Continuous-Flow Synthesis of Supported Magnetic Iron Oxide Nanoparticles for Efficient Isoeugenol Conversion into Vanillin

Article abstract of DOI:10.1002/cssc.201701884

This work presents the synthesis of iron oxide nanocatalysts supported on mesoporous Al-SBA-15 by using a continuous-flow setup. The magnetic nanomaterials were tested as catalysts in the oxidative disruption of isoeugenol by using hydrogen peroxide as a green oxidant, featuring high activities (63–88 % conversion) and good selectivities to vanillin (44–68 %). The catalytic systems exhibited good magnetic properties when synthesized under continuous-flow conditions at temperatures not exceeding 190 °C. The use of microwave irradiation significantly reduced times of reaction drastically but exerted negative effects on catalyst reusability.

Full text of DOI:10.1002/cssc.201701884

Accepted Article
Title: Continuous flow synthesis of supported magnetic iron oxide
nanoparticles for efficient isoeugenol conversion to vanillin
Authors: María Dolores Marquez, Pepijn Prinsen, Hangkong Li, Kaimin
Shih, Antiono Ángel Romero Reyes, and Rafael Luque
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemSusChem 10.1002/cssc.201701884
Link to VoR: http://dx.doi.org/10.1002/cssc.201701884
A Journal of
www.chemsuschem.org

ChemSusChem
10.1002/cssc.201701884
FULL PAPER
Continuous flow synthesis of supported magnetic iron oxide
nanoparticles for efficient isoeugenol conversion to vanillin
[
a]
[a]
[b]
[b]
[a]
M. Dolores Marquez-Medina, Pepijn Prinsen, Hangkong Li, Kaimin Shih, Antonio A. Romero,
[
a]
and Rafael Luque *
Abstract: This work presents the synthesis of iron oxide
materials are attractive porous materials due to their
hexagonally well-ordered mesopores with large specific surface
nanocatalysts supported on mesoporous Al-SBA-15 using
a
2
-1 [12,13]
continuous flow set-up. The magnetic nanomaterials were tested as
catalysts in the oxidative disruption of isoeugenol using hydrogen
peroxide as green oxidant, featuring high activities (63-88 %
conversion) and good selectivities to vanillin (44-68 %). The catalytic
systems exhibited good magnetic properties when synthesized
under continuous flow conditions at temperatures not exceeding
areas (> 900 m g ).
relatively simple protocol for the synthesis of MNPs using a
continuous flow process in which Fe nanoparticles were
Recently, Gawande et al. reported a
2
O
3
deposited on aluminosilicates in a single step. This setup can
provide greater process control, flexibility and productivity and
has practical potential as an alternative to industrial scale
synthesis processes based on traditional procedures.[14,15]
190 °C. The use of microwave irradiation significantly reduced times
of reaction drastically, but exerted negative effects on catalyst re-
usability.
Vanillin can be isolated from natural source derived extracts,
including vanilla beans and pods, pine woods (e.g. Pinus
tabuliformis) and roasted coffee.[16] It can also be obtained from
lignin fractions through depolymerization and subsequent
conversion and/or isolation. The increasing global demand of
Introduction
vanillin cannot be met anymore by the supply of vanilla orchid
An important part of current research in green chemistry focus
on the transition towards more sustainable manufacturing
processes that efficiently utilize resources and raw feedstocks,
reduce waste streams and avoid the use of toxic and hazardous
materials. To bring this into practice, the use of recyclable
catalysts has proven to be very useful for a wide range of
chemical processes.[1-3] The deposition of highly active catalytic
nanoparticles onto various organic or inorganic supports is
probably the most efficient strategy to recover nanocatalysts. In
this context, nanocatalysts recovery by magnetic fields could
enable more efficient separation as compared to conventional
decantation and filtration, especially when nanocatalysts are
expensive (i.e. noble metals) and difficult to separate.[4,5] The
incorporation of an iron oxide core in the nanocatalyst increases
its density and thus also its recovery efficiency by conventional
separation methods. The synthesis of magnetic nanoparticles
pods as the sole source.[17] Only ca. 1 % of the global production
of vanillin is derived from vanilla pods; the majority produced
synthetically using e.g. lignin[18] and (iso)eugenol[19] as starting
materials. The commercial value of vanillin extracted from vanilla
-1
[20]
pods varies between 1200-4000 USD kg , excluding inflation.
Although the oxidative disruption of isoeugenol (Scheme 1) was
already reported in literature, significant results were reported by
Shimoni et al. using Bacillus subtilis sp. strains as
biocatalysts,[21] resulting in vanillin production with 12-14 mol%
-1
yields (0.6-0.9 g L after 24-48 h at 30 °C). Enzymatic
conversion by lipoxygenases was also reported by Markus et
[22]
al.,
with yields varying between 10 and 15 mol%. Major
-1
improvements (32.5 g L vanillin after 72 h at 37 °C) were
reported by Zhao et al. using Bacillus fusiformis strains.
