Toward a Large-Scale Approach to Milnacipran Analogues Using Diazo Compounds in Flow Chemistry

Article abstract of DOI:10.1021/acs.oprd.5b00308

The safe use of diazo reagents for the preparation of a key structure in the synthesis of milnacipran analogues is described herein. Using continuous flow technology, the diazo reagent is synthesized, purified, dried, and subsequently used in semi-batch mode for an intramolecular cyclopropanation. Side products formed in the reaction are isolated and rationalized to optimize the process. Different separation techniques in flow are compared with regard to their ability to produce pure and dry diazo reagents. The studies yield a scalable process to a key intermediate in the syntheses of milnacipran and its possible substituted analogues.

Full text of DOI:10.1021/acs.oprd.5b00308

Subscriber access provided by UNIV OF ARIZONA
Full Paper
Towards a large scale approach to Milnacipran
analogues using diazo compounds in flow chemistry
Simon Müller, Aurelien Murat, Paul Hellier, and Thomas Wirth
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.5b00308 • Publication Date (Web): 29 Oct 2015
Downloaded from http://pubs.acs.org on October 30, 2015
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 18
Organic Process Research & Development
1
2
3
4
Towards a large scale approach to Milnacipran analogues using diazo compounds in flow chemistry
5
6
Simon T. R. Müller,1 Aurélien Murat,2 Paul Hellier,3 Thomas Wirth1,*
7
8
9
1 School of Chemistry, Cardiff University, Park Place, Cardiff CF10 3AT (UK), 2 Institut de Recherche
Pierre Fabre, 81603 Gaillac (France), 3 Pierre Fabre Médicament, Parc Industriel de la Chartreuse, 81106
Castres (France)
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
ACS Paragon Plus Environment

Organic Process Research & Development
Page 2 of 18
1
2
3
4
Graphics for the Table of Contents:
5
6
7
8
9
diazo transfer
in flow
O
N2
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
+
O
Ar-SO2N3
no isolation
O
O
L/L separator
drying trap
aq. waste
Rh2L4
intramolecular cyclopropanation
in semi-batch
2
ACS Paragon Plus Environment

Page 3 of 18
Organic Process Research & Development
1
2
3
4
5
6
Abstract
7
8
9
The safe use of diazo reagents for the preparation of a key structure in the synthesis of milnacipran
analogues is described herein. Using continuous flow technology, the diazo reagent is synthesized,
purified, dried and subsequently used in semiꢀbatch mode for an intramolecular cyclopropanation. Side
products formed in the reaction are isolated and rationalized to optimize the process. Different separating
techniques in flow are compared in their ability to produce pure and dry diazo reagents. The studies are
yielding a scalable process to a key intermediate in the synthesis of milnacipran and possible substituted
analogues.
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: Cyclopropanation, Diazoꢀcompounds, Diazoꢀtranzfer, FlowꢀChemistry
3
ACS Paragon Plus Environment

Organic Process Research & Development
Page 4 of 18
1
2
3
4
Introduction
5
6
7
8
The 1S,2R enantiomer of milnacipran is called ‘levomilnacipran’ 1, which is a serotoninꢀnorepinephrine
reuptake inhibitor (SNRI) used for the treatment of depression and fibromyalgia.1 It is a more balanced
reuptake inhibitor of serotonin and norepinephrine than other SNRIs.2 Substituted milnacipran analogues
have shown promising pharmaceutical properties and are interesting target molecules in drug
development.3 Several routes have been identified for the preparation of levominacipran 1. One is the
preparation of the bicyclic lactone 4 via the reaction of phenylacetonitrile 2 with nonꢀracemic
epichlorohydrin 3. In this sequence, the epoxide 3 undergoes nucleophilic addition by phenylacetonitrile 2
to make the cyclopropane moiety. Treatment with sodium hydroxide transforms the nitrile group into a
carboxylate which undergoes lactonization to 4 under treatment with hydrochloric acid. This compound is
then treated with diethylamine and nꢀbutyllithium to ringꢀopen the lactone. Subsequently, an Appel
reaction is used to make azide 5 which is transformed into levomilnacipran 1 by reduction using
hydrogen.4 Hu and Doyle showed that lactone 4 can also be obtained via a rhodiumꢀcatalyzed
stereoselective intramolecular cyclopropanation using allyl phenyldiazoacetate with up to 95% ee.5
Recently, such an approach to 4 was achieved through iron catalysis with good stereoselectivities using a
spirobox ligand.6 A diazoꢀbased scaleꢀup approach to 4 would give rapid access to functionalized
milnacipran analogues. However, diazo compounds are challenging reagents for large scale applications
because of their thermal properties.7
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
NEt2
O
NH2 • HCl
1
Figure 1: Levomilnacipran
O
O
1. NaNH2
O
2. KOH
CN
+
Cl
3. HCl (12 N)
2
3
4
1. Et2NH, n-BuLi
THF, –78 °C
2. CBr4, PPh3, LiN3
NEt2
NEt2
O
O
10% Pd-C, H2
MeOH
N3
NH2
1
5
4
ACS Paragon Plus Environment

