Oxidative cleavage of isoeugenol to vanillin under molecular oxygen catalysed by cobalt porphyrin intercalated into lithium taeniolite clay

Article abstract of DOI:10.1016/j.molcata.2012.05.005

Three types of Co-porphyrin complexes differing in their porphyrin-ring substituents and nonionic or cationic structure were intercalated into lithium taeniolite (LiTN) using a simple cation exchange method. All the intercalated catalysts displayed expanded clearance spaces (in the range of 0.30-0.51 nm) compared with that of the parent LiTN (0.24 nm). The catalysts were applied for the oxidative cleavage of isoeugenol to vanillin with molecular oxygen as the sole oxidant, and the reaction proceeded effectively under mild reaction conditions. The highest vanillin selectivity of 72% was obtained with CoTPyP/TN, which had the widest clearance space among the three Co-porphyrin catalysts. These intercalated catalysts showed high stability during oxidation, enabling their recycling for at least three runs.

Full text of DOI:10.1016/j.molcata.2012.05.005

Accepted Manuscript
Title: Oxidative cleavage of isoeugenol to vanillin under
molecular oxygen catalysed by cobalt porphyrin intercalated
into lithium taeniolite clay
Authors: Indri Badria Adilina, Takayoshi Hara, Nobuyuki
Ichikuni, Shogo Shimazu
PII:
DOI:
Reference:
S1381-1169(12)00148-3
doi:10.1016/j.molcata.2012.05.005
MOLCAA 8421
To appear in:
Journal of Molecular Catalysis A: Chemical
Received date:
Accepted date:
5-4-2012
2-5-2012
Please cite this article as: I.B. Adilina, T. Hara, N. Ichikuni, S. Shimazu, Oxidative
cleavageofisoeugenoltovanillinundermolecularoxygencatalysedbycobaltporphyrin
intercalated into lithium taeniolite clay, Journal of Molecular Catalysis A: Chemical
(2010), doi:10.1016/j.molcata.2012.05.005
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Oxidative cleavage of isoeugenol to vanillin under molecular oxygen catalysed by cobalt porphyrin
intercalated into lithium taeniolite clay
Indri Badria Adilinaa,b, Takayoshi Haraa, Nobuyuki Ichikunia, and Shogo Shimazua*
aGraduate School of Engineering, Chiba University, Japan
bResearch Centre for Chemistry, Indonesian Institute of Sciences, Indonesia
* Corresponding author. Tel./ Fax:+81 43 290 3379.
Email address: shimazu@faculty.chiba-u.jp (S. Shimazu)
Abstract
Three types of Co-porphyrin complexes differing in their porphyrin-ring substituents and nonionic or
cationic structure were intercalated into lithium taeniolite (LiTN) using a simple cation exchange method. All
the intercalated catalysts displayed expanded clearance spaces (in the range of 0.30 to 0.51 nm) compared
with that of the parent LiTN (0.24 nm). The catalysts were applied for the oxidative cleavage of isoeugenol to
vanillin with molecular oxygen as the sole oxidant, and the reaction proceeded effectively under mild reaction
conditions. The highest vanillin selectivity of 72% was obtained with CoTPyP/TN, which had the widest
clearance space among the three Co-porphyrin catalysts. These intercalated catalysts showed high stability
during oxidation, enabling their recycling for at least three runs.
Keywords: Vanillin synthesis; Isoeugenol oxidation; Cobalt porphyrin-clay; Molecular oxygen
1
Page 1 of 29

1. Introduction
Vanillin is the main component in vanilla and is widely applied as a flavouring agent in foods, beverages,
and pharmaceuticals [1]. The annual demand for vanillin is reported to be 15,000 tons, whereas only
approximately 50 tons are produced naturally by extraction from vanilla beans [2]. The remaining quantity is
supplied via industrial synthetic protocols such as the catechol route [3,4], which requires consecutive steps
and uses petroleum-based phenol as a starting material. The developments of alternative methods are strongly
desired, especially those using biomass-derived compounds, such as isoeugenol, a component of essential oil.
Although the biochemical transformation of isoeugenol to vanillin using a bacterial strain [5,6] and catalytic
conversions using a homogeneous Co-salen complex [7,8] have been reported previously, the yields of
vanillin have been low in such alternative processes.
Metalloporphyrins are well known for their catalytic activity, which mimics that of cytochrome P-450
enzymes in reactions such as the aerobic oxidation of olefins to epoxides, alcohols, aldehydes, and ketones
[9,10]. Among the metalloporphyrins, Co-porphyrins are reported to have high oxygen-binding activity,
making them efficient catalysts for the oxidation of various alkenes in an oxygen atmosphere [11-13]. The
only drawback of metalloporphyrins is their instability in solution; these compounds easily self-destruct
during catalytic reactions due to the oxidative dimerisation of metalloporphyrin molecules and the
decomposition of the porphyrin ring [14,15]. To increase the stability of metalloporphyrins and to prevent
catalyst deactivation, the intercalation of cationic and anionic metalloporphyrins into layered inorganic
supports, such as clay minerals has been investigated [16-18]. The intercalation of nonionic metalloporphyrins
2
Page 2 of 29

