5) in Scheme 1] was considerably faster than that
of 1CN at identical concentrations. The reaction of trans-
[PhPdF(PPh3)2] (5H)[15] (2 mm, Erped =À2.1 V) with 2-ThSnBu3 (b=
7.5 equiv vs. 5H) also took place within less than 3 min (as at-
tested by the formation of 4), confirming the following reactiv-
ity order with 2-ThSnBu3, irrespective of the Ar group:
Figure 3. Kinetics of the formation of [Pd0(PPh3)3] (4) in the reaction of trans-
[PhPdBr(PPh3)2] 1H (2 mm) with 2-ThSnBu3 (b equiv vs. 1H) and FÀ (a equiv
vs. 1H) in the presence of PPh3 (4 mm) in DMF at 258C: a) Plot of the
pseudo-first-order observed rate constant kobs versus a for b=7.5 and 10;
b) plot of kobs vs. b for a=6.
volved in two antagonistic kinetic effects. They are beneficial
for the reaction at low concentration ([FÀ]/[2-ThSnBu3]<1) but
they exert an inhibiting effect when their concentra-
tion is too high ([FÀ]/[2-ThSnBu3]ꢀ1; Figure 3a). A
decay of kobs versus b at constant a was also ob-
served (Figure 3b). This indicates that the reaction
became slower when the concentration of 2-ThSnBu3
exceeded that of the fluoride ions. In other words,
when [2-ThSnBu3]/[FÀ]>1, the fluoride ions are
quenched by 2-ThSnBu3 to generate an unreactive
species: anionic [2-ThSn(F)Bu3]À [Eq. (3) in Scheme 1].
The fluorophilicity of the tin centre is well
known.[13a–c] The formation of [2-ThSn(F)Bu3]À by reac-
tion of nBu4NF with 2-ThSnBu3 in DMF was proven by
19F NMR spectroscopy (d=À139.1 ppm; see the Sup-
porting Information, Figure S11b).[13d] Interestingly,
the concentration of [2-ThSn(F)Bu3]À increased when
the initial ratio [FÀ]0/[2-ThSnBu3]0 was increased (see
the Supporting Information, Table S1 and Figure S12).
These experiments support the reversible formation
of [2-ThSn(F)Bu3]À in DMF [Eq. (3) in Scheme 1], the
trans-½ArPdFðPPh3Þ2Š ꢁ trans-½ArPdBrðPPh3Þ2Š
Scheme 1. Mechanism of the transmetallation and reductive elimination in Stille reac-
lack of reactivity of which is established in Figure 3.
tions performed in the presence of fluoride ions.
Chem. Eur. J. 2015, 21, 18401 – 18406
www.chemeurj.org
18403
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
The formation of [Pd0(PPh3)3] when 5CN or 5H reacts with 2 libria: i) one equilibrium generates trans-[ArPdFL2] [Eq. (4)],
(see above) suggests that the reductive elimination from the which reacts with Ar’SnBu3 as a consequence of the fluorophi-
intermediate complexes trans-[{p-(CN)C6H4}Pd(2-Th)(PPh3)2] licity of Sn [Eq. (5)]; ii) the second equilibrium generates the
(6CN,Th) or trans-[PhPd(2-Th)(PPh3)2] (6H,Th) formed in the trans- unreactive [Ar’Sn(F)Bu3]À [Eq. (3)]. This mechanism is in agree-
metallation took place slowly in the absence of fluoride ions, ment with the kinetic law [Eq. (8)] for which the theoretical var-
in contrast to what was observed in previous works for related iation of kobs versus the fluoride concentration exhibits a maxi-
complexes trans-[ArPdAr’(PPh3)2] (Ar=Ar’=Ph; Ar=Ph, Ar’=p- mum.[18]
(CN)C6H4)[14] for which fluoride ions were required to promote
the reductive elimination.[7] This is due to the electron-donat-
ing property of the 2-Th group in 6CN,Th and 6H,Th, which fa-
vours the reductive elimination in those trans complexes.
