Communication
Organometallics, Vol. 29, No. 7, 2010 1519
strictly limited to a 2D-network due to the unique chemical pro-
perties of the copper(II) nitrate as connector. Hence, to achieve
more diverse structures and a more generalized synthetic route for
active metal site-containing networks, an improved and syste-
matic protocol was needed. To this end, it was rationalized that
two kinds of independent metal sources were compulsory: the
first to induce construction of the supramolecular networks by
connecting the building blocks, and the second to form the active
metal sites by reaction with the building blocks (Scheme 1d).
To realize this approach, new copper reagents having
reasonable reactivity toward imidazolium salts were requisite
to form the NHC-copper complex. Usually, in the literature,
NHC-copper complexes are prepared via two synthetic
routes:8,9 (i) treatment of imidazolium salts with an appro-
priate base and sequential reaction with copper sources to
yield the NHC-copper complexes;8 (ii) reaction of silver oxide
(Ag2O) with imidazolium salts to form the NHC-silver com-
plex, whereby silver moieties are replaced with copper sources
to form the NHC-copper complexes9 (Scheme 2).
In contrast to the literature methods, during the screening
of the reactivity of diverse copper reagents toward imidazo-
lium salts in the absence of base, we found that Cu2O reacted
directly with the imidazolium salts to form the NHC-copper
complex.10 As far as we are aware, this discovery was not
reported in the literature and can beapplied asa new and more
general synthetic route for copper-NHC complexes. Table 1
summarizes the reactivity of Cu2O toward several imidazo-
lium salts to form the corresponding NHC-copper complexes.
Cu2O showed very similar chemical properties to Ag2O
toward the imidazolium salts. It has been well established
that Ag2O possess good reactivity toward imidazolium
halides in the absence of base to form NHC-silver halide
complexes.11 As shown in entries 1-5 in Table 1, when the
counterions were halides, the NHC-copper complexes were
successfully formed in good yields, indicating that the cop-
per-halide bond formation is an important driving force in
this reaction. It is noteworthy that the same behavior was
observed for Ag2O.11 Powdery Cu2O is not soluble in con-
ventional organic solvents; thus, isolation of the product
Scheme 2. Synthetic Routes for NHC-Copper Complexes
from the reaction mixture was quite straightforward: simple
filtration to remove the unreacted Cu2O and evaporation of
the solvent to result in formation of microcrystalline solids.
Recently, 1,3-bis(2,6-diisopropylphenyl)imidazoline-copper
halides, denoted as (IPr)CuX, have shown unprecedented
reactivity in diverse organic transformations.12 Presently,
1,3-bis(2,6-diisopropylphenyl)imidazoline-copper halides
are also commercially available.13 Thus, we believe that
Cu2O routes can be effectively applied for mass production
of these complexes. In addition, in this work, two new NHC-
copper complexes (compounds 1 and 2), in entries 6 and 7 in
Table 1, were prepared and fully characterized.10 Figure S1
in the Supporting Information shows the single-crystal
X-ray structure of compound 2.14
Next, we applied the reaction between imidazolium salts and
Cu2O for the decoration of a 3D-supramolecular network with
coordinatively unsaturated NHC-copper moieties. Using Cu2O
as an NHC-Cu inducing reagent in the self-assembly process of
L1, the systematic approach for decoration of supramolecular
networks with NHC-copper species was developed; Scheme 3
summarizes the overall observations.
During screening of the diverse cationic connectors in the
self-assembly of L1, a 3D-network was formed when cerium-
(III) nitrate was used as a connector.15 As shown in Figure 1,
two cerium ions coordinate to eight carboxylate groups in the
building block. Four carboxylates were coordinated to one
cerium ion, and the other four carboxylates functioned as
bridging ligands by coordinating with the two cerium ions.
The coordination pattern of the eight building blocks resulted
in formation of four helical chains around two cerium ions.
ꢁ
(8) (a) Dıez-Gonzalez, S.; Stevens, E. D.; Nolan, S. P. Chem. Com-
mun. 2008, 4747. (b) Jurkauskas, V.; Sadighi, J. P.; Buchwald, S. L. Org.
Lett. 2003, 5, 2417.
(9) (a) Arnold, P. L.; Scarisbrick, A. C.; Blake, A. J.; Wilson, C.
Chem. Commun. 2002, 2340. (b) Hu, X.; Castro-Rodriguez, I.; Meyer, K. J.