as a starting molecule, however, has rarely been studied. LiTN (Li[(Mg2Li)(Si4O10)F2]), a mica-type clay, is
known to have a high cation-exchange capacity (C.E.C = 268 meq/100 g), a high swelling ability, and
controllable interlayer distances and has been reported to display potential as a catalyst support for a variety of
metal complexes [19-21].
Herein, we report a new method for vanillin synthesis through the oxidative cleavage of isoeugenol by use
of taeniolite-intercalated Co-porphyrin catalysts (CoTPP/TN, CoTPyP/TN, and CoTC12PBP/TN) and
molecular oxygen as the sole oxidant (Scheme 1). Three types of Co-porphyrin complexes, with different
aromatic substituents and either nonionic or cationic structures were synthesised and intercalated into LiTN
using a simple cation-exchange method to produce 5,10,15,20-meso-tetrakis(4-phenyl)porphyrin (CoTPP),
5,10,15,20-meso-tetrakis(4-pyridyl)porphyrin
(CoTPyP),
and
5,10,15,20-meso-tetrakis(4-(dodecylpyridinium)benzoate)porphyrin (CoTC12PBP). The structures of these
catalysts were characterised by induction-coupled plasma–atomic emission spectroscopy, powder X-ray
diffraction, and Raman spectroscopy, and their catalytic activity in the oxidative cleavage of isoeugenol to
vanillin was investigated under various reaction conditions.
3
Page 3 of 29

Scheme 1. Oxidative cleavage of isoeugenol to vanillin catalysed by various taeniolite-intercalated
Co-porphyrins.
2. Experimental
2.1. Materials
Isoeugenol, substituted aldehydes, propionic acid, and anhydrous cobalt chloride were purchased from
Wako Pure Chemical Ind., Ltd. Isoeugenol was distilled under vacuum before use. Pyrrole was obtained from
Sigma-Aldrich Co., LLC. All solvents were purified before use, as described elsewhere [23]. The solid
support LiTN was obtained from Topy Ind., Ltd., Japan.
2.2 Synthesis of porphyrins
2.2.1. Synthesis of 5,10,15,20-meso-tetrakis(4-phenyl)porphyrin (TPP)
TPP was prepared using the typical procedure for porphyrin synthesis described by Adler and Longo [24]:
benzaldehyde (7.1 g, 67.0 mmol) was dispersed in propionic acid (150 mL) and heated to 90 ºC. A solution of
4
Page 4 of 29

pyrrole (4.5 g, 67.0 mmol) in propionic acid (20 mL) was then added slowly through a dropping funnel to the
reaction mixture, which was then heated to 141 ºC. After 3 h, the reaction mixture was filtered, washed with a
mixture of hot water and methanol, dried in vacuo, and purple crystals were obtained.
UV-Vis λmax (EtOH): 415 nm; 1H-NMR (400 MHz, CDCl3): δ 7.73-7.80 (m, 12H), 8.21-8.23 (m, 8H), 8.84 (s,
8H), –2.78 (s, 2H); FT-IR(KBr): ν(N-H) 3315.
2.2.2. Synthesis of 5,10,15,20-meso-tetrakis(4-pyridyl)porphyrin (TPyP)
TPyP was synthesised in the same manner as for TPP synthesis except for the use of
4-pyridinecarbaldehyde (7.2 g, 67.0 mmol) as the starting substituted aldehyde. Purple crystals were obtained.
1
UV-Vis λmax(EtOH): 417 nm; H-NMR (400 MHz, CDCl3): δ 8.16-8.17 (m, 8H), 8.88 (s, 8H), 9.06-9.08 (m,
8H), –2.93 (s, 2H); FT-IR(KBr): ν(N-H) 3313, ν(C=Nmeso) 1562 cm-1.
2.2.3. Synthesis of 5,10,15,20-meso-tetrakis(4-dodecylpyridiniumbenzoate)porphyrin (TC12PBP)
The synthesis of TC12PBP was performed by first preparing the catalyst precursors 12-bromododecanol
(C12BrOH),
N-(ω-hydroxydodecyl)pyridium
bromide
(C12PB),
and
5,10,15,20-meso-tetrakis(4-carboxyphenyl)porphyrin (TCPP). The synthesis of C12BrOH was conducted by
diluting 1,12-dodecanediol (8.3 g, 41.0 mmol) in toluene (120 mL). HBr (3.9 g, 48.0 mmol) was then added
dropwise and stirred at 110 °C for 32 h. The resulting mixture was diluted with diethyl ether, dried with
MgSO4, and further purified by Kugelrohr distillation to yield white crystals. Subsequently, C12PB was
5
Page 5 of 29

