Synthesis of all four stereoisomers of 3-(tert-butoxycarbonyl)-3- azabicyclo[3.1.0]hexane-2-carboxylic acid

Article abstract of DOI:10.1021/jo4013282

A synthesis of all four stereoisomers of 3-(tert-butoxycarbonyl)-3- azabicyclo[3.1.0]hexane-2-carboxylic acid has been developed, thereby significantly shortening the known literature procedures for the syntheses of these unnatural amino acids. With a simple adjustment of the reaction conditions, we were able to obtain either pure cis or trans acid. Optical resolution was accomplished via diastereomeric salt formation or alternatively via chromatography on a chiral stationary phase. Finally, ab initio calculations gave an explanation for the observed cis selectivity in the initial step.

Full text of DOI:10.1021/jo4013282

Subscriber access provided by Library, Special Collections and Museums, University of Aberdeen
Article
Synthesis of all four stereoisomers of 3-(tert-
butoxycarbonyl)-3-azabicyclo[3.1.0]hexane-2-carboxylic acid
Bettina Bakonyi, Markus Furegati, Christian Kramer, Luigi La Vecchia, and Flavio Ossola
J. Org. Chem., Just Accepted Manuscript • DOI: 10.1021/jo4013282 • Publication Date (Web): 19 Aug 2013
Downloaded from http://pubs.acs.org on August 21, 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.
The Journal of Organic Chemistry 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 34
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Synthesis of all four stereoisomers of 3-(tert-
butoxycarbonyl)-3-azabicyclo[3.1.0]hexane-2-
carboxylic acid
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Bettina Bakonyi, † Markus Furegati,*,‡ Christian Kramer,§ Luigi La Vecchia, ‡ Flavio Ossola‡
†
Doetsch Grether AG, Falkensteinerstrasse 37, 4132 Muttenz, Switzerland
‡
Global Discovery Chemistry, Novartis Institutes for Biomedical Research, Preparations
Laboratories, Klybeckstrasse 141, 4057 Basel, Switzerland
E-mail: markus.furegati@novartis.com
§
Institute of General, Inorganic and Theoretical Chemistry, and Center for Molecular Biosciences
Innsbruck (CMBI), University of Innsbruck, Innsbruck, Austria
TABLE OF CONTENTS / ABSTRACT GRAPHIC
ABSTRACT
A synthesis of all four stereoisomers of 3-(tert-butoxycarbonyl)-3-azabicyclo[3.1.0]hexane-2-carboxylic
acid has been developed; thereby significantly shortening the known literature procedures for the
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 2 of 34
syntheses of these unnatural amino acids. With a simple adjustment of the reaction conditions, we were
able to obtain either pure cis or trans acid. Optical resolution was accomplished via diastereomeric salt
formation or alternatively via chromatography on a chiral stationary phase. Finally, ab-initio
calculations gave an explanation for the observed cis selectivity in the initial step.
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
KEYWORDS
scale-up, lithiation, diastereomeric salt formation, flow, ab-initio.
INTRODUCTION
The four stereoisomers of 3-azabicyclo[3.1.0]hexane-2-carboxylic acid (series 1), also referred to as
2-carboxy-3,4-methanopyrrolidines or 3,4-methanoprolines, belong to a unique class of amino acids.1
The naturally occurring non-proteinogenic amino acid (S)-cis-1 derived from L-proline was discovered
in 1969 in the fresh seeds of the American horse chestnut, Aesculus parviflora2 and also found in the
stem tissue of E. foeminea.3
In the first three decades after the discovery of the 3-azabicyclo[3.1.0]hexane-2-carboxylic acids only
little has been published about this compound class. In 1980, the racemic mixture of both the geometric
isomeric forms were found to be active as plant male gametocides.4 It has also been shown that (S)-cis-1
and (S)-trans-1 strongly inhibit the proline permease in E. Coli.5 (S)-trans-1 (referred to as exo) has
been marketed by Calbiochem AG as a male sterilant in wheat.6 Since 2004, all four stereoisomers have
been patented for a range of applications.7
Scheme 1. Syntheses A-E of 3-azabicyclo[3.1.0]hexane-2-carboxylic acid
ACS Paragon Plus Environment

Page 3 of 34
The Journal of Organic Chemistry
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
Shortly after the first isolation of (S)-cis-1, a chemical synthesis was published (Scheme 1A). The
synthesis begins with protection of functional groups of (S)-2,5-dihydro-1H-pyrrole-2-carboxylic acid
(2). N-trifluoroacetyl-3,4-dehydro-L-proline methyl ester (3) was then treated, neat, with copper (I)
chloride and excess diazomethane to give the cyclopropyl amino acids (S)-cis-1 and (S)-trans-1 in a cis
to trans ratio of 1:3.5 after deprotection.2b A second synthesis (Scheme 1B), starting from 3-
azabicyclo[3.1.0]hexane (4) involves addition of a carboxy group in position 2, introduced via
chlorination of the nitrogen, elimination of HCl to the imine, formation of the bisulfite adduct and
treatment with sodium cyanide followed by saponification with barium hydroxide (Scheme 1B).8 The
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 4 of 34
racemic mixture of cis- and trans-2-cyano-3-azabicyclo[3.1.0]hexanes (5) was separated and converted
to the corresponding acid mixture 6.9 The cis/trans mixture 5 has been also prepared by cyclising cis-1-
ethoxycarbonyl-2-formylcyclopropane (7) with ammonia and hydrogencyanide followed by the
reduction of the intermediate 2-cyano-3-azabicyclo[3.1.0]hexane-4-one (8) (Scheme 1C).10 Conversion
of cis-ethyl-2-cyanocyclopropylcarboxylate (9) to cis-2-aminomethylcyclopropyl-1,2-diethylacetal (10)
and ring closure to the pyrrolidine ring gave racemic (S)-cis-1 in 8 steps (Scheme 1D).4 Alternatively,
rac-cis-1 was prepared via ring contraction of 5-chloro-3-azabicyclo[4.1.0]heptane-4-one (12), which
was prepared from 3-hydroxyimino[3.1.0]hexane (11) in 3 steps (Scheme 1E).11
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. Syntheses F-I of 3-azabicyclo[3.1.0]hexane-2-carboxylic acid
Starting with the chiral synthons 13 and (R)-glycidyl triflate (14), (S)-cis-1 was obtained in 6 linear
steps in 34% overall yield (Scheme 2F) via intermediate 15.12 (S)-cis-1 was also prepared in a 10 step
ACS Paragon Plus Environment

Page 5 of 34
The Journal of Organic Chemistry
sequence containing a formal insertion of the cyclopropylidene (derived from 17) or a related carbenoid
into the CH-bond adjacent to nitrogen as the key step to form bicyclo system 18 (Scheme 2G). The
overall yield was 40%, however, starting material 16 had to be prepared via oxidation of 1,1-dibromo-2-
vinylcyclopropane with KMnO4 followed by resolution with dehydroabietylamine first.13 The HCl salt
of (S)-trans-1 was prepared in 11 steps (11% overall yield) starting from L-pyroglutamic acid (19)
(Scheme 2H). The rigid structure of the O,N-acetal 20 with the directing phenyl ring allowed the
stereoselective cyclopropanation to 21.14 A more recent route towards (S)-trans-1 also started from L-
pyroglutamic acid (19) (Scheme 2I). Protection of the carboxylic acid as the ortho-ester 22 avoided
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
reduction/oxidation steps. The double bond in 22 was introduced by
a
PhSeCl
substitution/oxidation/elimination sequence. A 1,3-dipolar cycloaddition with diazomethane followed
by a photoinduced ring contraction forming 23 were the key steps. (S)-trans-1 was obtained in overall
yield of 10% in 12 linear steps. (S)-cis-1 was prepared in 15 steps via the same intermediate 22
respectively.15 All the above described syntheses are rather lengthy and not all starting materials are
easily accessible. From a user’s standpoint, a short synthesis allowing access to all four stereoisomers is
desirable.
RESULTS AND DISCUSSION
In this contribution we would like to present two approaches to prepare enantiopure cis and trans 3-
(tert-butoxycarbonyl)-3-azabicyclo[3.1.0]hexane-2-carboxylic acids, (R)/(S)-cis- and (R)/(S)-trans-30.
We started our investigations with tert-butyl 3-azabicyclo[3.1.0]hexane-3-carboxylate (24) as it became
commercially available in kg quantities or can be prepared in 5 steps from commercial cyclopropyl-1,2-
dicarboxylic acid (Scheme 3).16 After lithiation and quenching with CO2 gas we obtained cis-3-(tert-
butoxycarbonyl)-3-azabicyclo[3.1.0]hexane-2-carboxylic acid (rac-cis-30) with a stereospecificity
greater than 99%.
Lithiation of prolines and the use of diamine additives. Barker et al. showed that the diamine free
lithiation of tert-butyl pyrrolidine-1-carboxylate (1.3 equiv. sec-BuLi in THF for 1 h at –40°C) and
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 6 of 34
subsequent reaction with benzaldehyde gave 64% of the addition product.17 In our hands, when these
conditions were translated to the lithiation of substrate 24 followed by quenching with CO2, neither the
cis- nor the trans-acid was observed (Scheme 3 and Table 1 entry 1).18
1
2
3
4
5
6
7
8
9
A mixture of sec-BuLi/TMEDA is widely used in lithiation reactions. For example, the lithiation of
tert-butyl pyrrolidine-1-carboxylate (Boc-pyrrolidine) followed by reaction with various electrophiles
has been extensively described in the literature,19 also with CO2 as the electrophile.20 The lithiation of
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
our substrate 24 in diethylether at –70°C and subsequent reaction with trimethylborate without
21
elucidation of the stereochemical outcome has been also described.16b,
The presence of a
stoichiometric amount of TMEDA accelerates and directs the deprotonation at low temperatures.22 Other
chelating diamines have been used for this purpose as well. For example 3,7-dipropyl-3,7-
diazabicyclo[3.3.1]nonane (27) in the lithiation of N-tert-butyloxycarbonyl-3-azabicyclo[3.3.0]octane
using sec-BuLi23 allows the direct formation of mostly the trans product.24 More literature about related
carbanion electrophilic substitutions is given.25
We performed two experiments in order to compare the efficiency of diamine 27 and the cheaper
TMEDA. Our substrate 24 was lithiated with sec-BuLi/27 or with sec-BuLi/TMEDA. The acids
obtained were converted to the benzyl esters in order to facilitate analysis. With diamine 27 the isolated
yield was 74% of pure rac-cis-30 (Table 1 entry 2). The trans isomer was not observed. Using TMEDA
the yield was 69% but a mixture rac-cis-25, rac-trans-25 and 26 was obtained (Table 1 entry 3). We
therefore concluded that diamine 27 was required to give high diastereoselectivity for carboxylation and
a maximum yield.
Scheme 3. Literature synthesis of 24, lithiation/CO2 quenching and benzylation sequence
ACS Paragon Plus Environment

