Redox reactions in palladium catalysis: On the accelerating and/or inhibiting effects of copper and silver salt additives in cross-coupling chemistry involving electron-rich phosphine ligands

Article abstract of DOI:10.1002/anie.201202504

Challenging a catalytic cycle: Pd⁰ catalysts are readily oxidized by Cu and Ag salts to give dinuclear Pd^I complexes and Cu^I or Ag^I cubanes (see scheme). The reactivities of the resulting Pd^I dimers are consistent with several observations of additive effects in cross-coupling chemistry. The results indicate the possibility for alternative catalytic cycles involving dinuclear Pd^I complexes over the currently accepted synergistic cycles involving Pd ⁰/Pd^II intermediates and Cu or Ag. Copyright

Full text of DOI:10.1002/anie.201202504

Angewandte
Chemie
DOI: 10.1002/anie.201202504
Catalysis
Redox Reactions in Palladium Catalysis: On the Accelerating and/or
Inhibiting Effects of Copper and Silver Salt Additives in Cross-
Coupling Chemistry Involving Electron-rich Phosphine Ligands**
Marialuisa Aufiero, Fabien Proutiere, and Franziska Schoenebeck*
Palladium-catalyzed cross-coupling reactions are widely used
alternative, Cu-free variants of the Sonogashira transforma-
tion, or processes involving Ag salts instead,[10] were sub-
sequently developed in recent years.[11]
[1]
ꢀ
ꢀ
to construct carbon carbon or carbon heteroatom bonds.
These transformations are frequently assisted by more than
one metal, and Cu/Pd is probably one of the most frequently
applied combinations, for example, in the Sonogashira and
Stille reactions.[2] The precise effects of Cu salts in these
transformations are not fully understood, but CuI is thought to
play a catalytic, accelerating role in the transfer of an alkynyl
group to Pd in the Sonogashira coupling (through formation
We herein report our observations of redox transforma-
tions of Pd0 catalysts to dinuclear PdI complexes in the
presence of Cu and Ag salts, and address the origin of
enhancement and/or inhibition of cross-coupling reactivity in
the presence of oxidizing salts. Our results suggest the
possibility of alternative cross-coupling cycles involving
bimetallic PdI over the currently accepted co-existing, syner-
gistic cycles involving Pd0/PdII and Cu (as shown in
Scheme 1).
ꢁ
of an organocopper species, [Cu(C CR)], which in turn is
more readily transferred to PdII, Scheme 1).[3] Other reports
We recently reported our study on the reactivity and
stability of the dinuclear PdI complex [{(PtBu3)PdBr}2]
(Scheme 2).[12,13] When this complex was used as precatalyst
Scheme 2. [{(PtBu3)PdIBr}2] activation.
Scheme 1. Generally accepted mechanism of Sonogashira coupling.
in Suzuki cross-coupling reactions, high reactivities were
observed, leading to quantitative conversion of aryl bromides
in as few as 15 minutes.[14,12] Aryl chlorides also reacted
rapidly under PdI-dimer conditions (ca. 80% conversion in
20 min), but the catalytic activity dropped quickly in these
reactions, and full conversion was not achieved as a conse-
quence of a competing precatalyst deactivation process.[12]
Our previous investigations showed that the reactivity
derived from this complex with aryl chlorides is consistent
with highly reactive monoligated Pd0PtBu3 as active catalytic
species in THF,[12] and the PdI dimer liberates the active
species more readily than Pd0(PtBu3)2, accounting for the
faster rates under PdI-dimer conditions[12,13c,14] (Scheme 2).[12]
Following these studies, we recently began to explore the
redox activity of [Pd0(PtBu3)2] against typical oxidizing agents
to study the propensity to form the dinuclear PdI complex in
the inverse process. We found that upon mixing one
equivalent of CuIIBr2 with one equivalent of [Pd(PtBu3)2],
the dinuclear complex [{(PtBu3)PdIBr}2] 1 was formed essen-
tially instantaneously, and according to 31P NMR spectrosco-
py the reaction was complete after 15 minutes at room
temperature (Figure 1).[15] Thus, the CuII salt triggered a one-
electron oxidation of Pd0 to give the dinuclear PdI complex 1.
