Observation and mechanistic study of facile C-O bond formation between a well-defined aryl-copper(III) complex and oxygen nucleophiles

Article abstract of DOI:10.1002/chem.201100608

A well-defined macrocyclic aryl-Cu^III complex (2) reacts readily with a variety of oxygen nucleophiles, including carboxylic acids, phenols and alcohols, under mild conditions to form the corresponding aryl esters, biaryl ethers and alkyl aryl

Full text of DOI:10.1002/chem.201100608

FULL PAPER
DOI: 10.1002/chem.201100608
ꢀ
Observation and Mechanistic Study of Facile C O Bond Formation between
a Well-Defined Aryl–Copper(III) Complex and Oxygen Nucleophiles
AHCTUNGTRENNUNG
Lauren M. Huffman,[a] Alicia Casitas,[b] Marc Font,[b] Mercꢀ Canta,[b] Miquel Costas,[b]
Xavi Ribas,*[b] and Shannon S. Stahl*[a]
ꢀ
Abstract: A well-defined macrocyclic
aryl–CuIII complex (2) reacts readily
with a variety of oxygen nucleophiles,
including carboxylic acids, phenols and
alcohols, under mild conditions to form
the corresponding aryl esters, biaryl
ethers and alkyl aryl ethers. The rela-
tionship between these reactions and
tigation of the stoichiometric C O
bond-forming reactions revealed nucle-
ophile-dependent changes in the mech-
anism. The reaction of 2 with carboxyl-
ic acids revealed a positive correlation
react more rapidly). The latter trend
resembles previous observations with
nitrogen nucleophiles. With carboxylic
acids and acidic phenols, UV-visible
spectroscopic data support the forma-
tion of a ground-state adduct between
2 and the oxygen nucleophile. Collec-
tively, kinetic and spectroscopic data
support a unified mechanism for aryl-
O coupling from the CuIII complex,
consisting of nucleophile coordination
to the CuIII center, deprotonation of
between the logACHTNUGRTNE(UNG kobs) and the pKa of
the nucleophile (less-acidic nucleo-
philes react more rapidly), whereas a
negative correlation was observed with
most phenols (more-acidic phenols
ꢀ
catalytic C O coupling methods is
demonstrated by the reaction of the
macrocyclic aryl–Br species with acetic
acid and p-fluorophenol in the pres-
ence of 10 mol% CuI. An aryl-CuIII-Br
species 2Br was observed as an inter-
mediate in the catalytic reaction. Inves-
ꢀ
Keywords: C O bond formation ·
copper · homogeneous catalysis ·
ꢀ
the coordinated nucleophile, and C O
ꢀ
(or C N) reductive elimination from
reductive elimination
CuIII.
Introduction
been developed, such as CuII-mediated oxidative coupling
reactions with arylboronic acids.[13–16]
Copper-mediated synthesis of diaryl ethers, diaryl amines, or
diaryl thioethers, commonly known as Ullmann or Ull-
mann–Goldberg coupling reactions, were discovered more
than a century ago.[1] These reactions result in the formation
ꢀ
ꢀ
ꢀ
of a C O, C N, or C S bond by the reaction of an aryl
halide with a phenol, an aniline, or a thiophenol, respective-
ly [Eq. (1)].[2–4] Reactions of this type have been the focus of
renewed interest in recent years, owing to the discovery of
new catalyst systems for these reactions, many of which uti-
lize nitrogen- and/or oxygen-based chelating ligands.[5–12]
These developments have significantly expanded the scope
and utility of these synthetic transformations. In addition, a
number of useful complementary coupling reactions have
In spite of these significant advances, the mechanisms of
these copper-catalyzed coupling reactions remain the subject
of investigation and debate, and at least three pathways
have been considered (Scheme 1). One of the most fre-
quently invoked catalytic mechanisms consists of a CuI/CuIII
cycle, analogous to the widely established Pd0/PdII cycle in-
volved in Pd-catalyzed cross-coupling reactions; however,
experimental evidence for this pathway is limited. Evidence
for a CuI/CuIII pathway with typical aryl halide substrates in-
cludes the lack of radical rearrangement products that could
arise
from
electron-transfer/atom-transfer
pathways
[a] Dr. L. M. Huffman,+ Prof. Dr. S. S. Stahl
Department of Chemistry, University of Wisconsin-Madison
1101 University Avenue, Madison, WI 53704 (USA)
Fax : (+1)608-262-6143
(Scheme 1),[17,18] but recent computational studies evaluating
different reaction pathways lead to divergent conclu-
sions.[19,20] Recently, we reported the first direct evidence for
a CuI/CuIII pathway in a Cu-catalyzed cross-coupling reac-
tion. An aryl–CuIII intermediate was directly observed in the
C-N coupling reaction between a macrocyclic aryl bromide
and pyridone.[21] Mechanistic questions have also been
raised concerning the mechanistic role of heteroatom nucle-
ophiles in the reaction. In some reactions, the nucleophile
has been shown to coordinate to the CuI center, for exam-
ple, forming a CuI-amidate or -alkoxide prior to oxidative
E-mail: stahl@chem.wisc.edu
[b] A. Casitas,+ M. Font, M. Canta, Dr. M. Costas, Dr. X. Ribas
QBIS group, Departament de Quꢀmica, Universitat de Girona
Campus Montilivi, 17071 Girona, Catalonia (Spain)
Fax : (+34)972-41-81-50
E-mail: xavi.ribas@udg.edu
[+] Contributed equally to this work.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201100608.
