Electronic Supplementary Material (ESI) for Green Chemistry.

This journal is © The Royal Society of Chemistry 2020

Electronic Supplementary Information

A transesterification-acetalization catalytic tandem process for the functionalization

of glycerol: the pivotal role of isopropenyl acetate

Davide Rigo, Alvise Perosa and Maurizio Selva*

Dipartimento di Scienze Molecolari e Nanosistemi dell’Università Ca’ Foscari,
Via Torino, 155 – 30172 – Venezia Mestre (Italy)
Email: selva@unive.it

S1
Table of contents
The acetylation of glycerol with AcOH. .............................................................................................S3
Figure S1............................................................................................................................................S3
The effect of the catalyst amount. ....................................................................................................S4
Table S1. ............................................................................................................................................S4
Figure S2............................................................................................................................................S6
The acetylation of glycerol with acetic anhydride.............................................................................S6
Figure S3............................................................................................................................................S7
Experiments with d4-isotope labelled acetic acid .............................................................................S7
Figure S4............................................................................................................................................S8
Scheme S1. ........................................................................................................................................S9
Figure S5…………………………………………………………………………………………………………………………………….S10
Figure S6..........................................................................................................................................S10
Figure S7..........................................................................................................................................S11
The reaction of glycerol with Ac2O and acetone .............................................................................S12
Table S2. ..........................................................................................................................................S13
The reaction of glycerol with AcOH and acetone............................................................................S13
Table S3. ..........................................................................................................................................S13
Scheme S2. ......................................................................................................................................S14
Scheme S3. ......................................................................................................................................S14
Figure S8..........................................................................................................................................S15
Figure S9..........................................................................................................................................S16

S2
The acetylation of glycerol with AcOH. A solution of glycerol (1.0 mmol) in acetic acid (10 mL; 0.1 M)
was set to react in the presence of Amberlyst-15 (15.0 mg; 15 wt%, with respect to glycerol), at 30,
50, and 70 °C, respectively. The reaction mixture was analyzed by GC at intervals from 0.5 up to 24
h. The conversion of glycerol and the products distribution, both evaluated by calibration with
standard solutions are shown in Figure S1.

O
OH OH OAc OH OAc OAc OAc
(175 mol equivs)
OH OH OAc + OH OAc + OAc
Amb-15
OH 30-70 °C OH OH OAc OH OAc
5/5' 6/6' 7

Glyc
100 100
Conversion (%)

Selectivity (%)
80 6/6' 80
60 60
40 70 °C 40
20 20
7 5/5'
0 0
0 4 8 12 24
100 Glyc 100
Conversion (%)

Selectivity (%)
80 50 °C 80
5/5'
60 60
40 6/6' 40
20 20
7
0 0
0 4 8 12 24
100 100
Glyc
Conversion (%)

5/5'
Selectivity (%)

80 80
60 60
40
30 °C 40
6/6'
20 20
0 0
0 4 8 12 24
t (h)
Figure S1. The reaction between glycerol and acetic acid at different temperatures of 30, 50 and 70 °C (bottom, mid,
and top, respectively). Conditions: solution of glycerol (1.0 mmol) in acetic acid (10 mL, 0.1 M), Amberlyst-15 (15.0 mg;
15 wt%) as a catalyst. (-■-) Glycerol conversion; (-•-) selectivity towards monoacetin (5/5’); (-▲-) selectivity towards
diacetin (6/6’); (-▼-) selectivity towards triacetin (7)

