Stoichiometric sensitivity and structural diversity in click-active copper(I) N,S-heterocyclic carbene complexes

Article abstract of DOI:10.1039/c3dt52059e

A series of novel Cu(I) N,S-heterocyclic carbene (NSHC) complexes [Cu(μ-Br)(NSHC)]₂, [Cu(μ-X)(NSHC)]₄ (X = Br or I), [(NSHC)₂Cu(μ-Br)₂Cu(NSHC)], and [(NSHC) ₂CuBr] have been isolated from in situ generated CuO^tBu and N-substituted benzothiazolium halides and characterized by X-ray crystallography. Five structural motifs were observed, viz. M_xL _y where x:y = 2:2, 4:4, 2:3, 1:2 and 2:4, with Cu...Cu separation traversing over a wide range of 2.5626(7) to 3.4725(7) A distances. A preliminary investigation of the catalytic activity of these compounds indicated that the unusual mononuclear complex 6 [(NSHC) ₂CuBr] is an active catalyst for the Huisgen 1,3-dipolar cycloaddition of azide and alkynes while complexes 1-5 and 7 were marginally less active. The Royal Society of Chemistry 2014.

Full text of DOI:10.1039/c3dt52059e

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Stoichiometric sensitivity and structural diversity
in click-active copper(I) N,S-heterocyclic carbene
complexes†
Cite this: Dalton Trans., 2014, 43,
1305
a,b
a,c
d,e
f
Xiaoyan Han, aZ,ehiqiang Weng, David James Young, Guo-Xin Jin and
T. S. Andy Hor*
A series of novel Cu(I) N,S-heterocyclic carbene (NSHC) complexes [Cu(μ-Br)(NSHC)]
2
, [Cu(μ-X)(NSHC)]
2 2
Cu(μ-Br) Cu(NSHC)], and [(NSHC) CuBr] have been isolated from in situ generated
4
(
X = Br or I), [(NSHC)
2
t
CuO Bu and N-substituted benzothiazolium halides and characterized by X-ray crystallography. Five
structural motifs were observed, viz. M
ration traversing over a wide range of 2.5626(7) to 3.4725(7) Å distances. A preliminary investigation of the
catalytic activity of these compounds indicated that the unusual mononuclear complex 6 [(NSHC) CuBr]
x y
L where x : y = 2 : 2, 4 : 4, 2 : 3, 1 : 2 and 2 : 4, with Cu⋯Cu sepa-
Received 30th July 2013,
Accepted 13th October 2013
2
DOI: 10.1039/c3dt52059e
is an active catalyst for the Huisgen 1,3-dipolar cycloaddition of azide and alkynes while complexes 1–5
and 7 were marginally less active.
www.rsc.org/dalton
We have previously reported the isolation of an unusual
bridging complex of Cu(I) with a close Cu⋯Cu contact. This
Introduction
8
Copper N-heterocyclic carbene (NHC) complexes are an finding suggested
a
hitherto unknown intramolecular
efficient and economical class of catalysts for a range of impor- migration of NHC across neighboring metal centers. The
tant C–C, C–H and C–N bond forming reactions of unsaturated passage of the ligand presumably requires an expandable and
1
–4
substrates. Investigations have concentrated on copper N,N- reversible coordination state. As an extension of that study and
heterocyclic carbene (NNHC) catalysts, while copper N,S- of our previous metal-NSHC work, we herein report a series of
heterocyclic carbenes (NSHC) have received little attention, Cu(I)-NSHC complexes that exhibit a remarkable range of stoi-
and there are relatively few published X-ray crystal structures. chiometry dependent coordination geometries, as determined
5
We previously prepared NSHC complexes of Ni, Cu, Au, Ru, by X-ray crystallography. Preliminary studies indicate that
Pd and Pt and demonstrated subtle differences in reactivity of these complexes are promising catalysts for the Huisgen
6
Pd(II)-NHSC catalyzed Suzuki coupling reactions. This differ- 1,3-dipolar cycloaddition of azides to alkenes.
ence may reflect the stronger metal–carbene bond that results
from a decreased carbenoid carbon stabilizing ability of S rela-
tive to N and therefore increased π back donation from
capable metals. This delicate electronic influence is manifest Experimental
in the formation of a Pt(II)-NSHC complex of stereochemistry
opposite to that of a comparable Pt(II)-NNHC complex.
General procedures
7
All experiments were performed under an inert Ar atmosphere
using a Labmate glove box or by Schlenk techniques. Tetra-
hydrofuran and diethyl ether were distilled over benzophenone
sodium-ketyl and kept under Ar. Acetone and dichloro-
a
Department of Chemistry, National University of Singapore, 3 Science Drive 3,
Kent Ridge, Singapore 117543, Singapore. E-mail: andyhor@nus.edu.sg
b
Testing and Analysis Center, Soochow University, Suzhou, 215123, P.R. China
methane were distilled over molecular sieves and CaCl
2
c
Department of Chemistry, Fuzhou University, Fuzhou, 350108, P.R. China
respectively and stored under Ar. Benzothiazole was purchased
from Sigma-Aldrich and distilled prior to use. Allyl bromide,
propyl bromide, crotyl bromide, iodomethane and allyl iodide,
d
Universiti Brunei Darussalam, Faculty of Science, Jln Tungku Link, Gadong,
BE1410, Brunei Darussalam
e
Institute of Materials Research & Engineering, A*Star, 3 Research Link, Singapore
t
1
17602, Singapore
NaO Bu and CuCl were purchased from Sigma-Aldrich and
f
Department of Chemistry, Fudan University, Shanghai, 200433, P.R. China
used as received. Phenylacetylene was purchased from Acros
†
Electronic supplementary information (ESI) available: Table S1 Selected bond
lengths (Å) for 1–7. Table S2 Selected crystal data, data collection and refinement
parameters of compounds 1–7. CCDC 952963–952971. For ESI and crystallo-
graphic data in CIF or other electronic format see DOI: 10.1039/c3dt52059e
t
and used as received. CuO Bu was prepared according to litera-
9
1
13
ture methods. H- and C-NMR spectra were recorded on
Bruker AMX 500 spectrometers using Me Si as an internal
4
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 1305–1312 | 1305

View Article Online
Paper
Dalton Transactions
standard. Elemental analyses were performed on a Perkin-
Elmer PE 2400 elemental analyzer at the Department of Chem-
istry, National University of Singapore.
