Mono- and bis-quinidine organocatalysts in the asymmetric methanolysis of cis-1,2,3,6-tetrahydrophthalic anhydride: A conformational and mechanistic NMR study

Article abstract of DOI:10.1002/chir.20993

The enantioselective organocatalytic methanolysis of cis-1,2,3,6- tetrahydrophthalic anhydride mediated by quinidine derivatives with pyridazine or anthraquinone core was investigated, carrying out a detailed nuclear magnetic resonance study of the confor

Full text of DOI:10.1002/chir.20993

CHIRALITY 23:784–795 (2011)
Mono- and bis-Quinidine Organocatalysts in the Asymmetric
Methanolysis of Cis-1,2,3,6-Tetrahydrophthalic Anhydride:
A Conformational and Mechanistic NMR Study
1
1,2
FEDERICA BALZANO,1 RAVINDRA P. JUMDE,1 ALESSANDRO MANDOLI,1,2 SOFIA MASI, DARIO PINI,
*
AND GLORIA UCCELLO-BARRETTA1,2
*
1Dipartimento di Chimica e Chimica Industriale, Universita` di Pisa, Pisa, Italy
2ICCOM (Section of Pisa), CNR, Pisa, Italy
Contribution to the Carlo Rosini Special Issue
ABSTRACT
The enantioselective organocatalytic methanolysis of cis-1,2,3,6-tetrahydroph-
thalic anhydride mediated by quinidine derivatives with pyridazine or anthraquinone core was
investigated, carrying out a detailed nuclear magnetic resonance study of the conformational
preferences of the alkaloid catalysts in the pure solvent and in the presence of the reaction
substrates and products. No significant interaction between the meso-anhydride and the alkaloid
derivatives was detected. In contrast, evidence for a considerable influence of the alcohol
reactant on the conformational state of some of the chiral organocatalysts could be obtained,
which lends support to the hypothesis of general-base catalysis mechanism, as opposed to the
nuclephilic one. The catalytic properties of the studied derivatives showed no obvious correla-
tion with their conformational prevalence in the resting state, suggesting that the alkaloid 9-O
substituent should have a more active role than merely enforcing the chiral fragments to adopt
a preferential reactive conformation. A strong enantioselective interaction between the
enantiomers of the hemiester product and the alkaloid derivatives was also observed, leading to
the conclusion that in the actual reaction conditions a relatively large fraction of the latter is in
C
VV
the protonated form. Chirality 23:784–795, 2011.
2011 Wiley-Liss, Inc.
KEY WORDS: enantioselective organocatalysis; desymmetrization; Cinchona alkaloids; ring
opening; conformation; nuclear magnetic resonance; chiral recognition; chiral
solvating agent
INTRODUCTION
phenanthrene and anthraquinone 9-O ether derivatives 3 and
4, an effective protocol could be disclosed whose landmarks
are wide substrate scope, mild reaction conditions, relatively
short reaction times and, especially, the possibility of using
just a limited amount (10 mol %) of the expensive alkaloid de-
rivative in the absence of any added achiral base.
Since then several improvements followed, including the
use of different alcohol nucleophiles,19–21 modified alkaloid
organocatalysts (either monofunctional20–22 or bifunc-
tional),23,24 recoverable variants,25–28 and additives.29
The desymmetrization of achiral meso substrates has
emerged in the past years as a powerful tool for the prepara-
tion of nonracemic polyfunctional chiral compounds.1,2
Amongst the many processes based on this concept, the
alcoholysis of meso-anhydride (Scheme 1) appears remark-
ably convenient in terms of experimental simplicity,
substrate scope, and synthetic versatility of the resulting
hemiester products.3–8
In this context, the use of Cinchona alkaloids and their deriv-
atives (Fig. 1)9,10 has played a major role since the pioneering
studies by Oda and Aitken’s groups in the mid 80’s.11–14 None-
theless, it was not until the turn of past century that this meth-
odology gained a new momentum, when Bolm et al. reported
the attainment of synthetically useful ee values by the use of an
excess of methanol and 1.1 equivalents of 1a or 1b, under
carefully optimized conditions.15,16 The same group succeeded
also in carrying out the reaction with just catalytic amounts of
the chiral organocatalyst 1a,16 but the need of one equivalent
of the expensive tertiary amine pimpedine, as an achiral auxil-
iary base to help keeping the catalyst in the not-protonated
form, somewhat limited the usefulness of the protocol.
Nonetheless, most of these efforts appear largely empiri-
cally-driven because the lack of a detailed mechanistic under-
standing of the reaction prevents a rational approach to the
design of improved catalytic systems. In this respect, it
should be noted that two alternate scenarios were proposed
Additional Supporting Information may be found in the online version of this
article.
Contract grant sponsor: MIUR; Contract grant number: PRIN2007 (Sintesi e
Stereocontrollo di Molecole Organiche per lo Sviluppo di Metodologie Inno-
vative di Interesse Applicativo).
Contract grant sponsor: MIUR; Contract grant number: FIRB Project
RBPR05NWWC.
*Correspondence to: Alessandro Mandoli, Dipartimento di Chimica
e
Chimica Industriale, Universita` di Pisa, Via Risorgimento 35, 56126, Pisa,
Italy. E-mail: alexm@dcci.unipi.it (or) Gloria Uccello-Barretta, Dipartimento
di Chimica e Chimica Industriale, Universita` di Pisa, Via Risorgimento 35,
56126, Pisa, Italy. E-mail: gub@dcci.unipi.it
Received for publication 2 March 2011; Accepted 14 June 2011
DOI: 10.1002/chir.20993
Nearly at the same time, Deng and coworkers gave another
fundamental contribution to the field,17 describing the organo-
catalytic use of some of the monomeric and dimeric alkaloid
ligands originally developed for the Sharpless’s osmium-medi-
ated asymmetric dihydroxylation.18 In particular, by using the
Published online 17 August 2011 in Wiley Online Library
(wileyonlinelibrary.com).
C
VV
2011 Wiley-Liss, Inc.

NMR INVESTIGATIONS OF CATALYSIS WITH QUINIDINE DERIVATIVES
785
dictated by the close relationship and not C2-symmetric
structure of 13 and 14, which made these compounds
appealing NMR probes, as well as by their lower performan-
ces with respect to 4 (vide infra), which could be indeed use-
ful for clarifying the basic requirements for an effective
catalyst. The results of this study are reported herein.
Scheme 1. Asymmetric alcoholysis of meso-anhydrides.
RESULTS
in the course of the years, where the alkaloid derivative
would act either as a nucleophilic or a general-base catalyst
for the alcoholysis reaction.3,4 Both hypotheses received
some degree of experimental support, with the latter prefer-
entially credited at present as a satisfactory mechanistic ra-
tionale.4,22,30
As witnessed by the caution exerted in several literature
contributions,3,23,30,31 this is not to say however that a satis-
factory settlement of the issue has been reached yet. In fact,
Synthesis and Characterization of the Alkaloid Derivatives
In the course of previous preparations of 11,34 thin layer
chromatography (TLC) analysis revealed the transient pres-
ence of a less retained compound 12 that, due to its lower
polarity, was likely to be one of the two possible regioiso-
meric mono-alkaloid intermediates en route to the final bis-
alkaloid product (Scheme 2). To isolate such an intermediate
and determine its actual structure, a regular preparation of
11 was stopped before complete conversion and a sample of
the resulting mixture was separated by column chromatogra-
phy. By these means 12 (17%) could be obtained as a TLC
homogeneous fraction, alongside with the expected deriva-
tive 11 (80%). To probe the identity of the newly synthesized
compound, 12 was subjected to electrospray mass spectrom-
even after
a recent mechanistic study by Deng and
coworkers on the catalytic reaction with 3–5,22 a number of
open questions still remained. Beside the fundamental
nature of the catalysis, these include how the accumulating
acidic reaction product could affect the chiral base organoca-
talyst3 and the specific role of the alkaloid 9-O substituent in
determining the overall catalytic properties. This latter aspect
appears particularly intriguing when one considers that the
switch from the effective mono- e bis-alkaloid organocatalysts
3 and 4 to similar systems like 6–8 generally leads to a dra-
matic drop of the reaction enantioselectivity (13–32% ee),17
whilst other apparently less related derivatives, like 9, may
still provide excellent ee values (87–95% ee).20,22
1
etry (ESI-MS) and H NMR analyses. Both techniques con-
firmed the introduction of a single alkaloid unit on the pyrid-
azine core and, in the case of the latter one, the absence of
any regioisomeric product. A preliminary rotating-frame over-
hauser enhancement spectroscopy (ROESY) experiment
revealed also the close spatial proximity of the quinoline H1
and pyridazine Ha protons, as well as the lack of any strong
dipolar interaction between the acetylenic side chain and the
alkaloid moiety (for the numbering scheme, compare Fig. 2).
These results are therefore consistent with the depicted
structure for 12 that, unsurprisingly, may form through
nucleophilic displacement of the least hindered chlorine
atom of the starting compound 10.
Even though the purified intermediates 11 and 12 could
be separately converted into the final derivatives,34 for the pur-
poses of this work the direct reaction of their mixture proved
equally effective. Indeed, after stirring the latter with an
excess benzylazide, copper(I) chloride, and the MonoPhos
ligand in tetrahydrofurane (THF),47 a single chromatographic
purification of the resulting crude mixture provided the new
In consideration of our longstanding and continuing inter-
est in the Cinchona alkaloids for supported asymmetric catal-
ysis32–34 and determination of enantiomer composition by
chromatographic35–40 and nuclear magnetic resonance
(NMR) methods,41–46 we were prompted to investigate fur-
ther the molecular basis of the said catalytic process by
exploiting the potential of NMR spectroscopy in the study of
molecular recognition processes. In addition to the commer-
cial anthraquinone ether 4, two alkaloid pyridazine deriva-
tives 13 and 14 were selected for this purpose, which were
recently introduced by our group within the framework of a
project for the development of polymer supported organoca-
talysts.34 For the aims of this work, such a choice was
Fig. 1. Cinchona alkaloid derivatives for the asymmetric alcoholysis of meso-anhydrides.
Chirality DOI 10.1002/chir