The increase of the temperature affected both the conversion and the products distribution. As T
was raised from 30 to 50 and 70 °C: i) a quantitative conversion of glycerol was reached after 24,
12, and 6 h, respectively; ii) prolonged experiments proceeded with the preferential formation of
products deriving from the multiple (bis and tris) acetylation of glycerol. Notably however, at
comparable high conversions (ca 90%), the ratio of mono- to di-acetins (5/5’:6/6’) displayed by the
three figures, was in a relatively narrow range from 55:40 to 65:35, thereby indicating that the
relative rates of mono- and bis-acetylation of glycerol were poorly influenced by the temperature,
S3
at least, until unconverted glycerol was present. This was manifest by comparing the selectivity
towards 5/5’ and 6/6’ reached at 30, 50 and 70 °C, after 24, 8, and 2 h, respectively, and led to
corroborate the observation that, regardless of the temperature, the selectivity was elusive using
AcOH as an acetylating agent.
Moreover, at the highest investigated temperature (70 °C), the amount of triacetin detected after 8
h (12%) was equivalent to that (13%) achieved in half the time (4 h) in the presence of iPAc/AcOH
(compare entry 3, Table 1 in the main text). This notwithstanding AcOH was in used in a very large
(175) molar excess, while iPAc was equimolar with respect to glycerol. The results suggested that
acetic acid was a far less active acetylating agent than iPAc.

The effect of the catalyst amount. Two experiments were carried out under the same conditions of
Figure 1 (glycerol: 1.00 mmol; iPAc: 7.50 mmol; AcOH: 8.00 mmol, 0. 5 mL; 30 °C), except for the
glycerol:catalyst weight ratio that was changed: the amount of Amberlyst-15 was reduced to 10
and 5 wt% with respect to glycerol. Results are reported in Table S1 after 24 h. For comparison, data
of Figure 1 (Amberlyst-15: 15 wt%) are also included.

Table S1. Effect of the catalyst amount on the reaction of glycerol with a mixture of iPAc/AcOH
Products
Amb-15 Conv.
Entry distribution (%)b
(wt%) (%)a
4/4’c 7
1 15 ≥99 50 50
2 10 ≥99 61 39
3 5 ≥99 67 33
Conditions: mixture of glycerol (1.00 mmol), iPAc (7.50 mmol), and AcOH (8.00 mmol, 0.5 mL) in the presence of variable
amounts of Amberlyst-15 (5.00, 10.00, and 15.00 mg; 5, 10, and 15 wt%), t = 24h. a Conversion of glycerol. b Selectivity
towards compounds 4/4’ and 7 was defined as reported in the main text. c Total amount of isomer products

Decreasing the catalyst amount brought about a drop of the rate of all the reactions involved: after
24 h, albeit acetal acetates 4/4’ and triacetin 7 were the only observed products in all cases, the
corresponding amounts increased and decreased, respectively, from 50 to 61 and 67% for 4/4’, and
from 50 to 39 and 33% for 7 (entries 1-3). Results were consistent with profiles of Figure 1 where at
the highest investigated catalyst amount (15 wt%), the (4/4’):7 ratio reached a maximum of ca 60:40
after 120 min, comparable to that achieved after 24 h with Amberlyst-15 at 10 wt%.
No further investigations were carried out on this aspect.

The reaction of iPAc and acetic acid. At 30 °C, after 24 h, the reaction of an equimolar mixture of
isopropenyl acetate (5.0 mmol) and acetic acid in the presence of Amberlyst-15 as a catalyst (45.0
mg; 15 wt% with respect to AcOH) mainly afforded acetic anhydride (80% yield, by GC), though
minor amounts by-products (≤10% with respect to Ac2O) were detected. The mass spectra of these
compounds are reported in Figure S2, and were consistent with the formation of acetylacetone, and
of products of the aldol condensation of acetone, both the aldol, 4-hydroxy-4-methylpentan-2-one,
and the two isomers of the α,β-unsaturated carbonyl derivative, 4-methylpent-3-en-2-one and 4-
methylpent-4-en-2-one. In addition, the formation of 4-oxopentan-2-yl acetate was observed.
Comparison with spectra of authentic compounds available from the NIST library of the GC/MS
ChemStation afforded a match quality >60% in all cases.

