The synthesis of (Z)-trisubstituted allylic alcohols by the selective 1,4-hydrogenation of dienol esters: Improved synthesis of (-)-β-santalol

Article abstract of DOI:10.1002/chem.201002729

(E)-Trisubstituted allylic alcohols are commonly prepared from the corresponding (E)-enals, themselves readily accessible by a simple aldol condensation reaction. We demonstrate that these very same (E)-enals can be converted into (Z)-trisubstituted allylic acetates (and thus alcohols) by a ruthenium-catalyzed 1,4-hydrogenation of the corresponding dienol acetates. This simple solution to a long-lasting problem was applied to an industrially feasible synthesis of (-)-β-santalol.

Full text of DOI:10.1002/chem.201002729

FULL PAPER
DOI: 10.1002/chem.201002729
The Synthesis of (Z)-Trisubstituted Allylic Alcohols by the Selective 1,4-
Hydrogenation of Dienol Esters: Improved Synthesis of (ꢀ)-b-Santalol
Charles Fehr,*[a] Iris Magpantay,[a] Magali Vuagnoux,[a, b] and Philippe Dupau[a]
Abstract: (E)-Trisubstituted allylic alcohols are commonly prepared from the cor-
responding (E)-enals, themselves readily accessible by a simple aldol condensation
reaction. We demonstrate that these very same (E)-enals can be converted into
(Z)-trisubstituted allylic acetates (and thus alcohols) by a ruthenium-catalyzed 1,4-
hydrogenation of the corresponding dienol acetates. This simple solution to a long-
lasting problem was applied to an industrially feasible synthesis of (ꢀ)-b-santalol.
Keywords: allylic alcohols
grances · hydrogenation · natural
products · ruthenium
· fra-
Introduction
to Coreyꢀs modified Wittig reaction in which the intermedi-
ate betaine was subjected to deprotonation, followed by hy-
droxyalkylation and departure of Ph3PO.[2] Although this el-
egant one-pot olefination reaction is highly selective for this
substitution pattern (Z/E=97:3), it is not generally applica-
ble. From an industrial point of view it also suffers from the
use of large amounts of both phosphorous reagents and or-
ganometallic reagents and low temperature conditions. Z-
Selective Wittig reactions,[3] which usually also produce sig-
nificant amounts of the E-isomer, and methods based on the
functionalization of acetylenes[4] or the oxidation of allyl
boranes[5] are no better suited for industrial applications.
We now present a more general catalytic procedure for
the synthesis of Z-trisubstituted allylic alcohols and their
esters.
Herein, we describe a new, stereoselective method for the
synthesis of (Z)-trisubstituted alcohols by the Ru-catalyzed
1,4-hydrogenation of dienol acetates and its application to
an industrially feasible synthesis of (ꢀ)-b-santalol.
Recently, we reported the first enantioselective, direct
synthesis of the highly prized natural sandalwood odorant
(ꢀ)-b-santalol ((ꢀ)-4),[1] in which the key step consisted of
an efficient Cu-catalyzed rearrangement (1!2; Scheme 1).
After partial hydrogenation, the aldehyde 3 was submitted
Results and Discussion
The idea is straightforward: (E)-enals are generally readily
accessible by aldol condensation (e.g., 5 from 3; Scheme 2)
and have been commonly used for the synthesis of (E)-allyl-
Scheme 1. Synthesis of (ꢀ)-b-santalol.[1]
[a] Dr. C. Fehr, I. Magpantay, Dr. M. Vuagnoux, Dr. P. Dupau
Firmenich SA, Corporate R&D Division
P. O. Box 239, CH-1211 Geneva 8 (Switzerland)
Fax : (+41)22-780-33-34
E-mail: charles.fehr@firmenich.com
cgfehr@bluewin.ch
[b] Dr. M. Vuagnoux
Present address: Syngenta Crop Protection
Route de lꢀIle au Bois, CH-1870 Monthey (Switzerland)
Scheme 2. Planned 1,4-hydrogenation of dienol acetate 6.
