Ligand-free Ullmann-type C-heteroatom couplings under practical conditions

Article abstract of DOI:10.1002/ejoc.201400033

A new practical ligand-free protocol for copper-catalyzed C-heteroatom cross-coupling reactions (Ullmann-type) is described. The use of dimethyl sulfoxide (DMSO) as the solvent overcomes the need to use organic auxiliary ligands; thus, DMSO is revealed as a nontoxic and superior solvent for Ullmann-type coupling reactions. This method allows the arylation of a wide range of amides, alcohols, and amines under practical conditions with bromobenzene and iodobenzene derivatives and will likely find direct application in current organic synthesis. The competitive reactivity among different functional groups is reported and rationalized, and the possibility to achieve selective arylation reactions is demonstrated. Copyright

Full text of DOI:10.1002/ejoc.201400033

Job/Unit: O40033
/KAP1
Date: 14-04-14 10:32:14
Pages: 9
FULL PAPER
DOI: 10.1002/ejoc.201400033
Ligand-Free Ullmann-Type C–Heteroatom Couplings Under Practical
Conditions
Imma Güell[a] and Xavi Ribas*[a]
Keywords: Copper / Cross-coupling / Ligand-free catalysis / Ullmann coupling / Dimethyl sulfoxide
A new practical ligand-free protocol for copper-catalyzed C–
heteroatom cross-coupling reactions (Ullmann-type) is de-
scribed. The use of dimethyl sulfoxide (DMSO) as the solvent
overcomes the need to use organic auxiliary ligands; thus,
DMSO is revealed as a nontoxic and superior solvent for
Ullmann-type coupling reactions. This method allows the
arylation of a wide range of amides, alcohols, and amines
under practical conditions with bromobenzene and iodo-
benzene derivatives and will likely find direct application in
current organic synthesis. The competitive reactivity among
different functional groups is reported and rationalized, and
the possibility to achieve selective arylation reactions is dem-
onstrated.
elusive reaction mechanism of Ullmann-type reactions.[4d,5]
Nevertheless, more effort is needed to unravel the mecha-
nism and to design better catalysts that work under mild
conditions. The latter approach is a long road that needs to
be followed, but other methods must be developed to find
a compromise between maximum efficiency under mild
conditions, which are sought for fully-designed catalysts,
and practical, straightforward methods amenable to use in
routine organic synthesis.
Cu-catalyzed C–heteroatom coupling reactions are ex-
tremely sensitive to experimental conditions, and the op-
erating mechanism can differ among distinct experimental
sets. Therefore, the optimization of the different experimen-
tal parameters is crucial to achieve good C–heteroatom
coupling results, and more efforts in this area are also ap-
preciated.[4c,6] In this sense, a few “ligand-free” procedures
have been described in the past decade[7] and, in some cases,
represent an enhanced practicality at the expense of mech-
anistic understanding. In general, the reported ligand-free
methods either work with N-based nucleophiles, which can
act as ligands for the Cu catalyst, or they use solvents with
coordinating abilities such as N-methylpyrrolidone (NMP)
or N,N-dimethylformamide (DMF).
Introduction
Diaryl amino and ether moieties are widely present in
many compounds with applications in numerous fields,
such as life sciences and the chemical, pharmaceutical, and
polymer industries, and are also found in natural products
and biologically active compounds.[1] The synthesis of these
molecules by combining aryl halides and N- or O-nucleo-
philes has been achieved by two different C–heteroatom
coupling methods: (a) copper-catalyzed Ullmann-type cou-
pling reactions[2] and (b) palladium-catalyzed Buchwald–
Hartwig couplings.[3] Importantly, the cost and inherent
toxicity of palladium catalysts has prompted different re-
search groups to refocus on the copper-mediated Ullmann
reactions in the past decade (Scheme 1).[1a,4]
Scheme 1. Copper-mediated Ullmann reaction.
Herein, we disclose a simple and practical ligand-free
procedure for the copper-catalyzed arylation of different
oxygen- and nitrogen-containing nucleophiles. We take ad-
vantage of the coordinating ability of dimethyl sulfoxide
(DMSO) and describe a new easy-to-use method that is
available for regular organic synthesis and is useful for a
wide range of substrates. Moreover, DMSO has been lately
considered a clean, nontoxic, and superior solvent for many
metal-catalyzed transformations,[8] including Ullmann cou-
plings with auxiliary ligands.[9] However, in most reported
studies, the evaluation of the reaction outcomes with
DMSO in the absence of auxiliary ligands is very scarce.[10]
Many approaches and methods have been reported since
then, most of which use auxiliary ligands, usually bidentate
amine or diketone compounds. These provide a major de-
gree of control over the coordination sphere of the copper
center and have also facilitated the investigation of the still-
[a] QBIS Group, Institut de Química Computacional i Catàlisi
(IQCC) and Departament de Química, Universitat de Girona,
Campus Montilivi, 17071 Girona, Catalonia, Spain
E-mail: xavi.ribas@udg.edu
www.udg.edu/qbis
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201400033.