Page 5 of 18
Organic Process Research & Development
1
2
3
4
Scheme 1: Shuto synthesis of levomilnacipran 1
5
6
7
8
Continuous flow microreactor technology promises to give rise to new opportunites for the large scale use
of dangerous reagents.8 Heat control is improved due to the high surfaceꢀtoꢀvolume ratio of
microreactors.9 Furthermore, no large quantities of dangerous materials are allowed to accumulate.
Accordingly, several protocols for the preparation of highly energetic diazo reagents10 such as
diazomethane11 and ethyl diazoacetate12 using microstructured devices have been developed. Recently,
we disclosed a flexible continuous flow protocol for the formation, purification and use of aryl
diazoacetates which are versatile donor/acceptor substituted carbene precursors.13 Considering the
structural similarities to those, it was anticipated that the diazo transfer onto allyl phenylacetate 6 leading
to diazo allyl phenylacetate 7 could be telescoped in flow with a selective extraction of 7 in an organic
solvent, so that a subsequent intramolecular cyclopropanation to 4 is possible. This protocol should then
be easily scalable to provide multigram quantities of key lactone 4. Substituted analogues should be
accessible via this route as well. The use of a chiral catalyst could provide enantioselective reactions to
generate nonꢀracemic 4.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
SO2N3
O
O
N2
O
Rh2L4
O
AcHN
O
O
DBU
6
7
4
Scheme 2: Diazo approach towards lactone 4
Results
Safety & Feasibility test
Initially, the thermal properties for the reaction of 6 to 7 were investigated. DSC (differential scanning
calorimetry) data for diazo transfer reagent pꢀABSA had been obtained previously.13 Diazo allyl
phenylacetate 7 was found to have a decomposition onꢀset temperature of 77 °C which corresponds to a
TD24 value of 6 °C. The enthalpy of the decomposition corresponds to ꢁH = ꢀ182 kJ/mol which gives a
ꢁTad of 523 °C. C80 calorimetry for studying the heat released within a reaction was performed. The heat
generated within the reaction and in a decomposition of the reaction mixture was recorded. It was found
that the ꢁH of the reaction was ꢀ150 kJ/mol corresponding to a ꢁTad of 39 °C. Given the standard reaction
temperature of 25 °C, the heat released would provide an operating temperature of 64 °C. This
temperature would surpass the TD24 calculated for this reaction (52 °C). If the decomposition would take
place at this time, it would lead to another ꢁTad of 67 °C. At this point, a thermal runaway would be
inevitable as the operating temperature would be at 131 °C and therefore well above the boiling point of
the solvent, acetonitrile (b.p. 81 °C). In accordance to chemical risk classes defined by Stoessel et al.,14
the diazo transfer described would have to be classified as risk class 5, the highest risk class for chemical
reactions. This analysis shows that the move to a flow chemistry environment will dramatically reduce the
risks associated with this process.
5
ACS Paragon Plus Environment

Organic Process Research & Development
Page 6 of 18
1
2
3
4
5
6
7
8
9
After this analysis, twoꢀstep batch methods were explored to investigate the feasibility of a twoꢀstep
protocol without purification via column chromatography of diazo reagent 7. Furthermore, the solubility
of the rhodium catalysts in different solvents was studied to test the possibility of a completely continuous
process. The diazo transfer from 6 to 7 was left stirring at 25 °C until no starting material could be
observed via HPLC (24 h).15 The yield of this reaction was 87%, providing diazo reagent 7 in pure form
after workꢀup (HPLC analysis). For the twoꢀstep batch protocol, yields varied between 44% and 79%.
Reactions with rhodium acetate as catalyst provided quite similar yields in all four solvent mixtures
tested, however only chlorinated solvent mixtures with acetonitrile provided completely homogenous
solutions (Table 1, entries 3 and 4). Therefore, these solvent mixtures are most promising for a continuous
flow approach. Toluene in combination with acetonitrile provided the lowest yield of all examples (Table
1, entry 2). Dichloromethane with acetonitrile (Table 1, entry 4) required a longer reaction time until all
diazo reagent was consumed as the operating temperature was limited by the boiling point of the solvents.
Rhodium octanoate as catalyst was more active for the transformation of 7 to 4 providing the lactone in
79% HPLC yield over two steps (Table 1, entry 5). With these promising results in hand, continuous flow
conditions for both steps were investigated.
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 1: Two step batch optimization from 6 to 4
Entry
Catalyst
Rh2(OAc)4
Rh2(OAc)4
Rh2(OAc)4
Rh2(OAc)4
Solvent
nꢀC7H16
C6H5CH3:CH3CN 1:1
CHCl3:CH3CN 1:1
CH2Cl2:CH3CN 1:1
nꢀC7H16
Catalyst solubility
Insoluble
Traces of solid
Homogeneous
Homogeneous
Insoluble
T [°C] Time [h]
Yield [%]a
1
2
3
4
5
60
60
60
35
60
1
1
1
3.5
1
59
44
59
51
79
Rh2(C7H15COO)4
a Two step yield as determined by HPLC assay.
Diazo transfer in flow
The diazo transfer reaction in flow was monitored using infrared inꢀline analysis techniques. The setꢀup is
shown in Scheme 3. The diazo transfer onto 6 using commercially available pꢀacetamidobenzene
sulfonylazide (pꢀABSA) worked smoothly. In short reaction times (14 min), high yields (86%) were
obtained at 65 °C (Table 2, entry 5) which could be increased by a longer residence time (Table 2, entry
6). A further increase in temperature however did not result in any additional diazo formation (Table 2,
entry 7). Lower temperatures did result in slightly lower yields (Table 2, entries 1ꢀ4). It seems that the
residence time of the reaction has a larger influence than the operating temperature. Interestingly, the
diazo transfer onto allyl phenylacetate 6 was slightly more efficient than onto methyl phenylacetate.13
SO2N3
Data analysis
AcHN
(1.2 M) in MeCN
N2
O
O
in line
IR analysis
flow cell
O
7
O
8.2 mL
d = 0.8 mm
6 (1 M)
DBU (1.4 M) in MeCN
6
ACS Paragon Plus Environment

