Asymmetric hydrogenation of substituted 2-pyrones

Article abstract of DOI:10.1021/jo982215l

Various substituted 2-pyrones have been hydrogenated with high enantioselectivity (up to 97% ee) to the corresponding 5,6-dihydropyrones using cationic ruthenium catalysts containing the (6,6'-dimethoxybiphenyl- 2,2'diyl)bis[3,5-di(tert-butyl)phenylphosphine] ligand. When substituents at position 3 are absent, 5,6-dihydropyrones are further hydrogenated to the fully saturated δ-lactones. In the case of 4,6-dimethyl-2H-pyran-2-one, the diastereoselectivity of the second hydrogenation step was determined by the chirality of the applied catalyst, while for the 4,5,6-trimethyl-2H-pyran2- one a double asymmetric induction effect was observed. Other cyclic substrates with endo- or exocyclic double bonds were hydrogenated, although with substantially lower enantioselectivity with respect to the 2-pyrones.

Full text of DOI:10.1021/jo982215l

5768
J . Org. Chem. 1999, 64, 5768-5776
Asym m etr ic Hyd r ogen a tion of Su bstitu ted 2-P yr on es
Matthias J . Fehr,† Giambattista Consiglio,*,† Michelangelo Scalone,‡ and Rudolf Schmid‡
Swiss Federal Institute of Technology, Department of Industrial and Engineering Chemistry,
ETH Zentrum, CH-8092 Zurich, Switzerland, and Pharmaceutical Division, Process Research & Catalysis,
F. Hoffmann-La Roche Ltd., CH-4070 Basel, Switzerland
Received November 5, 1998
Various substituted 2-pyrones have been hydrogenated with high enantioselectivity (up to 97% ee)
to the corresponding 5,6-dihydropyrones using cationic ruthenium catalysts containing the (6,6′-
dimethoxybiphenyl-2,2′diyl)bis[3,5-di(tert-butyl)phenylphosphine] ligand. When substituents at
position 3 are absent, 5,6-dihydropyrones are further hydrogenated to the fully saturated δ-lactones.
In the case of 4,6-dimethyl-2H-pyran-2-one, the diastereoselectivity of the second hydrogenation
step was determined by the chirality of the applied catalyst, while for the 4,5,6-trimethyl-2H-pyran-
2-one a double asymmetric induction effect was observed. Other cyclic substrates with endo- or
exocyclic double bonds were hydrogenated, although with substantially lower enantioselectivity
with respect to the 2-pyrones.
In tr od u ction
thylendioxolanones19 have been hydrogenated with high
enantioselectivity using Ru(II)-Binap catalysts.
Enantioselective homogeneous hydrogenation using
Ru(II) catalysts is a well-investigated method for stereo-
selective synthesis.1-4 High enantioselectivities have been
achieved for the hydrogenation of functionalized olefins
such as R-(acylamino)acrylic acids,5 enamides,6,7 R,â- or
â,γ-unsaturated carboxylic acids,8-12 and allylic and
homoallylic alcohols,13 especially when using complexes
containing atropisomeric bidentate phosphine ligands
such as Binap,14 Biphemp, and MeO-Biphep.15,16 Other
exocyclic olefinic substrates with aprotic functions as R,â-
unsaturated carbonyl compounds, alkylidene lactones,
alkenyl esters,17 3-alkylidene-2-piperidones,18 and R-me-
† ETH Zentrum.
‡ F. Hoffmann-La Roche Ltd.
(1) Schmid, R.; Broger, E. A.; Cereghetti, M.; Crameri, Y.; Foricher,
J .; Lalonde, M.; Mu¨ller, R. K.; Scalone, M.; Schoettel, G.; Zutter, U.
Pure Appl. Chem. 1996, 68, 131-138.
(2) Noyori, R. Acta Chem. Scand. 1996, 50, 380-390.
(3) Noyori, R. Asymmetric Catalysis in Organic Synthesis; J ohn
Wiley & Sons: New York, 1994; pp 16-94.
(4) Takaya, H.; Ohta, T.; Noyori, R. In Catalytic Asymmetric
Synthesis; Ojima, I., Ed.; VCH Publishers: New York, 1993; pp 1-39.
(5) Lubell, W. D.; Kitamura, M.; Noyori, R. Tetrahedron: Asymmetry
1991, 2, 543-554.
The hydrogenation of the R-pyrone 3a was the first
example of a regio- and enantioselective reduction of an
endocyclic enol ester type double bond.1,20 The hydroge-
nation of this type of substrate represents an interesting
and straightforward method for synthesizing optically
active 5,6-dihydro-2-pyrones and, if complete, δ-lactones,
which show interesting pharmaceutical properties.21-26
(6) Kitamura, M.; Hsiao, Y.; Ohta, M.; Tsukamoto, M.; Ohta, T.;
Takaya, H.; Noyori, R. J . Org. Chem. 1994, 59, 297-310.
(7) Noyori, R.; Ohta, M.; Hsiao, Y.; Kitamura, M.; Ohta, T.; Takaya,
H. J . Am. Chem. Soc. 1986, 108, 7117-7119.
(8) Ohta, T.; Takaya, H.; Noyori, R. Tetrahedron Lett. 1990, 31,
7189-7192.
(18) Chung, J . Y. L.; Zhao, D.; Hughes, D. L.; McNamara, J . M.;
Grabowski, E. J . J .; Reider, P. J . Tetrahedron Lett. 1995, 36, 7379-
7382.
(19) Le Gendre, P.; Braun, T.; Bruneau, C.; Dixneuf, P. H. J . Org.
Chem. 1996, 61, 8453-8455.
(20) Scalone, M.; Schmid, R.; Broger, E.; Burkart, W.; Cereghetti,
M.; Crameri, Y.; Foricher, J .; Hennig, M.; Kienzle, F.; Montavon, F.;
Schoettel, G.; Tesauro, D.; Wang, S.; Zell, R.; Zutter, U. In ChiraTech′97
Conference Proceedings; Philadelphia, 1997.
(21) Kim, J .-S.; Shin-ya, K.; Furihata, K.; Hayakawa, Y.; Seto, H.
Tetrahedron Lett. 1997, 38, 3431-3434.
(22) Thaisrivongs, S.; Romero, D. L.; Tommasi, R. A.; J anakiraman,
M. N.; Strohbach, J . W.; Turner, S. R.; Biles, C.; Morge, R. R.; J ohnson,
P. D.; Aristoff, P. A.; Tomich, P. K.; Lynn, J . C.; Horng, M.-M.; Chong,
K.-T.; Hinshaw, R. R.; Howe, W. J .; Finzel, B. C.; Watenpaugh, K. D.
J . Med. Chem. 1996, 39, 4630-4642.
(23) Tait, B. D.; Hagen, S.; Domagala, J .; Ellsworth, E. L.; Gajda,
C.; Hamilton, H. W.; Prasad, V. J . V. N.; Ferguson, D.; Graham, N.;
Hupe, D.; Nouhan, C.; Tummino, P. J .; Humblet, C.; Lunney, E. A.;
Pavlovsky, A.; Rubin, J .; Gracheck, S. J .; Baldwin, E. T.; Bhat, T. N.;
Erickson, J . W.; Gulnik, S. V.; Liu, B. J . Med. Chem. 1997, 40, 3781-
3792.
(9) Ashby, M. T.; Halpern, J . J . Am. Chem. Soc. 1991, 113, 589-
594.
(10) Zhang, X.; Uemura, T.; Matsumura, K.; Sayo, N.; Kumobayashi,
H.; Takaya, H. Synlett 1994, 501-503.
(11) Geneˆt, J . P.; Pinel, C.; Ratovelomanana-Vidal, V.; Mallart, S.;
Pfister, X.; Bischoff, L.; Can˜o De Andrade, M. C.; Darses, S.; Galopin,
S.; Laffitte, J . A. Tetrahedron: Asymmetry 1994, 5, 675-690.
(12) Ohta, T.; Takaya, H.; Kitamura, M.; Nagai, K.; Noyori, R. J .
Org. Chem. 1987, 52, 3174-3176.
(13) Takaya, H.; Ohta, T.; Sayo, N.; Kumobayashi, H.; Akutagawa,
S.; Inoue, S.; Kasahara, I.; Noyori, R. J . Am. Chem. Soc. 1987, 109,
1596-1597.
(14) Miyashita, A.; Yasuda, A.; Takaya, H.; Toriumi, K.; Ito, T.;
Souchi, T.; Noyori, R. J . Am. Chem. Soc. 1980, 102, 7932-7934.
(15) Schmid, R.; Foricher, J .; Cereghetti, M.; Scho¨nholzer, P. Helv.
Chim. Acta 1991, 74, 370-389.
(16) Schmid, R.; Cereghetti, M.; Heiser, B.; Scho¨nholzer, P.; Hansen,
H.-J . Helv. Chim. Acta 1988, 71, 897-929.
(17) Ohta, T.; Miyake, T.; Seido, N.; Kumobayashi, H.; Takaya, H.
J . Org. Chem. 1995, 60, 357-363.