Synthesis of complexes 1–7
2
Preparation of [Cu(μ-Br)(NSCH)] (NSHC = N-allylbenzothia-
t
zolin-2-ylidene) (1). A solution of NaO Bu (49 mg, 0.5 mmol)
in THF (2 mL) was added to a suspension of CuCl (50 mg,
0
.5 mmol) in THF (5 mL). The mixture was stirred for 1 h.
After filtration, a bright orange solution was obtained. A sus-
pension of N-allylbenzothiazolium bromide (128 mg,
t
0
.5 mmol) in THF (5 mL) was added to the CuO Bu solution to
stir for 4 h. Filtration through celite and concentration under
vacuum, followed by cooling of the THF solution to −20 °C
provided crystals suitable for X-ray diffraction (Fig. 1). Yield:
1
1
14 mg (0.179 mmol, 72%). H-NMR (500 MHz, DMSO-d
6
):
δ 8.11 (d, 2H, J = 8.2 Hz, Aryl-H), 7.98 (d, 2H, J = 8.2 Hz,
Aryl-H), 7.62 (t, 2H, J = 7.6 Hz, Aryl-H), 7.55 (t, 2H, J = 7.6 Hz,
2
Aryl-H), 6.11 (m, 2H, CHvCH ), 5.47 (d, J = 5.6 Hz, 4H,
CHvCH ), 5.28 (s, 2H, CH ), 5.25 (d, J = 5.7 Hz, 2H, CH ).
2
2
2
1
3
1
C-NMR ( H): δ 215.1 (NCS), 143.3 (s, Aryl-C), 134.3 (s, Aryl-C),
32.4 (s, CHvCH ), 127.7 (s, Aryl-C), 125.9 (s, Aryl-C), 123.5 (s,
1
2
Aryl-C), 119.6 (s, CHvCH ), 116.0 (s, Aryl-C), 56.7 (s, CH ).
2
2
Anal. calcd for 1, C20
H18Br
2
Cu
2
N
2
S
2
: C, 37.69; H, 2.85; N, 4.32.
Found: C, 37.73; H, 2.91; N, 4.32.
Preparation of [Cu(μ-Br)(NSCH)]2 (NSHC = N-propylbenzo-
thiazolin-2-ylidene) (2a). Compound 2a was prepared and pur-
ified using a method similar to that for 1. Yield: 125 mg
1
(
0.1 mmol, 78%). H-NMR (500 MHz, DMSO-d ): δ 8.11 (m, 8H,
6
Aryl-H), 7.64 (m, 4H, Aryl-H), 7.55(m, 4H, Aryl-H), 4.81 (m, 8H,
1
3
CH
(
2
), 1.94 (m, 8H, CH
H): δ 215.9 (NCS), 143.5 (s, Aryl-C), 134.3 (s, Aryl-C), 127.6 (s,
Aryl-C), 125.7 (s, Aryl-C), 123.5 (s, Aryl-C), 115.7 (s, Aryl-C), 55.8
s, CH ), 23.2 (s, CH CH ), 11.5 (s, CH ). Anal. calcd for 2a,
C H Br Cu N S : C, 37.45; H, 3.46; N, 4.37. Found: C, 37.76;
2 3 3
CH ), 0.94 (m, 12H, CH ). C-NMR
1
(
2
2
3
3
4
0
44
4
4 4 4
H, 3.54; N, 4.27.
Preparation of [Cu(μ-Br)(NSCH)]
4
(NSHC = N-crotylbenzo-
thiazolin-2-ylidene) (2b). Compound 2b was prepared and pur-
ified using a method similar to that for 1. Yield: 134.6 mg
1
(
0.101 mmol, 81%). H-NMR (500 MHz, DMSO-d
6
): δ 7.98 (d,
4
4
4
H, J = 8.2 Hz, Aryl-H), 7.94 (d, 4H, J = 8.2 Hz, Aryl-H), 7.59 (t,
H, J = 8.2 Hz, Aryl-H), 7.50 (t, 4H, J = 7.6 Hz, Aryl-H), 6.00 (m,
H, CHvCH), 5.83 (m, 4H, CHvCH), 5.38 (d, J = 6.3 Hz, 8H,
1
3
1
CH ), 1.72 (t, 12H, J = 6.3 Hz, CH ). C-NMR ( H): δ 212.4
2
3
(
NCS), 144.0 (s, Aryl-C), 135.5 (s, Aryl-C), 133.3 (s, CHvCH),
27.8 (s, Aryl-C), 126.1 (s, Aryl-C), 125.3 (s, Aryl-C), 123.5 (s,
CHvCH), 116.0 (s, Aryl-C), 57.1 (s, CH ), 17.9 (s, CH ). Anal.
1
2
3
Fig. 1 ORTEP view of complexes (1), (2a), (2b) and (3) with 40% thermal
ellipsoids.
calcd for 2b, C44
4 4 4 4
H44Br Cu N S : C, 39.71; H, 3.33; N, 4.21.
Found: C, 39.53; H, 2.84; N, 4.20.