[23]
Although these biocatalytic transformations are very attractive
from an environmental point of view, they possess inherent
drawbacks that include limited selectivity and stability in the
systems, along with rather long reaction times and the need for
nutrient media. Other catalytic systems were proposed by
Herrmann et al. using 1 mol% methyltrioxorhenium at short
reaction times (10 min). However, anhydrous hydrogen peroxide
was required in 3:1 molar excess and MnO was added to
2
prevent further oxidation to vanillic acid.[24] More recently,
Gusevskaya et al. reported the catalytic conversion of
(
MNPs) was proposed for several applications being currently a
hot topic in future catalyst design practises.[6-10] However, highly
active MNPs are rather difficult to synthesize due to their
tendency to aggregate and/or coalescence (sintering). This is
important, because the nanosize of iron particles can have a
significant impact on their catalytic activities. One way to
overcome this problem is to support MNPs on mesoporous
solids, which in turn also improve the stability of the metal or
metal oxide nanoparticles.[11] In particular, mesoporous SBA-15
2 2
isoeugenol, with H O as green oxidant in combination with
(
5
n-butyl) NVO
4
3
and pyrazine-2-carboxylic acid, reaching
0 mol% yields in acetonitrile at 40 °C after 2 h.[
25]
[
a]
Bsc. M.D. Marquez Medina, Dr. P. Prinsen, Prof. A. A. Romero,
Prof. R. Luque
Departamento de Química Orgánica
Universidad de Cordoba
Campus de Rabanales, Edificio Marie Curie (C-3), Ctra Nnal IV-A,
Km 396, E-14014 Cordoba, Spain
E-mail: q62alsor@uco.es
[
b]
MSc. H. Li, Dr. K. Shih
Department of Civil Engineering,
The University of Hong Kong
Haking Wong Building, Pokfulam Road, Hong Kong HK 852, China
Supporting Information and the ORCID identification number(s) for
the author(s) of this article can be found under
Scheme 1. Oxidative disruption of isoeugenol to vanillin and side-products.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201701884
FULL PAPER
Results and Discussion
Table 2. Elemental analysis (atomic percentages) according
to SEM/EDX and ICP-MS.
A
series of magnetic nanoparticle (MNP) catalysts were
SEM/EDX (%)
ICP-MS (%)
Si Fe
synthesized in a continuous flow set up (Scheme 2) using iron
oxide and Al-SBA-15 as support. This synthesis method
promotes the deposition of iron oxide particles on the outer
surface of Al-SBA-15 and promotes a close interaction of Fe
with Al, rather than forming iron oxide particles within Al-SBA-15
pores. A catalyst prepared via deposition of iron oxide on
Al-SBA-15 by classic wet impregnation (FeNMag-WI) was
additionally synthesized for comparative purposes. MNP
synthesis temperature was varied between 150 and 210 ºC to
study its effect on the magnetic properties (Table 1) and the
incorporation of iron (Table 2) in the final materials. FeNMag-WI
was comparably non-magnetic (on a macroscopic basis) as
classic wet impregnation proceeds at room temperature. The
largest magnetic susceptibility was reached when the synthesis
temperature was set at 170 ºC, while magnetic properties were
lost at an apparent cut-off temperature of 195 ºC. The elemental
composition results showed an iron (Fe) content in the whole
materials in the range of 8 to 19 wt% (ICP-MS), while the Fe
content at the surface was higher, with values ranging from 17 to
Catalyst
Al
0.8
-
Si
40.7
-
Fe
4.9
Al
0.8
0,8
0.8
0.8
-
FeNMag-WI[a]
22.3
28.2
22.2
22.3
-
13.3
8.4
11.7
13.3
-
FeMag-150
FeMag-160
FeMag-170
FeMag-180
FeMag-190
FeNMag-195
FeNMag-200
[a]
-
1.0
1.3
1.1
1.2
1.1
0.5
23.6
30.3
28.9
29.3
28.3
16.7
21.4
18.2
17.1
19.8
18.2
48.1
-
-
-
0.7
-
17.4
-
19.0
-
Prepared by wet impregnation at 25 ºC with Fe loadings identical to those in
FeMag-170.
The structure of the synthesized materials was studied by XRD.
Figure 1 shows the wide and low angle diffraction patterns of
different temperature synthesized materials. Low angle
diffractograms depict the presence of three peaks, one intense
peak at low 2Ɵ values (reflection line d100) and two weaker
peaks at higher angles (reflection lines d110 and d200),
characteristic of SBA-15 structures.[26] The decrease in the
intensity of these lines in FeNMag-200 indicates that Al-SBA-15
support underwent structural deterioration upon Fe incorporation.
The high iron content in FeNMag-200 (Table 2) was most
probably related to a partial collapse of the aluminosilicate SBA-
48 wt% (SEM/EDX). Fe content in the material prepared by
classic wet impregnation (FeNMagWI), in contrast, did not show
this surface enrichment effect. On the other hand, Fe content
increased with higher synthesis temperatures, coinciding with
the collapse of the SBA-15 structure in the materials synthesized
at temperatures above 195 ºC.
1
5 as mesopores could be partly blocked by the high amount of
iron oxide deposited on the aluminosilicate surface. These
hypotheses were in good agreement with the wide angle XRD
pattern in the materials in which the broad peak around 22°
almost completely disappeared. Next to Al-SBA-15, peaks from
2 3
Fe O were also visible. Diffraction lines from iron oxide in
FeNMagWI (Figure S1 in the Supporting Information) were
comparably hardly detectable, a most plausible indication of a
highly dispersed iron oxide nanoparticles. The iron oxide
diffraction pattern of FeNMag-200 could be indexed as a
2 3
hematite crystal structure (α-Fe O , ICPDS No. 33-0,664)
(
Figure 1a), while XRD patterns of FeMag-190 and FeMag-170
are more difficult to assign. All diffraction peaks of maghemite
and magnetite crystals are located within 1° difference.[27] The
diffraction patterns of iron oxide in FeMag-190 and FeMag-170
samples (Figure 1b and 1c) are likely to be maghemite crystals
(
2 3
γ-Fe O , JCPDS No. 39-1346). However, the absence of
magnetite cannot be excluded at the present signal-to-noise
ratio and resolution since iron oxide nanoparticles typically
exhibit noisy diffractograms,[28]. Crystal sizes of non-magnetic
Scheme 2. Continuous flow synthesis of magnetic nanoparticles (MNPs).