10.1021/jo982215l CCC: $18.00 © 1999 American Chemical Society
Published on Web 07/14/1999

Asymmetric Hydrogenation of Substituted 2-Pyrones
J . Org. Chem., Vol. 64, No. 16, 1999 5769
Sch em e 1
In fact, tetrahydrolipstatin, a potent antiobesity drug that
is accessible starting from lactone (3S,4S,6R)-5a , has
recently been introduced in the market.1,20,27-29 When
the catalyst precursor 1 (with the atropisomeric ligand
(6,6′-dimethoxybiphenyl-2,2′-diyl)bis([3,5-di(tert-butyl)-
phenylphosphine]) was used in the presence of HBF4, the
5,6-dihydropyrone 4a was obtained in high enantiomeric
excess (96% ee). The use of other MeO-Biphep ligands
gave substantially lower enantioselectivities.1,20 The ster-
ically demanding tert-butyl groups lead to effective
hydrogenation of 3a . The high enantioselectivity was
attributed to the comparatively high rigidity and the
slightly larger chiral pockets of complexes bearing the
above-mentioned ligand.30 NMR studies of selected com-
plexes revealed restricted rotation around several of the
P-C(ipso) bonds of the phosphorus substituents contain-
ing the bulky 3,5-di-tert-butyl groups.30-32 This effect,
which also led to improved enantioselectivity of several
other catalytic reactions,33-36 was named 3,5-dialkyl meta
effect by Pregosin and co-workers. Several other ruthe-
nium complexes containing the above-mentioned ligand,
e.g., [Ru(acetone)4(P∧P)](BF4)2, [Ru(DMF)4(P∧P)](BF4)2,
or [RuH(i-PrOH)2(P∧P)]BF4, although bearing only weakly
coordinating solvent molecules, exhibited no catalytic
activity for the hydrogenation of 3a in the absence of
HBF4.20 Therefore, the addition of HBF4 seems to be
essential, not only for the abstraction of the coordinated
acetate, but also for lowering the pH value to enable high
catalytic activity.
Furthermore, the synthesis and characterization of the
31
ruthenium hydrido complexes [RuH(i-PrOH)2(P∧P)]BF4
30
and [RuH(cymene)(P∧P)]BF4 (where P∧P represents
(6,6′-dimethoxybiphenyl-2,2′-diyl)bis[3,5-di(tert-butyl)-
phenylphosphine]) suggest possible coordination modes
of the 2-pyrone 3a during hydrogenation and possible
intermediates involved in the catalytic cycle. To explore
the potential of this atropisomeric ligand, different
substituted 2-pyrones (Scheme 1) and other substrates
with endocyclic C-C double bonds were synthesized and
subjected to hydrogenation. Here we present detailed
results of the catalytic hydrogenation using the catalysts
1 and 2 in the presence of HBF4. Some of the results
concerning the stereochemistry of the stepwise reduction
of pyrone 3g have been published in a preliminary form.37
Ta ble 1. Asym m etr ic Hyd r ogen a tion of P yr on e 3a w ith
1 or 2
product 4a
[HBF4]/
convn
entry s/ca
[Ru]b catalyst [%]c sel [%] ee [%]d confign
1
2
3
4
5
6
7
8
9
50
2
10
10
2
0
10
2.6
3
10
(S)-1
(S)-1
(S)-1
(S)-1
(R)-2
(R)-2
(R)-1
(R)-1
(S)-1
100
99
99
6
86
>99
>99
95.5
94.7
96.2
R
R
R
500
333
500
500
500
500
1000
500
7
98
90
73
100
>99
>99
>99
80
94.2
92.7
91.8
96.4
S
S
S
R
a
b
Molar substrate-to-catalyst ratio. Molar equivalents of HBF4
(50% aqueous) added. c Reaction time: 20 h. See Experimental
d
(24) Beck, G.; J endralla, H.; Kesseler, K. Synthesis 1995, 1014-
1018.
Section.
(25) Endo, A. J . Med. Chem. 1985, 28, 401-405.
(26) Bennett, F.; Knight, D. W.; Fenton, G. J . Chem. Soc., Perkin
Trans. 1 1991, 133-140.
Resu lts a n d Discu ssion
En a n tioselective Hyd r ogen a tion of 4-Hyd r oxy-2-
p yr on es. Suitable reaction conditions were found in
screening experiments using pyrone 3a (see Table 1).1
Hydrogenation reactions were usually carried out at 60
°C under 60 bar of hydrogen pressure in order to achieve
complete conversion within 20 h. The chosen solvent was
2-propanol.1 To achieve high catalytic activity, the addi-
tion of HBF4 to the reaction was essential when the
acetate complex 1 or the cyclooctadienyl complex 2 was
used. While for hydrogenations at a low substrate-to-
catalyst ratio (hereafter indicated as s/c) of 50, 2 molar
equiv of HBF4 have already produced good results, for
higher s/c ratios excess HBF4 had to be added to the
reaction mixture before high catalytic activity was ob-
served (entries 2-4 and 6, Table 1).
(27) Weibel, E. K.; Hadvary, P.; Hochuli, E.; Kupfer, E.; Lengsfeld,
H. J . Antibiot. 1987, 40, 1081-1085.
(28) Hochuli, E.; Kupfer, E.; Maurer, R.; Meister, W.; Mercadal, Y.;
Schmidt, K. J . Antibiot. 1987, 40, 1086-1091.
(29) Barbier, P.; Schneider, F. J . Org. Chem. 1988, 53, 1218-1221.
(30) Trabesinger, G.; Albinati, A.; Feiken, N.; Kunz, R. W.; Pregosin,
P. S.; Tschoerner, M. J . Am. Chem. Soc. 1997, 119, 6315-6323.
(31) Currao, A.; Feiken, N.; Macchioni, A.; Nesper, R.; Pregosin, P.
S.; Trabesinger, G. Helv. Chim. Acta 1996, 79, 1587-1591.
(32) Feiken, N.; Pregosin, P. S.; Trabesinger, G.; Scalone, M.
Organometallics 1997, 16, 537-543.
(33) Togni, A. Angew. Chem., Int. Ed. Engl. 1996, 35, 1475-1477.
(34) Blaser, H.-U.; Spindler, F. Topics Catal. 1997, 4, 275-282.
(35) Trost, B. M.; Murphy, D. J . Organometallics 1985, 4, 1143-
1145.
(36) Rajanbabu, T. V.; Ayers, T. A.; Casalnuovo, A. L. J . Am. Chem.
Soc. 1994, 116, 4101-4102.
(37) Fehr, M. J .; Consiglio, G.; Scalone, M.; Schmid, R. New J . Chem
1998, 1499-1504.

5770 J . Org. Chem., Vol. 64, No. 16, 1999
Fehr et al.
Sch em e 2
Ta ble 2. Asym m etr ic Hyd r ogen a tion of P yr on es 3a -3d
w ith Com p lex 1
product
[%] [%]c confignd
[HBF4]/
[Ru]a
convn
sel
ee
entry
s/ca
50
3a 500
3b 250
catalyst [%]b
1
2
3
4
5
3a
2
10
10
2
(S)-1
(S)-1
(S)-1
(R)-1
(R)-1
>99 4a
86 95.5 R(-)
99 4a >99 94.7 R(-)
>99 4b
88 4c
75 4d
96 97.2 R* (-)
91 91.4 S* (+)
95 93.0 S* (+)
3c
3d
50
50
2
a
See Table 1. b Reaction time: 20 h. c See Experimental Section.
d
Where the absolute configuration was not determined, the
expected absolute configuration is given (*). Sign of optical rotation
is given in parentheses.
Ta ble 3. Asym m etr ic Hyd r ogen a tion of P yr on e 3e w ith
Com p lex 1
F igu r e 1. Different stereoselectivitiy for the hydrogenation
of 3e (given as relative concentrations in percent) using the
catalyst (S)-1 with or without (values in brackets) HBF4.
5e sel. [%]
[%]b [%] [%] confignc cis trans
[HBF4]/
[Ru]a
convn sel. 4e ee
entry
catalyst
1d
2
10
0
(S)-1
(S)-1
93
92
9
25
73
75
R
S
11
32
38
35
absence of HBF4, unlike in the case of 4-hydroxy-2-
pyrones 3a -d , and gave mainly the cis- and trans-
tetrahydropyrones 5e. Pyrone 3e has a higher acidity
(pKa ) 4.73, see experimental part) than the pyrones
bearing an alkyl substituent in position 3 (pKa 3d ) 5.12).
The calculated pKa values (see experimental part) for the
pyrones 3e, 3d , and 3b were 4.96, 5.24, and 5.77,
respectively, and show the growing inductive and like-
wise shielding effect going from no substituent to methyl
and to n-hexyl.
a
b
See Table 1. Reaction time: 20 h. c Based on the sign of the
optical rotation of derivatized 7e.38 Low selectivity to 5e is due
to acid-catalyzed reaction with 2-propanol (see text).
d
However, with smaller solvent volumes good activity
with somewhat lower enantioselectivitiy was obtained
with a smaller amount of acid (entries 7 and 8, Table 1).
Therefore, a certain acidity of the reaction mixture seems
to be essential for catalytic activity. Enantioselectivity
in the formation of 4a reached up to 96% ee, as deter-
mined after derivatization to the 5,6-dihydro-4-methoxy-
2-pyrone 7a (Scheme 2). The hydrogenation proceeds
selectively to the dihydropyrone. Further hydrogenation
of the second double bond of the dihydropyrone 4a is
much slower. After 5 days (compare entry 9, Table 1) only
12% of tetrahydropyrone 5a (and 8% of 5,6-dihydro-3-
hexyl-4-(2-propyloxy)-6-undecyl-2H-pyran-2-one from re-
action of 4a with the solvent 2-propanol) was formed.