Preparation of [Cu(μ-I)(NSCH)]4 (NSHC = N-methylbenzo-
thiazolin-2-ylidene) (3). A suspension of N-methylbenzothiazo-
lium iodide (138 mg, 0.5 mmol) in THF (8 mL) was added to a obtain complex 3. X-ray diffraction quality crystals were
t
solution of CuO Bu (68 mg, 0.5 mmol) in THF (2 mL). After a obtained from a DCM solution at −20 °C (Fig. 1). Yield:
1
few minutes, the mixture turned clear, and a yellow precipitate 161 mg (0.186 mmol, 74%). H-NMR (500 MHz, DMSO-d
6
):
appeared. After stirring for an additional one hour, the yellow δ 8.06 (d, 4H, J = 8.2 Hz, Aryl-H), 7.94 (d, 4H, J = 8.2 Hz,
precipitate was collected and washed with diethyl ether to Aryl-H), 7.62 (m, 4H, Aryl-H), 7.52 (m, 4H, Aryl-H), 4.34
1306 | Dalton Trans., 2014, 43, 1305–1312
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
1
3
1
(
3
s, 12H, CH ), 3.58 (m, THF), 1.75 (m, THF). C-NMR ( H): bromide (270 mg, 1 mmol) in THF (8 mL) was added to
t
t
δ 216.4 (NCS), 144.4 (s, Aryl-C), 134.0 (s, Aryl-C), 127.5 (s, Aryl- CuO Bu (68 mg, 0.5 mmol) and NaO Bu (48 mg, 0.5 mmol) in
C), 125.7 (s, Aryl-C), 123.2 (s, Aryl-C), 115.6 (s, Aryl-C), 67.4 THF (4 mL). Filtration through celite and concentration under
(s, THF), 42.0 (s, CH
3
), 25.6 (s, THF). Anal. calcd for 3·0.5THF, vacuum, followed by cooling of the THF solution to −20 °C
C H I Cu O N S : C, 29.28; H, 2.31; N, 4.02. Found: C, provided crystals suitable for X-ray diffraction. Yield: 142 mg
3
4
32 4
4 0.5 4 4
1
2
9.21; H, 2.32; N, 3.94.
Preparation of [(NSCH)Cu(μ-Br)(μ-O Bu)Cu(NSCH)] (NSHC = 2H, J = 7.6 Hz, Aryl-H), 8.02 (d, 2H, J = 8.2 Hz, Aryl-H), 7.62 (m,
N-crotylbenzothiazolin-2-ylidene) (4). Compound 4 was pre- 2H, Aryl-H), 7.53 (m, 2H, Aryl-H), 5.90 (m, 2H, CHvCH), 5.80
pared and purified using a method similar to that for 1 using (m, 2H, CHvCH), 5.44 (d, 4H, J = 4.4 Hz, CH ), 3.58 (m, THF),
N-crotylbenzothiazolium bromide (135 mg, 0.5 mmol) and 1.75 (m, THF), 1.61 (d, 6H, J = 5.0 Hz, CH
6
(0.27 mmol, 54%). H-NMR (500 MHz, DMSO-d ): δ 8.12 (d,
t
2
1
3
1
3
). C-NMR ( H):
CuO Bu (102 mg, 0.75 mmol) as starting materials. Yield: δ 219.0 (NCS), 143.6 (s, Aryl-C), 134.5 (s, Aryl-C), 131.2 (s,
t
1
1
6
29.9 mg (0.195 mmol, 78%). H-NMR (500 MHz, DMSO-d ): CHvCH), 127.5 (s, Aryl-C), 125.6 (s, Aryl-C), 125.5 (s, Aryl-C),
δ 8.13 (d, 2H, J = 7.6 Hz, Aryl-H), 8.05 (d, 2H, J = 8.2 Hz, Aryl- 123.4 (s, CHvCH), 115.8 (s, Aryl-C), 67.4 (s, THF), 55.6 (s,
H), 7.66 (m, 2H, Aryl-H), 7.57 (m, 2H, Aryl-H), 5.92 (m, 2H, CH ), 25.6 (s, THF), 18.0 (s, CH ). Anal. calcd for 6,
2
3
CHvCH), 5.79 (m, 2H, CHvCH), 5.40 (d, 4H, J = 6.3 Hz, CH
2
3
),
C
22
H
22CuBrN
2
S
2
: C, 50.62; H, 4.25; N, 5.37. Found: C, 50.42;
3
1
.58 (m, THF), 1.75 (m, THF), 1.65 (d, 6H, J = 6.3 Hz, CH
), H, 4.21; N, 5.19.
Preparation of [Cu(μ-I)(NSCH) ] (NSHC = N-allylbenzothia-
1
3
1
.34 (s, 9H, C(CH ) ). C-NMR ( H): δ 210.3 (NCS), 143.1 (s,
3
3
2 2
Aryl-C), 134.2 (s, Aryl-C), 131.8 (s, CHvCH), 127.9 (s, Aryl-C), zolin-2-ylidene) (7). Compound 7 was prepared and purified
1
26.1 (s, Aryl-C), 125.2 (s, Aryl-C), 123.6 (s, CHvCH), 116.0 (s, using a method similar to that for 3 with N-allylbenzothiazo-
t
Aryl-C), 67.4 (s, THF), 56.5 (s, CH ), 36.0 (s, C(CH ) ), 25.6 (s, lium iodide (303 mg, 1 mmol), CuO Bu (68 mg, 0.5 mmol) and
2
3 3
t
THF), 18.0 (s, CH
3
), 14.0 (s, C(CH
: C, 47.62; H, 4.81, N, 4.21. Found: C, 65%). H-NMR (500 MHz, DMSO-d
Aryl-H), 8.06 (d, J = 8.2 Hz, 4H, Aryl-H), 7.66 (m, 4H, Aryl-H),
] (NSHC = N-allyl- 7.57 (m, 4H, Aryl-H), 5.92 (m, 4H, CHvCH ), 5.81 (m, 8H,
). C-NMR ( H): δ 143.4 (s, Aryl-C),
thiazolium bromide (64 mg, 0.25 mmol) in THF (5 mL) was 134.5 (s, Aryl-C), 131.9 (s, CHvCH ), 127.7 (s, Aryl-C), 125.9 (s,
3 3
) ). Anal. calcd for 4·0.1THF, NaO Bu (48 mg, 0.5 mmol). Yield: 175.8 mg (0.162 mmol,
1
C
26.4
H
31.8Cu
2
BrO1.1
N
2
S
2
6
): δ 8.15 (d, J = 8.2 Hz, 4H,
4
8.01; H, 4.34; N, 4.44.