Table 1. Magnetic properties of the synthesized MNPs.
(
FeNMag-200) vs magnetic samples (FeMag-190 and FeMag-
70) were measured to be 30-40 nm and < 10 nm, respectively.
1
FeNMag-200 and FeMag-170 samples were analyzed using a
quantitative XRD technique, allowing the quantification of the
amorphous material content. The content of the hematite phase
in FeNMag-200 sample was 50 % by weight, while the content
of the magnetic phase was less than 10 % by weight in
FeMag-170.
Magnetic susceptibility χ
m
Catalyst
Synthesis temperature (ºC)
×10-
6
3
-1
(
m
kg )
FeMag-150
FeMag-160
FeMag-170
FeMag-180
FeMag-190
FeNMag-195
FeNMag-200
FeNMag-205
150
160
170
180
190
195
200
205
73
140
206
161
190
[
[
[
a]
a]
a]
[
a] Non-magnetic.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201701884
FULL PAPER
Table 3. Porosity analysis and acidity determination of synthesized (magnetic)
nanoparticles.
Average
pore
diameter
Average
pore
Surface acidity
-1 [a]
S
m
BET
-1
(µmol g )
BrØnsted
Catalyst
2
(
g )
volume
-
1
Total
61
-
(nm)
(mL g )
8
5
Al-SBA-15
FeNMag-WI
FeMag-150
FeMag-160
FeMag-170
FeMag-180
FeMag-190
FeNMag-195
800
450
470
470
350
310
300
320
7.6
0.9
-
0.3
7.3
7.8
7.7
7.5
6.9
7.6
0.5
0.5
0.5
0.4
0.4
0.3
0.4
5
8
6
2
0
8
1
39
30
35
40
45
35
3
5
4
4
3
[
a] As determined by titration with base probes pyridine and 2,6-
dimethylpyridine for total and BrØnsted acidity, respectively.
Yepez et al. previously reported TEM images of pure
Al-SBA-15.[30] In comparison, TEM images (Figure 2) of
representative catalysts synthesized in the present study show
that the ordered porous structure of the Al-SBA-15 support was
preserved upon Fe incorporation in continuous flow. Hematite
(
FeNMag-200) and maghemite (FeMag-170) nanoparticles were
Figure 1. (Left) wide angle and (right) low angle XRD patterns of (a)
FeNMag-200, (b) FeMag-190 and (c) FeMag-170.
homogeneously distributed on the aluminosilicate support. High
resolution images of a single particle in the FeNMag-200
material along the plane zone (010) depict a periodic spacing of
0
.368 nm, consistent with the "d" spacing corresponding to
The adsorption-desorption isotherms of SBA-15 aluminosilicate
support and those of most representative synthesized
nanocatalysts are of type IV (see Figure S2 in the Supporting
Information), characteristic of mesoporous materials. All
synthesized solids exhibited relatively high specific surface
areas (≥ 300 m2 g-1), with pore sizes in the 7-8 nm range
α-Fe planes, indicative of single crystals with good
crystallinity. The periodic spacing between 0,489 nm bands in
FeMag-170 sample is also consistent with the corresponding "d"
2
O
3
2 3
spacing of the {111} plane of γ-Fe O .
(
Table 3). FeNMag-WI prepared by wet impregnation showed a
similar specific surface area and pore volume, but with a
significantly reduced pore diameter. These results pointed to the
deposition of a large part of the iron oxide nanoparticles into the
pores of Al-SBA-15. Importantly, both the specific surface area
and pore volume decreased with an increase of the temperature
employed in the continuous flow synthesis, whereas the pore
diameter was retained This fact can be explained by the
increasing iron oxide content (see Table 2), which contributes in
weight and partially obstructs the entrance to micropores in
mesoporous void spaces. Surface acid properties (Table 3), as
studied by the chromatographic pulse technique, pointed out no
drastic differences in total acidity for the different synthesized
nanocatalysts, whether magnetic or non-magnetic. In some
cases, the number of BrØnsted acid centers exceeded the
number of total acid centers, as DMPY is more basic than PY
(
pKb 7.4 and 8.8, respectively). The measured magnetic
susceptibilities ranged between 70-210×10- m3 kg
6
-1
for
magnetic nanocatalysts, consistent with the maghemite content
in the materials (10-15 wt%). Pure magnetite nanoparticles
generally show larger magnetic susceptibilities ca. 500×10 m3
kg .
-
6
Figure 2. (Left) TEM and (right) HRTEM images of (up) FeNMag-200 and
(bottom) FeMag-170 catalysts.
-1 [29]
These susceptibility values were large enough for the
magnetic nanocatalysts to be magnetically recovered from
reaction mixtures after reaction.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201701884
FULL PAPER
The synthesized MNPs were employed as heterogeneous
catalyst in the selective oxidative disruption of isoeugenol in
liquid phase using aqueous hydrogen peroxide as green oxidant,
both at room temperature and 90 ºC (conventional heating).
