New insights into the stoichiometric and catalytic reactivity of unsaturated Pd3(dppm)3COn+ clusters (n = 0, 1) towards halocarbons - First evidence for inorganic by-products

Article abstract of DOI:10.1002/ejic.200400594

The title clusters, Pd₃(dppm)₃(CO)⁺ and Pd₃(dppm)₃(CO)⁰ can be electrochemically generated from the 1- and 2-electron reductions, respectively, of the Pd ₃(dppm)₃(CO)²⁺ cluster [dppm = bis(diphenylphosphanyl)methane; Pd₃²⁺]. Pd ₃⁺ reacts in a stoichiometric ratio with methyl iodide, MeI, and benzyl bromide, BzBr, in THF to provide the corresponding Pd ₃(X)⁺ adducts (X = I, Br respectively) as inorganic products. Other products are Bz₂ and PhMe for BzBr but, for MeI, no organic product was observed (since they are too volatile). In the presence of the same substrates, Pd₃⁰ also reacts in a stoichiometric ratio to form the same organics and the Pd₃-(X)⁺ adducts (X = I and Br). However for MeI, the major inorganic product is the A-frame Pd₂(dppm)₂(Me)₂I⁺ binuclear complex. For BzBr, the corresponding A-frame complex Pd₂(dppm) ₂(Bz)₂Br⁺ could not be detected. The spin-trap agents, 2,2,6,6-tetramethylpiperidin-1-oxyl (TEMPO) and 5,5′-dimethyl-1- pyrroline N-oxide (DMPO), have been used to demonstrate the intermediacy of the radical Bz^.. The catalytic generation of "Bz^." was performed using two methods, i.e. 1) using a copper anode as the working electrode [Pd₃-(Br)⁺ + Cu - e^- → Pd ₃²⁺ + CuBr (s)] and 2) using a carbon cathode as the working electrode [Pd₃(Br)⁺ + 2e^- → Pd₃⁰ + Br^-]. The chemical yields for Bz ₂ vary between 50 and 56% and the Faradic yield is of the order of 90% for method 1 and between 52 and 59% for method 2 [taking into account the quantity of electricity necessary to reduce the catalyst Pd³-(Br) ⁺]. The X-ray structure of Pd₃(dppm)₃(CO)(Br) ⁺ is presented and the following parameters were recorded: monoclinic space group P2₁/n, a = 10.6546(2), b = 37.1091(7), c = 212714(7) A, β = 91.55(1)°, V = 8407,3(4) A³, Z = 4, R₁ = 0.0581 [I > 2σ(I)], wR₂ = 0.1478 (all data). Wiley-VCH Verlag GmbH & Co. KGaA, 2005.

Full text of DOI:10.1002/ejic.200400594

FULL PAPER
New Insights into the Stoichiometric and Catalytic Reactivity of Unsaturated
Pd3(dppm)3COn+ Clusters (n = 0, 1) Towards Halocarbons – First Evidence
for Inorganic By-Products
Dominique Lucas,*[a] Frédéric Lemaître,[a,b] Beatriz Gallego-Gómez,[a] Cyril Cugnet,[a]
Philippe Richard,[a] Yves Mugnier,[a] and Pierre D. Harvey*[b]
Keywords: Palladium cluster / Alkyl halide / Reduction / Electrocatalysis / Alkyl radicals
The title clusters, Pd3(dppm)3(CO)+ and Pd3(dppm)3(CO)0
can be electrochemically generated from the 1- and 2-elec-
to demonstrate the intermediacy of the radical Bz·. The cata-
lytic generation of “Bz·” was performed using two methods,
tron reductions, respectively, of the Pd3(dppm)3(CO)2+ cluster
i.e. 1) using a copper anode as the working electrode [Pd3-
+
2+
[dppm = bis(diphenylphosphanyl)methane; Pd32+]. Pd3 re- (Br)+ + Cu – e– Ǟ Pd3 + CuBr (s)] and 2) using a carbon
0
acts in a stoichiometric ratio with methyl iodide, MeI, and
benzyl bromide, BzBr, in THF to provide the corresponding
Pd3(X)+ adducts (X = I, Br respectively) as inorganic products.
Other products are Bz2 and PhMe for BzBr but, for MeI, no
organic product was observed (since they are too volatile).
cathode as the working electrode [Pd3(Br)+ + 2e– Ǟ Pd3
+
Br–]. The chemical yields for Bz2 vary between 50 and 56%
and the Faradic yield is of the order of 90% for method 1 and
between 52 and 59% for method 2 [taking into account the
quantity of electricity necessary to reduce the catalyst Pd3-
(Br)+]. The X-ray structure of Pd3(dppm)3(CO)(Br)+ is pre-
sented and the following parameters were recorded: mono-
clinic space group P21/n, a = 10.6546(2), b = 37.1091(7), c =
0
In the presence of the same substrates, Pd3 also reacts in a
stoichiometric ratio to form the same organics and the Pd3-
(X)+ adducts (X = I and Br). However for MeI, the major inor-
ganic product is the A-frame Pd2(dppm)2(Me)2I+ binuclear
complex. For BzBr, the corresponding A-frame complex
Pd2(dppm)2(Bz)2Br+ could not be detected. The spin-trap
agents, 2,2,6,6-tetramethylpiperidin-1-oxyl (TEMPO) and
5,5Ј-dimethyl-1-pyrroline N-oxide (DMPO), have been used
21.2714(7) Å, β = 91.55(1)o, V = 8407.3(4) Å3, Z = 4, R1
0.0581 [I Ͼ 2σ(I)], wR2 = 0.1478 (all data).
=
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2005)
Introduction
The C–X bond activation is an important process in or-
ganic synthesis.[1–7] In this respect, our groups have investi-
gated numerous thermal and electrochemical reactions be-
2+ [8–10]
tween Pd3(dppm)3(CO)2+ (1; Pd3
)
and halocarbons
(RX) and some important electrochemically induced or-
ganic catalyses have been reported.[11–13] The direct thermal
2+
reaction (Pd3 + RX) quantitatively produces the corre-
+
0
species Pd3 and Pd3 have only been briefly studied under
catalytic conditions.
sponding adducts Pd3(X)+ and “R+”.
