Evaluation of Amino Nitriles and an Amino Imidate as Organo?-catalysts in Aldol Reactions

Article abstract of DOI:10.1055/s-0039-1690150

The efficiency of l -valine and l -proline nitriles and a tert -butyl?- l -proline imidate as organocatalysts for the aldol reaction have been evaluated. l -Valine nitrile was found to be a syn -selective catalyst, while l -proline nitrile was found to be anti -selective, and gave products in modest to good enantioselectivities. tert -Butyl l -proline imidate was found to be a very efficient catalyst in terms of conversion of starting reagents to products, and gave good anti -selectivity. The enantioselectivity of the tert -butyl l -proline imidate was found to be good to excellent, with products being formed in up to 94percent enantiomeric excess.

Full text of DOI:10.1055/s-0039-1690150

SYNTHESIS
0
0
3
9
-
7
8
8
1
1
4
3
7
-
2
1
0
X
©
Georg Thieme Verlag Stuttgart · New York
2019, 51, A–G
paper
A
en
Synthesis
N. Vagkidis et al.
Paper
Evaluation of Amino Nitriles and an Amino Imidate as Organo-
catalysts in Aldol Reactions
Nikolaos Vagkidis
Alexander J. Brown
catalyst
(10 mol%)
O
OH
O
OH
O
O
(S)
(S)
(
R) Ar
+
(S) Ar
Paul A. Clarke*
0
0
0
0-0
0
0
3-3
9
5
2-
3
5
9
X
H
Ar
cyclohexane
9
examples
Department of Chemistry, University of York, Heslington, York
YO10 5DD, North Yorks, UK
up to 95% yield
up to 94% ee
paul.clarke@york.ac.uk
New catalysts
Ot-Bu
N
CN
N
H2N
CN
H
H
NH
Received: 13.06.2019
Accepted after revision: 18.07.2019
Published online: 08.08.2019
DOI: 10.1055/s-0039-1690150; Art ID: ss-2019-t0330-op
or
O
OH OH
O
N
H
CN
H2N
CN
1
2
H
H
O
(20 mol%)
+
+
OH
Abstract The efficiency of L-valine and L-proline nitriles and a tert-
butyl L-proline imidate as organocatalysts for the aldol reaction have
been evaluated. L-Valine nitrile was found to be a syn-selective catalyst,
while L-proline nitrile was found to be anti-selective, and gave products
in modest to good enantioselectivities. tert-Butyl L-proline imidate was
found to be a very efficient catalyst in terms of conversion of starting
reagents to products, and gave good anti-selectivity. The enantioselec-
tivity of the tert-butyl L-proline imidate was found to be good to excel-
lent, with products being formed in up to 94% enantiomeric excess.
OH
O
OH OH
H2O
H
H
OH
OH
anti:syn ratio 1.5:1–2:1
Scheme 1 Amino nitrile-catalysed formation of 2-deoxy-D-ribose
formation of the mixed anhydride and treatment with
methanolic ammonia. Dehydration of the amide to the ni-
Key words asymmetric synthesis, organocatalysis, aldol reaction,
amino nitrile, amino imidate
trile was achieved in 90% yield using TFAA and Et N. Finally
3
the Cbz-group was removed in 91% yield by hydrogenation
The synthesis and evaluation of new small molecules as
organocatalysts has become an important endeavour. From
over a Pd(OH) /C catalyst in EtOAc to give 1 (Scheme 2).
2
1
1) EtO2CCl, Et3N
the initial development of proline as a catalyst for the aldol
2
reaction by List and Barbas, and the imidazolidinones by
2) NH /MeOH (7 N)
THF, 87%
CbzHN
CO2H
3
CbzHN
CONH2
3
3
4
MacMillan for the Diels–Alder reaction, many novel contri-
butions have been made. The pyrrolidine ring of proline is
still by far the most abundant scaffold for these catalysts,
TFAA, Et3N
THF, 90%
Pd(OH) /C. H2
2
EtOAc, 91%
H2N
CN
4
CbzHN
CN
with the carboxylic acid being replaced with tetrazoles, si-
1
5
5
6
7
lyl ethers of tertiary alcohols, esters, and amides, all of
which bring subtle changes in catalytic ability and the type
of transformation which can be catalysed. Recently, we re-
ported the use of amino nitriles as catalysts for the forma-
tion of 2-deoxy-D-ribose under aqueous, potentially pre-
Scheme 2 Synthesis of L-valine nitrile 1
Boc-L-proline was converted to the primary amide in
85% yield by formation of the mixed anhydride and treat-
ment with methanolic ammonia. Dehydration of the amide
to the nitrile was achieved in 89% yield using TFAA and
8
biotic conditions (Scheme 1). The ability of amino nitriles
to catalyse this reaction inspired us to evaluate them as
more general aldol catalysts in organic solvents under more
conventional reaction conditions.
Et N. Removal of the Boc-group was achieved by treatment
3
with TFA in CH Cl at 0 °C to generate the TFA salt of 2 in a
2
2
Amino nitriles 1 and 2 were prepared from the parent
carbamate-protected amino acids (Schemes 2 and 3). Cbz-L-
valine was converted to the primary amide in 87% yield by
93% yield. The amine was free-based immediately before
use by stirring with solid NaHCO in CH Cl (Scheme 3).