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1

Job/Unit: O40033
/KAP1
Date: 14-04-14 10:32:14
Pages: 9
I. Güell, X. Ribas
FULL PAPER
terms of the nature of the inorganic base and of the final
concentration of the catalytic reaction are included in the
Supporting Information (Table S2). The most efficient re-
sults were obtained with K3PO4 and with a concentration
of 1 of 1.25 m. With minor CuI catalyst loadings (5 mol-%),
the yields of 3a dropped to 74% (Table 1, Entry 6). Ad-
ditionally, [Cu(CH3CN)4][CF3SO3] was also tested as the
copper(I) source, and very similar results as those for CuI
were obtained (Table 1, Entry 10 and Table S1). If no cop-
per salt is added as catalyst, the reaction does not occur
(Table 1, Entry 9).
Results and Discussion
To develop a practical Ullmann-type coupling procedure,
our initial study consisted of a set of experiments with iodo-
benzene (1) as a model substrate and benzamide (2a) as a
N-nucleophile. Firstly, CuI (5–10 mol-%) was chosen as the
copper(I) source, and K3PO4 (1–2 equiv.) was chosen as the
inorganic base. DMSO was the solvent of choice, and the
reaction outcome was screened at different temperatures
under a N2 atmosphere (Table 1, Entries 1–7). Yields were
obtained as the average of at least two runs and were deter-
mined by GC, unless otherwise stated, because the isolated
yields for different coupling products were similar to those
obtained by GC (see Tables 1, 2, and 4). The optimized
combination of 90 °C, CuI (10 mol-%), and K3PO4
(2 equiv.) afforded the coupling product benzanilide (3a) in
97% yield. Subsequently, we explored other copper sources
such as CuII(OTf)2 (OTf = trifluoromethanesulfonate), but
much longer reaction times were required to reach quantita-
tive yields (Table S1, Entries 1–5); this suggests the exis-
tence of a reduction step when a copper(II) source is used
and clearly indicates the involvement of CuI species in the
catalytic cycle. This was further confirmed by the observa-
tion of an almost complete quench of the reaction under
an O2 atmosphere, which precluded the in situ reduction of
CuII to CuI species (Table S1, Entry 5). The latter can be
explained by the presence of the DMSO solvent, which has
been previously reported to act as a reducing reagent in
other reaction systems.[11] The preoptimization results in
Table 2. Ligand-free N-arylation of amides with iodobenzene.
Table 1. Arylation of benzamide (2) with iodobenzene (1) under
different conditions.
Cu source
[mol-%]
Temp. Conv. Yield
[°C] [%]
Entry[a]
Solvent
[%][b]
1
2
3
4
5
6
7
CuI (10)
CuI (10)
CuI (10)
CuI (10)
CuI (10)
CuI (5)
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
50
60
70
90
90
90
90
46
53
78
97
82
74
90
46
53
77
97
82[c]
74
CuI (10)
88[d]
[a] Reaction conditions: 1 (1.75 mmol, 1.25 m), 2 (3.5 mmol). [b]
Yield was determined by GC. [c] 2.2 mmol of amide was used in-
stead of 3.5 mmol. [d] 30% isolated yield of 3e, 5% GC yield and
4% isolated yield of p-phenoxybenzamide (5i), and 6% GC yield
and 5% isolated yield of p-phenoxybenzamide-N-benzanilide (3eЈ)
were obtained (also see Scheme 2). With 1 mL of iodobenzene
(scale-up reaction), the same results were obtained. [e] 8% yield of
N,N-diphenylacetamide (3gЈ). [f] Isolated yield: 64%. [g] Scaled-up
reaction (1 mL of iodobenzene); 6% yield of N,N-diphenylacet-
amide (3gЈ). [h] Isolated yield: 76%. [i] Isolated yield: 52%; * 2i =
3a, now used as substrate.
