Structurally simple pyridine N-oxides as efficient organocatalysts for the enantioselective allylation of aromatic aldehydes

Article abstract of DOI:10.1021/jo052132m

A series of structurally simple pyridine N-oxides have readily been assembled from inexpensive amino acids and tested as organocatalysts in the allylation of aldehydes with allyl(trichloro)silane to afford homoallylic alcohols. (S)-Proline-based catalysts

Full text of DOI:10.1021/jo052132m

Structurally Simple Pyridine N-Oxides as Efficient Organocatalysts
for the Enantioselective Allylation of Aromatic Aldehydes
Luca Pignataro, Maurizio Benaglia,* Rita Annunziata, Mauro Cinquini, and Franco Cozzi
Dipartimento di Chimica Organica e Industriale, UniVersita’ degli Studi di Milano,
Via Golgi 19, I-20133 Milano, Italy
maurizio.benaglia@unimi.it
ReceiVed October 12, 2005
A series of structurally simple pyridine N-oxides have readily been assembled from inexpensive amino
acids and tested as organocatalysts in the allylation of aldehydes with allyl(trichloro)silane to afford
homoallylic alcohols. (S)-Proline-based catalysts afforded the products derived from aromatic aldehydes
in fair to good yields and in up to 84% enantiomeric excess (ee). The allylation of heteroaromatic,
unsaturated, and aliphatic aldehydes was less satisfactory. By running the reaction in the presence of
achiral and chiral additives and structurally different catalysts, we collected some insights into the
relationship between the stereochemical outcome and the catalyst’s structural features. Even if the ee’s
obtained are inferior to the best values observed with other catalysts, this work concurs to show that
structurally simple pyridine N-oxides can also promote the allylation reaction with satisfactory stereocontrol.
Introduction
part of a project devoted to the identification of new organo-
catalysts that fulfill the above-mentioned requirements, we
became interested in developing structurally simple N-oxides
as chiral Lewis base catalysts.1,3,4 Inspection of literature data5,6
led us to consider mono-N-oxides as promising candidates.5e-h,7
Mono-N-oxides are more easily obtained than bis-5a-d and tris-
N-oxides,5j the synthesis of which requires relatively long
procedures often involving a resolution process or a stereose-
lective synthesis as the key step.
In recent years, appropriately defined “the golden age of
organocatalysis”,1 the possibility of effectively replacing metal-
based enantioselective catalysts with wholly organic molecules
while maintaining high levels of chemical efficiency and
stereocontrol has been demonstrated. A key issue for the success
of this approach resides in the structural simplicity of the
organocatalyst2 that should be much more readily available than
its organometallic counterparts.
On the basis of some encouraging literature precedents5g,5k,7
and considering synthetic simplicity as a major goal of our effort,
it was also decided to avoid the presence of a stereogenic axis
but to maintain a biaryl bond in the proximity of the catalyst’s
active site. Indeed, embedding a stereogenic axis in the catalyst
An ideal organocatalyst should be a molecule present in the
chiral pool: proline and quinine are suitable examples. If a
nonnatural catalyst is required, it must be obtained by modifica-
tion of an inexpensive, commercially available, enantiopure
material whose manipulation must be kept to minimum. As a
(3) For leading references to the use of Lewis basic organocatalysts,
see: (a) Denmark, S. E.; Fu, J. J. Am. Chem. Soc. 2003, 167 and references
therein. (b) Rowlands, G. J.; Kentish Barnes, W. Chem. Commun. 2003,
2712. (c) Kobayashi, S.; Suigira, M.; Ogawa, C. AdV. Synth. Catal. 2004,
346, 1023.
(4) For a review describing the use of N-oxides as chiral metal ligands
in asymmetric catalysis, see: Chelucci, G.; Murineddu, G.; Pinna, G. A.
Tetrahedron: Asymmetry 2004, 15, 1373.
* Phone: +39.02.50314171. Fax: +39.02.50314159.
(1) (a) Dalko, P. I.; Moisan, L. Angew. Chem., Int. Ed. 2004, 43, 5138.
(b) Berkessel, A.; Gro¨ger, H. Asymmetric Organocatalysis; Wiley-VCH:
Weinheim, Germany, 2005.
(2) For a definition of organic catalyst, see: Benaglia, M.; Puglisi, A.;
Cozzi, F. Chem. ReV. 2003, 103, 3401.
10.1021/jo052132m CCC: $33.50 © 2006 American Chemical Society
Published on Web 01/14/2006
1458
J. Org. Chem. 2006, 71, 1458-1463

Pyridine N-Oxides as Efficient Organocatalysts
SCHEME 1. Synthesis of Catalysts 3b-7ba
has often proved to be a winning strategy in achieving high
levels of stereocontrol,5a-d,5f but it also made the synthesis of
the catalyst quite complex, calling for the stereoselective
formation of a difficult-to-control stereogenic element. Here,
we describe the facile synthesis of some structurally simple
amino-acid-derived pyridine N-oxides that catalyze the enan-
tioselective allylation of aromatic aldehydes with allyl(trichloro)-
silane8 in up to 84% enantiomeric excess (ee).
