(3S,4S)-N-substituted-3,4-dihydroxypyrrolidines as ligands for the enantioselective Henry reaction

Article abstract of DOI:10.1002/aoc.6175

The enantioselective Henry reaction is a very important and useful carbon–carbon bond forming reaction. The execution of this reaction requires the use of efficient chiral catalysts. In this work, in situ formed complexes of N-substituted dihydroxypyrrolidines, chiral ligands derived from L-tartaric acid and amines, were evaluated as catalysts in the enantioselective Henry reaction. The results showed that the nature of the N-substituent on the ligand significantly influences the outcome of the reaction. Best results were obtained using a Cu (II) complex of (3S,4S)-N-benzyl-3,4-dihydroxypyrrolidine, in the presence of DIPEA, for the reaction of aromatic aldehydes with nitromethane, at room temperature, originating products with er up to 92:8 (R:S) and conversions up to 96%. The interaction between the pyrrolidine ligand and the copper ion, in isopropanol, was followed by UV-vis spectrophotometry, showing a 1:1 stoichiometry and a binding constant of 4.4. The results obtained will contribute to the design and development of more efficient chiral catalysts for this type of reaction.

Full text of DOI:10.1002/aoc.6175

Received: 9 December 2020
DOI: 10.1002/aoc.6175
Revised: 6 January 2021
Accepted: 10 January 2021
F U L L P A P E R
(3S,4S)-N-substituted-3,4-dihydroxypyrrolidines as ligands
for the enantioselective Henry reaction
Márcia R.R. Rénio
Artur J.M. Valente
| Francisco J.P.M. Sousa
| M. Elisa da Silva Serra
| Nélia C.T. Tavares
| Dina Murtinho
|
CQC and Department of Chemistry,
University of Coimbra, Coimbra,
3004-535, Portugal
The enantioselective Henry reaction is a very important and useful carbon–
carbon bond forming reaction. The execution of this reaction requires the use
of efficient chiral catalysts. In this work, in situ formed complexes of
N-substituted dihydroxypyrrolidines, chiral ligands derived from L-tartaric acid
and amines, were evaluated as catalysts in the enantioselective Henry reaction.
The results showed that the nature of the N-substituent on the ligand signifi-
cantly influences the outcome of the reaction. Best results were obtained using
a Cu (II) complex of (3S,4S)-N-benzyl-3,4-dihydroxypyrrolidine, in the pres-
ence of DIPEA, for the reaction of aromatic aldehydes with nitromethane, at
room temperature, originating products with er up to 92:8 (R:S) and conver-
sions up to 96%. The interaction between the pyrrolidine ligand and the copper
ion, in isopropanol, was followed by UV-vis spectrophotometry, showing a 1:1
stoichiometry and a binding constant of 4.4. The results obtained will contrib-
ute to the design and development of more efficient chiral catalysts for this
type of reaction.
Correspondence
Dina Murtinho, Department of
Chemistry, University of Coimbra, Rua
Larga, Coimbra 3004-535, Portugal.
Email: dmurtinho@ci.uc.pt
Funding information
~
Fundaçao para a Ciência e a Tecnologia,
Grant/Award Numbers: PhD grant
PD/BD/128496/2017, PhD grant
PD/BD/135494/2018, Project
UID/QUI/00313/2020; Portuguese Agency
~
for Scientific Research, Fundaçao para a
Ciência e a Tecnologia (FCT), Grant/
Award Numbers: PD/BD/135494/2018,
PD/BD/128496/2017, UID/
QUI/00313/2020
K E Y W O R D S
asymmetric catalysis, complex stoichiometry, Henry reaction, pyrrolidines
1 | INTRODUCTION
described.[5,14–18] Ligands containing a pyrrolidine moiety
have been used in various catalytic reactions, with good
The Henry reaction allows the synthesis of β-nitro
alcohols, precursors of several types of important
compounds such as α-hydroxy ketones, carboxylic acids,
amino alcohols and amino acids, among others.[1,2]
results. Many of these ligands are based on L-proline,
and this chiral precursor has been extensively explored in
asymmetric catalysis, presenting high induction of
chirality.[19–23] In the asymmetric Henry reaction, the use
of proline derivatives and other pyrrolidine type chiral
ligands, either as organocatalysts or as ligands in organo-
metallic complexes, has also been reported, with promis-
ing results.[1,5,24–28]
Pyrrolidines derived from tartaric acid, which can be
easily prepared by reaction with amines, have been less
explored in the Henry reaction. Some mention has been
made in literature to the use of tartaric acid derived cata-
lysts for this reaction, although mostly regarding the use
of TADDOL and its derivatives.[29–33]
Although the Henry reaction is a classic C–C bond-
forming reaction, the asymmetric version was only
reported by Shibasaki et al.[3] Since then, various metal
complexes of chiral ligands and chiral organocatalysts
have been used in this reaction.[4,5] A variety of chiral
ligands with different types of functionalities (amino
alcohols, diols, diamines, salen, salan, amino phosphines,
among others) have been reported in the literature.[1,6–12]
The use of metals such as La, Mo, Zn, Co, Cr, Rh
and with special emphasis Cu[13] has also been
Appl Organomet Chem. 2021;35:e6175.
wileyonlinelibrary.com/journal/aoc
© 2021 John Wiley & Sons, Ltd.
1 of 10
https://doi.org/10.1002/aoc.6175

2 of 10
RÉNIO ET AL.
In the continuation of our studies on the use of
CH3); 2.87 (q, 2H, CH2, J = 6.4 Hz); 3.41 (t, 2H, CH2,
J = 6.4 Hz); 4.32 (d, 2H, CH, J = 4.0 Hz); 6.27 (approx. d,
2H, OH, J = 4.0 Hz); 7.40 (d, 2H, H–Ar, J = 8.2 Hz); 7.65
(d, 2H, H–Ar, J = 8.2 Hz); 7.75 (t, 1H, NH, J = 6.4 Hz).