Figure 3 depicts the conversion of isoeugenol and vanillin
selectivity observed after 40 min and 24 h. Blank runs (no
significantly influence final results as compared to Fe alone, with
the exception of a higher selectivity to vanillin achieved at room
temperature after 24 h. This may be the result of a caging effect
of iron oxide and isoeugenol inside the Al-SBA-15 pores, leading
to lower amounts of side products. GC/MS analysis of the
products showed the presence of a diphenyl ether structure as
main side product in all the experiments, produced by oxidative
radical coupling of quinone methide tautomers of isoeugenol (or
derivatives).
catalyst) and using Al-SBA-15 provided
conversion (<20 %). The use of pure iron oxide nanoparticles
Fe) or a mixture of Fe with Al-SBA-15 comparatively led to
a very low total
(
significant conversions in the systems, pointing to iron oxide as
catalytically active phase. The presence of Al-SBA-15 did not
Figure 3. Isoeugenol conversion (blue, mol%) and selectivity to vanillin (red, mol%) without catalyst (blank), with Al-SBA-15, with pure iron oxide (Fe), with Al-
SBA-15 + Fe (mix), with Fe deposited by wet impregnation (FeNMag-WI) and with Fe(N)Mag catalysts, at 20 °C after (a) 40 min. and (b) 24 h, and at 90 °C after
(c) 40 min. and (d) 24 h.
selectivity was slightly reduced after 24 h as compared to 40 min.
an indication that further oxidation to vanillic acid or other
undesired processes took place. FeNMag-WI catalyst in contrast,
prepared by wet impregnation, yielded vanillin amounts in the
range of the results obtained with pure iron oxide (Fe) or with
mixed Fe + Al-SBA-15 (Figure 3). No obvious relationship was
As opposed to room temperature experiments, the advantage of
heating related to an already close to the equilibrium conversion
and selectivity after 40 min. (comparing with data after 24 h).
Still, interesting results were obtained at room temperature after
24 h, with conversions ranging between 63 and 91 % at vanillin
selectivities of ca. ≥ 50 %. Whereas the conversions increased
consistently with reaction time up to 24 h, vanillin selectivity
increased during the first 8 h of reaction, to then remain
practically constant at ca. 50 %. At high temperatures, the
2 3
found between Fe content and Fe O phases present in
materials (maghemite or hematite) when comparing conversion
and selectivity obtained with all synthesized nanocatalysts under
continuous flow conditions at different temperatures. The
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201701884
FULL PAPER
catalytic activity was more related to their BrØnsted acidity. The
fact that there was no direct relationship between catalytic
activity and Fe O phase together with the low selectivity to
2 3
appreciate as it was partially overlapped with the broad band
-1
from Si-O bonds (1062 cm ) in Al-SBA-15. Although the small
band at 1600 cm-1 is representative of C=C stretching in
isoeugenol, the amount adsorbed is negligible as any other
bands arising from isoeugenol are hardly observable. In addition
to these results, TG-DTA experiments (Figure 5A, see also
supporting information) showed a significant mass loss (ca.
4-10%) in the 200-400ºC range accompanied by an endothermic
peak in the DTA spectra, plausibly related to the observed
aromatic compounds in ATR experiments. Importantly, a second
small mass loss at higher temperatures (400-500ºC, Figure S3)
is also present and likely to be due to the formation of heavier
molecular weight aromatics as side products from the oxidation
reaction.
vanillin observed for Al-SBA-15 leads us to propose active
centers similar to those described previously by our research
group,[31] which showed a Fe-Al synergistic effect on Fe
2 3
O
nanoparticles supported on SBA-15 type aluminosilicates, with
excellent catalytic activities in oxidations of benzyl alcohol to
benzaldehyde. This effect is most appreciable (Figure 3)
comparing vanillin yields obtained for FeMag-170 and those
obtained with physical mixtures of both compounds separately
(
Fe + Al-SBA-15) as catalyst. Next to conventional heating,
microwave irradiation was also employed, showing beneficial
effects (apart from faster and homogeneous heating).
Outstandingly, the equilibrium seemed to be reached almost
completely already after 5 min. when microwave heating was
utilised (Table 4), (60-80 % conversion and 54-65 % selectivity).
The differences with conventional heating after 24 h were small
It can be thus concluded that the observed deactivation in the
catalytic experiments is mostly due to adsorption of aromatics-
including vanillin and heavier molecular weight compounds-on
the catalyst surface (fouling). FeMag-190 spent catalyst (after 5
reuses) was regenerated by recalcination at 300 ºC (in air, 2 h).
XRD analysis (Figure S4, Supporting Information) pointed out
the presence of the key diffraction lines corresponding to the
iron oxide phase with the inherently amorphous nature of the
aluminosilicate material. Interestingly, catalytic experiments of
regenerated spent catalysts at 300ºC (e.g. FeMag-190) showed
a remarkable recovery in the initial catalytic activity after
regeneration (Figure 4), which in principle may be fully
recovered upon thermal treatment at 550-600ºC (according to
TG-DTA results, Figure 5A, a temperature at which all organics
are fully removed). However, thermal regeneration was not
conducted at such a high temperature to preserve the magnetic
properties of the materials. Nevertheless, the regeneration
studies clearly evidenced that the catalytic activity drop due to
surface fouling could be solved by thermal treatment of the
catalysts (300ºC) after which most of the initial catalytic activity
in the systems was recovered. These findings also pointed out
the strong deactivating role of adsorbed aromatic compounds on
the surface after subsequent reuses, a phenomenon that also
supports the observed strong deactivation of Fe-containing
systems under microwave irradiation.
(
< 10 %). Somewhat lower selectivity was observed after 15 min.
due to possible side reactions.
