Development and Scale-up of an Organocatalytic Enantioselective Process to Manufacture (S)-Pregabalin

Article abstract of DOI:10.1021/acs.oprd.5b00160

Herein is reported the development of a new process to manufacture (S)-pregabalin. The method comprises six steps, run under the catalysis of a recyclable polymer bound phase transfer catalyst, and afforded (S)-pregabalin in overall 54% yield, starting from building blocks acetylacetone, isovaleraldehyde, and nitromethane.

Full text of DOI:10.1021/acs.oprd.5b00160

Page 1 of 29
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Development and scale up of an organocatalytic
enantioselective process to manufacture (S)ꢀ
Pregabalin
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Ʒ
Ʒ
Maria Moccia,Ʒ* Mauro Cortigiani, Claudio Monasterolo, Federica Torri,Ʒ Claudia Del
Fiandra,Ʒ Geoffrey Fuller,Ʊ Brian Kelly,Ʊ Mauro F. A. AdamoƱ*
Ʒ
Centre for Synthesis and Chemical Biology (CSCB), Royal College of Surgeons in Ireland
(RCSI), 123 St Stephen’s Green, Dublin 2, Dublin. Tel +353 402 2208; *Email:
mariamoccia@rcsi.ie
Ʊ
Kelada Pharmachem. Ltd. 123 St Stephen’s Green, Dublin 2, Dublin. *Email:
mauro.adamo@keladapharmachem.com
ACS Paragon Plus Environment
1

Organic Process Research & Development
Page 2 of 29
1
2
3
4
Table of Contents Graphic
5
6
7
Y-
HO
H3CO
N
8
9
tBu
N
But
N
O2N
O
H2N
N
O
0.03 eq
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
COOH
K2CO3 2.5 eq.toluene
(0.5M, 21 °C
CH3NO2 2eq.
Y- = PolySulfonated
Polystyrene beads
NO2
NO2
(S)-Pregabalin
98% yield er 86 : 14 (S:R)
after crystallisation
70% Yield er 99.95 : 0.05 (S:R)
ACS Paragon Plus Environment
2

Page 3 of 29
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
ABSTRACT. Herein is reported the development of a new process to manufacture (S)ꢀ
Pregabalin. The method comprises six steps, run under the catalysis of a recyclable polymer
bound phase transfer catalyst and afforded (S)ꢀPregabalin in overall 54% yield, starting from
building blocks acetylacetone, isovaleraldehyde and nitromethane.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
KEYWORDS. (S)ꢀPregabalin, Phase Transfer Catalysis, 4ꢀnitroꢀ5ꢀstirylisoxazoles, Michael
addition.
ACS Paragon Plus Environment
3

Organic Process Research & Development
Page 4 of 29
1
2
3
4
1. Introduction
5
6
7
8
9
(S)ꢀPregabalin 1 is currently manufactured by the originator (Pfizer) as well as by other generic
pharmaceutical producers.1ꢀ9 The original Pfizer process involved six steps, the enantioselective
one relying on a Rh(DUPHOS) catalyzed hydrogenation.1 Rhodium and chiral phosphines are
becoming increasingly expensive and in order to gain competitiveness, Pfizer has recently
developed and patented a new process in which the chirality was generated via an enzymatic
step.2 Many other generic producers have recently filed patents describing their processes to
manufacture (S)ꢀPregabalin 1.3 A vast proportion of these technologies gave (RS)ꢀPregabalin as a
racemate, from which (S)ꢀ1 was then obtained via diastereoisomeric crystallization. This limited
the theoretical yield of (S)ꢀ1 at 50% with concomitant discard of 50% in mass of the unwanted
(R)ꢀisomer. Recently, Piramal developed a method to racemise (R)ꢀPregabalinꢀ 1 to (RS)ꢀ
Pregabalinꢀ 1 in order to recycle the unwanted isomer.4 Other methodologies have been
described to produce (S)ꢀ1. These relied either on bioenzymatic steps,5 diastereoselective
synthesis from natural biomass (Dꢀmannitol),6 organocatalysis,7,8 and flow chemistry.9 Although
these methodologies provided (S)ꢀ1 in high enantiomeric excess,5ꢀ9 no evidence has been
provided on the reliability of these syntheses in a multikg context. Therefore, the suitability of
these to manufacture (S)ꢀ1 at a large scale is still uncertain.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Phase Transfer Catalysis (PTC) has been effectively used to manufacture drugs intermediates
and unnatural αꢀaminoacids and drugs intermediates at scale.10a Cinchona derived phase transfer
catalysts, used to perform these transformations, can be prepared in a single step from
inexpensive Cinchona alkaloids which could be recovered and recycled at the end of reaction.10bꢀ
d In this context, it is noteworthy the immobilization of Cinchona derived phase transfer catalysts
on polystyrene beads, which could be used to perform an enantioselective step and recycled.11
ACS Paragon Plus Environment
4

Page 5 of 29
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Importantly, the reactions carried out with the polystyrene bound catalysts gave similar
conversion and enantioselectivity as the one carried out using the unbound agents. Therefore on
the bases of cost of reagents, chemical efficiency and precedents for large scale manufactures,
Phase Transfer Catalysis (PTC) was selected as the technology of choice for the development of
a new process to prepare (S)ꢀ1.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1: Retrosynthetic analysis for the preparation of (S)ꢀPregabalin 1.
O2N
O2N
H2N
N
O
COOH
COOH
NO2
1
2
3
O
N
N
O
O
CH3NO2
PTC
NO2
NO2
4
5
6
As part of their research, the Adamo’s group has developed alkenes 4 as optimal Michael
acceptors (Scheme 1).12 Compounds 4 underwent highly enantioselective addition when reacted
with nitromethane13,14 (Scheme 1), as well as with bromomalonates15 under phase transfer
catalysis. With this in mind, the synthesis of (S)ꢀ1 was planned around the PTC catalyzed
addition of nitromethane to compounds 4. In retrosynthetic fashion (Scheme 1), (S)ꢀ1 could be
obtained via reduction of nitroacid 2, which is also an intermediate for the synthesis of racemic
and enantiopure pregabalin.16 In turn, nitroacid 2 could be prepared from hydrolysis of
nitroisoxazole 3 via the SartiꢀFantoni reaction,15 a procedure in which the 4ꢀnitroisoxazole core
could be opened to display a carboxylate by treatment with excess hydroxide. Compound 3 was
ACS Paragon Plus Environment
5

Organic Process Research & Development
Page 6 of 29
1
2
3
4
5
6
then accessible via PTC catalyzed addition of nitromethane to alkene 4, which, in turn, could be
made by condensation of aldehyde 5 and 4ꢀnitroisoxazole 6.14
7
8
9
2. Result and Discussion
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 2: Preparation of (S)ꢀPregabalin 1.
H3C
NO2
H3C
1. NaOH 0.2
equiv) H2O : THF
: MeOH
O
O
N
H2SO4
HNO3
O
Na2CO3 0.5 equiv.
O
N
N
(NH3OH)SO4 0.5 equiv.
CH3
O
O
2. MsCl (1.2
equiv) /NEt3
(2 equiv) DCM
NO2
5
6
7
8
4
99% yield
99% yield
1.2 equiv
Y-
88% yield
HO
H3CO
N
tBu
NaOH (5 equiv)
H2O
But
0.03 eq
O2N
N
O2N
Ra-Ni, H2, 1 atm,
H2O rt,16h
H2N
N
O
9
COOH
COOH
K2CO3 2.5 eq.toluene
(0.5M, 21 °C
3
2
1
NO2
CH3NO2 2eq.
Y- = PolySulfonated
Polystyrene beads
98% yield er 86 : 14 (S:R)
after crystallisation
65% Yield er 99.95 : 0.05 (S:R)
95% yield er 99.99 : 0.01 (S:R)
99% yield er 99.99 : 0.01 (S:R)
In forward sense, the preparation of (S)ꢀ1 has been accomplished starting from acetylacetone 5,
aldehyde 8 and nitromethane as the carbon containing building blocks.17 Hence, acetylacetone 5
was reacted with hydroxylamine to obtain 3,5ꢀdimethylisoxazole 6 in excellent isolated yields;
compound 6, in turn was nitrated to 3,5ꢀdimethylꢀ4ꢀnitroisoxazole 7 which was equally obtained
in quantitative yields. The next step involved a two phase procedure to condensate 7 and
aldehyde 8 to desired Michael acceptor 4. Compound 4, which was obtained in 88% optimized
yield, was then subjected to enantioselective Michael addition with nitromethane, a reaction
which proceeded under the catalysis of Cinchona based ammonium salt 9 to provide adduct 3 in
excellent yield and good enantioselectivity (from 72 to 86% ee, vide infra). Compound 3 could
be recrystallized to enantiopure (99.7% ee) and then subjected to the SartiꢀFantoni reaction to
access nitroacid 2. Final reduction of the nitro group in 2 provided (S)ꢀPregabalin 1. The
conditions highlighted in Scheme 2 are optimized and proved to work at kg scale. This required
ACS Paragon Plus Environment
6

