Chiral periodic mesoporous copper(II) bis(oxazoline) phenylene-silica: A highly efficient and reusable asymmetric heterogeneous catalyst

Article abstract of DOI:10.1016/j.jcat.2014.09.017

We describe the preparation of an effective and reusable heterogeneous asymmetric catalyst. A novel chiral periodic mesoporous phenylene-silica containing high density of bis(oxazoline) moieties is prepared by co-condensation method with 1,4-bis(triethoxy

Full text of DOI:10.1016/j.jcat.2014.09.017

Journal of Catalysis 320 (2014) 63–69
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Chiral periodic mesoporous copper(II) bis(oxazoline) phenylene–silica:
A highly efficient and reusable asymmetric heterogeneous catalyst
a,
Mirtha A.O. Lourenço a, Liliana Carneiro b, Alvaro Mayoral c, Isabel Diaz d, Ana R. Silva b, , Paula Ferreira
⇑
⇑
a Department of Materials and Ceramics Engineering, CICECO, Universidade de Aveiro, 3810-193 Aveiro, Portugal
b Department of Chemistry, CICECO, Universidade de Aveiro, 3810-193 Aveiro, Portugal
c Laboratorio de Microscopias Avanzadas, Instituto de Nanociencia de Aragon, Universidad de Zaragoza, Edificio I+D, Mariano Esquillor, 50018 Zaragoza, Spain
d Instituto de Catálisis y Petroleoquímica, CSIC, c/Marie Curie 2, 28049 Madrid, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
We describe the preparation of an effective and reusable heterogeneous asymmetric catalyst. A novel chi-
ral periodic mesoporous phenylene–silica containing high density of bis(oxazoline) moieties is prepared
by co-condensation method with 1,4-bis(triethoxysilyl)benzene. After copper(II) coordination, the mate-
rial is extremely efficient on the kinetic resolution of the 1,2-diphenylethane-1,2-diol with persistent
high enantioselectivities (91 - > 99%) and yields (46–43% in maximum 50% resolution) at least for five
consecutive cycles. Characterization of the material after the catalytic experiments showed that the het-
erogeneous catalyst was very robust keeping the integrity of the structure.
Received 28 April 2014
Revised 23 August 2014
Accepted 22 September 2014
Available online 20 October 2014
Keywords:
Heterogeneous catalysis
Asymmetric catalysis
Homogeneous catalysis
Periodic mesoporous organosilicas
Co-condensation
Ó 2014 Elsevier Inc. All rights reserved.
1. Introduction
pore surfaces [17] and high recyclability [15]. The well dispersed
catalytic sites partially integrated on the wall of these materials
Bis(oxazoline) are very versatile chiral ligands in asymmetric
catalysis (Scheme 1) and a large number of organic transformations
have been enantioselectively catalyzed by their transition metal
complexes [1]. These ligands are commercially available, but are
expensive. Therefore there has been an intense research effort in
order to make them recyclable and reusable, thus economical
[2–6]. The most successful strategy has been the covalent immobi-
lization onto several supports [2–5]. Asymmetric heterogeneous
catalysts obtained using some organic polymers [7,8] or siliceous
mesocellular foams [9,10] as supports have been superior to those
using copper(II) bis(oxazoline) post-grafted onto ordered mesopor-
ous silicas [11–13] or mesoporous carbons [12,13]. The enantiose-
lectivity was influenced by the active phase density [6,12–14] and
the capping of the siliceous mesocellular foam silanols [9,10,14].
The design of periodic mesoporous organosilicas (PMOs) con-
taining catalytic active sites as part of the material framework is
a relatively new strategy for the preparation of heterogeneous cat-
alysts based on versatile and successful homogeneous catalysts
[15–23]. These materials show high density of active sites on the
offer reduced resistance to the diffusion of molecules within the
mesochannels and are more resistant to leaching than if they were
grafted onto the ordered mesoporous silica [17].
The synthesis of PMOs incorporating chiral privileged ligands is
still quite un-explored and the synthesis of chiral PMOs with high
catalytic activity and enantioselectivity remains a challenge, espe-
cially in the case of bis(oxazoline) ligands [22,23].
Hence, here we describe the synthesis and characterization of a
novel periodic mesoporous organosilica bearing a high density of
chiral bis(oxazoline) ligands by co-condensation of 1,4-bis
(triethoxysilyl)benzene (BTEB) with trimethoxypropylsilane di-
functionalized commercial bis(oxazoline) [13,24] (Scheme 2) in
the presence of a surfactant. After coordination of [Cu(OTf)2], this
material acts as a very efficient asymmetric heterogeneous catalyst
in the kinetic resolution of the 1,2-diphenylethane-1,2-diol with
high enantioselectivity and yield, keeping such high performance
for at least five consecutive cycles.
2. Experimental
⇑
Corresponding authors at: Departamento de Química, CICECO, Universidade de
2.1. Chemicals and reagents
Aveiro, Campus Universitário de Santiago, 3810-193 Aveiro, Portugal. Fax: +351 234
401 470 (A.R. Silva). Department of Materials and Ceramics Engineering, CICECO,
Universidade de Aveiro, 3810-193 Aveiro, Portugal (P. Ferreira).
