One catalyst for both enantiomers: Uncovering the inversion of enantioselectivity in cinchona-mediated desymmetrization of glutaric meso-anhydrides

Article abstract of DOI:10.1016/j.tet.2012.07.035

A puzzling inversion of enantioselectivity dependent on catalyst loading was observed during the quinine-mediated desymmetrization of glutaric meso-anhydrides. This study presents the improvement of the catalytic path by the inclusion of carboxylic acid additives up to synthetically useful levels. The novel protocol utilizing 0.1 equiv of alkaloid and xanthene-9-carboxylic acid at room temperature (rt) was found comparable to the protocol requiring 1.1 equiv of alkaloid at -30 °C. Thus, by altering the protocol the same catalyst produces the opposite enantiomer. This occurrence was rationalized by an extensive computational study of the interactions governing the molecular complexes formed by quinine, methanol, 3-methylglutaric anhydride, and the acetic acid. It was found that in a quinine catalyzed reaction the alcohol and the anhydride were directly hydrogen bonded to the catalyst. On the other hand, in the reaction with additive the acid intercalates between the alcohol and quinine. Due to this insertion the alcohol approaches the anhydride from the opposite face, in agreement with the observed inversion of enantioselectivity

Full text of DOI:10.1016/j.tet.2012.07.035

Tetrahedron 68 (2012) 8311e8317
Contents lists available at SciVerse ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
One catalyst for both enantiomers: uncovering the inversion of enantioselectivity
in cinchona-mediated desymmetrization of glutaric meso-anhydrides
b
a,
ꢀ ꢁ a
ꢀ ꢁ b
ꢀ
*
Trpimir Ivsic , Jurica Novak , NaCa Doslic , Zdenko Hamersak
a
ꢀ
ꢁ
Department of Organic Chemistry and Biochemistry, Rudjer Boskovic Institute, PO Box 180, 10002 Zagreb, Croatia
Department of Physical Chemistry, Rudjer Boskovic Institute, PO Box 180, 10002 Zagreb, Croatia
b
ꢀ
ꢁ
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 13 April 2012
Received in revised form 13 June 2012
Accepted 10 July 2012
Available online 21 July 2012
A puzzling inversion of enantioselectivity dependent on catalyst loading was observed during the qui-
nine-mediated desymmetrization of glutaric meso-anhydrides. This study presents the improvement of
the catalytic path by the inclusion of carboxylic acid additives up to synthetically useful levels. The novel
protocol utilizing 0.1 equiv of alkaloid and xanthene-9-carboxylic acid at room temperature (rt) was
found comparable to the protocol requiring 1.1 equiv of alkaloid at ꢀ30 ꢁC. Thus, by altering the protocol
the same catalyst produces the opposite enantiomer.
This occurrence was rationalized by an extensive computational study of the interactions governing the
molecular complexes formed by quinine, methanol, 3-methylglutaric anhydride, and the acetic acid. It
was found that in a quinine catalyzed reaction the alcohol and the anhydride were directly hydrogen
bonded to the catalyst. On the other hand, in the reaction with additive the acid intercalates between the
alcohol and quinine. Due to this insertion the alcohol approaches the anhydride from the opposite face,
in agreement with the observed inversion of enantioselectivity
Keywords:
Desymmetrization
Enantioselectivity
Density functional calculations
Asymmetric synthesis
Hydrogen bonding
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
The enantioselective desymmetrization of cyclic meso-anhy-
drides1 (Scheme 1) is of critical importance in the total syntheses of
numerous biologically active substances,2e5 including Pregabalin,6
Biotin,7 Baclofen,8 etc. The asymmetric opening of cyclic meso-an-
hydrides was first put forward by Oda9 and Aitken.10 The stoi-
chiometric protocol with natural cinchona alkaloids was further
developed by Bolm and co-workers.11 Alternatively, a catalytic
amount of modified cinchona alkaloids including ethers,12 ureas or
thioureas,13 and sulfonamides14 was also found to steer the reaction
toward the desired enantiomer. Modified cinchona alkaloids pro-
vide exceptionally high enantioselectivities with succinic anhy-
drides, but in the case of more demanding glutaric anhydrides
inferior results are obtained with the same catalyst loading.13 Thus,
in the case of glutaric anhydrides, especially when larger quantities
of products are needed, the use of easily available, inexpensive, and
recoverable unmodified alkaloids is still the best option.
Recently reported was a Pregabalin synthesis6a where the
quinine-mediated ring opening of 3-isobutylglutaric anhydride was
the key step. At low catalyst loading an unexpected inversion of
enantioselectivity has been observed.15 More generally it has been
* Corresponding author. Fax: þ385 14680108; e-mail address: hamer@irb.hr (Z.
