Reac Kinet Mech Cat (2016) 118:235–246

DOI 10.1007/s11144-016-0994-9

The solvent effect in b-pinene oxide rearrangement

Eliška Vyskočilová1 • Marek Malý1 • Atte Aho2 •

Jiřı́ Krupka1 • Libor Červený1

Received: 7 January 2016 / Accepted: 28 January 2016 / Published online: 11 February 2016
Ó Akadémiai Kiadó, Budapest, Hungary 2016

Abstract Perillyl alcohol preparation by the rearrangement of b-pinene oxide was

studied. Sn-MCM-41, USY and FeCl3 were used as catalysts. The solvent influence
was tested using basic polar solvents—dioxane, acetonitrile, DMF, DMA, DMSO
and N-methyl pyrrolidone. It was found that the solvent has significant influence
on the selectivity: Sn-MCM-41 was in previous works used for b-pinene oxide
rearrangement for the preparation of myrtenal or myrtenol in acetonitrile or nitro-
methane as solvent. When DMSO was used, a high yield of perillyl alcohol was
obtained. The highest selectivity to perillyl alcohol (66 %) was obtained when Sn-
MCM-41 was used as a catalyst under the following conditions: 70 °C, 10 wt% of
catalyst, volume ratio b-pinene oxide : solvent = 1:8. The selectivity may be
slightly increased by decrease of ratio b-pinene oxide : solvent in the case of Sn-
MCM-41. Moreover, Sn-MCM-41 may be reused at least four times without any
loss of selectivity or activity.

Keywords Sn-MCM-41
Perillyl alcohol
b-Pinene oxide
Solvent effect

Introduction

b-Pinene oxide is an important material for fine chemical production. Its

rearrangement (Fig. 1) using acid catalysts serves for the production of two
important compounds—myrtanal and perillyl alcohol. Especially the latter is
demanded in the industry due to its bactericide and fungicide properties [1]. The

& Eliška Vyskočilová

eliska.vyskocilova@vscht.cz
1
Department of Organic Technology, University of Chemistry and Technology Prague,
Technicka 5, 166 28 Prague 6, Czech Republic
2
Johan Gadolin Process Chemistry Centre, Åbo Akademi University, 20500 Turku/Åbo, Finland

123
236 Reac Kinet Mech Cat (2016) 118:235–246

O OH OH OH
O

+ + +

OH
-Pinene oxide Myrtanal Myrtenol Perillyl alcohol
p-Menth-1-ene-
7,8-diol

Fig. 1 b-Pinene oxide rearrangement

leading of the reaction is deeply dependent on the reaction conditions—especially

on the used catalyst. Myrtanal can be prepared with high selectivity using Sn
modified MCM-41 [2] or Zr, Sn, Ti, Ta, Nb, Al, Ge modified beta zeolite [3].
Modification by Sn was also successfully tested using beta zeolites as catalysts with
different Si/Al ratio [4]. For the preparation of perillyl alcohol from b-pinene oxide,
ammonium salts of nitric acid [5] or nitric acid modified pyridine resins [2] were
used. The selectivity to desired alcohol was up to 60 %. A synergic effect of
ammonium nitrate immobilized on hectorite clay was observed [6] in the
preparation of this compound. Perillyl alcohol may be also prepared from b-pinene
oxide by multistep synthesis using e.g. catex as a catalyst [7], this way includes the
acylation of b-pinene oxide to perillyl acetate followed by saponification. This way
may be performed also using different catalyst types [8].
In earlier work [4], it was found that the Brønsted acid centers are undesired for
the production of myrtanal (because in the presence of Brønsted acid sites,
formation of myrtenol occurs) and in the presence of both types of acid centers, the
production rate of perillyl alcohol increases. By adjusting the acid–base properties
of the catalyst with Brønsted centers, it was found [2] that a medium strong acid
with a comparatively strong amine supported the formation of perillyl alcohol,
whereas the combination of a stronger acid with a weaker base benefits the myrtenol
production.
In previous works, only low attention was paid to the solvents. The b-pinene
oxide rearrangement was performed in toluene (Sn-zeolite), acetonitrile [2, 3],
nitromethane [2] or dimethyl formamide [5, 6]. These solvents can be almost all
included to the group of basic polar aprotic solvents. The basicity of these solvents
may be measured by e.g. donor number, which was described by Gutmann [9].
Solvents can mediate the proton transfer reaction from Brønsted acidic surface
groups to organic substrates [10, 11] and Lewis sites on surface can interact with
polar molecules such as acetone to generate proton [12]. Detailed studies dealing the
catalytic activity as a function of solvent are not given because many factors can
influence the reaction [13–16] especially the reaction catalyzed by solid acids.
In this work, we would like to show the effect of chosen solvents to the
selectivity to perillyl alcohol in the b-pinene oxide rearrangement. The properties of
the used solvent are essential for the formation of the product in combination with