To investigate the steric influence of the substituents,
differently substituted 4-hydroxy-2-pyrones (3b-e) were
synthesized and hydrogenated using either (R)- or (S)-1.
An excess of HBF4 was used for reactions at s/c ratios
higher than 50 (vide infra). The results are shown in
Tables 2 and 3. Pyrone 3b (with two n-hexyl substituents
in positions 3 and 6) is hydrogenated with comparable
or somewhat higher enantioselectivity (97.2% ee) with
respect to 3a .
Substrates 3c and 3d , both bearing a methyl group in
position 3 of the pyrone ring, are hydrogenated with lower
enantioselectivity to the corresponding dihydropyrones
(91.4 and 92.9% ee). Thus, the substitution of the n-hexyl
substituent at position 3 for a methyl group has an
favorable effect on enantioselectivity.
4-Hydroxy-6-methylpyrone (3e) was hydrogenated un-
der identical conditions and gave the fully hydrogenated
tetrahydropyrone 5e as the main product, in contrast to
the pyrones 3a -d . The two hydrogenation reactions
shown in Table 3 were carried out under identical
conditions except that in entry 2 no additional acid was
used. The catalytic reaction proceeded even in the
The higher acidity of 3e may in itself lead to a higher
proton concentration in the solution and allows for
effective hydrogenation even in the absence of HBF4. The
low selectivity for the tetrahydropyrones in entry 1 (Table
3) is due to the slowly progressing acid-catalyzed alco-
holysis of the lactone 5e to give isopropyl 3,5-dihydroxy-
hexanoate. The progressing formation of the latter com-
pound makes the alternative mechanistic possibility of
hydrogenation of a â-keto ester, formed by ring opening,
followed by ring closure rather unlikely. Surprisingly, the
hydrogenation of 3e took place preferentially from the
opposite enantioface of the C-C double bond when no
acid was used, but occurred with comparable, although
relatively lower enantioselectivity in both cases (73-76%
ee). Kinetic resolution effects in the second hydrogenation
step, as shown by the enantioselectivity in the formation
of cis- and trans-5e, do not significantly influence these
enantioselectivity values. The absolute configuration of
dihydropyrone 4e was determined through derivatization
to 7e (according to Scheme 2), determination of its optical
rotation value and comparison with literature data.38 The
hydrogenation of either enantiomer of dihydropyrone 4e,
shown in detail in Figure 1, takes place with only very
low diastereoselectivity in favor of trans-5e in the pres-
ence of HBF4 (entry 1, Table 3). In the absence of HBF4
(entry 2, Table 3) essentially no diastereoselectivity of
the second hydrogenation step is observed.
(38) Nagumo, S.; Toyonaga, T.; Inoue, T.; Nagai, M. Chem. Pharm.
Bull. 1989, 37, 2621-2623.

Asymmetric Hydrogenation of Substituted 2-Pyrones
J . Org. Chem., Vol. 64, No. 16, 1999 5771
Ta ble 4. Asym m etr ic Hyd r ogen a tion of P yr on e 3g a n d 3h w ith (R)-1a
substrate
H2 [bar]
T [°C]
t [min]
convn [%]b
ee [%]c
de [%]d
de [%]e
% ee cis-5g/h
3g
3g
3g
3h
40
60
40
40
40
40
50
40
83
77
56
99
99
98
93
86.9
88.2
87.5
92.2g
88.1
89.3
62.7
65.3
97.3
97.8f
97.2
94.2
86.5
60.8
220
>98h
>98h
a
s/c ) 250 for substrate 3g, s/c ) 100 for substrate 3h , conversion and product selectivity determined after complete conversion by
means of GC. Conversion of dihydropyrones 4g or 4h . c Enantioselectivity in the hydrogenation of 3g to 4g (Figure 3). The value was
calculated as {[(4R,6S)-5g] + [(4S,6S)-5g] - [(4R,6R)-5g] - [(4S,6R)-5g]}/{Σ5g} at complete conversion. Diastereoselectivity in the
b
d
hydrogenation of (S)-4g to 5g (Figure 3). The de value was calculated as {[(4R,6S)-5g] - [(4S,6S)-5g]}/{[(4R,6S)-5g] + [(4S,6S)-5g]} at
complete conversion. e Diastereoselectivity in the hydrogenation of (R)-4g to 5g (Figure 3). The de value was calculated as {[(4R,6R)-5g]
- [(4S,6R)-5g]}/{[(4R,6R)-5g] + [(4S,6R)-5g]} at complete conversion. f E.g., for the second entry, the diastereomeric product ratio for 5g
g
(see Figure 3) is (4S,6R)/(4R,6R)/(4S,6S)/(4R,6S) ) 1.0:4.9:5.0:89.1. Initial enantiomeric excess (at 55% conversion with respect to 3h ).
h
Relative concentrations of other diastereomers of 5h were less then 1%.
F igu r e 2. Possible geometry of the adduct between 4-hydroxy-
2-pyrone and a ruthenium hydride complex with C2-symmetric
atropisomeric (S)-diphosphine (side view, ruthenium behind
the ring).
In contrast to the observed differences in chemoselec-
tivity for the hydrogenation of pyrone 3e, 3f (the methyl
ether thereof) selectively gives the corresponding dihy-
dropyrone 4f with enantiomeric excess between 86 and
90% upon hydrogenation under identical conditions.37
The fact that pyrone 3e was fully hydrogenated to the
tetrahydropyrone 5e, while 4-hydroxy-2-pyrones with
alkyl substituents in position 3 (3a -d ) and the 4-meth-
oxy-2-pyrone 3f stop at the corresponding dihydropyrone,
can be accounted for by assuming a possible chemose-
lective hydrogenation of a keto group of the tautomer of
dihydropyrone 4e. The keto form is, in fact, not accessible
to dihydropyrone 4f due to the methylation of the hydroxy
group and to the dihydropyrones 4a -d due to the steric
hindrance of the 3-alkyl substituent. The lower enanti-
oselectivities for the hydrogenation of 4-hydroxy-2-py-
rones with substituents at position 3, going from n-hexyl
to methyl and hydrogen, can be understood on the basis
of a coordination of the substrate like that shown in
Figure 2. This coordination is analogous to that shown
by the η6-coordinated cymene found in the crystal struc-
ture of [RuH(cymene)(P∧P)]BF430 (where P∧P represents
6,6′-dimethoxybiphenyl-2,2′diyl)bis[3,5-di(tert-butyl)phe-
nylphosphine), where the methyl and isopropyl substit-
uents of the ring are located in the region of the axial
phenyl rings bound to the phosporus atoms. The para-
type substitution pattern of the alkyl substituents (R3
and R6) of the planar pyrone ring in combination with
the nearly C2-symmetry of the backbone of a possible
ruthenium hydrido adduct complex, leads to a discrimi-
nation effect influenced by the interaction of both sub-
stituents.
F igu r e 3. Enantiomeric and diastereomeric excesses in the
hydrogenation of pyrone 5g using the catalyst (R)-1/HBF4
leading to an ee of up to 98% of the main product cis-5g (entry
2, Table 4).
tetrahydropyrones 5g (Figure 3). Kinetic effects led to
an increase in the ee of 4g and cis-5g during the reaction,
thus making it difficult to compare the actual stereose-
lectivity of the hydrogenation with other pyrones. From
the ratio of the four stereomers of 5g at complete
conversion, the enantioselectivity of the hydrogenation
of pyrone 3g and the diastereoselectivity of the hydro-
genation of the two enantiomers 4g was calculated (see
Table 4 and Figure 3). The enantioselectivity of the first
hydrogenation step is around 88% ee and lies in the
region of the value for pyrone 3f. The diastereoselective
hydrogenation of the major enantiomer 4g occurs with
diastereomeric excess between 86 and 89% de and yields
preferably the cis-5g tetrahydropyrone, while the minor
enantiomer is hydrogenated with somewhat lower dias-
tereoselectivity (60-65% de) mainly to the trans-stere-
omer of 5g. In contrast, when starting with pyrone 3h
(4,5,6-trimethyl-2H-pyran-2-one), the diastereoselectivity
of the second hydrogenation step is overwhelmingly
determined by the asymmetry of the substrate. The first
hydrogenation step leads, with 80% selectivity and 94%
ee, to the corresponding 5,6-dihydropyrone 4h , while 20%
3,6-dihydropyrone 6h with close to 81% ee were formed.
Therefore, the additional methyl group leads to an
increase in the enantioselectivity of the hydrogenation
compared to pyrone 3g (88 to 94% ee). The hydrogenation
proceeds smoothly, despite the double bond being tetra-
substituted. Unlike the hydrogenation of 4g, both
enantiomers of dihydropyrone 4h are hydrogenated
diastereospecifically to the corresponding all-cis-5h
tetrahydropyrone. Isolated dihydropyrone 4h was hydro-
En a n tioselective Hyd r ogen a tion of 4-Alk yl-2-p y-
r on es. The hydrogenation of 4,6-dimethylpyrone (3g),
together with the catalyst (R)-1, gives the corresponding
cis-(4R,6S)-tetrahydro-2-pyrone 5g as the main product
in up to 98% ee.37 The hydrogenation proceeds stepwise
from pyrone 3g to give the 5,6-dihydropyrone 4g as well
as smaller amounts of the 3,6-dihydropyrone, both of
which are hydrogenated further to the corresponding

5772 J . Org. Chem., Vol. 64, No. 16, 1999
Ta ble 5. Asym m etr ic Hyd r ogen a tion of Com p ou n d s 8, 9, 11 a n d 12 by (R)-1
Fehr et al.
ee [%]
entry
substrate
H2 [bar]
T [°C]
s/c
t [h]
convn [%]
product
1a
2a
3b
4a
8
9
11
12
60
60
40
60
40
40
40
60
100
100
250
150
24
20
0.5
0.5
>99
>99
17
(S)-10
(S)-10
(R)-13c
(R)-14c
65
56
76
69
46
a
b
Two equivalents of HBF4 (50% aqueous) used. Four equivalents of HBF4 (50% aqueous) used. c Determined at low conversion (no
fully hydrogenated alcohol present).