Chem. Eur. J. 2011, 17, 1257 – 1260
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1257

ic alcohols (e.g., (E)-b-santalol[6]). We recognized that this
very same aldehyde could be converted into the dienol ace-
tate isomeric mixture 6, and that 1,4-hydrogenation of the
Ru-complexed s-cis-conformer should then lead to the (Z)-
allylic acetate 7 and thus (ꢀ)-4.[7]
Recently, pentamethylcyclopentadienyl (Cp*) Ru com-
plexes have been found to be effective in the 1,4-hydrogena-
tions of dienes to afford Z-alkenes such as (Z)-3-hexenol,[8]
and this method has been further improved at Firmenich in
terms of catalyst loading and scope by the use of [(Cp*)Ru-
Scheme 3. Synthesis of (ꢀ)-b-santalol ((ꢀ)-4) by 1,4-hydrogenation of
dienol acetates
6 ((1-(E)/1-(Z)=79:21). Reagents and conditions:
a) (CH2)6NH2+PhCO2ꢀ(0.4 equiv), propanal (3.0 equiv), toluene, reflux,
3 h; b) Ac2O (2.0 equiv), NEt3 (1.0 equiv), AcOK (0.43 equiv), 808C,
12 h; c) catalyst 8 (0.5 mol%), maleic acid (8 mol%), acetone, 608C, H2
(4 bar), 12 h (92% conversion); d) K2CO3 (1.2 equiv), MeOH, 90 min.
ACHTUNGTRENNUNG(1,3-cod)]ACHTUNGTRENNUNG[BF4] (8; 1,3-COD=1,3-cyclooctadiene) in the
presence of an acid.[9]
For the synthesis of 4 by 1,4-
hydrogenation, aldehyde 3 was
condensed with propanal under
organocatalytic conditions and
5[10] was converted into dienol
acetates 6 (1-(E)/1-(Z)=79:21).
The hydrogenation of which
(4 bar H2) in the presence of
catalyst
8
(0.5 mol%) and
maleic acid (8 mol%) in ace-
tone at 608C for 12 h indeed af-
forded the desired (Z)-allylic
acetate 7 in 88% yield and ex-
cellent selectivity (Z/E=98:2)
and almost complete conver-
sion (see Scheme 3 and the Ex-
perimental Section).[11] Finally,
treatment of 7 with K2CO3 in
MeOH gave (ꢀ)-4 in 98%
yield, identical in all respects to
an authentic sample.
Scheme 4. Synthesis of “(Z)-Undecavertol” ((ꢁ)-9) by 1,4-hydrogenation of dienol acetate 13. Reagents and
conditions: a) Ac2O (2.0 equiv), NEt3 (1.0 equiv), AcOK (0.1 equiv), 808C, 11 h; b) KOtBu (1.2 equiv), N-
methylpyrrolidinone (12 equiv), THF, ꢀ788C, 90 min; then ClCO2CH3 (1.1 equiv), ꢀ788C to RT, 15 min;
c) compound 13+catalyst 8 (0.5 mol%), maleic acid (8 mol%), acetone, 608C, H2 (5 bar), 9 h (97% conver-
sion); d) compound 15+catalyst
e) K2CO3 (1.2 equiv), MeOH, 1 h, RT; f) compound 17+Dess–Martin reagent (15% in CH2Cl2; 1.5 equiv),
CH2Cl2, RT, 40 min; g) CH3A
(CH2)4MgCl (2m in Et2O; 1.2 equiv), THF, ꢀ78 to ꢀ608C, 30 min.
8 (0.5 mol%), maleic acid (8 mol%), acetone, 608C, H2 (4 bar), 3 h;
CTHUNGTRENNUNG
We next applied this methodology to the preparation of
the hitherto unknown (Z)-isomer 9 of the green-floral Gi-
vaudan fragrance ingredient Undecavertol (10).[12] Whereas
(E)-2-methyl pentenal 11 is the precursor of 10, (Z)-2-
methyl pentenal 12 was chosen as the precursor of 9
(Scheme 4). The formation of dienol acetate 13 and the 1,4-
hydrogenation were again very efficient (14: Z/E=99:1).
This protocol is not limited to
therefore illustrates the overall (E)- to (Z)-isomerization of
an a,b-unsaturated aldehyde.
Scheme 5 shows further applications and one limitation.
Hydrogenation of dienol acetate 19, readily prepared from
18[15] gave the (Z)-allylic acetate 20 with excellent selectivity.