123
Reac Kinet Mech Cat (2016) 118:235–246 237

the catalyst and that when using Sn modified MCM-41, previously used for selective
preparation of myrtanal, perillyl acohol may be prepared with high selectivity.

Experimental

For USY (Zeolyst Inc.), Si/Al ratio of zeolite catalysts was 15. MCM-41 was
prepared by hydrothermal synthesis [17] and modified by Sn using SnCl2.
Impregnation was followed by calcination (X-ray powder diffraction data were
collected at room temperature with an X’Pert PRO h-h powder diffractometer with
parafocusing Bragg–Brentano geometry using Cu Ka radiation). The amount of Sn
was 5.61 %. The amount of elements in the materials was confirmed by XRF
analysis (ARL 9400 XP sequential WD-XRF spectrometer was used). FeCl3 was
purchased from Penta.

FTIR

The Brønsted and Lewis acid sites of the catalysts were measured by infrared
spectroscopy (ATI Mattson FTIR) using pyridine as a probe molecule. A thin self-
supported wafer of catalyst was pressed and then placed into the FTIR-cell. The cell
was evacuated and the temperature was raised to 450 °C and kept for 1 h.
Thereafter, the temperature was decreased to 100 °C and the background spectra of
the sample were recorded. Pyridine was adsorbed on the sample for 30 min at
100 °C followed by desorption at 250, 350 and 450 °C for 1 h and the spectra of the
sample were recorded in between every temperature ramp. The scanning was
performed under vacuum at 100 °C. Spectral bands at 1545 cm-1 and at 1450 cm-1
were used to identify Brønsted (BAS) and Lewis (LAS) acid sites, respectively. The
quantitative amount of BAS and LAS was calculated with the constants of Emeis
[18].

TPD

Temperature programmed desorption (TPD) of pyridine was carried out using

Micromeritics Instrument, AutoChem II 2920. For desorbed pyridine detection, both
a thermal conductivity detector (TCD) and quadrupole mass spectrometer (MKS
Cirrus 2 Analyzer) with a capillary-coupling system were used. 0.09 g of catalyst
sample was placed in quartz made U shaped tube. Prior to the adsorption of
pyridine, the catalyst was heated under ultrahigh-purity-helium flow (30 mL/min)
up to 450 °C and kept at 450 °C for 1 h to activate the catalyst surface. In the
following step, the sample was cooled down to adsorption temperature of 150 °C.
This temperature was chosen to prevent the condensation of pyridine on the catalyst
surface. Measured pulses of pyridine vapor (pulse volume 5 mL) were injected into
helium gas and carried through the catalyst sample until saturated adsorption. Then
the sample was flushed with helium for 2 ho to remove physisorbed pyridine.
Afterwards, the linear temperature program (5 °C min-1) was started and sample

123
238 Reac Kinet Mech Cat (2016) 118:235–246

heated up to temperature of 800 °C. The desorbed amounts of pyridine were

determined by calibration of the intensity of 79 amu MS response.

Catalytic test

b-Pinene oxide (purity 80 %) was obtained from Aroma a.s. and purified by
distillation before use to obtain purity 96 %. N,N-Dimethyl formamide, 1,4-dioxane
and acetonitrile were purchased from Penta, N,N-dimethyl acetamide and N-methyl
pyrrolidone were purchased from Sigma-Aldrich and dimethylsulfoxide from
Lachema.
The solid catalysts were dried overnight at 300 °C under nitrogen atmosphere
prior to use. In a typical experiment, 1.25 g of b-pinene oxide (8.2 mmol) was
dissolved in 6 ml of a solvent (concentration of was 1.4 mol/l). The reaction
mixture was heated to the desired temperature (70 °C) and 0.125 g of catalyst was
added. Reactions were performed under nitrogen atmosphere. The samples taken
were analyzed by GC equipped with non-polar DB-5 column. The identification of
the products was performed using GC–MS, non-polar column DB-1.