Sch em e 3
F igu r e 4. Free energy minimized structure (consistent
valence force field) of the single molecule (S)-4g and (5S,6S)-
4h (view along the C(4)-C(3) bond).
Hyd r ogen a tion of Oth er Su bstr a tes. In contrast
with pyrones described above, hydrogenation experi-
ments with the substrates ethyl 4,6-dimethyl-2-oxo-2H-
pyran-5-carboxylate (3i) and 4,6-diphenyl-2-pyrone (3k ),
using the catalytic system 1, resulted in no hydrogenation
even when low s/c ratios were applied. Steric hindrance
of the substituents may have prevented the formation of
an adduct complex and thus a catalytic hydrogenation.
Despite their structural similarity to the 2-pyrones, the
two 2-pyridones (3l and 3m ) did not exhibit comparable
results in terms of activity and selectivity when used in
the asymmetric hydrogenation. While the 4,6-dimethyl-
2-pyridone (3l) could not be hydrogenated at all, 1,4,6-
trimethyl-2-pyridone (3m ) was hydrogenated only slowly
to the corresponding piperidone. After 46 h at 60 °C /60
bar H2 and s/c ) 25, complete conversion was possible.
The cis-piperidone 5m was formed with 84% selectivity
as the main product but with a very low enantioselec-
tivity of about 10% ee. The trans-piperidone, which was
formed with around 16% selectivity, showed 60% ee. The
reason for the comparatively low activity and enantiose-
lectiviy for the hydrogenation of these substrates is not
yet clear. However the lack of reactivity for hydrogena-
tion of pyridone 3l could be due to the predominant
existence of the pyridinol tautomeric form. This form is
not possible for the N-methyl-substituted compound 3m .
Hydrogenation of other related unsaturated hetero-
cycles with catalyst 1 was also studied (Scheme 3, Table
5). Hydrogenation of substrate 8 with an exocyclic and
substrate 9 with an endocyclic C-C double bond gave
the (S)-γ-valerolactone (S)-10 in 65 and 56% enantiomeric
excess, respectively, when the (R)-1 catalyst was used.
This values are substantially lower than the enantio-
meric excess reached for the hydrogenation of 2-pyrones.
Using Ru(II)-binap complexes, Takaya and co-workers
found 94 and 20% ee for the resulting product after the
hydrogenation of 8 and 9 respectively.17 This was ex-
plained with the difficulty of substrates bearing endocy-
clic double bonds to give chelate coordination as is the
case for substrate 9. In our case for substrate 9 (as well
as for all of the hydrogenated 2-pyrones), the ee is
substantially higher, although only endocylic bonds are
present for an adduct formation with the Ru(II) species.
genated using the opposite enantiomeric catalyst ((R)-1
and (S)-1) to give in all cases formation of all-cis-5h only.
3,6-Dihydropyrone 6h is not hydrogenated further. As
found for 4g,37 kinetic resolution effects play an impor-
tant role in the second hydrogenation step from 4h to
5h . The major enantiomer of 4h is hydrogenated faster
than its mirror image, resulting in a decrease in the
enantiomeric excess of 4h with increasing conversion. At
complete conversion, the enantiomeric excess of all-cis-
5h amounts to 94% ee. The diastereoselectivity of the
second hydrogenation step, which occurs as a result of
the additional methyl group at position five of the pyrone,
can be accounted for by looking at the structures of
dihydropyrone 4g and 4h (Figure 4). The methyl group
in allylic position to the C(4)-C(3) double bond of
dihydropyrone 4h is in a pseudoaxial position and
hinders a second hydrogenation to give the corresponding
tetrahydropyrone (4R,5S,6S)- or (4S,5R,6R)-5h . In con-
trast, for the hydrogenation of the 5,6-dihydropyrone 4g,
attack on both diastereofaces is possible, and the dias-
tereoselectivity is instead determined by the chirality of
the catalyst to give the trans- or cis-tetrahydropyrone 5g.
In the hydrogenation of the pyrone 3g, solvent, and
acid have a strong influence on the reaction rate. Use of
methanol as the solvent resulted in a 15 times lower
reaction rate than with 2-propanol, while hydrogenation
in tert-butyl alcohol showed a 1.5 times higher reaction
rate. The reaction was also carried out in 50/50 v/v %
mixture of 2-propanol and water and was as slow as that
in methanol. Using dichloromethane as the solvent, no
catalytic activity was observed. When CF3SO3H was used
instead of HBF4, the reaction proceeded at a similar rate,
while when H2SO4 was used the reaction rate decreased
by a factor of 10. When HCl or p-TsOH was used as the
acid, no catalytic hydrogenation was observed. Thus, the
importance of using strong acids with weakly coordinat-
ing counteranions was shown. Applying higher hydrogen
pressure led to an increase in the chemoselectivity for
the formation of 5,6-dihydropyrone 4h against 3,6-
dihydropyrone 6h . At 5 bar of hydrogen pressure, the
chemoselectivity was around 75%, while above 20 bar of
H2 the chemoselectivity was between 90 and 95%.

Asymmetric Hydrogenation of Substituted 2-Pyrones
J . Org. Chem., Vol. 64, No. 16, 1999 5773
purified by recrystallization or kugelrohr distillation. Fluo-
roboric acid solution (50% in water), palladium on activated
charcoal, and 2-propanol were purchased from Fluka. Hydro-
gen gas (purity 99.9999%) was used. The catalysts [Ru(OAc)2-
((S)- and (R)-3,5-di(tert-butyl)-MeO-Biphep)] ((S)-1 and (R)-1)
and 2-hexylacetoacetate were kindly provided by F. Hoffmann-
La Roche. [Ru(η5-cyclo-C8H11)((R)-3,5-di(tert-butyl)-MeO-Bi-
phep)]BF4 was synthesized as reported earlier.32
This might be explained by assuming a more rigid
structure of complex 1 with respect to the analogous
binap species. Hydrogenation of the R,â-unsaturated
cyclic ketones 11 and 12 succeeded with 76 and 69% ee,
respectively, yielding preferentially the (R)-enantiomer
when (R)-1 catalyst was used. The enantiomeric excess
was determined at low conversion due to the subsequent
hydrogenation of the keto group, which led to kinetic
resolution effects changing the ee of 13 and 14 with
progressing reaction. The relative topicity for the hydro-
genation of these two substrates was the same as for the
hydrogenation of (R)- or (S)-4g using the (R)-catalyst,
which gave the trans-(4R,6R)-5g and cis-(4R,6S)-5g
diastereomers.
3-Hexyl-4-hydroxy-6-undecyl-2H-pyran-2-one (3a) was kindly
provided by F. Hoffmann- La Roche. 4,6-Dimethyl-2H-pyran-
2-one(3g),40 4-methoxy-6-methyl-2H-pyran-one(3f),41 4,6-dim-
ethylpyridone (3l),42 1,4,6-trimethylpyridone (3m ),43 and 5,6-
dihydro-4-hydroxy-6-methyl-2H-pyran-2-one (4e)44,45 were
prepared according to published methods. 4-Hydroxy-6-methyl-
2H-pyran-2-one (3e), ethyl isodehydracetate (3i), 4,6-diphenyl-
2H-pyran-2-one (3k ), 4,5-dihydro-5-methylene-2(3H)-furanone
(8), 4-hydroxy-3-pentenoic acid γ-lactone (R-angelicalactone)
(9), 3-methyl-2-cyclohexenone (11), and 3-methyl-2-cyclopen-
tenone (12) were purchased from Fluka, Aldrich, Acros, or
Interchim S. A. and were purified by recrystallization or
kugelrohr distillation prior to use. Reference compounds of the
hydrogenation products γ-valerolactone (10), (R)-γ-valerolac-
tone ((R)-10), 3-methylcylcohexanone (13), (R)-3-methylcylco-
hexanone ((R)-13), 3-methylcyclopentanone (14), and (R)-3-
methylcyclopentanone ((R)-14) were purchased from Fluka,
Aldrich, or Acros. Synthesis and analytical data of 5,6-dihydro-
4-methoxy-6-methyl-2H-pyran-2-one (4f respectively 7e),37,38
5,6-dihydro-4,6-dimethyl-2H-pyran-2-one (4g), 3,6-dihydro-4,6-
dimethyl-2H-pyran-2-one (6g), and cis- and trans- 4,6-dim-
ethyl-3,4,5,6-tetrahydro-2H-pyran-2-one (5g) were described
previously.37
Con clu sion
The enantioselective hydrogenation of 3,6-dialkyl-4-
hydroxy-2-pyrones with the cationic ruthenium complex
of (6,6′-dimethoxybiphenyl-2,2′diyl)bis[3,5-di(tert-butyl)-
phenylphosphine] takes place selectively at positions 5
and 6 with generally high enantioselectivity (up to 97%
ee). Substitution at position 3 influences the enantiose-
lectivity of the process and renders further hydrogenation
of the double bond at positions 3 and 4 much slower, thus
bringing about good chemoselectivity. In the absence of
substitution at C(3), further hydrogenation to tetrahy-
dropyrones takes place. 4,6-Dialkyl-2-pyrones undergo
complete consecutive hydrogenation, first at the 5,6 and
then at the 3,4 double bond. Additional substitution at
position 5 strongly changes the diastereoselectivity of the
complete process causing selective formation of the all-
cis-product. An aspect which deserves further investiga-
tion is represented by the apparent 1,4-hydrogenation for
which precedents in the literature are rather scarce.
