Toward a Scalable Synthesis and Process for EMA401, Part II: Development and Scale-Up of a Pyridine- A nd Piperidine-Free Knoevenagel-Doebner Condensation

Article abstract of DOI:10.1021/acs.oprd.0c00216

During route scouting for EMA401 (1), an angiotensin II type 2 antagonist, we identified the synthesis of key amino acid intermediate 2 via its cinnamic acid derivative 3 as a streamlined option. In general, cinnamic acids can be synthesized from the corresponding aldehydes by a Knoevenagel-Doebner condensation in pyridine with piperidine as an organocatalyst. We aimed to replace both of these reagents and found novel conditions involving toluene as the solvent and morpholine as the organocatalyst. Scale-up of the process allowed the production of 25 kg of cinnamic acid 3 that was of the quality required for process development of the subsequent phenylalanine ammonia lyase-catalyzed step. The modified conditions were found to be widely applicable to alternative aldehydes and thus are of relevance to practitioners of chemical scale-up.

Full text of DOI:10.1021/acs.oprd.0c00216

Subscriber access provided by CORNELL UNIVERSITY LIBRARY
Communication
Towards a Scalable Synthesis and Process for EMA401.
Part II: Development and Scale-up of a Pyridine- and
Piperidine-free Knoevenagel-Doebner Condensation
Eric Sidler, Roger Humair, and Leo Albert Hardegger
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.0c00216 • Publication Date (Web): 20 Aug 2020
Downloaded from pubs.acs.org on August 20, 2020
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 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 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.
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 35
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Towards a Scalable Synthesis and Process for
EMA401. Part II: Development and Scale-up of a
Pyridine- and Piperidine-free Knoevenagel-Doebner
Condensation
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
Eric Sidler†, Roger Humair, Leo A. Hardegger*
Chemical and Analytical Development, Novartis Pharma AG, 4002 Basel, Switzerland.
ACS Paragon Plus Environment
1

Organic Process Research & Development
Page 2 of 35
1
2
3
4
5
Graphical abstract
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
2

Page 3 of 35
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
ABSTRACT During the route scouting of EMA401 (1), an angiotensin II type 2
antagonist, we identified the synthesis of key amino acid intermediate 2 via its cinnamic
acid derivative 3 as a streamlined option. In general, cinnamic acids can be synthesized
from the corresponding aldehyde in a Knoevenagel-Doebner condensation in pyridine
with piperidine as an organocatalyst. We aimed at replacing both of these reagents and
found novel conditions in toluene as the solvent and morpholine as the organocatalyst.
Scale-up of the process allowed production of 25 kg of the cinnamic acid 3, which was
of required quality for process development on the subsequent phenylalanine ammonia
lyase-catalyzed step. The modified conditions were found to be widely applicable to
alternative aldehydes, and so of relevance to practitioners of chemical scale-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
KEYWORDS Knoevenagel-Doebner, scale-up, cinnamic acid, pyridine-free, piperidine-
free
Introduction
ACS Paragon Plus Environment
3

Organic Process Research & Development
Page 4 of 35
1
2
3
4
5
6
7
8
9
During the route scouting of EMA401 (1), several potential routes to the key amino
acid intermediate 2 were evaluated.1 One of the options identified relied on the
phenylalanine ammonia lyase (PAL) catalyzed conversion of cinnamic acid 3 to amino
acid 2 (Scheme 1).2 To support our development program, we needed access to
kilogram quantities of cinnamic acid 3. We envisioned that the cinnamic acid could be
made by a condensation reaction with benzylated ortho-vanillin 4, an intermediate we
had available in several kilograms as it was as a common intermediate for many of the
proposed approaches to amino acid 2.1 In a first screening employing either Perkin
condensation conditions, the use of catalytic DBU,3 or the Doebner modification of the
Knoevenagel condensation,4 we found that the latter conditions proved most suitable for
our substrate (Table 1, entry 1). The standard condition for a Knoevenagel-Doebner
condensation employs pyridine as the solvent and piperidine as the organocatalyst.
Since their first reports, a plethora of reports for Knoevenagel and Knoevenagel-
Doebner condensation conditions replacing toxic or hazardous reagents have been
published. Recently, Allais and co-workers developed a Knoevenagel-Doebner
condensation to p-hydroxycinnamic acids with L-proline and malonic acid in EtOH,
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
4

