The large-scale synthesis of (S)-N-boc-bis(4-fluorophenyl)alanine

Article abstract of DOI:10.1021/op3002855

The synthesis of (S)-N-Boc-bis(4-fluorophenyl)alanine, an intermediate in the synthesis of denagliptin, is described from the synthesis of a 12 g proof of principle sample to a >900 kg cGMP manufacturing campaign. The chiral centre was established by the asymmetric hydrogenation of the sterically crowded precursor, ethyl 2-acetamido-3,3-bis(4-fluorophenyl)acrylate. The ability to isolate the various intermediates in a physical form that would readily allow filtration, washing, and ultimately purification underpinned the successful manufacturing campaign.

Full text of DOI:10.1021/op3002855

Subscriber access provided by FORDHAM UNIVERSITY
Full Paper
The large scale synthesis of (S)-N-Boc bis(4-fluorophenyl)alanine
Robert Appell, Lee Boulton, Edward D Daugs, Matt Hansen, Christopher H. Hanson,
Jerry Heinrich, Christel Kronig, Richard C. Lloyd, David H Louks, Mark Nitz, Celine
F Praquin, James A Ramsden, Helen Samuel, Mark Smit, and Matthew Willets
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/op3002855 • Publication Date (Web): 27 Nov 2012
Downloaded from http://pubs.acs.org on December 16, 2012
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 24
Organic Process Research & Development
1
2
3
4
5
6
The large scale synthesis of (S)ꢀNꢀBoc bis(4ꢀ
7
8
9
fluorophenyl)alanine
10
11
12
13
14
15
16
Robert B. Appell,† Lee T. Boulton,‡ Edward D. Daugs, † Matt Hansen, † Christopher H. Hanson, § Jerry
Heinrich, † Christel Kronig, ‡ Richard C. Lloyd, ‡ David Louks, † Mark Nitz, † Céline Praquin, ‡ James A
Ramsden,* ‡ Helen Samuel, ‡ Mark Smit,† Matthew Willets §
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 Dow Chemical Company, Midland, MI 48674
‡Dr Reddy’s Laboratories Ltd., Chirotech Technology Centre, Unit 410 Cambridge Science Park,
Milton Road Cambridge, CB4 0GH, United Kingdom;
§Dr Reddy’s Laboratories (Eu) Ltd., Steanard Lane, Mirfield, West Yorkshire, WF14 8HZ
Corresponding author jramsden@drreddys.com
Abstract
The synthesis of (S)ꢀNꢀBoc bis(4ꢀfluorophenyl)alanine, an intermediate in the synthesis of Denagliptin,
is described from the synthesis of a 12g proof of principle sample to a >900 Kg cGMP manufacturing
campaign. The chiral centre was established by the asymmetric hydrogenation of the sterically crowded
precursor ethyl 2ꢀacetamidoꢀ3,3ꢀbis(4ꢀfluorophenyl)acrylate. The ability to isolate the various
intermediates in a physical form that would readily allow filtration, washing and ultimately purification
ACS Paragon Plus Environment
1

Organic Process Research & Development
Page 2 of 24
underpinned the successful manufacturing campaign.
1
2
3
4
5
INTRODUCTION
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
(S)ꢀNꢀBocꢀ3,3ꢀbis(4ꢀfluorophenyl)alanine, (S)ꢀ1 is a key raw material in the synthesis of Denagliptin
tosylate 2, a dipeptidyl peptidase IV (DPPIV) inhibitor developed for the treatment of type II diabetes.
The final supply route selected by the innovator was based on the coupling of (4S)ꢀfluoroꢀLꢀproline and
NꢀBocꢀ(S)ꢀdifluorophenyl amino acid 1.1
The synthesis of small, highly substituted chiral molecules such as 1 is an enduring challenge. The
various substituents need to be brought together in an efficient manner while achieving the desired high
stereoselectivity. Within a commercial environment, this must be accomplished with due regard for the
overall process efficiency and cost. Furthermore, all this must be completed within a marketꢀdriven time
frame with the resources available. In developing an asymmetric synthesis of Bocꢀ3,3ꢀbis(4ꢀ
fluorophenyl)alanine, we considered the use of resolution of diastereomeric salts, biocatalysis and
asymmetric hydrogenation. Initial evaluation of these routes quickly led us to focus on asymmetric
hydrogenation as the means of introducing the key chiral centre.
1
Patterson, D. E.; Powers, J. D.; LeBlanc, M.; Sharkey, T.; Boehler, E.; Irdam, E.; Osterhout, M. H.
Org. Process Res. Dev. 2009, 13, 900.
ACS Paragon Plus Environment
2

Page 3 of 24
Organic Process Research & Development
During the initial route selection process it became clear that our chosen route still posed some
1
2
significant challenges. We had a great deal of experience in the synthesis of unnatural amino acids using
asymmetric hydrogenation; typically, this involved an Erlenmeyer condensation and the use of catalysts
such as [(EtDuPhos)Rh(COD)]BF4.2 However, this synthetic approach was not be suitable in this case.
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
RESULTS AND DISCUSSION
A number of hydrogenation substrates were prepared using chemistry outlined in Scheme 1. Reaction
of ethyl isocyanoacetate 4 with 4,4’ꢀdifluorobenzophenone 3 in the presence of base afforded the
required Nꢀformyl dehydroamino ester 5.3
O
OHCHN
CO2Et
O
C
tBuOK, Toluene
N
OEt
F
F
74%
F
F
3
4
5
BocHN
CO2Et
AcHN
CO2Et
1) Boc2O, DMAP, THF
1) Ac2O, Pyridine
5
2) K2CO3, MeOH
73%
2) K2CO3, MeOH
95%
F
F
F
F
7
6
Scheme 1. The synthesis of the Nꢀformyl 5, Nꢀacetyl 6 and NꢀBoc 7 hydrogenation substrates.
2 Cobley, C.J.; Johnson, N.B.; Lennon, I.C.; McCague, R.; Ramsden, J.A.; ZanottiꢀGerosa, A. The
application of DuPHOS Rhodium(I) Catalysts for Commercial Scale Asymmetric Hydrogenation, In
Asymmetric Catalysis on Industrial Scale Eds Blaser and Schmidt. WileyꢀVCH: Weinheim, Germany,
2004. Page 269.
3 Antihypercholesterolemic Tetrazolꢀ1ꢀyl Compounds, Sit, S.ꢀY.; Wright J.J. US patent 4870187, 1988.
ACS Paragon Plus Environment
3