Scheme 1. Desymmetrization of glutaric meso-anhydrides.
ꢀ
Hamersak).
0040-4020/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tet.2012.07.035

ꢀ ꢁ
T. Ivsic et al. / Tetrahedron 68 (2012) 8311e8317
8312
demonstrated that organic acids as additives also invert the di-
rection of enantioselectivity. Here we elaborate further on these
findings and explore the synthetic utility of the alkaloideacid co-
catalyst approach as our first goal.
The mechanism of desymmetrization of meso-cyclic anhydrides
by alcohol nucleophiles has been the subject of controversial dis-
cussions in the literature. In the nucleophilic catalysis,10,16 the re-
action is initiated by the attack of the amino group of the quinidine
on the sterically less-hindered side of the anhydride resulting in the
formation of a chiral acylammonium salt. The intermediate is sta-
bilized via a hydrogen bond between the OH group of the catalyst
and the carboxylate moiety. In the next step the alcohol reacts with
the salt yielding the ester product. Alternatively in the general base
catalysis proposed by Oda and co-workers9 the alcohol is depro-
tonated by the amino group of the catalyst. In the subsequent step
a nucleophilic attack of the resulting methoxide onto the anhydride
leads to ring opening. Again, the formation of a hydrogen bond
between the OH group of the catalyst and the carboxylate moiety
facilitates the reaction. Recently, Dedeoglu and co-workers17 shed
light on the energetics of the two pathways. Based on the DFT
calculations of a system consisting of a meso-cyclic anhydride and
a model catalyst, (1R,2R)-2-(piperidin-1-yl)cyclohexanol, they have
shown that the formation of the intermediate salt in the nucleo-
philic catalysis is energetically highly unfavorable. While the study
undoubtedly points toward general base catalysis, the experimen-
tally observed stereochemistry could not be explained on the basis
of model calculations.17
Fig. 1. Influence of catalyst loading on enantioselective opening of
reactions were performed in toluene (0.1 M with respect to anhydride) with 1.5 equiv
of benzyl alcohol at rt. For reactions with acid additives, acid/quinine molar ratio was 2.
1 and 3. All
Table 1
Influence of acid additives on the desymmetrization of anhydride (1)
No.
1
Acid
eea/% No.
Acid
eea/%
47
36
45
5
6
Therefore, the second goal of this paper is to unravel the in-
teractions in quinine-mediated ring opening of cyclic meso-anhy-
drides leading to the observed stereochemistry, as well as to
elucidate the role of organic acids in the inversion of enantiose-
lectivity. Specifically, we focus on the hydrogen bonding in-
teractions and show that both, the original enantioselectivity and
the inversed one can be traced back to the pro-chirality of the most
abundant reaction complexes.
2
3
38
55
53
63
7
8
2. Results and discussion
We present first the experimental results on the inversion of
enantioselectivity in alkaloid and alkaloideacid catalyzed opening
of glutaric anhydrides, and then explore the enantioselectivity of
both reactions.
4
54
Reaction conditions: 0.1 M anhydride 1 in toluene, 0.5 equiv of quinine, 1 equiv of
acid, 1.5 equiv of BnOH, rt.
a
2.1. Synthetic potential
Major enantiomer (S)-configuration, determined by chiral HPLC on chiralcel AS
(EtOH/hexane/TFA¼2:98:0.1).
As mentioned in Introduction, we reported an inversion of
enantioselectivity in the quinine catalyzed desymmetrization of
glutaric anhydrides at low catalyst loading.15 Specifically, in the
case of anhydride 1 it was found that a catalyst loading of 160%
produces about 40% ee of (R)-product, while a 10% loading yields
40% ee of (S)-product. In contrast, succinic anhydrides tend to
produce racemic products if less of alkaloid is loaded (see Fig.1).10,11
Recently, Bolm also reported a similar reversion of enantiose-
lectivity for alkaloid-mediated thiolyses of succinic meso-anhy-
that better enantioselectivities are obtained with aprotic solvents
(see Table 2).
On the other hand, as shown for urea and thiourea derived al-
kaloids, dilution effects the enantioselectivity.13,19,20 Lowering the
anhydride concentration from 0.125 to 0.025 M raises the enan-
tioselectivity from 85% ee to 96% ee.13 The same effect of dilution
was observed with the quinine/acid combination (Fig. 2). Although
high dilution improves the enantioselectivity, the reaction rate is
significantly lowereddat 0.005 M solution the reaction was found
incomplete after 6 days.