Page 5 of 35
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
yielding the naturally occurring acids in high yields.5,6 The preparation of cinnamic acids
in water was reported using microwave irraditation.7 In addition, “solvent-free”
Knoevenagel-Doebner conditions have also been reported recently, using ammonium
bicarbonate as a catalyst and yielding the corresponding cinnamic acids in high yields.8
We aimed at replacing both piperidine and pyridine for several reasons: 1) piperidine
is a precursor substance of narcotics,9 and so its usage is complicated by bureaucratic
tracking and control paper processes, 2) pyridine is possibly carcinogenic to humans
and hazardous to the aquatic environment,10,11 and 3) using an acidic quench as the
work-up would have yielded large amounts of aqueous waste not suitable for the waste
water treatment plant. With a large scale commercial manufacturing in mind, we aimed
at developing conditions using more benign reagents and solvents following published
solvent selection principles.12,13
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
5

Organic Process Research & Development
Page 6 of 35
1
2
3
4
Na
5
6
7
8
PAL
NH3
O
O
O
O
O
O
O
O
O
O
(S)
O
O
O
N
OH
OH
O
(S)
NH2
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
4
3
2
1
This publication
Scheme 1. Synthetic route to EMA401 (1) via the key intermediate amino acid 2. Amino
acid 2 can be prepared from cinnamic acid derivative 3, which is itself prepared in a
Knoevenagel-Doebner condensation starting from aldehyde 4.
Experimental Details
Several alternatives to piperidine as the organocatalyst could be identified in an initial
screening (Table 1). In particular, morpholine (entry 2), pyrrolidine (entry 3) and -
alanine (entry 4) showed comparable reactivity to piperidine (entry 1) in pyridine. Other
secondary amines such as piperazine or L-proline generally showed lower or no
conversion in our screenings. We selected morpholine as the reagent of choice,
considering the HSE properties, the large-scale availability and pricing, the performance
in the screening, as well as the compatibility with the process conditions (vide infra).
ACS Paragon Plus Environment
6

Page 7 of 35
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Table 1. Solvent and amine reagent screening for the conversion of aldehyde 4 to
cinnamic acid 3.
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
conditions
O
O
O
O
O
HO
OH
O
O
OH
O
4
5
3
Entry Solvent
Amine
Temperature Time
Conversion to
3
reagent
Pyridine
Pyridine
Pyridine
Pyridine
Piperidine
Morpholine
Pyrrolidine
-Alanine
-
100 °C
100 °C
80 °C
2 h
2 h
2 h
2 h
2 h
2 h
2 h
2 h
2 h
2 h
98%
>99%
>99%
>99%
5%
1
2
3
4
5
6
7
8
9
10
100 °C
100 °C
80 °C
Morpholine
Pyrrolidine
DBU
-
<1%
<1%
24%
12%
99%
Morpholine
Morpholine
Morpholine
Morpholine
85 °C
TEA
85 °C
DIPEA
4-
85 °C
105 °C
Methylmorpholine
ACS Paragon Plus Environment
7

Organic Process Research & Development
Page 8 of 35
1
2
3
4
5
6
7
8
9
THF
Morpholine
Morpholine
Morpholine
Morpholine
Morpholine
Morpholine
Morpholine
Morpholine
Morpholine
Morpholine
90 °C1)
80 °C
94 °C
94 °C
94 °C
80 °C
94 °C
110 °C
97 °C
94 °C
20 min1) 80%
11
12
13
14
15
16
17
18
19
20
MeTHF
CPME
3.5 h
7 h
95%
>99%
>99%
>99%
>95%
>99%
83%
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
t-Amyl alcohol
iPrOH
2 h
7 h
7 h
Toluene
Xylenes
Heptane
H2O
7 h
3.5 h
3 h
2%
7 h
<1%
Conditions: 1–5 v/w solvent with 1.05–1.2 eq. of malonic acid and 0.12–0.25 eq. of the
amine reagent. 1) microwave
We then tried to run the reactions in pure organocatalyst as the solvent. Running the
reaction in neat morpholine (entry 5) or pyrrolidine (entry 6), however, only led to minor
conversion with mostly starting material detected by HPLC. A screening of the ratio
pyridine/morpholine showed that only mixtures above 10:1 (v/v) were efficient at
catalyzing this reaction, with ratios up to 50:1 (v/v) giving good conversion, albeit at a
ACS Paragon Plus Environment
8

