Redesign of the Synthesis and Manufacture of an Azetidine-Bearing Pyrazine

Article abstract of DOI:10.1021/acs.oprd.7b00384

Commercial route definition for a glucokinase activator called for a re-evaluation of the synthesis and processes used to access multikilogram quantities of a pyrazine building block. The processes developed allowed a literature route to sodium 6-oxo-1H-pyrazine-3-carboxylate to be leveraged. One of these processes consisted of a highly selective decarboxylation that allowed the target building block to be accessed with complete regioselectivity in standard batch processing equipment. The presence of an azetidine ring in the target required the mitigation of impurity liabilities arising from the use of the hydrochloride salt of azetidine as an input material.

Full text of DOI:10.1021/acs.oprd.7b00384

Subscriber access provided by NATIONAL UNIV OF SINGAPORE
Full Paper
Redesign of the Synthesis and Manufacture of an Azetidine-Bearing Pyrazine
David R. J. Hose, Phillip Hopes, Alan Steven, and Christian Herber
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.7b00384 • Publication Date (Web): 18 Jan 2018
Downloaded from http://pubs.acs.org on January 18, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Organic Process Research & Development is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 19
Organic Process Research & Development
1
2
3
4
Redesign of the Synthesis and Manufacture of an Azetidine-Bearing
5
6
Pyrazine
7
8
David R. J. Hose†, Phillip Hopes‡, Alan Steven*,† and Christian Herber§
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
Pharmaceutical Technology & Development, AstraZeneca, Charter Way, Macclesfield,
SK10 2NA, United Kingdom
‡
Cyton Biosciences Ltd, 68 Macrae Road, Bristol, BS20 0DD, United Kingdom
§
DOTTIKON Exclusive Synthesis AG, P.O. Box CH – 5605 Dottikon, Switzerland
Alan.Steven@astrazeneca.com
ACS Paragon Plus Environment

Organic Process Research & Development
Page 2 of 19
1
2
3
Table of Contents graphic
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
ACS Paragon Plus Environment

Page 3 of 19
Organic Process Research & Development
1
2
3
ABSTRACT
4
5
6
7
8
Commercial route definition for a glucokinase activator called for a reꢀevaluation of the
synthesis and processes used to access multikilogram quantities of a pyrazine building block.
The processes developed allowed a literature route to sodium 6ꢀoxoꢀ1Hꢀpyrazineꢀ3ꢀ
carboxylate to be leveraged. One of these processes consisted of a highly selective
decarboxylation that allowed the target building block to be accessed with complete
regioselectivity in standard batch processing equipment. The presence of an azetidine ring in
the target required the mitigation of impurity liabilities arising from the use of the
hydrochloride salt of azetidine as an input material.
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
KEY WORDS
azetidine, decarboxylation, glucokinase activator, pyrazine, sulfolane
ACS Paragon Plus Environment

Organic Process Research & Development
Page 4 of 19
1
2
3
INTRODUCTION
4
5
6
7
8
AZD1656 (
1), a compound containing two pyrazine units, has been developed by
AstraZeneca as a glucokinase activator for the treatment of type 2 diabetes.1ꢀ3 This article will
9
discuss the development of the manufacturing processes used to obtain chloropyrazine
(Scheme 1) one of the key building blocks of AZD1656 ( ).
Scheme 1. Principal disconnections of AZD1656 (
2,
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
1
)
AZD1656 (1)
2
ArOH =
TACTICAL CHANGES TO MEDICINAL CHEMISTRY APPROACH
The chemistry utilised by medicinal chemistry colleagues is outlined in Scheme 2. Ester
3
was first converted to acid
chloride used oxalyl chloride and was catalysed by DMF that tracked through from the first
stage. When AZD1656 progressed into development, potassium carbonate in THF was
successfully introduced for the conversion of ester to acid Dichloromethane was
developed out of the acid chloride formation as part of a switch to the use of thionyl chloride
and catalytic tetrabutylammonium chloride in toluene. Acid chloride was then coupled
4 using lithium chloride in DMF. The conversion of acid 4 to acid
5
3
4.
5
with an aqueous solution of azetidine whose alkalinity was maintained above pH 9 using
ACS Paragon Plus Environment