S4
 C:\Users\rigod...p\Davide\p5.txt Injection 1 Function 1 MS + spectrum 0.00
110

85.000
100.00%
100
43.000
92.00% O O
90

80

70

100.000
60.00%
60

50

40

30

20

10 55.000 86.000
69.100 5.00% 101.000
4.00% 3.00% 3.00% 207.000
1.00%
0

-10
40 50 60 70 80 90Injection
C:\Users\rigod...p\Davide\p1.txt 100 110 120 130
1 Function 1 MS140 150 160
+ spectrum 0.00170 180 190 200 210 220 230 240 250 260 270 280
110 m/z (Da)
83.100
100.00%
100

O
90

80

70
55.100
61.00%
60

50
98.100
41.00%
40 43.000
35.00%

30

20
53.000
11.00%
10 51.000 84.100
5.00% 56.100 5.00% 100.100
3.00% 72.000 3.00% 207.000
1.00% 1.00%
0

-10
40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260
m/z (Da)

S5
 C:\Users\rigod...p\Davide\p2.txt Injection 1 Function 1 MS + spectrum 0.00
110

43.000
100.00%
100

90
O OH
80

70

60

50

40

101.100
30.00%
30 59.100
24.00%

20

103.000
10 8.00%
39.100 58.100 72.100 145.000
2.00% 2.00% 1.00% 2.00%

-10
40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280
C:\Users\rigod...p\Davide\p3.txt Injection 1 Function 1 MS + spectrum 0.00
m/z (Da)
110

43.000
100.00%
100
O
90

80

70

60

50

40

83.100
30 28.00%

98.100
20 18.00%

59.100
9.00%
10
103.000
61.000 3.00%
1.00%
0

-10
40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400 420 440 460 480 500
m/z (Da)

S6
 C:\Users\rigod...p\Davide\p4.txt Injection 1 Function 1 MS + spectrum 0.00
110

43.000
100.00%
100

90
O O
80 O
85.000
70 67.00%

60

50 100.000
45.00%

40

30

20

10
69.000 101.100
3.00% 3.00%

-10
40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400 420 440 460 480 500
m/z (Da)
Figure S2. Mass spectra of by-products of the reaction of iPAc and acetic acid

The acetylation of glycerol with acetic anhydride. A solution of glycerol (1.00 mmol) in acetic acid
(0.1 M, 10 mL) was set to react at 30 °C, in the presence of different amounts of acetic anhydride
and Amberlyst-15 (15.0 mg, 15 wt% with respect to glycerol) as the catalyst. Figure S3 reports the
results of four reactions in which the glycerol:Ac2O (W) molar ratio was 5, 2, 1 and 0.2, respectively.

S7
 Ac2O Ac2O
100 5/5' 100 100 100

80 80 80 80
5/5'
Conversion (%)

Selectivity (%)
60
W=5 60 60
Glyc W=2 60

40 Glyc 40 40 40

6/6'
20 6/6' 20 20 20

7
0 0 0 0

0 20 40 60 80 100 120 200 300 720 1440 0 25 50 75 100 200 300 720 1440

Ac2O Glyc
100 100 100 100
Glyc 7
80 80 80 80
W=1 W=0.2
Conversion (%)

Selectivity (%)
60 5/5' 60 60 60

40 40 40 40

6/6'
20 20 20 20

7 6/6'
5/5'
0 0 0 0

0 50 100 200 300 720 1440 0 25 50 75 100 200 300 720 1440
t (min) t (min)

O O
OH O OAc OH OAc OAc OAc
(0.2-5 mol equivs)
OH OH OAc + OH OAc + OAc
AcOH solvent
Amb-15, 30 °C
OH OH OH OAc OH OAc
5/5' 6/6' 7

Figure S3. The reaction of a solution of glycerol (1.00 mmol) in acetic acid (10 mL; 0.1 M) carried out at 30 °C, in the
presence of variable amounts of and acetic anhydride and Amberlyst-15 catalysis (15.0 mg; 15 wt%) as the catalyst. The
glycerol:Ac2O molar ratio (W) was 5, 2, 1 and 0.2: the corresponding four reaction profiles are shown in top left, top
right, bottom left, and bottom right figures, respectively. (-■-) conversion of glycerol; (-•-) conversion of acetic
anhydride; (-▲-) selectivity towards mono-acetins (5/5’); (-▼-) selectivity towards diacetin (6/6’); (-♦-) selectivity
towards triacetin (7)