Page 9 of 35
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
lower rate. Reactions using pyridine as the solvent at 5 g substrate scale were
performed with a pyridine/morpholine ratio of 10:1 (v/v). Conversion of aldehyde 4 was
usually complete within 2 h, but the reactions afforded very thick suspensions after a
very exothermic quench with aqueous acids, and the material generated under these
conditions tended to stick to the walls of the reactor. With mechanical cleaning of the
reactor, isolated yields of cinnamic acid 3 of around 90% could be obtained. We then
set out to screen for alternatives to pyridine as the reaction solvent, focusing first on
amine-based solvents. Alternative nitrogen bases led to low conversion (DBU, entry 7)
or the formation of gels (TEA, entry 8, DIPEA, entry 9). When 4-methylmorpholine was
used, a conversion of 99% was observed in 2 hours (entry 10). However, a self-ignition
temperature of 167 °C of 4-methylmorpholine would have required additional safety
measures for production in a pilot plant environment.14 We therefore continued our
screening with common organic solvents.12,13 A minimal boiling point of 60 °C was
imposed, since the decarboxylation of the dicarboxylic acid intermediate V (Scheme 2)
took place at around 70 °C. Gratifyingly, most of the solvents tested gave good to
excellent conversion. THF (entry 11), MeTHF (entry 12), CPME (entry 13), DMSO (entry
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
9

Organic Process Research & Development
Page 10 of 35
1
2
3
4
5
6
7
8
9
14), t-amyl alcohol (entry 15), iPrOH (entry 16), toluene (entry 17) and xylenes (entry
18) all gave conversions of >80% after 2–7 h at temperatures above 80 °C, while
heptane (entry 19) and water (entry 20), gave only minor conversion with mostly
unreacted starting material detected by HPLC. Although DMSO seemed to be the best
solvent for the given reaction, it was excluded from further screenings, due to safety
considerations and the formation of darkly colored solutions. Out of the solvents
screened, we favored MeTHF, t-amyl alcohol and toluene due to their fast reaction, and
their high and clean conversion.
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
When t-amyl alcohol was used on a 5 g scale, full conversion could be achieved in 2
h. However, upon cooling the reaction mixture, the desired product precipitated rapidly
as a hard solid, which caused stirrers to block and produced crusts which were not
easily washed out of the reactor. Modifying the conditions did not help to overcome
these issues in a manner we felt was acceptable for scale up.
Next, we ran the reaction in MeTHF on 5 g scale. The product was formed in 4 hours
at a jacket temperature (JT) of 94 °C. The high solubility of cinnamic acid 3 in MeTHF
(14.7wt%), further increased by the presence of morpholine (vide infra) meant that large
ACS Paragon Plus Environment
10

Page 11 of 35
Organic Process Research & Development
1
2
3
4
5
6
7
8
volumes of antisolvent, heptane in our case, were required to achieve a well stirrable
and transferable suspension with reasonable isolated yields of the product.
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
In an initial screening with toluene at 94 °C, the reaction took 7 h for full conversion.
When repeating the reaction and heating to JT = 110 °C, we were pleased to find that
full conversion with only very few by-products could be achieved in 2 h. In addition,
toluene is about seven times less expensive than MeTHF.
Based on these results, we went back to check the suitability of different secondary
amines as organocatalysts in toluene at a loading of 0.1 eq. Under these conditions,
only morpholine lead to full conversion in 5 h, while L-proline (26%), pyrrolidine (17%),
-alanine (8%) and piperazine (5%) gave significantly lower conversions.
Based on this initial screening for the conditions we then moved our attention to
developing a process on 10–20 g of aldehyde 4 in a double jacketed glass reactor.
When charging the reactor with all starting materials (aldehyde 4, malonic acid (5), and
morpholine) and suspending them in toluene, we noted that at temperatures between
25–50 °C, a gum-like substance was formed in situ, which was sticking to stirrers and
walls and impeded stirring, but dissolved again at temperatures above 50 °C. We
ACS Paragon Plus Environment
11