8
9
Cu(OTf)2 (10) DMSO
DMSO
CuOTf[e] (10) DMSO
90
90
45
0
43
0
–
10
11
12
13
14
15
16
90
90
90
93
90
57
77
81
17
70
93
90
43
77
81
16
70
CuOTf (10)
CuOTf (10)
CuOTf (10)
CuOTf (10)
CuI (10)
ACN
THF
THF/DMSO(9:1) 90
THF/DMSO(2:1) 90
DPSO
MPSO
90
90
CuI (10)
[a] Reaction conditions: 1 (1.75 mmol, 1.25 m), 2a (3.5 mmol), cop-
per salt (0.2 mmol), and K3PO4 (3.5 mmol). [b] Yield was deter-
mined by GC. [c] Scaled-up reaction (1 mL of iodobenzene). GC
yield 82%, isolated yield 76%. [d] 1.2 equiv. of K3PO4. [e] CuOTf
is [Cu(CH3CN)4][CF3SO3].
Next, we examined the effect of substituting DMSO by
other solvents. For example, if acetonitrile (ACN) is used,
yields decrease significantly at 50 and 70 °C, albeit yields of
90% of 3a were obtained at 90 °C (Table 1, Entry 11 and
2
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0

Job/Unit: O40033
/KAP1
Date: 14-04-14 10:32:14
Pages: 9
Ligand-Free Ullmann-Type C–Heteroatom Couplings
Table S1, Entries 9–11). The use of tetrahydrofuran (THF) the para and meta positions of phenols (Table 3, Entries 2–
also resulted in decreased yields, whereas mixtures of THF 4) increased the yields compared with those of the nonsub-
and DMSO afforded good yields at higher DMSO ratios stituted phenol (Table 3, Entry 1), the formation of product
(Table 1, Entries 12–14). Furthermore, the substitution of was almost quenched with electron-withdrawing groups at
one or two methyl groups of the sulfoxide for a phenyl the para position (Table 3, Entry 5). For aliphatic alcohols,
group [methylphenylsulfoxide (MPSO) and diphenylsulf- it was much more difficult to obtain the desired product
oxide (DPSO), Table 1, Entries 15 and 16] had a strong ef- in moderate yields owing to the formation of several side
fect on the reaction outcome and gave moderate and low products (Table S4, Entries 4 and 5). When CsF was used
yields, respectively. In general, the solvent screening shows as the base instead of K3PO4, the final products formed
that (a) highly polar aprotic solvents are beneficial for this without the presence of byproducts (Table 3, Entries 6–7)
reaction and enhance the solubility of the base and (b) the but in low or moderate-to-good yields. Trifluoroethanol
presence of the phenyl group affords poorer yields, most (4g) affords much better yields than ethanol (4f), and this
likely because of increased steric hindrance and the re- suggests that the deprotonation step has a clear role in the
duction of the electron-donating ability of the sulfoxide reaction (see below).
moiety as a ligand.
Table 3. Ligand-free O-arylation of phenols with iodobenzene.
To examine the scope of this reaction, iodobenzene was
subjected to N-arylation with various primary and second-
ary amides (Table 2). Aromatic primary amides analogous
to benzylamide (2a) such as 2-hydroxybenzamide (2d) and
nicotinamide (2f) afforded good-to-excellent yields, albeit at
higher reaction temperatures (130 °C; Table 2, Entries 1, 4,
and 6). The present method also tolerates aliphatic primary
amides such as acetamide (2g) and affords an 82% yield of
N-phenylacetamide (3g) in a gram-scale reaction (Table 2,
Entry 7). This reaction seemed to be sensitive to the steric
hindrance of the nucleophile, because although cyclic ali-
phatic secondary amides such as pyrrolidin-2-one (2b) af-
forded excellent yields of the coupling product 3b under the
optimized reaction conditions, the acyclic ones such as N-
methylacetamide (2h) and N-ethylacetamide (2j) afforded
moderate and low yields, respectively (Table 2, Entries 2, 8,
and 10). In any case, the 66% yield of 3h should be high-
lighted as acyclic secondary amides are usually poor coun-
terparts in arylation cross-coupling reactions. A copper-free
blank experiment was performed for 2b, and the reaction
did not proceed (0% conversion, Table S3, Entry 1). In con-
trast, aromatic secondary amides were less prone to aryl-
ation, and low or very low yields were obtained for 2c and
2i as nucleophiles (Table 2, Entries 3 and 9). The generally
low reactivity of secondary amides explains the high selec-
tivity encountered with primary amides. An exception for
these reactivity trends was found for 4-hydroxybenzamide
(2e), which gave the desired product in only 39% yield ow-
[a] Reaction conditions: 1 (1.75 mmol, 1.25 m), 4 (3.5 mmol). [b]
Yield was determined by GC. [c] Reaction temperature of 130 °C.
1
[d] CsF used as base. [e] Yield was determined by H NMR spec-
troscopy.