786
BALZANO ET AL.
Fig. 2. Numbering schemes of the three alkaloid derivatives 4, 13, and 14.
mono-alkaloid 14 and the known bis-alkaloid 13,34 in accepta-
ble yields (ꢀ75 and 54%, respectively, based on the composi-
tion of the mixture 11 1 12). The characterization of 14 was
carried out as described for 12 and confirmed the structure
of the derivative, including the substitution pattern of the py-
ridazine ring.
afforded results which were not appreciably different from
those provided by 13 at the same–alkaloid units- loading
(compare Table 1, entry 5). Finally, in the absence of any
organocatalyst rac-16 was formed in just minute amounts af-
ter 48 h at 2208C (Table 1, entry 9).
NMR Conformational Study of the Alkaloid Organocatalysts
Catalysis Experiments
The conformational analysis of the Cinchona alkaloids has
been the subject of extensive theoretical and spectroscopic
investigations during the last twenty years.48,49 These efforts
culminated in the protocol by Burgi and Baker that identified
four more populated conformers for the native alkaloids, cus-
tomarily named Closed(1), Closed(2), Open(3), and Open(4)
(Fig. 3).50 The Closed and Open families differ each other
for the value of the dihedral angle H8-C9-C8-H9, which is
ꢀ1758 for the former (corresponding to the quinuclidine
nitrogen directed toward the quinoline ring) and ꢀ788 for
the latter (with an antirelationship between the quinuclidine
nitrogen and the quinoline ring). Moreover, within each of
the previous families the rotation around the C9-C16 bond
provides the two alternate forms [e.g., Closed(1) vs.
Closed(2)], which hence differ in relative orientation of the
quinoline moiety. All of these conformers could be easily
distinguished on the basis of the angular dependence of the
For testing the performances of 13 and 14, the methano-
lytic desymmetrization of cis-1,2,3,6-tetrahydrophtalic anhy-
dride was chosen as the benchmark (Scheme 3). The condi-
tions were those of Deng and coworkers,17,22 with the use of
10 mol % of the bis-alkaloid 13 or 20 mol % of the mono-alkaloid
organocatalyst 14 and a fixed reaction time of 48 h. For
comparison purposes runs with the commercial anthraquinone
bis-hydroquinidyl ether 4 (10 mol %) were also carried out,
providing the additional results summarized in the Table 1.
An initial solvent screening with 13 revealed a strong
influence of the reaction medium on the methanolysis pro-
gress, with some notable differences in comparison with the
literature derivative 4. In fact, whilst diethyl ether and THF
proved substantially superior than toluene in the reactions
involving 4 (Table 1, entries 2,17 4, and 6), in the case of the
derivative 13 only the latter solvent could afford a reasona-
ble compromise between activity and enantioselectivity
(Table 1, entries 1, 3, and 5). For this reason additional
experiments were carried out in toluene, with the aim of
exploring the effect of catalyst loading, the influence of the
structure of the pyridazine derivative, and the rate of the
background reaction.
3
vicinal coupling constant J89 and the dipolar interactions
originated by the protons H8/H9 in rotating-frame Overhauser
effect (ROE) or nuclear Overhauser effect (NOE) experiments.
In the case of the derivatives 4, 13, and 14 the situation
is further complicated by the presence of two additional
degrees of freedom for each alkaloid unit, corresponding to
rotations around the two bonds connecting the 9-O oxygen
atom to the adjacent molecular fragments. Therefore, in
addition to the Burgi–Baker analysis described above, the
definition of the conformational features of the three chiral
organocatalysts of this work required also to evaluate the
In the first of these runs, the use of 50 mol % of 13 led to
an essentially complete conversion of 15 in the standard
time and a noticeable increase of the enantiomeric purity of
the product 16 (Table 1, entry 7). In contrast, the use of the
mono-alkaloid derivative 14 (20 mol %, Table 1, entry 8)
Scheme 2. Preparation of the compounds 11–14.
Chirality DOI 10.1002/chir

NMR INVESTIGATIONS OF CATALYSIS WITH QUINIDINE DERIVATIVES
787
Scheme 3. Asymmetric methanolysis of 15.
orientation of each chiral unit with respect to the 9-O aro-
matic substituent. With this aim 4, 13, and 14 were studied
by NMR, starting with the unperturbed alkaloid derivatives
in pure toluene-d8.
Preliminary variable-concentration experiments with 13 in
the 5–80 mM range revealed very small variations in the
chemical shift values and an essentially constant diffusion
coefficient D of the compound, as determined by diffusion or-
dered spectroscopy (DOSY) experiments.51
1
The H NMR spectrum of 13 (Fig. S1b, Supporting Infor-
mation) showed distinct resonances of the two protons H9
0
and H9 centered at 3.20 ppm/3.28 ppm, respectively.
In ROESY maps, both of them gave dipolar interactions
0
0
with the respective quinoline protons (H5/H1 or H5 /H1 ),
0
with a slightly larger effect at the frequencies of H1 and H1
0
(Fig. 4). Therefore, the protons H9 and H9 in 13 appear to
Fig. 3. The four limit conformers for quinidine (1a). [Color figure can be
viewed in the online issue, which is available at wileyonlinelibrary.com.]
lay out of the quinoline plane with H9 somewhat closer to H1
0
0
0
than to H5 and H9 closer to H1 than to H5 . The ROEs H9-H8
0
0
and H9 -H8 were significantly lower than those caused by
3
0
H9/H9 on the quinoline protons discussed above, indicative
0
(5.1 Hz) and J8’9 (7.0 Hz) could be measured and corre-
lated52 to dihedral angles values of 130 and 1408, for H8-C9-
C8-H9 and H8’-C9’-C8’-H9’ respectively. Hence, by taking also
into account the ROEs constraints discussed above, all these
results appear consistent with the spatial orientation depicted
in the Figure 5 where, for each alkaloid unit, the proton H8
(or H8’) and the pyridazine ring lay on the opposite side than
the quinuclidine moiety with respect to the quinoline plane
in a Closed(2) prevailing conformation. It is worth noting
that the tendency to attain the closed-like state appears more
pronounced for the alkaloid fragment adjacent to the achiral
pyridazine side chain.
The relative disposition of the two quinidine units of 13
was analyzed on the basis of the four limit structures of Fig-
ure 5: two ANTI-type conformers, where the quinoline rings
of the two chiral units lay in the opposite hemispaces defined
by the pyridazine plane, and two SYN-type conformers,
where the quinoline rings occupy the same hemispace and,
therefore, face each other.
of the prevalence of Closed conformers. Some differences
were also detected in the conformation of the two quinidine
0
0
units around the respective C8-C9 or C8 -C9 bonds, as the
0
0
ROE H9-H8 was more intense in comparison with the H9 -H8
0
0
one (compared to the respective ROEs H9-H1 or H9 -H1 ).
The protons H5/H5’ gave a very intense ROE at the frequen-
cies of the respective protons H8/H8’ and no significant dipo-
lar interaction with the other quinuclidine protons, which
strongly supports the prevalence of the Closed(2) conformer.
0
The resonances of the protons H8 and H8 , which fell in a
crowed spectral region, were extracted by 1D total correla-
tion spectroscopy (TOCSY) selective perturbations at the fre-
0
quencies of H9 and H9 , respectively (Fig. S2, Supporting In-
formation). In this way the two vicinal coupling constants 3J89
TABLE 1. Asymmetric methanolysis of 15
Entry Cat.(mol%) Solvent Conversion(%)a Yield(%)b Ee(%)c,d
1
2
3
4
5
6
7
8
9
13 (10)e Et2O
54
–
45
95
18
73
69
85
55
98
67
93
50
83
66
47
–
4 (7)f
13 (10)
4 (10)
13 (10)
4 (10)
13 (50)
14 (20)
–
Et2O
THF
THF
Toluene
Toluene
Toluene
Toluene
Toluene
25
78
70
88
>98
68
6
ꢀ quant.
66
–
aDetermined on the crude product, by 1H NMR.
bIsolated yield after flash chromatography.
cBy high performance liquid chromatography (HPLC) (Chiralcel OJ, 1 ml min21
n-hexane:IPA 5 95:510.1% trifluoroacetic acid, 210 nm).
dThe major enantiomer of 16 had (1)2(1R;2S)configuration.17
eThe catalyst was largely undissolved.
Fig. 4. 2D ROESY (600 MHz, toluene-d8, 258C, mix 600 ms) traces corre-
fFrom ref. 17
0
sponding to H9 (a) and H9 (b) protons of 13.
Chirality DOI 10.1002/chir