Organic Process Research & Development
Page 12 of 35
1
2
3
4
5
6
7
8
9
reasoned that this might be caused by the formation of morpholinium salts. Therefore,
we decided to dose morpholine as a solution in toluene at IT = 55–65 °C. Running the
process in this manner, the formation of the gum-like intermediate was no longer
observed.
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
In our small scale (5g) screening reactions, 1.05–1.20 eq. of malonic acid had been
sufficient to achieve full conversion. In our initial trials however, we were not able to
achieve full conversion of aldehyde 4 and observed the formation of a dark yellow color
when extending the reaction time.15 This led to the isolation of the desired product as a
yellow solid, however, with no impurity observed by HPLC or 1H NMR. The yellow color
could only be removed by extensive washing with toluene, incurring a consequent
reduction of the yield. We speculated that the yellow impurity might be caused by
polymerization. Indeed, when stirring a 1:1 mixture of aldehyde 4 and cinnamic acid
product 3 with morpholine at 110 °C, we observed the formation of a strong yellow
color, with concomitant consumption of both reaction partners and without apparent
formation of a new by-product detectable by HPLC. Heating both aldehyde 4 and
cinnamic acid 3 in separate reactions in the presence of morpholine to reflux did not
ACS Paragon Plus Environment
12

Page 13 of 35
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
lead to the formation of a yellow color and HPLC showed that the two compounds were
reasonably stable. Based on these observations, we pre-supposed that it is crucial to
achieve full conversion in a reasonable time, before the stated assumed polymerization
occurs in a follow-up reaction. Given that incomplete conversion occurred even when
employing 1.2 equivalents of malonic acid, we hypothesized that the decomposition of
malonic acid is a competing side-reaction and hence that the equivalents of malonic
acid are critical for achieving full conversion and high yield and purity of the isolated
cinnamic acid 3. This hypothesis was corroborated by the following experiments. A DSC
(see supporting information (SI)) of malonic acid (5) shows a first endotherm starting at
82 °C (minima at 92 °C), indicating decomposition, and a strong second endotherm
starting at 119 °C (minima at 136 °C), which corresponds to the reported melting point
of malonic acid.14 In a second experiment, a suspension of malonic acid and morpholine
were gradually heated in toluene and the stability of malonic acid assessed over 17 h.
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
By off-line H NMR analysis, no peak for the methylene group of malonic acid was
detected, but a singlet at 1.91 ppm was observed instead, indicating the formation of
acetic acid by CO2 elimination. Based on these results, we extended the equivalents of
ACS Paragon Plus Environment
13

Organic Process Research & Development
Page 14 of 35
1
2
3
4
5
6
7
8
malonic acid and found that 1.3 eq. are required to achieve full conversion of aldehyde
4 and prevent the formation of a strongly yellow colored product.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To remove the water formed during the reaction, a Dean-Stark trap was routinely used
in all reactions.16 For safety reasons, the release of CO2 was monitored by volume in
selected experiments and it was found that CO2 is formed over the entire reaction time
of approx. 1.5–2 h above ca. 80 °C, with no accumulation or sudden release of CO2
observed. Generally, after 2 h in refluxing toluene, full conversion of aldehyde 4 (< 1a%
by HPLC) was found. In the plant, it was deemed impractical to sample the reactor with
refluxing toluene and so for HSE reasons it was hence decided to increase the reaction
time at reflux to 3 h to ensure full conversion without the need for an in-process control
(IPC). Further experiments showed that up to 24 h at 110 °C did not have a detrimental
effect on yield and purity of the isolated product. After complete conversion, the mixture
was cooled to 20 °C with a gradient of 0.5 K/min. At a temperature of 30–40 °C, a
sudden precipitation of the voluminous, cotton-candy-like product occurred. We
observed that the product did not settle when stirring was stopped, and the addition of
more toluene did not afford a suspension which allowed emptying the reactor through its
ACS Paragon Plus Environment
14