13C NMR (DMSO-d6), δ (ppm): 20.9 (CH3), 37.8 (CH2),
39.4 (CH2), 74.2 (CH), 126.4 (C–Ar), 129.7 (C–Ar), 137.4
(C–Ar), 142.7 (C–Ar), 174.6 (CO). HRMS (ESI):
calculated for C13H17N2O6S [M + H]+ 329.0802, found
329.0805.
chiral pyrrolidines derived from tartaric acid for
enantioselective transformations,[34–36] we decided to
screen the potential of several dihydroxypyrrolidines in
the Henry reaction. Most of the synthesized ligands have
the advantage of being easily prepared in a synthetic
sequence of only two steps, starting from inexpensive
tartaric acid and amines.
2 | EXPERIMENTAL SECTION
2.1 | General
2.2.2 | (3S,4S)-N-(2-tosylaminoethyl)-
3,4-dihydroxypyrrolidine 2d
Commercially available compounds were used without
further purification. All solvents were dried prior to use
following standard procedures. Benzaldehyde was dis-
tilled prior to use and stored over 4 Å molecular sieves.
Melting points were determined using a FALC melting
point apparatus (open capillary method). Optical rota-
tions were measured with an Optical Activity AA-5 polar-
imeter. NMR spectra were recorded at room temperature
on a Bruker Avance III 400 MHz (100 MHz for 13C). TMS
was used as the internal standard and chemical shifts are
given in ppm. Infrared spectra were recorded on an
Agilent Technologies Cary 630 FTIR in the ATR mode.
High-resolution mass spectra (HRMS) were obtained on
a TOF VG Autospect M spectrometer with electrospray
ionization (ESI). Conversions for the Henry reactions
were determined by NMR. Enantiomeric ratios were
determined by HPLC using an Agilent 1100 series instru-
ment with a Chiralpack® IB column.
In a round bottom flask, cooled in an ice bath, lithium
aluminum hydride (11.5 mmol, 0.44 g) was slowly
added to (3R,4R)-N-(2-tosylaminoethyl)-3,4-dihydroxy-
2,5-dioxopyrrolidine (5 mmol, 1.64 g) in diethyl ether
(25 ml). The mixture was then refluxed for 48 h. After
cooling, in an ice bath, ethyl acetate was slowly added to
the reaction mixture, followed sequentially by water
(0.5 ml), NaOH, 15% (0.5 ml) and water again (1.5 ml).
The resulting mixture was stirred for 1 h, filtered with
celite and dried over anhydrous Na2SO4. After evapora-
tion of the solvent under reduced pressure, the resulting
oil was purified by column chromatography using silica
gel and CH2Cl2/MeOH (90:10) as eluent.
20
Yield: 18%. ½αꢀD = + 11.5 (c1.3, EtOH); m.p: 88ꢁC
(degradation). IR (cm−1): 3156 (OH), 1324 (SO2), 1152
1
(N–SO2), 1080 (C–O), 814 (p-Ar). H NMR (DMSO-d6),
δ (ppm): 2.22 (dd, 2H, CH2, J = 3.6 Hz, 9.6 Hz);
2.32–2.41 (m, 2H, CH2); 2.39 (s, 3H, CH3); 2.68
(dd, 2H, CH2, J = 5.6 Hz, 9.6 Hz); 2.76 (t, 2H, CH2,
J = 6.8 Hz); 3.76 (approx. d, 2H, CH, J = 4.4 Hz);
4.79 (d, 2H, OH, J = 4.8 Hz); 7.40 (d, 2H, H–Ar,
J = 8.0 Hz); 7.69 (d, 2H, H–Ar, J = 8.0 Hz). 13C NMR
(DMSO-d6), δ (ppm): 21.5 (CH3), 40.7 (CH2), 54.6
(CH2), 59.8 (CH2), 77.7 (CH), 127.1 (C–Ar), 129.8
(C–Ar), 136.6 (C–Ar), 143.5 (C–Ar). HRMS (ESI):
calculated for C13H21N2O4S [M + H]+ 301.1217, found
301.1222.
2.2 | General procedure for the synthesis
of ligands 2d and 2e
2.2.1 | (3R,4R)-N-(2-tosylaminoethyl)-
3,4-dihydroxy-2,5-dioxopyrrolidine 1d
To a suspension of tartaric acid (15 mmol, 2.25 g) in
25 ml of xylene, N-tosylethylenediamine (15 mmol,
3.21 g) was added, and the mixture was refluxed with stir-
ring in a round bottom flask equipped with a Dean–Stark
apparatus. The reaction was complete when the appropri-
ate amount of water was collected (30 mmol, 0.54 ml).
After cooling the reaction mixture, the precipitated prod-
uct was filtered and recrystallized in acetone/hexane to
2.2.3 | (3S,4S)-N-benzyl-
3,4-diacetoxypyrrolidine 2f
To acetic anhydride (25 mmol, 2.36 ml), compound 2a
(10 mmol, 1.93 g) and sodium acetate trihydrate
(10 mol%, 1 mmol, 0.14 g) were added. The mixture was
reacted at room temperature for 1 h and then diluted
with ethyl acetate (10 ml) and washed twice with a satu-
rated solution of NaHCO3. The organic phase was dried
over anhydrous Na2SO4, filtered and evaporated under
give a beige solid.
20
Yield: 54%. ½αꢀD
=
+
65 (c1, EtOH); m.
p. = 133–134ꢁC. IR (cm−1): 3274 (OH), 1713 (C O), 1688
(C O), 1331 (SO2), 1159 (N–SO2), 1065 (C–O),
1
816 (p-Ar). H NMR (DMSO-d6), δ (ppm): 2.39 (s, 3H,

RÉNIO ET AL.
