Improved copper(I)-NHC catalytic efficiency on huisgen reaction by addition of aromatic nitrogen donors

Article abstract of DOI:10.1002/chem.200900727

A study was conducted to demonstrate copper(I)-NHC catalytic efficiency improvement on Huisgen reaction by adding aromatic nitrogen donors. Several unsaturated nitrogen heterocycles such as l-histidine, a ubiquitous amino acid residue associated to all ty

Full text of DOI:10.1002/chem.200900727

DOI: 10.1002/chem.200900727
Improved Copper(I)–NHC Catalytic Efficiency on Huisgen Reaction by
Addition of Aromatic Nitrogen Donors
Marie-Laure Teyssot,[a] Aurꢀlien Chevry,[a] Mounir Traꢁkia,[a] Malika El-Ghozzi,[b]
Daniel Avignant,[b] and Arnaud Gautier*[a]
Dedicated to Professor Istvꢀn E. Markꢁ
The recent emergence of “click chemistry” as a new para-
digm for organic synthesis, invoking only simple, high-yield-
ing and easily workable transformations, has facilitated an
extraordinary increase in the number of molecules available
for medicinal chemistry, biology, and material science.[1]
Among these synthetic “click tools”, the Huisgen-catalyzed
cycloaddition, also called copper(I)-catalyzed azide alkyne
cycloaddition (CuAAC), has received unrivalled attention.[2]
Efficient CuAAC catalysts consist of a combination of
copper and amino ligands (TBTA, (BimC4A)3, bathophe-
nanthroline) in the presence of a sacrificial reducing reagent
(Scheme 1).[3] Detailed studies have demonstrated the re-
cruitment of a bimetallic species and showed that the li-
gands strongly accelerate the process.[3b,c]
Very few systems are effective on CuAAC without the in-
tervention of a reducing reagent. For example, Vincent and
co-workers took advantage of the high stability of [Cu-
AHCTUNGRTEG(NNUN C186tren)]Br in the presence of oxygen to synthesize 1,4-
triazoles at low catalyst loading with a high turnover
number in solution (Scheme 2).[4] Recently, Nolan and Dꢀez
have reported that N-heterocyclic carbenes (NHCs) are suit-
able ligands for copper(I) catalysis.[5] In a first report,
among the copper(I)–NHCs screened, they showed that
[CuXACTHUNRTGNEUNG(SIMes)] (X=Cl, Br) functions particularly well under
neat conditions; alternatively, the reaction can be carried
out “on water”.[5a] The Huisgen cycloaddition is an highly
exothermic reaction and, for obvious safety reasons, reac-
tions conducted in hydro-alcoholic solvents would be pref-
erable, especially for perform-
ing large-scale synthesis.
A
second article reports the im-
pressive catalytic activity of
[CuACHTNURTGNE(UNG Icy)2]PF6 that functions at
50 ppm loading, but still under
neat conditions.[5b] Nevertheless,
this catalyst suffers from the
same safety limitations as
[CuX
(SIMes)].
Furthermore,
Scheme 1. CuAAC accelerating ligands.
neat conditions are not adapted
to bioconjugation.
Closer examination of the
binding mode of TBTA and (BimC4A)3 reveals that they
share a central nitrogen s-donor ligand connected to three
[a] Dr. M.-L. Teyssot, A. Chevry, Dr. M. Traꢁkia, Dr. A. Gautier
Clermont Universitꢂ, SEESIB, UMR CNRS 6504
Universitꢂ Blaise Pascal
24 Avenue des Landais, 63177 Aubiꢃre CEDEX (France)
Fax : (+33)473-40-77-17
E-mail: Arnaud.Gautier@univ-bpclermont.fr
[b] Dr. M. El-Ghozzi, Dr. D. Avignant
Clermont Universitꢂ, LMI, UMR CNRS 6002
Universitꢂ Blaise Pascal
24 Avenue des Landais, 63177 Aubiꢃre CEDEX (France)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200900727.
Scheme 2. Oxygen stable copper(I) catalysts.