CuII was reduced to CuI in the same process. This latter event
suggest the facilitation of transmetalation by CuI salts via
organocopper species in the Stille and Suzuki reactions
also.[4,5] On the other hand, a ligand-scavenging effect was
also ascribed to CuI.[6,7] Another possibility of Cu effect was
added by Chen and co-workers recently, who reported
a transmetalation of a methyl group at platinum, involving
a PtII-CuI bimetallic species.[8] In contrast to these “accelerat-
ing” effects of Cu salts, Buchwald and co-workers also
reported an inhibitory effect of Cu salts in Sonogashira
cross-coupling reactions of aryl chlorides.[9] A number of
[*] M. Aufiero, F. Proutiere, Prof. Dr. F. Schoenebeck
Laboratory for Organic Chemistry, ETH Zꢀrich
Wolfgang-Pauli-Strasse 10, 8093 Zꢀrich (Switzerland)
E-mail: schoenebeck@org.chem.ethz.ch
Homepage: http://www.schoenebeck.ethz.ch/
[**] We thank the Swiss National Science Foundation and ETH Zꢀrich
for financial support. Calculations were performed on the ETH
High-Performance Cluster Brutus. We thank Dr. W. B. Schweizer for
X-ray crystal structure analyses and Prof. Peter Chen for helpful
discussions.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201202504.
Angew. Chem. Int. Ed. 2012, 51, 1 – 6
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
therefore be the origin of some of the frequently observed
accelerating effects of Cu additives in catalysis. To address
this issue, we performed reactivity studies and used the Suzuki
cross-coupling of aryl chloride 3 as our benchmark reaction
(Scheme 3). Our reactivity survey of dimer 1 versus [Pd-
(PtBu3)2] in the presence and absence of CuBrn (n = 1 or 2)
Scheme 3. Reactivity comparison of Pd/Cu system (see Table 1).
Figure 1. Oxidation of [Pd(PtBu3)2] by CuBr2 and X-ray structure of Cu
cubane 2.[16] After 15 minutes, [Pd(PtBu3)2] was fully converted to 1.
indeed shows increased reactivity under conditions that
would favor formation of PdI dimer, that is, higher conver-
sions are observed in the presence of CuBrn relative to
[Pd(PtBu3)2] in one hour (Table 1). To establish that the PdI
dimer 1 is also formed under the applied reaction conditions,
we mixed arylboronic acid (1 equiv), KF (3 equiv), [Pd-
(PtBu3)2] (5 mol%) and CuBr (5 mol%) in THF. Examina-
tion of the mixture by 31P NMR spectroscopy indeed showed
the 31P NMR signal of PdI dimer 1 after five minutes mixing
time.
became apparent upon crystallization of the product mixture.
We identified two different kinds of crystals. One type
corresponded to the PdI dimer crystals, which are dark-
green in color. The other type was white in color and was
identified as the Cu cubane complex 2[16] (see Figure 1 for the
crystal structure). The Cu cubane 2[16] gives a broad peak of
low intensity at 55–59 ppm in the 31P NMR spectrum in THF
at room temperature, and showed relatively high stability.[17]
Our DFT calculations[18] predict a driving force (DG) of
ꢀ5.4 kcalmolꢀ1 for monomeric “CuBrPtBu3” to assemble as
cubane 2 in THF.
Table 1: Comparison of the effects of the [Pd]/Cu system on the reactions
in Scheme 3.
Entry
[Pd]/Cu system[a]
Yield
Recov.
4 [%]
3 [%]
Having observed this very facile oxidation of [Pd-
(PtBu3)2], we turned our attention to CuI salts, as those are
more frequently employed in cross-coupling reactions. We
once again observed the formation of the dinuclear PdI
complex 1 upon employing one equivalent of CuBr along
with [Pd(PtBu3)2] (Figure 2). The PdI dimer 1 is already
detected after a reaction time of 15 minutes. We hypothesize
that the oxidation occurs as a consequence of the tendency of
CuI to disproportionate to Cu0 and CuII. The thus generated
CuII then triggers rapid oxidation of Pd0 to dinuclear complex
1.