Page 7 of 34
The Journal of Organic Chemistry
Reaction
rac-cis-25
conditions
1
2
3
4
5
6
7
8
Entry
Additive
none
rac-trans-25
26 a
1
2
3
5 h, –40°C
5 h, –70°C
5 h, –70°C
0
0
3
100 b
0
2
0
9
TMEDA
77
21
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 Structure suggestion based on MS data. b Area% HPLC at 215 nm. The numbers reflect the amounts of the different acids
in the crude mixture after the quenching with CO2.
Table 1. Results of the lithiation/CO2 quenching step depending on the amine additive
Enantioselective deprotonation. Kerrick et al. showed that the deprotonation of tert-butyl
pyrrolidine-1-carboxylate (28) with sec-BuLi/(–)-sparteine in Et2O and subsequent quenching with CO2
provided 29 in 55% yield with 88%ee.26 The same conditions were applied to 24, with the exception of a
prolonged reaction time; resulting in only 32% yield and 73%ee of (R)-cis-30 (Scheme 4). Due to this
discouraging result, other sparteine-like chiral amines27 were not tested. Rather than preparing the single
enantiomers directly, we synthesised the racemic cis- and trans-Boc-protected 3-
azabicyclo[3.1.0]hexane-2-carboxylic acids (rac-cis-30 and rac-trans-30) and resolved them using two
alternative appoaches discussed below.
Scheme 4. Enantioselective deprotonation. a) literature, b) own work
Approach 1 – Resolution by diastereomeric salt formation. Our first approach was to prepare rac-
cis-30 and rac-trans-30 and resolve them via diastereomeric salt formation. The salt forming conditions
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 8 of 34
were found by screening 192 different resolving agent and solvent combinations for both the cis and the
trans acids (see salt screen in the experimental section for more details).
1
2
3
4
5
6
7
8
Synthesis of the cis enantiomers. Rac-cis-30 was resolved with dehydroabietylamine into the (–)-
enantiomer (S)-cis-30 in 32% yield and with (–)-cinchonidine into the (+)-enantiomer (R)-cis-30 in 35%
yield (Scheme 5).
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
Synthesis of the trans enantiomers. We found that it is possible to epimerize rac-cis-30 to rac-trans-
30 via double deprotonation with LDA (cis/trans ratio of 1:9). We combined this epimerization step
with the already established rac-cis-30 synthesis (Scheme 5). Taking advantage of the fact, that rac-cis-
30 existed as the mono lithium salt in the reaction mixture, only one equivalent of LDA would be
necessary to perform the epimerization. We observed however, that excess CO2 partially quenched the
LDA28 and prevented the reaction reaching equilibrium. It was therefore necessary to not overdose the
CO2 and this was accomplished by sparging the CO2 in portions with careful observation of the
temperature. As soon as no exotherm was observed, the addition of CO2 was stopped. A sample was
then titrated with n-BuLi/fluorene in order to determine the excess CO2, which was in our case 0.6
equiv. Using this information, the required amount LDA was determined. On our largest scale, 389 g of
rac-trans-30 were prepared showing a 51% yield and containing less than 2% of the cis acid after two-
fold crystallization. Finally, rac-trans-30 was resolved in EtOAc with (R)- and (S)-phenylglycinol in
29% and 15% yield respectively. The solvent choice was crucial; EtOAc with 0.5% water was required
to give good crystallization.29
Scheme 5. Approach 1 – Resolution by diastereomeric salt formation
ACS Paragon Plus Environment

Page 9 of 34
The Journal of Organic Chemistry
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
Approach 2 – Resolution by chromatography. From previous experiments we learned that it is
possible to resolve benzylester rac-25 into its enantiomers (S)- and (R)-cis-25 by chromatography on a
chiral stationary phase. In our second approach we planned to convert the enantiomerically pure cis
compounds (S)- and (R)-cis-25 to the (R)- and (S)-trans-25 by epimerization followed by hydrogenation
in order to obtain (S)- and (R)-trans-30 (Scheme 6). We chose the benzylester because it made UV-
detection during chromatography easy and we expected that the mild deprotection conditions would not
affect the stereochemical integrity. We initially planned to synthesize the benzylester rac-cis-25 directly
by reacting the lithiated intermediate from step 1 with benzylchloroformate instead of CO2. In contrast
to the excellent cis selectivity for the reaction with CO2, the use of benzylchloroformate led to
significant formation of trans product rac-trans-25. The longer the reaction time and the higher the
temperature after the electrophile addition, the lower the cis/trans ratio and the smaller the sum of cis
and trans product compared to an internal standard. Due to this observed epimerization we decided to
follow a two-step protocol (Scheme 6, step 1). The benzylation was accomplished with the crude rac-
cis-30 using an excess of benzyl bromide/K2CO3 in acetone. After filtration and solvent exchange for
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 10 of 34
ammonia in methanol, the benzyl bromide was converted to benzylamine to facilitate removal by acid
extraction. Chromatography on silica gel provided 242 g rac-cis-25 in 58% yield. Next, we wanted to
epimerize rac-cis-25 to the trans ester rac-trans-25.
1
2
3
4
5
6
7
8
Scheme 6. Approach 2 – Resolution by chromatography
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
We identified LDA as the best base for the epimerization30 and quickly learned that short reaction times
were key to success. The short reaction time of <60 s necessary for batch reactions31 would have been
challenging for scaling up and we expected decreased yields for larger batches. With the aid of a
commercially available flow machine (Scheme 7) we were able to achieve these very short reaction
times and to produce 17 g of (S)-trans-25 with a productivity of 55 g/h (55% yield).32
Scheme 7. Flow setup for the epimerization of (S)-cis-25 and 38 using a Vapourtec flow machine
ACS Paragon Plus Environment