Nanocatalysts presented in this work outperform previously
reported results, at least in cost-effectiveness, when comparing
results with previously reported data on the catalytic oxidation of
vanillin using oxygen or hydrogen peroxide as oxidants
(
Table 5). Still, the high amount of operational variables makes
the comparison not straightforward. Gusevskaya et al. obtained
similar vanillin yields (50 mol%) under comparable conditions,
although they used
unfriendly catalyst in ca. 50 times less quantity (mol% of
n-Bu) NVO
+ pyrazine-2-carboxylic acid (PCA)).[25]
a homogeneous and environmentally
(
4
3
Finally, as the iron nanoparticles were recovered efficiently by
activation in an external magnetic field, the re-usability of the
magnetic catalysts was assessed (Figure 4). Although the
magnetic properties were retained in the recovered catalysts,
both conversion and selectivity significantly dropped, particularly
after 4-5 uses. The spent FeMag-190 catalyst recovered after 4
cycles of 24 h at room temperature exhibited significantly lower
Table 4. Fe(N)Mag catalyzed isoeugenol conversion and selectivity to
vanillin using microwave irradiation.
2
-1
porosity as compared to the fresh catalyst (Table 3), 150 m g
Catalyst
Conversion (mol%)
Selectivity (mol%)
specific surface area, 0.4 nm pore diameter and 0.3 mL g-1 pore
volume. Comparably, vanillin yield was stable over 3 re-
utilization experiments using conventional heating (90 ºC) while
the yield dropped drastically when using microwave irradiation.
This indicates that, although beneficial for the reactions kinetics,
microwave irradiation had detrimental effects on catalyst
performance (both activity and selectivity). In an attempt to
provide further insights into catalyst deactivation, the reused
catalyst was analyzed by TG-DTA and ATR (Figure 5). ATR
spectrum showed absorption bands characteristic of vanillin-like
5
min.
15 min.
5 min.
15
min.
17
9
Blank
Al-SBA-15
18
15
73
77
80
69
68
68
60
62
19
16
75
75
81
70
70
67
61
57
10
10
59
65
56
61
58
62
59
54
FeMag-150
FeMag-160
FeMag-170
FeMag-180
FeMag-190
FeNMag-195
FeNMag-200
FeNMag-205
58
64
57
57
54
56
65
55
-1
aromatics including O-H stretching (3330 cm ), C=O stretching
-1
(
1675 cm ), and the characteristic stretching vibration
absorption of benzene ring corresponding to three bands at
-
1
1
513, 1432 and 797 cm . The band originated from the bending
vibrations of phenolic OH around 1250 cm-1 was more difficult to
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201701884
FULL PAPER
Table 5. Comparison of catalytic performance of FeMag200 catalyst with
previously reported ones for the selective oxidation of isoeugenol to vanillin.
Conversion
Selectivity
(mol%)
-
Yield
(mol%)
43
Catalyst
Ref.
[32]
[33]
(mol%)
PhI(OAc)
2
-NaY[a]
-
Nocardia iowensis
-
-
60
[
b]
whole cells
TiO (P25
Degussa)
2
84
9
8
[34]
[35]
[
c]
[
d]
CoTPyP/TN
99
72
70
57
71
50
49
(
n-Bu)
PCA[e]
FeMag-180
4 3
NVO +
71
[25]
[
f]
85
This work
[
0
0
a] 0.3 M isoeugenol in 6 mL acetonitrile, 2 eq. of catalyst, 130 °C, 5 min. [b]
.01 M isoeugenol in 2 mL aqueous buffer, 40 g L cells, 130 °C 32 h. [c]
.5 mM isoeugenol in 150 mL aqueous phase, 0.8 g L-1 catalyst, 20 °C, 2 h.
-
1
[
d] 0.03 mM isoeugenol in 10 mL acetonitrile, 1 mol% cobalt (as cobalt
porphyrin/lithium taeniolite), 50 °C, 24 h,
isoeugenol in 0.9 mL acetonitrile, 2 eq. H
3
bars pure
O
2
.
[e] 0.2
NVO
3
M
+
2
O
2
, 0.025 mol% (n-Bu)
4
0.05 mol% PCA, 80 °C, 2 h. [f] 0.5 M isoeugenol in 8 mL acetonitrile, 2.3 eq.
2 2
H O , 12.5 mol% catalyst, 20 °C, 24 h.
Figure 5. (a) TGA analysis and (b) ATR analysis of deactivated FeMag-190
catalyst recovered after 5 catalytic reuses.
Conclusions
The synthesis of iron nanoparticles supported on Al-SBA-15 has
been successfully performed using a simple and reproducible
continuous flow method. Synthesized nanomaterials exhibited
suitable good magnetic properties when the synthesis
temperature stayed below an apparent cut-off temperature of
195 °C. Nanocatalysts obtained provided excellent catalytic
activities and promising selectivities (>50%) to vanillin in the
selective oxidation of isoeugenol using hydrogen peroxide as
green oxidant. Optimum results pointed to an 85 % conversion
of 0.5 M isoeugenol with 57 % selectivity to vanillin after 24 h at
room temperature. Reaction times were drastically reduced to
5-15 min under microwave irradiation, however with negative
effects on catalyst re-usability. The catalyst deactivation was
related with surface fouling, with thermally-treated regenerated
catalysts (300ºC) able to recover over 80% of the initial catalytic
activity of fresh catalysts.
Experimental Section
Figure 4. Vanillin yield (mol%) obtained with FeMag-190 catalyst over 5
re-utilization experiments for the catalytic oxidation of isoeugenol at 20 ºC and
at 90 ºC (using both conventional heating and microwave irradiation). The
activity after regeneration (300ºC, under air, 2h) is also included showing the
almost complete regeneration of the initial catalytic activity upon calcination,
particularly for the bottom plot (time extended reactions, 24 h).
