Development of a scalable synthesis of a pyridinyl-3-azabicyclononene, a novel nicotinic partial agonist

Article abstract of DOI:10.1021/op400002r

The process research and development of two syntheses of a novel nicotinic partial agonist, TC-8817 ((+)-5), are described. The original Medicinal Chemistry route had multiple flaws, making it unsuitable for further development. A second approach was expl

Full text of DOI:10.1021/op400002r

Subscriber access provided by Univ. of Tennessee Libraries
Full Paper
Development of a Scalable Synthesis of a Pyridinyl-3-
azabicyclononene, a Novel Nicotinic Partial Agonist
Scott R. Breining, John F Genus, J. Pike Mitchener, Timothy J Cuthbertson,
Ronald Heemstra, Matt S Melvin, Gary M Dull, and Daniel Yohannes
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/op400002r • Publication Date (Web): 09 Feb 2013
Downloaded from http://pubs.acs.org on February 18, 2013
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 23
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Development of a Scalable Synthesis of a Pyridinylꢀ
3ꢀazabicyclononene, a Novel Nicotinic Partial
Agonist
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
Scott R. Breining*1, John F. Genus2, J. Pike Mitchener2, Timothy J. Cuthbertson 2, Ronald
Heemstra1, Matt S. Melvin1, Gary M. Dull2 and Daniel Yohannes1,2
1Formerly, Department of Medicinal Chemistry; 2 Formerly, Department of Chemical Process
and Development. Targacept, Inc. 200 E. First St. Suite 300, WinstonꢀSalem, NC 27101
KEYWORDS: Nicotinic, smoking cessation, 3ꢀazabicyclo[3.3.1]nonene, asymmetric, Corey
catalyst
ABSTRACT: The process research and development of two syntheses of a novel nicotinic
partial agonist TCꢀ8817 ((+)ꢀ5), are described. The original Medicinal Chemistry route had
multiple flaws, making it unsuitable for further development. A second approach was explored
which was more amenable to optimization. The key steps were an intramolecular Lewis Acid
promoted cyclization, a dibromination/elimination sequence to provide a vinyl bromide and
subsequent Suzuki coupling with 3ꢀpyridineboronic acid. The overall yield of ~3ꢀ16% over nine
steps was offset by the low cost of goods and ease of synthesis. A major drawback was the need
for SMB chiral separation on the penultimate intermediate to afford the subsequently desired
single enantiomer version. A thirdꢀ generation, asymmetric variation afforded a key intermediate
ACS Paragon Plus Environment
1

Organic Process Research & Development
Page 2 of 23
1
2
3
4
in good yield and enantiomeric purity, providing proof of concept for a more efficient production
5
6
of the desired (+)ꢀenantiomer.
7
8
9
TOC GRAPHIC
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
N
9 Steps
CHO
~6% yield
N
5
INTRODUCTION
Smoking is the single largest preventable cause of disease and premature death. Smoking relatedꢀ
diseases are considered responsible for the deaths of one in 10 adults globally; this figure
continues to rise steadily.1 In 2006, Pfizer’s Chantix® (varenicline, Figure 1, 1), a novel
treatment for the nicotine addiction believed responsible for smoking behavior, was approved by
the FDA. Varenicline is characterized as an α4β2 nicotinic acetylcholine receptor partial agonist,
the first of this therapeutic class to have reached the market and validate this mechanism.2
Further characterization has since suggested the mechanism of action may be more complicated,
involving activity at additional subtypes of nicotinic receptors containing α7 and α6 subunits.3,4
Consequently, we chose to pursue novel triple pharmacology ligands (α4β2,α7, α6) with a range
of potency and efficacy across these three subtypes to provide additional therapeutic possibilities
for treatment of nicotine addiction. Our initial Discovery efforts identified the 3ꢀmethylꢀ7ꢀ
ACS Paragon Plus Environment
2

Page 3 of 23
Organic Process Research & Development
1
2
3
4
(pyridinꢀ3ꢀyl)ꢀ3ꢀazabicyclo[3.3.1]nonꢀ6ꢀene (5, Scheme 1) as a promising compound for further
5
6
development.5
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
Figure 1. Varenicline (1).
The original Medicinal Chemistry route is shown in Scheme 1. This route served to support
synthesis of >250 analogs as well as the initial multiꢀgram batches for behavioral assays.6 The
route was tedious, low yielding, extremely costly and deemed impractical for further scaleꢀup
development.
Scheme 1. Original Medicinal Chemistry Route to Racemic 5.
O
O
N
OTf
N
O
1). PyrB(OH)2
Pd(PhP)4, DME
N
7 steps
~7%
LDA, THF
PhN(Tf)2
2). HCO2H, CH2O
~60%
HO2C
CO2H
-78 C
60-80%
N
O
O
O
2
3
4
5
A new route was designed within the Medicinal Chemistry group which addressed many of the
issues inherent in the original route shown in Scheme 1. This route (Scheme 2) relied on three
key steps: an intramolecular Lewis Acid promoted cyclization of 9 to give 107, the elimination of
dibromide 11 to afford vinyl bromide 12, and the subsequent Suzuki8 coupling with 3ꢀ
ACS Paragon Plus Environment
3

Organic Process Research & Development
Page 4 of 23
1
2
3
4
pyridineboronic acid to give 13. While still lengthy, the sequence worked fairly well and
5
6
appeared amenable to further optimization and scaleꢀup to support a development campaign.
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
Scheme 2. Improved Medicinal Chemistry Route to Racemic 5.
RESULTS AND DISCUSSION
While the improved route shown in Scheme 2 was adequate to meet the initial needs of the
program with respect to preclinical profiling, any further advancement of this compound would
need a more robust, efficient and costꢀeffective synthetic access. Scheme 3 illustrates the route
refinements developed within the Process group. Due to the high cost and limited availability of
the nitrile 6, an alternative method to obtain amine 7 was sought. Conversion of the inexpensive,
readily available carboxaldehyde 14 to an oxime was performed. Initially, this was conducted by
ACS Paragon Plus Environment
4

Page 5 of 23
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
addition of hydroxylamine to sodium hydroxide solution and aldehyde. On larger scale, this
resulted in multiple impurities. A simple order of addition switch, adding sodium hydroxide
solution to a binary solution of aldehyde and hydroxylamine hydrochloride, resulted in a much
cleaner product. Extraction of product from the aqueous mixture with 2ꢀmethyltetrahydrofuran
(2ꢀMeTHF) gave a solution suitable for direct use in the subsequent lithium aluminum hydride
step. This reduction proceeded well on scale to give amine 7 without need for further
modification. The formation of carbamate 8 required a thorough investigation, as multiple side
products were formed, particularly the undesired urea byproduct. Under the initial conditions
employing triethylamine as base, urea formation could actually predominate and yields and
purity of carbamate were highly variable. Other carbamates (benzyl, tertꢀbutyl) were prepared
and carried through the subsequent steps, but these groups proved labile during the Lewis acid
cyclization step. A switch to bicarbonate or carbonate base in combination with the ethyl
chloroformate alleviated the issue of urea formation. A drawback on scale was the potential for
foaming due to carbon dioxide. Aqueous sodium hydroxide solution was shown to be an
effective base, provided the base was added slowly to a solution of amine 7 and ethyl
chloroformate to avoid exothermic decomposition of ethyl chloroformate in the presence of
excess base. A yield of 98% of carbamate 8 was achieved when this step was conducted on 64 kg
scale.
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 3. Process Route to Racemic 5.
ACS Paragon Plus Environment
5

Organic Process Research & Development
Page 6 of 23
1
2
3
4
5
6
7
8
9
O
CHO
(CH2O)n
NH2
N
O
ClCO2Et
98%
1). NH2OH-HCl
H
BF3, CH2Cl2
62%
2). LiAlH4
96%
14
8
7
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
N
Br
N
Br
N
Br
B(OH)2
N
Br2, DCM
93%
KOt-Bu, THF
N
40% molar yield
Pd(OAc)2, DPPF,
Dioxane, aq.
O
O
N
O
N
O
O
N
O
Na2CO3
67%
O
O
10
11
12
13
CH2O, HCO2H
85%
c. HCl
46%
N
H
N
15
5
Formation of azabicyclononene 10 originally called for a twoꢀstep sequence of
hydroxymethylation to 9, followed by cyclization (as illustrated in Scheme 2). Many conditions
were screened to optimize the hydroxymethylation step, including varying equivalents of
paraformaldehyde, base (cesium carbonate, potassium carbonate, potassium phosphate),
temperature and solvent. The reaction was prone to stalling and sublimation of paraformaldehyde
often clogged the reactor; thus, this method proved untenable on larger scales. It was
subsequently discovered that use of ~1 eq. of BF3ꢀetherate gave the desired product 10 in a
single step (62% yield), reducing reactor time and providing a more controllable reaction. The
ACS Paragon Plus Environment
6