Page 5 of 19
Organic Process Research & Development
1
2
3
4
5
potassium carbonate. With these changes in place, 130 kg of chloropyrazine
as part of initial development deliveries.
2 were delivered
6
7
8
Scheme 2. Medicinal chemistry approach to chloropyrazine
2
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
a
91 %
3
4
b
c
77 % over
2 steps
5
2
(a) LiCl, DMF; (b) (COCl)2, CH2Cl2;
(c) (i) (CH ) N HCl, Et N, (ii) chromatography
•
2 3
3
DEVELOPMENT OF AZETIDINE ACYLATION
As development progressed, the use of externally sourced acid
on cost grounds versus the continued use of ester as the starting material. Development
efforts focused at this stage on the amidation of acid chloride with azetidine ( ). Sourcing
4
proved to be more attractive
3
5
6
azetidine free base was ruled out due to its tendency to oligomerize,4 necessitating the
sourcing and use of the hydrochloride salt.5 Accelerating rate calorimetry of this salt
indicated an onset of exothermic activity from ~71 °C, and 29 °C was evaluated as the
temperature at which it would take 24 h for the decomposition to reach its maximum rate
under adiabatic conditions.6 It was shipped and stored in 100 kg lots kept at 5 °C.7
Compounds
manufactured to date, implied the opening of the azetidine ring at some stage. Adding the
acid chloride solution to a basic solution of azetidine was found to be important in limiting
the ringꢀopening of the azetidine ring of chloropyrazine by chloride, though at the expense
7 and 8, two of the impurities repeatedly present in amide 2 that had been
5
2
of the formation of impurity
9
. 3ꢀChloropropylamine (10) in the azetidine hydrochloride
ACS Paragon Plus Environment

Organic Process Research & Development
Page 6 of 19
1
2
3
4
5
6
7
8
input material was identified as the principal source of impurity
chloropropylamine reacts faster with acid chloride than azetidine so any increase in the
charge of azetidine relative to acid chloride , would increase levels of alkyl chloride 7.
7. It was also shown that 3ꢀ
5
5
9
Whilst a typical level of 3ꢀchloropropylamine in azetidine hydrochloride was of the order of
0.5–1.0% wt/wt, 10% wt/wt was detected in one batch that had been destined for a kilogram
lab manufacture. Fortuitously, it proved possible to control levels of 3ꢀchloropropylamine by
washing the aqueous azetidine solution with toluene prior to use in the coupling with acid
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
chloride 5.
7
8
9
10
Figure 1. Impurities in chloropyrazine
2
REVISED SYNTHESIS OF CHLOROPYRAZINE 2
Whilst a further 430 kg of chloropyrazine were delivered starting from acid
occasionally required drying on receipt before it could be successfully engaged in the
transformation to acid chloride and it was proposed that it would be cheaper to manufacture
2
4, the latter
5
it inhouse.8 Ideas for its synthesis, some of which use validated literature precedent,9ꢀ13 are
shown in Scheme 3. The route disclosed by Mano et al. was selected for development based
ACS Paragon Plus Environment

Page 7 of 19
Organic Process Research & Development
1
2
3
on its use of inexpensive and readily available diaminomaleonitrile (11) and glyoxylic acid
4
5
6
(12) as starting materials.
7
8
Scheme 3. Possible means of synthesizing acid
4
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
reference 11
references 9-10
4
reference 13
12
11
reference 12
Scheme 4. Revised synthesis of chloropyrazine
2
c
a
b
OH
•
2
14
Na 16
Na 13
•
11
•
e
f
d
13
5
4
g
5
2
Na 17
15
16
•
a) glyoxylic acid, sulfuric acid, water, 30 C; b) i) NaOH (aq), 10 C, ii) 100 C,
°
°
°
iii) 10 C, H SO (aq); c) sulfolane, toluene, Aliquat 336, 150 C; d) i) NaOH (aq), ii) HCl (aq);
°
°
2
4
e) i) PCl , toluene, 80 C, ii) water; f) i) nBu NCl, toluene, 73 C, ii) SOCl ; g) i) (CH ) NH Cl,
°
°
5
4
2
2 3
2
K CO , toluene, water, ii) acid chloride
5
2
3
ACS Paragon Plus Environment