788
BALZANO ET AL.
Fig. 5. Graphical representation of the four limit ANTI and SYN conformers of 13. [Color figure can be viewed in the online issue, which is available at wiley
onlinelibrary.com.]
The intense ROE effects caused by H1 and the vinyl proton
H20 at the frequencies of the pyridazine proton Ha and the si-
multaneous observation that the some selected protons of
In spite of the symmetry of 4, which limited its conforma-
tional analysis to some extent, several important deductions
could be made by the careful examination of NMR data.
Amongst these the most relevant conclusion is the strong
prevalence of Open-type conformations for this derivative,
even (vide infra) in a solution of pure toluene. Indeed, even
0
the achiral side chain gave ROEs on H1 of the nearby quino-
line ring and not onto the methine H8’ proton, ruled out a
large contribution by SYN-1 and identified the ANTI-1 as the
most populated conformer. The presence of SYN-2 as a
minor conformer was detected based on the weak ROE
between H5 and Ha. Because the SYN-2 may be generated
from ANTI-1 by rotation of the alkaloid fragment adjacent to
Ha with respect to the rest of the molecule, these findings
suggested that the chiral unit linked to the least hindered
side of the pyridazine ring should experience some degree
of conformational freedom.
0
though the resonance of H8 was so large (Dm 5 41 Hz, Fig.
½
S1a Supporting Information) that 3J89 could not be measured,
the observation of ROE effects between H9-H5 and H9-H8
remarkably stronger than for H9-H1 proved particularly in-
formative, as they could place the H8 and H9 protons in close
proximity of H5 and hence identify an Open(3) conformation.
In contrast, the ROEs originated by the protons Ha, H8, H5
and the CH3 of the ethyl group on the quinuclidine moiety
were quite diagnostic regarding the orientation of the anthra-
quinone moiety. Amongst them, the very large ROE H8-Ha
could well originate from the ANTI structure of Figure 6,
with the H8 protons of both quinidine moieties bent at the
corresponding Ha position. Furthermore dipolar interactions
between the proton Hb of the antraquinone moiety and the
ethyl group on the quinuclidine cage were detected, which
strengthen the above conclusions.
On the contrary, no analogous dipolar interaction could be
0
observed between the side chain protons and H5 of the
nearby quinoline ring, indicative of an essentially locked ori-
entation for the corresponding alkaloid residue. As it may be
deduced from the Figure 5, the prevalence of ANTI-1 and
SYN-2 is probably a consequence of the fact that, in these
conformers, the pyridazine side chain can be accommodated
0
in the spatial region facing H1 and, hence, devoid of major
steric repulsions with the closest quinidine fragment.
Organocatalyst-Reactants Interaction
The conformation of 14, containing the same pyridazine
group as 13 but only one quinidine unit, was similarly
defined through the analysis of spatial constraints imposed
by the dipolar interactions detected in the ROE spectra and
Given the strong influence of the solvent polarity on the
conformational prevalence of Cinchona alkaloids48,49 and to
gain details about the reaction mechanism, toluene-d8 solu-
tions containing 4, 13, or 14 and the anhydride 15 or
MeOH were investigated next.
3
by the vicinal coupling constant J89 of 6.2 Hz. Hence, the
large prevalence of the Closed conformer on the Open one
was revealed also in this case, to a degree that was very
similar to that found in the dimeric compound 13. In the
prevailing conformer of 14 the proton Ha of the pyridazine
moiety pointed at the quinoline proton H1 and the vinyl
proton H20, but a minor population of conformers with Ha
pointing at H5 was also detected, like in ANTI-1 and SYN-2 of
13, respectively.
At variance with the similar pyridazine derivative 13, dis-
cussed above, initial variable-concentration experiments with
the commercial bis-alkaloid anthraquinone ether 4 resulted
1
in dramatic changes in both the H NMR spectrum and the
diffusion coefficient D deduced from DOSY measurements.
Both results, clearly indicative of a strong propensity of 4 to
self-aggregate, appear somewhat surprising if one considers
that the alkaloid derivative 4 is devoid of hydrogen-bond
donor groups.53 However, at the concentration used in the
catalysis experiments (5 mM) the diffusion coefficients of 13
(D 5 4.9 3 10210 m2 s21) and 4 (D 5 5.5 3 10210 m2 s21
were similar, to indicate that the self-aggregation of 4 could
)
Fig. 6. Graphical representation of the ANTI-1 conformer of 4. [Color fig-
ure can be viewed in the online issue, which is available at wileyonlinelibrary.
com.]
be neglected.
Chirality DOI 10.1002/chir

NMR INVESTIGATIONS OF CATALYSIS WITH QUINIDINE DERIVATIVES
789
0
Fig. 7. 2D ROESY (600 MHz, toluene-d8, 258C, mix 600 ms) traces corresponding to H9 (a) and H9 (b) protons of 13 in the presence of CD3OD (500 mM).
In general, no significant interaction between the alkaloid
derivatives (5 mM) and the anhydride substrate (5–50 mM)
could be evidenced by these experiments, as deduced
from the lack of relevant intermolecular dipolar interaction in
the ROESY spectra and the observation of a diffusion coeffi-
cient for 15 in the mixtures that proved identical to that
of the pure compound in toluene-d8 alone (D15 5 15.0 3
10210 m2 s21). Furthermore, any significant complexation-
induced shift was not detected in the mixtures with respect
to the pure anhydride.
induced Closed to Open swap discussed above, probably
because of the repulsive steric interaction that would ensue
between the quinuclidine cage and the achiral side chain on
the pyridazine ring. Similarly, the latter result suggests that
the ANTI-1 to SYN-2 population ratio was essentially unaf-
fected by the presence of the alcohol. Interestingly, all these
conclusions were also confirmed for a sample of 13 in neat
MeOH-d4.
The mono-alkaloid derivative 14 showed features similar to
the least hindered half of the compound 13. In particular, the
addition of a large excess of MeOH-d4 (500 mM) caused the
increase of the population of the Open(3) conformer at
the expense of the Closed(2) one, as detected for example by
the increase of the H9-H8 and H9-H5 ROEs relative to the H9-H1
one. Moreover, much as for 13 the orientation of the pyrida-
zine ring with respect to the quinidine structure was not
affected significantly by the change in solvent composition.
Finally, in the presence of an excess of the alcohol
(500 mM) the anthraquinone derivative 4 showed the antici-
pated shift of several of the alkaloid’s resonances (Table 2)
but no relevant sign of conformational modifications. In this
respect, it is worth noting that even the ROESY spectra of 4
in neat MeOH-d4 retained the main features discussed above
Similarly, the addition of small amounts of MeOH (5 mM)
to the chiral derivatives (5 mM) caused no obvious differen-
ces with respect to the separate components, either in terms
of the diffusion coefficient of the alcohol (DMeOH 5 31.0 3
10210 m2 s21) or for what it concerns the conformers popula-
tion of the alkaloid organocatalysts. This latter aspect was
studied in detail for 13, where the quantitative analysis of
the relevant intramolecular ROE effects provided the same
large preference for the Closed(2) form of each alkaloid
residues and an identical relative weight of the ANTI-1 and
SYN-2 conformers as observed before in the pure solvent.
However, when the alcohol concentration was raised to
the level of the actual catalysis experiments (500 mM, in this
case added as its d4 deuterated form) some significant
changes in the spectral features of 4, 13, and 14 could be
seen. Several protons resonances were shifted in the NMR
spectra and, in the case of 13, the presence of a large excess
of MeOH-d4 led also to the enhancement of the dipolar inter-
action between the protons H9 and H8, as demonstrated by
the much stronger ROE between them (Fig. 7) in compari-
son with the sample in pure toluene-d8 (Fig. 4). In view of
the previous discussion this fact was clearly indicative of a
reduction of the H8-C9-C8-H9 dihedral angle, as confirmed
also by the analysis of additional ROEs. Hence these results
unequivocally demonstrated that, for the alkaloid residue on
the least hindered side of 13, the population of the Open(3)
conformer increases remarkably in the presence of MeOH-d4
at the expenses of the Closed(2) one.
TABLE 2. Chemical shift (600 MHz, 258C)
variations (Dd 5 d1 – d2, ppm) of selected
resonances of 4 (5 mM, toluene d8) in the
presence (500 mM, d1) and in the absence
(d2) of methanol
Proton
Dd
1
2
3
4
0.16
20.14
20.03
0.05
5
20.03
0.16
0.36
OMe
8
9
20.13
20.09
20.04
0.15
0
0
In contrast, no significant enhancement of the H9 -H8 dipo-
lar interaction or change in the relative intensities of the Ha-
H1 and Ha-H8 ones could be detected under these conditions.
The former results clearly points to a much lower propensity
of the second alkaloid unit of 13 to undergo the alcohol-
15
16
18
19
20.12
Chirality DOI 10.1002/chir