Indeed, it was previously established that reductive elimination
in trans-[ArPdAr’(PPh3)2] proceeded slowly in the absence of
In addition, fluoride ions catalyse the reductive elimination
fluoride ions when Ar or Ar’ was substituted by an electron-do- from trans-[ArPdAr’(PPh3)2] complexes [Eq. (6)]. This leads to
nating group, such as OMe[14] (interestingly, the reductive elimi- the catalytic cycle for the Stille reaction depicted in Scheme 2,
nation in those cases was even faster in the presence of fluo- which reveals the three different roles exerted by the fluoride
ride ions).[7]
ions.
The reaction of trans-[{p-(CN)C6H4}PdF(PPh3)2] (5CN; 2 mm)
with PhSnBu3 (36.2 equiv vs. 5CN) was subsequently tested in
the presence of PPh3 (4 mm) at 258C. The reaction was sub-
stantially slower, indicating that PhSnBu3 is less reactive than
2-ThSnBu3.
The
intermediate
complex
trans-[{p-
(CN)C6H4}PdPh(PPh3)2] (6CN,Ph) was formed and detected by its
reduction peak at À1.50 V.[16] The kinetics of formation of
[Pd0(PPh3)3] from 6CN,Ph [Eq. (6) in Scheme 1] was monitored by
cyclic voltammetry (see the Supporting Information, Fig-
ure S14b,c). The reductive elimination from 6CN,Ph was very
slow (Figure 4) but was considerably accelerated upon addition
of fluoride ions (6 equiv vs. 5CN; Figure 4 and Figure S14d in
the Supporting Information). This is a further confirmation of
what was previously established for the Suzuki[7a] and Hiyama[8]
reactions: fluoride ions promote the reductive elimination
from trans-[ArPdAr’(PPh3)2] [Eq. (6) in Scheme 1] by formation
of anionic pentacoordinated palladium(II).[17] This bypasses the
classical reductive elimination from the cis complex, which is
slower due to the endergonic trans/cis isomerisation [Eq. (7) in
Scheme 1].
Scheme 2. Three roles for fluoride ions in the Stille reaction.
The mechanism of the transmetallation and reductive elimi-
nation based on experimental data is shown in Scheme 1. The
two antagonistic kinetic roles of the fluoride ions evidenced in
The intrinsic lack of reactivity of [2-ThSn(F)Bu3]À with trans-
Figure 3 are due to their involvement in two competitive equi- [ArPdF(PPh3)2], established from kinetic data, can be explained
by the fact that the saturated Sn centre in [2-ThSn(F)Bu3]À is
no longer fluorophilic and so cannot react with trans-
[ArPdF(PPh3)2] as 2-ThSnBu3 does [Eq. (5)]. The intrinsic lack of
reactivity of [2-ThSn(F)Bu3]À with trans-[ArPdBr(PPh3)2] estab-
lished from kinetic data, can be easily rationalised by consider-
ing that the saturated Sn center in [2-ThSn(F)Bu3]À cannot be
bromophilic. 2-ThSnBu3 itself has no affinity for bromide ions
contrary to fluoride ions, which is why 2-ThSnBu3 did not react
with trans-[ArPdBr(PPh3)2] at room temperature (see above).
DFT calculations were performed to support the lack of reac-
tivity of the anionic [2-ThSn(F)nBu3]À from an electronic point
of view (see the Supporting Information). Computational work
was carried out by using Gaussian 09 (version B.01).[19a] Com-
Figure 4. Kinetics of the formation of [Pd0(PPh3)3] (4) in the reaction of trans-
puted electron densities were analyzed by using the natural
[{p-(CN)C6H4}PdF(PPh3)2] (5CN; 2 mm) with PhSnBu3 (36.2 equiv vs. 5CN) until
bond orbital (NBO) partition implemented in Gaussian 09.[19b]
t=90 min ( ) and then after addition of FÀ (6 equiv vs. 5CN
;
&
*
). Reaction per-
formed in the presence of PPh3 (4 mm) in DMF at 258C.
The 6-311+G(d,p) basis set was used for all atoms (C, H, O, Si,
Chem. Eur. J. 2015, 21, 18401 – 18406
www.chemeurj.org
18404 ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
F, S)[19c] except Sn, which was treated by using the SDD basis
set and associated pseudopotential.[19d] To take into account
the London dispersion forces that can have a strong contribu-
tion to the stability of group XIV hypervalent species, the B97D
functional developed by Grimme was used.[19e] This long-range
dispersion-correction functional already demonstrated its effi-
ciency in the investigation of the reactivity of silicon reagents.