Page 7 of 29
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
optimization of each step and development of opportune methods of purification. The
development and the optimization of each synthetic steps described in Scheme 2 will described
below.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The condensation of acetylacetone and hydroxylamine is a wellꢀknown reaction that has been
described in many papers.18 This reaction takes place in water and involves dissolving an
opportune hydroxylamine salt and a base to make an aqueous solution of hydroxylamine (pH ≈
8) to which acetylacetone 5 was slowly added under vigorous stirring. The addition of
acetylacetone 5 to a solution of hydroxylamine is exothermic and on a 10 kg scale the
temperature of the solution went from 18°C to 35ꢀ40 °C (5 added in portions during a period of
one hour). We have described the preferred use of hydroxylamine carbonate as this was, to our
knowledge, the cheapest source of hydroxylamine. However, we have tested other salts, for
example hydroxylamine hydrochloride, which worked equally well. The reaction of 5 and
hydroxylamine gave in our hands almost quantitative yields which could be explained as follows.
Firstly, the resulting 3,5ꢀdimethylisoxazole 6 is insoluble in water and upon cooling of the
reaction mixture separated out as an oily supernatant that was decanted. This procedure assured
≈85% weighted yield of 6, which was augmented by subjecting the water layer to distillation
under reduced pressure (T=58°C, P=37 mmHg). This operation, at R&D lab scale, could well be
performed using a rotavapor. Compound 6 distilled off from the water layer to provide an
additional 10ꢀ14% in mass. Key to optimization of this step was the concentration at which this
reaction should be performed. Hence, using a 3M molar concentration ensured separation of the
oily phase of 6 as a supernatant, which would not occupy promptly when the reaction was
performed in a diluted media. Compound 6 obtained from this step and purified as described
above could be carried through the next step without further purification.
ACS Paragon Plus Environment
7

Organic Process Research & Development
Page 8 of 29
1
2
3
4
5
6
7
8
9
Nitration of compound 6 to provide 3,5ꢀdimethylꢀ4ꢀnitroisoxazole 7, which occurred under
conventional aromatic nitration condition. This involved dissolving compound 6 in cold (0ꢀ5°C)
sulfuric acid then adding nitric and then excess of sulfuric acid keeping the temperature below
15ꢀ20°C. The subsequent heating of the reaction mixture generated an exotherm which brought
the temperature from 50 to 58°C. This data is in line with observation recorded for other
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
+
aromatic nitrations which is explained considering the generation of a NO2 ion. Quenching of
reaction using water ice bath gave almost quantitative yield of desired 7, which is highly water
insoluble and precipitated immediately. Noteworthy, to obtain acid free 7, suitable for the next
step, the acidic reaction mixture should be poured into a vortex formed by fast stirring of an
cooled aqueous bath media. This ensured fast precipitation of water insoluble 7 that was obtained
as a fine powder. Filtration of the cake of 7 and its washing with cold water gave compound 7
pure enough to be carried out in the next step. The high yield observed in the preparation of 7,
therefore is a reflection of its extremely low solubility in water. It should be noted that nitration
of 6 to 7 could also be performed using ammonium nitrate and trifluoroacetic acid,12a and that the
sulfonitric mixture was selected based on low cost of these reagents.
Scheme 3: Preparation of Michael acceptor 4.
H3C
NO2
1. NaOH 0.2
equiv) H2O : THF :
MeOH
N
2. MsCl (1.2
equiv) /NEt3
(2 equiv) DCM
N
OH
O
O
O
N
O
NO2
NO2
10
89% yield
7
8
4
99% yield
1.2 equiv
99% yield
The condensation of compound 7 and aldehyde 8 (Scheme 3) to provide aldol compound 10 was
studied under a vast number of conditions which included variation of temperature, solvent,
amount of hydroxide base, order of addition of reagents and concentration. The nucleophilic
ACS Paragon Plus Environment
8

Page 9 of 29
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
addition of deprotonated 7 to aldehyde 8 is thermodynamically favored. However, there is a
remarkable difference in the solubility of these reagents. Compound 7 lithium or sodium salt is
very soluble in water, but virtually insoluble in most organic solvents. On the other hand,
aldehyde 8 is very soluble in most organic solvents but very poor soluble in water. For this
reason, the optimization of this step required, principally, a careful selection of the solvent
media. After several attempts, a ternary mixture of water, tetrahydrofuran and methanol was
identified as the best media to ensure high yields of aldol compound 10. The use of
substoichiometric amounts of sodium hydroxide allowed limiting the loading of 8 to 1.2
equivalents. The limited amount of base reduced the extent of selfꢀcondensation of 8 and allowed
obtaining compound 10 in higher yields and purity. We have shown that preparation of 10 could
be performed in a range of temperatures varying from ꢀ5°C to ambient (20°C). However when
run at 0ꢀ5°C, the selfꢀcondensation of aldehyde 8 was suppressed and compound 10 obtained in
89% yield. The crude 1HꢀNMR of this reaction showed, after work up, that 10 contained starting
materials 7 (7%) and 8 (5%) together with a small amount of alkene 4 (8%) as the only
contaminants. Therefore the reaction of 7 and 8 under this condition was an equilibrium reaction
which cleanly produced desired 10 and small amounts of dehydrated 4. Crude compound 10
could be used in the next step without further purification. Therefore compound 10 was
dissolved in dichloromethane and treated with mesyl chloride and triethylamine. The optimized
procedure involved using 1.2 equivalents of mesyl chloride and two equivalents of triethylamine.
It should be noted that the addition of the first equivalent of triethylamine was exothermic and
therefore this first part of the reaction must be conducted with control of temperature. Also,
while the mesyl chloride could be charged in one portion, the addition of triethylamine has to
occur within a range of time (typically one hour when the reaction was conducted on a 10 kg
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
9

Organic Process Research & Development
Page 10 of 29
1
2
3
4
5
6
7
8
9
scale, but this may be variable with the design of the reactor). Noteworthy, this step was equally
fast when run at ꢀ30°C and at 20°C and in all cases only the Eꢀstereoisomer could be detected.12a
Compound 4 could be obtained in 96% purity (NMR) by simple wash with water and diluted
hydroxide which served to eliminate the remainder traces of compound 7 and 8 brought through
from the previous step. This product was pure enough to be carried through the enantioselective
to prepare 3. It should be noted that the following step is robust enough to allow using samples
of alkene 4 containing up to 15ꢀ20% impurities. These low grade samples reacted equally
providing desired Michael adduct 3 in identical enantioselectivity and reaction rates as a pure
sample. Although not necessary, compound 4 could be obtained in purity higher that 99.9% by:
(a) distilling crude 4 (120°C at 0.2mb); (b) performing a treatment with DMF and water which
retained the traces of unreacted 7 and 8.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The following step involved the optimisation of the enantioselective procedure to obtain
compound 3. The optimisation of this step has been achieved in three phases, which involved:
(a) the selection of an optimal catalyst, (b) the adjustment of condition to perform the addition of
nitromethane to 4 at 0.5M concentration; (c) the preparation of a polymer bound catalyst and its
recycling.
We have first determined that Quinidinium salts were providing the desired (S)ꢀ3 over the
unwanted (R)ꢀ3. This was established by converting (S)ꢀ3 to desired (S)ꢀ1 and measuring the
optical rotation of the final compound. In addition, toluene and potassium carbonate were found
as the best media to perform this reaction. Therefore, a number of quinidinium ammonium salts
were subsequently screened. A short list of the most relevant catalysts 11-15 is described in
Table 1 alongside the enantioselectivity of compounds 3 obtained. The addition of nitromethane
ACS Paragon Plus Environment
10