Preparation of [(NSCH)Cu(μ-Br)
benzothiazolin-2-ylidene) (5a). A suspension of N-allylbenzo- CHvCH
2
(NSCH)
2
2
1
3
1
2
), 5.39 (d, 8H, CH
2
2
added to a solution of 1 (159 mg, 0.25 mmol) in THF (5 mL), Aryl-C), 125.3 (s, Aryl-C), 123.5 (s, CHvCH
2 2 2 4 4
and then NaO Bu (24 mg, 0.25 mmol) in THF (1 mL) was 56.0 (s, CH ). Anal. calcd for 7, C40H36Cu I N S : C, 44.41; H,
2
), 116.0 (s, Aryl-C),
t
added. The mixture was stirred for 1 h. Filtration through celite 3.35; N, 5.18. Found: C, 43.92; H, 3.75; N, 5.14.
and concentration under vacuum, followed by cooling of the
THF solution to −20 °C provided crystals suitable for X-ray dif-
1
fraction. Yield: 166 mg (0.21 mmol, 82%). H-NMR (500 MHz,
Results and discussion
6
DMSO-d ): δ 8.13 (d, 3H, J = 8.2 Hz, Aryl-H), 7.98 (d, 3H, J = 8.2
Hz, Aryl-H), 7.63 (m, 3H, Aryl-H), 7.54 (m, 3H, Aryl-H), 6.15 (m, Copper(I)-NSHC complex [Cu(μ-Br)L ]
1
1
2
(L = N-allylbenzothia-
3
H, CHvCH ), 5.53 (d, 6H, J = 5.6 Hz, CHvCH ), 5.29 (m, 6H, zolin-2-ylidene) (1) (Scheme 1) was synthesized from the reac-
2
2
13
1
t
CH
NCS), 143.6 (s, Aryl-C), 134.4 (s, Aryl-C), 132.7 (s, CHvCH
27.5 (s, Aryl-C), 125.7 (s, Aryl-C), 123.4 (s, Aryl-C), 119.5 (s, room temperature reaction (typically within 2 h) was moni-
CHvCH ), 115.8 (s, Aryl-C), 67.4 (s, THF), 56.2 (s, CH ), 25.6 (s, tored by the disappearance of the N–CH–S proton in the
THF). Anal. calcd for 5a·0.5THF, C32 Br
5.29; H, 3.68; N, 4.95. Found: C, 45.34; H, 3.22; N, 5.29.
Preparation of [(NSCH)Cu(μ-Br) Cu(NSCH)
tylbenzothiazolin-2-ylidene) (5b). Compound 5b was prepared basic metal precursor, such as PdCl
2
), 3.58 (m, THF), 1.75 (m, THF). C-NMR ( H): δ 219.2 tion of N-allylbenzothiazolium bromide with CuO Bu,
t
(
1
2
), generated in situ from CuCl and NaO Bu. Completion of this
2
2
1
H31Cu
2
2
O
0.5
3 3
N S : C,
H-NMR spectrum.
Although Pd(II) and Ni(II) NSHC complexes are usually pre-
4
2
2
] (NSHC = N-cro- pared from direct reaction of the benzothiazolium salt with a
, Pd(OAc) , NiCl or
this methodology failed to yield the desired
2
2
2
5
d,6,7
and purified using a method similar to that for 5a. Yield: Ni(OAc)2,
1
10,11
1
68 mg (0.20 mmol, 80%). H-NMR (500 MHz, THF-d
8
): δ 7.92 product from CuCl, CuOAc or Cu
2
O
in THF at room temp-
(
(
5
(
(
d, 3H, J = 8.2 Hz, Aryl-H), 7.87 (d, 3H, J = 8.2 Hz, Aryl-H), 7.52 erature or at 60 °C. The alternative use of free carbene as the
1
2
m, 3H, Aryl-H), 7.43 (m, 3H, Aryl-H), 5.99 (m, 3H, CHvCH), substrate, which is also common for the preparation of NHC
.88 (m, 3H, CHvCH), 5.52 (m, 3H, CH ), 3.57 (m, THF), 1.72 complexes, was similarly unsuccessful. Thiazolinylidene
2
1
3
1
m, THF), 1.68 (d, 9H, J = 6.2 Hz, CH
3
). C-NMR ( H): δ 131.0 copper(I) complexes have previously been prepared in a com-
n
s, CHvCH ), 126.3 (s, Aryl-C), 125.3 (s, Aryl-C), 124.6 (s, Aryl- parable, albeit less convenient, procedure employing BuLi at
2
1
3
C), 122.3 (s, CHvCH
2
), 114.8 (s, Aryl-C), 67.2 (s, THF) 55.5 (s, low temperatures.
CH ), 25.1 (s, THF), 17.0 (s, CH ). Anal. calcd for 5b·0.5THF, Reaction of N-propyl- and N-crotyl-benzothiazolium bro-
2
3
t
C H Cu Br O N S : C, 47.19; H, 4.19; N, 4.95. Found: C, mides with CuO Bu gave the tetranuclear copper complexes
3
5
37
2
2 0.5 3 3
2
2
4
7.42, H, 4.29; N, 4.45.
[Cu(μ-Br)(L )]
4
4
(L = N-proylbenzothiazolin-2-ylidene) (2a) and
3
3
Preparation of [CuBr(NSCH)
2
] (NSHC = N-crotylbenzothiazo- [Cu(μ-Br)(L )]
(L
= N-crotylbenzothiazolin-2-ylidene) (2b)
lin-2-ylidene) (6). A suspension of N-crotylbenzothiazolium respectively, as determined by X-ray crystallography. The
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 1305–1312 | 1307

View Article Online
Paper
Dalton Transactions
1
3
butoxide bridges. The 5 ppm upfield shift in the carbene
C
NMR signal is commensurate with the more electronegative,
1
4a
bridging ancillary oxygen ligand.