3 of 10
reduced pressure. The product, an oil, was purified by
column chromatography using silica gel and EtOAc/hex-
ane (1:1) as eluent.
14.0 Hz); 5.08–5.12 (m, 1H, CH); 5.23–5.32 (m, 1H, CH);
7.27–7.54 (m, 5H, H–Ar). 13C NMR (CDCl3), δ (ppm):
20.8 (CH3), 20.9 (CH3), 50.4 (CH2), 52.9 (CH2), 73.7 (CH),
75.0 (CH), 127.3(C–Ar), 128.5 (C–Ar), 130.5 (C–Ar), 135.8
(C–Ar), 169.6 (CO), 169.7 (CO), 170.1 (CO). HRMS (ESI):
calculated for C15H18NO5 [M + H]+ 292.1179, found
292.1178.
20
Yield: 85%. ½αꢀD = + 35.7 (c1.1, CH2Cl2). IR (cm−1):
1
2798 (CH), 1734 (C O), 1073 (C–O), 745 (Ar). H NMR
(CDCl3), δ (ppm): 2.06 (s, 6H, CH3); 2.55 (dd, 2H, CH2,
J = 4.0 Hz, 10.8 Hz); 3.08 (dd, 2H, CH2, J = 5.6 Hz,
10.8 Hz); 3.62 (d, 1H, CH2, J = 13.0 Hz); 3.69 (d, 1H, CH2,
J = 13.0 Hz); 5.12(dd, 2H, CH, J = 4 Hz, 5.6 Hz);
7.25–7.35 (m, 5H, H–Ar). 13C NMR (CDCl3), δ (ppm):
21.0 (CH3), 58.1 (CH2), 59.7 (CH2), 77.7 (CH), 127.3
(C–Ar), 128.4 (C–Ar), 128.9 (C–Ar), 137.5 (C–Ar), 170.4
(CO). HRMS (ESI): calculated for C15H20NO4 [M + H]+
278.1387, found 278.1392.
2.2.6 | (3S,4S)-N-benzoyl-
3,4-dihydroxypyrrolidine 2e
Compound 2h (5 mmol, 1.46 g) and NaOH (10 mmol,
0.40 g), dissolved in water (25 ml), were stirred at room
temperature for 3 h. The reaction mixture was extracted
several times with CHCl3, dried with anhydrous Na2SO4,
filtered and the solvent was evaporated.
2.2.4 | (3S,4S)-diacetoxypyrrolidine 2g
1
Yield: 74%. H NMR (CDCl3), δ (ppm): 3.25 (approx.
In a round-bottom flask compound 2f (10 mmol, 2.77 g),
ammonium formate (20 mmol, 1.26 g) and Pd/C 10%
(0.3 g) in methanol (100 ml) were added. The mixture
was refluxed for 1 h, cooled to room temperature, filtered
with celite and washed several times with methanol.
After evaporation of the solvent under reduced pressure,
the product was obtained as an oil and was used directly
in the synthesis of 2h.
d, 1H, CH2, J = 11.4 Hz); 3.53 (approx. d, 1H, CH2,
J = 13.0 Hz); 3.67 (dd, 1H, CH2, J = 4.0 Hz, 11.4 Hz); 3.82
(dd, 1H, CH2, J = 4.0 Hz, 13.0 Hz); 4.02 (sl, 1H, CH); 4.14
(sl, 1H, CH); 7.19–7.38 (m, 5H, H–Ar).
2.3 | General procedure for the addition
of nitromethane to aldehydes
1
Yield: 98%. H NMR (CDCl3), δ (ppm): 2.08 (s, 6H,
CH3); 2.56 (dd, 2H, CH2, J = 4.2 Hz, 10.4 Hz); 3.12
(dd, 2H, CH2, J = 5.8 Hz, 10.4 Hz); 5.12 (dd, 2H, CH,
J = 4.2 Hz, 5.8 Hz).
Chiral ligand (0.08 mmol) and solvent (8 ml) were stirred
in a round bottom flask, until ligand dissolution was
complete. Then the metal salt (0.08 mmol) was added
and allowed to stir for 2 h, at room temperature (ultra-
sound was used for better solubilization when the solvent
was toluene). The substrate (0.8 mmol) and nitromethane
(44.8 mmol) were then added, and the mixture was
allowed to stir for 10 min. The base (0.04 mmol) was then
added, and the reaction mixture stirred for 48 h at room
temperature.
The solvent was removed under reduced pressure and
the residue was purified by column chromatography
using silica gel and ethyl acetate/hexane mixtures as
eluent.
The enantiomeric ratio of the products was deter-
mined by HPLC analysis (254 nm), using hexane/iso-
propanol mixtures as eluent. The absolute configurations
of the products were assigned by comparison to literature
values and by the sign of the optical rotation of the
products.[37,38]
2.2.5 | (3S,4S)-N-benzoyl-
3,4-diacetoxypyrrolidine 2h
In a two-necked round bottom flask, cooled in an ice
bath, compound 2g (9 mmol, 1.68 g) and triethylamine
(10.8 mmol, 1.5 ml) in dichloromethane (25 ml) were
added. In inert atmosphere and via syringe, benzoyl chlo-
ride (10.8 mmol, 1.25 ml) was slowly added and then the
reaction mixture was stirred for 4 h, at room temperature.
The mixture was extracted twice with water, the organic
phase was washed with a saturated solution of NaHCO3,
dried with anhydrous Na2SO4, filtered and the solvent
was evaporated. The product, an oil, was purified by col-
umn chromatography using silica gel and EtOAc/hexane
(1:1) as eluent.