However, acetate 21 possessing a benzene ring proved inert
under the same reaction conditions.[16] Apparently, the Ru-
acetates, as demonstrated by
the efficient transformation of
carbonate 15[13] into the (Z)-al-
lylic carbonate 16 (Z/E=98:2).
Some loss of yield for 14, 16,
and 17 can be attributed to
evaporation of the volatile
products during isolation (<
2% of distillation residue). Oxi-
dation of 17 using the Dess–
Martin periodinane and addi-
tion of pentylmagnesium chlo-
ride to (Z)-enal 12 accom-
plished the synthesis of 9.[14]
The new protocol of 1,4-hydro-
Scheme 5. 1,4-Hydrogenation of dienol acetates 19 and 23. Reagents and conditions: a) isopropenyl acetate
(solvent), pyridinium tosylate (2% by weight), 958C, 4 h; b) Ac2O (2.0 equiv), NEt3 (1.0 equiv), AcOK
genation of dienol acetates (0.2 equiv), 808C, 9 h; c) catalyst 8 (0.5 mol%), maleic acid (8 mol%), acetone, 608C, H2 (4 bar), 3 h.
1258
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 1257 – 1260

The Synthesis of (Z)-Trisubstituted Allylic Alcohols
FULL PAPER
The 1,4-hydrogenation proto-
col could also be applied to 3,4-
disubstituted dienol acetates.
Hydrogenation of the citral-de-
rived dienol acetate 26 with as
little as 0.025 mol% of 8 afford-
ed neryl acetate 27 with excel-
lent selectivity (Z/E>99:1) in
98% yield (Scheme 6).[19] Like-
wise, hydrogenation of 29, pre-
pared from 28,[20] gave isomeri-
cally pure 30.
Scheme 6. 1,4-Hydrogenation of 3,4-disubstituted dienol acetates 26 and 29. Reagents and conditions: a) Ac2O
(1.5 equiv), DMAP (2.5 mol%), 1358C, 4 h; fractional distillation of isomeric mixture containing 41% of 26
(yield: 27% of 99% pure 26); b) catalyst 8 (0.025 mol%), maleic acid (0.75 mol%), acetone, 608C, H2 (5 bar),
2 h; c) isopropenyl acetate (solvent), pTsOH·H2O (2% by weight), 958C, 4 h; d) catalyst 8 (0.05 mol%),
maleic acid (0.8 mol%), acetone, 608C, H2 (4 bar), 4 h (93% conversion).
The 1,4-hydrogenation pre-
sented here starts with a ligand
exchange between 1,3-COD
and
the
dienol
acetate
(Scheme 7). Oxidative addition
of H2 to complex A would then lead to B which undergoes a
concerted 1,4-hydrogenation.[8] Alternatively, stepwise re-
duction via C (or B) would lead to an allylic species D,[21]
followed by reductive elimination or protodemetalation.
This scenario would explain the beneficial role of added
acid, postulated to accelerate the protodemetalation and
therefore prevent competing isomerizations of allylic spe-
cies.
To gain further insight into the reaction mechanism, we
deuterogenated dienol acetate 32 using Ru catalyst 8
(10 mol%) and one equivalent of maleic acid. The reaction
was run at room temperature and atmospheric pressure and
was complete after 4 h (Scheme 8).
The obtained product mixture consisted of dideuterated
acetate 33 and two regioisomeric monodeuterated products
34 and 35 in a ratio of 18:1:1. The major product 33 is con-
sistent with a concerted reaction or a reductive elimination
of D, whereas the formation of monodeuterated products 34
and 35 can be rationalized by protodemetalation of D.
Scheme 7. Possible reaction pathways.
catalyst 8 is unavailable for 1,4-hydrogenation due to its
preferential complexation to the aromatic ring.
Conclusion
The hydrogenation of dienol acetates 23 ((1-(E)/1-(Z)=
83:17)[17] revealed that the 1-(E)-isomer is reduced much
faster than the 1-(Z)-isomer.[18] Moreover, with 23 no hydro-
genation of the more substituted dienic system nor consecu-
tive isomerization reaction occurred, and 24 was obtained as
the only hydrogenation product.
We have developed a new, general, highly efficient, and
highly selective Ru-catalyzed 1,4-hydrogenation of dienol
esters to give (Z)-trisubstituted allylic alcohols and their
esters thus providing a solution to a long-standing problem.