The dinuclear 1 : 1 Cu : NSHC complex 1 reacted with an
equimolar equivalent of N-allylbenzothiazolium bromide in
t
the presence of NaO Bu to yield the unsymmetrical tricarbene
1
complex, [Cu
2
(μ-Br)
2
(L )
3
] (5a), in the solid state. Complex 2b
is the dimeric form of 1 (albeit with different N-alkyl groups)
and reacted with two equivalents of N-crotylbenzothiazolium
t
3
bromide and NaO Bu to yield the analogous [Cu
2
(μ-Br)
2 3
(L ) ]
5b. While the NMR spectra of 5a and 5b are quite distinct
from those of the parent complexes 1 and 2b, there is no evi-
dence of two ligand environments in the NMR spectra for
either complex. This suggests that either the unsymmetrical
solid-state structures are not replicated in solution, or that a
rapid, dynamic process is occurring in solution. However,
cooling a d -DCM solution of 5b to 193 K failed to resolve any
2
extra signals.
Using an excess of N-crotylbenzothiazolium bromide in a
t
3
t
1
[
: 2 : 1 substrate ratio of CuO Bu : L HBr : NaO Bu provided
3
2
CuBr(L ) ] (6). This was the only mononuclear Cu(I) carbene
observed in this series and it is unusual to see a terminal
bromide in the solid-state despite its basicity and donor poten-
tial to a neighbouring Cu(I). The isolation of this complex
possibly suggests a low energy barrier between a mono- and
dinuclear form and that dissociation of these Cu(I) carbene
−
complexes can be anticipated in solution. Anions such as BF
4
−
and PF
6
are known to support cationic complexes of type
1
6,17
[Cu(carbene) ]Z.
2
t
The reaction of N-allylbenzothiazolium iodide with CuO Bu
and NaO Bu in a ratio of 2 : 1 : 1 provided [Cu(µ-I)(L ) ] , 7 in
2 2
Scheme 1
t
1
the solid-state. The larger halide of 7 presumably makes it
easier for two copper spheres to come into bonding contact.
1
13
aromatic H and C NMR signals of 2a and 2b in DMSO-d
6
were very similar to each other and were almost indistinguish- Its formation validates our earlier DFT based prediction that a
1
8
able from monomeric 1. It is difficult, therefore, to establish fourth carbene could be added to the [Cu I (L ) ] complex.
2
2
3
whether the solid-state structures of 2a and 2b are replicated
In general, 1, 2a, 2b, 4, 5b and 6 are more soluble in THF
in solution. It is possible that complexes 2a and 2b dissociate than 3, 5a and 7. All are however soluble in CH Cl and
2
2
to dinuclear or even mononuclear copper species in this DMSO. They are stable in air for a few hours and can be stored
highly coordinating solvent. N-methylbenzothiazolium iodide under argon at low temperature for several months. As dis-
4
4
1
13
as the substrate provided [Cu(μ-I)(L )]
thiazolin-2-ylidene) (3) in the solid state with unusual μ4 chiometric information and nuclearity may change on
capping iodides in place of μ bridging bromides. The aro- dissolution or, indeed, vary in different solvents. X-ray single-
4
(L = N-methylbenzo- cussed above, their NMR spectra ( H and C) carry little stoi-
3
1
13
matic H and C NMR signals for this complex were signifi- crystal diffraction analysis of complexes 1–7, however, revealed
cantly different from the corresponding signals in the remarkable structural diversity in the solid state depending on
preceding three complexes with, for example, a >0.5 ppm down- the reagent stoichiometry and inorganic ligand (bridging
1
3
field shift in the carbene C signal, reflecting a decrease in halide or alkoxide). These compounds can be classified into
the Lewis acidity of the metal by virtue of the increase in the five motifs, viz. M L where x : y = 2 : 2 (1) (4), 4 : 4 (2) (3), 2 : 3 (5),
x
y
1
4
σ-donor ability of the iodide ligands.
1 : 2 (6) and 2 : 4 (7). The Cu(I) geometry is either trigonal planar
(NHC)CuO Bu]-type complexes generated from [(NHC)- or tetrahedral with the metal carrying one or two carbene
CuCl] and NaO Bu or KO Bu are known to be active catalysts ligands. The final structure is determined by the preferred
t
[
t
t
for a range of reactions involving carbonyl and alkene sub- coordination mode of the inorganic ligand, viz. from terminal
strates, including with the control of absolute stereo- and unidentate (6) to doubly bridging µ (1, 2, 3, 4, 5, 7) and
2
1
5
t
3
chemistry. We therefore prepared [Cu (μ-Br)(μ-O Bu)(L ) ] (4) triply µ bromide bridging (2) to quadruply µ iodide capping
2
2
3
4
from the reaction of N-crotylbenzothiazolium bromide with 1.5 (3). Despite this diversity, the Cu–C carbene bond lengths cover
t
equivalents of CuO Bu. The product of this reaction was a a fairly narrow range (1.852(5)–1.932(4) Å, Table S1†) and
16
unique dinuclear carbene complex with mixed bromide and compare well with those of other copper NHC complexes.
1308 | Dalton Trans., 2014, 43, 1305–1312
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Complex 1 is dinuclear with two trigonal planar Cu(I)
atoms, each of which carries a carbene ligand (Fig. 1). Associ-
ation of the two Cu(I) spheres is unsymmetrical with two sig-
nificantly uneven Cu–Br bonds (2.3037(8) and 2.772(1) Å
respectively), thus also pointing to possible dissociation in
solution. Its monomeric form has not been isolated, except for
6
in which an additional carbene provides the needed stabili-
zation. The two NSHC planes are parallel to each other and
twisted by 122.5° from the [Cu Br ] coordination plane. This
structure is similar to the previously reported [Cu(µ-Cl)L] (L =
N-methyl-4,5-dimethylthiazolin-2-ylidene) complex which also
2
2
2
1
3
exhibits two unequal Cu–Cl bonds (1.888(6) and 2.122(2) Å).