3
2
2
©
2019. Thieme. All rights reserved. — Synthesis 2019, 51, A–G
Georg Thieme Verlag KG, Rüdigerstraße 14, 70469 Stuttgart, Germany

B
Synthesis
N. Vagkidis et al.
Paper
1) EtO2CCl, Et3N
O
OH
(
S)
N
CO2H 2) NH3/MeOH (7 N)
N
CONH2
1
(S)
Boc
Boc
6
THF, 85%
7
O
2
H N
CN
O
NO2
(10 mol%)
12-syn (major)
TFAA, Et3N
THF, 89%
TFA, CH Cl2
H
2
O
OH
solvent
NO2
N
CN
93%
N
H
CN
Boc
8
2·TFA
10
11
NaHCO3,
CH2Cl2
NO2
12-anti (minor)
Ot-Bu
63%
N
H
CN
N
H
Scheme 4 Aldol reactions catalysed by H N-L-Val-CN 1
NH
2
2
9·TFA
Scheme 3 Synthesis of L-proline nitrile 2
While it was disappointing that L-valine nitrile 1 was
not a better catalyst, it was very interesting that the syn-
diastereomer 12-syn was the major product under all con-
ditions studied. The formation of the syn-diastereomer as
the major product is most unusual in organocatalytic aldol
reactions which proceed via enamine catalysis, as the anti-
However, the Boc-deprotection of 8 was more challeng-
ing than expected, as the clean formation of 2·TFA was de-
pendent on the batch of TFA used. Some batches of TFA gen-
erated 2·TFA cleanly, while other batches also generated a
1
11
side product, which was identified by H NMR and MS as
diastereomer usually dominates. In order to determine if
9
the imidate 9·TFA. We rationalised that if the TFA was wet,
this diasteroeselectivity was a general feature of amino
nitrile catalysis L-proline nitrile 2 was investigated. It was
also rationalised that any formation of a L-proline nitrile 2/
4-nitrobenzaldehyde adduct would be less problematic due
to it being an iminium species rather than an imine and so
it would be slower to form and more easily hydrolysed.
Reactions were conducted with 10 mol% of catalyst 2,
with 5 equivalents of cyclohexanone to 1 equivalent of
4-nitrobenzaldehyde in a range of solvents (Table 1).
water could intercept the t-Bu-cation to form t-BuOH,
which then underwent an acid-catalysed addition to the ni-
trile 2·TFA to form imidate 9·TFA. Conducting the TFA-me-
diated deprotection in the presence of t-BuOH provided a
reliable method for the synthesis of imidate 9·TFA, and also
provided us with an additional new catalyst class to study.
The first reaction which was investigated was the stan-
dard test reaction for any new organocatalyst: the aldol re-
action of cyclohexanone with substituted benzaldehydes
4,7,10
(
Scheme 4).
All reactions used 10 mol% of catalyst, with
Table 1 Aldol Reactions Catalysed by HN-L-Pro-CN 2
5
equivalents of cyclohexanone to 1 equivalent of 4-nitro-
O
OH
benzaldehyde in a range of solvents. When L-valine nitrile 1
was used conversion to the aldol adduct was <15% for a
wide range of solvents covering both dipolar aprotic and
non-polar solvents. However, very interestingly the syn:anti
ratio of the products favoured the syn-isomer in all cases
(
S)
2
(R)
O
N
CN
O
H
(
10 mol%)
NO2
NO2
H
12-anti (major)
solvent
O
OH
NO2
(
S)
(CH Cl 4.5:1; DMF 2.3:1; 1.4-dioxane 1.3:1; THF 3.8:1; tol-
10
11
2
2
(S)
uene 5.3:1; cyclohexane 3.0:1; cyclohexanone 5.3:1) with
the highest ratio being in EtOAc >25:1. Due to the low con-
versions the enantioselectivity of these reactions were not
determined. It was rationalised that one possibility for the
low conversions was that the amino nitrile catalyst was be-
ing trapped as the 4-nitrobenzaldehyde imine. In order to
try and hydrolyse any imine back to 4-nitrobenzaldehyde
and amino nitrile, water (10 mol%) was added to the reac-
tion in toluene, which had provided the greatest conver-
sion. This had the marked effect on increasing the syn:anti
ratio from 5.3:1 to >25:1, but had no effect on the conver-
sion. The introduction of water (10 mol%) and TsOH (10
mol%) to this system did not improve the conversion and
gave products in a syn:anti ratio of >25:1. In this instance,
the enantioselectivity of the reaction was determined by
HPLC and the syn-product 12-syn was found to have a 34%
ee. The absolute stereochemistry of 12-syn was determined
12-syn (minor)
b
b
Entry Solvent
Conversion (%)a anti:syna % ee anti
% ee syn
1
2
3
4
5
6
7
8
9
CH
2
Cl
2
43
7
4.0:1
2.5:1
4.8:1
2.9:1
1.7:1
3.9:1
3.9:1
4.8:1
4.0:1
13
20
11
20
–c
11
18
11
20
–c
DMF
1,4-dioxane 55
MeCN
DMSO
THF
11
3
39
51
75
75
40
23
20
13
12
15
6
EtOAc
toluene
cyclohex-
ane
0
a
1
Determined by 400 MHz H NMR analysis, by integration of the aldehyde
proton and the carbinol protons of the aldol products.
10a
b
to be S,S, by comparison to the literature.
Determined by HPLC chiralpak IB column (see Supporting Information).
c
Not determined.
©
2019. Thieme. All rights reserved. — Synthesis 2019, 51, A–G

C
Synthesis
N. Vagkidis et al.
Paper
The results in Table 1 show that L-proline nitrile 2 is a
much more efficient catalyst than L-valine nitrile 1 in terms
of converting starting materials into products. Conversions
of 75% were reached in non-polar hydrocarbon solvents
such as toluene and cyclohexane (Table 1, entries 8 and 9).
The increased conversion is attributed to the greater cata-
lytic ability of the secondary amine of 2 compared to the
primary amine of 1, for the reasons mentioned earlier. In-
terestingly, the anti-diastereomer 12-anti was the major
adduct formed in all cases, showing that it is not the amino
nitrile function alone which was responsible for the switch
to the syn-diastereomer for L-valine nitrile 1. The difference
in the major diastereomer is probably down to the confor-
mation adopted by the enamine and its attack trajectory on
the aldehyde, to minimise steric interactions. The enantio-
selectivity of the reaction remained reasonably constant in
all solvents studies (~10–20%) with the exception of THF
Table 2 Aldol Reactions Catalysed by tert-Butyl Proline Imidate 9
O
OH
9
Ot-Bu
(S)
(R)
N
O
H
O
NH
10 mol%)
(
NO2
H
12-anti (major)
solvent
O
OH
NO2
(
S)
10
11
(
S)
NO2
12-syn (minor)
b
b
Entry Solvent
Conversion (%)a anti:syna % ee anti
% ee syn
1
2
3
CH
THF
toluene
cyclohexane
2
Cl
2
61
57
85
5.6:1
5.8:1
4.6:1
5.3:1
69
46
58
76
45
36
27
51
4
100
a
1
(entry 6), which generated 12-anti product in 40% ee. In
Determined by 400 MHz H NMR analysis, by integration of the aldehyde
proton and the carbinol protons of the aldol products.
general, the % ee of the anti-diastereomer was slightly
greater than that of the syn-diastereomer. The absolute
stereochemistry of the aldol products was determined as
b
Determined by HPLC chiralpak IB column (see Supporting Information).