6322
ꢄ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 6322 – 6326

COMMUNICATION
N-donor aromatic ligands. The helpful effect from N-donor
aromatic ligands is also illustrated by the acceleration ach-
ieved by using bathophenanthroline.[6]
To overcome the above-mentioned limitations of cop-
per(I)–NHC, we assumed that an association of the N-heter-
ocyclic carbene (as the strong s-donor ligand) with external
N-donor aromatic ligand(s) would increase the catalysis effi-
ciency. We thus selected several unsaturated nitrogen het-
erocycles: l-histidine, a ubiquitous amino acid residue asso-
ciated to all types of copper enzymes, N-methylimidazole
(NMI), 4-dimethylaminopyridine (4-DMAP), phenanthro-
line (Phen), and bathophenanthroline disulfonic sodium salt
(Bathophen) to evaluate their influence on the course of the
CuAAC catalyzed by [CuClACHTNUTGRNEUNG
(SIMes)] (1).[7] For this purpose,
Scheme 3. Triazoles obtained by 1/4-DMAP (1:2) or 1/Phen (1:1) cata-
we monitored the cycloaddition of benzyl azide and phenyl
lyzed CuAAC.
acetylene in a 2:1 tert-butyl alcohol/water mixture (Table 1).
Table 2. Catalyzed formation of triazoles.[a]
Table 1. Effect of additives.
2
3
4[b]
5
6
7
8
9
4-DMAP
Phen
84
78
91
77
76
78
68
97
78
71
0[c],21[d]
0[c],72[d]
77
78
72
60
[a] All reactions were performed on 2 mmol scale during 18 h in an open
vessel at room temperature using 1 mol% of 1 + 1 mol% of Phen or
2 mol% of 4-DMAP. Otherwise stated in supplementary material, prod-
ucts were isolated by simple filtration [b] In MeOH at 458C due to the
poor solubility of the alkyne in tBuOH/water. [c] Using the free acid.
[d] Using the triethylamonium salt.
Entry
Additive (ratio)
Loading
[mol%]
Time
[h]
T
[8C]
Yield
[%][a]
G
1
2
3
4
5
6
7
8
none
1
1
1
1
1
1
1
1
18
18
18
18
18
18
18
4
20
20
20
20
20
20
20
85
50
20
10
0
l-histidine (1:1)
NMI (1:1)
NMI (1:2)
4-DMAP (1:1)
76
24
63
84
78
83
73
69
Aromatic (2), electron-rich (3), electron-poor (4), and
functionalized aliphatic (5, 6) alkynes are well tolerated.
The exception is propionic acid (7). This could be expected
since propionic acid is likely to protonate the additives, thus
annihilating any favorable effect.[10] Yield was increased in
basic media (pH 9) by adding 1.1 equivalents of triethyla-
mine. The low yield obtained with 4-DMAP is probably due
to the competitive chelation of triethylamine, whereas the
bidentate phenanthroline acting as a better binder allowed
increased yield. Azidoacetic acid was also neutralized with
propargylamine that can participate directly in the catalytic
cycle; in this particular case a good yield was obtained (8).
Finally, the reaction also proceeded well with phenyl azide
as demonstrated by the formation of 9.
The rate-accelerating effect of the additives was evi-
denced by monitoring the formation of the soluble triazole 5
during the CuAAC reaction, by 1H NMR spectroscopy
(Scheme 4, Figure 1). After 7 h, a poor conversion was ob-
served for the CuAAC catalyzed by 1 (2 mol%). The bene-
ficial effect of the additive is evidenced by the simple addi-
tion of an equimolar amount of phenanthroline (2 mol%)
into the solution. Subsequently, immediate formation of tria-
zole 5 was observed.
4-DMAP
N
Phen (1:1)
Phen (1:1)
Phen (1:1)
Bathophen (1:1)
9
10
0.1
1
72
18
[a] Refers to pure isolated compounds.