The synthesis of the mesoporous aluminosilicate Al-SBA-15 was
carried out following the procedure described by Stucky and co-
workers. The triblock co-polymer Pluronic P123 was used as
[36]
directing agent and dissolved in aqueous HCl (2M, pH 1.5) for
2
h at room temperature. After complete dissolution, tetraethyl
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201701884
FULL PAPER
orthosilicate (25 mmol) was used as the silica source and the
corresponding amount of aluminum isopropoxide (10 mmol) was
slowly added. The mixture was stirred for 24 h at room
temperature. Subsequently, the mixture was submitted to
hydrothermal treatment in an oven at 100 °C for 24 h. The
obtained solid was filtered off, dried and subsequently calcined
under nitrogen atmosphere at 600 °C for 2 h, and then under air
atmosphere for 4 h. The obtained SBA solids (Si/Al molar ratio
composition of the synthesized materials was studied in two
ways. First, the surface of the solids was analyzed with an Inca
Energy 250 microanalysis system using an JEOL JSM 6300
Scanning Electron Microscope equipped with a window type
Si/Li detector (ATW2), in the boron to uranium detection range
(137 eV to 5.9 keV). Secondly, the quantitative elemental
analysis of the solids was determined with ICP-MS, using an
Elan DRC-e equipment (Perkin Elmer SCIEX, Waltham, USA).
The surface morphologies of the particles were examined by
TEM and HRTEM using FEI Tecnai G2 20 S-TWIN and 2010F
JEOL electron microscopes, respectively. The nitrogen
adsorption-desorption isotherms were recorded at -196 °C using
a Micromeritics ASAP 2000 automatic analyzer. The samples
were degassed for 24 h at 100 °C under vacuum (p < 10-20 Pa)
prior to analysis. The linear part of the Brunauer–Emmett–Teller
20) were recovered and stored.
The preparation of magnetic nanoparticles (MNPs) was carried
using a continuous flow setup (Scheme 2). First, a stainless
steel reactor (15 cm × 1 cm i.d.) was filled with Al-SBA-15
support (ca. 3.0 g) and fixed between two quartz wool stoppers
above and beneath to prevent the displacement of the solids.
The reactor was equipped with two stainless steel filters, one at
the entrance and another one at the exit to avoid obstructions.
The synthesis started by injecting ethanol at 0.5 mL min.-1 during
(
BET) equation (relative pressure between 0.05 and 0.30) was
used to determine the specific surface area. The pore size
distribution was calculated using the adsorption branch of the
adsorption-desorption isotherm, applying the method of Barrett,
Joyner and Halenda (BJH).[ The magnetic susceptibility χm
was determined at room temperature using an MS2 magnetic
susceptibility meter (Bartington Instruments Ltd., UK) equipped
with a MS2B dual-frequency (470 and 4700 Hz) laboratory
sensor. χm is equal to the ratio of the magnetization M within the
material to the applied magnetic field strength H. This ratio,
strictly speaking, is the volumetric susceptibility, because
37]
4
5 min. while reaching the desired synthesis temperature.
Subsequently, the ethanol flow was changed to a saturated
solution of Fe(NO .9H O (precursor salt, Sigma Aldrich) in
3
)
2
2
-1
ethanol, at 0.5 mL min. during 1 h. After completion, propionic
acid was pumped at a flow of 0.5 mL min.-1 during 45 min. Then,
after closing the valve in this loop, an air flow of 30 mL min.-1
was injected for approximately 3 h while the reactor was
submitted to a programmed temperature ramp of 1 °C min.-1 to
reach a final temperature of 300 °C, which then was kept
constant for 1 h. FeNMag-WI was prepared by wet impregnation
at room temperature using an Fe O load corresponding with the
2 3
same Fe content as determined by ICP-MS analysis in the
FeMag-170 catalyst. We assumed that in wet impregnation
magnetization essentially involves
a certain measure of
magnetism (dipole moment) per unit volume. It is a measure of
the magnetic response of a material to an external magnetic
field. The determination of the surface acidity was carried out at
250 °C, with pyridine (PY) and 2,6-dimethylpyridine (DMPY) as
titration bases for total and BrØnsted acidity, respectively. The
bases were introduced (2, 3, 4, 5 and 6 µL) in a GC injector
coupled to a stainless steel column (10 cm × i.d. 2 mm) packed
with the analyte (40-70 mg), which in turn was coupled to a 50
cm chromatographic column containing 5 wt% polyphenylether
in Chromosorb AW-MCS 80/100. The bases remaining after
elution through the packed column were quantified by gas
chromatography equipped with a flame ionization detector (GC-
FID). Thermogravimetric analysis (TGA) was carried out in
equipment using ca. 15 mg sample under dynamic air
100 % of the iron was retained in the FeNMag-WI material. The
Fe , Al-SBA-15 and the FeNMag-WI materials were also
2
O
3
treated with propionic acid which was added dropwise in batch,
followed by calcination in air using an oven at the same
temperature program as used in the continuous flow synthesis.