1
2
3
4
5
6
7
[{(PtBu3)PdBr}2] 1 (2.5%)
[Pd(PtBu3)2] (5%)/CuBr2 (5%)
[Pd(PtBu3)2] (5%)/CuBr (5%)
[Pd(PtBu3)2] (5%)
87
52
48
11
23
0
10
36
49
84
75
99
62
[Pd2dba3] (2.5%)/PtBu3 (5%)
cubane 2 (2.5%)
PdI 1 (2.5%)/cubane 2 (1.25%)
36
[a] Reaction conditions: KF (3.0 equiv), THF, RT, reaction time: 1 h.
Previous observations of reactivity enhancing effects of
CuI salts in Suzuki reactions with ArBr or ArI were presumed
to be a result of the formation of an organocopper species that
would facilitate the transmetalation step.[5] However, for the
coupling of aryl chlorides with [Pd0L2] (in which L is a bulky
ligand, such as PtBu3), the initial ligand dissociation step
([PdL2]![PdL] + L) to form the monoligated, highly active
Pd species ([PdL]) is considered to be the reactivity limiting
step.[19] Thus, facilitation of the transmetalation step (via an
organocopper species) should be mostly irrelevant for the
overall reaction rate, as it would not constitute the slowest
step in the catalytic cycle. However, our finding that a change
of the precatalyst from [Pd(PtBu3)2] to the more active
precatalyst [{(PtBu3)PdIBr}2] takes place (which in turn
liberates [PdL] more readily), is fully consistent with the
observed reactivity enhancement.
The use of dinuclear complex [{(PtBu3)PdIBr}2] (1) leads
to much higher rates of cross-coupling reactions with aryl
chlorides than [Pd(PtBu3)2].[12,14] The formation of 1 might
Considering the rapid formation of PdI dimer in the
presence of Cu salts, we wondered why we did not observe as
high conversions as under PdI dimer catalysis itself (compare
entries 1–3 in Table 1: ꢂ 50% vs. 87% conversion). For
Figure 2. 31P NMR analysis of the formation of 1 in the presence of
one equivalent of CuBr (relative to [Pd(PtBu3)2]) versus internal
standard (Me3PO4) in THF.
2
www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
efficient catalysis, the presence of PdI dimer at ppm levels
should be sufficient, and given that the dimer can be very
rapidly observed macroscopically by 31P NMR spectroscopy
(Figures 1 and 2), we questioned whether Cu cubane 2, which
is formed upon oxidation of [Pd(PtBu3)2] and co-exists in
solution, might influence the overall reactivity. We first
established that cubane 2 does not trigger the Suzuki coupling
reaction (Table 1, entry 6: no conversion was observed).
When we added cubane 2 to PdI dimer 1 and then studied the
transformation shown in Scheme 3, we observed lower
conversion than under PdI dimer catalysis in the absence of
cubane (compare entries 7 and 1). Through kinetic studies
(Figure 3), we established further that the reactivity of the PdI
dimer 1 in the presence of Cu cubane 2 (red) is particularly
toward small molecules, such as H2, than PdI–Br dimer 1;[13b]
however, information on the reactivity of 5 in cross-coupling
reactions is sparse.[20] We set out to explore the reactivity of
PdI–I dimer 5 in the identical Suzuki coupling of 4-chloroaryl
ketone 3 (Scheme 3 and Table 2). Essentially, PdI–I dimer 5
showed similar reactivity as [Pd(PtBu3)2], but much lower
reactivity than PdI–Br dimer 1. While the PdI–Br dimer
Table 2: Comparison of the effects of the [Pd]/Cu system on the reaction
in Scheme 3.
Entry
[Pd]/Cu system[a]
t [h]
Yield
Recov.
4 [%]
3 [%]
1
2
3
4
5
6
[{(PtBu3)PdI}2] 5 (2.5%)
[Pd(PtBu3)2] (5%)/CuI (5%)
[Pd(PtBu3)2] (5%)
[{(PtBu3)PdI}2] 5 (2.5%)
[Pd(PtBu3)2] (5%)/CuI (5%)
[Pd(PtBu3)2] (5%)
24
24
24
1
1
1
45
29
38
0
5
11
53
58
59
99
93
84
[a] Reaction conditions: KF (3.0 equiv), THF, RT.