Syn t h esis of P yr on es. 3-H exyl-4-h yd r oxy-6-u n d ecyl-
2H-p yr a n -2-on e (3a ). The compound was recrystallized from
diisopropyl ether under argon atmosphere prior to use. White
1
solid: mp 83 °C. H NMR (300.13 MHz, CDCl3, 25 °C): 0.86
(d, 6H), 1.25-1.65 (m, 26H), 2.41-2.46 (t, 4H), 6.06 (s, 1H),
8.59 (br, 1H, OH). 13C NMR (75.5 MHz, CDCl3, 25 °C): 14.11
(CH3), 22.69, 23.12, 26.86, 28.07, 29.08, 29.35, 29.52, 29.63,
31.83, 31.92, 33.56 (CH2), 100.83, 103.43, 163.63, 167.08,
168.29. Anal. Calcd for C22H38O3(350.54): C, 75.38; H, 10.93.
Found: C, 75.18; H, 11.05.
3,6-Dih exyl-4-h yd r oxy-2H-p yr a n -2-on e (3b). NaH (18.5
g, 0.44 mol, 55% pure in suspension in oil) was washed several
times with absolute hexane and suspended in 250 mL of
absolute THF in a three-necked flask with a mechanical
stirrer. Freshly distilled methyl 2-hexylacetoacetate (80.1 g,
0.4 mol) was added dropwise at 0 °C. The suspension was
cooled to -10 °C, and 260 mL (0.42 mol) of n-butyllithium (1.6
M in hexane) was added within an hour. The addition of 28.8
g (0.2 mol) of methyl heptanoate was done at 0 °C, also within
an hour. The reaction mixture was then stirred for 1 h at 0
°C. The solution was cooled to - 30 °C and neutralized with 1
L of HCl (1.2 M) under vigorous stirring. The organic phase
was separated, and the water phase extracted three times with
100 mL of diethyl ether. Organic phases were washed with
saturated NaHCO3 and water. The yellowish oil (methyl 3,5-
dioxo-2-hexylundecanoate) was purified by silica gel filtration.
Methyl 3,5-dioxo-2-hexylundecanoate (57.5 g, 0.184 mol) and
1.8-diazabicyclo[5.4.0]undec-7-ene (26 g, 0.17 mol) were re-
fluxed in 200 mL of toluene for 1 h. The reaction mixture was
cooled and neutralized with ice-cold diluted HCl. The organic
Exp er im en ta l Section
Gen er a l Rem a r k s. All hydrogenation reaction solutions
were prepared and transferred to the autoclave in a glovebox
under nitrogen (<1.5 ppm of oxygen). For homogeneous
hydrogenation reactions, a 60-mL steel autoclave with a
mechanical stirrer (stirring speed 700 rpm) and a pressflow
gas controller were used. Solution NMR spectra were mea-
sured using Bruker AM-300, AMX-400, or AMX-500 spectrom-
eters with a TMS internal standard. 2D NMR methods were
used for confirmation of structures. Optical rotation was
measured with a Perkin-Elmer polarimeter 241 using a 1-dm
quartz glass cell. Elemental analyses were performed at the
Laboratory of Organic Chemistry at ETHZ. Gas chromato-
graphic analyses were carried out on a Hewlett-Packard 5890
II GC equipped with a flame ionization detector on a Restek
Rtx 200 capillary column (30 m). Enantiomeric excess was
determined by means of GC using a Lipodex C or Lipodex D
(50m), R-, â-, and γ-dex-120 (30m) column or a column with
heptakis (6-O-TBDMS-2,3-O-methyl)-â-cyclodextrin as the
stationary phase or by means of HPLC on a Hewlett-Packard
HP 1050 using Chiralcel OD-H, OJ , and OB-H columns. GC-
MS analyses were conducted on a Hewlett-Packard 5890 II
GC equipped with a quadrupole mass spectrometer. B. Wag-
ner, Hoffmann-La Roche, kindly determined the pKa values
of 4-hydroxy-2-pyrones in water by potentiometric titration
with 0.5 M HCl using a 0.1 M KNO3 electrolyte. Calculations
of pKa values were made using the program Pallas 2.0
(CompuDrug Chemistry Ltd.). For minimization of the free
energy of the single molecules (S)-4g and (6S)-4h a consistent
valence force field39 was applied using the program Cerius2
(version 3.5).
(39) Dauber-Osguthorpe, P.; Roberts, V. A.; Osguthorpe, D. J .; Wolff,
J .; Genest, M.; Hagler, A. T. Proteins: Struct., Funct., Genet. 1988, 4,
31-47.
(40) Lohaus, G.; Friedrich, W.; J eschke, J . P. Chem. Ber. 1967, 100,
658-677.
(41) De March, P.; Moreno-Man˜as, M.; Pi, R.; Ripoll, I.; Sa´nchez-
Ferrando, F. J . Heterocycl. Chem. 1985, 22, 1537-1542.
(42) Kato, T.; Sato, M.; Noda, M.; Itoh, T. Chem. Pharm. Bull. 1980,
28, 2244-2247.
(43) Kuzuya, M.; Noguchi, A.; Kamiya, S.; Okuda, T. Chem. Pharm.
Bull. 1985, 33, 2313-2322.
(44) Nedjar, B.; Hamdi, M.; Pe´rie´, J .; He´rault, V. J . Heterocycl.
Chem. 1978, 15, 1153-1158.
Ch em ica ls. All solvents were dried using standard methods
and distilled under argon prior to use. All substrates were
(45) Bacardit, R.; Moreno-Man˜as, M. Tetrahedron Lett. 1980, 21,
551-554.

5774 J . Org. Chem., Vol. 64, No. 16, 1999
Fehr et al.
layer was separated, and the water phase extracted three
times with dichloromethane. Organic phases were washed with
water and dried with Na2SO4. The solvent was evaporated,
and the product was precipitated with hexane. Recrystalliza-
tion from diisopropyl ether under argon gave 17 g 3b as a
white solid (30% yield based on methyl heptanoate): mp 99-
100 °C. 1H NMR (300.13 MHz, CDCl3, 25 °C): 0.88 (t, 6H),
1.26-1.38 (m, 12H), 1.45-1.55 (m, 2H), 1.58-1.68 (m, 2H),
2.39-2.45 (t, 4H), 5.92 (s, 1H), 9-10 (br, 1H). 13C NMR (75.47
MHz, CDCl3, 25 °C): 13.99 (CH3), 14.11 (CH3), 22.47, 22.69,
23.12, 26.78, 28.09, 28.69, 29.37, 31.44, 31.83, 33.51 (CH2),
101.12, 103.33, 163.48, 167.78, 168.54. Anal. Calcd for C17H28O3
(280.41): C, 72.82; H 10.06. found C, 72.91; H, 10.11.
4-Hydr oxy-3-m eth yl-6-u n decyl-2H-pyr an -2-on e (3c). Ac-
cording to the synthesis of 3b, 43.2 g (0.3 mol) of ethyl
2-methylacetoacetate, 14.4 g (0.33 mol, 55%) of NaH, 200 mL
(0.31 mol) of n-butyllithium, and 34.3 g (0.15 mol) of ethyl
dodecanoate gave after acid workup 48.5 g of the raw inter-
mediate ethyl 3,5-dioxo-2-methylhexadecanoate, which was
refluxed using 18.7 g (0.12 mol) of DBU in 200 mL of toluene
for 120 min to give 3c. Recrystallization from diisopropyl ether
gave 11.1 g (26.4% yield based on ethyl dodecanoate) of 3c as
a white solid: mp 106-107 °C. 1H NMR (300.13 MHz, CDCl3,
25 °C): 0.87 (t, 3H), 1.26 (m, 16H), 1.62 (m, 2H), 1.97 (s, 3H),
2.44 (t, J ) 7.5 Hz, 2H), 6.17 (s, 1H), 9.7 (br, 1H). 13C NMR
(75.47 MHz, CDCl3, 25 °C): 8.09, 14.08 (CH3), 22.66, 26.84,
28.98, 29.28, 29.31, 29.47, 29.59, 31.88, 33.48 (CH2), 98.55,
100.65, 163.50, 166.88, 168.38. Anal. Calcd for C17H28O3
(280.41): C, 72.82; H 10.06. found C, 72.57; H, 10.28.
MS: 138 (M+, 64), 110 (100), 95 (52), 67 (95), 43 (70). Anal.
Calcd for C8H10O2 (138.17): C, 69.55; H, 7.29. Found: C, 69.48;
H, 7.42.
Hyd r ogen a tion of th e Su bstr a tes. Gen er a l P r oced u r e.