The enantioselecivity of the reaction can significantly be im-
proved by lowering the reaction temperature when the alkaloid is
employed in stoichiometric quantity. With the quinine/acid type of
catalyst, lowering the temperature causes a minor improvement in
the enantioselectivity, but a drastic reduction of the reaction rate.
Finally, the influence of quinine/acid ratio on enantioselectivity was
explored. For both acid additives the selectivity of the reaction
reached its maximum at a base/acid ratio of 1:2 (Fig. 3).
drides.18 The fact that at low catalyst loading
a specific
productebase complex inverts the direction of opening of glutaric
anhydrides encouraged us to verify whether some other organic
acid additives could improve the stereoselectivity. More than 25
different carboxylic acids were tested and a representative selec-
tion is compiled in Table 1.
The best enantioselectivities were achieved with acids pos-
sessing phenylacetic acid substructure (entries 3, 4 (X9C), and 8) as
well as with 2-thiophene-acetic acid (entry 7), which can be con-
sidered as phenylacetic acid congener. We also noted that the di-
rection of opening is no longer dependent on catalyst loading, and

ꢀ ꢁ
T. Ivsic et al. / Tetrahedron 68 (2012) 8311e8317
8313
Table 2
R'O
O
R'O
O
Solvent effect on opening selectivity
alkaloid(0.1 eq)
X9C (0.2 eq)
R'OH, RT
R
R
+
+
Entry
Solvent
ee/%
eea/%
OH
OH
1
2
3
4
5
6
Chloroform
Toluene
i-Pr2O
MTBE
Dioxane
THF
50.2
47.5
43.5
39.6
26.5
15
68
65
d
d
d
d
O
O
R
O
A
B
(S)-4,5,6
(R)-4,5,6
O
R'O
O
R'O
O
O
Reaction conditions: 0.1 M anhydride 1 solution, 0.5 equiv of quinine, 1 equiv of
R
R
benzoic acid, 1.5 equiv of BnOH, 65 h at rt.
alkaloid(1.1 eq)
R'OH, –30 °C
a
Xanthene-9-carboxylic acid as additive.
OH
OH
O
O
(S)-4,5,6
(R)-4,5,6
Scheme 2. Methodologies for enantioselective anhydride opening.
carboxylic acid was found to be the most promising acid additive.
The results are compiled in Table 3. With all anhydrides and all
catalysts comparable results were obtained. Enantioselectivities
were in the range of 56e73%, which is quite good bearing in mind
that unmodified, easily available, and recyclable catalysts were
used, and that glutaric anhydrides open with lower enantiose-
lectivities in comparison with succinic. We should mention that the
latter are typically used for catalytic activity testing.
Table 3
Enantioselective opening of anhydrides 1 and 2. Comparison of methodologiesa
Anhydride
Methodb
R0
T/ꢁC
Time
h
d/%
eee/%
Product
1
1
1
1
1
2
2
2
2
A(0.1QDþX)
B(1.1Q)
A(0.1QþX)
A(0.1QþX)
B(1.1Q)
Bn
Bn
Bn
Cinn
Cinn
Bn
Bn
Bn
rt
22 h
23 h
18 h
18 h
25 h
2 d
7 d
2 d
7 d
70
72
73
64
65
61
69
72
77
56
62
(R)-4
(R)-4
(S)-4
(S)-5
(R)-5
(R)-6
(R)-6
(S)-6
(S)-6
ꢀ25
Fig. 2. Dilution effect on opening selectivity. Reaction conditions: anhydride 1 in tol-
uene, 0.5 equiv quinine, 1 equiv acetic acid, 1.5 equiv BnOH, 65 h at rt.
rt
n.i.c
n.i.c
85
rt
ꢀ25
A(0.1QDþX)
rt
82
60
87
74
To verify the synthetic utility of the enantioselective opening of
3-substituted glutaric meso-anhydrides catalyzed by the alka-
loideacid complexes, this methodology was compared with the
stoichiometric approach (Scheme 2). Anhydrides 1 and 2 were
chosen because their hemiester products provide an easy access to
both enantiomers of pharmacologically interesting compounds
(Pregabalin, Baclofen) in a multigram scale, while xanthene-9-
B (1.1Q)
ꢀ35
rt
ꢀ35
A(0.1QþX)
B(1.1QD)
Bn
a
Reactions were carried out with 10 mmol of anhydrides, 0.1 M solution in tol-
uene, 1.5 equiv of alcohol until >90% conversion was reached.
b
Methods: (A) 0.1 equiv of quinidine (QD) or quinine (Q), 0.2 equiv of xanthene-
9-carboxylic acid (X); (B) 1.1 equiv of quinine (Q) or quinidine (QD).
c
Scale: 0.5 mmol.
Isolated yield.