(
S,R)-12-anti and (S,S)-12-syn by comparison with litera-
10a
ture data.
Table 3 Amino Imidate 9-Catalysed Aldol Reactions
Disappointingly it seems that the amino nitriles studied
are not useful catalysts for the formation of aldol products.
This is probably due to the lack of functionality, which can
allow for the controlled association or organisation of the
reagents via hydrogen bonding (as in the case of proline) or
large steric buttresses (as in the case of diaryl proline silyl
ethers) to control the facial selectivity of the attack.
tert-Butyl proline imidate 9, however, does contain both
a potential hydrogen bond donor in the form of the imidate
NH, and a sterically bulky t-Bu group and so this catalyst
could provide higher levels of enantioselectivity in the aldol
reaction. t-Bu-proline imidate 9 was initially screened using
our standard conditions: 10 mol% of catalyst, 5 equivalents
of cyclohexanone to 1 equivalent of 4-nitrobenzaldehyde in
several solvents (Table 2).
Pleasingly, amino imidate 9 is a much better catalyst
than amino nitrile 2 for the promotion of aldol reactions. As
can be seen from Table 2 the conversions are all substantial-
ly better, with hydrocarbon solvents like toluene (Table 2,
entry 3) and cyclohexane (entry 4) providing 85% and 100%
conversion of starting material to aldol product. The an-
ti:syn ratio is modest and very similar irrespective of the
solvent used, with the anti-diastereomer 12-anti being the
major product in all cases. Significantly, the enantioselec-
tivities were also much higher when amino imidate 9 was
used as a catalyst, with the highest for both the anti- and
syn-diastereomers (at 76% ee and 51% ee, respectively)
when the reaction was run in cyclohexane (entry 4). With
these encouraging results it was decided to screen a num-
ber of different aldehydes in the amino imidate 9-catalysed
reaction (Table 3).
O
OH
9
Ot-Bu
(S)
(R) Ar
N
H
O
NH
(10 mol%)
O
1
4-anti (major)
H
Ar
cyclohexane
O
OH
(
S)
10
13
(S) Ar
14-syn (minor)
a
14 anti:syna
b
Entry
13 Ar
Conversion (%)
% ee anti
a
b
c
d
e
f
2-NO
3-NO
2
C
6
6
H
4
100
100
98
96
94
100
99
90
69
0
4.7:1
3.0:1
5.0:1
3.0:1
2.7:1
7.0:1
2.5:1
3.0:1
3.5:1
–
75
63
76
67
57
69
71
61
67
–
2
C
H
4
2-ClC
3-ClC
4-ClC
6
6
6
H
H
H
4
4
4
2-BrC H
6
4
4
4
g
h
i
3-BrC
4-BrC
Ph
6
H
H
6
j
6 4
4-MeOC H
a
1
Determined by 400 MHz H NMR analysis, by integration of the aldehyde
proton and the carbinol protons of the aldol products.
Determined by HPLC chiralpak IB column (see Supporting Information).
b
As can be seen from Table 3, amino imidate 9 was able
to efficiently catalyse the aldol reaction of cyclohexanone
with a number of differently substituted aryl aldehydes
13a–i to afford the aldol products 14a–i. Excellent conver-
sions were obtained regardless of whether the aldehyde
was substituted in the 2, 3, or 4-positions with an electron-
©
2019. Thieme. All rights reserved. — Synthesis 2019, 51, A–G

D
Synthesis
N. Vagkidis et al.
Paper
withdrawing substituent (Table 3, entries a–h). However,
no reaction was observed when electron-donating 4-MeO
group was introduced (entry j). Unsubstituted, electronical-
ly neutral, benzaldehyde had the lowest conversion of those
aldehydes that underwent reaction at only 69% (entry i)
compared to the +90% conversions of the other aldehydes.
The reaction was modestly anti-selective in all cases, while
the enantioselectivities were modest to good with the high-
est being 75% ee (entry a) and 76% ee (entry c). In general,
higher enantioselectivities were seen for aldehydes with 2-
substitution than for 3- or 4-substitution (compare entries
a and b, entries c, d, and e), and with the exception of
portion of anti-product formed in the reaction. However,
the reduced rate of reaction at 0 °C, does lead to a reduced
conversion to adduct over the same period of time.
The final investigation focused on the use of cyclopenta-
none (15) and pyran-4-one (17) in the amino imidate 9-
promoted reaction with 4-nitrobenzaldehyde (11) (Scheme
5).
O
OH
9
Ot-Bu
N
O
H
O
NH
NO2
16-syn (major) 60% e.e.
OH
(
10 mol%)
H
O
4-chlorobenzaldehyde (entry e) were all above 60% ee.
cyclohexane
0 °C
NO2
In order to determine if the enantioselectivity could be
15
11
64%
syn:anti = 1.9:1
increased further, the reactions were run at 0 °C. The reac-
tion of cyclohexanone, 4-nitrobenzaldeyde in cyclohexane
at 0 °C, catalysed by 9 proceeded with a conversion of 40%
and a anti:syn ratio of 5.7:1. However, the enantioselectivi-
ty of the anti-product 12-anti was found to be 94% ee. En-
couraged by this significant increase in enantioselectivity
the use of other aldehydes was investigated. These results
can be seen in Table 4.