Page 7 of 23
Organic Process Research & Development
1
2
3
4
5
6
only drawback was the need to employ halogenated solvents such as dichloromethane; use of
THF, 2ꢀMeTHF or MTBE did not allow the reaction to go to completion.
7
8
9
Bromination to give dibromide 11 was conducted per the Medicinal Chemistry procedure but
with a critical change in workup. It was found that the thiosulfate quench originally employed
resulted in a suspension that was difficult to filter, and the product contained sulfur impurities
which had to be removed by column chromatography to avoid poisoning the palladium catalyst
in the later Suzuki reaction step. A switch to an ascorbic acid quench obviated the need for
purification. Dichloromethane was the solvent of choice, as the reaction had a tendency to stall
with other solvents. On 7 kg scale, a yield of 93% and purity of 88% were obtained.
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
Elimination of HBr to provide the vinyl bromide 12 was conducted in THF in the presence of 2.5
eq. of potassium tertꢀbutoxide. While near quantitative on smaller scales at room temperature,
the yields were reduced on scaleꢀup. On the large scale runs, heating was required in order to
achieve rapid completion (room temperature reactions took several days and tended to stall). The
increased temperature is a likely cause of reduced yields, though this reaction was not thoroughly
studied due to time constraints. A silica gel plug filtration was added on a kilogram scale run to
remove some minor impurities, but it was not strictly necessary. Material was obtained with 97%
mass recovery and 37% purity on an 11.8 kg scale (reported as a corrected molar yield of 40%)
and used directly in the subsequent step. Additional studies of the reaction indicated that an
isomeric impurity (likely the allylic bromide based on preliminary spectral data) present in the
vinyl bromide, but not readily removed by silica gel plug filtration, substantially reduced the
yield, reaction rate and product purity in the subsequent Suzuki coupling. Since the yields and
purities were much higher in smaller scale runs, and as this represents a major limitation to the
overall route, further study and optimization of this step are clearly needed.
ACS Paragon Plus Environment
7

Organic Process Research & Development
Page 8 of 23
1
2
3
4
5
6
7
8
9
The Suzuki coupling reaction of 12 was thoroughly explored, resulting in a switch from
Pd(Ph3P)4 to palladium(II) acetate/dppf and replacement of dimethoxyethane with dioxane.
Other variables were evaluated, such as temperature, stirring speed, choice of base, stoichiometry
and waterꢀsolvent ratio. No enhancements in yield or variability were achieved, though a DOE
approach was not undertaken to further improve this step. On a 2.95 kg scale, 13 was obtained in
a yield of 67% (corrected for the purity of the vinyl bromide 12) and a purity of 80.5%, but this
was dependent on scale and the quality of the vinyl bromide employed. The major impurity was
found to be ~14% of 3,3ꢀbipyridyl resulting from homocoupling of the 3ꢀpyridineboronic acid. It
should be possible to reduce this side reaction by charging the boronic acid gradually, avoiding
the presence of a large excess.
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
Removal of the ethyl carbamate protecting group from 13 required the very harsh conditions of
heating under reflux with conc. HCl. Unfortunately, attempts to alter pH and temperature did not
improve yields (46% on 3.9 kg scale; 57% corrected for purity of starting material) or variability,
though column chromatography was avoided by use of a neutral rinse followed by basification
and heptane slurry. When necessary, reduction of palladium levels was achieved through a
thiolized silica treatment of 15.
Finally, methylation under EschweilerꢀClark9 conditions afforded the desired Nꢀmethyl product 5
cleanly, rapidly and in high yield (85%). At this point, complete in vitro and in vivo evaluation
indicated a preference for the (+)ꢀenantiomer of 5, the eventual clinical candidate TCꢀ8817
(Scheme 4). An extensive classical resolution study was performed utilizing 46 chiral acids to
separate the desired enantiomer from the racemate. The only acid providing a significant degree
of enrichment was diꢀpꢀanisoylꢀDꢀtartaric. Attempts to find appropriate chromatographic
conditions for large scale chiral separation were also unsuccessful. It was discovered that the
ACS Paragon Plus Environment
8

Page 9 of 23
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
intermediate secondary amine 15 was more readily separable by chiral Supercritical Fluid
Chromatography (SFC) or Simulated Moving Bed (SMB) chromatographic techniques.
Ultimately, SMB chromatography was employed for the separation of material comprising the
batch for toxicology studies. It was necessary to make a salt selection at this point as well, as the
free base was an oil and the dihydrochloride salt utilized in early preclinical studies proved
hygroscopic. A salt screen provided the pꢀhydroxybenzoate as the best candidate, yielding
product as a white, crystalline solid (86% 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
Scheme 4. Chiral Separation Method for (+)-5 (TC-8817).
Chiral route process development
While chiral separation of the large batch was ongoing, investigations were conducted into the
feasibility of an asymmetric route. Literature precedent existed for the preparation of chiral ester
17, which would serve as a suitable starting point for elaboration to the single enantiomer version
of amine 7. Several methods, catalysts and conditions were explored, resulting in a facile and
robust method (Scheme 5). Cyclization of butadiene with trifluoroethyl acrylate 16 in the
ACS Paragon Plus Environment
9

Organic Process Research & Development
Page 10 of 23
1
2
presence of Corey’s catalyst, (R)ꢀ(+)ꢀoꢀtolylꢀCBSꢀoxazaborolidine trifluromethanesulfonimide10
afforded the desired ester 17 in good yield (74%). A threeꢀstep conversion to (-)-8 via aminolysis
to 18, reduction and ethyl carbamate formation proceeded in good yield (72% overall). Chiral
analytical methods were not yet developed for analysis of these intermediates, so material was
pushed through using the methods in Scheme 3 and the final product (+)-5 was evaluated for
chiral purity. One change to note was the use of a reductive removal of the carbamate in (+)-13
to directly give the desired (+)-5. This approach had the benefit of avoiding the difficult acid
hydrolysis. The overall yield from carbamate (-)-8 for this thirdꢀgeneration, asymmetric
synthesis of product (+)-5 was 19%, providing ~20 g with an e.e. of 70%. It was not clear if the
erosion of chirality was occurring in a single step or over multiple steps. In any event, classical
resolution through the diꢀpꢀanisoylꢀDꢀtartrate was sufficient to obtain the desired chiral purity
(>98% single enantiomer). Unfortunately, the project was discontinued before the asymmetric
route could be further optimized and implemented.
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
Scheme 5. Asymmetric Route to (+)-5 (TC-8817).
ACS Paragon Plus Environment
10

Page 11 of 23
Organic Process Research & Development
1
2
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
CONCLUSIONS
A scalable, nine step synthesis of (±)ꢀ5 was developed. The key steps were an intramolecular
Lewis Acid promoted cyclization, a dibromination/elimination sequence to provide a vinyl
bromide and subsequent Suzuki coupling with 3ꢀpyridineboronic acid. Though somewhat
lengthy and low yielding (~3ꢀ16% over nine steps), the route provided the necessary GMP
material on kilogram scale in racemic form, which would have been unobtainable by the
previous route. The major drawback was the need for chiral SMB separation on the penultimate
intermediate to afford the desired single enantiomer version, (+)ꢀ5, effectively further reducing
the yield by half. A third generation, asymmetric version was identified and explored on
laboratory scale. The (S)ꢀcyclohexenecarboxylic acid ester 17 was prepared in high yield and
enantiomeric purity according to known methods. This was subsequently converted to the (ꢀ)ꢀ
ACS Paragon Plus Environment
11