E-mail address: ana.rosa.silva@ua.pt (A.R. Silva).
Copper(II) trifluoromethanosulfonate (copper(II) triflate,
[Cu(OTf)2], 98%), 2,20-methylenebis[(4S)-4-phenyl-2-oxazoline]
http://dx.doi.org/10.1016/j.jcat.2014.09.017
0021-9517/Ó 2014 Elsevier Inc. All rights reserved.

64
M.A.O. Lourenço et al. / Journal of Catalysis 320 (2014) 63–69
(1, 97%), (S)-(–)-2,2’-isopropylidenebis(4-phenyl-2-oxazoline) (2,
97%), butyl lithium solution 1.6 M in hexane, 3-iodopropyltrimeth-
oxysilane (IPS, P95.0%), triethylamine (Et3N, P99%), dry tetrahy-
drofuran (THF, P99.9%), (R,R)-1,2-diphenyl-1,2-ethanediol (99%),
(S,S)-1,2-diphenyl-1,2-ethanediol (99%), N,N-diisopropylethyl-
amine (DIPEA, 99%), benzoyl chloride (99%), potassium hydroxide
(p.a.) and potassium bromide (FT-IR grade, P99%) were purchased
from Aldrich and used as received. Ethanol (p.a.) and methanol
(p.a.) were from Riedel de Häen. Dichloromethane, n-hexane,
isopropanol and ethyl acetate were from Romil and HPLC grade.
The octadecyltrimethylammonium bromide surfactant template
(ODTMABr, 98%) was obtained from Aldrich.
0.3 mL of triethylamine were dropped [13]. The mixture was
refluxed, under inert atmosphere, for 6 h followed by stirring at
room temperature for 48 h. The obtained material was filtered,
washed successively 2 ꢀ 20 mL of THF, 2 ꢀ 20 mL of CH2Cl2,
2 ꢀ 20 mL of MeOH, refluxed with 20 mL of THF for 6 h, under inert
atmosphere, and stirred at room temperature for 24 h. After filtra-
tion, the material was dried in an oven at 60 °C overnight.
2.5. Physical and chemical methods
Elemental analysis was performed in triplicate by ‘‘Servicio de
Análisis Instrumental’’, CACTI Vigo, Universidade de Vigo, Spain.
The copper ICP-AES was performed at ‘‘Laboratório Central de Aná-
lises’’ of the University of Aveiro, Portugal. FTIR spectra were col-
lected in the range 400–4000 cmꢁ1 at room temperature using a
resolution of 4 cmꢁ1 and 256 scans by attenuated total reflectance
(ATR) using a Bruker Tensor 27 spectrophotometer; the samples
were dried in an oven at 100 °C for 12 h prior to the analysis. Ther-
mogravimetry analyses (TGA) were performed under air flux with
a ramp of 5 °C/min in a TGA apparatus, model Shimadzu TGA-50.
Nitrogen adsorption isotherms at ꢁ196 °C were measured in an
automatic apparatus (Gemini V 2.00 instrument model 2380;
Micromeritics). Before the adsorption experiments the samples
were outgassed under vacuum overnight at 150 °C (to minimize
destruction of functionalities) to an ultimate pressure of 1024 mbar
and then cooled to room temperature. The specific surface areas
(SBET) were determined by the BET method. The pore size distribu-
tions (PSD, the differential volume adsorbed with respect to the dif-
ferential pore size per unit mass as a function of pore width) were
calculated by the BJH method using the modified Kelvin equation,
using the adsorption branch of the experimental data [27]. The
adsorption branch was used as it is known to be more accurate spe-
cially if the typical hysteresis of the mesoporous materials is of H2
type. In this case, the pore size distribution obtained from the
desorption branch of the isotherm noticeably deviates from that of
the adsorption branch and gives an incorrect pore size distribution
[28]. X-band electron paramagnetic resonance (EPR) was performed
at RIAIDT, Universidade de Santiago de Compostela, Spain, with fre-
quency of 9.434 GHz and 100 kHz field modulation at 120.5 K. The
reported EPR parameters were obtained by simulation using the
program Win EPR Simfonia (Bruker) assuming axial spin Hamiltoni-
ans. The values of g\ and A\ are less accurate because of their depen-
dence on the line widths (ꢂ100 G) used in the simulations.