Table 1, entry 1 shows the poor conversion obtained by
using 1 alone in solution. Whereas histidine acted as a dele-
terious additive (Table 1, entry 2), pleasingly, the addition of
one equivalent of NMI per copper–NHC afforded a good
isolated yield (Table 1, entry 3). Unexpectedly, the addition
of a second equivalent considerably decreased the yield
(Table 1, entry 4).[8] Regarding 4-DMAP, optimum efficiency
was reached for a ratio of ligand/1 of 2:1 (Table 1, entries 5
and 6). A beneficial effect is also observed with the biden-
tate ligands phenanthroline (Table 1, entry 7), and batho-
phenanthroline (Table 1, entry 10). Importantly, the reaction
also proceeds efficiently when the solvent was refluxed
(858C, Table 1, entry 8). This highlights the well recognized
ability of the NHC to protect copper(I) from oxidation. We
were also satisfied to observe that the catalyst loading may
be decreased to 0.1 mol% with an extended reaction time
(Table 1, entry 9).[9]
We then compared the efficiency of Phen, NMI, and 4-
DMAP (1 mol%) with the well known TBTA. For this,
TBTA and copper sulfate (1:1) were mixed in the presence
of 10 mol% of ascorbic acid, and completion of the reaction
occurred after about 12 h. Comparison of the reaction with
the different additives (without the addition of any reducing
Next, the scope of the reaction was examined with phe-
nanthroline and 4-DMAP as additives (Scheme 3, Table 2).
Chem. Eur. J. 2009, 15, 6322 – 6326
ꢄ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6323

A. Gautier et al.
plexes.[12] A comparison of the important structural features
of 10 and 1 is given in Table 3. Interestingly, the carbene–
copper bond remains unchanged, whereas the geometrical
switch from 1 to 10 results in an elongation of the copper–
chloride bond.
Inspection of the 1H and 13C NMR spectra of 1 in the
presence of additives revealed an extreme broadening of
Phen, 4-DMAP, or NMI signals,
Scheme 4. Catalyzed CuAAC formation of 5.
whereas the carbene signature
remained unchanged. This typi-
cal ligand-exchange phenomen-
on occurring at the NMR time
scale prompted us to use UV/
Vis spectroscopy to gain more
information. Among all ligands,
phenanthroline is unique be-
cause the typical copper(I)-to-
ligand d!p* charge-transfer
band can be clearly observed at
450 nm in dichloromethane
(Figure 3). This property ena-
bled the determination of both
the complex stoichiometry and
the value of the association
constant of phenanthroline with
[CuClACHTNUGRTNEUNG
(SIMes)].[13]
Evaluation of the complex
stoichiometry by Jobꢅs continu-
ous variation method, in di-
chloromethane, revealed the
Figure 1. Kinetic study of the catalyzed formation of triazole 5.
reagent) enabled us to establish a classification based on re-
action rates: 4-DMAP> Phen>TBTA>NMI. Impressively,
4-DMAP afforded 100% completion in only 1.7 h, whilst
6 h are necessary for Phen, and 12 h for TBTA and NMI.
Gratifyingly, we were able to obtain red single crystals of
Table 3. Representative distances and angles.
Bond lengths [ꢆ]
3
4
1
ꢀ
ꢀ
ꢀ
ꢀ
Cu Cl
2.347(2)
2.099(1)
Cu N
Cu N
Cu C
the complex [CuClACHTUNGTRENNUNG(Phen)ACHUTNGTREN(NNGU SIMes)] (10), suitable for X-ray
10
1
2.114(5)
–
2.128(5)
–
1.916(7)
1.882(4)
analysis, by saturating a solution of the complex in dichloro-
methane with diethyl ether. Representations of the structure
are depicted in Figure 2 (hydrogen atoms are omitted for
clarity).[11]
The copper atom lies in a distorted tetrahedral environ-
ment, which is in deep contrast with the linear structure of
Bond angles [8]
N3-Cu-N4
N4-Cu-Cl
99.9(1)
–
N3-Cu-Cl
102.0(2)
–
C1-Cu-Cl
120.5(2)
178.5(1)
10
1
77.4(2)
–
[CuCl
ACHTUNGTRENNUNG(SIMes)]. This is not surprising as the same type of ge-
ometry is found for mixed s, N-donor aromatic ligands com-
Figure 3. Copper(I) charge-transfer band increasing with the amount of
Figure 2. Views of [CuCl
G
G
phenanthroline (Curves from down to up: [Phen]/[1] from 0 to 50).