Organic Process Research & Development
Page 4 of 24
The coupling reaction was conducted using potassium tertꢀbutoxide to give ethyl 3,3ꢀbis(4ꢀ
fluorophenyl)ꢀ2ꢀformamidoacrylate 5 in 74% yield. Ethyl 2ꢀacetamidoꢀ3,3ꢀbis(4ꢀfluorophenyl)acrylate
6 and ethyl 2ꢀ((tertꢀbutoxycarbonyl)amino)ꢀ3,3ꢀbis(4ꢀfluorophenyl)acrylate 7 were prepared by forming
the diamide with the appropriate anhydride followed by treatment with methanolic potassium carbonate.
Preliminary attempts at achiral hydrogenation of 5 and 7 using [DiPFc Rh COD]BF4 gave no
conversion.4 However, hydrogenation of 6 with the same catalyst gave racemic ethyl NꢀAcꢀbis(4ꢀ
fluorophenyl)alanine 8. This became the focus of subsequent asymmetric hydrogenation studies vide
infra. Further studies on 5, 6 and later 7 with both [DiPFc Rh COD]BF4 and Pd/C showed that fairly
forcing conditions were required to hydrogenate the double bond. For example, at a molar S/C = 1,000
with [DiPFc Rh COD]BF4 the reaction took several hours at 50 °C. Similarly high temperatures were
required for reductions with 10% Pd/C at a 4% w/w loading.
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
Scheme 2. Synthesis of racemic NꢀBoc and Nꢀacetyl bis(4ꢀfluorophenyl)alanine.
Having demonstrated the nonꢀstereoselective hydrogenation of the Nꢀacetyl derivative with [DiPFc
Rh COD]BF4, we embarked on a screen of asymmetric hydrogenation catalysts. We chose to use similar
4
Smith, M.E.B.; Derrien, N.; Substituted cyclopentenes, their preparation and their use for chiral
scaffolds WO2001017952. Smith, M.E.B.; Derrien, N.; Lloyd, M.C.; Taylor, S.J.C.; Chaplin, D.A.;
McCague, R.; Tetrahedron Letters, 2001, 42, 1347. Sperry, J.; Moody, C.J.; Tetrahedron 2010, 66,
6483.
ACS Paragon Plus Environment
4

Page 5 of 24
Organic Process Research & Development
reaction conditions (10 Bar H2, 50 °C) as we expected the reaction to be very slow at lower pressures
and temperatures. Selected results are shown in Table 1.
1
2
3
4
5
6
Table 1. Results of screen of hydrogenation precatalysts against Nꢀacetyl substrate 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
Preꢀcatalyst
Solv
S/C Press Temp Conv e.e.
(bar)
250 10
(ºC)
50
50
50
50
50
50
50
50
50
50
(%)#
24
(%)
nd
5
1
2
3
4
5
6
7
8
9
[(R,R)ꢀEtꢀDuPhos Rh COD]BF4
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
EtOH
6
[(S,S)ꢀPhꢀBPE Rh COD]BF4
250 10
250 10
250 10
250 10
250 10
250 10
250 10
250 10
250 10
>95
16
91.8
nd
7
[(R,R)ꢀPhꢀBPM Rh COD]BF4
[(R,R)ꢀMeꢀBPE Rh COD]BF4
94
52.8
92.7
nd
8
[(R)ꢀPhanePhos Rh COD]BF4
>95
34
9
[(R,R)ꢀMeꢀ5ꢀFc Rh COD]BF4
[(R,R)ꢀiPrꢀ5ꢀFc Rh COD]BF4
[(R,R)ꢀMeꢀ5ꢀFc Rh COD]BF4
[(R,R)ꢀEtꢀ5ꢀFc Rh COD]BF4
>95
>95
>95
57
50.3
74.2
72.4
55.5
EtOH
EtOH
10
10 [(S,S,R,R)ꢀTangPhos Rh COD]BF4
EtOH
5 Burk, M.J.; Feaster, J.E.; Nugent, W.A.; Harlow, R.L. J. Am. Chem. Soc. 1993, 115, 10125
6 Pilkington, C.J.; ZanottiꢀGerosa, A.; Org. Lett. 2003, 5, 1273.
7 Jackson, M.; Lennon, I.C. Tetrahedron Letters, 2007, 48, 1831.
8
Pye, P.J.; Rossen, K.; Reamer, R.A.; Tsou, N.N.; Volante, R.P.; Reider P.J. J. Am. Chem. Soc. 1997,
119, 6207.
9 Burk, M.J.; Gross, M.F. Tetrahedron Letters, 1994, 35, 9363.
10 Tang, W.; Zhang, X. Angew. Chem., Int. Ed. Eng. I2002,41, 1612 ref
ACS Paragon Plus Environment
5

