Citral-to-Menthol Transformations in a Continuous Reactor over Ni/
Mesoporous Aluminosilicate Extrudates Containing a Sepiolite Clay
Binder
Irina L. Simakova, Zuzana Vajglová, Päivi Mäki-Arvela, Kari Eränen, Leena Hupa, Markus Peurla,
Ermei M. Mäkilä, Johan Wärnå, and Dmitry Yu. Murzin*
Cite This: Org. Process Res. Dev. 2022, 26, 387−403 Read Online
ACCESS Metrics & More Article Recommendations *sı Supporting Information
ABSTRACT: One-pot continuous synthesis of menthols from citral was performed over 5 wt % Ni supported on a mesoporous
aluminosilicate catalyst with sepiolite as a binder at 70 °C with a selectivity of 75% to menthols. Catalyst deactivation with time-on-
stream resulted in a decrease of the conversion and selectivity to menthols at the expense of higher selectivity to isopulegols.
Stereoselectivity to isopulegols and menthols only slightly changed with conversion and TOS. A kinetic model capable of describing
experimental data for transformations of citral to menthol in a continuous mode was developed. It was based on a detailed reaction
network and also comprised deactivation on both metal and acid sites. Numerical data fitting confirmed a good correspondence
between the experimental data and calculations.
KEYWORDS: citral, menthol, shaped Ni catalyst, clay binder, trickle-bed reactor
■ INTRODUCTION
Production of menthol, a compound with local anesthetic and
counterirritant properties, acting also as a weak k-opiod receptor
agonist, has attracted significant attention because of its
physiological behavior. Menthol, occurring naturally in
peppermint oil, can be manufactured by freezing this oil
followed by filtration of menthol crystals. Substantial amounts of
menthol are also produced by synthetic routes starting from
renewable sources. Manufacturing of menthol from myrcene
starts with formation of an allylic amine, followed by asymmetric
isomerization using a BINAP rhodium complex giving
enantiomerically pure cintronellal after hydrolysis.1 Subsequent
steps include cyclization to isopulegol and hydrogenation to
menthol. This route has been commercialized by Takasago
International Corporation.2 An alternative route involves m-
cresol, which is alkylated with propene to form thymol.3 The
latter is hydrogenated over heterogeneous catalysts resulting in a
mixture of menthol enantiomers,4−8 which are separated by
chiral resolution. A particular interesting option is the one-pot
synthesis of menthol starting from another renewable source,
citral. The latter is obtained by distillation of essential oils.
As can be seen from Figure 1, citral comprises three functional
groups, which can be hydrogenated on metal sites. Selective
hydrogenation of the double bond adjacent to the carbonyl
groups gives cintronellal9,10 on the metal sites. This reaction
should be selective not resulting in hydrogenation of either the
second double bond or the carbonyl group to citronellol10 and
further to 3,7-dimethyloctanol. Cyclization of citronellal to
isopulegol requires an acidic catalyst.11,12 The final step in this
reaction path is hydrogenation of isopulegol.13
Several reports are available in the literature on one-pot citral
transformations in batch reactors.14−20 High yields of menthol
(94%) were obtained starting from citral at 70 °C and 20 bar
hydrogen over 8 wt % Ni/Al-H-MCM-41 and 15 wt % Ni-
MCM-41-Zr-β.14,16 Ma ̈ki-Arvela et al. reported that the
selectivity of 54% to four stereoisomers of menthols was
achieved at 100% citral conversion over 15 wt %Ni-H-MCM-41
under 10 bar total pressure at 70 °C in cyclohexane.19 The
stereoselective ratio of (±)-menthols/(±)-neomenthols/
(±)-isomenthols was 71:25:4% with the absence of (±)-neo-
isomenthols in the reaction mixture.19 Noble metals, such as Pd,
Pt, or Ru, have also been applied in a quest for a delicate balance
between hydrogenation ability of the catalyst and its acid-
ity.15−20 These active metals could lead to overhydrogenation,
while too acidic catalysts can give rise to dehydration and
defunctionalized products. In addition to nickel, other nonnoble
metals, including, for example, copper,21 supported on a natural
ma t e r i a l s e p i o l i t e o f t h e f o l l ow i n g s t r u c t u r e
Si12O30Mg8(OH)4(OH2)48H2O,
22 gave 45% menthol yield in
heptane at 90 °C under 1 bar hydrogen in a batch reactor.21
In our recent work, we have reported mildly acidic
mesoporous Ru-MCM-41 with a relatively large size of
ruthenium clusters for one-pot synthesis of menthol from citral
in batch and continuous reactors.20 Such a large size (7−20 nm)
was intended to diminish the hydrogenation activity. The
catalyst along with microporous Ru-H-Y-80 was not very
Received: November 16, 2021
Published: January 25, 2022
Articlepubs.acs.org/OPRD
© 2022 The Authors. Published by
American Chemical Society
387
https://doi.org/10.1021/acs.oprd.1c00435
Org. Process Res. Dev. 2022, 26, 387−403
D
ow
nl
oa
de
d 
vi
a 
U
N
IV
 O
F 
TU
RK
U
 o
n 
A
pr
il 
22
, 2
02
2 
at
 0
5:
59
:2
5 
(U
TC
).
Se
e 
ht
tp
s:/
/p
ub
s.a
cs
.o
rg
/sh
ar
in
gg
ui
de
lin
es
 fo
r o
pt
io
ns
 o
n 
ho
w
 to
 le
gi
tim
at
el
y 
sh
ar
e 
pu
bl
ish
ed
 a
rti
cl
es
.
selective toward menthol, giving defunctionalized mentha-
trienes as the main products.23
Continuous menthol production from citronellal requires
utilization of either shaped catalyst bodies or structured reactors
to avoid high pressure drop, inevitable with powder catalysts.24
In a series of studies, various aspects of catalyst shaping by their
extrusion and utilization for menthol production have been
addressed, including selection of the binder (e.g., bentonite or
Ludox) and the metal location using Ru as the main
metal.20,23,25−27 Based on this previous experience, sepiolite
was selected as the binder, Ni as a metal for hydrogenation, and a
mildly acidic mesoporous aluminosilicate as a metal support and
simultaneously, an acid catalyst for citronellal cyclization. Ni/
MCM-41 catalysts have been previously demonstrated as rather
selective for one-pot synthesis of menthol from citral in a batch
reactor.14,16,17,19 It should, however, be mentioned that catalyst
shaping with an inorganic binder, which itself can be selective,
for example, in cyclization, can influence catalytic performance
of the final extrudate influencing not only activity, but more
importantly selectivity to the desired products.
As mentioned above, there are only a handful of papers in the
literature where one-pot continuous transformations of citral to
menthol have been reported. None of them consider any kinetic
analysis or discuss catalyst deactivation in a quantitative way. In
fact, the open literature is devoid of kinetic analysis of such one-
pot transformations even for experimental data generated in a
batch mode. Thus, the aim of this work was not only to study
performance of a nickel catalyst supported on a mesoporous
aluminosilicate in a continuous mode but also to develop a
kinetic model of this process, which will include deactivation of
the bifunctional catalyst.
■ RESULTS AND DISCUSSION
Characterization of Ni Catalysts. Nickel supported on
mesoporous aluminosilicate catalysts was characterized before
and after the reaction by inductively coupled plasma-optical
emission spectrometry (ICP-OES), transmission electron
microscopy (TEM), nitrogen physisorption technique, X-ray
diffraction method (XRD), scanning electron microscopy with
energy-dispersive X-ray microanalysis (SEM-EDX), Fourier
transform infrared spectroscopy (FTIR), and crush testing.
Nickel loading and particle size as well as some textural data
for Ni supported on mesoporous aluminosilicate catalysts are
presented in Table 1. Isotherms and pore size distribution of
catalysts are shown in Figures S1 and S2.
According to ICP-OES analysis (Table 1) of the fresh reduced
and spent Ni catalysts, nickel loading decreased during citral
transformations regardless of the operation mode (batch or
continuous). The nickel content in the powder Ni/MAS catalyst
decreased from 4.76 to 4.55%, while for the extruded catalyst it
changed from 6.30 to 5.95%, supposedly due to insignificant
metal leaching during citral conversion (Table 1).
