[Cu8(μ8-H){S2P (OEt)2} 6](PF6): A novel catalytic hydride-centered copper cluster for azide-alkyne cycloaddtion

Article abstract of DOI:10.1007/s10562-013-0993-7

A hydride-centered dithiophosphate cluster [Cu₈(μ ₄-H){S₂P(OEt)₂}₆](PF₆) (1)] previously developed by us was applied as a new catalyst to the 1,3-dipolar cycloaddition of organic azides and

Full text of DOI:10.1007/s10562-013-0993-7

Catal Lett (2013) 143:572–577
DOI 10.1007/s10562-013-0993-7
[Cu8(l4-H){S2P(OEt)2}6](PF6): A Novel Catalytic Hydride-
Centered Copper Cluster for Azide-Alkyne Cycloaddtion
•
Bo-Han Lee Cheng-Chieh Wu Xuan Fang
•
•
•
C. W. Liu Jia-Liang Zhu
Received: 17 December 2012 / Accepted: 7 March 2013 / Published online: 27 March 2013
Ó Springer Science+Business Media New York 2013
Abstract A hydride-centered dithiophosphate cluster
[Cu8(l4-H){S2P(OEt)2}6](PF6) (1)] previously developed
by us was applied as a new catalyst to the 1,3-dipolar
cycloaddition of organic azides and alkynes for preparing
substituted trizoles. With the required catalyst loading as
low as 0.4 mol%, the reactions of terminal alkynes with
BnN3 all proceeded smoothly at ambient temperature in
CH3CN to exclusively produce 1,4-triazoles in good yields.
For these reactions, it is assumed that the formation of the
requisite copper acetylide intermediate is facilitated by the
abstraction of the terminal hydrogen of alkynes by the
hydride released from the central of the cluster. With only
few examples being documented in literatures, the reac-
tions of a range of internal alkynes have also been realized
under the catalysis of 1 (0.8 mol%) in DMF at elevated
temperature, to yield 1,4,5-trisubstituted triazoles in mod-
erate to high yields. Our study has provided a preliminary
insight into the effect of sulfur-based ligands on the activity
of copper ion.
1 Introduction
The copper(I)-catalyzed 1,3-dipolar cycloaddition of
organic azides with alkynes (CuAAC) has emerged as an
extremely powerful method for preparing substituted tria-
zoles [1–3], which has found wide applications in numer-
ous fields [4–15] such as drug discovery [4–6], bio-
modifications [7–10], material chemistry [4, 11–14] and
surface science [15]. As compared with the thermal
cycloaddition of terminal alkynes (Huisgen reaction) usu-
ally producing an isomeric mixture [16], the cop-
per(I) catalysis leads to a major improvement in both
regioselectivity and rate to exclusively produce 1,4-tria-
zoles (vs 1,5-regioisomers) under relatively mild condi-
tions (Scheme 1), to thus fulfill the requirement of ‘‘click
chemistry’’. The previous works have demonstrated that
some commercially available copper(I) salts such as CuBr
[17] and CuI [18] or even metallic copper [19] can be used
to catalyze the reaction. Nevertheless, the innovation of
more efficient and/or robust catalytic systems has always
attracted considerable interests. Among the developed
methods, the most widely employed ones are those
involving the in situ generation of catalytically active
Cu(I) from Cu(II) salts, such as the combination of CuSO4/
Na-ascorbate [19, 20] and CuSO4/Cu [5]. Besides, the
applications of copper nanoparticles [21] or the p-donating
ligand-coordinated copper(I) species [22] to CuAAC have
also been well documented in literatures, as exemplified by
the recent advance in using N-heterocyclic carbene (NHC)-
stabilized copper complexes to catalyze the reaction [23–
25].
Keywords Copper(I) cluster ꢀ 1,3-Dipolar cycloaddition ꢀ
Alkynes ꢀ Triazoles ꢀ Click chemistry
Electronic supplementary material The online version of this
article (doi:10.1007/s10562-013-0993-7) contains supplementary
material, which is available to authorized users.