Organic Process Research & Development
Page 8 of 19
1
2
3
4
5
6
7
8
9
The studies of Mano et al., which ultimately take us as far as the sodium salt Na•13, start
with the condensation of equimolar amounts of diaminomaleonitrile and glyoxylic acid
monohydrate.9,14 Whilst their condensation to form bisnitrile 14 proceeds without the need
for exogeneous acid, the use of a catalytic amount of sulfuric acid was found to accelerate the
reaction. It also avoided the initial formation of a voluminous suspension with poor mixing
properties when no acid was used. The decision not to isolate bisnitrile 14 was made after
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
initial labꢀscale isolations provided a lowꢀmelting solid (mp <40 °C), which darkened on
drying. Compound 15, obtained from the partial hydrolysis of the orthoꢀdisposed nitriles of
bisnitrile 14 under basic conditions, was initially targeted. Whilst it proved possible to
convert compound 15 to diacid 16 using an acidic aqueous medium and heat, the material so
formed was difficult to isolate and dark in colour.
Attention thus turned to the
decarboxylation of diacid 16, salts of which could be accessed by heating bisnitrile 14 with
sodium hydroxide. A subsequent acidification to pH 2 permitted the isolation of the
monohydrate of the monosodium salt of diacid 16. In the presence of the sodium sulfate
produced as a coproduct of the neutralisation process, this material showed very low
solubility in the acidic aqueous media used for its isolation. In order to control the amount of
sodium sulfate produced, the sodium hydroxide charge was limited to 5 molar equivalents.
The acidification of the liquors to below pH 2 was found to promote the decarboxylation of
the small amount of bisacid 16 they contained, as well as darkening the colour of the
monosodium salt Na•16 isolated and negatively impacting its filtration properties.
For cost reasons, the glyoxylic acid monohydrate solid that was initially used was
replaced with its aqueous solution. This came with the drawback that this material is
typically contaminated with glyoxal and oxalic acid. Fortunately, whilst the former reacts
with diaminomaleonitrile (11), the pyrazineꢀ2,3ꢀdicarboxylic acid so formed (ca. 1.5 LC area
percent at the end of the reaction), is conveniently purged in the mother liquors from the
ACS Paragon Plus Environment

Page 9 of 19
Organic Process Research & Development
1
2
3
4
5
6
7
8
crystallisation of monosodium salt Na•16. The procedure was successfully used to process
890 kg diaminomaleonitrile (11), as part of a manufacturing campaign, resulting in the
isolation of sodium salt Na•16 in an average yield of 92%.
9
Sodium salt Na•16 could be decarboxylated to sodium salt Na•13 under forcing
thermal conditions (>180 °C), that were beyond the capability of standard reactor vessels.
Catalytic amounts of Brønsted acid were initially used in sulfolane in an attempt to partially
solubilise sodium salt Na•16 and to lower the temperature required for the decarboxylation to
one that could be achieved using standard plant equipment. Whilst successful, this approach
formed sodium salt Na•13 that was dark in colour and of low strength. The replacement of
the Brønsted acid by phaseꢀtransfer catalysts Aliquat HTAꢀ1 or Aliquat 336 allowed the
decarboxylation to proceed without excessive darkening of the reaction mixture. The latter
was used for cost reasons, despite its tendency to degrade under the reaction conditions.
Controlling the water content of the reaction mixture prior to the addition of the catalyst
helped to control the colour of the sodium salt Na•13 produced during the decarboxylation,
and was achieved through azeotropic distillation with toluene.
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
None of the regioisomeric 2,6ꢀdisubstituted product Na•17 arising from
decarboxylation at the 3ꢀposition of the pyrazine ring was detected during the conversion of
Na•16 to Na•13. A possible explanation for the regioselectivity of the reaction is the loss of
carbon dioxide to initially form a carbanion 18 that is in the plane of the σꢀframework of the
pyrazine ring. If such a species were to form at the 3ꢀposition, as in carbanion 19, then it
would be destabilised by a lone pairꢀlone pair repulsion (Scheme 5). Consistent with this
possible explanation, limited calculations performed using dimethyl sulfoxide as a surrogate
solvent for sulfolane, showed the free energy change was more unfavourable when the
decarboxylation took place at the 3ꢀposition, rather than the 2ꢀposition (Scheme 5). Whilst
the decarboxylation is an irreversible process, application of the BellꢀEvansꢀPolanyi principle
ACS Paragon Plus Environment

