Fe(III)-catalyzed α-terpinyl derivatives synthesis from β-pinene via reactions with hydrogen peroxide in alcoholic solutions

Article abstract of DOI:10.1039/c4ra13112f

In this study, a novel and environmentally benign Fe(iii)-catalyzed terpinyl derivatives synthesis using hydrogen peroxide in alcohol solutions (i.e. methyl, ethyl, propyl, isopropyl and butyl alcohols) was investigated. The use of Bronsted acid catalysts was avoided and β-pinene was used as the starting reactant. High conversions (ca. 90%) and combined selectivities for the α-terpineol and terpinyl alkyl ethers (ca. 70-73%) were obtained when Fe(NO₃)₃ was used as the catalyst. The role of each component catalyst system was studied with special focus on the solvent. The use of a biodegradable and renewable origin solvent (ethyl alcohol), which was added to an inexpensive and mildly toxic catalyst and a green oxidant are the main positive features of this process.

Full text of DOI:10.1039/c4ra13112f

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Fe(III)-catalyzed a-terpinyl derivatives synthesis
from b-pinene via reactions with hydrogen
peroxide in alcoholic solutions†
Cite this: RSC Adv., 2015, 5, 10529
*
M. J. da Silva, D. M. Carari and Adalberto Manoel da Silva
In this study, a novel and environmentally benign Fe(III)-catalyzed terpinyl derivatives synthesis using
hydrogen peroxide in alcohol solutions (i.e. methyl, ethyl, propyl, isopropyl and butyl alcohols) was
investigated. The use of Brønsted acid catalysts was avoided and b-pinene was used as the starting
reactant. High conversions (ca. 90%) and combined selectivities for the a-terpineol and terpinyl alkyl
ethers (ca. 70–73%) were obtained when Fe(NO3)3 was used as the catalyst. The role of each component
catalyst system was studied with special focus on the solvent. The use of a biodegradable and renewable
origin solvent (ethyl alcohol), which was added to an inexpensive and mildly toxic catalyst and a green
oxidant are the main positive features of this process.
Received 24th October 2014
Accepted 19th December 2014
DOI: 10.1039/c4ra13112f
www.rsc.org/advances
value to the feedstock.16,17 In addition to oxygen, hydrogen
peroxide is an environmentally benign oxidant that does not
Introduction
produce any waste with water as the only by-product, unlike
other corrosive and organic peroxides, which cannot be handled
easily.18,19 Moreover, hydrogen peroxide is a powerful liquid
oxidant and an inexpensive chemical commodity.20 Thus,
because of the high percentage of available oxygen, in addition
to the non-ammability and compatibility with a large scope of
catalysts, hydrogen peroxide is employed on an industrial
scale.21,22
Transition metal catalysis is of utmost importance to
convert renewable and cheap feedstock into more valuable
products. In general, homogenous catalysts are more active
and selective than heterogeneous catalysts, although many
signicant drawbacks are found when attempting to recover
and reuse these catalysts.23 Note that noble metals are broadly
used as homogeneous catalysts in organic synthesis and
industrial processes; nevertheless, developing alternative
processes based on cheap metal catalysts is a goal that should
be pursued always. Moreover, iron salts are inexpensive, envi-
ronmentally friendly and commercially available, and they
oen offer relevant advantages in terms of sustainable chem-
istry; thus, they are potential catalysts that can be employed in
organic synthesis.24,25
Considering all these aspects, in this study, we wish to
describe the synthesis of a-terpinyl derivatives through
Fe(NO3)3-catalyzed reactions of b-pinene with H2O2 in alcoholic
solutions. In particular, we have studied the reactions in ethyl
alcohol, which is a renewable, cheap, greener and less toxic
solvent. In addition, reactions were also carried out in other
alcohols (i.e. methyl, propyl, isopropyl and butyl alcohols),
which resulted in the selective formation of a-terpineol and
a-terpinyl alkyl ethers. The effects of the main parameters of
Monoterpenes are abundant raw materials present in all plants
and their valorization in the presence of active and cheap cata-
lysts is potentially a cost-effective and environmentally friendly
process that can supply feedstock to the ne chemical industries
such as avor, drug, pesticide and pharmaceutical industries.1–3
In addition, some oxygenated derivatives of monoterpenes have
agreeable organoleptic properties: a-terpinyl methyl or propyl
ethers have scents of fresh grass, a-terpinyl ethyl ether has citrus
fragrance, whereas a-terpinyl butyl ether has a woody aroma.4
Moreover, a-terpineol is a compound highly attractive for phar-
macological applications or as a food and fragrance ingredient,
and its synthesis has been extensively studied both in homoge-
neous as well as in heterogeneous catalytic processes.5–8
Therefore, a-pinene and limonene are always the starting
materials used in the synthesis of terpinyl derivatives.9,10 In
these reactions, majority of the heterogeneous catalysts are acid
solid catalysts such as zeolites, sulfated carbon activates and
heteropolyacids.9–12 Similar to a-pinene, b-pinene is also an
important constituent of turpentine oil, which is industrially
obtained by the distillation of pine tree resins or as a by-product
of the Kra cellulose pulping processes.13 Note that b-pinene is
a raw material for industrial processes that produce a number
of important chemicals and have been selected as model
molecules in several studies.14–16
In addition to the alkoxylation or hydroxylation processes,
the oxidation of monoterpenes is also a reaction used to add
Laboratory of Catalysis, Chemistry Department, Federal University of Vicosa, Minas
Gerais State, Brazil. E-mail: silvamj2003@ufv.br; Tel: +55 31 38993210
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c4ra13112f
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 10529–10536 | 10529

View Article Online
RSC Advances
Paper
reaction (i.e. reactants concentration, catalyst nature, tempera-
ture) and mainly the role of the solvent role were evaluated. A
brief discussion on iron-catalyzed isomerization and rear-
rangement of the carbon skeleton by b-pinene has been
included. Remarkably, aer a short time reaction (ca. 120 min),
the Fe(NO3)3-catalyzed reactions of b-pinene with H2O2 ach-
ieved both high conversions (ca. 80–97%) and high combined
selectivity (i.e. 60–70%) for two valuable products: a-terpineol
and a-terpinyl alkyl ethers.