Organic Process Research & Development
Page 12 of 23
1
2
3
4
5
6
7
8
9
enantiomer version of the key intermediate, amine 7. This material was processed according to
the optimized route, providing proof of concept for a more efficient production of the desired
(+)-5, the clinical candidate TCꢀ8817.
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
EXPERIMENTAL SECTION
General
Purities were determined by HPLC and reported as area % according to the appropriate method.
Analytical method for the determination of purity: Column: Xterra RPl 8, 3.5 ꢁm, 100 x 4.6
mm; Gradient elution of 0ꢀ100% B over 45 min; Mobile Phase A: 0.01M NH4HCO3 : MeCN
95/5(v/v); Mobile Phase B: 0.01M NH4HCO3 : MeCN 50/50 (v/v); Flow Rate of 1 mL/min;
Wavelength of detection: 220 nM; Run time: 58 min. Analytical method for the determination
of chiral purity: Column: Chiralpak IC, 250 x 4.6 mm, 5 ꢁm, Chiral Technologies; Eluant:
95:5:0.1 Heptane:Ethanol:Diethylamine. Analytical method for the determination of chiral purity
of (+)-14 as prepared by SMB: Pump: LPGꢀ3400A or equivalent quaternary pump; Detector:
VWDꢀ3400 or equivalent UV detector; Autosampler: WPSꢀ3000 or equivalent autosampler;
Column: Daicel Chiralpak ADꢀH, 250 x 4.6 mm; Column temperature: 40°C; Eluent:
Isopropanol / Ethanol / Diethylamine 800:200:1 (v/v/v); Flow: 0.7 ml/min; Run time: 30 min;
Detection: UV at 245 nm; Maximum pressure: 100 bar; Injection volume: 5 ꢂL
(Autosampler). Proton NMR spectra were recorded on a Varian VNMR at 300 or 400 MHz in
the solvent indicated.
Optimized SecondꢀGeneration Route
Cyclohex-3-enylmethanamine (7).
A reaction vessel was charged hydroxylamine hydrochloride (44.8 kg, 644.7 mol) and water (305
L) under a nitrogen atmosphere. Stirring was initiated and 2ꢀMeTHF (301.1 L) was charged.
Aldehyde 13 (71.3 kg, 647.3 mol) was charged followed by a rinse with 2ꢀMeTHF (11.0 L). A
solution of 50% NaOH (34 L, 425 mol) was charged over 2 h to keep the pH at 9 and the
reaction temperature between 22ꢀ29°C. The reaction mixture was held for 1.5 h at 23°C. Upon
completion of reaction (>97% by GC), stirring was stopped and the phases were allowed to
separate over 20 min. The lower aqueous phase was removed. A 26% (w/w) solution of NaCl
(46.5 kg) was charged to the reactor. Stirring was initiated and the phases were mixed for 20
min. Stirring was stopped and the phases were allowed to separate over 15 min. The lower
aqueous phase was removed. A second charge of a 26% (w/w) solution of NaCl (46.5 kg) was
added to the reactor. Stirring was initiated and the phases were mixed for 20 min. Stirring was
ACS Paragon Plus Environment
12

Page 13 of 23
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
stopped and the phases were allowed to separate over 20 min and the lower aqueous phase was
removed. 2ꢀMeTHF (220.0 L) was added to the reactor containing the productꢀrich 2ꢀMeTHF
fraction and stirring was initiated. The solvent was removed under reduced pressure (250→120
mbar) at 27°C over 10 h to collect 223 L of distillate. The yield of the cyclohexꢀ3ꢀ
enecarboxaldehyde oxime intermediate was 79.5 kg (98.1%) as a 28.9% (w/w) solution in 2ꢀ
MeTHF (275.1 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
A reaction vessel was purged with nitrogen and charged with a 2.3 M solution of lithium
aluminium hydride (295 kg, 763.2 mol) in THF. Stirring was initiated and the above solution of
cyclohexꢀ3ꢀenecarboxaldehyde oxime (274.5 kg) was charged at 19ꢀ23°C over 5.5 h. The
reactor contents were then heated to 56ꢀ57°C over 2 h, and the mixture was held for 12 h. The
reaction mixture was cooled to 24°C over 2 h. Water (29.3 L) was charged over 2 h keeping the
pot temperature between 20ꢀ48°C. A 15.5 % (w/w) solution of NaOH (33.2 kg) was charged at
35ꢀ38°C over 1 h. Water (28.9 L) was charged at 35ꢀ37°C over 40 min. The reaction mixture
was then cooled to 25°C over 1 h. The suspension was centrifuged (3 loads) to collect the
filtrate. The wet cake (180 kg) was discharged and returned to the reactor. THF (196 L) was
charged to the reactor and stirring was initiated. The mixture was stirred for 8 h at 21ꢀ22°C. The
suspension was centrifuged (3 loads) to collect the filtrate. The wet cake (151 kg) was
discharged and returned to the reactor. THF (175 L) was charged to the reactor and stirring was
initiated. The mixture was stirred for 2 h at 21ꢀ22 °C. The suspension was centrifuged to collect
the filtrate. The wet cake (151 kg) was discarded. The three filtrates (731 kg) were concentrated
under reduced pressure (700→32 mbar) at 28ꢀ40°C over 10 h, collecting 345 L of distillate. The
resulting 16% (w/w) solution (422.3 kg) of 7 (67.6 kg, 95.9% yield) was used directly for the
preparation of 8. An analytical sample was obtained by treatment of an aliquot with excess
1
methanolic HCl and evaporation of solvent. H NMR (400 MHz, CD3OD) 5.76ꢀ5.65 (m, 2H),
2.87 (d, J=7.1 Hz, 2H), 2.18 (d, J=6.9 Hz, 1H), 2.11 (s, 2H), 1.98ꢀ1.75 (m, 3H), 1.39ꢀ1.29 (m,
1H). 13C NMR (400 MHz, CDCl3) 129.2, 126.7, 36.9, 34.8, 31.0, 28.1, 26.5. MP: 218ꢀ220°C.
HRMS (ES) m/z Calcd. for C7H13N+H: 112.1126. Found: 112.1130.
Ethyl cyclohex-3-enylmethylcarbamate (8).
To an inerted reactor was charged a 32% NaOH solution (74.5 kg, 596.0 mol). The reactor
contents were cooled to 5°C with stirring. A 16% (w/w) solution of 7 (401.5 kg of solution, 64.2
kg, 577 mol) in 2ꢀMeTHF/THF (401.5 kg) was charged at 2ꢀ5°C over 1 h. Ethyl chloroformate
(64.5 kg, 594.3 mol) was charged at 0ꢀ5°C over 4 h. The reaction mixture was stirred for 1 h at
2ꢀ4°C. Upon completion of reaction (< 1.0% of 6 by GC), water (66.8 L) was charged at 2ꢀ4°C
over 10 min. A 26.2 % (w/w) solution of NaH2PO4 (12 kg) was charged to the reactor at 3ꢀ5°C
over 15 min. The reaction mixture was stirred for 7.5 h at 5ꢀ10°C. The reactor contents were
warmed to 22°C and stirring was stopped. The phases were allowed to separate over 30 min.
The lower aqueous layer (140 L) was removed. Water (67 L) was charged and the reactor
contents were mixed for 20 min. Stirring was stopped and the phases allowed to separate over
45 min at 24°C. The lower aqueous layer was removed (60 L). A 26% (w/w) brine solution (81
kg) was charged to the reactor, and the contents mixed for 20 min at 20ꢀ23°C. Stirring was
stopped and phases were allowed to separate over 30 min. The lower aqueous layer (90 L) was
removed. The organic phase was concentrated to dryness under reduced pressure (300→3 mbar)
at 36ꢀ40°C over 15 h, removing 290 L of solvent. The yield of 8 was 104 kg (98.3%) as a
ACS Paragon Plus Environment
13