results in nearly full conversion after one hour at room
temperature (87%, Table 1), the iodo analogue 5 leads to no
conversion at all in the same time. Even after 24 hours
reaction time, only moderate conversion was achieved (45%,
Table 2), similar to the reactivity with [Pd(PtBu3)2]. We
presume that the reason for the lower reactivity of the PdI–I
dimer 5 compared to its Br analogue 1 is the lower propensity
of the latter to form the active catalytic species (i.e.,
[PdPtBu3]) as a consequence of its higher stability. We are
currently exploring the precise pathway for precatalyst
activation. Our preliminary calculations of the driving force
for 1) the dissociation of the homolytic dimer to [PdIPtBu3I]
radicals, and 2) disproportionation of the complex to
[Pd0PtBu3] and [PdIII2PtBu3] emphasize the stability of the
PdI dimer 5 (26.8 kcalmolꢀ1 were calculated for the former
and 40.2 kcalmolꢀ1 for the latter process).[18,21] These results
suggest that CuBrx has a reactivity-enhancing effect in
reactions with aryl chlorides in Suzuki coupling because of
the formation of PdI–Br dimer 1, whereas CuI has an
inhibitory effect because of the formation of PdI–I dimer 5,
which in turn shows low reactivity in cross-coupling reactions.
We next considered the effect of AgBr. Ag salts have
recently been identified as proficient alternatives to Cu in
Sonogashira and Suzuki cross-coupling reactions.[10] With
AgBr as additive, we also found an acceleration of the Suzuki
coupling compared with additive-free [Pd(PtBu3)2] catalysis,
giving 53% conversion after one hour for the transformation
shown in Scheme 3. With 31P NMR investigations of a 1:1
mixture of AgBr and [Pd(PtBu3)2] in THF, we observed also
formation of PdI–Br dimer 1 within one hour at room
temperature (see the Supporting Information for the
31P NMR spectra, pages S12–S15). In this process, an analo-
gous Ag cubane 6 is formed, which we identified by X-ray
crystal structure analysis (Figure 4).[22] The latter is the first
crystallographic evidence that such a structure, involving the
bulky ligand PtBu3, is stable.
Figure 3. Kinetic study of reaction in Scheme 3. The conversion (Y) of
3 in the presence of different catalysts is plotted.
inhibited in the first 15 minutes, that is, the time window when
the PdI dimer usually shows greatest activity (blue). Separate
31P NMR investigations showed that the PdI dimer is stable in
the presence of CuI cubane for the time that we monitored the
mixture (4 h). We speculate that the transformation of the PdI
dimer to the active catalytic species is inhibited by the Cu
cubane, therefore leading to a lower reaction rate.
Since CuI is much more frequently used in cross-coupling
reactions than CuBr, we next explored the effect of CuI.
Upon mixing of one equivalent of CuI and one equivalent of
[Pd(PtBu3)2], we observed the analogous results, that is, the
dinuclear PdI–I complex [{(PtBu3)PdI}2] 5[13b] was formed (see
the Supporting Information, page S9). Through in situ
31P NMR analysis, we were also able to show that PdI–I
dimer forms rapidly also under Suzuki cross-coupling con-
ditions (see pages S16–S17 in the Supporting Information).
The PdI–I dimer 5 was previously studied by Mingos,
Vilar, and co-workers, and was found to be less reactive
To this point, we only discussed the effect of additives on
Suzuki cross-coupling reactions. What are the effects of
Angew. Chem. Int. Ed. 2012, 51, 1 – 6
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
dation of these species. The preliminary investigations
indicate however that the discussed oxidation might be
general in Pd catalysis across a range of electron-rich ligands.
Our results might therefore also be of relevance for other
types of coupling reactions, such as direct arylation reactions
[24]
(C H functionalization) or Wacker processes.[25] For these
ꢀ
processes, the formation of PdII species after oxidation of Pd0
by Cu/Ag salts, but not the formation of dinuclear PdI
complexes, is mechanistically assumed.