We also wanted to test the N-arylation with primary and
ing to the formation of side products (Table 2, Entry 5 and secondary amines as nucleophiles. We observed that the fi-
Scheme 2, E). In contrast, the ortho-substituted benzanilide nal products were obtained in modest yields under the same
3d did not afford byproducts related to the arylation of the optimized conditions (Table S5). As fluoride salts can accel-
alcohol group, probably because of steric hindrance erate some transition-metal-catalyzed coupling reactions,[12]
(Table 2, Entry 4).
we used CsF as a stronger base to replace the generally
After screening the amide family of N-nucleophiles un- used K3PO4 and were very pleased to find higher yields and
der optimized conditions, we then turned our attention to reduced side reactions after 24 h (Table 4). We also tested
O-nucleophiles. Various substituted phenols and aliphatic the use of Cs2CO3 as a base for the reactions with 6a, 6b,
alcohols were treated with iodobenzene (see Tables 3 and and 6d nucleophiles and obtained similar or lower yields
S3) at different temperatures; 110 °C is the optimal tem- compared to those with CsF.
perature for phenol derivatives. The copper-free reaction
Aromatic primary amines could be introduced in moder-
was tested for 4c, and the reaction did not occur (0% con- ate-to-excellent yields. The presence of an aromatic ring de-
version; Table S3, Entry 2). The electronic properties of the creased the reactivity, and aniline (6a) was less reactive
substituents played a crucial role in governing the product (Table 4, Entry 1). The introduction of para substituents
yield. Whereas the presence of electron-donating groups at also had a strong influence. In this case, the electron-de-
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3

Job/Unit: O40033
/KAP1
Date: 14-04-14 10:32:14
Pages: 9
I. Güell, X. Ribas
FULL PAPER
Table 4. Ligand-free N-arylation of amines with iodobenzene.
ficient arylamine 6b is more reactive than the electron-rich
6c, as the yield increased from moderate to excellent. The
aliphatic cyclohexanamine (6d) afforded very good aryl-
ation yields (7d; Table 4, Entry 4). Cyclic and acyclic sec-
ondary amines proved to be much more difficult nucleo-
philes for the C–N coupling catalysis. The acyclic ones did
not react or afforded poor yields (Ͻ5%; Table 4, Entries 5–
7), and the cyclic aliphatic piperidine (6h) afforded moder-
ate yields of the arylated product 7h (Table 4, Entry 8). In
contrast, conjugated N-heterocyclic secondary amines were
highly reactive and gave the desired products in excellent
yields even at 90 °C (Table 4, Entries 9 and 10). The reac-
tion time needed with our method is much shorter and the
results are comparable to those of previously reported li-
gand-free[10b] and non-ligand-free reactions.[13] Similarly,
heteroaromatic amines were highly reactive, as the corre-
sponding biaryl products were obtained as the major prod-
ucts (Table 4, Entries 11–14). Pyridin-2-amine (6k) afforded
full conversion of the double arylated product 7kЈ (70%
with respect to iodobenzene) at 110 °C, together with minor
amounts of the monoarylated one (7k, 28% yield). As ex-
pected, if 7k is used as the nucleophile, the reaction afforded
almost quantitative formation of 7kЈ (92%, Table 4, En-
try 12). For quinolin-8-amine (6l; Table 4, Entry 13), excel-
lent yields (94% with respect to iodobenzene) of the biaryl
coupling product 7lЈ were obtained at room temperature
and within 6 h (Table 4, Entry 13). Thus, the possible che-
lating ability of this nucleophile allows the reaction to occur
smoothly under very mild conditions. Finally, pyridazine-3-
amine (6m; Table 4, Entry 14) was also tested and afforded
70% yield (with respect to iodobenzene) of the biarylated
product 7mЈ at 110 °C. The reaction is fully copper-depend-
ent, as copper-free blank reactions afforded 0% conversion
(Table S3, Entry 3).
The ligand-free method described so far proved to be
very efficient for primary amides and amines, whereas reac-
tions with secondary amides and amines were much more
difficult. The new procedure described here also proved ef-
ficient for phenols (except those with strong electron-with-
drawing groups) and acidic aliphatic alcohols. Bearing in
mind the necessity to discriminate between functional
groups within a multifunctionalized compound, we con-
ducted competitive experiments among pairs of amides and
amines and also pairs of phenol and amines (Scheme 2).
The competitive reactions were performed under unified
optimized conditions: 130 °C, CsF as base, and iodo-
benzene as the limiting reagent. The competition between
nicotinamide (2f) and cyclohexanamine (6d) clearly showed
a much enhanced reactivity of the amide 2f to arylation
(Scheme 2, A; Table S7, Entry 1). This result indicates that
this method is orthogonal and clearly discriminates in favor
of primary amides with primary aliphatic amines unre-
acted.
