Electrostatic immobilisation of copper(I) and copper(II) bis(oxazolinyl)pyridine catalysts on silica: application to the synthesis of propargylamines via direct addition of terminal alkynes to imines

Article abstract of DOI:10.1016/j.tetlet.2007.04.095

Copper(I) and copper(II) complexes of two bis(oxazolinyl)pyridines were immobilized on silica via electrostatic interactions. The catalytic activity of the immobilized catalysts in the direct addition of terminal alkynes to imines leading to propargylamines was investigated under a variety of reaction conditions. The performance of the immobilized catalysts compares very well with their homogeneous equivalents. When used in toluene, the catalysts could be recycled a number of times and maintained activity. This study is the first such report of the immobilization on silica in this manner of any bis(oxazolinyl)pyridine (pybox) complex.

Full text of DOI:10.1016/j.tetlet.2007.04.095

Tetrahedron Letters 48 (2007) 4387–4390
Electrostatic immobilisation of copper(I) and copper(II)
bis(oxazolinyl)pyridine catalysts on silica: application
to the synthesis of propargylamines via direct addition
of terminal alkynes to imines
Chiara McDonagh,a Peter O’Conghaile,a Robertus J. M. Klein Gebbinkb
and Patrick O’Learya,*
aNational University of Ireland, Galway, Department of Chemistry, Galway, Ireland
bUtrecht University, Organic Chemistry and Catalysis, Faculty of Science, Padualaan 8, 3584 CH Utrecht, The Netherlands
Received 22 February 2007; revised 11 April 2007; accepted 18 April 2007
Available online 22 April 2007
Abstract—Copper(I) and copper(II) complexes of two bis(oxazolinyl)pyridines were immobilized on silica via electrostatic interac-
tions. The catalytic activity of the immobilized catalysts in the direct addition of terminal alkynes to imines leading to propargyl-
amines was investigated under a variety of reaction conditions. The performance of the immobilized catalysts compares very well
with their homogeneous equivalents. When used in toluene, the catalysts could be recycled a number of times and maintained activ-
ity. This study is the first such report of the immobilization on silica in this manner of any bis(oxazolinyl)pyridine (pybox) complex.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
were used successfully in Diels–Alder cycloadditions.
In these reactions the immobilised catalysts were found
to give similar yields and enantioselectivities to the
homogeneous catalyst. In certain Diels–Alder reactions,
the opposite enantiomer was produced using the immo-
bilised catalyst compared to the homogeneous catalyst.
The catalyst could be recycled successfully three times.
As enantioselective catalysis becomes more developed,
preparation of many of the ligands and catalytic species
involves multistep syntheses of increasing complexity
starting from chiral materials such as amino acids. This
in turn results in an increase in the cost of such catalysts.
Many of these catalytic species offer significant improve-
ments on their simpler predecessors in terms of yield,
enantioselectivity or reactions which can be accessed.
Some reactions can be carried out using a very low
catalyst loading and as such, the expense of the catalyst
becomes less important. However, as is the case in the
three component coupling to produce propargylamines
described here, there remain reactions which require a
significant loading of catalyst to progress satisfactorily.1
In seeking to determine the general applicability of our
electrostatic immobilisation procedure to such catalysts
and their use we decided to study the three component
coupling of an amine, an aldehyde and an acetylene
which is an excellent method of making propargylam-
ines (Scheme 1).1,3,4 Propargylamines are useful building
blocks in organic chemistry and have found use in syn-
thesis due to the fact that elements of their structure are
found in natural products5 and compounds of pharma-
ceutical interest.6 They have been used in the synthesis
We have recently described a new technique for the elec-
trostatic immobilisation of copper(II) bisoxazolines on
silica.2 Catalytic species immobilised by this method
Keywords: Copper(I) bis(oxazolinyl)pyridines; Electrostatic immobili-
sation; Propargylamine; Silica.