Determined by chiral HPLC.
d
e
The slightly lower yields on isolated monoesters of anhydrides 1
and 2 compared to the conversion can be attributed to slow ester
hydrolysis and the emulsification properties of their potassium
salts during the work-up, which comprises successive washing of
alkaline aqueous solution of monoester with organic solvent to
remove residual alcohol.
Altogether the alkaloidexanthene-9-carboxylic acid complex
was found to have several advantages when compared with natural
alkaloids in stoichiometric amount: lower catalyst loading, faster
reaction, and the most importantdan ambient reaction
temperature.
2.2. Computational study
The conformational space of cinchona alkaloids has been in-
tensively studied.21,22 Several combined experimental and theo-
retical studies indicate that the open conformer of cinchona
alkaloids, known as Open(3), is the most populated one at room
temperature in apolar solvents.22 In the following we focus on this
conformer.
Fig. 3. Quinine/acid ratio effect on opening selectivity. aReactions of anhydride 1 (6 mL
of 0.1 M toluene solution) and 1.5 equiv BnOH with 0.5 equiv of quinine and acetic acid
at rt. bReactions of anhydride 1 (6 mL of 0.1 M toluene solution) and 1.5 equiv BnOH
with 0.1 equiv of quinine and xanthene-9-carboxylic acid at rt.

ꢀ ꢁ
T. Ivsic et al. / Tetrahedron 68 (2012) 8311e8317
8314
In principle, a series of plausible initial geometries of the
the exposure of the pro-S carbonyl, elongation of the pro-S CeO
bond and to the likely formation of the (S)-product. Comparing the
geometries of the four conformers it is apparent that they are
stabilized by a comparable network of hydrogen bonding in-
teractions. A closer look at the geometries, however, reveals that
the stabilization of the pro-R conformers with respect to the pro-S
ones stems from more favorable steric interactions and in par-
ticular from the larger separation between methanol and the
anhydride hydrogen atoms in the axial position. For example in A
the distance between methanol O and the two axial atoms HA and
three molecular complex consisting of quinine, methanol, and
3-methylglutaric anhydride required for full geometry optimiza-
tions at the B3LYP-D3/6-31G(d,p) level of theory can be obtained
from classical molecular dynamics simulations. However, force
field calculations are not adequate for describing the hydrogen
bonding and dispersive interactions that determine the enantio-
selective ring opening, and the size of the system prevents the use
of long DFT/ab initio molecular dynamics runs needed to explore
the rather rough conformational landscape. Hence, our initial
structures were based on a series of educated guesses facilitating
both the nucleophilic and the general base catalysis. As the results
of full geometry optimizations at the B3LYP/6-31 G(d) level we
found nine distinct structures spanning an energy range of
13.97 kcal/mol. The four lowest structures plus another four
obtained by inverting the stereochemical face of the anhydride
have been reoptimized at the B3LYP-D3/6-31G(d,p) level. The rel-
ative energies, predicted stereochemistry of the final product, and
selected geometrical parameters are compiled in Table 4.
ꢂ
ꢂ
HD is 2.41 A and 2.51 A, respectively, while a closer contact with HE
0
ꢂ
of 2.33 A is found in A .
The addition of organic acids causes the inversion of enan-
tioselectivity. Preliminary exploration of the configurational
space of the acetic acid, quinine, methanol, and 3-methylglutaric
anhydride complex at the B3LYP/6-31G(d) level of theory yielded
18 distinct structures with an energy range of 14.7 kcal/mol.
Table 5 compiles the relative energies, predicted chirality, and
relevant geometrical parameters of the four lowest energy
structures found in the computation. The four lowest energy
conformers are shown in Fig. 5. As a general feature we found
that the acetic acid has replaced methanol in the hydrogen bond
with the quinuclidine amine. In all structures the quinuclidine N
is protonated. The methanol is now hydrogen bonded to the
acetate.
Generally we have found that all investigated structures are
stabilized by two hydrogen bonds, one between the alcohol and the
quinuclidine amine, and the other one between the oxygen of one
of the carboxyl groups and the quinine OH. The distance between
the quinuclidine N and the closest carbonyl C atom of the anhydride
ꢂ
ꢂ
is relatively large, in the range between 4.2 A and 4.6 A. In contrast,
the distance between the methanol O and the closest carbonyl C
In the lowest energy conformer C0 the pro-R carbonyl is H-
bonded to the catalyst OH so the pro-S carbonyl becomes the pre-
ferred site for the methoxide nucleophilic attack, in agreement
with the observed inversion of enantioselectivity. The corre-
sponding pro-(R) conformer C is found 2.23 kcal/mol higher in
energy, while the next two structures D and D0 featuring pro-R and
pro-S chirality are found 1.10 and 2.17 kcal/mol higher in energy,
respectively. Note also the different binding modes in C and D.