Compared to Figure S1, the acetylating capability of Ac2O proved much better than that of AcOH.
This was manifest even when sub-stoichiometric amounts of the anhydride were used: in the first
60-120 min of reaction, as long as Ac2O was present in the reaction mixture, both the conversion of
glycerol and the proportion of bis- versus mono- acetylated products [diacetins 6/6’ vs mono-acetins
5/5’] were higher than that achieved with AcOH (cfr Figure S1, bottom, to Figure S2, top). No to

S8
mention that in the same time interval, triacetin 7 was obtained as a single product when 5 molar
equivs of Ac2O were used (bottom right, Figure S2).
Notably, after the consumption of Ac2O, all reactions slowed down significantly: the further
conversion of glycerol and the formation of its multiple transesterification derivatives was
ascribable to the acetylating activity of AcOH.

Experiments with d4-isotope labelled acetic acid. Reactions were carried out under the conditions of
Figure 1 by replacing AcOH with its d4-isotope labelled analogue, CD3COOD. A mixture of glycerol
(1.0 mmol), iPAc (7.5 mmol), CD3COOD (0.5 mL, 8.75 mmol), and Amberlyst-15 (15.0 mg, 15 wt%)
was set to react at 30 °C, for 24 h. The GC/MS analysis of the reaction mixture proved the formation
of labelled derivatives with variable D contents for each of the expected products of glycerol.
Notwithstanding this complexity, at the end of the experiment (24 h), the formation of acetal
acetates and triacetin was detected in analogy to what observed in Figure 1. Acetal acetates were
present as two species (per each isomer), the non- and the tri-deuterated compounds, while
triacetin was obtained in the form of four species, the non-, the tri-, the hexa-, and the nona-
deuterated products, respectively (see Scheme 7, main text). Figure S4 reports the result by showing
enlarged regions of GC/MS analyses corresponding to the two signals of solketal acetate, the most
abundant 5-membered ring acetal ester (top), and the four signals of triacetin (bottom).

2500000
O
D 3C O O

2000000 O 4*
Intensity GC/MS (%)

O
1500000 H 3C O O

O 4
1000000

500000

24,0
t (min) O
O

D 3C O D 3C O
600000 D 3C O O CH3
D 3C O O CD3

O O O O
O

H 3C O
Intensity GC/MS (%)

400000
D 3C O O CH3

O O
O
200000
H 3C O
H 3C O O CH3

O O

21,6 21,8 22,0 22,2

t (min)

S9
Figure S4. Top: the red profile shows the GC/MS signals for solketal acetate; the corresponding d3- and non-deuterated
derivatives are indicated from left to right, respectively. Bottom: the red profile shows the GC/MS signals for triacetin;
the corresponding d9-, d6-, d3- and non-deuterated derivatives are indicated from left to right, respectively.
Green profiles for both top and bottom represent the deconvolution of the red profiles carried out by the “Gaussian
polynomial fitting” available in the OriginPro Software (ver. 9.1)

In the figure, red profiles were the authentic GC/MS analyses, while green profiles were obtained
by the deconvolution of the red ones, carried out by the “Gaussian polynomial fitting” available in
the OriginPro Software (ver. 9.1, Origin Lab. Corp.). Thanks to a more satisfactory resolution, signals
of solketal acetate (compounds 4 and 4*, top) allowed not only a more reliable integration of the
area under the corresponding peaks, but also a better reading/interpretation of GC/MS spectra,
than those of different triacetins.
The investigation was then continued considering only products 4 and 4*. From deconvolution, the
relative amounts of 4* and 4 were 67% and 33%, meaning that the quantity of the tri-deuterated
solketal acetate was twice as much the non-deuterated derivative. From the mass spectra recorded
in the full scan (TIC, 70 eV), fragment ions at m/z = 159 and 162 were recognized as the most
abundant and easily distinguishable ions for 4 and 4*, respectively (Figure S5).
The characteristic absence of the molecular ion peak (M+=174 and 177 for 4 and 4*, respectively)
was already noticed in a previous paper reported by us.1 The fragment ions 159 and 162 were
consistent with the loss of a methyl radical from the acetal ring (Scheme S1)