Page 11 of 34
The Journal of Organic Chemistry
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
Other approaches. As a mild alternative for the epimerization we utilized aldehyde promoted
racemization via an iminium intermediate33 on deprotected amines rac-cis-31 and rac-trans-31 (Scheme
8). Unfortunately, the thermodynamic equilibrium between rac-cis-31 and rac-trans-31 could not be
pushed past 1:1.
Scheme 8: Aldehyde promoted epimerization via iminium intermediate
The formation of esters or amides from Boc-protected anologue, rac-cis-30 using chiral alcohols or
amines gave diastereoisomers that were potentially separable by crystallization or chromatography.
However, the corresponding diastereomers obtained by coupling (–)-menthol, (–)-borneol, (+)-1-
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 12 of 34
phenylethanol and (–)-1-phenylethylamine with rac-cis-30 showed poor separation by chromatography
and were only obtained in yields between 27 and 49% after coupling with EDCI, DMAP, NEt3 in
dichloromethane.
1
2
3
4
5
6
7
8
9
Stereochemical assignment. We synthesized (S)-trans-1 starting from L-pyroglutamic acid (19)
following literature protocols (Scheme 2H)14 and obtained (S)-trans-25 after Boc protection. The sample
of (S)-trans-25 showed a negative optical rotation. A crystal structure of the (R)-phenylglycinol salt of
(R)-trans-30 confirmed the stereochemical assignments.
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 origin of cis-selectivity. Ab-initio calculations were carried out in order to explain the
unexpected cis selectivity during the deprotonation and CO2-addition sequence described (Scheme 5).
The energy of the thermodynamic trans-product rac-trans-30 is 3.46 kcal/mol lower than the energy for
the cis-product rac-cis-30, which is in qualitative agreement with the observed 9:1 excess for rac-trans-
30 after complete epimerization. For the intermediate complex of 24 with lithium and diamine 27, the
cis-conformation is 3.90 kcal/mol lower than the trans-conformation. The lithium coordinated complex
without diamine 27 is 3.55 kcal/mol lower in energy than the trans and this is in very good agreement
with the sole formation of cis-product rac-cis-30. The agreement between both basis set levels is very
good, with relative energy differences between the lithium intermediate and the final structures of 0.12
and 0.16 kcal/mol, and average root mean square deviation of RMS = 0.044 Å. The 3,7-dipropyl-3,7-
diazabicyclo[3.3.1]nonane complex of the intermediate was calculated with the 6-311G** basis set only
and with methyl carbamate instead of Boc, while the Li-only coordinated intermediate was calculated
using the PVTZ basis set.
The structures of the energy-minimized cis- and trans intermediate complexes are shown in Figure 1.
ACS Paragon Plus Environment

Page 13 of 34
The Journal of Organic Chemistry
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
Figure 1: Energy minimized a) cis- and b) trans Li-complex of 24 (diamine 27 not shown for clarity)
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
with atom naming convention used for describing dihedrals
In the overlay, both complexes are very similar, i.e. an almost perfectly planar Li-O=C-N-C-Li ring is
the dominating structural motif. The energetic preference for cis-Li-24 can be rationalized by comparing
dihedral angles between substituents of the 5-rings: In the minimum energy cis-structure, the dihedral
Li-C2-C3-C6 is 56.7º, the dihedral H-C2-C3-H is 48.9º, the dihedral H-C4-C5-H is 40.9º and the
dihedral C6-C5-C4-H1 is 70.2º. For the minimum energy trans-structure, the dihedral Li-C2-C3-H is
16.9º, the dihedral H-C2-C3-C6 is 30.7º, the dihedral H-C4-C5-H is 8.4º and the dihedral C6-C5-C4-H1
is 40.4°. Thus, the trans intermediate contains more eclipsed-like features which are energetically less
favorable than the more staggered conformations of the cis-intermediate.
Our calculations indicated that the cis-selectivity is introduced at the stage of the Li-24-diamine-27
complex and then maintained through a CO2-insertion mechanism with retention.
CONCLUSION
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 14 of 34
In conclusion we have developed a one-pot protocol for the synthesis of the cis and trans acids rac-
cis-30 and rac-trans-30, in 74% and 51% yield, respectively. They can be resolved via diastereomeric
salt formation into all four isomers. Alternatively, one can form the benzyl ester from rac-cis-30
followed by a resolution on a chiral stationary phase. Hydrogenation of these benzylesters delivered the
enantiomerically pure acids (R)- and (S)-cis-30 or, after epimerization, the enantiomerically pure acids
(R)- and (S)-trans-30.
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
General. All reagents were purchased and used as received unless otherwise noted. tert-Butyl 3-
azabicyclo[3.1.0]hexane-3-carboxylate (24) and 3,7-Dipropyl-3,7-diazabicyclo[3.3.1]nonane (27) were
commercially available. The latter was distilled at 0.04 mbar and 80–115°C prior to use. Cis-3-(tert-
butoxycarbonyl)-3-azabicyclo[3.1.0]hexane-2-carboxylic acid (rac-cis-30) was commercially available
and used (N-Boc protected)21 for the resolution via diastereomeric salt formation. HPLC method 1:
XDB-C18 column, 4.6 mm x 50 mm, 1.8 μm, using acetonitrile and water as eluent (both containing
0.05% TFA), column temperature of 35 °C, flow rate of 1.0 mL/min, detection at 216 nm. The standard
gradient used was 5 to 100% acetonitrile over 6 min, 100% acetonitrile for 1.5 min, followed by 100 to
5% acetonitrile over 0.5 min. HPLC method 2 (for ee determination): IC (IC00CE-LB014, 250x4.6 mm,
5 um) column, 100% dichloromethane, column temperature 25 °C, flow rate of 1.0 mL/min, detection at
250 nm, run time 10 min. HPLC method 3 (for ee determination): AD-H (250x4.6 mm, 5 um) column,
hexane/2-PrOH 93:7 + 0.1% TFA, column temperature r.t., flow rate of 0.5 mL/min, detection at 210
nm, run time 40 min. HPLC method 4 (for ee determination): IC (250x4.6 mm, 5 um) column,
heptanes/EtOH/MeOH 90:5:5 + 0.1% TFA, column temperature r.t., flow rate 1 mL/min, detection at
210 nm, run time 20 min. Purities were characterized with area% at the wave length declared for the
method used. GC: Silaren column (30 m x 0.32 mm ID, 0.12 μm film). The standard 12-min run started
at 40°C which was held for 0.3 min, followed by a temperature ramp at 25 °C/min up to 220 °C and a
second ramp of 40 °C/min up to 280 °C, at which the temperature was held for 3 min. The hydrogen
flow was 2 mL/min, the front inlet temperature was 220 °C, and the front detection temperature was 300
ACS Paragon Plus Environment

Page 15 of 34
The Journal of Organic Chemistry
°C. LC-MS method 1: Acquity HSS T3 1.8 um 2.1 x 50 mm column at 50 °C. Eluent A: water + 0.05%
formic acid + 3.75 mM ammonium acetate; eluent B: acetonitrile + 0.04% formic acid. Gradient: from 2
to 98% B in 1.4 min with a flow rate of 1.2 mL/min, detection at 215 nm. NMR was performed using a
1
2
3
4
5
6
7
8
9
1
13
400 MHz machine. H shifts were referenced to DMSO-d6 at 2.49 ppm and CDCl3 at 7.26 ppm. C
shifts were referenced to DMSO-d6 at 39.52 ppm and CDCl3 at 77.16 ppm. High resolution mass
spectrometry (HRMS) was performed using QTOF with Classic Acquity UPLC with PDA. Elemental
analyses were performed externally complying ISO 9001 standard. Glass vessels for small scale
lithiation reactions were heated up to >150°C and cooled in vacuo or in a stream of argon.
Specifications of the Vapourtec flow equipment were described earlier by our group;34 we used the
following micromixer: “Comet X-01” from Techno Applications Co., Ltd.,Tokyo, Japan).
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
Ab-initio calculations. All calculations were done using Jaguar Version 7.9 (Suite 2012, Schrödinger,
LLC, New York, NY, 2012). Geometries were initially optimized using the M06-2X DFT method at the
6-311G** basis set.35 Where possible, the key structures were fully reoptimized using the M06-2X
functional at the PVTZ basis set level, as recommended by Schenker et al.36
Cis-2-benzyl 3-tert-butyl 3-azabicyclo[3.1.0]hexane-2,3-dicarboxylate (rac-cis-25). Equipment:
Buechi 30 L hastelloy reactor (CR30) with FlexyALR Systag control. To a solution of 250 g (1.36 mol)
tert-butyl 3-azabicyclo[3.1.0]hexane-3-carboxylate (24) and 360 g (1.71 mol, 1.25 equiv.) 3,7-dipropyl-
3,7-diazabicyclo[3.3.1]nonane (27) in 5 L dry THF at –84°C were added within 15 min 1.46 L (2.05
mol, 1.5 equiv.) sec-BuLi solution (1.4 M in cyclohexane). The progress of deprotonation was
monitored with GC (quenching of a sample with excess benzaldehyde). After 3 h at –85°C a mixture of
dry CO2 gas was sparged through the reaction mixture in such a manner, that the internal temperature
did not exceed –76°C. To the reaction mixture was added 2 L of 20% KHSO4 solution (until pH = 7)
and 5 L water and warmed to ambient temperature. After 4 L THF were distilled off, the pH was
lowered to 4 and the residue was extracted 3 times with a total of 7 L MTBE. The combined organic
phases were dried over sodium sulfate, filtered and concentrated to yield 259.8 g of a dark brown oil.
This was dissolved in 5 L acetone at r.t. to which were added 173 g (1.25 mol) potassium carbonate and
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 16 of 34
135 mL (1.14 mol) benzyl bromide. The reaction mixture was stirred at r.t. overnight. The white
suspension was filtered over celite and concentrated to give 357 g of a dark, brown oil, which was
stirred with 1 L heptanes for 1 h at r.t., filtered and concentrated to give 330 g oil (with this step the
most polar impurities were removed). In order to remove excess benzyl bromide, the crude product was
dissolved in 1.6 L THF, then 320 mL of a 7N ammonia solution in MeOH were added and the reaction
mixture stirred at 40°C for 6 h. The reaction mixture was concentrated and treated with 2 L 10% KHSO4
solution, 1 L crushed ice and 2 L MTBE. The aqueous phase was extracted twice with each 0.8 L
MTBE. The combined organic phases were dried over sodium sulfate, filtered and concentrated to yield
286 g of an orange/brown oil. The crude product was purified on a 2 kg silica gel column (5 injections)
with an EtOAc/heptanes gradient to yield 242 g (56% yield over two steps) cis-acid benzyl ester. Purity
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
1
> 97% (HPLC method 1). H-NMR (600 MHz, DMSO-d6) δ 7.38–7.32 (m, 5H), 5.19–5.07 (m, 2H),
4.35–4.34 (m, 1H), 3.49–3.44 (m, 1H), 3.37 (d, J = 10.4, 1H), 1.93–1.87 (m, 1H), 1.68–1.62 (m, 1H),
13
1.33 and 1.23 (2s, 2 rotamers 4:6, 9H), 0.66–0.61 (m, 1H), 0.52–0.47 (m, 1H). C-NMR (151 MHz,
DMSO-d6) δ 170.7, 170.2, 154.1, 153.3, 136.1, 135.8, 128.5, 128.4, 128.2, 127.9, 127.7, 79.2, 66.0,
1
65.6, 60.2, 60.0, 49.6, 49.5, 28.0, 27.7, 20.5, 19.4, 16.3, 15.6, 8.3, 8.1 (rotamers). H-NMR (600 MHz,
DMSO-d6, 100°C) δ 7.41–7.31 (m, 5H), 5.20–5.13 (m, 2H), 4.38 (d, J = 5.3 Hz, 1H), 3.52 (dd, J =
10.4, 5.0 Hz, 1H), 3.40 (d, J = 10.5 Hz, 1H), 1.96–1.90 (m, 1H), 1.71–1.64 (m, 1H), 1.34 (s, 9H), 0.68–
0.63 (m, 1H), 0.58–0.54 (m, 1H). (LC-MS method 1, tR =1.19 min m/z 262 [M – butene + H]+, 318
[M+H]+, 335 [M+NH4]+, 652 [2M+NH4]+. HRMS (ESI) calcd for C18H23NO4Na [M+Na]+: 340.1525,
found: 340.1534.
Chromatographic resolution of cis-2-benzyl 3-tert-butyl 3-azabicyclo[3.1.0]hexane-2,3-
dicarboxylate (rac-cis-25). A 203.3 g batch of the racemic cis-benzyl ester rac-cis-25 was resolved.
Preparative method: IC 250 x 76 mm, 20 um; 100% dichloromethane; 270 mL/min at 25°C; detection at
250 nm. Analytical method: IC (IC00CE-LB014, 250 x 4.6 mm, 5 um); 100% dichloromethane; 1
mL/min at 25°C, detection at 250 nm. First eluting compound was the (–)-enantiomer, the second
eluting compound the (+)-enantiomer.
ACS Paragon Plus Environment