Besides the greater stability of the pro-R and pro-S conformers
in the quinine and quinineeacid catalyzed reactions, respectively,
the enantioselectivity is affected by the height of the activation
barriers. It is apparent, however, that in order to simulate the in-
teractions within the molecular complexes these need to be treated
in their full complexity. Reaction path calculations for such highly
fluxional systems are very demanding and they remain beyond the
scope of the present study.
ꢂ
atom is much shorter (2.7e3.2 A). The geometries of these com-
plexes point to the general base catalysis mechanism of desym-
metrization of cyclic anhydrides.
Table 4
Relative energies, predicted chirality, and selected geometry parameters of the four
lowest energy structures as obtained at the B3LYP-D3/6-31G(d,p) level of theory. The
prediction of chirality is based on the ratio of the bond lengths d(C1eO6)/d(C5eO6).
The numeration of atoms is given in Scheme 3
Conf. E/kcal molꢀ1 Predict. config. Distance/A
ꢂ
d1/d2a
(O,N)b (O,O)c (O,C)d
A
0.00
1.47
0.12
1.13
R
S
R
S
1.37/1.41 2.76
1.41/1.36 2.75
1.37/1.41 2.76
1.41/1.37 2.75
2.82
2.84
2.83
2.84
3.07
2.69
3.02
2.82
A0
B
B0
a
b
c
The stereochemistry of the product depends on whether
methoxide attacks the hydrogen bonded or the non-hydrogen
bonded carbonyl. The non-hydrogen bonded carbonyl is in our
opinion the site of the methoxide attack. By assuming this we were
able to rationalize the observed enantioselectivity in both quinine
and quinineeacid catalyzed reactions.
Bond lengths ratio d(C1eO6)/d(C5eO6).
H-bond distance between methanol OeH and quinuclidine N.
H-bond distance between quinine OH and the closest carbonyl O.
d
Distance between methanol OeH and the non H-bonded carbonyl C.
Fig. 4 displays the lowest energy structure (conformer A) and
the corresponding chiral structure (conformer A0). The latter was
obtained by reflecting the anhydride in the plane defined by the
atoms C(1)eO(6)eC(5) and subsequent full geometry optimization
(see Scheme 3). The energy difference between A (Fig. 4, top left)
and A0 conformers (Fig. 4, top right) is 1.47 kcal/mol, while the
energy difference between conformer B and conformer B0 is
1.01 kcal/mol.
Conformers A and B are stabilized by two hydrogen bonds, one
between the alcohol and the quinuclidine N, and the other one
between the pro-S carboxyl group and the catalyst OH. The natural
bonding orbital (NBO) analysis reveals that there is virtually no
difference in the charges between the two carbonyl C atoms. Be-
cause of the involvement of the pro-S carbonyl in the hydrogen
bond with quinine OH, the pro-R carbonyl is more flexible and
hence more prone to the methoxide nucleophilic attack.
3. Conclusion
A novel room temperature catalytic protocol for enantiose-
lective opening of 3-substituted glutaric meso-anhydrides utilizing
alkaloideacid complex was explored in detail.
Its synthetic utility was demonstrated in comparison to the
stoichiometric approach at ꢀ30 ꢁC in the formal syntheses of
sample pharmacological compounds. In both cases good enantio-
selectivities were found, particularly bearing in mind that glutaric
anhydrides typically open with lower enantioselectivities as com-
pared to the succinic.
The option of using a single catalyst, but two protocols to obtain
both enantiomers in excess is the most intriguing result, especially
as unmodified, easily available, and recyclable catalysts were used.
To rationalize this occurrence, an extensive computational study
on the molecular interactions in model compounds was carried out.
In all of the low energy structures, the carbonyl oxygen of the an-
hydride was H-bonded to the hydroxyl group of the catalyst. In
Assuming the base catalyzed mechanism we predict in both
cases a likely formation of the (R)-products. In an analogous way,
the less stable conformers A0 and B0 feature a hydrogen bond be-
tween the quinine OH and the pro-R carboxyl group. This leads to

ꢀ ꢁ
T. Ivsic et al. / Tetrahedron 68 (2012) 8311e8317
8315
Fig. 4. The structures of the lowest energy conformer (A) featuring pro-R chirality and the corresponding conformer A0 leading to the (S)-product as obtained with the B3LYP-D3/6-
31G(d,p) method. Sterical interactions governing the stability are marked above.
Table 5
Relative energies, predicted chirality and relevant geometry parameters of the four
lowest energy structures upon addition of acid as computed at the B3LYP/6-
31þG(d,p) level
E/kcal molꢀ1
Predict. config.