B.-H. Lee ꢀ C.-C. Wu ꢀ X. Fang ꢀ C. W. Liu (&) ꢀ
J.-L. Zhu (&)
Department of Chemistry, National Dong Hwa University,
Our research group previously reported [26] a novel
hydride-centered Cu8(I)-dithiophosphate (dtp) cluster for-
mulated as [Cu8(l4-H){S2P(OEt)2}6](PF6) (1) (Scheme 2),
which was readily prepared by mixing tetrakis(acetonitrile)
Hualien 97401, Taiwan
e-mail: chenwei@mail.ndhu.edu.tw
J.-L. Zhu
e-mail: jlzhu@mail.ndhu.edu.tw
123

Copper(I)-Cluster Catalyzed Azide-Alkyne Cycloaddition
573
Splitting patterns are described as singlet (s), doublet (d),
triplet (t), quartet (q) or multiplet (m). Coupling constants
(J) are reported in Hertz (Hz). The resonances of infrared
(IR) spectra are reported in wave numbers (cm-1). High
resolution mass spectra (HRMS) were determined in fast
atom bombardment (FAB) or electron impact (EI) modes.
Cluster 1 was prepared following our previously reported
procedure [26] and stored at 0 °C before each use. The
unknown products from this study (2c, 2i, 3c, 3d, 3f, 3h/3h0
and 3j/3j0) were all characterized by spectroscopic methods
as well as the X-ray analysis (2i, 3f and 3j, CCDC
831728-831730).
Scheme 1 1,3-Dipolar cycloaddition of terminal alkynes with azides
copper(I) hexafluorophosphate, diethyl dithiophosphate
ammonia salt and sodium borohydride together in a ratio of
8:6:1. The X-ray analysis revealed that 1 possessed an
interesting and rare coordination pattern, in which, the
hydride-centered tetracapped tetrahedral Cu8 core is sur-
rounded by an icosahedral sulfur cage composed of 12 S
atoms out of six dtp ligands. Furthermore, we observed that
cluster 1 exhibited high solubility in various organic sol-
vents (e.g. CH3CN, DMF, CH2Cl2, THF, etc.) and good
air-stability, which appeared to be suited for catalysis. A
literature search disclosed that the application of such
sulfur ligand-based complexes to the 1,3-dipolar cycload-
dition was very rare [27]. We therefore decided to attempt
1 to the reaction for exploring its catalytic activity. Herein
we wish to report our preliminary results.
2.2 Synthesis of 1
See ref. [26]
2.3 Typical Experimental Procedure for Cycloaddition
of Terminal Alkynes (2a)
To a stirred solution of 1 (8 mg, 0.4 mol%, 1.6 M) in
acetonitrile (2.5 mL), phenylacetylene (0.102 g, 1 mmol)
and benzyl azide (0.133 g, 1 mmol) were successively
added under N2. The mixture was stirred at room temper-
ature for 4 h then filtrated through a Celite pad, washed
with ethyl acetate (10 mL) and concentrated. The residue
was purified by flash chromatography on silica gel (hex-
2 Experimental
1
ane–ethyl acetate = 8:1) to afford 2a (214 mg, 91 %). H
2.1 Materials and Methods
NMR (400 MHz, CDCl3) d 7.82–7.81 (m, 2H), 7.41–7.29
(m, 9H), 5.56 (s, 2H); 13C NMR (100 MHz, CDCl3) d
134.7, 130.6, 129.1, 128.9, 128.8, 128.2, 128.1, 125.7,
54.3.