Results and discussion

High basicity polar aprotic solvents show wide variation in their ability to interact
with cations [19]. The measurement of their basicity, donor number, may divide
them to two groups: with high donor numbers and with low donor numbers. We
have chosen some members of both groups as can be seen in Table 1. From the
group possessing the high ones, N,N-dimethyl formamide, N-methyl pyrrolidone,
N,N-dimethyl acetamide and dimethyl sulfoxide were used. From the other group,
acetonitrile and dioxane were chosen. The polarity of the solvent also plays an
important role in many reactions. Dioxane was the least polar of all solvents
(relative permittivity 2.2), the highest polarity from studied group had dimethyl
sulfoxide. The differences between polarities of other solvents were more or less
comparable.
The solid catalysts applied here were commercial USY in H? form, which was
used without any treatment and its characterization was often studied in details. Sn-
Table 1 Chosen properties of the solvents used
Solvent Dielectric constant Donor number Boiling
(relative permittivity er) (kJ/mol) [9] point (°C)
[20]

Acetonitrile 37.5 59 82
Dioxane 2.2 62 101
Dimethyl formamide (DMF) 36.7 111 152
N-Methyl pyrrolidone (NMP) 32 114 202
Dimethyl acetamide (DMA) 37.8 116 165
Dimethylsulfoxide (DMSO) 46.7 125 189

123
Reac Kinet Mech Cat (2016) 118:235–246 239

MCM-41 was characterized by XRD to confirm that MCM-41 structure (Fig. 2) was
preserved. Fig. 2 shows low angle XRD patterns with the evident diffraction peak at
2h values of 2.3° corresponding to (100) reflection together with small reflections at
3.9° and 4.6°, which are characteristic for the mesoporous structure of MCM-41.

Catalysts characterization

The basic information necessary for the evaluation of b-pinene oxide rearrangement
is the acidity of the catalysts. Concentrations of Brønsted and Lewis acid sites were
determined by FTIR using pyridine as a probe molecule and supporting acidity
measurement was performed by TPD of pyridine. The comparison of these two
methods is possible [21, 22]. The distribution of Brønsted (BAS) and Lewis (LAS)
acid sites for the catalysts is given in Table 2.
From FTIR measurements, it was obvious that the significantly higher Brønsted
acidity had zeolite USY as it was supposed. Sn-MCM-41 has only few BAS, but
possessed by weak and middle strong Lewis acid sites. The total amount of LAS of
Sn modified MCM-41 was slightly higher than LAS of USY, which has lower total
amount of Lewis acid center, but had also the strong ones. According to the
literature, [2] the type and amount of acid sites has significant influence on the
selectivity of b-pinene oxide rearrangement.
The total acidity of used materials was measured also by TPD of pyridine. This
technique showed lower total acidity of the surfaces, but this phenomenon was not
unusual [21] and reflected the fact, that the acidities were not determined under
identical experimental conditions. Differences could be the consequence of different
responses of the catalysts in holding the probe under the action of a carrier gas
rather than under vacuum. The trends in total acidity are the same using both of the
techniques. A TPD spectrum of pyridine from Sn-MCM-41 was composed of one
peak (Fig. 3). This showed the existence of one type of acidic centers of same
acidity strength. The temperature maximum (264 °C) indicated weaker acid centers.
From comparison with the results of pyridine-FTIR, it could be deduced that these
are Lewis acid centers in the groups LAS-250 °C and LAS 350 °C. Under TPD
conditions, pyridine desorbed from the surface of USY in the range 150–800 °C

Fig. 2 XRD of MCM-41 15000

modified by Sn
Intensity (a.u.)

