Determination of the Absolute Configuration of (S)-N-(1-Aryl-allyl)-3,5-dinitrobenzamides and Their Elution Order on Brush-Type Chiral Stationary Phases

Article abstract of DOI:10.1002/ejoc.201800346

A series of ten enantiomerically pure (S)-N-(1-aryl-allyl)-3,5-dinitrobenzamides (S-DNBs) was prepared using enzymatic resolution and chiral chromatography. Enzymatic resolution of corresponding 1-aryl-allylamines using Candida antarctica lipase B (CaLB) was efficient for amines with no steric hindrance near the stereogenic center and S-DNB amides were prepared by acylation of the obtained S-amine. When steric effects interrupted enzymatic resolution, racemic DNB amides were resolved using a brush-type chiral column (CSP-A) developed in our laboratory. Previously reported behavior of CaLB in kinetic resolution of amines was considered a starting point for the determination of absolute configuration (AC). The AC of prepared S-DNB amides was anticipated using the elution order of prepared DNB amides on CSP-A and commercial Whelk-O1 columns and comparison with DNB amides obtained after acylation of (S)-amines. The comparison between experimental electronic circular dichroism (ECD) spectra with those obtained by conformational analysis and ECD calculations of representative compounds allowed us to verify the AC of prepared DNB amides.

Full text of DOI:10.1002/ejoc.201800346

A Journal of
Accepted Article
Title: Absolute configuration determination of (S)-N-(1-aryl-allyl)-3,5-
dinitrobenzamides and elution order on brush-type chiral
stationary phases
Authors: Anamarija Knežević, Jurica Novak, Gennaro Pescitelli, and
Vladimir Vinković
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Eur. J. Org. Chem. 10.1002/ejoc.201800346
Link to VoR: http://dx.doi.org/10.1002/ejoc.201800346
Supported by

10.1002/ejoc.201800346
European Journal of Organic Chemistry
FULL PAPER
Absolute configuration determination of (S)-N-(1-aryl-allyl)-3,5-
dinitrobenzamides and elution order on brush-type chiral
stationary phases
Anamarija Knežević,*[a] Jurica Novak,[b] Gennaro Pescitelli,[c] and Vladimir Vinković[a]
Abstract: A series of ten enantiomerically pure (S)-N-(1-aryl-allyl)-
3,5-dinitrobenzamides (S-DNBs) was prepared using enzymatic
resolution and chiral chromatography. Enzymatic resolution of
corresponding 1-aryl-allylamines using Candida antarctica lipase B
(CaLB) was efficient for amines with no steric hindrance near the
stereogenic center and S-DNB amides were prepared by acylation of
the obtained S-amine. When steric effects interrupted enzymatic
resolution, racemic DNB amides were resolved using a brush-type
chiral column (CSP-A) developed in our laboratory. Previously
reported behavior of CaLB in kinetic resolution of amines was
considered a starting point for the determination of absolute
configuration (AC). The AC of prepared S-DNB amides was
anticipated using the elution order of prepared DNB amides on CSP-
A and commercial Whelk-O1 columns and comparison with DNB
amides obtained after acylation of (S)-amines. The comparison
between experimental electronic circular dichroism (ECD) spectra
with those obtained by conformational analysis and ECD calculations
of representative compounds allowed us to verify the AC of prepared
DNB amides.
brush-type CSP used nowadays, such as Whelk-O1, I or ULMO,
II, possess 3,5-DNB moiety together with another aromatic group.
In the past few years, compounds bearing 3,5-DNB moiety
have also been investigated as potential antituberculosis (anti-
TB) agents. N-substituted 3,5-dinitrobenzamide derivatives, such
as
V and VI (Figure 1), showed high activity against
Mycobacterium tuberculosis.[6-12] These compounds share a
common mechanism of action with a very potent anti-TB
candidate
BZT043
(VII)
and
inhibit
the
enzyme
decaprenylphosphoryl-β-D-ribose 2′-epimerase (DprE1).[7-10] Due
to their potential toxicity, nitroaromatic compounds are frequently
avoided in medicinal chemistry. However, it was demonstrated
that some of these compounds have an acceptable safety index,
in vivo stability and bio-availability.[11,12] Recently, several N-aryl
3,5-dinitrobenzamides, such as VIII, showed inhibition activity
toward human 5-lipoxygenase (5-LOX), a proven target for anti-
inflammatory therapy.[13]
Introduction
Compounds bearing 3,5-dinitrobenzoyl (DNB) together with a
second aryl (Ar) moiety are very intriguing since they posses both
a π-acceptor acid (DNB) and π-donor base (Ar) group. In the
1
1960s and 1970s the rapid expansion of HNMR chiral solvating
agents (CSA) for the determination of optical purity[1,2] led to the
development of a great number of brush-type chiral stationary
phases (CSP) for enantioselective chromatography in Pirkle's
laboratiories.[3,4] There are several examples of 3,5-DNB
derivatives of amines, amino alcohols and amino acids which are
used as CSPs (I, II, III) or CSAs (IV)[5] for the separation of
enantiomers (Figure 1). Particularly, CSPs with 3,5-DNB
derivatives of amino acids were the first commercially available
HPLC chiral stationary phases in the 1980s. Also, most commonly
[a]
Division of Organic Chemistry and Biochemistry
Ruđer Bošković Institute
Bijenička cesta 54, Zagreb, Croatia
E-mail: Anamarija.Knezevic@irb.hr
Division of Physical Chemistry
Ruđer Bošković Institute
Bijenička cesta 54, Zagreb, Croatia
Dipartimento di Chimica e Chimica Industriale
Università di Pisa
Figure 1. Structures of commercially available CSPs (I, II), CSP previously
prepared in our laboratory (III), CSA for NMR analysis of protected amines (IV),
known DNB based anti-TB agents (V-VII), and 5-LOX inhibitor (VIII).
[b]
[c]
Via Moruzzi 13, Pisa, Italy
When the allyl group is introduced into a molecule already
bearing DNB and aromatic groups, it widens its synthetic potential.
The terminal double bond enables the binding to a silica surface
Supporting information for this article is given via a link at the end of
the document
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800346
European Journal of Organic Chemistry
FULL PAPER
and preparation of brush-type CSP. Also, it allows the introduction
of various functional groups using simple methodologies. The
common feature necessary for all the above mentioned
implementations of 3,5-dinitrobenzoyl-aryl-allylamines is the
demand for enantiomerically pure compounds.
A10 with (E)-configuration of double bond were prepared from
aromatic aldehydes and reduced to allyl alcohols B2–B10. The
Overman rearrangement of allyl trichloroacetimidates C1–C10,
obtained from allyl alcohols, afforded 2,2,2-trichloro-N-(1-
arylallyl)acetamides D1–D10 in high yield. Racemic amines 1–10
were obtained after amide hydrolysis.
Previously, we reported the preparation of a series of (S)-1-
aryl-allylamines.[14] The kinetic resolution of racemic amines using
Candida antarctica lipase B (CaLB) in organic solvent provided
amines with excellent enantiomeric excess for amines with no
steric hindrance near the stereogenic center. Herein, we broaden
this methodology to 1-aryl-allylamines bearing 9-phenanthryl, 2-
methylnaphthalen-1-yl and naphthalen-2-yl aromatic groups and
describe the preparation of enantiomerically pure S-3,5-DNB
amides from these (S)-1-aryl-allylamines. Also, we demonstrate
the preparation of enantiomerically pure 3,5-DNB amides from
racemic amides using chiral chromatography on a brush-type
CSP. The absolute configuration of prepared S-DNB amides was
anticipated using the elution order of prepared DNB amides on
brush-type chiral columns and comparison with DNB amides
obtained after acylation of (S)-amines. The configuration was then
confirmed by means of electronic circular dichroism (ECD)
spectroscopy and quantum chemistry calculations.
Results and Discussion
Racemic 1-aryl-allylamines were prepared
from the
corresponding aromatic aldehydes in five steps according to the
procedure described previously (Scheme 1).[14] Acrylic esters A2–
Scheme 1. Preparation of 1-aryl-allylamines.