Chem. Eur. J. 2011, 17, 10643 – 10650
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
10643

phenol in CH3CN at 258C. Addition of 10 mol% [CuI-
(CH3CN)4](CF3SO3) resulted in formation of the diaryl
A
ACHTUNGTRENNUNG
ether coupling product in 80% yield within 24 h (Scheme 2).
A similar reaction took place between 1-Br and acetic acid,
affording near-quantitative yield of the aryl ester. The reac-
tion with acetic acid proceeded about two-fold faster when
acetic acid was added slowly to the reaction mixture via a
syringe pump, rather than all at once at the beginning of the
reaction. This slight inhibitory effect of AcOH will be dis-
ꢀ
cussed further below in the context of stoichiometric C O
bond-forming reactions. When these reactions were moni-
tored by UV-visible spectroscopy, the oxidative-addition
product aryl-CuIII-Br (2Br) was observed as a steady-state in-
termediate (see Figure S1 and Supporting Information for
details).
ꢀ
Scheme 1. Possible mechanistic pathways for aryl heteroatom bond-
forming Ullmann-type coupling reactions.
addition of the aryl halide.[17,18,22] In other cases, however,
the nucleophile has been proposed to react with the CuIII
center, after oxidative addition of the aryl halide.[20,21]
In the context of these mechanistic discussions, we[23–25]
In the context of this catalytic reactivity, we undertook a
systematic investigation of the stoichiometric reactivity of
[aryl–CuIII]
ACHTUNTRGNEUNG(ClO4)2 complex 2 with different oxygen nucleo-
and others[26] have begun investigating carbon heteroatom
philes, HO-Nuc, depicted in Scheme 3. These nucleophiles
react with 2 in CH3CN solution from 15–508C to afford the
corresponding aryl-O-Nuc coupling product (as an HClO4
ꢀ
bond-forming reactions involving well-defined aryl–CuIII
complexes. Such studies can provide valuable insights into
the fundamental reactivity of aryl–CuIII species, and they es-
tablish a foundation for assessment of the potential role of
such species in Ullmann and related Cu-mediated cross-cou-
pling reactions. Here, we report the observation and mecha-
nistic investigation of reactions between a well-defined aryl–
CuIII complex with oxygen nucleophiles, including carboxylic
acids, phenols and alcohols, which form the corresponding
aryl esters, diaryl ethers and aryl alkyl ethers. Kinetic and
spectroscopic studies reveal mechanistic differences among
the different classes of oxygen nucleophiles, and these re-
sults are compared to recent results obtained with nitrogen
nucleophiles. A unified mecha-
ꢀ
Scheme 2. Catalytic C O cross-coupling reactions. Conditions: [1-Br]=
5 mm, [Cu
A
G
1H NMR yield using 1,3,5-trimethoxybenzene as internal standard in
CD3CN.
nism is proposed that accounts
for all of the results with both
O- and N-nucleophiles.
Results and Discussion
ꢀ
Catalytic C O bond formation
involving an aryl–CuIII complex:
Our previous studies of stoi-
chiometric and catalytic reac-
tions of aryl–CuIII species with
amide-type nitrogen nucleo-
philes[21,23] prompted us to in-
vestigate analogous reactions
with oxygen nucleophiles (HO-
Nuc). Preliminary studies re-
vealed that the macrocyclic aryl
bromide 1-Br is capable of un-
dergoing catalytic coupling with
a phenol or carboxylic acid nu-
cleophile (Scheme 2). The aryl
bromide 1-Br was combined
III
ꢀ
with two equivalents of p-F- Scheme 3. Reactions of aryl Cu complex 2 with nitrogen and oxygen nucleophiles.
10644
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 10643 – 10650

ꢀ
C O Bond Formation
FULL PAPER
adduct) and [CuI
ACHTUNGTRENNUNG(CH3CN)4]ClO4. The reactions were per-
Table 1. Aryl-O-Nuc product yields from the reaction of 2 with carboxyl-
ic acids.[a]
formed under N2 to prevent any side-reactions related to
the oxidation of the CuI product. Qualitatively, the carboxyl-
ic acids reacted faster than phenols, and the latter faster
than aliphatic alcohols. We recently communicated our work
on the stoichiometric reactivity of 2 with nitrogen nucleo-
philes.[23] The key results of this study will be summarized
below, followed by a systematic investigation of the reactivi-
ty of 2 with the different oxygen nucleophiles.
