“This document is the Accepted Manuscript version of a Published Work that appeared in final form in ACS Catalysis, copyright © American Chemical Society after peer review and technical editing by the publisher. To access the final edited and
published work see http://pubs.acs.org/doi/abs/10.1021/cs501573c .”

A Double-Supported Silica – Metal-Organic Framework Palladium Nanocatalyst for Aerobic Oxidation of Alcohols under
Batch and Continuous Flow Regimes
Vlad Pascanu,a,c§ Antonio Bermejo Gómez,a,c§ Carles Ayats,d Ana Eva Platero-Prats,b,c Fabian
Carson,b,c Jie Su,b,c Qingxia Yao,b,c Miquel À. Pericàs,d,e Xiaodong Zou,b,c* Belén MartínMatutea,c*
†

a

Department of Organic Chemistry, Stockholm University, Stockholm, SE-10691, Sweden.

b

Department of Materials and Environmental Chemistry, Stockholm University, Stockholm, SE-10691, Sweden.

c

Berzelii Center EXSELENT, Stockholm University, Stockholm, SE-10691, Sweden.

d

Institute of Chemical Research of Catalonia, E-43007, Tarragona, Spain.

e

Department of Organic Chemistry, University of Barcelona, E-08028 Barcelona, Spain.

ABSTRACT: The stable and easily synthesized metal-organic framework MIL-88B-NH represents an attractive support for catalysts employed in oxidation reactions, which are typically performed under relatively harsh conditions.
However, MIL-88B-NH , the thermodynamic polymorph of the more popular MIL-101-NH , has been rarely employed in
catalytic applications due to a difficult impregnation process caused by the flexible nature of the framework. We report
herein a new catalyst denoted Pd@MIL-88B-NH (8 wt% Pd), the first example of metallic nanoparticles successfully
impregnated in the pores of MIL-88B-NH . Further, by enclosing the MOF crystals in a tailored protective coating of
SiO nanoparticles, an even more enduring material was developed and applied to the aerobic oxidation of benzylic
alcohols. This double supported catalyst Pd@MIL-88B-NH @nanoSiO , displayed high activity and an excellent performance in terms of endurance and leaching control. Under batch conditions, a very convenient and efficient recycling
protocol is illustrated, using a “teabag” approach. Under continuous flow, the catalyst was capable to withstand 7
days of continuous operation at 110 C, without deactivation. During this time, no leaching of metallic species was
observed and the material maintained its structural integrity.
2

2

2

2

2

2

2

2

°

KEYWORDS: Palladium Nanoparticles, MIL-88B-NH2, Aerobic Oxidation, Teabag catalysis, Flow Chemistry.
Introduction
The aerobic oxidation of alcohols is an important industrial process. The sustained interest in the field is partially driven by
the prospect of applying aerobic oxidation to biomass conversion, which is an abundant source of alcohols.1 Recent advances in the area aim to use mild conditions and replace hazardous reagents with air as the oxidant.2 Supported heterogeneous oxidation catalysts3 have also received widespread attention because they can be recovered and recycled,4 and pal6
ladium is often the transition metal chosen to perform the oxidation reaction.5 For example, Pd@ARP (amphiphilic resin) was
used for the aerobic oxidation of secondary alcohols in water and Pd@SBA-15 (SBA = functionalized Santa Barbara Amorphous silica material) was used to oxidize benzylic alcohols at temperatures as low as 80 °C. Hutchings and co-workers
reported that Au-Pd bimetallic core-shell nanoparticles supported on TiO2 could selectively oxidize benzylic and allylic alco–1
hols with exceptional turnover frequencies (269000 h ) under neat conditions at 100–160 °C.8
7

Metal–organic frameworks (MOFs) are porous crystalline materials assembled from metallic ions or clusters and organic
multi-topical linkers.9 MOFs have gained popularity as supports in heterogeneous catalysis10 11 because of their high surface
areas and large pore volumes, which can accommodate the active catalytic sites. Another advantage is the possibility to
post-synthetically modify MOFs in order to tune the properties of the framework.12 Metallic nanoparticles can be stabilized in
the pores of MOFs both through mechanical and charge transfer interactions.13 Through these mechanisms, highly active,
small nanoparticles with a narrow size distribution can be generated and their leaching can be avoided, facilitating the recycling of the catalyst. Based on these properties, combining the benefits of MOFs and flow chemistry can be regarded as a
natural direction for progress in the field and we recently reported the first example of metallic nanoparticles supported on
MOFs for flow processes.
,