Page 7 of 18
Organic Process Research & Development
1
2
3
4
Scheme 3: Setꢀup for diazo transfer reaction optimization
5
6
Table 2: Diazo transfer optimization in flow
7
8
9
Entry
Reaction time [min]
Temperature [°C]
Yield [%]a
1
2
3
4
5
6
7
8
14
8
14
14
26
14
40
50
50
60
65
65
75
64
80
66
84
86
93b
86
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 determined by IR spectroscopy; b conversion 1HNMR
Intramolecular cyclopropanation in flow
The intramolecular cyclopropanation of the pure diazo reagent 7 was investigated under continuous flow
conditions. The setꢀup is illustrated in Scheme 4. Solvents could be superheated in this continuous setꢀup
(Table 3, entries 1 ꢀ 3, 7) due to the use of a back pressure regulator. Again, rhodium octanoate proved to
be a more efficient catalyst for the diazo decomposition than rhodium acetate (Table 3, entries 1 and 3).
Several reactions had only a limited solubility of the catalyst (Table 3, entries 1, 3, 5, 6) in the respective
solvent as the precipitation of the catalyst could be observed in the syringe. In the case of nꢀhexane this
led to a complete lack of reactivity (only starting material was recovered). To improve the solubility of
the catalyst, a mixture of dichloromethane and acetonitrile was used (Table 3, entry 2) which improved
the yield of the reaction slightly compared to pure dichloromethane (Table 3, entry 1). The best result was
obtained using toluene as solvent (Table 3, entry 4).
N2
O
10 min, 60 °C
O
0.1 ml/min
O
O
7
0.2 M in solvent
1 mL
d = 0.8 mm
bpr
2.67 bar
Catalyst
2.5 mol%
4
Scheme 4: Setꢀup for optimization of cyclization in flow
Table 3: Optimization of cyclization to 4 in flow
Solventa
CH2Cl2
CH2Cl2/CH3CN
9/1
Entry
1
Catalyst [2.5 mol%]
Yield [%]b
Rh2(OAc)4
27
2
Rh2(OAc)4
31
3
4
5
6
7
CH2Cl2
CH3CN
Toluene
nꢀC6H14
CH2Cl2
Rh2(C7H15COO)4
Rh2(C7H15COO)4
Rh2(C7H15COO)4
Rh2(C7H15COO)4
Fe(III)TPPCl
63
56
82
0
0
a all solvents were dried before use; b isolated yield
7
ACS Paragon Plus Environment

Organic Process Research & Development
Page 8 of 18
1
2
3
4
5
6
7
8
9
The best conditions for the oneꢀstep continuous flow protocol were used with the substituted diazo allyl
ester 8 for the intramolecular cyclopropanation. Cyclopropane 9 was obtained in 91% yield within just 10
minutes residence time in toluene. However, again not all the rhodium octanoate dissolved in the syringe
for the flow reaction which could prove problematic when scaling this reaction in a continuous flow setꢀ
up.
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
OMe
OMe
10 min, 60 °C
0.1 ml/min
O
O
O
O
N2
1 mL
d = 0.8 mm
O
Rh2(C7H15COO)4
O
8
9
0.1 M in toluene
91% yield
Scheme 5: Cyclization of substituted diazoester 8 in flow
Semi-batch and continuous two-step protocols
With the results obtained for each of the steps in batch and flow, a semiꢀbatch setꢀup was designed (setꢀup
A) with the formation of the diazoester in continuous flow, followed by an inꢀline liquid/liquid extraction
and the use of a stream of the diazo compound in the organic layer into a flask containing the rhodium
catalyst in the reaction solvent (Scheme 6). No significant accumulation of the diazo reagent was
observed in the flask (HPLC analysis), making this process inherently safe. The formation of the diazo
reagent was monitored using inline infrared analysis to verify that the diazo transfer reaction was stable
over a longer period of time. Directly after the infrared analysis an aqueous sodium nitrite solution was
added to quench residual azide moieties. The reaction mixture was extracted using a feed of nꢀheptane
into a glass mixer. The mixture was then separated inꢀline and subsequently, the organic layer was
pumped into a flask containing the rhodium catalyst in nꢀheptane.
NaNO2 in H2O
Data analysis
O
O
0.29 mL/min
S
Ar
N3
n-heptane
0.39 mL/min
1.2 M
O
O
organic layer
N2
O
mixing system LTF
0.41 mL/min
in line
IR analysis
flow cell
5.25 mL
d = 0.8 mm
0.3 ml/min
18 min res. time
65 °C
CO2R
O
1.0 M
6
4
7
L/L phase separation
Rh2L4 (1 mol%)
in n-heptane
65 °C
DBU 1.4 M in MeCN
aq layer,
waste
Scheme 6: Continuous flow into semiꢀbatch setꢀup A
For the extraction, a gravitational separating system was designed (Figure 2). A narrow bore glass column
was filled with water and then stabilized with the reaction stream running into the column. The organic
layer was pumped out of the top of the gravitational separator whereas the lower aqueous layer was
8
ACS Paragon Plus Environment

Page 9 of 18
Organic Process Research & Development
1
2
3
4
5
displaced from the separator via a siphon outlet. With this method, it was possible to have a stable
interface of aqueous and organic layer at midꢀheight in the column.
6
7
8
9
Inlet
Outlet
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
water/n-heptane
interface
aq. waste
Figure 2: Gravitational separator left: photo; right: Scheme of gravitational separator16
Using this setꢀup, lactone 4 was obtained in low yields (31% yield; Table 4, entry 1). Considering the
higher yields in the experiments developed before, this was surprising. The first step of the transformation
worked well as could be monitored via inꢀline infrared analysis during the reaction. To determine if the
liquid/liquid extraction was the source of the reduced yield of the cyclization, an alternative setꢀup with a
membrane based liquid/liquid extraction was used (Figure 3).
Figure 3: Membrane separator
In the setꢀup using the membrane separator (Scheme 7), smaller quantities of reagents could be used as
there was no need for the relatively large hold up volumes of the liquid/liquid extractor of setꢀup A.
Surprisingly, in setꢀup B rhodium acetate was not capable of transforming 6 into lactone 4. However,
rhodium octanoate now provided lactone 4 in 33% isolated yield.
9
ACS Paragon Plus Environment