Organic Process Research & Development
Page 6 of 24
11 [(R)ꢀiPrꢀPhanePhos Rh COD]BF4
12 [(R,R)ꢀEtꢀBPE Rh COD]BF4
13 [(R,R)ꢀMeꢀDuPhos Rh COD]BF4
14 [(S,S)ꢀ2,5ꢀdiMeꢀ1ꢀPhꢀPhospholane]2
[Rh COD]BF4
EtOH
EtOH
EtOH
EtOH
250 10
250 10
250 10
250 10
50
50
50
50
96
91
14
94
24.2
10.3
nd
1
2
3
4
5
6
7
8
52.3
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
#Conversion assessed by analysis of 1H NMR spectrum of the crude material. nd: Not determined.
Quite clearly [PhꢀBPE Rh COD]BF4 and [PhanePhos Rh COD]BF4 (entries 2 and 5) provide both
high conversion and selectivity. Further evaluation of these two catalyst systems was conducted with the
intention of increasing the substrate to catalyst (S/C) ratio and enantiomeric excess. The results are
shown in Table 2. Initial results at S/C = 1,000 showed [PhanePhos Rh COD]BF4 gave slightly higher
activity, particularly in ethanol where 41% conversion was achieved (entry 6). We continued to modify
the process in order to target full conversion at a reasonably low catalyst loading. Unfortunately at this
stage we had to concede that high conversion could only be achieved by increasing the amount of
catalyst used (Entry 10).
Table 2: Preliminary optimization of the hydrogenation of 6.
Preꢀcatalyst
Solv
S/C
Pressure
(bar)
Temp. Conv. e.e.
(ºC)
50
(%)a
13
(%)
1
2
3
[(R,R)ꢀPhꢀBPE Rh COD]BF4
[(R,R)ꢀPhꢀBPE Rh COD]BF4
MeOH
EtOH
1000 10
1000 10
1000 10
96.8
95.3
95.4
50
15
[(R)ꢀPhanePhos Rh COD]BF4 MeOH
50
20
ACS Paragon Plus Environment
6

Page 7 of 24
Organic Process Research & Development
4
[(R)ꢀPhanePhos Rh COD]BF4 EtOH
1000 10
1000 10
1000 10
1000 10
1000 10
1000 20
50
70
70
30
30
50
50
50
50
50
54
10
41
15
43
69
96
86
18
35
94.3
94.5
94.1
96.2
95.6
94.5
94.9
83.9
42.4
65.7
1
2
3
4
5
6
7
8
5
[(R)ꢀPhanePhos Rh COD]BF4 MeOH
[(R)ꢀPhanePhos Rh COD]BF4 EtOH
[(R)ꢀPhanePhos Rh COD]BF4 MeOH
[(R)ꢀPhanePhos Rh COD]BF4 EtOH
[(R)ꢀPhanePhos Rh COD]BF4 EtOH
[(R)ꢀPhanePhos Rh COD]BF4 EtOH
[(S)ꢀPhanePhos Rh COD]BF4 PhMe
6
7
9
8
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
9
10
11
12
13
350
14
1000 10
[(S)ꢀPhanePhos Rh COD]BF4 PhMe/EtOH 1000 10
[(S)ꢀPhanePhos Rh COD]BF4 PhMe/MeOH 1000 10
aConversion assessed by analysis of 1H NMR spectrum of the crude material.
Some competition reactions were conducted in order to try to determine why the asymmetric
hydrogenation reaction required a relatively high catalyst loading even though a significant amount of
conversion was observed after a few minutes. Firstly, the reaction was conducted in which equimolar
amounts of reduced material was added to the reaction to check for product inhibition (Table 3, entry 2),
Table 3: Results of competition studies in the hydrogenation of 6.
Substance A
Substance B
Conv
Sub
Conv
Sub
A
B
(%)#
(%)#
O
AcHN
OEt
1
79
F
F
ACS Paragon Plus Environment
7

Organic Process Research & Development
Page 8 of 24
O
O
O
1
2
3
4
5
6
7
8
AcHN
AcHN
AcHN
OEt
OEt
2
3
4
70
N/A
>95
N/A
F
F
F
F
F
F
O
AcHN
OEt
OMe
9
85
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
O
O
AcHN
AcHN
OEt
OMe
>95
F
F
#Conversion assessed by analysis of 1H NMR spectrum of the crude material.
The reactions were performed at S/C = 500. At this loading, Entry 2 would be expected to proceed to
between 80 and 90% conversion. Entry 2 suggests that the addition of product to the reaction causes
some inhibitory effect albeit rather small. In Entry 3, the reaction was run in the presence of another
hydrogenation substrate, methyl acetamidocinnamate (MAC). This was completely reduced in the
presence of 6 and 8. While these results are not exhaustive, they do indicate that product inhibition of
the catalyst had a fairly minor effect.
Taking the results we had in hand, we scaled up the hydrogenation reaction to 15 g (Scheme 3). The
hydrogenation of the substrate 6 was initiated at a molar substrate to catalyst ratio (S/C) of 500, but
additional catalyst was added after 18 hours in order to achieve complete conversion, giving a final
substrate to catalyst ratio of 406. Interestingly, on this scale the reaction was observed to proceed to
88% conversion after a relatively short time (45 minutes), and the remaining material took a
considerable time to convert. The reason for the apparent drop in enantiomeric excess compared to the
results obtained during the screen was unclear.
ACS Paragon Plus Environment
8

Page 9 of 24
Organic Process Research & Development
O
O
1
2
AcHN
AcHN
[(S)-PhanePhos Rh COD]BF4 (S/C = 406)
OEt
OEt
3
4
H2 (12 bar), EtOH, 50oC, 21 hrs
5
6
F
F
F
F
98% conv, 88% e.e.
6
S-8
7
8
9
Scheme 3. The asymmetric hydrogenation of 6 with [(S)ꢀPhanephos Rh(COD)]BF4.
10
11
12
13
14
15
16
Complete deprotection and formation of the hydrochloride salt was accomplished by refluxing in 6M
hydrochloric acid (Scheme 4).
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
O
O
AcHN
HCl.H2N
OEt
OH
6M HCl, reflux
Quant.
F
F
F
F
-8
-9
S
S
Scheme 4. Hydrolytic deprotection of (S)-8.
The crude hydrochloride salt (S)ꢀ9 was then dissolved in water, the pH was adjusted and the nitrogen
was Boc protected to provide the desired material (Scheme 5). Although the enantiomeric excess
achieved in the hydrogenation step was lower than expected, the crystallization of (S)ꢀ1 gave material of
high e.e.
O
O
HCl.H2N
BocHN
1) Boc2O, THF, water, NaOH
2) Recrystallisation
OH
OH
F
F
F
F
S-9
S-1
77% yield, 99% e.e., 97.3% purity
Scheme 5. NꢀBoc formation and final crystallization of (S)ꢀ1.
ACS Paragon Plus Environment
9