[a] Reaction conditions: 1 (1.75 mmol, 1.25 m), 6 (3.5 mmol). [b]
Yield was determined by GC. [c] 12% yield of side product N,N-
diphenylaniline. [d] 15% yield of side product 4-methoxy-N,N-di-
phenylaniline. [e] Isolated yield: 87%. [f] Product not detected by
GC–MS. [g] Yield was determined by 1H NMR spectroscopy. [h]
Reaction temperature of 90 °C. [i] Isolated yield: 84%. [j] Isolated
yield: 81%. [k] K3PO4 used as base. [l] Reaction temperature of
110 °C. [m] 28% yield of side product N-phenylpyridin-2-amine
(7k). [n] 8 h reaction time. [o] 6 h at room temperature, traces of
monoarylated product 7l were found. [p] 15% yield of N-phenyl-
pyridazin-3-amine (7m).
When the competition is done between 2f and the most
reactive para-nitro-substituted aniline (6b; Table 4), the
arylation of the amide occurred with a slight preference
over the amine (1.4:1; Scheme 2, B). On the other hand, the
competition between phenol (4a) and cyclohexanamine (6d)
4
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0

Job/Unit: O40033
/KAP1
Date: 14-04-14 10:32:14
Pages: 9
Ligand-Free Ullmann-Type C–Heteroatom Couplings
Scheme 2. Competitive cross-coupling studies among different nucleophiles.
clearly showed an enhanced selectivity (7:1) for the aryl- tron-withdrawing or electron-donating groups at the para
ation of the aliphatic primary amine (Scheme 2, C; Table position of the iodobenzene ring did not result in an impor-
S7, Entry 2). On the contrary, when the competition is be- tant effect on the reactivity at high temperatures, although
tween phenol (4a) and aniline (6a), only a slight preference better results were obtained for the para-nitro derivative at
for the formation of the diphenyl ether (5a) is observed (3:2; lower reaction temperatures (Tables 5 and 6).
Scheme 2, D). As shown in Scheme 2, part E, the reactivity
On the other hand, we were delighted to observe that
with 4-hydroxybenzamide (2e) is a showcase for the com- bromobenzene is a good electrophilic partner in these cross-
parison of the reactivity between two functional groups coupling reactions when aromatic alcohol and amine nu-
(phenol and benzamide moieties) in the same molecule. cleophiles are used; however, it fails for amide nucleophiles
Again, the enhanced reactivity of the amide moiety over (Table 7). Excellent results were obtained for phenol deriva-
other functional groups (in this case the phenol moiety) was tives, and remarkable yields were obtained for amines (up
clearly observed.
to 62% for p-nitroaniline). With chlorobenzene, the reactiv-
With these good results in hand and to substantiate the ity was almost completely suppressed for all nucleophiles
generality of this new method, we increased the substrate tested (Table 8).
scope to other aryl halides. We selected two new aryl iodide
Regarding the possible reaction mechanism in the exam-
derivatives p-nitroiodobenzene and p-iodotoluene to check ples reported, the deprotonation of the nucleophile is not
the effect of electron-withdrawing and electron-donating relevant for amides (Table 2), as the yield of the reaction is
substituents and selected bromobenzene and chlorobenzene not affected over a wide range of pKa values (values in
to study the reactivity of different halides (Tables 5, 6, DMSO), that is, between 18.5 (indolin-2-one, 2c) to 25.5
and 7). For these reactions, we used the N- and O-nucleo- (acetamide, 2g). On the contrary, phenol substrates bearing
philes that gave the best results with iodobenzene (2a, 2b, electron-donating substituents (pKa ≈ 18–19) afforded bet-
4a, 4c, 4d, 6b, and 6d). On one hand, the presence of elec- ter yields than those with strong electron-withdrawing
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5

Job/Unit: O40033
/KAP1
Date: 14-04-14 10:32:14
Pages: 9
I. Güell, X. Ribas
FULL PAPER
Table 5. Ligand-free N- and O-arylation with p-iodonitrobenzene.
Table 7. Ligand-free N- and O-arylation with bromobenzene.
[a] Reaction conditions: p-iodonitrobenzene (1.75 mmol, 1.25 m),
nucleophile (3.5 mmol). [b] Yield was determined by GC. [c] CsF
1
used as base. [d] Yield was determined by H NMR spectroscopy.
[e] At 130 °C, lower yields of 7k were obtained owing to side reac-
tions.
[a] Reaction conditions: bromobenzene (1.75 mmol, 1.25 m), nu-
cleophile (3.5 mmol). [b] Yield was determined by GC. [c] CsF was
used as the base.