*
Scheme 1. Three component coupling for the synthesis of
Corresponding author. Tel.: +353 91 492476; fax: +353 91 525700;
propargylamines.
e-mail: patrick.oleary@nuigalway.ie
0040-4039/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2007.04.095

4388
C. McDonagh et al. / Tetrahedron Letters 48 (2007) 4387–4390
of pyrrolidines,7 oxazoles8 and dynemicins.9 The obvi-
ous advantage of multi-component reactions is the syn-
thesis of a complex molecule from simple starting
materials with complete atom economy. A wide variety
of compounds are easily accessible given the ease with
which the starting materials can be varied. Enantioselec-
tive catalysts for this reaction have been developed lar-
gely since 200210 and avoid the traditional methods for
the synthesis of propargylamines which involved stoichio-
metric amounts of organometallic reagents reacting with
an imine.1 Some catalytic enantioselective processes
have been reported for this reaction but they lack gener-
ality and use a large catalyst loading.11,3,4 We felt that
recyclable catalysts applicable to this reaction would
be of use.
matographic grade) which had been dried under vacuum
for 2 h at 70 ꢁC.13 The heterogeneous mixture was stir-
red until the colour disappeared from the solvent and
the silica became coloured (ꢀ1 min). The organic layer
was decanted and the solid washed with two aliqouts
of the solvent to be used in the catalytic reaction. Typi-
cally, theoretical loadings of 2.8% copper were used in
this study.
The electrostatically immobilised Cu(I) (phenylpybox)
triflate was tested in the reaction of benzaldehyde, ani-
line and phenylacetylene. Initially, the amine and the
aldehyde were combined, without solvent, and heated
to 60 ꢁC for 2 h and the imine formed was added to
the catalyst along with phenylacetylene and toluene.
The reaction mixture was then heated to reflux and stir-
red for 20 h at which stage the crude reaction mixture
could be removed from the catalyst by decanting off
the liquid and washing the solid with toluene
(2 · 5 ml). The solid catalyst was now ready to be reused
and the product was isolated from the solution by
removing the solvent at reduced pressure. The results
are shown in Table 1.
Copper(I) complexes of bis(oxazolinyl)pyridines
(pybox) are catalytically active for this reaction. Singh
has reported a large number of individual reactions for
which 5 or 10 mol % copper(I) pybox hexafluorophos-
phates can be used resulting in the isolation of products
in high yields and often exceptional enantioselectivity.3
Similarly, Li has studied copper(I) pybox trifluoro-
methanesulfonates and found that they also catalysed
the reaction giving high yields and enantiomeric excesses
of over 90% in many cases.11 We are aware of one cova-
lently bound copper catalyst which has been used in the
enantioselective catalysis of this reaction.12 The neces-
sity to alter the catalyst structure to covalently immobi-
lise it limits any direct comparison which can be made
with the homogeneous catalyst. The covalent immobili-
sation process is often a very arduous process as it
requires the synthesis of a special ligand.
The homogeneous Cu(I)(phenylpybox)triflate catalyst
gave 100% conversion and 51% ee when the reaction
was conducted overnight in refluxing toluene. The
Cu(I)(phenylpybox)triflate catalyst electrostatically
immobilised on silica gave very similar results in terms
of percentage conversion. The enantioselectivity of the
reaction, though reduced in the third use of the catalyst
was similar to that of the homogeneous catalyst. This is
We envisaged electrostatic immobilisation of copper(I)
and copper(II) pybox trifluoromethane sulfonates on sil-
ica which would be the first example of immobilisation
of a pybox complex in this way, and also the first exam-
ple of the immobilisation of such copper(I) complexes
and would provide a recyclable catalyst for the synthesis
of propargylamines. This would broaden the applicabil-
ity of this very simple immobilisation technique, the pri-
mary advantages of which are the absence of any
complexity in the immobilisation process, the wide-
spread availability of silica (chromatographic grade)
and the lack of any need to alter the catalyst or ligand
to facilitate the immobilisation (Fig. 1).
Table 1. Immobilised Cu(I) (phenylpybox) catalysis of propargyl-
amine synthesis
Catalyst type Theoretical Cu
loading of silica (%)
Cycle Conva eeb (%)
(%)
Homo
Hetero
Hetero
Hetero
N/A
N/A
1st
2nd
3rd
100
97
98
51
2.8
2.8
2.8
47.5
44.5
35.5
91
The phenyl-substituted copper(I) pybox catalyst was
first generated in dichloromethane from the commer-
cially available ligand and copper(I) triflate, and this
solution was added to silica (silica gel 40–63 lm, chro-
a With respect to the aldehyde determined by 1H NMR analysis.
b Determined by HPLC (Daicel Chiralcel OD, 254 nm), hexane (0.1%
diethylamine):isopropyl alcohol, 90:10.