6324
www.chemeurj.org
ꢄ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 6322 – 6326

Improved Copper(I)–NHC Catalytic Efficiency
COMMUNICATION
Scheme 5. Equilibrium between 1, Phen and 10.
Figure 5. Titration curve of complex 10, in CH2Cl2.
Table 4. Stoichiometry, association constants of 10, in CH2Cl2.
Stoichiometry Jobꢅs plot
K [mꢀ1
]
K [mꢀ1
]
Creswell-Allred
Benesi–Hildebrand
1:1
254ꢂ7
241ꢂ3
likely that 10 may simultaneously accommodate azide and
alkyne in a five-coordinate copper(I) complex and thus
function as the sole catalytic species. Indeed, it seems rather
difficult to propose an exact picture of the species involved,
because of the particular complexity of the alkyne mode of
complexation with copper(I). The most frequent arrange-
ment reported includes three CuI ions in a tetrahedral ar-
rangement, but bimetallic complexes can also be found. In
Figure 4. Stoichiometry: Jobꢅs plot for complex 10.
ꢀ
ꢀ
typical formation of a 1:1 (host:guest) complex in agreement
with our X-ray data (Scheme 5, Figure 4, and Table 4).
Surprisingly, 10 is a weak complex, based on a comparison
(Figure 5 and Table 4) with literature values reported for
the particular case of [(IPr)Cu CꢁC Ph], a 1:1 copper/
acetylene stoichiometry has been proposed, but to the best
of our knowledge, no X-ray structure analysis has been re-
ported.[15] For the moment, the exact nature of the copper
acetylene complex 11 formed from 1 and phenyl acetylene
[CuACHTUNGTRENNUNG ACHTUNGTRENNUGN
(Phen)]+ and [Cu(Phen)2]+.[14] Inspection of the distan-
ces deduced from single-crystal
X-ray diffraction rules out the
involvement of a steric effect to
account for this low association.
The weakness of complex 10
could be attributed to different
electronic effects from the two
ligands. It is important to re-
member here that a weak asso-
ciation of copper with its li-
gands is the key feature of the
rate-accelerating effect dis-
played by tris-triazole, tris-ben-
zimidazole, and related li-
gands.[3]
Regarding
the
CuAAC
mechanism (Scheme 6), we pro-
pose that additional ligands
form the saturated complex 10
(in which the chloride may be
substituted by a solvent mole-
cule). At this point, it seems un- Scheme 6. Outline of the assumed mechanism.
Chem. Eur. J. 2009, 15, 6322 – 6326
ꢄ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6325

A. Gautier et al.
Importantly, histidines, and some derivatives in stoichiometric quan-
tities, considerably accelerate the CuAAC reaction: b) K. Tanaka,
C. Kageyama, K. Fukase, Tetrahedron Lett. 2007, 48, 6475–6479.
Beneficial effects of the imidazole moiety are also encountered in
copper aerobic oxidation of alcohols: c) I. E. Markꢈ, A. Gautier, R.
Dumeunier, K. Doda, F. Phillipart, S. M. Brown, C. J. Urch, Angew.
Chem. 2004, 116, 1614–1617; Angew. Chem. Int. Ed. 2004, 43, 1588–
1591; for bioinspired oxidation catalysts: d) L. Que, Jr., W. B.