Results and Discussion
The catalysts were readily assembled as described in Scheme
1. A two-step procedure involving Suzuki coupling of 2,6-
dimethoxyphenyl boronic acid with the corresponding 2-bro-
mopyridines to afford 1a and 1b followed by demethylation
with pyridine hydrochloride allowed 2-(2,6-dihydroxyphenyl)-
pyridines 2a and 2b in 86 and 70% overall yields, respectively.
Condensation of 2a with N-Cbz-protected (S)-valine, (S)-
phenylalanine, and (S)-proline under standard conditions (HOBt,
EDC) afforded the corresponding bisesters 3a-5a in 50-60%
unoptimized yield. Proline-derived compound 5a (that eventually
led to the most stereoselective catalyst, see below) was better
obtained by esterification of 2a with N-Cbz prolinoyl chloride9
(97% average yield of several runs), a reaction that also allowed
the synthesis of adduct 7a from bisphenol 2b. N-Boc proline
derivative 6a was prepared in 76% yield by a condensation
carried out in the presence of DCC and catalytic DMAP.
Conversion of 3a-7a to the corresponding N-oxides 3b-7b
occurred uneventfully by reaction with m-CPBA in 75-85%
yield.
a Reagents and conditions: (i) 2,6-dimethoxyphenyl boronic acid, 0.1
mol equiv of Pd(OAc)2, 0.4 mol equiv of PPh3, 2 M aqueous Na2CO3,
DME, reflux, 70 h; (ii) excess Py‚HCl, neat, 200 °C, 3 h; (iii) 2 mol equiv
of N-Cbz-protected amino acid, 2.4 mol equiv each of HOBt and EDC,
CH2Cl2/DMF 1:1, room temperature, 15 h; (iv) 1.5 mol equiv of 70%
m-CPBA, CH2Cl2, room temperature, 15 h; (v) 2 mol equiv of N-Cbz
prolinoyl chloride, THF, reflux, 15 h; (vi) 2 mol equiv of N-Boc proline,
2.1 mol equiv of DCC, 0.2 mol equiv of DMAP, THF, room temperature,
15 h.
The catalytic activity of the N-oxides was established using
the allylation of benzaldehyde to afford homoallylic alcohol 8a
as the model reaction (Scheme 2). A typical experiment involved
the use of 0.1 mol equiv of catalyst, 1.2 mol equiv of allyl-
(trichloro)silane, and 3 mol equiv of DIPEA in acetonitrile10
for 48 h at different temperatures. Isolated yields and ee’s, as
determined by HPLC, are collected in Table 1; the (S) absolute
configuration was assigned to the predominant isomer of 8a by
comparison of optical rotation.
As can be seen from the reported data, catalyst 5b, derived
from N-Cbz proline, was more stereoselective than those
obtained from valine (3b) or phenylalanine (4b) (entry 3 vs 1
and 2). By lowering the reaction temperature to 0 °C, a
satisfactory 68% ee was obtained with catalyst 5b (entry 4),
although the reaction proceeded in moderate yield (45%). The
addition of tetrabutylammonium iodide did not improve the yield
and slightly depressed the ee (entry 5). The yield was increased
to 75% and the ee was slightly decreased to 63% by doubling
the amount of catalyst (entry 6). The use of 0.1 mol equiv of
catalyst 6b derived from N-Boc proline (entry 7) allowed us to
increase the yield to 87% while maintaining a good level of
stereocontrol (62% ee). The latter was increased to 67% by
further lowering the reaction temperature to -20 °C (entry 8).
The removal of the methyl substituent on the pyridine ring as
in catalyst 7b affected very slightly the outcome of the reaction
(entry 9 vs 4).
(5) For the use of N-oxides as organocatalysts in the allylation reaction,
see: (a) Nakajima, M.; Saito, M.; Shiro, M.; Hashimoto, S. J. Am. Chem.
Soc. 1998, 120, 6419. (b) Shimada, T.; Kina, A.; Ikeda, S.; Hayashi, T.
Org. Lett. 2002, 4, 2799. (c) Shimada, T.; Kina, A.; Hayashi, T. J. Org.
Chem. 2003, 68, 6329. (d) Kina, A.; Shimada, T.; Hayashi, T. AdV. Synth.
Catal. 2004, 346, 1169. (e) Malkov, A. V.; Orsini, M.; Pernazza, D.; Muir,
K. W.; Langer, V.; Meghani, P.; Kocovsky, P. Org. Lett. 2002, 4, 1049. (f)
Malkov, A. V.; Dufkova`; Farrugia, L.; Kocovsky, P. Angew. Chem., Int.
Ed. 2003, 42, 3674. (g) Malkov, A. V.; Bell, M.; Vassieu, M.; Bugatti, V.;
Kocovsky, P. J. Mol. Catal. A: Chemical 2003, 196, 179. (h) Malkov, A.
V.; Bell, M.; Orsini, M.; Pernazza, D.; Massa, A.; Herrmann, P.; Meghani,
P.; Kocovsky, P. J. Org. Chem. 2003, 68, 9659. (i) Malkov, A. V.; Bell,
M.; Castelluzzo, F.; Kocovsky, P. Org. Lett. 2005, 7, 3219. (j) Wong, W.-
L.; Lee, C.-S.; Leung, H.-K.; Kwong, H.-L. Org. Biomol. Chem. 2004, 2,
1967. (k) Pignataro, L.; Benaglia, M.; Cinquini, M.; Cozzi, F.; Celentano,
G. Chirality 2005, 17, 396. (l) Traverse, J. F.; Zhao, Y.; Hoveyda, A. H.;
Snapper, M. L. Org. Lett. 2005, 7, 3151.