The sorption isotherm of catalysts is of type IV according to
IUPAC classification (Figure S1a). Characteristic features of the
Type IV isotherm are its hysteresis loop, which is associated with
capillary condensation taking place in mesopores and the
limiting uptake over a range of high p/p°.28 A type H2 hysteresis
loop indicated that the samples are mesoporous materials with
cage-like mesopores formed due to disordered blocked pore
shapes.28 Prior to the reaction, the fresh reduced powder Ni/
MAS catalyst exhibited a specific surface area (SBET) of 447m
2/g
and total pore volume (V) of 0.47 cm3/g (Table 1). After citral
conversion, the surface area and pore volume of the catalyst
decreased to 280 m2/g (SBET) and 0.36 cm
3/g, respectively
(Table 1 and Figure S1a). The same trend was also noticed for
Figure 1. Transformations of citral to menthol.
Table 1. Concentration and Particle Size of Nickel and Textural Properties of the Catalystsa
catalyst [Ni] (wt %) dNi (XRD) (nm) dNi (TEM) (nm) SBET (m
2/g) SD−R (m
2/g) V (cm2/g) Vm (%)
fresh, reduced powder Ni/MAS 4.76 8 447 412 0.47 71.1
spent, powder Ni/MAS 4.55 13 280 317 0.36 66.3
fresh, reduced crushed extrudates Ni/(MAS + sepiolite) 6.30 12 743 756 0.85 71.0
fresh, reduced extrudates Ni/(MAS + sepiolite) 337 329 0.39 75.5
spent, extrudates Ni/(MAS + sepiolite) 5.95 11 14 149 214 0.31 81.0
a[Ni]concentration of Ni determined by ICP-OES, dNi (XRD)average particle size of Ni determined by XRD, dNi (TEM)median particle
size of Ni determined by TEM, Sspecific surface area calculated by BET and Dubinin−Radushkevich equations, Vtotal pore volume (micro +
mesopore volume) calculated by the density functional theory (DFT) method, and Vmpercentage of the mesopore volume.
Organic Process Research & Development pubs.acs.org/OPRD Article
https://doi.org/10.1021/acs.oprd.1c00435
Org. Process Res. Dev. 2022, 26, 387−403
388
the extrudates during citral transformations in the continuous
reactor under the same reaction conditions giving a decrease of
the surface area from 337 to 149 m2/g (SBET) and pore volume
from 0.39 to 0.31 cm3/g (V) (Table 1 and Figure S1b). A lower
surface area and pore volume of Ni/(MAS + sepiolite)
extrudates compared to powder Ni/MAS catalysts can be
assigned to sepiolite that clogged pores of the pristine
aluminosilicate during extrusion. Interestingly, crushing of the
fresh reduced extrudates Ni/(MAS + sepiolite) significantly
increased the surface area from 337 to 743 m2/g (SBET),
indicating that the groud catalyst is able to expose much more
available sites for N2 molecule adsorption. All samples exhibited
more than 66% of the mesopore volume. After the reaction, the
fraction of the mesopore volume of extrudates increased;
however, the total pore volume decreased, which can be
explained by preferential blocking of micropores in line with
pore size distribution (Figure S2).
Pristine support along with the fresh reduced and spent
extrudates (Figure 2) as well as powdered Ni/MAS catalyst
(fresh calcined, fresh reduced, and spent samples) (Figure S3)
were studied by powder X-ray diffraction (XRD).
The observed lines indicated the presence of sepiolite phase
Mg4Si6O15(OH)2(H2O)6 (PDF # 04-014-4557) (Figure 2a).
XRD patterns of the studied materials at low angles showed that
the positions and intensities of the observed diffraction lines
correspond to the MCM-41 material (Figure 2b).29 The
Figure 2.XRD patterns of (a) fresh reduced extrudates, with sepiolite reflectionsmarked in red, (b) support (black) and freshly reducedNi-containing
(blue) and spent (red) extrudate samples at low angle, and (c) spent extrudates.
Organic Process Research & Development pubs.acs.org/OPRD Article
https://doi.org/10.1021/acs.oprd.1c00435
Org. Process Res. Dev. 2022, 26, 387−403
389
Figure 3. continued
Organic Process Research & Development pubs.acs.org/OPRD Article
https://doi.org/10.1021/acs.oprd.1c00435
Org. Process Res. Dev. 2022, 26, 387−403
390
interplanar distances d100 for the mesoporous support, freshly
reduced, and spent extrudates were respectively 32.4, 31.8, and
31.5 Å. The corresponding values of the hexagonal lattice
parameter defined as 2d100/√3 were 37.4, 36.7, and 36.4 Å. A
significant decrease in the reflections for the spent catalyst
compared to the MCM-41 phase can be associated with more
prominent structural disordering after the reaction (Figure 2b).
XRD patterns of the spent extrudates are presented in Figure 2c.
A l o n g w i t h t h e l i n e s f r om s e p i o l i t e p h a s e
Mg4Si6O15(OH)2(H2O)6, the lines from the metallic nickel
Ni0 phase (PDF No. 04-0850) are presented. The average
coherent scattering domain size (CSD) of Ni0 DNi is 11.0 nm
that is consistent with TEM data (Table 1 and Figure 3l). The
determined value of the Ni lattice parameter a = 3.524 Å. Note
Figure 3. TEM images and Ni particle size distribution: (a, b) Ni/MAS fresh nonreduced powder catalyst, (c, d) Ni/MAS fresh reduced powder
catalyst, (e, f) Ni/MAS spent powder catalyst, (g, h) Ni/(MAS + sepiolite) fresh nonreduced extrudates, (i, j) Ni/(MAS + sepiolite) fresh reduced
extrudates, and (k, l) Ni/(MAS + sepiolite) spent extrudates.
Organic Process Research & Development pubs.acs.org/OPRD Article
https://doi.org/10.1021/acs.oprd.1c00435
Org. Process Res. Dev. 2022, 26, 387−403
391
there are no additional lines in XRD patterns of the fresh
reduced and spent catalysts other than characteristic lines of
MCM-41, indicating stability of the support texture after
deposition of nickel and subsequent treatment as well as upon
exposure to the reaction media.
The irregular shape of Ni particles was observed for all Ni
catalysts by TEM analysis (Figure 3). The median Ni particle
size for the fresh and spent powder Ni/H-MCM-41 catalyst
without a binder was the same, 13 nm, with a broad range from
0.2 to 56 nm. TEM analysis of Ni extrudates clearly confirmed
that Ni was deposited randomly on both H-MCM-41 and the
sepiolite binder. The median Ni particle size was 9 nm for the
fresh extrudates. This value is smaller than that for the powder
catalyst without a binder and can be attributed to the impact of
extrusion per se and the additional thermal treatment after
shaping being similar to previous observations for Pt
extrudates.30 Such a behavior was explained by variations in
the location of metal particles on the support and the binder, and
the influence of the organic binder methylcellulose. The latter,
upon calcination, is removed acting as a reducing agent for metal
particles, thereby increasing the metal dispersion by forming
small metal particles.30 On the contrary, a slightly larger median
Ni particle size of 14 nm for the spent extrudates, compared to
the fresh extrudates, can be an indication of minor
agglomeration.
FTIR results for the Ni/MAS fresh nonreduced powder
catalyst and Ni/(MAS + sepiolite) fresh reduced extrudates are
reported in Table 2 together with the pristine H-MCM-4127
powder catalyst for comparison. A slightly higher value of weak
Lewis acid sites, 35 μmol/g, but lower total acid sites and lower
Table 2. Brønsted and Lewis Acid Sites of Ni/MAS Fresh Nonreduced Powder Catalysta
catalyst BAS LAS TAS B/L
w m s ∑ w m s ∑ μmol/g
H-MCM-4127 41 19 24 84 20 14 21 56 140 1.5
fresh, reduced powder Ni/MAS 18 25 9 51 35 13 7 55 106 0.9
fresh, reduced extrudates Ni/(MAS + sepiolite) 23 7 0 30 24 11 0 35 65 0.9
aBASBrønsted acid sites; LASLewis acid sites; TAStotal acid sites; sstrong acid sites, data at 450 °C; mmedium acid sites, data at 350
°C minus data at 450 °C; and wweak acid sites, data at 250 °C minus data at 350 °C; and B/Lratio of Brønsted-to-Lewis acid sites.
