Steric versus electronic effects of the ligand in the enantioselective palladium-catalyzed allylic alkylation with chiral oxazolinylpyridines

Article abstract of DOI:10.1016/S0957-4166(99)00368-7

Chiral oxazolinylpyridines bearing an oxazolinyl [bis(oxazolinyl)pyridines] or a cyano group in the 6-position of the pyridine ring were prepared and assessed in the enantioselective palladium-catalyzed allylic substitution of 1,3-diphenylprop-2-enyl acet

Full text of DOI:10.1016/S0957-4166(99)00368-7

Pergamon
Tetrahedron: Asymmetry 10 (1999) 3803–3809
Steric versus electronic effects of the ligand in the
enantioselective palladium-catalyzed allylic alkylation with chiral
oxazolinylpyridines
Giorgio Chelucci,a,∗ Sebastiano Deriu,a Gerard A. Pinna,b Antonio Saba a and
Raffaela Valenti a
aDipartimento di Chimica, Università di Sassari, via Vienna 2, I-07100 Sassari, Italy
bDipartimento Farmaco Chimico Tossicologico, Università di Sassari, Via Muroni 23, I-07100 Sassari, Italy
Received 30 July 1999; accepted 17 August 1999
Abstract
Chiral oxazolinylpyridines bearing an oxazolinyl [bis(oxazolinyl)pyridines] or a cyano group in the 6-position
of the pyridine ring were prepared and assessed in the enantioselective palladium-catalyzed allylic substitution of
1,3-diphenylprop-2-enylacetate with dimethyl malonate. The electronic and steric properties of substituents hardly
affected either the catalytic activity or the enantioselectivity of the substitution reaction. Enantioselectivities up to
94% were obtained. © 1999 Elsevier Science Ltd. All rights reserved.
1. Introduction
Oxazolinylpyridine derivatives have been shown to induce high stereoselectivity in the enantioselective
palladium-catalyzed allylic substitutions.1 Thus, in the alkylation of 1,3-diphenylprop-2-enyl acetate with
dimethyl malonate, very high enantiomeric excesses have been obtained using oxazolinylpyridines with
two stereocentres.2 Recently, a number of chiral oxazolinylpyridines 1 with a single stereocentre having
different substituents on the pyridine and oxazoline rings have been prepared and assessed in this process
(Scheme 1: R,R0,R00=alkyl, aryl; R000=electron-donating or withdrawing groups) and their effects on
the catalytic activity and stereoselectivity (steric control and electronic control) have been disclosed.3
Moreover, we have more recently attempted to obtain a ring-size control in this reaction, using the
modified ligands 2 and 3.4
Concerning the electronic control of the reaction, we synthesized, in order to differentiate the electronic
effects from steric ones, oxazolinylpyridines in which the electronic differentiating substituents were
in the 4-position of the pyridine ring. The results of this study showed that the catalytic activity was
∗
Corresponding author. E-mail: chelucci@ssmain.uniss.it
0957-4166/99/$ - see front matter © 1999 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(99)00368-7

3804
G. Chelucci et al. / Tetrahedron: Asymmetry 10 (1999) 3803–3809
Scheme 1.
dramatically affected by the variation of the electronic property of the ligand but not the stereoselectivity,
which depends mainly on steric effects.
Pursuing our work in this field,3,4 we have been evaluating the potential in this process of oxazo-
linylpyridines bearing an electronic differentiating substituent on the 6-position of the pyridine ring,
taking into account that in this case both the electronic and steric properties of the substituent should be
considered. The oxazolinyl and the cyano groups were selected for this purpose. In the former case we
dealt with bis(oxazolinyl)pyridines, ligands extensively used for asymmetric reactions,5 but whose use
for enantioselective palladium-catalyzed allylic substitution has not been described.
We report here the synthesis of 2-cyano-6-(oxazolinyl)pyridines 5, a new direct way to obtain
bis(oxazolinyl)pyridines 6 and the results obtained with these ligands in the palladium-catalyzed allylic
alkylation of 1,3-diphenylprop-2-enyl acetate with dimethyl malonate.
2. Results and discussion
2.1. Synthesis of ligands
2-Cyano-6-(oxazolinyl)pyridines 5 were prepared by heating under reflux a chlorobenzene solution of
2,6-dicyanopyridine 4 with the appropriate aminoalcohol in the presence of a catalytic amount of zinc
chloride.6 To obtain the preferential formation of mono-oxazolines 5, a 1:1 molar ratio of 4 and the
appropriate aminoalcohol was used (Scheme 2). Notwithstanding this, ligands 5 were obtained in low
yield (26–44% yield) because of the formation of a relevant amount of the compounds of bis-anellation
6 (20–30% yield). Instead, the formation of the bis(oxazolinyl)pyridines 6 was exclusive if a 1:3 molar
ratio between 4 and the aminoalcohol was used. The bis(oxazolinyl)pyridine 6a was obtained in 60%
yield when 4 was allowed to react with (S)-(+)-2-amino-3-methyl-1-butanol. Therefore, this method
opens a direct access to this very important class of ligands which up to now have been obtained through
long reaction sequences.7 Moreover, it is important to note that this route appears particularly attractive
because 4 is easily accessible from a variety of pyridine derivatives.8
Scheme 2.

