Design and synthesis of binuclear Co-salen catalysts for the hydrolytic kinetic resolution of epoxides

Article abstract of DOI:10.1016/j.catcom.2015.05.006

Three binuclear Co(III)-salen complexes have been synthesized based on a series of hydrophilic cyclic frameworks with different ring sizes. The catalytic performance of Co-salen complexes have further been evaluated in the hydrolytic kinetic resolution of racemic epoxides. And kinetic studies reveal that the binuclear Co-salen catalysts show a higher reactivity and better enantioselectivity in comparison to monometallic reference complex, indicating the two Co-salen units on the cyclic framework may work in a cooperative manner. Specifically, Co3, with the most flexible cyclic framework exhibits the best catalytic performance among the three catalysts, due to the efficient cooperative interactions between two cobalt centers.

Full text of DOI:10.1016/j.catcom.2015.05.006

Catalysis Communications 68 (2015) 101–104
Contents lists available at ScienceDirect
Catalysis Communications
journal homepage: www.elsevier.com/locate/catcom
Short communication
Design and synthesis of binuclear Co-salen catalysts for the hydrolytic
kinetic resolution of epoxides
a
c
Fengshou Wu a,b, , Kai Wang , Zaoying Li , Xunjin Zhu
⁎
b,⁎⁎
a
Key Laboratory for Green Chemical Process of the Ministry of Education, Wuhan Institute of Technology, Wuhan, PR China
Department of Chemistry, Hong Kong Baptist University, Hong Kong, PR China
College of Chemistry and Molecular Sciences, Wuhan University, Wuhan, PR China
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
Three binuclear Co(III)-salen complexes have been synthesized based on a series of hydrophilic cyclic
frameworks with different ring sizes. The catalytic performance of Co-salen complexes have further been
evaluated in the hydrolytic kinetic resolution of racemic epoxides. And kinetic studies reveal that the binuclear
Co-salen catalysts show a higher reactivity and better enantioselectivity in comparison to monometallic
reference complex, indicating the two Co-salen units on the cyclic framework may work in a cooperative manner.
Specifically, Co3, with the most flexible cyclic framework exhibits the best catalytic performance among the three
catalysts, due to the efficient cooperative interactions between two cobalt centers.
Received 13 March 2015
Received in revised form 7 May 2015
Accepted 8 May 2015
Available online 10 May 2015
Keywords:
Asymmetric catalysis
Ring-opening
Kinetic resolution
Cobalt
© 2015 Elsevier B.V. All rights reserved.
Salen ligands
1. Introduction
However, the oligomeric or polymeric catalyst system shows the less
hydrophobic character, which resulted in limited solubility in the epox-
The preparation of optically pure terminal epoxides has long stood
as a most significant target for asymmetric chiral building-block synthe-
sis, as they are valuable intermediates in the syntheses of a large variety
of pharmaceutical compounds [1–4]. Terminal epoxides available are
very inexpensive as racemic mixtures, and kinetic resolution is an
attractive strategy for the production of optically active epoxides
[5–7]. Employing chiral Co(salen) complexes as catalysts, Jacobsen
and co-workers reported the hydrolytic kinetic resolution (HKR) of
terminal epoxides with water as the nucleophile [8]. This method
provided direct access to unreacted chiral epoxides and 1,2-diols (also
valuable) in high enantiomeric excess and yield. Mechanistic investiga-
tion displayed a second-order dependence on the catalyst concentra-
tion, which is suggested to arise from the dual activation pathway in
the reaction. Generally, one cobalt center is proposed to bind to the
hydroxide and the other activates the epoxide in a catalytic cycle
[9–11]. Based on this cooperative bimetallic mechanism, several
bimetallic, poly-metallic, oligomeric catalysts, and dendrimeric Co(salen)
catalysts were prepared and examined in the HKR of racemic epoxides
with improved reactivity relative to monometallic catalysts [12–21].
ide–water mixtures of the HKR and required solvent to solubilize the
catalyst, complicating the otherwise simple isolation of the volatile
epoxide. Besides, due to the uncontrollable features of both cyclization
reactions, these support systems are produced as mixtures of cyclic
oligomers of different ring sizes ranging from dimer to decamer, which
prevented to fully understand of the origin of this improvement [22,23].
And the synthetic utility of multimeric catalysts prepared in this manner
is much limited by the inefficiency of their syntheses. In an ideal case, a
catalyst should be determined in structure and easily synthesized with
both high reactivity and selectivity.