Tolman, Nature 2008, 455, 333–340.
remains unclear. Therefore, it is more likely that the phe-
nanthroline-containing complex 10 participates in the cata-
lytic cycle as an additional copper center (whatever the
nature of 11), which associates through a p coordination of
the acetylide in the rate-determining step.[3,16]
In conclusion, we report that simple addition of aromatic
amines increases CuAAC catalytic activity of [CuClACTHNUGRTNEUNG(SIMes)]
at a large range of temperatures in such a way that efficient
catalysis can safely take place in hydro-alcoholic solvents.
Phenanthroline and 4-DMAP are the additives of choice be-
cause the intrinsic stability of copper(I)–NHC may be main-
tained along the whole process, allowing a homogeneous
CuAAC to proceed without the intervention of a reducing
reagent. Current research in our laboratory focuses on the
screening of a larger collection of additives and on delineat-
ing their effects on the CuAAC reaction catalyzed by cop-
per(I)–NHC.
[8] A red precipitate forms rapidly using excess of NMI. A possible de-
composition path could involve oxidation by molecular oxygen.
Indeed, we have recently demonstrated that [CuClACTHUNRTGNEUNG(SIMes)] acts as a
Fenton reagent. See: M.-L Teyssot, A. S. Jarrousse, A. Chevry, A.
De Haze, C. Beaudoin, M. Manin, S. P. Nolan, S. Dꢀez-Gonzꢇlez, L.
Morel, A. Gautier, Chem. Eur. J. 2009, 15, 314–318.
[9] Blank experiment revealed that no trace of product was formed
without catalyst at 858C; only the 1–4 regioisomer was observed.
[10] Interestingly, 7 has been synthesized using [CuClACTHNUGRTENUNG(SIMes)] in solvent-
less conditions: A. Maisonial, P. Serafin, M. Traꢁkia, E. Debiton, V.
Thꢂry, D. J. Aitken, P. Lemoine, B. Viossat, A. Gautier, Eur. J.
Inorg. Chem. 2008, 298–305. This reinforces the assumption that
protonation of nitrogen additives annihilates the catalytic activity.
[11] Crystal data for 10: C33H34ClCuN4, Mr =585.65, trigonal, space
3
¯
group R3, a=42.0084(19), c=9.3049(4) ꢆ, V=14220.5(1) ꢆ , Z=
Acknowledgements
ACTHNUTRGNEUNG
18, 1calcd =1.231 gcmꢀ3, m=0.80 mmꢀ1, T=293(2) K, R(F2>2s(F2))=
0.053, Rw (F2, all data)=0.209, S=1.10 for 3274 unique data (q<
20.88; Bruker APEX-II CCD diffractometer, Moka radiation, l=
0.71073 ꢆ) and 358 refined parameters; final difference synthesis
The authors thank Rachid Mahiou, Lionel Nauton, Oscar and Sara Mam-
moliti for helpful discussions.
D1max =0.54eꢆꢀ3
,
D1min =ꢀ0.63eꢆꢀ3
. CCDC-720889 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
Keywords: azides
· click chemistry · cycloaddition ·
homogeneous catalysis · N-heterocyclic carbenes
[12] Selected examples; a)
McMillin, W. R. Robinson, D. R. Powell, A. T. McKenzie, S. Chen,
Inorg. Chem. 1985, 24, 3928–3933; b) [ClCu(Bipy)(PPh3)]: B. E.
Green, C. H. L. Kennard, G. Smith, B. D. James, A. H. White, Acta
Crystallogr. Sect. 1984, 40, 426–428; c) [ClCu-
(Phen)(PPh2CH2OH]: Q.-Y. Cao, W.-F. Fua, Z.-L Wang, Acta Crys-
tallogr. Sect. E 2004, 60, 987–989.
(PPh3)2]+ ion: J. R. Kirchhoff, D. R.
ACHUTGTNRNEUG[N CuPhenACHTUGNTRENNUGN
A
R
ACHTUNGTRENNUNG
[1] a) C. W. Tornøe, M. Meldal in Peptides: The Wave of the Future
(Eds.: M. Lebl, R. A. Houghten), Kluwer Academic, Dordrecht,
2001, pp. 263–264; b) H. C. Kolb, M. G. Finn, K. B. Sharpless
Angew. Chem. 2001, 113, 2056–2075; Angew. Chem. Int. Ed. 2001,
40, 2004–2021; c) C. W. Tornøe, C. Christensen, M. Meldal, J. Org.