Page 15 of 35
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
bottom valve. A screening of different additives indicated that only the addition of EtOH
helped to break the surface tension and ensured that the suspension could be
transferred onto the filter. Since the cinnamic acid 3 is highly soluble in ethanol, the
amount employed was minimized to 0.25 v/w with respect to aldehyde 2. As the
solubility of cinnamic acid 3 in toluene is also significantly increased in presence of
morpholine (vide infra, SI), heptane was added as an anti-solvent to decrease the loss
of product in the mother liquor. Twice the volume of toluene was used to reduce the
solubility of cinnamic acid 3 in the reaction solvent system. To purge residual product,
the lab glass reactor was rinsed with a mixture of toluene, ethanol and heptane
corresponding to the final solvent ratio obtained before filtration. When going to the
plant, a stainless steel reactor was used and it was found that the suspension could be
entirely transferred to the filter, with no crusts or residual product visible in the reactor.
For all development done after the initial scale-up, the reactor was consequently
washed with heptane only.
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
After drying, the cinnamic acid 3 thus obtained was pure by HPLC, but contained 3–5
1
mol% of morpholine as determined by H NMR. Since at this point in the project, and
ACS Paragon Plus Environment
15

Organic Process Research & Development
Page 16 of 35
1
2
3
4
5
6
7
8
9
more specifically of enzyme evolution,2 it was not clear if morpholine was compatible
with the subsequent enzyme-catalyzed step, we aimed at further purifying the cinnamic
acid 3. This was done in a straightforward manner by suspending the crude compound
in an aqueous acid, followed by filtration and drying. Hydrochloric acid proved to purge
morpholine efficiently in an initial finding, but due to its incompatibility with some
materials of construction and the potential to form ethyl chloride in the presence of
EtOH, we searched for alternative acids. Employing sulfuric acid lead to the formation of
yellow impurities during drying of cinnamic acid 3, even with only traces of sulfuric acid
present in the wet product after extended washing of the filter cake with H2O. When
using 10% citric acid, we were able to purge morpholine, and one additional wash with
H2O was sufficient to purge the citric acid. We serendipitously had also noted that the
addition of acetonitrile helped purging the by-product VI (Scheme 2). Therefore, a 10%
citric acid solution was mixed with acetonitrile (85:15 v/v) for the first wash, followed by
a second wash with H2O. Upon treatment of the compound with the aqueous layer, the
aspect changed from voluminous needles to broad platelets. In order to obtain a
powder-like substance, cinnamic acid 3 was finally re-slurried in heptane and filtered.
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
16

Page 17 of 35
Organic Process Research & Development
1
2
3
4
5
Subsequent drying afforded the desired compound in 86–92% yield with >99.8% purity,
6
7
8
1
and no morpholine detected by H NMR. Crucially for scale-up, starting materials,
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
product, solutions and suspensions were not found to have a critical heat release in
safety experiments. Only neat malonic acid showed an exotherm in safety experiments,
which was safe to control under the process conditions.
Mechanistic considerations
The mechanism of the Knoevenagel-Doebner condensation outlined in Scheme 2 is
adapted from textbooks as well as a computational study investigating p-
hydroxycinnamic acid and the formation of vinyl phenols,17 similar to the recently
reported work on L-proline mediated condensations.5 After formation of iminium I, and
attack by malonate II, intermediate III can undergo three different fragmentation
pathways to cinnamic acid 3. Pathway A by direct fragmentation, pathway B through
formation of the morpholine adduct IV and subsequent elimination, or pathway C via
dicarboxylic acid intermediate V. We observed both intermediates IV and V by LC-MS,
ACS Paragon Plus Environment
17

Organic Process Research & Development
Page 18 of 35
1
2
3
4
5
6
7
8
9
while intermediate III was never observed. Intermediate V was shown by LCMS to
accumulate at lower temperatures, which is consistent with our finding that no CO2
release was observed below 70 °C. Compound V could be prepared and it was shown
to convert to cinnamic acid 3 in 2 h (86% conversion) in the presence of morpholine,
while <1% conversion was observed when no morpholine was added to the reaction
mixture, highlighting the dual catalytic activity of the organocatalyst as base and
nucleophile. Dicarboxylic acid VI was observed as a by-product, presumably formed by
the 1,4-addition of malonic acid to cinnamic acid 3 and subsequent de-carboxylation.
Both acids V and VI were found as trace impurities (<0.1a%) in some batches of
isolated cinnamic acid 3, but were not deemed critical as they were not expected to
interfere with the subsequent enzyme-catalyzed step.
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
18