Organic Process Research & Development
Page 10 of 24
The chemistry discussed above gave rise to a putative process for the synthesis of (S)-1. The next step
was to establish a synthesis which could be operated on a multi Kg scale. With a route established and
economically evaluated, it became apparent that the key step was the hydrogenation (Stage 4, Scheme
3). Not only was the [(S)ꢀPhanePhos Rh COD]BF4 catalyst the main raw material cost, at the 15ꢀg scale
an extra aliquot of catalyst was required to drive this reaction to completion (final S/C 406:1). For the
development of an economical and robust commercial process, optimisation of this particular step was a
necessity.
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
A design of experiment plan was set up to determine the important factors governing the
hydrogenation. Design parameters at a substrate to catalyst ratio (S/C) of 500:1.were:
Substrate Concentration: Low= 20 g/L; High= 100 g/L
H2 Pressure: Low= 4 Bar; High = 12 Bar
Reaction Temperature: Low= 20 °C; High = 50 °C
Umetrics software was used to process the data.11 As the product e.e. did not vary significantly during
these experiments (from 94 to 96%), the substrate conversion was used as the significant response. The
generated model showed good reproducibility, the model validity was correct (above 0.25) and the
ANOVA (analysis of variance) plot also showed a good model. The main factor was determined to be
the substrate concentration, followed by H2 pressure. The model predicted that better substrate
conversion would be obtained under conditions of high substrate concentration, high H2 pressure and
high reaction temperature
11 Version MODDE 6.0. www.umetrics.com
ACS Paragon Plus Environment
10

Page 11 of 24
Organic Process Research & Development
1
2
3
4
5
6
7
8
Conversion %
25
20
15
10
5
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
Conversion %
0
-5
-10
-15
-20
Figure 1. Coefficient plots to determine the effects of three factors (Temperature, Pressure of H2 and
Substrate Concentration) on the hydrogenation of 6.
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
As large scale hydrogenation would be carried out at a maximum pressure of 8 Bar due to equipment
limitations, it was decided to fix the pressure at this value for further experiments. The next set of
experiments was used to determine the effect of different catalyst loadings. S/C ratios of 500, 600, 750
and 1000:1 were used at 120 g/L substrate concentration at 50 °C and 8 Bar H2. At both S/C 500 and
600:1, >98% conversion was obtained. A slight drop to 95% conversion was observed at 750:1 and
about 87% at 1000:1. This suggests that the inhibition of the catalyst would prevent use at higher S/C
ratios.
Further studies at S/C of 1000:1 were carried out to investigate the effects of the use of a coꢀsolvent
(toluene, ethyl acetate, MTBE, THF), higher concentration, or temperature on conversion.
Unfortunately, it soon became apparent that the reproducibility of the hydrogenation at this catalyst
ACS Paragon Plus Environment
11

Organic Process Research & Development
Page 12 of 24
loading was very poor; for example at 120 g/L, 50 °C, 8 Bar, the conversion varied from 70 to 92%.
1
2
3
4
5
6
7
8
The effect of the presence of residues from previous steps in the route was investigated by the
addition of small amounts of these compounds to hydrogenations which were run using the standard
conditions used previously (120 g/L substrate concentration, 50 °C, 8 Bar H2). It was found that when
4,4’ꢀdifluorobenzophenone 3 or the NꢀFormyl product 5 from stage 1 were added, > 98% conversion
was obtained; however, only 4% conversion was obtained when the NꢀFormylꢀNꢀAcetyl compound 10
(the product of Stage 2) was added. This last experiment was repeated and confirmed that the presence
of this compound inhibited the hydrogenation.
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 2. The NꢀFormylꢀNꢀAcetyl intermediate 10.
One batch of substrate could not be hydrogenated at all. As the initial analysis (NMR and SFC) did
not identify an impurity (it was free of 10 contamination), it was thought that an inorganic contaminant
might be responsible for the poor quality hydrogenation. Consequently, a reaction with good substrate
containing 5% of potassium carbonate was run, and no reaction at all was observed. It is thus thought
that the presence of small amounts of potassium carbonate, which is used in the previous stage, also
prevented the hydrogenation of 6.
The troublesome batch was hydrogenated in presence of acetic acid (3% v/v in ethanol) at S/C 477 but
this did not give any conversion. However, after purification by recrystallization from ethanol/acetic
acid (8/1) this batch could be hydrogenated.
ACS Paragon Plus Environment
12

Page 13 of 24
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
As we were concerned that the catalyst could be deactivated by impurities in the substrate, we
examined the process conditions that were likely to arise at commercial scale, and particularly how they
differed from small scale lab conditions. Of particular concern was the longer time required at larger
scale for the reactor to reach operating pressure. In order to replicate this, a series of laboratory
experiments were carried out in which the reactor was brought to operating pressure over the course of
1h. This was repeated at various catalyst loadings and reaction temperatures (Table 4).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 4. Results of hydrogenations run under pilot plant conditions.
Entry
temp
S/C Time
Conv ee
1
2
3
50 °C 600 +1Hr
35 °C 600 +1Hr
35 °C 900 +1Hr30
+4Hr41
93%
99%
82% *
88%
97%
92%
94%
93%
97%
4
35 °C 750 +1Hr23
* The inꢀprocess sample was not homogeneous as the substrate is not completely soluble in ethanol.
The results, shown in Table 4, suggest that at 50 °C the catalyst was not sufficiently stable, since
under laboratory conditions at a S/C 750:1, complete conversion was obtained within 20 minutes.
Higher conversions were obtained at 35 °C. At S/C = 900, 750 and 600, conversions were 88%, 97%,
and 99%, respectively.
Hydrogenation of 6 containing 0.02% of 10 at 35 °C and a S/C = 595 gave only 65% conversion after
2 hours at 8 Bar. Additional catalyst (S/C of 433) with slow pressurization gave 76% conversion within
ACS Paragon Plus Environment
13