Having thus identified the proline-derived catalysts 5b-7b
as the more efficient ones, we extended their use to the allylation
of other aromatic aldehydes to afford alcohols 8b-8h (Scheme
2 and Table 2). The reported data show that ee’s equal to or
greater than 60% could be obtained, with a maximum value of
84% ee observed in the case of the 3-nitrophenyl-substituted
alcohol 8f (entry 5).
(6) For the use of chiral N-oxides as catalysts or promoters in other
reactions, see: (a) Denmark, S. E.; Fan, Y. J. Am. Chem. Soc. 2002, 124,
4233. (b) Denmark, S. E.; Fan, Y.; Eastgate, M. D. J. Org. Chem. 2005,
70, 5235. (c) Liu, B.; Feng, X.; Chen, F.; Zhang, G.; Cui, X.; Jiang, Y.
Synlett 2001, 1551. (d) Jiao, Z.; Feng, X.; Liu, B.; Chen, F.; Zhang, G.;
Jiang, Y. Eur. J. Org. Chem. 2003, 3818. (e) Nakajima, M.; Yokota, T.;
Saito, M.; Hashimoto, S. Tetrahedron Lett. 2004, 45, 61.
(7) Reports appeared while this work was being completed that confirmed
the validity of this approach: see ref 5i and 5l.
The dependence of the chemical yield on the electronic nature
of the aryl substituents in the aldehydes was not easily
rationalized. However, independently of the catalyst employed,
higher ee’s were observed starting from electron-poor aldehydes
(alcohols 8e, 8f, and 8h; entries 4, 5, 7, 10, and 11). As in the
(8) For a review on the allylation of aldehydes with ally(trichloro)silane,
see: Hosomi, A. Acc. Chem. Res. 1988, 21, 2090.
(9) Huang, J.; Corey, E. J. Org. Lett. 2003, 5, 3455.
(10) The use of other solvents (toluene, dichloromethane, dichloroethane,
THF, DME) led to lower yields and ee. The best result (25% yield, 42%
ee) was observed in dichloromethane.
J. Org. Chem, Vol. 71, No. 4, 2006 1459

Pignataro et al.
SCHEME 2. Synthesis of Alcohols 8 and 9 and Catalysts
10-12
yields and slightly lower ee’s than catalyst 5b (entries 2-4 vs
8-11).
Extension of the reaction promoted by catalyst 5b to electron-
poor heteroaromatic aldehydes under the conditions of entry 4,
Table 1, afforded the corresponding alcohols in good yield (2-
pyridylcarbaldehyde, 88%; 2-thiazolylcarbaldehyde, 70%) but
in virtually racemic form (ee < 10%). The presence of the
coordinating heterocyclic nitrogen on these substrates can
account for this poor stereocontrol. Electron-rich heteroaromatic
aldehydes reacted sluggishly, if at all. The 5b-catalyzed ally-
lation of cinnamaldehyde and 3-phenylpropanal was also
attempted. Although the latter proved to be poorly reactive (20%
yield), the former gave adduct 8i in fair yield (61%) and low
ee (31%). Thus, the reactivity trend observed with catalyst 5b
(aromatic aldehydes > R,â-unsaturated aldehydes . aliphatic
aldehydes) does not differ from that of the other N-oxides
described in the literature.5 Finally, the use of catalyst 5b was
extended to the reaction of benzaldehyde with an 8:2 mixture
of (E)- and (Z)-crotyl(trichloro)silane (conditions of entry 4,
Table 1; Scheme 2). An 8:2 mixture of diastereoisomeric
alcohols anti-9 and syn-9 was obtained in 40% yield, with the
anti isomer having a 69% ee. The fact that the anti:syn
diastereoisomeric ratio reflected the (E):(Z) ratio of the starting
silane is generally considered5 a strong indication that a six-
membered cyclic chairlike transition structure is involved in the
allylation (see below).
With the aim of understanding how the various structural
components of the catalysts affected the stereochemical outcome
of the reaction, the allylation of benzaldehyde was carried out
in the presence of different catalysts and additives under the
conditions of entry 4, Table 1.
TABLE 1. Stereoselective Synthesis of Homoallylic Alcohol 8a
yield
%
ee
%
First of all, it was demonstrated that the presence of the
N-oxide function was essential for the reaction to take place:
no reaction was observed with pyridine 5a as the catalyst. It is
generally assumed that the N-oxide oxygen coordinates the
silicon atom to form the catalytically active species.5 However,
our catalysts present other sites potentially capable of silicon
coordination, namely the oxygens of the ester and carbamate
groups present in the catalyst’s sidearms. To check whether
coordination by these sites affects the course of the reaction,
two syntheses of 8a were performed in the presence of 0.1 mol
equiv of 5b and 0.4 mol equiv of N-Cbz pyrrolidine or (S)-
methyl N-Cbz prolinate as additives, respectively. Even with
this large additive/catalyst ratio, only a slight decrease in the
ee was observed in both cases (from 68 to 63%), thus suggesting
that either the ester’s or the carbamate’s oxygens do not interact
with the silicon atom or, if coordination does indeed occur, that
this should not be a decisive factor to achieve stereoselectivity.
entry catalyst amino acid
temp
abs conf.