Figure 1. Schematic representation of the immobilisation of a copper(I) pybox catalyst on silica.

C. McDonagh et al. / Tetrahedron Letters 48 (2007) 4387–4390
4389
Table 3. Immobilised Cu(II)(phenylpybox) catalysis of propargyl-
amine synthesis
a significant result as it represents a number of new
achievements. This is the first reported immobilisation
of a copper(I) species in this manner. It is also the first
reported immobilisation of a pybox ligand by this meth-
od. In addition, it is one of very few enantioselective
methods for the catalysis of this reaction using an elec-
trostatically immobilised reusable catalyst. When the
immobilised catalyst was refluxed in toluene and the tol-
uene removed and used as solvent for the reaction (in
the absence of any further catalyst) no propargylamine
was detected indicating that catalysis is generally occur-
ring on the surface and there was minimal catalyst
leaching.
mol (%) Cu(II) Catalyst Cycle Solvent
type
Conva eeb (%)
(%)
5
5
5
5
5
5
10
10
10
10
Homo
Hetero
Hetero
Homo
Hetero
Hetero
Homo
Hetero
Hetero
Hetero
N/A
1st
2nd
N/A
1st
2nd
N/A
1st
2nd
3rd
DCM
DCM
DCM
Toluenec
Toluene 100
Toluene 100
DCM
DCM
DCM
DCM
32
52
29
82
83.5
83.5
81.5
62
56
68
85
80
79
81.5
98
99
76
52
a With respect to the aldehyde determined by 1H NMR analysis.
b Determined by HPLC (Daicel Chiralcel OD, 254 nm, hexane (0.1%
diethylamine):isopropyl alcohol, 90:10.
We sought to develop this method by varying the
amount of catalyst used, the oxidation state of the
metal, the substitution on the pybox ligand and the start-
ing aldehyde. The results obtained when the reaction
was conducted using Cu(I) (isopropylpybox) triflate
and its silica immobilised equivalent are shown in Table
2. The results using the isopropyl-substituted copper(I)
pybox complex were disappointing for both the homo-
geneous catalyst and its immobilised equivalent. The
supported catalyst displayed variable and generally
disappointing enantioselectivity but the activity was
enhanced as conversions in excess of 90% were observed
in all three runs. Our previous study of Cu(II) BOX
catalysis of Diels–Alder reactions had found that
though bis(oxazoline) BOX complexes bearing a phenyl
group were stable on immobilisation over multiple uses,
c This reaction was conducted at 40 ꢁC over 48 h.
run it was improved on the second use of the catalyst to
give a similar result to the homogeneous catalyst. This
result in itself is surprising given the homogeneous reac-
tion was conducted at 40 ꢁC and the heterogeneous reac-
tion was conducted at reflux.
Previous reports of the synthesis of propargylamines
using homogeneous copper catalysts have used
10 mol% of copper so we also investigated the immobi-
lised catalysts performance under these conditions (the
loading of copper on the silica was maintained at the
same level for this study). The runs using the homo-
geneous catalyst and the heterogeneous catalyst on first
use both gave very high conversions and enantioselectiv-
ities of 85% and 80% ee, respectively. Though the con-
version dropped on the second and third use of the
immobilised catalyst, the enantioselectivity was main-
tained. It is likely that a slightly longer reaction time
in the second and third run would allow for high conver-
sions to be achieved in these cases also.
t
those bearing a butyl group, decomposed during the
first use of the immobilised catalyst. In this case the
alkyl-substituted pybox ligand survived the immobilisa-
tion and several uses.
A study was conducted to see how the Cu(II) pybox cat-
alysts would perform on immobilisation (Table 3).