Organic Process Research & Development
Page 10 of 19
1
2
3
4
5
6
7
8
9
might reasonably lead to an expectation that the respective activation energy barriers reflect
this difference in free energy. The solution calculations additionally indicated that the
carboxylate virtually exclusively resides at the 2ꢀposition, rather than the 3ꢀposition, further
biasing the outcome of the decarboxylation.
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 5
CO
−
2
3
2
Na 16
•
G
rxn = +57.6 kJ/mol
ꢀ
2
−
18
G = 27.2
kJ/mol
ꢀ
−
Calculations performed at
IEFPCM//B3LYP/6-31+G(d,p)
level of theory
CO
−
2
−
3
3
2
Grxn = +90.0 kJ/mol
ꢀ
19
Initial attempts at isolating sodium salt Na•13 involved adding ethyl acetate to the sulfolane
(mp 27.5 C) solution to allow the slurry to be manipulated at ambient temperatures. This led
°
to sodium salt Na•13 displaying a very small particle size, to the detriment of the filtration
time. Washing the cake with ethyl acetate failed to completely displace all of the sulfolane,
and was implicated in the formation of further fines which reduced the filtration performance
still further. These issues were overcome by filtering the slurry produced at the end of the
reaction at 70 °C, in order to avoid the freezing of the sulfolane solvent, and then washing the
cake with toluene.
The presence of residual sulfolane in sodium salt Na•13 on scaling up was implicated in
balling observed when in a dryer, as well as significantly lowering the yield as part of its
subsequent conversion to acid 4. The development of a process that would completely rid
isolated sodium salt Na•13 of sulfolane was mindful of the lack of recrystallization options
afforded by its insolubility in organic solvents. The discovery that the yield for the
ACS Paragon Plus Environment

Page 11 of 19
Organic Process Research & Development
1
2
3
conversion to acid
4
could be increased from 60–68% to 85–90% by switching from
4
5
6
7
8
(sulfolaneꢀfree) sodium salt Na•13 to the corresponding free acid as the starting material,
guided the decision to target the isolation of acid 13 itself. This was conveniently done by
dissolving sodium salt Na•13 off the filter with an alkaline solution, washing with toluene, to
remove any residual Aliquat 336 and the products of its decomposition, before acidifying
with hydrochloric acid to precipitate the free acid in a form that filtered easily. This
delivered 831 kg of high strength (98% wt/wt) acid 13 in an average yield of 77% from
sodium salt Na•16. It was recognised that the longꢀterm use of hydrochloric acid for the
acidification ran the risk of corroding parts of the plant, though its use was retained for
development manufacture, as the use of sulfuric acid led to the contamination of the sodium
salt Na•13 isolated with sodium sulfate.
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 double chlorination of acid 13 to acid chloride
5 using just thionyl chloride did
not proceed to completion. The finding that it could be accelerated using phosphorus
pentachloride as an additive led to the use of phosphorus pentachloride in toluene. In the
optimised procedure, the reaction was initially run at 50
°
C to selectively complete the
before it was heated to 85 C in order to chlorinate the
formed from acid in this manner was not deemed to be of
. To circumvent this, the
phosphorus oxychloride byꢀproduct of the chlorination was distilled off and the initially
formed acid chloride was hydrolysed to acid , a material that crystallised out of aqueous
solution. This initially produced a reddish brown solid, though the treatment of the acid
chloride solution with charcoal, prior to the hydrolysis, allowed material that was a lighter
brown and displayed a strength of >99.5% wt/wt to be reproducibly isolated. The procedure
formation of the acid chloride
pyrazinone ring. Acid chloride
5
°
5
4
sufficient quality for use in the manufacture of chloropyrazine
2
5
4
5
was used to deliver 698 kg of acid
manufacturing campaign. The acid
4
in an average yield of 86% from acid 13 as part of a
4
so isolated was elaborated to chloropyrazine using the
2
ACS Paragon Plus Environment