All reagents were purchased from commercial suppliers
and used without further purification. All reactions were
performed under an atmosphere of nitrogen except other-
wise noted. Tetrahydrofuran was freshly distilled from
sodium-benzophenone, and dichloromethane was freshly
distilled from CaH2, acetonitrile, hexane and acetone were
2.4 Typical Procedure for Cycloaddition of Internal
Alkynes (3a)
˚
distilled from molecular sieves (4 A) before use. TLC
To a stirred solution of 1 (16 mg, 0.8 mol%, 1.6 M) in
DMF (5 mL) under N2, diethylcarboxylate acetylene
(0.170 g, 1 mmol) and benzyl azide (0.133 g, 1 mmol)
were successively added. The mixture was then heated at
120 °C for 12 h and cooled to rt. After concentration under
reduced pressure, the residue was purified by flash chro-
matography on silica gel (hexane–ethyl acetate = 10:1,
analysis was carried out on glass-backed silica gel plates,
and visualized by UV light. The products were purified by
1
flash chromatography using silica gel (70–230 mesh). H,
13C NMR spectra were recorded on 400 MHz spectrome-
ters using deuteriochloroform (CDCl3) as solvent. Chemi-
cal shifts measurements are reported in delta (d) units.
Scheme 2 Preparation and
thermal ellipsoid plot of 1 (with
ethyl groups omitted for clarity)
123

574
B.-H. Lee et al.
1
5:1) to afford 3a (273 mg, 90 %) as a yellowish solid. H
NMR (400 MHz, CDCl3) d 7.31–7.29 (m, 3H), 7.23–7.21
(m, 2H), 5.77 (s, 2H), 4.40 (q, J = 7.2 Hz, 2H), 4.28 (q,
J = 7.3 2H), 1.36 (t, J = 7.2 Hz, 3H), 1.25 (t, J = 7.3 Hz,
3H). 13C NMR (100 MHz, CDCl3) d 160.4, 158.8, 140.4,
134.0, 129.9, 128.9, 128.7, 128.0, 62.8, 61.8, 53.7, 14.1,
13.8.
3 Results and Discussion
Our initial investigation was directed to the cycloaddition
of terminal alkynes. By using 0.4 mol% of 1, we first
screened several solvents including hexane, dichloro-
methane, acetone, tetrahydrofuran (THF), N,N-dimethyl-
formamide (DMF) and acetonitrile by the reaction between
phenylacetylene and benzyl azide (BnN3) (Table 1, entries
1–6). At room temperature, all reactions completed in 4 h
to afford 1-benzyl-4-phenyl-1,2,3-triazole (2a) as the sole
regioisomer in 52–91 % yields, and the best conversion
was obtained in acetonitrile (entry 6). On this basis, we
subsequently carried out the reactions respectively in 0.8,
0.3 and 0.2 mol% catalytic loadings. As shown in entries
7–9, the use of 0.8 mol% of 1 did not improve the yield
(entry 7) as compared with the yield in entry 6, while both
reduced loadings gave the poorer conversions (entries 8
and 9). Therefore the conditions in entry 6 (0.4 mol% of 1/
acetonitrile/rt) were considered to be our optimal choice.
Scheme 3 Copper cluster-catalyzed 1,3-dipolar cycloaddition of
terminal alkynes [a] 1.0 equiv of BnN3 was used. [b] 3.0 equiv of
BnN3 were used
To verify the role of 1, we also carried out a blank reaction
without the catalyst and found that no cycloaddition could
take place even after prolonged time ([48 h) (entry 10).
Under the optimized conditions, we continued to con-
duct the reactions of a range of selected terminal alkynes
with BnN3 (1 equiv), and found that all reactions proceeded
efficiently to give triazoles 2b–h in good to excellent iso-
lated yields (78–91 %) (Scheme 3). For the reaction of 1,3-
diethynylbenzene, the use of 1 equiv of BnN3 led to the
formation of mixed mono-triazole 2i and bis-triazole 2i0 in
54 and 15 % isolated yield, respectively. When 3 equiv of
BnN3 was employed, the full consumption of the triple
bonds was achieved in giving 2i0 as the sole product (iso-
lated: 90 %).