Entry
Structure
pKa(DMSO)
Yield [%][b]
100
1
12.6
2
3
12.8
12.9
100
100
Summary of previous work with nitrogen nucleophiles:
Complex 2 reacts readily with a number of different amide-
4
9.1
100
type nitrogen nucleophiles in acetonitrile at 508C to afford
[23]
ꢀ
the corresponding C N coupled product (cf. Scheme 3).
5
6
7
–
100
100
100
Kinetic studies of these reactions were consistent with a
simple bimolecular rate law, first-order in both [2] and [HN-
Nuc]. Analysis of the effect of the pKa of the nitrogen nu-
cleophile on the rate of the reaction revealed a reasonable
Brønsted correlation with a negative slope, indicating that
more-acidic nucleophiles react more rapidly (Figure 1; pyri-
done does not fit the correlation, as discussed previously[23]).
Two different products were observed from the reactions of
the less-acidic nucleophiles (benzamide, oxazolidinone and
11
11.2
8
11.4
100
ꢀ
acetanilide): the intermolecular C N cross-coupling product
ꢀ
and an intramolecular C N reductive elimination product 3,
involving the aryl moiety and the central tertiary amine ni-
[a] General conditions: [2]=2.2 mm, [HO-Nuc]=2.2 mm in 0.4 mL
CD3CN. All reactions are complete at 248C in less than 10 min.
[b] 1H NMR yield, int. std.=1,3,5-trimethoxybenzene.
trogen (Scheme 4).[27]
Carboxylic acids as HO-Nuc: The combination of aryl–CuIII
complex 2 with a number of different carboxylic acids in
acetonitrile at room temperature resulted in fast (<10 min)
and quantitative formation of the corresponding aryl esters
(Table 1). By cooling the reactions to 0–158C, it was possible
to monitor the reaction by UV-visible spectroscopy, follow-
ing the decay of the 450 nm charge-transfer band of complex
2. In the presence of 5–10 equiv of HO-Nuc, the disappear-
ance of 2 exhibits an exponential time course, whereas,
higher [HO-Nuc] resulted in a slower non-exponential reac-
tion profile. Representative time-course plots for acetic acid
and benzoic acid are shown in Figure 2. At relatively low
[HO-Nuc] (ꢁ10–15 equiv), the reaction rate is approximate-
ly independent of [HO-Nuc], whereas the reaction is inhibit-
ed at higher [HO-Nuc] (Figure 2; see Supporting Informa-
tion for other carboxylic acids). Analysis of the effect of the
pKa of the carboxylic acids on the rate of the reaction re-
vealed a reasonable Brønsted correlation, but this time the
plot exhibits a positive slope (Figure 3; cf. Figure 1), indicat-
ing that less-acidic carboxylic acids react more rapidly.
Figure 1. Brønsted plot for the reaction of aryl–CuIII complex 2 with vari-
ous nitrogen nucleophiles, showing that more-acidic nuclophiles react
more rapidly. Reaction conditions: [2]=0.8 mm, [HN-Nuc]=8.0 mm,
CH3CN, 508C[1] (this plot is derived from data presented in ref. [23]).
Phenols and aliphatic alcohols as HO-Nuc: The combination
of aryl–CuIII complex 2 with phenols in acetonitrile at 508C
resulted in quantitative formation of the corresponding
diaryl ethers (Table 2, entries 1–7). As evident from the
higher temperature, phenols are less reactive than carboxylic
acids, and aliphatic alcohols are even less reactive. Acidic
aliphatic alcohols, 2,2,2-trifluoro- and 2,2,2,-tribromoethanol
reacted in good yield at 508C (Table 2, entries 8 and 9);
however, simple aliphatic alcohols, cyclohexanol and
tBuOH, failed to react. In the latter cases, complex 2 was
ꢀ
Scheme 4. Reaction of 2 with nitrogen nucleophiles, resulting in C N
bond coupling. In reactions with less-acidic nucleophiles, formation of
ꢀ
the intramolecular C N reductive elimination product 3 competes with
ꢀ
intermolecular C N bond formation.
Chem. Eur. J. 2011, 17, 10643 – 10650
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10645

S. S. Stahl, X. Ribas et al.
Table 2. Aryl-O-Nuc product yields from the reaction of 2 with phenols
and aliphatic alcohols.[a]
Entry
HO-Nuc
pKa(DMSO)
Yield [%][b]
100
1
10.8
2
3
4
13.2
15.3
16.7
100
100
100
5
6
7
18.0
18.0
19.1
100
100
100
8
9
CF3CH2OH[c]
CBr3CH2OH[d]
23.5
–
75
85
10
–
0[e]
11
12
G
32.2
29.0
0[f]
100
[a] General conditions: [2]=12 mm, [HO-Nuc]=12.5 mm, 508C, CH3CN.