Organic Process Research & Development
Page 12 of 19
1
2
3
4
5
previously established procedure which involved reforming acid chloride
with azetidine.
5 and then coupling
6
7
8
9
CONCLUSIONS
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 means of manufacturing chloropyrazine
2 on a large scale has been developed. One of the
challenges that had to be overcome involved handling azetidine, a material that poses process
safety concerns either as its free base or hydrochloride salt, and whose facile ringꢀopening by
chloride had to be addressed. Due to concerns over the handling and transport of azetidine
hydrochloride, the inhouse hydrogenolysis of Nꢀbenzhydrylazetidine was envisaged as a
more appropriate means of accessing azetidine over the longꢀterm. Issues associated with
using sulfolane as solvent for a high temperature decarboxylation also had to be overcome as
part of the development of an economically viable, regioselective and scalable means of
accessing acid 13. Whilst the number of chemical transformations (excluding salt breaks) in
the route to chloropyrazine
deliver 647 kg of chloropyrazine
of 44.3% yield from diaminomaleonitrile (11). With further development, there may well be
the opportunity to access chloropyrazine directly from acid 13 without the need to isolate
acid
EXPERIMENTAL SECTION
2
is high (7), it is replete with control points. It was used to
2
in a single manufacturing campaign, with an overall yield
2
4
.
General. NMR spectra were recorded on a Bruker instrument with tetramethylsilane (TMS)
as internal reference, unless otherwise stated. Chemical shifts are expressed in ppm (δ)
relative to TMS, whilst coupling constants (J) are in Hz. Infrared spectra were recorded with
diamond ATR sampling on an FTIR spectrometer using powdered samples. Melting points
were recorded using a Q2000 differential scanning calorimeter from TA Instruments. High
ACS Paragon Plus Environment

Page 13 of 19
Organic Process Research & Development
1
2
3
4
5
resolution mass spectra were recorded on a Waters Synapt UPLCꢀMS instrument. Unless
stated otherwise, commercial grade materials were used.
6
7
8
9
Sodium 3-carboxy-6-oxo-1H-pyrazine-2-carboxylate monohydrate (Na•16).
Aqueous glyoxylic acid solution (244 kg of 50% wt/wt, 1.65 kmol)15 was added over at least
30 min. to a solution of diaminomaleonitrile (178 kg, 1.65 kmol) and sulfuric acid (35.6 kg of
50% wt/wt, 181 mol) in water (1041 L). The batch held for a further 30 min. at 30 °C before
it was polish filtered, washing through with more water (89 L).16 1H NMR (referenced to
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
water peak at δH = 4.79, 500 MHz, 27
°
C) 8.24 (1H, s). 13C NMR (unreferenced in water,
125.7 MHz, 27 C) 160.6, 144.5, 129.8, 120.6, 114.3, 112.6.
°
Sodium hydroxide solution (660 kg of 50% wt/wt, 8.25 kmol) was then added to the
filtrate with cooling such that the temperature did not rise above 10 °C during the addition of
the first 60% of the charge.17 The batch was then aged for 6 h at 100 °C before being cooled
to 60 °C and diluted with water (712 L).18 After cooling to 10 °C,19 the pH was adjusted to 2
using sulfuric acid solution (ca. 712 kg of 50% wt/wt, 3.63 kmol),20 resulting in the
crystallisation of monosodium salt Na•16. The batch was allowed to desupersaturate for a
further 30 min., before it was filtered and the cake sequentially washed with water (823 L)
and isopropanolꢀwater (826 L of 67%v/v). Vacuum drying at 45 °C gave monosodium salt
monohydrate Na•16 as a tan solid (339 kg at 99.1% wt/wt, 90.8%). Mp 144.0 °C. IR 3158,
1
1669, 1594, 1524, 1464, 1441, 1345, 1096, 964, 835, 814, 781, 759, 710, 643 cmꢀ1. H NMR
(DMSOꢀd6, 500 MHz, 27
°
C)
δ
11.21 (1H, br s), 8.12 (1H, s), 3.34 (2H, br s). 13C NMR
H
(DMSOꢀd6, 125.7 MHz, 27
°
C)
δ 163.4, 161.3, 154.7, 150.1, 134.8, 125.4. HRMS
C
elemental calculated for C6H3N2O5 (M–H+ where M is diacid 16): 183.0042; found:
183.0039.
ACS Paragon Plus Environment