Distance/A
ꢂ
Conf.
(O,N)a
(O,O)b
(O,O)c
(O,C)d
C0
C
0.00
2.23
1.10
2.17
S
R
R
S
2.61
2.63
2.58
2.58
2.83
2.83
2.78
2.79
2.74
2.71
2.65
2.66
3.45
3.67
4.98
5.19
D
D0
a
H-bond distance between acetic acid OeH and quinuclidine N.
H-bond distance between quinine OH and the closest carbonyl O.
H-bond distance between methanol OH and acetate C]O.
b
c
d
Distance between methanol O and anhydride non-hydrogen bonded carbonyl C.
Scheme 3. Numeration of atoms in model anhydride.
complexes encompassing quinine, methanol, 3-methylglutaric an-
hydride, and acetic acid was initially explored at the B3LYP/6-
31G(d) level. Several low energy conformers were located. After
inverting the stereochemical face of the anhydride, the eight lowest
energy structures were reoptimized by using the empirical, dis-
persion corrected B3LYP-D3 method with the 6-31G(d,p) basis
set.23 The quinine catalyzed opening of meso-cyclic anhydrides is
governed by medium-strong hydrogen bonding interactions.24,25
However, we found that the geometries and energetics of the
complexes are sensitive to the description of the dispersion in-
teractions. In general, the B3LYP-D3 method yields more compact
structures then B3LYP, thus enhancing the energy difference be-
tween enantiomers. The calculations were performed by using the
Gaussian 0326 and Turbomole27 program packages.
quinine catalyzed reactions the alcohol was directly H-bonded to
the quinuclidine nitrogen. Upon acid addition it was found that the
acid intercalates between the alcohol and the quinuclidine nitro-
gen. The net result is that the alcohol approaches the anhydride
from the opposite face, in agreement with the observed inversion of
enantioselectivity.
4. Experimental section
4.1. Computational details
The calculations were performed using density functional the-
ory (DFT). The conformational landscape of the molecular

ꢀ ꢁ
T. Ivsic et al. / Tetrahedron 68 (2012) 8311e8317
8316
Fig. 5. The structures of the lowest energy conformers of the reaction complexes formed between acetic acid, quinine, methanol, and 3-methylglutaric anhydride. C0 and C feature
pro-S and pro-R chiralities, respectively, with C0 being 2.23 kcal/mol more stable than C. The structure of the second most stable conformer (D) featuring pro-R chirality and the
corresponding conformer D0 leading to the (S)-product is also given. Note that C0 is 1.10 kcal/mol more stable than D.
4.2. General methods
4.4. Method B
All reactions were conducted under an argon atmosphere. An-
hydride 2 was prepared according to literature.28All other reagents
and solvents were purchased from commercial sources and used
without purification. 1H and 13C NMR were recorded on Bruker AV
300 spectrometer. Chemical shifts (dH and dC) are quoted in parts
per million (ppm), referenced to TMS. High-resolution mass spec-
trometry (HRMS) was performed on 4800 Plus MALDI TOF/TOF
Analyzer. Optical rotations were measured using Optical Activity
AA-10 automatic polarimeter. Melting points were determined on
Electrothermal 9100 apparatus in open capillaries and are not
corrected. For the chemical purity determination, and monitoring
of the progress of the reactions Nucleosil 100-5 C18 column was
used (50% methanol in water with 0.5% H3PO4 to 100% methanol).
ees of monoesters 4 and 6 were determined on Chiralpak AS col-
umn (hexane/EtOH/TFA¼98:2:0.1) and Chiralpak OD column
(hexane/EtOH/TFA¼95:5:0.1), respectively.
To the cold 0.1 M toluene solution of anhydride (10 mmol), al-
kaloid (1.1 equiv) and alcohol (1.5 equiv) were added. The reaction
mixture was stirred until >90% conversion was reached (see Table
3) and the reaction was stopped by the addition of 5% HCl. The
organic layer was washed once more with 5% HCl and evaporated.
Oily residue was dissolved in 2% K2CO3 and washed successively
with EtOAc. Aqueous solution was acidified with HCl to pH 2 and
extracted with EtOAc. The organic extracts were dried over Na2SO4
and evaporated in vacuo.