790
BALZANO ET AL.
In view of its possible relevance for the understanding of
the reaction mechanism, the interaction between the alkaloid
organocatalysts and the reaction products was finally investi-
gated. In this respect, it is interesting to note that in the
course of the kinetic measurements all of the recorded spec-
tra showed a noticeable splitting of most of the product’s
resonances. This fact was clearly indicative of a chiral recog-
nition phenomenon, probably related to the ability of the Cin-
chona alkaloid derivatives to act as NMR chiral solvating
agents toward suitable chiral substances.41–46,54–56 Control
experiments with rac-16 (50 mM) and either 4 or 13 (5
mM) in toluene-d8, confirmed this hypothesis, showing that
even such a small concentration of the alkaloid derivatives
was sufficient for inducing a rather large anisochrony in
most of the protons of the enantiomers of 16 (e.g., Dd 5
0.011 ppm for the methoxy groups of rac-16 in the presence
of 13, Fig. 8a). This effect was largely preserved even in the
presence of an excess of MeOH (250 mM) (Fig. 8b) thereby
suggesting that a quite strong interaction between the acidic
products and the alkaloid catalyst could arise under the
actual methanolysis conditions.
Fig. 8. 1H NMR (600 MHz, toluene-d8, 258C) spectra of 13 (5 mM)/rac-
To strengthen this conclusion DOSY and ROESY measure-
ments were carried out during the reaction progress and on
the final mixtures. The DOSY experiments confirmed a con-
siderable degree of interaction between the enantiomers of
16 and the chiral organocatalysts, as demonstrated by the
fairly large reduction of the diffusion coefficient of the former
(D16 5 6.8 3 10210 m2 s21 at 22% conversion) with respect
to the free racemic product in neat toluene-d8 (Df16 5 9.0 3
10210 m2 s21). Interestingly, under these conditions the
interaction proved somewhat enantioselective, with the major
(1R; 2S)-16 enantiomer retarded slightly less than the (1S;
2R)-16 minor one (DD 5 D(1R; 2S)-16 2 D(1S; 2R)-16 5 0.4 3
10210 m2 s21).
Further information came from the ROESY maps, where
several cross-peaks connecting the protons of the enantiom-
ers of 16 and those of the chiral organocatalysts could be
clearly seen. As an example, the methoxy protons of 13 pro-
duced relevant dipolar interactions with the olefinic and
methine/methylene protons of 16 (Fig. 9).
16 (50 mM) (a) and at the end of the methanolysis of 15 (50 mM) with
MeOH (250 mM) (b).
for the sample in toluene-d8, thus witnessing the substantial
preservation of the preferential bis-alkaloid Open(3)-ANTI
arrangement in spite of the large differences in the polarity
of the medium.
Reaction Monitoring and Organocatalyst-Product
Interaction
To explore in more detail the reaction under exam, the pro-
gress of the methanolysis of 15 in the presence of the orga-
nocatalysts 4 or 13 was followed by NMR. With this aim the
runs were carried out in toluene-d8 at 258C with 10 mol % of
the alkaloid derivative and 5 equivalents of MeOH, monitor-
ing the disappearance of the olefinic signals of the substrate
15 and the growth of the corresponding resonances of the
product 16. These experiments revealed that under these
conditions the methanolysis proceeded much faster than in
the regular runs at 2208C (Table 1) and led to the complete
conversion of 15 in less than 20 h. Nonetheless, a clear dif-
ference of catalytic performances between the two alkaloid
derivatives was maintained here too, both in terms of activity
and enantioselectivity. The former conclusion is proved by
the half-lives of the substrate 15 in the reaction mixtures con-
taining 4 or 13, which resulted t1/2 ꢀ 20 and ꢀ 80 min,
respectively. Similarly, the greater asymmetric induction of 4
vs. 13 (Table 1, entries 5 and 6) was also preserved under
these conditions, eventually leading to the product (1)-(1R;
2S)-16 with 80 and 42% ee, respectively.
For what it concerns the former component, it is worth
noting that the nuclei responsible of the observed intermo-
lecular dipolar interactions were mainly situated in proximity
of the edge between the quinoline and quinuclidine moiety.
Therefore, a rather specific docking between the acidic prod-
ucts and the alkaloid organocatalysts appears to take place in
Fig. 9. 2D ROESY (600 MHz, toluene-d8, 258C, mix 600 ms) trace corre-
sponding to OMe/OMe’ protons of 13 (5 mM) at the end of the methanoly-
sis of 15 (50 mM) with CD3OD (500 mM). *Resonances of 16.
Fig. 10. 2D ROESY (600 MHz, toluene-d8, 258C, mix 600 ms) traces cor-
0
responding to H9 (a) and H9 (b) protons of 13 (5 mM) at the end of the
methanolysis of 15 (50 mM) with CD3OD (500 mM).
Chirality DOI 10.1002/chir