Organic Process Research & Development
Page 14 of 19
1
2
3
4
5
6
7
8
9
6-Oxo-1H-pyrazine-3-carboxylic acid (13). A slurry of monosodium salt Na•16
(95.0 kg at 99.1% wt/wt, 420 mol) in sulfolane (420 kg) and toluene (40 L) was prepared and
vacuum distilled at ca. 60 °C to remove some of the toluene (10 L). Aliquat 336 in toluene
(28 kg of 50% wt/wt, 34.6 mol) was added, washing through with more toluene (5 L). The
batch was vacuum distilled at 60 °C, using a DeanꢀStark arrangement with a tolueneꢀfilled
receiver, until the water content was ≤ 0.3% wt/wt (ca. 2 L water removed). The batch was
then aged for 2.5 h at 150 °C before being cooled to 70 °C, and filtered hot.21 The cake was
washed with toluene (137 L), before being dissolved off the filter with sodium hydroxide
solution (825 kg of 2.36% wt/wt, 487 mol). The small organic phase was separated off,
before the aqueous phase was washed with more toluene (58 L). The aqueous phase was
adjusted to pH 1.0 using hydrochloric acid (ca. 90 kg of 33% wt/wt, 815 mol). The
temperature was kept below 45 °C during the addition, until the pH was 6.5 whereupon the
temperature was adjusted to 75 °C for the remainder of the addition. Activated charcoal (3.5
kg Chemviron PWA) was stirred with the batch for 90 min whilst it was at 75 °C. It was then
warmed to 90 °C and filtered, washing through with hydrochloric acid solution (32 kg of
2.1% wt/wt, 18 mol). The temperature of the filtrate was then lowered to 5 °C over at least
90 min. and the slurry filtered. The cake was washed with water (3 x 67 L), before it was
vacuum dried below 55 °C using a paddle dryer. This afforded acid 13 as a light brown solid
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
(45.4 kg, 77.1%). Mp 269.2 °C. IR 2889, 1724, 1614, 1389, 1301, 1249, 1200, 1135, 877,
1
790, 785, 729, 633 cmꢀ1. H NMR (DMSOꢀd6, 500 MHz, 27
°
C)
δ
8.05 (1H, d, J=1.1 Hz),
H
7.98 (1H, d, J=1.1 Hz). 13C NMR (DMSOꢀd6, 125.7 MHz, 27
°C) δ 164.6, 155.9, 147.8,
C
133.1, 124.0. HRMS elemental calculated for C5H3N2O3 (M–H+): 139.0144; found:
139.0143.
5-Chloropyrazine-2-carboxylic acid (4). A slurry of acid 13 (48.0 kg, 343 mol) in
toluene (195 L) was added to a solution of phosphorus pentachloride (150 kg, 720 mol) in
ACS Paragon Plus Environment

Page 15 of 19
Organic Process Research & Development
1
2
3
toluene (200 L) that was being stirred at 0
°C. The addition was washed through with more
4
5
6
toluene (85 L). After being heated for 2 h at 80 °C, a put (480 L) and take (ca. 530 L)
7
8
9
vacuum distillation was performed at 70
was stirred with the batch for 30 min at the same temperature, before the batch was cooled to
20 C and filtered into a receiver containing water (240 L), washing through with more
toluene (48 L). The mixture was stirred at 20 C for 10 h, and at least 60 min at 0 C before
it was filtered. The cake was sequentially washed with water (48 L) and toluene (48 L), and
°C. Activated charcoal (1.2 kg Chemviron PWA)
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
°
°
°
dried below 45
°C to give acid
4
as a white solid (46.5 kg, 85.6%). Mp 157.5 °C. IR 1730,
1
1319, 1310, 1264, 1158, 1120, 1029, 911, 844, 826, 799, 728, 720 cmꢀ1. H NMR (DMSOꢀ
d6, 500 MHz, 27
125.7 MHz, 27
°
C)
δ
δ
9.01 (1H, s, J=1.3 Hz), 8.92 (1H, s, J=1.3 Hz). 13C NMR (DMSOꢀd6,
H
°C)
164.4, 151.1, 145.5, 144.4, 142.5. HRMS elemental calculated for
C
C5H2ClN2O2 (M–H+): 156.9805; found: 156.9801.
5-Chloropyrazine-2-carbonyl chloride (5). A slurry of acid
4 (320 kg, 2.02 kmol)
and tetrabutylammonium chloride (6.4 kg, 23 mol) was slurried in toluene (1220 L) and
heated to 73 °C. Thionyl chloride (341 kg, 2.87 kmol) was dosed in over at least 60 min and
rinsed through with more toluene (230 L). After stirring for a further 4 h, a put (1530 L) and
take (ca. 1700 L) vacuum distillation was performed at 50 °C. The put (1700 L) and take (ca.
1700 L) vacuum distillation was then repeated, before more toluene (2856 L) was added.
This solution of acid chloride
5
(ca. 9.0% wt/wt) could be stored at 20 °C. A sample was
1
analysed by 1H NMR after solvent swapping from toluene to CDCl3. H NMR (CDCl3, 400
MHz, 27 °C) δ 9.11 (1H, d, J=1.3 Hz), 8.82 (1H, d, J=1.3 Hz).
H
Azetidin-1-yl(5-chloropyrazin-2-yl)methanone (2). An aqueous potassium carbonate
solution (508 kg of 12% wt/wt, 0.44 kmol) was added to a slurry of azetidine hydrochloride
(83 kg, 0.89 kmol)22 in toluene (777 L) such that the temperature remained below 20
°C.
ACS Paragon Plus Environment

