M. Casiello et al. / Journal of Molecular Catalysis A: Chemical 381 (2014) 99–106
101
The carbonylation of glycerol under the other conditions dis-
cussed in the paper were carried out in the same way. The
comparative experiment carried out under the safety conditions
(CO:O2 = 15:1 molar ratio), was performed on a DMA solution
(3.0 mL) of glycerol (2,6 mmol), catalyst (CuCl2, 0.26 mmol) and
pyridine (0.13 mmol). The vial was introduced into the autoclave,
which was sealed, purged and charged with O2 (0.25 MPa) and CO
up to a total pressure of 4 MPa. Under these conditions. The mixture
was heated at 130 ◦C and allowed to react for 8 h. After this time,
the reaction mixture was analyzed by IR and GC–MS for assessing
conversions and selectivities as reported below.
CH2CH2OSiMe3+), 73 (97, SiMe3+), 59(16, MeCO2+), 43 (83,
MeCO+).
O
CH3 C O
O Si
O
Si
S-3b
Tetramethyl urea (4): m/z (%) 116 (6, M+), 87 (1, M+−2Me), 72
(100, C(O)NMe2+), 44 (23, Me2N+).
O
Me2N C NMe2
4
Silylated 2,3-dihydroxypropyl dimethylcarbamate (S-5): m/z (%)
292 (2, M+−Me), 263 (1, M+−NMe2), 218 (2, M+−HOC(O)NMe2),204
(52, M+−CH2OSiMe3), 147 (27, CH2CHOCH2OSiMe3+), 73 (51,
SiMe3+), 72 (100, C(O)NMe2+).
To obtain reproducible and resolved chromatographic peaks
of 1 and 2 we decided to convert them into the correspond-
ing silylated compounds. In a typical derivatization procedure
[17],
a
weighed amount of the reaction mixture (150 mg
10 mL centrifuge tube with excess
O
ca.) was treated into
a
Me2N C O
O Si
amounts (1.00 g ca.) of the silylating reagent hexamethyldisi-
lazane:trimethylchlorosilane:pyridine in a 3:1:9 weight ratio. The
ture for 30 min causing the precipitation of NH4Cl as a white solid.
Then, 50 mg of 4-methylanysole was added to the mixture as an
external standard. After centrifugation at 2500 rpm for 5 min, the
clear supernatant was analyzed by GC–MS to assess the conver-
formed carbonate 2.
Besides 1 and 2, the silylation procedure allowed the detection of
the main by-products arising from side-reactions involving the sol-
vent [Section 3.1, Eqs. (3)–(6)]. The mass spectra of all compounds
detected by GC–MS, including the ones formed upon derivatization,
are given below:
O
S-5
Si
(2-oxo-1,3-dioxolan-4-yl)methyldimethylcarbamate (6): m/z (%)
189 (10, M+), 102(5, CH2OC(O)NMe2+), 88(10, CH2OC(O)NMe2+),
72 (100, C(O)NMe2+), 58 (22, NCO2+), 44 (43, NMe2+).
O
O C NMe2
O
O
6
O
2.4. Catalyst stability and scale-up
The glass vial was charged with 70 mmol of glycerol (4.600 g),
5.0 mL of DMA, 0.70 mmol of CuCl2 (1 mol%), and 0.35 mmol of pyri-
dine (0.5 mol%). The vial was introduced into the autoclave, which
was sealed and charged with O2 (0.7 MPa) and CO up to a total pres-
sure of 4.0 MPa. The autoclave was heated at 130 ◦C for 5 h. After
this time, the autoclave was cooled down to room temperature,
evacuated from the residual gas and sampled for conversion and
selectivity. Then, the mixture was recharged with a fresh mixture
of CO/O2 and allowed to react at 130 ◦C for other 5 h. This procedure
was repeated nine times until the conversion remained constant
indicative of the fact that the carbonylation process was stopped.
To verify if the incomplete conversion was due to the cata-
lyst deactivation, the organic products were distilled out from
the reaction mixture and the solid residue was tested for a new
carbonylation experiment by adding fresh starting materials (glyc-
erol, pyridine and DMA) under 4 MPa of CO/O2. The carbonylation
started again indicating that the catalyst is stable under the reac-
tion conditions and that the apparent deactivation is due to the
side-reaction of CO oxidation to CO2 promoted by Cu(II) and water
(by-product of carbonylation) which becomes predominant during
the course of reaction (see Section 3.3).
Silylated glycerol (S-1): m/z (%) 307 (1, M+−1) 293 (3, M+−CH3,),
263 (2, M+−3Me), 218 (10, M+−6Me), 205 (44, M+−CH2OSiMe3),
191 (5, M+−CH2OSiMe3−Me), 175 (5, M+−CH2OSiMe3−2Me),
147 (66, M+−OSiMe3−SiMe3), 133 (17, M+−CH2OSiMe3−SiMe3),
117 (31, CHCH2OSiMe3+), 103 (24, CH2OSiMe3+), 101 (5,
CHCH2OSiMe2+), 89 (4, OSiMe3+), 73 (100, SiMe3+).
Si O
O Si
O
Si
S-1
Silylated glycerol carbonate (S-2): m/z (%) 189 (1, M+−1), 175
(1, M+−Me), 160 (1, M+−2Me), 145 (1, M+−3Me), 131 (48,
M+−CH2CO), 116 (4, M+−SiMe3), 103 (26, CH2OSiMe3+), 101 (100,
CHCH2OSiMe2+), 89 (3, OSiMe3+), 73 (76, SiMe3+).
O Si
O
O
S-2
O
Silylated glycerol monoformate (S-3a): m/z (%) 243 (1, M+−Me),
191 (10, M+−CH2OC(O)H), 161 (55, M+−CH2OSiMe3), 147
(65, M+−OSiMe3−HCO), 133 (20, OCH2CH2OSiMe3+), 117 (40,
CH2CH2OSiMe3+), 73 (100, SiMe3+), 59 (25, MeCO2+).
2.5. Carbonylation mechanism: reductive step
A glass vial was charged with glycerol (280 mg, 3.7 mmol), an
equimolar amount of CuCl2 (496 mg, 3.7 mmol), K2CO3 (780 mg,
7.4 mmol) in DMA (5 mL). The vial was introduced into the auto-
clave, which was sealed, charged with 2.5 MPa of CO and allowed
to react at 100 ◦C for 5 h. After this time, the autoclave was cooled
to room temperature and evacuated, and the reaction mixture was
treated with the silylating reagent and analyzed by GC–MS. A con-
version of 50% ca. was detected. The reaction was repeated with
O
H C O
O Si
O
Si
S-3a
Silylated glycerol monoacetate (S-3b): m/z (%) 263 (1, M+−Me),
205 (9, M+−CH2OC(O)Me), 175 (100, M+−CH2OSiMe3), 147
(69, M+−OSiMe3−MeCO), 133 (48, OCH2CH2OSiMe3+), 117 (43,