Page 11 of 29
Organic Process Research & Development
1
2
to alkene 4 was also performed under the catalysis of Maruoka’s catalysts.10a These experiments
3
4
5
6
gave desired adduct 3 in almost racemic mixtures (3ꢀ8% ee).
7
8
9
Table 1: catalyst screening
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Br
Br
Br
HO
H CO
3
HO
H CO
N
HO
C(CH )
3
N
C(CH )
H CO
3 3
3
3 3
N
CF
3
N
N
N
C(CH )
C(CH )
3 3
3 3
13
CF
11
12
3
Br
Br
HO
H CO
HO
3
N
Br
H CO
3
N
CH
3
N
N
Br
CH
3
14
15
Entry(a) Catalyst Conv% of 4 Yield% of 3
Ee% of 3
Er S : R of 3
86 : 14
93 : 7
1
2
3
4
5
100
100
100
100
100
95
97
96
97
91
72
86
85
64
62
11
12
13
14
15
92 : 8
82 : 18
81 : 19
(a)
Alkene 4 (20 mg, 0.096 mmol) toluene (9.6 mL), catalyst 11-15 (0.1 equiv, 10mol%), and nitromethane
(30mg, 0.48 mmol, 5 equiv), temperature 0ºC, K2CO3 (66mg, 0.48 mmol, 5 equiv) added in one portion. The
reaction was stirred for 32 hours, then quenched with sat NH4Cl (10 mL), extracted with toluene (2 x 10 mL),
dried over MgSO4, filtered over celite and evaporated to give pure compound in yield and ee listed in table 1.
t
As evidenced in Table 1, catalysts 12ꢀ13 holding two butyl groups ensured the highest
enantioselectivity. Compound 3 was therefore obtained in up to 86%ee when the reaction was
conducted at 0˚C and in a diluted media. Reactions carried out at lower temperatures (up to ꢀ
28˚C) provided desired 3 in significantly lower ees. With the best catalyst in hand, we have then
studied the variation of enantioselectivity of 3 as a function of temperature and reagents
concentration.
ACS Paragon Plus Environment
11

Organic Process Research & Development
Page 12 of 29
1
2
3
4
Table 2: Effect of concentration on the enantioselectivity of 3
5
6
7
Entry(c)
Conc. of 4
Conv. of 4 (h)
Ee % of 3
8
9
1(a)
2(b)
5(b)
6(b)
7(b)
8(b)
9(b)
10(d)
0.01M
0.01M
0.04M
0.08M
0.16M
0.32M
0.50M
0.50M
100% (96h)
90% (96h)
75% (40h)
80% (20h)
83% (20h)
92% (19h)
100% (24h)
60% (96h)
86%
86%
85%
83%
80%
74%
72%
72%
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(a) Reaction conducted with 0.1 equiv of 12. (b) Reaction conducted with 0.05 equiv of 12.
(c)
Alkene 4 (20 mg, 0.096 mmol) toluene, catalyst 12 (0.05 equiv, 5mol%),
nitromethane (5 equiv), temperature 0ºC, K2CO3 (5 equiv) added in one portion. The
reaction was stirred time given in Table 2, then quenched with sat NH4Cl, extracted
with toluene, dried over MgSO4, filtered over celite and evaporated to give compound 3
in ee listed in table 2. (d) Reaction conducted with 0.03 equiv of 12.
The results collected in Table 2 pointed out that reaction rate increased with the concentration,
however the enantioselectivity of 3 dropped from 86% at 0.01M to 72% at 0.5M. This data
identified 0.08M as a threshold for the concentration to still obtain high enantioselectivity and it
also showed that complete conversion was only obtained with high catalyst loading. In order to
reduce the catalyst loading further, we have conducted a series of experiments at room
temperature. This set of experiments showed that compound 4 was fully converted in 24 hours
and with only 0.03 equiv of 0˚C 12 affording compound 3 in high yields and in 72% ees.
Importantly under this new set of condition, the amount of nitromethane could be reduced to
only two equivalents. It has been found beneficial to purity of compound 3 to prepare a solution
of nitromethane and catalyst 12 and then add this dropꢀwise to a suspension of alkene 4 and base
ACS Paragon Plus Environment
12

Page 13 of 29
Organic Process Research & Development
1
2
3
4
5
6
in toluene. This limited the time of contact between base and nitromethane which suffered of
base catalyzed polymerization.
7
8
9
Finally, with the aim of recycling catalyst 12 and show this could be used in additional cycles,
we have immobilized 12 on a sulfonylated polystyrene matrix as described by Itsuno.11 Hence
compound 12 was reacted with pꢀsulfonylstyrene and the ammonium sulfonate salt so obtained
was subsequently subjected to radical polymerization. The catalyst loading was determinated by
elemental analysis and resulted to be of 0,57mmol/g. The resulting material was employed in the
reaction of 4 with nitromethane under the optimized conditions identified (Scheme 2).
Delightfully, the reaction proceeded under the catalysis of 0.03 equiv of polystyrene bound 12
providing compound 3 in the high yields and 72% ee. In relation to polymer bound 12, we have
found the following experimental facts: (a) the polymerized product must be washed with
toluene before being employed; indeed unpurified material obtained via the Itsuno procedure
may still contain dimethyformamide which interferes with the reaction of 4, leading in the first
cycle to compound 3 in a decreased ee‘s; (b) The polymer bound catalyst could be reused up to
10 times without loss of enantioselectivity and of conversion; (c) as reported by others,11 the
polymer elemental content of Nitrogen remained constant at each run, hence demonstrating no
loss of ammonium salt from the polymer. Hence, polymer bound catalyst 12 was a robust and
practical catalyst that promoted the formation of desired adduct 3, could be separated via simple
filtration and reused.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Compound 3 could be recrystallysed to >99%ee and then carried through to final (S)ꢀPregabalin
1. In alternative, enantioenriched compound 3 (72%ee) cold be carried through the two following
steps to give (S)ꢀ1 in identical 72%ee. This latter could be then subjected to partial resolution via
reported methodology.3
ACS Paragon Plus Environment
13

Organic Process Research & Development
Page 14 of 29
1
2
Compound 3 was then subjected to the SartiꢀFantoni reaction,14 which entailed mixing neat
compound 3 with a 5N NaOH solution and then heating to reflux for 2 hours. This reaction gave
excellent yields of compound 2, which was fully characterized. However, compound 2 is
remarkably acid sensitive and it is prone to decompose rapidly if heated or if brought to a pH <3.
On the contrary, sodium carboxylate salt of 2 is very stable and could be isolated and stored for
long periods. We have found that the best procedure to obtain high yield of desired (S)ꢀ1
involved submitting an aqueous solution of 2 sodium salt to reduction using Ni/Ra and hydrogen.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The reduction of acid 2 to (S)ꢀ1 has been reported in a number of papers and patents20 and could
be conducted with loadings of Ni/Ra in dependence of the pressure of H2 employed and of the
quality and make of the catalyst. With the low grade Ni/Ra available to us and in order to operate
at 1atm, 2 equiv of Ni/Ra were necessary to obtain quantitative yields of compound 1. The high
loading of Ni/Ra was required to avoid the formation of products of partial reduction. We have
observed that the starter nitroacid 2 disappeared just after 2h, however the purity and quantity of
the reduced compound, i.e. Pregabalin 1, increased with elongation of reaction time. We have
also verified that Ni/Ra could be reused for at least 8 cycles before it lost in part its activity and
required regeneration. We have also demonstrated that Pd/C and H2 or ammonium formate are
equally good reductants allowing to obtain desired (S)ꢀ1.
3. Conclusion
In conclusion, we have developed and reported a new process for the manufacture of (S)ꢀ1 in six
steps and in overall 54% yields. The method made use of a recyclable organocatalyst and
employed acetylacetone, isovaleraldehyde and nitromethane as the carbon containing building
blocks. The Chincona alkaloids could be sourced at average 25ꢀ30€$/kg from the Asian market
ACS Paragon Plus Environment
14