MNP solids obtained were characterized by X-ray diffraction
(
XRD), nitrogen adsorption-desorption porosimetry, magnetic
-1
atmosphere (10 mL min ). Attenuated total reflectance infrared
Fourier transform spectroscopy (ATR-IR) spectra were recorded
susceptibility, scanning electron microscopy (SEM), inductively
coupled plasma mass spectroscopy (ICP-MS), (high resolution)
transmission electron microscopy and pulse chromatography (to
study the surface acid properties). XRD spectra were recorded
on a Bruker D8 X-ray diffractometer (10-80° 2θ range) in the
Bragg-Brentano geometry and in reflection mode, using a Cu X-
ray tube, a rotating platform, a monochromatic primary beam
and a high sensitivity detector. In addition, some samples were
analyzed using an advanced D8 X-ray diffractometer (Bruker
AXS GmbH, Germany) with a LYNXEYE detector and Cu Kα
irradiation at 40 kV and 40 mA, using a count time of 1 second
for phase identification and 1 second for phase quantification.
The quantitative analysis (QXRD) was carried out using TOPAS
TM
using a Spectrum Two instrument (Perkin Elmer, Waltham,
USA), from 4000 to 450 cm- with a resolution of 4 cm-1.
1
The reactions with isoeugenol were carried out by conventional
heating at one hand and by microwave heating at the other hand.
The conventional heated reactions were performed using a
multiple parallel synthesis system (Carrusel Reaction Station™,
Radleys Discovery Technologies Ltd., UK) equipped with
magnetic stirring at 1000 rpm. The experiments were run at 20
and at 90 °C, using 5 mmol isoeugenol and 11.7 mmol hydrogen
peroxide (1.2 mL of 33 % w/v H
with 100 mg catalyst, which were added when the desired
reaction temperature was reached (time 0). For the
2 2
O in water) in 8 mL acetonitrile
4
.2 software (Bruker AXS). All samples were mixed with 20 wt%
CaF (Merck, Germany) as internal standard for the
quantification of the amorphous content. The elemental
2
=
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201701884
FULL PAPER
experiments with Al-SBA-15+Fe
amounts of Al-SBA-15 (56.7 mg) and Fe O
2 3
2
O
3
-170, the corresponding
-170 (43.3 mg) were
[7]
S. Shylesh, V. Schünemann, W. R. Thiel, Angew. Chem. Int. Ed. 2010,
49, 3428–3459.
[
[
[
8]
Y. Zhu, L. P. Stubbs, F. Ho, R. Liu, C. P. Ship, J. A. Maguire, N. S.
Hosmane, ChemCatChem 2010, 2, 365–374.
used (to reach an equal Fe content in the reaction as compared
to the experiment with FeMag-170 catalyst, based on the Fe
content according to the ICP-MS analysis in Table 2). The
experiments with FeNMag-WI were carried out with 100 mg
catalyst, as it contained the same amount of Fe as in
FeMag-170. The reactions using microwave heating were
carried out in a CEM-Discover focused microwave, controlled
and monitored by a computer in standard mode ("Discover")
under pressure which allows control of the irradiation power,
temperature and pressure. The reactants used were 1.25 mmol
9]
V. Polshettiwar, R. Luque, A. Fihri, H. Zhu, M. Bouhrara, J. M. Basset,
Chem. Rev. 2011, 111, 3036–3075.
10] R. B. N. Baig, R. S. Varma, Chem. Commun. 2013, 49, 752–770.
[11] N. I. Cuello, V. R. Elías, E. Winkler, G. Pozo-López, M. I. Oliva, G. A.
Eimer, J. Magn. Magn. Mater. 2016, 407, 299–307.
[12] C. T. Kresge, M. E. Leonowicz, W. J. Roth, J. C. Vartuli, J. S. Beck,
Nature 1992, 359, 710–712.
[
13] J. S. Beck, J. C. Vartuli, W. J. Roth, M. E. Leonowicz, C. T. Kresge, K.
D. Schmitt, C. T.-W. Chu, D. H. Olson, E. W. Sheppard, S. B. McCullen,
J. B. Higgins, J. L. Schlenkert, J. Am. Chem. Soc. 1992, 114, 10834–
2 2
isoeugenol and 2.9 mmol H O in 2 mL acetonitrile, with 25 mg
10843.
catalyst at 90 °C, using 300 W irradiation power. In all
experiments the reaction temperature was reached after ca. 1
min without any significant overheating. The reaction mixture
was analyzed before, during and after the reaction by gas
chromatography on an Agilent Technologies 7890 A GC System
equipped with a Petrocol™ DH column (100 m x 0.25 mm x
[
14] M. B. Gawande, Y. Monga, R. Zboril, R. K. Sharma, Coord. Chem. Rev.
2015, 288, 118–143.
[15] M. B. Gawande, R. Luque, R. Zboril, ChemCatChem 2014, 6, 3312–
3313.
[
[
[
16] M. J. W. Dignum, J. Kerler, R. Verpoorte, Food Rev. Int. 2001, 17, 119-
120.
17] N. J. Walton, A. Narbad, C. B. Faulds, G. Williamson, Curr. Opin.
Biotechnol. 2000, 11, 490–496.
0.50 μm i.d.) and a FID detector. The temperature of the column
was set at 200 ºC (70 min. hold time) and the temperature of the
injector and detector at 300 ºC. The nitrogen gas flow was set at
18] M. Fache, B. Boutevin, S. Caillol, ACS Sustainable Chem. Eng. 2016,
4(1), 35–46.
-
1
3
3
mL min . The retention times of isoeugenol and vanillin were
[19] A. Franco, S. De, A. M. Balu, A. Garcia, R. Luque, Beilstein J Org.
Chem. 2017, 13, 1439-1445.