H 3C H 3C CH3
CH3 CH3
O O O
O -e O - CH3 O
X 3C O X 3C O X 3C O

O O O
m/z= 174, X=H m/z= 159, X=H
m/z= 177, X=D m/z= 162, X=D

Scheme S1. Plausible fragmentation pathway for the formation of ions 159 and 162 from compounds 4 and 4*

S10
 C:\Users\rigod...tic cd3cood.txt Injection 1 Function 1 MS + spectrum 0.00
110

43.100
100.00%
100

159.100
90.00% H 3C
90 CH3
O
H 3C O O
80
O

70

60

50

40

30

101.100
20 18.00% 160.100
15.00%

44.100 72.100
9.00%
10 8.00% 99.000
55.100 5.00% 115.100 161.000 207.000
2.00% 2.00% 2.00% 1.00%
0

-10
40 50 60 70 80
90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340
m/z (Da)
C:\Users\rigod...to di testo.txt Injection 1 Function 1 MS + spectrum 0.00
110

46.100
100.00%
100

90
162.100
81.00% O
80 D 3C O O

O
70

60
163.100
52.00%

50

40

30
102.100
21.00%
20
57.100 73.100 101.100 164.100
11.00% 12.00% 12.00% 12.00%

10 74.000 103.100
6.00% 99.000 5.00% 165.100
3.00% 121.200 3.00% 206.900
1.00% 1.00%
0

-10
40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360
m/z (Da) Figu
re S5. Mass spectra (TIC) of product 4 (top) and 4* (bottom)

S11
To improve sensitivity, mass spectra were then acquired in the SIM mode on the characteristic
fragment ions m/z = 159 and 162 (70 eV) (Figure S6).
Abundance

Ion 159.00 (158.70 to 159.70): SPP117 SIM 159 E 162.D\data.ms

4500000

4000000

3500000

3000000

2500000

2000000

1500000

1000000

500000

0
18.00 18.50 19.00 19.50 20.00 20.50 21.00
Time-->
Abundance

Ion 162.00 (161.70 to 162.70): SPP117 SIM 159 E 162.D\data.ms

4500000

4000000

3500000

3000000

2500000

2000000

1500000

1000000

500000

0
18.00 18.50 19.00 19.50 20.00 20.50 21.00
Time-->

Figure S6. SIM chromatograms of selected fragment ions m/z = 159 and 162 for products 4 (top) and 4* (bottom),
respectively.

Integration of signals in the SIM chromatogram proved that relative areas under the peaks of target
ions 159 and 162 were 65% and 35%, respectively, thereby confirming that the parent compounds
4* and 4 were formed in a ratio of about 2:1.
The GC/MS analysis of the same reaction mixture also indicated the formation of three species of
acetic anhydride, i.e. the non-, the tri-, and hexa-deuterated products (Figure S7).

O O

D 3C O CH3
1600000 O O

1400000 D 3C O CD3

1200000 O O
Intensity GC/MS (%)

1000000 H 3C O CH3

800000

600000

400000

200000

-200000
7,6 7,8 8,0
t (min)
Figure S7. The red profile shows the GC/MS signals for acetic anhydride: the corresponding d6-, d3-, and non-deuterated
derivatives are indicated from left to right, respectively. The green profile represents the deconvolution of the red
profiles carried out by the “Gaussian polynomial fitting” available in the OriginPro Software (ver. 9.1)
S12
From deconvolution (green profile), the integration of the area under the corresponding peaks
allowed to estimate that non-deuterated, d3- and d6- species of acetic anhydride were present in a
1:3:2 ratio, respectively.

The reaction of glycerol with Ac2O and acetone. Glycerol (1.00 mmol) was set to react with different
mixtures of Ac2O (1.00-10.00 mmol) and acetone (1.00-20.00 mmol) using Amberlyst-15 as a catalyst
(15 wt%). Acetic acid (1.5 equivs. with respect to glycerol) was employed as a solvent. Experiments
were carried out at 30°C, for 24 h. Results are shown in Table S2.