Am. Chem. Soc. 2003, 125, 12237. (c) Hu, X.; Castro-Rodriguez, I.; Hu, X.;
Castro-Rodriguez, I.; Olsen, K.; Meyer, K. Organometallics 2004, 23, 755.
(d) Larsen, A. O.; Leu, W.; Oberhuber, C. N.; Campell, J. E.; Hoveyda, A. H.
J. Am. Chem. Soc. 2004, 126, 11130. (e) Alexakis, A.; Winn, C. L.; Guillen,
F.; Pytkowicz, J.; Roland, S.; Mangeney, P. Adv. Synth. Catal. 2003, 345,
345.
(12) Selected examples: Jurkauskas, V.; Sadighi, J. P.; Buchwald, S.
L. Org. Lett. 2003, 5, 2417. (b) Liu, R.; Herron, S. R.; Fleming, S. A. J. Org.
ꢁ
Chem. 2007, 72, 5587. (c) Díez-Gonzalez, S.; Stevens, E. D.; Nolan, S. P.
(10) Typical procedure for the synthesis of NHC-copper complexes:
imidazolium salt (0.19 mmol) and Cu2O (0.14 mmol) in 4 mL of dioxane
were heated at 100 °C for 16 h. Then the reaction mixture was cooled to
room temperature. The unreacted Cu2O was removed by filtration, and
the remaining solvent was evaporated to form precipitates. The solid was
washed with water to remove any salts and then dried under vacuum.
The spectroscopic characterization data of (IPr)CuX in entries 1-3
matched well with those in the literature (ref 12). Characterization data
of compound 1: 1H NMR (300 MHz, CDCl3) δ 7.60 (s, 4H), 7.11 (s, 2H),
2.45 (septet, J = 6.9 Hz, 4H), 1.27 (d, J = 6.9 Hz, 12H), 1.20 (d, J = 6.9,
12H) ppm; 13C NMR (75 MHz, CDCl3) δ 180.6, 147.8, 134.2, 133.9,
123.4, 97.6, 28.8, 24.7, 23.8 ppm. Anal. Calcd for C27H34N2I2CuCl: C,
43.86; H, 4.63. Found: C, 43.91; H, 4.48. Characterization data of
compound 2: 1H NMR (300 MHz, CDCl3) δ 7.99 (s, 4H), 7.19 (s, 2H),
4.00 (s, 6H), 2.58 (septet, J = 6.9 Hz, 4H), 1.33 (d, J = 6.6 Hz, 12H), 1.27
(d, J = 6.9, 12H) ppm; 13C NMR (75 MHz, CDCl3) δ 180.2, 166.4,
146.2, 137.9, 132.3, 125.8, 123.3, 52.5, 29.0, 24.7, 23.8 ppm. Anal. Calcd
for C31H40O4N2CuCl: C, 61.68; H, 6.68. Found: C, 61.70; H, 6.60.
(11) Garrison, J. C.; Youngs, W. J. Chem. Rev. 2005, 105, 3978.
Chem. Commun. 2008, 4747. (d) Liu, J.; Zhang, R.; Wang, S.; Sun, W.; Xia,
C. Org. Lett. 2009, 11, 1321.
(13) Catalogue of Aldrich Chemical Co. and STREM Chemicals Inc.
(14) The crystals of compound 1 have a plate-like morphology and
were not suitable for single-crystal X-ray analysis. Crystallographic data
for compound 2: Mr = 603.64, monoclinic, space group C2/c, a =
15.847(4) A, b = 9.503(2) A, c = 21.959(4) A, β = 106.577(6)o, V =
3170(1) A3, Z = 4, F = 1.265 g cm-3, completeness 96.3%, 13 113
collected reflections, 3818 crystallographically independent reflections
(Rint = 0.0560), R1 = 0.0557 (I > 2σ(I)), wR2 = 0.1509 (all data).
(15) L1 (0.113 g, 0.220 mmol) and cerium nitrate (0.115 g, 0.265
mmol) were dissolved in DMF (4 mL), and the reaction mixture was
heated at 110 °C for 3 days. The colorless crystal was formed on the wall
of the vial. Crystal data for (L1-Cl)2CeNO3: Mr = 1153.31, monoclinic,
space group P21/c, a = 14.868(4) A, b = 25.174(6) A, c = 22.523(6) A,
β = 98.458(4)o, V = 8338(4) A3, Z = 4, F = 0.919 g cm-3, completeness
99.1%, 30 887 collected reflections, 8508 crystallographically indepen-
dent reflections (Rint = 0.0794), R1 = 0.0500 (I > 2σ(I)), wR2 = 0.1446
(all data).