The elucidation of the mechanism of C–X bond cleavage
for alkyl halides and acid chlorides has been studied in
+
We now wish to report the reactivity of the Pd3 and
0
Pd3 species under stoichiometric conditions towards MeI
2+
some detail.[14] However, whereas the reactivity of Pd3
and BzBr and evidence for the slow formation of the d8–d8
towards RX has been fully investigated under both stoichio-
metric and catalytic conditions, the corresponding reduced
0
A-frame Pd2(dppm)2(Me)2I+ complex during the Pd3
+
MeI reactions has been obtained for the first time, indicat-
ing the presence of a parallel process which may explain the
catalyst poisoning in some cases.
The use of BzBr was made since the presence of the po-
tential Bz+ and Bz· intermediates can easily be deduced
from the lesser volatile products, hence providing key infor-
mation on the mechanism. Similarly, the MeI substrate is
well known to undergo facile oxidative addition with Pd
species in the low oxidation states, often resulting in the
[a] Laboratoire de Synthèse et dЈElectrosynthèse Organométal-
liques, CNRS UMR 5188, Faculté des Sciences Mirande, Uni-
versité de Bourgogne,
9 avenue Alain Savary, 21100 Dijon, France
Fax: +33-380-39-36-64
E-mail: lucasd@u-bourgogne.fr
[b] Département de Chimie, Université de Sherbrooke,
Sherbrooke, Québec, Canada J1K 2R1
Fax: +1-819-821-8017
E-mail: pharvey@courrier.usherb.ca
Eur. J. Inorg. Chem. 2005, 1011–1018
DOI: 10.1002/ejic.200400594
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1011

D. Lucas, P. D. Harvey et al.
FULL PAPER
volving the “Bz·” species which produces Bz2 by homo-
coupling and PhMe by hydrogen abstraction from the sol-
vent.[24] The addition of an excess of the spin trap agent
5,5Ј-dimethyl-1-pyrroline N-oxide (DMPO, 10 equiv.) to the
Pd3+/BzBr reaction yields the stable Bz-DMPO spin adduct
as readily observed by EPR spectroscopy which exhibits the
H
expected 6-line signal:[25]aβ = 20.66 G, aN = 14.16 G.[26,27]
The absence of Bz2 in reaction (1) suggests a quantitative
reaction.
formation of a Pd–Me bond.[15–18] Finally, the efficient
catalytic generation of “Bz·” was examined.
Results and Discussion
+
Stoichiometric Reactivity of Pd3
As reported earlier,[11,14,19] Pd3+ can be quantitatively ob-
tained from a 1-electron reduction of the corresponding di-
cationic compound. Its RDE voltammogram is depicted in
Figure 1 (trace a) and exhibits two closely spaced waves for
oxidation and reduction labelled AЈ1 and A2, respectively
[E1/2 (A1Ј) = –0.24 V and E1/2 (A2) = –0.51 V vs. SCE]. Line
1b results from the addition of an excess of methyl iodide
to Pd3+. The disappearance of waves AЈ1 and A2 can be
The two possible mechanisms for the production of the
+
organic radical [Pd3 + RX Ǟ Pd3(X)+ + R·] are 1) an
+
outer-sphere electron transfer from Pd3 to RX, followed
by X– elimination from the intermediate R–X·– (X– is then
trapped by the resultant Pd32+) or 2) a halide abstraction
process occurring within a host-guest Pd3···(RX)+ assembly
in which the homolytic C–X bond cleavage takes place, as
described by reactions (2) and (3).
+
observed, demonstrating that the Pd3 complex has com-
pletely reacted. On the other hand, the current waves A3
and A4 indicate the formation of the previously known io-
dide coordination cluster Pd3(I)+ [E1/2 (A3) = –0.80 V and
E1/2 (A4) = –1.25 V vs. SCE] which was confirmed by 31P
NMR spectroscopic analysis of the reaction solution
in which only the signal of Pd3(I)+ can be seen (δ =
–6.4 ppm).[20–22]
The driving force in reaction (3) is the formation of the
very stable Pd3(X)+ adduct. The former outer-sphere
mechanism is highly unlikely since the oxidation (E1/2
=
–0.29 V) and reduction potentials (E1/2 Ϸ –2.5 V vs.
+
SCE)[28,29] for Pd3 and R–X, respectively, are not thermo-
dynamically favourable for such a process. On the other
hand, the second proposal is more likely to occur based
2+
upon the well-established mechanism reported for the Pd3
2+
cluster [Pd3
+ RX Ǟ Pd3···(XR)2+ Ǟ Pd3(X)+
+
“R+”].[11–14]
Going back to the Pd3+/MeI reaction, the absence of ob-
served organics is due to their high volatility and, based on
the observation made for the Pd3+/BzBr reaction, one can
suggest that these are Me2 and MeH.
0
Reactivity with Pd3
The 2-electron reduction of Pd32+ leads to Pd30 quantita-
tively. Addition of an excess of MeI to the electrochemically
0
generated Pd3 -containing solution monitored by RDE vol-
+
Figure 1. RDE voltammogram of Pd3 (0.81 mM) in THF with
tammetry (Figure 2) results in the complete disappearance
0.2 m Bu4NPF6: (a) alone, (b) immediately after addition of 1.1
0
of the two 1-electron oxidation waves of the Pd3 species
molar equiv. of CH3I.