2.2. Synthesis of the 1-Ph-PMO material
This method has three main steps: (i) functionalization of ligand
1 [13], (ii) hydrolysis and co-condensation of functionalized 1 with
1,4-bis(triethoxysilyl)benzene (BTEB) [25] and (iii) extraction of the
octadecyltrimethylammonium bromide (OCTMABr) template. First,
0.60 mmol of 1 were reacted with 1.49 mmol of butyl lithium
(1.6 M in hexane, BuLi) in 10 ml of dry tetrahydrofuran, under inert
atmosphere. Then this solution was cooled with liquid nitrogen and
1.97 mmol of 3-iodopropyltrimethoxysilane were added drop-wise
during 30 min. After stirring for 3 days at room temperature, under
inert atmosphere, the color changed to red-brown [13]. Secondly,
the synthesis of 1-Ph-PMO was performed using a mixture with
molar ratios BTEB: functionalized 1: ODTMABr: NaOH: H2O: THF
of 1: 0.23: 1.15: 5.07: 647: 7.44. In a typical synthesis the ODTMABr
was dissolved in a mixture of ultrapure water and 6 M aqueous
sodium hydroxide at 20–60 °C. The precursors mixture (functional-
ized 1 and BTEB) previously dissolved in THF, was subsequently
added drop-wise under vigorous stirring, at room temperature. This
solution was kept for 10 min in an ultrasonic vessel and stirred for
24 h at room temperature. After 24 h ageing in a Teflon flask at
100 °C, the resultant precipitate was recovered by filtration and
dried at 60 °C. Finally, the surfactant was removed through solvent
extraction. 0.5 g as-synthesized sample was stirred in 125 mL eth-
anol, for 12 h at 90 °C. The powder was filtered, washed with etha-
nol and water and dried overnight at 60 °C.
2.3. Preparation of the Cu@1-Ph-PMO material
To 0.24 g of the 1-Ph-PMO material a solution of 0.21 mmol of
copper(II) triflate in 40 mL of methanol were added. After refluxing
the mixture for 24 h, the solid was isolated by filtration, washed
with methanol and dried under vacuum at 60 °C for 6 h. A light
green material was obtained.
Powder X-ray diffraction (PXRD) data were collected with a
Phillips X’Pert MPD diffractometer using Cu K
a
radiation. 13C,
29Si and spectra were recorded on a Bruker Avance III 400 spec-
trometer operating at 9.4 T. 13C cross-polarization magic-angle
spinning (CP MAS) NMR spectra were recorded with 4
ls
1H 90°
mR) of 8 kHz
pulse, a contact time (CT) of 1 ms, a spinning rate (
2.4. Preparation of the graf_Cu-1@Ph-PMO material (post-grafting)
and recycle delay (RD) of 4 s. 29Si MAS NMR spectra were collected
employing a 40° flip angle pulse, a mR of 5 kHz and RD of 60 s. 29Si
The phenylene PMO (Ph-PMO) was synthetized following the
literature procedures [26] by hydrolysis and condensation of BTEB
in the presence of an ODTMABr surfactant. The template extraction
was made using an ethanol–HCl solution. The Ph-PMO (0.37 g) was
dried at 120 °C under vacuum atmosphere during 3 h to activate
the pores. Then 10 mL of dry dichloromethane containing
0.089 mmol of dissolved copper(II) functionalized 1 (see 2.i) and
CP MAS NMR spectra were acquired with a 4 l
s 1H 90° pulse, a CT
of 8 ms, a mR of 5 kHz and RD of 5 s. The 13C and 29Si NMR spectra
were quoted in ppm from trimethylsilane. The samples for trans-
mission electron microscopy (TEM) analysis were finely crushed
with mortar and pestle and dispersed in ethanol. Cs-corrected
STEM-HAADF observations were carried out in a FEI Titan X-FEG
300-60, operated at 80 kV, equipped with a CEOS spherical aberra-
tion corrector for the electron probe, an EDAX EDS detector and a
Gatan Tridiem Energy Filter for spectroscopy measurements.
H3C CH3
O
O
O
O
N
N
2.6. Catalysis experiments
N
N
Ph
Ph
Ph
Ph
All the catalytic reactions of the prepared materials were per-
formed in batch reactors at atmospheric pressure and with constant
stirring. The kinetic resolution of 1,2-diphenylethane-1,2-diol (3)
1
2
Scheme 1. Chiral bis(oxazoline) ligands.

M.A.O. Lourenço et al. / Journal of Catalysis 320 (2014) 63–69
65
(i) BuLi, IPS, THF, 3 days
Ph
Ph
O
(ii) BTEB, H2O, NaOH, ODTMABr
O
O
O
O
Si
Si
(iii) reflux, EtOH, overnight
N
N
O
N
N
O
Ph
Ph
O
O
1
O
1-Ph-PMO
Scheme 2. Synthesis of the periodic mesoporous organosilica 1-Ph-PMO.