prepared by dissolving the synthesised C12BrOH (0.5 g, 1.9 mmol) in pyridine (1 mL, 12.4 mmol) followed
by heating to 110 °C for 24 h. The resulting white crystals were filtered and washed with diethyl ether. The
synthesis of TCPP was conducted as described for TPP except for the use of 4-carboxybenzaldehyde (10.1 g,
67.0 mol) as a starting substituted aldehyde. Finally, TC12PBP was prepared by refluxing the synthesised
TCPP (0.2 g, 0.2 mmol) with SOCl2 (4.8 g, 160 eq. relative to TCPP) at 80 °C for 8 h. The mixture was then
evaporated under vacuum, and the previously synthesised C12PB (0.4 g, 6.5 eq. relative to TCPP) along with
CH3CN (50 mL) and triethylamine (0.4 g, 16 eq. relative to TCPP) were added, and the mixture was stirred at
90 °C for 36 h. The resulting solid was filtered, washed with acetone, and dried in vacuo to obtain green
crystals.
1
UV-Vis λmax (EtOH): 421 nm; H-NMR (400 MHz, DMSO): δ 1.11-1.58 (m, 64H), 1.84-1.95 (m, 16H),
4.30-4.60 (m, 16H), 8.11 (s, 8H), 8.38 (m, 16H), 8.64 (t, 4H), 8.85 (s, 8H), 9.07 (d, J = 5.3, 8H), –2.91 (s, 2H);
FT-IR(KBr): ν(N-H) 3303, ν(C=Omeso) 1710, ν(C=N+meso) 1631 cm-1.
2.2.4. Synthesis of cobalt porphyrin complexes
CoTPP was synthesised by reacting TPP (0.15 g, 0.25 mmol) with CoCl2 (0.5 g, 15 eq.) in CH3CN (50 mL).
The solution was refluxed for 24 h, and metalation was confirmed by the red shift of the Soret Band and the
disappearance of three or four visible Q bands [22]. CoTC12PBP was synthesised from TC12PBP in the same
manner, whereas CoTPyP was synthesised by the reaction of TPyP (0.15 g, 0.25 mmol) with CoCl2 (0.5 g, 15
eq. relative to TPyP) in DMF (15 mL) with refluxing in air for 72 h. The mixture was filtered, washed with
6
Page 6 of 29

water, and dried in vacuo to produce reddish-brown crystals.
2.3. Preparation of taeniolite-intercalated cobalt porphyrin catalysts
A suspension of LiTN (0.25 g) in H2O (20 mL) was prepared and allowed to swell at 30 ºC for 1 h. An
aliquot of CoTPP or CoTC12PB (0.005-0.150 g) dissolved in H2O (30 mL) was then added to the LiTN
suspension followed by stirring for 7 days. In the case of CoTPP, an aliquot of DMF (2 mL) was also added to
the solution to increase its solubility. The intercalation of CoTPyP was conducted in the same manner using
0.3 M of CoCl2 aqueous solution as the intercalation solvent with stirring for 24 hours. The procedure
afforded brown powders of the TN-intercalated Co-porphyrin catalysts (CoTPP/TN, CoTPyP/TN, and
CoTC12PBP/TN).
2.4. Characterisation of the catalysts
The synthesised Co-porphyrin complexes (CoTPP, CoTPyP, and CoTC12PBP) were analysed by UV-Vis
spectroscopy on a Shimadzu UV-2101 and by Fourier transform infrared spectroscopy on a Horiba FT-720
equipped with a TGS detector. The successful metalation of the Co-porphyrin complexes was monitored by
the red shift of the Soret Bands and the disappearance of ν(N-H) bonds in the porphyrin structure [22]. The
taeniolite-intercalated Co-porphyrin catalysts (CoTPP/TN, CoTPyP/TN, and CoTC12PBP/TN) were further
characterised by powder X-ray diffraction (XRD) acquired at room temperature from 2º to 16º on an MAC
SCIENCE MXP3V system. The diffractometer was operated at 25 kV and 20.0 mA with Cu Kα radiation. The
7
Page 7 of 29

evaluation of the active sites of the catalysts was performed with resonance Raman spectroscopy on a JASCO
NRS-1800 equipped with a charge-coupled plasma device detector (CCD) excited at 457.9 nm by an Ar laser.
The loading amount of Co intercalated into the TN interlayers was determined by induction-coupled
plasma–atomic emission spectroscopy (ICP-AES) on a Seiko Instruments SPS 1700 HVR after dissolving the
samples in 0.1 M HNO3 (35%).
2.5. General procedure for the oxidative cleavage of isoeugenol
An intercalated Co-porphyrin catalyst (Co: 1 mol%) was placed with a mechanical stirrer in a Schlenk
flask equipped with a reflux condenser. Isoeugenol (0.0492 g, 0.3 mmol) and a sufficient amount of
acetonitrile as the solvent were then added, and the solution was kept under a pure O2 atmosphere (0.1 MPa)
at 50 ºC for 24 h. The product composition was confirmed by GC analysis using biphenyl as an internal
1
standard, GC-MS, and H NMR analysis. GC analyses were performed on a Shimadzu GC-17A Gas
Chromatograph packed with a Thermon 3000 or KOCL-3000T column. The mass spectral analyses were
conducted with a Shimadzu GC-MS 2010 equipped with an Rtx-5MS column. Proton NMR measurements
were performed on a JEOL JNM-GSV 400 Hz using CDCl3 or DMSO as a solvent. Oxidations under elevated
O2 pressures were performed in a stainless steel autoclave following the same procedure.
2.6. Catalyst reuse tests
After the first catalytic reaction, catalyst was separated each time by centrifugation, washed with
8
Page 8 of 29

acetonitrile (3 × 10 mL), and dried in vacuo before reuse. The recovered catalyst was then characterised by
XRD and FT-IR analyses. Each succeeding cycle of oxidation was performed as described for the first run.
3. Results and Discussion
3.1. Characterisation of taeniolite-intercalated cobalt porphyrin catalysts
3.1.1. ICP-AES elemental analyses and powder X-ray diffraction studies
The loading amount of Co determined by ICP-AES analysis showed that the Co content was 0.04 mmol g-1
in both CoTPP/TN and CoTPyP/TN (Table 1). In the case of CoTC12PBP/TN, a lower loading amount of
0.01 mmol g-1 was observed, possibly due to the higher charge (4+) of the Co-porphyrin complex (Scheme 1).
The XRD patterns of the CoTPyP/TN, CoTPP/TN and CoTC12PBP/TN catalysts are presented in Fig. 1, and
their analytical parameters are summarised in Table 1.
9
Page 9 of 29

Fig. 1. Powder X-ray diffraction patterns of the catalysts; (a) LiTN, (b) CoTPyP/TN, (c) CoTPP/TN, and (d)
CoTC12PBP/TN.
Table 1. Analytic parameters and estimated elevation angles of the Co-porphyrins in the TN interlayer.
An expansion of the clearance space (CS = basal spacing (d001) – layer thickness (0.96 nm)) in TN was
observed for all the catalysts after the intercalation of the Co-porphyrin complexes. The d001 peaks of these
catalysts shifted to lower diffraction angles between 5.9º to 7.0º, indicating that the CSs of these catalysts
(ranging from 0.30 to 0.51 nm) had increased over that of the parent LiTN (0.24 nm). The widest CS of 0.51
10
Page 10 of 29

nm was obtained for CoTPyP/TN, whereas CoTPP/TN and CoTC12PBP/TN showed CSs of 0.32 and 0.30 nm,
respectively. Our results suggest that the CS expansion of these intercalated catalysts is related to the
orientation of the interlayer Co-porphyrin complexes. The orientation of a Co-porphyrin complex in the TN
interlayer can be estimated in terms of the elevation angle (θ) towards the TN silicate sheet. We found that the
Co-porphyrin molecule in the CoTPyP/TN catalyst is apparently inclined at an angle of 27º with respect to the
TN layer sheet, whereas lower values of θ were observed for CoTPP and CoTC12PBP, at 17º and 4º,
respectively. In contrast, we first predicted that CoTC12PBP would have a larger CS due to the pillarisation of
alkyl groups in the porphyrin ring, as we previously observed with Rh(I)-bipyridine complexes intercalated
into LiTN [20].
3.1.2. Raman spectra
The cationic Co-porphyrin complex (CoTC12PBP) was readily able to enter the interlayer of TN by cation
exchange because it dissociated into cationic [CoIITC12PBP]4+ species in the intercalation solution.
Conversely, nonionic Co-porphyrin complexes (CoTPyP and CoTPP) in general hardly intercalate into layered
cation exchangers due to a lack of positive charge. In the present study, however, it was evident that CoTPyP
and CoTPP intercalated into LiTN, as indicated by the ICP-AES analyses and the XRD studies. Several
mechanisms have been proposed for the formation of positively charged species in metalloporphyrins, e.g.,
protonation of the substituent in the porphyrin ring or oxidation of the metal centre [25,26]. To identify the
active species in the CoTPyP/TN catalyst, the oxidation state of the Co centre of CoTPyP in the TN interlayer
11
Page 11 of 29

was investigated using Raman spectroscopy (Fig. 2).
Fig 2. Raman spectra of the (a) CoIITPyP and (b) CoIIITPyP/TN catalysts.
Raman spectroscopy was a suitable technique for our purposes because the spectrum is free from matrix
interference, as clay minerals are poor light scatters and do not absorb in the visible range [27]. The
vibrational bands of CoTPyP were detected at 1567 and 1369 cm-1 for the ν2 and ν4 modes, respectively (Fig.
2(a)). After the intercalation of CoTPyP into LiTN, these vibrational bands shifted to higher wavenumbers, i.e.,
at 1576 cm-1 (ν2) and 1371 cm-1 (ν4). It is known that this ν2 mode arises from the in-phase ν(CβCβ) stretching
of the porphyrin ring together with a smaller contribution from the ν(CαCm), whereas the ν4 mode is a response
of the pyrrole half-ring symmetric stretching modes consisting of the ν(CαN) and ν(CαCβ) vibrations of the
12
Page 12 of 29

pyrrole ring [28, 29]. These modes have been reported to be markers of the oxidation states of the metal
centres in Co- and Fe-porphyrins, where shifts (of up to 12 cm-1) to higher wavenumbers are normally
observed following changes in the metal centres from lower to higher oxidation states [30, 31]. The shifts of
the ν2 and ν4 modes in CoTPyP/TN indicated that the oxidation state of Co in the CoTPyP complex changed
after intercalation. Specifically, the Raman study indicates a change in the oxidation state of the Co centre
from CoIITPyP to CoIIITPyP. A similar trend was previously observed in a study of a water-soluble porphyrin
TMPyP, where an upward shift (of 6 cm-1) in the ν2 mode was observed when the Co centre was oxidised from
CoIITMPyP to CoIIITMPyP [30]. Consequently, a positively charged species [CoIIITPyP]+ was intercalated in
the interlayer of LiTN by cation exchange. The same mechanism presumably takes place for the CoTPP/TN
catalyst when the cationic [CoIIITPP]+ is intercalated into the TN interlayer.
3.2. Oxidative cleavage of isoeugenol in molecular oxygen
A wide range of oxidation products have been previously reported in the catalytic conversion of isoeugenol
using a homogeneous Co-salen complex or a biochemical transformation by a bacterial strain, e.g., vanillin
[5,6,7,8], isoeugenol epoxide [5,6], isoeugenol diol [5,6], vanillyl methyl ketone [32], dehydrodiisoeugenol
[5,6,7,8], acetovanillone [8], vanillylmandelic acid [8], vanillic acid [5,6], protocatechuic acid [5], guaiacol
[6], and vanillyl alcohol [6]. In this study of Co-porphyrin/TN-catalysed oxidative cleavage of isoeugenol, the
catalytic reaction was performed using CoTPP/TN, CoTPyP/TN, or CoTC12PBP/TN under pure O2 at
atmospheric pressure with acetonitrile as the solvent. Only four byproducts accompanied the formation of the
13
Page 13 of 29

desired vanillin (1), viz., isoeugenol epoxide (2), isoeugenol diol (3), vanillyl methyl ketone (4), and
dehydrodiisoeugenol (5), as summarised in Table 2.
Table 2. Oxidative cleavage of isoeugenol using various Co-porphyrin/TN catalysts.
1
The oxidation products were identified by H NMR and their mass spectra; here, vanillic acid was not
formed, in contrast to the biochemical transformation of isoeugenol to vanillin [5]. The CoTC12PBP/TN
catalyst showed the lowest conversion (22%) of isoeugenol, with a yield of only 1% (1) (entry 1). The low
activity of this catalyst was attributed to the flat bilayer of CoTC12PBP/TN, which resulted in steric hindrance
between the substrate and the Co-porphyrin centre. In contrast, a high activity of 96% isoeugenol conversion
was observed for CoTPP/TN, although the selectivity for (1) was low (6%), yielding (4) (50%) and (5) (36%)
as the main products (entry 2). Sletzinger reported that (4) (48% yield) was formed by the oxidation of
14
Page 14 of 29

isoeugenol in formic acid followed by subsequent hydrolysis [32], whereas (5) (20% yield) was produced by
auto-oxidation via a free radical mechanism, which normally became the main reaction pathway at a low
concentration of the Co-salen catalyst [7,33]. Among all the catalysts tested, CoTPyP/TN showed the highest
isoeugenol conversion, at >99%, with a 34% yield of (1) (entry 3). The high activity of the CoTPyP/TN
catalyst is presumably caused by its wide CS. The CoTPyP/TN catalyst, which had the most expanded CS,
provided a larger reaction space, allowing for easier access of isoeugenol into the TN interlayer. This
relationship was also observed for montmorillonite-intercalated Co-porphyrin complexes, where the oxidation
of cyclohexene was best achieved using catalysts having the highest CS [34]. Test reactions with the catalyst
precursors CoCl2, LiTN, and Co/TN displayed no catalytic activity in the oxidation of isoeugenol (entries 4, 5
and 6). Additionally, in control reactions under a nitrogen atmosphere and without a catalyst, no conversion of
isoeugenol was detected in either case (entries 7 and 8). These results indicate that the formation of (4) and (5)
was in fact initiated by the novel catalysts.
Solvents, such as toluene, dichloroethane, methanol, tetrahydrofuran, and dimethylformamide, were also
tested with the CoTPyP/TN catalyst; however, other than acetonitrile, only toluene favoured the oxidative
cleavage of isoeugenol, although the yield of (1) was less than 10% (Fig. 3).
15
Page 15 of 29

Fig. 3. Solvent effects on the vanillin yield.
Reaction conditions: cis-, trans-isoeugenol, 0.3 mmol; CoTPyP/TN catalyst, Co: 0.003 mmol; O2, 0.1 MPa;
50 ⁰C, 24 h; solvent, 1 mL. Yield determined by GC and 1H NMR analyses.
We investigated the effects of temperature on isoeugenol oxidation with the novel catalysts in acetonitrile;
the results are presented in Fig. 4. The yield of (1) increased as the reaction temperature was increased from
35 to 50 ⁰C and then decreased at 60 ⁰C, when the byproducts dominated. In an attempt to improve the yield
of (1) by reducing the amounts of (4) and (5) produced, a detailed study on the oxidative cleavage of
isoeugenol under various conditions was conducted using the CoTPyP/TN catalyst, given that this catalyst
showed the highest activity and selectivity among the three novel catalysts.
16
Page 16 of 29