Figure 4. SEM images: (a) Ni/MAS fresh nonreduced powder catalyst, (b) Ni/MAS fresh reduced powder catalyst, (c) Ni/(MAS + sepiolite) fresh
nonreduced extrudates, (d) Ni/(MAS + sepiolite) fresh reduced extrudates, and (e) Ni/(MAS + sepiolite) spent extrudates.
Organic Process Research & Development pubs.acs.org/OPRD Article
https://doi.org/10.1021/acs.oprd.1c00435
Org. Process Res. Dev. 2022, 26, 387−403
392
Brønsted and Lewis acid sites ratio (B/L), 0.9, were obtained for
5 wt % Ni/MAS. When Ni is loaded in zeolites and mesoporous
catalysts, typically, the amount of strong acid Brønsted sites
decreases as confirmed by Kubicǩa et al.31 The same B/L ratio
and overall lower values for both Brønsted and Lewis acid sites
were obtained for Ni/(MAS + sepiolite) extrudates compared to
the powder catalyst without sepiolite binder.
SEM images of fresh powder catalysts, nonreduced (Figure
4a) and reduced (Figure 4b), are in line with the results from
nitrogen physisorption (Table 1) and clearly show structural
reordering after the reduction procedure (350 °C). The same
trend was observed in case of extrudates. The spent extrudates
exhibited a more compacted surface, which could be related to
additional structural reordering and the blockage of pores after
the reaction. The Si/Al weight ratio was 5.15 for both
nonreduced and reduced powder catalysts. A slightly higher
Si/Al weight ratio of 5.86 was detected for extrudates containing
the sepiolite binder, which is consistent with a lower acidity of
Ni/(MAS + sepiolite) compared to Ni/MAS detected by FTIR
analysis of pyridine adsorption (Table 2).
Themechanical strengths of extrudates were determined to be
32 ± 7 and 4 ± 1 bar in the vertical and horizontal positions,
respectively. These are relatively high compared to the
mechanical strengths of aluminosilicate extrudates32 calcined
at 600 and 800 °C, i.e., 4.2 and 25 bar, respectively. A higher
value for Ni/(MAS + sepiolite) extrudates in the current work
can be attributed to the sepiolite binder and the presence of the
metal in the extrudates. Pure sepiolite extrudates exhibited
mechanical strengths of 56 ± 6 and 35 ± 4 bar in vertical and
horizontal positions, respectively.33 For Pt/H-β-25 extrudates
containing SiO2 Bindzil binder or Bentonite binder, the
mechanical strengths were 29−45 and 32−48 bar in the vertical
position.30,34
Activity and Selectivity of the Powder Ni Catalyst in
the Batch Experiment. The catalytic results from a batch
experiment with 5 wt % Ni/MAS powder catalyst are displayed
in Figure 5a−c. Citral conversion (X) increased with time to
53% in 5 h, while the liquid mass balance closure (MB) slightly
decreased to 88%. Previously, Ni on MCM-4119 was used for
one-pot transformations of citral to menthol also in cyclohexane
at the same temperature and hydrogen pressure as in the current
work. The catalyst-to-substrate ratio was, however, much higher
(6-fold), resulting in complete conversion already after 2 h. In
addition, it should be pointed out here that Mak̈i-Arvela et al.19
Figure 5. (a) Citral conversion (X) and liquid phase mass balance closure (MB), (b) concentrations of different products in a batch reactor, and (c)
selectivity as a function of conversion. Legend: ACPacyclic hydrogenation products, IPsisopulegol isomers, MEsmenthol isomers, DFP
defunctionalization products, and DMdimeric ethers and heavy components. Conditions: 70 °C, 10 bar of H2, 0.086M initial concentration of citral
in cyclohexane, and 0.2 g of the Ni/MAS catalyst. Notation: (■) ACP, (●) IPs, (▲) MEs, (▼) DFP, and (⧫) DM.
Organic Process Research & Development pubs.acs.org/OPRD Article
https://doi.org/10.1021/acs.oprd.1c00435
Org. Process Res. Dev. 2022, 26, 387−403
393
reported both citral and solvent were preheated and saturated
with hydrogen prior to their injection into the reactor, which
contained the prereduced catalyst. This pretreatment ensured
rapid start up and also faster reaction, while in the current case,
prereduced catalyst, reactant, and solvent were loaded into the
reactor and the reaction was started when reaching the desired
Figure 6. Scheme of menthol synthesis from citral with potential side reactions. Reproduced from Vajglova ́ et al.20 with permission from the Royal
Society of Chemistry. Copyright [2021] [Royal Society of Chemistry].
Organic Process Research & Development pubs.acs.org/OPRD Article
https://doi.org/10.1021/acs.oprd.1c00435
Org. Process Res. Dev. 2022, 26, 387−403
394
temperature and pressure. Although the initial hydrogenolysis
rate over 5 wt % Ni-MCM-41 was negligible compared to the
initial hydrogenation rate, the selectivity to DFP products after
330 min at 100% conversion was 40%. However, the selectivity
to menthols was 54% at the same point and only traces of acyclic
hydrogenation products were formed.19 This catalyst exhibited
5.8 nm Ni particles and the support exhibited a BAS/LAS ratio
of 0.53.
In the current work, already during the first minute of the
experiment, citral was selectively hydrogenated to citronellal (S
= 96%), which was partially converted to isopulegols (S = 4%).
The concentration profiles for different product groups, i.e.,
acyclic hydrogenation products (ACP), pulegols (IPs),
menthols (MEs), and defunctionalized products (DFP) are
shown in Figure 5b. The scheme of citral transformation
demonstrating all potential reactions in detail is shown in Figure
6. The ratio between the initial rates for IPs, ACP, and DFP, i.e.,
r0,IPs/r0,ACP/r0,DFP, is 1:4.6:27.7, while menthols are only visible
at 300 min. This result indicates that hydrogenated products
were already formed during the heating period as a large amount
of citronellal was present at 1 min. Furthermore, a maximum
concentration of hydrogenation products was visible at 120 min
after which mainly citronellal reacted further to cyclic products.
It can also be stated that IPs andMEs were intermediates, which
reacted very rapidly to menthenes, while no hydrogenated
menthanes were formed. The selectivity to DFP increased with
increasing conversion reaching 78% after 5 h (Figure 5c). At the
same time, the selectivity to citronellal, which was the only
acyclic hydrogenation product, decreased. This result indicates
that hydrogenation of citral proceeded during 5 h reaction time;
however, its rate decreased because only 53% conversion was
achieved. No other hydrogenation products than citronellal
were formed over a bifunctional Ni catalyst, which agrees well
with data reported byMak̈i-Arvela et al.19 It is worth to note that
also a bifunctional Pd/MCM-41 extensively promoted the
formation of 3,7-dimethyloctanol.19 Extensive hydrogenolysis of
menthols occurred during 5 h; however, only unsaturated
menthene derivatives, i.e., α-terpinolene, mentha-2,8-diene, p-
menth-4(8)-ene, p-mentha-1,3,8-triene, p-mentha-1,5,8-triene,
and o-cymene, were formed. The high selectivity to DFP
products can be explained by a relatively high BAS/LAS ratio of
0.93 (Table 2) as well as larger Ni particle size, 8 nm, in
comparison to ref 19.
Selectivities to isopulegols and menthols after 5 h were only
11 and 6%, respectively (Figure 5c). Stereoselectivity to the
desired menthol isomer was 40%. This relatively low value could
be attributed to an overall low selectivity to menthols at the citral
conversion level achieved after 5 h of the reaction (i.e., 53%).
Stereoselectivity tomenthol was only 40% after 5 h reaction time
in a batch reactor, while in ref 19 it was 70%. This result might
also be explained by the slow reaction rate in comparison to ref
19. Typically, (±)-menthol exhibits stereoselectivity in the range
of 66−71%.19,27
Activity and Selectivity of Ni Extrudates in the
Continuous Operation Mode.One-pot citral transformation
experiments at 70 °C, 10 bar of hydrogen, and 0.4 mL/min of
feed in the trickle-bed reactor over Ni/(MAS+sepiolite)
extrudates demonstrated rather high citral conversion and
exceptionally high selectivity to menthols of ca. 75% (Figure 7).