1
ꢀ
[b] H NMR yield of C O coupling product, int. std. =1,3,5-trimethoxy-
benzene. [c] 3 equiv of 2,2,2-trifluoroethanol. [d] 5 equiv of 2,2,2-tribro-
ꢀ
moethanol. [e] 5 equiv of cyclohexanol; 47% yield of intramolecular C
ꢀ
N coupling product, 3. [f] 52% yield of intramolecular C N coupling
product, 3. [g] MeOH used as the solvent; see ref. [25].
Figure 2. Reaction time course profiles and [HO-Nuc] dependence of the
reaction of aryl–CuIII complex 2 with acetic acid (a) and benzoic acid (b)
(based on monitoring the 450 nm band of 2 by UV-visible spectroscopy).
[2]=0.55 mm, [HO-Nuc]=0.8 mm to 40 mm 158C for benzoic acid deriva-
tives, 58C for acetic acid.
The dependence of the reaction rate on the [phenol] was
investigated with several para-substituted phenols. The reac-
tions of complex 2 with p-methoxy- and p-fluorophenol
(pKa(DMSO) =19.1 and 18.0, respectively) exhibited a first-
order dependence of the rate on [p-X-phenol] for both nu-
cleophiles (Figure 4a and S3a). In contrast, the reactions of
p-trifluoromethyl- and p-cyanophenol (pKa(DMSO) =15.3 and
13.2, respectively) exhibited a saturation dependence on [p-
X-phenol] (Figure 4b and S3b). With all four of these nucle-
ophiles, the reactions exhibited a clean exponential time
course for all nucleophile concentrations. Finally, the most
acidic phenol investigated, p-nitrophenol (pKa(DMSO) =10.8),
exhibited kinetic behavior reminiscent of the carboxylic
acids: an exponential time course and a rate relatively inde-
pendent of [phenol] at low [phenol] (5–10 equiv), and inhibi-
tion of the rate at higher [phenol] (Figure 4c).
The effect of phenol pKa was evaluated by comparing the
reaction rates obtained when the reaction was carried out in
the presence of 10 equivalents of nucleophile. At this nucle-
ophile concentration, the rate is approximately first-order in
[phenol]. The resulting Brønsted plot showed a clear trend
for all of the nucleophiles (except p-nitrophenol), with the
increased rates observed with more-acidic nucleophiles
(Figure 5). This trend matches the qualitative behavior of ni-
trogen nucleophiles (Figure 1), but it is opposite to the
trend observed with carboxylic acids (Figure 3).
Figure 3. Brønsted plot for the reaction of aryl–CuIII complex 2 with car-
boxylic acid nucleophiles, showing that less-acidic nucleophiles react
more rapidly. Reaction conditions: [2]=0.8 mm, [HO-Nuc]=8 mm,
CH3CN, 158C.
partially recovered from the reaction mixture, and it reacts
primarily to form the intramolecular C-N coupling product
3[27] (Table 2, entries 10 and 11). Product 3 was not observed
with the other, more-reactive oxygen nucleophiles. If aceto-
nitrile is replaced with methanol as the solvent, complex 2
reacts quantitatively to afford the corresponding aryl methyl
ether, as described previously.[25]
10646
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 10643 – 10650

ꢀ
C O Bond Formation
FULL PAPER
Figure 5. Plot of logACHTUNGTRNEUNG(kobs) versus pKa(DMSO) (Brønsted plot) with the sever-
al p-substituted phenols. Conditions: [2]=1 mm, [HO-Nuc]=10 mm,
CH3CN, 258C.
2X (X=Cl, Br, I) that can undergo aryl–X reductive elimina-
tion;[21] however, analogous complexes with nitrogen and
oxygen nucleophiles have not been observed previously. Ef-
forts to prepare the aryl-CuIII-ONuc intermediates by addi-
tion of deprotonated nucleophiles in excess (e.g., AcOꢀ or
PhOꢀ) to 2 were unsuccessful. Such reactions resulted in
rapid formation of a deeply colored solution, and the species
formed under these conditions do not undergo straightfor-
ward formation of the aryl-O-Nuc product. The origin of
these observations is not yet known, but we speculate that
the anionic nucleophiles react with 2 by deprotonating one
of the secondary amines, rather than coordinating to the Cu
center.[28] Efforts to characterize this reactivity are ongoing.
The potential interaction of several carboxylic acids and
phenols with 2 was probed by UV-visible spectroscopy at re-
duced temperature (0 ! ꢀ308C) by titrating the HO-Nuc
(1–40 equiv) into an acetonitrile solution of 2 (Figure 6).
Addition of acetic acid and benzoic acid into a solution of 2
resulted in a noticeable change in the spectrum (Figure 6a
and b). The solution containing acetic acid started bleaching
during the course of the experiment, consistent with forma-
tion of the aryl–OAc product, even at ꢀ308C. In the case of
benzoic acid, isosbestic points were observed at 430 and
710 nm, consistent with clean formation of a new species.