11d

Among the first MOF-supported catalysts employed for the oxidation of alcohols, Li and co-workers developed a Pd catalyst supported on MIL-101-NH2.14 This system displayed a good efficiency if molecular oxygen was bubbled through the
reaction mixture at a sufficiently high rate. The authors also suggest an important contribution of the acid sites in the framework to the outcome of the reaction. More recently, a modified version of UiO-66 was also shown to tolerate a very wide
scope of substrates and to be recyclable in the presence of sacrificial oxidants (e.g. TBHP, H2O2).15 Around the same time, it
was shown that TiO nanoparticles incorporated in the mesoporous HKUST-1 were capable of performing a photocatalytic
oxidation of alcohols in moderate yields under sunlight irradiation.16 However, it is well known that many MOFs suffer from
poor mechanical strength compared to denser supports, such as ceramics.17 Based on this limitation we envisioned that
coating the MOF with a protective layer could improve its stability and reusability for catalytic processes. An interesting solution consists of surrounding the catalyst with SiO nanoparticles that contain mesopores between the particles, which were
recently shown to improve the diffusion into the MOF.18 In this way, the mechanical properties of the MOF are improved
while also allowing reactants to reach the active sites.
2

2

Herein we report the development of a new palladium nanocatalyst protected by an innovative double-layered,
MOF@Silica support. For the first time, the immobilization of metallic nanoparticles could be achieved in the flexible functionalized MIL-88B-NH2 framework. This long-elusive goal was accomplished after investigating the manner in which the
material responds to the polarity of its surrounding environment, by alternating its open/close conformations. MIL-88B-NH2
is the denser, thermodynamic polymorph of the notorious MIL-101-NH2 and thus the proposed catalyst is expected to withstand much harsher reaction conditions than previously reported Pd@MIL-101-type catalysts.14,19 A good dispersion of the
Pd nanoparticles is achieved in MOFs containing nitrogen functional groups in the organic linkers. These coordinating
groups also contribute to reduce the metal leaching during catalysis.20 Importantly, the functionalized MIL-88B-NH2 MOF
could be more conveniently synthesized directly from 2-aminoterephthalic acid, thereby avoiding wasteful post-synthetic
modification procedures required for its MIL-101 analog.21 The MOF was subsequently enclosed by SiO2 nanoparticles to
produce a more resistant material (Scheme 1) labelled Pd@MIL-88B-NH2@nano-SiO2. This material was also more suitable
for stabilizing small nanoparticles and preventing metallic leaching than simple Pd@SiO2 catalysts. c The catalytic activity of
this novel material was evaluated in the oxidation of benzylic alcohols to their corresponding carbonyl compounds using only
air as the oxidant, at ambient pressure without using any bubbling device. The benefits of this new support are also demonstrated by the outstanding properties of the catalyst under a continuous flow regime. When packed in a microflow reactor,
the material can endure more than 7 days of continuous operation without diminishing its efficiency. Metal leaching is minimized throughout the whole experiment and high total turnover numbers are achieved. Therefore this is a more durable catalyst then previously reported, synthesized in a more economical manner, fulfilling the main directives of green chemistry
through catalysis: efficiency, sustainability, and recyclability.22-25
5

Scheme 1. Graphical illustration of the synthesis of Pd@MIL-88B-NH @nano-SiO
2

2

Results and Discussion
Synthesis of Pd@MIL-88B-NH2@nano-SiO2
MIL-88B-NH2 was synthesized under solvothermal conditions and activated according to an adapted procedure from Lin
and co-workers.26 The crystallinity of the MOF was confirmed by X-ray powder diffraction (XRPD) and is shown in Figure 1.
Impregnation of this MOF with palladium can be challenging due to the flexible nature of the framework. A screening experiment using solvents with different polarities confirmed that the impregnation is more efficient with polar solvents and the
best results were obtained in MeOH (see ESI, Table S1).27 The structure of MIL-88B expands in MeOH,28 which facilitates
introduction of Pd(II) complexes into the MOF pores. Higher loadings of palladium were achieved with Na2PdCl4 when compared to PdCl2(CH3CN)2, and after 72 h loadings between 7.15 and 8.60 wt% were achieved, as determined by ICP-OES
analysis. This corresponds to a Pd:Cr ratio of 1:4 and ~26% functionalization of the aminoterephthalate linkers.29 NaBH4 was
then used to reduce Pd(II) and form palladium nanoparticles in the MOF [Pd(0)@MIL-88B-NH2]. The XRPD pattern matched
with that of MIL-88B-NH2, confirming the integrity of the framework upon functionalization (Figure 1).30

2

Figure 1. a) XRPD patterns of MIL-88B-NH2, Pd(II)@MIL-88B-NH2 and Pd(0)@MIL-88B-NH2 and b) Schematic polyhedral
representation of Pd(II) impregnation into MIL-88B-NH2. The purple ball represents Pd. The blue and light green balls represent nitrogen and chloride, respectively. The Pd(II) species lead to MIL-88B-NH2 adopting a semi-open form, highlighted in
the XRPD patterns by asterisks (*).
Pd K-edge extended X-ray absorption fine structure (EXAFS) spectroscopy was performed on Pd(II)@MIL-88B-NH2(Cr) to
determine the coordination environment of palladium (Figure 2a). The data were best fit using a cis square-planar geometry
at the Pd(II) center, with the coordination environment comprising chloride anions and two amino groups (Figure 2b). The cis
conformation appears to be more likely than trans (see ESI, Section S2), which agrees well with palladium(II) complexes
located in the pores of the MOFs closed form (see Figure 2). The data fitting indicated Pd bonds to two N atoms at 2.06(1) Å
and two Cl atoms at 2.307(5) Å. These distances are in agreement with crystallographic data of reported [Pd(II)(ArNH )2Cl2]
complexes.31 The nearest-neighbour carbon atom from the phenyl ring was determined at a distance of 3.38(5) Å, which is
consistent with a semi-open form of this MOF (crystallographic value for the closed form is 3.052 Å).32
2