Page 15 of 29
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
and are a cheap source of chirality mostly when considering their recyclability up to 10 fold.
Therefore they compare well, priceꢀwise with the phosphines normally employed in the synthesis
of γꢀaminoacids which price range from 40.000$ to 80.000$ per kg. In addition, the best methods
of resolution of Pregabalin gave only 25ꢀ30% yields.1ꢀ4 Therefore, the present method, that
provides (S)ꢀ1 in 65% yield is competitive when benchmarked against resolution. Considering
that all the reagents could be sourced at cheap price and the solvents and catalysts recycled, this
is a cost efficient and competitive method for the large scale manufacture of (S)ꢀ1.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4. Experimental Section
General Experimental Details
13
1H, C, NMR spectra were recorded on a Varian AS 300, Bruker 400 and 600 spectrometer.
1
13
Chemical shifts (δ) are reported in ppm relative to residual solvent signals for H and C NMR
(1H NMR: 7.26 ppm for CDCl3; 13C NMR: 77.0 ppm for CDCl3. 13C NMR spectra were
1
acquired with H broad band decoupled mode. DMSOꢀd6 (referenced to 2.52 and 3.35 ppm for
1H and 40.0 for 13C). Coupling constants (J) are in Hz. Multiplicities are reported as follows: s,
singlet, d, doublet, dd, doublets of doublets, t, triplet, q, quartet, m, multiplet, c, complex, and br,
1
1
13
broad. HꢀNMR spectral assignments are supported by Hꢀ1H COSY and Cꢀ1HꢀCOSY where
necessary. Carbon spectra are supported by DEPT analysis where necessary.
Melting points were determined using a Stuart scientific melting point apparatus and are
uncorrected. Infrared spectra (IR) were recorded as KBr discs using a Bruker Tensor27 FTꢀIR
instrument. Absorption maximum (νmax) was reported in wave numbers (cmꢀ1) and only
selected peaks are reported. High resolution mass spectra were obtained on a Waters Micro mass
ACS Paragon Plus Environment
15

Organic Process Research & Development
Page 16 of 29
1
2
3
4
5
6
7
8
9
LCT and low resolution mass spectra were recorded on Waters Micro mass Quattro LCꢀMS
spectrometers at 70 eV. Tetrahydrofuran was freshly distilled over sodium benzophenone prior
to use according to standard procedure. All other reagents and solvents were used as purchased
from Aldrich. Reactions were checked for completion by TLC (EM Science, silica gel 60 F254)
which were visualized by quenching of u.v. fluorescence (λmax = 254nm) or by staining with
either 10% w/v ammonium molybdate in 2M sulphuric acid or basic potassium permanganate
solution (followed by heat) as appropriate. The enantiomeric excess (ee) of the products was
determined by chiral stationary phase HPLC (Daicel Chiralpak AD, Chiracel OJ, Chiracel OD,
Chiralpak AS columns), using a UV detector operating at 254 nm. Retention factors (Rf) are
reported to ±0.05.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Preparation of 3,5ꢀdimethylisoxazole 6: In a round bottomed flask (20L) fitted with a magnetic
stirrer were put 7L of deꢀionised water, followed by (NH3OH)2SO4 (1710.2g, Mw = 164, 10.44
mol) followed by addition of small portions of Na2CO3·10H2O (3000g, Mw= 286, 10.48 mol).
After evolution of CO2 ceased and pH was ~8, acetylacetone 5 (2070g, Mw= 101, 20.7 mol)
was added in one portion and the reaction mixture heated at 65°C for 16h. At the end of this time
an oily supernatant was noted which was decanted and proven to be pure 3,5ꢀdimethylisoxazole
(1730 g). The water layer was loaded in a rotary evaporator, heated at 58°C and kept at a
pressure of 37 mmHg to distill a second fraction of 3,5ꢀdimethylisoxazole (360 g). The combined
fractions could be used in the next step without further purification. Optionally they could be
purified by distillation using a rotavapor (58 °C, 37 mmHg), to give pure title compound (2040g,
99% yield). 3,5ꢀDimethylisoxazole 6 so obtained is stable at room temperature and did not show
signs of decomposition after 12 months.
ACS Paragon Plus Environment
16

Page 17 of 29
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Preparation of 3,5ꢀdimethylꢀ4ꢀnitroisoxazole 7: In a round bottomed flask (20L), fitted with
overhead stirrer and kept at ꢀ5 to 0°C by an ice salt bath, was put H2SO4 (95ꢀ98% ACS reagent,
3240 mL; d=1.83) then 3,5ꢀdimethylisoxazole (1200 g). This addition must be carried out slowly
to ensure that the temperature stays close to 0°C. Then HNO3 conc. previously cooled at
temperature between 0ꢀ5°C (1200 mL; d=1.413) was added and then additional H2SO4 conc.
(5064 mL) was added in small portions. The solution was then heated at 60ꢀ65°C for 6 hour then
allowed to reach room temperature and poured dropꢀwise in a basin (30L) containing ice and
water under vigorous stirring. A fine precipitate was obtained then filtered, washed (suspended
in 20L of water and stirred vigorously for 10 minutes, then filtered (operation to be repeated until
water is neutral) and dried (1720 g; yield 99%). This product could be used in next step without
further purification and stored at room temperature for long periods (over 24 months).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Preparation of 4ꢀmethylꢀ1ꢀ(3ꢀmethylꢀ4ꢀnitroꢀisoxazolꢀ5ꢀyl)ꢀpentanꢀ2ꢀol 10: In a 10L round
bottomed flask fitted with a magnetic stirrer were put 3,5ꢀdimethylꢀ4ꢀnitroisoxazole (1000g,
Mw=142, 7.042 mol) and 800 mL of THF and the resulting solution stirred at 0°C. To this
solution, were then added 2000 mL of MeOH : H2O (7 : 3). A freshly made solution of NaOH
(56g, Mw = 40, 1.4 mol, 0.2 equiv.) in 500 mL of H2O was then charged in a dropping funnel
and added dropꢀwise over 10 minutes. Upon this addition, the solution becomes yellow to dark
brown. It was noted the formation of a precipitate of sodium 3,5ꢀdimethylꢀ4ꢀnitroisoxazolate that
will dissolve during the course of the reaction. At this point, a solution of isovaleraldehyde
(726g, Mw = 86, 8.45 mol, 1.2 equiv.) in 200 mL of THF was added drop wise at 0°C, over the
period of 50 min. After this period the ice bath was removed and the reaction allowed reaching
room temperature under vigorous stirring. Conversion was monitored after 5h (60%), 10h (83%),
16h (92%) since end of addition of aldehyde. A sample of the reaction mixture was kept under
ACS Paragon Plus Environment
17