NO2
16-anti (minor) 51% e.e
9
O
OH
Ot-Bu
N
H
O
O
O
NH
(
10 mol%)
O
NO2
H
18-anti (major) 74% e.e
cyclohexane
0 °C
NO2
O
OH
7
4%
17
11
anti:syn = 9.1:1
Table 4 Amino Imidate 9-Catalysed Aldol Reactions at 0 °C
O
NO2
18-syn (minor) 23% e.e.
O
OH
9
Scheme 5 Aldol reaction of cyclopentanone and pyran-4-one with 4-
nitrobenzaldehyde catalysed by amino imidate 9
Ot-Bu
(S)
(
R)
Ar
N
H
O
NH
(10 mol%)
O
14-anti (major)
H
Ar
cyclohexane
°C
O
OH
Cyclopentanone (15) underwent aldol condensation to
generate aldol adducts 16-syn and 16-anti, with the syn-di-
astereomer dominating. The major product 16-syn was
formed in 60% ee, while the minor product 16-anti was
formed in 51% ee. The use of pyran-4-one (17) as the aldol
donor resulted in the formation of 18-anti as the major dia-
stereomer in 74% ee, while the minor 18-syn-diastereomer
was formed in 23% ee. The ratio of anti:syn was a good
0
(
S)
1
0
13
(S) Ar
1
4-syn (minor)
Conversion (%)a
14 anti:syna
% ee anti
b
Entry 13 Ar
a
b
c
2-NO
3-NO
2
2
C
6
H
H
4
4
87
80
46
39
47
47
54
57
4.8:1
5.7:1
6.8:1
5.3:1
4.8:1
6.6:1
4.8:1
5.8:1
3.8
82
51
79
72
77
69
74
76
73
C
6
9.1:1.
2-ClC
3-ClC
4-ClC
6
6
6
H
H
H
4
An investigation has been conducted into the catalytic
d
e
f
4
4
efficiency of amino nitriles and an amino imidate for aldol
condensations. L-Valine nitrile 1 was not efficient as a cata-
lyst in terms of reaction yields and enantioselectivity, al-
though it did exhibit unusual syn-diastereomer selectivity.
L-Proline nitrile 2 was more efficient in terms of both con-
version and the enantioselectivity of the products, with the
major anti-diastereomer being formed in up to 76% ee,
when cyclohexane was used as the reaction solvent. How-
ever, the serendipitous discovery of L-proline imidate 9, and
its use as an organocatalyst led to synthetically useful con-
versions and anti:syn ratios of products in line with other
organocatalysts. The enantioselectivities of the major anti-
products were good (60–75%). The enantioselectivity of the
L-proline imidate-catalysed reaction and the anti:syn ratio
of the products could be increased further when the reac-
tion was run at 0 °C, with the anti-product being formed in
2-BrC
3-BrC
4-BrC
Ph
6
6
6
H
H
H
4
4
4
g
h
i
10
a
1
Determined by 400 MHz H NMR analysis, by integration of the aldehyde
proton and the carbinol protons of the aldol products.
Determined by HPLC chiralpak IA, IBN-5, and IC columns (see Supporting
b
Information).
Reducing the temperature of the reaction to 0 °C does
have a beneficial effect on % ee in almost all cases, raising it
by as much as 18% in the case of aldehyde 11. It also has a
beneficial effect on the anti:syn ratio, increasing the pro-
©
2019. Thieme. All rights reserved. — Synthesis 2019, 51, A–G

E
Synthesis
N. Vagkidis et al.
Paper
1
3
as high as 94% ee. These enantioselectivities are on a par
with other common proline-derived catalysts, which have
been used in similar aldol reactions. Proline amides gave
products with % ees in the mid-70% to high 90% range, de-
C NMR (400 MHz, DMSO-d ):  = 173.2, 156.2, 137.2, 128.4, 127.8,
6
1
27.3, 65.4, 60.1, 30.2, 19.4, 18.0.
+
HRMS (ESI): m/z [M + Na] calcd for C13H18N O Na: 273.1210; found:
2
Spectroscopic data are identical to that reported in the literature.8
2
3
73.1210.
12
pending on the amine used. Proline tetrazole gave % ees
4
up to the low 90% range, whereas ring-substituted prolines
with parent carboxylic acid gave products with % ees up to
Cbz-L-Valine Nitrile 5
12
the high 90% range. Amino imidates based on proline are a
new class of organocatalyst, which have the potential to be
efficient and highly enantioselective aldol catalysts. Further
work is underway to modify the proline imidate in order to
increase the enantioselectivity further.
A flask was flame dried and was allowed to cool at r.t. under a N at-
2
mosphere. Cbz-L-valine amide 4 (1.75 g, 7.00 mmol), dissolved in an-
hyd THF (30 mL) was added to the flask. The flask was cooled at 0 °C,
and Et N (2.18 mL, 2.2 equiv) was added and the solution was stirred.
3
After 30 min, TFAA (1.50 mL, 10.5 equiv) was added and the reaction
was continued to be stirred at 0 °C for 1 h and a further 17 h at r.t. The
reaction was deemed complete by TLC (90:10 DCM/MeOH) and the
stirring was stopped. The solvent was removed in vacuo and the
crude oil was re-dissolved in EtOAc. The crude mixture was washed
with aq 2 M HCl and extracted with EtOAc (3 × 10 mL). The organic
layers were combined and washed with sat. aq NaHCO
then washed with brine. The aq brine layer was extracted with EtOAc
(1 × 10 mL). The organic extracts were combined, dried (MgSO ), fil-
tered, and the solution was concentrated in vacuo to give the crude
product as a red translucent oil. The crude product was then further
purified by column chromatography (90:10 hexane/EtOAc) to afford
the pure Cbz-protected amino nitrile 5 as a red solid; yield: 1.47 g
Unless otherwise noted, all compounds were bought from commer-
cial suppliers and used without further purification. NMR spectra
were recorded on a Jeol ECS-400 spectrometer at ambient tempera-
ture; chemical shifts are quoted in parts per million (ppm) and were
(3 × 10 mL),
3
1
referenced as follows: CDCl 7.26 ppm for H NMR; CDCl3 77.0 ppm
4
3
for 13C NMR. Coupling constants (J) are quoted in Hertz. IR absorbanc-
es were recorded on a PerkinElmer UATR Two FT-IR spectrometer us-
ing NaCl plates. Mass spectrometry was performed by the University
of York mass spectrometry service using electron spray ionisation
1
4
20
3
–1
(
ESI) technique. Optical rotations were carried out using a JASCO-
(90%, 6.30 mmol); mp 49–51 °C (Lit. mp 53 °C); []
D
(deg cm g
–1
2
–1
dm–1) –43.07 (c 1.0 g cm–3 in MeOH) {[]D25 (deg cm
3
g
–1 dm−1) Lit.