To further study the functionalization of MIL-88B-NH2 with palladium complexes, we performed N2 sorption experiments
and LeBail refinements of the XRPD patterns (see ESI, Figure S3). Pd(II)@MIL-88B-NH2 contains two forms of MIL-88B; a
3
3
closed form and a semi-open form in a ratio of ~70:30 (V = 1976(1) Å and 2099(1) Å , respectively, see Table S4). We attribute the presence of this semi-open form to Pd(II) attached to the amino groups in the MOF, which agrees with the results
from EXAFS and ICP-OES (Figure 1). The underlying net of MIL-88B is acs-a,33 which consists of packing six-membered
rings in a chair conformation. MIL-88B is flexible and palladium incorporation into the MOF changes the degree of pore
2
opening. In addition, the BET surface area increases from 7 to 30 m /g after palladium impregnation, which implies that the
32	
MOF is partially locked in a semi-open form (see ESI, Section S4).

3

3

Figure 2. a) Pd K-edge XAFS spectra for Pd(II)@MIL-88B-NH2 and k-weighted XAFS, χ(k) ×k (detail). b) Experimental (solid
3
–1
line) and fitted (dashed line) FT[c(k)×k ] functions (bottom) with Δk ≈ 2–14 Å . Palladium coordination environment used to fit
the EXAFS data (hydrogen atoms were omitted for clarity). The purple ball represents palladium. The blue and light green
balls represent nitrogen and chloride, respectively.
Reduction of Pd(II), following a similar procedure than the described for Pd(0)@MIL-101-NH , afforded Pd(0)@MIL-88BNH . A second layer of protection was achieved by coating the Pd(0)@MIL-88B-NH2 crystals with nano-sized SiO2 particles
(see ESI, Section S1 for detailed synthetic procedure). The material was washed with HCl (aq, pH = 2) to remove residual
TEOS, which can affect its catalytic performance. Traces of acid were neutralized with an aqueous solution of Et N. The integrity of the MOF framework was confirmed by XRPD, which contained a large background hump attributed to the silica
(Figure 3a). This new material, denoted Pd@MIL-88B-NH2@nano-SiO2, contained 0.51 wt% Pd, as measured by ICP-OES
analysis. It is noteworthy that a similar coating strategy applied to Pd@MIL-101-NH2 led to the decomposition of the MOF
and the formation of a completely amorphous material.
19a

2

2

3

The Pd nanoparticles were evenly distributed in the MOF with an average particle size of 2–3 nm, as shown by transmission electron microscopy (TEM, see ESI, Section S5). Furthermore, the presence of nano-sized SiO particles surrounding the
Pd@MOF crystal surface is evident in Figure 3b. Some larger Pd agglomerates were visible on the MOF surface subsequent
to silica coating (Figure 3b).
2

N2 sorption experiments on Pd(0)@MIL-88B-NH showed very low gas uptake, indicating the closed form of the MOF is
predominant. In contrast, the SiO -supported material displayed a Type IV isotherm with significant N uptake (Figure 3c).34
This hysteresis is due to mesopores between the silica nanoparticles. The BET surface area and pore volume for Pd@MIL2
3
88B-NH2@nano-SiO2 were calculated as 603±4 m /g and 1.82 cm /g, respectively (see ESI, Section S4).
2

2

2

The TGA analyses of all materials showed a considerably improved thermal resistance (ca. 60 °C) of the double-supported
silica-MOF compared to that of the original Pd@MOF catalyst (see ESI, Section S6).

4

Figure 3. a) XRPD patterns of Pd(0)@MIL-88B-NH2 and Pd(0)@MIL-88B-NH2@nano-SiO2, b) TEM micrograph of Pd(0)@MIL88B-NH2@nano-SiO2 and c) N2 sorption isotherms of Pd(0)@MIL-88B-NH2 and Pd(0)@MIL-88B-NH2@nano-SiO2.