The yield of menthols of 63% was previously reported over 8 wt
% Ni on heteropolyacid with a mesoporous support17 after 24 h
of the reaction at 80 °C and 10 bar of hydrogen. Lower yields
(35−50%) were reported for other studied catalysts. The
authors17 concluded that the presence of strong Lewis and
medium Brønsted acid sites is required for high menthol
selectivity. Note that for the Ru/MCM-41 catalyst20 also tested
in the form of extrudates with the Bindzil binder, mainly
defunctionalized products were obtained (ca. 25−30% yield)
and the highest yield of menthol was just 6% with a
stereoselectivity of 66%. Analogous to the previous work on
the ruthenium catalyst in a trickle-bed reactor, substantial
catalyst deactivation was seen resulting not only in a decrease of
the conversion but also significant selectivity changes with
increasing time-on-stream (TOS) as conversion was decreased
because of deactivation (Figure 7).
A decrease in the catalytic activity can be caused by Ni
leaching as well as by pore blocking (Table 1). The subsequent
changes in selectivity are in fact expected for a consecutive
reaction network.
The citral conversion and selectivity to the desired menthols
(MEs) decreased by 0.9 and 1.1% per hour of TOS, respectively.
On the contrary, selectivity to isopulegols (IPs) and the acyclic
hydrogenation products (ACP) during the same time increased
by 0.7 and 0.3%, respectively. Selectivity to the defunctionaliza-
tion products (DFP) was constant, not changing with
conversion, indicating that such products are formed in parallel
Figure 7.Transformations of citral in a continuous reactor: (a) citral conversion (X) and liquid phase mass balance closure (MB) and (b) selectivity as
a function of conversion. Legend: ACPacyclic hydrogenation products, IPsisopulegol isomers, MEsmenthol isomers, DFP
defunctionalization products, and DMdimeric ethers and heavy components. Conditions: 70 °C, 10 bar of H2, 0.4 mL/min of feed, 0.086 M
initial concentration of citral in cyclohexane, 1 g of the Ni/(MAS + sepiolite) catalyst, and 11.5 min of residence time.
Organic Process Research & Development pubs.acs.org/OPRD Article
https://doi.org/10.1021/acs.oprd.1c00435
Org. Process Res. Dev. 2022, 26, 387−403
395
from all reactants and products as illustrated in Figure 6. Neither
dimeric ethers nor heavy components (DM) were detected in
the reaction mixture. The carbon balance and liquid phase mass
balance closure (MB) were close to 100%. Note that much lower
mass balance closure was reported for one-pot citral trans-
formations over Ru/MCM-41 containing extrudates finally
reaching 60−80% depending on the type of catalyst.20
Moreover, even the activity decrease was more pronounced in
the case of Ru/MCM-41 as conversion dropped from ca. 90−
100 to 65−70% just within 3 h.20
As expected, detailed analysis of the acyclic hydrogenation
products (ACP, Figure 8) revealed an increased selectivity to
citronellal, formed in the first step from citral, with TOS at the
expense of the subsequent hydrogenation product 3,7-
dimethyloctan-1-ol. Other ACP (including 2,6-dimethyloctane)
were formed in minor quantities without a visible trend in TOS.
Stereoselectivity to isopulegol and menthol isomers as a
function of time-on-stream is displayed in Figures 9 and 10.
In both cases, stereoselectivity only slightly changed with
conversion and TOS. Among cyclization products, selectivity to
isopulegol and isoisopulegol increased, while selectivity to
neoisopulegol clearly decreased with TOS. In 50 h of TOS, the
ratio of isopulegol/neoisopulegol/isoisopulegol/neoisopulegol
changed from 54/46/0/0 to 61/32/7/0. Often, the distribution
of isopulegols follows the thermodynamic equilibrium,35,36
which means that stereoselectivity to (±)-isopulegol is ca. 70−
75%. At the same time, stereoselectivity to isopulegol depends
on acidity and over such Lewis acids as Zr-β or zinc chloride, it
can reach 90−94% in citronellal cyclization.35,37 Lower values
than thermodynamic ones have also been reported; thus, the
stereoselectivity distribution of ca. 55/ 35/10% for IP/ NIP/ IIP
was observed for ruthenium bearing extrudates with a random
Ru deposition on both H-MCM-41 and Bindzil in one-pot citral
transformations to menthol.20
Changes in the stereoselectivity distribution of isopulegols
were concomitant with the ratio of menthols; thus, the
formation of menthol and neomenthol followed selectivity
patterns for isopulegol and neoisopulegol. For menthol/
neomenthol/isomenthol/neoisomenthol, their ratio changed
from 66/26/2/6 to 71/15/0/14 in 50 h of TOS.
Stereoselectivity to the desired (±)-menthol of 68−70% was
reported previously for Ru/H-MCM-41 extrudates containing
the Bindzil binder,20 while somewhat lower stereoselectivity to
menthol (42%) was obtained for Ru/Y-80 extrudates when
Bindzil was used as a binder.23 Note that the total yield of
menthols in previous reports20,23 for continuous one-pot
transformations of citral was much lower (just between 2 and
6%) than in the current work.
Figure 11 shows citral conversion as a function of time-on-
stream at different temperatures (40−80 °C) and the same
pressure of hydrogen (10 bar), flowrate of feed (0.4 mL/min),
and initial citral concentration in cyclohexane (0.086 M) over 1
g of Ni-containing extrudates.
Figure 11 also demonstrates catalyst deactivation visible
through a decrease of the conversion with time-on-stream at
isothermal conditions as well as by the difference in the level of
conversion when the same temperature was tested after several
hours of operation.
Figure 12 displaying citral conversion as a function of time-on-
stream at different hydrogen pressures (2.5−10 bar) and the
same temperature (70 °C) shows almost zero-order dependence
in hydrogen for hydrogenation of citral. Somewhat more
prominent deactivation was observed at a lower hydrogen
pressure that could be probably related to slower recovery of
metallic nickel at a lower H2 pressure.
Figure 8. Selectivity to acyclic hydrogenation products as a function of
total time-on-stream in a continuous reactor. Legend: CLAL
citronellal, NRLnerol, GRLgeraniol, CLOLcitronellol,
DMOL3,7-dimethyloctan-1-ol, and DME2,6-dimethyloctane.
Conditions: 70 °C, 10 bar of H2, 0.4 mL/min of feed, 0.086 M initial
concentration of citral in cyclohexane, 1 g of the Ni/(MAS + sepiolite)
catalyst, and 11.5 min of residence time.
Figure 9. Stereoselectivity to isopulegol isomers (SSIPs) in a continuous reactor as a function of (a) total time-on-stream and (b) conversion. Legend:
IPisopulegol, NIPneoisopulegol, IIPisoisopulegol, andNIIPneoisoisopulegol. Conditions: 70 °C, 10 bar of H2, 0.4mL/min of feed, 0.086M
initial concentration of citral in cyclohexane, 1 g of the Ni/(MAS + sepiolite) catalyst, and 11.5 min of residence time.
Organic Process Research & Development pubs.acs.org/OPRD Article
https://doi.org/10.1021/acs.oprd.1c00435
Org. Process Res. Dev. 2022, 26, 387−403
396
■ KINETIC MODELING
Kinetic analysis of one-pot transformations of citronellal to
menthols in a batch reactor has been reported for the Ru/β-25
catalyst,37 while such analysis is absent for citral transformations
to menthol for both batch and continuous operation modes.
In the current work, we have thus developed a kinetic model
capable of describing experimental data for transformations of
citral to menthol in a continuous mode.
The reaction network is an extension of the network proposed
previously for one-pot transformations of citronellal to
menthol38 and includes hydrogenation of citral to citronellal
and side transformations to defunctionalized products (Figure
13).
The noncompetitive adsorption of organic compounds and
hydrogen allows the introduction of the rate expressions for the
following type
α ρ=r k c P
D D1
1 Citral
H
1 B
2 (1)
Figure 10. Stereoselectivity to menthol isomers (SSMEs) in a continuous reactor as a function of (a) total time-on-stream and (b) conversion (X).