Organic Process Research & Development
Page 14 of 23
1
2
1
3
yellow oil. H NMR (400 MHz, CDCl3) 5.68 (m, 2H), 4.70 (br s, 1H), 4.12 (q, J=6.6 Hz, 2H),
3.10 (br s, 2H), 2.12ꢀ2.06 (m, 3H), 1.77ꢀ1.62 (m, 4H), 1.24 (t, 6.7 Hz, 3H). 13C NMR (400 MHz,
CDCl3) 159.1, 129.7, 128.3, 63.2, 49.0, 36.9, 31.8, 28.8, 27.3, 17.3. HRMS (ES) m/z Calcd. for
C10H17NO2+H: 184.1338. Found: 184.1337.
4
5
6
7
8
9
Ethyl 3-azabicyclo[3.3.1]non-7-ene-3 carboxylate (10).
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
To an inerted reaction vessel was charged paraformaldehyde (18.9 kg, 630 mol) and
dichloromethane (399.0 L). Stirring was initiated and 8 (104 kg, 567.5 mol) was charged to the
reactor at 25°C. Boron trifluoride etherate (32.8 kg, 231.0 mol) was charged over 6 h at 25ꢀ26°C
and held for 30 min at 25°C. Upon completion of reaction (< 1% of 8 by GC), the suspension
was filtered over a pressure filter. The reactor was rinsed with dichloromethane (43.0 L). The
wet cake (1.5 kg) was discarded. A 7.9 % (w/w) solution of NaHCO3 (308.6 kg) was charged at
20ꢀ23°C over 2 h. The reactor contents were stirred at 23°C for 5 h. More NaHCO3 solution (4
kg) was charged to bring the pH up to 7. Stirring was stopped and the phases were allowed to
separate over 15 min. The top aqueous layer was removed (285.0 L). Water (285.0 L) was
charged and the mixture was stirred at 20ꢀ22°C for 20 min. Stirring was stopped and the phases
were allowed to separate over 15 min. The lower organic layer (477.2 L) was separated from the
aqueous layer and charged to a different reactor. The organic layer was dried over MgSO4 (17.9
kg) with stirring for 1 hour at 20°C. The MgSO4 was removed via filtration, and the filter cake
was washed with dichloromethane (27 L). The organic solutions were combined and charged to
a different 630 L reactor. The solvent was removed under reduced pressure (832→18 mbar) at
35ꢀ40°C over 12 h, removing 470 L of solvent. The solution was further concentrated under
vacuum (18→9 mbar) at 40°C over 30 min, removing a total of 5.0 L of solvent. Crude 10
(108.5 kg) was discharged and transferred to a thinꢀfilm evaporator. Olefin 10 was purified by
vacuum distillation at 108ꢀ120°C (jacket temperature), at 0.1 mbar, at a feed rate of 2.5ꢀ3.5 L/h.
1
The yield was 69.0 kg (62.3%) of 10 as a clear, colorless liquid. H NMR (400 MHz, CDCl3)
5.60ꢀ5.45 (m, 2H), 4.17ꢀ3.75 (m, 4H), 2.97ꢀ2.63 (m, 2H), 2.24ꢀ2.07 (br s, 2H), 2.05ꢀ1.82 (m,
2H), 1.63 (dd, J=12.2, 36 Hz, 2H), 1.10 (t, J=6.5 Hz, 3H). 13C NMR (400 MHz, CDCl3)
(complicated by presence of rotamers; individual rotamers not assigned) 159.7, 158.8, 131.9,
131.2, 130.4, 129.95, 63.4, 53.8, 53.6, 49.7, 33.6, 32.6, 32.5, 17.2. HRMS (ES) m/z Calcd. for
C11H17NO2+H: 196.1338. Found: 196.1345.
Ethyl 6,7-dibromo-3-azabicyclo[3.3.1]nonane-3 carboxylate (11).
To an inerted reactor vessel was charged 10 (7.0 kg, 35.8 mol) and dichloromethane (49 L).
Stirring was initiated and the solution was cooled to 10ꢀ15°C. A 42% (w/w) solution of bromine
(6.07 kg, 38.0 mol) in dichloromethane (11.9 kg) was charged over 30 min maintaining the
temperature around 15°C. The reaction mixture was warmed to 20°C and was held for 2 h at
20°C. To the reaction mixture was charged a 9% (w/w) solution of ascorbic acid (31.0 kg) over
5 min keeping the temperature below 25°C. The reaction mixture was held for 10 min before
stirring was stopped and the layers were allowed to phase separate over 10 min. The lower
organic layer was transferred to different reactor, and the upper aqueous layer was discarded.
Water (28.0 L) was charged to the reactor containing the organic layer. Stirring was initiated and
the phases were mixed for 10 min. Stirring was stopped and the layers allowed to phase separate
over 10 min. The upper aqueous phase was discarded. The lower organic layer was drummed
ACS Paragon Plus Environment
14

Page 15 of 23
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
and then transferred portionꢀwise to a rotary evaporator for concentration to dryness under
reduced pressure (630→400 mbar) at 30ꢀ35°C over 13 h. THF (139.9 L) was charged to the
reactor, and the reactor contents were stirred for 10 min. The contents were drummed and then
transferred to a rotary evaporator where the solution was concentrated under reduced pressure
(116 mbar) at 30ꢀ35°C over 5 h until the distillate flow ceased. The yield of 11 was 11.84 kg
(92.7%) as an amber oil. The purity by HPLC was 88.2% for the combined areas of both
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
rotamers. H NMR (400 MHz, CDCl3) 4.81ꢀ4.70 (m, 1H), 4.70ꢀ4.68 (br s, 1H), 4.20ꢀ4.12 (q,
J=17.0 Hz, 2H), 4.12ꢀ3.95 (br s, 2H), 3.18ꢀ3.15 (d, J=11 Hz,1H), 3.09ꢀ3.06 (d, J=11 Hz, 1H),
2.54ꢀ2.51 (m, 2H), 2.44ꢀ2.37 (dt, J=3 Hz, 17 Hz, 1H), 2.16ꢀ2.12 (d, J=13 Hz, 1H), 2.02 (s, 1H),
1.57ꢀ1.55 (m, 1H), 1.30ꢀ1.26 (t, J=7 Hz, 3H). 13C NMR (400 MHz, CDCl3) 155.4, 62.8, 62.0,
49.5, 48.5, 47.2, 39.2, 39.0, 31.0, 26.5, 15.3.
Ethyl 7-bromo-3-azabicyclo[3.3.1]non-7-ene-3-carboxylate (12).
In a reaction vessel equipped with temperature probe and reflux condenser, a solution of
potassium tertꢀbutoxide (9.32 kg, 83.05 mol) in anhydrous THF (70.8 L) was cooled to 3°C. In
another reaction vessel, a solution of 11 (11.84 kg, 33.34 mol, 88.2% purity) in anhydrous THF
(23.6 L) was prepared and added to the potassium tertꢀbutoxide solution over a minimum of 60
min, while maintaining the temperature at 5°C. Following addition, the suspension was heated at
40°C for a total of 4 h. An IPC by HPLC indicated complete consumption of 11. The reaction
mixture was cooled to 20°C and was quenched into 10% NH4Cl solution (35.4 L). Following
phase separation, the organic layer was concentrated under vacuum at 28ꢀ29°C to approximately
half volume, removing 45 L of THF. The aqueous layer was extracted with MTBE (47 L), the
layers separated and the organic layer was combined with the residue from the concentrated THF
layer. The MTBEꢀTHF solution was dried over MgSO4 (5.9 kg) with stirring for 45 min. The
mixture was filtered under vacuum via pressure filter fitted with a polypropylene filter disk, and
the filter cake was washed with MTBE (5.7 L). The filtrate was concentrated to dryness under
vacuum on a rotary evaporator at 29°C, collecting 77 L of distillate. The residue was then
dissolved in 1,4ꢀdioxane (16.6 L) and filtered via pressure filter, fitted with Whatman Grade 1
filter paper to remove residual inorganic material. The filter bed was washed with 1,4ꢀdioxane
(4.6 L), and the combined organic filtrate was concentrated to dryness under vacuum (174 mbar)
on a rotary evaporator at 35°C. The raw yield of 12 was 8.88 kg with an HPLC purity of 36.6%.
Based on the purity of the starting material and product, the corrected molar yield was 40.3%.
1H NMR (400 MHz, CDCl3) 6.02 (dd, J=5.8 Hz, 15.6 Hz, 1H), 4.32ꢀ3.87 (m, 4H), 3.01ꢀ2.67 (m,
3H), 2.42 (d, J=18 Hz, 2H), 2.03 (br d, J=12.6 Hz, 1H), 1.75 (dd, J=12.6 Hz, 36 Hz, 2H), 1.26
(m, 3H). 13C NMR (400 MHz, CDCl3) (complicated by presence of rotamers; individual
rotamers not assigned) 159.4, 159.1, 132.2, 131.9, 126.5, 125.7, 63.9, 53.5, 53.3, 49.6, 49.4,
43.3, 42.9, 35.3, 35.2, 32.6, 32.5, 31.9, 17.5. HRMS (ES) m/z Calcd. for C11H16NO2Br+H:
274.0443. Found: 274.0444.
Ethyl 7-(pyridin-3-yl)-3-azabicyclo[3.3.1]non-7-ene-3-carboxylate (13).
ACS Paragon Plus Environment
15