Organic Process Research & Development
Page 10 of 18
1
2
3
4
5
6
NaNO2 in H2O
0.1 mL/min
Data analysis
O
O
O
O
S
aq layer,
waste
Ar
N3
n-heptane
0.138 mL/min
1.2 M
organic layer
7
4
8
9
N2
O
mixing system LTF
0.175 mL/min
in line
IR analysis
flow cell
2.21 mL
d = 0.8 mm
0.084 ml/min
26 min res. time
65 °C
L/L phase separator
CO2R
O
1.0 M
6
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
7
Rh2L4 (1 mol%)
in n-heptane
65 °C
DBU 1.4 M in MeCN
Scheme 7: Continuous flow into semiꢀbatch setꢀup B
Table 4: Results of the two step continuous flow into semiꢀbatch approach
Entry
Set-up
Catalyst (1 mol%)
Rh2(OAc)4
Rh2(OAc)4
Rh2(C7H15COO)4
Yield [%]
1
2
3
31a
0
A
B
B
33b
a HPLC yield; b isolated yield
As entry 3 provided the best results obtained to that point, it was briefly verified if substituted analogues
of 4 could be synthesized via this route. The corresponding substituted alkenes were obtained by reacting
phenylacetyl chloride with the substituted allylic alcohols. These were then subjected to reaction
conditions of setꢀup B. Lactone 11 and 13 were obtained from substituted allyl phenylacetates 10 and 12
in 32% and 30% yield over two steps, respectively. Both reagents were obtained as single diastereomers
(Scheme 8).17
O
O
1. p-ABSA, DBU
O
2. Rh2(OOct)4
O
10
11 (32%)
O
O
O
1. p-ABSA, DBU
H
O
2. Rh2(OOct)4
12
13 (30%)
Scheme 8: Substituted lactones prepared
It was hoped that a completely continuous process could increase the yield of this transformation.
Therefore, a setꢀup utilizing the extraction system from setꢀup A was designed in which the outlet stream
was fed into a continuous flow system. Rhodium octanoate was used as the catalyst. Dichloromethane
was the solvent of this reaction as it provided a better solubility of the catalyst than nꢀheptane or toluene.
10
ACS Paragon Plus Environment

Page 11 of 18
Organic Process Research & Development
1
2
3
4
5
However, the completely continuous experiment did not improve the yield of the twoꢀstep protocol,
providing lactone 4 in only 27% yield (determined by HPLC) over two steps.
6
7
Considering the inconsistent results obtained to that point, we decided to do a side product analysis to
understand the competing reactions in the preparation of lactone 4.
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
Side product and error analysis
LC/MS analysis of the crude reaction products and flash chromatography for isolation of the side
products provided evidence for the main side products formed within the reactions. It was found that
relatively large quantities of compounds 14 and 15 could be observed in the majority of reactions. We
speculated that 14 could be explained by an OꢀH insertion reaction of the carbene into residual water
within the reaction. It would then be possible that 14 could react with another molecule of 7 to generate
15. Structure 16 was harder to explain. However, it is known that diazo compounds can react with nitriles
to form oxazoles.18 Therefore, it could be imagined that residual acetonitrile stemming from the diazo
transfer reacts with diazo compound 7 to furnish oxazole 16.
O
O
O
Me
O
N
Ph
O
OH
O
O
O
O
14
15
16
Figure 4: Isolated side products
To test this hypothesis for the formation of 16, diazo methyl phenylacetate 17 was subjected to a mixture
of acetonitrile in dichloromethane with 1 mol% rhodium acetate. Oxazole 18 was obtained in 38% yield
which showed that acetonitrile could indeed react with aryl diazoacetates to yield oxazoles (Scheme 9).
Me
N2
N
2 eq. MeCN
Rh2(OAc)4 1 mol%
O
CO2Me
OMe
CH2Cl2
25 °C, 12 h
Ar atmosphere
17
18 (38%)
Scheme 9: Formation of oxazoles from diazo methylphenylacetate
After isolation of the sideꢀproducts, it was investigated which factors could have played a role in the
formation of these sideꢀproducts. The quality of the rhodium catalysts could have an impact on the
transformation, therefore different batches of rhodium acetate were investigated for their ability to
transform 6 into 4 (see supporting information). However, no differences were observed.
11
ACS Paragon Plus Environment

Organic Process Research & Development
Page 12 of 18
1
2
3
4
5
6
7
8
9
Another possible explanation of the results was that the extractions were inefficient and that too much
water came through the workꢀup systems in setꢀup A and B. This was surprising as a completely
continuous setꢀup based on setꢀup B had been efficiently employed for intermolecular reactions of diazo
methylphenylacetate 17.13 To test this hypothesis, three batch experiments were performed using the
diazo reagent 7 directly after the batch extraction protocol. One of the reactions was performed in
acetonitrile (technical grade) to test if more of oxazole 16 could be observed (entry 1). The other two
reactions used 0.1 (entry 2) and 1 equivalent (entry 3) of water respectively. The reactions were
performed in nꢀheptane with rhodium acetate as catalyst in all three cases. The HPLC ratios of these
reactions are shown below. Acetonitrile as solvent did reduce the formation of lactone 4 significantly
which could be contributed in parts to the formation of larger quantities of oxazole 16 but also to the
water content of the solvent. Although 0.1 equivalents of water are still tolerated by the reaction, 1
equivalent of water shuts down the reaction completely, only providing a mixture of alcohol 14 and dimer
15. This means that even very small quantities of water passing by the liquid/liquid extraction would have
a detrimental effect on the formation of lactone 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 5: Side product analysis
Entry
1
Conditions
Lactone 4
[%]
Alcohol 14
Dimer 15 [%] Oxazole 16 [%]
[%]
8
CH3CN as
solvent
45
11
26
2
3
0.1 eq H2O added
1 eq H2O added
73
1
5
57
6
27
10
1
With the understanding of the influence of water for the reaction, we developed new approaches for the
twoꢀstep continuous into semiꢀbatch protocol (Setꢀup C).
Semi-batch protocols 2nd generation and scale-up
A simple initial alteration to the semiꢀbatch protocol of the first generation was the use of dried solvents
for the cyclization in batch. Beforehand solvents of technical grade had been used in the twoꢀstep protocol
to reduce costs for an eventual scaleꢀup. The flask was kept under argon atmosphere with molecular
sieves as drying agents. Rhodium octanoate was used as the catalyst as it had outperformed rhodium
acetate in previous reactions consistently. Furthermore, the gravitational separator was used as the
separating system as it had given better results than the membrane separating system (see Table 4 entry 1
vs 3). A smaller column was used for the liquid/liquid extraction to reduce the scale needed to test the
reaction setꢀup. Setꢀup C is shown in Scheme 10.
12
ACS Paragon Plus Environment