20
Yield: 70%. ½αꢀD = − 4.10 (c4.9, CH2Cl2). IR (cm−1):
The HPLC conditions and retention times for the
different products of the Henry reaction are
1-phenyl-2-nitroethanol, hexane:iPrOH (90:10), 1 ml/min,
tR = 5.8 and 6.5 min; 1-(2-nitrophenyl)-2-nitroethanol,
hexane:iPrOH (90:10), 1 ml/min, tR = 8.1 and 8.7 min;
1-(3-nitrophenyl)-2-nitroethanol, hexane:iPrOH (90:10),
1741 (C O), 1628 (C O), 1060 (C–O), 731 (Ar). 1H NMR
(CDCl3), δ (ppm): δ 2.05 (s, 3H, CH3); 2.12 (s, 3H, CH3);
3.51 (approx. d, 1H, CH2, J = 12.4 Hz); 3.80 (approx.
d, 1H, CH2, J = 14.0 Hz); 3.91 (dd, 1H, CH2,
J = 4.0 Hz, 12.4 Hz); 4.06 (dd, 1H, CH2, J = 4.4 Hz,

4 of 10
RÉNIO ET AL.
1 ml/min, tR = 12.5 and 14.1 min; 1-(4-nitrophenyl)-
3 | RESULTS AND DISCUSSION
2-nitroethanol, hexane:iPrOH (85:15), 0.8 ml/min tR
=
10.4 and 12.0 min; 1-(2-chlorophenyl)-2-nitroethanol,
hexane:iPrOH (90:10), 1 ml/min, tR = 17.6 and 18.9 min;
1-(4-chlorophenyl)-2-nitroethanol, hexane:iPrOH (85:15),
0.8 ml/min tR = 9.3 and 11.4 min; 1-(2-methoxyphenyl)-
3.1 | Ligand synthesis
Ligands 2a-c were prepared according to previously
described procedures.[34,40,41] Dihydroxydioxopyrrolidines
(1a-c) were synthesized in good yields (62%–92%),
through the condensation of tartaric acid with
several amines (benzylamine, cyclohexylamine and
2-nitroethanol, hexane:iPrOH (90:10), 0.8 ml/min tR
=
6.7 and 7.2 min; 1-(3-methoxyphenyl)-2-nitroethanol,
hexane:iPrOH (90:10), 1 ml/min tR = 8.1 and 9.0 min;
1-(2-methylphenyl)-2-nitroethanol, hexane:iPrOH (90:10),
1 ml/min tR = 4.7 and 5.6 min; 1-(3-methylphenyl)-
2-nitroethanol, hexane:iPrOH (90:10), 1 ml/min tR = 4.9
and 5.4 min.
naphthymethylamine) in refluxing xylene, using
a
Dean–Stark apparatus (Scheme 1). Reduction of 1a-c
with LiAlH4, using diethyl ether as solvent, allowed
the synthesis of the dihydroxypyrrolidines 2a-c in
moderate yields (45%–48%), after
a 48 h reflux.
Dihydroxypyrrolidine 2d was synthesized using a similar
procedure, starting from N-tosylethylenediamine and
tartaric acid (1d 54% yield, 2d 18% yield).
2.4 | UV-visible spectroscopy
measurements
Ligand 2e, already described by Siedlecka et al.,[42]
was prepared using a slightly different strategy, starting
from 2a, Scheme 2. Briefly, 2a was acetylated with acetic
anhydride in the presence of sodium acetate, at room
temperature, to give 2f in quantitative yield. 2g was pre-
pared in 94% yield by hydrogenolysis of the benzyl group
with ammonium formate and Pd/C 10%, in methanol
reflux. Reaction of 2g with benzoyl chloride, in the pres-
ence of NEt3, gave 2h in 70% yield. Finally, hydrolysis
of the acetate group with aqueous NaOH allowed the
synthesis of ligand 2e (74% yield).
The UV-visible spectra of solutions of Cu (II), in the
absence and presence of the pyrrolidine, were recorded
on a Shimadzu 2450 UV-vis spectrophotometer. For Job's
method of continuous variation,[39] the total concentra-
tion of the two species is kept constant and equal to
4 mM. For the determination of the binding constant, a
titration of a 4 mM Cu (II) solution was carried out by
using a 2a solution. The concentration of the latter varied
between 0.5 and 8 mM. All solutions were prepared by
using isopropanol as solvent.
SCHEME 1 Pyrrolidine synthesis
SCHEME 2 Synthesis of N-benzoylpyrrolidine

RÉNIO ET AL.
5 of 10
3.2 | Catalytic studies
tested as solvents for the reaction and conversions higher
than 96% were obtained in all cases. However, low er
resulted with all the alcohols tested. Polar aprotic
solvents such as dichloromethane, diethyl ether and THF
gave lower conversions, but higher er. Toluene, an apolar
solvent, gave the best results, with almost complete con-
version and an er of 87:13 (R:S), Table 1. This same trend,
in which aromatic solvents lead to better induction of
chirality, has been previously observed.[6,47–49]
The use of different metal salts for the Henry reaction
is reported. Thus, copper, zinc, nickel and cobalt salts
were complexed with ligand 2a and tested in the reaction
of benzaldehyde with nitromethane. The results are sum-
marized in Table 2.
Good conversions were obtained with all the metal
salts but, in some cases, the main product was the
nitroalkene, instead of the nitroaldol. Better results were
obtained using Cu (OAc)2.H2O as the metal salt, as has
been previously referred.[7,13,50,51]
Alkaline additives can be used in the Henry reaction
to deprotonate nitromethane, forming the corresponding
nitronate and thus improving the outcome of the reac-
tion. Bases such as carbonates and amines are widely
used.[10,11,49,52] Accordingly, several bases (5 mol% with
respect to the aldehyde) were screened as additives for
the Henry reaction, as reported in Table 3. Except for
NEt3, very good conversions were obtained with all bases
(greater than 92%). Although bases can also deprotonate
the nitroaldol product and thus favor the formation of
the nitroalkene, this did not occur with those studied.