On the other hand, it has been suggested in the literature that the
activity and selectivity of a cyclic catalytic system might be dependent
on the orientation, distance and flexibility of the involved Co(salen)
catalytic centers [22]. To this end, oligomers with different ring sizes
might have different catalytic reactivity. In this contribution, we
designed and synthesized a new class of binuclear salen ligands, in
which the salen moieties were attached to a series of hydrophilic cyclic
frameworks with different ring sizes, such as piperazine, 1,7-diaza-12-
crown-4 and 1,4,10,13-tetraoxa-7,16-diazacyclooctadecane, for the
purpose of tuning of the flexibility of cyclic organic compounds. The
synthesized compounds and intermediates were well-characterized
by NMR and mass spectroscopy (MS). The catalytic performance of
the three binuclear salen ligands, after cobalt(II) metalation, was
evaluated for the HKR of epoxides.
⁎
Correspondence to: F. Wu, Key Laboratory for Green Chemical Process of the Ministry
of Education, Wuhan Institute of Technology, Wuhan, PR China.
⁎⁎ Corresponding author.
E-mail addresses: wfs42@126.com (F. Wu), xjzhu@hkbu.edu.hk (X. Zhu).
http://dx.doi.org/10.1016/j.catcom.2015.05.006
1566-7367/© 2015 Elsevier B.V. All rights reserved.

102
F. Wu et al. / Catalysis Communications 68 (2015) 101–104
2. Experimental
2.2. General procedure for the preparation of binuclear Co(II)-salen
complexes
2.1. General procedure for the synthesis of dimeric salen ligands
A 20 mL flask equipped with a condenser was charged with bisalen
The synthetic procedures of compound 1, compound 2 and
bialdehydes were present in supporting information.
ligand (0.32 mmol) and anhydrous CH2Cl2 (5 mL) under a nitrogen atmo-
sphere. A degassed solution of Co(OAc)2·4H2O (100 mg, 0.40 mmol) in
5 mL methanol was then added to the flask with careful exclusion of
air. The red solution was heated at reflux for 24 h under a nitrogen
atmosphere, followed by the slow addition of degassed methanol
(2 mL). The precipitate was collected by filtration under the protection
of nitrogen, washed with degassed dichloromethane/methanol (1:10,
10 mL), and then dried under vacuum for 24 h to afford the binuclear
Co(II) salen as a dark red solid.
Co1: 95% yield, elemental analysis calcd (%) for C70H98Co2N6O4: C
69.75, H 8.19, N 6.97; found C 69.42, H 8.08, N 7.22.
Co2: 92% yield, elemental analysis calcd (%) for C74H106Co2N6O6: C
68.71, H 8.26, N 6.50; found C 68.96, H 8.18, N 6.81.
A 50 mL flask was charged with 3,5-di-tert-butylsalicylaldehyde
(327 mg, 1.39 mmol), (R,R)-1,2-diaminocyclohexane mono(hydrogen
chloride) (210 mg, 1.39 mmol), and activated 4 Å molecular sieves
(150 mg) under a nitrogen atmosphere. 10 mL anhydrous methanol
and 10 mL anhydrous ethanol were then added. After 4 h of stirring
at room temperature, a solution of bialdehyde (0.695 mmol) in
20 mL anhydrous CH2Cl2 were added, followed by the slow addition
of anhydrous triethylamine (0.39 mL, 2.8 mmol). The solution was
stirred for an additional 18 h, filtered, and the residue was washed
with CH2Cl2. The combined filtrates were concentrated under
reduced pressure and the solid was isolated by column chromatogra-
phy on silica gel (30/70, dichloromethane/hexane) to afford a yellow
powder.
Co3: 90% yield, elemental analysis calcd (%) for C77H112Co2N6O8: C
67.62, H 8.25, N 6.15; found C 67.03.73, H 8.45, N 6.07.
bisalen-1: 56% yield, 1H NMR (CDCl3, 400 MHz) δH ppm 13.82 (s,
2H, OH), 13.67 (s, 2H, OH), 8.29 (s, 2H, CHN), 8.26 (s, 2H, CHN), 7.29
(br, 2H, Ar salen-H), 7.12 (d, 2H, Ar salen-H), 6.96 (m, 4H, Ar
salen-H), 3.32 (s, 8H, NCH2), 2.36 (m, 8H, ArCH2 and NCH),
1.92–1.65 (br, 16H, CH2), 1.40 (s, 18H, (CH3)3), 1.38 (s, 18H,
(CH3)3), 1.26 (s, 18H, (CH3)3); exact mass calcd for C70H102N6O4
1089.65, found 1089.80.