We are actively working on further applications of this se-
lective 1,4-hydrogenation reaction.
Experimental Section
Representative procedure: A stainless-
steel 50 mL autoclave equipped with a
magnetic stirring bar was charged in
the glove box with dienol ester 6 ((1-
(E)/1-(Z)=4:1)(6.80 g;
26.1 mmol),
Ru catalyst (56 mg; 0.13 mmol),
8
maleic acid (241 mg; 2.08 mmol) and
degassed acetone (20 mL). The auto-
Scheme 8. 1,4-Deuterogenation of 32. Reagents and conditions: a) isopropenyl acetate (solvent), pTsOH·H2O
(2% by weight), 958C, 4 h; b) catalyst 8 (10 mol%), maleic acid (1.0 equiv), acetone, RT, D2 (1 bar), 4 h.
Chem. Eur. J. 2011, 17, 1257 – 1260
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1259

C. Fehr et al.
clave was purged with H2, pressurized with H2 (4 bar), and heated to
608C. After 4 and 8 h the pressure had to be re-established (4 bar). After
12 h (8% of remaining 6(Z)) the reactor was cooled, and the reaction so-
lution poured into 5% aq NaOH and pentane. The organic phase was
washed (2ꢂsat. aq NaCl), dried (Na2SO4), and concentrated. Bulb-to-
bulb distillation (oven temp. 1258C/0.04 mbar) afforded acetate 7 (Z/E=
98:2; 6.81 g; 88% pure; containing 8% of 6(Z); 88%). The distillate was
dissolved in MeOH (70 mL), treated with K2CO3 (3.80 g; 27.4 mmol) and
stirred for 90 min at RT. After partial concentration, the product was ex-
tracted (pentane/sat. aq NaCl), washed (sat. aq NaCl), dried (Na2SO4),
and concentrated. Purification by chromatography[22] using silica gel
(240 g) and cyclohexane/AcOEt 9:1 afforded pure (ꢀ)-4 (98%) identical
in all respects with an authentic sample. [a]2D0 =ꢀ104.3 (c=0.76 in
CHCl3) (lit. [3c]: [a]2D0 =ꢀ109.5 (c=0.86 in MeOH).
mann, Liebigs Ann. Chem. 1979, 743; H. Shimizu, N. Sayo, T. Saito,
Synlett 2009, 1295.
[7] For two examples of stereospecific Cr-catalyzed 1,4-hydrogenations
of dienol acetates (20–29 mol% of catalyst; 100–140 bar, 458C), see:
M. Sodeoka, T. Iimori, M. Shibasaki, Tetrahedron Lett. 1985, 26,
6497; A. Takahashi, M. Shibasaki, J. Org. Chem. 1988, 53, 1227.
[8] B. Drießen-Hçlscher, J. Heinen, J. Organomet. Chem. 1998, 570,
141; S. Steines, U. Englert, B. Driessen-Hçlscher, Chem. Commun.
2000, 217.
[9] P.
Dupau,
L.
Bonomo
(Firmenich),
WO
2008120175
(prior.03.04.2007); P. Dupau, L. Bonomo (Innovaroma), WO
2008120174 (prior.03.04.2007).
[10] For two lengthy syntheses of (ꢁ)-5, see ref. [6].
[11] In the absence of maleic acid, the Z/E-selectivity dropped to 86:14.
[12] R. Kaiser, D. Lamparsky (Givaudan Roure), EP45453, 1982, (prior.
31.07.1980).
[13] Prepared according to: P. F. De Cusati, R. A. Olofson, Tetrahedron
Lett. 1990, 31, 1405; G. Cahiez, V. Habiak, O. Gager, J. Org. Chem.
2008, 73, 6871.
[14] Compound 9 possesses a weak odor reminiscent of Undecavertol
and violet leaves.