Page 13 of 18
Organic Process Research & Development
1
2
3
4
5
NaNO2 in H2O
0.1 mL/min
Data analysis
O
O
O
O
S
n-heptane
Ar
N3
0.138 mL/min
6
1.2 M
organic layer
7
4
8
9
N2
O
mixing system LTF
in line
IR analysis
flow cell
2.21 mL
d = 0.8 mm
0.084 ml/min
26 min res. time
65 °C
0.175 mL/min
L/L phase separator
CO2R
O
1.0 M
6
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
Rh2L4 (1 mol%)
in toluene (dry)
under Ar atmosphere
4 Å MS
DBU 1.4 M in MeCN
aq. layer
waste
65 °C
Scheme 10: Continuous flow into semiꢀbatch setꢀup C
Setꢀup C provided the lactone in 40% yield (Table 6, entry 1). To further improve the yield of the
flow/semiꢀbatch protocol, a column containing MgSO4 as water trap was included between the
liquid/liquid extraction and the addition to the round bottom flask. This improved the yield, providing
lactone 4 in 53% yield over two steps (Table 6, entry 2). Increasing the amount of the MgSO4 in the trap
did not lead to a further improvement in product formation (Table 6, entry 3).
Table 6: Results of the two step continuous flow into semiꢀbatch approach
Entry
MgSO4 trap
n.A.
4 cm x 6 mm (1 g)
11 cm x 6 mm (2.7 g)
Yield [%]
1
2
3
40
53
50
Having established a setꢀup with a significantly improved yield and being equipped with a good
understanding of the dynamics of the twoꢀstep reaction, we proceeded to scale up the continuous flow
into semiꢀbatch system.
For the scaleꢀup setꢀup C was employed with only few modifications. All flow rates were increased by a
factor of 2.4 and a larger reactor coil (10 ml) was used for the diazo transfer reaction to account for the
higher flow rates (Figure 5). The setꢀup was run for 8 hours with a final collection time of 6 hours (360
min). 2.04 g (33%) of 4 were obtained. The main sideꢀproduct was dimer 15 of which 1.69 g (26%) were
isolated. This result suggests that more water was passing through the system than in the small scale
reaction. One possible explanation is that the MgSO4 trap was saturated with water after a certain reaction
time. Alternatively, the higher flow rates employed led to a less efficient trapping of residual water in the
MgSO4 column.
13
ACS Paragon Plus Environment

Organic Process Research & Development
Page 14 of 18
1
2
3
4
5
MgSO4
trap
Batch
reactor
6
7
reactor
8
9
Phase
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
separator
pumps
mixer
Figure 5: Upscale for multistep protocol to bicyclic lactone 4
Conclusion
In conclusion, we have presented herein a diazo based continuous flow approach to the bicyclic lactone 4
which is a key intermediate on the route to milnacipran. All steps were carefully optimized in continuous
flow and a 2ꢀstep process was established which circumvents the use of column chromatography of the
diazo reagent. By analyzing the side products formed, it was found that water had a very strong negative
impact on the reaction. Although a solution to trap the water from in the inꢀline liquid/liquid separation
was found for small scale reactions using an inꢀline MgSO4 trap, this system did not proof efficient for
larger scale. This work presents a proof of concept for a flexible, safe and rapid generation of molecular
complexity from simple starting materials which should give access to highly substituted milnacipran
analogues. Albeit providing a reduced yield, the system was scalable to gram quantities of materials.
Experimental Section
General Information. Reactions were carried out in standard reagent grade solvents. All chemicals were
purchased from Sigma Aldrich and Alfa Aesar and used without further purification. KR Analytical Ltd
Fusion 100 Touch syringe pumps and Vapourtec E series pumps were used for cyclopropanation in flow
as well as for setꢀup B and setꢀup C. Syrris Asia pumps and Ismatec piston pumps were used for diazo
transfer optimization studies and setꢀup A. Mettler Toledo IR spectroscopy with flow diamond probe and
flow silicon probe was used for reaction optimization and kinetics studies. Thermal analyses were
performed with a Setaram C80 calorimeter. Purifications were performed on a Biotage® 25 g SNAP
cartridge / Biotage Isolera system. Thin layer chromatography was performed on Merck Silica gel 60 F254.
Dried solvents were directly used from a SPS dry solvent system (toluene, heptane, acetonitrile) or
distilled over CaH2 (dichloromethane). H and 13C spectra were recorded on Bruker Fourier 300 and
1
Bruker DPX 400 and referenced to the residual proton solvent peak (1H: CDCl3, δ 7.26 ppm) and solvent
13C signal (CDCl3, δ 77.16). Signals were reported in ppm, multiplicity (s = singlet, d = doublet, t =
triplet, q = quartet, quin = quintet, sex = sextet, hep = septet, m = multiplet, b = broad), coupling constants
in Hertz.
Optimization diazo transfer in flow for Allyl 2-diazo-2-phenylacetate (7) Allyl phenylacetate (35.24
g, 200 mmol) and DBU (42.6 ml, 280 mmol, 1.4 eq) were dissolved in acetonitrile for a total volume of
14
ACS Paragon Plus Environment