Page 17 of 34
The Journal of Organic Chemistry
(–)-Cis-(1R,2S,5S)-2-benzyl 3-tert-butyl 3-azabicyclo[3.1.0]hexane-2,3-dicarboxylate ((S)-cis-25).
104.0 g (51% recovery) of a colorless viscous oil were obtained. The compound partially crystallized
after several months. Mp 42–45°C. Purity (HPLC method 1, tR = 5.53 min) 90.0%. Enantiopurity (HPLC
1
2
3
4
5
6
7
8
1
method 2, tR = 4.26 min) > 99.9%ee. H-NMR (400 MHz, DMSO-d6) δ 7.38–7.31 (m, 5H), 5.19–5.06
9
(m, 2H), 4.35–4.34 (m, 1H), 3.49–3.43 (m, 1H), 3.39-3.36 (m, 2H, partially overlaid by H2O signal),
1.93–1.85 (m, 1H), 1.67–1.61 (m, 1H), 1.35 and 1.23 (2s, 2 rotamers 3:5, 9H), 0.66–0.59 (m, 1H), 0.51–
0.46 (m, 1H). [α]24D = –106 (c = 0.5, CHCl3).
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
(+)-Cis-(1S,2R,5R)-2-benzyl 3-tert-butyl 3-azabicyclo[3.1.0]hexane-2,3-dicarboxylate ((R)-cis-25).
99.3 g (49% recovery) of a colorless viscous oil were obtained. Purity (HPLC method 1, tR =5.53 min)
1
98.5%. Enantiopurity (HPLC method 2, tR =5.77 min) >99.9%ee. H-NMR (400 MHz, DMSO-d6) δ
7.38–7.31 (m, 5H), 5.19–5.06 (m, 2H), 4.35–4.34 (m, 1H), 3.49–3.40 (m, 1H), 3.37-3.36 (m, 2H,
partially overlaid by H2O signal), 1.93–1.85 (m, 1H), 1.67–1.61 (m, 1H), 1.35 and 1.23 (2s, 2 rotamers
3:5, 9H), 0.66–0.60 (m, 1H), 0.51–0.46 (m, 1H). HRMS (ESI) calcd for C18H23NO4Na [M+Na]+:
340.1525, found: 340.1534. Anal. Calcd for C18H23NO4: C, 68.12; H, 7.30; N, 4.41. Found: C, 67.27; H,
7.08; N, 4.26. [α]24D = +108 (c = 0.6, CHCl3).
Trans-2-benzyl 3-tert-butyl 3-azabicyclo[3.1.0]hexane-2,3-dicarboxylate (rac-trans-25) in batch
mode. Preparation of the LDA solution: In a 10 mL round bottom flask under argon containing 4.55 mL
(6.78 mmol, 2.2 equiv.) of a 1.49 M diisopropylamine solution in THF were added at 0°C 4.14 mL (6.16
mmol, 2.0 equiv.) of a 1.49 M BuLi solution in hexane (freshly titrated). After 5 min this solution was
cooled in a dry ice /acetone bath to –36°C. The LDA solution was then quickly transferred via cannula
to a solution of 0.98 g (3.08 mmol) racemic cis-benzyl ester rac-cis-25 in 15.4 mL THF, which was
cooled to –32°C. The internal temperature of the reaction mixture immediately increased from –32 to –
21°C and then decreased again. After 60 s of intense stirring the reaction mixture (–23°C) was quenched
with 25 mL sat. aq, NaHCO3 solution and warmed to r.t. After addition of 30 mL of water the mixture
was extracted three times with EtOAc. The combined organic phases were washed with brine, dried
over sodium sulfate and concentrated to yield 0.97 g of an oil. This was purified on 100 g silica gel with
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 18 of 34
EtOAc/heptanes to yield 36 mg (3.7% yield) of the starting material and 0.69 g (71% yield) product rac-
trans-25 as a colorless viscous oil, that crystallized after a few days upon standing. Mp 49–53°C. Purity
1
2
3
4
5
1
1
(HPLC method 1, tR = 5.67 min) 89.5% (looked pure in H-NMR). H-NMR (600 MHz, DMSO-d6) δ
7.38–7.32 (m, 5H), 5.22–5.10 (m, 2H), 4.25 and 4.19 (2s, 1H, 2 rotamers), 3.43–3.39 (m, 2H), 1.64–
1.58 (m, 2H), 1.35 and 1.22 (2s, 2 rotamers, 9H), 0.76–0.73 (m, 1H), 0.26–0.26 (m, 1H). 13C-NMR (151
MHz, DMSO-d6) δ 171.6, 171.4, 154.1, 153.7, 135.8, 128.5, 128.4, 128.2, 128.1, 127.7, 79.3, 79.1,
66.1, 66.0, 61.2, 60.9, 48.3, 48.3, 28.0, 27.8, 19.6, 18.7, 15.3, 14.5, 9.0, 8.8 (rotamers). LC-MS method
1, tR =1.24 min m/z 262 [M – butene + H]+, 318 [M+H]+, 335 [M+NH4]+, 652 [2M+NH4]+. HRMS (ESI)
calcd for C18H23NO4Na [M+Na]+: 340.1525, found: 340.1536.
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
(–)-Trans-(1S,2S,5R)-2-benzyl 3-tert-butyl 3-azabicyclo[3.1.0]hexane-2,3-dicarboxylate ((S)-
trans-25) in flow mode. The flow reactor was configured using a combination of the R2C pump module
and R4 chiller module. A 10 mL and a 5-mL PFA tubing (internal diameter 1 mm) reactor were
installed in the R4 module, along with an 8-bar ceramic back-pressure regulator fitted in-line between
the reactor outflow and the collection valve. The solvent bottle was filled with anhydrous THF, and the
reagent stock bottles were filled with n-BuLi in hexane (1.44 M), diisopropylamine in THF (1.49 M)
and substrate (R)-cis-25 in THF (1.00 M), respectively (Scheme 7). Pump 1 delivered 7.19 mL/min n-
BuLi solution (2.0 equiv.) and pump 2 delivered 7.64 mL/min diisopropylamine solution (2.2 equiv.)
through the 10 mL reactor with a residency time of 40 s at –20°C. The outlet flow of this LDA solution
(14.83 mL/min) was combined with the substrate flow (1.0 equiv.) delivered from pump 3 with 5.17
mL/min through the 5 mL reactor to give a residency time of 15 s at –15°C. The second Y-piece
(between the two reactors) was immersed into an external cooling bath and kept at –20°C. The outlet
stream was collected for 18 min (365 mL) and poured into 400 mL of a well stirred aq. 10% ice cold
NH4Cl solution. This was then extracted three times with 150 mL MTBE. The combined organic phases
were washed with brine, dried over sodium sulfate and concentrated to give 27.1 g of an orange oil. The
cis/trans ratio of the two benzylesters was 3:97. The crude product was purified on 2 kg silica gel
(heptanes/EtOAc) to yield 16.83 g (56% yield) of the (–)-trans benzyl ester (S)-trans-25 as a pale
ACS Paragon Plus Environment