Spectroscopy data of major products
1H NMR (300 MHz, CDCl3) d 1.15 (s, 3H, H-9 or H-10); 1.17 (s,
3H, H-9 or H-10); 1.20–1.28 (m, 1H, H-5); 1.42–1.54 (m, 1H, H-4);
1.55 (s, 1H, OH); 1.64 (s, 3H, H-7); 1.70–2.10 (m, 5H, H-3/H-5/
H-6); 5.36 (br s, 1H, H-2) (Fig. 1).
13C NMR (75 MHz, CDCl3): d 23.3 (C7); 23.9 (C5); 26.2 (C10),
26.8 (C3), 27.4 (C9), 31.0 (C6); 44.9 (C4); 72.7 (C8); 120.5 (C2);
133.9 (C1).
MS (m/z/int.rel.): 154/0.08 (M+c); 136/54; 121/46; 107/6; 93/37;
81/21; 67/19; 59/100; 55/18; 43/52; 41/27; 39/17; 31/10.
FT-IR (KBr) nmax cmꢀ1: 3381, 2965, 2923, 2833, 1631, 1438,
1376, 1157.
Experimental procedures
Materials and physical methods
1H NMR (300 MHz, CDCl3) d: 1.09 (s, 3H, H-9 or H-10), 1.11 (s,
3H, H-9 or H-10), 1.14 (t, 3H, J ¼ 7.0 Hz, H-12), 1.18–1.32 (m, 1H,
H-4), 1.64 (s, 3H, H-7), 1.65–1.99 (m, 6H, H-3/H-5/H-6), 3.37
(quart., 2H, J ¼ 14.0, J ¼ 7.0 Hz, H-11), 5.38 (br s, 1H, H-2) (Fig. 2).
13C NMR (75 MHz, CDCl3) d: 16.2 (C12); 22.7 (C7); 23.0 (C10);
23.4 (C9); 24.0 (C5); 26.9 (C3); 31.1 (C6); 41.6 (C4); 55.8 (C11);
76.4 (C8); 120.9 (C2); 133.9 (C1).
All the chemicals and solvents were used as received.
Fe(NO3)3$9H2O (98.5% w/w), Fe2(SO4)3$5H2O (97% w/w),
FeCl3$6H2O (97% w/w) and FeSO4$7H2O (99% w/w) were
acquired from Sigma-Aldrich. b-Pinene (99%) and the alkyl
alcohols (i.e. methyl, ethyl, propyl, isopropyl and butyl alcohols,
99.5–99.8% w/w) were also purchased from Sigma-Aldrich. The
active oxygen content of the hydrogen peroxide (aqueous solu-
tion, 34% w/w, Vetec) was quantied by titration with KMnO4
solution (ca. 0.10 mol Lꢀ1) before use.
MS (m/z/int.rel.): 180/0 (M+c); 167/2; 136/58; 121/35; 107/4;
93/22; 87/70; 79/8; 67/11; 59/100; 53/11; 43/34; 41/31; 39/12;
31/11.
FT-IR (KBr) nmax cmꢀ1: 2972, 2926, 2892, 2874, 1631, 1455,
1378, 1159, 1071.
1H NMR (300 MHz, CDCl3) d: 1.09 (s, 3H, H-13 or H-12); 1.10
(s, 3H, H-9 or H-10); 1.10 (s, 3H, H-9 or H-10); 1.12 (s, 3H, H-12
or H-13); 1.20–1.25 (m, 1-H, H-5); 1.53–1.59 (m, 1H, H-4); 1.65 (s,
3H, H-7); 1.72–1.97 (m, 5H, H-3/H-5/H-6); 3.76–3.88 (m, 1H,
H-11); 5.40 (br s, 1H, H-2) (Fig. 3).
Catalytic runs
Catalytic runs were carried out under air in a glass reactor
(50 mL) tted with a reux condenser and sampling septum
with heating using a magnetic stirrer. Typically, adequate
amount of iron catalyst and b-pinene were dissolved in alcohol
(ca. 25 mL), and the reactor temperature was adjusted to 333 K.
Then, the reaction started by the addition of H2O2 solution
(ca. 34% w/w).
The progress of the reaction was observed by taking aliquots
at regular time intervals and analyzing them using gas chro-
matography in a Shimadzu 2010 plus gas chromatograph
instrument tted with a FID detector and a CP-WAX capillary
chromatographic column (25 m ꢁ 0.32 mm ꢁ 0.30 mm). Note
that toluene was the internal standard, and the conversions
were calculated by comparing the corresponding GC peak areas
to those provided by the calibrating curve, including the peaks
of the substrate and the main pure products.
13C NMR (75 MHz, CDCl3) d: 22.8 (C10); 23.4 (C7); 23.7 (C9);
24.0 (C5); 25.0 (C13); 25.2 (C12); 27.2 (C3); 31.1 (C6); 43.0 (C4);
62.7 (C11); 77.1 (C8); 120.9 (C2); 134.1 (C1).