4.5. Compounds characterization
4.5.1. (R)-3-Isobutyl-pentanedioic
acid
monobenzyl
ester
(4). Yellowish oil; [
a
]
26 ꢀ1.9 (c 3.166, EtOH) for 73% ee; nmax (KBr)¼
D
3600e2800 (br), 3035, 2960, 2870, 1738, 1707, 1456, 1388, 1236,
1163, 753, 698 cmꢀ1 1H NMR (300 MHz, CDCl3, 25 ꢁC, TMS):
;
d
¼9.88 (br s, 1H), 7.28e7.42 (m, 5H), 5.08e5.14 (s, 2H), 2.35e2.45
(m, 5H), 1.52e1.68 (m, 1H), 1.16e1.24 (m, 2H), 0.82e0.92 (dd, 6H,
13
4.3. Method A
J1¼6.7 Hz, J2¼3.2 Hz) ppm; C NMR (75 MHz, CDCl3, 25 ꢁC, TMS):
d
¼178.6, 172.4, 135.9, 128.6, 128.3, 128.2, 66.3, 43.3, 38.6, 38.4, 29.8,
To the 0.1 M toluene solution of anhydride (10 mmol), alkaloid
(0.1 equiv), xanthene-9-carboxylic acid (0.2 equiv), and alcohol
(1.5 equiv) were added. The reaction mixture was stirred until
>90% conversion was reached (see Table 3) and the reaction was
stopped by the addition of 5% HCl. The organic layer was washed
once more with 5% HCl and evaporated. The oily residue was dis-
solved in 2% K2CO3 and washed successively with EtOAc. The
aqueous solution was then carefully acidified with H3PO4 to pH 5.4
and extracted with toluene. The organic extracts were dried over
Na2SO4 and evaporated in vacuo.
25.2, 22.5, 22.5 ppm; HRMS (MALDI): MKþ, found 317.1145.
C16H22O4 requires 317.1149.
4.5.2. (R)-3-(4-Chloro-phenyl)-pentanedioic acid monobenzyl ester
25
(6). White powder; mp 143.2e144.8 ꢁC; [
a]
ꢀ4.6 (c 1.192, EtOH)
D
for 77% ee; nmax (KBr)¼3600e2800 (br), 2940, 1733, 1719, 1675,
1493, 1415, 1232, 1152, 825, 733, 692 cmꢀ1 1H NMR (300 MHz,
CDCl3, 25 ꢁC, TMS):
¼10.60 (br s, 1H), 7.08e7.39 (m, 9H), 4.95e5.12
(s, 2H), 3.55e3.70 (m, 1H), 2.55e2.85 (m, 4H) ppm; 13C NMR
(75 MHz, CDCl3, 25 ꢁC, TMS):
¼177.0, 171.1, 140.5, 135.5, 132.9,
;
d
d

ꢀ ꢁ
T. Ivsic et al. / Tetrahedron 68 (2012) 8311e8317
8317
14. Park, S. E.; Nam, E. H.; Jang, H. B.; Oh, J. S.; Some, S.; Lee, Y. S.; Song, C. E. Adv.
Synth. Catal. 2010, 352, 2211e2217.
15. Ivsic, T.; Hamersak, Z. Tetrahedron: Asymmetry 2009, 20, 1095e1098.
16. Bigi, F.; Carloni, S.; Maggi, R.; Mazzacani, A.; Sartori, G.; Tanzi, G. J. Mol. Catal., A:
Chem. 2002, 182, 533e539.
128.8, 128.7, 128.5, 128.3, 128.3, 66.5, 40.5, 40.0, 37.5 ppm; HRMS
(MALDI): MNaþ found 355.0707. C18H17O4Cl requires 355.0707.
ꢀ ꢁ
ꢀ
Acknowledgements
17. Dedeoglu, B.; Catak, S.; Houk, K. N.; Aviyente, V. ChemCatChem 2010, 2,
1122e1129.
18. Schmitt, E.; Schiffers, I.; Bolm, C. Tetrahedron 2010, 66, 6349e6357.
19. Oh, S. H.; Rho, H. S.; Lee, J. W.; Lee, J. E.; Youk, S. H.; Chin, J.; Song, C. E. Angew.
Chem., Int. Ed. 2008, 47, 7872e7875.
20. Rho, H. S.; Oh, S. H.; Lee, J. W.; Lee, J. Y.; Chin, J.; Song, C. E. Chem. Commun.
2008, 1208e1210.
We thank the Ministry of Science, Education and Sports of the
Republic of Croatia for financing (grant nos. 098-0982933-2908 and
098-0352851-2921).
21. Caner, H.; Biedermann, P. U.; Agranat, I. Chirality 2003, 15, 637e645.
22. (a) Burgi, T.; Baiker, A. J. Am. Chem. Soc. 1998, 120, 12920e12926; (b) Meier, D.