Page 15 of 18
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
200 ml. pꢀAcetamidobenzenesulfonyl azide (57.66 g, 240 mmol, 1.2 eq) was dissolved in a total volume
of 200 ml acetonitrile and subsequently filtered to remove residual solids. The two solutions were
pumped through the flow system using Asia syringe pumps with flow rates of 0.5 ml/min each (combined
flow rate 1 ml/min) for a residence time of 8.2 min and of 0.3 ml/min each (combined flow rate 0.6
ml/min) for a residence time of 13.7 min. The reaction mixture was analyzed using inꢀline infrared with
the consumption of starting material used to determine the yield of the transformation. Calibration of allyl
phenylacetate gave a R2 of 0.993. Extraction of allyl 2ꢀdiazoꢀ2ꢀphenylacetate was performed using nꢀ
heptane (15 ml) and 1M NaNO2 (11 ml) solution per 1 g allyl phenylacetate used. 964 mg Allyl 2ꢀdiazoꢀ
2ꢀphenylacetate 7 (collection time 19 min, 4.77 mmol) was obtained without further purification in 84%
yield as yellow oil. 1H NMR (400 MHz, CDCl3): δ = 7.46ꢀ7.51 (m, 2H), 7.36ꢀ7.42 (m, 2H), 7.16ꢀ7.22 (m,
1H), 5.92ꢀ6.04 (m, 1H), 5.37 (ddd, J = 17.2, 3.0, 1.5 Hz, 1H), 5.28 (ddd, J = 10.4, 2.5, 1.2 Hz, 1H), 4.78
(dt, J = 5.6, 1.4 Hz, 1H) ppm; 13C (100 MHz, CDCl3): δ = 165.2, 132.1, 128.9, 125.9, 125.4, 123.9, 118.4,
65.4 ppm.
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
Optimization cyclopropanation in flow for 1-phenyl-3-oxabicyclo[3.1.0]hexan-2-one (4) Allyl 2ꢀ
diazoꢀ2ꢀphenylacetate 7 (202 mg, 1 mmol) was dissolved in the reaction solvent to give the total volume
of 5 ml. Catalyst (0.025 mmol, 2.5 mol%) was dissolved in the reaction solvent to give the total volume
of 5 ml. Both solutions were charged into 5 ml syringes and pumped using a KR Analytical LtD Fusion
100 Touch syringe pump with a combined flow rate of 0.1 ml/min through a 1 ml reaction coil (residence
time 10 minutes). The system was stabilized for 30 min and then collected for 60 min by running through
a silica plug to quench the reaction mixture. The silica plug was washed with CH2Cl2 and the solvent was
evaporated in vacuo. The crude reaction mixture was purified using column chromatography (10 g SNAP
Ultra, 9:1 nꢀHexane:EtOAc). 1ꢀPhenylꢀ3ꢀoxabicyclo[3.1.0]hexanꢀ2ꢀone was obtained as colorless oil (for
yields see Table 3). 1H NMR (400 MHz, CDCl3): δ = 7.48ꢀ7.27 (m, 5H), 4.47 (dd, J = 9.27, 4.6 Hz, 1H),
4.31 (d, J = 9.27 Hz, 1H), 2.57 (dt, J = 7.8, 4.6 Hz, 1H), 1.66 (dd, J = 7.8, 4.86 Hz, 1H), 1.38 (t, J = 4.58
Hz, 1H) ppm. 13C NMR (100 MHz, CDCl3): δ = 176.5, 134.5, 129.1, 128.8, 128.2, 68.6, 32.2, 25.5, 20.6
ppm
5-(((4-Methoxybenzyl)oxy)methyl)-1-phenyl-3-oxabicyclo[3.1.0]hexan-2-one (9) The reaction mixture
was collected for 60 minutes (combined flow rate 0.1 ml/min) and subsequently purified as described
above. 178 mg (91%) of 9 were obtained as colorless oil. 1H NMR (400 MHz, CDCl3): δ = 7.24 (m, 5H),
6.98 (m, 2H), 6.74 (m, 2H), 4.45 (d, J = 9.06 Hz, 1H), 4.22 (d, J = 11.56 Hz, 1H), 4.15 (d, J = 11.56 Hz,
1H), 3.70 (s, 3H), 3.40 (d, J = 10.73 Hz, 1H), 3.15 (d, J = 10.73 Hz, 1H), 1.63 (d, J = 5.13 Hz, 1H), 1.31
(d, J = 5.13 Hz, 1H). 13C NMR (75 MHz, CDCl3): δ = 176.5, 159.3, 131.2, 130.4, 129.6, 129.2, 128.6,
128.2, 113.8, 72.9, 70.1, 68.3, 55.3, 36.8, 35.3, 20.5 ppm; IR (neat): ν 3958m, 2908m, 2858m, 1770s,
1612m, 1512s, 1454w, 1361m, 1303w, 1249s, 1091s, 1037m, 1014m, 910s cmꢀ1; HMRS: exact mass
calcd for C20H20O4 [M]+: 324.1362, Found: 324.1360.
2-Step continuous flow / semi-batch protocol Allyl ester (4 mmol) was dissolved in acetonitrile and
DBU (0.84 ml, 5.6 mmol, 1.4 eq) was added to give a total volume of 4 ml. pꢀABSA (1.15 g, 4.8 mmol,
1.2 eq) was dissolved in acetonitrile to give a total volume of 4 ml. Both solutions were charged into 5 ml
syringes. Flow rates were put to 0.042 ml/min for the ester and sulfonyl azide solution on a syringe pump.
1 M aqueous sodium nitrite quench solution was pumped via Vapourtec pump (blue tubing) with a flow
rate of 0.101 ml/min and nꢀheptane for the extraction was pumped with a flow rate of 0.158 ml/min. After
passing the L/L phase separator, the aqueous layer was discarded and the organic layer was added to the
15
ACS Paragon Plus Environment