Organic Process Research & Development
Page 16 of 19
1
2
3
After washing through with water (327 L), the biphasic mixture was stirred for 30 min at 20
4
5
6
7
8
°
C before being optionally filtered. The phases were cut and the aqueous phase washed with
toluene (2 x 777 L). Potassium carbonate (229 kg, 1.66 kmol) was added to the aqueous
phase after the first toluene wash. A known fraction of acid chloride solution in toluene
9
5
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
(1598 kg at 9.0% wt/wt, 0.81 kmol) prepared above was then added over 60 min., before
being washed through with toluene (159 L). After a further 30 min, the reaction mixture was
filtered and the filter cake rinsed with toluene (230 L). The phases were cut and the organic
phase sequentially washed with water (767 L), hydrochloric acid solution (140 kg of 33%
wt/wt diluted with 767 L water), and then sodium chloride solution (896 kg of 25% wt/wt).
A vacuum distillation at about 50
Heptanes (1635 L) were then added and the batch treated with activated charcoal (16 kg) for
60 min at 95 C. The batch was filtered hot, washing through with hot heptane (166 L). The
temperature was then adjusted to 55 C, prior to seeding with chloropyrazine (0.1 kg)
°C was then used to remove some of the toluene (2270 L).
°
°
2
slurried in heptane (7 L). Desupersaturation to a thick slurry was allowed to proceed
whereupon the jacket temperature was linearly ramped over at least 3 h so that the final batch
temperature was below
before the slurry was isolated and the cake washed with petroleum ether (2 x 184 L of
fraction 140–155 C). The wet cake was vacuum dried at 45 C to give chloropyrazine as a
white solid (117 kg, 73.4% from acid , based on the use of 40.0% of the acid chloride
−5 °C. Desupersaturation was allowed to proceed for a further 3 h,
°
°
2
4
5
solution). Mp 104.2 °C. IR 1624, 1434, 1297, 1140, 1122, 1024, 991, 920, 791, 707, 610
1
cmꢀ1. H NMR (DMSOꢀd6, 500 MHz, 27
°
C)
δ
8.91 (1H, d, J=1.3 Hz), 8.83 (1H, d, J=1.3
H
Hz), 4.53 (2H, dd, J=7.7, 7.7 Hz), 4.10 (2H, dd, J=8.1, 7.6 Hz), 2.29 (2H, dddd, J=8.1, 7.7,
7.7, 7.6 Hz). 13C NMR (DMSOꢀd6, 125.7 MHz, 27
°
C) δ 162.1, 149.7, 145.0, 144.4, 142.8,
C
54.1, 48.9, 16.1. HRMS elemental calculated for C8H9N3OCl (M+H+): 198.0434; found:
198.0440.
ACS Paragon Plus Environment

Page 17 of 19
Organic Process Research & Development
1
2
3
4
5
ASSOCIATED CONTENT
Supporting information available: H NMR spectra, C NMR spectra, infrared spectra and
1
13
6
7
8
DSC traces for sodium salt Na•16, acid 13, acid 4 and chloropyrazine 2.
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
AUTHOR INFORMATION
Corresponding Author
Eꢀmail: Alan.Steven@astrazeneca.com
ORCID Alan Steven: 0000ꢀ0002ꢀ0134ꢀ0918
ORCID David R. J. Hose: 0000ꢀ0003ꢀ3872ꢀ7996
Notes
The authors declare no competing financial interest.
References
McKerrecher, D.; Waring, M. J., Chapter One ꢀ PropertyꢀBased Design in the
1.
Optimisation of Benzamide Glucokinase Activators: From Hit to Clinic. In Progress in
Medicinal Chemistry, Lawton, G.; Witty, D. R., Eds. Elsevier: 2013, 52, 1.
2.
Waring, M. J.; Clarke, D. S.; Fenwick, M. D.; Godfrey, L.; Groombridge, S. D.;
Johnstone, C.; McKerrecher, D.; Pike, K. G.; Rayner, J. W.; Robb, G. R.; Wilson, I., Property
based optimisation of glucokinase activators ꢀ discovery of the phase IIb clinical candidate
AZD1656. Med. Chem. Commun. 2012, 3, 1077.
3.
Processes for making AZD1656 are covered in the following patents. a) Butters, M.
Crabb, J.; Hopes, P.; Patel, B. Process for preparation of benzoyl amino heterocyclic
compounds. WO 2010092387, August 19, 2010. b) Bowden, S. A.; Hoile, D. P.; Lövqvist,
K. Crystalline polymorphic form 631. WO 2010092386, August 19, 2010. c) Briggner, L.ꢀ
ACS Paragon Plus Environment