Table S2. The reaction of glycerol with acetone and acetic anhydride
Glyc:Ace:Ac2Oa Conversionb Selectivityc (%)
Entry Solvent
(mol:mol:mol) (%) 3/3’ 4/4’ 5/5’ 6/6’ 7
1 1:5:3 ≥99 61 1 38
2 1:1:1 ≥99 10 30 34 23 1
3 1:1:10 ≥99 30 1 69
AcOH
4 1:3:10 ≥99 58 1 41
(1.5 equivs.)
5 1:5:10 ≥99 68 1 31
6 1:20:5 ≥99 82 18
7 1:20:2.5 ≥99 15 78 3 5
Conditions: T=30 °C, Amberlyst-15 (15 mg; 15 wt%, with respect to glycerol), t=24 h. a Molar ratio of the reactants. b
Conversion of glycerol, by GC. c Products selectivity (by GC) calculated as described in the main text.

Under the conditions of Scheme 9 of the main text, when iPAc was replaced by Ac2O, the selectivity
towards 4/4’ did not exceed 61% (entry 1: compared to 91% achieved with iPAc). The optimization
of the reactants molar ratio allowed to obtain the desired acetal acetates 4/4’ in a 82% amount only
by using a significantly large excess of both acetone and acetic anhydride (20 and 5 molar
equivalents, respectively), with respect to glycerol (entry 6).

The reaction of glycerol with AcOH and acetone. Glycerol (1.00 mmol) was set to react with different
mixtures of AcOH (1.00-10.00 mmol) and acetone (1.00-20.00 mmol) using Amberlyst-15 as a
catalyst (15 wt%). Experiments were carried out at 30°C, for 24 h. Results are shown in Table S2.

Table S3. The reaction of glycerol with acetone and acetic acid
Glyc:Ace:AcOHa Conversionb Selectivityc (%)
Entry
(mol:mol:mol) (%) 3/3’ 4/4’ 5/5’ 6/6’ 7
1 1:1:1 70 56 12 30 2 -
2 1:1:10 95 8 15 51 25 1
3 1:3:10 96 33 26 33 7 -
4 1:5:10 ≥99 48 32 15 5 -
5 1:20:5 ≥99 95 5 - - -
6 1:20:2.5 ≥99 97 3 - - -
Conditions: T=30 °C, Amberlyst-15 (15.00 mg; 15 wt%, with respect to glycerol), t=24 h. a Molar ratio of the reactants. b
Conversion of glycerol, by GC. c Products selectivity (by GC) calculated as described in the main text

The reaction of an equimolar mixture of glycerol, acetone and acetic acid showed that the
acetalization of glycerol was favored over its acetylation: compounds 3/3’ were the most abundant
products (56%; entry 1). The situation reversed using an excess AcOH (10 molar equivs) which

S13
brought about the predominant formation of glyceryl esters as a mixture of mono- and di-acetins
(5/5’: 51% and 6/6’: 25%, respectively: entry 2). Thereafter, attempts to tune the products
distribution by further modifying the reactants molar ratio were unsuccessful. Increasing acetone
with a constant volume of AcOH allowed to increase both acetals 3/3’ and acetals esters 4/4’ up to
a maximum of 48% and 32% amount, respectively (entry 4); however, further increasing the ketone
and decreasing acetic acid resulted in the highly selective formation of acetals 3/3’ (95-97%, entries
5-6). Results led to conclude that the mixture of acetone and acetic acid could in no way replace
the use of iPAc for the selective synthesis of acetals acetates 4/4’.

The reaction with d6-acetone. The synthesis of acetal acetates was explored using d6-acetone in
place of acetone under the conditions of Scheme 9. A mixture of glycerol (1.00 mmol), isopropenyl
acetate (4.00 mmol), d6-acetone (5.00 mmol), acetic acid (1.50 mmol), and Amberlyst-15 (15.00 mg;
15 wt% with respect to glycerol) was set to react for 16 h at 30 °C. The experiment indicated the
formation of acetal acetates in the form of two species (per each isomer), the non- and the hexa-
deuterated compounds, respectively. The attention was focused on solketal acetate which was by
far the most abundant isomer (97%). Scheme S2 offers a pathway for the formation of the d6- and
the non-deuterated solketal acetate.
OH
HO OAc