Table 1 Optimization of reaction conditions for cycloaddition of
terminal alkynes
Entrya,b
Catalytic conditions
Yield (%)c
1
2
3
4
5
6
7
8
9
10
a
0.4 mol% of 1/hexane
0.4 mol% of 1/CH2Cl2
0.4 mol% of 1/acetone
0.4 mol% of 1/THF
0.4 mol% of 1/DMF
0.4 mol% of 1/CH3CN
0.8 mol% of 1/CH3CN
0.3 mol% of 1/CH3CN
0.2 mol% of 1/CH3CN
None of 1/CH3CN
59
52
71
82
60
91
90
77
To rationalize the catalytic mechanism, we indepen-
dently treated phenylacetylene and BnN3 with 1 equiv of 1
in acetonitrile. After stirring for 4 h at room temperature,
the TLC analysis indicated no change of BnN3, whereas the
disappearance of phenylacetylene along with the formation
of a high polar complex. Therefore, it is speculated that the
catalytic process should begin with the formation of the
copper(I) acetylide intermediate as generally accepted for
the reaction of terminal alkynes (Scheme 4) [19, 28–30].
Regarding the hydride-donating capability of 11, we herein
58
_
All reactions were performed under a N2 atmosphere. For a given
catalytic system, it was observed that the reactions carried out under a
N2 atmosphere often resulted in better and more consistent yields of
2a than the reactions without the N2 protection, e.g. 71–84 % yields
for the reactions under the catalytic system of 0.4 mol % of 1/CH3CN/
rt (Table 1, entry 6) without the N2 protection
1
The hydride-donating ability of 1 has recently been realized by its
ability to reduce 2-cyclohexenone into cyclohexanone. Treatment of
2-cyclohexenone (10 mg, 0.104 mmol) with 1 (624 mg) in THF
(25 mL) at ambient temperature for 12 h resulted in 30 % conversion
into cyclohexanone as monitored by gas chromatography
b
All reactions were performed by using 1.6 M of 1 and 1 equiv of
BnN3
c
Isolated yield
123

Copper(I)-Cluster Catalyzed Azide-Alkyne Cycloaddition
575
Table 2 Optimization of reaction conditions for cycloaddition of
internal alkynes
Entrya
Catalytic conditions
Yield (%)b
1
0.4 mol% of 1/CH3CN/rt
0.8 mol% of 1/CH3CN/rt
0.8 mol% of 1/hexane/rt
0.8 mol% of 1/CH2Cl2/rt
0.8 mol% of 1/THF/rt
0.8 mol% of 1/H2O/rt
0.8 mol% of 1/DMF/rt
1 mol% of 1/DMF/rt
61
77
64
48
56
68
80
76
2
Scheme 4 Proposed mechanism for 1-catalyzed cycloaddition of
terminal alkynes
3
4
5
assume that the initial acetylide-formation in our case can
be particularly facilitated with the abstraction of the ter-
minal hydrogen by the hydride released from the central of
the cluster. The resulting acetylide intermediate then
engages in an addition with azide to afford a six-membered
metallacycle. The ring contraction of which produces a
copper-metallated triazole, which subsequently undergoes
the hydrogen-exchange with another alkyne to give the
triazole product. The released catalyst then participates in
another cycle of cycloaddition.
6
7
8
9
0.4 mol% of 1/DMF/rt
None of 1/DMF/rt
69
_
10
11
12
a
0.8 mol% of 1/DMF/120 °C
None of 1/DMF/120 °C
90
57
All reactions were performed with 1.6 M of 1 and 1 equiv of BnN3
b
Isolated yield
After completing the aforementioned works, we then
turned our attention to investigating the copper-catalyzed
1,3-dipolar cycloaddition of internal alkynes. So far, there
are only few examples on this type of reactions being
documented in literatures [23, 24]. Although the same goal
can nowadays be attended by ruthenium- or zinc-catalysis
[31, 32], yet the copper-based catalysts should be of more
use in practice in term of relatively low cost.