With regard to er, the best results were obtained using
In the initial experiments, ligand 2a was used to optimize
several reaction parameters. The ligand was reacted in
.
situ with Cu (OAc)2 H2O to form the catalyst, and subse-
quently tested in the enantioselective Henry reaction
between benzaldehyde and nitromethane, at room tem-
perature for 48 h, using isopropanol as solvent, and
Na2CO3 as additive (no product was observed in the
absence of a base). Lower temperatures (0ꢁC) resulted in
lower conversions and er (84% and 81.5:18.5, respec-
tively). For lower reaction times, less than 48 h, a signifi-
cant amount of reagent was detected by TLC. Using
10 mol% of the complex, in a 1:1 ligand:metal stoichiome-
try, 90% conversion of the substrate was obtained, with
86% of chiral nitroaldol product 3a and 14% of
nitroalkene 3b, resulting from the elimination of a water
molecule (Scheme 3).
An enantiomeric ratio (er) of only 54.5:45.5 (R:S)
was obtained using these conditions. With 15 or
20 mol% of catalyst, conversions were higher (96 and
97%, respectively) but percentages of the chiral product
were lower (72% and 46%, respectively). Using 5 mol% of
catalyst a lower er 51.5:48.5 (R:S) was obtained. Thus,
it was decided to use 10 mol% of catalyst to proceed the
studies.
The solvent is another parameter that can influence
the outcome of the Henry reaction. The use of polar and
apolar solvents is mentioned in the literature, although
polar protic solvents such as alcohols usually give better
conversions and er.[43–46] Therefore, several alcohols were
SCHEME 3 Henry reaction products
TABLE 1 Solvent effect on the Henry reactiona
Solvent
MeOH
EtOH
iPrOH
BuOH
CH2Cl2
Et2O
Conversion (%)b
Chiral product (%)b
er (R:S)c
54:46
97
96
97
97
47
78
84
89
95
81
81
93
90
92
91
51:49
56:44
59:41
n.d.
86.5:13.5
68:32
THF
Toluene
>99
87:13
aBenzaldehyde (0.8 mmol), ligand 2.2a (0.08 mmol), Cu (OAc)2.H2O (0.08 mmol), CH3NO2 (44.8 mmol), Na2CO3 (0.04 mmol), solvent (8 ml), 48 h.
bDetermined by 1H NMR.
cDetermined by chiral HPLC.

6 of 10
RÉNIO ET AL.
TABLE 2 Metal salt effect on the
Henry reactiona
Metal
Conversion (%)b
>99
Chiral product (%)b
er (R:S)c
87:13
Cu (OAc)2.H2O
CuCl
91
>99
98
>99
65:35
CuCl2.2H2O
Zn (OTf)2
30
75
87
70
10
n.d.
96
50:50
Zn (OAc)2.2H2O
Ni (OAc)2.4H2O
Co (OAc)2.4H2O
95
46.5:53.5
47.5:52.5
n.d.
>99
98
aBenzaldehyde (0.8 mmol), ligand 2.2a (0.08 mmol), metal salt (0.08 mmol), CH3NO2 (44.8 mmol), Na2CO3
(0.04 mmol), toluene (8 mL), 48 h.
bDetermined by 1H NMR.
cDetermined by chiral HPLC.
TABLE 3 Base effect on the Henry reactiona
Base
Conversion (%)b
Chiral product (%)b
er (R:S)c
85.5:14.5
72.5:27.5
82:18
DABCOd
DBUe
99
94
78
92
93
95
96
91
91
95
NEt3
DIPEAf
Na2CO3
Cs2CO3
89.5:10.5
87:13
>99
>99
83:17
aBenzaldehyde (0.8 mmol), ligand 2.2a (0.08 mmol), Cu (OAc)2.H2O (0.08 mmol), CH3NO2 (44.8 mmol), base (0.04 mmol), toluene (8 mL), 48 h.
bDetermined by 1H NMR.
cDetermined by chiral HPLC.
d1,4-diazabicyclo[2.2.2]octane.
e1,8-diazabicyclo[5.4.0]undec-7-ene.
fDiisopropylethylamine.
diisopropylethylamine (DIPEA), albeit the conversion is
less than that obtained with Na2CO3.
state. Another feature that may explain this marked dif-
ference in the er is that the carbonyl group can remove
electron density from the nitrogen atom, and this may
affect the coordination to the metal. These two aspects
can thus affect the relative energies of the transition
states in the Henry reaction and lead to products with
lower selectivity.
Ligand 2d, unlike the other ligands, can coordinate
with copper in a tridentate manner. Because references
are made in the literature to the use of tridentate ligands
with good results, we thought that it would be interesting
to synthesize and test this ligand. As can be seen from
the results in Table 4, ligand 2d gave low conversion and
a racemic product.
Using our most efficient ligand 2a, the scope of the
reaction was then extended to other aromatic substrates
and the results are summarized in Table 5. In general,
aromatic substrates with electron-withdrawing groups
presented better conversions than those containing
electron-donating groups. On the contrary, better er were
obtained with aromatic substrates containing electron-
donating groups.
In an attempt to improve the er, using DIPEA as
additive, the reaction temperature was lowered to 0ꢁC.
Under these reaction conditions, a lower conversion was
obtained (84%) and the er decreased to 82.5:17.5 (R:S).
Using the optimized reaction conditions, the other
ligands prepared were screened for their catalytic activity
in the Henry reaction (Table 4). Comparing ligands 2b-c
with 2a, it can be concluded that the presence of a
naphthylmethyl or a cyclohexyl group on the nitrogen of
the hydroxypyrrolidine slightly decreases the er of the
reaction.