Organic Process Research & Development
Page 18 of 29
1
2
3
4
5
6
7
8
9
stirring for further two hours and conversion measured again (92%) indicating the reaction
mixture had reached the equilibrium. The reaction was quenched by addition of distilled H2O
(5000 mL), stirred for 10 minutes, then extracted twice with DCM (first extraction 3000 mL;
second extraction 2000 mL). The organic layers were combined, washed with H2O (3000 mL at
least to avoid formation of emulsion), then with sat. NaHSO3 (2000 mL) then again with H2O
(3000 mL). The organic layer was evaporated at the rotavapor (49°C, 44mb) to give 1610g of
material which contains the title compound alongside 7% of alkene 4 (vide infra), 8% of
unreacted 3,5ꢀdimethylꢀ4ꢀnitroisoxazole 7 and 5% of isovaleraldehyde 8. Estimated weight in
title alcohol was 1430g (89% yield). Colorless liquid; Rf = 0.2 (Petroleum Ether/Ethyl Acetate,
90:10); δH (400 MHz, CDCl3) 4.23ꢀ4.20 ( 1H, m), 3.36 (1H, dd, J = 15 J = 4), 3.29 ( 1H, dd, J =
15 J = 7), 2.55 (3H, s), 1.87ꢀ1.76 (1H, m), 1.58ꢀ1.53 (1H, m), 1.40ꢀ1.34 (1H, m), 0.95 (3H, d, J =
7), 0.93 (3H, d, J = 7); δC (100.6 MHz, CDCl3) 172.9, 155.8, 130.9, 68.0, 46.8, 36.1, 24.8, 23.3,
22.0, 11.8. HRMS: m/z found [M+H]+ 229.1154, C10H17N2O3 requires 229.1188.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Synthesis of 3ꢀmethylꢀ5ꢀ(4ꢀmethylꢀpentꢀ1ꢀenyl)ꢀ4ꢀnitroꢀisoxazole 4: In a 20L round bottomed
flask fitted with an overhead stirrer (to overcome difficult stirring caused by formation of solid
HNEt3Cl during the reaction) were loaded 800g of 4ꢀmethylꢀ1ꢀ(3ꢀmethylꢀ4ꢀnitroꢀisoxazolꢀ5ꢀyl)ꢀ
pentanꢀ2ꢀol (about 90% pure, 720g, 3.165 mol, Mw = 229) and 10 L of DCM. The solution was
cooled at 0°C by ice water, then mesyl chloride (433.1g, Mw = 114, 3.8 mol, 1.2 equiv) was
added dropwise over 30 min. To the resulting solution was then added NEt3 (638.5g, 6.33 mol,
Mw = 101, 2.0 equiv) dropꢀwise over the period of 2h 30 min. To be noted that addition of the
first equivalent of NEt3 is exothermic, therefore addition should be carried out at such a rate to
keep temperature between 0°C and 10°C maximum. This is to avoid formation of side products
and ensure high purity of alkene product. After the addition was completed the reaction was
ACS Paragon Plus Environment
18

Page 19 of 29
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
allowed to reach room temperature and stirred for 2h (from the end of addition of NEt3). The
reaction was quenched with H2O (3000 mL), washed with additional H2O (2 x 5000 mL), then
with 5% NaOH in water, the organic layer dried over Na2SO4 and then evaporated under reduced
pressure to give 1390g of crude product that could be used in the next step (estimated purity 96%
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
by HꢀNMR). The product so obtained could be taken through the next step without further
purification. Optionally, the crude product (1390g) could be thinned with 500 mL of petroleum
ether (40ꢀ60°C) and passed through a plug of silica gel (flash type) (100g) to give 1320g of title
compound. Pale yellow liquid; Rf = 0.8 (Petroleum Ether/Ethyl Acetate, 90:10); δH (400 MHz,
CDCl3) 7.12ꢀ 7.00 (2H, m ), 2.56 (3H, s), 2.29ꢀ2.26 (2H, m), 1.89ꢀ1.82 (1H, m), 0.97 (6H, d, J =
6), δC (100.6 MHz, CDCl3) 167.0, 156.0, 130.1, 115.6, 43.0, 28.3, 22.5, 12.0. HRMS: m/z found
[M]+ 196.0815, C9H14N2O3 requires 196.0848.
Note 1: Whilst the next step proceeds well with alkene containing variable levels of impurities
(1 to 15%) Product 4 can be obtained over 99.9% pure via distillation, collecting the fraction
boiling over 120°C at 0.2mb. The yield was ca 65ꢀ70% of pure alkene.
Note 2: Purification via DMF wash: the alkene may be obtained over 99.9% pure by dissolving
it in a DMF : H2O (20: 80) mixture and then extracting with petroleum ether. Typically, 120 g of
crude material were dissolved in 100 mL of DMF, then 400 mL of water were added and the
resulting yellow solution was extracted with petroleum ether (40 : 60), (2x 500 mL). Optionally,
the alkene (20g) was dissolved in petroleum ether (40 : 60), (50 mL), then treated with 3 mL of
DMF, stirred for 20 minutes, then washed with water (2 x 50 mL).
Synthesis of (S)ꢀ3ꢀmethylꢀ5ꢀ(4ꢀmethylꢀ2ꢀnitromethylꢀpentyl)ꢀ4ꢀnitroꢀisoxazole 3 using a solution
of catalyst 12: in a 20L round bottomed flask, fitted with an overhead stirrer were charged
ACS Paragon Plus Environment
19

Organic Process Research & Development
Page 20 of 29
1
2
3
4
5
6
7
8
9
toluene (6600 mL) and K2CO3 (1518g, 11 mol, 2.5 equiv). A solution of alkene 4 (924g, 4.4
mol), nitromethane (536g, 8.8 mol, 2 equiv) and catalyst 12 (217.8g, 0.03 equiv) in toluene
(2200 mL) was charged in a dropping funnel and added dropwise over a 2h period. The reaction
was stirred at room temperature (21°C) until conversion is completed (8 to 24h) (Note:
Conversion rates have been found variable depending on the particle size of potassium
carbonate. Smaller particle size gave faster reactions). Then the organic liquid phase was
filtered, washed with NH4Cl dil solution (1000 mL) then with water (2000 mL). The organic
layers were evaporated to give compound 3 in 98% yield (1168g) as a sticky liquid and in
enantiomeric ratio (S : R). The ee of the product was determined by CSPꢀHPLC using a
Chiralcel OD column (nꢀhexane/iꢀPrOH 90:10, flow rate 1 mL/min). Retention times: (S)ꢀ3, 34.4
min, (R)ꢀ3 32.1 min.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Reꢀcrystallization of compound 3 from 72% ee to enantiopure +99 ee: pure compound 3 (400g)
was dissolved in minimum amounts of hot isopropanol or hot mixtures of isopropanol petroleum
ether (1 : 1), the second typically 1000 mL. The resulting solution was cooled at 0°C to give
compound 3 as needles, which were filtered, dried and weighted 280g (70%) and were 99.95%
ee. Colorless solid, m.p. 55°C (isopropanol/hexane); δH (400 MHz, CDCl3) 4.38 (2H, d, J = 6),
3.35 (1H, dd, J = 15, J = 6), 3.28 (1H, dd, J = 15, J = 7), 2.87 (1H, sept, J = 7), 2.56 (3H, s),
1.68 (1H, sept, J = 7), 1.36ꢀ1.23 (2H, m), 0.92 (3H, d, J = 4), 0.90 (3H, d, J = 4), δC (100.6 MHz,
CDCl3) 172.1, 122.9, 78.5, 40.9, 33.9, 30.0, 25.1, 22.4, 22.3, 11.7. HRMS: m/z found [M+H]+
272.1212, C11H18N3O5 requires 272.1246.
ACS Paragon Plus Environment
20