8
–37.3
DIP370 polarimeter and []D values are given in 10 deg·cm ·g . TLC
was performed on aluminum sheets coated with Merck Silica gel 60
F254. The plates were developed using ultraviolet light, basic aq
KMnO4 or ethanolic anisaldehyde. Liquid chromatography was per-
formed using forced flow (flash column) with the solvent systems in-
dicated. The stationary phase was silica gel 60 (220–240 mesh) sup-
plied by Sigma-Aldrich. Anhydrous solvents were acquired from a
PureSolv PS-MD7 solvent tower. High-performance liquid chromatog-
raphy (HPLC) was performed using an Agilent 1200 series instrument
using the chiral columns indicated and a range of wavelengths from
–3
(c 0.97 g cm in MeOH)}.
IR (ATR): 3298 (N–H),3064, 3032, 2970, 2930, 2877 (C–H), 2459 (CN),
1686 (C=O), 1213 (C–N), 1176 cm
–1
(C–O).
1
H NMR (400 MHz DMSO-d ):  = 8.22 (1 H, br d, J = 8.0 Hz), 7.39–7.31
6
(
(
1
5 H, m), 5.09 (2 H, s), 4.40 (1 H, app t, J = 8.0 Hz), 1.98 (1 H, m), 1.00
3 H, d, J = 6.8 Hz), 0.94 (3 H, d, J = 6.8 Hz).
3
C NMR (400 MHz, DMSO-d ):  = 155.5, 135.7, 128.8, 128.6, 128.4,
6
117.8, 67.9, 49.1, 31.9, 18.7, 18.0.
+
210–280 nm for detection.
HRMS (ESI): m/z [M + Na] calcd for C13H16N O Na: 255.1104; found:
2
2
255.1105.
Cbz-L-Valine-Amide 4
Spectroscopic data are identical to that reported in the literature.8
A flask was flame dried and was allowed to cool at r.t. under a N at-
2
mosphere. Cbz-L-valine 3 (2.0 g, 7.96 mmol) was added to the flask. To
L-Valine Nitrile 1
this flask Et N (1.2 mL, 1.1 equiv) and anhyd THF (40 mL) were added.
3
A flask was flame dried and was allowed to cool at r.t. under a N at-
2
The solution was cooled at 0 °C and stirred. After 10 mins, ethyl chlo-
roformate (0.8 mL, 1 equiv) was added and the reaction was contin-
mosphere. Cbz-L-valine nitrile 5 (200 mg, 0.86 mmol) in EtOAc (7.5
mL) and Pearlman’s reagent (20% b.w., 60 mg, 0.1 equiv) were placed
in the flask and the flask was evacuated. Then the flask was placed
ued to be stirred at 0 °C. After 1 h, NH in MeOH (7 N) was added (1.66
3
mL, 1.5 equiv) and the mixture was continued to be stirred at 0 °C for
another 1 h. After 1 h, the reaction was allowed to warm at r.t. and
was continued to be stirred. After a further 17 h, the reaction was
deemed complete by TLC (90:10 DCM/MeOH) and the stirring was
stopped. The solvent was removed in vacuo and the white precipitate
under a H atmosphere (60 psi) and the reaction was stirred. After 1.5
2
h of stirring, the reaction was deemed complete by TLC (95:5
DCM/MeOH) and the stirring was stopped. The mixture was filtered
through a pad of Celite and the Celite was washed thoroughly with
EtOAc (50 mL). Aq 4 M HCl in 1,4-dioxane (1.0 mL) was added and the
mixture was stirred for 30 min turning the solution cloudy. Upon
evaporation, the salt of the amine was isolated as a white-yellow sol-
id. The free amine 1 was liberated by dissolving the salt in DCM and
was filtered and washed with ice cold H O to give the pure Cbz-pro-
2
tected amide 4 as a white solid; yield: 1.73 g (87%, 6.92 mmol); mp
1
3
20
3
–1
–1
2
1
06–209 °C (Lit. mp 205–208 °C); [] (deg cm g dm ) +24.7 (c
D
–1
.0 g cm in DMF) {[]D25 (deg cm g dm–1) Lit. +25.0 (c 1.0 g cm
–3
3
8
–3
stirring over NaHCO for 30 min before filtering and concentrating in
3
in DMF)}.
20
3
vacuo, as a yellow oil; yield: 76 mg (91%, 0.78 mmol); [] (deg cm
D
–
1
dm ) –6.37 (c 1.0 g cm in DCM) {[]D (deg cm g dm ) Lit.8
–1
–3
25
3
–1
–1
IR (ATR): 3374, 3315 (N–H), 3201, 3030, 2972, 2958, 2895, 2872 (C–
g
H), 1681, 1654 (C=O), 1243 cm–1 (C–O).
–3
–8.3 (c 0.83 g cm in DCM)}.
1
IR (ATR): 3384 (N–H), 2228 (CN), 1098 cm–1 (C–N).
H NMR (400 MHz, DMSO-d ):  = 7.38–7.28 (6 H, m), 7.16 (1 H, d, J =
6
8
1
.9 Hz), 7.03 (1 H, br s), 5.03 (2 H, s), 3.80 (1 H, dd, J = 8.9, 6.6 Hz),
.99–1.28 (1 H, app oct, J = 6.6 Hz), 0.86 (3 H, d, J = 6.6 Hz), 0.83 (3 H,
1
H NMR (400 MHz, CDCl ):  = 3.52 (1 H, d, J = 5.6 Hz), 1.93 (1 H, dspt,
3
J = 6.8, 5.6 Hz), 1.64 (2 H, br s), 1.07 (3 H, d, J = 6.8 Hz), 1.06 (3 H, d, J =
6
d, J = 6.6 Hz).