Solvent effects (DMF in this case) were taken into account by
using the PCM model,[19f,g] also implemented in Gaussian 09.
All structures were computed without geometrical constraints
and were characterised as local minima (no imaginary vibration
frequency).
The calculated charge repartition in [2-ThSn(F)nBu3]À re-
vealed that the highest negative charge is not located on the
2-Th group but on the fluoride atom: q(nBu)=À0.502 e,
À0.497 e, À0.481 e; q(F)=À0.865 e; q(2-Th)=À0.702 e;
q(Sn)= +2.047 e. This suggests that the more nucleophilic
centre is the coordinated F and not 2-Th, thus explaining why
[2-ThSn(F)nBu3]À cannot react with trans-[ArPdX(PPh3)2] (X=Br,
F) by direct transfer of the 2-Th group.[20a] Moreover, it is inter-
esting to note that the anionic adduct [2-ThSn(F)nBu3]À incor-
porates a 3-center/4-electron bond (w-bond) between the fluo-
ride, the Sn center and the metallated C2 of the thienyl group
(Figure 5). The formation of this bond can be confirmed by the
presence of two occupied molecular orbitals where the elec-
tron density arising from combination of axially symmetric p
orbitals (2p for C and F, and 5p for Sn) is located on F, Sn, and
C2 (Y1) and onto F and C2 with no contribution on Sn (Y2,
which only exhibits a bonding contribution between Sn and
the nBu groups; Figure 5).
with the antibonding 2-ThÀSn orbital in [2-ThSn(F)nBu3]À is es-
timated to lead to a stabilisation of approximately 24.2 kcal
molÀ1
.
A similar calculation for [PhSi(F)(OMe)3]À also revealed that
the highest negative charge is located on the F atom and the
lowest on the Ph group.[20b] This helps to explain why
[PhSi(F)(OMe)3]À was also found to be unreactive with trans-
[ArPdX(PPh3)2] (X=Br, F).[8] [PhSi(F)(OMe)3]À also incorporates
a 3-center/4-electron bond, which confirms its lack of reactivity
in the above-described transmetallation step (see the Support-
ing Information, Figure S15). Similarly to the tin analogue, the
interaction of the occupied lone pair of the fluoride anion with
the antibonding PhÀSi orbital is estimated to lead to a stabilisa-
tion of approximately 25.1 kcalmolÀ1.
Conclusion
The role of fluoride ions in the Stille reaction (cross-coupling of
ArX with Ar’SnBu3) has been established. Fluoride ions pro-
mote the rate-determining transmetallation by formation of
more reactive trans-[ArPdF(PPh3)2], which reacts with Ar’SnBu3
at room temperature, in sharp contrast to unreactive trans-
[ArPdBr(PPh3)2]. The fluoride ions also promote the reductive
elimination from trans-[ArPdAr’(PPh3)2] generated in the trans-
metallation. However, the concentration ratio [FÀ]/[Ar’SnBu3]
must not be too high because of the formation of unreactive
anionic [Ar’Sn(F)Bu3]À. This explains the two kinetically antago-
nistic roles exerted by the fluoride ions. Consequently, fluoride
ions play three roles in the Stille reaction, similar to those es-
tablished in the mechanisms of the Suzuki and Hiyama reac-
tions.
The presence of the fluoride ion on the Sn center in [2-
ThSn(F)nBu3]À weakens the 2-ThÀSn bond, but also fills the
electronic vacant levels of Sn. This prevents the transmetalla-
tion step, in which Lewis-acidic character is required for the Sn
centre to to interact with the fluoride of trans-[ArPdF(PPh3)2]
(Scheme 1). The decrease in the Lewis-acidic character of the
Sn center in [2-ThSn(F)nBu3]À can be quantified by using NBO
analysis: the interaction of the lone pairs of the fluoride anion
Experimental Section
Typical procedure for the kinetics of the reaction of trans-
[PhPdBr(PPh3)2] with 2-ThSnBu3 in the presence of nBu4NF
in DMF, as monitored by chronoamperometry at a rotating
disk electrode
Experiments were performed under argon atmosphere in a thermo-
stated three-electrode cell connected to a Schlenk line at 258C.