Page 19 of 34
The Journal of Organic Chemistry
1
yellow, viscous oil. Purity (HPLC method 1, tR = 5.68 min) 98.7%. H-NMR (400 MHz, DMSO-d6) δ
7.38–7.30 (m, 5H), 5.23–5.09 (m, 2H), 4.25 and 4.19 (2s, 2 rotamers, ratio 2:3, 1H), 3.44–3.39 (m, 2H),
1.65–1.54 (m, 2H), 1.35 and 1.22 (2s, 2 rotamers, ratio 2:3, 9H), 0.77–0.71 (m, 1H), 0.25–0.20 (m, 1H).
[α]24D = –83.4 (c = 0.5, CHCl3).
1
2
3
4
5
6
7
8
9
(+)-Trans-(1R,2R,5S)-2-benzyl 3-tert-butyl 3-azabicyclo[3.1.0]hexane-2,3-dicarboxylate ((R)-
trans-25) in flow mode. The same procedure was applied as for (S)-trans-25 but with (S)-cis-25 as the
starting material. However, the first reactor clogged and had to be cleaned. As a consequence, the yield
was lower with 46%. (R)-trans-25 (12.0 g ) was obtained as a pale yellow, viscous oil. Purity (HPLC
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
method 1, tR = 5.68 min) 95.5%. H-NMR (400 MHz, DMSO-d6) δ 7.38–7.30 (m, 5H), 5.23–5.09 (m,
2H), 4.25 and 4.19 (2s, 2 rotamers, ratio 2:3, 1H), 3.44–3.36 (m, 2H), 1.65–1.54 (m, 2H), 1.35 and 1.22
(2s, 2 rotamers, ratio 2:3, 9H), 0.77–0.71 (m, 1H), 0.26–0.26 (m, 1H). [α]24D = +75.5 (c = 0.6, CHCl3).
Determination of the cis/trans ratio of cis/trans-30 via benzyl ester formation. General procedure.
To 50 mg (ca 0.2 mmol) acid 30 or 0.5 mL of the reaction mixture (concentrated) were added 0.2 mL
BnBr, 0.25 g potassium carbonate and 3 mL acetone. After stirring for at least 2 h at r.t. the cis/trans
ratio of the benzyl ester 25 was determined with HPLC method 1.
Cis-3-(tert-butoxycarbonyl)-3-azabicyclo[3.1.0]hexane-2-carboxylic acid (rac-cis-30) – method A.
Equipment: 400 mL reaction flask with 4 necks and mechanical stirrer, internal thermometer and argon
inlet. To a solution of 12 g (65.5 mmol) tert-butyl 3-azabicyclo[3.1.0]hexane-3-carboxylate (24) and
17.2 g (82 mmol, 1.25 equiv.) 27 in 262 mL dry THF, with the temperature kept below –60°C was
added 56.1 mL (79 mmol, 1.2 equiv.) sec-BuLi solution (1.4 M in cyclohexane) over 12 min. The
progress of deprotonation was checked with GC (quenching of a sample with excess benzaldehyde).
After 5 h at –60°C the reaction mixture was cooled to –68°C and dry CO2 gas was sparged through the
reaction mixture. The internal temperature went up to –45°C within 1-2 min. To the reaction mixture
was added ca 150 mL of water (pH > 10). Most of the THF was distilled off and the aq. phase extracted
twice with 150 mL MTBE. To the aq. phase ca 300 mL 25% aq. KHSO4 solution and ice were added
until the pH was < 3, then it was extracted three times with 200 mL MTBE. The combined organic
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 20 of 34
phases were washed with 200 mL brine, dried over sodium sulfate, filtered and concentrated. Some
dichloromethane was added and the product completely dried on vacuum over night to yield 11.6 g rac-
cis-30 (78%) as an almost colorless resin that solidified during 1 week (a 3 L round bottom flask was
used due to strong foaming on drying). The purity according to the 1H-NMR (CDCl3)was estimated as >
1
2
3
4
5
6
7
8
9
1
95%. No trans acid could be detected (as its benzylesters). Mp 119–124°C. H-NMR (400 MHz,
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
DMSO-d6) δ 12.4 (br, 1H), 4.18–4.14 (m, 1H), 3.47–3.31 (m, ?H, together with H2O signal), 1.88–1.81
(m, 1H), 1.65–1.58 (m, 1H), 1.35 and 1.30 (2s, 2 rotamers ca 2:3, 9H), 0.64–0.58 (m, 1H), 0.52–0.49
(m, 1H). 1H-NMR (400 MHz, CDCl3) δ 8.67 (br, 1H), 4.43–4.35 (m, 1H), 3.63–3.55 (m, 2H), 1.93–1.87
13
(m, 1H), 1.68–1.62 (m, 1H), 1.44 and 1.40 (2s, 2 rotamers ca 1:2, 9H), 0.79–0.68 (m, 2H). C-NMR
(100 MHz, CDCl3) δ 177.5, 80.7, 80.6, 60.5, 60.4, 50.0, 49.7, 28.3, 28.2, 20.6, 19.7, 16.7, 16.1, 8.64,
8.56 (rotamers). HRMS (ESI) calcd for C11H16NO4 [M–H]–: 226.1079, found: 226.1087. Anal. Calcd for
C11H17NO4: C, 58.14; H, 7.54; N, 6.16. Found: C, 58.24; H, 7.38; N, 6.15.
Cis-3-(tert-butoxycarbonyl)-3-azabicyclo[3.1.0]hexane-2-carboxylic acid (rac-cis-30) – method B.
Equal to method A but instead of 27 12.18 g (105 mmol, 1.6 equiv.) TMEDA was used. Rac-cis-30
1
(12.1 g, 81% yield) as an almost colorless resin was obtained. According to H-NMR the compound
contained 15% MTBE. Benzylation of a small sample revealed the existence of 2% of the trans acid and
22% of a diacid. Based on these data the purity was estimated to be 60%. 1H-NMR (400 MHz, DMSO-
d6) δ 4.18–4.14 (m, 1H), 3.48–3.27 (m, ?H, overlaid by a broad signal from 4.0-3.2 ppm), 1.88–1.82 (m,
1H), 1.65–1.53 (m, 1H), 1.35 and 1.30 (2s, 2 rotamers ca 2:3, 9H), 0.65–0.58 (m, 1H), 0.52–0.49 (m,
1H).
Cis-3-(tert-butoxycarbonyl)-3-azabicyclo[3.1.0]hexane-2-carboxylic acid (rac-cis-30) – diamine
free deprotonation. Equipment: 50 mL three neck flask with magnetic stirrer, internal thermometer and
argon inlet. To a solution of 0.5 g (2.73 mmol) tert-butyl 3-azabicyclo[3.1.0]hexane-3-carboxylate (24)
in 19 mL dry THF at –40°C were added dropwise 2.53 mL (3.55 mmol, 1.3 equiv.) sec-BuLi solution
(1.4 M in cyclohexane). The bright yellow reaction mixture was stirred between –45 and –38°C. The
progress of deprotonation was monitored with GC (quenching of a sample with excess benzaldehyde)
ACS Paragon Plus Environment