Supplementary data
€
ꢃ
€
M.; Urakawa, A.; Turra, N.; Ruegger, H.; Baiker, A. J. Phys. Chem. A 2008, 112,
€
6150e6158; (c) Urakawa, A.; Meier, D. M.; Ruegger, H.; Baiker, A. J. Phys. Chem. A
Supplementary data associated with this article can be found in
the online version, at http://dx.doi.org/10.1016/j.tet.2012.07.035.
These data include MOL files and InChiKeys of the most important
compounds described in this article.
2008, 112, 7250e7255; (d) Balzano, F.; Jumde, R. P.; Mandoli, A.; Masi, S.; Pini,
D.; Uccello-Barretta, G. Chirality 2011, 23, 784e795.
23. (a) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. J. Chem. Phys. 2010, 132,
154104e154123; (b) Grimme, S.; Ehrlich, S.; Goerigk, L. J. Comput. Chem. 2011,
32, 1456e1465; (c) Hujo, W.; Grimme, S. Phys. Chem. Chem. Phys. 2011, 13,
13942e13950.
ꢀ
24. (a) Hadzi, D. Theoretical Treatments of Hydrogen Bonding; John Wiley Sons:
References and notes
ꢁ
Chichester, UK, New York, NY, 1997; (b) Giese, K.; Petkovic, M.; Naundorf, H.;
€
Kuhn, O. Phys. Rep. 2006, 430, 211e276.
1. (a) Rodriguez-Docampo, Z.; Connon, S. J. ChemCatChem 2012, 4, 151e168; (b)
Atodiresei, I.; Schiffers, I.; Bolm, C. Chem. Rev. 2007, 107, 5683e5712; (c) Chen,
Y.; McDaid, P.; Deng, L. Chem. Rev. 2003, 103, 2965e2984.
ꢁ
ꢀ ꢁ
ꢁ
25. (a) Matanovic, I.; Doslic, N.; Mihalic, Z. Chem. Phys. 2004, 306, 201e207; (b)
ꢁ
ꢀ ꢁ
Petkovic, M.; Novak, J.; Doslic, N. Chem. Phys. Lett. 2009, 474, 248e252.
26. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.;
Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.;
Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.;
Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao,
O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken,
V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A.
J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G.
A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.;
Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Fores-
man, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov,
B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.;
Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P.
M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian
03, Revision E.01; Gaussian, Inc.: Pittsburg PA, 2003.
2. Starr, J. T.; Koch, G.; Carreira, E. M. J. Am. Chem. Soc. 2000, 122, 8793e8794.
€
3. Tanyeli, C.; Ozc¸ ubukc¸ u, S. Tetrahedron: Asymmetry 2003, 14, 1167e1170.
4. Mittendorf, J.; Benet-Buchholz, J.; Fey, P.; Mohrs, K. H. Synthesis 2003, 1,
136e140.
5. Keen, S. P.; Cowden, C. J.; Bishop, B. C.; Brands, K. M. J.; Davies, A. J.; Dolling, U.
H.; Lieberman, D. R.; Stewart, G. W. J. Org. Chem. 2005, 70, 1771e1779.
ꢀ
ꢁ
ꢁ
6. (a) Hamersak, Z.; Stipetic, I.; Avdagic, A. Tetrahedron: Asymmetry 2007, 18,
1481e1485; (b) Peschiulli, A.; Procuranti, B.; O’Connor, C. J.; Connon, S. J. Nat.
Chem. 2010, 2, 380e384.
7. Huang, J.; Xiong, F.; Chen, F. E. Tetrahedron: Asymmetry 2008, 19, 1436e1443.
8. Ji, L.; Ma, Y.; Li, J.; Zhang, L.; Zhang, L. Tetrahedron Lett. 2009, 50, 6166e6168.
9. Hiratake, J.; Inagaki, M.; Yamamoto, Y.; Oda, J. J. Chem. Soc., Perkin Trans. 1 1987,
1053e1058.
10. Aitken, R. A.; Gopal, J.; Hirst, J. A. J. Chem. Soc., Chem. Commun. 1988, 632e634.
11. Bolm, C.; Schiffers, I.; Atodiresei, I.; Hackenberger, C. P. R. Tetrahedron Asym-
metry 2003, 14, 3455e3467.
€
€
€
27. Ahlrichs, R.; Bar, M.; Haser, M.; Horn, H.; Kolmel, C. Chem. Phys. Lett. 1989, 162,
165e169.
12. Chen, Y.; Tian, S.-K.; Deng, L. J. Am. Chem. Soc. 2000, 122, 9542e9543.
13. Peschiulli, A.; Gun’k, Y.; Connon, S. J. J. Org. Chem. 2008, 73, 2454e2457.
28. Smith, W. T.; Kort, P. G. J. Am. Chem. Soc. 1950, 72, 1877e1878.