1
2
3
4
5
6
7
8
9
3b
4b
5b
5b
5b
5b
6b
6b
7b
Val
Phe
Pro
Pro
Pro
Pro
Pro
Pro
Pro
room temp
room temp
room temp
0 °C
0 °C
0 °C
0 °C
-20 °C
0 °C
85
53
51
45
45a
75b
87
40
50
0
6
25
68
61
63
62
67
60
(S)
(S)
(S)
(S)
(S)
(S)
(S)
a In the presence of 1.2 mol equiv of tetrabutylammonium iodide. b In
the presence of 0.2 mol equiv of catalyst.
TABLE 2. Stereoselective Synthesis of Homoallylic Alcohols
8b-8h
yield
%
ee
%
entry
catalyst
Ar
product
abs conf.
1
2
3
4
5
6
7
8
9
5b
5b
5b
5b
5b
5b
5b
6b
6b
6b
7b
4-MePh
4-BrPh
4-FPh
4-NO2Ph
3-NO2Ph
3-MePh
C6F5
4-BrPh
4-FPh
4-NO2Ph
4-NO2Ph
8b
8c
8d
8e
8f
8g
8h
8c
8d
8e
8e
65
46
50
38
60
45
40
56
68
60
51
71
71
62a
76
84
73
73a
66
60a
76
73
(S)
(S)
(S)
(S)
(S)
(S)
(S)
(S)
(S)
(S)
(S)
To try to understand if the phenolic oxygen can participate
in silicon coordination (a possibility recently examined by
Kocovsky et al.5g,5i), two other allylations of benzaldehyde were
carried out. One was performed in the presence of 0.1 mol equiv
each of N-oxide 5b and the achiral, methoxy-bearing N-oxide
10 (Scheme 2); from this reaction compound, 8a was obtained
in a virtually racemic form (2% ee, 70% yield). The other
experiment was similarly performed in the presence of 0.1 mol
equiv each of N-oxide 5b and the achiral, methyl-bearing
N-oxide 11 (Scheme 2) that has roughly the same steric
hindrance as that of 10 but lacks the phenolic oxygens; from
this reaction compound, 8a was obtained in 75% yield and 30%
ee. On the basis of these experiments, it seems possible that
the N-oxide oxygen and one of the phenolic oxygens of 5b are
10
11
a Determined by HPLC on the corresponding acetates obtained by
quantitative reaction with Ac2O in pyridine at room temperature.
case of the reaction involving benzaldehyde, those involving
other aldehydes catalysts 6b and 7b led to somewhat higher
1460 J. Org. Chem., Vol. 71, No. 4, 2006

Pyridine N-Oxides as Efficient Organocatalysts
By implementing the results of this conformational analysis
with those of the above-mentioned experiments, transition
structure C (Figure 1) can be proposed to tentatively explain
the stereochemical result of the allylation reaction promoted by
catalyst 5b. In this model, the hypervalent silicon atom
(commonly believed to be involved in this type of reaction)5,6
is coordinated by the pyridine N-oxide oxygen and the phenolic
oxygen of one sidearm. The bulky proline residue effectively
blocks one side of the adduct and accommodates the aldehyde
better than the sterically more requiring allyl residue as its cis
substituent. The proposed model of stereoselection is in agree-
ment with (i) the observed formation of (S) homoallylic alcohols,
(ii) the scarce influence exerted on the stereoselectivity by the
presence of the methyl substituent of the pyridine ring of 7b
and by the steric hindrance of the proline N-protective group,
and (iii) the higher stereoselection observed in the reactions
catalyzed by 5b relative to those catalyzed by 12; this observa-
tion seems to be due to the fact that the stereodirecting proline
group is forced to be closer to the core of the transition structure
in the former than in the latter because of the presence of the
ester bond.
FIGURE 1. Conformational analysis and proposed transition structure.
involved in silicon coordination and that this coordination helps
to stereocontrol the reaction. Indeed, the achiral N-oxide 10,
which provides the same pattern of chelating sites as that of
5b, effectively competes with the chiral catalyst in promoting
the reaction, and in its presence, a virtually racemic product is
obtained; the achiral but not doubly coordinating N-oxide 11 is
less effective in competing with 5b, and the ee of product 8a is
decreased but not erased.11
To further substantiate this hypothesis, N-oxide 12 (Scheme
2) lacking the ester carbonyl oxygen but still featuring the
phenolic oxygen was prepared by a Mitsunobu reaction of
phenol 2a with commercially available N-Boc prolinol followed
by pyridine N-oxidation (45% unoptimized overall yield,
Scheme 2). When the allylation of benzaldehyde was carried
out using this catalyst under the conditions of entry 6, Table 1,
alcohol (S)-8a was obtained in 90% yield and 45% ee. Thus,
the yield was very similar to that observed with the related
catalyst 6b (90 vs 87%), but the ee was lower (45 vs 62%).