(waves A1Ј and A2Ј) and the appearance of waves A3 and
+
Addition of BzBr (1.1 equiv.) to Pd3 produces Pd3(Br)+ B1 (–1.18 V vs. SCE) which are due to the anticipated Pd3-
as confirmed by the comparison of the electrochemical (I)+ species and the known dinuclear d8–d8 A-frame com-
[E1/2 = –0.7 V (A3)] and 31P NMR spectroscopic data (δ = plex Pd2(dppm)2(Me)2I+, respectively.[30] Based upon the
–6.14 ppm in [D6]acetone) as well as the formation of Bz2 current amplitudes of waves A3 and B1 and the relative in-
(44% based upon GCMS data) and PhMe.[23] During the tensity of the corresponding 31P NMR signals (at –6.40 and
course of this investigation, crystals of Pd3(Br)+ suitable for +19.76 ppm), the relative proportions are Ϸ 20–25% and
X-ray analysis were obtained and investigated (see below). Ϸ 80–75%, respectively. This result is different from that
+
The organic products indicate a radical-type reactivity in- of Pd3 for which no A-frame complex was observed and
1012
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2005, 1011–1018

Reactivity of Pd3(dppm)3COn+ Clusters (n = 0, 1) Towards Halocarbons
FULL PAPER
0
suggests the presence of oxidative addition of MeI to Pd3
as a parallel pathway.
+
A similar mechanism to that for Pd3 can be proposed;
see reaction (7).
The halocarbon molecule penetrates inside the cluster
cavity forming a transient reactive host-guest complex
Pd3···(XR)0 [reaction (5)], in which a homolytic C–X bond
cleavage takes place, giving the corresponding alkyl radical
“R·” [reaction (6)]. The resultant Pd3(X)0 intermediate is
unstable and eliminates X– to form Pd3+ [reaction (7)]. This
latter process is well known and has been previously estab-
lished for X– = Br– and I–.[8,22] Finally, the paramagnetic
0
Figure 2. RDE voltammogram of Pd3 (1.2 mM) in THF with
0.2 M Bu4NPF6: (a) alone, (b) immediately after addition of an
+
excess (4 molar equiv.) of CH3I.
Pd3 cluster undergoes reactions (2) and (3).
A second mechanism can also be considered and this in-
volves the heterolytic C–X bond cleavage of the species
Pd3···(RX)0 followed by the reductive elimination of “R·”;
see reactions (8) and (9).
0
The addition of BzBr (2 equiv.) to a Pd3 -containing
solution leads to a slower reaction with respect to that of a
+
similar reaction with Pd3 (as monitored by the RDE
method; Figure 3) but the presence of Bz2 (GC-MS yields
Ϸ 44%) and PhMe as organic products could be detected
in about the same relative amount.
+
Again, Pd3 undergoes reactions (2) and (3). Reac-
tion (8), which differs from reaction (6), finds its driving
force from the electrostatic attraction of a potential carboc-
0
ation “R+” with the electron-rich Pd3 centre. However, re-
actions 8 and 9 are both unlikely. The postulated Pd3(R)+
intermediate exhibits a Pd32+ centre and a carbanion which
follows the octet rule similar to the Pd3X+ adducts. Its
+
dissociation into Pd3 and “R·” [reaction (9)] would be far
too high in energy. In addition, the formation of the Pd3-
(R)+ species would prove difficult since the approach of the
0
Pd3 cluster towards the alkyl carbon atom in BzBr is steri-
cally encumbered. Finally, such alkyl adducts of Pd32+ have
never been observed and appear highly reactive.[31]
In order to support this hypothesis, DFT computations
were
performed
on
the
model
compounds
Pd3(H2PCH2PH2)3(CO)(Me)+ and Pd3(H2PCH2PH2)3-
(CO)(Br)+. The first observation is that the computed total
energies are –17067.91 and –19590.39 a. u. (arbitrary units),
respectively, indicating the greater stability of the Pd3(Br)+
species. The second observation pointing in the same direc-
0
Figure 3. RDE voltammogram of Pd3 (0.75 mM) in THF with
0.2 M Bu4NPF6: (a) alone, (b) immediately after addition of 2 mo-
lar equiv. of PhCH2Br, (c) after 15 min at room temperature.
On the other hand, Pd3(Br)+ was quantitatively obtained tion is that the computed energies for individual MOs are
and no trace of the d8–d8 Pd2(dppm)2(Bz)2Br+ complex was systematically higher for the Pd3(Me)+ species. As an exam-
detected on this occasion (see below). The addition of ple, a partial comparative MO diagram for the LUMO and
DMPO to the solution prior to reaction leads to the same HOMO is presented in Figure 4. The MO’s have been de-
spin adduct as described above and no Bz2 was observed, scribed previously elsewhere and this is not within the scope
also suggesting completion of the spin-trap reaction. The of this work.[32,33] In conclusion, DFT supports the hypoth-
overall reaction for the radical formation can be written as esis that Pd3(Me)+ is a less stable species than Pd3(Br)+, Br–
2+
in reaction (4).
or Pd3 .
Eur. J. Inorg. Chem. 2005, 1011–1018
www.eurjic.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1013

D. Lucas, P. D. Harvey et al.
FULL PAPER
Some questions remain such as how does the structure
of the halogenated substrate affect the occurrence of the A-
frame product? What is the main factor between the halo-
gen and the alkyl group? In order to address these points,
Pd30 was treated with a slight excess of BzI in a cross experi-
ment. This led to quantitative formation of Pd3(I)+ meaning
that reaction (4) proceeds solely.
Thus, the steric hindrance of the alkyl group is likely
playing an important role in the oxidative addition of RX
0
to Pd3 . In order to confirm this proposal, a series of alkyl
0
iodides, RI, were tested in the reaction with Pd3 (R = Et,
iPr, tBu). In no case was the A-frame product observed and
Pd3(I)+ was always the sole inorganic product. Hence, reac-
tion (10) may be extremely sensitive to the steric encum-
brance of the alkyl group since it operates only when R =
Me.
Figure 4. Left: Partial MO diagram of the HOMO and LUMO of
the model compounds Pd3(H2PCH2PH2)3(CO)(Me)+ and
Pd3(H2PCH2PH2)3(CO)(Br)+. Right: MO drawing for the HOMO
and LUMO for Pd3(H2PCH2PH2)3(CO)(Me)+.