Table 1. Enzymatic resolution of 1-aryl-allylamines using CaLB
Entry
Amine
Ar
Solvent
Acyl donor[b]
Time [d]
Amine, ee [%]
Yield[c] [%]
1
2
3
4
5
1
2
3
4
5
phenyl[a]
hexane
hexane
hexane
MTBE
iso-propyl acetate
ethyl methoxyacetate
iso-propyl acetate
ethyl methoxyacetate
iso-propyl acetate
2
3
7
8
4
> 99
99
81
63
-
45
39
43
n.i.
0[d]
n.i.
45
48
n.i.
0[d]
49
42
4-methylphenyl[a]
3,5-dimethylphenyl[a]
2-methylphenyl[a]
2,4,6-trimethylphenyl[a]
hexane
6
7
8
5
6
7
2,4,6-trimethylphenyl[a]
MTBE
MTBE
MTBE
MTBE
hexane
hexane
MTBE
ethyl methoxyacetate
ethyl methoxyacetate
ethyl methoxyacetate
ethyl methoxyacetate
iso-propyl acetate
4
5
2
6
3
9
> 99
> 99
14
1-napthalenyl[a]
4-methylnaphthalen-1-yl[a]
2-methylnaphthalen-1-yl
2-methylnaphthalen-1-yl
2-napthalenyl
9
8
10
11
12
8
-
9
10
iso-propyl acetate
ethyl methoxyacetate
11
1
98
99
9-phenanthryl
[a] Described in Reference [14]. [b] Reactions in EtOAc proceed with both enantiomers of amine acylated to amides. [c] Maximum yield is 50 %. [d] Neither of
the enantiomers react. n.i. = not isolated.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800346
European Journal of Organic Chemistry
FULL PAPER
The enzyme-catalyzed resolutions of racemic amines were
performed using CaLB immobilized on
a
polymer carrier
When 1-aryl-allylamines were obtained with excellent ee’s
using CaLB, the corresponding 3,5-DNB derivatives were easily
prepared using 3,5-dinitrobenzoyl chloride. Racemic DNB amides
were also prepared in order to determine the conditions for
enantioseparation on a brush-type chiral column and the elution
order of prepared S-DNB amides. Recrystallization of the
prepared DNB amides increased in some cases their optical purity.
Higher ee’s were obtained for S-DNB-2, S-DNB-9 and S-DNB-10
(Table 2). The preparation of brush-type chiral column (CSP-A)
used for the analysis of DNB amides was previously described in
Ranogajec et al. (CSP 7).[18] This column (Figure 1, III) proved
suitable for the enantioseparation of compounds bearing a 3,5-
dinitrobenzoyl moiety.
(Novozym 435). Consistent with previous research, in the case of
1-aryl-allylamines with no significant steric hindrance, excellent
enantiomeric excesses (ee’s) were obtained (Table 1). When a
methyl group is attached at position 2 of the aromatic ring Ar,
however, it is near the stereogenic center and this disturbs the
enantiorecognition process inside the enzyme. Therefore, modest
or no enantioselectivity was gained for amines 4, 5 and 8.
The absolute configuration of isolated amines was
presumed to be (S), according to previously reported results in
which CaLB always preferentially transformed the (R)-enantiomer
of the amine.[15,16] This was confirmed by comparing the
measured specific rotation of (S)-phenylallylamine
1 with
literature data,[17] which we discussed in a previous paper.[14] It is
reasonable to assume that structurally similar 1-aryl-allylamines
behave in the same way when the reaction is catalyzed by CaLB.
Table 2. Preparation of S-DNB amides and enantioseparation using
analytical CSP-A column.
η
Mobile phase
tR1
tR2
AC[a]
ee[a]
[min]
[min]
[%]
[%]
S-DNB-1
S-DNB-2
81
92
Ψ(hexane, EtOH)
= 6 : 4
Ψ(hexane, 2-
PrOH) = 1 : 1
MeOH
MeOH
Ψ(hexane, EtOH)
= 6 : 4
4.7
6.7
5.6
8.7
S
S
> 99
> 99
S-DNB-6
S-DNB-7
S-DNB-9
90
78
73
3.7
4.2
7.9
4.5
5.5
9.3
S
S
S
> 99
> 99
99
S-DNB-
10
79
86
MeOH
5.3
8.6
9.2
S
S
> 99
> 99
DNB-3
DNB-4
DNB-5
DNB-8
Ψ(hexane, DCM,
MeOH) = 50 : 100 :
1
Ψ(hexane, DCM,
MeOH) = 40 : 60 :
1
Ψ(hexane, DCM,
MeOH) = 65 : 35 :
1.5
Ψ(hexane, THF,
DCM) = 3 : 1 : 1
11.3
Figure 2. Enantioseparation of DNB-4 using CSP-A, Ψ(hexane, DCM, MeOH)
= 40 : 60 : 1; 254 nm; a) semi-preparative column (300 mm x 8 mm), 5 mL/min,
injection volume 20 μL; c) semi-preparative column, 5 mL/min, injection volume
700 μL.
97
91
88
6.3
7.4
5.6
7.8
8.3
7.3
S
S
S
> 99
> 99
99
When enzymatic resolution of 1-aryl-allylamines did not
proceed with high ee’s, racemic DNB amides were prepared. It
was demonstrated that CSP-A could be used for the separation
of enantiomers on a semi-preparative scale (Figure 2). Although
CSP-A was available only in the semi-preparative form, its great
[a] For the first eluting enantiomer after chiral chromatography.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800346
European Journal of Organic Chemistry
FULL PAPER
advantage is a wide range of solvents suitable for mobile phase
since it is a brush-type column. Prepared DNB amides have poor
solubility in nonpolar organic solvents and are soluble only in
solvent mixtures with high content of DCM or THF. Eventually, all
the remaining DNB amides were successfully resolved on CSP-
A, and excellent ee’s were obtained for the first eluting
enantiomers (Table 2).
Using brush-type column CSP-A for the separation of
enantiomers, also enables the prediction of absolute configuration
(AC) of enantiomers. The assumption is that CaLB preferentially
acylates the (R)-amine of the prepared 1-aryl-allylamines
according to Kazlauskas rule.[15] The progress of these kinetic
resolutions was monitored using Chiralcel OD-H. Polysaccharide
CSPs have a very complicated enantioseparation mechanism
In order to substantiate this finding, electronic circular
dichroism (ECD) spectra of obtained DNB amides were recorded
in methanol at room temperature (Figure 3). It is clear from the
recorded spectra that all DNBs bearing a substituted phenyl ring
(DNB-1 to DNB-5) follow the same pattern. The S-enantiomer is
characterized by a positive signal near 260 nm and a negative
signal below 240 nm. The spectra of DNBs bearing substituted 1-
naphthalenyl ring are much more complicated and diverse. When
the methyl group is at position 4, the spectrum is analogous to the
spectrum of S-DNB-6, a species with no substituents on the ring.
On the other hand, the spectrum of S-DNB-8 (with methyl
attached at position 2) demonstrates a completely different shape
of the curve. The ECD of S-DNB-9, featuring a 2-naphthalenyl
substituent, behaves analogous to ECDs of phenyl series of DNB
and the elution order of enantiomers is hard to predict.[19] Contrary, amides, while ECD of S-DNB-10 has again an entirely different
the mechanism of brush-type CSP is often plain and more
studied.[20] For the set of compounds differing only in the methyl
substituents on the aromatic ring, the elution order of enantiomers
at room temperature should remain unchanged.[21-23] Indeed, the
S-DNB amides prepared by acylation of S-amines obtained after
enzymatic resolution, were always the first eluting enantiomers on
CSP-A column. Also, prepared DNBs were analyzed on a
commercial brush-type column Whelk-O1 (Table 3). It is evident
that the resolution of prepared DNBs on Whelk-O1 is excellent.