Extension of the use of 12 to the synthesis of (S)-8b (73% yield,
45% ee; cfr. entry 1, Table 2) and (S)-8e (92% yield, 56% ee;
cfr. entry 4, Table 2) confirmed this trend. By considering these
results together with those obtained by running the reaction in
the presence of different catalysts and additives (see above), it
seems possible that the higher stereoselectivity observed with
catalysts 5b and 6b with respect to 12 can in part be due to the
reduced conformational mobility of the sidearms of 5b and 6b
that contain an ester rather than an ether bond.
Molecular mechanics (MM2) calculations on model com-
pounds confirmed this hypothesis. Indeed, 2-tert-butyl-1-
acetoxybenzene (Figure 1), a model for the ester-bearing
catalysts such as 5b, was shown to exist almost exclusively as
in conformation A, featuring the CdO bond virtually eclipsed
with the aromatic ring and an s-cis arrangement of the ester
moiety; the s-trans conformation A′ is about 8 kcal/mol less
stable than A. On the other hand, in 2-tert-butyl-1-ethoxyben-
zene (a model for 12), conformation B is only 0.9 kcal/mol
more stable than B′, which is therefore largely accessible at the
reaction temperature.
Conclusions
In conclusion, a series of pyridine mono-N-oxides, readily
assembled from inexpensive amino acids, have been prepared
and tested as organocatalysts in the allylation of aldehydes with
allyl(trichloro)silane to afford homoallylic alcohols. The use of
(S)-proline-based catalysts allowed us to obtain the products
from aromatic aldehydes in fair to good yields and ee’s. The
allylation of heteroaromatic, unsaturated, and aliphatic aldehydes
was less satisfactory. By running the reaction in the presence
of achiral and chiral additives and structurally different catalysts,
some insights into the dependence of the stereochemical
outcome on the catalyst’s structure were obtained. Even if the
ee’s obtained are lower than the best values observed so far
with other catalysts, this work has demonstrated the possibility
that structurally simple pyridine N-oxides can also promote the
allylation reaction with satisfactory stereocontrol. Work is in
progress to test the catalysts in other enantioselective reactions
and to identify other readily obtained pyridine N-oxides as chiral
organocatalysts.
Experimental Section
Synthesis of 1a and 2a. A. Suzuki Coupling of 2-Bromo-3-
methylpyridine with 2,6-Dimethoxyphenyl Boronic Acid. To a
stirred solution of 2-bromo-3-methylpyridine (1.00 g, 5.52 mmol)
in DME (110 mL) were added Pd(OAc)2 (124 mg, 0.552 mmol)
and PPh3 (0.58 g, 2.21 mmol) in this order. After 20 min of stirring
at room temperature, 2,6-dimethoxyphenyl boronic acid (1.005 g,
5.52 mmol) and Na2CO3 (23 mL of a 2 M aqueous solution) were
added to the yellow solution. The resulting mixture was refluxed
under vigorous stirring for 70 h. Most of the solvent was evaporated,
and the residue was extracted with AcOEt (3 × 50 mL). The
combined organic phases were dried over Na2SO4, filtered, and
concentrated under vacuum to afford the crude product. This was
purified by flash chromatography with an 8:2 dichloromethane
(DCM)/AcOEt mixture as eluant to afford 2-(2,6-dimethoxyphenyl)-
3-methylpyridine 1a in 91% yield as a white solid (mp 93-94 °C).
IR (DCM): 1610, 1450, 1298, 1131 cm-1. H NMR (300 MHz,
1
CDCl3): δ 8.54 (d, J ) 4.8 Hz, 1H), 7.53 (d, J ) 7.6 Hz, 1H),
7.30 (t, J ) 8.4 Hz, 1H), 7.14 (dd, J ) 7.6, 4.8 Hz, 1H), 6.66 (d,
J ) 8.4 Hz, 2H), 3.69 (s, 6H), 2.07 (s, 3H). 13C NMR (75 MHz,
CDCl3): δ 157.7, 154.3, 146.6, 137.0, 133.2, 129.4, 121.9, 104.0,
(11) It must be noted that both 10 and 11 are less sterically hindered
and, hence, are likely to be better catalysts than 5b because they can have
a stronger N-oxide/silicon interaction.
J. Org. Chem, Vol. 71, No. 4, 2006 1461

Pignataro et al.
55.8, 18.6. Elemental anal. calcd for C14H15NO2: C, 73.34; H, 6.59;
N, 6.11. Found: C, 73.08; H, 6.48; N, 6.33.
148.1, 147.5, 138.7, 136.9, 134.1, 130.5, 129.4, 129.3, 127.1, 126.1,
123.2, 120.9, 119.5, 67.7, 59.6, 30.7, 18.8, 18.5, 17.6, 17.0.
Elemental anal. calcd for C38H41N3O9: C, 66.75; H, 6.04; N, 6.15.
Found: C, 66.88; H, 6.02; N, 6.09.