10000

5000

0
2 4 6 8
2 Theta (°)

123
240 Reac Kinet Mech Cat (2016) 118:235–246

Table 2 Surface acid sites per gram of dry catalyst and overall surface acidity, determined by FTIR and
TPD analysis of pre-adsorbed pyridine
Catalyst FTIR TPD

BAS LAS R AS R AS

250 °C 350 °C 450 °C 250 °C 350 °C 450 °C lmol/g lmol/g

USY 211 170 108 41 16 6 552 284

Sn-MCM-41 6 2 0 62 21 0 91 67

(Fig. 3); the spectrum is composed from three peaks. Therefore, USY contains
active centers characterized by different acidity strength and the majority of them is
represented by a peak with maximum at 564 °C. By the deconvolution of the TPD-
MS curve (incorporated in the evaluation software), it was observed that the peaks
represent 15, 5 and 80 % of area of pyridine peak and their maxima can be
attributed to the temperatures 254, 360 and 564 °C. This relative distribution is in
conflict with the result obtained by FTIR, where the majority of acidic centers
correspond to weak and medium acid centers (Table 2).

b-Pinene oxide rearrangement

As it was mentioned above, from b-pinene oxide rearrangement (Fig. 1), a large
variety of the products may be obtained. The most important of them are myrtanal,
myrtenol and perillyl alcohol. In the presence of water, p-menth-1-ene-7,8-diol can
also be present.

USY Sn-MCM-41
500

400
TPD-MS Signal (a.u.)

300

200

100

150 200 250 300 350 400 450 500 550 600 650 700 750 800
Temperature (°C)

Fig. 3 TPD spectra of pyridine (m/z = 79 fragment) from USY zeolite and Sn-MCM-41 (0.09 g of
sample, pyridine adsorption at 150 °C, heating ramp 5 °C/min)

123
Reac Kinet Mech Cat (2016) 118:235–246 241

From the literature, [2] the presence of amine is recommended for the increase of
the amount of perillyl alcohol in the reaction mixture. We have supposed that by the
replacement of amine (base) by Lewis basic solvent (Table 1) the formation of
perillyl alcohol could be also increased by the same mechanism and the difficult
catalyst modification could be excluded. Acetonitrile and dioxane were chosen as
low basic solvents and the rest of solvents as high basic solvents. From highly basic
solvents, the use of DMSO is desired from the point of view of use in
pharmaceutical industry.
We tested three types of catalysts in the b-pinene oxide rearrangement: (1)
Classic Lewis acid catalyst FeCl3. According to the literature, [2] in the presence of
a Lewis acid, a mixture of isomers is present in the reaction mixture in nitromethane
(DN = 11.3 kJ/mol) as a solvent. (2) Solid Brønsted acid USY and (3) Solid Lewis
acid—Sn-modified MCM-41. The results are summarized in Table 3.
In the case of ferric chloride and USY, it is obvious from Table 3 that with
increasing donor number, i.e. basicity of the solvent, the selectivity to perillyl
alcohol increased. The highest selectivity was in both cases obtained using DMSO
as a solvent. Ferric chloride should serve as Lewis acid, but the presence of
p-menth-1-ene-7,8-diol in the reaction mixture shows to the presence of water in the
reaction mixture. These traces of water may decompose this catalyst to HCl,
Brønsted acid. Moreover, similar results were obtained using hydrate of ferric

Table 3 Testing of polar basic aprotic solvents in b-pinene oxide rearrangement

Catalyst Solvent Conversion Selectivity to Selectivity to Selectivity to Other
of b-pinene myrtanal (%) perillyl myrtenol (%) (%)
oxide (%) alcohol (%)

FeCl3 Acetonitrile 97.3 63.6 5.9 2.3 28.2

Dioxane 98.3 63.4 6.1 1.6 28.9
DMF 96.1 21.6 31.9 8.3 38.2
DMA 78 21.0 36.1 7.0 35.8
DMSO 96 9.1 49.1 14.0 27.8
USY Acetonitrile 88.4 66.5 14.9 5.7 12.9
Dioxane 92.6 59.2 18.7 6.4 15.7
DMF 95.4 5.5 49.6 22.5 22.4
NMP 82.1 15.7 52.2 8.9 23.2
DMA 89.6 14.6 53.7 10.2 19.6
DMSO 98.3 12.4 58.6 14.6 14.4
Sn-MCM-41 Acetonirile 85 53.5 13.8 3.5 29.2
Dioxane 93.6 57 34.2 6.6 2.2
DMF 11.3 38.2 30.4 6.6 24.8
NMP 44 31.5 26.2 6.8 35.5
DMA 40.5 38.5 21.3 11.7 28.5
DMSO 98.4 8.2 66.1 12 13.7