Page 21 of 29
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Synthesis of (S)ꢀ3ꢀmethylꢀ5ꢀ(4ꢀmethylꢀ2ꢀnitromethylꢀpentyl)ꢀ4ꢀnitroꢀisoxazole 3 using polymer
supported 12: in a 20L round bottomed flask, fitted with an overhead stirrer were charged
toluene (6600 mL) and K2CO3 (1518g, 11 mol, 2.5 equiv), toluene (2200 mL), alkene 4 (924g,
4.4 mol), nitromethane (536g, 8.8 mol, 2 equiv) and polymer supported catalyst 12 (429g, 0.03
equiv). The reaction was stirred at room temperature (21°C) until conversion is completed (8 to
24h) (Note: Conversion rates have been found variable depending on the particle size of
potassium carbonate. Smaller particle size gave faster reactions). Then the organic phase was
separated, the solids were washed with water to dissolve the inorganic salts to give 428g of
polymer bound catalyst (99% recovery). Ten consecutive runs using the same catalyst gave
compound 3 in identical yield and enantiomeric ratio.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Preparation of sodium 5ꢀmethylꢀ3ꢀnitromethylꢀhexanoate 2: neat compound 3, (392g, 1.44 mol)
was charged in a 5L round bottomed flask and treated with a freshly made aqueous solution (5N,
1440 mL, 5 equiv.) of NaOH (288g, 7.2 mol). The resulting dark solution was heated (T = 60ꢀ
65°C) for 2h, then allowed to reach room temperature. The dark aqueous solution of 2 sodium
salt was carried through the following reduction without requirement for isolation. Isolation and
purification of free acid 2 involved: cooling the aqueous solution of 2 sodium salt, adjustment of
pH ≈ 3 and extraction with EtOAc. Dark yellow liquid, 1H NMR (400 MHz, CDCl3) δH 4.50
(1H, dd, J = 12, J = 7), 4.44 (1H, dd, J = 12, J = 6) 2.67 (1H, sept, J = 6), 2.49 (2H, d, J = 6),
1.66 (1H, sept, J = 7), 1.28 (2H, m), 0.93 (3H, d, J = 7), 0.91 (3H, d, J = 7). 13C NMR δC (100.1
MHz, CDCl3) 177.6, 78.4, 40.3, 35.6, 31.7, 24.9, 22.3, 22.0.
Preparation of (S)ꢀPregabalin 1: in a 10L round bottomed flask were charged RaneyꢀNi (2
equiv, 493g of slurry 50% RaneyꢀNi in water) and methanol (2500 mL). To this suspension was
ACS Paragon Plus Environment
21

Organic Process Research & Development
Page 22 of 29
1
2
3
4
5
6
7
8
9
added a solution of nitroacid 2 in water and base obtained via hydrolysis of compound 3 (1.44
mol). The suspension obtained was stirred at room temperature under H2 (1 atm) for 6h, then the
liquid phase decanted, the solid RaꢀNi rinsed with methanol and reused for 10 times. The
methanolic aqueous solution was evaporated to give an aqueous solution of (S)ꢀPregabalin 1,
which was precipitated by adjustment of pH ≈ 6 via addition of diluted hydrochloric acid (95%
yield) as a colorless solid and in enantiomeric ratio of (S : R) of 99.99 :0.01. The optical purity of
(S)ꢀ1 was assessed via HPLC. Compound 1 was reacted with Nꢀαꢀ5ꢀfluoroꢀ2,4ꢀdinitrophenylꢀ5ꢀ
Lꢀalanine amide. The two diastereoisomers so obtained could be separated by an ordinary
chromatography column (Inertsil ODSꢀ2.5 ꢁm, 250 mm × 4.6 mm i.d.). Phosphoric acid buffer
and acetonitrile (55:45, v/v) were used as mobile phase, the separation carried out at 1.0 mL/min
flow rate, at room temperature and the detector wavelength fixed at 340 nm. Procedure: 8mg of
(S)ꢀPregabalin 1 were accurately weighted in a 10 mL calibrated flask and dissolved in 1N HCl.
The pH of the system was adjusted to 7 by addition of 1N NaOH in water. The volume was
brought to 10 mL by addition of distillated water. 50 ꢀL of this solution were mixed with 200 ꢁL
of the FDNPAA solution (100.0 mg FDNPAA was dissolved by acetone into a 10ꢀmL calibrated
flask and stored in the refrigerator) and 20 ꢁL 1N NaHCO3. The solution was closed and heated
for 1 h at 40°C, then 10 ꢁL 2N HCl was added, and the solution evaporated under vacuum. The
residue was dissolved in 4 mL of dimethyl sulfoxide (DMSO) (equivalent to 10.0 ꢁg/mL
Pregabalin sample). The sample solution was filtered through a membrane filter, and an aliquot
(20 ꢁL) of the sample was injected for HPLC analysis. Retention times: derivative of (S)ꢀ1, 9.3
min; derivative of (R)ꢀ1 10.2 min.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
22

Page 23 of 29
Organic Process Research & Development
1
2
Preparation of Nꢀ(3,5ꢀditertbutylbenzyl)quinidinium bromide 12:18 in a round bottomed flask
fitted with a magnetic stirrer and a reflux condenser were put sequentially quinidine (1.0 equiv),
acetone to make a 0.18ꢀ0.20M solution and finally 3,5ꢀditertbutylbenzyl bromide (1.05 equiv).
The resulting solution was heated at 60ꢀ65°C for 2h. The reaction mixture was then allowed to
reach room temperature, the solvent evaporated to give a solid which was suspended in
petroleum ether, stirred for 30 minutes, then filtered and dried to give pure catalyst (NMR).
Optionally the crude product may be dissolved in the minimum quantity of DCM and then
precipitated by addition of Et2O. Yield 94ꢀ98%. NOTE 1: Commercial sources of quinidine
always contain variable amounts of dihydroquinidine. This implies that catalysts may contain
variable amounts of the corresponding dihydroquinidinium salt. Dihydroquinidinium salts
performed equally well in the synthesis of 3 as the quinidinium salts. NOTE 2: Catalysts may
form sticky viscous oils when in the presence of small amounts of organic solvents (e.g.
Acetone, DCM or Et2O); drying under reduce pressure (rotavapor) gave compound 12 as a fine
powder. Colorless powder; δH (400 MHz, CDCl3) 8.61ꢀ8.59 (1H, m), 7.97ꢀ7.94 (1H, m), 7.68ꢀ
7.67 (1H, m), 7.60ꢀ7.59 (1H, m), 7.53ꢀ7.51 (1H, m), 7.3ꢀ7.26 (2H, m), 6.79ꢀ6.77 (1H, m), 6.58ꢀ
6.55 (1H, m), 6.00ꢀ5.92 (1H, m), 5.83ꢀ5.80 (1H, m), 5.19ꢀ5.15 (2H, m), 4.85ꢀ4.76 (1H, m), 4.62ꢀ
4.57 (1H, m), 4.10ꢀ4.04 (1H, m), 3.93 (3H, s), 3.88ꢀ3.83 (1H, m), 3.50ꢀ3.44 (1H, m), 3.07ꢀ3.02
(1H, m), 1.89ꢀ1.85 (1H, m), 1.80ꢀ1.75 (1H, m), 1.08ꢀ1.01 (1H, m); ); δC (100.6 MHz, CDCl3)
157.9, 152.0, 147.4, 144.2, 143.0, 135.8, 131.8, 128.3, 126.2, 126.1, 124.4, 121.0, 120.5, 118.1,
102.2, 68.4, 65.3, 64.3, 56.8, 56.0, 54.2, 38.3, 35.0, 31.4, 30.9, 27.8, 24.2, 21.5.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
General preparation of quinidium bromides 11ꢀ15: In a round bottomed flask fitted with a
magnetic stirrer and a reflux condenser were put sequentially quinidine (1.0 equiv), a suitable
benzyl bromide (1.05 equiv) and acetone to make a 0.18ꢀ0.20M solution. It was noted that at 50ꢀ
ACS Paragon Plus Environment
23