.8 Hz).
©
2019. Thieme. All rights reserved. — Synthesis 2019, 51, A–G

F
Synthesis
N. Vagkidis et al.
Paper
1
3
C NMR (400 MHz, CDCl ):  = 121.1, 49.7, 32.8, 18.8, 17.5.
L-Proline Nitrile Trifluoroacetate Salt (2·TFA)
A flask was flame dried and was allowed to cool at r.t. under a N at-
2
3
+
HRMS (ESI): m/z [M + H] calcd for C H N : 99.0917; found: 99.0919.
5
11
2
Spectroscopic data are identical to that reported in the literature.8
mosphere. Boc-L-proline nitrile 8 (364 mg, 1.7 mmol) and TFA (3.6
mL, 25 equiv) in anhyd DCM (5 mL) were added to the flask and the
flask was cooled at 0 °C. The solution was stirred until the reaction
was deemed complete by TLC (90:10 DCM/MeOH). The stirring was
Boc-L-Proline Amide 7
A flask was flame dried and was allowed to cool at r.t. under a N at-
2
stopped and solvent was removed in vacuo. Trituration with Et O pro-
2
mosphere. Boc-L-proline 6 (2.0 g, 9.2 mmol) and anhyd THF (30 mL)
vided the pure TFA salt of L-proline nitrile 2·TFA; yield: 318 mg (93%,
were added to the flask. To this flask, Et N (1.43 mL, 1.1 equiv) was
16
20
3
–1
3
1.58 mmol); mp 90–92 °C (Lit. mp 92–94 °C); []D (deg cm g
added and the solution was stirred at r.t. After 15 min, ethyl chlorofor-
mate (0.86 mL, 1 equiv) was added and the reaction was continued to
dm ) –11.6 (c 1.0 g cm in MeOH) {[]D25 (deg cm g dm ) Lit.8
–1
–3
3
–1
–1
–3
–
16.7 (c 1.0 g cm in MeOH)}.
be stirred at r.t. After 1 h, NH in MeOH (7 N, 2 mL), was added and the
3
IR (ATR): 3323 (N–H), 2943, 2831, 2269 (CN), 1665 cm–1 (C=O).
reaction was continued to be stirred for a further 14 h. After that, the
reaction was deemed complete by TLC (70:30 hexane/EtOAc) and the
stirring was stopped. The solvent was removed in vacuo and the solu-
1
H NMR (400 MHz, CD OD):  = 4.60 (1 H, t, J = 7.4 Hz), 3.62–3.43 (2 H,
3
m), 2.58–2.47 (1 H, m), 2.27–1.97 (3 H, m).
13
tion was washed with H O (10 mL) and extracted with DCM (5 × 10
C NMR (400 MHz, CD OD):  = 161.8 (q, J = 34.7 Hz, CF ), 115.2, 46.8,
3 3
2
mL). The combined organic layers were dried (MgSO ) and the solu-
45.8, 29.9, 23.2.
4
tion was concentrated in vacuo to give the title compound 7 as a
+
HRMS (ESI): m/z [M + H] calcd for C H N : 97.0760; found: 97.0759.
5
9
2
2
5
3
–1
–1
white solid; yield: 1.67 g (85%, 7.8 mmol); []
(deg cm g dm )
D
–1
Spectroscopic data are in agreement with the literature.8
–3
25
3
–1
15
–
44.7 (c 1.0 g cm in MeOH) {[]D (deg cm g dm ) Lit. –42.4 (c
–3
1.0 g cm in MeOH)}.
The free amine 2 was liberated by dissolving the salt in DCM and stir-
IR (ATR): 3344 (N–H), 1676 (C=O), 1164 cm–1 (C–O).
ring over NaHCO for 30 min before filtering and concentrating in vac-
uo; yield: 90 mg (63%, 1.07 mmol).
3
1
H NMR (400 MHz, CDCl ):  = 6.85 (1 H, s), 5.40–6.10 (1 H, m), 4.35–
3
1
H NMR (400 MHz, CD OD):  = 4.07 (1 H, dd, J = 7.9, 4.7 Hz), 3.10–
4.15 (1 H, m), 3.55–3.25 (2 H, m), 2.40–1.80 (4 H, m), 1.45 (9 H, s).
3
2.85 (2 H, m), 2.15 (1 H, m), 2.07–1.74 (3 H, m).
+
HRMS (ESI): m/z [M + Na] calcd for C10H18N O Na: 237.1210; found:
2
3
2
37.1209.
L-Proline Imidate Trifluoroacetate Salt (9·TFA)
Spectroscopic data are in agreement with the literature.8
A flask was flame dried and was allowed to cool at r.t. under a N at-
2
mosphere. Boc-L-proline nitrile 8 (200 mg, 1.02 mmol) dissolved in
TFA (3.55 mL, 45 equiv) was added to this flask and the flask was
cooled at 0 °C. Upon consumption of the starting material (TLC check),
t-BuOH (0.2 mL, 2 equiv) was added and the reaction was allowed to
warm at r.t. The reaction was left stirring overnight. Stirring was
stopped and the solvent was removed in vacuo. Trituration with hot
diisopropyl ether provided the TFA salt of the L-proline imidate 9·TFA;
Boc-L-Proline Nitrile 8
A flask was flame dried and was allowed to cool at r.t. under a N at-
mosphere. Boc-L-proline amide 7 (625 mg, 2.92 mmol) in anhyd THF
2
(
20 mL) and Et N (0.9 mL, 2.2 equiv) were added to the flask. The flask
3
was cooled at 0 °C and stirred. After 30 min of stirring, anhyd TFAA
kept in a dry ampule (1.0 g, 1.5 equiv) was added and the reaction
was continued to be stirred at 0 °C. After 2 h, the reaction was
warmed at r.t. and was continued to be stirred. After a further 16 h,
the reaction was deemed complete by TLC (90:10 DCM/MeOH) and
the stirring was stopped. The solvent was removed in vacuo. The
crude yellow oil was re-dissolved in EtOAc and was washed with aq 2
M HCl and extracted with EtOAc (3 × 10 mL). The organic layers were
25
(deg cm3
g
–1
yield: 217.5 mg (75%, 0.77 mmol); mp 88–90 °C; []
dm ) –47.23 (c 1.0 g cm in DCM).