Legend: MEmenthol, NMEneomenthol, IMEisomenthol, and NIMEneoisomenthol. Conditions: 70 °C, 10 bar of H2, 0.4 mL/min of feed,
0.086 M initial concentration of citral in cyclohexane, 1 g of the Ni/(MAS + sepiolite) catalyst, and 11.5 min of residence time.
Figure 11. Citral conversion as a function of total time-on-stream at different temperatures in a continuous reactor. Conditions: 40−80 °C, 10 bar of
H2, 0.4 mL/min of feed, 0.086M initial concentration of citral in cyclohexane, 1 g of the Ni/(MAS + sepiolite) catalyst, and 11.5 min of residence time.
Figure 12. Citral conversion as a function of total time-on-stream at different pressures in a continuous reactor. Conditions: 70 °C, 2.5−10 bar of H2,
0.4 mL/min of feed, 0.086 M initial concentration of citral in cyclohexane, 1 g of the Ni/(MAS+sepiolite) catalyst, and 11.5 min of residence time.
Organic Process Research & Development pubs.acs.org/OPRD Article
https://doi.org/10.1021/acs.oprd.1c00435
Org. Process Res. Dev. 2022, 26, 387−403
397
where k1 is the lumped rate constant, which includes adsorption
constants of hydrogen and citral, P is the hydrogen pressure, ρB
is the catalyst bulk density (0.217 g/cm3), r1 is the reaction rate
along path 1, ccitral is the concentration of citral (mol/L), and D
and DH2 are denominators
= + + + + +
+ + + + + +
D K c K c K c c c
c K c c c c K c
1 (
) ( )
A Citral B Citronellal C IP NIP IIP
NIIP D MNT NM IM NIM F DFP
(2)
= +D K P1H H2 2 (3)
In eqs 2 and 3, KA, KB, KC, KD, KF, and KH2 are equilibrium
constants for adsorption of citral, citronellal, isopulegols,
menthols, defunctionalized products, and hydrogen, respec-
tively.
The term α1 reflects catalyst deactivation, which was modeled
in a semiempirical way24,39,40 by introducing an exponential
decay function of metal sites with TOS
α = −e k1 TOSd1 (4)
where kd1 is the temperature-dependent deactivation constant.
For the cyclization reaction routes (reactions 21−24), the
expressions for the observed rates can be written as follows
η α ρ=− −
−r
k c
D2,1 4 2,1 4
2,1 4 Citronellal
2 B (5)
With another deactivation function, specific for acidic sites
α = −e k2 TOSd2 (6)
where kd2 is the temperature-dependent deactivation constant
for acidic sites. During the parameter estimation, it turned out,
however, that the temperature dependence of this and the first
deactivation constant could be neglected. In eq 5, η is the
catalyst effectiveness factor for a particular reaction reflecting the
experimental data in this work were generated in the regime
where the external and internal mass transfer cannot be
neglected. In general, for nonfirst-order reactions, the catalyst
effectiveness factor changes during the reaction as the Thiele
modulus depends on the concentration.
In the current work for hydrogenation reactions of isopulegols
to menthols on metallic sites, the rate expressions are
η α ρ η α ρ
η α ρ η α ρ
= =
= =
r
k c P
D D
r
k c P
D D
r
k c P
D D
r
k c P
D D
; ;31 31
31 IP
H
1 B 32 32
32 NIP
H
1 B 33
33
33 IIP
H
1 B; 34 34
34 NIIP
H
1 B
2 2
2 2 (7)
The initial kinetic model included step 4; however, because the
yields of citronellol were very low, this step was excluded from
further considerations. Moreover, numerical data fitting
indicated that because of lumping of defunctionalized products
into one group, it is sufficient to include only the rate of reaction
5 in the model, finally giving
η α ρ=r k c
D
( )
5 5
5 Citral
1 B (8)
η α ρ= + + +r k c c c c P
D D
( )
7 7
7 IP NIP IIP NIIP
H
1 B
2 (9)
η α ρ=r k c P
D D8 8
8 Citronellal
H
1 B
2 (10)
The plug flow reactor model was used, solving the rate equations
along with mass balances
= − −
= − − − − −
τ τ
dc
d
r r
dc
d
r r r r r r
;Citral 1 5
Citronellal
1 21 22 23 24 8
= − −
+ + +
= − −
+ + +
dc
d
r r r
c
c c c c
dc
d
r r r
c
c c c c
( )
;
( )
IP
21 31 7
IP
IP NIP IIP NIIP
NIP
22 32 7
NIP
IP NIP IIP NIIP
= − −
+ + +
= − −
+ + +
dc
d
r r r
c
c c c c
dc
d
r r r
c
c c c c
( )
;
( )
IIP
23 33 7
IIP
IP NIP IIP NIIP
NIIP
24 34 7
NIIP
IP NIP IIP NIIP
Figure 13. Reaction network for one-pot transformations of citral to menthol. Legend: IPisopulegol, NIPneoisopulegol, IIPisoisopulegol,
NIIPneoisoisopulegol, MNTmenthol, NMneomenthol, IMisomenthol, NIMneoisomenthol, DMOdimethyloctanol, DFP
defunctionalized products, and CLOLcintronellol.
Organic Process Research & Development pubs.acs.org/OPRD Article
https://doi.org/10.1021/acs.oprd.1c00435
Org. Process Res. Dev. 2022, 26, 387−403
398
= = =
= = +
dc
d
r
dc
dt
r
dc
dt
dc
dt
r
dc
dt
r
dc
dt
r r
; ; ; ;MNT 31
NM
32
IM NIM
34
DMO
8
DFP
7 5 (11)
where τ is the residence time in min. For the sake of simplicity,
the rate constants for transformations of different isomers of
isopulegols to defunctionalized products were considered to be
equal to each other; thus, the rates of step 7 in eq 11 contain the
term corresponding to the fraction of a particular isomer.
During the parameter estimation, average values of the
catalyst effectiveness factors for different rates were calculated
neglecting their concentration dependencies, as a rather close to
first-order kinetic curve could be deduced from the batch data
(Figure 5).
The temperature dependence was modeled with the modified
Arrhenius equation, where T̅ is the mean temperature used in
the experiments (i.e., 70 °C)
= − −
i
k
jjj
y
{
zzzk k ej j
E
R T T
0
1 1a j
mean
,
(12)
Kinetic modeling was performed using ModEst software,41
minimizing the residual sum of squares between the calculated
and experimental data with the Levenberg−Marquardt and
simplex algorithms implemented in the software. The degree of
explanation reflecting a comparison between the residuals given
by the model with the residuals of the simplest model, i.e., the
average value of all data points,40 served as a measure of the fit
quality.
Application of eq 11 along with the expressions for the rates
for various steps in the reactionmechanism (eqs 1,5, 7−10) gave
a reliable description of experimental data (Figures 14 and S4)
with the degree of explanation equal to 97.1%. The values of
parameters are presented in Table 3.
In general, the parameters are rather well-identified apart from
Ea34 and activation energy for deactivation steps, exhibiting a
very minor temperature dependence. Due to a large number of
parameters, errors for some constants were in the range of 50%
and a certain correlation between them was seen. Figure 15
illustrates examples of such correlations.
Elongated contour plots visible in some of the graphs in
Figure 15 reflect a strong correlation between some parameters.
It should be, however, considered that this study represents a
first example of successful one-pot transformation of citral to
menthol in a continuous reactor over a supported nickel catalyst
shaped to extrudates along with a binder and apparently more
Figure 14.Comparison between experimental and calculated data. Time-on-stream (TOS, h) dependence of (a) citral, (b) isopulegol isomers, and (c)
menthol isomers concentrations in mol/L.
Organic Process Research & Development pubs.acs.org/OPRD Article
https://doi.org/10.1021/acs.oprd.1c00435
Org. Process Res. Dev. 2022, 26, 387−403
399
work is needed to determine the values of kinetic parameters in a
more statistically reliable way.
■ CONCLUSIONS
A 5% Ni on mesoporous aluminosilicate and sepiolite (as a
binder) catalyst was prepared by extrusion and tested in one-pot
menthol synthesis from citral at 40−80 °C under varied
hydrogen pressures between 2.5 and 10.0 bar. After deposition
of nickel, the mesoporous structure of the support was preserved
in the catalyst, which exhibited the size of metal clusters of ca. 10
nm.