Complex 2a is tetranuclear in the solid state with the
[
Cu Br ] assembled in a step fashion. The two internal Cu(I)
4
4
are tetrahedral whereas the two external Cu(I) are trigonal
planar and each metal atom bears a carbene ligand. This
complex can be viewed as the dimeric form of 1 with two triply
bridging bromides linking the subunits. The Cu–C bond at the
three-coordinated Cu(I) [Cu(1)–C(1) 1.900(2) Å] is only margin-
ally shorter than the corresponding bond of the four-
coordinated Cu(I) [Cu(2)–C(11) 1.910(3) Å)], but both are
slightly longer than their counterparts in 1 (1.887(4) Å) and
1
6
those in the literature. The structure of 2b is similar to that
of 2a. However, Cu(2)–Br(2A) and Cu(2A)–Br(2) in 2b (2.7138(9)
and 2.7139(9) Å respectively) are significantly longer than
other Cu–Br bonds (2.4396(7), 2.4155(8), 2.4263(8), 2.6435(8) Å
respectively) and their counterparts in 2a (2.5828(5) and
2
.5827(5) Å), indicating weaker secondary, bridging inter-
actions between the two monomeric subunits.
Complex 3 is also tetranuclear but the four Cu(I) atoms are
planar (the dihedral angle between the planes of Cu(1)Cu(2)-
Cu(1A) and Cu(2)Cu(1A)Cu(2A) is 0°) and capped above and
below by quadruply capping iodides and two additional edge-
bridging iodides. This is the only complex in this series with
tetrahedral-only Cu(I) centers, each of which carries one
carbene ligand. The quadruply bridging iodides I(2) and I(2A)
bring the two non-bonding Cu⋯Cu (2.5626(7) Å) into close
proximity. This distance is significantly shorter than all other
Cu⋯Cu lengths in this series. The methyl N-substituents on
Fig. 2 ORTEP view of complexes (4), (5a) and (5b) with 40% thermal
ellipsoids.
the parallel NSHC planes across Cu(1)–Cu(2A) are oriented bromides. The Cu–C(carbene) bond is understandably stronger
opposite to one another to avoid steric interaction.
in the copper monocarbene moiety (Cu(1)–C(1) 1.899(3) Å)
Complex 4, like 1, has purely trigonal planar Cu(I) centers than in the copper dicarbene moiety (Cu(2)–C(11) 1.914(3) Å
Fig. 2). The contrast in the size and electronegativity of the and Cu(2)–C(21) 1.931(3) Å). Similarly, the Cu–Br bonds on the
(
bridging atoms imparts a sharp distortion to the central copper dicarbene are significantly weaker (Cu(1)–Br(1)
Cu
BrO] ring with strong Cu–O (Cu(1)–O(1) = 1.874(3) Å and 2.4343(5) Å and Cu(1)–Br(2) 2.4538(5) Å) on Cu(1) compared to
Cu(2)–O(1) = 1.907(3) Å) bonds and longer Cu–Br bonds 2.5794(5) and 2.7097 (5) Å for Cu(2)). The NSHC ring with the
[
2
(
(
2.7460(9) Å) that draw the two copper atoms closer in 4 longest Cu–C carbene bond (Cu(2)–C(21) 1.931(3) Å) is almost
2.9964(8) Å) than in 1 (3.400 Å). This mixed-ligand formu- perpendicular to the other two NSHC rings, the Cu(1)Br(1)-
lation with a butoxide bridge appears to give a more acidic Cu(2) and Cu(1)Br(2)Cu(2) planes (dihedral angles of 111.4°,
Cu(I) leading to the strongest Cu–C carbene bonds (1.869(5) 102.1°, 98.6° and 83.4°, respectively), consistent with an
and 1.852(5) Å) in this series.
Two independent molecules co-crystallized in the unit cell complex 1.
adduct-like addition of free NSHC ligand to electron deficient
1
1
of 5a. Their structures are similar to that of [L Cu(μ-I) CuL ]
It is unusual, although not unprecedented, to obtain a
2
2
8
previously reported. There are two copper geometries, viz. tri- mononuclear dicarbene Cu(I) complex with a terminal halide
gonal planar with one carbene (16 electron) and tetrahedral (complex 6).
1
6,17
The trigonal planar geometry of Cu(I) in 6 is
with two carbenes (18 electron) connected by the two distorted with a large C(1)–Cu(1)–C(12) angle of 142.8(1)°
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 1305–1312 | 1309

View Article Online
Paper
Dalton Transactions
in solution or a fast shuttle of the ligands between the metals
could be operating. Discrepancy between solid-state and solu-
tion structures for NHC complexes has been reported
1
8
previously.
CuX(NHC) (X = Cl and Br)
1
9,20
and [Cu(NHC) ]Z (Z = PF
2
6
1
e,21
and BF
4
)
are very efficient catalysts for the Huisgen 1,3-
dipolar cycloaddition of azides to alkynes (including internal
alkynes), particularly in aqueous solvents and under solvent
free conditions. We therefore conducted some preliminary
screening of our compounds in this respect. Complex 1 was
chosen as a model catalyst in the “click reaction” of benzyl
azide and phenylacetylene to optimize conditions (Table 1).
3
Yields were highest in aqueous CH CN and DMF (entries 2 &
3). Surprisingly, solvent-free conditions or reactions in water,
aqueous EtOH or aqueous THF were significantly lower yield-
ing (entries 1, 4 & 7). These results are at odds with the per-
formance of CuX(NHC) and [Cu(NHC) ]Z under solvent-free
2
conditions and/or in water or in a variety of aqueous sol-
1
9,20
vents.