D
–1
–3
IR (ATR): 3300 (N–H), 2967, 2872, 1658 cm–1 (C=N).
1
H NMR (400 MHz, CD OD):  = 8.00 (1 H, br s), 4.15 (1 H, dd, J = 8.4,
3
6.8 Hz), 3.44–3.32 (2 H, m), 2.48–2.34 (1 H, m), 2.09–1.89 (3 H, m),
1.36 (9 H, s).
combined, washed with sat. aq NaHCO and the aqueous layers were
extracted with EtOAc (3 × 10 mL). The organic layers were combined,
washed with brine and the aq brine layer was extracted with EtOAc (3
3
13
C NMR (400 MHz, CDCl ):  = 167.2, 59.9, 51.4, 51.2, 46.1, 30.1, 29.7.
3
HRMS (ESI): m/z [M + H]+ calcd for C H N O: 171.1492; found:
9
19
2
×
10 mL). The organic layers were combined, dried (MgSO ) and fil-
4
171.1491.
tered. The solution was concentrated in vacuo to give the crude prod-
uct as an orange oil. The crude oil was further purified by column
chromatography (20:80 EtOAc/hexane) to give the title compound 8
The free L-proline imidate 9 was liberated by dissolving the salt in
DCM and stirring over NaHCO for 30 min before filtering and concen-
3
2
0
3
trating in vacuo; yield: 31 mg ( 55%, 0.18 mmol).
IR (ATR): 3300 (N–H), 2967, 2872, 1658 cm–1
.
as a pale yellow oil; yield: 508 mg (89%, 2.60 mmol); []
(deg cm
D
–
1
–1
–3
25
3
–1
–1
g
dm ) –91.15 (c 1.3 g cm in MeOH) {[]D (deg cm
g
dm
)
15
–3
Lit. –95.5 (c 1.3 g cm in MeOH).
1
H NMR (400 MHz, CDCl ):  = 7.44 (1 H, br s), 3.69 (1 H, dd, J = 8.8, 5.6
3
IR (ATR): 2976, 2239 (CN), 1797, 1692 cm–1 (C=O).
Hz), 3.10–2.86 (3 H, m) 2.18–2.05 (1 H, ddt, J = 12.6, 8.8, 7.1 Hz), 1.92–
1
1.81 (1 H, m), 1.78–1.64 (1 H, m), 1.33 (9 H, s).
H NMR (400 MHz, CDCl ):  = 4.60–4.40 (1 H, m), 3.58–3.25 (2 H, m),
.30–1.95 (4 H, m), 1.50–1.45 (9 H, m).
3
13C NMR (400 MHz, CDCl
):  = 173.5, 61.1, 50.4, 47.2, 30.8, 28.8, 26.1.
2
3
13
HRMS (ESI): m/z [M + H]+ calcd for C
H N O: 171.1492; found:
9 19 2
C NMR (400 MHz, CDCl ):  = 153.1, 119.3, 81.6, 47.3, 45.8, 31.8,
3
28.4, 23.9.
171.1491,
+
HRMS (ESI): m/z [M + Na] calcd for C10H16N O Na: 219.1104; found:
2
2
Aldol Reaction Catalysed by L-Proline Imidate; General Procedure
A flask was flame dried and was allowed to cool at r.t. under a N at-
mosphere. Ketone (1.25 mmol) was added to this flask. The catalyst
·TFA (0.025 mmol, 0.1 equiv) was dissolved in cyclohexane (1 mL)
2
19.1105.
Spectroscopic data are in agreement with the literature.8
2
9
©
2019. Thieme. All rights reserved. — Synthesis 2019, 51, A–G

G
Synthesis
N. Vagkidis et al.
Paper
and was added to the flask. Solid NaHCO (0.025 equiv) was then add-
Supporting Information
3
ed to the flask and the contents of the flask were stirred. After 5 min,
aldehyde (0.25 mmol) was added and the reaction was continued to
be stirred for a further 24 h. The stirring stopped after 24 h and the
Supporting information for this article is available online at
https://doi.org/10.1055/s-0039-1690150.
S
u
p
p
orit
n
g Inform ati
o
n
S
u
p
p
orit
n
g Inform ati
o
n
reaction was quenched with aq NH Cl and the solvent was removed
4
in vacuo at r.t. The crude product was re-dissolved in DCM and
References
washed with H O (5 mL) and extracted with DCM (3 × 10 mL). The
2
combined organic layers were dried (Na SO ) and filtered. The solu-
2
4
(
1) (a) Hughes, D. L. Org. Process Res. Dev. 2018, 22, 574. (b) de
tion was then concentrated in vacuo. The conversion of the reaction
Gama Oliveira, V.; do Carmo Cardoso, M. F.; da Silva Magalhães
Forezi, L. Catalysis 2018, 8, 605.
1
was determined by integrating the H NMR of the crude reaction mix-
ture using the aldehyde peak as a reference. Syn/anti ratio was deter-
(
(
(
2) List, B.; Lerner, R. A.; Barbas, C. F. III. J. Am. Chem. Soc. 2000, 122,
1
mined by integrating the H NMR of the crude reaction mixture and
2
395.
3) Ahrendt, K. A.; Borths, C. J.; MacMillan, D. W. C. J. Am. Chem. Soc.
000, 122, 4243.
4) (a) Cobb, A. J. A.; Shaw, D. M.; Ley, S. V. Synlett 2004, 558.
b) Cobb, A. J. A.; Shaw, D. M.; Longbottom, D. A.; Gold, J. B.; Ley,
by comparing the two CHOH peaks. The enantiomeric excess of the
crude product was analysed by HPLC using a chiralpak IA, IBN-5, IC,
IB, and AD-H column. Representative data for 12-syn and 12-anti are
given below. See Supporting Information for data on 14a–i, 16, and
2
(
18.