The time-on-stream conversion of citral, which was initially
complete, decreased to ca. 60%. Simultaneously high selectivity
to menthols (above 70%), substantially exceeding previously
reported values for one-pot transformations with citral as the
reactant in a continuous reactor, decreased because of lower
conversion. In addition to the main product and the reaction
intermediates, the defunctionalized products were also formed
with ca. 10% overall selectivity. Selectivity to these products was
constant with conversion as a result of parallel pathways of their
generation from all reactants and products.
Stereoselectivity to menthols and isopulegols only slightly
changed with conversion and time-on-stream (TOS) being in
the range of 66−71% for the desired (±)-menthol.
Citral conversion as a function of time-on-stream was almost
independent of hydrogen pressure at 70 °C with somewhat
more prominent deactivation at a lower hydrogen pressure.
Kinetic analysis for citral transformations to menthol for both
batch and continuous operation modes is absent in the open
literature; thus, in the current work, a kinetic model capable of
describing experimental data for transformations of citral to
menthol in a continuous mode was developed for the first time.
The model, including formation of the main reaction as well as
side products, takes into account catalyst deactivation on two
types of sites of the bifunctional catalyst. Numerical data fitting
performed for the whole experimental data set confirmed
applicability of the advanced kinetic model.
■ EXPERIMENTAL SECTION
Preparation of the Ni Catalyst. Mesoporous aluminosi-
licate (MAS) as a support and sepiolite (Mg2H2SiO9 xH2O,
Sigma-Aldrich) as a binder were dried overnight at 100 °C and
physically mixed in the mass ratio = 70/30, e.g., 4.2 g and 1.8 g,
respectively. Methylcellulose (1.0 wt %, e.g., 0.06 g) as an
organic binder (viscosity: 4000 cP, Sigma-Aldrich) was
dissolved in 3 mL of water and added into the mixture of
MAS and sepiolite to control the rheological properties of the
final slurry. Distilled water was slowly dropped into the mixture
under stirring at room temperature. The extrudates were
fabricated in a cylindrical shape with a diameter of 1.4 mm
using the extruder equipped with a 1.5-mm diameter-hole die
plate (TBL-2, Tianjin Tianda Beiyang Chemical Co. Ltd.,
China) driven by a rotational velocity of 1400 rpm.
Subsequently, the extrudates were dried in an oven at 100 °C
overnight, calcined at 500 °C for 4 h in a muffle oven, and cut to
a length of ca. 0.6−1.0 cm.
Ni-containing extrudates were prepared by an incipient
wetness impregnation with nickel nitrate aqueous solution
Ni(NO3)2·6H2O (Sigma-Aldrich, >97%) to achieve the nominal
nickel loading 5.7 wt %, dried overnight at 100 °C and calcined
at 450 °C for 6 h in a muffle oven. Before catalytic run, the
synthesized Ni/(MAS+Sepiolite) extrudates were reduced in
the reactor in hydrogen flow (100mL/min) at 350 °Cduring 2.5
h with a temperature ramp of 2 °C/min.
The 5 wt % Ni/MAS powder catalyst was prepared by the
incipient wetness impregnation with an aqueous solution of
Ni(NO3)2·6H2O (CJSC Souzchimprom). Thereafter, the
sample was dried at 100 °C overnight and calcined at 450 °C
for 6 h with a temperature ramp of 2 °C/min controlling
decomposition of nickel nitrate by measuring the pH of the
outgoing gases. Before a catalytic run, calcined Ni/MAS powder
catalysts were activated in hydrogen flow at 350 °C during 2 h
with the temperature ramp of 2 °C/min.
Characterization of Ni Catalysts. Fresh and spent Ni
catalysts in the powder and shaped forms were characterized in
detail. Inductively coupled plasma-optical emission spectrom-
etry determined the metal concentration in the entire volume of
the catalyst with the PerkinElmer Optima 5300 DV device.
Transmission electron microscopy using JEOL JEM-
1400Plus was used to investigate the metal shape, particle size,
and metal location.
Morphological studies were performed by scanning electron
microscopy (SEM, Zeiss Leo Gemini 1530). Elemental analysis
of microporous materials was conducted by energy-dispersive X-
ray microanalysis using the same instrument.
Nitrogen physisorption was used to determine the pore size
distribution, pore volume, and specific surface area with the
Micromeritics 3Flex-3500 machine.
Fourier transform infrared spectroscopy with pyridine as a
probe molecule was applied to quantify the amount and strength
of Brønsted and Lewis acid sites using an ATI Mattson FTIR
Infinity series spectrometer.
Powder X-ray diffraction (XRD) patterns of extrudates were
obtained on a Bruker D8 Advance diffractometer (Cu Kα
radiation, λ = 0.15418 nm) equipped with a LynxEye position-
sensitive detector. The data were collected in the 2θ range of 5−
65° with a step of 0.05° and a collection time of 3 s. Phase
identification was performed using the ICDD PDF-2 database
Table 3. Values of Kinetic Parameters
constant value error (%) units
η1k1 52.9 56.8 mol/(g bar min)
η21k21 8.11 45.6 mol/(g min)
η22k22 3.54 45.5 mol/(g min)
η24k24 1.18 51.3 mol/(g min)
η31k31 77.6 56.8 mol/(g bar min)
η32k32 45.0 56.4 mol/(g bar min)
η34k34 40.2 57.6 mol/(g bar min)
η5k5 4.7 59.2 mol/(g bar min)
η7k7 4.07 60.0 mol/(g bar min)
η8k8 7.1 56.3 mol/(g bar min)
Ea1 80.7 4.7 kJ/mol
Ea21 38.1 10.4 kJ/mol
Ea22 37.9 13.0 kJ/mol
Ea24 92.2 18.7 kJ/mol
Ea31 45.5 7.9 kJ/mol
Ea32 48.8 13.3 kJ/mol
Ea34 not reliable not reliable kJ/mol
Ea5 56.8 11.5 kJ/mol
Ea7 111 20.2 kJ/mol
Ea8 40.7 20.7 kJ/mol
KC 749 52.3 L/mol
KH 4.5 41.6 1/bar
kd1 0.037 2.8 l/min
kd2 0.0175 8.8 l/min
Organic Process Research & Development pubs.acs.org/OPRD Article
https://doi.org/10.1021/acs.oprd.1c00435
Org. Process Res. Dev. 2022, 26, 387−403
400
[Powder Diffraction File database PDF-2, International Centre
for Diffraction Data, 2009]. The average crystallite size of phases
was calculated by the line broadening analysis according to the
Scherrer equation.
The mechanical strength of the shaped catalyst was measured
by the crush tester (SE 048, Lorentzen & Wettre) for 10
extrudates in the vertical position and 10 extrudates in the
horizontal position.
Details of characterization methods and instruments are
reported in our previous publications.20,25,26,30,42−44
Catalytic Tests. The experiments of citral transformations
over Ni/(MAS + Sepiolite) extrudates were carried out in a
continuous trickle-bed reactor (internal diameter 1.25 cm,
empty reactor volume 14.7 mL, catalyst bed volume 4.6
mL)20,23,25−27 at different temperatures 40, 50, 60, 70, and 80
°C under varied hydrogen pressures of 2.5, 5.0, and 10.0 bar. A
quartz wool was placed at the bottom of the reactor, then 1 g of
Ni/(MAS + Sepiolite) extrudates as well as inert quartz beads of
the size in the range 0.2−0.8 mm (∼18 g) were loaded inside the
reactor, filling the voids between extrudates and in the reactor.
Citral (cis-/trans-isomer ∼1/1, ≥95.0%, Sigma-Aldrich) in
cyclohexane (≥99.9%, Alfa Aesar) solution (0.086 M) was fed
into the reactor with a feeding rate of 0.4 mL/min, while the
hydrogen rate was 100 mL/min.
A 5 wt %Ni/MAS powder catalyst was tested in the autoclave
(Parr, 300 mL). The experiment was performed over 0.2 g of the
catalyst at 70 °C, 10 bar ofH2 with 0.086M citral in cyclohexane,
and 900 rpm.