MS (m/z/int.rel.): 196/0 (M+c); 181/1; 136/20; 121/19; 101/32;
93/17; 81/10; 67/6; 59/100; 43/16; 41/11.
FT-IR (KBr) nmax cmꢀ1: 2985, 2928, 1548, 1450, 1351, 1119,
1004.
1H NMR (300 MHz, CDCl3) d 0, 90 (t, 3H, J ¼ 7.4 Hz, H-13);
1.11 (s, 3H, H-9 or H-10); 1.14 (s, 3H, H-9 or H-10); 1.22–1.30
(m, 1H, H-5); 1.46–1.58 (m, 2H, H-12); 1.66 (s, 3H, H-7); 1.71–
2.00 (m, 6H, H-3/H-4/H-5/H-6); 3.26 (t, 2H, J ¼ 6.7 Hz, H-11);
5.41 (br s, 1H, H-2) (Fig. 4).
13C NMR (75 MHz, CDCl3) d: d 10.8 (C13); 22.6 (C10); 22.9
(C9); 23.4 (C7); 23.8 (C12); 24.0 (C5); 26.8 (C3); 31.1 (C6), 41.9
(C4); 62.3 (C11); 76.2 (C8); 121.0 (C2); 134.0 (C1).
Products chromatographic separation and characterization
The main products of the reaction were identied by GC/MS
analyses (Shimadzu MS-QP 2010 ultra, mass spectrometer,
electronic impact mode at 70 eV, coupled to a Shimadzu 2010
plus, GC) and by comparison with the authentic samples. The
major products were isolated by column chromatography
(silica, 60 G) and then characterized by 1H, 13C NMR (Varian
300 spectrometer), and IR spectroscopy analyses (Varian FTIR
660). The NMR spectra were obtained in CDCl3 solutions using
the Varian 300 spectrometer at 300.13 and 75.47 MHz. Chemical
shis were expressed in d (ppm) relative to tetramethylsilane,
which was used as an internal reference.
Fig.
1 Structure and carbon numeration of a-terpineol (1a) (2-
(4-methylcyclohex-3-enyl)propan-2-ol).
10530 | RSC Adv., 2015, 5, 10529–10536
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
FT-IR (KBr) nmax cmꢀ1: 2967; 2926; 2828; 1438; 1378; 1362;
1074.
MS (m/z/int.rel.): 153/2; 136/34; 121/23; 93/29; 73/100; 59/12;
55/13; 43/31; 41/28.
1
The H and 13C NMR spectra (i.e. HSQC, COSY, and DEPT
135) are provided in the ESI.†
Results and discussion
General aspects
Fig. 2 Structure and carbon numeration of 4-(2-ethoxypropan-2-yl)-
1-methylcyclohex-1-ene (1b).
Nowadays, the concepts of atom economy and green chemistry
have resulted in the development of catalytic processes, in
which all the components have a minimum environmental
impact. Thus,
a system to oxidize monoterpenes using
Fe(III)/H2O2 in a methyl alcohol solution was recently described
by us.26 Nevertheless, we realized that the products obtained in
that system were alkoxylation and hydroxylation products.
Herein, the Fe(NO3)3/H2O2 system was used to synthesize
alkoxide derivatives of b-pinene, which is a renewable raw
material of great interest for several chemical industries. Our
focus was to study the reactions in ethyl alcohol, which is an
inexpensive and green solvent; moreover, we also investigated
the reactions with other alcohols.
Fig. 3 Structure and carbon numeration of 4-(2-isoproxypropan-
2-yl)-1-methylcyclohex-1-ene (1c).
A deeper assessment revealed that Fe(III) cations successfully
promoted the b-pinene isomerization under non-corrosive and
Brønsted acid-free conditions. Note that the Fe(NO3)3/H2O2/
C2H5OH system employs a Fe(III) salt catalyst, which is an
inexpensive, easy to handle and affordable reactant of low
toxicity. The alkoxylation reactions were effective mainly when
they were carried out in C2H5OH, which is a renewable solvent.
Moreover, hydrogen peroxide, an environmentally friendly and
highly efficient-atom oxidant, has an important role in the
reactions that were studied.
Fig. 4 Structure and carbon numeration of 4-(2-proxypropan-2-yl)-
1-methylcyclohex-1-ene (1d).
Assessment of the role of system components in the Fe(NO3)3-
catalyzed b-pinene reaction with hydrogen peroxide in
C2H5OH
MS (m/z/int.rel.):196/0 (M+c); 181/1; 136/26; 121/20; 101/46;
93/18; 81/9; 67/6; 59/100; 43/19; 41/13.
FT-IR (KBr) nmax cmꢀ1: 2961, 2918, 2870 (ns e nas –CH2
e
In general, the majority of reactions that synthesizes a-terpineol
starting from a-pinene work with water as the nucleophilic
reactant and Brønsted acid as the catalyst.10–12 Under acidic
conditions, a-pinene undergoes carbon skeletal rearrangement
to generate a tertiary carbocation (Fig. 6). Nucleophiles as water
or alcohol present in the reaction medium may then attack this
intermediate, resulting in a-terpineol or alkyl ether as the main
product.
–CH3); 1664 (n C]C); 1458, 1367 (d-CH2 e –CH3); 1078 (nas C–O);
1003 (ns C–O).
1H NMR (300 MHz, CDCl3) d: 5.39 (s, 1H, H-2); 3.17 (s, 2H,
H-11); 2.05–1.65 (m, 6H, H-3/H-4/H-6); 1.64 (s, 3H, H-11); 1.35–
1.10 (m, 1H, H-5); 1.09 (s, 6H, H-9/H-10) (Fig. 5).