We suggest that these more strongly Cu(I)-coordinat-
ing solvents DMF and CH CN may be needed to promote
3
dissociation of 1 into catalytically active [CuBr(NSHC)] and this
proposition is supported by a comparison of different catalysts
below (Table 2).
Aqueous acetonitrile (in the ratio of 1 : 1) was chosen as the
solvent system for preliminary screening of our different
copper NSHC complexes. Not surprisingly, given the solvent
dependence study above, the marginally more efficient catalyst
proved to be mononuclear complex 6. While the solution struc-
tures of some complexes are not known precisely, one
interpretation of these results is that cuprophilic coordinating
solvents result in dissociation to give mononuclear [CuX-
Fig. 3 ORTEP view of complexes 6 and 7 with 40% thermal ellipsoids.
+
(
NSHC)] or [Cu(NSHC) ] species which then react rapidly with
2
1
e
phenyl acetylene to provide intermediate [Cu(NSHC)(CCPh)].
(Fig. 3). The shorter of the Cu–C(carbene) bonds (1.904(3) Å This hypothesis implies that the yields displayed in Table 2
and 1.924(3) Å) are significantly shorter than their counter- may reflect the rate of dissociation and portends the use of
1
6
parts in the corresponding NHC analogue (1.930(3) Å). The these dinuclear and tetranuclear complexes as slow-release
two NSHC planes are nearly perpendicular to each other with catalysts for a range of applications. Quantitative yield was
the dihedral angle of 111.5°. This twist allows the two carbene achieved by raising the loading of precatalyst 1 to 1% (Table 2,
moieties to draw closer to the coordination sphere enabling a entry 1).
stronger Cu–C contact but without invoking undesirable inter-
ligand repulsions.
Complex 7 can be viewed as the dimeric form of 6. Such Table 1 Influence of solvent on yield
associations tend to produce a more electron rich Cu(I) and
weaken the Cu–C carbene link ((1.923(3) & 1.925(3) Å in 7 rela-
tive to 1.904(3) & 1.924(3) Å in 6). The four Cu–I bonds fall in a
narrow range of 2.8330(5)–2.8594(5) Å, and are significantly
longer than the Cu–I bonds in the corresponding [CuI(NHC) ]
2
1
6
complex (2.7623(7) Å). The dihedral angles of the Cu(1)I(1)-
Cu(1A)I(1A) to C(1)N(1)S(1) and C(11)N(2)S(2) planes are 69.3°
and 103.8° respectively, which support minimum inter-ligand
interaction. The C(1)N(1)S(1) and C(11)N(2)S(2) planes are
twisted by 62.8° to avoid non-bonding contacts. The Cu⋯Cu
separation (3.473 Å) is, unsurprisingly, the longest in this
series. However, as discussed above, only one set of signals
was observed in the NMR spectra of these complexes, indicat-
ing that the observed solid structures might not be significant
a
b
Entry
Solvent
Yield /%
1
2
3
4
5
6
7
EtOH–H
CH CN–H
DMF–H
THF–H
CH CN
2
O
2
O
O
43
95
97
23
19
60
13
3
2
2
O
3
2
H O
Nil
a
b
Mixed solvent in the ratio of 1 : 1. Isolated yields.
1310 | Dalton Trans., 2014, 43, 1305–1312
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Table 2 Comparison of different Cu(I)-NSHC complexes
accommodate more than one carbene ligand on each metal. It
+
is notable that a linear [Cu(NHSC) ] moiety was not observed;
2
however, the formation of mononuclear dicarbene 6 with a
coordinating halide ligand was obtained for the first time. The
Cu⋯Cu separation over a wide range of 2.5626(7) to 3.4725(7)
Å distances points to the propensity for Cu(I) carbene dis-
sociation and aggregation. This presents an ideal model for us
to explore the metal–metal cooperativity and its response to
migratory ligands and creation of vacant sites for substrate
entry in a typical catalytic pathway. We are currently exploring
the applications of this potentially useful property of bridged
Cu(I) carbenes.
a
Entry
Cat.
Yield /%
b
1
2
3
4
5
6
7
8
9
[CuBr(NSHC)]
[CuBr(NSHC)]
[CuBr(NSHC)]
2
4
4
(1)
(2a)
(2b)
95, 100
81
75
40
40
25
69
98
36
[CuI(NSHC)]
4
(3)
2
Br(O Bu)(NSHC) ] (4)
t
[Cu
[Cu
[Cu
2
2
2
Br
Br
2
(NSHC)
(NSHC)
3
](5a)
] (5b)
2
3
[CuBr(NSHC)
[CuI(NSHC) ]
2
] (6)
(7)
2 2
Notes and references
a
b
Isolated yields. 1 mol% Cu, 14 h.
1
(a) T. Ohishi, L. Zhang, M. Nishiura and Z. Hou, Angew.
Chem., Int. Ed., 2011, 50, 8114–8117; (b) H. Jang,
A. R. Zhugralin, Y. Lee and A. H. Hoveyda, J. Am. Chem.
Soc., 2011, 133, 7859–7871; (c) L. Ackermann, Angew.
Chem., Int. Ed., 2011, 50, 3842–3844; (d) G. C. Fortman,
A. M. Z. Slawin and S. P. Nolan, Organometallics, 2010, 29,
Table 3 Comparison of different alkyne and aryl bromide substrates
3966–3972; (e) S. Díez-González and S. P. Nolan, Angew.
Chem., Int. Ed., 2008, 47, 8881–8884.
2
(a) M. R. L. Furst and C. S. J. Cazin, Chem. Commun., 2010,
a
Entry
R
1
R
2
Yield /%
46, 6924–6925; (b) M. Yoshida, H. Ohmiya and
M. Sawamura, J. Am. Chem. Soc., 2012, 134, 11896–11899.
S. Gaillard, C. S. J. Cazin and S. P. Nolan, Acc. Chem. Res.,
2012, 45, 778–787.