Entry 2 (Table 1) is taken as an example. In an N2 atmosphere
in the glovebox, 1.5 mL of a solution of fluoroboric acid in
2-propanol (351 mg HBF4 (50% in water) in 50 mL 2-propanol)
was added to a solution of 7.5 mg (0.006 mmol) of [Ru((S)-5)-
(OAc)2] in 3.5 mL of 2-propanol. The solution was stirred for
1.5 h. 6-Hexyl-4-hydroxy-3-undecyl-2H-pyran-2-one (3a ) (1.05
g, 3 mmol) was weighed into the glass insert of the autoclave,
the catalyst solution was added, and the solution was filled
up to 30 mL with 2-propanol. The autoclave was closed and
pressurized at 60 bar of H2 and heated to 60 °C. After 20 h
the autoclave was cooled to room temperature and the residual
gas was released. After the solvent was removed from the
reaction mixture, the crude product was purified by silica gel
filtration in order to remove the catalyst and acid. Conversion
was determined by means of GC on a rtx 200 (restek) column
after derivatization with MBTFA in pyridine for 10 h or
directly by means of 1H NMR. Slightly different conditions
were used for the following entries: Table 1, entry 1, 1 mmol
of 3a in 10 mL of 2-propanol; Table 1, entry 7, 3 mmol of 3a
in 5 mL of 2-propanol; Table 1, entry 8, 3 mmol of 3a in 3 mL
of 2-propanol; Table 1, entry 9, the reaction was left for 5 days
to check whether further hydrogenation takes places. The final
reaction mixture composition was 12% 5a and 8% 5,6-dihydro-
3-hexyl-4-(2-propyloxy)-6-undecyl-2H-pyran-2-one from reac-
tion of 4a with 2-propanol were formed: Table 2, entry 3, 1.5
mmol of 3b in 15 mL of 2-propanol; Table 2, entry 4, 0.5 mmol
3c in 10 mL of 2-propanol; Table 2, entry 4, 1 mmol of 3d in
10 mL of 2-propanol.
3,6-Dim eth yl-4-h yd r oxy-2H-p yr a n -2-on e (3d ).46 6-Meth-
yl-4-hydroxy-2H-pyran-2-one (20.11 g, 0.16 mol) was sus-
pended in 160 mL of acetic acid and heated to dissolve at 50
°C. Urea (9.6 g, 0.16 mol) dissolved in 25 mL of DMF was
added. The solution was then heated to 60 °C and, within 10
min, 24.4 mL (0.223 mol) of trimethyl orthoformate was added
dropwise. The solution was left for 30 min at 60 °C and then
cooled to room temperature. The precipitated 6-methyl-3-
ureidomethylen-2H-pyran-2-one was filtered off, washed with
acetic acid and water, and dried under high vacuum. The
colorless 6-methyl-3-ureidomethylen-2H-pyran-2-one (cis/trans
mixture) (23.5 g, 0.12 mol) was suspended in 200 mL of acetic
acid and warmed to 35 °C. Dimethylamineborane (9.18 g, 0.16
mol) in 200 mL acetic acid and added dropwise. After the
addition, the solution was refluxed for 1 h. Most of the acetic
acid was evaporated under high vacuum. After addition of ice-
cold water the product precipitated and was filtered off.
Recrystallization from dioxane/acetone gave 6.24 g (37.2%
Ch a r a cter iza tion of Hyd r ogen a tion P r od u cts. (S)-5,6-
Dih yd r o-3-h exyl-4-h yd r oxy-6-u n d ecyl-2H -p yr a n -2-on e
(4a ). Compound 4a was isolated as a white solid (80%) after
hydrogenation of 1.05 g (3 mmol) of pyrone 3a in 2-propanol
under 60 bar of hydrogen at 60 °C for 20 h in the presence of
(R)-2/HBF4 as the catalyst. Purification by column chroma-
tography using dichloromethane gave a white solid: mp 112
1
°C. H NMR (300.13 MHz, CDCl3, 25 °C, keto form): 0.88 (t,
6H), 1.26-1.97 (m, 30H), 2.36-2.46 (dd, J ) 18.7, 11.6 Hz,
1H), 2.68-2.75 (dd, J ) 18.8, 2.8 Hz, 1H), 3.40 (t, J ) 5.4 Hz,
1H), 4.67 (m, 1H). 13C NMR (75.5 MHz, CDCl3, 25 °C, keto
form) 14.07 (CH3), 14.13 (CH3), 22.60, 22.69, 23.08, 24.74,
27.22, 29.27, 29.34, 29.42, 29.52, 29.61, 31.57, 31.92, 34.48,
43.63, 56.75, 74.08, 169.40 (C(2)), 201.39 (C(4)). [R]25 ) + 33
D
(c ) 0.53, MeOH, 94.2% ee). Anal. Calcd for
C22H40O3-
1
(352.56): C, 74.95; H, 11.44. Found: C, 74.85; H, 11.45.
yield) 3d as a colorless solid: mp 210 °C. H NMR (200 MHz,
DMSO, 25 °C): 1.74 (s, 3H), 2.14 (s, 3H), 5.98 (s, 1H), 11.08
(br, 1H, OH). 13C NMR (50.32 MHz, DMSO, 25 °C): 8.24 (C(3)-
CH3), 19.11 (C(6)-CH3), 96.21 (C(3)-C), 99.66 (C(5)-CH), 159.25,
164.84, 165.01. Anal. Calcd for C7H8O3 (140.14): C, 60.00; H
5.75. Found C, 59.95; H, 5.79.
(-)-3,6-Dih e xyl-5,6-d ih yd r o-4-h yd r oxy-2H -p yr a n -2-
on e (4b). Compound 4b was isolated as a white solid after
hydrogenation of 0.42 g (1.5 mmol) of pyrone 3b in 2-propanol
under 60 bar of H2 at 60 °C for 20 h in the presence of (S)-1/
HBF4 as the catalyst. It was purified by column chromatog-
raphy using dichloromethane: mp 122 °C. 1H NMR (300.13
MHz, CDCl3, 25 °C, keto form): 0.87 (t, 6H), 1.26-1.95 (m,
20H), 2.40 (dd, J ) 18.7, 11.7 Hz, 1H), 2.70 (dd, J ) 18.7, 2.9
Hz, 1H), 3.38 (t, J ) 5.5 Hz, 1H), 4.65 (m, 1H), 9-10 (br, 1H).
13C NMR (75.47 MHz, CDCl3, 25 °C, keto form): 14.02 (CH3),
22.21, 22.57, 23.06, 24.68, 27.20, 28.91, 29.34, 31.55, 31.57,
4,5,6-Tr im eth yl-2H-p yr a n -2-on e (3f).47 Iodomethane (6.84
g, 48 mmol) was added to 3.7 g (30 mmol) of pyrone 3g in 10
mL of sodium methoxide solution (30% in methanol) over 10
min. The reaction mixture was stirred for 2 h at 50 °C and
neutralized by adding concentrated acetic acid. After 10 mL
of water was added, the reaction mixture was extracted three
times with dichloromethane. The organic phases were dried
with NaSO4 and evaporated to dryness. The product was
purified by kugelrohr distillation from higher boiling side
products followed by column chromatography using ethyl
acetate/dichloromethane (1:5). Recrystallization from hexane
gave 600 mg (15% yield) of a colorless solid: mp 74 °C. 1H
NMR (300.13 MHz, CDCl3, 25 °C):1.92 (s, 3H, C(5)-CH3), 2.10
(s, 3H), 2.22 (s, 3H), 6.00 (s, 1H). 13C NMR (75.47 MHz, CDCl3,
25 °C): 12.36 (C(5)-CH3), 17.60 (C(6)-CH3), 20.55 (C(4)-CH3),
111.47, 111.53, 156.90 (C(6)), 157.39 (C(4)), 162.77 (C(2)). EI-
34.45, 43.60, 56.72, 74.05, 169.36, 201.37. [R]25 ) -40 (c )
D
0.49, MeOH, 97.2% ee). Anal. Calcd for C17H30O3 (282.42): C,
72.30; H 10.71. found C, 72.07; H, 10.68.
(+)-5,6-Dih yd r o-4-h yd r oxy-3-m eth yl-6-u n d ecyl-2H-p y-
r a n -2-on e (4c). Compound 4c was isolated as a white solid
after hydrogenation of 0.14 g (0.5 mmol) of pyrone 3c in
2-propanol under 60 bar of H2 at 60 °C for 20 h in the presence
of (R)-1/HBF4 as the catalyst. Purification by column chroma-
tography using dichloromethane gave a white solid: mp 115-
116 °C.1H NMR (300.13 MHz, CDCl3, 25 °C, keto form): 0.88
(t, 3H), 1.27-1.89 (m, 20H), 2.43 (dd, J ) 19.0, 11.8 Hz, 1H),
2.74 (dd, J ) 19.0, 2.9 Hz, 1H), 3.57 (q, J ) 6.7 Hz, 1H), 4.70
(m, 1H). 13C NMR (75.47 MHz, CDCl3, 25 °C, keto form): 7.78,
14.12 (CH3), 22.69, 22.74, 29.27, 29.34, 29.42, 29.52, 29.61,
(46) Ziegler, E.; Wolfbeis, O. S.; Trummer, I. Z. Naturforsch. 1982,
37b, 105-107.
(47) Pavlik, J . W.; Patten, A. D.; Brolin, D. R.; Bradford, K. C.;
Clennan, E. L. J . Org. Chem. 1984, 49, 4523-4531.