Page 21 of 34
The Journal of Organic Chemistry
after 2 h and 5 h (in both cases mostly starting material was detected). Dry CO2 gas was then sparged
through the reaction mixture. The internal temperature went up to –36°C, then the reaction mixture was
allowed to warm to r.t., when its consistency became jelly-like. The work up was identical as described
for method A. 486 mg starting material were recovered from the basic extraction and 23 mg of a
colorless resin were obtained after extraction at pH < 3 and benzylated for HPLC analysis. Neither cis-
nor trans benzylester were detectable, the major signal corresponded to the bis-benzylated diester (26).
Salt screen. Resolution of racemic acids via formation of diastereomeric salts. General
procedure. 192 combinations consisting of 16 chiral bases and 12 solvents were applied. For each
combination 0.05 mmol racemic acid and a stoichiometric amount of a chiral base were used in a
concentration range starting from maximum 250 down to 25 mM. The following bases were used: (S)-(–
)-1-phenylethylamine, (–)-ephedrine, (+)-pseudoephedrine, (–)-norephedrine, (R)-(–)-2-amino-1-
butanol, (R)-(–)-phenylglycinol, (S)-(+)-1,2,3,4-tetrahydro-1-naphtylamine, brucine, strychnine, (+)-
cinchonin, (–)-cinchonidine, (+)-quinidine, quinine, L-lysine, (+)-dehydroabietylamine 60%, (1R,2R)-(–
)-1,2-diaminocyclohexane. The following solvents were used: water, EtOH/water 50:50, EtOH/water
96:4, EtOH abs., MeOH, 2-PrOH, acetone, 2-butanone, EtOAc, EtOAc sat. with water, acetone/CHCl3
1:1, acetonitrile. The chiral bases as well as the racemic acid were distributed as freshly prepared
solutions in MeOH or dichloromethane into 192 x 2 mL HPLC vials. After removal of the solvent the
12 different solvents were added and the vials closed with a lid. After heating at 80°C for 1-2 h (manual
shaking from time to time) the vials containing clear solutions were placed on a second plate and
allowed to cool to r.t. To the remaining vials were added solvents in 0.2 to 0.5 mL portions and the
heating and picking steps repeated until the vials were full. After a day, sometimes longer, crystals were
separated from the mother liquor (filtered or decanted, no washing) and dried in vacuo. In a second
round the vials were stored at 6°C for a few days and new crystals collected. The yields were calculated
based on the weight assuming a 1:1 salt was formed and the ee determined via HPLC (the acid was
liberated before).
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

The Journal of Organic Chemistry
Page 22 of 34
Salt screen of rac-cis-30. The best results were obtained with (a) (R)-(–)-phenylglycinol in EtOAc (c
= 250 mM), crystals collected at 6°C, yield 33%, enantiopurity (HPLC method 3, tR = 16.65 min)
71%ee. (b) brucine in water (c = 14 mM), crystals collected at 24°C, yield 32%, enantiopurity (HPLC
method 3, tR = 17.56 min) 71%ee. (c) (–)-cinchonidine in acetonitrile (c = 250 mM), crystals collected at
6°C, yield 46%, enantiopurity (HPLC method 3, tR = 16.65 min) 78%ee. (d) (+)-dehydroabietylamine in
2-butanone (c = 63 mM), crystals collected at 24°C, yield 53%, enantiopurity (HPLC method 3, tR =
17.56 min) 60%ee. (e) (+)-dehydroabietylamine in EtOAc sat. with water (c = 31 mM), crystals
collected at 24°C, yield 32%, enantiopurity (HPLC method 3, tR = 17.56 min) 99%ee. Condition (e) was
used to obtain the (–)-enantiomer followed by condition (c) for the (+)-enantiomer.
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
Salt screen of rac-trans-30. The best results were obtained with (a) (R)-(–)-phenylglycinol in EtOAc
(c = 55 mM), crystals collected at 6°C, yield 56%, enantiopurity (HPLC method 4, tR = 7.06 min)
74%ee. (b) (R)-(–)-phenylglycinol in acetone/CHCl3 1:1 (c = 250 mM), crystals collected at 6°C, yield
32%, enantiopurity (HPLC method 4, tR = 7.06 min) 93%ee. (c) (1R,2R)-(–)-1,2-diaminocyclohexane in
2-butanone (c = 250 mM), crystals collected at 24°C, yield 29%, enantiopurity (HPLC method 4, tR =
7.06 min) 89%ee. Condition (a) was used to obtain the (+)-enantiomer followed by condition (a) using
(S)-(+)-phenylglycinol for the (–)-enantiomer.
(–)-Cis-(1R,2S,5S)-3-(tert-butoxycarbonyl)-3-azabicyclo[3.1.0]hexane-2-carboxylic acid ((S)-cis-
30) via diastereomeric salt formation. To a well stirred solution of 121.5 g (0.54 mol) rac-cis-30 in 20
L EtOAc (sat. with water) was added a solution of 164 g (0.52 mol) 90% pure (+)-dehydroabietylamine
in 4 L EtOAc (sat. with water) at r.t. After the addition a precipitation formed. The suspension was
heated to reflux for one hour. The clear solution was cooled to 20°C over 3 hours. A pale brown
suspension was obtained that was stirred for another 3 hours. After filtration the filter cake was washed
twice with 2 L cold EtOAc and dried 18 h at 10 mbar and 30°C. 86 g of the dehydroabietylamine salt
were obtained. Mp 213–219°C. Anal. Calcd for C31H48N2O4: C, 72.62; H, 9.44; N, 5.46. Found: C,
72.29; H, 9.06; N, 5.37. [α]23 = –33.3 (c = 0.50, MeOH). To a well stirred suspension of 95 g
D
dehydroabietylamine salt (combined with a trial batch) in 3 L deionized water were added 60 g
ACS Paragon Plus Environment

Page 23 of 34
The Journal of Organic Chemistry
potassium carbonate and 3 L MTBE. After 5 min the two clear phases were separated. The aqueous
phase was extracted twice with 1 L MTBE. The combined organic phases were extracted twice with 0.5
L water, dried over magnesium sulfate, filtered and concentrated to yield 70 g of a brown oil of
recovered dehydroabietylamine. To the combined aq. phases, while vigorously stirred, were added 3 L
EtOAc and 150 g citric acid. After 10 min the aq. phase was separated and extracted two more times
with 1 L EtOAc. The combined organic phases were washed twice with 1 L water, dried over
magnesium sulfate, filtered and concentrated at 45°C in vacuo. The resulting resin was triturated with
low boiling petroleum ether for one hour in order to crystallize. The solid was filtered off and dried at
50°C for 5 h at 1 mbar to yield 38.6 g (32% yield, based on 100 g rac-cis-30) of colorless, crystalline
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
1
(S)-cis-30. Mp 89–91°C. Enantiopurity (HPLC method 3, tR = 17.56 min) > 99%ee. H-NMR (600
MHz, DMSO-d6) δ 12.41 (s, 1H), 4.19–4.16 (m, 1H), 3.47–3.42 (m, 1H), 3.35 (d, J=10.1 Hz, 1H),
1.88–1.83 (m, 1H), 1.65–1.60 (m, 1H), 1.36 and 1.31 (2s, 2 rotamers ca 1:2, 9H), 0.64–0.60 (m, 1H),
0.53–0.50 (m, 1H). Anal. Calcd for C11H17NO4: C, 58.14; H, 7.54; N, 6.16. Found: C, 57.85; H, 7.09; N,
6.16. [α]20D = –121.7 (c = 1.0, CHCl3). The mother liquor of the first crystallization was concentrated to
give 207 g of a pale brown resin that were suspended in 3 L of water. After the addition of 130 g
potassium carbonate and 3 L MTBE, 75 g (+)-enantio enriched (R)-cis-30 were isolated in analogy to
the above procedure. Via hydrogenation of benzylester (S)-cis-25. (S)-cis-25 (6.37 g, 17.7 mmol) was
dissolved in 90 mL EtOAc and 0.94 g Pd/C 10% were added. The reaction mixture was hydrogenated
for 41 h at r.t. and atmospheric pressure. The catalyst was filtered off and the product solution extracted
with 150 mL sat. aq. NaHCO3 solution. The aq. phase was extracted with another 100 mL EtOAc, then
the pH was lowered by addition of citric acid and again extracted with 3 x 100 mL EtOAc. These three
EtOAc portions were combined, washed with brine, dried over sodium sulfate and concentrated to yield
3.61 g (90% yield) (S)-cis-30 as a colorless resin, that started to crystallized after one week. Mp 85–
1
88°C. H-NMR (400 MHz, DMSO-d6) δ 12.39 (s, 1H), 4.18–4.14 (m, 1H), 3.47–3.41 (m, 1H), 3.35–
3.33 (m, overlaid by H2O signal, ca 2H), 1.89–1.81 (m, 1H), 1.65–1.58 (m, 1H), 1.35 and 1.30 (2s, 2
rotamers, ratio 3:5, 9H), 0.64–0.58 (m, 1H), 0.52–0.49 (m, 1H). [α]24D = –128.7 (c = 0.53, CHCl3).
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 24 of 34
(+)-Cis-(1S,2R,5R)-3-(tert-butoxycarbonyl)-3-azabicyclo[3.1.0]hexane-2-carboxylic acid ((R)-cis-
30) via diastereomeric salt formation. To 73.0 g of the enantio enriched (R)-cis-30 in 2 L acetonitrile
were added 95 g (–)-cinchonidine. While heating to reflux, the suspension turned into a yellow solution.
The oil bath was removed and the solution allowed to cool to r.t. After a few hours the crystals were
filtered and washed with 0.5 L cold acetonitrile. The wet crystals were recrystallized in 1.3 L
acetonitrile to yield 103.7 g of the cinchonidine salt after drying at 45°C in vacuo. Mp 154.1–155.8°C.
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
Anal. Calcd for C30H39N3O5: C, 69.07; H, 7.54; N, 8.06. Found: C, 68.86; H, 7.35; N, 7.95. [α]23 = –
D
16.3 (c = 0.52, MeOH). The salt was suspended in a solution of 4 L deionized water and 63 g citric acid
and then extracted with 6 L MTBE. The MTBE phase was washed twice with 2 L water. The combined
aq. phases were extracted twice with 2 L MTBE. The combined organic phases were dried over
magnesium sulfate, filtered and concentrated to yield 44.3 g of a greenish resin which crystallized after
trituration in 500 mL warm heptanes. After standing overnight at r.t., the crystals were filtered, washed
twice with 100 mL cold heptanes and dried to yield 34.0 g crystalline (R)-cis-30 (35% yield, based on
1
100 g rac-cis-30). Mp 82–84°C. Enantiopurity (HPLC method 3, tR = 16.65 min) 97.4%ee. H-NMR
(400 MHz, DMSO-d6) δ 12.38 (s, 1H), 4.18–4.14 (m, 1H), 3.47–3.40 (m, 1H), 3.35–3.30 (m, 1H), 1.88–
1.81 (m, 1H), 1.64–1.57 (m, 1H), 1.34 and 1.29 (2s, 2 rotamers ca 1:2, 9H), 0.64–0.58 (m, 1H), 0.52–
0.49 (m, 1H). Anal. Calcd for C11H17NO4: C, 58.14; H, 7.54; N, 6.16. Found: C, 57.92; H, 7.06; N, 6.28.
[α]24D = +121.4 (c = 1.0, CHCl3).
(+)-cis-(1S,2R,5R)-3-(tert-butoxycarbonyl)-3-azabicyclo[3.1.0]hexane-2-carboxylic acid ((R)-cis-
30) via enantioselective deprotonation. Equipment: 100 mL reaction flask with 4 necks and
mechanical stirrer, internal thermometer and argon inlet. 5.4 g (23.1 mmol, 1.25 equiv.) (–)-sparteine
(distilled prior to use) were placed into the reaction flask and diluted with 230 mL dry diethylether. This
0.1 M solution was cooled below –70°C then 16.5 mL (23.06 mmol, 1.25 equiv.) of 1.4 M sec-BuLi
solution in cyclohexane were added within 5 min. After being stirred for 15 min below –70° a solution
of 3.38 g (18.45 mmol, 1 equiv.) 3-azabicyclo[3.1.0]hexane-3-carboxylic acid tert-butyl ester (24) in 15
mL diethylether were added to the reaction mixture with a syringe. The pale yellow reaction mixture
ACS Paragon Plus Environment