In conclusion, we have shown that oxidation of Pd0
catalysts to dinuclear PdI complexes takes place in the
presence of Cu and Ag salts. In the same process, CuI or
AgI cubanes are formed, which in turn effect the activity of
the dinuclear PdI complexes. The PdI complexes react readily
with alkynes and initiate polymerization reactions, thus
accounting for the inhibitory effects of Cu salts in Sonogashira
reactions with ArCl. For Suzuki cross-coupling reactions with
ArCl, the effect of oxidants depends on the anion. If X = I
(i.e., CuI), the dinuclear PdI–I complex 5 is formed, which
causes moderate transformations in cross-coupling as a con-
sequence of its high stability and low propensity to form the
active species [Pd0PtBu3]. If X = Br, a more active precatalyst
(1) results, which liberates [Pd0PtBu3] more readily and hence
leads to much higher reaction rates than [PdL2]. Our results
suggest the possibility of alternative cycles involving bimet-
allic PdI [26] over the currently accepted co-existing, synergistic
catalytic cycles involving Pd0/PdII and Cu. In our future
studies, we will focus on the latter and also expand our
investigations to alternative ligands and additives (e.g., with
noncoordinating counterions or anions that are less common
as bridging halides).
Figure 4. Reaction of [Pd(PtBu3)2] with AgBr (1:1) and X-ray structure
of Ag cubane 6.[22]
precatalyst transformation for the Sonogashira reaction?
Buchwald and co-workers reported, for example, that CuI
would inhibit the Sonogashira coupling of aryl chlorides, as
the alkyne cross-coupling partner was consumed in a side
reaction.[9] The oxidation of [PdL2] by CuX to the PdI complex
would be completely consistent with these observations. PdI–I
dimer 5 and PdI–Br dimer 1 react with alkynes to give
polymerization side products.[13b] When we exposed phenyl-
acetylene to PdI dimers 1 and 5 along with diisopropylamine
as base (at 508C in dioxane), we found that the alkyne was
nearly fully consumed within one hour with dimer 1 (3%
alkyne left; with dimer 5: 30% left).[23] Thus, oxidation of the
[PdL2] to PdI dimer opens an alternative reaction pathway for
alkyne polymerization, which competes with the “standard”
cross-coupling pathway, and for more challenging substrates
(such as ArCl, for which the oxidative addition is slower than
for, e.g., ArBr) the side reaction wins over the cross-coupling
process (Scheme 4).
Received: March 30, 2012
Published online: && &&, &&&&
Keywords: copper · redox reactions · silver · Sonogashira ·
.
Suzuki
[1] a) Handbook of Organopalladium Chemistry for Organic Syn-
thesis (Eds.: E.-I. Negishi, A. de Meijere), Wiley, New York,
2002; b) Metal-Catalyzed Cross-Coupling Reactions, 2nd ed.
(Eds.: F. Diederich, A. de Meijere), Wiley-VCH, Weinheim,
2004.
[2] For recent reviews on bimetallic catalysis, see: a) M. H. Pꢀrez-
Temprano, J. A. Casares, P. Espinet, Chem. Eur. J. 2012, 18, 1864;
b) A. E. Allen, D. W. C. MacMillan, Chem. Sci. 2012, DOI: 10.
1039/c2sc00907b.
[3] a) N. J. Long, C. K. Williams, Angew. Chem. 2003, 115, 2690;
Angew. Chem. Int. Ed. 2003, 42, 2586; b) Y. Fujikura, K.
Sonogashira, N. Hagihara, Chem. Lett. 1975, 1067; c) H.
Ogawa, T. Joh, S. Takahashi, K. Sonogashira, J. Chem. Soc.
Chem. Commun. 1988, 561; d) K. Osakada, R. Sakata, T.
Yamamoto, Organometallics 1997, 16, 5354; e) K. Osakada, T.
Yamamoto, Coord. Chem. Rev. 2000, 198, 379.
[4] P. Espinet, A. M. Echavarren, Angew. Chem. 2004, 116, 4808;
Angew. Chem. Int. Ed. 2004, 43, 4704.
[5] a) J. Z. Deng, D. V. Paone, A. T. Ginnetti, H. Kurihara, S. D.
Dreher, S. A. Weissman, S. R. Stauffer, C. S. Burgey, Org. Lett.
2009, 11, 345; b) L. S. Liebeskind, J. Srogl, J. Am. Chem. Soc.
2000, 122, 11260.
Scheme 4. Polymerization side reaction in Sonogashira coupling as
a consequence of PdI dimer formation.