Catalytic Formation of “Bz·” Radicals
Two new questions now arise. Firstly, why can the radical
Bz· intermediate be observed in quantitative amounts
The chemistry of free radicals has experienced an impor-
tant growth in recent years[40,41] but the use of organotin
compounds such as R3SnH remains the first choice for rad-
ical generation.[42] However, because their use must be stoi-
chiometric, their toxicity has encouraged chemists to find
alternate and more environmentally friendly methods.[43]
Thus, the indirect reduction of RX using an electron trans-
fer catalyst mediating the RX conversion to “R·” is appeal-
ing.[44]
0
(based upon observed Bz2 and PhMe) for both Pd3 and
Pd3+ species and, secondly, why can two inorganic products
0
be observed for the Pd3 /MeI reaction (Pd3 adduct and A-
frame) while the Pd3+/MeI reaction leads to only one inor-
ganic product (Pd3 adduct)? These observations imply that
another competitive process must occur which can be for-
malised as shown in reaction (10).
In an attempt to render the stoichiometric reactions for
“Bz·” described above catalytic, two methodologies were ex-
amined. The first involves Pd3(Br)+ as a precatalyst in a
single compartment electrochemical cell equipped with a
copper anode as the controlled-potential working electrode,
a platinum cathode as the counter-electrode and CH2Cl2 as
the solvent. The electrolysis proceeds as shown in Scheme 1.
The 1-electron oxidation at the copper anode produces
The answer to the first question can be found in the con-
siderable stability of the Pd3(X)+ adducts[22] [reaction (3)]
which readily drives the reactions. Thus, the same driving
force may be present in the case of Pd3+. The answer to the
0
second question can be found in the synthesis of Pd3
itself.[34] The reduction of Pd(dppm)Cl2 or Pd2(dppm)2Cl2
leads to the zero-valent “Pd2(dppm)2” species. In the pres-
2+
2+
CuBr(s) and Pd3 from Pd3(Br)+. Pd3 diffuses towards
+
ence of a source of CO, this very reactive binuclear complex
the cathode where it is reduced to Pd3 which itself diffuses
0 [35,36]
in the solution and reacts with BzBr to generate Pd3(Br)+
and “Bz·”. Bz2 is the sole organic product observed and no
PhMe is formed. The reason is that, unlike THF, CH2Cl2
is not an H-donor. Table 1 summarises the results (entry
provides Pd3 .
On the other hand, the synthesis of
Pd2(dppm)2(Me)2(I)+ can be performed by addition of MeI
[37]
to Pd2(dppm)3
and the latter can dissociate into
“Pd2(dppm)2” and dppm.[38] These previous results suggest
0
that Pd3 should dissociate to form the highly reactive spe-
cies “Pd2(dppm)2” which would react with MeI.
On the other hand, the absence of reactivity in the case
of BzBr, to form the A-frame product, suggests an impor-
0
tant difference between the rates of the Pd3 /BzBr and
“Pd2(dppm)2”/BzBr reactions which would favour the first
reaction. It should be noted that the A-frame product
Pd2(dppm)2(Bz)2(Br)+ has been previously reported in the
literature[39] but the synthesis was not through the BzBr oxi-
dative addition pathway as for MeI. This method was at-
tempted but with no success which is consistent with our
results. The preferred route was the reaction between
Pd(dppm)Br2 and the Grignard reagent BzMgBr.
Scheme 1.
1014
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2005, 1011–1018

Reactivity of Pd3(dppm)3COn+ Clusters (n = 0, 1) Towards Halocarbons
FULL PAPER
Table 1. Controlled-potential electrolysis of Pd3(Br)+ in the presence of BzBr.
Working
Entry electrode
(WE)
Solvent[a]
Q (F per
mol of
BzBr)
EWE (V vs.
SCE)
Molar ratio
BzBr/cluster
% Chemical % Faradic
yield
Spin trap
Product(s)
yield[b]
1
2
3
4
Cu anode[c]
+0.6
–0.9
–1.3
–0.9
CH2Cl2
THF
THF
25
14
14
14
0.84
1.24
1.21
1.14
none
none
dibenzyl
dibenzyl
dibenzyl
spin adduct
dibenzyl
spin Adduct
76[d]
90
56
50
C cathode[e]
C cathode[e]
C cathode[e]
70[d]
none
60[d]
[g]
THF
DMPO[f]
42[d]
30[h]
38
28
5
C cathode[e]
–0.9
THF
14
1.09
TEMPO[f]
[a] In the case of entry 1, CH2Cl2 was used as the solvent and no PhMe was observed; for entries 2–5, THF was used and PhMe was
detected by GC–MS but not quantified because of strong overlap with the solvent peak in the chromatogram. [b] Faradic yield = Chemical
Yield/Q. [c] The platinum counter-electrode was not separated, being directly immersed in the solution in order for it to electrolyse. [d]
Measured by gas chromatography. [e] Standard assembly with an isolated platinum counter-electrode located in a separate compartment.
[f] Quantity = 1 molar equivalent with respect to BzBr. [g] Only observed by ESR with no quantification. [h] Isolated yield.
1). Both the chemical (76%) and Faradic (90%) yields are the Pd–X distances indicate strong ionic interactions and
good.
The second method involves the use of a carbon cathode
vary as expected (Cl Ͻ Br Ͻ I).
0
where the single 2-electron reduction of Pd3(Br)+ in Pd3
(at about –1 V vs. SCE) occurs in THF. The catalytic cycle
is shown in Scheme 2 based on the findings of the stoichio-
metric reactions above. This method leads to slightly lower
chemical yields of Bz2 and the presence of PhMe was noted.
The data are summarised in Table 1.
Scheme 2.
