Page 9 - Organogold Reactivity with Palladium, Nickel, and Rhodium: Transmetalation, Cross-Coupling, and Dual Catalysis

Organogold Dual Metal Catalysis Hirner et al.

second metal outcompetes the single-electron oxidation by
Au. In our nickel-catalyzed cross-coupling reaction (section
VII), 32 only electron poor aryl bromides were sufficiently fast
I
oxidants for Ni to allow for the desired reaction to occur;
electron rich aryl bromides, which undergo oxidative addition
at a lower rate, 37 resulted predominately in single-electron
reduction of the organogold(I) compounds leading to homo-
FIGURE 2. Possible transition states for the Au/Rh transmetalation
coupling of the organogold fragment.
reaction with varying coordination environments about Rh.
Recently, Gagn e reported a competitive reduction and
homocoupling of organogold compounds by palladium that
reactions with gold may be the absence of a unified design 0
provides additional insight. During an attempted Pd -cata-
strategy to overcome the challenges of compatible reactivity
lyzed cross-coupling of aryl triflates with organogold
and simultaneous catalyst turnover. We have found it help-
compounds, 57 the low-valent Pd intermediates were oxi-
ful to follow three guiding principles in our design of dual- I
dized exclusively by Au , resulting in Au mirror and the
metal reactions:
homocoupled product, possibly because the COTf oxida-
1. Gold Is a Versatile Transmetalation Partner. Gold has 41,58
tive addition was sluggish. Notably, the use of the NHC
been shown to undergo transmetalation reactions with
Ti, 18,45,46 Zr, 45 Hf, 46 V, 46 Nb, 46 Cr, 46,47 Re, 48 Fe, 18 Co, 18 ligands IPr and IMes on Au inhibited this redox reactivity,
59
Rh, 39 Ir, 49 Ni, 18,32 Pd, 1719,25 Pt, 18,50 Au, 18 Hg, 51,52 indicating that strongly donating ligands may reduce
single-electron reduction of catalytic Au in the presence of
11,53,54
13
55,56
B, Si, and Sn. We believe that future studies will
redox active metals. 60
extend this compatibility of Au to many of the remaining d
Promoting the desired oxidative-addition reaction via
block elements. The breadth of this accessible transmetala-
Lewis acid catalysis 31d,e is a successful strategy for outcom-
tion reactivity strongly suggests an abundance of potential
peting the potental undesired single-electron transfer to
methods to intercept and functionalize catalytic organogold
gold. This strategy was demonstrated in our dual-catalyzed
intermediates by harnessing the unique reactivity of many
preparation of butenolides and isocoumarins (section V),
different transition metals.
wherein the gold Lewis acid lowered the barrier for palla-
2. Ligands Must Be Chosen Judiciously. The weakly
coordinating counterions typically required for gold(I) cata- dium-catalyzed deallylation/oxidative addition.
lysis (e.g., OTf, SbF 6 ) render the catalyst subject to deactiva-
X. Conclusions and Outlook
tion by labile phosphine ligands on the second metal
through the formation of the bisphosphine Au cation (e.g., In the past five years, our group has developed the compat-
þ
[Au(PPh 3 ) 2 ] ). In our Au and Pd dual-catalyzed synthesis of ibility and reactivity of Au with other transition metals.
butenolides and isocoumarins (section V), 25 the olefinic Mechanistic understanding gained through the investiga-
ligand dba was found to be the optimal Pd ligand to avoid tion of the Au/Pd transmetalation allowed us to develop
Au catalyst deactivation. several Pd-catalyzed and Au/Pd dual-catalyzed reactions.
Our investigation of the RhAu dual-metal system re- The broad applicability of the Pd-catalyzed cross-coupling
vealed the need for an open coordination site on Rh prior to reaction of stoichiometric organogold compounds has since
transmetalation. We postulate that a coordinatively unsatu- been demonstrated by the groups of Hashmi, 21,61 Sarand-
rated intermediate may also be a prerequisite in the trans- eses and P erez Sestelo, 62 and Sarkar. 63 Echavarren et al.
metalation of gold with other transition metals. Therefore, have further noted the possibility that a reaction previously
ligands that favor a coordinatively saturated second transi- thought to be catalyzed only by Au is in fact cocatalyzed by
tion metal, such as chelating phosphines or nonlabile ha- trace Pd impurities. 64,65 We envision that our investigations
lides, may inhibit the interception of catalytic organogold of the Au/Ni cross-coupling and Au/Rh transmetalation
intermediates. reactions could form a similar foundation for dual-catalyzed
3. Oxidative Steps Involving the Second Metal Compete reactivity. Such functionalization methods are accessible
with Single-Electron Transfer to Au. Organogold com- even to acid-sensitive catalytic intermediates through un-
pounds are not redox innocent. Dual-metal reactions with derstanding the basicities of organogold compounds. All of
Au must be designed such that any desired reactivity of the these advances were made possible by first gaining new

Vol. 44, No. 8 ’ 2011 ’ 603–613 ’ ACCOUNTS OF CHEMICAL RESEARCH ’ 611