Page 19 of 35
Organic Process Research & Development
1
2
3
O
O
4
5
6
7
8
9
HO
OH
5
START HERE
O
O
H2O
N
R1
O
R1 N+
H
R1 = OBn
R2 = OMe
R2
R2
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
HO
OH
I
4
II
H+
O
+
O
H
N
O
R1
O
OH
O
N
H
R2
R1
OH
R2
OH
pathway C
O
O
-CO2
VI
III
-CO2
pathway A
O
-CO2
+
O
O
H
N
R1
O
HO
OH
pathway B
R1
O
R2
5
OH
R1
O
R2
OH
OH
R2
OH
IV
O
3
V
O
N
H
Scheme 2. Catalytic cycle of the Knoevenagel-Doebner condensation with the key
intermediates III, IV and V for the synthesis of cinnamic acid 3, and the proposed
explanation for the formation of dicarboxylic acid by-product VI.
Process intensification after first scale-up
ACS Paragon Plus Environment
19

Organic Process Research & Development
Page 20 of 35
1
2
3
4
5
6
7
8
9
With the experience gained in the first scale-up, we aimed at further streamlining the
process. A screening performed on the subsequent PAL-catalyzed step revealed that
morpholine concentrations higher than those generally observed in isolated cinnamic
acid 3 (3–5mol%) were acceptable.2 Hence, the aqueous washes and the final heptane
wash of the filter cake of cinnamic acid 3 were no longer required. This allowed usage
of a standard filter instead of a stirred filter dryer. In addition, as described above, the
filter cake only required a displacement wash of the mother liquor with heptane when
using a stainless steel reactor. The product thus generated in the lab was successfully
employed in use-tests within the subsequent PAL-catalyzed step.
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
In addition to the changes above, we used a Crystal16 to determine the cloud point of
the precipitation of cinnamic acid 3 during cooling of the mixture at the end of the
reaction. We aimed at avoiding a spontaneous precipitation of product 3 upon cooling
when running the reaction at higher concentration, and before the addition of EtOH.
While the transition points of cinnamic acid 3 in toluene without EtOH were found to be
79–85 °C (4 v/w toluene wrt aldehyde 4, 0.15 eq. morpholine, see SI), the cloud point
could be significantly reduced to 44–46 °C after the addition of EtOH (0.1 v/w wrt 4),
ACS Paragon Plus Environment
20

Page 21 of 35
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
while the clear point remained at 81–82 °C. As the cloud point was not critical anymore
once EtOH had been added, we set up another experiment to determine the
dependency of the cloud point on the dilution in toluene and the equivalents of
morpholine (see SI). Based on this second screening, we selected 0.165 eq. of
morpholine at 3 v/w dilution since the cloud point of <70 °C is low enough to prevent
spontaneous precipitation before EtOH addition. This could be confirmed on a 20 g
scale. We further found that the equivalents of malonic acid could be reduced to 1.2 eq.
as a consequence of the higher concentration, which led to reduced formation of the
dicarboxylic acid by-product VI. This proved to be crucial as VI and other less soluble
impurities led to spontaneous nucleation at temperatures >70 °C. With this solvent
system combined with heptane as anti-solvent at 70–80 °C, precipitation occurred after
approx. 33% of the total heptane volume (4 v/w wrt aldehyde 4, see SI), resulting in a
thick but stirrable suspension, which was further mobilized by the addition of the
remaining 67% of the heptane. In conclusion, the concentration could be increased by a
factor of 2, with a reduction of the PMI from 47.1 to 7.3 (Scheme 3), and following this
protocol on 20 g scale, cinnamic acid 3 could be isolated in 98% yield.18
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
21