13C NMR (75 MHz, CDCl3) d: 134.0 (C1); 120.8 (C2); 76.6 (C8);
48.7 (C11); 41.5 (C4); 31.1 (C6); 26.8 (C3); 23.9 (C5); 23.4 (C7);
22.3 (C9 ou C10); 21.8 (C10 ou C9).
Herein, Fe(III) cations were used as the Lewis acid catalyst,
whereas b-pinene was used as the substrate instead of a-pinene.
Fig. 5 Structure and carbon numeration of 4-(2-methoxypropan- Fig. 6 Brønsted acid-catalyzed skeletal rearrangement of a-pinene
2-yl)-1-methylcyclohex-1-ene (1e).
(adapted).27
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 10529–10536 | 10531

View Article Online
RSC Advances
Paper
In addition, as an aqueous solution of hydrogen peroxide
(34% w/w) was used, the two nucleophiles (i.e. H2O2 or H2O)
could contribute for the formation of a-terpineol; moreover, as
iron nitrate is also an oxidant, it might oxidize the olen. All
these key aspects are addressed in Table 1 with the reactions
being carried out under different experimental conditions.
Furthermore, we found that, in the absence of Fe(NO3)3
catalyst, conversion of b-pinene was not detected, regardless of
the reaction time (Runs 1 and 4, Table 1) or the presence of
H2O2 (Run 2, Table 1). In addition, despite the low conversion
achieved (ca. 21%) in ethyl alcohol solutions, Fe(NO3)3 con-
verted b-pinene to a-terpinyl ethyl ether (1b) with high selec-
tivity (ca. 86%) in the absence of water or peroxide (Run 3,
Table 1). Remarkably, the combination of Fe(NO3)3/H2O2 con-
verted the b-pinene into two main products with good selectivity
(ca. 32% and 44% for a-terpineol (1a) and a-terpinyl ethyl ether
(1b), respectively) with high conversion (ca. 84%, Run 6, Table
1). It can be concluded that the presence of H2O2 in the
Fe(NO3)3-catalyzed reactions is more favorable to form
a-terpineol compared with the presence of H2O alone; more-
over, the selectivity of a-terpineol increased considerably
(ca. 8–32%) when we replaced water by hydrogen peroxide.
Moreover, water required a larger reaction time compared with
the presence of peroxide (ca. 24 against 8 hours, Runs 5 and 6,
Table 1).
Fig. 7 Main products obtained in the Fe(NO3)3-catalyzed b-pinene
reaction with H2O2 in C2H5OH.
There are some recent studies that describe the activity of
iron catalysts under heterogeneous conditions (i.e. iron-
modied mesoporous silicates; Fe-MCM) in acid-catalyzed
reactions such as a-pinene oxide isomerization.29,30
Gusevskaya et al. showed that the acidic properties of MCM
samples were signicantly increased by the addition of iron
cations, which were nominated by them as catalytically active
species (i.e. Fe+2-MCM or Fe+3-MCM) in the acid-catalyzed
transformations undergone by the substrate (i.e. isomeriza-
tion, skeletal rearrangement, opening ring).30 Actually, they
veried that the unmodied MCM-41 samples have no suffi-
cient acidity to promote any transformation of the moiety of
a-pinene oxide.
The mechanism of Fe(III)-catalyzed isomerization of b-pinene
remains unclear. Although H+ cations could be generated
herein from the reaction of H2O2 with Fe(III) species via the
Effect of catalyst concentration on the Fe(NO3)3-catalyzed b-
pinene reaction with hydrogen peroxide in C2H5OH
Habber–Weiss mechanism, it is important to note that, even in To evaluate the activity of Fe(NO3)3 catalyst in the reactions in
the absence of peroxide, the a-terpinyl ethyl ether (1b) was C2H5OH, were accomplished reactions within broad range of
selectively obtained probably via reaction between the tertiary concentration (i.e., 0.010 to 0.612 mmol). The data of conver-
carbocation (Fig. 6) and ethyl alcohol.28 Indeed, the high sion, TON and selectivity are shown in Table 2.
selectivity for the formation of a-terpinyl tertiary carbocation
Any signicant changes in reaction selectivity were triggered
derivatives (i.e., 1a and 1b products) may be attributed to the by varying the catalyst concentration. The formation of these
stabilization of this intermediate by the polar solvent (i.e. ethyl products always occurred in virtually equimolar amounts (i.e. 1a
alcohol) (Fig. 7). Nevertheless, the mechanism for the formation
of this carbocation in the presence of only Fe(NO3)3 and alcohol
Table 2 Effect of the catalyst concentration on the Fe(NO3)3-cata-
lyzed b-pinene reaction with H2O2 in CH3CH2OHa
is being investigated and is unclear at present.