Lastly, we also performed analogous investigations with
alternative ligands, such as PCy3 and (2-dicyclohexylphos-
phino)-biphenyl phosphine ligands. Our 31P NMR analyses
showed rapid formation of new species in the 31P NMR
spectrum upon mixing the corresponding [PdL2] with CuI or
CuBrx (see pages S18–S25 in the Supporting Information).
Our future research is directed toward the structural eluci-
4
www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
[6] a) V. Farina, S. Kapadia, B. Krishnan, C. Wang, L. S. Liebeskind,
J. Org. Chem. 1994, 59, 5905.
[7] A. L. Casado, P. Espinet, Organometallics 2003, 22, 1305.
[8]
M.-E. Moret, D. Serra, A. Bach, P. Chen, Angew. Chem. 2010,
122, 2935; Angew. Chem. Int. Ed. 2010, 49, 2873.
[9] D. Gelman, S. L. Buchwald, Angew. Chem. 2003, 115, 6175;
Angew. Chem. Int. Ed. 2003, 42, 5993.
[10] J.-M. Weibel, A. Blanc, P. Pale in Silver in Organic Chemistry
(Ed.: M. Harmata), Wiley, Hoboken, 2010, p. 285. For an early
example, see: J. Uenishi, J.-M. Beau, R. W. Armstrong, Y. Kishi,
J. Am. Chem. Soc. 1987, 109, 4756.
[11] For selected examples, see: a) Y. Liang, Y.-X. Xie, J.-H. Li,
J. Org. Chem. 2006, 71, 379; b) B. H. Lipshutz, D. W. Chung, B.
Rich, Org. Lett. 2008, 10, 3793; c) A. R. Gholap, K. Venkatesan,
R. Pasricha, T. Daniel, R. J. Lahoti, K. V. Srinivasan, J. Org.
Chem. 2005, 70, 4869.
Supporting Information for details on calculation; b) M. J. Frisch
et al. Gaussian09, RevisionA.01—full reference in the Support-
ing Information.
[19] Recent computational studies suggest an activation free energy
barrier of ca. 30 kcalmolꢀ1 for an associative displacement of
PtBu3 from [PdL2]: a) C. L. McMullin, J. Jover, J. N. Harvey, N.
Fey, Dalton Trans. 2010, 39, 10833. See also: b) S. Kozuch,
J. M. L. Martin, ACS Catal. 2011, 1, 246; c) F. Proutiere, F.
Schoenebeck, Angew. Chem. 2011, 123, 8342; Angew. Chem. Int.
Ed. 2011, 50, 8192; d) F. Barrios-Landeros, B. P. Carrow, J. F.
Hartwig, J. Am. Chem. Soc. 2009, 131, 8141.
[20] A carbonylation was recently reported, catalyzed by PdI – I
dimer at high temperature on ArX (X = Br, I), see: G. Buscemi,
P. W. Miller, S. Kealey, A. D. Gee, N. J. Long, J. Passchier, R.
Vilar, Org. Biomol. Chem. 2011, 9, 3499.
[21] We calculated only 3 kcalmolꢀ1 lower dissociation energies for
the PdI–Br dimer 1 (for both pathways), thus suggesting that an
alternative activation mechanism is likely. Compare refer-
ence [12].
[12] F. Proutiere, M. Aufiero, F. Schoenebeck, J. Am. Chem. Soc.
2012, 134, 606.
[13] The complex was first synthesized by Vilar, Mingos et al.: a) R.
Vilar, D. M. P. Mingos, C. J. J. Cardin, J. Chem. Soc. Dalton
Trans. 1996, 4313; b) V. Durꢁ-Vilꢁ, D. M. P. Mingos, R. Vilar,
A. J. P. White, D. J. Williams, J. Organomet. Chem. 2000, 600,
198; for further studies, see: c) U. Christmann, R. Vilar, A. J. P.
White, D. J. Williams, Chem. Commun. 2004, 1294; d) U.
Christmann, D. A. Pantazis, J. Benet-Buchholz, J. E. McGrady,
F. Maseras, R. Vilar, J. Am. Chem. Soc. 2006, 128, 6376.
[14] a) J. P. Stambuli, R. Kuwano, J. F. Hartwig, Angew. Chem. 2002,
114, 4940; Angew. Chem. Int. Ed. 2002, 41, 4746; b) T. Hama, X.