G. Chelucci et al. / Tetrahedron: Asymmetry 10 (1999) 3803–3809
3805
2.2. Palladium-catalyzed allylic alkylation
Allylic substitutions were carried out employing Trost’s procedure, which demands the
use of [Pd(η3–C3H5)Cl]2 as the procatalyst and a mixture of dimethyl malonate, N,O-
bis(trimethylsilyl)acetamide (BSA) and potassium acetate in methylene chloride at room or reflux
temperature.9
The results obtained in a set of experiments with the new ligands are summarized in Table 1, in which
the results previously obtained with the related ligands 1a–c bearing the methyl instead of the cyano
group (Scheme 1: 1a: R=Me, R00=i-Pr, R0=R000=H2; 1b: R=Me, R000=Ph, R0=R00=H; 1c: R=Me, R00=t-
Bu, R0=R000=H) are reported too. According to Table 1 the following considerations can be made:
(i) The bis(oxazolinyl)pyridine 6a, which was selected to determine the reactivity of this class of
ligands, provided a very ineffective palladium catalyst. Thus, no reaction was obtained after 7
days at room temperature (entry 8) whereas 80% of the starting material was converted after 2
days (entry 9) at reflux. This catalytic outcome could be rationalized considering that this ligand
behaves as the analogue 2,20:60,200-terpyridine which forms allyl palladium(II) complexes that in
solution are present in a dynamic equilibrium between η1- and η3-allyl isomeric forms.10 In the
former case the terpyridine behaves as a terdentate ligand, whereas in the latter it coordinates the
palladium atom in a bidentate fashion. Likewise, ligand 6a could form η1- and η3-allyl palladium
complexes of which the former, catalytically unreactive, is prevailing at room temperature, whereas
the latter, catalytically reactive, is slowly formed at higher temperatures. In this case, ligand 6a
forms, on account of its C2-symmetry, only one palladium η3-allyl complex where the central
pyridine coordinates the palladium with either of the oxazoline nitrogens. The low enantioselectivity
(26%) obtained in this process indicates that the steric and electronic effects of the substitution are
detrimental to the stereoselectivity of the reaction. On this basis the use of this kind of ligand was
abandoned.
(ii) A dramatic effect on the catalytic activity is observed by the presence in the pyridine ring of a cyano
group. This electron-withdrawing substituent decreases the reaction rate with respect to the related
6-methyl substituted oxazolinylpyridines 1a–c (entries 2, 5 and 7 versus 1, 4 and 6). The results
obtained can be tentatively explained considering the catalytic cycle of the reaction which begins
by the oxidative addition of the allylic substrate to the palladium(0) catalyst. The rate of this step
is dependent on the nature of the ligand and, in general, decreases with reducing basicity (ability to
donate electrons to the metal) of the ligand, which decreases the electronic density on the palladium
and so its nucleophilic ability. These considerations are consistent with the observation that the
cyano group drastically decreases the reaction rate. However, the rate of the oxidative addition step
can be decreased not only by reducing the electronic density on the palladium but also by improving
the steric requirement of the ligand, as we observed previously when the steric demands of both the
substituent on the oxazoline and the pyridine rings were increased.3a Though both these factors
should be considered we prefer the former explanation because the steric demands of the cyano and
the methyl groups are not so different to justify a dramatic change in the reactivity.
(iii) In contrast, the enantioselectivity of the substitution reaction appears to be favourably affected by
the presence of the cyano group. Thus, an improved stereoselectivity has always been obtained
with ligands 5a–c with respect to the related 1a–c. The best enantioselectivity (94% ee) has been
obtained with the ligand 5c bearing the tert-butyl group on the oxazoline moiety. The stereochemical
result can be tentatively explained by considering the related π-allyl palladium complexes. The
accepted mechanism for palladium-catalyzed allylic substitutions, which proceed through a meso
η3-allyl intermediate, foresees that the nucleophile attacks the allylic termini of two alternative