31.91, 34.44, 43.12, 51.66, 74.20, 170.00, 201.58. [R]25 ) +
D

Asymmetric Hydrogenation of Substituted 2-Pyrones
J . Org. Chem., Vol. 64, No. 16, 1999 5775
39 (c ) 0.32, MeOH, 91.4% ee). Anal. Calcd for C17H30O3
(282.42): C, 72.30; H, 10.71. Found C, 72.40; H, 10.94.
(+)-5,6-Dih yd r o-3,6-d im et h yl-4-h yd r oxy-2H -p yr a n -2-
on e (4d ). Compound 4d was isolated as a white solid after
hydrogenation of 0.28 g (2 mmol) of pyrone 3d in 2-propanol
under 60 bar of H2 at 60 °C for 20 h in the presence of (R)-1 as
catalyst. It was purified by column chromatography using
dichloromethane/ethyl acetate ) 1: 1 to give a white solid:
mp 151-153 °C.1H NMR (200.13 MHz, DMSO, 25 °C): 1.28
(d, J ) 6.3 Hz, 3H), 1.61 (s, 3H), 2.41-2.52 (dd, 2H), 4.30-
4.47 (m, 1H), 10.66 (br, 1H). 13C NMR (50.32 MHz, DMSO, 25
°C): 8.60 (C(3)-CH3), 20.26 (C(6)-CH3), 34.37 (C(5)-CH2), 70.35
H2 at 60 °C for 5 h in the presence of (R)-1 as the catalyst.
Purification by kugelrohr distillation gave 80 mg of a colorless
liquid. [R]25D ) - 24 (c ) 0.23, MeOH) (88% ee by GC (Lipodex
D)).1H NMR (500.14 MHz, CDCl3, 25 °C): 0.84 (d, J ) 7.2 Hz,
3H), 1.02 (d, J ) 6.5 Hz, 3H), 1.33 (d, J ) 6.6 Hz, 3H), 1.75
(m, 1H), 2.14 (dd, J ) 17.0, 12.5 Hz, 1H), 2.20 (m, 1H), 2.54
(dd, J ) 17.0, 4.7 Hz, 1H), 4.52 (dq, J ) 6.6, 2.5 Hz, 1H). 13C
NMR (75.47 MHz, CDCl3, 25 °C): 4.50, 18.91, 18.98, 31.95,
33.88, 36.39, 81.02, 171.67. EI-MS: 142 (M+, 2), 98 (26), 70
(13), 56 (100), 43 (21). Anal. Calcd for C8H14O2 (142.20): C,
67.57; H, 9.92. found: C, 67.37; H, 10.18.
1,4,6-Tr im eth ylp ip er id on e (5m ). Compound 5m was
isolated after hydrogenation of 69 mg (0.5 mmol) of pyridone
3m in 2-propanol under 60 bar of H2 at 60 °C for 46 h in the
presence of (S)-1 as the catalyst. Purification by chromatog-
raphy (dichloromethane/ethyl acetate 1:1) gave 57 mg of a
colorless liquid (84:16 cis/ trans). Yield: 57 mg (80%, 84% cis-
5m ). 1H NMR (300.13 MHz, CDCl3, 25 °C): 0.97 (d, J ) 6.1
Hz, 3H), 1.15 (m, 1H), 1.24 (d, J ) 6.3 Hz, 3H), 1.72 (m, 1H),
1.85 (dd, 1H), 1.90 (m, 1H), 2.45 (dd, 1H), 3.42 (m, 1H). 13C
NMR (75.47 MHz, CDCl3, 25 °C): 21.20, 21.78, 26.62, 31.09,
40.819, 40.844, 54.25, 170.63. EI-MS: 141 (M+, 33), 126 (100),
73 (10), 69 (74), 58 (28), 42 (23). Anal. Calcd for C8H15NO
(141.21): C, 68.04; H, 10.71; N, 9.92. Found: C, 66.15; H, 9.00;
N, 9.53.
Deter m in a tion of En a n tiom er ic Excess. 5,6-Dih yd r o-
4-h yd r oxy-2H-p yr a n -2-on es (4a -e) (com p a r e Sch em e 2).
A sample of the hydrogenation products 4a -e (0.5 mmol) was
refluxed in 5 mL of methanol, 400 mg (3.8 mmol) of trimethyl
orthoformate, and 10 mg (0.05 mmol) of p-toluenesulfonic acid
monohydrate for 10 h. The solvent of the reaction mixture was
evaporated, and the raw products 5,6-dihydro-4-methoxy-2H-
pyran-2-ones 7a -e were purified by column chromatography.
Determination of enantiomeric excesses was achieved by
means of GC or HPLC using chiral columns. Racemic reference
mixtures of 4a -e were obtained by hydrogenation of 3a -e in
ethanol with Pd on charcoal as catalyst at room temperature
and 1-5 bar of hydrogen pressure for a few hours. Derivati-
zation of the racemic 4a -e gave correspondent 7a -e as above-
described.
(C(6)-CH), 96.89 (C(3)), 165.32 (C(4)), 168.24 (C(2)). [R]25
)
D
+ 130 (c ) 0.42, MeOH, 92.9% ee). Anal. Calcd for C7H10O3
(142.15): C, 59.14; H, 7.09. Found C, 59.09; H, 7.17.
5,6-Dih yd r o-4-h yd r oxy-6-m eth yl-2H-p yr a n -2-on e (4e).
Compound 4e was obtained according to literature meth-
ods.44,45 Colorless solid: mp 118-120 °C. 1H NMR (500.14
MHz, CDCl3, 25 °C, keto form): 1.52 (d, J ) 6.2 Hz, 3H), 2.47
(dd, J ) 18.3, 11.4 Hz, 1H), 2.72 (d, J ) 18.1 Hz, 1H), 3.43 (d,
J ) 18.8 Hz, 1H), 3.58 (d, J ) 18.7 Hz, 1H), 4.81 (m, 1H). 13C
NMR (125.76 MHz, CDCl3, 25 °C): 20.49 (CH3), 45.01, 46.84,
71.97, 167.34 (C(2)), 200.07 (C(4)). Anal. Calcd for C6H8O3
(128.13): C, 56.25; H, 6.29. Found: C, 56.08; H, 6.29.
cis- a n d tr a n s-4-Hyd r oxy-6-m eth yl-3,4,5,6-tetr a h yd r o-
2H -p yr a n -2-on e (5e). A diastereomer mixture of cis- and
trans-5e was isolated after hydrogenation of 126 mg (1 mmol)
of pyrone 3e in 2-propanol under 60 bar of H2 at 60 °C for 20
h, in the presence of (S)-1 as the catalyst. Purification by
column chromatography using ethyl acetate gave 45 mg of a
colorless liquid (73% trans-5e, 27% cis-5e).
trans-5e.26 1H NMR (500 MHz, CDCl3, 25 °C): 1.39 (d, J )
6.4 Hz, 3H), 1.70 (ddd, J ) 14.5, 11.3, 3.3 Hz, 1H), 2.00 (dddd,
J ) 14.5, 3.8, 3.2, 1.7 Hz, 1H), 2.61 (ddd, J ) 18.4, 3.6, 1.7 Hz,
1H), 2.68 (dd, J ) 18.4, 4.9 Hz, 1H), 4.35 (dddd, J ) 4.9, 3.8,
3.6, 3.3 Hz, 1H), 4.85 (ddq, J ) 11.1, 6.5, 3.1 Hz, 1H). 13C NMR
(125.76 MHz, CDCl3, 25 °C): 21.38 (CH3), 37.42 (C(5)), 38.31
(C(3)), 62.42 (C(4)), 72.74 (C(6)), 171.51 (C(2)).
cis-5e.26 1H NMR (500 MHz, CDCl3, 25 °C): 1.41 (d, J ) 6.4
Hz, 3H), 1.57 (ddd, J ) 13.8, 11.6, 9.1 Hz, 1H), 2.28 (dddd, J
) 13.8, 5.6, 3.0, 1.3 Hz, 1H), 2.44 (dd, J ) 17.1, 7.6 Hz, 1H),
2.87 (ddd, J ) 17.1, 6.0, 1.3 Hz, 1H), 4.24 (dddd, J ) 9.2, 7.6,
5.6, 5.9 Hz, 1H), 4.35 (ddq, J ) 11.6, 6.5, 3.0 Hz, 1H). 13C NMR
(125.76 MHz, CDCl3, 25 °C): 21.38 (CH3), 39.21 (C(3)), 39.43
(C(5)), 63.52 (C(4)), 74.03 (C(6)), 171.70 (C(2)).
7a . White solid; mp 62 °C. 1H NMR (200.13 MHz, CDCl3,
25 °C): 0.88 (t, 6H), 1.26-1.95 (m, 28H), 2.28 (t, 2H), 2.30-
2.60 (dd, 2H), 3.76 (s, 3H), 4.23-4.35 (m, 1H). 13C NMR (50.32
MHz, CDCl3, 25 °C): 22.36 (CH3), 23.15, 24.60, 28.31, 28.78,
28.88, 29.03, 29.13, 29.19, 29.26, 29.32, 31.44, 31.60. 34.60
(CH2), 55.00 (OCH3), 74.33 (CH), 107.93 (C(3)), 165.65, 168.22.
(-)-5,6-Dih yd r o-4,5,6-tr im eth yl-2H-p yr a n -2-on e (4f).48
Compound 4f was isolated from a hydrogenation of 138 mg (1
mmol) of pyrone 3f in 2-propanol under 10 bar of hydrogen at
40 °C for 3 h in the presence of (S)-1 as the catalyst.