Organic Process Research & Development
Page 16 of 23
1
2
3
4
5
6
7
8
9
Water (11.92 L) and sodium carbonate (3.19 kg, 30.1 mol) were charged to a reaction vessel
equipped with a temperature probe and reflux condenser. The mixture was stirred and 3ꢀ
pyridinylboronic acid (1.319 kg, 10.73 mol) and 1,4ꢀdioxane (17.67 L) were then added. A
solution of crude 12 (7.97 kg, containing 2.95 kg actual 12, 10.75 mol) in 1,4ꢀdioxane (2.95 L)
was prepared and was added to the reactor, rinsing the transfer lines with additional 1,4ꢀdioxane
(2.99 L). Bis(diphenylphosphino)ferrocene (dppf) (0.291 kg, 0.525 mol) was then added. By
using a nitrogen dip tube, the reactor contents were sparged with nitrogen for 30 min while
stirring. Palladium(II) acetate (0.116 kg, 0.52 mol) was then added, and under a nitrogen
atmosphere, the reaction mixture was heated under reflux (80°C) for 3 h and then cool to ≤ 50°C.
An IPC by 1H NMR showed complete consumption of 12. The reaction mixture was then cooled
to ≤ 25°C and was filtered through an inꢀline filter bag, washing with 1,4ꢀdioxane (5.93 L).
Because of equipment limitations, the workꢀup was done in two additional 50 L vessels; the
reaction mixture was divided into two equal portions for the workꢀup. Each half was processed
as follows: Most of the organic solvents were removed by vacuum distillation at 35 °C,
collecting ~15 L of distillate and leaving the product residue in water. Water (7.39 L) and 2ꢀ
MeTHF (16.27 L) were added, and the mixture was stirred for 10 min. Following phase
separation, the lower aqueous layer was separated. Water (6.40 L) and concentrated HCl
solution (1.37 kg) were then added to the remaining organic layer, and the mixture was stirred
for a minimum of 10 min. Following phase separation, the lower aqueous layer was separated
and 2ꢀMeTHF (7.26 L) and water (1.47 L), were added as a line rinse. The mixture was basified
by the addition of a solution of NaOH (0.784 kg) in water (1.5 L), keeping the temperature ≤
30°C. After stirring for a minimum of 5 min, and allowing for phase separation, the upper
organic phase was separated from the lower aqueous phase (pH 13.27). The organic phase was
dried over MgSO4 (1.18 kg) for 30 min and filtered under vacuum via bag filtration. The filter
cake was washed with 2ꢀMeTHF (3.05 L) and added to the filtrate. Once the workꢀup was
completed, the product rich organic layers from each half were combined and concentrated to
dryness under vacuum on a rotary evaporator at a temperature of ≤ 35°C. The yield of 13 was
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
1.965 kg (67.1%) as a brown, viscous oil. The purity by HPLC was 80.5%. H NMR (400 MHz,
CDCl3) 8.60 (s, 1H), 8.45 (s, 1H), 7.63 (s, 1H), 7.22 (m, 1H), 6.11 (d, J=19.6 Hz, 1H), 4.36 (d,
J=13.3 Hz, 0.5 H), 4.18 (d, J=12.9 Hz, 0.5 H), 4.08 (d, J=12.9 Hz, 0.5 H), 3.97 (m, 2.5 H), 3.09ꢀ
2.89 (m, 2H), 2.78ꢀ2.40 (m, 3H), 2.19 (br s, 1H), 1.86ꢀ1.82 (m, 2H), 1.07 (m, 3H). 13C NMR
(400 MHz, CDCl3) (complicated by presence of rotamers; individual rotamers not assigned)
159.8, 159.1, 150.7, 149.5, 149.3, 139.7, 139.0, 138, 135.2, 135.1, 129.7, 129.2, 125.7, 125.6,
63.7, 53.9, 53.8, 50.3, 50.1, 35.5, 35.3, 33.4, 33.3, 32.4, 30.6, 30.5, 17.3. HRMS (ES) m/z Calcd.
for C16H20N2O2+H: 273.1603. Found: 273.1611.
7-(pyridin-3-yl)-3-azabicyclo[3.3.1]non-6-ene (15).
Water (3.9 L), concentrated HCl (21.45 L), and 13 (3.90 kg, 14.32 mol) were charged to a vessel
equipped with a temperature probe and reflux condenser. The mixture was heated at ~100°C for
a total of 30 h until the IPC by HPLC showed < 1% of 13 remaining. The mixture was cooled to
25°C. In a separate vessel, a 33% solution of NaOH was prepared by dissolving NaOH (10.04
kg) in water (19.97 L), keeping the internal temperature ≤ 45°C. Because of equipment
limitations, the workꢀup was done in two vessels; the reaction mixture was divided into two
equal portions for the workꢀup. Each half was processed as follows: The reaction mixture was
brought to pH 7.0 with 33% NaOH solution (12.08 kg), keeping the internal temperature < 30°C.
ACS Paragon Plus Environment
16

Page 17 of 23
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
After stirring for 10 min, the aqueous solution was washed with toluene (2 x 7.8 L), stirring each
extraction for 15 min. Following the 2nd toluene wash, the lower productꢀrich aqueous layer was
separated. Toluene (7.8 L) was added to this aqueous layer, and the mixture was basified with
33% NaOH solution (1.40 kg) to pH 13.1, keeping the internal temperature < 30°C. After
stirring for 15 min and allowing for phase separation, the lower aqueous layer was separated and
extracted with toluene (7.8 L). The productꢀrich toluene extracts were combined. Once the
workꢀup was completed, the product rich organic layers from each half were combined and
azeotropically dried by vacuum distillation (76 mbar) at 27→24°C, removing 8 L of distillate.
At 24°C, MgSO4 (1.99 kg) and thiolized silica gel (Silicycle Cat. No. R51030B) (0.4 kg) were
added. This was done to remove traces of palladium (725 ppm by ICPꢀMS), residual water, and
black, insoluble, tarry material. The mixture was stirred for 30 min before filtering using a 1 ꢁm
filter bag, washing the filter cake with toluene (4 L). The filtrate was concentrated under
reduced pressure (43 mbar) at 52→27°C until a volume of 7.4 L remained. Heptane (19.91 L)
was added. The resulting slurry was concentrated under reduced pressure (51 mbar) at 30°C
until a volume of about 13.2 L remained. Heptane (11.7 L) was added, and the reactor contents
were cooled to 24°C. The product was collected by vacuum filtration using a bag filter. The
solids were washed with heptane (7.81 L) and dried under vacuum on a rotary evaporator at
35°C for 19 h. The yield of 13 was 1.328 kg (46.3%) as a brown solid. The HPLC purity was
99.8%. Based on the purity of the starting material and product, the corrected molar yield was
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
57.4%. The palladium content was 46 ppm as determined by ICPꢀMS. H NMR (400 MHz,
CDCl3) 8.71 (d, J=2 Hz, 1H), 8.54 (m, 2H), 7.90 (dd, J=5.8 Hz, 8.2 Hz, 1H), 6.49 (d, J=6.3 Hz,
1H), 3.27 (d, J=13 Hz, 1H), 3.19ꢀ3.11 (m, 3H), 2.9ꢀ2.81 (m, 2H), 2.48ꢀ2.43 (m, 2H), 1.79 (dd,
J=13.3 Hz, 26 Hz, 2H). HRMS (ES) m/z Calcd. for C13H16N2+H: 201.1392. Found: 201.1394.
3-Methyl-7-(pyridin-3-yl)-3-azabicyclo[3.3.1]non-6-ene (5).
Thiolized silica gel (75 g) was added to a 29% (w/w) solution of 15 in isopropanol (1.725 kg of
solution, 0.5 kg, 2.5 mol). This treatment was done to reduce the level of palladium. The
suspension was heated to 40°C for 2 h, cooled to 20°C and then stirred for 16 h. The solids were
filtered through a pressure filter fitted with Whatman Grade 1 filter paper, washing the filter cake
with isopropanol (1 L). The filtrate was diluted with heptane (1.8 L) and the resulting solution
was concentrated under reduced pressure (180 mbar) at 45°C until about 2.5 L of the solvent was
removed. The isopropanol was then solvent swapped with three separate charges of heptane (1.8
L, 2.5 L, 2.9 L), with vacuum distillation (161ꢀ169 mbar) to remove 50% of the solvent
following each of three heptane charges. Precipitation was observed, and the flask contents were
then vacuum distilled to dryness at 45°C (~50 mbar), leaving 0.465 kg (2.32 mol) of material in
the flask. The material was treated with formaldehyde (1.27 kg of 37% aqueous solution, 0.47
kg, 15.7 mol) and formic acid (1.42 kg, 30.7 mol), keeping the temperature < 40°C. The solution
was stirred and heated under reflux for 2 h and then cooled to 45°C. An IPC by HPLC indicated
the absence of 15. The reaction mixture was cooled to 20°C. Isopropyl acetate (2.3 L) was
added followed by 10 M NaOH solution (4.36 kg) to bring the pH to 13ꢀ14, keeping the internal
temperature < 40°C. The phases were separated and the lower aqueous phase was extracted with
isopropyl acetate (0.9 L). The combined isopropyl acetate extracts were washed with 2 M NaOH
solution (1 kg), followed by a water (0.9 L) wash. The isopropyl acetate phase was separated
and dried over MgSO4 (0.47 kg). The resulting suspension was filtered through a pressure filter
fitted with Whatman Grade 1 filter paper, washing the filter cake with isopropyl acetate (0.6 L).
ACS Paragon Plus Environment
17