NMR INVESTIGATIONS OF CATALYSIS WITH QUINIDINE DERIVATIVES
791
Scheme 4. Nucleophilic and general-base catalysis mechanisms.
the systems under exam, possibly as a consequence of the
formation of a tight ion-pair resulting from the proton trans-
fer between 16 and the most basic sites of 4 and 13. Inter-
estingly, this latter hypothesis was indirectly supported by
the analysis of intramolecular ROE effects of the bis-alkaloid
pyridazine derivative 13 in the final reaction mixture (Fig.
10). At variance with the unperturbed derivative, where the
most hindered alkaloid unit remains in the Closed(2)
arrangement even in neat MeOH-d4 (vide supra), in the pres-
ence of an excess of 16 both chiral fragments appeared to
attain a fully Open(3) conformation as clearly witnessed by
presence of 3 or 4 (kMeOH/kMeOD 5 1.4 and 1.8, respec-
tively).22 Moreover, to the best of our knowledge, no specific
effort has been reported in the literature for probing the
interaction between the alkaloid organocatalyst and the
nucleophilic alcohol, implied by the general-base mechanistic
scheme.
From this point of view, the DOSY- and ROE-based experi-
ments discussed in the previous section also failed to detect
the presence of defined molecular aggregates involving ei-
ther the anhydride 15 or MeOH and the chiral organocata-
lysts 4, 13, or 14. Therefore, it may be estimated that if any
complex is formed between the reaction partners and alka-
loid derivatives (e.g., of the kind depicted in the Deng’s
study)22 the corresponding equilibrium constant should be
0
0
0
the very large prevalence of the ROEs H9(9 )-H5(5 ) or H9(9 )
-
0
0
0
H8(8 ) on the ROE H9(9 )-H1(1 ). This result has been previ-
ously related to the involvement of the quinuclidine nitrogen
atom as complexation site.55
lower than about 1–10 mM .
21 57 Nonetheless, the observation
of large shifts of the resonances of 4, 13, and 14 only in the
presence of an excess of MeOH-d4 (100 equiv.), but not of
the anhydride 15 (10 equiv.), appears indicative of a much
stronger interaction involving the former. Although this
result could be partially ascribed to the larger amount of the
alcohol reagent, it has to be pointed out that the effect is
undoubtedly specific for what it concerns the site of interac-
tion (apparently the quinuclidine sites of 4, 13, and 14) and
significant in terms of its magnitude (see, e.g., Table 2). In
this respect it is worth noting that, while in the case of the
least hindered alkaloid units of the pyridazine derivatives 13
and 14 the observed Closed-to-Open transition could largely
contribute to the chemical shift variations, analogous effects
are displayed also by the locked chiral unit of 13 and by the
anthraquinone derivative 4, which appear not to undergo
any significant conformational change.
As anticipated, this latter aspect constitutes a second rele-
vant proof of the substantial interaction between the alcohol
and the alkaloid derivatives 13 and 14 under the actual reac-
tion conditions. In fact, the stabilization of the Open conform-
ers of Cinchona alkaloids in polar protic solvents has been al-
ready reported48,49 and is confirmed, for the pyridazine deriv-
atives of this work, by the results obtained in neat MeOH-d4.
Nonetheless, a comparison of the different behavior of the
two chiral units of 13 seems worthwhile because it demon-
strates that minute amounts of the alcohol (ꢀ2% v/v) in the
actual methanolysis mixtures may have, or not, a profound
influence on the (resting state of) chiral organocatalysts, as a
function of even subtle differences in the structure of the 9-O
substituent.
DISCUSSION
As anticipated in the Introduction, the two alternate mech-
anisms contrasted in the Scheme 4 can be considered for the
alcoholysis of meso anhydrides.3,4
The nucleophylic catalysis mechanism looks particularly
appealing as the formation of an acylamonium intermediate
would involve the anhydride itself as the primary site of mo-
lecular recognition by the alkaloid.13,14,16 However, direct
clues in favor of this hypothesis appear limited to the mass-
spectrometric detection of an anhydride-1a adduct, by Car-
loni and coworkers,25 which unfortunately could prove nei-
ther the structure of such an intermediate nor its actual
placement on the catalytic cycle pathway.
In contrast, a somewhat larger body of evidence has been
provided in the course of the years for the alternate general-
base catalysis explanation. Besides the comparison with
model systems,3 as well as theoretical calculations (albeit for
a rather different chiral organocatalyst),30 this include the
detection by Oda and coworkers of large kinetic isotopic
effects (kMeOH/kMeOD 5 2.3)11,12 in the methanolysis of cis-
2,4-dimethylglutaric anhydride catalyzed by cinchonine.11,12
Because the magnitude of the latter effect appeares incompa-
tibile with a nucleophilic catalysis mechanism, this result has
long been considered a particularly stringent proof in sup-
port of the general-base one.
However, in the recent study by Deng and coworkers sus-
btantially attenuated rate differences were observed for the
methanolysis of cis-2,3-dimethylsuccinic anhydride in the
Chirality DOI 10.1002/chir

792
BALZANO ET AL.
This observation naturally leads to another intriguing as-
former was responsible for the observed catalysis, even for 4
and 14, the lack of a significant performance differential
between 13 and the latter derivative clearly tends to militate
against this hypothesis.
pect of the reaction under exam, i.e., the role of the latter
group in determining the alkaloid catalytic properties. In the
recent contribution by Deng and coworkers, a detailed study
of the designed rigid derivative 5 led to the conclusion that
an app-closed conformation (Closed(2) in the Burgi–Baker
notation)50 is well suited for catalyzing the asymmetric meth-
anolysis of cis-2,3-dimethylsuccinic anhydride (93% ee).22
Moreover, on the basis of the similar result afforded by 3
(96% ee) the inference was also made that the common
unconstrained analogs, for which all the four major conform-
ers were detected in solution by NMR, adopts the same app-
closed arrangement when mediating the highly enantioselec-
tive methanolysis of meso anhydrides. On this ground, a
model was eventually worked out that, on the assumption of
general-base type of mechanism, could nicely explain the
asymmetric induction sense and other experimental features
of the reaction.22 However, in this rationalization the role of
alkaloid’s 9-O substituent was deemed to be limited to the
enforcement of the app-closed conformation in virtue of its
bulkiness, a conclusion that appears difficult to reconcile
with the findings of the present work.
As described in the previous section, 13 and 14 resulted
less effective than the commercial organocatalyst 4 (Table 1,
entries 5, 6, and 8), although they could surprisingly induce
substantially higher ee values than the related phthalazine
bis-hydroquinidyl ether 6 (which affords 18% ee in the reac-
tion of cis-2,3-dimethylsuccinic anhydride).17 Interestingly,
considerable differences were observed in terms of both
reaction rate and enantioselectivity, to witness once again
the influence of the 9-O substituent on the overall catalytic
properties of the alkaloid derivatives. In principle, the two
effects could be related each other because when the activity
of the chiral organocatalyst is low the uncatalyzed (back-
ground) process tends to afford a higher amount of racemic
product, in relative terms. However, the observation of a very
limited anhydride conversion in the standard run without
any catalyst (Table 1, entry 9) rules out such an explanation
and points therefore to the differences in the intrinsic chiral
discrimination ability of the various derivatives as the more
likely reason of the observed enantioselectivity changes.
In this context, the comparison of the results presented in
the previous section with the conclusion of the Deng’s study
leads to a rather paradoxical situation. In fact, in view of the
NMR results discussed above the alkaloid units of 4 and 14
exist predominantly in the Open(3) conformation in solution
(or easily attain it, in the presence of MeOH), whilst the
derivative 13 is the only one that presents one quinidine
residue in the expectedly optimal Closed(2) arrangement,
already in the resting state. If the adoption of a Closed(2)
conformation were sufficient, per se, for optimal catalytic per-
formances, the prediction could then be made that 13 would
be the best organocatalyst of the series investigated herein.
In contrast, the experimental results unequivocally demon-
strated that 13 was not only considerably less active and
enantioselective of 4, but also substantially equivalent to the
mono-alkaloid derivative 14 that is devoid of the persistent
Closed(2) unit.
Finally, the interaction between the alkaloid organocata-
lysts and the reaction products deserves a brief comment. As
noted by Spivey and Andrews,3 one of the striking features
of the catalytic protocol by Deng and coworkers is the possi-
bility of attaining high conversions and ee values without the
need of adding an achiral base for scavenging the acidic
hemiester product. To explain this remarkable result, two
hypotheses were put forward: (i) the possibility that both pro-
tonated and unprotonated forms of the alkaloid derivative
can mediate the enantioselective alcoholysis reaction or (ii)
that the equilibrium concentration of the latter is sufficient
for the catalysis. Even if the former suggestion appears very
stimulating, also in view of recent studies on the effect of
achiral acidic additives,29 it stands against the long-estab-
lished limited effectiveness of the alkaloid salts.3,13 Concern-
ing the alternate possibility, the same authors made the pro-
voking suggestion that in toluene the unprotonated alkaloid
could be the major species at equilibrium, due to the plausi-
ble higher acidity of R3NH1 with respect to RCO2H in such
an apolar solvent.3 Although in the present work no specific
attempt was made to determine the stoichiometry or the
equilibrium constant of the observed adducts between the
enantiomers of 16 and either 4 or 13, the dramatic changes
in the spectra of the alkaloid derivatives in the presence of
an excess of the product tend to suggest a substantial degree
of protonation of the former. In this regard, it is interesting
to note that preliminary heterogeneous catalysis experi-
ments, with the polystyrene-supported analog of 1334 in tolu-
ene, demonstrated also the propensity of the immobilized
alkaloid derivative to scavenge a large fraction of the product
(34%) at the end of the reaction. The latter was released only
by treatment with an excess of diethylamine in toluene.
From this point of view, the considerably better results
attained on rising the loading of the organocatalyst 13 from
10 to 50 mol % (Table 1, entries 5 and 7) appear totally antici-
pated. Given the evidence discussed above, most notably its
enantioselective nature, this phenomenon is likely to involve
the basic catalytic sites in the alkaloid structures, i.e., the
quinuclidine nitrogen atoms.
MATERIALS AND METHODS
All the reactions involving sensitive compounds were carried out
under dry nitrogen, in flame-dried glassware. Before the use, toluene,
CH2Cl2, and MeOH were freshly distilled under dry nitrogen from the
proper drying agent. 3,6-Dichloro-4-(hex-5-yn-1-yl)pyridazine (10) was
prepared as described.34 The other compounds were commercially avail-
able and used as received. Quinidine (1a, Aldrich) contained about 5%
1
of hydroquinidine (2a) by H NMR. HPLC analyses were carried out on
a Jasco PU-1580 chromatograph, equipped with an UV-1575 detector. Op-
tical rotation was measured as solutions in 1-dm cells at the sodium D
line, using a Jasco DIP360 polarimeter.
NMR measurements were performed on a spectrometer operating at
600 MHz for 1H. The temperature was controlled to 60.18C. The 2D
NMR spectra were obtained by using standard sequences with the mini-
mum spectral width required. Proton 2D gradient correlated spectros-
copy (gCOSY) spectra were recorded with 256 increments of 4 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 collected. The 2D ROESY experiments were performed in the
This latter comparison appears particularly important as it
provides a kind of internal consistency proof for the previous
reasoning. In fact, even if the demonstrated conformational
flexibility of the alkaloid systems does not allow us to
exclude that a minor, yet much more active, Closed(2) con-
Chirality DOI 10.1002/chir

NMR INVESTIGATIONS OF CATALYSIS WITH QUINIDINE DERIVATIVES
793
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 Pan-
dora Software. The selective 1D TOCSY spectra were acquired with 256
scans in 32-K data points with a 1 s relaxation delay and a mixing time of
80 ms. The gradient 1H,13C gradient heteronuclear single quantum cor-
relation (gHSQC) and gradient heteronuclear multiple bond correlation
(gHMBC) spectra were recorded with 256 or 128 time increments of 24
scans. The gradient HMBC experiments were optimized for a long-range
1H-13C coupling constant of 8 Hz and a delay period of 3.5 ms for sup-
pression of one-bond correlation signals. No decoupling was used during
the acquisition. DOSY experiments were carried out by using a stimu-
lated echo sequence with self-compensating gradient schemes, a spec-
tral width of 6400 Hz and 64 K data points. Typically, a value of 50 ms
was used for D, 2.0 ms for d, and g was varied in 20 steps of four transi-
ents each to obtain an ꢀ90–95% decrease in the resonance intensity at
the largest gradient amplitudes. The baselines of all arrayed spectra
were corrected before processing the data. After data acquisition, each
FID was apodized with 1.0-Hz line broadening and Fourier transformed.
The data were processed with the DOSY macro (involving the determi-
nation of the resonance heights of all the signals above a pre-established
threshold and the fitting of the decay curve for each resonance to a
Gaussian function) to obtain pseudo two-dimensional spectra with NMR
chemical shifts along one axis and calculated diffusion coefficients along
the other.
AcOEt : MeOH 5 8 : 2 1 0.5% Et2NH) to give 14 (0.112 g, ꢀ75%) and
13 (0.556 g, ꢀ54%) as colorless solid foams. The chromatographic and
spectral properties of 13 were identical to the previously reported
ones.34 For 14: TLC Rf 5 0.53 (SiO2, AcOEt : MeOH 5 8 : 2 1 0.5%
1
Et2NH). H NMR (300 MHz, CDCl3) d 5 8.65 (1H, d, J 5 4.50 Hz); 7.97
(1H, d, J 5 9.21 Hz) ; 7.52 (1H, d, J 5 2.44 Hz); 7.40–7.29 (5H, m); 7.27–
7.17 (2H, m); 7.13–7.04 (1H, m); 6.89 (1H, s); 6.05 (1H, ddd, Ja 5 17.32,
Jb 5 10.45, Jc 5 7.24 Hz); 5.46 (2H, s); 5.14–5.03 (2H, m); 3.95 (3H, s);
3.51–3.32 (1H, m); 3.11–2.78 (4H, m); 2.78–2.67 (3H, m); 2.63 (1H t, J 5
7.38 Hz,) 2.63 (1H, m); 2.28 (1H, q, Ja5b 5 7.66 Hz, Jc 5 7.48 Hz); 2.03
(1H,dd, Ja 5 12.52 Hz, Jb 5 9.43 Hz); 1.89–1.79 (1H, m), 1.79–1.62 (4H,
m), 1.62–1.52 (3H, m); 1.35–1.15 (1H, m). 13C NMR (75 MHz,CDCl3) d
5 163.80, 157.92, 152.54, 147.84, 147.32, 144.69, 144.48, 143.81, 140.39,
134.86, 131.68, 129.09, 128.69, 127.99, 127.21, 121.91, 120.71, 118.30,
114.94, 101.69, 77.58, 77.15, 76.73, 59.74, 55.77, 54.06, 49.97, 49.40, 39.76,
32.24, 28.92, 27.96, 27.33, 26.42, 25.29, 23.40, 23.40. MS (ESI1) m/z 5
651.3 [M1H1]. [a]D25 5 277.6 (c 5 0.75, CH2Cl2).
(1)-(1R; 2S)-6-(Methoxycarbonyl)cyclohex-4-
enecarboxylic Acid by the Asymmetric Methanolysis
of cis-1,2,3,6-tetrahydrophthalic Anhydride.
General procedure
A 25-ml Schlenk tube was charged under dry nitrogen with the alka-
loid catalyst 4, 13, or 14 (for the exact amounts see Table 1) and 15
(0.0761 g, 0.50 mmol). After cooling to 2208C, the dry solvent (10 ml)
was added and the mixture was kept stirring for 15 min, whereupon the
complete dissolution of the solids was generally observed. Dry MeOH
(202 lL, 5 mmol, and 10 equiv.) was added and the resulting solution
was stirred at the same temperature for 48 h. The reaction was
quenched by adding 2N HCl (4 ml) and, after separation of the layers,
the organic one was washed with a second portion of 2N HCl (4 ml).
The aqueous washings were back-extracted with AcOEt (2 3 5 ml) and
the combined organic phases dried (Na2SO4). The volatiles were
removed with a rotary evaporator, and the reaction conversion was deter-
mined by 1H NMR of the residue. The crude product was then purified
by flash chromatography (SiO2, pet. ether:AcOEt 5 3:1 1 0.5% trifluoro-
acetic acid) to give (1)-(1R; 2S)-16 as a viscous oil that solidified on
standing. The absolute configuration of the product was confirmed by
comparing its optical rotatory power with the literature value.17 After dis-
solution in iso-propyl alcohol (IPA, 0.1 ml), the enantiomeric composition
of the sample was determined by HPLC analysis (210 nm, Daicel Chiral-
cel OJ, 1 ml min21 n-hexane:IPA 5 95:510.1% trifluoroacetic acid):
tR[(1R; 2S)-16] 5 10.7 min; tR[(1S; 2R)-16] 5 16.2 min.
4-(Hex-5-yn-1-yl)-3,6-bis(9-O-quinidinyl)pyridazine (11)
and 6-Chloro-4-(Hex-5-yn-1-yl)-
3-(9-O-quinidinyl)pyridazine (12)
A 500-ml three-necked flask, fitted with a Dean–Stark apparatus, was
charged under nitrogen with 10 (3.09 g, 13.5 mmol), 1a (8.76 g, 27.0
mmol, 2 equiv.), KOH pellets (1.55 g, 27.6 mmol, 2.05 equiv.), and tolu-
ene (225 ml). The mixture was refluxed for 3 h with azeotropic removal
of water, cooled to room temperature, and then treated with water (150
ml). The organic components were extracted with Et2O (3 3 150 ml)
and the organic phases were washed with brine (3 3 75 ml). After dry-
ing (Na2SO4) the volatiles were removed with a rotary evaporator to give
a brownish solid foam (10.1 g) that was directly used in the next step.
For characterization purposes, a sample of the crude product (0.450 g)
was subjected to flash chromatography (SiO2, AcOEt:MeOH 5 7:3 1
0.5% Et2NH) to give 12 (0.053 g, 17%) and 11 (0.386 g, 80%)34 as a color-
less solid foams. The chromatographic and spectral properties of 11
were identical to the previously reported ones.34 For 12: TLC Rf 5 0.75
(SiO2, AcOEt : MeOH 5 7 : 3 1 0.5% Et2NH). 1H NMR (600 MHz,
CDCl3), 250C, d 5 8.66 (1H, d, J 5 4.6 Hz); 7.98 (1H, d, J 5 6.7 Hz); 7.52
(1H, d, J 5 2.6 Hz); 7.37 (1H, d, J 5 4.6 Hz); 7.35 (1H, dd, Ja 5 6.7 Hz,
Jb 5 2.6 Hz); 7.11 (1H, d, Ja 5 4.4 Hz); 6.91 (1H, s); 6.05 (1H, ddd, Ja 5
17.4 Hz, Jb 5 10.4 Hz, Jc 5 7.3 Hz); 5.12 (1H, ddd,); 5.09 (1H, ddd, Ja 5
17.4 Hz, Jb5c 5 1.6 Hz); 3.97 (3H, s); 3.39 (1H, m); 3.06 (1H, m); 2.95
(1H, m); 2.86 (1H, m); 2.78 (1H, m); 2.65 (2H, m); 2.27 (1H, m); 2.25
(2H, m); 2.06 (1H, m); 1.97 (1H, t, Ja5b 5 2.7 Hz); 1.84 (1H, m); 1.76
(2H, m); 1.61 (2H, m); 1.60–1.58 (2H, m); 1.54 (1H, m). 13C NMR
(150 MHz, CDCl3) d 5 163.7; 157.9; 152.6; 147.3; 144.7; 144.3; 143.6;
140.3; 131.7; 127.2; 121.9; 118.5; 118.1; 114.9; 101.7; 83.6; 76.5; 69.0; 59.7;
55.7; 49.9; 49.3; 39.7; 31.9; 27.9; 27.7; 26.7; 26.6; 23.1; 18.1. MS (ESI1)
m/z 5 517.3 [M1H1]. [a]D25 5 285.7 (c 5 1.28, CH2Cl2).
CONCLUSIONS
In summary, several features of the enantioselective meth-
anolysis of meso-anhydrides mediated by three quinidine
derivatives were explored by combining organocatalyst varia-
tion, catalytic desymmetrization runs, and detailed NMR
measurements.
Even if the lack of spectroscopic evidences for the interac-
tion between the anhydride substrate and the alkaloid deriva-
tives does not necessarily rule out the nucleophilic catalysis
path, in terms of general mechanistic scheme the experimen-
tal observation of a strong influence of the alcohol reactant
on the NMR properties of the alkaloid derivatives, provided
herein, appears to be better accommodated by the assump-
tion of a general-base type of catalysis.
Moreover, the in-depth investigation of the solution prop-
erties of chiral organocatalysts revealed a somewhat unantici-
pated strong influence of the alkaloid’s 9-O substituent on
the conformational prevalence of the chiral units, as well as
the tendency of the latter to undergo Closed-to-Open transi-
tions in the presence of the alcohol reactant. At variance with
previous literature indications, no obvious correlation could
be found, however, between the propensity of the different
4-(4-(1-Benzyl-1H-1,2,3-triazol-4-yl)butyl)-
3,6-bis(9-O-quinidinyl)pyridazine (13) and
4-(4-(1-benzyl-1H-1,2,3-triazol-4-yl)butyl)-6-chloro-
3-(9-O-quinidinyl)pyridazine (14)
A 25-ml Schlenk tube was charged under dry nitrogen with the crude
mixture of 11 and 12 (1.00 g, ꢀ1.1 and 0.23 mmol, respectively), benzyl
azide (0.33 g, 2.4 mmol), CuCl (0.0013 g, 13 lmol, ꢀ1 mol %), (R)-Mono-
Phos (0.0048 g, 13 lmol, 1 mol %) and THF (11 ml). After stirring over-
night at r.t., the solution was concentrated with a rotary evaporator and
the dark residue (1.123 g) subjected to flash chromatography (SiO2,
Chirality DOI 10.1002/chir