3806
G. Chelucci et al. / Tetrahedron: Asymmetry 10 (1999) 3803–3809
Table 1
Allylic alkylation of 1,3-diphenylprop-2-enyl acetate with dimethyl malonatea
diastereomeric π-allyl palladium complexes which interconvert through a π–σ–π mechanism and
which are present at the equilibrium in a different ratio (for instance 9 and 10 for ligands with (S)
configuration: Scheme 3). Because both ligands 1a–c and 5a–c give the substitution product 8 with
the same prevailing configuration, it is reasonable to assume that the reactive transition states are
always the same.
Scheme 3.
Since it has usually been observed that the prevailing reaction product comes from a nucleophilic attack
on one terminus of the allylic unit of the most abundant complex11 (in this case the allylic terminus trans
to the oxazoline nitrogen in the more stable diastereomer 9), the ability of the ligand to stabilize one of
the two diasteromeric complexes is a crucial point in determining the stereochemical outcome. Thus, the
better stereochemical results obtained with ligands 5a–c with respect to ligands 1a–c seem to indicate
that the cyano group is able to better stabilize the reactive diasteromeric complex.

G. Chelucci et al. / Tetrahedron: Asymmetry 10 (1999) 3803–3809
3807
3. Conclusions
In this paper the catalytic activity in the palladium-catalyzed allylic substitution of 1,3-diphenylprop-
2-enyl acetate with dimethyl malonate using chiral oxazolinylpyridines bearing, on the 6-position of
the pyridine ring, a substituent which has both electronic and steric properties, is reported. The results
obtained show that bis(oxazolinyl)pyridines are unable to give a catalytic active palladium catalyst. In
contrast, an improved stereoselectivity is obtained by the cyano group, whose favourable steric demand
overcomes its electron withdrawing property which reduces the catalytic activity. An interesting and
direct way to obtain bis(oxazolinyl)pyridines, a very important class of ligands, is also reported.
4. Experimental
4.1. General methods
Boiling points are uncorrected. Melting points were determined on a Buchi 510 capillary apparatus and
are uncorrected. The 1H NMR (300 MHz) spectra were obtained with a Varian VXR-300 spectrometer.
Optical rotations were measured with a Perkin–Elmer 241 polarimeter in a 1 dm tube. Elemental
analyses were performed on a Perkin–Elmer 240 B analyser. (R)-(−)-2-Amino-2-phenylethanol, (S)-
(+)-2-amino-3,3-dimethyl-1-butanol and (S)-(+)-2-amino-3-methyl-1-butanol were commercial products
(Aldrich A.G.). 2,6-Dicyanopyridine 4 was prepared following the literature procedure.8
4.2. General procedure for the preparation of oxazolinylmethylpyridines 5
In a 25 ml two-necked flask, zinc chloride (14 mg, 0.10 mmol) was melted under high vacuum and
cooled under argon. After cooling to room temperature, anhydrous chlorobenzene (10 ml) was added
followed by the 2,6-dicyanopyridine 4 (2 mmol) and the aminoalcohol (2.0 mmol). The resulting mixture
was heated under reflux for 24 h and then the solvent was evaporated under reduced pressure. The residue
was dissolved in CH2Cl2 (10 ml) and the resulting solution was washed with water (3×4 ml). The
aqueous solution was extracted with CH2Cl2 (10 ml), the combined organic phases were dried over
anhydrous Na2SO4 and the solvent evaporated. The residue was purified by chromatography on a silica
gel column.
4.3. (S)-2-Cyano-6-[4,5-dihydro-4-(1-methylethyl)oxazol-2-yl]pyridine 5a
Chromatographic eluent: benzene:acetone=8:2; 0.135 g (31%); mp 154°C; [α]25 −58.7 (c 1.1,
D
1
CHCl3); H NMR (CDCl3) δ: 8.31 (d, 1H, J=7.8 Hz), 7.95 (t, 1H, J=7.8 Hz), 7.80 (d, 1H, J=7.8 Hz),
4.56 (t, 1H, J=9.0 Hz), 4.30–4.12 (m, 2H), 1.97–1.86 (m, 1H), 1.06 (d, 3H, J=6.6), 0.96 (d, 3H, J=6.6
Hz). Anal. calcd for C12H13N3O: C, 66.96; H, 6.09; N, 19.52. Found: C, 66.86; H, 6.19; N, 19.32.
4.4. (R)-2-Cyano-6-(4,5-dihydro-4-phenyloxazol-2-yl)pyridine 5b
Chromatographic eluent: petroleum ether:ethyl acetate=1:2; 0.502 g (26%); mp 121–122°C; [α]25
D
+123.5 (c 1.4, CHCl3); 1H NMR (CDCl3) δ: 8.39 (dd, 1H, J=7.8, J=1.0 Hz), 7.98 (t, 1H, J=7.8 Hz), 7.84
(dd, 1H, J=7.8, 1.0 Hz), 7.40–7.29 (m, 5H), 5.48 (dd, 1H, J=10.3, 8.6 Hz), 4.94 (dd, 1H, J=10.3, J=9.0