was performed at 0 °C using 0.48 mmol (R,R)-1,2-diphenylethane-
1,2-diol, 0.48 mmol (S,S)-1,2-diphenylethane-1,2-diol, 1.00 mmol
were reached by post-grafting trimethoxypropylsilane functional-
ized 1 onto conventional ordered mesoporous silicas [13]. After
DIPEA (170
ll), amount of heterogeneous catalyst containing 1.0%
coordination of [Cu(OTf)2] complex,
a
copper loading of
mol Cu and 0.50 mmol of benzoyl chloride (58
ll) in 5.00 ml of
120 mol/g onto the 1-Ph-PMO (Cu@1-Ph-PMO) was obtained by
l
dichloromethane [13,29]. The mixture was stirred for 24 h and after
filtration of the heterogeneous catalyst the solvent was evaporated
from the filtrate and the monobenzoylated product (4, Scheme 3)
isolated by column chromatography over silica gel using n-hex-
ane/ethyl acetate 3:1 as eluent. The 4 enantiomeric excess was
determined by HPLC at 254 nm using a Chiralcel OD column
ICP-AES, yielding an excess of 1 to copper essential to the catalytic
asymmetric process (Table 1). For comparison, a similar material
was prepared via secondary grafting, i.e. the trimethoxypropylsi-
lane functionalized copper(II) bis(oxazoline) onto [13] the
phenylene PMO. Herein, the grafted material is denoted graf_Cu-1@
Ph-PMO. In this material the ligand and the copper loadings were
(250 mm ꢀ 4.6 ID, 5
l
m) and n-hexane/isopropanol 9:1 as eluent.
calculated to be 96 and 57 lmol/g (Table 1), respectively, which
are much lower values than for Cu@1-Ph-PMO.
The retention times of the (R)-4 and (S)-4 enantiomers were identi-
fied by comparison with those of a racemic 4 (see Figs. S1 and S2,
SD). The reaction selectivity (S) was calculated based on the isolated
yields of 4 and respective enantiomeric excess by using the formu-
lae: ln[1 ꢁ yield(1 + ee)]/ln[1 ꢁ yield (1 ꢁ ee)]. The isolated materi-
als at the end of the reactions were washed extensively with the
appropriate solvent, dried under vacuum and reused in another
cycle using the same experimental procedure. Control experiments
were also performed using the same experimental procedure in
homogeneous phase with equimolar quantities of [Cu(OTf)2] plus
1 or 2 in order to compare with the heterogeneous ones.
Due to the high 1 ligand content, the characteristic C = N
stretching vibration can be clearly seen at 1649 cmꢁ1 in the ATR–
FTIR spectrum of the novel 1-Ph-PMO (Fig. 1), besides of other
characteristic vibrations at 3065 cmꢁ1 (Fig. S3, SD), between
1500–1400 and 800–700 cmꢁ1. The C=N stretching vibration is
shifted to lower energy when compared with the free 1 ligand.
New very strong Si–O stretching vibrations of the organosilica
framework can now be also observed around 1043 cmꢁ1 in the
1-Ph-PMO spectrum. It is noteworthy that the presence of isolated
silanols cannot be distinguished at 3750 cmꢁ1 (Fig. S3, SD). The
ATR–FTIR spectrum of the Cu@1-Ph-PMO is very similar to that
of 1-Ph-PMO, but the presence of S=O stretching vibrations from
the sulfonate group of the triflate anion can be detected at
3. Results and discussion
638 cmꢁ1
.
Initially, the commercial bis(oxazoline) ligand 1 (Scheme 1) was
functionalized with two trimethoxypropylsilane groups (Scheme 2),
using a reported procedure [13,24]. The 1-Ph-PMO (Scheme 2) was
synthesized following a typical co-condensation synthesis [30]
with minor modifications in the presence of octadecyltrimethylam-
monium bromide (ODTMABr), which was extracted at the end of
the reaction.
Elemental analysis of 1-Ph-PMO shows that the material has
nitrogen, probing the presence of the bis(oxazoline) ligand 1. Tak-
ing into consideration that each 1 ligand contains two atoms of
The X-band EPR spectrum of the Cu@1-Ph-PMO material was
taken at 120.5 K in the solid state and is depicted in Supplementary
Data (Fig. S4, SD), together with the simulated spectra. It shows a
signal typical of magnetically diluted copper species where the
hyperfine structure from the coupling of the free electron with
the Cu(II) nucleus (I = 3/2) can be clearly seen. The spectrum is
axial with g|| and g\ of 2.265 and 2.075, respectively, and
A|| = 182 ꢀ 10ꢁ4 cmꢁ1 obtained by simulation of the spectra
(Fig. S4, SD). When compared to the reported parameters for
[Cu(OTf)2] in frozen solution (g|| = 2.412, g\ = 2.083 and
A|| = 134.5 ꢀ 10ꢁ4 cmꢁ1), the g values are lower and the higher A||
indicates that there are nitrogen atoms in the coordination sphere.
The simulated parameters are also close to the ones reported for
nitrogen, a loading of 816 lmol/g can be calculated (Table 1). This
value is much higher than the amount of l post-grafted onto other
ordered mesoporous silicas such as SBA-15 [13]. Hence the 1 den-
sity in the 1-Ph-PMO is 1.1
l
mol/m2, whereas 0.4–0.05 mol/m2
l
Ph
O
O
O
O
Si
Si
N
OTf
O
Cu
O
O
O
N
OTf
Ph
Ph
Ph
Ph
HO
Ph
1-0.7% mol
PhCOCl (0.5 eq)
DIPEA (1.0 eq)
0°C, CH2Cl2
HO
OOCPh
OH
4
( 3)
Scheme 3. Kinetic resolution of 1,2-diphenylethane-1,2-diol (3) by Cu@1-Ph-PMO.