Reaction conditions: 70 °C, 10 wt% of catalyst, 24 h, except FeCl3 and all catalysts in DMSO, volume
ratio b-pinene oxide : solvent = 1:8

123
242 Reac Kinet Mech Cat (2016) 118:235–246

chloride instead of anhydrous one. USY contains a majority of Brønsted acid sites
(Table 2), but the reaction may be catalyzed by both types of acid centers. Zeolite
USY was more successful for the preparation of perillyl alcohol and using solvents
DMF, NMP and DMSO only slight increase of selectivity to perillyl alcohol was
observed (from 49.6 to 58.6 %).
Totally different was the behavior of Sn-MCM-41. When using low polarity
solvents (acetonitrile and dioxane), myrtanal was the product with the highest
concentration in the reaction mixture as it was described in literature [2]. When
more basic solvents were used, the production of myrtanal was suppressed, but also
the concentration of perillyl alcohol was low (around 20–30 %). The concentration
of other products increased. In the reaction mixture using NMP as a solvent, the
following products were detected with the selectivity depicted in parentheses:
isoperillyl alcohol (10.6 %), p-mentha-1,4-dien-7-ol (9.7 %), p-mentha-2,5-dien-7-
ol (7.2 %) and myrtenal (4.3 %) (Fig. 4). It should be mentioned that these products
were detected also in other cases. Moreover, very low conversion compared to other
reactions were obtained using DMF, NMP and DMA as solvents. The low
conversion may be caused by the lowest concentration of acid sites in Sn-MCM-41
comparing other catalysts used. The relatively high concentration of perillyl alcohol
isomers can be attributed to isomerization reactions taking place in the reaction
mixture.
When DMSO was used as a solvent, the results confirming the results obtained
using other two catalysts was observed. High reaction rate and high selectivity to
perillyl alcohol was obtained. The selectivity to desired perillyl alcohol was the
highest from all tested catalysts (66 %). We suppose the synergic behavior of polar
Lewis aprotic base and weak Lewis acid in the formation of perillyl alcohol and the
synergy between solid Lewis acid and base was even more successful. A similar
synergy was also observed before [6]. Supposing that the solvent may participate in
the reaction mechanism, the combination of relatively strong Lewis base and acid
(Brønsted or Lewis) could serve for proton transfer favoring the formation of
perillyl alcohol (Fig. 5). The proposed mechanism confirms the mechanism offered
before [2]. The explanation of the synergic effect of acid and high basicity polar
aprotic solvents is a very complex problem that is influenced by many factors given
by use of solid catalyst and also by properties of solvents [16].
Another explanation may be also given—the participation of solvent in
mechanism is excluded. When FeCl3 was used as a catalyst the formation of
organic complexes with solvent [23, 24] could occur e.g. Fe(III)2Cl5(DMSO)3,

Fig. 4 Other compounds in the OH OH OH

reaction mixture a isoperillyl
alcohol, b p-mentha-1,4-dien-7- O
ol, c p-mentha-2,5-dien-7-ol and
d myrtenal

(a) (b) (c) (d)

123
Reac Kinet Mech Cat (2016) 118:235–246 243

Fe(III)2Cl4(DMF)4, Fe(III)Fe(II)Cl8(DMA)6, [Fe(II)(acetonitrile)6][Fe(III)Cl4]2.

The different nature of these complexes may influence both selectivity and reaction
rate.
Both solid catalysts (USY and Sn-MCM-41) have Lewis and Brønsted acid sites.
The Brønsted acid sites may be blocked by basic solvents; the higher donor number
of the solvent (basicity) the larger amount of Brønsted acid sites is blocked and,
therefore, the higher selectivity to perillyl alcohol is obtained. The highest
selectivity and reaction rate in the case when DMSO was used as a solvent may be
probably explained by its highest basicity (compared to other solvents). Moreover,
this behavior may be probably supported by stronger polar interaction of oxygen
atom of DMSO with Sn atom—this can affect the Lewis acidity and, therefore, the
selectivity of reaction.
The influence of reaction conditions was studied using Sn-MCM-41 in DMSO as
a solvent—the amount of the solvent in the reaction mixture, temperature and
catalyst amount. It was found out that the presence of the solvent is necessary
because without solvent reaction mixture polymerization occurred. The temperature
(40–70 °C) and catalyst amount (5–15 wt%) influence on the reaction rate. Slightly
higher formation of polymers was detected only at 80 °C, but ranged only in the
measurement error. The results using solvent amounts are depicted in Table 4 and it
is obvious that this factor had a significant influence on selectivity and especially on
polymer formation. Using low amounts of solvent, the conversion of b-pinene oxide
was low and this was connected with high formation of polymers. With an
increasing amount of solvent in the reaction mixture, the conversion and selectivity
increased and polymers were not observed in the reaction mixture. The highest
selectivity to perillyl alcohol (71 %) was observed when volume ratio b-pinene
oxide : DMSO was 1:16, but this value differed from value of selectivity (69.7)