The addition of the spin trap DMPO to the solution dur-
ing catalysis (entry 4) leads to the formation of the spin
adduct as shown in reaction (1). On the other hand, ad-
dition of 2,2,6,6-tetramethylpiperidin-1-oxyl (TEMPO)
[45–48]
(entry 5) leads to a partial trapping where Bz2 and
the diamagnetic spin adduct can be observed [reaction (12)].
Figure 5. Top: ORTEP representation of Pd3(Br)+. The PF6
–
counter-anion, crystallisation toluene molecules and H-atoms are
not shown for clarity. The ellipsoids are shown at the 50% prob-
ability level. Bottom: Drawing showing the skeleton of the cluster
and the atom numbering.
X-ray Structure of Pd3(Br)+
This compound completes the series for Pd3(X)+ (Fig-
ure 5 and Table 2) where X = Cl[49,50] and I.[21] A compari-
son of the skeleton (Table 3) indicates a fairly robust core
in which the nature of the halide does not perturb the bond
Conclusions
The stoichiometric reactivity toward halocarbons of the
+
0
distances within the experimental uncertainties. In all cases, Pd3 and Pd3 species has been investigated for the first
Eur. J. Inorg. Chem. 2005, 1011–1018
www.eurjic.org © 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 1015

D. Lucas, P. D. Harvey et al.
FULL PAPER
Table 2. Selected bond distances [Å] and angles [°] for Pd3(Br)+.
was synthesized by mixing stoichiometric amounts of Bu4NOH
(40% in water) and HPF6 (60% in water). After filtration, the salt
was recrystallised twice from ethanol and dried at 80 °C for at least
two days.
Distances
Angles
Pd(1)–C(4) 2.143(10)
Pd(2)–C(4) 2.139(10)
Pd(3)–C(4) 2.145(9)
Pd(1)–Pd(2)–Pd(3)
Pd(1)–Pd(3)–Pd(2)
Pd(2)–Pd(1)–Pd(3)
O–C(4)–Pd(1)
O–C(4)–Pd(2)
O–C(4)–Pd(3)
C(4)–Pd(1)–P(1)
C(4)–Pd(1)–P(6)
C(4)–Pd(2)–P(2)
C(4)–Pd(2)–P(3)
C(4)–Pd(3)–P(4)
C(4)–Pd(3)–P(5)
P(1)–Pd(1)–P(6)
P(2)–Pd(2)–P(3)
P(4)–Pd(3)–P(5)
P(1)–Pd(1)–Br
P(6)–Pd(1)–Br
P(2)–Pd(2)–Br
P(3)–Pd(2)–Br
P(5)–Pd(3)–Br
P(4)–Pd(3)–Br
59.63(2)
60.26(2)
60.11(2)
132.7(9)
141.0(9)
132.9(10)
114.8(2)
112.3(2)
115.2(2)
118.4(2)
113.6(2)
112.1(2)
109.55(7)
108.44(7)
108.97(7)
104.46(6)
113.22(6)
104.95(6)
98.83(6)
113.85(6)
104.09(6)
Apparatus: NMR spectra were measured on a Bruker WM 300
spectrometer (31P NMR: 121.497 MHz). The chemical shifts are
reported with respect to H3PO4 (31P NMR). The GC–MS data
were collected on a Hewlett–Packard 6890 Series instrument. EPR
measurements were carried out on a Bruker ESP 300 spectrometer;
field calibration was made with DPPH (g = 2.0037).
Pd(1)–Br
Pd(2)–Br
Pd(3)–Br
3.027(9)
2.8276(10)
2.9666(11)
Pd(1)–Pd(2) 2.5991(8)
Pd(1)–Pd(3) 2.5826(8)
Pd(2)–Pd(3) 2.5950(8)
Pd(1)–P(1) 2.315(2)
Pd(1)–P(6) 2.314(2)
Pd(2)–P(2) 2.315(2)
Pd(2)–P(3) 2.311(2)
Pd(3)–P(4) 2.309(2)
Pd(3)–P(5) 2.305(2)
Electrochemical Experiments: All manipulations were performed
using Schlenk techniques in an atmosphere of dry oxygen-free ar-
gon. The supporting electrolyte was degassed under vacuum before
use and then dissolved to a concentration of 0.2 m. For cyclic vol-
tammetry experiments, the concentration of the analyte was almost
10–3 m. Voltammetric analyses were carried out in a standard three-
electrode cell with a Tacussel PJT24–1 potentiostat connected to a
Tacussel GSTP4 waveform generator. The reference electrode was
a saturated calomel electrode(SCE)separated from the solution by
a sintered glass disk. The auxiliary electrode was a platinum wire.
For all voltammetric measurements, the working electrode was a
vitreous carbon electrode (φ = 3 mm). Under these conditions,
when operating in THF, the formal potential for the ferrocene+/–
couple was found to be +0.56 V versus the SCE. The controlled
potential electrolysis was performed with an Amel 552 potentiostat
coupled with an Amel 721 electronic integrator. High scale electro-
lyses were performed in a cell with three compartments separated
with fritted glasses of medium porosity. Carbon gauze was used as
the working electrode, a platinum plate as the counter-electrode
and a saturated calomel electrode as the reference electrode. The
electrolyses corresponding to the copper method were performed
with a copper plate as the working electrode (anode), a platinum
plate as the counter-electrode (cathode) and a saturated calomel
electrode as the reference electrode, the latter being separated from
the others in a two-compartment cell.
C(4)–O
1.104(10)
Table 3. Comparison of selected bond distances [Å] for the Pd3-
(X)+ clusters.