66
M.A.O. Lourenço et al. / Journal of Catalysis 320 (2014) 63–69
Table 1
Chemical and textural characteristics of the PMO materials.
Sample
Concentration (
Cua
l
mol/g)
[Ligand]/[Cu]
Textural properties
Density (
Cu
l
mol/m2)
Ligand
Ligandb
ABET (m2/g)
Aext (m2/g)
1-Ph-PMO
Cu@1-Ph-PMO
graf_Cu-1@Ph-PMO
816
740
96
637
684
664
1033
1049
834
1.3
1.1
0.2
120
57
6.2
1.7
0.2
0.1
a
Determined by ICP-AES.
Determined by EA from the nitrogen content.
b
Fig. 1. ATR–FTIR spectra of the 1, 1-Ph-PMO, Cu@1-Ph-PMO and graf_Cu-1@Ph-
PMO.
the Cu-1 complex in frozen solution indicating that Cu(II) coordi-
nation to the bis(oxazoline) nitrogen atoms took place (Scheme 2)
[31].
Fig. 2. 13C CP-MAS NMR spectra of 1-Ph-PMO and Cu@1-Ph-PMO. Asterisks depict
spinning side bands. Cardinals represent ethanol impurities. S is used to identify the
surfactant residual peaks.
Solid-state 13C CP MAS NMR spectra of 1-Ph-PMO and Cu@1-Ph-
PMO materials are shown in Fig. 2. The 1-Ph-PMO spectrum shows
resonances between 9 and 20 ppm that can be assigned to the CH2
carbons of the propylsilane ligands. The peak at 35.0 ppm corre-
sponds to the quaternary carbon that binds the two oxazoline rings
to the propylsilane ligands. The two peaks at 69.5 and 75.2 ppm are
related with the sp3 carbons of the oxazoline rings, linked to,
respectively, the N and the O atoms. In the same spectrum it is pos-
sible to observe resonances at 128.2, 133.5 and 142.2 ppm
assigned to the sp2 carbons from the phenyl groups. The peak at
142.2 ppm should correspond to the sp2 carbons of the phenyl
rings directly bounded to the oxazolines. The peak at 133.5 ppm
is assigned to the sp2 carbons of the phenylene bridge from the
BTEB precursor. Finally, the peak at 170.1 ppm is attributed to
the sp2 carbons directly linked to both N and O atoms of the oxaz-
olines. Other small peaks in the spectrum between 10 and 75 ppm
may correspond to residual carbons of the alkyl chain of the surfac-
tant. The 13C CP MAS NMR spectrum of Cu@1-Ph-PMO is very sim-
ilar to the spectrum of 1-Ph-PMO but it seems slightly cleaner
probably because the copper coordination is carried out in metha-
nol at 80 °C, allowing a complete surfactant removal. The similar
spectrum results from the non-coordinated bis(oxazoline) ligands.
As Cu(II) is a paramagnetic nucleus it precluded the observation of
the carbon nuclei close to it and only the non-coordinated can be
observed.
1 is observed. The 1-Ph-PMO spectrum shows three intense peaks
at ca. ꢁ62, ꢁ71 and ꢁ81 ppm assigned to T1, T2 and T3 organosili-
ceous species of BTEB. The T environments of the 1 ligand are less
intense and seem to be slightly shifted to low field under the sig-
nals of the BTEB.
The structural order before and after copper coordination was
evaluated by powder X-ray diffraction (PXRD) and transmission
electron microscopy (TEM). PXRD pattern of 1-Ph-PMO (Fig. 3)
reveals the characteristic low angle diffraction peaks at
d = 4.50 nm (2h = 1.96°) usually associated with mesoporous mate-
rials. In addition
a medium-range reflection is observed at
d = 0.764 nm (2h = 11.58°) due to the molecular-scale periodicity
along the wall.
In this way the 1-Ph-PMO material displays both meso- and
molecular-scale periodicities. Upon copper coordination and after
use of Cu@1-Ph-PMO as catalyst, the materials seem to preserve
the structure (Fig. 3). The Cu@1-Ph-PMO before and after catalysis
present a first low angle reflection at d100 = 4.46 and 4.41 nm,
respectively (Table S1, SD). At a medium scattering angle, these
two materials display peaks at d = 0.765 (2h = 11.57°) and
0.762 nm (2h = 11.61°), respectively (Fig. 3).