The counter electrode was a platinum wire of approximately 1 cm2
apparent surface area; the reference was a saturated calomel elec-
trode (SCE) separated from the solution by a bridge filled with
2 mL of a 0.3m nBu4NBF4 solution in DMF. A solution of degassed
DMF (18 mL) containing nBu4NBF4 (0.3m) as supporting electrolyte
was poured into the cell followed by trans-[PhPdBr(PPh3)2]
(28.3 mg, 0.036 mmol, 2 mm) and PPh3 (18.9 mg, 0.072 mmol). 2-
ThSnBu3 (85 mL, 0.27 mmol) was then introduced into the cell. The
kinetic measurements were performed at a rotating gold disk elec-
trode (d=2 mm, inserted into a Teflon holder, EDI 65109, radiome-
ter) with an angular velocity of 105 radsÀ1. The rotating electrode
was polarised at +0.1 V on the oxidation wave of [Pd0(PPh3)3].
nBu4NF (216 mL of a commercial 1m mother solution in THF,
0.216 mmol) was then added into the cell and the increase of the
oxidation current of [Pd0(PPh3)3] was recorded versus time up to
a limit value, attesting the end of the reaction. The solution turned
yellow, the color of [Pd0(PPh3)4] in DMF. Cyclic voltammetry at the
same steady electrode was then performed towards oxidation po-
Figure 5. [2-ThSn(F)nBu3]À: a) simplified 3-center/4-electron bond resonance
arising from the interaction of 3 axially symmetric p orbitals; b) occupied 3-
center/4-electron bond MOs for [2-ThSn(F)nBu3]À. Y1: HOMOÀ7, À0.254 eV;
Y2: HOMOÀ1, À0.189 eV. The third and empty MO resulting from the w-
bond is not represented due to the usual DFT drawbacks related to the pre-
cise computation of unoccupied levels.
Chem. Eur. J. 2015, 21, 18401 – 18406
www.chemeurj.org
18405
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
tentials and revealed the oxidation peak of [Pd0(PPh3)3]. Its yield
(97%) was determined from the increase of its oxidation peak cur-
rent after addition of an authentic sample of [Pd0(PPh3)4] (21 mg,
0.018 mmol). A cyclic voltammetry was then performed towards re-
duction potentials and revealed the reversible reduction peak of 2-
ThÀPh. Its yield (95%) was determined from the increase of its re-
duction peak current after addition of a commercially available au-
thentic sample of 2-ThÀPh (5 mg, 0.031 mmol).
ionic strength due to the continuous release of halide ions from ArX in
the course of the reactions. The addition of fluoride ions in noncatalytic
amounts also contributes to increasing the ionic strength. One can con-
sider that the CV experiments are performed under experimental condi-
tions that are close to those of the catalytic reactions.
[10] C. Amatore, A. Jutand, F. Khalil, M. A. M’Barki, L. Mottier, Organometallics
1993, 12, 3168–3178.
[11] For the synthesis of 1H and 1CN, see: P. Fitton, E. A. Rick, J. Organomet.
Chem. 1971, 28, 287–291.
[12] a) 3H is commercially available; b) for the synthesis of 3CN, see: C. Ama-
tore, A. Jutand, S. Negri, J. Organomet. Chem. 1990, 390, 389–398.
[13] a) Chemistry of Tin, 2nd ed. (Ed.: P. J. Smith), Blackie Academic & Profes-
sional, London, 1997; b) The formation of [PhSn(F)Me3]À [nBu4N]+ was
previously suggested by Nolan and co-workers (see ref. [5b]), who re-
ported the emergence of a new 19F-containing species in 19F NMR from
the reaction of nBu4NF with PhSnMe3; c) for DFT calculations, see
ref. [5h]; d) A 19F resonance in the same range (d=À137.1 ppm) was
detected for [PhSn(F)nBu3]À with the expected patterns due to J(F,Sn)
couplings, after addition of nBu4NF to PhSnBu3 (see the Supporting In-
formation, Figure S11 a).
Acknowledgements
This work was supported in part by the Centre National de la
Recherche Scientifique (UMR CNRS-ENS-UPMC 8640), the Minis-
t›re de la Recherche (Ecole Normale SupØrieure), the Hercules
Foundation, and the University of Antwerp (BOF). We thank
Johnson Matthey for a loan of palladium salt.