Spectral changes were also observed upon addition of p-cya-
nophenol to a solution of 2 (Figure 6c). A plot of volume-
corrected Abs550nm versus [p-CN-phenol] exhibits a satura-
tion curve (Figure 7) that resembles the saturation kinetic
profile for formation of the corresponding aryl-OAr product
(Figure 4b). No spectral changes were observed upon
adding p-methoxyphenol to a solution of 2 (Figure 6d). This
observation is consistent with the fact that the reaction of 2
with p-methoxyphenol never deviates from a first-order ki-
netic dependence (cf. Figure 4a). Overall, the data in
Figure 6 supports the presence of a ground-state interaction
between 2 and the more-acidic (and reactive) nucleophiles,
carboxylic acids and acidic phenols. The precise identity of
this adduct is not known; however, the relatively small spec-
troscopic changes in Figure 6a–c suggest that the adduct
does not arise from coordination of a deprotonated nucleo-
phile to the CuIII center. Such species would be expected to
exhibit relatively intense ligand-to-metal charge-transfer
bands, analogous to those observed for aryl-CuIII-halide
complexes.[21] An alternative structure for the adduct is dis-
cussed below, in the context of the proposed mechanism.
Figure 4. Nucleophile concentration dependence with several para-substi-
tuted phenols: a) p-MeO-phenol b) p-CN-phenol and c) p-NO2-phenol.
Conditions: [2]=1 mm, CH3CN, 258C except for the reaction of p-NO2-
phenol, which was performed at 158C.
Efforts to identify CuIII-nucleophile adducts: In the reac-
tions described above, a zero-order kinetic dependence on
[HO-Nuc] is observed at low [HO-Nuc] for carboxylic acids
and p-nitrophenol, and at high [HO-Nuc] for p-CN- and p-
CF3-phenol. These observations could be explained by the
formation of a ground-state adduct between the nucleophile
and the aryl–CuIII complex. We recently reported structural
characterization of five-coordinate aryl-CuIII-X complexes
Chem. Eur. J. 2011, 17, 10643 – 10650
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10647

S. S. Stahl, X. Ribas et al.
Scheme 5. Proposed mechanism for the reaction of aryl–CuIII complex 2
with HO-Nuc (acetic acid used in the depicted proposal).
III
ꢀ
phile (step 2), and C O reductive elimination from Cu , re-
sulting in formation of the coupling product and CuI.
According to this proposal, the ability of more-acidic nu-
cleophiles to coordinate to 2 is rationalized by their ability
to undergo hydrogen bonding to an acetonitrile molecule
from the solvent. This interaction will enhance the anionic
character of the HO-Nuc, thereby making it a better ligand.
This type of weak CuIII/HO-Nuc interaction rationalizes the
small spectroscopic changes observed upon nucleophile
binding (Figure 6). Formation of adduct A as the resting
state species in solution would account for the zero-order ki-
netic dependence on [carboxylic acid] at low concentration
Figure 6. Volume corrected UV-visible spectra of the titration of a) acetic
acid, b) benzoic acid, c) p-CN-phenol and d) p-MeO-phenol nucleophiles
into an acetonitrile solution of 2. Conditions for acetic, benzoic acid and
p-methoxyphenol: [2]=0.8 mm, [HO-Nuc]=0.8–16 mm, ꢀ308C. Condi-
tions for p-CN-phenol: [2]=1.5 mm, [Nuc]=1.5 mm ꢀ75 mm, 08C.
ꢀ
(Figure 2) because loss of a proton from A followed by C
O reductive elimination will be unaffected by excess carbox-
ylic acid in solution. The ability of acetonitrile to act as a
base in step 2 of the mechanism (k2) finds precedent in the
literature.[29,30] The inhibitory behavior observed at elevated
[carboxylic acid] is not fully understood, but it might arise
from dissociation of the coordinated nucleophile via forma-
Figure 7. Plot of absorbance (volume corrected) versus [p-CN-phenol] at
550 nm for the titration of p-CN-phenol into a cooled solution of 2 at
08C. Initial [2]=1.5 mm, 1 cm path length.
ꢀ
tion of intermediate C, which has a carboxylic acid carbox-
ylate hydrogen-bonded dimer as a counterion. Charge-assist-
ed hydrogen bonds of this type can be quite strong.[31]
The diverse kinetic behavior observed with phenols
(Figure 4) can be explained by the relative propensity of the
different phenols to form the ground-state adduct A. p-Ni-
trophenol is sufficiently acidic to form A at low concentra-
tions (i.e., K1 @1) and p-cyanophenol favors formation of A
at high concentrations. In these concentration regimes, the
Mechanistic proposal: A mechanistic pathway that accom-
modates all of the experimental data described above is de-
picted in Scheme 5, using acetic acid as a representative
HO-Nuc. The key steps in the mechanism consist of pre-
equilibrium formation of a CuIII-(HO-Nuc) adduct A, stabi-
ꢀ
lized by hydrogen bonding between the O H group and
CH3CN (step 1), deprotonation of the coordinated nucleo-
reaction exhibits
[ArOH]. An adduct of p-methoxyphenol is never detected,
a zero-order kinetic dependence on
10648
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 10643 – 10650

ꢀ
C O Bond Formation
FULL PAPER
even at high concentrations, and the reaction exhibits a
simple first-order dependence on [p-methoxyphenol] at all
concentrations.
