A.A. Mekkaoui, H. Ben El Ayouchia, H. Anane et al.
Journal of Molecular Structure 1235 (2021) 130221
ane were purchased from commercial sources of chemical reagent
grade and used as received, unless otherwise indicated.
4
.2. Preparation of limonene oxide (2)
The limonene oxide (2) was obtained by an epoxidation reac-
tion of R-limonene (1) [22]. A certain amount of ruthenium com-
plex (S/C = 650, 0.012 mmol) in 5 ml of dichloromethane is in-
troduced into a 100 ml three-necked flask equipped with a ball
condenser surmounted by an oil bubbler. The mixture is stirred
at room temperature for 5 min under a stream of oxygen before
adding isobutyraldehyde (1.72 mg, 24 mmol). After 4 min of stir-
ring, limonene (7.8 mmol) is added, and the mixture is stirred for
1.5 h under a stream of oxygen. The obtained residue is filtered
through a column of basic alumina (5% Et O/hexane) affording the
2
limonene oxide (2) in a yield of 91%. The trans/cis ratio are deter-
mined by GC analysis.
4
.3. Ozonolysis of limonene oxide (2)
A solution of limonene oxide (2) (1 g, 6.58 mmol) mixture in
dry CH Cl2 (25 mL) was magnetically stirred at −78 °C and a
2
stream of ozone in oxygen was bubbled through it, and the course
of the reaction was monitored by TLC. When the starting material
was consumed (3 h), the solution was flushed out with an argon
stream for eliminating the ozone excess. Then triphenylphosphine
(
2 g, 7.63 mmol) was added at −78 °C. The solution was stirred
for 4 h, letting the temperature increase to room temperature.
Removal of the solvent under vacuum afforded a crude product
which was purified by column chromatography (5% Et O/hexane)
2
to afford the epoxide 3 (932 mg, 92%) as colourless oil. The mix-
ture was identified by NMR and GC–MS analysis (Supplementary
Material, Figures S1-S7) and was found to be the same as reported
by Delay et al. [23].
Fig. 4. Optimized geometries (in DCM) of the stationary points involved in the de-
oxygenation reactions of the epoxide 3 with ZnCl2. Distances are given in A˚ .
The polar nature of the deoxygenation reaction was analysed
evaluating the GEDT at the TSs. Along this reaction, the values of
the GEDT are 0.10, 0.09, 0.005 and 0.004 e at TS1, TS2, TS3 and
TS4, respectively. The calculated GEDT values indicate that TS1 and
TS2 have some polar character. While the GEDT at the TS3 and
TS4 is negligible, indicating the non-polar nature of the protona-
tion/elimination reaction of iodohydrin.
4
.4. Limonaketone (4) from (3)
The preparation of limonaketone (4) was performed by a mod-
ified procedure of Cornforth et al. [24]. A solution of epoxide
(0.1 g, 0.65 mmol), in CH Cl2 (2.5 mL) was added at 0 °C
3
2
over 5 min to a stirred mixture of a zinc precursor (0.127 g,
.95 mmol, 3 eq.), NaI (0.97 g, 6.49 mmol, 10 eq.), NaOAc (0.266 g,
.24 mmol,), and acetic acid (0.7 mL) in CH Cl . After being stirred
1
The geometries of the TSs and the intermediate involved in the
deoxygenation reactions of the epoxide 3 with ZnCl2 are given in
Fig. 4. At the complex 1 associated with the nucleophilic attack of
the epoxide 3 by oxygen atom O1 at the zinc atom of ZnCl2 com-
3
2
2
at room temperature for 3 h, the mixture was filtered and ex-
tracted with CH Cl . The separated organic layer was washed with
2
2
˚
water, saturated NaHCO3, and saturated NaCl. The organic layer
was dried over MgSO4, concentrated to give pure limonaketone (4)
(0.89 g) (Supplementary Material, Figures S8-S12).
plex, the length of the Zn–O1 forming bond is 2.02 A and the C2–
˚
O1 bond length is 1.502 A. At the first TS of this stepwise mech-
˚
anism, the lengths of the I–C2 forming bond is 3.06 A, while the
Limonaketone (4): [α]D25= + 120.7 (C = 2,3; CHCl3); 1H NMR
˚
Zn–O1 and C2–O1 bonds length becomes 1.936 and 2.387 A re-
(
300 MHz, CDCl ): δ 5.34 (m, 1H; =CH), δ = 2.48 (m, 1H; -CH),
spectively. The Zn–O1 and I–C2 bonds length at the corresponding
3
˚
δ = 2.15 (m, 2H; -CH2), δ = 2.12 (s, 3H; -CH2), δ = 1.96 (m, 2H;
intermediate complex (IC1) are 1.923 and 2.228 A respectively. At
13
-
CH ), δ = 1.85 (m, 2H), 1.60 (d, 3H; -CH ). C NMR (75 MHz,
the second TS associated with the protonation of IC1 by acetic acid,
2
3
˚
CDCl3): δ: 211.57, 133.70, 119.24, 47.16, 29.45, 27.83, 27.01, 24.87,
the lengths of the O1–H forming bond is 1.370 A. At the TS3 asso-
2
3.31. MS (EI): m/z = 138 [M]+
.
ciated with the protonation of iodohydrin by hydrogen iodide, the
˚
lengths of the O1–H forming bond is 1.504 A. Finally, at TS4 the
˚
4.5. Computational details
lengths of the I–I forming bond is 3.11 A.
The quantum chemical calculations reported in this paper were
performed using M06–2X functional with the 6–311G(d, p) basis
set for C, N, O, and H nuclei and the LANL2DZ type basis set
for Zn included in the GAUSSIAN 09 package [30]. The stationary
points were characterised by frequency calculations at the same
computational level in order to ensure all reactants, intermedi-
ates and products have no imaginary frequency and that TSs have
one imaginary frequency. In order to confirm that TS is correctly
connect to two associated minima, the intrinsic coordinate (IRC)
4
. Experimental section
4
.1. Materials
(
R)-(+)-limonene (97%), triphenylphosphine (Pφ ), zinc powder
3
(
Zn), sodium iodide (NaI), sodium acetate (NaOAc), magnesium sul-
fate (MgSO ), Sodium bicarbonate (NaHCO ), acetic acid (AcOH),
4
3
dichloromethane (CH Cl ), diethyl ether, ethyl acetate and hex-
2
2
5