Organic Process Research & Development
Page 16 of 18
1
2
3
4
5
6
7
8
9
solution. Rh2(OOct)4 (10.1 mg, 1 mol%) was dissolved in nꢀheptane (5 ml) and stirred at 65 °C. Addition
of organic layer stream was performed for 30 minutes and the mixture stirred at 65 °C for another 30 min.
Subsequently, solvent was evaporated in vacuo and the reaction mixture was purified via flash
chromatography on an automated purification system using a 20g silica column.
5-Methyl-1-phenyl-3-oxabicyclo[3.1.0]hexan-2-one (11)19 63 mg (32% yield; 161 mg of 198 mg of the
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
crude reaction mixture was used for purification) of 10 were obtained as colorless oil. H NMR (250
MHz, CDCl3): δ = 7.35ꢀ7.15 (m, 5H), 4.30 (d, J = 9.1 Hz, 1H), 4.12 (d, J = 9.1 Hz, 1H), 1.55 (d, J = 4.9
Hz, 1H), 1.32 (d, J = 4.9, 1H), 1.09 (s, 3H) ppm. 13C NMR (63 MHz, CDCl3): δ = 177.6, 132.2, 130.4,
128.9, 128.2, 73.4, 37.1, 31.4, 22.8, 15.4 ppm.
6-Methyl-1-phenyl-3-oxabicyclo[3.1.0]hexan-2-one (13)20 66 mg (30% yield; 196 mg of 211 mg of the
1
crude reaction mixture was used for purification) of 11 were obtained as colorless oil. H NMR (400
MHz, CDCl3): δ = 7.44ꢀ7.28 (m, 5H), 4.42 (dd, J = 9.1, 4.5 Hz, 1H), 4.32 (d, J = 9.1 Hz, 1H), 2.30 (t, J =
13
4.5 Hz, 1H), 1.50ꢀ1.42 (m, 1H), 0.80 (d, J = 6.1 Hz, 3H) ppm. C NMR (100 MHz, CDCl3): δ = 176.7,
130.6, 128.8, 128.2, 68.7, 37.2, 28.9, 26.4, 13.6 ppm.
Allyl 2-hydroxy-2-phenylacetate (14)21 Obtained as colorless oil. 1H NMR (300 MHz, CDCl3): δ = 7.48ꢀ
7.41 (m, 2H), 7.39ꢀ7.28 (m, 3H), 5.90ꢀ5.73 (m, 1H), 5.20 (s, 2H), 5.16 (dq, J = 5.14, 1.27 Hz, 1H), 4.60ꢀ
4.68 (m, 2H), 3.40 (bs, 1H) ppm. 13C NMR (75 MHz, CDCl3): δ = 173.3, 138.3, 131.2, 128.6, 128.5,
126.6, 118.7, 72.9, 66.4 ppm.
Diallyl 2,2’-oxybis(2-phenylacetate) (15) Obtained as pale yellow oil. 1H NMR (300 MHz, CDCl3): δ =
7.33ꢀ7.50 (m, 5H), 5.76ꢀ5.90 (m, 1H), 5.13ꢀ5.24 (m, 2H), 5.04 (d, J = 2.0 Hz, 1H), 4.57ꢀ4.64 (m, 2H)
ppm. 13C NMR (75 MHz, CDCl3): δ = 169.8, 135.5, 131.5, 128.9, 127.6, 127.5, 118.4, 78.7, 65.7 ppm; IR
(neat): 3066m, 2085m, 1817w, 1801w, 1747s, 1685m, 1647w, 1496m, 1454m, 1361w, 1246m, 1203m,
1173s, 1103m, 987m, 935m, 732m, 698m cm–1; HMRS: exact mass calcd for C22H25O5 [M]+: 366.1467,
Found: 366.1468.
5-Methoxy-2-methyl-4-phenyloxazole (18) Methyl 2ꢀdiazoꢀ2ꢀphenylacetate 15 (176 mg, 1 mmol) was
dissolved in dry CH2Cl2 (4 ml) and added via syringe pump over 1 h to a solution containing Rh2(OAc)4
(4.4 mg, 0.01 mmol; 1 mol%) in dry CH2Cl2 (4 ml) under argon atmosphere (Schlenk dried flask) at
reflux. After completion of the reaction (via TLC analysis) the reaction mixture was cooled down to room
temperature and then the solvent was evaporated in vacuo. The crude reaction mixture was immediately
purified via flash column chromatography (12 g silica column with 0%ꢀ10% EtOAc in nꢀhexane over 25
1
column volumes) to furnish 71 mg of 16 (38%) as colorless oil. H NMR (300 MHz, CDCl3): δ = 7.74ꢀ
7.79 (m, 2H), 7.30ꢀ7.38 (m, 2H), 7.18 (tt, J = 8.37, 1.27 Hz, 1H), 4.00 (s, 3H), 2.41 (s, 3H) ppm. 13C
NMR (75 MHz, CDCl3): δ = 154.3, 151.9, 131.4, 128.5, 126.3, 124.8, 114.5, 60.1, 14.3 ppm. IR (neat)
3028w, 2949w, 1743s, 1647s, 1599s, 1500w, 1449m, 1369m, 1221s, 1173m, 1016m, 766m, 731m cm–1.
HRMS: Exact mass calc for C11H12NO2 [M+H]+: 190.0868, Found: 190.0862.
Author information
Corresponding author
*Eꢀmail: wirth@cf.ac.uk
16
ACS Paragon Plus Environment