[14] For the characterisation of trans-[ArPdAr’(PPh3)2] complexes, see: a) C.
Adamo, C. Amatore, I. Ciofini, A. Jutand, H. Lakmini, J. Am. Chem. Soc.
2006, 128, 6829–6836; b) C. Amatore, A. Jutand, G. Le Duc, Chem. Eur.
J. 2011, 17, 2492–2503 and Ref. [7].
Keywords: fluoride · kinetics · palladium · Stille reaction ·
transmetallation
[15] For the synthesis of trans-[PhPdF(PPh3)2], see: M. C. Pilon, V. V. Grushin,
Organometallics 1998, 17, 1774–1781.
[16] The formation of complex 6CN,Ph has already been reported, by reaction
of trans-[{p-(CN)C6H4}PdF(PPh3)2] with PhB(OH)3 (see ref. [7a]) and with
PhSi(OMe)3 (see ref. [8]).
[17] For reductive elimination induced by a fifth ligand in square planar d10
complexes, see: a) R. Giovannini, P. Knochel, J. Am. Chem. Soc. 1998,
120, 11186–11187; b) For pentacoordinated anionic PdII complexes fa-
vouring reductive elimination, see: R. Gramage-Doria, D. Armspach, D.
Matt, L. Toupet, Chem. Eur. J. 2012, 18, 10813–10816.
[1] a) M. Kosugi, K. Sasazawa, T. Migita, Chem. Lett. 1977, 301–302; b) D.
Milstein, J. K. Stille, J. Am. Chem. Soc. 1978, 100, 3636–3638; c) V. Farina,
G. P. Roth in Advances in Metal-Organic Chemistry, Vol. 5 (Ed.: L. S. Liebe-
skind), JAI Press, Greenwich, CT, 1996, pp. 1–53; d) V. Farina, V. Krishna-
murthy, X. J. Scott, Organic Reactions, Vol. 50: The Stille Reaction, Wiley,
New York, 1997; e) C. Amatore, A. Bahsoun, A. Jutand, G. Meyer, A. Nde-
di Ntepe, L. Ricard, J. Am. Chem. Soc. 2003, 125, 4212–4222; f) P. Espi-
net, A. M. Echavarren, Angew. Chem. Int. Ed. 2004, 43, 4704–4734;
Angew. Chem. 2004, 116, 4808–4839; g) G. Eidamshaus, J. D. Birch, Org.
Lett. 2008, 10, 4211–4214; h) M. M. Heravi, E. Hashemi, F. Azimian, Tetra-
hedron 2014, 70, 7–21; i) T. O. Ronson, J. R. Carney, A. C. Whitwood,
R. J. K. Taylor, I. J. S. Fairlamb, Chem. Commun. 2015, 51, 3466–3469.
[2] For recent approaches for Ar’SnR3 synthesis, see: D. Qui, H. Meng, L.
Jin, S. Wang, S. Tang, X. Wang, F. Mo, Y. Zhang, J. Wang, Angew. Chem.
Int. Ed. 2013, 52, 11581–11584; Angew. Chem. 2013, 125, 11795–11798.
[3] a) L. S. Liebeskind, R. W. Fengi, J. Org. Chem. 1990, 55, 5359–5364;
b) G. P. Roth, V. Farina, L. S. Liebeskind, E. Pena-Cabrera, Tetrahedron
Lett. 1995, 36, 2191–3194; c) V. Farina, S. Kapadia, B. Krishnan, C. Wang,
L. S. Liebeskind, J. Org. Chem. 1994, 59, 5905–5911.
[18] This kinetic law was established in refs. [7a] and [8].
[19] a) M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H.
Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G.
Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Ha-
segawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven,
J. A. Montgomery, Jr. , J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E.
Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Ra-
ghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N.
Rega, N. J. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo,
J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R.
Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakr-
zewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Dan-
iels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian
09, revision B.01; Gaussian, Inc.: Wallingford, CT, 2010; b) J. P. Foster, F.
Weinhold, J. Am. Chem. Soc. 1980, 102, 7211–7218; c) R. Krishnan, J. S.