Organic Process Research & Development
Page 24 of 29
1
2
3
4
5
6
7
8
9
55°C the reaction became clear. The resulting solution was heated at 60ꢀ65°C for 2ꢀ8h depending
on scale. The reaction mixture was then allowed to reach room temperature, the solvent
evaporated to give a solid which was suspended in petroleum ether, stirred for 30 minutes, then
filtered and dried to give pure quinidinium bromides 11ꢀ15. Optionally compounds 11ꢀ15 may be
dissolved in the minimum quantity of DCM and then precipitated by addition of Et2O.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
NOTE 1: Commercial sources of quinidine always contain variable amounts of
dihydroquinidine. This implies that catalysts 11ꢀ12 and 14ꢀ15 may contain variable amounts of
the corresponding dihydroquinidinium salt.
NOTE 2: In the synthesis of compounds 13ꢀ15, it was noted the formation of a precipitate from
hot acetone, minutes from refluxing at 60ꢀ65°C.
NOTE 3: Compounds 11ꢀ12 may form sticky viscous oils when wet (Acetone, DCM or Et2O);
Drying under reduce pressure (rotavapor) gave compounds 11ꢀ15 as a fine powder.
Nꢀ(3,5ꢀditrifluoromethylbenzyl)quinidinium bromide 11: light yellow powder; δH (400 MHz,
CDCl3) 8.39ꢀ8.36 (1H, m), 8.25ꢀ8.24 (2H, m), 7.74 (1H, s), 7.67 (1H, d, J = 8 Hz), 7.61 (1H, d, J
= 4 Hz), 7.49 (1H, d, J = 4 Hz), 6.95ꢀ6.92 (1H, m), 6.48ꢀ6.47 (1H, m), 6.12ꢀ6.09 (1H, m), 6.10
(1H, d, J = 8 Hz), 5.82 (1H, d, J = 8 Hz), 5.80ꢀ5.74 (1H, m), 5.16ꢀ5.10 (2H, m), 4.56ꢀ4.50 (1H,
m), 4.34ꢀ4.29 (1H, m), 4.15ꢀ4.10 (1H, m), 3.68 (3H, s), 3.08ꢀ3.02 (1H, m), 2.64ꢀ2.59 (1H, m),
2.39ꢀ2.19 (2H, m), 1.77ꢀ1.73 91H, m), 0.85ꢀ0.82 (1H, m); δC (100.6 MHz, CDCl3) 157.9, 147.0,
144.0, 142.0, 134.9, 133.9, 132.6, 132.3, 131.6, 130.4, 126.0, 123.9, 121.2, 120.4, 120.0, 118.5,
103.2, 68.1, 67.1, 60.4, 60.2, 56.5, 56.3, 54.5, 37.9, 27.0, 23.8, 21.9.
ACS Paragon Plus Environment
24

Page 25 of 29
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Nꢀ(3,5ꢀditertbutylbenzyl)dihydroquinidinium bromide 13: colourless powder; δH (400 MHz,
CDCl3) 8.69ꢀ8.68 (1H, m), 8.03ꢀ8.00 (1H, m), 7.75ꢀ7.74 (1H, m), 7.59ꢀ7.58 (2H, m), 7.54ꢀ7.53
(1H, m), 7.63ꢀ7.29 (2H, m), 6.82ꢀ6.80 (1H, m), 6.66ꢀ6.62 (1H, m), 4.76ꢀ4.73 (1H, m), 4.42ꢀ4.36
(1H, m), 3.94 (3H, s), 3.74ꢀ3.65 (2H, m) 3.55ꢀ3.46 (1H, m), 3.14ꢀ3.06 (1H, m), 2.54ꢀ2.48
(1H,m), 1.94ꢀ1.88 (1H, m), 1.81ꢀ1.77 (1H, m), 1.69ꢀ1.55 (3H, m), 0.88ꢀ0.84 (1H, m); ); δC
(100.6 MHz, CDCl3) 158.0, 152.1, 147.7, 144.3, 143.1, 132.2, 128.3, 126.2, 126.0, 124.5, 120.6,
120.5, 102.3, 68.9, 64.9, 64.6, 57.1, 56.0, 55.9, 36.2, 35.0, 31.4, 31.3, 24.9, 24.6, 24.3, 21.3, 11.4.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Nꢀ(3,5ꢀdimethylbenzyl)quinidinium bromide 14: colourless powder; δH (400 MHz, CDCl3)
8.45ꢀ8.43 (1H, m), 7.89ꢀ7.86 (1H, m), 7.71ꢀ7.70 (1H, m), 7.20ꢀ7.19 (1H, m), 7.20ꢀ7.17 (2H, m),
6.89ꢀ6.88 (1H, m), 6.69ꢀ6.67 (1H, m), 6.47ꢀ6.45 (1H, m), 5.89ꢀ5.81 (1H, m), 5.73ꢀ5.69 (1H, m),
5.17ꢀ5.12 (2H, m), 4.53ꢀ4.48 (1H, m), 3.91ꢀ3.86 (1H, m), 3.80 (3H, s), 3.43ꢀ3.38 (1H, m), 2.93ꢀ
2.90 (1H, m), 2.39ꢀ2.34 92H, m), 2.30 (6H, s), 1.80ꢀ1.65 (4H, m), 0.93ꢀ0.80 (1H, m); δC (100.6
MHz, CDCl3) 157.8, 147.3, 144.3, 142.6, 138.7, 135.6, 131.9, 131.7, 131.4, 126.8, 126.4, 120.7,
120.6, 118.0, 102.8, 68.0, 66.9, 62.9, 56.6, 56.0, 54.0, 38.2, 27.2, 24.0, 21.7, 21.3.
Nꢀ(3,5ꢀdibromobenzyl)quinidinium bromide 15: colourless powder; δH (400 MHz, CDCl3) 8.36ꢀ
8.35 (1H, m), 7.82ꢀ7.74 (4H, m), 7.66ꢀ7.65 (1H, m), 7.04ꢀ7.01 (1H, m), 6.58ꢀ6.56 (1H, m), 6.44ꢀ
6.40 (1H, m), 6.17ꢀ6.15 (1H, m), 5.86ꢀ5.79 (1H, m), 5.64ꢀ5.59 (1H, m), 5.28ꢀ5.22 (2H, m), 4.55ꢀ
4.52 (1H, m), 4.27ꢀ4.25 (2H, m), 3.70 (3H, s), 3.23ꢀ3.20 (1H, m), 2.79ꢀ2.76 (1H, m), 2.40ꢀ2.38
(1H, m), 2.25ꢀ2.21 (1H, m), 1.85ꢀ1.77 (3H, m), 0.93ꢀ0.85 (1H, m); ); δC (100.6 MHz,
CDCl3)157.7, 147.0, 144.1, 142.0, 136.2, 135.1, 135.0, 131.6, 131.1, 126.3, 123.5, 120.5, 119.5,
118.2, 103.8, 67.7, 59.8, 56.3, 56.0, 54.2, 37.9, 27.0, 23.8, 22.0, 15.3.
ACS Paragon Plus Environment
25

Organic Process Research & Development
Page 26 of 29
1
2
3
4
5
6
7
8
9
Preparation of polymer supported 12: this was prepared following the procedure reported by
Itsuno.11 Typically, compound 12 (12.12g, 20 mmol) was dissolved in DCM (10 mL) and styrene
sodium sulfonate added (1 equiv, 20 mmol). The resulting solution was stirred for 20 min, then
washed with water (2 x 10 mL) and the organic phase evaporated to give the correspondent
ammonium sulfonate salt (Mw = 710, 14.19g, 99.99% yield). The ammonium sulfonate
(14.19g, 20 mmol) was then dissolved in DMF (5g), the solution charged in a round bottomed
flask fitted with a magnetic stirrer and styrene (18.7g, Mw = 104, 9 equiv.), divinylbenzene 13
(520mg, Mw = 130, 0.2 equiv) and AIBN (660 mg, 0.05 equiv) consecutively added.
Copolymerization was run at 60°C for 6 hours, then the polymer was collected and washed with
toluene and methanol to remove the DMF to give polymer supported 12 (33.3g, 99% yield).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The loading of the catalyst was established by elemental analysis:
Elem. Anal. Calculated for C11.6H11.7N0.2O0.4S0.1 : C, 85.20%; H, 7.21%; N,1.71%.
Found: C, 82.66%; H, 7.19%; N,1.58%. Corresponding at 0.57 mmol N/g.
AUTHOR INFORMATION
Present Addresses: Royal College of Surgeons in Ireland (RCSI), 123 St Stephen’s Green,
Dublin 2, Dublin
Author Contributions
The manuscript was written through contributions of all authors. All authors have given approval
to the final version of the manuscript.
ACS Paragon Plus Environment
26