tant foundation for understanding the behavior of different
nucleophiles in catalytic reactions. Whereas some Cu-cata-
lyzed coupling reactions have been shown to proceed via
CuI–nucleophile intermediates,[17,22] others have been pro-
posed to proceed via reaction of a nucleophilic coupling
partner with an aryl–CuIII intermediate.[20,32] A growth in un-
derstanding of the fundamental reactivity of high-valent
organocopperACTHUNTGRNEUNG(III) species should play an important role in
the development of new or improved copper-catalyzed syn-
thetic methods.
Brønsted plots revealed sharp differences among different
types of nucleophiles: nitrogen and most phenol nucleo-
philes exhibit a negative Brønsted-plot slope (more-acidic
nucleophiles react faster) whereas carboxylic acids exhibit a
positive Brønsted-plot slope (less-acidic nucleophiles react
faster). These results can be rationalized by a change in the
rate-limiting step for different nucleophiles, from rate-limit-
ing deprotonation (k2) for less-acidic nucleophiles to rate-
ꢀ
limiting C O reductive elimination (k3) for more-acidic nu-
cleophiles. This conclusion is consistent with the kinetic data
for different nucleophiles noted above.
Experimental Section
As noted above, we have not been able to prepare inter-
mediate B directly via addition of acetate (or other depro-
tonated nucleophiles) to 2 because a side reaction takes
place, possibly involving deprotonation of the secondary
amines of 2. However, the mechanism in Scheme 5 suggest-
ed that it might be possible to promote the reaction by
adding acetate to a solution of the pre-formed acetic acid
adduct A. Addition of 1 equiv of Bu4NOAc after the addi-
tion of 4 equiv of acetic acid to a solution of 2 at 58C result-
ed in significantly (twelve-fold) faster decay of the aryl–CuIII
species (kobs =30.3ꢂ10ꢀ2 sꢀ1) relative to the reaction of 2
with 5 equiv of acetic acid (kobs =2.5ꢂ10ꢀ2 sꢀ1) (Figure 8).
These observations suggest that acetate can serve as a more
effective base than CH3CN in step 2, thereby favoring pre-
equilibrium deprotonation (K2) and enhancing the rate of
Ligand 1 and aryl–CuIII complex 2 were synthesized following published
procedures.[25,33]
General procedure for catalytic experiments: In an inert-atmosphere
glove box, a vial was loaded with ligand 1-Br (0.5 mL, 30 mm in CH3CN)
and CuACHTUNRTGENNU(G CH3CN)4ACHTUNGTRNE(NUGN CF3SO3) was added (10 mol%, 0.2 mL of stock solu-
tion 7.5 mm in CH3CN). The colorless solution became red indicating
that oxidative addition took place obtaining the corresponding complex
aryl-CuIII-Br, complex 2Br
. Then HO-nucleophile (2.3 mL, 13 mm in
CH3CN) was added. Final concentrations: [1-Br]=5 mm, [Cu]=0.5 mm
and [HO-nucleophile]=10 mm. After stirring the reaction mixture for
24 h, 1,3,5-trimethoxybenzene (50 mL, 3 mm in CH3CN) as internal stan-
dard was added and the solvent was removed under vacuum. The sample
was redissolved in CD3CN (0.5 mL) and NMR yields were obtained by
1H NMR using integration of benzylic protons respect to 1,3,5-trimethox-
ybenzene. In the case of acetic acid the corresponding equivalents were
added by syringe pump in a period of 5 h.
ꢀ
Synthesis and characterization of aryl esters (C O coupling with carbox-
ylic acids): In an inert-atmosphere glove box, a sample of the aryl–CuIII
complex 2 (9 mg, 17 mmol) was dissolved in CD3CN (2 mL). A portion of
this solution (0.1 mL) was loaded into an NMR tube, and the carboxylic
acid nucleophile (1 to 10 equiv) was added into the tube. The tube was
sealed with a screw-cap and the reaction was monitored by 1H NMR
spectroscopy at room temperature. 1H and 13C NMR spectra and mass
spectrometric analysis were obtained without isolation of the C-O cou-
pling product. Reaction yields were obtained by integration of the
1H NMR spectra of the crude reaction mixtures relative to 1,3,5-trime-
thoxybenzene as an internal standard (2.5 mm in CD3CN).
ꢀ
C O bond formation.
