Monomeric and Dimeric 9-O Anthraquinone and Phenanthryl Derivatives of Cinchona Alkaloids as Chiral Solvating Agents for the NMR Enantiodiscrimination of Chiral Hemiesters

Article abstract of DOI:10.1002/chir.22488

Mono- and bis-alkaloid chiral auxiliaries with anthraquinone or phenanthryl cores were probed as chiral solvating agents (CSAs) for the enantiodiscrimination of chiral cyclic hemiesters. The dimeric anthraquinone derivative and the monomeric phenanthryl o

Full text of DOI:10.1002/chir.22488

CHIRALITY 27:693–699 (2015)
Special Issue Article
Monomeric and Dimeric 9-O Anthraquinone and Phenanthryl
Derivatives of Cinchona Alkaloids as Chiral Solvating Agents for the
†
NMR Enantiodiscrimination of Chiral Hemiesters
GLORIA UCCELLO BARRETTA,* ALESSANDRO MANDOLI, FEDERICA BALZANO, FEDERICA AIELLO, BEATRICE DE NICOLA, AND
ALESSANDRO DEL GRANDE
Dipartimento di Chimica e Chimica Industriale, Università di Pisa, Pisa, Italy
ABSTRACT
Mono- and bis-alkaloid chiral auxiliaries with anthraquinone or phenanthryl
cores were probed as chiral solvating agents (CSAs) for the enantiodiscrimination of chiral cyclic
hemiesters. The dimeric anthraquinone derivative and the monomeric phenanthryl one showed
remarkable efficiency in the nuclear magnetic resonance (NMR) differentiation of enantiomeric
mixtures of hemiesters. An anthraquinone analogous with a single alkaloid unit was remarkably
less effective. The conformational prevalence of the chiral auxiliaries were ascertained by NMR.
Chirality 27:693–699, 2015. © 2015 Wiley Periodicals, Inc.
KEY WORDS: NMR spectroscopy; cinchona alkaloids; hemiesters; enantiodiscrimination
phenomena; chiral solvating agents
Applications of cinchona alkaloids as chiral auxiliaries in
organic chemistry date back to late 1853, with their use as
conditions. In an attempt to rationalize the enantiodiscrimination
1
processes, the stereochemical preferences of CSAs were com-
pared by NMR and some relevant complexation parameters
were evaluated.
resolving agents. Since then a great number of applications
could be found spanning from asymmetric synthesis and
2
–7
7–9
catalysis,
to chromatographic
and spectroscopic enan-
tioseparations.1
0,11
Cinchona alkaloids enantiodifferentiating
MATERIALS AND METHODS
ability may be ascribed to their highly preorganized struc-
ture, which makes several interaction sites available for the
stabilization of supramolecular aggregates with different
classes of enantiomeric substrates. In the field of chiral
auxiliaries for the analytical enantiodiscrimination by nuclear
magnetic resonance (NMR) spectroscopy, both quinine or
quinidine and their derivatives at the C-9 hydroxyl site,
carbon–carbon double bond, or quinuclidine nitrogen have
been extensively exploited as chiral solvating agents
All the reactions involving sensitive compounds were carried out under dry
nitrogen, in flame-dried glassware. Before use, THF, MeOH, and benzyl alco-
hol were freshly distilled under dry nitrogen from the proper drying agent.
The other compounds were commercially available and used as received.
High-performance liquid chromatography (HPLC) analyses were
carried out on a Jasco (Tokyo, Japan) PU-1580 chromatograph, equipped
with an UV-1575 detector. Optical rotations were measured as solutions in
1
dm cells at the sodium D line, using a Jasco DIP360 polarimeter. For the
1 13
intermediates in the synthesis of 3, H and C NMR spectra were
recorded using a Varian (Palo Alto, CA) XL-300 spectrometer operating
10,11
(
CSAs).
Often, stimulating hints in this respect have been
1
13
at 300 MHz and 75 MHz for H and C, respectively. NMR measure-
ments were performed on a Varian Inova 600 spectrometer operating at
obtained from the related field of cinchona alkaloid-based
selectors for chiral chromatography applications,
7–9
1
2
13
19
as well
600 MHz, 92 MHz, 150 MHz, and 564 MHz for H, H, C, and F,
respectively. The temperature was controlled to ±0.1ꢀC. The 2D NMR
spectra were obtained using standard sequences with the minimum spec-
tral width required. Proton 2D gradient correlated spectroscopy (gCOSY)
spectra were recorded with 256 increments of four scans and 2K data
points. The relaxation delay was 2 s. 2D TOCSY spectra were recorded
by employing a mixing time of 80 ms. The pulse delay was maintained
at 1 s; 256 increments of four scans and 2K data points each were
as from the growing applications of alkaloid derivatives for
asymmetric catalysis. The latter is indeed the case of mono-
and bis-quinidine derivatives examined in the present contri-
bution. In the course of a spectroscopic investigation on the
mechanism of the asymmetric methanolysis of achiral meso-
12
anydrides, the possibility was suggested of exploiting the
chiral discrimination ability of the organocatalytic systems
for the in situ NMR determination of enantiomeric purities
of reaction products. Therefore, we were encouraged to go
deeper in their potentialities as CSAs for the differentiation
of NMR signals of chiral hemiesters, focusing in particular
[This article is part of the Thematic Issue: “Chirality in Separation Science
and Molecular Recognition Honoring Prof. F. Gasparrini.” See the first articles
for this special issue previously published in Volume 27:9. More special arti-
cles will be found in this issue as well as in those to come.]
on the commercial derivatives (DHQD) AQN (1) and
*Correspondence to: G. Uccello-Barretta, Dipartimento di Chimica e Chimica
Industriale, Università di Pisa, via Giuseppe Moruzzi 13, 56124 Pisa, Italy.
E-mail: gloria.uccello.barretta@unipi.it
2
DHQNPHEN (2), as well as DHQDAQN (3) as an on-
purpose synthesized monomeric analog of 1 (Scheme 1).
NMR enantiodiscrimination experiments were performed
with different hemiesters 4–8 (Scheme 2), some of which
have fluorinated or deuterated probes that lead to remark-
able spectral simplification and, hence, better analytical
†
th
Dedicated to Prof. Francesco Gasparrini, Rome, on the occasion of his 70
birthday.
Received for publication 28 May 2015; Accepted 9 July 2015
DOI: 10.1002/chir.22488
Published online 11 August 2015 in Wiley Online Library
(wileyonlinelibrary.com).
©
2015 Wiley Periodicals, Inc.

6
94
BARRETTA ET AL.
Scheme 1. CSAs 1–3 with numbering scheme for NMR analysis.
Scheme 2. Synthesis of hemiesters 4–8 with numbering scheme for NMR analysis.
collected. The 2D ROESY experiments were performed in the phase-
sensitive mode, by employing a mixing time of 0.6/0.3 s. The pulse delay
was maintained at 1 s; 256 increments of 16 scans and 2K data points each
were collected. Proton 1D TOCSY spectra were recorded using selective
pulses generated by means of the Varian Pandora software. The selective
(SiO
as an orange powder (0.164 g, 25%). H NMR (300 MHz, CDCl
8.20-8.13 (m, 2H), 7.75-7.67 (m, 2H), 7.56 (d, J = 8.4 Hz, 2H), 7.44-7.36
(m, 5H), 5.25 (s, 2H). C NMR (75 MHz, CDCl , δ): 181.9, 173.6, 172.9,
3
157.6, 155.6, 154.1, 136.1, 134.3, 134.1, 133.6, 133.2, 128.8, 128.4, 127.0,
2 2 2
/CH Cl - n-hexane 2:1) to give 1-benzyloxy-4-fluoroanthraquinone
1
3
, δ):
1
3
1
D TOCSY spectra were acquired with 256 scans in 32K data points with
126.6, 124.2, 123.8, 122.6, 122.4, 122.3, 122.2, 107.9, 71.8. MS (ESI, m/z):
= [M+H] 333, [M+MeCN+H] 374.
1
13
+
+
a 1-s relaxation delay and a mixing time of 80 ms. The H- C gradient
heteronuclear single quantum correlation (gHSQC) and gradient
heteronuclear multiple bond correlation (gHMBC) spectra were re-
corded with 256 or 128 time increments of 24 scans. The gHMBC exper-
iments were optimized for a long-range H- C coupling constant of 8 Hz
and a delay period of 3.5 ms for suppression of one-bond correlation sig-
nals. No decoupling was used during the acquisition. The stoichiometry
of (1S,2R)-4/1, (1R,2S)-4/1, (1S,2R)-4/2, and (1R,2S)-4/2 complexes
was determined by employing Job’s method,13 by maintaining the total
concentration constant (5 mM), and by varying the molar fraction of 4
from 0 to 1. Racemic 4–8 and enantiomerically enriched (1R,2S)-4 and
Synthesis of 1-benzyloxy-4-[9-O-(10,11-dihydroquinidinyl)]an-
thraquinone (3). A 50-mL two-necked flask was charged under nitro-
gen with dihydroquinidine (0.147 g, 0.45 mmol), NaH 60% in oil (0.027
g, 0.67 mmol), and 3 mL of DMF. The reaction mixture was heated to
60ꢀC for 30 min, then 1-benzyloxy-4-fluoroanthraquinone (0.149 g, 0.45
mmol) was added. The resulting solution was kept at 60ꢀC overnight,
1
13
monitoring the reaction course by TLC (SiO
The mixture was quenched by adding 5 mL of water and 10 mL of
HCl (10%) and then extracted with CH Cl (3 × 10 mL). The combined
organic layers were washed with water (3 × 10 mL), dried (Na SO ),
and concentrated with a rotary evaporator. The crude product (0.245
g) was purified by flash chromatography (SiO /AcOEt - CH Cl 7:1) to
give 4 as an orange powder (0.171 g, 59%). H NMR (600 MHz, CDCl
δ): 8.65 (d, J2-1 = 4.6 Hz, 1H, H2), 8.23 (m, 1H, Hb’), 8.20 (m, 1H, Hb),
.05 (d, J3-4 = 9.2 Hz, 1H, H3), 7.77 (m, 1H, Hc’), 7.76 (m, 1H, Hc),
2
, AcOEt:MeOH = 7:1).
2
2
(1S,2R)-4 were obtained by alcoholysis of the corresponding anhydride,
2
4
12,14
according to published methods.
The enantiomeric purity of the latter
-1
was checked by chiral HPLC (Chiralcel OJ, 1 mL min n-hexane :
-propanol : trifluoroacetic acid = 95:5:0.1, 210 nm: t1R,2S-4 = 11.5 min,
2
2
2
1
2
3
,
1
2,14
t1S,2R-4 = 14.5 min).
8
Synthesis of 1-benzyloxy-4-fluoroanthraquinone. A 100-mL two-
7.61 (d, J1-2 = 4.6 Hz, 1H, H1), 7.61 (br s,1H, H5) 7.50 (d, Je-f = 7.5 Hz,
2H, He), 7.42 (dd, J4-3 = 9.2 Hz, J4-5 = 2.3 Hz, 1H, H4), 7.36 (t, Jf-g = Jf-e
= 7.5 Hz, 2H, Hf), 7.28 (d, Jg-f = 7.5 Hz, 1H, Hg), 7.07 (br s, 2H,
Ha/Ha’), 6.75 (br s, 1H, H8), 5.14 (m, 2H, Hd/Hd’), 4.06 (s, 3H,
OMe), 3.36 (m, 1H, H9), 3.25 (m, 1H, H19), 3.13 (m, 1H, H16), 3.06
(m, 1H, H18), 2.91 (m, 1H, H15), 2.85 (m, 1H, H10), 1.95 (m, 1H,
H12), 1.76 (m, 1H, H20), 1.66 (m, 2H, H13/H14), 1.60 (m, 1H, H17),
necked flask was charged under nitrogen with 210 μL (2.0 mmol) of ben-
zyl alcohol, 5 mL of THF, and 0.115 g (2.8 mmol) of NaH 60% in oil. The
reaction mixture was heated to 60ꢀC and kept under stirring for 30 min.
Then 0.491 g (2.0 mmol) of 1,4-difluoroanthraquinone were added and
the resulting solution was heated to 75ꢀC overnight. Since the starting
2 2 2
anthraquinone was still present (TLC, SiO , CH Cl ), a second portion
1
3
of 0.050 mL (0.5 mmol) of benzyl alcohol and 0.0261 g (0.6 mmol) of
NaH (60% in oil) were added and the solution was kept at 75ꢀC overnight.
The mixture was quenched by addition of 5 mL of water, and the aqueous
1.46 (m, 1H, H11), 0.92 (t, J21-20 = 7.2 Hz, 3H, H21). C NMR (150
MHz, CDCl , δ): 177.9, 177.2, 142.1, 139.1, 130.8, 128.8, 128.7, 127.9,
3
127.7, 126.6, 123.1, 122.4, 121.4, 121.2, 120.9, 120.5, 118.3, 118.0, 117.1,
layers were extracted with CH
washed with water (10 mL), dried (Na
2
Cl
2
(3 × 20 mL). The organic layers were
SO ), and concentrated with a ro-
115.4, 113.7, 95.0, 71.8, 66.3, 54.7, 45.2, 44.4, 20.9, 18.6. MS (ESI, m/z):
+
25
3
À1
À1
À3
2
4
= [M+H] 639. [α]
D
= -310 deg cm g dm (c = 0.5 and gcm in
tary evaporator. The crude product was purified by flash chromatography
dichloromethane).
Chirality DOI 10.1002/chir

NMR ENANTIODISCRIMINATION OF CHIRAL HEMIESTERS
695
RESULTS AND DISCUSSION
Synthesis of Chiral Auxiliaries and Substrates
ROEs between anthraquinone Ha/H8 protons and Hb/methyl
group of the ethyl groups.
With the plan of carrying out chiral discrimination
measurements also in a solvent less expensive than toluene-d8
CSAs 1 and 2 are commercially available, whereas for the
preparation of the monomeric compound 3 a synthetic route
analogous to that of Zhang and colleagues was followed.15
Accordingly, the sodium salt of benzyl alcohol was reacted
in THF with 1,4-difluoroanthraquinone. After work-up, the
monosubstituted product was separated from the disubsti-
tuted one and from unreacted substrate by flash chromatog-
raphy. A second nucleophilic substitution with the sodium
salt of 10,11-dihydroquinidine in DMF afforded 3 in 59% yield
after chromatographic purification.
(
vide infra), at the outset of the present study the conforma-
tional characterization of 1 was repeated in CDCl . Even
3
though extensive signal superimposition was found under
these conditions, a selective H1–H10 dipolar interaction
could be detected, which is diagnostic for the Open(3)
conformation.
Regarding the monomeric derivatives, a strong preference
for the Open(3) conformation was ascertained also in the case
of 2 on the basis of the analysis of ROEs. In particular, some
dipolar interactions which support the Open(3) conformer
were detected (Fig. 2a–c), like the ROE between H11
quinuclidine proton and H1 quinoline moiety (Fig. 2a). The
orientation of the phenantryl moiety with respect to the alka-
loid fragments was defined on the basis of inter-ROEs Ha-H8
Racemic hemiesters 4–8 (Scheme 2) were obtained by
alcoholysis of the corresponding anhydride with an excess
of MeOH, CD OD, or CF CH OH; in the case of 5 and 6,
3
3
2
quinuclidine (5 mol%) was added as an achiral catalyst.
Enantiomerically enriched (1R,2S)-4 (95% ee) and (1S,2R)-4
(
91% ee) were obtained by asymmetric methanolysis of cis-
,2,3,6-tetrahydrophthalic anhydride in Et O, in the presence
2
12,14
(
Fig. 2b) and Hi-H11,H14 (Fig. 2e), which suggest that the
1
former lays almost perpendicular to the quinoline ring with
its Ha proton bent at H8 nucleus and its Hi proton in proxim-
ity of the quinuclidine moiety.
of 1 (10 mol%) and 2 (20 mol%), respectively.
A similar conformational analysis could not be performed
for the anthraquinone monomeric derivative 3, due to the
extensive line-broadening and signals superimposition that
made impossible the univocal interpretation of ROEs
interactions.
Conformational Analysis of Chiral Auxiliaries
Following the well-known Burgi’s protocol,16 stereochemical
preference of cinchona alkaloids is commonly described by
means of mostly populated conformers identified as Closed
(
1), Closed(2), Open(3), and Open(4) (Fig. 1). Closed confor-
mations have almost anti H8-C9 and C8-H9 bonds, leading
quinuclidine nitrogen in proximity of the quinoline ring. On
the contrary, the same bonds are cisoid (dihedral angle
H8-C9-C8-H9 about of 78ꢀ) in Open-like conformations, mak-
ing the basic quinuclidine nitrogen better predisposed for
the interaction with suitable hydrogen bond donor groups. In-
side each family of conformations (Closed or Open), rotation
about the C9-C16 bond is possible, leading the C9-H8 bond
bent at the H1 quinoline proton in Closed(1) or Open(4) or
Enantiodiscrimination Experiments
NMR spectra of pure hemiesters 4–8 and their mixtures
with the chiral auxiliaries 1–3 were compared in enan-
tiodiscrimination experiments. The efficiency of each chiral
auxiliary was evaluated in terms of the chemical shift non-
equivalence ( δ), i.e., the magnitude of the separation of
corresponding signals of the two enantiomeric substrates in
their mixtures with the CSA. In this regard it is noteworthy
that within each chiral substrate a suitable probe nucleus
was present, which originates very simple NMR signals: the
methoxy singlet for 4, 7, and 8 or the deuterium or fluorine
nucleus in 5 and 6, respectively.
at its H5 proton Closed(2) or Open(3). The values of the
3
vicinal coupling constant J
and through space dipolar
H8-H9
interactions allow the definition of the conformational pref-
erence, which is strongly affected by the nature of any
derivatizing group at the hydroxyl function or quinuclidine
nitrogen and is often remarkably dependent on the solvent.
Such an analysis has been described in the case of
Preliminary optimization of the analytical conditions was
performed for racemic hemiester 4, starting with solvent
selection. Comparison of the results in toluene-d with those
8
12
derivative 1. In toluene-d , the solvent employed in meso-
in CDCl revealed significant advantages in the use of the
8
3
anhydride desymmetrizzation experiments, the Open(3)
conformation of each alkaloid unit prevails. In spite of the
latter. In the equimolar mixture containing the racemic sub-
strate 4 and the dimeric anthraquinone auxiliary 1 at 10 mM
concentration, a relevant differentiation of 0.055 ppm of
methoxy resonances of rac-4 was observed in toluene-d8,
which nearly doubled when the spectrum was recorded in
molecular C symmetry, which leads to complete isochro-
2
nism of alkaloids units, an anti-disposal of the two DHQD
moieties could be nonetheless ascertained thanks to reciprocal
Fig. 1. Limit conformers according to Burgi’s protocol for quinidine.
Chirality DOI 10.1002/chir

6
96
BARRETTA ET AL.
TABLE 1. Nonequivalences (Δδ = |δent-1 À δent-2|, ppm) mea-
sured for the OMe group of rac-4 in rac-4/CSAs mixtures
[
rac-4]
Entry
CSA
Solvent
rac-4/CSA
(mM)
Δδ (ppm)
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
2
2
2
2
2
2
3
Toluene-d
8
8
1:1
1:1
1:1
1:1
5:1
10:1
1:1
1:1
1:1
1:1
5:1
10:1
1:1
10
10
5
0.7
10
10
10
10
5
0.7
10
10
5
0.055
0.100
0.101
0.092
0.035
0.013
0.029
0.076
0.077
0.043
0.020
0.070
0.020
CDCl
CDCl
CDCl
CDCl
CDCl
3
3
3
3
3
Toluene-d
CDCl
CDCl
CDCl
CDCl
CDCl
CDCl
3
3
3
3
3
3
10
11
12
13
closer to catalysis experiments, since methoxy signals of 4
were clearly differentiated even with a sub-stoichiometric
amount (0.1–0.2 equivalents) of the chiral auxiliary 1 or 2
(
Table 1, entries 5, 6 and 11, 12, respectively).
The change from the dimeric structure of 1 to its mono-
meric counterpart 3 caused a 5-fold decrease of nonequiva-
lence of the methoxy protons of the enantiomers of 4 (0.020
ppm in the presence of 3 to be compared to 0.100 ppm in
the presence of 1, Table 1, entries 13 and 3); nonetheless,
the splitting magnitude was still high and suitable for the
accurate determination of enantiomeric purity of the chiral
analyte.
Further enantiodiscrimination experiments in CDCl with
3
substrates 5–8 (Supporting Table S1) demonstrated the
generality of the approach and confirmed the superior
performance of chiral auxiliaries 1 and 2 with respect to 3.
In detail, the presence of the double bond in the chiral
substrates did not turn out to be mandatory for enantiomer
differentiation, since substrate 7, analogous of 4, but with a
saturated cyclic moiety, underwent remarkably large dou-
bling of the methoxy resonance both in the presence of 1
(
3
1
0.092 ppm) and 2 (0.106 ppm). The anthraquinone monomer
(0.019 ppm) was confirmed less effective in comparison to
and 2.
The more rigid bicyclic hemiester 8 showed the highest
Fig. 2. 2D ROESY traces (600 MHz, CDCl
3
, 25ꢀC) of selected protons of 2
nonequivalences among the substrates examined in this
study. As already found for 7, in this case the optimal CSA
was found to be the phenanthryl ether 2, which caused a
chemical shift nonequivalence of the methoxy groups of 8
(0.155 ppm) remarkably larger than observed in the presence
of the anthraquinone dimer 1 (0.104 ppm). As before, the
anthraquinone monomer 3 proved to be the least effective
CSA in the series (0.014 ppm), in accordance with the general
trend noted above.
The anthraquinone moiety in the dimer structure of 1 and
phenanthryl moiety in the monomer alkaloid 2 caused analo-
gous enantiomer differentiation, but diverse complexation
shifts (Table 2). As a matter of fact, both chiral auxiliaries
strongly shielded the methoxy nuclei of both enantiomers of
4, but the low-frequency shifts were remarkably higher in
the presence of 1 than in the presence of 2. By contrast, 3,
the monomeric analog of 1, not only caused lower differenti-
ation of the enantiomers of 4, but also lower complexation
(
5 mM): (a) H11, (b) H8, (c) H9, (d) H5, and (e) Hi.
CDCl (0.100 ppm, Table 1, entries 1 and 2). An analogous
3
solvent effect was found for the chiral auxiliary 2, with an
increase from 0.029 ppm in toluene-d to 0.076 ppm in CDCl3
Table 1, entries 7 and 8).
The effect of total concentration was negligible in the 10-0.7
mM range for 1 and 10-5 mM range for 2 (Table 1, entries 2–4
and 8–9, respectively). In the case of the latter, the nonequiv-
8
(
alence in CDCl was still remarkable at the 0.7 mM concentra-
3
tion (0.043 ppm, Table 1, entry 10). For practical reasons,
however, most of the subsequent enantiodiscrimination
experiments were performed at a 5 mM concentration.
Concerning the influence of the analyte/CSA ratio, it is
noteworthy that accurate enantiomeric purity determinations
could be performed also in experimental conditions that are
Chirality DOI 10.1002/chir

NMR ENANTIODISCRIMINATION OF CHIRAL HEMIESTERS
697
TABLE 2. Complexation shifts of enantiomers (ent-1 and ent-2) of hemiesters 4, 7 and 8 (Δδ = δmix À δ
f
, ppm) in the presence of
3
equimolar amount of CSAs 1-3 (5 mM, CDCl )
Substrate/1
Substrate/2
Substrate/3
Entry
Substrate
Δδent-1
Δδent-2
Δδent-1
Δδent-2
Δδent-1
Δδent-2
1
2
3
4 (OMe)
7 (OMe)
8 (OMe)
–0.45
–0.45
–0.44
–0.55
–0.54
–0.54
–0.22
–0.16
–0.14
–0.29
–0.27
–0.29
–0.13
–0.14
–0.13
–0.15
–0.16
–0.14
1
Fig. 3. H NMR (600 MHz, CDCl
3
, 25ꢀC) of OMe resonance of 4 (5 mM) in (a) pure rac-4, (b) rac-4/CSA 1:1, (c) (1R,2S)-4/CSA 1:1, and (d) (1S,2R)-4/CSA 1:1.
2
shifts (Table 2, entry 1). The same general trend was
observed in the analysis of the hemiesters 7 and 8 (Table 2,
entries 2 and 3).
In order to correlate the position of the observed reso-
nances with the absolute configuration of the corresponding
stereoisomer of the substrate, mixtures of 1 and 2 with
enantiomerically enriched (1R,2S)-4 or (1S,2R)-4 were exam-
ined next (Fig. 3).
Substrate 5, the deuterated analog of 4, was selected for H
NMR enantiodiscrimination experiments (Fig. 4). The influ-
ence of the structure of the chiral auxiliary on the magnitude
1
of nonequivalence reproduced the trend found in the H
NMR spectra of 4. In particular, the effectiveness of the differ-
ent CSAs in splitting the CD3 2H NMR signals of 5 was found
to vary in the order: 1 > 2 > 3, with 1 producing a 5-fold
larger nonequivalence (0.103 ppm) than its monomeric
counterpart 3 (0.020 ppm) and the phenanthryl ether 2
(0.079 ppm) just between the two anthraquinone derivatives.
Moreover, the correlation between the sense of nonequiva-
lence and absolute configuration of each enantiomer of 5
was the same as that found for 4.
Both 4 and 5 showed analogous nonequivalences of the
olefin proton in the presence of each chiral auxiliary
(Supporting Table S1) which, however, were less useful for
analytical purposes due to their complex spectral patterns
and, in some cases, partial superimposition of the resonances.
Chirality DOI 10.1002/chir
The two chiral auxiliaries 1 and 2 have opposite absolute
configurations at the C8-C9 alkaloid sites and behave as
1
2,14
pseudo-enantiomers in catalysis.
Nonetheless, in the
NMR experiments they gave the same sense of nonequiva-
lence, i.e., relative positions of the enantiomeric substrates.
In particular, (1R,2S)-4 had the more shielded methoxy
resonance in the presence of either 1 or 2. Given the demon-
strated uniform conformational preferences of 1 and 2, this
result could be the consequence of the different nature of
the high-anisotropy groups bound to 9-O.

6
98
BARRETTA ET AL.
Fluorinated hemiester 6 gave us the opportunity of detect-
ing enantiodiscrimination phenomena in the F spectra. Pure
originated a triplet centered at –10.18 ppm due to the
trifluoromethyl group. This signal underwent analogous split-
tings in the presence of 1 and 2 (0.032 ppm in 1/6 and 0.026
ppm in 2/6, Fig. 5), with complexation shifts that were
remarkably larger in the presence of 1 than in the mixture
containing 2. Once again, the monomeric ether 3 was less
effective of the corresponding dimer in differentiating the
two enantiomers of the substrate, with a nonequivalence of
1
9
0.016 ppm only. In view of the spectral simplification attained
2
19
6
in H and F experiments, application of these results for in
situ investigation of catalytic processes may be foreseen.
Finally, the complexation stoichiometry was determined for
the monomeric CSA 2 and compared with that of the dimeric
CSA 1. Job’s plots (Fig. 6) clearly demonstrated that a 1:1
supramolecular assembling was given by each stereoisomer
of the substrate 4 and the monomeric CSA 2, whereas a 1:2
2
Fig. 4. H NMR (92 MHz, CDCl
3
, 25ꢀC) of OCD
3
resonance of 5 (5 mM) in (a) pure rac-5, (b) rac-5/CSA 1:1, and (c) (1R,2S)-5/CSA 1:1.
1
9
Fig. 5. F NMR (564 MHz, CDCl
Chirality DOI 10.1002/chir
3 3
, 25ꢀC) of CF resonance of 6 (5 mM) in (a) pure rac-6, (b) rac-6/CSA 1:1, and (c) (1R,2S)-6/CSA 1:1.

NMR ENANTIODISCRIMINATION OF CHIRAL HEMIESTERS
699
Fig. 6. Determination of stoichiometry for (a) (1S,2R)-4/1, (b) (1R,2S)-4/1, (c) (1S,2R)-4/2, and (d) (1R,2S)-4/2 complexes.
2
3
. Marcelli T, Hiemstra H. Cinchona alkaloids in asymmetric organo-
catalysis. Synthesis 2010; 2010: 1229–1279.
CSA/substrate complexation stoichiometry was involved in
the interaction with the dimeric chiral auxiliary 1.
. Duan J, Li P. Asymmetric organocatalysis mediated by primary amines
derived from cinchona alkaloids: recent advances. Catal Sci Technol
2014; 4: 311–320.
. Zheng S, Schienebeck CM, Zhang W, Wang HY, Tang W. Cinchona alka-
loids as organocatalysts in enantioselective halofunctionalization of
alkenes and alkynes. Asian J Org Chem 2014; 3: 366–376.
. Babu SA, Anand RV, Ramasastry SSV. Cinchona alkaloid-catalyzed
stereoselective carbon-carbon bond forming reactions. Recent Pat Catal
2013; 2: 47–67.
CONCLUSION
4
Chiral auxiliaries 1 and 2 are very efficient in the differen-
tiation of corresponding nuclei of enantiomeric hemiesters
and, hence, can be reliably employed in the determination
of enantiomeric purity for this class of substrates. Very high
nonequivalences were obtained, which encourage their use
for ex situ and, potentially, in situ chiral analysis of enantio-
meric products. Application of the latter idea for monitoring
of the course of desymmetrization reactions of prochiral anhy-
drides is currently under examination.
5
6. Boratyński P. Dimeric Cinchona alkaloids. Mol Divers 2015; 19: 385–422.
7. Kacprzak K, Gawronski J. Resolution of racemates and enantioselective
analytics by cinchona alkaloids and their derivatives. In: Song CE editor,
Cinchona alkaloids in synthesis and catalysis. Wiley-VCH: Weinheim,
Germany; 2009. p 421–469.
The dimer vs. monomer arrangement of alkaloids units
seem fundamental in the enantiodiscrimination phenomena
in the case of 1 vs. 3, whereas, in selected cases, phenanthryl
monomer 2 was even more effective than anthraquinone di-
mer 1. It is noteworthy that, at least for CSAs 1 and 2, the
prevalence of open conformations was demonstrated, where
the quinuclidine nitrogen is more available for the interaction
with hydrogen bond donor groups of the enantiomeric
substrates, accounting for the relevant enantiodiscriminating
efficiency of both chiral auxiliaries. Noticeably, different com-
plexation shifts produced similar nonequivalence in the mix-
8
. Lammerhofer M, Lindner W. Liquid chromatographic enantiomer separa-
tion and chiral recognition by cinchona alkaloid-derived enantioselective
separation materials. Adv Chromatogr (Boca Raton, FL) 2008; 46: 1–107.
9. McCalley DV. Analysis of the cinchona alkaloids by high-performance
liquid chromatography and other separation techniques. J Chromatogr
A 2002; 967: 1–19.
1
0. Uccello-Barretta G, Balzano F. Chiral NMR solvating additives for differ-
entiation of enantiomers. Top Curr Chem 2013; 341: 69–131.
1
1. Uccello Barretta G, Vanni L, Balzano F. Nuclear magnetic resonance
approaches to the rationalization of chromatographic enantiorecognition
processes. J Chromatogr A 2010; 1217: 928–940.
12. Balzano F, Jumde RP, Mandoli A, Masi S, Pini D, Uccello Barretta G.
Mono- and bis-quinidine organicatalysis in the asymmetric methanolysis
of cis-1,2,3,6-tetrahydrophtalic anhydride: a conformational and mechanis-
tic NMR study. Chirality 2011; 23: 784–795.
3. Homer J, Perry MJ. Molecular complexes. Part 18. A nuclear magnetic
resonance adaptation of the continuous variation (Job) method of stoichi-
ometry determination. Chem Soc Faraday Trans 1 1986; 82: 533–543.
4. Li H, Liu X, Wu F, Tang L, Deng L. Elucidation of the active conformation
of cinchona alkaloid catalyst and chemical mechanism of alcoholysis of
meso anhydrides. Proc Natl Acad Sci U S A 2010; 107: 20625–20629.
5. Cheng SK, Zhang SY, Wang PA, Kuang YQ, Sun XL. Homogeneous
catalytic asymmetric dihydroxylation of olefins induced by an efficient
and recoverable polymer-bound ligand QN-AQN-OPEG-OMe. Appl
Organometal Chem 2005; 19: 975–979.
6. Burgi T, Baiker A. Conformational behavior of cinchonidine in different
solvents: a combined NMR and ab initio investigation. J Am Chem Soc
1998; 120: 12920–12926.
1
1
tures containing 1 and 2, which once again demonstrated
the relevance of stereochemical differentiation rather than
thermodynamics differentiation in NMR enantiodiscrimina-
tion processes.
1
1
1
SUPPORTING INFORMATION
Additional supporting information may be found in the
online version of this article at the publisher’s web-site.
LITERATURE CITED
1
1
. Jacques J, Collet A, Wilen SH. Enantiomers, racemates and resolutions.
New York: John Wiley & Sons; 1981.
Chirality DOI 10.1002/chir