Organic Process Research & Development
Page 14 of 24
1 hr at 8 Bar and no further uptake of hydrogen after 30 minutes. Pressurization to 9 Bar drove the
reaction to 91% conversion (95% ee) overnight at 35 °C.
1
2
3
4
5
6
7
8
9
The experiments carried out demonstrate that variations in substrate concentration and temperature
both effect the hydrogenation; high concentration (>100 g/L) coupled with high temperature (>35 °C)
gave better conversion of substrate to product. However, at 50 °C the catalyst became unstable leading
to poor overall conversion despite an initial higher reaction rate.
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
Proposed pilot plant conditions were 35 °C, 8 Bar hydrogen pressure, at a concentration of 120 g
substrate/L.
PILOT PLANT CAMPAIGN
Although we had only made slight additional improvements to the overall process and to the
hydrogenation step in particular, it was felt that we had enough confidence in the process overall to
scale up the process to a multi Kg pilot plant campaign.
Problems were experienced with stage 1 of the reaction. The longer contact times between 4,4’ꢀ
difluorobenzophenone and potassium tertꢀbutoxide lead to the formation of significant amounts of 4,4’ꢀ
bisꢀ(tertꢀbutoxy)benzophenone. Reꢀexamination of samples previously made in the lab also showed
small amounts of 4,4’ꢀbis(tertꢀbutoxy)benzophenone, but not in the amounts observed in the pilot plant.
Modification of the order of addition (i.e. gradual addition of potassium tertꢀbutoxide in THF solution to
a mixture of 4,4’ꢀdifluorobenzophenone and ethyl isocyanoacetate in THF) gave a much improved
product purity and yield.
It also became very clear that the physical form of some of the intermediates were going to create
ACS Paragon Plus Environment
14

Page 15 of 24
Organic Process Research & Development
problems. Typically the materials formed thick slurries which were difficult to stir and to isolate by
filtration. The slurries with NꢀFormyl 5 and NꢀAcetyl 6 acrylates were both thick pastes. A consequence
of the poor physical form was the extended drying time and high residual solvent content of the
products. The loss on drying for 6 was approximately 20%; in order to save time, the product was
discharged from the dryer and analyzed to determine ethanol and water content. This was then used to
adjust the solvent charges in the product recrystallization. Fortunately, after recrystallization the
material was more tractable. The recrystallization step proceeded smoothly to generate a product which
filtered and dried very rapidly.
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
The full scale hydrogenation batch ran very smoothly, with hydrogen uptake being complete within 1
hour. Reaction occurred almost immediately after hydrogen addition, and it was estimated that when the
desired 8 bar pressure was achieved, the reaction was at approximately 80% complete based on gas
absorption. In process analysis showed 98.7% conversion and 94.1% ee, as expected.
The hydrolysis of the ester and amide groups was somewhat slow, requiring additional charges of
28% hydrochloric acid and extended reflux to obtain complete conversion. No new impurity peaks were
detected as a result of this.
The Boc protection step worked very well, being complete within 2 hours. Further problems,
however, occurred when acidifying the mixture as the product separated out as an oil. After a short
laboratory investigation, the batch was treated with 32% sodium hydroxide to give a two phase mixture,
and the lower aqueous layer discarded. The organic layer was diluted with water and partially reꢀ
acidified, giving a thick, unstirrable slurry. Due to the difficulty in obtaining a consistent pH throughout
the slurry, the mixture was filtered and washed with water and then the filtrates acidified in three
aliquots which were filtered and washed separately. In all, the crude material was obtained in four parts.
The problems with product physical form were repeated on the purification step, during which the
cooled slurry became very thick and then immobile. The mixture was transferred to the filter with
ACS Paragon Plus Environment
15

Organic Process Research & Development
Page 16 of 24
difficulty, and although the filtration rate was reasonable, when the flow of filtrates stopped, LOD
analysis was approximately 70%.
1
2
3
4
5
6
7
8
Despite the physical form difficulties, the recrystallization gave significant enhancement of
enantiomeric excess; analysis of the filtrates showed that the product waste stream had a composition of
50% e.e. (S).
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
MANUFACTURING CAMPAIGNS
Progressing to a 300 Kg campaign, more care was given to controlling cooling profiles for the
recrystallization of the hydrogenation substrate 6. While hydrogen pressure was expected to be an
important parameter, laboratory batches showed that the processing could be completed at 60 psig rather
than the 90 psig that had been used in the previous campaign. Laboratory studies of the hydrogenation
reaction using representative batches of 6 indicated that a catalyst loading of S/C = 1,200 could be
achieved. The first plant batch was carried out at S/C = 800. After implementation of handling
improvements, subsequent batches were run with S/C = 1,000 – 1,200. During process development
studies for this campaign, it was discovered that purification of the final product was best accomplished
by carrying out a preliminary partial purification of 1 as a mixed salt; adjusting the pH in the final
crystallization then gave the product in excellent purity. The NꢀBoc crude product could be difficult to
handle during filtration and washing steps, but this could be circumvented by the concurrent addition of
32% HCl and the basic Boc protected intermediate solution to a pH 3 water solution. The concurrent
addition precipitated the product to provide an easily isolated wet cake.
In the subsequent 900 Kg campaign, significant improvements were made to the process (Scheme 6).
By switching to a mixture of toluene and THF for the first step, it was possible to telescope the first two
steps. In previous runs after the acetylation step, the reaction mixture was quenched with water and the
intermediate NꢀformylꢀNꢀacetyl 10 was isolated. A rapid water addition with insufficient cooling led to
ACS Paragon Plus Environment
16