ꢀ
Synthesis and characterization of diaryl ethers and aryl alkyl ethers (C
O coupling with alcohols): In an inert-atmosphere glove box, a sample of
the aryl–CuIII complex 2 (49.2 mg, 17 mmol) was dissolved in CD3CN
(4.6 mL). A portion of this solution (0.4 mL) was loaded into an NMR
tube, and 1.1 equiv of the corresponding phenol nucleophile was added
to the tube (0.25 mL, 35 mm). 1,3,5-Trimethoxybenzene (50 mL) was
added as an internal standard. Final concentrations: [2]=12 mm and [al-
cohol]=12.5 mm. The tube was sealed with a screw-cap and the reaction
was heated at 508C and monitored by 1H NMR spectroscopy until reac-
tion completion. 1H and 13C NMR spectra and mass spectrometric analy-
Figure 8. Reaction profiles for the reaction of 2 with 5 equiv of acetic
acid (circles) and with 4:1 mixture of acetic acid/NBu4OAc (triangles).
Conditions: [2]=0.8 mm, [AcOH] + [AcOꢀ]=4.0 mm, 58C, N2 atmos-
phere, CH3CN.
Conclusion
ꢀ
sis were obtained without isolation of the C O coupling product. Reac-
tion yields were obtained by integration of the 1H NMR spectra of the
crude reaction mixtures relative to 1,3,5-trimethoxybenzene.
Aryl–copperACHTUNGTRENNUNG(III) intermediates have been proposed in a
General procedure for monitoring kinetics by UV/Vis and NMR spec-
troscopy: A UV-visible cuvette equipped with a Teflon stopcock was
dried in an oven and cooled under vacuum. Stock solutions of the O-nu-
cleophile (40 mm) and the CuIII–aryl complex 2 (7.5 mm) were prepared
in dry acetonitrile (50 mL). After backfilling the cuvette with dry N2,
0.5 mL of the nucleophile stock solution was added via syringe, and it
was diluted with acetonitrile to a total volume of 2.5 mL. The cuvette
was inserted into the spectrometer and the temperature was allowed to
wide range of Cu-catalyzed and Cu-mediated reactions;
however, very few species of this type are known, and inves-
tigations of their reactivity are very limited. In this study we
have observed and mechanistically characterized a number
of different reactions between a well-defined aryl–CuIII com-
plex and heteroatom nucleophiles. The results showcase im-
portant systematic mechanistic changes that arise from
changing the pKa of the heteroatom nucleophile. Fundamen-
tal insights of the type described here can provide an impor-
equilibrate. The reaction was initiated by adding the copperACTHNUTRGENUGN(III) stock so-
lution (0.3 mL) to the cuvette followed by rapid mixing of the combined
solutions. (Final concentrations: [2]=0.8 mm, [nucleophile]=8.0 mm).
Chem. Eur. J. 2011, 17, 10643 – 10650
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10649

S. S. Stahl, X. Ribas et al.
The average dead time was estimated to be 7 s. The reaction progress
was monitored by measuring the change in absorbance at 450 nm, and
the data were fit to an exponential decay curve.
[11] G. Evano, N. Blanchard, M. Toumi, Chem. Rev. 2008, 108, 3054.
[12] D. S. Surry, S. L. Buchwald, Chem. Sci. 2010, 1, 13.
[13] D. M. T. Chan, K. L. Monaco, R.-P. Wang, M. P. Winters, Tetrahe-
dron Lett. 1998, 39, 2933.
General procedure for obtaining kinetic data by NMR spectroscopy: In
an inert-atmosphere glove box, a stock solution of the CuIII–aryl complex
2 (26 mm) and tri-tert-butyl benzene (1 mg, 4 mmol) as an internal stan-
dard was prepared in CD3CN (1.0 mL). A stock solution of O-nucleo-
phile (75 mm) in CD3CN (2.0 mL) was prepared. The concentration of
each stock solution was confirmed by 1H NMR spectroscopy. Pulse
widths and relaxation times were determined by using standard methods.
To acquire the kinetic data, 0.3 mL of the nucleophile stock solution was
added to a NMR tube and sealed with a septum, and the sample was
placed in the NMR probe for 5 min to allow temperature equilibration.
The reaction was initiated by addition of 0.1 mL of the CuIII–aryl stock
solution to the NMR tube via syringe. The solution was mixed rapidly
and the tube was inserted into the probe to begin data acquisition. The
relaxation delay between acquisitions was set to 5ꢂT1 (T1 ~2 s); Final
concentrations: [2]=6.5 mm, [nucleophile]=56 mm.
[14] D. A. Evans, J. L. Katz, T. R. West, Tetrahedron Lett. 1998, 39, 2937.
[15] P. Y. S. Lam, C. G. Clark, S. Saubern, J. Adams, M. P. Winters,
D. M. T. Chan, A. Combs, Tetrahedron Lett. 1998, 39, 2941.
[16] J. X. Qiao, P. Y. S. Lam, Synthesis 2011, 829.
[17] J. W. Tye, Z. Weng, A. M. Johns, C. D. Incarvito, J. F. Hartwig, J.
Am. Chem. Soc. 2008, 130, 9971.
[18] J. W. Tye, Z. Weng, R. Giri, J. F. Hartwig, Angew. Chem. 2010, 122,
2231; Angew. Chem. Int. Ed. 2010, 49, 2185.