B. Demethylation.12 Pyridine hydrochloride was prepared by
adding HCl (417.7 mL of a 37% w/w aqueous solution, 213 mmol)
to pyridine (15.8 mL, 195 mmol). Water was removed under
vacuum, and the resulting melted residue was added to a flask
containing 2-(2,6-dimethoxyphenyl)-3-methylpyridine (1.15 g, 5.01
mmol). The mixture was stirred at 200 °C for 3 h. After the flask
was cooled at about 80 °C, hot water (8 mL) was added, and the
mixture was poured into warm water (18 mL). The pH was adjusted
to neutrality by addition of 10% w/w NaOH, and the aqueous phase
was extracted with EtOAc (3 × 30 mL). The combined organic
phases were dried over Na2SO4, filtered, and concentrated under
vacuum to afford the crude product. This was purified by flash
chromatography with a 6:4 DCM/EtOAc mixture as eluant to afford
2-(2,6-hydroxyphenyl)-3-methylpyridine 2a in 95% yield as a white
Synthesis of 5a and 5b. A. Esterification. To a stirred solution
of 2-(2,6-hydroxyphenyl)-3-methylpyridine (100 mg, 0.497 mmol),
TEA (0.554 mL, 3.476 mmol), and a catalytic amount of DMAP
in dry THF (6 mL) was added a solution of crude N-Cbz prolinoyl
chloride (prepared from 263 mg, 1.04 mmol, of N-Cbz proline)9 in
dry THF (5 mL). The mixture was refluxed for 15 h under nitrogen,
and the solvent was evaporated. The residue was purified by flash
chromatography with a 98:2 DCM/MeOH mixture as eluant. (S,S)-
3-Methyl-2-[2′,6′-bis(N-Cbz-prolinoxy)phenyl]pyridine 5a (0.481
mmol, 97%) was obtained as a white solid (mp 64 °C). [R]23 -140.0
(c 0.3 in DCM). IR: 1772, 1703 cm-1 1H NMR (300 MHz,
.
CDCl3): δ 8.50 (bd, J ) 4.2 Hz, 1H), 7.40-7.60 (m, 1H), 7.30-
7.40 (m, 10H), 7.15-7.30 (m, 2H), 6.70 (q, J ) 5.9 Hz, 1H), 5.04-
5.20 (m, 4H), 4.25-4.40 (m, 2H), 3.30-3.58 (m, 4H), 2.18, 2.14,
2.12, and 2.07 (4s, 3H overall), 1.70-1.95 (m, 2H), 1.60-1.75
(m, 2H), 1.20-1.40 (m, 4H). 13C NMR (75 MHz, CDCl3): δ 170.2,
170.0, 154.5, 151.5, 148.5, 146.6, 137.6, 134.5, 129.3, 128.4, 128.0,
127.9, 127.0, 122.9, 120.5, 120.0, 67.0, 66.9, 59.0, 58.5, 46.7, 46.1,
30.2, 29.2, 23.9, 22.9, 19.4, 18.3. Elemental anal. calcd for
C38H37N3O8: C, 68.76; H, 5.62; N, 6.33. Found: C, 68.95; H, 5.58;
N, 6.78.
1
solid (mp 184 °C). IR (DCM): 3350, 1654, 1465, 1028 cm-1. H
NMR (300 MHz, DMSO): δ 9.05 (bs, 2H), 8.38 (d, J ) 4.2 Hz,
1H), 7.60 (d, J ) 7.5 Hz, 1H), 7.18 (dd, J ) 7.5, 4.2 Hz, 1H), 6.97
(t, J ) 7.5 Hz, 1H), 6.39 (d, J ) 9.0 Hz, 2H), 2.06 (s, 3H). 13C
NMR (75 MHz, DMSO): δ 155.6, 155.0, 146.0, 136.7, 132.8,
128.6, 121.7, 115.3, 106.4, 18.4. Elemental anal. calcd for C12H11-
NO2: C, 71.63; H, 5.51; N, 6.96. Found: C, 71.48; H, 5.41; N,
7.03.
Synthesis of 3a and 3b. A. Esterification. To a stirred solution
of N-Cbz valine (200 mg, 0.79 mmol) in a DMF/DCM mixture (2
+ 2 mL) at 0 °C were added EDC (183 mg, 095 mmol) and HOBT
(146 mg, 0.95 mmol). After 20 min, 2-(2,6-hydroxyphenyl)-3-
methylpyridine 2a (80 mg, 0.397 mmol) was added and the reaction
mixture was stirred for 24 h under nitrogen at room temperature.