The liquid samples were analyzed by gas chromatography
with flame-ionization detection using Agilent GC 6890 N. All
samples were diluted with cyclohexane as a solvent before
analysis. The column was DB-1 column with a length of 30 m,
diameter of 250 μm, and film thickness of 0.5 μm. Temperature
program: 110 °C−0.4 °C/min−130 °C−13 °C/min−200 °C (5
min). Details are given in our previous publications.20,23,25−27
■ ASSOCIATED CONTENT
*sı Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.oprd.1c00435.
Isotherms of adsorption/desorption (powdered and
extruded catalysts), pore size distribution (powdered
and extruded catalysts), parity plots for all reaction
components, and XRD patterns of powdered Ni/MAS
catalysts (PDF)
Figure 15. Examples of correlations between kinetic parameters. The units of parameters are shown in Table 3.
Organic Process Research & Development pubs.acs.org/OPRD Article
https://doi.org/10.1021/acs.oprd.1c00435
Org. Process Res. Dev. 2022, 26, 387−403
401
■ AUTHOR INFORMATION
Corresponding Author
Dmitry Yu. Murzin− Johan Gadolin Process Chemistry Centre,
Åbo Akademi University, Turku/Åbo 20500, Finland;
orcid.org/0000-0003-0788-2643; Email: dmurzin@
abo.fi
Authors
Irina L. Simakova − Boreskov Institute of Catalysis, 630090
Novosibirsk, Russia; orcid.org/0000-0002-5138-4847
Zuzana Vajglová − Johan Gadolin Process Chemistry Centre,
Åbo Akademi University, Turku/Åbo 20500, Finland
Päivi Mäki-Arvela − Johan Gadolin Process Chemistry Centre,
Åbo Akademi University, Turku/Åbo 20500, Finland;
orcid.org/0000-0002-7055-9358
Kari Eränen − Johan Gadolin Process Chemistry Centre, Åbo
Akademi University, Turku/Åbo 20500, Finland
Leena Hupa − Johan Gadolin Process Chemistry Centre, Åbo
Akademi University, Turku/Åbo 20500, Finland
Markus Peurla − University of Turku, FI-20520 Turku,
Finland
Ermei M. Mäkilä − University of Turku, FI-20520 Turku,
Finland; orcid.org/0000-0002-8300-6533
Johan Wärnå − Johan Gadolin Process Chemistry Centre, Åbo
Akademi University, Turku/Åbo 20500, Finland
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.oprd.1c00435
Notes
The authors declare no competing financial interest.
■ ACKNOWLEDGMENTS
The authors are grateful to the Academy of Finland for funding
through the project: Synthesis of spatially controlled catalysts
with superior performance. Electron microscopy samples were
processed and analyzed at the Electron Microscopy Laboratory,
Institute of Biomedicine, University of Turku, which receives
financial support from Biocenter Finland. I.L.S. is grateful for the
support from the Ministry of Science and Higher Education of
the Russian Federation, under the governmental order for
Boreskov Institute of Catalysis (Project No. AAAA-A21-
121011390055-8).
■ REFERENCES
(1) Noyori, R. Asymmetric Catalysis: Science and Opportunities
(Nobel Lecture 2001). Adv. Synth Catal. 2003, 345, 15−32.
(2) Otsuka, S.; Tani, K.; Yamagata, T.; Akutagawa, S.; Kumobayashi,
H.; Yagi, M. Process for the preparation of enamines or imines.
EP0068506A11983.
(3) Fleischer, J.; Bauer, K.; Hopp, R. Resolution of menthol,
neomenthol or isomenthol racemic mixtures by esterification with
benzoic acid derivatives. DE2109456A11972.
(4) Besson, M.; Bullivant, L.; Nicolaus, N.; Gallezot, P. Stereoselective
Thymol Hydrogenation: I. Kinetics of Thymol Hydrogenation on
Charcoal-Supported Platinum Catalysts. J. Catal. 1993, 140, 30−40.
(5) Allakhverdiev, A. I.; Kul’kova, N. V.; Murzin, D.Yu. Liquid-Phase
Stereoselective Thymol Hydrogenation over Supported Nickel
Catalysts. Catal. Lett. 1994, 29, 57−67.
(6) Allakhverdiev, A. I.; Kul’kova, N. V.; Murzin, D.Yu. Kinetics of
Thymol Hydrogenation over a Ni-Cr2O3 Catalyst. Ind. Eng. Chem. Res.
1995, 34, 1539−1547.
(7) Cardena, G.; Oliva, R.; Reyes, P. Catalytic Hydrogenation of
Thymol over Pd/MgO Prepared by SMADMethod. J. Chil. Chem. Soc.
2006, 51, 1053−1056.
(8) Dudas, J.; Hanika, J.; Lepuru, J.; Barkhuysen, M. Thymol
Hydrogenaton in a Bench Scale Trickle Bed Reactor. Chem. Biochem.
Eng. Quart. 2005, 19, 255−262.
(9) Aumo, J.; Oksanen, S.; Mikkola, J.-P.; Salmi, T.; Murzin, D. Yu.
Hydrogenation of Citral over Activated Carbon Cloth Catalyst. Ind.
Eng. Chem. Res. 2005, 44, 5285−5290.
(10) Mäki-Arvela, P.; Tiainen, L.-P.; Lindblad, M.; Demirkan, K.;
Kumar, N.; Sjöholm, R.; Ollonqvist, T.; Väyrynen, J.; Salmi, T.; Murzin,
D. Yu. Liquid-Phase Hydrogenation of Citral for Production of
Citronellol: Catalyst Selection. Appl. Catal., A 2003, 241, 271−288.
(11) Mäki-Arvela, P.; Kumar, N.; Nieminen, V.; Sjoholm, R.; Salmi,
T.; Murzin, D. Yu. Cyclization of Citronellal over Zeolites and
MesoporousMaterials for Production of Isopulegol. J. Catal. 2004, 225,
155−169.
(12) Chuah, G. K.; Liu, S. H.; Jaenicke, S.; Harrison, L. J. Cyclisation
of Citronellal to Isopulegol Catalysed by Hydrous Zirconia and Other
Solid Acids. J. Catal. 2001, 200, 352−359.
(13) Gralla, G.; Heydrich, G.; Bergner, E. J.; Ebel, K. Process for
preparation of menthol by hydrogenation of isopulegol. U.S. Patent
US2010191021A12010.
(14) Nie, Y. T.; Jaenicke, S.; Chuah, G. K. Zr-Zeolite Beta: A New
Heterogeneous Catalyst System for the Highly Selective Cascade
Transformation of Citral to (+/-)-Menthol. Chem. Eur. J. 2009, 15,
1991−1999.
(15) Negoi, A.; Teinz, K.; Kemnitz, E.; Wuttke, S.; Parvulescu, V. I.;
Coman, S. M. Bifunctional Nanoscopic Catalysts for the One-Pot
Synthesis of (+/-)-Menthol from Citral. Top Catal. 2012, 55, 680−687.
(16) Trasarti, A. F.; Marchi, A. J.; Apesteguia, C. R. Design of Catalyst
Systems for the One-Pot Synthesis of Menthols from Citral. J. Catal.
2007, 247, 155−165.
(17) Shah, A. K.; Maitlo, G.; Shah, A. A.; Channa, I. A.; Kandhro, G.
A.; Maitlo, H. A.; Bhatti, U. H.; Shah, A.; Memon, A. Q.; Jatoi, A. S.;
Park, Y. H. One Pot Menthol Synthesis via Hydrogenations of Citral
and Citronellal over Montmorillonite-Supported Pd/Ni-Heteropoly
Acid Bifunctional Catalysts. React. Kinet., Mech. Catal. 2019, 128, 917−
934.
(18) Virtanen, P.; Salmi, T. O.; Mikkola, J. P. Supported Ionic Liquid
Catalysts (SILCA) for Preparation of Organic Chemicals. Top Catal.
2010, 53, 1096−1103.
(19) Mäki-Arvela, P.; Kumar, N.; Kubicka, D.; Nasir, A.; Heikkilä, T.;
Lehto, V.-P.; Sjöholm, R.; Salmi, T.; Murzin, D. Yu. One-pot Citral
Transformation to Menthol over Bifunctional Micro- and Mesoporous
Metal Modified Catalysts: Effect of Catalyst Support and Metal. J. Mol.
Catal. A. Chem. 2005, 240, 72−81.