Organic Process Research & Development
Page 18 of 19
1
2
3
4
5
E.; Lindfors, L.; Smith, C. M. Pharmaceutical formulations. WO 2012007758, January 19,
2012.
6
7
8
4.
Goethals, E. J.; Schacht, E. H.; Bogaert, Y. E.; Ali, S. I.; Tezuka, Y. Polym. J. (Tokyo)
9
1980, 12, 571.
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
5.
Alabanza, L. M.; Dong, Y.; Wang, P.; Wright, J. A.; Zhang, Y.; Briggs, A. J. Org.
Process Res. Dev. 2013, 17, 876.
6.
Azetidine hydrochloride was found to decompose with an energy of –820 J/g, though
without the attendant gas evolution that would indicate a detonation or deflagration, and there
was no sensitivity to impact in a BAM Fallhammer test.
7.
temperature controlled”).
8. Alternative means of accessing chloropyrazine
ideated but were not investigated due to the positive and established experience that existed
around the conversion of acid to chloropyrazine . A further attraction of accessing
chloropyrazine from acid and acid chloride was the stability of toluene solutions of the
latter which enabled it to be stored for months at 21 °C.
The material was classified as a solid with UN number 3238 (“selfꢀreactive solid type,
2
that did not route via acid 4 were
4
2
2
4
5
9.
Mano, M.; Seo, T.; Imai, K. Chem. Pharm. Bull. 1980, 28, 3057.
10.
Hosogai, T.; Nishida, T.; Stoi, K.; Takagi, T. 2ꢀSubstituted 5,6ꢀdicyanoꢀ3ꢀ
hydroxypyrazines. JP 50059379, May 22, 1975.
11.
12.
13.
814.
14.
Sato, N.; Arai, S. J. Heterocyclic Chem. 1982, 19, 407.
Sato, N.; Fujii, M., J. Heterocyclic Chem. 1994, 31, 1177.
Kiener, A.; Roduit, J. P.; Tschech, A.; Tinschert, A.; Heinzmann, K. Synlett 1994, 10,
The use of the conditions of Mano et al. led in our hands to sodium salt Na•13 that
filtered slowly as part of its isolation and was dark in colour. Also, whilst the HPLC purity
ACS Paragon Plus Environment

Page 19 of 19
Organic Process Research & Development
1
2
3
4
was high (>95 LC area percent), the assay varied from 30
−85% wt/wt. As a result of these
5
6
characteristics, redesign of the literature conditions was necessary.
7
8
9
15.
0.95–1.05 eq. window versus with respect to the amount of diaminomaleonitrile.
16. Filtering after the reaction mixture has been left for longer or held at a higher reaction
Bisnitrile 14 shows lower purity when the amount of glyoxylic acid used is outside a
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
temperature will remove insoluble intermediates arising from the hydrolysis of the nitrile
functionality of bisnitrile 14 to the corresponding amides. This acts to reduce the reaction
yield as these amides would otherwise be converted to the monosodium salt Na•16
17. Controlling the exotherm associated with the sodium hydroxide addition (T ≤ 10 °C),
prevents dark coloured sodium salt Na•16 from being isolated.
18. Diluting the batch with water whilst its temperature is above ambient prevents the
precipitation of the trisodium salt of acid 16
19. The sodium sulfate was found to precipitate out, whereupon it was challenging to
remove, if the temperature was below 10 °C during the desupersaturation.
20. No sodium sulfate was adjudged to have coprecipitated, according to a barium
chloride test to assess sulfate content.
21. Small quantities of unreactive disodium salt meant the reaction could never be pushed
beyond 98% conversion.
22. Azetidine hydrochloride can deliquesce, depending on its mode of manufacture.
.
.
ACS Paragon Plus Environment