O O O
Nu Added to the mixture
O H 3C CH3 D 3C CD3

Nu

H 3C CD3
CH3 D 3C
O O
AcO O O OAc
4 4d6
Scheme S2. Pathways for the formation the non-deuterated and the d6- solketal acetate (4 and 4d6, respectively)

The acetylation of any nucleophilic species available in the reaction environment (Nu: glycerol,
mono- or di-acetins) by iPAc, releases acetone (CH3COCH3) which competitively reacts with d6-
acetone (CD3COCD3, added to the mixture) for the acetalization of glycerol, affording the non-
deuterated 4 and the d6- solketal acetate 4d6, respectively).
GC/MS analyses recorded in the full scan (TIC, 70 eV) allowed to identify fragment ions at m/z = 159
and 162 as the most abundant and easily distinguishable ions for 4 and 4d6, respectively (Figure S5).

In analogy to Scheme S1, the fragment ions 159 and 162 were consistent with the loss of a methyl
radical from the acetal ring (Scheme S3).

X 3C X 3C CX3
CX3 CX3
O O O
O -e O - CH3 O
H 3C O H 3C O H 3C O

O O O
m/z= 174, X=H m/z= 159, X=H
m/z= 180, X=D m/z= 162, X=D
S14
 Scheme S3. Plausible fragmentation pathway for the formation of ions 159 and 162 from compounds 4 and 4d6

Although ions with m/z = 159 and 162 were the same selected when labelled acetic acid was used
(cf. Figure S5), the analysis of mass spectra proved that compound 4* bearing the isotopic marking
on the acetyl group underwent a different fragmentation pathway with respect to compound 4d6
where the marking was on the methyl group of the acetal ring. The comparison of Figure S5 and
Figure S8 highlights this difference. The characteristic fragment ions for 4* were 162 (81), 102 (21),
73 (12), 57 (11), 46 (100), 43 (66), while those for 4d6 were 162 (100), 107 (30), 76 (18), 65 (12), 46
(88), 43 (86).
C:\Users\rigod...ato acetone.txt Injection 1 Function 1 MS + spectrum 0.00
110

162.100
100.00%
100

46.100
90 88.00% D 3C
CD3
O
O O
80
O

70

60

50

40

107.100
30.00%
30

49.100 78.100
20 18.00% 18.00%
65.100
12.00% 163.100
79.100 9.00%
10 7.00% 102.100
4.00% 115.000 164.100
1.00% 1.00%
0

-10
40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350
m/z (Da)

Figure S8. Mass spectra of product 4d6. For the comparison with mass spectra of product 4 see Figure S4, top

Mass spectra were then acquired in the SIM mode on the fragment ions m/z = 159 and 162 for
compounds 4 and 4d6, respectively. (Figure S9). Integration of signals in the SIM chromatograms
proved that relative areas under the peaks of target ions 159 and 162 were 65% and 35%,
respectively, thereby confirming that the parent compounds 4d6 and 4 were formed in a ratio of
about 2:1.

S15
Abundance

Ion 159.00 (158.70 to 159.70): SPP140 B (SIM 159 E 162).D\data.ms

8000000
7000000
6000000
5000000
4000000
3000000
2000000
1000000
0
16.50 17.00 17.50 18.00 18.50 19.00 19.50 20.00 20.50 21.00 21.50 22.00
Time-->
Abundance

Ion 162.00 (161.70 to 162.70): SPP140 B (SIM 159 E 162).D\data.ms

8000000
7000000
6000000
5000000
4000000
3000000
2000000
1000000
0
16.50 17.00 17.50 18.00 18.50 19.00 19.50 20.00 20.50 21.00 21.50 22.00
Time-->
Figure S9. SIM chromatograms of fragment ions m/z = 159 and 162 from products 4 and 4d6, respectively.

S16
References
1R. Calmanti, M. Galvan, E. Amadio, A. Perosa, M. Selva, ACS Sustainable Chem. Eng. 2018, 6(3),
3964-3973, DOI: 10.1021/acssuschemeng.7b04297.

S17