(20) Vajglová, Z.; Mäki-Arvela, P.; Eränen, K.; Kumar, N.; Peurla, M.;
Murzin, D. Y. Catalytic Transformations of Citral in a Continuous Flow
over Bifunctional Ru-MCM-41 Extrudates. Catal. Sci. Technol. 2021,
11, 2873−2884.
(21) Deliy, I. V.; Danilova, I. G.; Simakova, I. L.; Zaccheria, F.;
Ravasio, N.; Psaro, R. Tuning Selectivity through the Support in the
Hydrogenation of Citral over Copper Catalysts. In Catalysis of Organic
Reactions, Book Series: Chemical Industries; Prunier, M. L., Ed.; CRC
Press: Boca Raton, Florida, USA, 2009; Vol. 123, pp 87−92.
(22) McKeown, D. A.; Post, J. E.; Etz, E. S. Vibrational Analysis of
Palygorskite and Sepiolite. Clays Clay Miner. 2002, 50, 667−680.
(23) Vajglová, Z.; Navas, M.; Mäki-Arvela, P.; Eränen, K.; Kumar, N.;
Peurla, M.; Murzin, D. Y. Transformations of Citral over Bifunctional
Ru-H-Y-80 Extrudates in a Continuous Reactor. Chem. Eng. J. 2022,
429, No. 132190.
(24) Murzin, D. Yu. Engineering Catalysis; De Gruyter: Berlin, Boston,
2020; pp 1−414.
(25) Vajglová, Z.; Kumar, N.; Mäki-Arvela, P.; Eränen, K.; Peurla, M.;
Hupa, L.; Murzin, D. Yu. Effect of Binders on the Physicochemical and
Catalytic Properties of Extrudate-Shaped Beta Zeolite Catalysts for
Cyclization of Citronellal. Org. Process Res. Dev. 2019, 23, 2456−2463.
(26) Vajglová, Z.; Kumar, N.; Mäki-Arvela, P.; Eränen, K.; Peurla, M.;
Hupa, L.; Nurmi, M.; Toivakka, M.; Murzin, D. Yu. Synthesis and
Physicochemical Characterization of Shaped Catalysts of Beta and Y
Organic Process Research & Development pubs.acs.org/OPRD Article
https://doi.org/10.1021/acs.oprd.1c00435
Org. Process Res. Dev. 2022, 26, 387−403
402
Zeolites for Cyclization of Citronellal. Ind. Eng. Chem. Res. 2019, 58,
18084−18096.
(27) Vajglová, Z.; Kumar, N.; Peurla, M.; Eränen, K.; Mäki-Arvela, P.;
Murzin, D. Yu. Cascade Transformations of (±)-Citronellal toMenthol
over Extruded Ru-MCM-41 Catalysts in a Continuous Reactor. Catal.
Sci. Technol. 2020, 10, 8108−8119.
(28) Sing, K. S. W.; Everett, D. H.; Haul, R. A. W.; Moscou, L.;
Pierotti, R. A.; Rouquerol, J.; Siemieniewska, T. Reporting
Physisorption Data for Gas/Solid Systems with Special Reference to
the Determination of Surface Area and Porosity (Recommendations
1984). Pure Appl. Chem. 1985, 57, 603−619.
(29) Beck, J. S.; Vartuli, J. C.; Roth,W. J.; Leonowicz,M. E.; Kresge, C.
T.; Schmitt, K. D.; Chu, C.-T.-W.; Olson, D. H.; Sheppard, E. W.;
McCullen, S. B.; Higgins, J. B.; Schlenker, J. L. A New Family of
MesoporousMolecular Sieves Prepared with Liquid Crystal Templates.
J. Am. Chem. Soc. 1992, 114, 10834−10843.
(30) Vajglová, Z.; Kumar, N.; Peurla, M.; Hupa, L.; Semikin, K.;
Sladkovskiy, D. A.; Murzin, D.Yu. Effect of the Preparation of Pt-
Modified Zeolite Beta-Bentonite Extrudates on Their Catalytic
Behavior in n-Hexane Hydroisomerization. Ind. Eng. Chem. Res. 2019,
58, 10875−10885.
(31) Kubicǩa, D.; Kumar, N.; Venäläinen, T.; Karhu, H.; Kubicǩová,
I.; Österholm, H.; Murzin, D. Yu. Metal- Support Interactions in
Zeolite-Supported Noble Metals: Influence of Metal Crystallites on the
Support Acidity. J. Phys. Chem. B. 2006, 110, 4937−4946.
(32) Özcan, A.; Kalipcilar, H. Preparation of Zeolite A Tubes from
Amorphous Aluminosilicate Extrudates. Ind. Eng. Chem. Res. 2006, 45,
4977−4984.
(33) Vajglová, Z.; Simakova, I. L.; Eränen, K.; Mäki-Arvela, P.; Kumar,
N.; Peurla, M.; Tolvanen, S.; Efimov, A.; Hupa, L.; Peltonen, J.; Murzin,
D. Yu. The Physicochemical and Catalytic Properties of Clay
Extrudates in Cyclization of Citronellal. Appl. Catal., A 2022, 629,
No. 118426.
(34) Vajglová, Z.; Kumar, N.; Peurla, M.; Hupa, L.; Semikin, K.;
Sladkovskiy, D. A.;Murzin, D. Yu. Deactivation and Regeneration of Pt-
Modified Zeolite Beta−Bindzil Extrudates in n-Hexane Hydro-
isomerization. J. Chem. Technol. Biotechnol. 2021, 96, 1645−1655.
(35) Plößer, J.; Lucas, M.; Claus, P. Highly Selective Menthol
Synthesis by One-Pot Transformation of Citronellal Using Ru/H-BEA
Catalysts. J. Catal. 2014, 320, 189−197.
(36) Mertens, P.; Verpoort, F.; Parvulescu, A. N.; de Vos, D. Pt/H-
Beta Zeolites as Productive Bifunctional Catalysts for the One-Step
Citronellal-To-Menthol Conversion. J. Catal. 2006, 243, 7−13.
(37) Yongzhong, Z.; Yuntong, N.; Jaenicke, S.; Chuah, G. K.
Cyclisation of Citronellal over Zirconium Zeolite Beta - a Highly
Diastereoselective Catalyst to (±)-Isopulegol. J. Catal. 2005, 229, 404−
413.
(38) Plößer, J.; Lucas, M.; Wärnå, J.; Salmi, T.; Murzin, D. Yu.; Claus,
P. Kinetics of the One-Pot Transformations of Citronellal to Menthols
on Ru/H-BEA Catalysts. Org. Process Res. Dev. 2016, 20, 1647−1653.
(39) Hughes, R. Deactivation of Catalysts; Academic Press: London,
1984; pp 1−265.
(40) Ostrovskii, N. M. General Equation for Linear Mechanisms of
Catalyst Deactivation. Chem. Eng. J. 2006, 120, 73−82.
(41) Haario, H. Modest 6.0 - a User’s Guide; ProfMath: Helsinki,
Finaland, 2001.
(42) Zhivonitko, V. V.; Vajglová, Z.; Mäki-Arvela, P.; Kumar, N.;
Peurla,M.;Murzin, D. Yu. DiffusionMeasurements of Hydrocarbons in
Zeolites with Pulse-Field Gradient Nuclear Magnetic Resonance
Spectroscopy. Russ. J. Phys. Chem. A. 2021, 95, 547−557.
(43) Vajglová, Z.; Kumar, N.; Peurla, M.; Peltonen, J.; Heinmaa, I.;
Murzin, D. Yu. Synthesis and Physicochemical Characterization of Beta
Zeolite-Bentonite Composite Materials for Shaped Catalysts. Catal. Sci.
Technol. 2018, 8, 6150−6162.
(44) Azkaar, M.; Mäki-Arvela, P.; Vajglová, Z.; Fedorov, V.; Kumar,
N.; Hupa, L.; Hemming, J.; Peurla, M.; Aho, A.; Murzin, D. Yu.
Synthesis of Menthol from Citronellal over Supported Ru- and Pt-
Catalysts in Continuous Flow. React. Chem. Eng. 2019, 4, 2156−2169.
Organic Process Research & Development pubs.acs.org/OPRD Article
https://doi.org/10.1021/acs.oprd.1c00435
Org. Process Res. Dev. 2022, 26, 387−403
403