Average
X = Br
(this work)
X = Cl[49]
X = Cl[50]
X = I[21]
distance[a]
d(Pd–Pd)
d(Pd–P)
d(Pd–C)
d(Pd–X)
2.595(17)
2.311(30)
2.103(10)
2.861
2.591(11)
2.310(11)
2.133(50)
2.899
2.592(10)
2.311(6)
2.142(3)
2.940
2.596(5)
2.314(9)
2.163(29)
3.022
[a] The uncertainties are based on the maximum difference between
the average value and experimental data.
time and has allowed the discovery of a new reactivity, par-
ticularly with respect to the inorganic products evolved
Procedures for the Electrochemical Experiments. Conventional High-
Scale Electrolysis: The [Pd3(dppm)3(CO)](PF6)2 cluster (12.5 mg,
7×10–3 mmol, 6 mol-% per mol of BzBr) and BzBr (12.45 μL,
0.105 mmol) were added to the cathodic compartment of the cell
0
from the Pd3 species. The resultant A-frame product is
stable since its reduction potential is higher than that for
+
0
Pd3 and Pd3 . Although there is no evidence in this work,
the parallel pathway leading to the A-frame product may containing a 0.2 m solution of Bu4NPF6 in THF (10 mL). The an-
odic compartment and the reference electrode compartment were
filled with the Bu4NPF6–THF solution. The potential was set to
–0.9 V vs. SCE. The electrolysis was stopped after the current had
dropped to less than 0.5 mA. After filtration of the mixture, the
solvent was removed and the residue extracted with diethyl ether
(3×5 mL). The internal standard method was used to measure the
GC yields of the products. These were identified by comparison of
the GC–MS spectra and GC retention times with those of available
authentic samples. The electrolysis was also performed in the pres-
ence of TEMPO (in an equimolar amount to BzBr) which led to
the isolation of the spin adducts Bz–DMPO (30%) and Bz2 (42%).
These were separated by column chromatography (solvent:
CH2Cl2) and identified by mass spectrometry, 1H NMR spec-
troscopy and by comparison with literature data.[53]
cause catalyst poisoning. Moreover, the catalytic generation
of free radicals has been performed in good yields and the
electrochemical methods may represent interesting alterna-
tives to other stoichiometric reagents such as R3SnH which
may exhibit some undesired toxicity. Applications of such
species in organic syntheses, especially those relative to C–
X bond activation, are many. Due to the nature of the radi-
cal-type reactions that these species lead to, many other ap-
plications should be possible.
Experimental Section
Materials: The [Pd3(dppm)3(CO)](PF6)2 complex was prepared ac-
cording to the literature.[20,51] Dichloromethane was distilled under
Ar over P2O5 and tetrahydrofuran (THF) was distilled under Ar
over Na (Aldrich). Benzyl bromide (BzBr), methyl iodide (MeI)
and the spin trap agents TEMPO (2,2,6,6-tetramethylpiperidin-1-
oxyl) and DMPO (5,5Ј-dimethyl-1-pyrroline N-oxide) were pur-
chased from Aldrich and used as received. Benzyl iodide (BzI) was
prepared according to a literature method.[52] The Bu4NPF6 salt
2+
High-Scale Electrolysis. The Copper Method: Pd3
cluster
(11.6 mg, 6.5×10–3 mmol, 4 mol-% per mol of BzBr) and BzBr
(20 μL, 0.168 mmol) were added to the anodic compartment con-
taining a 0.2 m solution of Bu4NPF6 in CH2Cl2 (10 mL ). The
Bu4NPF6–CH2Cl2 solution (10 mL) was poured into the reference
electrode compartment. The copper anode potential was set to
+0.6 V vs. SCE and the electrolysis was stopped after one Faraday
1016
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2005, 1011–1018

Reactivity of Pd3(dppm)3COn+ Clusters (n = 0, 1) Towards Halocarbons
FULL PAPER
per mol of BzBr had passed. The quantification procedure was the
same as described above.
The minor component of the disorder was isotropically refined
with some least-squares restraints to ensure equivalent geometrical
parameters for both components. Among the three toluene solvate
molecules, only one was anisotropically refined, the others were
each refined with an overall isotropic thermal parameter and one
of them, close to an inversion centre, was refined with an occu-
pancy of 0.5. All other non-hydrogen atoms were anisotropically
refined and the hydrogen atoms were included using a riding
model. Crystallographic data are reported in Table 4.
Theoretical Computations: The Density Functional Theory calcula-
tions (DFT) were performed using the commercially available
Gaussian 98[54] software and a Pentium 4 (1.6 GHz) computer. The
computations were performed at the RB3LYP level [Becke’s three
parameter hybrid functional[55] using the LYP (Lee–Yang–Parr)
correlation functional].[56,57] The basis set for the polarisation and
diffuse function was the 3-21G* set.[58–60] To save computation
time, H2PCH2PH2 was selected instead of dppm. In order to test
the methodology, the computed bond distances and angles of the
model cluster Pd3(H2PCH2PH2)3(CO)(Br)+ were checked against
those of the crystal structure of Pd3(dppm)3(CO)(Br)+ [denoted as
Pd3(Br)+ in the text]. The comparison was satisfactory.
CCDC-243382 contains the supplementary crystallographic data
for this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
X-ray Structure of [Pd3(dppm)3(CO)Br](PF6): Single crystals of
Acknowledgments
–
Pd3Br+ (as its PF6 salt) were obtained from acetone/toluene/hex-
ane, mounted in inert oil and transferred to the cold gas stream
(110 K) of a Nonius Kappa CCD diffractometer located at the
Université de Bourgogne. The structure was solved by the heavy
atoms method (SHELXS-97) and refined by full-matrix least-
squares methods (SHELXL-97)[61] with the aid of the WINGX
program.[62] The capping CO molecule and the capping Br atom
were found to be slightly disordered (inversion of both ligands with
respect to the Pd3 plane), the occupancies converged to 0.93/0.07.
P. D. H. thanks the NSERC (Natural Sciences and Engineering Re-
search Council) for funding. Y. M. is grateful to the CNRS (Centre
National de la Recherche Scientifique) and the Conseil Régional
de Bourgogne for funding.
[1] Comprehensive Organometallic Chemistry II: A Review of the
Literature 1982–1994 (Eds.: E. W. Abel, F. G. A. Stone, G. Wil-
kinson), 14-volume set, 1995.