Catalytic properties
The properties of the new material were then investigated in terms of robustness and catalytic activity in the oxidation of
secondary benzylic alcohols. Pd@MIL-88B-NH2@nano-SiO2 was demonstrated to be highly active and selective, using air as
the oxidant in p-xylene at 150 °C, tolerating a wide variety of substitution patterns (Table 1). With typical substrates as 1phenylethanol (1a), 1-(4´-methylphenyl)ethanol (1b), 1-(4´-methoxyphenyl)ethanol (1c), and 1-phenylpropanol (1d), Pd@MIL88B-NH2@nano-SiO2 (2 mol% Pd) returned the corresponding ketones (2a-2d) in excellent yields within reasonably short
reaction times (Table 1, entries 1-4). Substrates containing cyclic motifs (Table 1, entry 6, 2f) and bulky substituents on either
side of the hydroxyl group (Table 1, entries 5 and 7-9, 2e and 2g-2i) were easily converted to the desired carbonyl compounds in reaction times ranging from 15 to 20 h, and were isolated in excellent yields. More functionalized substrates, such
as diol 1j (Table 1, entry 10) and benzoin 1k (Table 1, entry 11), were also transformed into the corresponding ketones (2j-2k)
with no difficulties and in very good isolated yields. Although these last two substrates with electron-withdrawing groups
afforded good results (1j-1k), other benzylic alcohols substituted with nitrogen-containing electron withdrawing groups at
the para position (2l and 2m) could not be oxidized (Table 1, entries 12 and 13). On the other hand, alcohol 1n, with a5

pyridyl substituent, afforded the product in good yield when 4 mol% of Pd was used (Table 1, entry 14). Importantly, when 1phenylethanol (1a) was reacted under neat conditions in a 10 mmol scale, the catalyst loading could be lowered to 0.04
mol% Pd and a full conversion could still be achieved within 12 h with complete selectivity towards the oxidation product
(Table 1, entry 1). A blank silica material synthesized in a similar manner without the addition of Pd@MIL-88B-NH2 was employed under the standard reaction conditions, but no conversion was observed. When the reaction of 1a was carried out
without catalyst under an atmosphere of air at 150 °C for 20 h, 2a was obtained in less than 5% yield.

Recycling experiments using a catalytic teabag
The beneficial effect of the innovative silica protective layer over the recyclability of the new material was demonstrated in a
“teabag” experiment (Figure 4, Movie S1) showing that Pd@MIL-88B-NH2@nano-SiO2 is remarkably easy to handle and recycle. In contrast, separation of the particles of Pd@MIL-88B-NH2 (i.e. without silica) from the reaction mixtures is tedious,
which is due to the small size of the silica-free particles, and can only be achieved through centrifugation. The catalyst (200
mg of 0.51 Pd wt%, 1 mol%) in a teabag firmly tied with a Teflon wire was inserted into a solution of 1-phenylethanol in pxylene. After 16 h at 150 °C, the catalyst was pulled out using the attached Teflon wire and subsequently immersed in a fresh
reaction batch. This facile recycling process could be repeated for 4 times obtaining very good yields (Table 2) until a significant drop in activity was observed, which was accompanied by an intense yellow coloring of the solution.
Although the teabag approach led to a convenient handling of the material, its rather rapid deactivation under these conditions was not satisfactory. One of the main limitations of the reaction in the conditions described above is the low solubility
of air in BTX solvents (benzene, toluene, xylene). This limitation was aggravated by the fact that the stirring was significantly
impeded and no additional air bubbling device was used.

Table 1. Substrate scope
OH

O
R2

Pd@MIL-88B-NH2@SiO2
(0.51 wt%)

R2

p-xylene
air, 150 °C

1

R1

R1

Cat. loading
[mol%]

98

12

97

15

90

12

63

2

15

97

2

15

95

2

12

94

2

20

97

2

15

87

2

1

10

2

O

Isolated
Yield [%]

2

Product

Time
[h]

2

Entry

15

97

0.04

2a

2

a

O

2
2b

O

3
2c

O

O

4
2d
O

5
2e

O

6
2f

O

7
2g

Ph

O
8

2h
O

9
2i

6

O
10

2

15

68

5

10

85

2

68

n.d. (<10)

b

4

44

n.d. (<10)

b

2

68

n.d. (<10)

b

4

44

n.d. (<10)b

2

44

n.d. (42)

4

24

71

2j

O
O

11
O

2k
O

12
O2N

2l

O

13
NC

2m

O

14
N
a

b

2n

Reaction run in neat conditions. No by-product formation. Yield estimated by H NMR spectroscopy.
b

1

Figure 4. a) Ingredients for preparation of catalytic teabag, b) catalytic teabag before use, c) reaction set-up using a Radley
carousel for reactions open to air; d) catalytic teabag after 5 runs; e) change in color upon deactivation of the catalyst.