794
BALZANO ET AL.
ing with benzyl alcohol and its application in the synthesis of highly
enantiomerically enriched b-amino acids. Tetrahedron Asymmetry
2003;14:3455–3467.
chiral derivatives to attain a Closed(2) conformation and their
catalytic properties. Therefore, the observed changes in the
catalytic efficiency of the different organocatalysts cannot be
traced back to a single structural feature, leading us to pro-
pose that the achiral 9-O derivatizing group could have a
more active role in the catalytic process (e.g., by virtue of its
stereoelectronic properties) than merely enforcing the alka-
loid units in a preferential reactive conformation.
Studies are currently underway to further investigate the
structure-activity relationship for Cinchona alkaloid derivatives,
elucidate the effects of products-organocatalyst interaction, and
address the use of polymer-supported analogs of 13 and 14 as
heterogeneized organocatalysts in the asymmetric ring-open-
ing of meso-anhydrides. Moreover, the possibility of exploiting
the chiral discrimination ability of the alkaloid organocatalysts
for the in situ NMR monitoring of product ee evolution along
the reaction course is actively investigated at present.
20. Deng L, Liu X, Chen Y, Tian S. Preparation of cinchona-alkaloid deriva-
tives for use as chiral catalysts for asymmetric mono-esterification/alco-
holysis of prochiral and meso cyclic anhydrides patent PCT Int Appl WO
2,004,110,609 A2, 2004.
21. Ishii Y, Fujimoto R, Mikami M, Murakami S, Miki Y, Furukawa Y. Practi-
cal syntheses of chiral a-amino acids and chiral half-esters by kinetic
resolution of urethane-protected a-amino acid N-Carboxyanhydrides and
desymmetrization of cyclic meso-anhydrides with new modified cinchona
alkaloid catalysts. Org Process Res Dev 2007;11:609–615.
22. Li H, Liu X, Wu F, Tang L, Deng L. Elucidation of the active conforma-
tion of cinchona alkaloid catalyst and chemical mechanism of alcoholysis
of meso anhydrides. Proc Natl Acad Sci U S A 2010;107:20625–20629.
23. Peschiulli A, Gun’ko Y, Connon SJ. Highly enantioselective desymmetri-
zation of meso anhydrides by a bifunctional thiourea-based organocata-
lyst at low catalyst loadings and room temperature. J Org Chem 2008;73:
2454–2457.
24. Connon SJ. Asymmetric catalysis with bifunctional cinchona alkaloid-
based urea and thiourea organocatalysts. Chem Commun 2008:2499–
2510.
LITERATURE CITED
1. Ward RS. Nonenzymic asymmetric transformations involving symmetri-
cal bifunctional compounds. Chem Soc Rev 1990;19:1–19.
25. Bigi F, Carloni S, Maggi R, Mazzacani A, Sartori G, Tanzi G. Homogene-
ous versus heterogeneous approach to the catalytic desymmetrisation of
meso-anhydrides promoted by cinchona alkaloids. J Mol Catal A Chem
2002;182–183:533–539.
2. Willis MC. Enantioselective desymmetrization. J Chem Soc Perkin Trans
1 1999;1765–1784.
26. Woltinger J, Krimmer H-P, Drauz K. The potential of membrane reactors
in the asymmetric opening of meso-anhydrides. Tetrahedron Lett
2002;43:8531–8533.
3. Spivey AC, Andrews BI. Catalysis of the asymmetric desymmetrization of
cyclic anhydrides by nucleophilic ring-opening with alcohols. Angew
Chem Int Ed Engl 2001;40:3131–3134.
27. Kim HS, Song Y-M, Choi JS, Yang JW, Han H. Heterogeneous organoca-
talysis for the asymmetric desymmetrization of meso-cyclic anhydrides
using silica gel-supported bis-cinchona alkaloids. Tetrahedron 2004;60:
12051–12057.
4. Chen Y, McDaid P, Deng L. Asymmetric alcoholysis of cyclic anhy-
drides. Chem Rev 2003;103:2965–2983.
5. France S, Guerin DJ, Miller SJ, Lectka T. Nucleophilic chiral amines as
catalysts in asymmetric synthesis. Chem Rev 2003;103:2985–3012.
28. Song Y-M, Choi JS, Yang JW, Han H. Silica gel-supported bis-cinchona
alkaloid: a chiral catalyst for the heterogeneous asymmetric desymmetri-
zation of meso-cyclic anhydrides. Tetrahedron Lett 2004;45:3301–3304.
6. Tian S-K, Chen Y, Hang J, Tang L, McDaid P, Deng L. Asymmetric or-
ganic catalysis with modified cinchona alkaloids. Acc Chem Res
2004;37:621–631.
29. Ivsic T, Hamersak Z. Inversion of enantioselectivity in quinine-mediated
desymmetrization of glutaric meso-anhydrides. Tetrahedron Asymmetry
2009;20:1095–1098.
7. Berkessel A, Groerger H. Desymmetrization and kinetic resolution of anhy-
drides; desymmetrization of meso-epoxides and other prochiral substrates.
Asymmetric organocatalysis. Weinheim: Wiley-VCH; 2005. p 347–55.
30. Dedeoglu B, Catak S, Houk KN, Aviyente V. A Theoretical study of the
mechanism of the desymmetrization of cyclic meso-anhydrides by chiral
amino alcohols. ChemCatChem 2010;2:1122–1129.
8. Atodiresei I, Schiffers I, Bolm C. Stereoselective anhydride openings.
Chem Rev 2007;107:5683–5712.
9. Lee JW, Jang HB, Lee JE, Song CE. Cinchona-based catalysts for desym-
metrization of meso-compounds and (dynamic) kinetic resolution of race-
mic compounds. In: Song CE, editor. Cinchona Alkaloids in Synthesis
and Catalysis. Weinheim: Wiley-VCH; 2009. p 325–357.
31. Palomo C, Oiarbide M, Lopez R. Asymmetric organocatalysis by chiral
Bronsted bases: implications and applications. Chem Soc Rev 2009;38:
632–653.
32. Pini D, Petri A, Nardi A, Rosini C, Salvadori P. Heterogeneous catalytic
asymmetric dihydroxylation of olefins with the OsO4/poly(9-O-acylqui-
nine-co-acrylonitrile) system. Tetrahedron Lett 1991;32:5175–5178.
10. Marcelli T, Hiemstra H. Cinchona alkaloids in asymmetric organocataly-
sis. Synthesis 2010;1229–1279.
11. Hiratake J, Yamamoto Y, Oda Ji. Catalytic asymmetric induction from
prochiral cyclic acid anhydrides using cinchona alkaloids. J Chem Soc
Chem Commun 1985:1717–1719.
33. Mandoli A, Pini D, Fiori M, Salvadori P. Asymmetric dihydroxylation
with recoverable cinchona alkaloid derivatives: a warning note and an
improved, insoluble polymer-bound ligand architecture. Eur J Org Chem
2005;1271–1282.
12. Hiratake J, Inagaki M, Yamamoto Y, Oda J. Enantiotopic-group differen-
tiation. Catalytic asymmetric ring-opening of prochiral cyclic anhydrides
with methanol, using Cinchona alkaloids. J Chem Soc Perkin Trans 1
1987;1053–1058.
34. Jumde RP, Mandoli A, De Lorenzi F, Pini D, Salvadori P. Simple prepara-
tion of dimeric cinchona alkaloid derivatives on polystyrene supports and
a
highly enantioselective catalytic heterogeneous dimerization of
13. Aitken RA, Gopal J, Hirst JA. Catalytic asymmetric synthesis of highly
functionalized compounds with six contiguous stereocenters. J Chem
Soc Chem Commun 1988:632–634.
ketenes. Adv Synth Catal 2010;352:1434–1440.
35. Rosini C, Bertucci C, Pini D, Altemura P, Salvadori P. Cinchona alkaloids
for preparing new, easily accessible chiral stationary phases. I. 11-(10,11-
Dihydro-6’-methoxycinchonan-9-ol)-thiopropylsilanized silica. Tetrahedron
Lett 1985;26:3361–3364.
14. Aitken RA, Gopal J. Catalytic asymmetric ring-opening of bridged tricy-
clic anhydrides. Tetrahedron Asymmetry 1990;1:517–520.
15. Bolm C, Gerlach A, Dinter CL. Simple and highly enantioselective nonen-
zymic ring opening of cyclic prochiral anhydrides. Synlett 1999;195–196.
36. Pini D, Rosini C, Bertucci C, Altemura P, Salvadori P. Chromatographic
resolution of alkyl aryl carbinols on a silica-supported quinine chiral sta-
tionary phase. Gazz Chim Ital 1986;116:603–606.
16. Bolm C, Schiffers I, Dinter CL, Gerlach A. Practical and highly enantiose-
lective ring opening of cyclic meso-anhydrides mediated by cinchona alka-
loids. J Org Chem 2000;65:6984–6991.
37. Rosini C, Bertucci C, Pini D, Altemura P, Salvadori P. Chiral stationary
phases based on Cinchona alkaloids for the HPLC resolution of race-
mates. Chromatographia 1987;24:671–676.
17. Chen Y, Tian S-K, Deng L. A highly enantioselective catalytic desymmet-
rization of cyclic anhydrides with modified cinchona alkaloids. J Am
Chem Soc 2000;122:9542–9543.
38. Salvadori P, Rosini C, Pini D, Bertucci C, Altemura P, Uccello-Barretta G,
Raffaelli A. A novel application of Cinchona alkaloids as chiral auxiliaries:
preparation and use of a new family of chiral stationary phases for the chro-
matographic resolution of racemates. Tetrahedron 1987;43:4969–4978.
18. Kolb HC, VanNieuwenhze MS, Sharpless KB. Catalytic Asymmetric dihy-
droxylation. Chem Rev 1994;94:2483–2547.
19. Bolm C, Schiffers I, Atodiresei I, Hackenberger CPR. An alkaloid-medi-
39. Salvadori P, Rosini C, Pini D, Bertucci C, Uccello-Barretta G. Resolution
ated desymmetrization of meso-anhydrides via a nucleophilic ring open-
of some b-hydroxyphenethylamines as N-(3,5-dinitrobenzoyl)amides on
Chirality DOI 10.1002/chir