Page 17 of 24
Organic Process Research & Development
a temperature rise to 86 °C. HPLC analysis showed partial formation of the Nꢀacetyl 6, indicating an
acidic hydrolysis of the Nꢀformyl group. By adding water and heating to 70 °C, 60% conversion to 6
could be achieved. Using ethanol for the acetic anhydride quench, the hydrolysis product 6 may be
isolated directly by crystallization, affording a one pot procedure for the conversion of the Nꢀformyl 5 to
the Nꢀacetyl 6 (isolation # 1). This was followed by the recrystallization of the hydrogenation substrate
6 (isolation # 2). The overall yield for these first stages including the crystallization was slightly higher
than achieved in the previous campaigns. Studies of the downstream chemistry and purification found
that conversion in the hydrogenation reaction as low as 90% could be tolerated. In order to minimize the
amount of catalyst used, the first batch was run at S/C = 1,800 compared to the S/C = 1,200 used in the
previous 300 Kg campaign. The reaction proceeded to 93 % conversion and the downstream processed
as expected. Ten subsequent hydrogenation batches were run with catalyst loadings of between 1,600
and 2,700; all proceeded to 97 – 100 % conversion. The hydrogenation, hydrolysis and Boc protection
steps (6 → 1) were telescoped (isolation # 3). It had been previously noted that when sodium hydroxide
was used as base in the Boc addition step, the resulting solid was difficult to process, it was troublesome
to isolate on the centrifuge, and it had a very high loss on drying (35ꢀ40%). Furthermore, it gave an
inconsistent mixture of carboxylic acid and sodium salt. This complicated unit ratio calculation for the
next step and gave an inconsistent recovery in the crystallization. The process was improved by
switching to aqueous potassium hydroxide. The resulting solid, which was predominantly the carboxylic
acid form, was more easily handled. This eliminated the concurrent addition of acid and isolation of the
basic 1 intermediate used in the 300 Kg campaign.
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
It was further discovered that if the subsequent crystallization was carried out as a mixture of the
sodium salt and carboxylic acid, a polar mixture of water and 2ꢀpropanol could be used as solvent. This
solvent mixture greatly facilitated the removal of organic impurities, and the residual rhodium level
from the hydrogenation catalyst was typically reduced from 160 to 3 ppm. It was found that the optimal
recovery was achieved when 0.5 equivalents of sodium hydroxide was used (isolation # 4). In the final
ACS Paragon Plus Environment
17

Organic Process Research & Development
Page 18 of 24
crystallization, the pH was again adjusted to below 2.0 to give exclusively the carboxylic acid form
(ROI <0.1%) (isolation # 5). This material was essentially free of any rhodium residue (<1 ppm).
Unfortunately, after successfully completing this project, Phase III clinical trials of Denagliptin were
suspended.
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
Scheme 6. Overall scheme for the manufacture of 900 Kg of (S)ꢀNꢀBoc bis(4ꢀfluorophenyl)alanine
(S)-1.
CONCLUSIONS
The synthesis of (S)ꢀNꢀBoc bis(4ꢀfluorophenyl)alanine, (S)ꢀ1 has been developed from a lab sample to a
>900 Kg cGMP manufacturing campaign. The overall synthetic route, including the choice of
asymmetric hydrogenation catalyst, remained essentially unchanged throughout the process. With
ACS Paragon Plus Environment
18

Page 19 of 24
Organic Process Research & Development
hindsight it is clear that we had a sound chemical synthesis in place. The success of the project
depended on minimising the formation of process related impurities such as 4,4’ꢀbis (tertꢀ
butoxy)benzophenone and ensuring complete conversion of intermediates particularly the NꢀFormylꢀNꢀ
Acetyl compound 10. While selection of reaction conditions helped minimise the amount of these
impurities and intermediates, their presence could not be completely avoided. It was therefore necessary
to further lower the amount of these materials to a level that the rhodium catalyst would tolerate and the
final product would meet the customer’s specification. Critical to this was the identification of
crystalline intermediates with physical properties that readily allowed filtration and washing to give
material of high quality without sacrificing yield. These purifications were carried out at key stages in
the sequence. A purification immediately prior to the asymmetric hydrogenation step allowed the
amount of catalyst required to be lowered to an economically viable level. Careful selection of bases
during the isolation of the final product gave material of sufficient purity to satisfy the customer
specifications. With these successful isolations, it was possible to telescope several steps, reducing the
number of synthetic steps and additional purification steps from twelve in the preliminary sample
synthesis to five in the final manufacturing campaign.
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
EXPERIMENTAL SECTION
2-Acetylamino-3,3-bis(4-fluorophenyl)acrylic acid ethyl ester 6.
A 400 L reactor was loaded with 4,4’ꢀdifluorobenzophenone (159.7 Kg, 731.8 mol) followed by
toluene (756 Kg) and ethyl isocyanoacetate (95.9 Kg, 848 mol), and the load lines were rinsed with
toluene (47 Kg). The reactor contents were cooled to < 5 °C, and a solution of potassium tertꢀbutoxide
(20 % in THF, 431 Kg, 768 mol) was added at a rate to maintain the internal temperature at less than 10
°C. The reactor contents were warmed to 15 °C and held for 2 hours before sampling and analyzing for
conversion (0.15% unreacted 4,4’ꢀdifluorobenzophenone). A solution of ammonium chloride (10 %,
ACS Paragon Plus Environment
19