Page 25 of 34
The Journal of Organic Chemistry
was stirred under argon overnight. After 38 h at –78°C dry CO2 gas was sparged through the reaction
mixture for 3 min. The temperature immediately went up to –68°C. The reaction mixture was allowed to
warm to r.t., then 25 mL of sat. aq. Na2CO3 solution were added and the two phases separated (another
70 mL water had to be added in order to dissolve precipitated salts). The aq. phase was extracted with
60 mL MTBE and the org. phase discarded. The pH of the aq. phase was set < 3 with 25% aq. KHSO4
solution and then extracted three times with 60 mL MTBE. The combined org. phases were washed with
brine, dried over sodium sulfate, filtered and concentrated to yield 1.53 g (37% yield) product as a
colorless resin, that contained 11% of MTBE (according to 1H-NMR). Enantiopurity (HPLC method 3,
tR = 15.85 min) 73%ee. [α]24D = +68 (c = 1.5, CHCl3). 98% of the sparteine was recovered.
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
Trans-3-(tert-butoxycarbonyl)-3-azabicyclo[3.1.0]hexane-2-carboxylic
acid
(rac-trans-30).
Equipment: Buechi 30 L hastelloy reactor (CR30) with FlexyALR Systag control. Procedure: The
reactor was inertized with argon. To a cooled solution of tert-butyl 3-azabicyclo[3.1.0]hexane-3-
carboxylate (24) (600 g, 3.27 mol) and 3,7-dipropyl-3,7-diazabicyclo[3.3.1]nonane (27) (867 g, 4.12
mol) in 13 L THF at –80°C was pumped sec-BuLi (1.4 M in cyclohexane, 2.88 L, 4.03 mol) within 45
min allowing the internal temperature not to exceed –75°C. After the complete addition the pump was
purged with 0.5 L THF and the reaction mixture was warmed to –57°C within 30 min and stirred at this
temperature. The color turned from bright yellow to dark orange/brown during this time. IPC samples
were taken as follows: With a pipet that was cooled in the reaction mixture a 1 mL aliquot was taken
and poured into 1.5 mL methyl iodide. After 2 min this mixture was extracted with 10 mL MTBE and 2
mL water, the organic phase was filtered through a plug of sodium sulfate and analyzed by GC. After 5
h the ratio of starting material to methylated starting material was 3:7 (the same ratio was already
observed after 4 h). Therefore, a second portion of sec-BuLi (0.320 L, 0.45 mol) was added. However,
one hour later the ratio did not change. A stream of dry CO2 gas was sparged through the reaction
mixture. After the internal temperature went up from –57°C to –42°C within a few minutes, the CO2
stream was switched off (the jacket temperature was kept constant at –60°C) and the internal
temperature allowed to decrease to –47°C. The CO2 was switched on again followed by a second
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 26 of 34
exotherm. At –40°C the CO2 was switched off and on again. No exotherm was observed anymore, the
reaction was assumed to be complete. The reaction mixture was purged with argon for 10 min and kept
at –50°C overnight (not necessary). In order to determine how much excess LDA was necessary, the
excess CO2 in the reactor was determined via titration. For this, an aliquot of 5 mL was taken out of the
reactor and placed into a 25 mL round bottom flask with a magnetic stirrbar (prior heated and cooled
under vacuum, then filled with argon). Fluorene as the indicator (ca 5 mg) was added to the pale yellow
solution, which was then titrated with the same batch n-BuLi solution, that was later used for the
formation of LDA. 0.36 mL were consumed until the color turned orange. The total amount of BuLi
solution used in the next step therefore was: 2.05 L (1 equiv.) + 1.31 L (for quenching excess CO2) +
0.672 L (20% safety margin) -> total of 4.09 L (=2.78 kg). Diisopropylamine (1.027 L, 7.20 mol) was
added to the reaction mixture at –57°C. After 10 min n-BuLi solution (1.6 M in hexane, 4.09 L, 6.55
mol) was added during 30 min. After 15 min the reaction mixture was slowly warmed to –10°C and
stirred at –7°C for 1.5 h. The reaction mixture was then quenched with 0.5 L water during 15 min
followed by the addition of 3.6 kg KHSO4 dissolved in 5 L water (almost saturated solution). The pH
dropped to 2-3. The reaction mixture was extracted with 1 x 5 L and 2 x 3 L MTBE. The combined
organic phases were washed twice with 3 L brine, dried over sodium sulfate (with addition of 1.2 mL
Octastat 5000 as an anti electrostatic agent) and filtered. The cis/trans ratio was determined to be 9:91
(as the benzylesters). The brown, clear solution was concentrated to give 780 g of a brown solid with
liquid parts in it.
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
Purification: The crude material was dissolved in 7 L EtOAc at 70°C. After being completely
dissolved, half of the solvent was distilled off (crystallization already started) and replaced by 4 L
heptanes. This mixture was cooled to r.t., then after 30 min to 0°C. After another hour, the crystals were
filtered off. The beige crystals were washed with 1 L cold EtOAc/heptanes 1:2 mixture, then dried for 6
h at 45°C in vacuo to yield 420 g (crop 1). The cis/trans ratio was determined as = 5:95. The mother
liquor was concentrated, dissolved in 0.6 L EtOAc and diluted with 0.5 L heptanes. An oil was formed,
that did not crystallize even after addition of 0.4 g seed crystals from the first crop. However, overnight
ACS Paragon Plus Environment