3.4 and 30.2 min., respectively. Isoeugenol calibration was
carried out in the 2.00-60.00 g L-1 range (r 0.98). The reaction
mixture was also analyzed with GC/MS, The standard deviation
on conversion and selectivity, as determined from two
independent experiments with catalyst FeNMag-200 at 90 ºC
was 7 and 2 % after 40 min. and 24 h, respectively. The
corresponding deviations on selectivity were 3 and 8 %,
2
[20] N. J. Walton, M. J. Mayer, A. Narbad, Phytochemistry 2003, 63, 505–
515.
[
21] E. Shimoni, U. Ravid, Y. Shoham, J. Biotechnol. 2000, 78, 1-9.
22] P. H. Markus, A. L. J. Peters, R. Roos, Process for the preparation of
phenylaldehydes. European patent EP 542348 (1992).
[
[23] L. Q. Zhao, Z. -H. Sun, P. Zheng, L. -L. Zhu, Biotechnol. Lett. 2005, 27,
1505–1509.
respectively.
The recycle experiments were carried out
[24] W. A. Herrmann, T. Weskamp, J. P. Zoller, R. W. Fischer, J. Mol. Catal.
A.-Chem. 2012, 363-364, 140-147.
identically. The catalyst was recovered with a hand magnet,
washed with acetonitrile and dried at 100 ºC. Regeneration of
the spent catalyst was performed by recalcination under
identical conditions as used in their synthesis (300 ºC, air, 2 h).
[25] E. V. Gusevskaya, L. Menini, L. A. Parreira, R. A. Mesquita, Y. N.
Kozlov, G. B. Shul’pin, J. Mol. Catal. A.-Chem. 2012, 363-364, 140-147.
[
26] L. Y. Shi, Y. M. Wang, A. Ji, L. Gao, Y. Wang, J. Mater. Chem. 2005,
5, 1392–1396.
27] F. Fajaroh, H. Setyawan, A. Nur, I. W. Lenggor, Adv. Powder Technol.
013, 24, 507-511.
1
[
2
Acknowledgements
[28] G. C. Papaefthymou, E. Devlin, A. Simopoulos, D. K. Yi, S. N. Riduan,
S. S. Lee, J. Y. Ying, J. Physical Review B 2009, 80, 024406, 1-9.
[
29] C. Peters, M. J. Dekkers, Phys. Chem. Earth 2003, 28, 659-665.
30] A. Yepez, J. M. Hidalgo, A. Pineda, R. Černý, P. Jíša, A. Garcia, A. A.
Romero, R. Luque, Green Chem. 2015, 17, 565-572.
R.L. gratefully acknowledges MINECO as well as FEDER funds
for funding under project CTQ2016-78289-P and financial
support from Programa Propio at University of Cordoba (Spain).
[
[31] A. M. Balu, A. Pineda, K. Yoshida, J. M. Campelo, P. L. Gai, R. Luque,
A. A. Romero, Chem. Commun. 2010, 46, 7825–7827.
Keywords: Isoeugenol • Oxidation • Vanillin • Magnetic •
[32] H. Marquez Alvarez, D. P. Barbosa, A. Tinoco Fricks, D. A. G. Aranda,
R. H. Valdés, O. A. C. Antunes, Org. Process Res. Dev. 2006, 10, 941-
Nanoparticles
943.
[
[
[
[
[
33] R. Seshadri, A. S. Lamm, A. Khare, J. P.N. Rosazza, Enzyme Microb.
[
[
[
1]
2]
3]
F. Lu, J. Ruiz, D. Astruc, Tetrahedron Lett. 2004, 45, 9443–9445.
D. Astruc, F. Lu, J. Ruiz, Angew. Chem. Int. Ed. 2005, 44, 7852–77872.
D. Astruc, Transition-metal nanoparticles in catalysis, Wiley-VCH,
Weinheim, Germany, 2008.
Technol. 2008, 43, 486–494.
34 V. Augugliaro, G. Camera-Roda, V. Loddo, G. Palmisano, L. Palmisano,
F. Parrino, M. A. Puma, Appl. Catal., B 2012, 111–112, 555– 561.
35] I. Badria Adilina, T. Hara, N. Ichikuni, S. Shimazu, J. Mol. Catal. A:
Chem. 2012, 361–362, 72–79.
[
4]
M. Iranmanesha J. Hulliger, Chem. Soc. Rev. 2017, Advance Article,
DOI: 10.1039/C7CS00230K.
36] D. Zhao, J. Feng, Q. Huo, N. Melosh, G. H. Fredrickson, B. F. Chmelka,
G. D. Stucky, Science 1998, 279, 548-542.
[
5]
6]
Q. M. Kainz, O. Reiser, Acc. Chem. Res. 2014, 47(2), 667–677.
A. H. Lu, E. L. Salabas, F. Schüth, Angew. Chem. Int. Ed. 2007, 46,
[
37] E. P. Barrett, L. G. Joyner, P. P. Halenda, J. Am. Chem. Soc. 1951, 61,
1222-1244.
373-380.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201701884
FULL PAPER
FULL PAPER
Magnetic iron oxide nanoparticles
supported on Al-SBA-15, prepared
under continuous flow, are highly
active and selective for the disruptive
oxidation of isoeugenol to vanillin
using hydrogen peroxide as green
oxidant.
María Dolores Marquez, Pepijn
Prinsen, Hangkong Li, Kaimin Shih,
Antonio Angel Romero, Rafael Luque*
Page No. – Page No.
Continuous flow synthesis of
supported magnetic iron oxide
nanoparticles for efficient
isoeugenol conversion to vanillin
This article is protected by copyright. All rights reserved.