The presence of mesopores in the novel 1-Ph-PMO, Cu@
1-Ph-PMO and graf_Cu-1@Ph-PMO is also supported by the nitro-
gen adsorption–desorption isotherms collected at ꢁ196 °C (see SD,
Fig. S6). Type IV N2 adsorption–desorption isotherms characteristic
of mesoporous materials were observed together with narrow pore
size distribution (PSD). The BET specific surface area of 637 m2 gꢁ1
The 29Si MAS and CP MAS NMR spectra of 1-Ph-PMO and
Cu@1-Ph-PMO are shown in Fig. S5 of SD. An overlapping of the
T environments [Tm = RSi(OSi)m(OH)3ꢁm] resulting from the con-
densation of the BTEB and trimethoxypropylsilane functionalized

M.A.O. Lourenço et al. / Journal of Catalysis 320 (2014) 63–69
67
arrangement of pores (Fig. 4b – inset), corroborating the lack (or
low intensity) of the (110) and (200) peaks characteristics of a
two-dimensional hexagonal symmetry (p6 mm) lattice in the
XRD patterns (Fig. 3). The pore size calculated from the TEM image
is approximately 3 nm, which is in good agreement with the pore
size obtained from the N2 sorption isotherms (3 nm). Fig. S9 (SD)
displays the micrograph images in which the ordered microstruc-
ture of the walls is evident. A low atomic percentage (below 1%)
of Cu was verified by EDS measurement (Fig. S10).
The thermogravimetric analyses of 1-Ph-PMO and Cu@1-Ph-
PMO (Fig. S11, SD) exhibit a first weight loss below 100 °C due to
desorption of physisorbed water. The 1-Ph-PMO material has a
thermal stability up to 200 °C. Above this value, the decomposition
and release of the 1 organic bridges from the framework takes
place. Above 500 °C, occurs the decomposition of the phenylene
organic moieties. In the case of the Cu@1-Ph-PMO material, the
decomposition of the 1 occurs at the same range of temperature,
followed by the degradation of the triflate anion. CuO should be
formed which may catalyze the combustion of the phenylene
organic moieties at a lower temperature than in the copper free
1-Ph-PMO material (starts to occur at ca. 450 °C).
Copper(II) complexes with ligand 2 act as efficient homoge-
neous catalysts in the kinetic resolution of 1,2-diols [29]. Hence,
we decided to test the Cu@1-Ph-PMO material as a heterogeneous
catalyst in the asymmetric benzoylation of hydrobenzoin
(Scheme 3). The results are compiled in Table 2, together with con-
trol experiments without the addition of catalyst and in homoge-
neous phase using ligand 1 and 2. It can be concluded that the
Cu@1-Ph-PMO material acts as a very efficient heterogeneous cat-
alyst with high enantioselectivity and yield, using only 1% mol of
Cu. Furthermore, the high enantioselectivities (91 ꢁ >99%) and
yields (46–43% in maximum 50% resolution) could be sustained
upon five consecutive reuses with 1–0.7% mol based on Cu. The
enantioselectivities are also higher than the Cu(II) complex with
the starting commercial bis(oxazoline) 1, in 1% mol and similar
to that of the methylated bis(oxazoline) 2 (Scheme 1). Thus propy-
lation of the carbon bridge yields a more enantioselective hetero-
geneous catalyst than the corresponding homogeneous catalyst.
These results are comparable to the reported heterogeneous cata-
lysts based on the Cu(II) aza-bis(oxazoline) supported onto
poly(ethyleneglycol) [33] and magnetic nanoparticles coated with
silica [34] and carbon [35]. Post-grafting the Cu(II) complex with
trimethoxypropylsilane functionalized 1 onto periodic mesopor-
ous phenylenesilica resulted in lower catalyst density at the sur-
face of the material and thus in a less efficient heterogeneous
catalyst with significantly lower selectivity, enantioselectivity
and yield. The high bis(oxazoline) density seems to be an impor-
tant factor when compared with similar catalysts in the literature
[13].
At the end of the five consecutive cycles the Cu@1-Ph-PMO
material was also characterized by ATR–FTIR (SD, Fig. S12) and
XRD (Fig. 3). Both techniques showed that there were not
significant differences between the original Cu@1-Ph-PMO
and after the catalytic cycles, confirming the chemical and struc-
tural integrity of the mesoporous material (Table S1, SD), respec-
tively, and that it acts as a reusable asymmetric heterogeneous
catalyst.
The higher metal density, catalytic activity and stability of the
Cu@1-Ph-PMO in comparison to the graf_Cu-1@Ph-PMO may
result from the co-condensation methodology of synthesis. As
it was observed recently in SBA-15 [36], the introduction of
the organic ligands during the formation of the mesoporous
structure originate homogeneous materials with local geometries
and environments that are prone to metal bind, creating stable
catalytic sites comparable to the ones present in homogeneous
catalysts.
Fig. 3. XRD patterns of 1-Ph-PMO, Cu@1-Ph-PMO before and after catalysis, and
graf_Cu-1@Ph-PMO.
(Table 1) obtained for the 1-Ph-PMO is much higher than a previ-
ous attempt to synthesize a hybrid organic–inorganic material by
co-condensation of tetraethoxysilane (TEOS), instead of BTEB, with
silane-functionalized bis(oxazoline) ligands.[32] The coordination
of [Cu(OTf)2] complex leads to an increase of the BET specific sur-
face area to 684 m2 gꢁ1, supporting the 13C CP MAS NMR evidence
that the use of methanol as solvent helps to clean the material
from surfactant residues.