Purification by column chromatography (hexane/ethyl acetate
[R]25 ) + 26 (c ) 0.21, MeOH) ((S)-7a , 92.5% ee). Enantio-
D
meric excess determined by HPLC (Chiralcel OD-H, hexane/
2-propanol 99:1, 0.8 mL/min, UV(254 nm), (R)-enantiomer
eluted first).
7b. 1H NMR (300.13 MHz, CDCl3, 25 °C): 0.89 (t, 6H), 1.27-
1.90 (m, 18H), 2.28 (t, J ) 6.6 Hz, 2H), 2.43 (dd, J ) 16.9,
11.4 Hz, 1H), 2.54 (dd, J ) 16.9, 4.3 Hz, 1H), 3.76 (s, 3H),
4.28 (m, 1H), 5.13 (d, J ) 1.5 Hz, 1H). 13C NMR (75.47 MHz,
CDCl3, 25 °C): 14.06, 14.13 (CH3), 22.55, 22.66, 23.53, 24.85,
28.65, 29.08, 29.12, 29.21, 31.67, 31.76, 34.93 (CH2), 55.25
(OCH3), 74.51 (C(6)-CH), 108.74 (C(3)), 165.37 (C(4)), 168.29
(C(2)). [R]25D ) -26 (c ) 0.08, MeOH) (97.2% ee). Enantiomeric
excess determined by HPLC (Chiralcel OJ , hexane/2-propanol
165:1, 0.4 mL/min, UV(254 nm), (+)-enantiomer eluted first).
1: 1) gave 40 mg of a colorless liquid. [R]25 ) - 23 (c ) 0.76,
D
MeOH) (89% ee by GC (γ-dex-120). 1H NMR (300.13 MHz,
CDCl3, 25 °C): 1.07 (d, J ) 7.1, 3H), 1.36 (d, J ) 6.6, 3H),
2.00 (d, J ) 1.4 Hz, 3H), 2.10 (qd, J ) 7.1, 3.2 Hz, 1H), 4.56
(dq, J ) 6.6, 3.2 Hz, 1H), 5.75 (q, 1H). 13C NMR (75.47 MHz,
CDCl3, 25 °C): 10.40, 17.19, 21.47, 38.24, 75.94, 115.50, 163.63,
165.39. EI-MS: 140 (M+, 2), 96 (100), 95 (30), 81 (34), 67 (25),
53 (16), 43 (16).
3,6-Dih yd r o-4,5,6-tr im eth yl-2H-p yr a n -2-on e (6f). Com-
pound 6f was isolated after hydrogenation of 138 mg (1 mmol)
of pyrone 3f in 2-propanol under 10 bar of H2 at 40 °C for 3 h
in the presence of (S)-1 as the catalyst. Purification by column
chromatography (hexane/ethyl acetate 1: 1) gave 10 mg of a
colorless liquid. 1H NMR (300.13 MHz, CDCl3, 25 °C): 1.41
(d, J ) 6.7, 3H), 1.69 (s, 6H), 2.96 (d, 2H), 4.79 (q, J ) 6.7 Hz,
1H). 13C NMR (75.47 MHz, CDCl3, 25 °C): 17.89, 20.44, 35.23,
53.46, 79.90, 121.47, 126.58, 170.27. EI-MS: 140 (M+, 21), 125
(17), 112 (12), 97 (42), 81 (22), 43 (100).
1
7c. H NMR (500.14 MHz, CDCl3, 25 °C): 0.88 (t, J ) 6.8
Hz, 3H), 1.26-1.92 (m, 20H), 1.78 (s, 3H), 2.43 (dd, J ) 17.0,
11.9 Hz, 1H), 2.54 (dd, J ) 17.0, 4.0 Hz, 1H), 3.78 (s, 3H),
4.30 (m, 1H). 13C NMR (125.76 MHz, CDCl3, 25 °C): 8.87, 14.12
(CH3), 22.70, 24.91, 29.20, 29.35, 29.40, 29.49, 29.57, 29.62,
29.64, 31.93, 34.93 (CH2), 55.40 (OCH3), 74.67 (C(6)-CH),
103.49 (C(3)), 165.56 (C(4)), 169.00 (C(2)). [R]25 ) + 23 (c )
D
0.08, MeOH) (91.4% ee). Enantiomeric excess determined by
HPLC (Chiralcel OD-H, hexane/2-propanol ) 99:1, 0.7 mL/
min, UV(254 nm), (+)-enantiomer eluted first).
a ll-cis-(-)-3,4,5,6-Tetr ah ydr o-4,5,6-tr im eth yl-2H-pyr an -
2-on e (5f). Compound 5f was isolated after hydrogenation of
138 mg (1 mmol) of pyrone 3f in 2-propanol under 60 bar of
1
7d . H NMR (200.13 MHz, CDCl3, 25 °C): 1.46 (d, J ) 6.3
Hz, 3H), 1.78 (s, 3H), 2.34-2.51 (m, J ) 17.0, 11.3 Hz, 1H),
2.58 (dd, J ) 17.0, 4.4 Hz, 1H), 3.78 (s, 3H), 4.37-4.54 (m,
1H). 13C NMR (50.32 MHz, CDCl3, 25 °C): 8.85 (C(3)-CH3),
20.80 (C(6)-CH3), 30.86 (C(5)-CH2), 55.39 (OCH3), 70.95 (C(6)),
(48) Willson, T. M.; Kocienski, P.; J arowicki, K.; Isaac, K.; Faller,
A.; Campbell, S. F.; Bordner, J . Tetrahedron 1990, 46, 1757-1766.

5776 J . Org. Chem., Vol. 64, No. 16, 1999
Fehr et al.
125.89 (C(3)), 165.26, 168.70. [R]25 ) + 98 (c ) 0.09, MeOH)
5,6-Dih yd r o-4,5,6-tr im eth yl-2H-p yr a n -2-on e (4h ). Enan-
tiomeric excess determined by GC (Lipodex C, (+)-enantiomer
eluted first).
3,6-Dih yd r o-4,5,6-tr im eth yl-2H-p yr a n -2-on e (6h ). Enan-
tiomeric excess determined by GC (Lipodex D).
a ll-cis-3,4,5,6-Tetr a h yd r o-4,5,6-tr im eth yl-2H-p yr a n -2-
on e (5f). Enantiomeric excess determined by GC (Lipodex D,
(+)-enantiomer eluted first).
1,4,6-Tr im et h yl-2-p ip er id on e (5m ). Enantiomeric ex-
cesses of cis- and trans-diastereomers were determined by GC
(γ-Dex-120). Absolute configuration and optical rotation values
could not be determined.
γ-Va ler ola cton (10). Enantiomeric excess determined by
GC (Lipodex C, (R)-(+)-enantiomer eluted first).
3-Meth ylcycloh exa n on e (13). Enantiomeric excess deter-
mined by GC (R-Dex-120, (R)-enantiomer eluted first).
3-Meth ylcyclop en ta n on e (14). Enantiomeric excess de-
termined by GC (R-Dex-120, (R)-enantiomer eluted first).
D
(92.9% ee).Enantiomeric excess determined by GC (6-O-
TBDMS-2,3-O-methyl)-â-cyclodextrin, (+)-enantiomer eluted
first).
7e.38 1H NMR (500 MHz, CDCl3, 25 °C): 1.44 (d, J ) 6.3,
3H), 2.35 (dd, J ) 17.0, 4.0, 1H), 2.47 (ddd, J ) 17.0, 11.6,
1.6, 1H), 3.75 (s, 3H), 4.53 (ddq, J ) 11.6, 4.0, 6.3 Hz, 1H),
5.14 (d, J ) 1.5 Hz, 1H). 13C NMR (100.61 MHz, CDCl3, 25
°C): 20.54 (CH3), 34.58 (C(5)), 55.98 (OCH3), 72.24 (C(6)), 90.23
(C(3)), 167.34, 172.74. [R]25 ) -123 (c ) 0.53, MeOH) ((R)-
D
7e,38 89% ee). Enantiomeric excess determined by GC (6-O-
TBDMS-2,3-O-methyl)-â-cyclodextrine, (S)-(+)-enantiomer elut-
ed first).
5,6-Dih yd r o-4-m eth oxy-2H-p yr a n -2-on es (4f). As dihy-
dropyrone 7e (see above).
5,6-Dih yd r o-4,6-d im eth yl-2H-p yr a n -2-on e (4g). Enan-
tiomeric excess determined by GC (Lipodex C, (S)-(+)-enan-
tiomer eluted first).37
3,6-Dih yd r o-4,6-d im eth yl-2H-p yr a n -2-on e (6g). Enan-
tiomeric excess determined by GC (Lipodex C, (S)-enantiomer
eluted first).37
t r a n s-4,6-Dim e t h yl-3,4,5,6-t e t r a h yd r o-2H -p yr a n -2-
on e (5g). Enantiomeric excess determined by GC (γ-Dex-120,
(4S,6S)-(-)-enantiomer eluted first).37
cis-4,6-Dim eth yl-3,4,5,6-tetr ah ydr o-2H-pyr an -2-on e (5g).
Enantiomeric excess determined by GC (Lipodex C, (4S,6R)-
(+)-enantiomer eluted first).37
Ack n ow led gm en t. The pioneering work of Dr. E.
Broger on the asymmetric hydrogenation of 2-pyrones
is gratefully acknowledged. We thank also Dr. U.
Widmer for helpful discussions. M.J .F. is grateful to
Hoffmann-La Roche for the generous support of his
Ph.D. work.
J O982215L