Table 2. Conversion and metallic leaching over 5 runs during the catalyst recycling experiment in the oxidation
of 1a.
Run

Conversion

Pd leaching

Cr leaching

1

85%

0.2 ppm

0.3 ppm

2

86%

0.1 ppm

0.3 ppm

3

82%

0.8 ppm

0.3 ppm

4

71%

11.1 ppm

0.8 ppm

5

38%

0.8 ppm

0.4 ppm

7

Aerobic oxidation under continuous flow
We predicted that a better mixing of the liquid and gaseous phases would drastically improve the reaction rate and would
allow us to decrease the reaction temperature and prolong the lifetime of the catalyst. This hypothesis was implemented and
evaluated in a continuous-flow set-up, which led to an exceptional improvement over the reaction outcome.
1-Phenylethanol (1a) was chosen as a model substrate for oxidation under a continuous-flow regime and synthetic air
(20.9% ± 1% pure O2) was used as the oxidizing agent. The instrumental set-up was adapted for a 3-phase reaction at hightemperature, as schematically represented in Figure 5 (see Experimental Section for detailed procedure). One of the essential
components in the system is a microchip mixing device (1 mL volume, 1.6 m length microchannel) where the liquid phase
(0.1 M, 1-phenylethanol in toluene) and gaseous phase (synthetic air) were pumped simultaneously (Movie S2). By precisely
regulating the flow rates of the two components, a very effective mixing is achieved before the reaction mass reaches the
catalyst. This was decisive in order to achieve a satisfactory conversion in one pass through the reactor at only 110 °C,
compared to 150 °C under batch conditions (vide supra).
The packed-bed reactor consisted of a fritted, glass chromatography column loaded with 0.51 wt% Pd(0)@MIL-88BNH2@nano-SiO2 (800 mg, 0.0385 mmol of Pd, ca. 40 mm bed height), and kept at a constant temperature of 110 °C. In order
to avoid toluene evaporation, the reactor outlet was connected to a supplementary pressure controller that provided a backpressure of 0.9 bar. From the pressure controller, the reaction mixture was directed further to a collection flask. The residence time was determined to be ca. 10 min. Samples were collected regularly and analyzed by achiral gas chromatography
for determining the conversion and by ICP-OES for measuring the level of leached metallic species.
Initially, toluene was passed through the column for several minutes, until the pressure was stabilized, followed by methanol (see SI for details). It was found that this treatment inhibits the formation of acetals as by-products, presumably by methanol coordination to the free Lewis-acidic sites of the chromium clusters.35 Subsequently, the column was washed again with
toluene and the reaction mixture was pumped in the microchip where it was intensively mixed with the airflow, and the experiment was allowed to run continuously at 150 μL/min (50 μL/min and 100 μL/min for liquid and gaseous phases, respectively) for 1 week during which the catalyst displayed a surprisingly stable behavior. The conversion stabilized after the first 6
h and remained almost constant around 80% (≈75% acetophenone + ≈5% ethylbenzene) until the experiment was stopped
after 7 days (168 h). The evolution of the yields of acetophenone and ethylbenzene in time is plotted in Figure 6 (see ESI,
Table S5 for all data points).

Pressurized input
1 bar

OH

1 bar
air

Pressure
controller
0.9 bar

Pd@MOF@SIO2

2 bar
air

50 µL·min-1

O

Toluene, 0.1 M

8 bar
air

microchip

Mass flow
controller
100 µL·min-1

Figure 5. Instrumental set-up for aerobic oxidation under continuous flow.
The product was collected between 6 and 168 h and isolated by distillation from toluene using a Vigreux column, followed
–
by column chromatography to yield 31.02 mmol (3.722 g) of isolated acetophenone, corresponding to a TOF of 4.97 h . Most
importantly, throughout the whole experiment (samples measured at 6, 24, 96, and 168 h) the level of leached palladium
remained below the detection limit of the ICP-OES technique (<0.1 ppm), while chromium levels in solution were constant at
0.2 ppm.
1

The recovered catalyst was washed with toluene, twice with ethanol and then characterized. The XRPD pattern showed
that the recycled Pd@MIL-88B-NH2@nano-SiO2 material remained crystalline (see ESI, Figure S4). The ICP-OES analysis
revealed a limited degradation of the framework. The C, H, N contents were slightly lower in the recycled material while the
content of metallic species increased from 0.51 to 0.62 wt% for palladium and from 0.96 to 1.22 wt% for chromium (see ESI,
Table S6). This result suggests that the silica matrix is very efficient at containing both metallic species and preventing the
contamination of the final product. This special property of the silica is also confirmed by the TEM images of the recycled
material, which show a tendency of Pd to agglomerate at the interface between the MOF and the SiO2 particles (see ESI,
Figure S9).

8

Figure 6. Conversion and metallic leaching in the aerobic oxidation of 1-phenylethanol catalyzed by 0.51 wt% Pd(0)@MIL88B-NH2@SiO2 over 7 days.
The activation period observed during the experiments under continuous flow regime may be explained by an incomplete
reduction of the Pd(II) complexes upon treatment with NaBH4. This was further supported by running a control experiment
performed under anaerobic conditions. A conversion of 7% of 1-phenylethanol to acetophenone was measured after running
the reaction for 16 h at 150 °C under an argon inert atmosphere, where only the Pd(II) species can act as an oxidant. The
limitations of NaBH4 as a reducing agent for the synthesis of Pd nanoparticles supported on MOFs were recently highlighted
in the literature36 and a detailed investigation of this scenario is currently underway in our laboratories.