NMR INVESTIGATIONS OF CATALYSIS WITH QUINIDINE DERIVATIVES
795
chiral stationary phases derived from cinchona alkaloids. Chirality 1989;
1:161–166.
molecular mechanics and x-ray approach. J Am Chem Soc 1989;111:
8069–8076.
40. Uccello-Barretta G, Pini D, Rosini C, Salvadori P. Stereochemical features
of the stereoselective interaction between (R)- or (S)-7,7’-bis(1-propen-3-
oxy)-2,2’-dihydroxy-1,1’-binaphthyl and quinine. J Chromatogr A 1994;
666:541–548.
49. Dijkstra GDH, Kellogg RM, Wynberg H. Conformational study of cin-
chona alkaloids. A combined NMR and molecular orbital approach. J Org
Chem 1990;55:6121–6131.
50. Buergi 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.
41. Rosini C, Uccello-Barretta G, Pini D, Abete C, Salvadori P. Quinine: an
inexpensive chiral solvating agent for the determination of enantiomeric
composition of binaphthyl derivatives and alkylarylcarbinols by NMR
spectroscopy. J Org Chem 1988;53:4579–4581.
51. Johnson CS Jr. Diffusion ordered nuclear magnetic resonance spectros-
copy: principles and applications. Prog Nucl Magn Reson Spectrosc 1999;
34:203–256.
42. Salvadori P, Pini D, Rosini C, Bertucci C, Uccello-Barretta G. Chiral dis-
criminations with Cinchona alkaloids. Chirality 1992;4:43–49.
52. Govil G, Hosur RV. Conformation of biological molecules: new results
from NMR. In: Diehl P, Fluck E, Kosfeld R, editors. NMR Basic Princi-
ples and Progress. Berlin: Springer-Verlag 1982. Vol. 20.
43. Uccello-Barretta G, Pini D, Mastantuono A, Salvadori P. Direct NMR
assay of enantiomeric purity of chiral b-hydroxy esters by using qui-
nine as chiral solvating agent. Tetrahedron Asymmetry 1995;6:1965–
1972.
53. Uccello-Barretta G, Di Bari L, Salvadori P. Autoaggregation phenomena
in quinine solutions. Magn Reson Chem 1992;30:1054–1063.
44. Uccello-Barretta G, Bardoni S, Balzano F, Salvadori P. Versatile chiral
auxiliaries for NMR spectroscopy based on carbamoyl derivatives of dihy-
droquinine. Tetrahedron Asymmetry 2001;12:2019–2023.
54. Uccello-Barretta G, Balzano F, Bardoni S, Vanni L, Giurato L, Guccione
S. Chiral discrimination processes by C9 carbamate derivatives of dihy-
droquinine: interaction mechanisms of diastereoisomeric 9-O-[(S)- or (R)-
1-(1-naphthyl)ethylcarbamate]dihydroquinine and the two enantiomers of
N-(3,5-dinitrobenzoyl)alanine methyl ester. Tetrahedron Asymmetry
2008;19:1084–1093.
45. Uccello-Barretta G, Balzano F, Salvadori P. Rationalization of the multire-
ceptorial character of chiral solvating agents based on quinine and its
derivatives: Overview of selected NMR investigations. Chirality 2005;17:
S243–S248.
55. Uccello-Barretta G, Balzano F, Quintavalli C, Salvadori P. Different enan-
tioselective interaction pathways induced by derivatized quinines. J Org
Chem 2000;65:3596–3602.
46. Uccello-Barretta G, Vanni L, Balzano F. NMR enantiodiscrimination phe-
nomena by quinine C9-carbamates. Eur J Org Chem 2009:860–869.
47. Campbell-Verduyn LS, Mirfeizi L, Dierckx RA, Elsinga PH, Feringa BL.
Phosphoramidite accelerated copper(i)-catalyzed [3 1 2] cycloadditions
of azides and alkynes. Chem Commun 2009:2139–2141.
56. Uccello-Barretta G, Mirabella F, Balzano F, Salvadori P. C11 versus C9
carbamoylation of quinine: a new class of versatile polyfunctional chiral
solvating agents. Tetrahedron Asymmetry 2003;14:1511–1516.
48. Dijkstra GDH, Kellogg RM, Wynberg H, Svendsen JS, Marko I, Sharp-
less KB. Conformational study of cinchona alkaloids. A combined NMR,
57. Fielding L. Determination of association constants (Ka) from solution
NMR Data. Tetrahedron 2000;56:6151–6170.
Chirality DOI 10.1002/chir