When the propargylamine synthesis was conducted in
dichloromethane at reflux using homogeneous
Cu(II)(phenylpybox), the conversion was very poor
but the enantioselectivity was good. When the catalyst
was electrostatically immobilised on silica, the perfor-
mance of the catalyst improved. The percentage conver-
sion increased by 20% and the enantioselectivity was
maintained. On reuse, the immobilised catalyst was less
active, and nearly identical to the homogeneous reaction
in terms of conversion and enantioselectivity. A similar
study of the immobilised catalyst activity in toluene at
reflux found that the reaction went to completion and
though there was a drop in enantioselectivity in the first
The decrease in activity of the immobilised catalyst on
reuse in dichloromethane is thought to be due to copper
leaching. Copper was detected by AAS in the product of
the reaction corresponding to <5% leaching. Reactions
in which the solid catalyst was removed at mid reaction
did not react further on continued refluxing. This
indicates that the copper which does leach is not catalyt-
ically active. A similar study with both copper(I) and
copper(II) catalysts reacting in toluene showed minimal
leaching (<1%). Any copper which did leach was found
to be catalytically inactive as above. These results
account for the catalysts activity being maintained on
reuse in toluene but diminishing activity on reuse in
dichloromethane.
Table 2. Immobilised Cu(I) (isopropylpybox) catalysis of propargyl-
amine synthesis at reflux in toluene
Catalyst type Theoretical Cu
loading of silica (%)
Cycle Conv eeb (%)
(%)a
Finally, to confirm that these results were not just
specific to the synthesis of the unsubstituted triphenyl
propargylamine the same reaction but using 4-bromo-
benzaldehyde was studied (Table 4). The heterogeneous
catalyst was used for two cycles over the course of
which the catalyst remained active even though only
5 mol % was used. Most encouragingly, the enantiomeric
excess of the product was found to be as good with the
Homo
Hetero
Hetero
Hetero
N/A
N/A
1st
2nd
3rd
31
98
97
95
33
25
17
30.5
2.8
2.8
2.8
a With respect to the aldehyde determined by 1H NMR analysis.
b Determined by HPLC (Daicel Chiralcel OD, 254 nm, hexane (0.1%
diethylamine):isopropyl alcohol, 90:10.

4390
C. McDonagh et al. / Tetrahedron Letters 48 (2007) 4387–4390
Table 4. Immobilised Cu(I)(phenylpybox) catalysis of the synthesis of
the bromosubstituted triphenyl propargylamine from 4-bromo-
benzaldehyde
then allowed to cool to room temperature and the silica
gel catalyst was allowed to settle. The solvent layer was
removed with a pasteur pipette and filtered through
Celite, the remaining solid was then washed with dry
solvent (2 · 5 ml) and the washings filtered through
Celite. The silica gel catalyst was left under dry solvent
(1 ml) and nitrogen for subsequent catalytic runs. The
combined extracts were evaporated under reduced pres-
sure to give a crude product as a yellow oil, which was a
mixture of starting materials and product. The percent-
Catalyst type
Cyclea
Convb (%)
eec (%)
Cu(I) bromide
Homogeneous
Heterogeneous
Heterogeneous
N/A
N/A
1st
100
94
87
N/A
51
53.5
66
2nd
71.3
a Reactions were conducted in toluene at reflux for 20 h.
b With respect to the aldehyde determined by 1H NMR analysis.
c Determined by HPLC (Daicel Chiralcel OD, 254 nm), hexane (0.1%
diethylamine):isopropyl alcohol, 90:10.
1
age conversion was determined by H NMR spectral
analysis. The product enantiomeric excess was deter-
mined by HPLC (Daicel Chiralcel OD, 254 nm), hex-
ane (0.1% diethylamine):isopropyl alcohol, 90:10 at a
flow rate of 0.5 ml/min. The major enantiomer was
detected at a retention time of 13.3 min and the minor
enantiomer at 16.4 min. The absolute configuration of
the product was assigned by analogy to the work of
Benaglia.4
heterogeneous catalyst as with the homogeneous equiv-
alent. Indeed, increased enantiomeric excess was ob-
served on the second cycle with the heterogeneous
catalyst which was also superior to the enantioselectivity
with the homogeneous catalyst.
In conclusion, the electrostatically immobilised bis-
(oxazolinyl)pyridine catalysts prepared in this study
are the first to be immobilised on silica by this method.