EtOAc was added, and the organic phase was washed with water
and a saturated aqueous solution of NaHCO3. The organic phase
was dried over Na2SO4, and the solvent was evaporated. The residue
was purified by flash chromatography with a 60:40 hexanes/EtOAC
mixture as eluant. (S,S)-3-Methyl-2-[2′,6′-bis(N-Cbz-valinoxy)-
phenyl]pyridine 3a (0.24 mmol, 60%) was obtained as a white solid
B. N-Oxidation. Following the procedure described for 3b, (S,S)-
3-methyl-2-[2′,6′-bis(N-Cbz-prolinoxy)phenyl]pyridine N-oxide 5b
was obtained in 85% yield as a white solid after flash chromatog-
raphy with a 96:4 DCM/MeOH mixture as eluant (mp 60 °C). [R]23
-10.3 (c 0.36 in DCM). IR(DCM): ν 2923, 2854, 1769, 1701,
1
1459, 1376, 1125 cm-1. H NMR (300 MHz, DMSO, 77 °C): δ
8.17 (d, J ) 6.5 Hz, 1H), 7.51 (t, J ) 8.3 Hz, 1H), 7.30-7.40 (m,
10H), 7.28 (t, J ) 6.5 Hz, 1H), 7.16 (d, J ) 6.5 Hz, 1H), 7.07 (bd,
J ) 8.1 Hz, 2H), 5.00-5.17 (m, 4H), 4.30-4.40 (m, 2H), 3.27-
3.43 (m, 4H), 3.15-3.25 (m, 1H), 2.97 (bs, 1H), 2.00-2.18 (m,
2H), 1.91 (s, 3H), 1.40-1.80 (m, 4H). 13C NMR (75 MHz, DMSO,
77 °C): δ 170.0, 169.5, 155.0, 154.5, 149.5, 142.5, 138.5, 137.2,
136.9, 136.5, 136.54, 130.6, 130.4, 128.4, 127.9, 127.8, 126.7,
126.3, 120.4, 120.1, 120.0, 67.0, 66.9, 59.2, 58.5, 46.8, 46.7, 30.5,
29.2, 24.0, 23.2, 18.5, 18.3. Elemental anal. calcd for C38H37N3O9:
C, 67.15; H, 5.49; N, 6.18. Found: C, 67.48; H, 5.62; N, 6.09.
(mp 65 °C). [R]23 -82.0 (c 0.4 in DCM). IR: 1765, 1720 cm-1
.
1H NMR (300 MHz, CDCl3): δ 8.44 (d, J ) 4.8 Hz, 1H), 7.50 (d,
J ) 5.8 Hz, 1H), 7.48 (m, 1H), 7.30-7.40 (m, 10H), 7.15-7.20
(m, 2H), 7.12 (m, 1H), 5.20-5.10 (m, 2H), 5.12 (m, 2H), 4.25 (m,
1H), 2.08 (s, 3H), 1.9 (m, 1H), 1.8 (m, 1H), 0.9 (d, J ) 5.8 Hz,
3H), 0.85 (d, J ) 5.8 Hz, 3H), 0.5 (m, 6H). 13C NMR (75 MHz,
CDCl3): δ 169.6, 156.1, 150.0, 148.4, 148.3, 146.5, 138.3, 136.0,
134.3, 129.5, 128.4, 128.3, 128.1, 126.2, 123.2, 120.4, 120.5, 67.5,
59.0, 30.6, 18.8, 18.7, 16.6. Elemental anal. calcd. for
C38H41N3O8: C, 68.35; H, 6.19; N, 6.29. Found: C, 68.45; H, 6.18;
N, 6.28.
Synthesis of (S,S)-2-[2′,6′-Bis(2-methyloxy-N-Boc-pyrrolidi-
ne)phenyl]-3-methylpyridine N-oxide 12. A. Mitsunobu Reac-
tion. To a stirred solution of 2-(2,6-hydroxyphenyl)-3-methyl-
pyridine 2a (0.201 g, 1.0 mmol), N-Boc prolinol (0.402 g, 2.0
mmol), and triphenylphosphine (0.629 g, 2.4 mmol) in dry THF
(43 mL) cooled at 10 °C was added dropwise DEAD (1.1 mL of
a 40% solution in toluene, 2.4 mmol). The cooling bath was
removed, and the progress of the reaction was monitored by TLC.
After 24 h of stirring at room temperature, volatile materials were
removed under vacuum and the residue was filtered through a short
column of silica gel for flash chromatography using EtOAc as
eluant. The resulting crude product was used as such for the
N-oxidation.
B. N-Oxidation. To a stirred solution of the diester (160 mg,
0.24 mmol) in DCM (5 mL) cooled at 0 °C was added 70%
m-CPBA (90 mg, 0.36 mmol). After 15 min of stirring at 0 °C, the
reaction mixture was allowed to warm to room temperature and
stirring was continued for 15 h. A saturated aqueous solution of
NaHCO3 was then added, and the aqueous phase was extracted
twice with DCM. The combined organic phases were repeatedly
washed with saturated NaHCO3, dried over Na2SO4, filtered, and
concentrated under vacuum to afford the crude product. This was
purified by flash chromatography with a 98:2 DCM/MeOH mixture
as eluant to afford (S,S)-3-methyl-2-[2′,6′-bis(N-Cbz-valinoxy)-
phenyl]pyridine N-oxide 3b in 77% yield as a white solid (mp 73
°C). [R]23 -9.3 (c 0.31 in DCM). IR(DCM): ν 1766, 1721, 1710
B. N-Oxidation. Following the procedure described for 3b,
product 12 was obtained (45% overall yield from 2a) as a white
solid after flash chromatography with a 9:1 DCM/MeOH mixture
as eluant (mp 68-70 °C). [R]23 -38.4 (c 0.1 in DCM). IR: 1685,
1
1400, 1265 cm-1. H NMR (300 MHz, CDCl3, 47 °C): δ 8.25
1
cm-1. H NMR (300 MHz, CDCl3): δ 8.10 (d, J ) 3.5 Hz, 1H),
(bm, 1H), 7.35 (t, J ) 7.6 Hz, 1H), 7.21 (bm, 2H), 6.72 (d, J )
7.7 Hz, 1H), 6.71 (bd, J ) 7.7 Hz, 1H), 4.20 (B part of AB system,
J ) 9.1 and 3.3 Hz, 1H), 4.03 (A part of AB system, J ) 9.1 and
1.9 Hz, 1H), 4.00 (m, 2H), 3.95 (m, 1H), 3.80 (m, 1H), 3.25 (bm,
4H), 2.03 (s, 3H), 1.84 (B part of AB system, 2H), 1.82 (A part of
AB system, 2H), 1.69 (m, 4H), 1.45 (bs, 18H). 13C NMR (75 MHz,
CDCl3, 47 °C): δ 157.0, 154.3, 145.5, 138.0, 137.0, 131.1, 126.3,
123.6, 110.0, 105.8, 105.2, 79.4, 69.5, 69.0, 68.4, 55.8, 47.0, 46.7,
46.4, 28.9, 28.5, 28.0, 27.5, 23.8, 23.4, 23.0, 22.5, 18.8. Elemental
7.50 (t, J ) 6.3 Hz, 1H), 7.30-7.40 (m, 10H), 7.20 (m, 2H), 6.95
(m, 2H), 5.17 (m, 2H), 5.05 (m, 2H), 4.30 (m, 1H), 4.20 (m, 1H),
2.01 (s, 3H), 1.90-2.00 (m, 2H), 0.85 (d, J ) 5.5 Hz, 3H), 0.80
(d, J ) 5.5 Hz, 3H), 0.70 (d, J ) 5.2 Hz, 3H), 0.65 (d, J ) 5.1 Hz,
3H). 13C NMR (75 MHz, CDCl3): δ 169.9, 156.9, 150.1, 148.8,
(12) Carina, R. F.; Dietrich-Buchecker, C.; Sauvage, J. P. J. Am. Chem.
Soc. 1996, 118, 9110.
1462 J. Org. Chem., Vol. 71, No. 4, 2006

Pyridine N-Oxides as Efficient Organocatalysts
anal. calcd for C32H45N3O7: C, 65.84; H, 7.77; N, 7.20. Found:
C, 66.01; H, 7.83; N, 7.05.
were purified by flash chromatography with different hexane/EtOAc
mixtures as eluants. The yield and ee for each reaction are indicated
in the Tables. The assignment of the (S) absolute configuration to
the predominant isomer formed in each reaction rests on comparison
of signs of optical rotation with those reported in the literature.
Allylation reaction. General procedure. To a stirred solution
of catalyst (0.03 mmol) in acetonitrile (2 mL) kept under nitrogen
were added an aldehyde (0.3 mmol) and diisopropylethylamine
(DIPEA, 0.154 mL, 0.9 mmol), in this order. The mixture was then
cooled to 0 °C, and allyl(trichloro)silane (0.054 mL, 0.36 mmol)
was added dropwise by means of a syringe. After 48 h of stirring
at 0 °C, the reaction was quenched by the addition of a saturated
aqueous solution of NaHCO3 (1 mL). The mixture was allowed to
warm to room temperature, and water (2 mL) and EtOAc (5 mL)
were added. The organic phase was separated, and the aqueous
phase was extracted 3 times with EtOAc. The combined organic
phases were dried over Na2SO4, filtered, and concentrated under
vacuum at room temperature to afford the crude products. These
1-Phenyl-3-buten-1-ol 8a. This product was purified with a
1
hexane/EtOAc 90:10 mixture as eluant. It had H NMR data in
agreement with those reported in the literature.5c The enantiomeric
excess was determined by HPLC on a Chiracel OD column (hexane/
2-propanol 95:5; flow rate 0.8 mL/min; λ 220 nm): tR ) 10.4 min,
tS ) 11.3 min.
Acknowledgment. The authors thank MIUR (Progetto
Nazionale: Strereoselezione in sintesi organica. Metodologie
ed applicazioni) for financial support. We thank Prof. Laura
Raimondi for the MM2 calculations.
(13) Kakiuchi, F.; Tsuchiya, K.; Matsumoto, M.; Mizushima, E.; Chatani,
N. J. Am. Chem. Soc. 2004, 126, 12792.
(14) Wadamoto, M.; Ozasa, N.; Yanagisawa, A.; Yamamoto, H. J. Org.
Chem. 2003, 68, 5593.
(15) Hosomi, A.; Kohra, S.; Ogata, K.; Yanagoi, T.; Tominaga, Y. J.
Org. Chem. 1990, 55, 2415.
(16) Chalicki, Z.; Guibe-Jampel, E.; Plenkiewicz, J. Synth. Commun.
1997, 27, 1217.
(17) Ramachandran, P. V.; Padiya, K. J.; Rauniyar, V.; Ram Reddy, M.
V.; Brown, H. C. J. Fluorine Chem. 2004, 125, 615.
Supporting Information Available: Synthesis and character-
ization of catalysts 1b, 2b, 4a, 4b, 6a, 6b, 7a, and 7b, synthesis of
N-oxides 10 and 11,13 and HPLC analysis details for products
8a-i14-17 and 9. This material is available free of charge via the
Internet at http://pubs.acs.org.
JO052132M
J. Org. Chem, Vol. 71, No. 4, 2006 1463