2.3. General procedure for the hydrolytic kinetic resolution of (rac)-
epoxides [13]
The precatalyst Co(II) (0.01 mmol or 0.08 mmol on the basis of
cobalt) was dissolved in dichloromethane (1 mL) in a 2.5 mL flask.
Glacial acetic acid (20 μL) was added and the mixture was stirred in
the open air for 30 min. The solvent and the excess acetic acid were
roughly removed in vacuo. The residue was pumped under vacuum
overnight to give the Co(III)-salen-(OAc) as a dark brown solid. Racemic
epoxide (10 mmol) and chlorobenzene (100 μL, internal reference)
were added to dissolve the activated catalyst and the flask was im-
mersed into a water bath at RT. Deionized water (0.60 equiv., 108 μL,
6 mmol) was added to the system to start the reaction. The resolution
process was monitored by chiral GC. At each designed time, 2 μL sam-
ples were taken from the reaction mixture with a micro-syringe, diluted
with anhydrous diethyl ether (2 mL), and passed through a plug of silica
gel to remove the catalyst and water. The conversions and enantiomeric
excesses of epichlorohydrin were measured by chiral GC.
bisalen-2: 52% yield, 1H NMR (CDCl3, 400 MHz) δH ppm 13.83 (s,
2H, OH), 13.65 (s, 2H, OH), 8.30 (s, 2H, CHN), 8.25 (s, 2H, CHN), 7.29
(br, 2H, Ar salen-H), 7.24 (d, 2H, Ar salen-H), 6.97 (m, 4H, Ar
salen-H), 3.51 (s, 8H, OCH2), 3.30 (m, 8H, NCH2), 2.66 (m, 8H,
ArCH2 and NCH), 1.94–1.67 (br, 16H, CH2), 1.40 (s, 36H, (CH3)3),
1.22 (s, 18H, (CH3)3); exact mass calcd for C74H110N6O6 1177.45,
found 1177.90.
bisalen-3: 52% yield, 1H NMR (CDCl3, 400 MHz) δH ppm 13.83 (s,
2H, OH), 13.66 (s, 2H, OH), 8.31 (s, 2H, CHN), 8.25 (s, 2H, CHN), 7.28
(br, 2H, Ar salen-H), 7.23(d, 2H, Ar salen-H), 6.94 (m, 4H, Ar salen-H),
3.57 (s, 8H, OCH2), 3.32 (m, 8H, NCH2), 2.69 (m, 8H, ArCH2 and NCH),
1.95–1.61 (br, 16H, CH2), 1.42 (s, 48H, (CH3)3), 1.23 (s, 18H, (CH3)3);
exact mass calcd for C78H118N6O8 1265.53, found 1265.90.
Scheme 1. Synthesis of binuclear Co(III)-salen complexes.

F. Wu et al. / Catalysis Communications 68 (2015) 101–104
103
Table 1
HKR of racemic terminal epoxides catalyzed by binuclear Co-salen catalysts.
d
Entry
1
R
Co loading [mol%]a
0.1
Catalyst
Time [min]
Conversion [%]b
ee [%]c
Initial TOF [min−1
]
CH2Cl
Co1
Co2
Co3
Co1
Co2
Co3
Co1
Co2
Co3
650
525
410
405
380
51
52
52
51
50
50
43
42
47
N99
N99
N99
N99
N99
N99
82
3.2
4.2
7.7
3.5
4.6
8.6
1.5
1.1
2.0
2
3
(CH3)3CH2
Ph
0.1
0.8
235
1440
1440
1440
79
87
a
Catalyst loading on a per Co basis relative to epoxides.
Determined by chiral GC.
The ee corresponds to the enantiomeric excess of the unreacted epoxide determined by chiral GC.
Calculated from plot of conversion vs. reaction time.
b
c
d
3. Results and discussion
3.2. HKR using racemic epoxides
3.1. Preparation of binuclear Co(III)-salen complexes
First, the catalytic performances of three binuclear [Co(salen)]
catalysts were examined in the HKR of racemic epichlorohydrin as a
model substrate. The reaction was carried out at room temperature in
the presence of 0.1 mol% catalyst based on the total cobalt concentration,
with 0.6 equivalents of water as the nucleophile and chlorobenzene as an
internal standard. The conversions and enantiomeric excesses of the
substrate were monitored by chiral GC analysis. The catalytic results are
summarized in Table 1 and the kinetic plots of the conversions and ee
versus the reaction time are presented in Figs. 1 and 2, respectively. In
the previous work, when the standard monometallic [Co(salen)]
complex was employed as catalyst, the resolution proceeded to an ee of
only 20% for the remaining epichlorohydrin with 19% conversion in 1 h.