More importantly, the elution order remains the same with the first
eluting enantiomer being the one obtained after acylation of S-
amine prepared by enzyme resolution. Regarding the previous
analysis, it is reasonable to conclude that with brush-type CSP-A
and Whelk-O1 for the set of prepared DNB amides, the first
eluting enantiomer is always the (S)-enantiomer.
ECD curve. Since these species possess unique chromophores
in our set of compounds, it is understandable that their ECD
spectra are not comparable to other prepared DNBs. However,
these compounds were obtained after enzymatic resolution of 1-
aryl-allylamines. Supported by the fact that the enantiomers
whose spectra are shown in Figure 3 were also the first eluting
enantiomers on the CSP-A and Whelk-O1 columns, we can
confidently assume that these enantiomers are S-DNBs. This
assumption was definitely proved by ECD calculations.
Table 3. Analysis of DNB amides on Whelk-O1.
[a]
Mobile phase
tR1
tR2
tR*
AC[b]
[min]
[min]
[min]
DNB-1
DNB-2
DNB-3
DNB-4
DNB-5
DNB-6
DNB-7
DNB-8
DNB-9
DNB-10
Ψ(hexane, EtOH) = 6 : 4
Ψ(hexane, EtOH) = 6 : 4
Ψ(hexane, EtOH) = 6 : 4
Ψ(hexane, EtOH) = 6 : 4
Ψ(hexane, EtOH) = 6 : 4
Ψ(hexane, EtOH) = 2 : 8
Ψ(hexane, EtOH) = 2 : 8
Ψ(hexane, EtOH) = 2 : 8
Ψ(hexane, EtOH) = 2 : 8
Ψ(hexane, EtOH) = 2 : 8
14.1
14.6
15.0
11.1
10.5
11.3
13.2
9.8
19.3
22.1
28.0
18.4
27.8
34.7
53.2
26.2
66.1
59.2
13.9
14.2
14.9
11.2
10.3
11.2
13.0
10.0
17.4
18.8
S
S
S
S
S
S
S
S
S
S
18.0
18.8
[a] retention time of compounds for which ECD spectra are presented in
Figure 3 – DNB amides prepared by acylation of amines obtained after
enzymatic resolution or first eluting enantiomer in chiral chromatography on
CSP-A. [b] AC of the first eluting enantiomer after chiral chromatography.
Figure 3. Experimental ECD spectra of synthesized compounds recorded in
methanol, at room temperature.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800346
European Journal of Organic Chemistry
FULL PAPER
As demonstrated in Figure 3, the ECD spectra of our DNBs
are mainly dictated by the specific aromatic chromophore
attached at C-3. In fact, ECD is a consequence of the
chromophore(s) present in the chiral molecule, and it is strongly
dependent on both the absolute configuration and the
conformation assumed by the molecule in the solution.[24]
Therefore, in order to obtain a calculated ECD spectrum to
compare with the experimental one, first it is necessary to
investigate the molecular conformational space, followed by the
calculation of ECD of relevant conformers. The aim of the ECD
study was not only to confirm the absolute configuration of DNBs,
but also to rationalize better the impact of steric factors on the
ECD spectra, with special attention to the presence of bulky
groups near the stereogenic center. Especially intriguing is the
fact that for a given chromophore, a different substitution pattern
leads to very different ECD spectra. Our set of compounds
includes two families with phenyl and 1-naphthyl substituent
represented by 5 and 3 members, respectively. Therefore, we
focused on S-DNB-2 and S-DNB-5 as representative of the
phenyl-substituted DNB family, exemplifying respectively the
compound with the smaller and larger steric hindrance of the
series. For the same reason, S-DNB-6 and S-DNB-8 were chosen
as representative of the 1-naphthyl-substituted series.
geometrical difference between close-lying structures is a slightly
different relative orientation of two chromophores inside molecule
(Figure S2). NOESY spectra were recorded in d4-methanol to
further support calculated conformers (Figure 4). Interaction of H3
with H7 and absence of interaction between H3 and H2 in DNB-2
and DNB-5 is an indicator of the position of both aryl moieties in
relation to H3 and is consistent with the calculated structures. Also,
NOESY correlations for DNB-8 between H3 and both H7 and H8
clearly demonstrate the presence of two types of rotamers,
differing in the rotation around C*-Caryl bond (Figure S3). These
interactions are not possible simultaneously in a single conformer.
According to DFT calculations, the most stable conformer of DNB-
8
is a structure with π stacking interactions (‘sandwich’
conformation, Figure S4). However, there are no correlations in
the NOESY spectrum proving interactions between H2 and H3
hydrogens (only 1.88 Å apart in the DFT minimum) and H2 and
H8 (3.22 Å). It must be stressed that, in the presence of multiple
conformers, NOE measurements tend to underestimate the
effective distance because of the <r–6> dependence. This means
that the importance of conformers with the shorter contacts is
amplified at the expense of those with longer contacts. Therefore,
in the absence of NOE correlations, at least the former ones may
be safely excluded. It appears that DFT calculations with both
B3LYP and M06-2X functionals overestimate stability of this
conformer. Therefore, this geometry was excluded from
subsequent computations, otherwise it would definitely worsen
the agreement between experimental and calculated spectra (see
Figure S5). This finding emphasizes the necessity of checking the
reliability of calculated structures with all available experimental
means.[25,26]
ECD calculations may be run with different computational
methods.[27] The time-dependent DFT method (TD-DFT) has
intrinsic problems with unphysically low-lying charge transfer
states expected in multi-chromophore systems like the present
ones.[28] This is especially true for hybrid functionals like the
popular B3LYP because of their incorrect asymptotic behavior.[29]
The problem may be relieved by using range-separated
functionals like CAM-B3LYP[30] and ωB97XD.[31]
A more
computationally costly but more accurate approach requires the
use of correlated wavefunction-based second-order approximate
coupled cluster singles and doubles method (CC2) for calculation
of ECD spectra.[29,32,33] Calculation details are reported in the
Experimental Section. The level of agreement between calculated
and experimental ECD spectra of compounds S-DNB-2, S-DNB-
5, S-DNB-6 and S-DNB-8, supported with previous experimental
evidences, permits us to determine the AC (Figure 5). This is
especially true by looking at CC2 calculation results and, at a
lesser extent, to TDDFT results with CAM-B3LYP and ωB97XD
functionals. For DNB-5 both CAM-B3LYP and ωB97XD
outperform the computationally demanding RI-CC2 method, but
for all other compounds RI-CC2 is the method of choice for the
ECD spectra simulation. Minor differences between RI-CC2
calculated and experimental spectra concern S-DNB-5 (missing
shoulder). The overall satisfactory agreement of spectral line
shapes and relative intensities of the ECD spectra predicted with
RI-CC2, backed with experimental evidences, support our AC
determination for the series of substituted DNBs.
Figure 4. NOESY correlations for DNB-2, DNB-5, DNB-6 and DNB-8. Dashed
lines represent the correlations of lower magnitude in NOESY spectrum.
In vacuo conformational analysis followed by DFT geometry
optimizations indicates the presence of several conformers with
sizable Boltzmann population at 298 K (see Table S1 in
Supporting Information), ranging from two to six conformers
respectively for S-DNB-6 and S-DNB-8. When solvent effects are
included in the DFT optimizations, only slight changes (below 5%)
in conformer populations are observed. The only exception is
DNB-2, where MIN-B becomes the lowest energy structure, being
0.28 kcal mol-1 below MIN-A structure, and conformer MIN-D of
DNB-8, which doubled its population. In the most cases the main
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800346
European Journal of Organic Chemistry
FULL PAPER
Figure 5. Experimental (red) and calculated ECD spectra of S-DNB-2 (upper panel, left), S-DNB-5 (upper panel, right), S-DNB-6 (lower panel, left) and S-DNB-8
(lower panel, right). See Computational Methods for scaling factors of theoretical spectra.