Binkley, R. Seeger, J. A. Pople, J. Chem. Phys. 1980, 72, 650–654; d) L.
von Szentpµly, P. Fuentealba, H. Preuss, H. Stoll, Chem. Phys. Lett. 1982,
93, 555–559; e) S. Grimme, J. Comput. Chem. 2006, 27, 1787–1799; f) S.
[4] S. Verbeeck, C. Meyers, P. Franck, A. Jutand, B. U. W. Maes, Chem. Eur. J.
2010, 16, 12831–12837.
[5] a) A. F. Littke, G. C. Fu, Angew. Chem. Int. Ed. 1999, 38, 2411–2413;
Angew. Chem. 1999, 111, 2568–2570; b) G. A. Grasa, S. P. Nolan, Org.
Lett. 2001, 3, 119–122; c) A. F. Littke, L. Schwarz, G. C. Fu, J. Am. Chem.
Soc. 2002, 124, 6343–6348; d) W. Su, S. Urgaonkar, P. A. McLaughlin,
J. G. Verkade, J. Am. Chem. Soc. 2004, 126, 16433–16439; e) A. Echavar-
ren, Angew. Chem. Int. Ed. 2005, 44, 3962–3965; Angew. Chem. 2005,
117, 4028–4031; f) for reaction of preformed [R’3SnF2]À, see: A. García
Martínez, J. Osío Barcina, M. Colorado Heras, A. de Fresno Cerezo, Orga-
nometallics 2001, 20, 1020–1023; g) J. R. Naber, S. L. Buchwald, Adv.
Synth. Catal. 2008, 350, 957–961; h) for DFT calculations on the role of
fluoride ions, see: A. Ariafard, B. F. Yates, J. Am. Chem. Soc. 2009, 131,
13981–13991.
[6] a) S. P. H. Mee, V. Lee, J. E. Baldwin, Angew. Chem. Int. Ed. 2004, 43,
1132–1136; Angew. Chem. 2004, 116, 1152–1156; b) S. P. H. Mee, V. Lee,
J. E. Baldwin, Chem. Eur. J. 2005, 11, 3294–3308.
[7] a) C. Amatore, A. Jutand, G. Le Duc, Angew. Chem. Int. Ed. 2012, 51,
1379–1382; Angew. Chem. 2012, 124, 1408–1411; b) C. Amatore, G.
Le Duc, A. Jutand, Chem. Eur. J. 2013, 19, 10082–10093.
ˇ
Miertus, E. Scrocco, J. Tomasi, Chem. Phys. 1981, 55, 117–129; g) G. Scal-
mani, M. J. Frisch, B. Mennucci, J. Tomasi, R. Cammi, V. Barone, J. Chem.
Phys. 2006, 124, 094107.
[20] a) Previous DFT calculations proposed the formation of [PhPdF(PtBu3)],
which reacts with CH=CHÀSnMe3. This step is more favoured than the
reaction of [PhPdCl(PtBu3)] with anionic [CH=CHÀSn(F)Me3]À (see
ref. [5h]); b) In [PhSi(F)(OMe)3]À: q(OMe)=À0,639 e, À0,641 e, À0,640 e;
q(F)=À0,711 e; q(Ph)=À0,549 e.; q(Si)= +2,182 e. Similarly, in
[PhB(OH)3]À the highest negative charge is located on the OH group
and the lowest on the Ph group: q(OH)=À0.528 e, À0.524 e,
À0.523 e.; q(Ph)=À0.444 e.; q(B)= +1.019 e. This supports the fact
that [PhB(OH)3]À was also found to be unreactive with trans-
[ArPdX(PPh3)2] (X=Br, F; see ref. [14b]).
[8] C. Amatore, L. Grimaud, G. Le Duc, A. Jutand, Angew. Chem. Int. Ed.
2014, 53, 6982–6985.
[9] a) For the use of electrochemical techniques to investigate the mecha-
nism of palladium-catalyzed reactions, see: A. Jutand, Chem. Rev. 2008,
108, 2300–2347; b) CV experiments require the use of a supporting
electrolyte so that experiments are performed under a high ionic
strength. This mimics catalytic reactions that are performed under high
Received: August 20, 2015
Published online on November 9, 2015
Chem. Eur. J. 2015, 21, 18401 – 18406
www.chemeurj.org
18406
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim