Continuous flow synthesis of menthol: Via tandem cyclisation-hydrogenation of citronellal catalysed by scrap catalytic converters

Article abstract of DOI:10.1039/c9gc03299a

A continuous flow synthesis of menthol starting from citronellal catalysed by scrap catalytic converters is reported. The reaction was conducted in a tandem system connecting in series two catalytic systems, with the first having Lewis acid properties (favouring the cyclisation of citronellal to isopulegols) and the second having hydrogenation catalytic activity (catalysing the hydrogenation of isopulegols to menthols). A Lewis acid catalyst was prepared by supporting iron oxide nanoparticles over a waste material, i.e. the ceramic core of scrap catalytic converters (SCATs) via a microwave assisted method. Most importantly, SCATs, containing a low residual noble metal content, could be directly employed in the second step as hydrogenation catalysts. The reaction was performed studying the influence on the yield and selectivity to (-)-menthol of various reaction parameters (T, p and flow rate). Under the best reaction conditions (at a flow rate of 0.1 mL min^-1 and at 373 K and 413 K for cyclisation and hydrogenation steps respectively) a conversion of >99% of (+)-citronellal to (-)-menthol with 77% final yield was achieved.

Full text of DOI:10.1039/c9gc03299a

Green Chemistry
View Article Online
View Journal
PAPER
Continuous flow synthesis of menthol via tandem
cyclisation–hydrogenation of citronellal catalysed
by scrap catalytic converters†
Cite this: DOI: 10.1039/c9gc03299a
a
a
b
b,c,d
Alessio Zuliani,
Antonio A. Romero
Camilla Maria Cova, Roberta Manno, Victor Sebastian,
a
a,e
and Rafael Luque
*
A continuous flow synthesis of menthol starting from citronellal catalysed by scrap catalytic converters is
reported. The reaction was conducted in a tandem system connecting in series two catalytic systems,
with the first having Lewis acid properties (favouring the cyclisation of citronellal to isopulegols) and the
second having hydrogenation catalytic activity (catalysing the hydrogenation of isopulegols to menthols).
A Lewis acid catalyst was prepared by supporting iron oxide nanoparticles over a waste material, i.e. the
ceramic core of scrap catalytic converters (SCATs) via a microwave assisted method. Most importantly,
SCATs, containing a low residual noble metal content, could be directly employed in the second step as
hydrogenation catalysts. The reaction was performed studying the influence on the yield and selectivity to
Received 19th September 2019,
Accepted 20th November 2019
(
−)-menthol of various reaction parameters (T, p and flow rate). Under the best reaction conditions (at a
DOI: 10.1039/c9gc03299a
−1
flow rate of 0.1 mL min and at 373 K and 413 K for cyclisation and hydrogenation steps respectively) a
conversion of >99% of (+)-citronellal to (−)-menthol with 77% final yield was achieved.
rsc.li/greenchem
9,10
tasting properties, being the most valuable isomer.
−)-menthol being primarily extracted from natural oils, typically
−)-Menthol (Fig. 1) is one of the most produced solid-flavour of Mentha piperita and Mentha arvensis, alternative synthetic
compounds with worldwide demand of more than routes have been developed in order to stabilise the market
0.000 metric tons per year.
Its interaction with human receptors confers a fresh taste
Despite
1
. Introduction
(
(
a
1
,2
11
3
fluctuations resulting from unpredictable bad harvests.
Main industrial processes include the Symrise process
and a cooling effect to a large variety of products ranging from (known as the Haarmann & Reimer process) and the Takasago
oral hygiene, drugs and tobacco to confectionary, cosmetics process.
3
–5
and foodstuffs.
Moreover, (−)-menthol has been proved to
In the Symrise process, thymol is hydrogenated to menthols
be an analgesic substance, potentially expanding its market to ( i.e. all diastereomers namely isomenthol, neomenthol, neoi-
6
,7
drugs.
somenthol and menthol) and sequentially (−)-menthol is
Remarkably, the commonly used term “menthols” refers to recovered by distillation and crystallization. In the Takasago
the group of eight stereoisomeric forms: menthol, neo-
menthol, isomenthol and neoisomenthol, presented in
2
,8
Fig. 1.
However, (−)-menthol has enhanced cooling and
a
Departamento de Quımica Organica, Universidad de Cordoba, Edificio Marie-Curie
(
C-3), Ctra Nnal IV-A, Km 396, Cordoba, Spain. E-mail: rafael.luque@uco.es
b
Nanoscience Institute of Aragon and Chemical and Environmental Engineering
Department, University of Zaragoza, Pedro Cerbuna 12, 50009 Zaragoza, Spain
c
Networking Research Center CIBER-BBN, 28029 Madrid, Spain
d
Aragón Materials Science Institute, ICMA, CSIC, University of Zaragoza, Pedro
Cerbuna 12, 50009 Zaragoza, Spain
e
Peoples Friendship University of Russia (RUDN University), 6 Miklukho Maklaya
str., 117198 Moscow, Russia
†
Electronic supplementary information (ESI) available: Details of the continu-
ous flow tandem systems and additional experimental data. See DOI: 10.1039/
c9gc03299a
Fig. 1 Structure of (−)-menthol and its isomers.
This journal is © The Royal Society of Chemistry 2019
Green Chem.

View Article Online
Paper
Green Chemistry
obtained. To the best of our knowledge, there are no reports
on a similar approach in the literature to date. Flow conditions
were selected as an safe and controllable alternative to batch
2
7,28
synthesis.
In addition, flow chemistry has been reported as
one of the “Top Ten Emerging Technologies” with the poten-
2
9
tial to turn the planet more sustainable. In order to maxi-
mize each step of the reaction, avoiding the formation of side
products, flow conditions were carefully designed and con-
trolled stepwise in a tandem protocol by connecting in series
two cartridges, one containing a Lewis acid catalyst (designed
for the cyclisation of citronellal) and one containing a hydro-
genation catalyst (for the sequential hydrogenation of isopule-
gols). The continuous flow apparatus was set up by linking an
H-Cube® Mini Plus (Thalesnano Inc.) to an X-Cube™
(
Thalesnano Inc.). The apparatus was allowed to individually
control the temperature of each cartridge.
Scheme 1 Synthetic strategies for the sequential cyclisation and hydro-
genation of citronellal.
The Lewis acid catalyst was prepared by supporting iron
over a cheap waste material, i.e. the ceramic core of scrap auto-
motive catalytic converters (denoted as “SCATs”) to obtain iron
oxide nanoparticles. SCATs were selected due to both low cost
process (Scheme 1A), myrcene is transformed into (+)-citronel- (and availability) and the presence of noble metal nano-
3
0
lal, which is sequentially cyclised and hydrogenated to particles (<0.5 wt%). Furthermore, the direct applicability of
1
2
(
−)-menthol.
SCATs in catalysis has been previously demonstrated in simple
3
1–33
Similar to the Takasago process, in 2012, BASF started a batch experiments.
new plant where (−)-menthol is synthesized from (+)-citronellal
derived from the upgrading of citral oils (e.g. by the enantio-
1
3
selective hydrogenation of geraniol and nerol). The cyclisation 2. Experimental sections
of citronellal to isopulegols is normally carried out using zinc
bromide or tris(2,6-diarylphenoxy)aluminium as a catalyst.
1
4
Iron(II) chloride tetrahydrate (FeCl ·4H O, 99.99% trace metal
2 2
basis), iron(III) chloride hexahydrate (FeCl ·6H O, 99.99% trace
3 2
metal basis), toluene (≥99.8%), acetone (≥99.5%), sodium
hydroxide (NaOH, ≥98%), (±)-citronellal (≥95%), (+)-citronellal
(>96%), octane (98%), (−)-isopulegol (≥99%, enantiomeric
ratio: ≥99.5 : 0.5), (−)-menthol (≥99%, enantiomeric ratio:
However, the homogeneous nature of catalysts and
corrosion issues derived from the utilisation of bromide salts
has driven research in the field to investigate heterogeneous
1
5–18
systems.
These systems have been further explored in
order to design bi-functional catalysts having Lewis-acid and
hydrogenating properties for the one-step synthesis of
≥99.5 : 0.5), silica, and ethanol (≥98%) were purchased from
1
9–23
Sigma-Aldrich Inc., St Louis, MO, USA. All reagents were used
without any further purification. Scrap ceramic-cores of auto-
motive catalytic converters (SCATs) were collected from
Provaluta España Reciclaje de Metales, S.L., Córdoba (ES).
SCATs were ground by Provaluta S.L. via a grinding process
and provided in the form of powder.
(
−)-menthol starting from (+)-citronellal.
For example, Ru supported over zeolites (H-BEA) has been
recently reported as an effective catalytic system for menthol
production (up to 87% operating at 372 K under 25 bar H
2
24
pressure in 1 h). Oldani et al. employed Pd and Ru nano-
particles supported over perfluorinated superacid polymers
obtaining 99% yield to (−)-menthol at 353 K under 10 bar H
pressure after 23 h of reaction using water as the solvent. More
2
25
2.1 Setup of the tandem apparatus
recently, Pt/W bifunctional catalysts supported over mesoporous The tandem apparatus was set up by linking an H-Cube® Mini
silica (TUD-1) have led to 96% yield of menthols at 353 K for Plus (Thalesnano Inc.) to an X-Cube™ (Thalesnano Inc.), as
26
1
6 h under 20 bar H
2
pressure. Despite these interesting illustrated in Fig. 2. In detail, the outlet connector of the car-
results, the proposed synthetic strategies still have a number of tridge of the H-Cube® (Cartridge 1) was connected to the inlet
practical limitations, mainly due to the conventional batch con- of the cartridge of the X-Cube (Cartridge 2). The outlet of the
ditions, long reaction times and the need to use expensive noble cartridge of the X-Cube was connected to the pressure valve of
metals and/or sophisticated catalytic systems (Scheme 1B).
Herein, an alternatively simple and efficient approach for
the selective synthesis of (−)-menthol starting from (+)-citro-
the H-Cube®. More details can be seen in Fig. S1 of the ESI.†
2.2 General procedure for the synthesis of catalysts
nellal performed under continuous flow (tandem) conditions Before utilisation, SCATs were washed and dried in order to
catalysed by waste-derived catalysts was explored (Scheme 1C). remove the superficial carbonaceous residuals and all other
Under optimum reaction conditions, a maximum of 92% yield pollutant residues. More in detail, 50 g of SCATs were dis-
of menthols (84% selectivity to (−)-menthol) could be persed in 100 mL of distilled water and ethanol (1 : 1).
Green Chem.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Green Chemistry
Paper
cartridges sealed on both sides with sealing systems made of
graphite filled PTFE sealing rings, stainless steel filters and
PTFE membranes (CatCatrs® Thalesnano Inc.). Two different
types of cartridges were employed: 30 mm and 70 mm-long
ThalesNano CatCarts®. The cartridges were filled with 150 mg
and 450 mg of catalysts respectively. Prior to utilisation, the
filled cartridges were washed in the flow apparatus at 30 bar
liquid pressure with toluene at 373 K for 30′ at a flow rate of
mL min− , in order to remove all the eventual residues and
1
8
all iron nanoparticles not fixed/slightly fixed on the SCATs.
After washing cycles, fresh fluxed toluene was analysed by GC
and ICP-MS analyses in order to confirm the cleanliness of the
system and that no leaching of the catalysts occurred.
Fig. 2 Diagram of the setup for the continuous flow synthesis of
menthols from citronellal.
2.4 Catalyst characterization
Sequentially, the mixture was ultrasonicated in a US bath for
Powder X-ray diffraction (XRD) patterns were recorded with a
Bruker D8 DISCOVER A25 diffractometer (PanAnalytic/Philips,
Lelyweg) using CuKα (λ = 1.5418 Å) radiation. The patterns
were collected over a 2θ value ranging from 10° to 80° with a
step size of 0.018° and a counting time of 5 s per step.
2
h (due to US thermal effects, the temperature was found to
be ∼323 K). Finally, the sonicated SCATs were filtered on a
vacuum filter, washed several times with water, acetone and
ethanol and dried in a 373 K oven for 12 h.
A series of Fe/SCAT catalysts were prepared by varying the
wt% of Fe adapting a microwave-assisted synthesis developed
The metallic composition of the catalysts was determined
by Microwave Plasma Atomic Emission Spectroscopy (MP-AES)
3
4
by Moores et al. In detail, a fixed amount of washed SCATs
1.15 g) was added to 20 mL of a previously stirred (10′) homo-
geneous aqueous solution of FeCl ·4H O, FeCl ·6H O and
(
Agilent 4100 MP-AES). The analysis was carried out by digest-
ing 20 mg of the selected catalyst with 5 mL of nitric acid
HNO ) and hydrochloric acid (HCl) in a volume ratio of 1 : 3 at
73 K for 20′ in a microwave oven (Milestone Ethos Plus,
(
2
2
3
2
(
4
3
4
.7 mL of a solution of 1.25 M NaOH. Three samples (A, B and
C) were produced by varying the quantities of the iron precur-
sors, as reported in Table 1.
Sequentially, the mixture was heated in a MW-oven (Ethos
Microwave, Milestone Srl) at 383 K for 1 h (5′ ramp). The result-
ing material was filtered, washed several times with water and
ethanol, and dried in a 373 K oven for 12 h.
Milestone Srl). Before analysis, the digested sample was fil-
tered with hydrophilic syringe filters (0.2 µm) in order to
discard any fragmented particles and diluted with Milli-Q
water to a final volume of 30 mL.
The aberration corrected Scanning Transmission Electron
Microscopy (Cs-corrected STEM) images were acquired on a
FEI XFEG TITAN electron microscope operating at 300 kV and
equipped with a CETCOR Cs-probe corrector (CEOS Company),
allowing the formation of an electron probe of 0.08 nm. The
samples were prepared by US dispersion in ethanol and by
pipetting 10 μL of the obtained suspension onto a TEM copper
grid having a continuous carbon film. After complete evapor-
ation of the solution, the samples were analysed by high-angle
Following the same procedure, three additional samples of
Fe/SiO were prepared with varying quantities of the iron pre-
2
cursor as reported in Table 2 (having the same theoretical
1
2 wt% iron content of sample A, the most active Fe/SCATs, as
explained in section 3.2).
2
.3 Preparation of cartridges
In order to perform the continuous flow tests, washed SCATs, annular dark-field scanning transmission electron microscopy
Fe/SCATs or Fe/SiO
samples were charged in stainless steel (HAADF-STEM).
Elemental analysis was performed with an EDS (EDAX) detec-
2
tor which allows performing EDS experiments in the scanning
mode. The analysis was conducted at the Laboratory of Advanced
Microscopies, LMA-INA-University of Zaragoza. Z-Contrast
Table 1 Quantities of Fe precursors for the theoretical x wt% Fe/SCATs
Sample
Theoretical x wt% Fe
2 2
FeCl ·4H O/mg
3 2
FeCl ·6H O/mg
(achieved by using the HAADF detector) was employed in order
to enable an image contrast based on the atomic number.
X-ray photoelectron spectroscopy (XPS) analysis was per-
formed with an Axis Ultra DLD spectrometer (Kratos Tech). In
order to perform the analysis, the samples were mounted on a
sample rod placed in the pretreatment chamber of the spectro-
meter and sequentially evacuated at room temperature. The
A
B
C
12%
6%
3%
188
94
47
510
255
127
Table 2 Quantities of Fe precursors for the theoretical 12 wt% Fe/SiO
2
Sample
Theoretical x wt% Fe
FeCl
2 2
·4H O/mg
FeCl
3
·6H
2
O/mg spectra were excited by using a monochromatized AlKα source
at 1486.6 eV and subsequently run at 12 kV and 10 mA. Survey
spectra were recorded at a pass energy of 160 eV and for the
individual peak regions, the spectra were recorded with a pass
D
E
F
12%
12%
12%
188
556
—
510
—
270
This journal is © The Royal Society of Chemistry 2019
Green Chem.

View Article Online
Paper
Green Chemistry
energy of 20 eV. The analysis of the peaks was performed with
The elemental compositions of SCATs and Fe/SCATs were
the CasaXPS software using a weighted sum of Lorentzian and analysed by ICP-MS (Inductively Coupled Plasma Mass
Gaussian component curves after Shirley background subtrac- Spectrometry) and MP-AES (Microwave Plasma Atomic
tion. The binding energies were referenced to the internal C 1s Emission Spectroscopy). Al, Si and Mg as well as Ce, Fe and
standard at 284.9 eV.
traces of Pt (<0.2 wt%) were detected in the SCAT matrix. On
the other hand, the three samples of Fe/SCAT catalysts con-
tained 7.5, 3.4 and 1.8 wt% Fe (for a detailed list of elemental
analysis, please see the ESI†). Remarkably, the support of iron
nanoparticles on SCATs was achieved with an important loss
of iron of ∼40–50 wt% during the preparation phase, in
accordance with the published procedure. After preliminary
studies, described below, the sample 7.5 wt% Fe/SCATs was
selected as the best one, and will henceforth be referred to as
2
.5 Catalytic experiments
The catalytic tests were performed using the continuous appar-
atus described previously. Before starting the reaction, the
system was rinsed with a solution of 20 mM (+)-citronellal (or
3
4
2
0 mM (±)-citronellal or 20 mM (−)-isopulegol) and 20 mM
octane (internal standard) in toluene at a flow rate of 1 mL
−
1
min for 20′. The GC analysis of the outline solution con-
“Fe/SCATs” in this article. No leaching of the employed cata-
firmed the cleanliness of the system. Sequentially, the reaction
lysts was detected by ICP-MS and MP-AES analyses after
utilisation.
As shown in XRD patterns in Fig. 3, some diffraction lines
corresponding to silica, one of the major components of the
catalytic converter, were observed in both SCATs and Fe/SCATs
before and after utilisation.
conditions were set. H was generated by in situ water electroly-
2
sis using H-Cube® equipment. The reactions were performed
for 2 h, collecting samples every 15′ for further analysis. The
data were reported after reaching stationary state conditions.
Time “0” was set up as the first outcome under operative reac-
tion conditions passed through the apparatus.
More in detail, the most intense peaks of Fe/SCATs and
SCATs at 2θ of 21.72°, 28.49° and 54.67° were attributed to the
The conversion and selectivity were determined by gas
chromatography (GC) on an Agilent 6890N gas chromatograph
1
(
1 0 0), (0 1 1), and (2 0 2) planes of SiO with a hexagonal
2
(
60 mL min−
N
2
carrier flow, 1.38 bar (20 psi) column top
3
9
structure (JCPDS 00-011-0252). No specific peaks assignable
to iron oxides were clearly observed in the XRD patterns of Fe/
SCATs. Also, no relevant changes in the XRD patterns before
and after utilisation of the catalysts were detected.
head pressure) using a flame ionization detector (FID). A
Restek Rt®-yDEXsa chiral capillary column (30 m × 0.25 mm ×
0
.25 μm) was employed. The calibration curve was obtained
using octane as the internal standard. Standard solutions of
+)-citronellal, (−)-isopulegol and (−)-menthol (from 5 to
0 mM) and 20 mM octane in toluene were analysed by GC to
High-angle annular dark-field scanning transmission elec-
tron microscopy (HAADF-STEM) and Energy-dispersive X-ray
Spectroscopy (EDS) of SCAT and Fe/SCAT materials were sub-
sequently conducted to elucidate the location of metal (oxide)
nanoparticles. As depicted in Fig. 4A and B (SCATs), the pres-
ence of laminar smashed structures of silica and alumina
derived from the honeycomb structure of the catalytic conver-
ter could be observed.
The same laminar structure could also be visualised homo-
geneously covered by iron in Fe/SCATs (Fig. 4D and E). With
EDS analysis (Fig. 4C and H), selected surface areas (L1 and L2
in Fig. 1B, L3 in Fig. 4E and L5 in Fig. 4F) could be associated
with Pt and Ce in both SCATs and Fe/SCATs, proving that the
MW treatment did not influence the morphology of the sup-
(
2
2
give linear regressions (R > 0.999). Gas chromatography-mass
spectroscopy (GC-MS) analysis was also performed using an
Agilent 7820A GC/5977B High Efficiency Source (HES) MSD, in
order to identify the obtained products in comparison with
commercial standards.
The optical rotation was measured using a Zuzi Polarimeter
Model 412 (20 °C, 589.44 nm wavelength).
3
3
. Results and discussion
.1 Characterization of the catalysts
The structure of the SCATs was reported to be composed of a porting material. Other selected surface areas could be associ-
3
5
sequence of coats, made of alumina, silica, titania and ceria,
and offers the possibility to support metal nanoparticles, as
3
6
demonstrated in recent work. Iron oxide nanoparticles were
then supported on SCATs adapting the microwave-assisted
3
4
methodology previously described. Washed SCATs (Cartridge
in Fig. 2B) were directly employed as catalysts for the hydro-
2
genation step, in order to exploit the presence of trace quan-
tities of noble metals in the matrix, potentially highly active
3
7,38
for hydrogenation reactions.
In order to evaluate the leaching effects and the change in
the oxidation states of the metals or morphology variations,
HAADF-STEM, MP-AES (in order to check the metal content
and metal leaching), XPS and XRD analyses were performed
both before and after the utilisation of the catalysts.
Fig. 3 XRD patterns for SCATs and Fe/SCATs.
Green Chem.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Green Chemistry
Paper
high-binding energy could be assigned to iron oxide Fe(III) and
4
0,41
the low-binding energy ones to Fe(II).
3.2 Selecting the iron content in Fe/SCATs
Firstly, the three different Fe/SCAT samples (denoted as
samples A, B and C, previously reported in Table 1) were tested
in the one-pot cyclisation–hydrogenation of (±)-citronellal to
menthols (analysing the diastereoisomers together) in order to
select the most active catalyst, as illustrated in Scheme 2. For
this purpose, only the H-Cube® Mini Plus was employed (as
illustrated in Fig. S1C in the ESI†). The catalysts (150 mg) were
charged in a 30 mm-long cartridge. The reactions were run
using 20 mM (±)-citronellal in toluene, at 30 bar H pressure
2
and at 373 K, with a flow rate of 0.1 mL min− . As illustrated in
Table 3, entry 4, the best performance, specifically ∼95% con-
version of (±)-citronellal with ∼53% selectivity to menthols and
1
∼40% selectivity to isopulegols, was achieved employing
Fig. 4 (A) and (B) STEM-HAADF images of SCATs. (C) EDS analysis of sample A containing 7.5 wt% Fe.
selected locations (L1 and L2) of SCATs. (D), (E), (F) and (G) STEM-HAADF
images of Fe/SCATs. (H) EDS analysis of selected locations (L3-L6) of Fe/
SCATs.
As a result, sample A was selected for the following tests.
Higher contents of iron were discharged due to the complexity
in fixing the iron nanoparticles on the supporting material
which would have had required much effort, not necessary
against the high conversion already obtained (considering that
ated with Fe in sample Fe/SCATs (L4 in Fig. 4F and L6 in iron nanoparticles catalyse only the cyclisation step).
Fig. 4G). Homogenously distributed iron oxide nanoparticles Remarkably, Fe/SCATs were able to catalyse both the cyclisa-
of an average size of 2.4 ± 0.4 nm with a well-defined crystal- tion of citronellal and the sequential hydrogenation of isopule-
line structure could be clearly observed in Fe/SCATs (Fig. 4I). gols (Table 3, entries 2–4). SCATs also showed some activity
No relevant changes were observed by HAADF-STEM and EDS (Table 3, entry 1), but the low conversion of only 13% at 373 K
of SCATs and Fe/SCATs after utilisation (please see Fig. S2 in demonstrated the need for two cartridges.
the ESI† for the HAADF-STEM images and EDS analysis).
3.3 Preliminary studies
XPS analysis (Fig. 5) evidenced the presence of two intense
peaks at binding energies of ca. 712.7 eV and 726.3 eV, which Sequentially, the tandem system was set up (please see Fig. S1-
confirmed the presence of an iron oxide phase, as reported by A and S1B in the ESI†). The reactions were run fixing Cartridge
3
4
Moores et al. The fitting of multiplets and satellites with 1 (150 mg of Fe/SCATs, 30 mm-long cartridge) to the
Fig. 5 XPS patterns and data of (A) Fe/SCATs before and (B) after the reaction.
This journal is © The Royal Society of Chemistry 2019
Green Chem.

View Article Online
Paper
Green Chemistry
Scheme 2 Cyclisation of citronellal to isopulegols catalysed by Lewis acid and hydrogenation of isopulegols to menthols. (A) Most relevant side
42
products derived from the hydrogenation of citronellal. (B) Proposed catalytic mechanism adapted from Sakamoto et al.
Table 3 One-pot cyclisation–hydrogenation of (±)-citronellal to menthols catalysed by Fe/SCATs
Entry Sample
Conversion %
9.6
S
isopulegols
%
S
menthols
%
S
3,7-dimethyloctanol
%
S
3,7-dimethyloctanal
%
S
citronellol
%
menthols
Y %
1
2
3
4
SCATs
38.5
60.2
44.8
40.0
9.6
39.4
0.8
0.6
1.3
0
12.4
1.5
1.4
1.0
0.9
C (1.8 wt% Fe/SCATs) 66.2
B (3.4 wt% Fe/SCATs) 84.5
A (7.5 wt% Fe/SCATs) 94.4
36.0
52.0
56.1
1.6
1.3
1.6
23.8
43.9
52.9
maximum operative temperature of the H-Cube® Mini Plus Cartridge 2 was discharged as it gave almost the same yield to
373 K), while the temperature of Cartridge 2 (450 mg of menthols.
SCATs, 70 mm-long cartridge) and the flow rate were varied. A quick study varying the flow rate confirmed that the best
The reactions were carried out at 30 bar H pressure. As sum- operative conditions were obtained at 0.1 mL min (please
(
−
1
2
−
1
marized in Fig. 6, at a flow rate of 0.1 mL min , the conver- see Table S1, entries 5,7 and 8 in the ESI†).
sion of citronellal improved from 74.8% up to ∼100% and the
selectivity to menthols enhanced from 65.2% to 83.1% by only
increasing the reaction temperature in Cartridge 2 (for a com-
plete list of the trials please see Table S1, entries 1–6 in the
ESI†). The best yield of menthol (∼83%) was obtained at 373 K
in Cartridge 1 and at 423 K in Cartridge 2. A higher T in
3
.4 Optimising the reaction
In the second phase, in order to maximise the cyclisation step,
a 70 mm-long cartridge was employed as Cartridge 1. The best
operative conditions of the preliminary study were employed
as starting parameters for the optimisation phase. The influ-
ence of the reaction parameters of T (of Cartridge 2), pressure
and flow rate on (±)-menthol yields was determined. As shown
in Table 4, an improvement in the yields by increasing the
temperatures could be observed, reaching a plateau at 413 K
(
Table 4, entries 4–6). The results confirmed the data obtained
in the preliminary study.
Table 4 Tandem cyclisation–hydrogenation of (±)-citronellal to
(
±)-menthol. Reaction parameters: 20 mM citronellal in toluene, 30 bar
H
2
pressure, and 0.1 mL min− flow rate
1
a
b
Entry
T /K
Y
ms /%
S
(±)-m/%
Y(±)-m/%
1
2
3
4
5
6
348
373
398
413
423
448
76.7
86.3
87.6
91.1
89.8
92.0
81.1
80.7
79.7
79.5
80.6
77.8
62.2
69.7
69.8
72.4
72.4
71.5
a
Fig. 6 Performances of the tandem reaction carried out using 30 mm-
long Cartridge 1 and 70 mm-long Cartridge 2.
Temperature of Cartridge 1 fixed at 373 K – temperature of Cartridge
b
2 reported in the table. Yield % to menthols in the stationary state.
Green Chem.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Green Chemistry
Paper
Sequentially, the temperature of Cartridge 2 was set at the reduction of the activity was due to the adsorption of reac-
4
13 K (the minimum value of temperature of the plateau Y(±)-m tants/materials on the active surface of the catalysts rather
2
vs. T of Table 4, entries 4–6) and the H pressure of the system than deactivation. Almost identical yields were observed in the
was varied. The best performance was achieved at 5 bar H
pressure (Table 5, entry 4).
2
second cycle, most likely confirming that the slight drop in
activity after 72 h could be due to the adsorbed materials on
H
2
pressure was then set at 5 bar and the flow rate was sub- the catalysts. The ICP-MS analysis of the collected outcome
sequently optimized. As expected, a linear reduction in yields liquid was also performed detecting no traces of iron or other
−
1
−1
from 77.3% at 0.1 mL min to 30.4% at 0.5 mL min was metals, proving that no metal leaching occurred even in long
noticed (please see Table S2 in the ESI† for the complete list term experiments. In order to confirm the efficiency of the pro-
of experiments).
posed catalytic system, optimum reaction conditions were
A long-term stability analysis was eventually performed to selected to produce (−)-menthol starting from enantiomeri-
demonstrate the stability of the tandem catalytic system under cally pure (+)-citronellal. The final yield to (−)-menthol of
the investigated optimum reaction conditions (Fig. 7). Despite 77.1% (side products include isomers of menthols) at 373 K
that almost no change in the yields to (±)-menthol in the first (Cartridge 1, cyclisation step), 413 K (Cartridge 2, hydrogen-
pressure and 0.1 mL min− flow rate was
1
7
h was noticed, a decrease of the yield to (±)-menthol (down ation step), 5 bar H
2
to 64.4% from an initial value of 75.6%) was observed after calculated by GC analysis on a chiral column and confirmed
4 h, and a plateau was noticed up to 72 h of reaction. the versatility of the proposed system and the potential to
A washing cycle was subsequently performed by pumping extend the protocol to multiple tandem flow reactions. The
2
toluene in the continuous flow apparatus in order to check if menthols (having a high concentration of (−)-menthol) were
sequentially isolated by distillation (in order to remove the
solvent and low boiling substances) and chromatographic puri-
fication on a silica column. The optical rotation of the isolated
±)-menthol. Reaction parameters: 20 mM citronellal in toluene, menthols, derived from the optical rotation of (−)-menthol
Table 5 Tandem cyclisation–hydrogenation of (±)-citronellal to
(
−1
20
43
20
Cartridge 1 at 373 K, Cartridge 2 at 413 K, and 0.1 mL min flow rate
(
[α] = −50) and its isomers, was found to be [α]D = −38.
D
In order to validate the proposed system in comparison
with a conventional supporting material and with commer-
cially available catalysts, the best reaction conditions were uti-
lised using Fe/SiO and commercial 5 wt% Pt/C and 5 wt% Pd/
2
C as catalysts, as illustrated in Table 6.
a
b
c
Entry
H
2
p/bar
Y
ms /%
S
(±)-m /%
Y
(±)-m /%
1
2
3
4
5
6
30
15
10
5
2
0
91.1
90.1
90.4
91.8
89.4
—
79.5
82.2
82.9
84.2
80.9
—
72.4
74.0
74.9
77.3
72.4
—
Entries 2–4 demonstrate that the best performances were
achieved using a mixture of Fe(II) and Fe(III) as the iron precur-
a
b
Yield
%
to menthols in the stationary state. Selectivity to sor (that is the case of sample D, entry 2). This behaviour was
c
34
(
±)-menthol. Yield to (±)-menthol.
also observed by Moores et al. The slightly increased activity
of Fe/SCATs (A) compared to that of sample D (having the
same iron content but supported on SiO ), reported in entry 1,
2
can be explained in terms of higher stability of the iron nano-
particles over the smashed catalytic converters, as recently
3
6
demonstrated by supporting Ru over SCATs. In addition, as
reported in Table 3 (entry 1), SCATs themselves showed some
activity in the cyclization of citronellal to isopulegols, thereby
increasing the final yield to menthols, whereas silica does not
have catalytic activity.
Table 6 Tandem cyclisation–hydrogenation of (±)-citronellal to
(
±)-menthol. Reaction parameters: 20 mM citronellal in toluene,
−1
Cartridge 1 at 373 K, Cartridge 2 at 413 K, 0.1 mL min flow rate, and 5
2
bar H pressure
a
b
Entry
Cartridge 1
Fe/SCATs (A)
Cartridge 1
Y
ms /%
Y
(±)-m /%
1
2
3
4
5
6
SCATs
SCATs
SCATs
SCATs
5% Pd/C
5% Pt/C
91.8
86.4
69.1
61.4
88.1
88.9
77.3
70.2
50.4
40.3
70.3
82.3
Fe/SiO
Fe/SiO
Fe/SiO
2
2
2
(D)
(E)
(F)
Fig. 7 Selectivity and yields to (±)-menthol in the tandem cyclisation–
hydrogenation of (±)-citronellal. Reaction parameters: 20 mM citronellal
Fe/SCATs (A)
Fe/SCATs (A)
−1
in toluene, Cartridge 1 at 373 K, Cartridge 2 at 413 K, 0.1 mL min flow
nd
2
rate, and 5 bar H pressure. The 2 cycle was performed after washing
a
b
the continuous flow apparatus with toluene for 20’.
Yield % to menthols in the stationary state. Yield to (±)-menthol.
This journal is © The Royal Society of Chemistry 2019
Green Chem.

View Article Online
Paper
Green Chemistry
7
Y. H. Luo, W. P. Sun, X. J. Feng, X. Y. Ba, T. Liu, J. Guo,
L. Z. Xiao, J. Jiang, Y. Hao, D. L. Xiong and C. Y. Jiang,
Biochem. Biophys. Res. Commun., 2019, 516, 825–830.
4
. Conclusions
In conclusion, a continuous flow tandem system for the syn-
thesis of (−)-menthol starting from (+)-citronellal using low
cost waste-derived catalysts was proposed for the first time.
The efficiency of the instrumental setup in both steps of the
reaction was proved under different reaction conditions, high-
8 J. Hartner and U. M. Reinscheid, J. Mol. Struct., 2008, 872,
145–149.
9 J. C. Leffingwell and R. E. Shackelford, Cosmet. Perfum.,
1974, 89(6), 69–89.
lighting the influence of flow parameters on the final yield to 10 R. Eccles, D. H. Griffiths, C. G. Newton and N. S. Tolley,
menthols. The catalytic system was also found to be highly Clin. Otolaryngol., 1988, 13, 25–29.
robust as demonstrated in long-term stability tests, with no 11 B. Etzold and A. Jess, Chem. Eng. Technol., 2008, 31, 1282–
metal leaching nor change in the morphology or the oxidation 1289.
state of the metal. Under optimized conditions, (−)-menthol 12 B. Etzold, A. Jess and M. Nobis, Catal. Today, 2009, 140,
(
(
∼77% yield) could be produced from enantiomerically pure
+)-citronellal. The high versatility of the proposed system com- 13 A. Stolle, T. Gallert, C. Schmoeger and B. Ondruschka, RSC
Adv., 2013, 3, 2112–2153.
additional chemistries under continuous flow conditions 14 M. Emura and H. Matsuda, Chem. Biodivers., 2014, 11,
30–36.
bined with its efficiency will pave the way to a number of
which will be reported in due course.
1688–1699.
15 P. Kocovsky, G. Ahmed, J. Srogl, A. V. Malkov and J. Steele,
J. Org. Chem., 1999, 64, 2765–2775.
16 T. Iwata, Y. Okeda and Y. Hori, Takasago International
Corporation, US Patent, 6774269B2, 2004.
Conflicts of interest
The authors declare no conflicts of interest.
17 S. Telalovic, A. Ramanathan, J. F. Ng, R. Maheswari,
C. Kwakernaak, F. Soulimani, H. C. Brouwer, G. K. Chuah,
B. M. Weckhuysen and U. Hanefeld, Chem. – Eur. J., 2011,
1
7, 2077–2088.
Acknowledgements
1
8 P. R. S. Braga, A. A. Costa, E. F. de Freitas, R. O. Rocha,
J. L. de Macedo, A. S. Araujo, J. A. Dias and S. C. L. Dias,
J. Mol. Catal. A: Chem., 2012, 358, 99–105.
The authors thank Mr Rafael Ángel Gómez Haro and Provaluta
España Reciclaje de Metales, S. L., Córdoba (ES), for the
supply of scrap automotive catalytic converters. Rafael Luque 19 C. Milone, C. Gangemi, S. Minico, S. Galvagno and G. Neri,
gratefully acknowledges the funding from the MINECO under Chem. Ind., 1998, 75, 571–576.
project CTQ2016-78289-P, co-financed by the FEDER and the 20 Y. T. Nie, W. Niah, S. Jaenicke and G. K. Chuah, J. Catal.,
contract for Camilla Maria Cova associated with this project. 2007, 248, 1–10.
This project received funding from European Union’s Horizon 21 F. Neatu, S. Coman, V. I. Parvulescu, G. Poncelet, D. De Vos
020 research and innovation programme under Marie and P. Jacobs, Top. Catal., 2009, 52, 1292–1300.
Skłodowska-Curie grant agreement no. 721290. This publi- 22 Y. T. Nie, S. Jaenicke and G. K. Chuah, Chem. – Eur. J.,
cation reflects only the author’s view, exempting the 2009, 15, 1991–1999.
Community from any liability. Project website: http://cosmic- 23 A. Negoi, S. Wuttke, E. Kemnitz, D. Macovei,
2
etn.eu/. HRSTEM studies were conducted at the Laboratorio de
Microscopias Avanzadas, Instituto de Nanociencia de Aragon,
V. I. Parvulescu, C. M. Teodorescu and S. M. Coman,
Angew. Chem., Int. Ed., 2010, 49, 8134–8138.
Universidad de Zaragoza, Spain. The publication was prepared 24 J. Plosser, F. Dedeaga, M. Lucas and P. Claus, Appl. Catal.,
with support from RUDN University program 5-100.
A, 2016, 516, 100–108.
2
2
5 C. Moreno-Marrodan, F. Liguori, P. Barbaro, S. Caporali,
L. Merlo and C. Oldani, ChemCatChem, 2017, 9, 4256–4267.
6 J. ten Dam, A. Ramanathan, K. Djanashvili, F. Kapteijn and
U. Hanefeld, RSC Adv., 2017, 7, 12041–12053.
References
1
https://www.basf.com/us/en/media/science-around-us/the-cool- 27 G. Jas and A. Kirschning, Chem. – Eur. J., 2003, 9, 5708–
freshness-of-menthol.html. 5723.
G. P. P. Kamatou, I. Vermaak, A. M. Viljoen and 28 J. Wegner, S. Ceylan and A. Kirschning, Adv. Synth. Catal.,
2
B. M. Lawrence, Phytochemistry, 2013, 96, 15–25.
R. Eccles, J. Pharm. Pharmacol., 1994, 46, 618–630.
S. J. Anderson, Tob. Control, 2011, 20, 1120–1128.
S. Kizil, N. Hasimi, V. Tolan, E. Kilinc and U. Yuksel,
Turk. J. Field Crops, 2010, 15, 148–153.
2012, 354, 17–57.
3
4
5
29 F. Gomollón-Bel, Chem. Int., 2019, 41, 12–16.
30 C. H. Kim, S. I. Woo and S. H. Jeon, Ind. Eng. Chem. Res.,
2000, 39, 1185–1192.
31 M. Zengin, H. Genc, T. Demirci, M. Arslan and
M. Kucukislamoglu, Tetrahedron Lett., 2011, 52, 2333–2335.
6
L. J. Macpherson, S. W. Hwang, T. Miyamoto, A. E. Dubin,
A. Patapoutian and G. M. Story, Mol. Cell. Neurosci., 2006, 32 E. Mieczynska, A. Gniewek and A. M. Trzeciak, Appl. Catal.,
2, 335–343. A, 2012, 421, 148–153.
3
Green Chem.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Green Chemistry
Paper
3
3
3 H. Genc, Catal. Commun., 2015, 67, 64–67.
4 A. Y. Li, M. Kaushik, C. J. Li and A. Moores, ACS Sustainable
Chem. Eng., 2016, 4, 965–973.
39 W. A. Bassett and D. M. Lapham, Am. Mineral., 1957, 42,
548–555.
40 M. Omran, T. Fabritius, A. M. Elmandy, N. A. Abdel-Khalek,
M. El-Aref and A. Elmanawi, Appl. Surf. Sci., 2015, 345, 127–
140.
3
3
3
3
5 J. Kaspar, P. Fornasiero and N. Hickey, Catal. Today, 2003,
7
7, 419–449.
6 C. M. Cova, A. Zuliani, M. J. Muñoz-Batista and R. Luque, 41 I. Fanlo, F. Gervilla, E. Mateo and S. Irusta, Eur. J. Mineral.,
Green Chem., 2019, 21, 4712–4722. 2008, 20, 125–129.
7 P. Gallezot and D. Richard, Catal. Rev., 1998, 40, 42 H. Itoh, H. Maeda, S. Yamada, Y. Hori, T. Mino and
1–126. M. Sakamoto, Org. Chem. Front., 2014, 1, 1107–1115.
8 A. Gutierrez, R. K. Kaila, M. L. Honkela, R. Slioor and 43 F. Reinscheid and U. M. Reinscheid, J. Mol. Struct., 2016,
A. O. I. Krause, Catal. Today, 2009, 147, 239–246. 1103, 166–176.
8
This journal is © The Royal Society of Chemistry 2019
Green Chem.