On the contrary, the use of B3LYP was largely
unsatisfactory for all compounds, as it cannot reproduce neither
the shape nor the relative intensities of recorded spectra. The
analysis of the character of transitions confirms the well
documented enormous underestimation of the charge transfer
excitations by B3LYP.[29]
All S-DNBs have weak positive ECD signal above 300 nm,
independent of the aryl substituent on the stereogenic carbon
atom. According to our calculations and orbital analysis of the
most stable conformer, these signals originate from the excitation
from a nonbonding orbitals of nitro groups to the π* orbital of the
DNB moiety and are common for all DNBs studied. Systems with
substituted phenyl or naphthyl moieties on asymmetric carbon
atom have strong positive (S-DNB-2, S-DNB-5, S-DNB-6) or
negative (S-DNB-8) ECD signals with maxima in the range from
254 nm to 260 nm. These excitations are of ππ* type, with
dominant contributions from π orbital of phenyl (naphthyl) and
DNB moiety to the π* orbital of DNB (Figures S6-S13).
C2 dihedral angle -93°, and N-C*-Cα-Cβ of only -7° (Figure S14).
S-DNB-5 has two close lying conformers, MIN-A (N = 0.335) and
MIN-B (N = 0.304). N-C*-C1-C2 dihedral angle of MIN-A conformer
is -55°, while the ethenyl group is oriented towards the phenyl
moiety (N-C*-Cα-Cβ dihedral angle is 138°, Figure S14). All those
geometrical differences are mirrored in ECD spectra. In the
spectrum of S-DNB-5, a negative signal centered at 280 nm is
missing, replaced by a broad positive shoulder. The same effect
is even more pronounced for compounds with naphthyl
substituents (Figure 6). Comparing the minimum energy
structures of S-DNB-6 and S-DNB-8 one can see that a single
methyl moiety on position 2 causes reorientation of naphth-1-yl
ring. Dihedral angle N-C*-C1-C2 is -100° in S-DNB-6 and 122° in
S-DNB-8, with the ethenyl group on asymmetric carbon atom
adjusted to minimize steric repulsions. 1-Naphthyl rings easily
adjust the rotamerism around sp2−sp3 junctions between the
limiting syn- and anti-periplanar conformations.[34] ECD spectra of
these two compounds are almost completely inverted. In fact, the
strong positive signal around 260 nm of S-DNB-6 turns to a weak
negative signal of S-DNB-8; the weak negative signal centered at
280 nm for S-DNB-6 becomes a broad, structured, strong and
positive signal in S-DNB-8.
It is interesting to examine how is the addition of methyl
groups on the aromatic ring reflected on the conformational
equilibrium and subsequently on ECD spectra. Substituting two
hydrogen atoms of phenyl group of S-DNB-2 at ortho position
results in S-DNB-5. MIN-A conformer of S-DNB-2 is dominant
conformer (Boltzmann factor N = 0.367 at 298 K), with N-C*-C1-
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800346
European Journal of Organic Chemistry
FULL PAPER
at position 4 on the aromatic ring. CaLB preferentially acylates the
R-amine according to Kazlauskas rule, leaving the S-amine in the
solution. We demonstrated that racemic 3,5-DNB amides can be
resolved using brush-type CSP-A, previously created in our
laboratory, as well as commercial Whelk-O1 column. The
absolute configuration of prepared S-DNB amides was presumed
using the elution order of prepared DNB amides on CSP-A and
Whelk-O1 columns and comparison with retention times of DNB
amides obtained after acylation of (S)-amines. In order to confirm
the proposed absolute configuration, ECD spectra of all prepared
compounds were recorded and simulated by CC2 calculations.
The agreement of experimental ECD spectra and calculated ECD
of representative compounds allowed us to verify the absolute
configuration of prepared DNB amides. The evaluation of CSPs
obtained after attachment of DNB amides onto silica is in progress.
Also, prepared compounds are being tested for anti-TB activity at
the time.
Cβ
Cα
C2
C*
C1
S-DNB-6
Experimental Section
Cβ
C1
Cα
General Methods: 1H and 13C NMR were recorded on a Bruker AV 300
spectrometer. Chemical shifts (δH and δC) are quoted in parts per million
(ppm), referenced to TMS. Melting points were determined on
Electrothermal 9100 apparatus in open capillaries and are not corrected.
IR spectra were recorded on a Bruker ABB Bowen instrument. Elemental
analyses were done on Perkin-Elmer 2400 CHNS Elemental Analyzer.
Optical rotations were measured using Optical Activity AA-10 automatic
polarimeter. Chiral HPLC analysis were performed using a Shimadzu
Prominence System (Pump LC-20AT, DGU-20A5 Degasser, UV detector
SPD-20A) or Knauer system (Pump Knauer 64, 4-Port Knauer Degasser,
UV detector Knauer Variable Wavelength Monitor, Interface Knauer, and
CD detector Jasco CD-2095). For chiral HPLC analysis of prepared 3,5-
dinitrobenzamides the brush-type CSP-A (Figure 1) and Whelk-O1
columns were used. For ee determination the analytical CSP-A column
(250 mm x 4.6 mm) was used and for preparative HPLC the dimensions
of semi-preparative column used were 300 mm x 8 mm. ECD spectra were
recorded on Jasco J-815 Circular Dichroism (CD) Spectropolarimeter at
room temperature. The samples (1 mg) were dissolved in MeOH (1 mL)
and 60 μL of the solution was dilluted with MeOH to the volume of 2 mL.
C*
C2
S-DNB-8
Figure 6. Structures of the lowest energy conformer (MIN-A) of S-DNB-6 and
S-DNB-8 calculated with DFT, showing the different reciprocal orientation of the
DNB and naphthalene chromophores.
The influence of solvent on ECD spectra was estimated by
the change of relative populations of conformers which is to be
reflected in the total spectrum (see Table S1). But, as can be seen
from Figure S15, this approach demonstrates that in our case for
the determination of the AC we can neglect the solvent effects.
This analysis demonstrates the success of
a
multidisciplinary approach combining experimental techniques
with state-of-the-art methods of computational chemistry to
determine the absolute configuration of a series of molecules with
potential anti-TB activity, and to explain how a small structural
change like methylation can induce a large conformational
rearrangement and cause major differences in the overall ECD
spectra.
General procedure for the synthesis of 3,5-dinitrobenzamides: To a
solution of corresponding amine (1 equiv) in dry THF (7 mL per 1 mmol of
amine) under argon 3,5-dinitrobenzoyl chloride (1.1 equiv) was added. The
resulting solution was cooled to 0 °C and propylene oxide (1 mL per 1
mmol of amine) was added dropwise. The reaction mixture was stirred for
1 h at 0 °C, then for 2 h at room temperature and concentrated under
reduced pressure. The crude product was purified by recrystallization or
column chromatography.
Conclusions
(S)-N-(1-Phenylallyl)-3,5-dinitrobenzamide (S-DNB-1): White solid
(0.20 g, 81 %) starting from (S)-1-phenyl-prop-2-en-1-amine (0.10 g).
Purified by recrystallization from methanol. νmax/cm-1 3300, 1637, 1539,
1342, 1078, 731, 702; 1H NMR (300 MHz, CDCl3) δ 9.15 (1H, t, J = 2.0 Hz,
DNBAr), 8.95 (2H, d, J = 2.0 Hz, DNBAr), 7.42–7.29 (5H, m, Ar), 6.72, (1H,
br s, NH), 6.14 (1H, ddd, J = 5.5, 10.4, 17.1 Hz, CH=CH2), 5.84 (1H, m,
CHNH), 5.38 (1H, d, J = 10.4 Hz, CH2=CH), 5.32 (1H, d, J = 17.1 Hz,
CH2=CH); 13C NMR (75 MHz, CDCl3) δ 161.98, 148.81, 139.46, 137.80,
136.26, 129.22, 128.49, 127.50, 127.36, 121.36, 117.42, 56.67;
C16H13N3O5 requires: C, 58.72; H, 4.00; N, 12.84, found: C, 58.54; H, 4.12;
In conclusion, we described the preparation of ten
enantiomerically pure (S)-N-(1-aryl-allyl)-3,5-dinitrobenzamides
using enzymatic resolution of corresponding 1-aryl-allylamines or
resolution of racemic DNB amides on chiral stationary phases.
The set of prepared DNBs include an aryl moiety with substituted
phenyl and naphthalen-1-yl rings, naphthalen-2-yl and
phenanthryl groups. Enzymatic resolution was achieved with
Candida antarctica lipase B (CaLB) in organic solvent and was
suitable for compounds with no substituents or methyl substituent
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800346
European Journal of Organic Chemistry
FULL PAPER
N, 12.65; ee > 99 %, CSP-A column, hexane-EtOH, 6 : 4, 1 mL/min, 254
nm, rt, tR(S) = 4.7 min., tR(R) = 5.6 min, [α]D25 – 21.2 (c 1.0, THF).
8.92 (2H, d, J = 1.9 Hz, DNBAr), 6.88 (2H, s, Ar), 6.82, (1H, br s, NH),
6.30–6.24 (1H, m, CHNH), 6.16 (1H, ddd, J = 17.2, 10.4, 3.8 Hz, CH=CH2),
5.30 (1H, d, J = 10.4 Hz, CH2=CH), 5.13 (1H, d, J = 17.2 Hz, CH2=CH),
2.45 (6H, s, ArCH3), 2.26 (3H, s, ArCH3); 13C NMR (75 MHz, CDCl3) δ
161.88, 148.70, 137.84, 137.69, 136.34, 136.31, 132.43, 130.32, 127.03,
121.09, 116.34, 52.49, 20.93, 20.76; C19H19N3O5 requires: C, 61.78; H,
5.18; N, 11.38, found: C, 61.39; H, 5.48; N, 10.91
(S)-N-(1-(2,4,6-trimethylphenyl)allyl)-3,5-dinitrobenzamide was prepared
using semi-preparative chiral HPLC chromatography CSP-A column,
hexane-DCM-MeOH, 65 : 35 : 1.5, 5 mL/min, 254 nm, rt. Racemic amide
(100 mg) was dissolved in DCM- hexane, 2 : 3 (12 mL) and 0.3 – 0.4 mL
was injected. Approximately 34 runs were performed to resolve 100 mg of
(S)-N-(1-(4-methylphenyl)allyl)-3,5-dinitrobenzamide
(S-DNB-2):
White solid (1.19, 92%) starting from (S)-1-(4-methylphenyl)-prop-2-en-1-
amine (0.56 g). Purified by recrystallization from methanol. mp 191.9–
192.8 °C; νmax/cm-1 3317, 1639, 1543, 1342, 1077, 921, 730, 722; 1H NMR
(300 MHz, CDCl3) δ 9.14 (1H, t, J = 2.0 Hz, DNBAr), 8.97 (2H, d, J = 2.0
Hz, DNBAr), 7.25 (2H, d, J = 8.1 Hz, Ar), 7.17 (2H, d, J = 8.1 Hz, Ar), 6.95
(1H, br s, NH), 6.14 (1H, ddd, J = 17.1, 10.4, 5.5 Hz, CH=CH2), 5.79 (1H,
m, CHNH), 5.37 (1H, d, J = 10.4 Hz, CH2=CH), 5.31 (1H, d, J = 17.1 Hz,
CH2=CH), 2.34 (3H, s, ArCH3); 13C NMR (75 MHz, CDCl3) δ 161.76,
148.66, 138.21, 137.75, 136.36, 136.27, 129.74, 127.30, 127.20, 121.17,
116.95, 56.29, 21.11; C17H15N3O5 requires: C, 59.82; H, 4.43; N, 12.31,
amide. mp 66.2–67.5 °C; ee > 99 %, CSP-A column, hexane-DCM-MeOH,
25
65 : 35 : 1.5, 1 mL/min, 254 nm, rt, tR(S) = 7.4 min., tR(R) = 8.3 min, [α]D
78.4 (c 1.2, DCM)
-
found: C, 59.44; H, 4.49; N, 12.28; ee > 99 %, CSP-A column, hexane-2-
25
propanol, 1:1, 1 mL/min, 254 nm, rt, tR(S) = 6.7 min, tR(R) = 8.7 min, [α]D
24,2 (c 1.0, DCM).
-
(S)-N-(1-Naphthalen-1-yl-allyl)-3,5-dinitrobenzamide (S-DNB-6): White
solid (1.21 g, 90 %) starting from (S)-1-(1-naphthalenyl)-prop-2-en-1-
amine (0.65 g). Purified by recrystallization from methanol with 1 %
acetone. mp 210.1–211.7 °C; νmax/cm-1 3280, 1644, 1545, 1344, 1074, 922,
805, 782, 731, 717; 1H NMR (300 MHz, CDCl3) δ 9.09 (1H, t, J = 2.2 Hz,
DNBAr), 8.91 (2H, d, J = 2.2 Hz, DNBAr), 8.01 (1H, m, Ar), 7.88–7.83 (2H,
m, Ar), 7.56–7.43 (4H, m, Ar), 6.92 (1H, br s, NH), 6.55 (1H, m, CHNH),
6.26 (1H, ddd, J = 17.1, 10.5, 4.6 Hz, CH=CH2), 5.45 (1H, d, J = 10.5 Hz,
CH2=CH), 5.34 (1H, d, J = 17.1 Hz, CH2=CH); 13C NMR (75 MHz, CDCl3)
δ 161.28, 147.97, 136.68, 135.38, 134.39, 133.47, 130.51, 128.78, 128.46,
126.68, 126.61, 125.79, 124.75, 124.70, 122.48, 120.64, 116.35, 51.77;
C20H15N3O5 requires: C, 63.66; H, 4.01; N, 11.14; found: C, 63.40; H, 3.98;
(S)-N-(1-(3,5-dimethylphenyl)allyl)-3,5-dinitrobenzamide (S-DNB-3):
White solid (0.20, 86%) starting from 1-(3,5-dimethylphenyl)-prop-2-en-1-
amine (0.10 g). Purified by recrystallization from methanol. mp 202.7–
204.2 °C; νmax/cm-1 3313, 1649, 1543, 1345, 1080, 917, 849, 729, 698; 1H
NMR (300 MHz, CDCl3) δ 9.16 (1H, t, J = 2.1 Hz, DNBAr), 8.95 (2H, d, J
= 2.1 Hz, DNBAr), 6.98 (3H, s, Ar), 6.56 (1H, br s, NH), 6.12 (1H, ddd, J =
17.0, 10.3, 5.5 Hz, CH=CH2), 5.77 (1H, m, CHNH), 5.36 (1H, d, J = 10.3
Hz, CH2=CH), 5.31 (1H, d, J = 17.0 Hz, CH2=CH), 2.33 (6H, s, ArCH3); 13
C
NMR (75 MHz, CDCl3) δ 161.68, 148.70, 139.27, 138.82, 137.82, 136.36,
130.02, 127.17, 125.15, 121.13, 116.70, 56.52, 21.30; C18H17N3O5
requires: C, 60.84; H, 4.82; N, 11.83, found: C, 61.14; H, 5.08; N, 11.53
(S)-N-(1-(3,5-dimethylphenyl)allyl)-3,5-dinitrobenzamide was prepared
using semi-preparative chiral HPLC chromatography, CSP-A column,
hexane-DCM-MeOH, 50 : 100 : 1, 5 mL/min, 254 nm, rt. Racemic amide
(100 mg) was dissolved in DCM-hexane, 4 : 1 (20 mL) and 0.8 – 0.9 mL
was injected. Approximately 23 runs were performed to resolve 100 mg of
amide. mp 175.0–175.6 °C; ee > 99 %, CSP-A column, hexane-DCM-
methanol, 50:100:1, 1 mL/min, 254 nm, rt, tR(S) = 8.6 min, tR(R) = 11.3 min,
[α]D25 -33.7 (c 0.9, DCM).
N, 10.85; ee > 99 %, CSP-A column, MeOH, 1 mL/min, 254 nm, rt, tR(S)
3.7 min, tR(R) = 4.5 min, [α]D25 -14.3 (c 1.0, DCM).
=
(S)-N-(1-(4-methylnaphthalen-1-yl)allyl)-3,5-dinitrobenzamide
(S-
DNB-7): Light yellow solid (0.34 g, 78%) starting from (S)-1-(4-
methylnaphthalen-1-yl)-prop-2-en-1-amine (0.22 g). Purified by
recrystallization from MeOH with 1 % DCM. mp 186.0–186.7 °C; νmax/cm-
1 3340, 1642, 1543, 1524, 1343, 921, 766, 732, 721; 1H NMR (300 MHz,
DMSO) δ 9.77 (1H, d, J = 7.9 Hz, NH), 9.14 (2H, d, J = 1.9 Hz, DNBAr),
8.93 (1H, t, J = 1.9 Hz, DNBAr), 8.17–8.13 (1H, m, Ar), 8.09–8.04 (1H, m,
Ar), 7.61–7.55 (2H, m, Ar), 7.50 (1H, d, J = 7.2 Hz, Ar), 7.40 (1H, d, J = 7.2
Hz, Ar), 6.52 (1H, m, CHNH), 6.31 (1H, ddd, J = 17.0, 10.4, 5.4 Hz,
CH=CH2), 5.34 (1H, d, J = 10.4 Hz, CH2=CH), 5.31 (1H, d, J = 17.0 Hz,
CH2=CH), 2.65 (3H, s, ArCH3); 13C NMR (75 MHz, DMSO) δ 161.34,
148.15, 137.25, 136.42, 134.33, 134.09, 132.41, 130.87, 127.79, 126.19,
125.97, 125.72, 124.86, 124.77, 123.65, 120.96, 116.24, 51.71, 19.13;
C21H17N3O5 requires: C, 64.45; H, 4.38; N, 10.74, found: C, 63.67; H, 4.65;
(S)-N-(1-(2-methylphenyl)allyl)-3,5-dinitrobenzamide
(S-DNB-4):
White solid (0.70 g, 97 %) starting from 1-(2-methylphenyl)-prop-2-en-1-
amine (0.31 g). Purified by column chromatography in DCM-MeOH, 100 :
1 (Rf = 0.63). mp 160.2–161.5 °C; νmax/cm-1 3289, 1641, 1540, 1344, 1078,
920, 731, 720; 1H NMR (300 MHz, CDCl3) δ 9.15 (1H, t, J = 2.0 Hz, DNBAr),
8.95 (2H, d, J = 2.0 Hz, DNBAr), 7.33–7.28 (1H, m, Ar), 7.25–7.20 (3H, m,
Ar), 6.68 (1H, br s, NH), 6.14 (1H, ddd, J = 16.9, 10.2, 4.9 Hz, CH=CH2),
6.03 (1H, m, CHNH), 5.37 (1H, d, J = 10.2 Hz, CH2=CH), 5.25 (1H, d, J =
16.9 Hz, CH2=CH), 2.41 (3H, s, ArCH3); 13C NMR (75 MHz, CDCl3) δ
161.71, 148.68, 137.55, 137.27, 136.53, 136.02, 131.08, 128.31, 127.17,
126.65, 126.53, 121.20, 116.74, 53.09, 19.23; C17H15N3O5 requires: C,
59.82; H, 4.43; N, 12.31, found: C, 59.63; H, 4.29; N, 11.99
(S)-N-(1-(2-methylphenyl)allyl)-3,5-dinitrobenzamide was prepared using
semi-preparative chiral HPLC chromatography, CSP-A column, hexane-
DCM-MeOH, 40 : 60 : 1, 5 mL/min, 254 nm, rt. Racemic amide (100 mg)
was dissolved in DCM-hexane, 4 : 1 (15 mL) and 0.7 mL was injected.
Approximately 21 runs were performed to resolve 100 mg of amide. mp
202.3–203.2 °C; ee > 99 %, CSP-A column, hexane-DCM-methanol,
40:60:1, 1 mL/min, 254 nm, rt, tR(S) = 6.3 min, tR(R) = 7.8 min, [α]D25 -23.5
(c 1.0, DCM)
N, 10.54; ee > 99 %, CSP-A column, MeOH, 1 mL/min, 254 nm, rt, tR(S)
4.2 min, tR(R) = 5.5 min, [α]D25 -21.8 (c 1.2, DCM).
=
(S)-N-(1-(2-methylnaphthalen-1-yl)allyl)-3,5-dinitrobenzamide
(S-
DNB-8): Yellow solid (1.04 g, 88%) starting from 1-(2-methylnaphthalen-
1-yl)-prop-2-en-1-amine (0.59 g). Purified by recrystallization from
methanol with 5 % DCM. mp 233.4–234.2 °C; νmax/cm-1 3321, 1638, 1543,
1347, 1074, 819, 730, 718; 1H NMR (300 MHz, DMSO) δ 9.90 (1H, d, J =
5.4 Hz, NH), 9.06 (2H, d, J = 2.0 Hz, DNBAr), 8.91 (1H, t, J = 2.0 Hz,
DNBAr), 8.37 (1H, d, J = 8.4 Hz, Ar), 7.86 (1H, d, J = 7.6 Hz, Ar), 7.76 (1H,
d, J = 8.4 Hz, Ar), 7.52–7.40 (2H, m, Ar), 7.36 (1H, d, J = 8.4 Hz, Ar), 6.50
(1H, ddd, J = 17.0, 10.4, 4.9 Hz, CH=CH2), 6.41 (1H, m, CHNH), 5.25 (1H,
d, J = 10.4 Hz, CH2=CH), 5.04 (1H, d, J = 17.0 Hz, CH2=CH), 2.68 (3H, s,
ArCH3); 13C NMR (75 MHz, DMSO) δ 161.93, 148.04, 137.03, 136.55,
134.74, 133.09, 132.82, 130.67, 129.48, 128.65, 127.83, 127.80, 125.55,
125.11, 124.47, 120.85, 116.50, 53.07, 20.78; C21H17N3O5 requires: C,
64.45; H, 4.38; N, 10.74, found: C, 63.80; H, 4.49; N, 10.47
(S)-N-(1-(2,4,6-trimethylphenyl)allyl)-3,5-dinitrobenzamide (S-DNB-5):
White solid (0.66 g, 91%) starting from 1-(2,4,6-trimethylphenyl)-prop-2-
en-1-amine (0.34 g). Purified by column chromatography in DCM (Rf =
0.42). mp 186.3–187.8 °C; νmax/cm-1 3362, 1643, 1541, 1346, 1076, 919,
(S)-N-(1-(2-methylnaphthalen-1-yl)allyl)-3,5-dinitrobenzamide
was
1
prepared using semi-preparative chiral HPLC chromatography, CSP-A
731, 719; H NMR (300 MHz, CDCl3) δ 9.15 (1H, t, J = 1.9 Hz, DNBAr),
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800346
European Journal of Organic Chemistry
FULL PAPER
column, hexane-THF-DCM, 3 : 1 : 1, 5 mL/min, 254 nm, rt. Racemic amide
(100 mg) was dissolved in the mixture of THF (11 mL) and mobile phase
(6 mL) and 0.6 mL was injected. Approximately 28 runs were performed to
resolve 100 mg of amide. mp 191.2–192.0 °C; ee 99 %, CSP-A column,
where ΔE is transition energy, Δ is half the bandwidth at 1/e peak height
expressed in energy units, and ΔEi and Ri are the transition energy and
rotatory strength for transition i, respectively.[42] For each conformer
rotatory strengths were scaled by Boltzmann factor, and for the half the
bandwidth at 1/e peak height value of 15 nm is used. The conformer with
π stacking interactions (‘sandwich’ conformation) was excluded from
subsequent computations since there are no correlations in NOESY
spectrum for interactions between H2 and H3 (only 1.88 Å apart) and H2
and H8 (3.22 Å) (Figure S4). Simulated spectra were scaled and
wavelength shifted to maximize overlap with experimental ECD spectra
(Table 4). All geometry optimizations, B3LYP and RI-CC2 calculations
were run in Turbomole 7.0,[43] with default grid size and convergence
criteria. For M06-2X single point energy calculations, and CAM-B3LYP
and ωB97X-D excited states calculations Gaussian 09 was used.[44]
hexane-THF-DCM, 3 : 1 : 1, 1 mL/min, 254 nm, rt, tR(S) = 5.6 min., tR(R)
7.3 min, [α]D25 -24.6 (c 1.2, DCM)
=
(S)-N-(1-Naphthalen-2-yl-allyl)-3,5-dinitrobenzamide
(S-DNB-9):
Yellow solid (0.30 g, 73 %) starting from (S)-1-(2-naphthalenyl)-prop-2-en-
1-amine (0.20 g). Purified by recrystallization from methanol with 5 % DCM.
mp 193.0–193.9 °C; νmax/cm-1 3298, 1636, 1541, 1346, 1077, 935, 920,
825, 731, 721; 1H NMR (300 MHz, DMSO) δ 9.81 (1H, d, J = 8.1 Hz, NH),
9.15 (2H, d, J = 2.1 Hz, DNBAr), 8.96 (1H, t, J = 2.1 Hz, DNBAr), 7.95–
7.88 (4H, m, Ar), 7.57 (1H, d, J = 8.5 Hz, Ar), 7.53–7.49 (2H, m, Ar), 6.29
(1H, ddd, J = 17.1, 10.3, 6.4 Hz, CH=CH2), 5.95 (1H, m, CHNH), 5.34 (1H,
d, J = 17.1 Hz, CH2=CH), 5.32 (1H, d, J = 10.3 Hz, CH2=CH); 13C NMR
(75 MHz, DMSO) δ 162.06, 148.64, 138.90, 137.96, 137.19, 133.28,
132.73, 128.58, 128.30, 128.21, 127.97, 126.77, 126.46, 126.09, 125.95,
121.42, 117.18, 56.59; C20H15N3O5 requires: C, 63.66; H, 4.01; N, 11.14;
found: C, 63.35; H, 3.95; N, 10.97; ee 99 %, CSP-A column, hexane-EtOH,
6 : 4, 1 mL/min, 254 nm, rt, tR(S) = 7.9 min, tR(R) = 9.3 min, [α]D25 -32.1 (c
1.1, DCM).
Table 4. Scaling (y = intensity) and shifting wavelength (x) of simulated ECD
spectra.
B3LYP
CAM-B3LYP
ωB97X-D
RI-CC2
S-DNB-2
S-DNB-5
S-DNB-6
S-DNB-8
x - 20 nm
y*3
x + 35 nm
y/3
x + 25 nm
(S)-N-(1-Phenanthren-9-yl-allyl)-3,5-dinitrobenzamide
(S-DNB-10):
y*5
Yellow solid (0.19 g, 79%) starting from (S)-1-(phenanthren-9-yl)-prop-2-
en-1-amine (0.13 g). Purified by recrystallization from MeOH. mp 252.4–
253.6 °C; νmax/cm-1 3423, 1622, 1532, 1347, 731, 725, 717; 1H NMR (300
MHz, DMSO) δ 9.82 (1H, d, J = 7.9 Hz, NH), 9.16 (2H, d, J = 2.1 Hz,
DNBAr), 8.93 (1H, t, J = 2.1 Hz, DNBAr), 8.92–8.88 (1H, m, Ar), 8.83 (1H,
d, J = 7.8 Hz, Ar), 8.21–8.17 (1H, m, Ar), 8.01 (1H, d, J = 7.6 Hz, Ar), 7.92
(1H, s, Ar), 7.74–7.63 (4H, m, Ar), 6.57 (1H, m, CHNH), 6.43 (1H, ddd, J
= 16.9, 10.3, 5.3 Hz, CH=CH2), 5.43 (1H, d, J = 10.3 Hz, CH2=CH), 5.42
(1H, d, J = 16.9 Hz, CH2=CH); 13C NMR (75 MHz, DMSO) δ 161.49, 148.16,
136.85, 136.31, 134.35, 130.74, 130.25, 129.83, 129.66, 128.61, 127.83,
127.23, 127.18, 127.10, 126.73, 126.08, 123.89, 123.57, 122.77, 120.99,
116.62, 51.88; C24H17N3O5 requires: C, 67.44; H, 4.01; N, 9.83, found: C,
66.95; H, 4.27; N, 9.32; ee > 99 %, CSP-A column, MeOH, 1 mL/min, 254
nm, rt, tR(S) = 5.3 min, tR(R) = 9.2 min, [α]D25 -172.2 (c 1.3, THF).
x - 10 nm
x + 35 nm
y/3
x + 30 nm
y/2
x + 5 nm
y*2
x + 10 nm
y*2
x + 35 nm
y/3
x + 15 nm
x + 10 nm
x + 35 nm
x + 15 nm
Acknowledgements
This work was supported by the Croatian Science Foundation
under project no. IP-2016-06-1142 (LightMol). The authors
acknowledge generous computer time provided by the Croatian
National Grid Infrastructure (CRONGI).
Computational Methods: Conformational analysis of S-DNB-2, S-DNB-
5, S-DNB-6 and S-DNB-8 have been carried out using the Monte Carlo
algorithm and the Merck molecular force field (MMFF)[35] with Spartan 14
package.[36] For all in vacuo MMFF minima, DFT-based optimizations
exploiting B3LYP hybrid functional[37] and def-SVP basis set were
performed, followed by high level optimization with Ahlrichs def2-TZVPP
basis set.[38] Minnesota global hybrid functional M06-2X[39] with 54 % of the
exact exchange was used to calculate relative energies of the conformers
to evaluate room temperature Boltzmann populations. Only for those
conformers with Boltzmann population above 1 %, 15 lowest singlet
excited states and associated rotatory strengths in length gauge are
calculated using three DFT functionals (B3LYP,[37] CAM-B3LYP,[30]
ωB97X-D[31]) and state-of-the-art approximate coupled cluster singles and
doubles model (CC2)[32] with the resolution of the identity approximation
for two-electron integrals (RI)[40].and def2-TZVPP basis set. When B3LYP
functional was used, 25 states were calculated. Possible change of
conformer populations due to solvation was investigated by performing
single point energy calculations on M06-2X/def2-TZVPP level, while the
Polarizable Continuum Model (PCM)[41] was used to incorporate solvent
effects. To simulate electronic circular dichroism (ECD) spectra electronic
Keywords: Ab initio calculations • Chiral resolution • Circular
dichroism • Configuration determination • Liquid chromatography
[1]
[2]
[3]
[4]
[5]
[6]
T. J. Wenzel, C. D. Chisholm, Prog. Nucl. Mag. Res. Sp. 2011, 59, 1–63.
D. Parker Chem. Rev. 1991, 91, 1441-1457.
C. J. Welch, J. Chromatogr. A 1994, 666, 3-26.
F. Gasparrini, D. Misiti, C. Villani, J. Chromatogr. A. 2001, 906, 35–50.
D. P. Iwaniuk, C. Wolf, J. Org. Chem. 2010, 75, 6724–6727.
B. Kappes, T. Neumann, E. Unger, H.-M. Dahse, U. Mollmann, B.
Schlegel, Patent DE-10158057, 2003, Chem Abstr 105.236177.
C. Trefzer, M. Rengifo-Gonzalez, M.J. Hinner, P. Schneider, V. Makarov,
S.T. Cole, K. Johnsson, J. Am. Chem. Soc. 2010, 132, 13663-13665.
S.M. Batt, T. Jabeen, V. Bhowruth, L. Quill, P.A. Lund, L. Eggeling, L.J.
Alderwick, K. Fütterer, G.S. Besra, P. Natl. Acad. Sci. USA, 2012, 109,
11354-11359.
[7]
[8]
[9]
T. Christophe, M. Jackson, K.J. Hee, D. Fenistein, M. Contreras-
Dominguez, J. Kim, A. Genovesio, J.-P. Carralot, F. Ewann, E.H. Kim,
S.Y. Lee, S. Kang, M.J. Seo, J.P. Eun, H. Škovierová, H. Pham, G.
Riccardi, N.J. Youn, L. Marsollier, M. Kempf, M.-L. Joly-Guillou, T. Oh,
K.S. Won, Z. No, U. Nehrbass, R. Brosch, S.T. Cole, P. Brodin, PLoS
Pathog 2009, 5, e1000645
transitions
are
convoluted
by
Gaussian
functions
ퟐ
푬−∆푬
∆
풊
ퟏ
ퟏ
×
∑ ∆푬풊푹풊풆−(
풊
)
( )
∆휺 푬 =
ퟐ. ퟐퟗퟔ × ퟏퟎ−ퟑퟗ
∆ 흅
√
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800346
European Journal of Organic Chemistry
FULL PAPER
[10] A.L. de Jesus Lopes Ribeiro, G. Degiacomi, F. Ewann, S. Buroni, M.L.
Incandela, L.R. Chiarelli, G. Mori, J. Kim, M. Contreras-Dominguez, Y.-
S. Park, S.-J. Han, P. Brodin, G. Valentini, M. Rizzi, G. Riccardi, M.R.
Pasca, PLoS One 2011, 6, e26675.
[29] A. Dreuw, M. Head-Gordon, Chem. Rev. 2005, 105, 4009–4037.
[30] T. Yanai, D.P. Tew, N.C. Handy, Chem. Phys. Lett., 2004, 393, 51-57.
[31] J.-D. Chai, M. Head-Gordon, Phys. Chem. Chem. Phys., 2008, 10, 6615-
6620.
[11] G. Munagala, K.R. Yempalla, S.K. Aithagani, N.P. Kalia, F. Ali, I. Ali, V.S.
Rajput, C. Rani, R. Chib, R. Mehra, A. Nargotra, I.A., Khan, R.A.,
Vishwakarma, P.P. Singh, MedChemComm 2014, 5, 521-527.
[12] R. Tiwari, G.C. Moraski, V. Krchňák, P.A. Miller, M. Colon-Martinez, E.
Herrero, A.G. Oliver, M.J. Miller, J. Am. Chem. Soc. 2013, 135, 3539-
3549.
[32] O. Christiansen, H. Koch, P. Jørgensen, Chem. Phys. Lett. 1995, 243,
409–418.¸
[33] Z. Brkljača, M. Mališ, D. M. Smith, A.-S. Smith, J. Chem. Theory Comput.
2014, 10, 3270–9.
[34] D. Casarini, L. Lunazzi, A. Mazzanti, J. Org. Chem., 1997, 62, 3315–
3323.
[13] E. Shang, Y. Liu, Y. Wu, W. Zhu, C. He, L. Lai, Bioorg. Med. Chem. 2014,
22, 2396–2402.
[35] T. A. Halgren, J. Comput. Chem. 1996, 17, 490-519.
[36] Spartan'14, Wavefunction, Inc., Irvine, CA
[14] A. Knežević, G. Landek, I. Dokli, V. Vinković, Tetrahedron: Asymmetry
2011, 22, 936-941.
[37] A. D. Becke, J. Chem. Phys. 1993, 98, 5648-5652.
[38] F. Weigend, M. Häser, H. Patzelt, R. Ahlrichs, Chem. Phys. Lett. 1998,
294, 143-152.
[15] R.J. Kazlauskas, A.N.E. Weissfloch, A.T. Rappaport, L.A. Cuccia, J. Org.
Chem. 1991, 56, 2656–2665.
[39] Y. Zhao, D.G. Truhlar, Theor. Chem. Account 2008, 120, 215-241.
[40] C. Hättig, F. Weigend, J. Chem. Phys. 2000, 113, 5154-5161.
[41] J. Tomasi, B. Mennucci, R. Cammi, Chem. Rev., 2005, 105, 2999-3094.
[42] P. J. Stephens, N. Harada, Chirality 2010, 22, 229–233.
[43] F. Furche, R. Ahlrichs, C. Hättig, W. Klopper, M. Sierka, F. Weigend,
Wiley Interdiscip. Rev. Comput. Mol. Sci. 2014, 4, 91–100.
[44] Gaussian 09, Revision D.01, M. J. Frisch, G. W. Trucks, H. B. Schlegel,
G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
G. A. Petersson, H. Nakatsuji, X. Li, M. Caricato, A. Marenich, J. Bloino,
B. G. Janesko, R. Gomperts, B. Mennucci, H. P. Hratchian, J. V. Ortiz,
A. F. Izmaylov, J. L. Sonnenberg, D. Williams-Young, F. Ding, F.
Lipparini, F. Egidi, J. Goings, B. Peng, A. Petrone, T. Henderson, D.
Ranasinghe, V. G. Zakrzewski, J. Gao, N. Rega, G. Zheng, W. Liang, M.
Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T.
Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, K. Throssell, J. A.
Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E.
Brothers, K. N. Kudin, V. N. Staroverov, T. Keith, R. Kobayashi, J.
Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J.
Tomasi, M. Cossi, J. M. Millam, M. Klene, C. Adamo, R. Cammi, J. W.
Ochterski, R. L. Martin, K. Morokuma, O. Farkas, J. B. Foresman, D. J.
Fox, Gaussian, Inc., Wallingford CT, 2016.
[16] V. Gotor-Fernández, E. Busto, V. Gotor, Adv. Synth. Catal. 2006, 348,
797-812. and references cited therein
[17] M. Pallavicini, E. Valoti, L. Villa, O. Piccolo, Tetrahedron: Asymmetry
2000, 11, 4017–4025.
[18] A. Ranogajec, D. Kontrec, V. Vinković, V. Šunjić, J. Liq Chromatogr.,
Relat. Technol. 2003, 26, 63-83.
[19] M. Lämmerhofer, J. Chromatogr. A 2010, 1217, 814-856.
[20] C. Fernandes, M.E. Tiritan, M. Pinto, Chromatographia 2013, 76, 871-
897.
[21] C. Wolf, L. Pranatharthiharan, E. C. Volpe, J. Org. Chem. 2003, 68,
3287-3290.
[22] B. Liu, Y. Zhou, G.-S. Yang, H.Y. Aboul-Enein, Chirality 2013, 25, 101–
106.
[23] W.H. Pirkle, K.Z. Gan, L.J. Brice, Tetrahedron: Asymmetry 1996, 7,
2813-2816.
[24] G. Pescitelli, L. Di Bari, N. Berova, Chem. Soc. Rev. 2011, 40, 4603-
4625.
[25] G. Pescitelli, T. Bruhn, Chirality 2016, 28, 466–474.
[26] A. Mazzanti, D. Casarini, WIREs Comput. Mol. Sci. 2012, 2, 613–641.
[27] M. Srebro-Hooper, J. Autschbach, Annu. Rev. Phys. Chem. 2017, 68,
399-420.
[28] C. Adamo, D. Jacquemin, Chem. Soc. Rev. 2013, 42, 845–856.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800346
European Journal of Organic Chemistry
FULL PAPER
FULL PAPER
Enantiomerically pure (S)-N-(1-aryl-
allyl)-3,5-dinitrobenzamides were
prepared using enzymatic resolution
and chiral chromatography. Absolute
configuration of prepared S-DNB
amides was anticipated using the
elution order on brush-type chiral
columns and verified by agreement
of experimental and calculated ECD
spectra.
Absolute configuration
determination
Anamarija Knežević,* Jurica Novak,
Gennaro Pescitelli and Vladimir
Vinković
Page No. – Page No.
Absolute configuration
determination of (S)-N-(1-aryl-allyl)-
3,5-dinitrobenzamides and elution
order on brush-type chiral
stationary phases
This article is protected by copyright. All rights reserved.