Page 17 of 18
Organic Process Research & Development
1
2
3
4
Notes
5
6
The authors declare no competing financial interest.
7
8
Acknowledgments
9
Financial support by Pierre Fabre and the School of Chemistry, Cardiff University, is gratefully
acknowledged. We thank the EPSRC National Mass Spectrometry Facility, Swansea, for mass
spectrometric data.
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
Supporting Information
The supporting information for this paper contains the results of reactions using different rhodium acetate
batches.
References
1. Kasper, S.; Pletan, Y; Solles, A.; Tournoux, A. Int. Clin. Psychopharmacol. 1996, 11, 35ꢀ39.
2. LopezꢀIbor, J; Guelfi, J. D.; Pletan, Y.; Tournoux, A.; Prost, J. F. Int. Clin. Psychopharmacol. 1996, 11, 41ꢀ46.
3. Tamiya, J.; Dyck, B.; Zhang, M.; Phan, K.; Fleck, B. A.; Aparicio, A.; Jovic, F.; Tran, J. A.; Vickers, T.; Grey,
J.; Foster, A. C.; Chen, C. Bioorg. Med. Chem. Lett. 2008, 18, 3328ꢀ3332.
4. a) Shuto, S.; Takada, H.; Mochizuki, D.; Tsujita, R.; Hase, Y.; Ono, S.; Shibuya, N.; Matsuda, A. J. Med. Chem.
1995, 38, 2964ꢀ2968; b) Shuto, S.; Ono, S.; Hase, Y.; Kamiyama, N.; Takada, H.; Yamasihita, K.; Matsuda, A.
J. Org. Chem. 1996, 61, 915ꢀ923.
5. Doyle, M. P.; Hu, W. Adv. Synth. Catal. 2001, 343, 299ꢀ302.
6. Shen, J.ꢀJ.; Zhu, S.ꢀF.; Cai, Y.; Xu, H.; Xie, X.ꢀL.; Zhou, Q.ꢀL. Angew. Chem. Int. Ed. 2014, 53, 13188ꢀ13191.
7. De Boer, T. J.; Backer, H. J. Org. Synth. 1956, 36, 16ꢀ19.
8. Gutmann, B.; Cantillo, D.; Kappe, C. O. Angew. Chem. Int. Ed. 2015, 54, 6688ꢀ6728.
9. Wirth, T. Microreactors in Organic Chemistry and Catalysis, 2nd edition, Wiley: Weinheim, 2013.
10. For reviews on diazo compounds in flow: a) Deadman, B. J.; Collins, S. G.; Maguire, A. R. Chem. Eur. J. 2015,
21, 2298ꢀ2308; b) Müller, S. T. R.; Wirth, T. ChemSusChem 2015, 8, 245ꢀ250.
11. a) Procter, L. D.; Warr, A. J. Org. Process Res. Dev. 2002, 6, 884ꢀ892; b) Ferstl, W. F.; Schwarzer, M. S.;
Lӧbbecke, S. L. Chem. Ing. Tech. 2004, 76, 1326ꢀ1327; c) Struempel, M.; Ondruschka, B.; Daute, R.; Stark, A.
Green Chem. 2008, 10, 41ꢀ43; d) Maurya, R. A.; Park, C. P.; Lee, J. H.; Kim, D.ꢀP. Angew. Chem. Int. Ed. 2011,
50, 5952ꢀ5955; e) Martin, L. J.; Marzinzik, A. L.; Ley, S. V.; Baxendale, I. R. Org. Lett. 2011, 13, 320ꢀ323; f)
Rossi, E.; Woehl, P.; Maggini, M. Org. Process Res. Dev. 2012, 16, 1146ꢀ1149; g) Mastronardi, F.; Gutmann,
B.; Kappe, C. O. Org. Lett. 2013, 15, 5590ꢀ5593; h) Pinho, V. D.; Gutmann, B.; Miranda, L. S. M.; de Souza, R.
O. M. A.; Kappe, C. O. J. Org. Chem. 2014, 79, 1555ꢀ1562.
12. a) Bartrum, H. E.; Blakemore, D. C.; Moody, C. J.; Hayes, C. J. J. Org. Chem. 2010, 75, 8674ꢀ8676; b) Delville,
M. M. E.; van Hest, J. C. M.; Rutjes, F. P. J. T. Beilstein J. Org. Chem. 2013, 9, 1813ꢀ1818; c) Maurya, R. A.;
Min, K.ꢀI.; Kim, D.ꢀP. Green Chem. 2014, 16, 116ꢀ120; d) Müller, S. T. R.; Smith, D.; Hellier, P.; Wirth, T.
Synlett 2014, 25, 871ꢀ875.
13. Müller, S. T. R.; Murat, A.; Maillos, D.; Lesimple, P.; Hellier, P.; Wirth, T. Chem. Eur. J. 2015, 21, 7016ꢀ7020.
14. See Stoessel, F.; Fierz, H.; Lerena, P.; Killé, G. Org. Proc. Res. Dev. 1997, 1, 428ꢀ434.
15. Volumes correspond to the amount of starting material used. For a reaction using 1 g of starting material, 11 ml
of nꢀheptane for the extraction were used.
16. a.) M. O’Brien, P. Koos, D. L. Browne, S. V. Ley, Org. Biomol. Chem. 2012, 10, 7031ꢀ7036; b.) D. X. Hu, M.
O’Brien, S. V. Ley, Org. Lett. 2012, 14, 4246ꢀ4249.
17. Traces of another diastereomer were possibly observed in case of 13 in the crude 1H NMR, however, isolation of
this compound was unsuccessful and therefore determination of the structure was not possible.
17
ACS Paragon Plus Environment

Organic Process Research & Development
Page 18 of 18
1
2
3
4
5
6
7
8
9
18. Doyle, M. P.; Buhro, W. E.; Davidson, J. G.; Elliott, R. C.; Hoekstra, J. W.; Oppenhuizen, M. J. Org. Chem.
1980, 45, 3657ꢀ3664.
19. Doyle, M. P.; Davies, S. B.; Hu, W. Org. Lett. 2000, 2, 1145ꢀ1147.
20. Barluenga, J.; Aznar F.; Gutiérrez, I.; Martin, J. A. Org. Lett. 2002, 4, 2719ꢀ2722.
21. Alamsetti, S. K.; Sekar, G. Chem. Commun. 2010, 46, 7235ꢀ7237.
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
18
ACS Paragon Plus Environment