S
OH
O
H
O H H

(I)

S S OH
O O+
Sn Sn H
H H
O
O

(II)

Fig. 5 Proposed mechanism for the formation of perillyl alcohol in the presence of solid Brønsted (I) or
Lewis acid (II) and Lewis base (DMSO)

123
244 Reac Kinet Mech Cat (2016) 118:235–246

Table 4 Effect of solvent volume on conversion of b-pinene oxide and selectivity to products
b-Pinene oxide/ Conversion of b- Selectivity to Selectivity to Selectivity to
DMSO (vol./vol.) pinene oxide (%) perillyl alcohol (%) myrtanal (%) polymers (%)

1:1 27.3 49.3 12.1 30.2

1:2 73.3 56.1 15.7 15.8
1:4 97.2 64.1 12.6 1.5
1:8 92.6 66.0 8.2 0
1:10 99.2 69.7 7.5 0
1:16 98.6 71.2 6.9 0
1:32 97.9 71.3 7.1 0
1:8a 96.3 67.1 8.3 0

Reaction conditions: catalyst Sn-MCM-41, solvent DMSO, temperature, 70 °C, 180 min
a
Dropwise addition of b-pinene oxide during 60 min

when ratio 1:10 was used in the range of measurement error. These results
confirmed the possible blockage of Brønsted acid sites and interaction of DMSO
with Lewis acid sites as it was mentioned above.
For the confirmation of optimal reaction conditions, the experiment was performed
on a larger scale (120 ml of DMSO, 1.5 g of Sn-MCM-41, 15 g of b-pinene oxide,
70 °C). The reaction rate of this experiment was lower compared to the kinetic
experiments (initial reaction rates 52.5 mmol/g/min and 0.55 mmol/g/min). The
selectivity to the desired perillyl alcohol was comparable in both cases. The
composition of experiment in a larger scale at the end of batch operation was: 3 %
of b-pinene oxide, 15 % of myrtanal, 10 % of myrtenol, 63 % of perillyl alcohol and
15 % of others compounds. The catalyst was separated by filtration and the reaction
mixture was separated from DMSO by extraction by water and diethyl ether. Using
elemental analysis, 50 ppm of sulfur was detected in a sample as a residue of DMSO.
For following hydrogenation it could be detrimental for some types of hydrogenation
catalyst. The catalyst was re-activated by calcination in the mixture O2:N2 = 30:70 at
550 °C for 8 h and reused in the reaction under the same conditions. The results are
listed in Table 5.
As it is visible from Table 5, the b-pinene oxide rearrangement was successful
with no significant loss of activity and selectivity after 4 cycles of reaction. All
differences are in the range of measurement error. In five experiments, 60 g of b-
Table 5 Reuse of Sn-MCM-41 as a catalyst in b-pinene oxide rearrangement
Cycle Conversion of b- Selectivity to perillyl Others Selectivity to Selectivity to
pinene oxide (%) alcohol(%) (%) myrtanal (%) myrtenol(%)

0 97.1 63.0 14.1 13.1 9.6

1 93.7 65.2 11.5 14.6 8.7
2 96.8 64.7 13.2 12.9 9.2
3 94.5 63.6 11.5 15.3 9.6

Reaction conditions: solvent DMSO, 4 h, 70 °C, volume ratio b-pinene oxide : DMSO 1:8, catalyst was
after each run filtered, washed by solvent and calcined in the mixture O2:N2 = 30:70 at 550 °C for 8 h

123
Reac Kinet Mech Cat (2016) 118:235–246 245

pinene oxide was transformed with 1.5 of Sn-MCM-41 which corresponds to

TON = 650 (turnover number defined as mol of reactant per mol of metal).