Page 27 of 29
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Acknowledgments: We acknowledge funding from the Irish Research Council for Science
Engineering and Technology (IRCSET) to Claudia del Fiandra (RS/2010/2812), Mauro
Cortigiani (EPSPG/2011/243), Federica Torri (EPSPG/2012/741), Claudio Monasterolo
(EPSPG/2012/741). We also acknowledge Science Foundation Ireland (SFI) for support to maria
Moccia (11/TIDA/B2032).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
13
Supporting information: Copies of HꢀNMR, CꢀNMR of compounds 11ꢀ15, 4, 10, 3, 2, 1;
HPLC of compounds 3 and 1; Additional table of results for the optimization of the preparation
of 3 using other phase transfer catalysts.
5. References
1. Hoekstra, M. S.; Sobieray, D. M.; Schwindt, M. A.; Mulhern, T. A.; Grote, T. M.;
Huckabee, B. K.; Hendrickson, V. S.; Franklin, L.C.; Granger, Eric J.; Karrick, G. L.
Org. Process Res. Dev. 1997, 1, 26.
2. Debarge, S.; McDaid, P.; O’Neill, P.; Frahill, P.; Wong, J. W.; Carr, D.; Burrell, A.;
Davies S.; Karmilowicz, M.; Steflik J. Org. Process Res. Dev., 2014, 18, 109.
3. See for example, Pradhan, B. S. in Process for the preparation of Rꢀ(ꢀ)ꢀ3ꢀ
(carbamoylmethyl)ꢀ5ꢀmethylhexanoic acid and the intermediates, WO2012/093411.
4. Harad, A.; Jagtap, A.; Roy, M.; HariꢀHaran S. in An improved process for racemization
of (s)ꢀ3ꢀ(carbamoylmethyl)ꢀ5ꢀmethylhexanoic acid, WO2014/016776A1.
5. Felluga, F.; Pitacco, G.; Valentin, E.; Venneri, C. D. Tetrahedron: Asymmetry 2008, 19,
945.
6. Izquierdoa, S.; Aguileraa, J.; Buschmann, H. H.; García, M.; Torrens A.; Ortuño R. M.
Tetrahedron: Asymmetry 2008, 19, 651.
ACS Paragon Plus Environment
27

Organic Process Research & Development
Page 28 of 29
1
2
3
4
7. Gotoh, H.; Ishikawa, H.; Hayashi, Y. Org. Lett., 2007, 9, 5307.
5
6
7
8
8. Peschiulli, A.; Procuranti, B.; O' Connor, C. J.; Connon, S. J. Nature Chem. 2010, 2, 380.
9. Ghislieri, D.; Gilmore, K.; Seeberger, P. H. Angew. Chem. Int. Ed. 2015, 54, 678.
10. (a)Shirakawa, S.; Maruoka, K. Angew. Chem. Int. Ed. 2013, 52, 4312. (b) Wang, X.; Yin,
L.; Yang, T.; Wang, Y. Tetrahedron:Asymmetry 2007, 18,108. (c) Itsuno, S.; Paul, D.K.;
Salam, M.A.; Haraguchi, N. J. Am. Chem. Soc. 2010,132(9), 2824. (d) Arakawa, Y.;
Haraguchi, N.; Itsuno, S. Angew. Chem. 2008, 47,8232.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
11. Itsuno, S.; Mehadi H. RSC Adv., 2014, 4, 5202.
12. (a) Wells, R.; Moccia, M,; Adamo, M. F. A. Tetrahedron Lett. 2014, 55, 803. (b) Piras,
L.; Moccia, M.; Cortigiani, M.; Adamo M. F. A. Catalysts 2015, 5, 595. (c) Salazar Illera,
D.; Suresh, S.; Moccia, M.; Bellini, G.; Saviano, M.; Adamo, M. F.A. Tetrahedron Lett.
2012, 53, 1808. (d) Bruschi, S.; Moccia M.; Adamo, M. F. A. Tetrahedron Lett. 2011, 52,
3602. (e) Dorschner, K. V.; Toomey, D.; Brennan, M. P.; Heinemann, T.; Duffy, F. J.;
Nolan, K. B.; Cox, D.; Adamo, M. F. A.; Chubb A. J. J. Chem . Inf. Model. 2011, 51 986.
(f) Adamo, M. F. A.; SartiꢀFantoni, P.; Chimichi, S.; Sandrelli, A. Tetrahedron Lett,
2010, 51, 6310. (g) Adamo, M. F. A.; Suresh, S.; Piras, L. Tetrahedron, 2009, 65, 5402.
(h) Adamo, M. F. A.; Suresh, S. Tetrahedron, 2009, 65, 990. (i) Adamo, M. F. A.;
Bruschi, S.; Suresh, S.; Piras, L. Tetrahedron Lett., 2008, 49, 7406. (l) Adamo, M. F. A.;
Konda, V. R. Tetrahedron Lett., 2008, 49, 6224. (m) Adamo, M. F. A.; Nagabelli, M.
Org. Lett., 2008, 10, 1807. (n) Adamo, M. F. A.; Donati, D.; SartiꢀFantoni, P.; Buccioni
A. Tetrahedron Lett., 2008, 49, 941. (o) Adamo, M. F. A.; Konda, V. R.; Donati, D.;
SartiꢀFantoni, P.; Torroba, T. Tetrahedron, 2007, 63, 9741. (p) Adamo, M. F. A.;
Nagabelli, M. Tetrahedron Lett., 2007, 4703. (q) Adamo, M. F. A.; Donati, D.; Duffy, E.
ACS Paragon Plus Environment
28

Page 29 of 29
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
F.; SartiꢀFantoni, P. Tetrahedron, 2007, 63, 2684. (r) Adamo, M. F. A.; Duffy, E. F.;
Konda, V. R.; Murphy, F. Heterocycles, 2007, 71, 1173. (s) Adamo, M. F. A.; Konda, V.
R. Org. Lett., 2007, 9, 303. (t) Adamo, M. F. A.; Duffy, E. F. Org. Lett., 2006, 8, 5157.
(u) Adamo, M. F. A.; Donati, D.; Duffy, E. F.; SartiꢀFantoni P. J. Org. Chem., 2005, 70,
8395.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
13. Baschieri, A.; Bernardi, L.; Ricci A.; Suresh, S.; Adamo, M. F. A. Angew. Chem. Int. Ed.
2009, 48, 9342.
14. R. Wells; M. Moccia, M. F. A. Adamo, Org. Biomol. Chem., 2015, 13, 2192.
15. Del Fiandra, C.; Piras, L.; Fini, F.; Disetti, P.; Moccia, M.; Adamo, M. F. A. Chem.
Commun. 2012, 48, 3863.
16. Saswata, L.; Hussain, M.; Datta, D. in Preparation of enantiomerically enriched gammaꢀ
nitro acid and pregabalin, WO 2010004577A1.
17. Adamo, M. F. A. in Process for the preparation of gamma amino acids and intermediates
used in said process, WO 2013076225A1.
18. Zhang, J.; Liu, X.; Ma, X.; Wang, R. Chem. Commun., 2013, 49, 9329.
19. Matoba, K.; Kaway, H.; Furukawa, T.; Kusuda, A.; Tokunaga, E.; Nakamura, S.; Shiro,
M.; Shibata, N. Angew. Chem. Int. Ed. 2010, 49, 5762.
20. (a) Pradhan, B.S. in Process for preparing pregabalin, WO2011016052A2. (b) Gore, V.;
Gadakar, M.; Shinde, D. in Novel and efficient method for synthesis of an aminoacid,
WO2009147434A1. (c) Saswata, L.; Hussain, M.; Datta, D. in Preparation of
enantiomerically enriched gammaꢀnitroacid and pregabalin, WO2010004577.
ACS Paragon Plus Environment
29