Fig. 4. Temperature effects on vanillin yield.
Reaction conditions: cis-, trans-isoeugenol, 0.3 mmol; CoTPyP/TN catalyst, Co: 0.003 mmol; O2, 0.1 MPa;
50 ⁰C, 24 h; CH3CN, 1 mL. Yield determined by GC and 1H NMR analyses.
3.3. Effects of various reaction conditions
3.3.1. Effects of oxygen gas and isoeugenol concentrations
The effects of varying the concentrations of oxygen gas and isoeugenol substrate were investigated
because auto-oxidation producing (5) was previously reported to occur at high oxygen uptake and depend on
the substrate concentration [7, 35]. We thus hypothesised that the yield of vanillin can be improved when
auto-oxidation is suppressed. The relationship between the concentration of oxygen gas and vanillin yield was
evaluated using O2/N2 gas mixtures containing 10%, 20%, and 50% O2 (v/v); the results are shown in Fig. 5.
17
Page 17 of 29

Fig. 5. The effects of oxygen concentration on isoeugenol conversion, vanillin yield, and vanillin selectivity.
Reaction conditions: cis-, trans-isoeugenol, 0.3 mmol; CoTPyP/TN catalyst, Co: 0.003 mmol; O2/N2 gas
mixture, 0.1 MPa; 50 ⁰C, 24 h; CH3CN, 1 mL. Conversion and yield determined by GC 1H NMR analyses.
When the concentration of O2 was reduced from 100% (v/v) to 10% (v/v), the total yield of byproducts was
reduced to 29%. However, a low yield of (1) (9%) was observed at 10% O2, resulting in a low vanillin
selectivity of only 23%. The yield of (1) rose to 33% at 50% (v/v), and the isoeugenol conversion was 87%;
however, the yield was lower than that obtained with 100% O2, although a slight increase in the selectivity for
(1) was observed (going from 34 to 38%). Thus, decreasing the concentration of O2 did not improve the
selectivity for (1) in this system.
In contrast, when the isoeugenol concentration was lowered from 0.3 to 0.015 mol L-1, a significant change
in selectivity was achieved (Fig. 6). Byproduct formation was reduced at the lower concentration of
isoeugenol, and the highest yield of (1) (45%) was obtained when the concentration was 0.03 mol L-1.
Accordingly, the selectivity increased from 34% (0.3 mol L-1) to 75% (0.03 mol L-1). At a very low
18
Page 18 of 29

concentration of 0.015 mol L-1, however, the selectivity was slightly lower (66%) than that at 0.03 mol L-1
because the yields of the byproducts also decreased. In view of these promising results, an isoeugenol
concentration of 0.03 mol L-1 was then selected for the subsequent investigation of the effects of varying the
oxygen pressure and the amount of catalyst.
Fig. 6. Effects of the concentration of isoeugenol on vanillin yield and vanillin selectivity.
Reaction conditions: cis-, trans-isoeugenol, 0.015 - 0.3 mmol; CoTPyP/TN catalyst, Co: 0.003 mmol; O2, 0.1
MPa; 50 ⁰C, 24 h; CH3CN, 1-10 mL. Yield determined by GC and 1H NMR analyses.
3.3.2. The effects of oxygen pressure and catalyst/substrate ratio
The effects of O2 pressure and amount of catalyst were investigated at 0.03 mol L-1 isoeugenol
concentration; the results are listed in Table 3.
19
Page 19 of 29

Table 3. CoTPyP/TN catalysed oxidative cleavage of isoeugenol at various O2 pressures and catalyst amounts.
To our surprise, no formation of (5) was detected when the O2 pressure was increased from 0.1 to 0.3 MPa
(entries 2 and 3) at an isoeugenol concentration of 0.03 mol L-1, and the yield of (1) increased from 45 to 50%
(entry 3). As a consequence of the elevation of O2 pressure, however, the yield of (2) increased drastically. We
then considered whether (1) was generated from (2). To investigate this reaction pathway, (2) was applied as a
substrate for the oxidation [36]. The reaction did not proceed, and no (1) was detected (entry 4). Thus, we
concluded that the formation of vanillin did not involve isoeugenol epoxide as a precursor but that (1) was
formed directly by the oxidative cleavage of isoeugenol. Earlier studies on the reaction mechanism of the
oxidative cleavage of isoeugenol catalysed by a Co-salen complex indicated that an attack by a Co-O2 species
at the -carbon of the allyl substituent in the isoeugenol radical leads to a dioxetane species, which then
20
Page 20 of 29