Organic Process Research & Development
Page 18 of 23
1
2
3
4
5
The concentration of 5 in this isopropyl acetate solution was determined to be 13.48% (w/w).
The yield of 5 was 0.456 kg (85.3%). H NMR (400 MHz, CDCl3) 8.72ꢀ8.71 (d, J=3.0 Hz, 1H),
1
8.43ꢀ8.41 (dd, J=3 Hz, J=15 Hz, 1H), 7.73ꢀ7.70 (dt, J=8 Hz, J=3 Hz, 1H), 7.21ꢀ7.18 (dd, J =7 Hz,
J=4 Hz, 1H), 6.41ꢀ6.40 (dd, J=7 Hz, J=1Hz, 1H), 2.86ꢀ2.83 (d, J=11 Hz, 1H), 2.79ꢀ2.76 (dt, J=10
Hz, J=1Hz, 1H), 2.70ꢀ2.64 rotamers (d, J=8 Hz, 1H), 2.50ꢀ2.46 (m, 2H) 2.24ꢀ2.20 (m, 1H), 2.18
(s, 3H), 2.17ꢀ2.13 (dt, J=13Hz, J=3 Hz, 1H), 2.04ꢀ2.02 (dd, J=11 Hz, J=3 Hz, 1H), 1.74ꢀ1.69
rotamers (m, 1H), 2.04ꢀ2.02 (dd, J=13 Hz, J=3 Hz, 1H). 13C NMR (400 MHz, CDCl3) 147.2,
146.4, 136.8, 135.6, 132.2, 129.1, 123.2, 63.7, 59.0, 47.0, 33.5, 31.1, 29.8, .28.5.
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
(1R,5S)-7-(pyridin-3-yl)-3-azabicyclo[3.3.1]non-6-ene ((+)-15).
Eight positions of a Simulated Moving Bed (SMB) chromatograph were equipped with Supelco
Chiralpak AD 20 ꢂm columns. A solution of isopropanol/diethylamine (2.92 kg, 1000:1 (v/v))
was added to 15 (1.30 kg) and the solids were dissolved. The resulting solution was transferred
to a flask, followed by an isopropanol/diethylamine (0.14 kg, 1000:1) rinse. The solution was
stirred for 3 h 14 min and was applied to the SMB system, eluting with isopropanol/diethylamine
(1000:1). During the separation, an IPC by chiral HPLC was done on the extracts. Eight
fractions of raffinate with a total volume of 267 L were collected. The raffinate fractions were
combined and were concentrated to dryness under vacuum (338→20 mbar) at 45ꢀ49°C. The
concentrate was further dried under vacuum (14ꢀ19 mbar) at 49°C for 2 h, followed by further
drying at 49→17°C (heating switched off) for 14 h. Isopropanol (2.27 L) was added to the
concentrate via cartridge filter, and the solids were dissolved at 29 °C. The product solution was
filtered over a POR3 Nutsche filter, followed by a rinse with isopropanol (0.14 L). The resulting
clear, dark brown solution (2.61 kg) was stirred until homogeneous. A titration analysis
indicated a 23.1% (w/w) solution containing 0.61 kg (46.9%) of (+)-15. The achiral purity was
99.3%, and the chiral purity was 99.9%. The palladium content was < 10 ppm as determined by
1
ICPꢀOES. Analytical data were obtained on a sample of the diꢀHCl salt. H NMR (400 MHz,
CD3OD) (br s, 1H), 8.78 (s, 1H), 8.76 (s, 1H), 8.10 (br s, 1H), 6.65 (d, J=4.2 Hz, 1H), 3.43 (d,
J=12.9 Hz, 1H), 3.35ꢀ3.25 (m, 3H), 3.01 (dd, J=6.3 Hz, 18 Hz, 1H), 2.93 (s, 1H), 2.69 (d, J=18
Hz, 1H), 2.58 (s, 1H), 2.05 (d, J=13 Hz, 1H), 1.94 (d, J=13.3 Hz, 1H). 13C NMR (400 MHz,
D2O) 142.8, 139.5, 139.3, 137.8, 135.2, 129.9, 126.9, 49.9, 45.7, 31.0, 27.3, 26.1, 24.8. HRMS
(ES) m/z Calcd. for C13H16N2+H: 201.1392. Found: 201.1389. [α]D20 = +36.70° (c= 1.0,
MeOH).
(1R,5S)-3-methyl-7-(pyridin-3-yl)-3-azabicyclo[3.3.1]non-6-ene ((+)-5) (TC-8817).
Thiolized silica gel (62.8 g) was added to a 23.1% (w/w) solution of (+)-15 in isopropanol (2.60
kg of solution, 0.60 kg, 2.99 mol). This treatment was done as a precautionary measure to
further reduce the level of palladium. The suspension was heated to 42°C for 2 h, cooled and
then stirred at 22°C for 14 h. The solids were filtered through a pressure filter fitted with
Whatman Grade 1 filter paper, washing the filter cake with isopropanol (0.35 L). The filtrate
was concentrated under reduced pressure (180 mbar) at 45 °C until about 50% of the solvent was
removed. The isopropanol was then solvent swapped with three separate charges of heptane (3 x
3 L), with vacuum distillation at 161ꢀ169 mbar to remove 50% of the solvent following each
heptane charge. The flask contents were then vacuum distilled to dryness at 52 mbar, leaving
0.588 kg of material in the flask. Formaldehyde (1.636 kg of 37% aqueous solution, 0.61 kg,
ACS Paragon Plus Environment
18