Ligand 2e was prepared aiming to study the influence
of the presence of a more sterically demanding and elec-
tron attracting benzoyl group on the nitrogen atom of the
ligand. It was observed that the presence of this group
decreases the conversion of the reaction but, most impor-
tantly, it has a drastic effect on the er. This may be due to
the fact that the carbonyl group is more sterically
demanding than the methylene group of ligand 2a and
this may lead to high steric hindrance in the transition

RÉNIO ET AL.
7 of 10
TABLE 4 Henry reaction using
L-tartaric acid derived ligandsa
Ligand
Conversion (%)b
Chiral product (%)b
er (R:S)c
92
91
89.5:10.5
81
84
94
83: 17
>99
82.5:17.5
77
78
72
92
50:50
51:49
aBenzaldehyde (0.8 mmol), ligand (0.08 mmol), Cu (OAc)2.H2O (0.08 mmol), CH3NO2 (44.8 mmol), DIPEA
(0.04 mmol), toluene (8 mL), 48 h.
bDetermined by 1H NMR.
cDetermined by chiral HPLC.
TABLE 5 Substrate effect on the Henry reactiona
active species, a different set of experiments were carried
out. Since the Cu (II) solution shows a well-defined UV–
visible spectrum, with a maximum at 702 nm, the
method of continuous variation was applied to evaluate
the stoichiometry of the interaction. It can be seen from
the analysis of Figure 1b that this method shows a maxi-
mum at 0.46, suggesting a 1:1 stoichiometry between 2a
and Cu (OAc)2. A possible structure for this complex is
presented in Figure 2. It can also be seen that Job's plot
does not show a reverse V-shape, but instead a Gaussian-
like trend. Such a behavior has been discussed else-
where[53] and suggests a lower binding constant between
the Cu (II) and 2a. Based on this stoichiometry, and tak-
ing into account the suggested catalytic cycle of similar
compounds,[46] the following equation is suggested:
Aldehyde
Conversion (%)b
er (R:S)c
89.5:10.5
57:43
Benzaldehyde
92
82
69
87
96
55
53
27
10
49
42
2-Nitrobenzaldehyde
3-Nitrobenzaldehyde
4-Nitrobenzaldehyde
2-Chlorobenzaldehyde
4-Chlorobenzaldehyde
2-Methoxybenzaldehyde
3-Methoxybenzaldehyde
4-Methoxybenzaldehyde
2-Methylbenzaldehyde
3-Methylbenzaldehyde
58:42
63.5:36.5
76.5:23.5
75.5:24.5
92:8
86:14
n.d.
89.5:10.5
82.5:17.5
aBenzaldehyde (0.8 mmol), ligand 2.2a (0.08 mmol), Cu (OAc)2.H2O
Cu ðOAcÞ2 + 2a−H⇄CuðOAcÞð2aÞ + HOAc
ð1Þ
(0.08 mmol), CH3NO2 (44.8 mmol), DIPEA (0.04 mmol), Toluene (8 mL),
48 h.
bDetermined by 1H NMR.
cDetermined by chiral HPLC.
For the sake of simplicity, we have highlighted the hydro-
gen atom of the pyrrolidine, in the equation, which is
deprotonated in the presence of copper acetate to form
the complex Cu (OAc)(2a) (Figure 2).
Our
best
result
was
obtained
using
Based on equation (1), and considering the [Cu (OAc)
(2a)] = [HOAc], the corresponding binding constant, K,
can be written as
2-methoxybenzaldehyde as substrate, giving the product
with an er of 92:8 (R:S).
À
Â
ÃÁ
2
K = ½CuðOAcÞð2aÞꢀ = ½2aꢀ CuðOAcÞ2
ð2Þ
3.3 | Determination of the pyrrolidine-
copper complex stoichiometry
The variation of the experimental absorbance, ΔA,
defined as a difference of the absorbance of the Cu
(II) solution in the presence and absence of pyrrolidine
In order to get insight on how our best ligand, pyrrolidine
2a, coordinates with copper to form the catalytically

8 of 10
RÉNIO ET AL.
FIGURE 1 (a) UV-visible spectra of 4 mM Cu (II) solutions in the presence of increasing concentrations of 2a; (b) Job's plot for 2a and
Cu (II) mixtures at different 2a molar fractions; and (c) effect of 2a concentration on the absorbance of a Cu (II) solution, at 702 nm
(Figure 1a). Solid lines in b and c were obtained by fitting the experimental data to a Gram–Charlier peak function[55] and to Equations 3
and 4, respectively
Figure 1c shows the fitting of Equations 3 and 4 to the
experimental ΔA, using
a non-linear least-squares
algorithm. The computed fitting parameters are: K = 4.4
( 1.9), ΔACu(OAc)(2a) = 0.20 ( 0.01), for a determination
coefficient of 0.9967 and a χ2 = 1.2 × 10−5. These results
show that, under the experimental conditions used
(i.e., in a 4 mM equimolar mixture of Cu (OAc)2 and 2a)
39% of the ligand remains uncomplexed. This is in close
agreement with the shape of Job's plot as well as with the
absence of a clear plateau of DA as a function of [2a] in
Figure 1c. It should be noticed that stability for Cu (II)-
ligand interaction was measured in a low dielectric
constant solvent, which justified the magnitude of the
binding constant.[54]
FIGURE 2 Possible structure of the 2a: copper 1:1 complex
2a, is dependent on the concentration of the complex, as
described in Equation 3:
ΔACuðOAcÞð2aÞ
Â
Ã
ΔA =
½CuðOAcÞð2aÞꢀ
ð3Þ
CuðOAcÞ2
T
4 | CONCLUSIONS
where [Cu (OAc)2]T is the total concentration of Cu
(II) in solution and ΔACu(OAc)(2a) corresponds to the
change of absorbance due to the complex; it should be
noted that the absorbance has a contribution of free and
complexed Cu (II) species.