Conclusions
In summary, we have developed a double-supported nano-palladium catalyst and report the first successful immobilization of catalytically active metallic nanoparticles in the pores of the flexible and robust MIL-88B-NH2 framework. The resulting
Pd@MOF material is coated with a customized protective layer of mesoporous silica through an innovative approach that
drastically increases the durability and recyclability of the catalyst.
Pd(0)@MIL-88B-NH2@nano-SiO2 can catalyze the aerobic oxidation of benzylic alcohols with very good efficiency while
completely avoiding the formation of by-products, by simply performing the reaction in tubes open to air, avoiding the hazardous yet common bubbling of oxygen in a reaction mixture containing Pd and refluxing organic solvents. Importantly, no
additional oxidants or additives were required. Pd(0)@MIL-88B-NH2@nano-SiO2 was also used for preparing a “catalytic
teabag” which allowed us to recycle and reuse the catalyst in an extremely simple procedure which would be very easily
scaled-up. Even though the catalyst was deactivated fairly quickly when re-used for 5 batches at 150 °C, the “catalytic teabag” method is extremely efficient at recovering all the catalyst from the reaction mixture with minimal leaching of metallic
species, and is a promising solution for large-scale processes.
By adapting the reaction to a continuous-flow process in which a much better mixing of air and starting materials was
achieved, we were able to decrease the temperature to 110 °C while maintaining the same performance of the catalyst. In
these conditions, Pd(0)@MIL-88B-NH2@nano-SiO2 could be used for at least 7 days, showing no signs of deactivation in
–1
continuous operation at turnover frequencies of 5 h .
The extraordinary stability of Pd(0)@MIL-88B-NH2@nano-SiO2 in these conditions represents a conclusive proof that the
deposition of the MOF in the silica matrix has a tremendous beneficial effect over the framework. The protective properties of
the SiO2 coating together with the adaptation of the catalytic system to operation under continuous flow, lead to a radical
improvement in the methodology for aerobic oxidation of benzylic alcohols. This reaction was redesigned as a safer and
cleaner process, which could be more attractive for scale-up.

ASSOCIATED CONTENT
Supporting Information
Further experimental details, spectroscopic measurements and analysis. This material is available free of charge via the Internet at http://pubs.acs.org

AUTHOR INFORMATION
Corresponding Author
E-mail: belen@organ.su.se
E-mail: xzou@mmk.su.se
† Inquiries related to the work on continuous flow should be addressed to mapericas@iciq.es

Author Contributions
§ These authors contributed equally.

Notes
The authors declare no competing financial interest.

9

ACKNOWLEDGMENT
This project was supported by the Swedish Governmental Agency for Innovation Systems (VINNOVA) and AstraZeneca through the Berzelii Center EXSELENT, by the Swedish Research Council (VR) through project grants to B.M.-M and
X.Z. and for the MATsynCELL project through the Röntgen Ångström Cluster, and by the Knut and Alice Wallenberg Foundation (Catalysis in selective organic synthesis). The work on flow chemistry was funded by MINECO (grant CTQ2012-38594C02-01), DEC (grant 2009SGR623) and the ICIQ Foundation. ICIQ authors also thank EU-ITN network Mag(net)icFun (PITNGA-2012-290248) for financial support.
B.M.-M. was supported by VINNOVA through a VINNMER grant. V.P. is grateful to the EU-ITN network
Mag(net)icFun (PITN-GA-2012-290248) for a secondment at ICIQ. C.A. thanks MICINN for a Juan de la Cierva postdoctoral
fellowship.