Organic Process Research & Development
Page 20 of 24
887 Kg) was added to the reaction mixture while maintaining the temperature at < 30 °C. The organic
phase was diluted with acetone (638 Kg), and the mixture was heated to 40 °C to complete the
dissolution of the intermediate. The aqueous phase was removed and discarded. The organic phase was
washed with 10 % sodium chloride solution (887 Kg). The organic phase was concentrated, ca 1450 Kg
of distillates were removed leaving the reactor contents as a thick slurry. The reactor was charged with
acetic anhydride (302 Kg, 2959 mol), and the reactor was cooled with a jacket temperature of 0 °C. The
load line was rinsed with toluene (9 Kg). Triethylamine (146 Kg, 1443 mol) was loaded over 30 – 60
min maintaining the reaction temp at < 10 °C, and the load line was rinsed with toluene (10 Kg). The
reactor contents were warmed to 20 °C and held for 5ꢀ6 hours. The reactor was sampled and analyzed
for conversion (0.7% stage 1 material). With the jacket temperature of 20 °C, the initial portion of
ethanol was added slowly. There was an initial exotherm to 60 °C and once this subsided the remaining
ethanol was added (823 Kg total). The reaction mixture was heated to reflux for 28 hours. The mixture
was sampled and analyzed for conversion (0.14% stage 2 intermediate). The mixture was cooled to 2 °C
over 7 h and then held at <5 °C for 4 h. The slurry was transferred to a centrifuge in two lots, washing
each load with chilled ethanol (118 Kg). About 220 Kg of wet cake was isolated (LOD 2.5%) and used
in the next step (dry weight 214 Kg, 620 mol, 85% yield).
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
Crystallization
A 2000 L reactor was charged with the wet cake (219 Kg, 210 Kg dry weight, 608 mol) and ethanol
(1158 Kg) and heated to reflux. The contents were cooled to 80 °C at 5 °C/min. After verifying that no
solids were present the mixture was cooled to 5 °C over 5 hr. The slurry was held at < 5 °C for at least 1
hr. The slurry was transferred to a centrifuge in three loads, and each load was washed with chilled
1
ethanol (56 Kg). A total of 209 Kg of wet cake was isolated (3% LOD, 587 mol, 96% yield) H NMR
(400 MHz, CDCl3) δ 7.24ꢀ7.12 (m, 2H), 7.12ꢀ7.04 (m, 4H), 7.02ꢀ6.96 (m, 2H), 6.78 (s, 1H), 4.03 (q, J =
13
7Hz, 2H), 2.02 (s, 3H), 0.97 (t, J = 7 Hz, 3H). C NMR (101 MHz, CDCl3) δ 168.5, 165.6, 162.7 (J =
249Hz), 162.6 (J = 250Hz), 135.4 (J = 4Hz), 134.3, 134.2 (J = 4Hz), 131.7 (J = 8Hz), 130.0 (J = 8Hz),
ACS Paragon Plus Environment
20

Page 21 of 24
Organic Process Research & Development
19
126.0, 116.0 (J = 22Hz), 115.3 (J = 22Hz), 61.4, 22.8, 13.5. F NMR (376 MHz, CDCl3) δ ꢀ113.1, ꢀ
111.6. HPLC Phenomenex LUNA 5ꢁ, Phenylꢀhexyl 4.6 x 150 mm, temp 24 °C, Eluent A 10/90
MeCN/water 0.1 % TFA, Eluent B 90/10 MeCN/water 0.1% TFA, gradient: 70% A to 30% A over 30
minutes. Flow rate 2 mL/min, injection vol 5ꢁL. Wavelength 254 nm. Retention times 2ꢀAcetylaminoꢀ
1
2
3
4
5
6
7
8
9
3,3ꢀbis(4ꢀfluorophenyl)acrylic acid ethyl ester 6, 9.9 min;
ethyl 3,3ꢀbis(4ꢀfluorophenyl)ꢀ2ꢀ
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
formamidoacrylate 5, 10.4 min; 4,4’ꢀdifluorobenzophenone 3 12.8 min; ethyl 3,3ꢀbis(4ꢀfluorophenyl)ꢀ
2ꢀ(Nꢀformylacetamido)acrylate 10, 16.0 min. The purity for the three loads from the centrifuge was 100
area %, 100 area% and 99.7 area %. The liquors contained 83 area % of 6 and 16 area % 5. The material
was use tested in a hydrogenation reaction before progressing to the next stage.
(S)-2-tert-Butoxycarbonylamino-3,3-bis(4-fluorophenyl)propionic acid, S-1.
A 2000 L pressure reactor was charged with 2ꢀacetylaminoꢀ3,3ꢀbis(4ꢀfluorophenyl)acrylic acid ethyl
ester 6 wet cake (120.3 Kg, 2.5% LOD, 348 mol) and ethanol 3C (530 Kg). The reactor was sealed and
leak tested.12 [(S)ꢀPhanePhos Rh COD]BF4 (115g, 0.131 mol, molar S/C = 2,650) was charged to a shot
tube in a glove box and then attached to the reactor. The catalyst charge was dropped into the reactor
and the charging apparatus was flushed with degassed ethanol (59 Kg). Addition of hydrogen to the
reactor was initiated, and the uptake of hydrogen was monitored as an indication of reaction progress.
The mixture was maintained at 40 °C, 60 psig hydrogen; conversion was 98.9% complete by HPLC
after 12 h. The reactor was vented and purged with nitrogen. The reaction mixture was transferred to a
1200 L vessel, rinsing with additional ethanol (118 Kg). The majority of the ethanol (614 Kg) was
removed by distillation. Water (307 Kg) and hydrochloric acid (32%, 186 Kg, 1,632 mol) were added
and rinsed in with additional water (12 Kg), and the mixture was heated to above 80 °C to allow the
removal of the ethanol byꢀproduct. Two additional water charges (113 Kg) were added. Distillation
12
The reactor was inerted by two cycles of evacuating to 200 mm/Hg then backfilling with nitrogen.
This was followed by three cycles of pressurising the vessel to 60 psig then venting.
ACS Paragon Plus Environment
21