From the mass balances for Equation 1 and after
some algebraic manipulation,[53] the concentration of the
complex can be computed from the following equation:
Several metal complexes of chiral N-substituted
dihydroxypyrrolidines derived from L-tartaric acid and
amines were prepared, in short and simple synthetic
sequences, and evaluated in the enantioselective Henry
reaction. It was observed that the nature of the substituent
on the pyrrolidine nitrogen significantly influences
the outcome of the reaction. Under optimized reaction
ꢀ
ꢁ
1=2
ÀÂ
Ã
Á
ÀÂ
Ã
Á
Â
Ã
2
CuðOAcÞ2 + ½2aꢀT
−
CuðOAcÞ2 + ½2aꢀT −4K −1 CuðOAcÞ2 ½2aꢀT
K
T
T
T
½CuðOAcÞð2aÞꢀ =
ð4Þ
2K −1
K

RÉNIO ET AL.
9 of 10
[4] J. Boruwa, N. Gogoi, P. P. Saikia, N. C. Barua, Tetrahedron:
Asymmetry 2006, 17, 3315.
[5] N. Ananthi, S. Velmathi, Chem. - Sect. B Org. Med. Chem.
2013, 52, 87.
[6] F. Xu, C. Lei, L. E. I. Yan, J. Tu, G. Li, Chirality 2015, 27, 761.
[7] X. Wang, W. Zhao, G. Li, J. Wang, G. Liu, L. Liu, R. Zhao, M.
Wang, Appl. Organomet. Chem. 2014, 28, 892.
[8] L. Androvicˇ, P. Drabina, I. Panov, B. Frumarová, A.
Kalendová, M. Sedlák, Tetrahedron: Asymmetry 2014, 25, 775.
[9] S. J. Canipa, A. Stute, P. O'Brien, Tetrahedron 2014, 70, 7395.
[10] L. Filippova, Y. Stenstrøm, T. V. Hansen, Molecules 2015, 20,
6224.
[11] Y. Fan, Y. Ren, J. Li, C. Yue, H. Jiang, Inorg. Chem. 2018, 57,
11986.
[12] N. H. Khan, S. H. R. Abdi, R. Tak, T. Menapara, M. K.
Choudhary, R. I. Kureshy, J. Mol. Catal. A: Chem. 2016,
421, 161.
conditions and using a Cu (II) complex of (3S,4S)-N-ben-
zyl-3,4-dihydroxypyrrolidine, conversions up to 96% and er
up to 92:8 (R:S) were obtained for the reaction of aromatic
aldehydes with nitromethane, in the presence of DIPEA,
at room temperature. The stability constant for Cu (II)-
(3S,4S)-N-benzyl-3,4-dihydroxypyrrolidine was addition-
ally evaluated by UV–vis spectroscopy and has been
computed as equal to 4.4, considering a 1:1 stoichiometry.
DATA AVAILABILITY STATEMENT
Data available on request from the authors.
ACKNOWLEDGEMENTS
Thanks are due to Coimbra Chemistry Centre (CQC),
supported by the Portuguese Agency for Scientific
~
[13] S. Zhang, Y. Li, Y. Xu, Z. Wang, Chinese Chem. Lett. 2018,
29, 873.
[14] S. Saranya, N. A. Harry, S. M. Ujwaldev, G. Anilkumar, Asian
J. Org. Chem. 2017, 6, 1349.
[15] J. Dimroth, M. Weck, RSC Adv. 2015, 5, 29108.
[16] W. Li, J. Zhang, Chem. Soc. Rev. 2016, 45, 1657.
[17] P. Anitha, R. Manikandan, P. Vijayan, S. Anbuselvi, P.
Viswanathamurthi, J. Organomet. Chem. 2015, 791, 244.
[18] O. Kammoun, W. Rekik, T. Bataille, K. T. Mahmudov, M. N.
Kopylovich, H. Naïli, J. Organomet. Chem. 2013, 741–742, 136.
[19] S. K. Panday, Tetrahedron: Asymmetry 2011, 22, 1817.
[20] Q. Yang, Y. Li, Z. Wei, Y. Lin, Y. Song, H. Duan, R. Wang, E.
Xu, Can. J. Chem. 2019, 97, 97.
Research, Fundaçao para a Ciência e a Tecnologia (FCT)
through Project UID/QUI/00313/2020. Nélia C. T.
Tavares (PhD grant PD/BD/128496/2017) and Márcia
Rénio (PhD grant PD/BD/135494/2018) also thank FCT
for funding. We also acknowledge the UC-NMR facility
for obtaining the NMR data (www.nmrccc.uc.pt).
AUTHOR CONTRIBUTIONS
Márcia Rénio: Investigation. Francisco Sousa:
Investigation. Nélia Tavares: Investigation. Artur
Valente: Conceptualization; supervision. M. Elisa da
Silva Serra: Conceptualization; supervision. Dina
Murtinho: Conceptualization; supervision.
[21] A. G. Doyle, E. N. Jacobsen, Chem. Rev. 2007, 107, 5713.
[22] D. Wang, M. Wang, X. Wang, Y. Chen, A. Gao, L. Sun,
J. Catal. 2006, 237, 248.
CONFLICT OF INTEREST
There are no competing interests to declare.
[23] Y.-J. Cao, Y.-Y. Lai, X. Wang, Y.-J. Li, W.-J. Xiao, Tetrahedron
Lett. 2007, 48, 21.
[24] A. Mondal, S. Bhowmick, A. Ghosh, T. Chanda, K. C.
Bhowmick, Tetrahedron: Asymmetry 2017, 28, 849.
[25] P. Niedziejko, M. Szewczyk, P. Kalicki, Z. Kałuża, Tetrahe-
dron: Asymmetry 2015, 26, 1083.