Conclusion

An attempt to study the influence of solvent basicity in b-pinene oxide

rearrangement was performed. For the studied reaction, three catalysts were used
for comparison—classic Lewis acid FeCl3 and two solid acids USY with high
amount of Brønsted acid sites and Sn modified MCM-41 characterized by the
presence of Lewis acid sites. The solvents were chosen due to their basicity defined
by the donor number. It was found that with increasing basicity of the solvent the
selectivity to perillyl alcohol increased in the case of FeCl3 and USY. FeCl3 was not
serving as typical Lewis acid, but was also decomposed in the reaction mixture of
HCl and comparable results were obtained when hydrate of this compound was used
as catalyst. Using Sn-MCM-41 the highest selectivity was obtained using DMSO,
the most basic solvent, but the trend of increasing selectivity was not observed. This
phenomenon may be explained by different mechanism of b-pinene oxide
rearrangement in the presence of Lewis acids and also by effect of other properties
of solvent like polarity. The influence of reaction conditions was tested and the
volume ratio substrate : solvent influenced significantly the reaction rate and the
selectivity. The highest selectivity to perillyl alcohol (66 %) was observed when Sn-
MCM-41 was used as catalyst together with DMSO as a solvent, 70 °C. From all
solvents DMSO is the most promising solvents for the preparation of perillyl
alcohol of all up to date tested solvents.

References
1. Chastain DE, Sanders WE, Sanders CC (1992) US5110832A
2. Corma A, Renz M, Susarte M (2009) Top Catal 52(9):1182–1189
3. de la Torre O, Renz M, Corma A (2010) Appl Catal A 380(1–2):165–171
4. Maki-Arvela P, Kumar N, Diaz SF, Aho A, Tenho M, Salonen J, Leino AR, Kordas K, Laukkanen P,
Dahl J, Sinev I, Salmi T, Murzin DY (2013) J Mol Catal A 366:228–237
5. Li Q, Wang X, Yin D, Xiao Y, Duan C (2002) Xiangliao Xiangjing Huazhuangpin 2:10–12
6. Lusticka I, Jansa P, Abelova L, Vyskocilova-Leitmannova E, Cerveny L (2013) React Kinet Mech
Cat 108(1):69–79
7. Chastain DE, Mody N, Majetich G (1999) US5994598A
8. Stekrova M, Paterova-Dudkova I, Vyskocilova-Leitmannova E, Cerveny L (2012) Res Chem
Intermed 38(8):2075–2084
9. Gutmann V (1976) Coord Chem Rev 18(2):225–255
10. Zimmermann Y, Spange S (2002) J Phys Chem 106:12524–12530
11. Haw JB, Nikolas JB, Xu T, Beck LW, Ferguson DB (1996) Acc Chem Res 29:259–267
12. Haw JF, Xu T, Nicholas JB, Gohuen PW (1997) Nature 389:832–835
13. Luo W, Deka U, Beale AM, van Eck ERH, Bruijnincx PCA, Weckhuysen BM (2013) J Catal
301:175–186
14. Schnapperelle I, Bach T (2013) ChemCatChem 5(11):3232–3236
15. Hronec M, Fulajtarova K, Liptaj T (2012) Appl Catal A 437–438:104–111
16. Fajt V, Kurc L, Cerveny L (2008) Int J Chem Kinet 40(5):240–252

123
246 Reac Kinet Mech Cat (2016) 118:235–246

17. Vrbkova E, Vyskocilova E, Cerveny L (2015) React Kinet Mech Cat 114(2):675–684
18. Emeis CA (1993) J Catal 141:347–354
19. Cox BG (2013) Acids and bases: solvent effects on acid-base strength. OUP, Oxford
20. Michigan State University, Department of Chemistry: virtual textbook of organic chemistry. http://
www.cem.msu.edu/*reusch/OrgPage/solvent.htm. Accessed 25 April 2015
21. Selli E, Forni L (1999) Microporous Mesoporous Mater 31:129–140
22. Lónyi F, Valyon J (2001) Microporous Mesoporous Mater 47:293–301
23. Garnovskii AD, Kharisov BI (1999) Direct synthesis of coordination and organometallic compounds.
Elsevier, Netherlands
24. Cotton SA (1972) Coord Chem Rev 8:185–223

123