Organic Process Research & Development
Page 22 of 35
1
2
3
4
5
6
O
7
8
9
N
H
O
O
O
O
O
Toluene
110 °C, 3 h
then cool to 80 °C
EtOH, heptane
HO
OH
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
OH
O
4
5
3
Scale-up Process
Streamlined Process
Charge aldehyde 4 (1.0 eq.)
Charge malonic acid (1.3 eq.)
Charge toluene (5 v/w)
Heat to 110 °C
Charge aldehyde 4 (1.0 eq.)
Charge malonic acid (1.2 eq.)
Charge toluene (2.75 v/w)
Heat to 110 °C
Charge morpholine (0.2 eq.) as a solution in toluene (1
v/w) at 55-65 °C
Charge morpholine (0.165 eq.) as a solution in toluene
(0.25 v/w) at 55-65 °C
Stir at 110 °C for 3 h
Stir at 110 °C for 3 h
Cool to 80 °C
Cool to 80 °C
Add EtOH (0.25 v/w)
Add EtOH (0.1 v/w)
Add heptane (12 v/w)
Add heptane (4 v/w)
Cool to 20 °C and stir for 2 h
Filter the suspension
Cool to 20 °C and stir for 0.5 h
Filter the suspension
Rinse reactor and wash filter cake with a mixture of
toluene/EtOH/heptane 1:0.04:2 (total 5 v/w)
Re-slurry the filter cake with 10% citric acid/MeCN
(85:15 v/v, total 10 v/w) and filter
Re-slurry filter cake with H2O (10 v/w) and filter
Re-slurry filter cake with heptane (10 v/w) and filter
Dry wet product
Rinse reactor and wash filter cake with heptane (2 v/w)
Dry wet product
PMI =47.1
PMI = 7.3
Scheme 3. Comparison of first process used in scale-up and second process after
additional streamlining.
Scope of the newly developed conditions
ACS Paragon Plus Environment
22

Page 23 of 35
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
The applicability of the reaction conditions developed for the actual intermediate of
EMA401 were then tested on a series of aromatic aldehydes (Table 2). Since the work-
up procedure was optimized for the desired cinnamic acid 3 and not applicable for all
types of substrates, screening products were not isolated and conversions are stated
according to LC-MS or HPLC data. The conditions gave good to excellent conversion
for electron-rich (entry 1–5) and electron-poor (entry 6–13) aromatic compounds, as well
as heterocyclic substrates (entry 14+15). Like for the actual substrate, substituents in
the ortho-position to the aldehyde were well tolerated. Interestingly, pyrrole-2-
carboxaldehyde (entry 16) did not afford the desired product but underwent full
conversion to the corresponding analogue of intermediate V. Terephthalaldehyde (entry
17) gave the corresponding dicarboxylic acid in 24%, with many other intermediates
observed by LCMS.
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 2. Screening of various aldehydes using the newly developed conditions in
toluene/morpholine.
ACS Paragon Plus Environment
23

Organic Process Research & Development
Page 24 of 35
1
2
3
4
5
6
7
8
Entry
1
Product
Conversion / a%1)
Reaction Time / h
OBn
O
>992)
2
O
OH
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
O
2
3
993)
3
4
OH
O
902)
OH
O
703)
482)
1.5
3
O
O
4
5
OH
Br
O
OH
Me2N
Br
O
O
6
7
922)
753)
2.5
3
OH
OH
Cl
Cl
O
8
9
873)
913)
22
3
OH
F
O
O
OH
N
H
ACS Paragon Plus Environment
24

Page 25 of 35
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
O
10
11
873)
3
O
OH
O
N
H
O
813)
22
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
OH
MeO
O
O
12
13
703)
3
2
OH
N
O
892)
OH
O2N
O
O
O
O
14
15
16
822)
4
OH
OH
OH
773)
22
3
S
Dicarboxylated
product observed3),
H
N
4)
H
N
O
O
OH
OH
O
17
243)
3
OH
HO
O
ACS Paragon Plus Environment
25

Organic Process Research & Development
Page 26 of 35
1
2
3
4
5
6
7
8
9
Conditions: Aldehyde (1.0 eq., 20 mmol) in toluene (20 mL) with morpholine (0.2 eq.)
and malonic acid (1.3 eq.) heated to 110 °C for 3 h. 1) For the conversion, the
integration of all peaks except the solvent peak and the injection peak were taken into
account, 2) Samples analyzed with HPLC at 228 nm, 3) Samples analyzed with LCMS
at a wavelength range of 210–450 nm; 4) structure proposed based on LCMS 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
Experimental Section: Synthesis of (2E)-3-[2-(Benzyloxy)-3-methoxyphenyl]prop-2-
enoic acid (3)
Scale-Up Process run on 2 times 12.5 kg
Aldehyde 4 (1.0 eq.) and malonic acid (5) (1.3 eq) were suspended in toluene (5 v/w
wrt 4) and heated to IT = 110±5 °C. At IT = 55–65 °C, morpholine (0.2 eq) in toluene (1
v/w wrt 4) was added over 3–5 min. The mixture was heated using a Dean-Stark trap at
IT = 110±5 °C for 3 h and cooled to IT = 75–80 °C. EtOH (0.25 v/w wrt 4) was then
added over 5 min, followed by heptane (12 v/w wrt 4) added over 30 min, and the
mixture was cooled to 20 °C over 2 h. A white precipitate was formed. The mixture was
stirred at 20 °C for 2 h and filtered.
ACS Paragon Plus Environment
26