Page 27 of 34
The Journal of Organic Chemistry
crystals formed, that were filtered and washed. Then the 55 g (wet weight) were recrystallized from
EtOAc/heptanes to yield 28.9 g of beige powder (crop 2). cis/trans ratio was determined to be 8:92.
Crop 1 and 2 were combined and recrystallized from 5.5 L hot EtOAc (stirred at 70°), then 2 L were
distilled off at 200 mbar. When the crystallization started, 1 L heptanes was added and the suspension
slowly cooled to r.t. overnight, while stirring. The suspension was stirred for 2 h at 5°C and then
filtered. The filter cake was washed with 1 L ice cold EtOAc/heptanes 2:1. The pale beige crystals were
dried in vacuo at 45°C for 3 h to yield 389 g (51%) of rac-trans-30 with a cis/trans ratio of 1.6 : 98.4.
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
1
Mp 161–163°C. H-NMR (400 MHz, CDCl3) δ 10.12 (br, 1H), 4.41 and 4.30 (2s, 1H, 2 rotamers),
3.63–3.50 (m, 2H), 1.78–1.67 (m, 1H), 1.58–1.52 (m, 1H), 1.45 and 1.41 (2s, 2 rotamers ca 1:1, 9H),
0.81–0.76 (m, 1H), 0.38–0.30 (m, 1H). 13C-NMR (100 MHz, CDCl3) δ 178.0, 176.7, 155.9, 154.8, 80.8,
80.5, 61.3, 61.0, 48.6, 48.3, 28.4, 28.3, 19.8, 18.6, 15.5, 14.9, 9.2, 9.0 (rotamers). HRMS (ESI) calcd for
C11H16NO4 [M–H]–: 226.1079, found: 226.1088. Anal. Calcd for C11H17NO4: C, 58.14; H, 7.54; N, 6.16.
Found: C, 58.11; H, 7.32; N, 6.06.
Benzylation: All mother liquors were combined and evaporated to yield 301 g of a dark brown honey,
which were dissolved in 3 L acetone, then 456 g potassium carbonate (3.3 mol) and 514 g benzyl
bromide (3 mol) were added and the reaction mixture was stirred at r.t. overnight. Celite (300 g) was
added and the reaction mixture filtered. The solvent was removed in vacuo and replaced by 2.5 L THF
then 1 L 7N ammonia in MeOH was added and the reaction mixture stirred overnight. According to
HPLC all excess BnBr was converted to benzylamine. The THF was distilled off and 3 L EtOAc and 3
L water were added. After mixing, the phases were separated and the aq. phase extracted two more
times with 1 L EtOAc. The combined organic phases were washed twice with 2.5 L 20% aq. KHSO4 aq.
solution. (After the first phase separation, some dark oil was removed with the water phase too, pH of
the water phase was < 2). The brown organic phase was washed with 2.5 L brine, dried over sodium
sulfate, concentrated, re-dissolved in 1 L EtOAc/heptanes 1:1 and filtered through a 5 cm thick plug of
silica gel (diameter 11 cm) and rinsed with 3 L EtOAc/heptanes 1:2. After removal of the solvent 321.6
g of a brown orange oil was obtained . This material was purified on 2 kg silica gel with
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 28 of 34
EtOAc/heptanes in 3 portions. The first eluting fractions resulted in 72.7 g (6.5% yield) trans benzyl
ester rac-36 with a purity of 93.5% at 215 nm. Later another 122.0 g of a yellow/orange, viscous oil
eluted. According to HPLC it contained 33% of the cis-benzyl ester rac-cis-25 and 61% of dibenzyl
ester 26.
1
2
3
4
5
6
7
8
9
Recycling of diamine 27: The aq. phase of the first reaction work up was filtered (ca 1.5 L of wet
solids were removed). To the filtrate were added 5 L ice followed by 5 L 30% aq. NaOH solution. The
milky mixture was extracted with 3 x 4 L MTBE. The combined organic phases were washed with 3 L
brine, dried over sodium sulfate and concentrated to yield 857 g crude diamine as an orange brown oil.
After distillation at 0.1 mbar and 68-80°C, 808 g 27 were obtained as a clear liquid with a GC purity of
100% (93% recovery).
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
(+)-Trans-(1R,2R,5S)-3-(tert-butoxycarbonyl)-3-azabicyclo[3.1.0]hexane-2-carboxylic acid ((R)-
trans-30) via diastereomeric salt formation. Rac-trans-30 (180 g, 0.79 mol) and (R)-2-phenylglycinol
(109 g, 0.79 mol) were placed in a 20 L evaporation flask of a Buechi Rotavapor R-220 and suspended
in 10 L EtOAc containing 0.5% water. The mixture was heated for 10 min at 70°C when a clear solution
was obtained. The heating of the water bath was switched off and the solution was allowed to cool to r.t.
while gently stirring. After 6.5 h (internal temperature was 27°C) the suspension was filtered off and the
pale brown crystals dried under vacuum and re-crystallized two more times at 70°C in each 10 L EtOAc
to yield 109 g of the (R)-2-phenylglycinol salt as a colorless solid. Mp 154.0–156.6°C. Anal. Calcd for
C19H28N2O5: C, 62.62; H, 7.74; N, 7.69. Found: C, 62.44; H, 7.54; N, 7.54. [α]23 = +54.8 (c = 0.50,
D
MeOH). This salt was treated with 2 L of cold 10% aq. KHSO4 solution and 1.5 L EtOAc. The aq. phase
was extracted with 1 L EtOAc. The combined organic phases were washed with 0.5 L 10% aq. KHSO4
solution, twice with 1 L brine, dried over magnesium sulfate and concentrated to dryness. (R)-trans-30
(53 g, 29% yield) as a colorless solid was obtained. Mp 189–191°C (decomp.). Enantiopurity (HPLC
1
method 4, tR = 7.06 min) > 99%ee. H-NMR (600 MHz, DMSO-d6) δ 4.09 and 4.05 (2s, 1H, 2
rotamers), 3.41–3.35 (m, ca 2H, within H2O signal), 1.62–1.54 (m, 2H), 1.37 and 1.32 (2s, 2 rotamers ca
4:6, 9H), 0.74–0.70 (m, 1H), 0.23–0.17 (m, 1H). [α]24 = +122 (c = 0.8, CHCl3). The mother liquors
D
ACS Paragon Plus Environment

Page 29 of 34
The Journal of Organic Chemistry
were combined and the enantio enriched acid liberated as described for the pure (+)-enantiomer, 104 g
of a brown solid were obtained.
1
2
3
4
5
6
7
8
(–)-Trans-(1S,2S,5R)-3-(tert-butoxycarbonyl)-3-azabicyclo[3.1.0]hexane-2-carboxylic acid ((S)-
trans-30) via diastereomeric salt formation. The 104 g of the above described (–)-enantio enriched
acid plus 4 g of the same from a different batch (total 0.475 mol) and 65.2 g (0.475 mol) (S)-2-
phenylglycinol were placed in a 20 L evaporation flask of a Buechi Rotavapor R-220 and suspended in
8 L EtOAc containing 0.5% water. The mixture was heated for 10 min at 70°C when a clear yellow
solution was obtained. The heating of the water bath was switched off and the solution was allowed to
cool to r.t. while gently stirring. After the weekend (internal temperature was 21°C) the suspension was
filtered off and the pale brown crystals dried and re-crystallized one more time at 70°C in 8 L EtOAc
containing 0.5% water to yield 59 g of the (S)-2-phenylglycinol salt as a colorless solid. The salt was
treated with 1 L of cold 10% aq. KHSO4 solution and 1 L EtOAc. The aq. phase was extracted with 0.5
L EtOAc. The combined organic phases were washed with 0.3 L 10% aq. KHSO4 solution, twice with
0.5 L brine, dried over magnesium sulfate and concentrated to dryness. (S)-trans-30 (28 g, 15% yield
based on 180 g rac-trans-30) as a colorless solid was obtained. Mp 187–190°C (decomp.).
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
1
Enantiopurity (HPLC method 4, tR = 6.27 min) 95.6%ee. H-NMR (600 MHz, DMSO-d6) δ 4.13 and
4.09 (2s, 1H, 2 rotamers ca 2:3), 3.45–3.42 (m, ca 2H, within H2O signal), 1.66–1.58 (m, 2H), 1.41 and
1.36 (2s, 2 rotamers ca 4:6, 9H), 0.78–0.74 (m, 1H), 0.26–0.21 (m, 1H). [α]23 = –121 (c = 0.86,
D
CHCl3). Via hydrogenation of benzylester (S)-trans-25. (S)-trans-25 (5.0 g, 15.8 mmol) was dissolved
in 80 mL EtOAc and 0.5 g Pd/C 10% were added. The reaction mixture was hydrogenated for 9 h at r.t.
and atmospheric pressure. The catalyst was filtered off and washed with MeOH, because product
precipitated out. The solvent was removed and the crude product crystallized from hot EtOAc. 2.88 g
1
(S)-trans-30 (80% yield) were obtained. Mp 182–184°C (decomp.). H-NMR (400 MHz, DMSO-d6) δ
12.74 (br 1H), 4.08 and 4.04 (2s, 2 rotamers, ratio 5:7, 1H), 3.43–3.35 (m, overlaid by H2O signal, ca
2H), 1.62–1.50 (m, 2H), 1.36 and 1.31 (2s, 2 rotamers ca 4:5, 9H), 0.74–0.68 (m, 1H), 0.22–0.15 (m,
1H). [α]24D = –143.3 (c = 0.51, CHCl3).
ACS Paragon Plus Environment