Organic Process Research & Development
Page 22 of 24
continued until the reaction was complete (>95% by HPLC, 32 hr). The pH was adjusted to between 9
and 10 using aqueous potassium hydroxide (45%, 217 Kg, 1,740 mol), keeping the temperature below
25 °C. A solution of Boc anhydride (82 Kg, 377 mol) in acetone (93 Kg) was added. Additional aqueous
potassium hydroxide was added (82 Kg, 658 mol) to maintain the solution at pH 9 – 10. The solution
was checked for reaction completion (>99 % by HPLC) after 1 hr. The pH was lowered to 2.6 using
32% HCl (96 Kg, 843 mol) and rinsed in with water (23 Kg). The product began to precipitate at a pH
below 4.5 with significant carbon dioxide evolution. The slurry was filtered by centrifugation in three
loads and washed with water (380 Kg). The product was collected as a wet cake (216 Kg, LOD 40%,
129 Kg dry weight, 342 mol, 98 % yield).
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
Crystallizations of (S)-2-tert-Butoxycarbonylamino-3,3-bis(4-fluorophenyl)propionic acid, S-1:
A 4500 L nitrogenꢀinerted reactor was charged with 2ꢀpropanol (816 Kg) and 392 Kg of crude S-1
wet cake (equivalent to 235 Kg, 623 mol dry weight) was added. After sealing and inerting the reactor, a
additional 627 Kg of 2ꢀpropanol was added. The mixture was heated to above 65 °C. Sodium hydroxide
(50%, 29 Kg, 363 mol) was added over 1 hour followed by water (29 Kg). The mixture was heated to 80
°C for 1 h, then cooled to ꢀ5 °C over 16 h. The resulting slurry was held at ꢀ5 °C for about 1 hr. The
mixed salt wet cake was collected and washed with chilled 2ꢀpropanol (236 Kg) to give 348 Kg of wet
cake.
A 1200 L reactor was loaded with ethanol (526 Kg) and cooled to less than 4 °C. Water (72 Kg) and
the mixed salt wet cake (339 Kg) were added. The solution was heated to about 75 °C and aqueous
hydrochloric acid (32%, 30.9 Kg, 271 mol) was added in portions until the pH dropped below pH 2. The
solution was passed through a preheated polish filter into a 2000 L reactor at 65 °C and rinsed through
with ethanol (73 Kg). Water (658 Kg) was added and cooled to ꢀ5 °C over 14 h and the slurry was held
2 hr. The wet cake was collected and washed with ethanol/water (50/50 w/w, 228 Kg). Analysis found a
chemical purity 99.7%, and an optical purity 99.6% e.e.,for the 175 Kg (dry basis) of final isolated
ACS Paragon Plus Environment
22

Page 23 of 24
Organic Process Research & Development
product (464 mol, 74% yield). The wet cakes from 2 to 3 centrifuge loads were dried under vacuum (50
°C 70 – 180 mmHg) to generate (S)ꢀ2ꢀtertꢀButoxycarbonylaminoꢀ3,3ꢀbis(4ꢀfluorophenyl)propionic
acid, Sꢀ1, average yield 77%. 1H NMR (400 MHz, DMSO) δ 12.4 (1H), 7.40ꢀ7.31 (m, 4H), 7.29 (d, J =
1
2
3
4
5
6
7
8
13
9Hz, 1H) 7.15ꢀ7.04 (m, 4H), 4.78 (dd, J = 9 & 11 Hz, 1H), 4.31 (d, J = 12 Hz, 1H), 1.28 (s, 9H). C
9
NMR (101 MHz, CDCl3) δ 172.5, 160.9 (J = 243Hz), 160.8 (J = 243Hz), 155.1, 137.41, 137.38, 130.1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
19
(J = 8Hz), 130.0 (J = 8Hz), 115.1 (J = 21Hz), 114.8 (J = 21Hz), 78.2, 56.9, 51.0, 28.0. F NMR (376
MHz, CDCl3) δ ꢀ116.7, ꢀ116.2. Distinct NMR signals associated with the minor rotamer (ca 15%) were
1
observed at: H NMR (400 MHz, CDCl3) δ 6.90ꢀ6.84 (m), 4.70ꢀ4.61 (m), 4.25 (d, J = 11Hz), 1.31 (s).
13C NMR (101 MHz, CDCl3) δ 78.6, 27.8. 19F NMR (376 MHz, CDCl3) δ ꢀ116.5, ꢀ116.1. HPLC
Column Zorbax Eclipse XDBꢀC18, 3.5 ꢁ, 4.6 x 150 mm, eluent A water 0.05% TFA, eluent B MeCN
0.05% TFA. Gradient 100% A to 0% A over 25 min, flow rate 1 mL/min, detection 254 nm, column
temp 40 °C injection vol 1ꢁL. The sample was prepared at 1 mg/mL in 9:1 MeCN:water. Retention time
(S)ꢀ1 11.2 min, Purity 99.7 area %, impurities at RRT 0.98 0.1 area % and RRT 1.16, 0.2 area %. Loss
on drying <0.1 % residue on ignition <0.1%, water by Karl Fisher 0.1% Chiral HPLC method Column
OJꢀRH, 5ꢁ, 4.6 x 150 mm, eluent A water 0.05% TFA, eluent B MeCN 0.05% TFA; isocratic 35 % B
over 20 min, flow rate 1 ml/min, detection 254 nm, column temp 40 °C injection vol 1ꢁL. The sample
was prepared at 1 mg/ml in 9:1 MeCN:water. Retention time (S)ꢀ1 11.2 min, (R)ꢀ1 10.1 min, 99.5% e.e.
Acknowledgements
We thank Paul H Moran and Justine Peterson for support in the manufacture of catalyst.
TOC
ACS Paragon Plus Environment
23

Organic Process Research & Development
Page 24 of 24
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
ACS Paragon Plus Environment
24