Selectivityc (%)
Table
1 Fe(NO3)3-catalyzed b-pinene reaction with hydrogen
Run Fe(NO3)3 (mmol) TON Convers.b (%) Isomers 1a 1b ni
peroxide in C2H5OHa
Selectivityd (%)
Run Fe(NO3)3 (mmol) H2O2 (mmol) Convers.d (%) 1a 1b Iso. ni
1
2
3
4
5
6
7
8
9
0.612
0.412
0.212
0.112
0.070
0.040
0.030
0.020
0.010
8
11
21
38
56
80
95
143
255
475
97
94
90
84
79
64
57
57
51
95
17
16
15
16
18
17
18
17
18
14
34 34 15
36 35 13
38 35 12
37 34 13
35 35 12
33 37 13
33 37 12
32 38 13
38 32 12
39 34 13
1b
2
—
—
0.04
—
0.21
0.21
—
30.0
—
—
—
0
0
21
0
92
84
—
—
—
—
8
—
—
86 10
—
—
—
4
3
4c
5c
6
—
36 33 23
20.0
32 44 17
7
10d 0.010
a
a
Reaction conditions: b-pinene (5 mmol), ethyl alcohol solution
Reaction conditions: b-pinene (5 mmol), H2O2 (30.0 mmol), ethyl
b
b
(15 mL), 333 K, 8 hours reaction. Run carried out for 24 hours. alcohol solution (15 mL), 333 K, 8 hours reaction. Determined via
c
c
Test catalytic carried out for 24 hours in the presence of water GC analyses. Determined via GC-MS analyses; a-terpineol (1a),
d
(ca. 20.0 mmol). Conversion determined by CG; a-terpineol (1a), a-terpinyl ethyl ether (1b); terpinene isomers (ca. equimolar amounts
a-terpinyl ethyl ether (1b); terpinene isomers (equimolar amounts of of a- and g-terpinene); ni (complex mixture of minority products).
d
a- and g-terpinene); and ni (complex mixture of minority products).
24 hours reaction.
10532 | RSC Adv., 2015, 5, 10529–10536
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
and 1b) with the combined selectivity ranging between ca. 68%
and 73%.
Table 2 shows that the reaction conversions decreased when
the Fe(NO3)3 load is reduced, which demonstrates that the
reaction equilibrium was not reached aer 8 hours. In addi-
tion, regardless of the catalyst concentration, any signicant
changes on selectivity occurred aer this reaction time.
Moreover, the TON of reactions with low catalyst concentration
was higher than ones obtained in the Bronsted acid-catalyzed
reactions.31 When we carried out the reactions with the lowest
catalyst concentration for 24 hours (run 10, Table 2), we
observed that high conversion was attained (ca. 97%). This
suggests that the catalyst remained active even if the reaction
time was longer.
To assess the catalyst activity we are taking into account the
reaction initial rates achieved aer 30 minutes using different
catalyst concentrations. These data were obtained from the
slopes of the conversion versus time data collected at the
beginning of the reaction (ca. 30 min). From the curves shown in
Fig. 8, we built a linear plot of ln (reaction initial rate) calculated
in terms of mmol b-pinene consumed/3600 seconds versus ln
(Fe(NO3)3 concentration), expressed in mmol, as shown in Fig. 9.
In principle, the slope of the curve ln (conversion) versus ln
(Fe(NO3)3 concentration) provided the order in relation to the
Fig. 9 Order of reaction with respect to catalyst concentration: (a)
0.010–0.612 mmol; (b) 0.010–0.070 mmol; and (c) 0.112–
0.612 mmol.
Several aspects may cause this change in behavior. Prob-
catalyst concentration. The data shown in Fig. 9(a) suggest that ably, the peroxide concentration variation or pH values are key
the effect of the catalyst concentration on the reaction conver- factors that may affect these reactions. Volumetric titration
sion varied in a non-linear way within the concentration range measurements were carried out with KMnO4, and they showed
of 0.010–0.612 mmol. An attempt was made to linearize the that aer the rst 30 minutes of the reaction, almost all
results with a low linear coefficient (R2 ¼ 0.90), which was hydrogen peroxide was consumed. It was possible that the
omitted by simplication (Fig. 9(a)).
peroxide intermediates could have formed in the reaction.
However, a careful analysis reveals that a split in two of Shul'pin et al. proposed a test to verify the presence of
the concentration ranges seems be more adequate, as shown hydroperoxides intermediates that are not detectable by GC
in Fig. 9b and c. The slope of the curve ln (reaction initial analysis via adding phosphine.35 In accordance with these
rate) versus ln (Fe(NO3)3 concentration) was approximately 0.53 authors, the hydroperoxydes are decomposed by the phosphine
(R2 ¼ 0.985) when a low concentration of Fe(NO3)3 was resulting in alcohol, increasing the GC peaks area of these
employed (ca. 0.010 to 0.070 mmol). In addition, using a catalyst compounds. Thus, we added phosphine to the aliquots of the
load higher than 0.070 (0.112–0.612 mmol), a dependence equal reaction to verify if the GC peak of a-terpineol has undergone
to 0.19 (R2 ¼ 0.990) in relation to the Fe(NO3)3 concentration changes. We found that no alkylperoxide intermediate was
can be considered (Fig. 9c). The high linearity coefficient indi- formed.
cates that this was an acceptable approach.
Although during the beginning of the reaction, the pH values
measured were almost constant (ca. 1.35–1.44 aer 30 minutes
reaction), this initial value depends on the initial amount of
Fe(NO3)3. We believe that both the Fe(III) and (H+) ions may act
to catalyze this reaction. Thus, when we have a lower concen-
tration of cations Fe(III), occurs a lower generation of cations H+;
consequently, the dependence in relation to Fe(NO3)3 catalyst is
higher. However, in the presence of a high amount of iron
catalyst, a lot of H+ ions are produced; therefore, a low depen-
dence in relation to the iron catalyst concentration was
observed.
Analyzing the kinetic curves of Fig. 8, we can note the runs
with the high reaction initial rates, were those that reached the
high conversions aer 8 hours reaction.