Spherical aberration (Cs) corrected scanning transmission elec-
tron microscopy (STEM) observations were performed at 80 kV
due to the beam sensitivity of C-containing mesoporous materials
under the electron beam. The STEM-HAADF mode was chosen in
an attempt to distinguish the pending groups with metal atoms
using Z contrast. However, the observations did not yield enough
evidences and therefore EDS was employed systematically to
probe the presence of Cu and to corroborate a homogeneous dis-
tribution throughout the Cu@1-Ph-PMO material (see Fig. S7, SD).
Fig. 4a shows the morphology of the 1-Ph-PMO with the pore sys-
tem perpendicular to the electron beam. Cu@1-Ph-PMO is pre-
sented in Fig. 4b again along the [110] orientation. In the inset,
the microstructure of the walls can be easily observed with a
spacing of d = 0.795 nm (top left) and the pore openings (bottom
left).
Nevertheless, using EELS nitrogen can also be detected in the
1-Ph-PMO material corroborating the elemental analysis (Fig. S8,
SD). It is possible to confirm based on the TEM micrographs
(Fig. 4) that the 1-Ph-PMO and Cu@1-Ph-PMO materials are
mainly long range disordered, displaying only local hexagonal

68
M.A.O. Lourenço et al. / Journal of Catalysis 320 (2014) 63–69
Fig. 4. Cs-STEM-HAADF micrograph images of (a) 1-Ph-PMO and (b) Cu@1-Ph-PMO. The top left inset represents the wall microstructure, while the bottom left shows the top
view of the hexagonal arrangement of pores.
Table 2
Kinetic resolution of 1,2-diphenylethane-1,2-diol by the Cu@1-Ph-PMO and in 1 and 2 in homogeneous phase (Scheme 3).a
Run
%molb
Cu
%yieldc
%eed
Se
TONf
Ligand
Without catalyst
21
0
1 + [Cu(OTf)2]
2 + [Cu(OTf)2]
1.0
1.0
1.0
1.0
46
47
84
>99
27
>581
46
45
Cu@1-Ph-PMO
1st
1.0
0.9
0.9
0.9
0.7
6.0
5.6
5.4
5.4
4.5
45
44
46
46
43
94
>99
98
97
91
76
460
264
172
43
46
47
53
52
58
2nd
3rd
4th
5th
graf_Cu-1@Ph-PMO
1st
2nd
0.8
0.8
0.8
0.7
38
37
66
64
7
7
50
50
a
Reactions performed for 24 h at 0 °C using 0.48 mmol (R,R)-3, 0.48 mmol (S,S)-3, 1.00 mmol DIPEA, 1.0% mol based on Cu and 0.50 mmol of benzoyl chloride in 5.0 ml of
CH2Cl2.
b
Percentage of copper and ligand 1 in the catalyst in relation to 3 (see Table 1); for the recycling experiments corrected for the loss of heterogeneous catalyst weight.
Isolated yield of 4 (Scheme 3).
Enantiomeric excess of 4, determined by HPLC.
Selectivity (S) = ln[1 ꢁ yield (1 + ee)]/ln[1 ꢁ yield (1 ꢁ ee)].
TON = moles of isolated 4/moles of Cu.
c
d
e
f
Ciência 2008. ARS and PF acknowledge IF/01300/2012 and IF/
00327/2013, respectively. MAOL thanks the PhD grant SFRH/BD/
80883/2011. The research leading to these results has received
funding from the European Union Seventh Framework Programme
under Grant Agreement 312483 – ESTEEM2 (Integrated Infrastruc-
ture Initiative–I3).
4. Conclusions
A novel mesoporous chiral Cu(II) bis(oxazoline) phenylene-
silica with narrow distribution of pores is proved to be easily pre-
pared by a co-condensation methodology. Upon coordination of
Cu(II) triflate complex, this material acts as a highly efficient and
reusable catalyst on the kinetic resolution of 1,2-diphenylethane-
1,2-diol. Moreover, the material shows improved catalytic activity
and superior enantioselectivity comparing to the homogeneous
counterpart, while being recyclable at least up to four consecutive
cycles. The preparation method seems to be extremely important
on the definition of the final properties of the heterogeneous cata-
lysts. The co-condensation of bis-silylated precursors showed to be
efficient on avoiding the leaching of the metal complex.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.jcat.2014.09.017.
References
[1] G. Desimoni, G. Faita, K.A. Jorgensen, Chem. Rev. 111 (2011) 284.
[2] D. Rechavi, M. Lemaire, Chem. Rev. 102 (2002) 3467.
[3] A.F. Trindade, P.M.P. Gois, C.A.M. Afonso, Chem. Rev. 109 (2009) 418.
[4] J.M. Fraile, J.I. Garcia, J.A. Mayoral, Chem. Rev. 109 (2009) 360.
[5] J.M. Fraile, J.I. Garcia, C.I. Herrerias, J.A. Mayoral, E. Pires, Chem. Soc. Rev. 38
(2009) 695.