decomposes into vanillin and acetaldehyde [7]. By analogy, this mechanism can be postulated for the
CoTPyP/TN-catalysed reaction. We speculate that metallaoxetane intermediates are formed when
metalloporphyrin derivatives are utilised as catalysts [37,38], resulting in either the oxidative cleavage of the
C=C bond to yield vanillin or epoxidation to form isoeugenol epoxide and the corresponding diol.
Further elevation of the O2 pressure to 0.5 MPa yielded no significant improvement of the oxidation process
compared with the reaction under 0.3 MPa (entry 5). However, when larger amounts of catalyst (2 mol% and
3 mol% Co) were applied, increased yields of (1) (67 and 72%, respectively) were obtained, and the formation
of both (2) and (3) was further reduced (entries 6 and 7). The best experimental conditions were determined to
be the CoTPyP/TN catalyst at 3 mol% Co under an O2 pressure of 0.3 MPa, resulting in >99% isoeugenol
conversion and 72% vanillin yield (entry 7). Reaction profiling of the oxidative cleavage for 21 h and the
effect of hot filtration of the catalyst (catalyst removal) at a reaction time of 5 h were then performed (Fig. 7).
After the removal of the catalyst, the reaction stopped, with no further formation of vanillin, confirming the
necessity of the presence of the catalyst for the reaction to proceed.
21
Page 21 of 29

Fig. 7. Reaction profile (●) and the effect of hot filtration (catalyst removal) at 5 h (○).
Reaction conditions: cis-, trans-isoeugenol, 0.3 mmol; CoTPyP/TN catalyst, Co: 0.003 mmol; O2, 0.1 MPa;
50 ⁰C, 24 h; CH3CN, 1 mL. Yield determined by GC and 1H NMR analyses.
3.4. Reusability of the catalyst
After the first run, the used CoTPyP/TN catalyst was recovered and analysed by XRD (Fig. 8).
22
Page 22 of 29

Fig. 8. Powder X-ray diffraction profiles of (a) fresh CoTPyP/TN and (b) recovered CoTPyP/TN catalysts.
The recovered CoTPyP/TN catalyst displayed the d001 peak at the same position (25.9º) as the fresh
catalyst, indicating that the CS of the CoTPyP/TN catalyst was maintained even after the catalytic oxidation.
Moreover, when the reaction solution was analysed by UV-Vis spectroscopy, no Soret Band peak was detected,
confirming that the leaching of CoTPyP molecules from the clay interlayer did not occur during the oxidation.
The catalyst was tested for at least three consecutive runs and showed no significant loss of activity, providing
vanillin yields 72, 71, and 71%, in the first, second, and third runs, respectively (Fig. 9).
23
Page 23 of 29

Fig. 9. Recycle of catalyst.
Reaction conditions: cis-, trans-isoeugenol, 0.3 mmol; CoTPyP/TN catalyst, Co: 0.003 mmol; O2, 0.3 MPa;
50 ⁰C, 24 h; CH3CN, 10 mL. Yield determined by GC and 1H NMR analyses.
4. Conclusions
A new synthetic method for preparing vanillin from isoeugenol was developed using taeniolite-intercalated
Co-porphyrin catalysts (CoTPP/TN, CoTPyP/TN, and CoTC12PBP/TN) with molecular oxygen as the sole
oxidant. These catalysts were synthesised from nonionic or cationic Co-porphyrin complexes, and all the
novel catalysts yielded expanded CS after intercalation compared with that of the parent LiTN. The best result
for the oxidative cleavage of isoeugenol (72% vanillin yield) was achieved using the CoTPyP/TN catalyst
(CS= 0.51 nm) at an isoeugenol concentration of 0.03 mol L-1 and 100% (v/v) O2 at a pressure of 0.3 MPa.
The formation of byproducts was controlled by decreasing the concentration of isoeugenol and increasing the
O2 pressure. The intercalation of Co-porphyrin complexes into the LiTN interlayer enhanced catalyst stability,
24
Page 24 of 29

and the catalyst was reusable for at least three runs.
Acknowledgements
The authors thank Professor Saburo Aimoto and Mr. Mitsuo Ohama (Osaka University, Japan) for their
assistance with the Raman spectroscopy measurements and our profitable discussions.
References
[1] N. J. Walton, M. J. Mayer, A. Narbad, Phytochemistry 63 (2003) 505–515.
[2] R. Wyers, The world of food ingredients (2011) 34-36.
[3] P. Metivier, US Patent (2001) 6184421.
[4] P. Metivier, US Patent (2001) 5783737.
[5] D. Hua, C. Mab, S. Lin, L. Song, X. Denga, Z. Maomy, Z. Zhang, B. Yu, P. Xu, J. Biotech. 130 (2007)
463-470.
[6] Y. Zhang, P. Xu, S. Han, H. Yan, C. Ma, Appl. Microbiol. Biotechnol. 73 (2006) 771-779.
[7] S. Drago, B.B. Corden, C. W. Barnes, J. Am. Chem. Soc. 108 (1986) 2453-2454.
[8] A. Salanti, M. Orlandi, E. Tolppa, L. Zoia. Int. J. Mol. Sci. 11 (2010) 912-926.
[9] I. Tabushi, N. Koga, J. Am. Chem. Soc. 101 (1979) 6456-6458.
[10] B. Meunier, Chem. Rev. 92 (1992) 1411-1456.
[11] J. P. Collman, J. I. Brauman, K. M. Doxsee, T. R. Halbert, S. E. Hayes, K.S. Suslick, J. Am. Chem. Soc.
100 (1978) 2761-2766.
25
Page 25 of 29