This report also contains the first example of a copper(I)
based catalyst immobilised by this method. The immobi-
lised catalysts were used in the enantioselective synthesis
of propargylamines. High conversions were obtained
using the copper(I) catalyst and copper(II) catalysts over
a number of cycles. Enantioselectivities were in line with
those obtained in the corresponding homogeneous reac-
tion. In toluene, the catalyst activity was maintained
over several cycles. In dichloromethane, excellent
enantioselectivities were obtained over three cycles but
some deactivation due to leaching was observed. This
study demonstrates the feasibility of using such electro-
static immobilisation to allow reuse of more specialist
(non commercially available) catalysts.
Acknowledgements
The authors gratefully acknowledge NUI-Galway for a
fellowship (C.M.D.) and Professor M. Hynes for assis-
tance with AAS. This publication has emanated from re-
search conducted with the financial support of Science
Foundation Ireland (RFP).
References and notes
1. Zani, L.; Bolm, C. Chem. Commun. 2006, 4263–4275.
2. O’Leary, P.; Krosveld, N. P.; De Jong, K. P.; van Koten,
G.; Klein Gebbink, R. J. M. Tetrahedron Lett. 2004, 45,
3177–3180.
3. Bisai, A.; Singh, V. K. Org. Lett. 2006, 8, 2405–2408.
4. Colombo, F.; Benaglia, M.; Orlandi, S.; Usuelli, F.;
Celentano, G. J. Org. Chem. 2006, 71, 2064–2070.
5. Jiang, B.; Xu, M. Angew. Chem., Int. Ed. 2004, 43, 216–
218.
2. Experimental; heterogeneous catalysis: representative
procedure
6. Shibasaki, M.; Ishida, Y.; Iwasaki, G.; Timori, T. J. Org.
Chem. 1987, 52, 3488–3489.
A mixture of benzaldehyde (50 mg, 4.72 · 10À4 mol) and
aniline (53 mg, 5.70 · 10À4 mol) was heated at 60 ꢁC for
2 h. Pybox ligand (5 mol %) and copper(I) triflate
(5 mol %) were dissolved in dry dichloromethane
(1 ml) in a dry Schlenk tube under nitrogen and stirred
for 1 h at rt. The catalyst complex solution was filtered
and the filtrate was added to silica gel (50 mg, 40–
63 lm pre dried for 1 h under vacuum at 70 ꢁC) in a
dry Schlenk tube under nitrogen. The mixture was stir-
red until the colour had disappeared from the solution
(ꢀ1 min). The silica gel, now coloured, was allowed to
settle and washed with dry dichloromethane (2 · 5 ml)
then left under the appropriate dry reaction solvent
(1 ml) and nitrogen. The imine mixture was added to
the solvent containing the immobilised catalyst complex
under nitrogen along with a dry solvent rinse (0.5 ml),
followed by phenylacetylene (74 mg, 7.29 · 10À4 mol).
The reaction mixture was stirred at reflux overnight,
7. Havey, D. F.; Sigano, D. M. J. Org. Chem. 1996, 61,
2268–2272.
8. Arcadi, A.; Cacchi, S.; Cascia, L.; Fabrizi, G.; Marinelli,
F. Org. Lett. 2001, 3, 2501–2504.
9. Yoon, T.; Shair, M. D.; Danishefsky, S. J.; Schulte, G. K.
J. Org. Chem. 1994, 59, 3752–3754.
10. Wei, C.; Li, C. J. J. Am. Chem. Soc. 2002, 124, 5638–5639.
11. Wei, C.; Mague, J. T.; Li, C.-J. Proc. Nat. Acad. Sci. 2004,
101, 5749–5754.
12. Weissberg, A.; Halak, B.; Portnoy, M. J. Org. Chem. 2005,
70, 4556–4559.
13. Our previous work2 involved a Diels–Alder reaction which
Evans et al.14 have reported fails if water is present on the
copper reaction centre. The method of pretreatment of the
silica reported here was found to be sufficient to prevent
catalyst deactivation indicating that at least free H2O was
removed.
14. Evans, D. A.; Johnson, J. S.; Olhava, E. J. J. Am. Chem.
Soc. 2000, 122, 1635–1649.