S. V. Org. Biomol. Chem. 2005, 3, 84.
(
5) Franzén, J.; Marigo, M.; Fielenbach, D.; Wabnitz, T. C.;
Kjæsgaard, A.; Jørgensen, K. A. J. Am. Chem. Soc. 2005, 127,
2
anti)
-[Hydroxy(4-nitrophenyl)methyl]cyclohexanone (12-syn and 12-
18296.
Yield: 89%, yellow oil; diastereomeric ratio: anti/syn: 5.3:1; 76% anti
ee.
(
6) (a) Burroughs, L.; Vale, M. E.; Gilks, J. A. R.; Forintos, H.; Hayes, C.
J.; Clarke, P. A. Chem. Commun. 2010, 46, 4776. (b) Burroughs, L.;
Clarke, P. A.; Forintos, H.; Gilks, J. A. R.; Hayes, C. J.; Vale, M. E.;
Wade, W.; Zbytniewski, M. Org. Biomol. Chem. 2012, 10, 1565.
12-syn Diastereomer
IR (ATR): 3517, 2940, 1700, 1516, 1346 cm–1
(7) Mase, N.; Nakai, Y.; Ohara, N.; Yoda, H.; Takabe, K.; Tanaka, F.;
.
Barbas, C. F. III. J. Am. Chem. Soc. 2006, 128, 734.
1
H NMR (400 MHz, CDCl ):  = 8.21 (2 H, m), 7.49 (2 H, m), 5.49 (1 H,
3
(
8) Steer, A. M.; Bia, N.; Smith, D. K.; Clarke, P. A. Chem. Commun.
017, 53, 10362.
br s), 3.18 (1 H, br s), 2.66–2.59 (1 H, m), 2.52–2.46 (1 H, m), 2.45–
2
2.35 (1 H, m), 2.15–2.08 (1 H, m), 1.89–1.82 (1 H, m), 1.76–1.65 (2 H,
(
9) A single report of 9·TFA exists; however, no data are reported:
Wang, Z.; Wei, P.; Xizhi, X.; Liu, Y.; Wang, L.; Wang, Q. J. Agric.
Food. Chem. 2012, 60, 8544.
m), 1.63–1.47 (2 H, m).
13
C NMR (400 MHz, CDCl ):  = 214.0, 149.1, 147.1, 126.7, 123.8, 70.2,
3
56.9, 42.7, 28.0, 26.0, 25.0.
(
10) For representative examples, see: (a) Owolabi, I. A.; Reddy, U. V.
S.; Chennapuram, M.; Seki, C.; Okuyama, Y.; Kwon, E.; Uwai, K.;
Tokiwa, M.; Takeshita, M.; Nakano, H. Tetrahedron 2018, 74,
HRMS (ESI): m/z calcd for C13H15NO Na: 272.0893; found: 272.0875.
4
4
705. (b) Luo, S.; Xu, H.; Li, J.; Zhang, L.; Cheng, J.-P. J. Am. Chem.
1
2-anti Diastereoisomer
Soc. 2007, 129, 3074. (c) Jacoby, C. G.; Vontobel, P. H. V.; Bach,
M. F.; Schneider, P. H. New J. Chem. 2018, 42, 7416. (d) Serra-
Pont, A.; Alfonso, I.; Solà, J.; Jimeno, C. Org. Biomol. Chem. 2017,
IR (ATR): 3510, 2939, 1693, 1520, 1346 cm–1
.
1
H NMR (400 MHz, CDCl ):  = 8.21 (2 H, m), 7.51 (2 H, m), 4.89 (1 H,
3
dd, J = 3.2, 8.35 Hz), 4.08 (1 H, d, J = 3.2 Hz), 2.64–2.54 (1 H, m), 2.53–
15, 6584.
2.46 (1 H, m), 2.42–2.31 (1 H, m), 2.15–2.08 (1 H, m), 1.89–1.79 (1 H,
(
11) (a) For mechanistic explanation, see: Bahmanya, S.; Houk, K. N.;
Martin, H. J.; List, B. J. Am. Chem. Soc. 2003, 125, 2475. For
reviews, see: (b) Guillena, G.; Nájera, C.; Ramón, D. J. Tetrahe-
dron: Asymmetry 2007, 18, 2249. (c) Heravi, M. M.; Zadsirjan,
V.; Dehghani, M.; Hosseintash, N. Tetrahedron: Asymmetry
m), 1.74–1.64 (1 H, m), 1.63–1.47 (2 H, m), 1.45–1.34 (1 H, m).
HRMS (ESI): m/z calcd for C13H15NO Na: 272.0893; found: 272.0879.
Spectroscopic data are in agreement with the literature.10a
4
Retention times for the syn and anti stereoisomers: syn-diastereo-
mer: minor enantiomer tR = 27.7 min, major enantiomer tR = 30.0
min; anti diastereomer: major enantiomer t = 34.6 min, minor enan-
tiomer t = 43.0 min.
2
017, 28, 587.
(12) Trost, B. M.; Brindle, C. S. Chem. Soc. Rev. 2010, 39, 1600.
(13) Van Zon, A.; Beyerman, H. C. Helv. Chim. Acta 1973, 56, 1729.
(14) Hoang, C. T.; Alezra, V.; Guillot, R.; Kouklovsky, C. Org. Lett.
R
R
2007, 9, 2521.
(15) Mangette, J. E.; Johnson, M. R.; Le, V.-D.; Shenoy, R. A.; Roark, H.;
Funding Information
Stier, M.; Belliotti, T.; Capiris, T.; Guzzo, P. R. Tetrahedron 2009,
65, 9536.
We thank the Department of Chemistry, The University of York for fi-
nancial support.()
(
16) Caputo, C. A.; Carneiro, F. d. S.; Jennings, M. C.; Jones, N. D. Can. J.
Chem. 2007, 85, 85.
©
2019. Thieme. All rights reserved. — Synthesis 2019, 51, A–G