[12] H. Tang, C. Shen, M. Lin, A. Sen, Inorg. Chim. Acta 1109 (2000) 300–302.
[13] A. K. Mandal, J. Iqbal, Tetrahedron 53 (1997) 7641-7648.
[14] S. Ozawa, Y. Watanabe, S. Nakashima, T. Kitagawa, I. Morishima. J. Am. Chem. Soc. 116 (1994)
634-641.
[15] R. F. Pasternack, P. R. Huber, P. Boyd, G. Engasser, L. Francesconi, E. Gibbs, P. Fasella, G. C. Venturo,
L. de C. Hinds, J. Am. Chem. Soc. 94 (1972) 4511-4517.
[16] S. Nakagaki, G. S. Machado, M. Halma, A. A. Marangon, K. A. Castro, N. Mattoso, F. Wypych. J. Catal.
242 (2006) 110-117.
[17] C. Crestini, A. Pastorini, P. Tagliatesta, J. Mol. Catal. A. 208 (2004) 195-202.
[18] T. J. Pinnavaia, P. K. Welty J. Am. Chem. Soc. 97 (1975) 3819-3820.
[19] T. Hara, M. Hatakeyama, A. Kim, N. Ichikuni, S. Shimazu. Green Chem. DOI: 10.1039/c2gc16437j.
[20] N. Yamaguchi, S. Shimazu, N. Ichikuni, T. Uematsu, Chem. Lett. 33 (2004) 208-209.
[21] Y. Permana, S. Shimazu, N. Ichikuni, T. Uematsu, Catal. Comm. 6 (2005) 426-430.
[22] A. Wolberg, J. Manassen, J. Am. Chem. Soc. 92 (1970) 2982-2991.
[23] W. L. F. Armarego, C. L. L. Chai, Purification of Laboratory Chemicals, sixth ed., Elsevier Inc.,
Massachusetts, 2005.
[24] A. D. Adler, F. R. Longo, W. J. Shergalis, J. Am. Chem. Soc. 86 (1964) 3145-3149.
[25] W. R. J. Scheidt, J. A. Cunningham, J. L. Hoard, J. Am. Chem. Soc. 95 (1973) 8289-8294.
[26] C. Shi, F. C. Anson, J. Am. Chem. Soc. 113 (1991) 9564-9570.
26
Page 26 of 29

[27] P. M. Dias, D. L. A. De Faria, V. R. L. Constantino, J. Incl. Phenom. Mol. Rec. Chem. 38 (2000)
251-266.
[28] K. Prendergast, T. G. Spiro, J. Am. Chem. Soc. 114 (1992) 3793-3801.
[29] S. Choi, T. G. Spiro, K. C. Langry, K. M. Smith, D. L. Budd, G. N. La Mar, J. Am. Chem. Soc. 104
(1982) 4345-4351.
[30] S. N. Terekhov, S. G. Kruglik, V. L. Malinovskii, V. A. Galienvsky, V. S. Chirvony, P. Turpin, J. Raman
Spectrosc. 34 (2003) 868-881.
[31] N. Parthasarathi, C. Hansen, S. Yamaguchi, T. G. Spiro, J. Am. Chem. Soc. 109 (1987) 3865-3871.
[32] M. Sletzinger. US Patent (1982) 4337360.
[33] I. J. Miller, Tetrahedron Lett. 49 (1972) 4955-4958.
[34] H. Kameyama, F. Narumi, T. Hattori, H. Kameyama, J. Mol. Catal. A. 258 (2006) 172-177.
[35] A. Zombeck, R. S. Drago, B. B. Corden, J. H. Gaul, J. Am. Chem. Soc. 103 (1981) 7580-7585.
[36] Isoeugenol epoxide (2) was synthesized according to literature procedures: M. Imuta, H. Ziffer, J. Org.
Chem. 44 (1979) 1351-1352.
[37] J. P. Collman, T. Kodadek, S. A. Raybuck, J. I. Brauman, L. M. Papazian, J. Am. Chem. Soc. 107 (1985)
4343-4345.
[38] J-E. Baeckvall, F. Boekman, M. R. A. Blomberg. J. Am. Chem. Soc. 114 (1992) 534-538.
27
Page 27 of 29





Co-porphyrin complexes were intercalated into taeniolite clay by cation exchange.
The intercalated catalysts displayed expanded clearance spaces.
Oxidative cleavage of isoeugenol to vanillin proceeded under molecular oxygen.
A selectivity for vanillin of 72% was achieved with the catalyst having the highest
clearance space.

The catalyst was reusable for at least three runs.
Page 28 of 29

*Graphical Abstract (for review)
Expanded
reaction space
LiTN clay
Co-porphyrin
Intercalation
Isoeugenol
O2
Co
0.51 nm
0.24 nm
Vanillin
Yield: 72%
Page 29 of 29