[6] D. Rechavi, M. Lemaire, Org. Lett. 3 (2001) 2493.
[7] R. Annunziata, M. Benaglia, M. Cinquini, F. Cozzi, M. Pitillo, J. Org. Chem. 66
(2001) 3160.
[8] M.I. Burguete, J.M. Fraile, J.I. Garcia, E. Garcia-Verdugo, C.I. Herrerias, S.V. Luis,
J.A. Mayoral, J. Org. Chem. 66 (2001) 8893.
[9] S.S. Lee, S. Hadinoto, J.Y. Ying, Adv. Synth. Catal. 348 (2006) 1248.
[10] S.S. Lee, J.Y. Ying, J. Mol. Catal. A-Chem. 256 (2006) 219.
Acknowledgments
Authors are grateful to the Fundação para a Ciência e a Tecno-
logia (FCT), Fundo Europeu de Desenvolvimento Regional (FEDER),
QREN-COMPETE, the European Union, and the Associate Labora-
tory CICECO for funding the Projects PTDC/QUI-QUI/64770/2006,
FCOMP-01-0124-FEDER-015644
FCOMP-01-0124-FEDER-037271 (PEst-C/CTM/LA0011/2013). ARS
thanks FCT, FSE and POPH for the contract under the program
(PTDC/QUI-QUI/113678/2009),

M.A.O. Lourenço et al. / Journal of Catalysis 320 (2014) 63–69
69
[11] A.R. Silva, H. Albuquerque, S. Borges, R. Siegel, L. Mafra, A.P. Carvalho, J. Pires,
[25] R.J.P. Corriu, J.J.E. Moreau, P. Thepot, M. W. Chi Man 4 (1992) 1217.
[26] S. Inagaki, S. Guan, T. Ohsuna, O. Terasaki, Nature 416 (2002) 304.
Micropor. Mesopor. Mater. 158 (2012) 26.
[12] A.R. Silva, V. Guimarães, A.P. Carvalho, J. Pires, Catal. Sci. Technol. 3 (2013) 659.
[13] A.R. Silva, L. Carneiro, A.P. Carvalho, J. Pires, Catal. Sci. Technol. 3 (2013) 2415.
[14] J. Lim, S.N. Riduan, S.S. Lee, J.Y. Ying, Adv. Synth. Catal. 350 (2008) 1295.
[15] P. Borah, X. Ma, K.T. Nguyen, Y.L. Zhao, Ang. Chem.-Int. Ed. 51 (2012) 7756.
[16] K.M.T. Seki, C.M. Crudden, Chem. Commun. (2012) 6369.
[17] M. Waki, N. Mizoshita, T. Tani, S. Inagaki, Ang. Chem.-Int. Ed. 50 (2011) 11667.
[18] X. Liu, P.Y. Wang, Y. Yang, P. Wang, Q.H. Yang, Chem. Asian J. 5 (2010) 1232.
[19] C. Baleizao, B. Gigante, D. Das, M. Alvaro, H. Garcia, A. Corma, J. Catal. 223
(2004) 106.
[27] M. Kruk, V. Antochshuk, M. Jaroniec, A. Sayari, J. Phys. Chem. B 103 (1999)
10670.
[28] S.J. Gregg, K.S.W. Sing, Adsorption, Surface Area and Porosity, second ed.,
Academic Press, London, 1982.
[29] Y. Matsumura, T. Maki, S. Murakami, O. Onomura, J. Am. Chem. Soc. 125 (2003)
2052.
[30] Q. Yang, M.P. Kapoor, S. Inagaki, J. Am. Chem. Soc. 124 (2002) 9694.
[31] M.E. Owen, E. Carter, G.J. Hutchings, B.D. Ward, D.M. Murphy, Dalton Trans. 41
(2012) 11085.
[20] B. Gigante, C. Baleizão, D. Das, M. Alvaro, H. Garcia, A. Corma, Chem. Commun.
(2003) 1860.
[21] N. Mizoshita, T. Tani, S. Inagaki, Chem. Soc. Rev. 40 (2011) 789.
[22] P. Van Der Voort, D. Esquivel, E. De Canck, F. Goethals, I. Van Driessche, F.J.
Romero-Salguero, Chem. Soc. Rev. 42 (2013) 3913.
[32] J.M. Fraile, J.I. Garcia, C.I. Herrerias, J.A. Mayoral, Chem. Commun. (2005) 4669.
[33] A. Gissibl, M.G. Finn, O. Reiser, Org. Lett. 7 (2005) 2325.
[34] A. Schatz, M. Hager, O. Reiser, Adv. Funct. Mater. 19 (2009) 2109.
[35] A. Schatz, R.N. Grass, Q. Kainz, W.J. Stark, O. Reiser, Chem. Mater. 22 (2010)
305.
[23] A.R. Silva, Curr. Org. Chem. 18 (2014) 1225.
[36] M.P. Conley, C. Coper´et, C. Thieuleux, ACS Catal. 4 (2014) 1458.
[24] R.J. Clarke, I.J. Shannon, Chem. Commun. (2001) 1936.