Page 27 of 35
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
The reactor was rinsed with a mixture of toluene/EtOH/heptane 1:0.04:2 (v/v/v, total 5
v/w wrt 4). The filter cake was dried under nitrogen and suspended in a mixture of
10wt% citric acid/acetonitrile 85:15 (v/v; total 10 v/w wrt 4), stirred for 10 min, and
filtered. The filter cake was suspended in H2O (10 v/w wrt 4), stirred for 10 min, filtered
and dried under nitrogen. The white solid was suspended in heptane (10 v/w wrt 4),
stirred for 10 min, filtered and dried under vacuum at 50 °C until dryness affording the
cinnamic acid 3 as a white, voluminous powder (84–87% yield).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
M.p.: 156 °C;19 1H NMR (400 MHz, (CD3)2SO): δ = 3.86 (s, 3 H), 4.98 (s, 2 H), 6.44 (d,
J = 16.1 Hz, 1 H), 7.06–7.18 (m, 2 H), 7.28–7.40 (m, 4 H), 7.40–7.46 (m, 2 H), 7.81 (dd,
J = 16.3, 1.4 Hz, 1 H), 12.31 ppm (s, 1 H); 13C NMR (101 MHz, (CD3)2SO): δ = 55.9,
74.6, 114.5, 118.8, 120.1, 124.5, 128.1, 128.2, 128.3 (2 C), 128.3 (2 C), 137.0, 138.3,
+
146.1, 152.8, 167.6 ppm; ESI-MS: m/z = 285.1 ([M + H]+, calcd for C17H17O4 : 285.1)
Final process after streamlining
In a double jacketed reactor equipped with a Dean-Stark trap, aldehyde 4 (20 g, 1.0
eq.) and malonic acid (5) (10.31 g, 1.2 eq.) were suspended in toluene (55 mL) and
ACS Paragon Plus Environment
27

Organic Process Research & Development
Page 28 of 35
1
2
3
4
5
6
7
8
9
heated to JT = 110 °C with a gradient of 0.5 K/min. At IT = 55 °C, a solution of
morpholine (1.19 mL, 0.165 eq.) in toluene (5 mL) was added dropwise. The mixture
was stirred for 3 h at JT = 110 °C (IPC showed <1a% of aldehyde 4), the Dean-Stark
trap was removed and the mixture cooled to IT = 80 °C with a gradient of 0.5 K/min.
EtOH (2 mL) was added, followed by the addition of heptane (80 mL) over 30 min,
maintaining an IT of 70–80 °C. After around 1/3 of the heptane volume, precipitation
was observed. The suspension was then cooled to 20 °C with 0.5 K/min, stirred for 30
min, and filtered. The filter cake was washed with heptane (40 mL) to afford the product
3 as a white solid (23 g, 98%), containing approx. 4mol% morpholine.
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
Conclusion
In summary, we developed conditions for the Knoevenagel-Doebner condensation in
toluene with morpholine as the organocatalyst, avoiding the undesired pyridine and
piperidine. To our knowledge, these conditions have not been used for a Knoevenagel-
Doebner condensation. These novel conditions were successfully scaled to 2 batches of
ACS Paragon Plus Environment
28

Page 29 of 35
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
12.5 kg each. Cinnamic acid 3 could be obtained meeting all specifications. After
additional process development, the conditions were further improved to yield cinnamic
acid 3 in a 98% yield. The optimized conditions in toluene, with morpholine as
organocatalyst appear to be generally applicable for a broad range of substrates, as
shown in a screening with various aromatic aldehydes.
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
ASSOCIATED CONTENT
The following files are available free of charge.
Solubility data and NMR spectra of cinnamic acid 3, HPLC or LCMS data for the
screening of various aromatic compounds (PDF)
AUTHOR INFORMATION
Corresponding Author
ACS Paragon Plus Environment
29