R. Y. Tan, F. S. N. Chiu, A. Hadzovic and D. T. Song, Organo-
metallics, 2012, 31, 2184–2192.
5 (a) S. K. Yen, L. L. Koh, H. V. Huynh and T. S. A. Hor,
Dalton Trans., 2007, 3952–3958; (b) N. Ding and
T. S. A. Hor, Dalton Trans., 2010, 39, 10179–10185;
(c) X. Han, L. L. Koh, Z. Weng and T. S. A. Hor, Dalton
Trans., 2009, 7248–7252; (d) N. Ding, J. Zhang and
T. S. A. Hor, Dalton Trans., 2009, 1853–1858; (e) S. K. Yen,
L. L. Koh, H. V. Huynh and T. S. A. Hor, Chem.–Asian J.,
2008, 3, 1649–1656.
1
2
3
4
5
6
7
8
9
H
H
H
4-NO
4-NO
4-CN
4-CN
4-CN
4-CN
4- Bu
4- Bu
4- Bu
H
H
2-F
Ph
4-MeOPh
4-pip-CH
Ph
4-MeOPh
Ph
>99
>99
77
98
98
87
75
29
78
>99
85
45
70
92
3
4
2
2
2
4-MeOPh
SiMe
AcO
Ph
AcO
SiMe
AcO
SiMe
Ph
3
t
10
11
12
13
14
15
t
t
3
3
44
a
Isolated yields.
6
7
8
9
S. K. Yen, L. L. Koh, H. V. Huynh and T. S. A. Hor,
Aust. J. Chem., 2009, 62, 1047–1053.
S. K. Yen, D. J. Young, H. V. Huynh, L. L. Koh and
T. S. A. Hor, Chem. Commun., 2009, 6831–6833.
X. Han, L. L. Koh, Z.–P. Liu, Z. Weng and T. S. A. Hor,
Organometallics, 2010, 29, 2403–2405.
(a) T. H. Lemmen, G. V. Goeden, J. C. Huffman,
R. L. Geerts and K. G. Caulton, Inorg. Chem., 1990, 29,
A one-pot CuAAC protocol with precatalyst 1 (1 mol% in
aqueous CH CN) generally provided good yields for a wide
3
range of substrates (Table 3). Reactions of ethynyltrimethyl-
silane were less reliable, presumably due to steric hindrance of
the bulky Me Si group.
3
3680–3685; (b) T. Tsuda, T. Hashimoto and T. Saegusa,
J. Am. Chem. Soc., 1972, 94, 658–659.
Conclusions
1
0 (a) J. Chun, H. S. Lee, G. Jung, S. W. Lee, H. J. Kim and
S. U. Son, Organometallics, 2010, 29, 1518–1521;
(b) M. R. L. Furst and C. S. J. Cazin, Chem. Commun., 2010,
46, 6924–6925.
A remarkable array of Cu(I) NSHC complexes have been pre-
pared by variation in the nitrogen substituent of the hetero-
cycle and the stoichiometry of reagents. The solid-state
structures of these complexes highlight the geometrical and 11 A. A. D. Tulloch, A. A. Danopoulos, S. Kleinhenz,
electronic flexibility of Cu(I), facile aggregation through brid-
ging and capping halides and, importantly, the ability to
M. E. Light, M. B. Hursthouse and G. Eastham, Organo-
metallics, 2001, 20, 2027–2031.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 1305–1312 | 1311

View Article Online
Paper
Dalton Transactions
1
2 N. P. Mankad, T. G. Gray, D. S. Laitar and J. P. Sadighi, 16 G. Venkatachalam, M. Heckenroth, A. Neels and
Organometallics, 2004, 23, 1191–1193. M. Albrecht, Helv. Chim. Acta, 2009, 92, 1034–1045.
3 H. G. Raubenheimer, S. Cronje, P. H. van Rooyen, 17 (a) S. Díez-González, E. D. Stevens, N. M. Scott,
1
P. J. Olivier and J. G. Toerien, Angew. Chem., Int. Ed. Engl.,
994, 106, 687–688.
4 (a) M. V. Baker, P. J. Barnard, S. K. Brayshaw, J. L. Hickey,
J. L. Petersen and S. P. Nolan, Chem.–Eur. J., 2008, 14, 158–
168; (b) S. Díez-González, N. M. Scott and S. P. Nolan,
Organometallics, 2006, 25, 2355–2358.
1
1
1
B. W. Skelton and A. H. White, Dalton Trans., 2005, 37–43; 18 S. Díez-González, E. C. Escudero-Adán, J. Benet-Buchholz,
(b) W. A. Herrmann, O. Runte and G. Artus, J. Organomet.
E. D. Stevens, A. M. Z. Slawin and S. P. Nolan, Dalton
Trans., 2010, 39, 7595–7606.
19 S. Díez-González, E. D. Stevens and S. P. Nolan, Chem.
Commun., 2008, 4747–4749.
Chem., 1995, 501, Cl–C4; (c) Y. Han, H. V. Huynh and
G. K. Tan, Organometallics, 2007, 26, 6447–6452.
5 (a) S. Díez-González and S. P. Nolan, Acc. Chem. Res., 2008,
4
1, 349–358; (b) K. Takatsu, R. Shintani and T. Hayashi, 20 S. Díez-González, A. Correa, L. Cavallo and S. P. Nolan,
Angew. Chem., Int. Ed., 2011, 50, 5548–5552; (c) R. Shintani, Chem.–Eur. J., 2006, 12, 7558–7564.
K. Takatsu and T. Hayashi, Chem. Commun., 2010, 46, 21 F. Lazreg, A. M. Z. Slawin and C. S. J. Cazin, Organo-
822–6824. metallics, 2012, 31, 7969–7975.
6
1312 | Dalton Trans., 2014, 43, 1305–1312
This journal is © The Royal Society of Chemistry 2014