Table 6. Ligand-free N- and O-arylation with p-iodotoluene.
Table 8. Ligand-free N- and O-arylation with chlorobenzene.
[a] Reaction conditions: p-iodotoluene (1.75 mmol, 1.25 m), nucleo-
phile (3.5 mmol). [b] Yield was determined by GC. [c] CsF used as
1
base. [d] Yield was determined by H NMR spectroscopy.
[a] Reaction conditions:
(3.5 mmol). [b] Yield was determined by GC. [c] CsF used as base.
1 (1.75 mmol, 1.25 m), nucleophile
groups (p-NO2-phenol, pKa = 10.8). The latter suggests that
the mechanism of action for phenols (and probably for
amides) does not involve a rate-limiting deprotonation step, within the aromatic amines. The same effect is observed for
but mostly indicates that the stabilization of a high oxi- ethanol (4f, pKa = 29.8) and trifluoroethanol (4g, pKa =
dation state copper center (as a putative aryl–CuIII–nucleo- 23.5), as the yield of 5g is much higher than that of 5f.
phile intermediate) is mandatory for the reaction to pro-
The determination of the precise nature of the active cop-
ceed.[4d] On the other hand, p-NO2-aniline (6b, pKa = 20.9) per species that accounts for the catalytic coupling reactions
showed exactly the opposite effect and afforded the best reported here is extremely difficult to achieve. The combina-
yield compared to the other anilines (6a, pKa = 30.6 and tion of DMSO with a base at high temperatures generates
6c, pKa Ͼ30). The latter suggest that the rate-limiting step the dimsyl anion in the reaction mixture.[14] High resolution
has now switched to the deprotonation of the nucleophile mass spectrometry studies allowed us to propose the forma-
6
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0

Job/Unit: O40033
/KAP1
Date: 14-04-14 10:32:14
Pages: 9
Ligand-Free Ullmann-Type C–Heteroatom Couplings
–
were performed in a N2 drybox (Jacomex-GP-Concept-II-P) with
O2 and H2O concentrations of Ͻ1 ppm. 1H and 13C NMR spectra
were recorded with a Bruker 400 MHz or Bruker 300 MHz NMR
spectrometer. Chemical shifts (δ) are reported in ppm and were
directly referenced to the solvent signal. GC product analyses were
performed with an Agilent 7820A gas chromatograph equipped
with an HP-5 capillary column (30 mϫ0.32 mmϫ0.25 μm) and a
flame ionization detector. GC–MS analyses were performed with an
Agilent 7890A gas chromatograph equipped with an HP-5 capillary
column interfaced with an Agilent 5975C mass spectrometer. The
tion of the reactive dimsyl anion (CH3SOCH2 ) under the
experimental conditions used and it might have a relevant
role in all catalytic reactions. After 2 h of heating a suspen-
sion of K3PO4 in DMSO at 130 °C, an aliquot of the brown
solution was filtered and injected into the HRMS spectro-
meter, and a clear peak at m/z = 94.9791 corresponds to
–
methanesulfonate species (CH3SO3 ; see Figure S1), which
is one of the thermal decomposition products of the dimsyl
anion.[15] Similar brownish mixtures were observed for most
catalytic reactions. Therefore, on the basis of the clear posi- electron ionization (EI) source was set at 70 eV. High resolution
mass spectra (HRMS) were recorded with a Bruker MicrOTOF-Q
IITM instrument with ESI or Cryospray ionization sources at the
Serveis Tècnics de Recerca of the University of Girona. Samples
were introduced into the mass spectrometer ion source by direct
injection through a syringe pump and were externally calibrated by
using sodium formate.
tive effect of DMSO on our catalysis, we favor that the
dimsyl anion is produced in situ and can act as the actual
base for the deprotonation of the nucleophiles and as the
reducing agent in the experiment with CuII salts instead of
CuI salts; we also cannot exclude the possibility that it par-
ticipates as an anionic ligand for the catalytically active
copper(I) species, together with neutral DMSO molecules
and deprotonated nucleophiles.
In comparison to other methods reported, our procedure
avoids the use of a large excess of strong bases such as KOH
and, thus, tolerates functional groups sensitive to basic hy-
drolysis (Table S2). This is exemplified by the shorter reac-
tion times and higher yields for the arylation of benzamide.
In general, the ligand- and metal-free procedure reported
by Yus and co-workers does not require the use of copper
catalyst, but 48–96 h reaction times are required and the
efficacy of their method for the arylation of amides is not
documented (with the exception of benzamide, 68% yield,
72 h at 120 °C).[10b] On the other hand, the method used by
Nageswar and co-workers combines KOH as base, DMSO
as solvent, and CuO supported on alumina as the catalyst
and affords the arylation of phenols in good yields with
reaction times below 24 h, but the nucleophile scope re-
ported is limited to phenols.[10a]
General Procedure for Catalytic Experiments: A vial was loaded
with the base (3.5 mmol) and the solid nucleophile (3.5 mmol).