Liu, D. A. Culkin, J. F. Hartwig, J. Am. Chem. Soc. 2003, 125,
11176; c) M. Prashad, X. Y. Mak, Y. Liu, O. Repic, J. Org. Chem.
2003, 68, 1163.
[22] Selected X-ray data for 6: C55H116Ag4Br4P4, M = 1652.48, clear
ꢀ
pale bronze prism; space group R3; a = 13.9482(4), b =
13.9482(4), c = 60.5717(13) ꢂ; V= 10205.6(5) ꢂ3; Z = 6; 1calcd
=
1.613 Mgcmꢀ3
; T= 100(2) K; reflections collected: 120779,
independent reflections: 5243 [R(int) = 0.0432], R1 = 0.0267,
wR2 = 0.0556, Flack parameter= 0.0(0). CCDC 874110 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.
[23] We also reacted phenylacetylene with [Pd(PCy3)2] in the
presence of 1) CuI or 2) CuBr2 (with base in dioxane at 508C).
Consumption of alkyne was also seen; 1% of alkyne was
recovered in (1) and 2% in (2) after one hour.
[24] a) L. Ackermann, R. Vicente, A. R. Kapdi, Angew. Chem. 2009,
121, 9976; Angew. Chem. Int. Ed. 2009, 48, 9792; b) D. Alberico,
M. E. Scott, M. Lautens, Chem. Rev. 2007, 107, 174.
[25] a) J. A. Keith, P. M. Henry, Angew. Chem. 2009, 121, 9200;
Angew. Chem. Int. Ed. 2009, 48, 9038; b) E. H. P. Tan, G. C.
Lloyd-Jones, J. N. Harvey, A. J. J. Lennox, B. M. Mills, Angew.
Chem. 2011, 123, 9776; Angew. Chem. Int. Ed. 2011, 50, 9602.
[26] Recent reports claim a direct involvement of PdI dimers and
[15] After five minutes, approximately 50% conversion to PdI dimer
1 was observed. See the Supporting Information.
[16] The cubane was previously synthesized from CuIBr and PtBu3:
R. G. Goel, A. L. Beauchamp, Inorg. Chem. 1983, 22, 395. The
propensity of CuI to form higher-ordered structures has been
noted, see: C. E. Holloway, M. Melnik, Rev. Inorg. Chem. 1995,
15, 147.
[17] We tested the stability by mixing cubane 2 with 1) arylboronic
acid in THF, 2) arylboronic acid and KF in THF, and 3) [Pd2-
(dba)3] (1:1) in THF. We did not observe any change in the
31P NMR spectrum and cubane 2 remained stable.
ꢀ
suggest a catalytic cycle in which the Pd Pd bond stays intact,
see: a) R. K. Das, B. Saha, S. M. W. Rahaman, J. K. Bera, Chem.
Eur. J. 2010, 16, 14459; b) T. Murahashi, K. Takase, M. Oka, S.
Ogoshi, J. Am. Chem. Soc. 2011, 133, 14908.
[18] a) CPCM (THF) M06L/6-31 + G(d)//B3LYP/6-31 + G(d) with
SDD (for Cu or Pd) or QM/QM’ method was applied. The
standard state was converted to 1m in solution. See also the
Angew. Chem. Int. Ed. 2012, 51, 1 – 6
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

.
Angewandte
Communications
Communications
Catalysis
M. Aufiero, F. Proutiere,
F. Schoenebeck*
&&&& — &&&&
Redox Reactions in Palladium Catalysis:
On the Accelerating and/or Inhibiting
Effects of Copper and Silver Salt Additives
in Cross-Coupling Chemistry Involving
Electron-rich Phosphine Ligands
Challenging a catalytic cycle: Pd0 cata-
lysts are readily oxidized by Cu and Ag
salts to give dinuclear PdI complexes and
CuI or AgI cubanes (see scheme). The
reactivities of the resulting PdI dimers are
consistent with several observations of
additive effects in cross-coupling
chemistry. The results indicate the possi-
bility for alternative catalytic cycles
involving dinuclear PdI complexes over
the currently accepted synergistic cycles
involving Pd0/PdII intermediates and Cu
or Ag.
6
www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 1 – 6
These are not the final page numbers!