An ee of greater than 99% was reached upon extending the reaction
time to 15 h [18]. In contrast, all of the synthesized binuclear [Co(salen)]
catalysts afford greater than 99% ee with 50% conversion in a shorter
reaction time under the same conditions, indicating the higher activity
and enantioselectivity relative to Jacobsen's catalyst. Additionally, the
HKR reactions catalyzed by Co3 (initial TOF = 7.7 min−1) occurred at a
The procedure for the synthesis of Co(III)-salen complexes is
depicted in Scheme 1. The bisalen ligands were prepared by following
the one-pot protocol described in the literature [24,25], which
employed the condensation of one equivalent of dialdehyde, two
equivalents of HCl-protected diamine (2), and two equivalents of
3,5-di-tert-butyl-2-hydroxybenzaldehyde in the presence of 4 Å
molecular sieves. Addition of anhydrous triethylamine allowed the
condensation to be complete, after which the unsymmetrical bisalen
ligands were isolated by column chromatography. The bisalen ligands
were characterized by NMR spectroscopy and mass spectrometry.
Metalation using cobalt(II) acetate tetrahydrate with bisalen ligands in
a dichloromethane/methanol mixture afforded the desired binuclear
Co(II) salen in good yields. The CoII precatalysts were oxidized prior to
use by stirring in dichloromethane solution under air in the presence
of glacial acetic acid, affording the corresponding activated catalysts in
CoIII form with AcO− as the counterion.
50
40
30
100
80
60
Co1
Co2
Co3
20
40
Co1
Co2
Co3
10
0
20
0
0
100
200
300
400
500
600
700
0
100
200
300
400
500
600
700
Time / min
Time / min
Fig. 1. Plot of conversion vs. reaction time in the HKR of epichlorohydrin.
Fig. 2. Plot of ee vs. reaction time in the HKR of epichlorohydrin.

104
F. Wu et al. / Catalysis Communications 68 (2015) 101–104
100
50
40
30
20
10
0
80
60
Co1
Co2
Co3
40
20
0
Co1
Co2
Co3
0
100
200
300
400
500
0
100
200
Time / min
300
400
500
Time / min
Fig. 3. Plot of conversion vs. reaction time in the HKR of 1,2-epoxyhexane.
Fig. 4. Plot of ee vs. reaction time in the HKR of 1,2-epoxyhexane.
higher rate than that of Co1 (initial TOF = 3.2 min−1) and Co2 (initial
TOF = 4.2 min−1), indicating the flexible cyclic framework could
improve the activity of the catalyst to some extent. This is probably due
to the fact that the attached Co-salen units on flexible cyclic compound
are allowed to rotate more freely, potentially leading to enhanced
cooperative interactions. As shown in Table 1, the time required to
reach greater than 99% ee with 50% conversion increased from 410 min
to 650 min in the HKR of epichlorohydrin when the catalyst changed
from Co3 to Co2. Co1 shows the lowest activity among three catalysts,
probably owing to its more constrained cooperative geometrical environ-
ment. To assess the versatility of the catalysts, we also tested the activity
of the catalysts with other terminal epoxides. As shown in Table 1, the
HKR of 1,2-epoxyhexane was completed in 235 min with 0.1 mol% of
cobalt loading, affording an enantiomeric excess of 99% (conversion
51%) when Co3 was used. The order of reactivity of the catalysts for
HKR of 1,2-epoxyhexane was Co3 N Co2 N Co1, which is consistent
with that of epichlorohydrin. The kinetic plots of the conversions and
ee versus the reaction time for 1,2-epoxyhexane were presented in
Figs. 3 and 4, respectively. In addition, a more challenging substrate,
such as styrene oxide was resolved by three catalysts. As shown in
Table 1, the HKR was not completed even if the reaction was carried
out for a prolonged period of time with the cobalt loading elevated to
0.8 mol%. However, Co3 is still the most active catalyst among three
binuclear catalysts, which could result in an ee of 87% and conversion of
47% in 24 h. The kinetic plots for styrene oxide are placed in the ESI†
for convenience.
Acknowledgment
Financial supports from the National Natural Science Foundation of
China (20902071) and the Project of Educational Commission of
Hubei Province of China (D20141506) are acknowledged. XZ thanks
Hong Baptist University (FRG2/14-15/034 and FRG1/14-15/058) for
the financial support.
Appendix A. Supplementary data
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.catcom.2015.05.006.