The application of the conditions in entry 11 of Table 2
to the reactions between benzyl, 1-adamantyl and
4-methoxybenzyloxycarbonyl azides and a variety of
internal alkynes led to the formation of triazoles 3b–3m/m0
in moderate to excellent yields (45–98 %) after isolation
(Scheme 5). As can be seen, the reactions proceeded effi-
ciently not only for the alkynes bearing the electron-with-
drawing carboxylate groups but also for the alkynes with
the alkyl and/or phenyl substituents known to be less
reactive for the 1,3-dipolar cycloaddition. Among which,
the yields of the reactions with BnN3 were uniformly
higher than those of the corresponding reactions with
1-adamantyl and 4-methoxybenzylcarbonyl azides (e.g. 3a
vs 3c and 3d), which could possibly be attributed to the less
favorable steric and/or electronic natures of the azides. In
addition, the formation of the mixtures of two regioisomers
(3g/3g0?3m/3m0) was observed for all unsymmetrically
substituted alkynes.
At first, the optimal conditions for terminal alkynes
(0.4 mol% of 1/acetonitrile) were attempted to the reaction
between BnN3 and diethyl acetylenedicarboxylate. After
proceeding for 12 h at room temperature, the reaction
produced triazole 3a in 61 % yield (Table 2, entry 1). A
better conversion could be obtained in acetonitrile when
the catalytic loading was increased to 0.8 mol% (entry 2,
77 %). In 0.8 mol% loading, we further tested several
solvents including hexane, dichloromethane, THF, water
and DMF (entries 3–7), and received the best yield from
DMF (entry 7, 80 %). In DMF, it was also observed that
the increment on the catalytic amount to 1 mol% did not
improve the yield (entry 8, 76 %) while the reduction to
0.4 mol% gave the poorer conversion (entry 9, 69 %).
Notably, no cycloaddition could be detected in DMF at
room temperature in the absence of 1 (entry 10), thus
validating the essential role that 1 played on the cycload-
dition. To further improve the yield, we then carried out the
reaction at 120 °C in DMF, and found that 3a could be
formed in an excellent yield under such conditions (entry
11, 90 %). A blank reaction at the same temperature still
could produce 3a but in much poorer yield (entry 12, 57 %)
to therefore confirm the effect of the cluster on the
cycloaddition.
Unlike observed for terminal alkynes, we found that
none of the employed internal alkynes could form detect-
able complex with 1 (1 equiv) alone in DMF. However,
after the azides were introduced to the reaction mixtures,
the gradual formation of the triazole products becomes
noticeable as monitored by TLC analysis. It is therefore
proposed that the cycloaddition can only be induced after
the cluster 1 simultaneously coordinates with both com-
ponents to give the formation of a six-membered metalla-
cyclic intermediate [22]. From which, the triazole
product(s) is generated accompanied by the release of the
cluster for the next catalytic cycle (Scheme 6). This pro-
posal agrees with the formation of the isomeric mixtures
123

576
B.-H. Lee et al.
Scheme 5 Copper cluster-
catalyzed 1,3-Dipolar
cycloaddition of internal
alkynes [a] Isolated yield.
[b] Ratio was determined by
integration of 400 MHz NMR
spectrum
Scheme 6 Proposed
mechanism for 1-catalyzed
cycloaddition of internal
alkynes
for the unsymmetrically substituted alkynes since the initial
addition step should not have a strong regioselective bias.