The isomerization of monoterpenes has been described in
the presence of Lewis or Brønsted acid catalysts.32,33 In this
study, H+ cations were not added at the beginning of the reac-
tion because only Fe(III) cations were used as catalysts. However,
Fig. 8 Kinetic curves of Fe(NO3)3-catalyzed b-pinene reaction with
H2O2 in C2H5OH using different catalyst load. Reaction conditions:
b-pinene (5 mmol), H2O2 (30.0 mmol), ethyl alcohol solution (15 mL),
333 K, 8 hours reaction.
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 10529–10536 | 10533

View Article Online
RSC Advances
Paper
it is well known that Fe(III) cations can react with peroxides to
give Fe(II) and H+ cations and OOH radicals;34 consequently,
either Lewis or Brønsted catalysts may have promoted the
b-pinene isomerization.
The kinetic curves show that the initial rates of the reaction
were notably affected by the decrease in the catalyst concen-
tration (Fig. 8). However, when the catalyst load was between
21.2 and 0.112 mmol (i.e., 12.0–4.0 mol%), high conversions
between 91% and 97% were achieved. Moreover, even though a
decrease in the catalyst concentration resulted in a lower
conversion, the TONs obtained were higher than those achieved
in the Brønsted acid-catalyzed a-pinene alkoxylation/
hydroxylation reactions.9–11
Fig. 10 Effect of H2O2 amount on the Fe(NO3)3-catalyzed b-pinene
reaction with H2O2 in C2H5OH solution. Reaction conditions:
b-pinene (5.0 mmol); Fe(NO3)3 (0.212 mmol); ethyl alcohol (15 mL);
333 K; 4 hour reaction.
Fe(NO3)3-catalyzed b-pinene reactions with hydrogen peroxide
in C2H5OH: effect of hydrogen peroxide concentration
The effect of H2O2 concentration on the conversion and selec- H2O2 was lower than 30.0 mmol. This can be attributed to the
tivity of the b-pinene reactions was investigated in the range of conversion rate of the alkyl intermediate to products, which is
30.0–5.0 mmol, and the main results are summarized in Table 3. lower when a low concentration of peroxide is used.
An increase in the amount of H2O2 aqueous solution may
favor the formation of water addition products (i.e. a-terpineol)
as the water content in the reaction medium was also increased.
Indeed, the selectivity for a-terpineol decreased from ca. 9 to
26% when the H2O2 load was increased from 5 to 30 mmol.
Moreover, competitive reactions such as isomerization seem to
have been favored, and a higher selectivity was observed for
isomers when the amount of aqueous hydrogen peroxide was
lowered from 30.0 to 5.0 mmol.
In addition, with the change in the conversion and reaction
selectivity, an increase in the H2O2 concentration also resulted
in an improved initial rate of the reaction, as shown in Fig. 10.
In addition, proceeding with H2O2 concentration equal to
20.0 mmol for more than 4 hours (i.e. 8 hours, Table 2), it was
possible to observe that the reaction selectivity of the main
products (i.e. 1a and 1b products) were enhanced (i.e. 65–73% of
combined selectivity).
Fe(NO3)3-catalyzed b-pinene reaction with H2O2 in CH3OH:
effect of temperature
The decrease in temperature notably affected the nal conver-
sion of the reactions: while 85% was achieved the reaction was
heated to 333 K, only a poor conversion of 37% was achieved at
room temperature (Table 4). Similarly, the selectivity of the
products also drastically changed: at room temperature, the
selectivity for a-terpinyl ethyl ether (1b) was the lowest (Run 1,
ca. 7%). It should be noted that the solubility of Fe(NO3)3 at
room temperature in ethyl alcohol was only partial, which is a
fact that certainly compromised the catalyst activity.
The kinetic curves obtained at different temperatures show
that the initial rates of the reaction were relatively close within
the initial 45 minutes of reaction. Subsequently, the run per-
formed at the highest temperature was visibly more favored
(Fig. 11).
In general, we note that unlike Fig. 8, in which aer the
initial period of the reaction, the conversion remains almost
constant, in Fig. 10, we observe a slight increase in the
conversion throughout the reaction when the concentration of
Table 4 Effect of temperature on the conversion and selectivity of the
Fe(NO3)3-catalyzed b-pinene reaction with hydrogen peroxide in ethyl
alcohola
Table 3 Effect of H2O2 concentration in Fe(NO3)3-catalyzed b-pinene
reactions in C2H5OHa
Selectivityc (%)
Selectivityc (%)
Run Temperature (K) Conversionb (%) Isomers 1a 1b ni
Run
H2O2 (mmol)
Conversionb (%)
Isomers
1a
1b
ni
1
2
3
4
298
313
323
333
37
49
61
85
21
25
25
23
33
7
31
1
2
3
4
30.0
20.0
10.0
5.0
85
76
67
73
17
18
24
30
26
28
17
9
28
37
40
41
29
17
19
20
34 24 17
35 26 15
36 27 14
a
Reaction conditions: b-pinene (5.0 mmol); Fe(NO3)3 (0.212 mmol);
hours reaction.
Determined via GC analyses. Determined via GC-MS analyses;
a
Reaction conditions: b-pinene (5.0 mmol), Fe(NO3)3 (0.21 mmol), ethyl H2O2 (20.0 mmol); ethyl alcohol (15 mL);
8
b
c
alcohol solution (15 mL), 333 K, 4 hours reaction. b Determined via GC
c
analyses. Determined via GC-MS analyses; a-terpineol (1a), a-terpinyl a-terpineol (1a), a-terpinyl ethyl ether (1b); terpinene isomers
ethyl ether (1b); terpinene isomers (ca. equimolar amounts of a- and (ca. equimolar amounts of a- and g-terpinene); ni (complex mixture of
g-terpinene); ni (complex mixture of minority products).
minority products).