3808
G. Chelucci et al. / Tetrahedron: Asymmetry 10 (1999) 3803–3809
Hz), 4.44 (t, 1H, J=8.6 Hz). Anal. calcd for C15H11N3O: C, 72.28; H, 4.45; N, 16.86. Found: C, 72.18;
H, 4.55; N, 16.77.
4.5. (S)-2-Cyano-6-[4,5-dihydro-4-(1,1-dimethylethyl)oxazol-2-yl]pyridine 5c
Chromatographic eluent: benzene:acetone=8:2; 0.202 g (44%); mp 154–156°C; [α]25 −86.1 (c 1.0,
D
CHCl3); 1H NMR (CDCl3) δ: 8.35 (dd, 1H, J=7.8, 0.9 Hz), 7.95 (t, 1H, J=7.8 Hz), 7.81 (dd, 1H, J=7.8,
0.9 Hz), 4.50 (dd, 1H, J=10.5, 8.7 Hz), 4.36 (t, 1H, J=8.7 Hz), 4.15 (dd, 1H, J=10.5, 8.7 Hz), 0.98 (s,
9H). Anal. calcd for C13H15N3O: C, 68.10; H, 6.59; N, 18.33. Found: C, 68.15; H, 6.71; N, 18.35.
4.6. 2,6-Bis[(4S)-(+)-isopropyl-2-oxazolin-2-yl]pyridine 6a
The procedure used for the synthesis of 5 was followed. In this case a 1:3 ratio of 2,6-dicyanopyridine
(4) (2.0 mmol) and of (S)-(+)-2-amino-3-methyl-1-butanol (6.0 mmol) was used. After usual work-up
the residue was purified by flash chromatography (ethyl acetate) to give 6a (0.36 g, 60% yield) whose
spectroscopic data were consistent with an authentic sample (Aldrich A.G.).
4.7. Allylic alkylation of 1,3-diphenyl-2-propenyl acetate with dimethyl malonate: general procedure
A solution of ligand (0.04 mmol, 10 mol%) and [{Pd(η3–C3H5)Cl}2] (4 mg, 2.5 mol%) in dry CH2Cl2
(2 ml) was stirred at room temperature for 15 min. This solution was treated successively with a solution
of rac-(E)-1,3-diphenyl-2-propenyl acetate (0.4 mmol) in CH2Cl2 (1 ml), dimethyl malonate (1.2 mmol),
N,O-bis(trimethylsilyl)acetamide (1.2 mmol) and anhydrous potassium acetate (3.5 mol%). The reaction
mixture was stirred for the appropriate time (see Table 1) until conversion was complete as shown by TLC
analysis (light petroleum:ether=3:1). The reaction mixture was diluted with ether (25 ml), washed with
ice-cold saturated aqueous ammonium chloride. The organic phase was dried (Na2SO4) and concentrated
under reduced pressure. The residue was purified by flash chromatography (light petroleum:ether=3:1)
to afford dimethyl 1,3-diphenylprop-2-enyl malonate. The enantiomeric excess was determined from the
1H NMR spectrum in the presence of enantiomerically pure shift reagent Eu(hfc)3; splitting of the signals
for one of the two methoxy groups was observed.
Acknowledgements
This work was supported by financial assistance from the Ministero dell’Università e della Ricerca
Scientifica e Tecnologica (M.U.R.S.T.-Roma).
References
1. For reviews, see: Trost, M. B.; Van Vranken, D. L. Chem. Rev. 1996, 96, 395. Hayashi, T. In Catalytic Asymmetric
Synthesis; Ojima, I., Ed.; VCH: Weinheim, 1993. Frost, C. G.; Howarth, J.; Williams, J. M. J. Tetrahedron: Asymmetry
1992, 3, 1089. For more recent references on palladium-catalyzed allylic alkylation reactions, see: Koning, B.; Meetsma,
A.; Kellogg, R. M. J. Org. Chem. 1998, 63, 1604.
2. Chelucci, G. Tetrahedron: Asymmetry 1997, 8, 2667. Nordström, K.; Macedo, E.; Moberg, C. J. Org. Chem. 1997, 62,
1604.
3. (a) Chelucci, G.; Medici, S.; Saba, A. Tetrahedron: Asymmetry 1999, 10, 543. (b) Chelucci, G.; Deriu, S.; Saba, A.; Valenti,
R. Tetrahedron: Asymmetry 1999, 10, 1457. (c) Brenberg, U.; Rahn, F.; Moberg, C. Tetrahedron: Asymmetry 1998, 9, 3437.