References
1. Tornøe CW, Christensen C, Meldal M (2002) J Org Chem
67:3057
2. Tornøe CW, Meldal M (2001) Peptidotriazoles: Copper(I)-cata-
lyzed 1,3-dipolar cycloaddition on solid-phase, peptide 2001. In:
Proceedings american peptide symposium. American Peptides
Society and Kluwer, San Diego, pp 263–264
3. Rostovtsev VV, Green LG, Fokin VV, Sharpless KB (2002)
Angew Chem Int Ed 41:2596
4. Lahann J (ed) (2009) Click chemistry for biotechnology and
material science. Wiley, Chichester
5. Whiting M, Tripp JC, Lin YC, Lindstrom W, Olson AJ, Elder JH,
Sharpless KB, Fokin VV (2006) J Med Chem 49:7697
6. Araga˜o-Leoneti V, Campo VL, Gomes AS, Field RA, Carvalho I
(2010) Tetrahedron 66:9475
7. Bock VD, Perciaccante R, Jansen TP, Hiemstra H, van Maar-
seveen JH (2006) Org Lett 8:919
8. Angell YL, Burgess K (2007) Chem Soc Rev 36:1674
9. Kalesh KA, Shi H, Ge J, Yao SQ (2010) Org Biomol Chem
8:1749
10. Amblard F, Cho JH, Schinazi RF (2009) Chem Rev 109:4207
11. Lutz JF (2007) Angew Chem Int Ed 46:1018
12. Johnson JA (2002) Biomacromolecules 37
4 Conclusion
In summary, the cluster 1 has been successfully applied as
a new and efficient catalyst to the 1,3-dipolar cycloaddition
for both terminal and internal alkynes. As compared with
those previously developed nitrogen- or phosphorus-based
ligands, our study has demonstrated that sulfur ligands may
have the comparable or even higher capability in enhancing
the activity of copper(I) ion, and consequently open a new
perspective for the development of novel catalytic systems.
At current stage, the exact role(s) that the sulfur ligand
within the cluster played on the catalysis is still not clear
for us. An investigation to solve this puzzle is ongoing in
our group.
Acknowledgments We are grateful to the National Science Council
of Republic of China (Taiwan) for financial support.
13. Qin A, Lam JWY, Tang BZ (2010) Chem Soc Rev 39:2522
123

Copper(I)-Cluster Catalyzed Azide-Alkyne Cycloaddition
577
¨
14. Iha RK, Wooley KL, Nystrom AM, Burke DJ, Kade MJ, Hawker
CJ (2009) Chem Rev 109:5620
15. Evans RA (2007) Aust J Chem 60:384
25. Lazreg F, Slawin AMZ, Cazin CSJ (2012) Organometallics
31:7969
26. Liao PK, Sarkar B, Chang HW, Wang JC, Liu CW (2009) Inorg
Chem 48:4089
27. Wang F, Fu H, Jiang Y, Zhao YF (2008) Green Chem 10:452
28. Hein JE, Tripp JC, Krasnova LB, Sharplesss KB, Fokin VV
(2009) Angew Chem Int Ed 48:8018
16. Huisgen R, Szeimies G, Moebius L (1967) Chem Ber 100:2494
17. Cuevas F, Oliva AI, Pericas M A (2010) Synlett 1873
18. Fu X, Albermann C, Zhang CC, Thorson JS (2005) Org Lett
7:1513
19. Himo F, Lovell T, Hilgraf R, Rostovtsev VV, Noodleman L,
Sharpless KB, Fokin VV (2005) J Am Chem Soc 127:210
20. van Dijk M, Rijkers DTS, Liskamp RM, van Nostrum CF,
Hennink WE (2009) Bioconjugate Chem 20:2001
21. Jin T, Yan M, Yamamoto Y (2012) ChemCatChem 4:1217
22. Meldal M, Tornøe CW (2008) Chem Rev 108:2952
29. Alonso F, Moglie Y, Radivoy G, Yus M (2010) Eur J Org Chem
2010:1875
30. Kuang GC, Guha PM, Brotheron WS, Simmons JT, Stankee LA,
Nguyen BT, Clark RJ, Zhu L (2011) J Am Chem Soc 133:13984
31. Boren BC, Narayan S, Rasmussen LK, Zhang L, Zhao H, Lin Z,
Jia G, Fokin VV (2008) J Am Chem Soc 130:8923
32. Meng X, Xu X, Gao T, Chen B (2010) Eur J Org Chem
2010:5409
´
´
23. Dıez-Gonzalez S, Correa A, Cavallo L, Nolan SP (2006) Chem
Eur J 12:7558
24. Li PH, Wang L, Zhang Y (2008) Tetrahedron 64:10825
123