Then, in an inert-atmosphere glovebox, copper(I) (10 mol-%) in
DMSO and the aryl iodide (1.8 mmol) were added. Liquid nucleo-
philes were added after the aryl iodide. The vial was sealed, and the
reaction mixture was kept under an inert atmosphere and placed in
a preheated oil bath at the required temperature. After the reaction
mixture was stirred for 24 h, 1,3,5-trimethoxybenzene (400 μL,
1.5 m in DMSO) as internal standard was added. Subsequently, the
reaction was quenched by the addition of AcOEt (10 mL). The
workup consisted of the filtration of 400 μL of the crude product
through silica gel with AcOEt as eluent. All samples were analyzed
by gas chromatography. The GC yields were obtained through cali-
bration curves obtained with authentic sample of all products with
1
1,3,5-trimethoxybenzene as the internal standard. H NMR spec-
troscopic yields were obtained for very volatile products, also with
1,3,5-trimethoxybenzene as the internal standard.
Supporting Information (see footnote on the first page of this arti-
cle): Catalysis optimization tables and full characterization data for
all compounds.
Acknowledgments
Conclusions
We have developed an effective, general, and straightfor-
ward CuI-catalyzed ligand-free method for the arylation of
a wide range of amides, alcohols, and amines at moderate
temperatures with para-substituted iodobenzene derivatives.
Furthermore, bromobenzene is also tolerated for cross-cou-
pling with phenols and amines. Moreover, we disclosed the
possibility to discriminate between these functional groups
in a rational way. As no auxiliary ligands are required and
CuI is cheap and widely available, this practical method will
likely find direct application in current organic synthesis.
More mechanistic investigations are underway in our labo-
ratory to fully understand these copper-catalyzed couplings.
The authors acknowledge financial support from the European Re-
search Council for Starting Grant Project (ERC-2011-StG-277801),
from the Spanish Ministerio de Economía y Competitividad (Con-
solider-Ingenio CSD2010-00065, INNPLANTA project INP-2011-
0059-PCT-420000-ACT1) and from the Generalitat de Catalunya
(2009SGR637). X. R. also thanks the Institució Catalana de Re-
cerca i Estudis Avançats (ICREA) for an ICREA-Acadèmia award.
[1] a) G. Evano, N. Blanchard, M. Toumi, Chem. Rev. 2008, 108,
3054–3131; b) F. Monnier, M. Taillefer, Angew. Chem. Int. Ed.
2009, 48, 6954–6971; Angew. Chem. 2009, 121, 7088; c) E. N.
Pitsinos, V. P. Vidali, E. A. Couladouros, Eur. J. Org. Chem.
2011, 1207–1222; d) A. Imramovsky, M. Pesko, K. Kralova,
M. Vejsova, J. Stolarikova, J. Vinsova, J. Jampilek, Molecules
2011, 16, 2414–2430.
[2] F. Ullmann, J. Bielecki, Ber. Dtsch. Chem. Ges. 1901, 34, 2174.
[3] a) J. P. Wolfe, S. Wagaw, J.-F. Marcoux, S. L. Buchwald, Acc.
Chem. Res. 1998, 31, 805–818; b) J. F. Hartwig, Acc. Chem.
Res. 1998, 31, 852–860.
Experimental Section
General Methods: The reagents and solvents used were commer-
cially available reagent quality unless indicated otherwise. Solvents
were purchased from SDS and were purified and dried by passing
them through an activated alumina purification system (MBraun
SPS-800). The preparation and handling of air-sensitive materials
[4] a) I. P. Beletskaya, A. V. Cheprakov, Coord. Chem. Rev. 2004,
248, 2337–2364; b) D. Ma, Q. Cai, Acc. Chem. Res. 2008, 41,
1450–1460; c) D. S. Surry, S. L. Buchwald, Chem. Sci. 2010, 1,
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
7

Job/Unit: O40033
/KAP1
Date: 14-04-14 10:32:14
Pages: 9
I. Güell, X. Ribas
FULL PAPER
13–31; d) A. Casitas, X. Ribas, Chem. Sci. 2013, 4, 2301–2318;
e) C. Sambiagio, S. P. Marsden, A. J. Blacker, P. C. McGowan,
Chem. Soc. Rev. 2014, DOI: 10.1039/c3cs60289c.
a) G. O. Jones, P. Liu, K. N. Houk, S. L. Buchwald, J. Am.