[26] H. Atoholi Sema, G. Bez, S. Karmakar, H. A. Sema, Appl.
Organomet. Chem. 2014, 28, 290.
ORCID
Márcia R.R. Rénio https://orcid.org/0000-0002-8850-
7660
Francisco J.P.M. Sousa https://orcid.org/0000-0002-
5246-4232
[27] P. Kumar, M. S. Chauhan, G. D. Yadav, S. Singh, Synlett 2016,
27, 267.
Nélia C.T. Tavares https://orcid.org/0000-0001-6034-
7733
[28] D. Scharnagel, F. Prause, J. Kaldun, R. G. Haase, M. Breuning,
Chem. Commun. 2014, 50, 6623.
Artur J.M. Valente https://orcid.org/0000-0002-4612-
[29] T. Akiyama, J. Itoh, K. Fuchibe, Adv. Synth. Catal. 2006,
348, 999.
[30] Y. Zhang, X. Y. Wu, J. Han, Chinese Chem. Lett. 2019, 30,
1519.
7686
M. Elisa da Silva Serra https://orcid.org/0000-0002-
4562-7072
Dina Murtinho https://orcid.org/0000-0002-1391-9855
[31] C. Gan, Can. J. Chem. 2008, 86, 261.
[32] J. Kousalová, O. Pytela, F. Bureš, ARKIVOC 2012, 2012, 35.
[33] K. Balaraman, R. Vasanthan, V. Kesavan, Synthesis 2012, 44,
2455.
REFERENCES
[1] A. Dixit, P. Kumar, G. D. Yadav, S. Singh, Inorganica Chim.
Acta 2018, 479, 240.
[34] A. M. D. A. Rocha Gonsalves, M. E. S. Serra, D. Murtinho,
~
[2] A. V. Gurbanov, S. Hazra, A. M. Maharramov, F. I. Zubkov,
F. I. Guseinov, A. J. L. Pombeiro, J. Organomet. Chem. 2018,
869, 48.
[3] H. Sasai, T. Suzuki, S. Arai, T. Arai, M. Shibasaki, J. Am.
Chem. Soc. 1992, 114, 4418.
V. F. V. F. Silva, A. M. M. Beja, J. A. A. Paixao, M. R. R. Silva,
L. A. Alte da Veiga, J. Mol. Catal. A: Chem. 2003, 195, 1.
[35] M. E. S. Serra, D. Murtinho, A. Goth, ARKIVOC 2010, 64.
[36] M. E. S. Serra, D. Murtinho, A. Goth, Tetrahedon: Asymmetry
2013, 24, 315.

10 of 10
RÉNIO ET AL.
[37] H. Maheswaran, K. L. Prasanth, G. G. Krishna, K. Ravikumar,
B. Sridhar, M. L. Kantam, Chem. Commun. 2006, 4066.
[38] N. Sanjeevakumar, M. Periasamy, Tetrahedron: Asymmetry
2009, 20, 1842.
[49] M. Liu, N. Ji, L. Wang, P. Liu, W. He, Tetrahedron Lett. 2018,
59, 999.
ꢀ
[50] E. Wolinska, Tetrahedron: Asymmetry 2014, 25, 1122.
[51] L. Yao, Y. Wei, P. Wang, W. He, S. Zhang, Tetrahedron 2012,
[39] J. S. Renny, L. L. Tomasevich, E. H. Tallmadge, D. B. Collum,
Angew. Chemie - Int. Ed. 2013, 52, 11998.
[40] U. Nagel, E. Kinzel, J. Andrade, G. Prescher, Chem. Ber. 1986,
119, 3326.
68, 9119.
[52] J.-L. Li, L. Liu, Y.-N. Pei, H.-J. Zhu, Tetrahedron 2014, 70,
9077.
[53] A. J. M. Valente, O. Söderman, Adv. Colloid Interface Sci. 2014,
205, 156.
[41] A. M. d'Albuquerque R. Gonsalves, M. E. da S. Serra, M. R.
~
Silva, A. M. Beja, J. A. Paixao, L. A. da Veiga, J. Mol. Catal. A
Chem. 2001, 168, 53.
[54] H. M. Felmy, K. T. Bennett, S. B. Clark, J. Chem. Thermodyn.
2017, 114, 83.
ꢀ
[42] R. Siedlecka, E. Wojaczynska, J. Skarżewski, Tetrahedon:
[55] R. A. Carvalho, H. A. Correia, A. J. M. Valente, O. Söderman,
M. Nilsson, J. Colloid Interface, Sci. 2011, 354, 725.
Asymmetry 2004, 15, 1437.
[43] C. Ao, J. Men, Y. Wang, T. Shao, Y. Huang, J. Huo, G. Gao,
Tetrahedron: Asymmetry 2016, 27, 589.
[44] R. Boobalan, G.-H. Lee, C. Chen, Adv. Synth. Catal. 2012, 354,
2511.
[45] J. Cho, M. K. Chun, S. Nayab, J. H. Jeong, Appl. Organomet.
Chem. 2019, 33, 1.
[46] K. Y. Spangler, C. Wolf, Org. Lett. 2009, 11, 4724.
[47] T. Arai, Y. Yamamoto, Org. Lett. 2014, 16, 1700.
[48] M. K. Choudhary, R. Tak, R. I. Kureshy, A. Ansari, N. H.
Khan, S. H. R. Abdi, H. C. Bajaj, J. Mol. Catal. A: Chem. 2015,
409, 85.
How to cite this article: Rénio MR, Sousa FJPM,
Tavares NC, Valente AJM, da Silva Serra ME,
Murtinho D. (3S,4S)-N-substituted-3,4-
dihydroxypyrrolidines as ligands for the
enantioselective Henry reaction. Appl Organomet
Chem. 2021;35:e6175. https://doi.org/10.1002/aoc.
6175