Page 19 of 23
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
20.2 mol) and formic acid (1.835 kg, 39.9 mol) were added, and the solution was stirred and
heated at reflux (85ꢀ87 °C) for 2 h and then cooled to 41 °C. An IPC by HPLC indicated < 0.5%
of (+)-15 remaining. The reaction mixture was cooled to room temperature. Isopropyl acetate (3
L) was added followed by 10 M NaOH solution (4.46 kg), keeping the internal temperature < 40
°C. Additional 10 M NaOH solution (1.0 kg) was added to bring the pH to 13. The phases were
separated and the lower aqueous phase was extracted with isopropyl acetate (0.87 L). The
combined isopropyl acetate extracts were washed with 2 M NaOH solution (1.60 kg), followed
by a water (1.5 L) wash. The isopropyl acetate phase was separated and dried over MgSO4 (0.63
kg). After stirring for 30 min, the solids were filtered through a pressure filter fitted with
Whatman Grade 1 filter paper, washing the filter cake with isopropyl acetate (0.6 L). The
concentration of (+)-5 in this isopropyl acetate solution (4.109 kg) was determined to be 14.04%
(w/w). The yield of (+)-5 was 0.577 kg (89.9%), and the purity by HPLC was 99.0%. The
palladium content was < 10 ppm in the solid as determined by ICPꢀOES. Analytical data were
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
obtained on a sample of the monoꢀfumarate salt. H NMR (400 MHz, CD3OD) (br s, 1H), 8.46
(br s, 1H) 7.95 (d, J=7.8 Hz, 1H), 7.43 (br s, 1H), 6.60 (s, 2H), 6.43 (d, J=6.3 Hz, 1H), 3.55 (d,
J=12.5 Hz, 1H), 3.46 (d, J=12 Hz, 1H), 3.22 (dd, J=3Hz, 12.5 Hz, 1H), 3.17 (dd, J=3.6 Hz, 12.1
Hz, 1H), 2.94 (dd, J=12.5 Hz, 19.5 Hz, 1H), 2.89 (br s, 1H), 2.81 (s, 3H), 2.67 (d, J=19.4 Hz,
1H), 2.57 (br s, 1H), 1.97ꢀ1.89 (m, 2H). HRMS (ES) m/z Calcd. for C14H18N2+H: 215.1548.
Found: 215.1544. [α]D20 = +22.27° (c= 1.0, MeOH, HCl salt).
(1R,5S)-3-Methyl-7-(pyridin-3-yl)-3-azabicyclo[3.3.1]non-6-ene p-Hydroxybenzoate ((+)-5
p-Hydroxybenzoate) (TC-8817 p-Hydroxybenzoate).
A 14.04% w/w solution of (+)-5 in isopropyl acetate (4.095 kg of solution, 0.575 kg, 2.68 mol)
was filtered through GF/F paper into reactor. The inꢀline filter was washed with isopropyl
acetate (0.27 L) and this wash was added to the filtrate. This solution was then heated under
reflux (~88°C). A solution of pꢀhydroxybenzoic acid (0.37 kg, 2.68 mol) in isopropanol (1.58 L)
was prepared at 40°C ±5°C. The pꢀhydroxybenzoic acid solution was then cooled to 20°C ±5°C
and filtered through GF/F paper, rinsing the inline filter with isopropanol (0.16 L). With
vigorous stirring, the filtered pꢀhydroxybenzoic acid solution was then added to the hot solution
of (+)ꢀ5 in isopropyl acetate over a minimum of 5 min. The solution was maintained under
reflux for 5ꢀ10 min before cooling to 0°C ± 5°C over a period of 3.5 h. Stirring was continued at
0°C ±5°C for 2.2 h. The solids were filtered under vacuum through Whatman Grade 1 filter
paper and washed with a solution of isopropyl acetateꢀisopropanol (2.5:1 (v/v), 0.84 L). The
solids were dried under vacuum for 30 min and then further dried under rotary evaporatoration to
constant weight at 40°C ±5°C. The yield was 0.813 kg (86.0%) of a white, crystalline solid. The
achiral HPLC purity was 100.5%. The chiral HPLC purity was determined to be 99.83% ((+)ꢀ5
1
retention time: 16.4 minutes, (ꢀ)ꢀ5 retention time: 17.7 minutes). H NMR (400 MHz, CDCl3) δ
8.44ꢀ8.43 (d, J=1.5 Hz, 1H), 8.27ꢀ8.25 (dd, J=4 Hz, J=1 Hz, 1H), 7.76ꢀ7.73 (dt, J=9 Hz, J = 2Hz,
1H), 7.63ꢀ7.60 (d, J =9 Hz, 2H), 7.28ꢀ7.25 (m, 1H), 6.75ꢀ6.71 (d, J=8 Hz, 2H), 6.23ꢀ6.11 (d, J= 6
Hz, 1H), 3.38ꢀ3.35 (dd, J=13 Hz, J=1Hz, 1H), 3.29ꢀ3.26 (d, J=12 Hz, 1H) 3.09ꢀ3.05 (dd, J=16
Hz, J=1Hz, 1H), 3.03ꢀ3.00 (dd, J=16 Hz, J=3Hz, 1H) 2.76ꢀ2.70 (m, 2H), 2.67 (s, 3H), 2.41ꢀ2.37
(br s, 2H), 1.73ꢀ1.71 (m, 2H). 13C NMR (400 MHz, CDCl3) δ 175.0, 158.3, 147.2, 145.0, 137.5,
135.6, 134.0, 131.1, 128.1, 125.4, 124.0, 114.7, 60.8, 56.7, 44.0, 31.2, 28.4, 26.4, 26.0. Calcd.
ACS Paragon Plus Environment
19

Organic Process Research & Development
Page 20 of 23
1
2
3
4
5
for C21H24N2O3: C, 71.57; H, 6.86; N, 7.95. Found: C, 71.43, 71.45; H, 6.74, 6.77; N, 7.88,
7.88.
6
7
8
ThirdꢀGeneration Asymmetric Route
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
(S)-2,2,2-Trifluoroethyl cyclohex-3-enecarboxylate (16).
Corey’s catalyst ((R)ꢀ(+)ꢀoꢀTolylꢀCBSꢀoxazaborolidine trifluromethanesulfonimide [(R)ꢀ(+)ꢀ3,3ꢀ
diphenylꢀ1ꢀoꢀtolylꢀtetrahydropyrrolo(1,2ꢀc)(1,3,2)oxazaboroleꢀ1,1,1ꢀtrifluoroꢀNꢀ
((trifluoromethyl)sulfonyl)methanesulfonamide], (26.0 g, 0.0438 mol), prepared according to
known procedures,10,11 and CH2Cl2 (70 mL) were charged to a pressure vessel. The mixture was
cooled to ꢀ78°C and 1,3ꢀbutadiene gas was passed in through a long needle until the required
quantity was condensed (58 g, 1.07 mol). 2,2,2ꢀTrifluoroethyl acrylate (67.4 g, 0.438 mol) was
added, and the pressure vessel was sealed with a screw cap. The resulting mixture was stirred at
room temperature for 2 days. TLC and GCꢀMS indicated that the reaction was complete. The
reaction was quenched by addition of triethylamine (6 mL) and the solvent was evaporated. The
crude product was purified by silica gel chromatography, eluting with 5% ether in hexanes.
Concentration of selected fractions afforded 66.5 g (74.0%) of 16 as a colorless oil. The
enantiomeric purity was determined using chiral GC (gammaꢀTA column, 60oC). The chiral
purity was 98.5% for the (S)ꢀenantiomer (RT 29.385 min); the (R)ꢀenantiomer (RT 30.900 min)
1
was present at a level of 1.5%. H NMR (400 MHz, CDCl3) δ 5.75ꢀ5.65 (m, 2H), 4.56ꢀ4.42 (m,
2H), 2.72ꢀ2.65 (m, 1H), 2.37ꢀ2.28 (m, 2H), 2.16ꢀ2.06 (m, 2H), 2.05ꢀ2.01 (m, 1H), 1.79ꢀ1.69 (m,
1H). 13C NMR (400 MHz, CDCl3) δ 176.7, 129.3, 127.3, 125.5 (q, J=277 Hz), 62.8 (q, J=36
Hz), 41.5, 29.8, 27.4, 26.8.
(S)-Cyclohex-3-enecarboxamide (17).
To a heavyꢀwalled pressure reactor equipped with a stir bar, stirrerꢀhotplate, and heating mantle
was charged 16 (20.0 g, 0.096 mmol), THF (120 mL) and concentrated ammonium hydroxide
(60 mL, 462 mmol). Stirring was initiated and the reactor was sealed. The turbid biꢀphasic
mixture was warmed to give a clear solution, which was held for 4 h before sampling for COR
(TLC; hexanes : EtOAc (1:1) and 1H NMR). TLC indicated the absence of the high Rf starting
material spot. Examination of the reaction mixture by 1H NMR also indicated the absence of
starting material. The reaction mixture was cooled to room temperature and CHCl3 (100 mL)
was added. After mixing for 10 min, the layers were separated. The lower organic phase was
collected and the aqueous layer was extracted with CHCl3 (75 mL). After mixing for 10 min, the
lower organic phase was collected. The combined CHCl3 layers were dried over Na2SO4 (10 g)
and filtered. The filtrate was concentrated under reduced pressure at 45ꢀ50°C to give 10.0 g
1
(83.1%) of 17 as a white solid. H NMR (400 MHz, CDCl3) δ 6.15 (br s, 1H), 5.70 (br s, 2H),
2.47ꢀ2.39 (m, 1H), 2.25 (m, 2H), 2.17ꢀ2.05 (m, 2H), 1.98ꢀ1.95 (m, 1H), 1.75ꢀ1.65 (m, 1H). 13C
NMR (400 MHz, CDCl3) δ 181.4, 129.5, 127.9, 43.3, 30.7, 28.3, 27.2. Mp: 153.5ꢀ154.1 °C.
[α]D20 = ꢀ69° (c= 1.0, CHCl3). HRMS (ES) m/z Calcd. for C7H11NO+H: 126.0919. Found:
126.0922.
ACS Paragon Plus Environment
20