10534 | RSC Adv., 2015, 5, 10529–10536
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
Fig. 11 Effect of temperature on the Fe(NO3)3-catalyzed b-pinene
reaction with H2O2 in ethyl alcohol. Reaction conditions: b-pinene
(5.0 mmol), Fe(NO3)3 (0.212 mmol), ethyl alcohol (15.0 mL), 6 h
reaction.
Fig. 12 Kinetic curves of the b-pinene reactions with H2O2 in ethyl
alcohol with varying amount of the iron catalyst. Reaction conditions:
b-pinene (5.0 mmol); catalyst (0.612 mmol); H2O2 (20.0 mmol); ethyl
alcohol (15 mL); 333 K; 8 h.
Table 6 Effects of the nature of alcohol in the Fe(NO3)3-catalyzed
b-pinene reaction with H2O2
Effect of the nature of iron catalyst on the reaction of b-pinene
with H2O2 in C2H5OH
a
Selectivityc (%)
Table 5 shows the results obtained using different iron salts as
catalysts. The solubility of the catalysts drastically affected their
efficiency, and consequently compromised the conversions of
the b-pinene throughout the reactions.
Run Alcohol
Conver.b (%) Isom. 1a Alkyl terpinyl ether ni
1
2
3
4
5
Methyl
Ethyl
Propyl
Isopropyl 75
Butyl <5
97
84
88
17
23
17
26
—
18 50
36 27
33 34
15 36
15
14
16
23
100
The lowest conversion was achieved in the presence of
Fe2(SO4)3 catalyst, which was virtually insoluble in the reaction
medium (Run 3, Table 5). Similarly, FeSO4 was not completely
soluble, and thus the conversion was low (Run 4, Table 5). The
higher iron content per mol of iron(III) sulfate explains the
higher activity compared with that obtained using iron(II)
sulfate. This different solubility hampers a straight comparison
between the activity of these catalysts or a possible discussion
about the effect of the iron oxidation state on the reactions. In
addition, both catalysts FeCl3 and Fe(NO3)3 were totally soluble;
however, when they were used at the same concentration,
Fe(NO3)3 was much more efficient than FeCl3 (Fig. 12).
Although the motive for this high activity remains unclear,
we believe that the nitrate anions might have become more
electrophilic as compared with the Fe(III) cations, which favors
the cleavage of the 4 members-ring of b-pinene and could be a
key step in the formation 1a or 1b.
—
—
a
Reaction conditions: b-pinene (5.0 mmol); catalyst (0.212 mmol); H2O2
(20.0 mmol); alcoholic solution (15 mL); 333 K; 8 h. b Determined via GC
analyses. Determined via GC-MS analyses; a-terpineol (1a), alkyl
c
terpinyl ether; isom. (ca. equimolar amounts of a- andg-terpinene); ni
(complex mixture of minority products).
Effect of the nature of alcohol on the Fe(NO3)3-catalyzed b-
pinene reaction with H2O2
The impact of alcohol chain on conversion and selectivity of the
reactions was assessed (Table 6). As the catalyst was insoluble,
the reaction does not proceed in butyl alcohol.
The results obtained in the reaction with methyl alcohol are
in agreement with those previously published by us.26 In
Table 5 Effects of the nature of iron catalyst on the b-pinene reaction
with H2O2 in ethyl alcohola
Selectivityc (%)
Run
Catalyst
Conversionb (%)
1a
1b
Isomers
ni
1
2
3
4
Fe(NO3)3
FeCl3
Fe2(SO4)3
FeSO4
91
63
55
32
34
16
20
13
34
22
27
17
17
11
16
12
15
51
37
58
a
Reaction conditions: b-pinene (5.0 mmol); catalyst (0.612 mmol); H2O2
b
(20.0 mmol); ethyl alcohol (15 mL); 60 ꢂC; 8 h. Determined via GC
analyses. Determined via GC-MS analyses; a-terpineol (1a), a-terpinyl
c
Fig. 13 Kinetic curves of the b-pinene reactions with H2O2 in alcoholic
solutions. Reaction conditions: b-pinene (5.0 mmol); catalyst
(0.212 mmol); H2O2 (20.0 mmol); ethyl alcohol (15 mL); 333 K; 8 h.
ethyl ether (1b); terpinene isomers (ca. equimolar amounts of a- and
g-terpinene); ni (complex mixture of minority products).
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 10529–10536 | 10535

View Article Online
RSC Advances
Paper
addition, although not novel, they are included herein for 10 J. E. Castanheiro, I. M. Fonseca, A. M. Ramos, R. Oliveira and
comparison. Note that methyl alcohol quickly reacts to give the J. Vital, Catal. Today, 2005, 104, 296.
respective alkyl ether with the highest selectivity and an almost 11 M. C. Avila, H. F. Norberto, N. F. Esther and I. P. Martha, J.
complete conversion (Run 1, Table 6). Chil. Chem. Soc., 2008, 53, 1460.
Ethyl and propyl alcohols reacted with similar rates, which 12 M. C. Avila, A. C. Nora, E. R. Castelon and A. J. Lopez, J. Mol.
provided the respective alkyl ethers with very close selectivity Catal. A: Chem., 2010, 32, 106.
(Fig. 13). However, unlike isopropyl alcohol, the steric 13 U. Neuenschwander, F. Guignard and I. Hermans,
hindrance from the tertiary carbon bonded to the hydroxyl ChemSusChem, 2010, 3, 75.
group hampered the formation of the a-terpinyl isopropyl ether 14 L. M. Pisarenko, D. A. Krugovov, A. N. Shchegolikhin and
(Run 4, Table 6) thus, we can conclude that the system of O. T. Kasaikina, Russ. Chem. Bull. Int. Ed., 2008, 57, 83.