Compound 33: 1H NMR (400 MHz, CDCl3, 258C, TMS): d=5.41 (d, J=
7.3, 1H); 4.56 (brs, 1H), 2.05 (s, 3H), 2.00–2.08 (m, 3H), 1.37–1.48 (m,
2H), 1.27–1.36 (m, 4H), 0.84–0.94 ppm (m, 6H); 13C NMR (100 MHz,
CDCl3, 258C, TMS): d=13.7 (q), 14.0 (q), 21.0 (q), 21.2 (t), 22.2 (t), 27.1
(t (C-D), J=19.2 Hz), 32.0 (t), 37.4 (t), 61.7 (t (C-D), J=22.5); 131.4 (d),
133.4 (s), 171.2 ppm (s); characteristic signal for 34: 13C NMR (100 MHz,
CDCl3, 258C, TMS): d=27.4 ppm (t); characteristic signal for 35:
1H NMR (400 MHz, CDCl3, 258C, TMS): d=4.58 ppm (s, 2H); 13C NMR
(100 MHz, CDCl3, 258C, TMS): d=62.0 ppm (t); compounds 33/34/35
(90:5:5): MS (70 eV): m/z (%): 159 (2), 158 (17), 157 (3).
[15] J. A. Bajgrowicz, K. Berg-Schultz, G. Brunner, Bioorg. Med. Chem.
2003, 11, 2931.
[16] Under forcing conditions, (Z)-2-phenylbut-2-en-1-ol (Z/E=94:6)
was obtained in 13% yield: 1) 8 (2 mol%), maleic acid (0.32 equiv),
acetone, 608C, H2 (20 bar), 20 h; then further addition of
8
(2 mol%), H2 (30 bar), 20 h (16% conversion); 2) K2CO3
(1.2 equiv), MeOH, 1 h, RT. Compound 21 (E/Z=59:41) was pre-
pared from the parent aldehyde (Ac2O (2.0 equiv), NEt3 (1.0 equiv),
AcOK (0.2 equiv), 80–1108C; 81%). For the synthesis of (E)-2-
phenyl-2-butenal, see: K. Clinch, C. J. Marquez, M. J. Parrott, R.
Ramage, Tetrahedron 1989, 45, 239.
Acknowledgement
We thank Dr. L. Saudan for helpful advice and assistance.
[17] O. Isler, M. Montavon, R. Rꢃegg, P. Zeller, Helv. Chim. Acta 1956,
39, 259.
[18] The same result was observed in the absence of maleic acid.
[19] Hydrogenation without added maleic acid led to a diminished selec-
tivity (Z/E=85:15).
[20] Prepared from 2,2-dimethyl-1-ethynyl-cyclohexanol according to:
R. L. Snowden, S. M. Linder, M. Wꢃst, Helv. Chim. Acta 1989, 72,
892.
[1] C. Fehr, I. Magpantay, J. Arpagaus, X. Marquet, M. Vuagnoux,
Angew. Chem. 2009, 121, 7357; Angew. Chem. Int. Ed. 2009, 48,
7221.
[2] E. J. Corey, H. Yamamoto, J. Am. Chem. Soc. 1970, 92, 226.
[3] a) C. Sreekumar, K. P. Darst, W. C. Still, J. Org. Chem. 1980, 45,
4260; b) W. C. Still, C. Gennari, Tetrahedron Lett. 1983, 24, 4405;
c) A. Krotz, G. Helmchen, Liebigs Ann. Chem. 1994, 601.
[4] E. J. Corey, H. A. Kirst, J. A. Katzenellenbogen, J. Am. Chem. Soc.
1970, 92, 6314; E. J. Corey, P. L. Fuchs, Tetrahedron Lett. 1972, 13,
3769; M. H. Kerrigan, S.-J. Jeon, Y. K. Chen, L. Salvi, P. J. Carroll,
P. J. Walsh, J. Am. Chem. Soc. 2009, 131, 8434 and ref. cited therein.
[5] R. J. Ely, J. P. Morken, J. Am. Chem. Soc. 2010, 132, 2534; M.
Schlosser, G. Zhong, Tetrahedron Lett. 1993, 34, 5441.
[21] Two isomeric allylic species can be formed. For clarity, only one
isomer is drawn. The more crowded 1-(Z), 3-(E)-dienol acetates are
supposed to follow the same reaction pathway, but the decreased re-
activity can be explained by a weaker diene–Ru complex.
[22] On a large scale, (ꢀ)-b-santalol is readily separated from recovered
aldehyde 5 by distillation.
[6] M. Baumann, W. Hoffmann (BASF), EP 0010213, 1980 (prior.
07.10.1978) [Chem. Abstr. 1980, 93, 185844]; M. Baumann, W. Hoff-
Received: September 22, 2010
Published online: December 7, 2010
1260
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 1257 – 1260