G. Chelucci et al. / Tetrahedron: Asymmetry 10 (1999) 3803–3809
3809
4. Chelucci, G.; Gladiali, S.; Saba, A. Tetrahedron: Asymmetry 1999, 10, 1393.
5. Nishiyama, H.; Soeda, N.; Naito, T.; Motoyama, Y. Tetrahedron: Asymmetry 1998, 9, 2865, and references cited therein.
6. Bolm, C.; Veickhardt, K.; Zehnder, M.; Ranff, T. Chem. Ber. 1991, 124, 1173.
7. Nishiyama, H.; Kondo, M.; Nakamura, T.; Itoh, K. Organometallics 1991, 10, 500.
8. The dinitrile 4 is easily accessible from a variety of pyridine derivatives: (a) from 2,6-dibromopyridine by cyanation
with CuCN (44%) [Pearse, G. A.; Wisowaty, J. J. Heterocyclic Chem. 1973, 10, 647]; (b) from the N-oxide of
2-pyridinecarbonitrile by cyanation with KCN (84%) [Tani, H. Chem. Pharm. Bull. 1959, 7, 930]; (c) from 2,6-
pyridinedicarboxamide by dehydration (52%) [Bancks, R.; Brookes, R. Chem. Ind. (London) 1974, 15, 617]; (d) from
2,6-dimethylpyridine by oxidative ammonolysis (75%) [Glubokovskikh, L. K.; Kan, I. I.; Suvorov, B. V.; Berstenev, S. V.
Izv. Akad. Nauk. Kaz. SSR, Ser. Khim. 1985, 3, 90; C.A. 1986, 104, 88395g].
9. Trost, B. M.; Murphy, D. J. Organometallics 1985, 4, 1143.
10. Ramdeehul, S.; Barloy, L.; Osborn., J. A.; De Cian, A.; Fischer, J. Organometallics 1986, 15, 5442.
11. Blöchl, P. E.; Togni, A. Organometallics 1996, 15, 4125.