Fe(NO3)3/H2O2 is efficient in converting alcohol of short carbon 15 P. Gallezot, Catal. Today, 2007, 121, 76.
chains and carbon less sterically hindered from their terpinyl 16 L. Menini, M. J. da Silva, M. F. F. Lelis, J. D. Fabris,
derivatives (i.e. alkoxides or alcohol products).
R. M. Lago and E. V. Gusevskaya, Appl. Catal., A, 2004, 269,
117.
17 M. J. da Silva, P. A. Robbles-Dutnhefner, L. Menini and
E. V. Gusevskaya, J. Mol. Catal. A: Chem., 2003, 201, 71.
18 D. Narog, A. Szczepanik and A. Sobkowiak, Catal. Lett., 2008,
120, 320.
19 J. M. Campos-Martin, G. Blanco-Brieva and J. L. G. Fierro,
Angew. Chem., Int. Ed., 2006, 45, 6962.
20 R. Noyori, M. Aoki and K. Sato, Chem. Commun., 2003, 1977.
21 P. T. Anastas, L. B. Bartlett, M. M. Kirchhoff and
T. C. Williamson, Catal. Today, 2000, 55, 11.
22 R. A. Sheldon, Chem. Commun., 2008, 3352.
23 D. Habibi, A. R. Faraji, M. Arshadi and J. L. G. Fierro, J. Mol.
Catal. A: Chem., 2013, 372, 90.
24 L.-X. Liu, Curr. Org. Chem., 2010, 14, 1099.
25 S. Chang-Liang, L. Bi-Jie and S. Zhang-Jie, Chem. Rev., 2011,
111, 1293.
Conclusion
A novel process to convert b-pinene to terpinyl derivatives using
non-corrosive Brønsted acid-free conditions was developed. In
the presence of catalytic amounts of Fe(NO3)3 dissolved in
alcoholic solutions containing hydrogen peroxide, b-pinene was
converted to a-terpineol and a-terpinyl alkyl ethers with high
conversions (ca. 90%) and combined selectivity ranging from
60% to 80% for the two main products (i.e. a-terpinyl alkyl
ethers and a-terpineol). The high TON achieved reveals that this
homogeneous catalyst is notably more effective than the other
catalysts described in previously reported studies. Among the
iron salts that were investigated, Fe(NO3)3 was the most active
catalyst. The use of a biorenewable solvent (ethyl alcohol), an
environmentally benign oxidant (i.e. H2O2), and an inexpensive
catalyst are positive features of this process.
26 M. J. da Silva and D. M. Carari, Catal. Lett., 2014, 144, 615.
27 H. Hartati, M. Santoso, S. Triwahyono and D. Prasetyoko,
Bull. Chem. React. Eng. Catal., 2013, 8, 14.
28 Y. N. Kozlov, A. D. Nadezhdin and A. P. Purmal, Int. J. Chem.
Kinet., 1974, 6, 383.
Acknowledgements
The authors are grateful for the nancial support from CAPES,
CNPq and FAPEMIG (Brazil).
29 N. A. Fellenz, J. F. Bengoa, S. G. Marchetti and A. Gervasin,
Appl. Catal., A, 2012, 435, 187.
30 J. V. Coelho, A. L. P. de Meireles, K. A. S. Rocha, M. C. Pereira,
L. C. A. Oliveira and E. V. Gusevskaya, Appl. Catal., A, 2012,
443, 125.
Notes and references
1 W. Schwab, C. Fuchs and F.-C. Huang, Eur. J. Lipid Sci.
Technol., 2013, 115, 3.
´
31 M. Roman-Aguire, A. F. W. De la Tore-Saenz, A. Robau-
2 K. A. D. Swi, Top. Catal., 2004, 27, 143.
3 R. Saladino, V. Neri, A. R. Pelliccia and E. Mincione,
Tetrahedron, 2003, 59, 7403.
4 M. Stanciulescu and M. Ikura, J. Anal. Appl. Pyrolysis, 2006,
75, 217.
5 S. P. Tia, Food Chem. Toxicol., 2008, 46, 5275.
6 D. H. Pybus and C. S. Sell, The chemistry of fragrances, RSC
Paperbacks, Cambridge, 2001.
´
´
Sanchez and A. A. Elguezabal, Catal. Today, 2005, 107, 310.
´
´
´
32 N. Flores-Holguın, A. Aguilar-Elguezabal, L. M. Rodrıguez-
Valdez and D. Glossman-Mitinik, J. Mol. Struct., 2008, 854,
81.
33 Y. S. Solkina, S. Reshetnikov, M. Estrada, A. V. Simakov,
D. Y. Murzin and I. L. Simakova, Chem. Eng. J., 2011, 176, 42.
34 M. Nakanishia and C. Bolma, Adv. Synth. Catal., 2007, 349,
861.
7 Y. Yuasa and Y. Yoko, Org. Process Res. Dev., 2006, 10, 1231.
8 M. K. Yadav, M. V. Patil and R. V. Jasra, J. Mol. Catal. A:
Chem., 2009, 297, 101.
9 J. Vital, A. M. Ramos, I. F. Silva, H. Valente and
J. E. Castanheiro, Catal. Today, 2001, 67, 217.
35 G. B. Shul'pin, Y. N. Kozlov, G. V. Nizova, G. Suss-Fink,
S. Stanislas, A. Kitaygorodskiy and V. S. Kulikova, J. Chem.
Soc., Perkin Trans. 2, 2001, 1351.
10536 | RSC Adv., 2015, 5, 10529–10536
This journal is © The Royal Society of Chemistry 2015