Chem. 2002, 67, 3057–3064.
[2] a) M. Meldal, C. W. Tornøe, Chem. Rev. 2008, 108, 2952–3015;
b) V. D. Bock, H. Hiemstra, J. H. van Maarseveen, Eur. J. Org.
Chem. 2006, 51–68; c) H. C. Kolb, K. B. Sharpless, Drug Discovery
Today 2003, 8, 1128–1137.
C
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
[13] a) K. Hirose, J. Inclusion Phenom. Macrocyclic Chem. 2001, 39,
193–209; b) L. Fielding, Tetrahedron 2000, 56, 6151–6170; c) H. A.
Benesi, J. H. Hildebrand, J. Am. Chem. Soc. 1949, 71, 2703–2707;
d) C. G. Creswell, A. L. Allred, J. Phys. Chem. 1962, 66, 1469–1472;
e) P. Job, Ann. Chim. 1928, 9, 113.
[14] Copper(I) and phenanthroline are well known to form 1:2 com-
plexes. For a recent determination of these constants, (log K1 =6
and log K2 =5.) see: S. V. Pakhomova, M. A. Proskurnin, V. V. Cher-
nysh, M. Yu. Kononets, E. K. Ivanova, J. Anal. Chem. 2001, 56, 910–
917.
[3] a) V. O. Rodionov, S. I. Presolski, S. Gardinier, Y.-H. Lim, M. G.
Finn, J. Am. Chem. Soc. 2007, 129, 12696–12704; b) V. O. Rodionov,
S. I. Presolski, D. Daz Daz, V. V. Fokin, M. G. Finn, J. Am. Chem.
Soc. 2007, 129, 12705–12712; c) M. Ahlquist, V. V. Fokin, Organo-
metallics 2007, 26, 4389–4391.
[15] a) L. A. Goj, E. D. Blue, C. Munro-Leighton, T. B. Gunnoe, J. L. Pe-
tersen, Inorg. Chem. 2005, 44, 8647–8649; b) see also ref. [5b].
[16] a) In our experimental conditions, the charge transfer band remains
measurable during the reaction progress, attesting to the presence of
complex 10 in solution; b) for a discussion on the association modes
of acetylene and copper(I) see the excellent review from Meldal,
ref. [2a]. For alternative mechanistic considerations on click chemis-
try, see ref. [2a,b].
[4] a) N. Candelon, D. Lastꢂcouꢃres, A. Khadri Diallo, J. Ruiz Ara-
nzaes, D. Astruc, J.-M. Vincent, Chem. Commun. 2008, 741–743. b)
S. Dꢀez-Gonzꢇlez, S. P. Nolan, Aldrichimica Acta 2008, 41, 43–51.
[5] a) S. Dꢀez-Gonzꢇlez, A. Correa, L. Cavallo, S. P. Nolan, Chem. Eur.
J. 2006, 12, 7558–7564; b) S. Dꢀez-Gonzꢇlez, S. P. Nolan, Angew.
Chem. 2008, 120, 9013–9016; Angew. Chem. Int. Ed. 2008, 47, 8881–
8884; c) S. Dꢀez-Gonzꢇlez, E. D. Stevens, S. P. Nolan, Chem.
Commun. 2008, 4747–4749.
[6] W. G. Lewis, F. G. Magallon, V. V. Fokin, M. G. Finn, J. Am. Chem.
Soc. 2004, 126, 9152–9153.
[7] Histidine is recognized to tune enzymeꢅs reactivity: a) E. I. Solomon,
U. M. Sundaram, T. E. Machonkin, Chem. Rev. 1996, 96, 2563–2606.
Received: March 20, 2009
Published online: May 21, 2009
6326
www.chemeurj.org
ꢄ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 6322 – 6326