REFERENCES

Davis, S. E.; Ide, M. S.; Davis, R. J. Green Chem. 2013, 15, 17-45.
For a review on the scope and limitations of oxidation reactions using O2 as oxidant, see: Shi, Z.; Zhang, C.; Tanga, C.; Jiao, N.
Chem. Soc. Rev. 2012, 41, 3381–3430, (184 references).
3
For a comprehensive and comparative review on aerobic oxidation of alcohols catalyzed by various transition metals, see: Parmeggiani, C.; Cardona, F. Green Chem. 2012, 14, 547–564.
4
Arends, I. W. C. E.; Sheldon, R. A. Appl. Catal. A, 2001, 212, 175–187.
5
(a) Shokouhim, M.; Shin, K.-Y.; Lee, J. S.; Hackett, M. J.; Jun, S. W.; Oh, M. H.; Jang, J.; Hyeon, T. J. Mater. Chem. 2014, 2,
7593–7599. (b) Karami, K.; Ghasemi, M.; Haghighat Naeini, N. Catal. Commun. 2013, 38, 10–15. (c) Johnston, E. V.; Verho, O.;
Kärkäs, M. D.; Shakeri, M.; Tai, C.-W.; Palmgren, P.; Eriksson, K.; Oscarsson, S.; Bäckvall, J.-E. Chem. Eur. J., 2012, 18, 12202–
12206. (d) Wu, H.; Zhang, Q.; Wang, Y. Adv. Synth. Catal. 2005, 10, 1356–1360.
6
Uozumi, Y.; Nakao, R. Angew. Chem. Int. Ed. 2003, 42, 194–197.
7
Karimi, B.; Abedi, S.; Clark, J. H.; Budarin, V. Angew. Chem. Int. Ed. 2006, 45, 4776-4779.
8
Enache, D. I.; Edwards, J. K.; Landon, P.; Solsona-Espriu, B.; Carley, A. F.; Herzing, A. A.; Watanabe, M.; Kiely, C. J.; Knight, D.
W.; Hutchings, G. J. Science, 2006, 311, 362–365.
9
(a) Metal-Organic Frameworks: Design and Application Eds: Eddaoudi, M.; Eubank, J. F. MacGillivray, L. R. John Wiley and
Sons, 2010; 37. (b) For a dedicated issue comprising 18 review articles covering MOF structures, applications, synthesis and properties see: Chem. Rev. 2012, 112, 673-674.
10
(a) For a concise description of the fundamental principles of catalysis with MOFs see: Ranocchiari, M.; van Bokhoven, J. A.
Phys. Chem. Chem. Phys. 2011, 13, 6388–6396. (b) For a comparative review on MOF and zeolite catalysts see: Dhakshinamoorthy,
A., Alvaro, M.; Corma, A.; Garcia, H. Dalton Trans. 2011, 40, 6344–6360. (c) For a comprehensive manual on all aspects of MOFs
synthesis, characterization and applications in catalysis see: Metal Organic Frameworks as Heterogeneous Catalysts Eds: Llabrés i
Xamena, F. X., Gascon, J. RSC Publishing, 2013; 237-424.
11
(a) Gustafsson, M.; Bartoszewicz, A.; Martín-Matute, B.; Sun, J.; Grins, J.; Zhao, T.; Li, Z.; Zhu, G.; Zou, X. Chem. Mat. 2010, 22,
3316–3322. (b) Carson, F.; Agrawal, S.; Gustafsson, M.; Bartoszewicz, A.; Moraga, F.; Zou, X.; Martín-Matute, B. Chem. Eur. J.
2012, 18, 15337–15334. (c) Platero-Prats, A. E.; Bermejo Gómez, A.; Samain, L.; Zou, X.; Martín-Matute, B. Chem. Eur. J. 2014, DOI:
10.1002/chem.201403909. (d) Pascanu, V.; Hansen, P.; Bermejo Gómez, A.; Ayats, C.; Platero-Prats, A. E.; Johansson, M. J.;
Pericàs, M. À.; Martín-Matute, B. ChemSusChem, 2014, 7, DOI: 10.1002/ cssc.201402858. For a related article, see: (e) Tao Yang,
T.; Bartoszewicz, A.; Ju, J.; Sun, J.; Liu, Z.; Zou, X.; Wang, Y.; Li, G.; Liao, F.; Martín-Matute, B.; Lin, J. Angew. Chem. Int. Ed. 2011,
50, 12555–12558.
12
(a) (Review) Tanabe, K. K.; Cohen, S. M. Chem. Soc. Rev. 2011, 40, 498–519. (b) Song, Y.-F.; Cronin, L. Angew. Chem. Int. Ed.
2008, 47, 4635–4637. (c) Savonnet, M.; Cockrick, E.; Camarata, A.; Bazer-Bachi, D.; Bats, N.; Lecocq, V.; Pinel, C.; Farrusseng, D.
New J. Chem. 2011, 35, 1892–1897.
13
(a) (Review) Dhakshinamoorthy, A.; Garcia, H. Chem. Soc. Rev. 2012, 41, 5262–5284. (b) Zhu, Q.-L.; Li, J.; Xu, Q. J. Am. Chem.
Soc. 2013, 135, 10210–10213. (c) Zahmakiran, M.; Dalton Trans. 2012, 41, 12690–12696. (d) Jiang, H.-L.; Akita, T.; Ishida, T.;
Haruta, M.; Xu, Q. J. Am. Chem. Soc. 2011, 133, 1304–1306.
14
Chen, G.; Wu, S.; Liu, H.; Jiang, H.; Li, Y. Green Chem. 2013, 15, 230–235.
15
Fei, H.; Shin, J.; Meng, Y. S.; Adelhardt, M.; Sutter, J.; Meyer, K.; Cohen, S. M. J. Am. Chem. Soc., 2014, 136, 4965–4973.
16
Abedi, S.; Morsali, A. ACS Catal. 2014, 4, 1298-1403.
17
Tan, J. C.; Cheetham, A. K. Chem. Soc. Rev. 2011, 40, 1059–1080.
18
Li, Z.; Zheng, H. C. J. Am. Chem. Soc. 2014, 136, 5631–5639.
19
(a) Pascanu, V.; Yao, Q.; Bermejo Gómez, A.; Gustafsson, M.; Yun, Y.; Wan, W.; Samain, L.; Zou X.; Martín-Matute, B. Chem.
Eur. J. 2013, 19, 17483–17493. (b) Pan, Y.; Yuan, B.; Li, Y.; He, D. Chem. Commun. 2010, 46, 2280–2282. (c) Yuan, B.; Pan, Y.; Li,
Y.; Yin, B.; Jiang, H. Angew. Chem. Int. Ed. 2010, 24, 4054–4058.
20
(a) Chen, L.; Chen. H.; Luque, R.; Li, Y. Chem. Sci. 2014, 5, 3708-3714; (b) Zhang, D.; Guan, Y.; Hensen, E.; Xue, T.; Wang, Y.
Catal. Sci. Technol. 2014, 4, 795-802; (c) Huang, Y.; Zheng, Z.; Liu, T.; Lü, J.; Lin, Z.; Li, H,; Cao, R. Catal. Commun. 2011, 14, 2731.
21
Bernt, S.; Guillerm, V.; Serre C.; Stock, N. Chem. Commun. 2011, 47, 2838–2840.
1
2