Chem. Soc. 2010, 132, 6205–6213; b) S. E. Creutz, K. J. Lotito,
G. C. Fu, J. C. Peters, Science 2012, 338, 647–651.
[9] a) D. Maiti, S. L. Buchwald, J. Org. Chem. 2010, 75, 1791–
1794; b) X. Lv, W. Bao, J. Org. Chem. 2007, 72, 3863–3867.
[10] a) K. Swapna, S. N. Murthy, M. T. Jyothi, Y. V. D. Nageswar,
Org. Biomol. Chem. 2011, 9, 5978–5988; b) R. Cano, D. J.
Ramón, M. Yus, J. Org. Chem. 2011, 76, 654–660.
[11] a) C. Qi, X. Sun, C. Lu, J. Yang, Y. Du, H. Wu, X.-M. Zhang,
J. Organomet. Chem. 2009, 694, 2912–2916; b) G. A. Russell,
E. G. Janzen, E. T. Strom, J. Am. Chem. Soc. 1964, 86, 1807–
1814.
[5]
[6]
Y. Zhang, X. Yang, Q. Yao, D. Ma, Org. Lett. 2012, 14, 3056–
3059.
[7] a) L. Zhu, G. Li, L. Luo, P. Guo, J. Lan, J. You, J. Org. Chem.
2009, 74, 2200–2202; b) E. Sperotto, J. G. de Vries, G. P. M.
van Klink, G. van Koten, Tetrahedron Lett. 2007, 48, 7366–
7370; c) M. Taillefer, N. Xia, A. Ouali, Angew. Chem. Int. Ed.
2007, 46, 934–936; Angew. Chem. 2007, 119, 952; d) N. Panda,
A. K. Jena, S. Mohapatra, S. R. Rout, Tetrahedron Lett. 2011,
52, 1924–1927; e) R. Zhu, L. Xing, X. Wang, C. Cheng, D. Su,
Y. Hu, Adv. Synth. Catal. 2008, 350, 1253–1257; f) A. Correa,
C. Bolm, Adv. Synth. Catal. 2007, 349, 2673–2676; g) J. W. W.
Chang, X. Xu, P. W. H. Chan, Tetrahedron Lett. 2007, 48, 245–
248; h) Y. Yuan, I. Thome, S. H. Kim, D. T. Chen, A. Beyer, J.
Bonnamour, E. Zuidema, S. Chang, C. Bolm, Adv. Synth. Ca-
tal. 2010, 352, 2892–2898.
[12] a) T. Okitsu, K. Iwatsuka, A. Wada, Chem. Commun. 2008,
6330–6332; b) D. P. Phillips, X.-F. Zhu, T. L. Lau, X. He, K.
Yang, H. Liu, Tetrahedron Lett. 2009, 50, 7293–7296.
[13] a) Y. Liu, Q. Zhang, X. Ma, P. Liu, J. Xie, B. Dai, Z. Liu, Int.
J. Org. Chem. 2013, 3, 185–189; b) M. Yang, F. Liu, J. Org.
Chem. 2007, 72, 8969–8971.
[14] a) G. A. Russell, G. J. Mikol, J. Am. Chem. Soc. 1966, 88, 5498–
5504; b) J. C. Trisler, C. S. Aaron, J. L. Frye, J. Y. Park, J. Org.
Chem. 1968, 33, 1077–1080; c) J. I. Hayami, N. Ono, A. Kaji,
Tetrahedron Lett. 1970, 11, 2727–2728; d) B. A. Trofimov, Sul-
fur Rep. 1992, 207.
[15] E. J. Corey, M. Chaykovsky, J. Am. Chem. Soc. 1965, 87, 1345–
1353.
[8] R. P. Vignes, Abstr. Pap. Am. Chem. Soc. 2000, 220, U426–
U426.
Received: January 8, 2014
Published Online: ᭿
8
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0

Job/Unit: O40033
/KAP1
Date: 14-04-14 10:32:14
Pages: 9
Ligand-Free Ullmann-Type C–Heteroatom Couplings
Cross-Coupling
We describe a new easy-to-use method for
the ligand-free copper-catalyzed arylation
of amides, alcohols, and amines that af-
fords good cross-coupling product yields
for a wide range of substrates (bromo-
benzene and iodobenzene derivatives) and
nucleophiles.
I. Güell, X. Ribas* ............................ 1–9
Ligand-Free Ullmann-Type C–Heteroatom
Couplings Under Practical Conditions
Keywords: Copper / Cross-coupling / Lig-
and-free catalysis / Ullmann coupling / Di-
methyl sulfoxide
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
9