Page 21 of 23
Organic Process Research & Development
1
2
3
4
Ethyl cyclohex-3-enylmethylcarbamate ((-)-8).
5
6
7
8
To an inerted reactor equipped with a stir bar, stirrerꢀhotplate, thermocouple, rubber septum, and
nitrogen inlet adapter was charged 17 (26.1 g, 0.209 mol) and 2ꢀMeTHF (312 mL). Stirring was
initiated and a 2.0 M solution of lithium aluminum hydride (156 mL, 0.312 mol) was added
dropꢀwise over 45 min while keeping the temperature below 45°C. The reaction mixture was
held for 90 min at 45°C. Upon completion of reaction (<1% of 17 by GCꢀMS), the reaction
mixture was cooled to 5°C. Water (9.0 mL) was charged dropꢀwise over 10 min, followed by
addition of a 10% NaOH solution (9.0 mL) over 10 min, then water (20 mL) over 10 min.
During the quench, the temperature was held below 40°C. The suspension was stirred for 30
min at 25°C before filtering the mixture. The wet cake was washed with 2ꢀMeTHF (2 x 80 mL)
and combined with the filtrate. The resulting solution of (1S)ꢀcyclohexꢀ3ꢀenylmethanamine (ꢀ)ꢀ7
(510 mL) was used directly in the next step without further purification.
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
To an inerted reactor equipped with a stir bar, stirrerꢀhotplate, thermocouple, rubber septum, and
nitrogen inlet adapter, was charged a 13.1% (w/w) NaHCO3 solution (240 g of solution, 31.6 g,
0.37 mol) and 2ꢀMeTHF (100 mL). Stirring was initiated and ethyl chloroformate (39 mL, 0.408
mmol) was charged. A solution of the above amine in 2ꢀMeTHF (510 mL) was slowly added
over 2 h keeping the temperature at 20ꢀ25°C. The reaction mixture was held at 25°C for 45 min.
Upon completion of reaction (< 1% amine remaining by GCꢀMS), stirring was stopped and the
layers were allowed to separate over 10 min. The lower aqueous layer was discarded, and the
upper organic layer was added back to the reactor along with a 10% (w/w) NaOH solution (200
mL). The mixture was stirred for 30 min. Stirring was stopped and the layers were allowed to
separate over 10 min. The lower aqueous layer was discarded, and the upper organic layer was
added back to the reactor along with a saturated NH4Cl solution (250 mL). The mixture was
stirred for 30 min. Stirring was stopped and the phases were allowed to separate over 10 min.
The lower aqueous layer was discarded, and the upper organic layer was dried over sodium
sulfate (10 g) for 2 h. The drying agent was filtered, and the filtrate was concentrated under
reduced pressure at 42ꢀ46°C to afford 33.25 g (87.0%) of (-)-8 as a pale yellow oil. [α]D20 = ꢀ54°
(c= 1.0, CHCl3). Spectral characteristics were identical to racemic 8.
(1R,5S)-3-methyl-7-(pyridin-3-yl)-3-azabicyclo[3.3.1]non-6-ene ((+)-5).
To a flask equipped with a stir bar, stirrer/hotplate, septum, thermocouple, and nitrogen inlet line
was charged (1R,5S)ꢀethyl 7ꢀ(pyridinꢀ3ꢀyl)ꢀ3ꢀazabicyclo[3.3.1]nonꢀ6ꢀeneꢀ3ꢀcarboxylate 13
(prepared from (-)-8 according to the same procedures reported above for the racemic material)
(0.6 g, 2.20 mmoles) and Tetrahydrofuran (10 mL). Stirring was initiated and the mixture cooled
to 0°C. Dropꢀwise addition of Lithium Aluminum Hydride (1.8 mL of 2M in THF; 3.60
mmoles;) was performed via a syringe and needle. Immediate offꢀgassing was observed and the
solution lightened (deep redꢀblack to a lighter red). After complete addition, the mixture was
warmed to 20ꢀ25°C and held overnight. An aliquot was analyzed for completion of reaction: 1
drop of the reaction mixture added to 1 drop of 10% NaOH and diluted with methanol to 1 mL
and filtered through a 0.45ꢂm filter. LCꢀMS indicated a ratio of product to starting material
of >95:5. (Rt =1.85 min, 215 AMU for product; 2.91 min, 274 AMU for SM). The reaction
mixture was cooled to 0°C and 10% NaOH (0.5 mL) was charged dropꢀwise, followed by water
(1 mL). The resulting suspension was stirred for 15 minutes then filtered, washing solids with
ACS Paragon Plus Environment
21

Organic Process Research & Development
Page 22 of 23
1
2
3
4
5
THF (2 x 5 mL). The filtrates were concentrated to a red, viscous oil (0.67 g), which was
purified by chromatography on 12 g of silica gel, eluting
with Dichloromethane:Methanol:Ammonium hydroxide (8:2:0.2). Desired fractions were
collected and concentrated to a viscous oil (0.42 g) whose spectral data were identical to that of a
(+)-5 reference standard. Analysis for chiral purity by HPLC indicated an 85:15 mixture in
favour of (+)ꢀ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
AUTHOR INFORMATION
Corresponding Author
*Scott R. Breining (srbreining@gmail.com).
Author Contributions
The manuscript was written through contributions of all authors. All authors have given approval
to the final version of the manuscript.
ABBREVIATIONS
IPC, in process control; COR, completion of reaction; TLC, Thin Layer Chromatography; SFC,
Supercritical Fluid Chromatography; SMB, Simulated Moving Bed chromatography.
REFERENCES
1
DeBiasi, M.; Salas, R. Exp. Biol. Med. 2008, 233, 917ꢀ929.
2
Rollema, H.; Coe, J. W.; Chambers, L. K.; Hurst, R. S.; Stahl, S. M.; Williams, K. E. Trends Pharm. Sci. 2007, 28,
316ꢀ325.
3
Mihalak, K. B.; Carroll, F. I.; Luetje, C. W. Mol. Pharmacol. 2006, 70, 801ꢀ805.
4 Bordia, T.; Hrachova, M.; Chin, M.; McIntosh, J. M.; Quik, M. JPET 2012, 342, 327–334.
5
Breining, S. R.; Melvin, M. S.; Bhatti, B. S.; Byrd; G. D.; Kiser; M. N.; Hepler; C. D.; Hooker, D. N.; Zhang, J.;
Reynolds, L. A.; Benson, L. R.; Fedorov, N. B.; Sidach, S. S.; Mitchener, J. P.; Lucero, L. M.; Lukas, R. J.;
Whiteaker, P. and Yohannes, D. J. Med. Chem. 2012, 55(22), 9929–9945.
6
See Ref. 5 and Breining, S. R.; Bhatti, B. S.; Hawkins, G. D.; Miao, L.; Mazurov, A.; Phillips, T. Y.; Miller, C. H.
US 7,402,59.
ACS Paragon Plus Environment
22

Page 23 of 23
Organic Process Research & Development
1
2
3
4
5
7
Similar cyclizations have been previously conducted under different conditions. See, for example: Udding, J. H.;
Papin, N.; Hiemstra, H.; Speckamp, W. N. Tetrahedron, 1994, 50 (29), 8853ꢀ62 and Shishido, K.; Hiroya, K.;
Fukumoto, K.; Kametani, T. J. Chem. Res. 1989, 100ꢀ101.
6
7
8
8
9
Eastwood, P. R. Tetrahedron Lett. 2000, 41, 3705ꢀ3708.
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
9 Pine, S. H.; Sanchez, B. L. J. Org. Chem. 1971, 36 (6), 829ꢀ832.
10 Ryu, D. H.; Corey, E. J. J. Am. Chem. Soc. 2003, 125 (21), 6388ꢀ6390.
11 Liu, D.; Hong, S.; Corey, E. J. J. Am. Chem. Soc. 2006, 128 (25), 8160ꢀ8161 and supporting information.
ACS Paragon Plus Environment
23