[2] R. J. Kulawiec, R. H. Crabtree, Coord. Chem. Rev. 1990, 99,
Table 4. Crystallographic data.
89–115.
[Pd3(dppm)3(CO)Br]PF6·
2.5toluene
[3] J. K. Kochi, Organometallic Mechanisms and Catalysis; Aca-
demic Press, New York, 1978.
[4] J. K. Stille, K. S. Y. Lau, Acc. Chem. Res. 1977, 10, 434–442.
[5] J. P. Collman, L. S. Hegedus, J. R. Norton, R. G. Finke, Or-
ganometallic Chemistry of Transition Metals: Principles and
Use, Part 2, University Science Books, Mill Valley, 1989.
[6] Applied Homogeneous Catalysis with Organometallic Com-
pounds: Developments, 2nd ed., vols. 2 and 3 (Eds.: B. Cornils,
W. A. Herrmann), Wiley-VCH, Weinheim, Germany, 2002.
[7] J. Tsuji, Transition Metal Reagents and Catalysts: Innovations
in Organic Synthesis; Wiley: Chichester, U.K., 2000.
[8] P. D. Harvey, Y. Mugnier, D. Lucas, D. Evrard, F. Lemaitre, A.
Vallat, J. Clust. Sci. 2004, 15, 63–90.
[9] P. D. Harvey, J. Clust. Sci. 1993, 4, 377–402.
[10] R. J. Puddephatt, L. Manojlovic-Muir, K. W. Muir, Polyhedron
1990, 9, 2767–2802.
[11] D. Brevet, D. Lucas, H. Cattey, F. Lemaitre, Y. Mugnier, P. D.
Harvey, J. Am. Chem. Soc. 2001, 123, 4340–4341.
[12] D. Brevet, Y. Mugnier, F. Lemaitre, D. Lucas, S. Samreth, P. D.
Harvey, Inorg. Chem. 2003, 42, 4909–4917.
Empirical formula
Fw
C76H66BrF6OP7Pd3·2.5C7H8
1955.52
red prism
0.3×0.1×0.1
110(2)
monoclinic
P21/n
10.6546(2)
37.1091(7)
21.2714(7)
90
91.55(1)
90
8407.3(4)
4
1.545
ω and φ
3948
1.305
2.96–23.77
–11 Յ h Յ 11
–42 Յ k Յ 32
–24 Յ l Յ 24
15640
Description
Crystal size [mm]
Temperature [K]
Crystal system
Space group
a [Å]
b [Å]
c [Å]
α [°]
β [°]
γ [°]
V [Å3]
Z
ρ(calcd.) [gcm–3]
Scan mode
F(000)
[13] F. Lemaitre, D. Lucas, Y. Mugnier, P. D. Harvey, J. Org. Chem.
2002, 67, 7537–7540.
[14] F. Lemaitre, D. Lucas, K. Groison, P. Richard, Y. Mugnier,
P. D. Harvey, J. Am. Chem. Soc. 2003, 125, 5511–5522.
[15] R. van Asselt, K. Vrieze, C. J. Elsevier, J. Organomet. Chem.
1994, 480, 27–40.
μ(Mo-Kα) [mm–1]
θ-range [°]
Limiting indices
Reflections collected
[16] P. K. Byers, A. J. Canty, B. W. Skelton, A. H. White, J. Chem.
Soc. Chem. Commun. 1986, 1722–1724.
Independent reflections
Data/restraints/parameters
Goodness-of-fit on F2
Final R indices [I Ͼ 2σ(I)]
11002
11002/25/949
[17] R. S. Paonessa, A. L. Prignano, W. C. Trogler, Organometallics
1985, 4, 647–657.
1.038[c]
[a]
R1 = 0.0581
[18] K. J. Klabunde, J. S. Roberts, J. Organomet. Chem. 1977, 137,
[b]
wR2 = 0.1325
113–130.
[a]
R indices (all data)
R1 = 0.0816
[19] I. Gauthron, Y. Mugnier, K. Hierso, P. D. Harvey, Can. J.
Chem. 1997, 75, 1182–1187.
[b]
wR2 = 0.1479
Largest diff. peak
1.14
–1.12
[20] L. Manojlovic-Muir, K. W. Muir, B. R. Lloyd, R. J. Pudde-
phatt, J. Chem. Soc. Chem. Commun. 1983, 1336–1337.
–3
and hole (eÅ
)
[a] R1 = ∑||Fo| – |Fc||/ ∑ |Fo |. [b] wR2 = (∑[w(Fo – Fc ) ]/∑[Fo4])1/ [21] B. R. Lloyd, L. Manojlovic-Muir, K. W. Muir, R. J. Pudde-
2
2 2
2. [c] Goodness of fit = [Σw(Fo – Fc ) /(No – Nv)]1/2. Weight for
phatt, Organometallics 1993, 12, 1231–1237.
[22] F. Lemaitre, D. Brevet, D. Lucas, A. Vallat, Y. Mugnier, P. D.
Harvey, Inorg. Chem. 2002, 41, 2368–2373.
2
2 2
[Pd3(dppm)3(CO)Br]PF6·2.5toluene = 1/[σ2(Fo2) + (0.051×P)2
+
54.0×P] where P = [Max(Fo2,0) + 2×Fc ]/3.
2
Eur. J. Inorg. Chem. 2005, 1011–1018
www.eurjic.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1017

D. Lucas, P. D. Harvey et al.
FULL PAPER
[23] PhMe was not quantified because its peak was not resolved
from that of the solvent in the chromatogram.
[24] T. Nishino, T. Watanabe, M. Okada, Y. Nishiyama, N. Sonoda,
J. Org. Chem. 2002, 67, 966–969.
[25] EPR hyperfine splitting for the benzyl spin adduct: found
(THF): aβH = 20.47 G, aN= 14.12 G; literature values refs.[26,27]
[26] E. G. Janzen, J. I. P. Liu, J. Magn. Reson. 1973, 9, 510–512.