10

22
(a) Poliakoff, M.; Fitzpatrick, J. M.; Farren, T. R.; Anastas, P. T. Science, 2002, 297, 807–810. (b) Collins, T. Science, 2001, 291,
48–49.
23
(Review) Anastas, P.; Eghbali, N. Chem. Soc. Rev. 2010, 39, 301–312.
24
(a) Russell, D. A. M.; Shiang, D. L. ACS Sustainable Chem. Eng. 2013, 1, 2−7. (b) Hansen, K. B.; Hsiao, Y.; Xu, F.; Rivera, N.;
Clausen, A.; Kubryk, M.; Krska, S.; Rosner, T.; Simmons, B.; Balsells, J.; Ikemoto, N.; Sun, Y.; Spindler, F.; Malan, C.; Grabowski, E.
J. J.; Armstrong III, J. D. J. Am. Chem. Soc. 2009, 131, 8798–8804. (c) Calvo-Flores, F. G. ChemSusChem 2009, 2, 905–919. (d)
Sheldon, R. A. Green Chem. 2008, 10, 359–360. (e) Horvath, I.; Anastas, P. Chem. Rev. 2007, 107, 2169−2173.
25
Office of Pollution Prevention and Toxics, The Presidential Green Chemistry Challenge Awards Program, Summary of 2008
Award Entries and Recipients, US Environmental Protection Agency, Washington DC, EPA 744R08002, 2008.
26
Shih, Y.-H.; Lo, S.-H.; Yang, N. S.; Singco, B.; Cheng, Y.-J.; Wu, C.-Y.; Chang, I.-H.; Huang, H.-Y.; Lin, C.-H. ChemPlusChem,
2012, 77, 982–986.
27
Non-dry methanol was used as the solvent for impregnation of Pd(II) into MIL-88B-NH2.
28
Serre,C.; Mellot-Draznieks, C.; Surblé, S.; Audebrand, N.; Filinchuk, Y.; Feréy, G. Science, 2007, 315, 1828–1831.
29
Elemental analysis: Found: Cr, 14.15; Pd, 7.40. Calculated: Cr, 15.45; Pd, 7.38. Based on [Cr3O(O2C-C6H3NH2CO2)3F(H2O)2(PdCl2)0.7•CH3CN•4H2O].
30
No peaks corresponding to crystalline palladium could be observed in the diffraction pattern.
31
Search on the Cambridge Crystallographic Database over 1372 reported crystal structures of palladium complexes with the
proposed coordination environment [Pd(II)(ArNH2)2Cl2] give average values of Pd–N = 2.03(6) Å and Pd–Cl = 2.30(2) Å.
32
Horcajada, P.; Salles, F.; Wuttke, S.; Devic, T.; Heurtaux, D.; Maurin, G.; Vimont, A.; Daturi, M.; David, O.; Magnier, E.; Stock,
N.; Filinchuk, Y.; Popov, D.; Riekel, C.; Férey, G.; Serre, C. J. Am. Chem. Soc. 2011, 133, 17839–17847.
33
Carson, F.; Su, J.; Platero-Prats, A. E.; Wan, W.; Yun, Y.; Samain, L.; Zou, X. Cryst. Growth Des. 2013, 13, 5036–5044.
34
King, K. S. W.; Everett, D. H.; Haul, R.; Moscou, L.; Pierotti, R. A.; Rouquerol, J.; Siemieniewska, T. Pure Appl. Chem. 1985, 57,
603–619.
35
When methanol was used as a co-solvent (10% v/v) throughout the whole experiment, a quick deactivation of the catalyst was
observed within the first 4–5 h.
36
Zhang, D.; Guan, Y.; Hensen, E. J. M.; Xue, T.; Wang, Y. Catal. Sci. Technol. 2014, 4, 795–802.

11