References
[1] H. Zhang, M.W. Grinstaff, J. Am. Chem. Soc. 135 (2013) 6806–6809.
[2] G. Wu, P. Xu, X. Lu, Y. Zu, S. Wei, W. Ren, D.J. Darensbourg, Macromolecules 46
(2013) 2128–2133.
[3] M.P.F. Pepels, M. Bouyahyi, A. Heise, R. Duchateau, Macromolecules 46 (2013)
4324–4334.
[4] B. Gao, R. Duan, X. Pang, X. Li, Z. Qu, Z. Tang, X. Zhuang, X. Chen, Organometallics 32
(2013) 5435–5444.
[5] J.A. Birrell, E.N. Jacobsen, Org. Lett. 15 (2013) 2895–2897.
[6] E.D. Cross, L.E.N. Allan, A. Decken, M.P. Shaver, J. Polym. Sci. A Polym. Chem. 51
(2013) 1137–1146.
[7] G. Wu, P. Xu, Y. Zu, W. Ren, X. Lu, J. Polym. Sci. A Polym. Chem. 51 (2013) 874–879.
[8] M. Tokunaga, J.F. Larrow, F. Kakiuchi, E.N. Jacobsen, Science 277 (1997) 936–938.
[9] L.P.C. Nielsen, C.P. Stevenson, D.G. Blackmond, E.N. Jacobsen, J. Am. Chem. Soc. 126
(2004) 1360–1362.
[10] J.A. Kalow, A.G. Doyle, J. Am. Chem. Soc. 133 (2011) 16001–16012.
[11] J. Park, K. Lang, K.A. Abboud, S. Hong, J. Am. Chem. Soc. 130 (2008) 16484–16485.
[12] S.J. Wezenberg, A.W. Kleij, Adv. Synth. Catal. 352 (2010) 85–91.
[13] X. Zheng, C.W. Jones, M. Weck, Chem. Eur. J. 12 (2005) 576–583.
[14] C.S. Gili, K. Venkatasubbaiah, N.T. Phan, M. Weck, C.W. Jones, Chem. Eur. J. 14 (2008)
7306–7313.
4. Conclusion
[15] K. Venkatasubbaiah, C.S. Gili, T. Takatani, C.D. Sherrill, C.W. Jones, Chem. Eur. J. 15
(2009) 3951–3955.
[16] X. Zheng, C.W. Jones, M. Weck, Adv. Synth. Catal. 350 (2008) 255–261.
[17] X. Zhu, K. Venkatasubbaiah, M. Weck, C.W. Jones, J. Mol. Catal. A Chem. 329 (2010)
1–6.
[18] X. Zhu, K. Venkatasubbaiah, M. Weck, C.W. Jones, ChemCatChem 2 (2010)
1252–1259.
[19] X. Zheng, C.W. Jones, M. Weck, J. Am. Chem. Soc. 129 (2007) 1105–1112.
[20] R.E. Key, K. Venkatasubbaiah, C.W. Jones, J. Mol. Catal. A Chem. 366 (2013) 1–7.
[21] L. Canali, E. Cowan, H. Deleuze, C.L. Gibson, D.C. Sherrington, J. Chem. Soc. Perkin
Trans. 1 (2000) 2055–2066.
[22] Y. Liu, J. Rawlston, A.T. Swann, T. Takatani, C.D. Sherrill, P.J. Ludovice, M. Weck,
Chem. Sci. 2 (2011) 429–438.
[23] R.M. Haak, S.J. Wezenberg, A.W. Kleij, Chem. Commun. 46 (2010) 2713–2723.
[24] M. Holbach, X. Zheng, C. Burd, C.W. Jones, M. Weck, J. Org. Chem. 71 (2006)
2903–2906.
In this context, we have prepared and characterized three novel
binuclear Co(II)-salen complexes based on different cyclic frameworks.
Upon aerobic oxidation under acidic conditions, the catalysts showed
moderate to good activity in the HKR of a variety of racemic terminal
epoxides under neat conditions. The kinetic data revealed that the
catalyzed reaction may follow an intramolecular cooperative pathway.
In addition, the ring size of the cyclic organic compounds has an effect
on the catalytic efficiency and selectivity. Co3 exhibited a better catalytic
performance in comparison to other two binuclear Co-salen catalysts in
all cases, probably due to its more flexible character. We believe this
rational design and analysis of linked metal systems will shed light on
the basis of these phenomena and allow their extension into general
principles for catalyst design.
[25] E.J. Campbell, S.T. Nguyen, Tetrahedron Lett. 42 (2001) 1221–1225.