[27] E. G. Janzen, Y.-K. Zhang, D. L. Haire, J. Am. Chem. Soc.
1994, 116, 3738–3743.
[46] E. G. Rozantsev, Free Nitroxyl Radicals, Plenum Press, New
York, 1970.
[47] Molecular Biology: Spin Labelling. Theory and Applications
(Ed.: L. J. Berliner), Academic Press, New York, 1976.
[48] S. Yamago, H. Miyazoe, J.-I. Yoshida, Tetrahedron Lett. 1999,
40, 2339–2342.
[49] P. D. Harvey, R. Provencher, J. Gagnon, T. Zhang, D. Fortin,
K. Hierso, M. Drouin, S. M. Socol, Can. J. Chem. 1996, 74,
2268–2278.
[28] E1/2,ox = –0.29 V and E1/2,red = –2.5 V vs. SCE and the outer-
sphere mode is not feasible at the limit of –0.3 V.[29]
[29] O. N. Efimov, V. V. Strelets, Coord. Chem. Rev. 1990, 99, 15–
53.
[50] L. Manojlovic-Muir, K. W. Muir, B. R. Lloyd, R. J. Pudde-
phatt, J. Chem. Soc., Chem. Commun. 1985, 536–537.
[51] B. R. Lloyd, R. J. Puddephatt, Inorg. Chim. Acta 1984, 90, 77–
78.
[52] A. Kowalczuk-Bleja, B. Trzebicka, H. Komber, B. Voit, A.
Dworak, Polymer 2004, 45, 9–18.
[30] D. Evrard, M. Drouin, Y. Mugnier, P. D. Harvey, Inorg. Chim.
Acta 2003, 350, 442–448.
[31] Attempts to prepare Pd3(R)+ from Pd3 and n-butyllithium
[53] V. F. Patel, G. Pattenden, D. M. Thompson, J. Chem. Soc. Per-
kin Trans. 2 1990, 2729–2734.
2+
failed, leading only to cluster dissociation.
[32] P. D. Harvey, S. M. Hubig, T. Ziegler, Inorg. Chem. 1994, 33,
3700–3710.
[33] P. D. Harvey, R. Provencher, Inorg. Chem. 1993, 32, 61–65.
[34] I. Gauthron, Y. Mugnier, K. Hierso, P. D. Harvey, New J.
Chem. 1998, 22, 237–246.
[35] This reactive complex has not been isolated so far but the re-
lated and crystallographically characterised Pd2(iPr2PCH2-
PiPr2)2 complex has been known for ten years.[36]
[36] A. Doehring, R. Goddard, G. Hopp, P. W. Jolly, N. Kokel, C.
Krueger, Inorg. Chim. Acta 1994, 222, 179–192.
[37] A. L. Balch, C. T. Hunt, C.-L. Lee, M. M. Olmstead, J. P. Farr,
J. Am. Chem. Soc. 1981, 103, 3764–3772.
[38] R. U. Kirss, R. Eisenberg, Inorg. Chem. 1989, 28, 3372–3378.
[39] R. A. Stockland Jr., G. K. Anderson, N. P. Rath, Organomet-
allics 1997, 16, 5096–5101.
[54] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A. Jr.
Montgomery, R. E. Stratmann, J. C. Burant, S. Dapprich, J. M.
Millam, A. D. Daniels, K. N. Kudin, M. C. Strain, O. Farkas,
J. Tomasi, V. Barone, M. Cossi, R. Cammi, B. Mennucci, C.
Pomelli, C. Adamo, S. Clifford, J. Ochterski, G. A. Petersson,
P. Y. Ayala, Q. Cui, K. Morokuma, D. K. Malick, A. D. Rab-
uck, K. Raghavachari, J. B. Foresman, J. Cioslowski, J. V. Or-
tiz, A. G. Baboul, B. B. Stefanov, G. Liu, A. Liashenko, P. Pis-
korz, I. Komaromi, R. Gomperts, R. L. Martin, D. J. Fox, T.
Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, C. Gon-
zalez, M. Challacombe, P. M. W. Gill, B. G. Johnson, W. Chen,
M. W. Wong, J. L. Andres, M. Head-Gordon, E. S. Replogle,
J. A. Pople, Gaussian 98 (Revision A.6), Gaussian, Inc., Pitts-
burgh PA, 1998.
[40] W. B. Motherwell, D. Crich, Free Radical Chain in Organic
Synthesis, Academic, London, 1992.
[55] A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652.
[56] C. Lee, W. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785–789.
[57] A. D. Miehlich, A. Savin, H. Stoll, H. Preuss, Chem. Phys.
Lett. 1989, 157, 200.
[41] C. P. Jasperse, D. P. Curran, T. L. Fevig, Chem. Rev. 1991, 91,
1237–1286.
[42] A. G. Davies, Organotin Chemistry, Wiley-VCH, Weinheim,
1997.
[58] K. D. Dobbs, W. J. Hehre, J. Comput. Chem. 1986, 7, 359–378.
[59] K. D. Dobbs, W. J. Hehre, J. Comput. Chem. 1987, 8, 880–893.
[60] K. D. Dobbs, W. J. Hehre, J. Comput. Chem. 1987, 8, 861–879.
[61] G. M. Sheldrick, SHELX-97 (Includes SHELXS-97,
SHELXL-97), Release 97–2, Programs for Crystal Structure
Analysis, University of Göttingen, Germany, 1998.
[62] L. J. Farrugia, J. Appl. Crystallogr. 1999, 32, 837.
Received: July 9, 2004
[43] P. A. Baguley, J. C. Walton, Angew. Chem. Int. Ed. 1998, 37,
3072–3082.
[44] E. Dunach, D. Franco, S. Olivero, Eur. J. Org. Chem. 2003,
1605–1622.
[45] A. R. Forrester, J. M. Hay, R. H. Thomson, Organic Chemistry
of Stable Free Radicals, Academic Press, New York, 1968.
1018
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2005, 1011–1018