[19] G. O. Jones, P. Liu, K. N. Houk, S. L. Buchwald, J. Am. Chem. Soc.
2010, 132, 6205.
[20] H.-Z. Yu, Y.-Y. Jiang, Y. Fu, L. Liu, J. Am. Chem. Soc. 2010, 132,
18078.
[21] A. Casitas, A. E. King, T. Parella, M. Costas, S. S. Stahl, X. Ribas,
Chem. Sci. 2010, 1, 326.
[22] E. R. Strieter, B. Bhayana, S. L. Buchwald, J. Am. Chem. Soc. 2009,
131, 78.
[23] L. M. Huffman, S. S. Stahl, J. Am. Chem. Soc. 2008, 130, 9196.
[24] A. Casitas, A. Poater, M. Solꢄ, S. S. Stahl, M. Costas, X. Ribas,
Dalton Trans. 2010, 39, 10458.
Acknowledgements
We are grateful for financial support from the MICINN of Spain
(CTQ2009-08464/BQU to M.C.; PhD grants to A.C. and M.F.), US De-
partment of Energy (DE-FG02-05ER15690) and Bristol-Myers Squibb
(to S.S.S.). M.C. and X.R. thank the Generalitat de Catalunya for
ICREA Academia awards and 2009 SGR637. We also thank STRꢃs from
UdG for NMR and ESI-MS technical support.
[25] A. E. King, L. M. Huffman, A. Casitas, M. Costas, X. Ribas, S. S.
Stahl, J. Am. Chem. Soc. 2010, 132, 12068.
[26] B. Yao, D.-X. Wang, Z.-T. Huang, M.-X. Wang, Chem. Commun.
2009, 2899.
[27] For a mechanistic study of the intramolecular C-N reductive elimi-
nation reaction to form 3, see: L. M. Huffman, S. S. Stahl, Dalton
Trans., DOI: 10.1039/C1DT10463B, in press.
[28] Characterization of this reactivity is described elsewhere: A. Casitas,
N. Ioannidis, G. Mitrikas, M. Costas, X. Ribas, Dalton Trans., DOI:
10.1039/C1DT10428D, in press.
[29] A. C. Legon, D. J. Millen, H. M. North, J. Phys. Chem. 1987, 91,
5210.
[30] J. E. Del Bene, J. Am. Chem. Soc. 1993, 115, 1610.
[31] P. Gilli, V. Bertolasi, V. Ferretti, G. Gilli, J. Am. Chem. Soc. 1994,
116, 909.
[1] a) F. Ullmann, Ber. Dtsch. Chem. Ges. 1903, 36, 2382; b) F. Ullmann,
A. Lehner, Ber. Dtsch. Chem. Ges. 1904, 37, 853; c) I. Goldberg,
Ber. Dtsch. Chem. Ges. 1906, 39, 1691; d) I. Goldberg, Ber. Dtsch.
Chem. Ges. 1907, 40, 4541; e) F. Ullmann, C. Wagner, Justus Liebigs
Ann. Chem. 1907, 355, 359.
[2] S. V. Ley, A. W. Thomas, Angew. Chem. 2003, 115, 5558; Angew.
Chem. Int. Ed. 2003, 42, 5400.
[32] Ongoing studies of CuII-catalyzed oxidative coupling reactions
(“Chan-Lam” reactions) in our lab reveal diverse nucleophile ef-
fects, including examples consistent with the reaction of nucleophile
with an intermediate aryl–CuIII species. For an initial study of these
reactions, see: A. E. King, T. C. Brunold, S. S. Stahl, J. Am. Chem.
Soc. 2009, 131, 5044.
[3] F. Monnier, M. Taillefer, Angew. Chem. 2009, 121, 7088; Angew.
Chem. Int. Ed. 2009, 48, 6954.
[4] E. Sperotto, G. P. M. van Klink, G. van Koten, J. G. de Vries, Dalton
Trans. 2010, 39, 10338.
[5] R. K. Gujadhur, C. G. Bates, D. Vekataraman, Org. Lett. 2001, 3,
4315.
[33] X. Ribas, D. A. Jackson, B. Donnadieu, J. Mahꢀa, T. Parella, R.
Xifra, B. Hedman, K. O. Hodgson, A. Llobet, T. D. P. Stack, Angew.
Chem. 2002, 114, 3117; Angew. Chem. Int. Ed. 2002, 41, 2991.
[6] M. Wolter, G. Nordmann, G. E. Job, S. L. Buchwald, Org. Lett. 2002,
4, 973.
[7] A. Shafir, P. A. Lichtor, S. L. Buchwald, J. Am. Chem. Soc. 2007,
129, 3490.
[8] D. Ma, Q. Cai, Acc. Chem. Res. 2008, 41, 1450.
[9] X. Lv, W. Bao, J. Org. Chem. 2007, 72, 3863.
[10] N. Xia, M. Taillefer, Chem. Eur. J. 2008, 14, 6037.
Received: February 24, 2011
Published online: August 16, 2011
10650
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 10643 – 10650