Enantioselective synthesis of (-)-dihydrocodeinone: A short formal synthesis of (-)-morphine

Article abstract of DOI:10.1021/jo0513008

The radical cyclization approach to the morphine alkaloids has been applied in an asymmetric synthesis of (-)-dihydrocodeinone. A chiral cyclohexenol (R-32), from the CBS reduction of the enone, is the source of chirality. The first key step, tandem closure in which stereochemistry is controlled by geometric constraints, (-)-15b → (+)-16, was followed by an unprecedented reductive hydroamination, completing the synthesis of (-)-dihydroisocodeine ((-)-17) in 13 steps from commercially available materials.

Full text of DOI:10.1021/jo0513008

VOLUME 71, NUMBER 2
JANUARY 20, 2006
© Copyright 2006 by the American Chemical Society
Enantioselective Synthesis of (-)-Dihydrocodeinone:
A Short Formal Synthesis of (-)-Morphine1,†
Kathlyn A. Parker*,‡ and Demosthenes Fokas§
Department of Chemistry, Brown UniVersity, ProVidence, Rhode Island 02912
kparker@notes.cc.sunysb.edu
ReceiVed June 24, 2005
The radical cyclization approach to the morphine alkaloids has been applied in an asymmetric synthesis
of (-)-dihydrocodeinone. A chiral cyclohexenol (R-32), from the CBS reduction of the enone, is the
source of chirality. The first key step, tandem closure in which stereochemistry is controlled by geometric
constraints, (-)-15b f (+)-16, was followed by an unprecedented reductive hydroamination, completing
the synthesis of (-)-dihydroisocodeine ((-)-17) in 13 steps from commercially available materials.
Introduction
desirable activity but exhibit reduced side effects continues at
a steady pace.3
Morphine Alkaloids as Targets for Total Synthesis.
Morphine (1a)2 is the principal active component of opium, the
extract of the immature seed capsule of the opium poppy.
Morphine and related opioids are analgesics.
Morphine itself is a potent compound. Despite serious side
effects, including physical addiction, it continues to be one of
the most widely used drugs for the treatment of severe pain.
Codeine (1b), a weak analgesic, is also an antitussive and
commonly used in cough medicines. Research focused on the
discovery of morphine derivatives or analogues which maintain
The compact, functional-group-rich structures of this class
continue to provoke the interest and effort of synthetic chemists.
Among many imaginative approaches which have been ex-
plored, a number have led to total syntheses or formal total
syntheses.4
† This paper is dedicated to the memory of Arthur G. Schultz.
‡ Current address: Department of Chemistry, State University of New York
at Stony Brook, Stony Brook, New York 11794-3400.
§ Current address: Department of Chemistry, ArQule Inc., 19 Presidential
Way, Woburn, MA 01801. E-mail: dfokas@arqule.com.
(1) A preliminary report of part of this work has appeared: Parker, K.
A.; Fokas, D. J. Am. Chem. Soc. 1992, 114, 9688.
(2) In their 1925 paper in which they deduce the structures of codeine
and thebaine, Gulland and Robinson designated the carbocyclic rings of
the morphine skeleton as I, II, and III. In our 1992 communication we
extended this convention by labeling the nitrogen-containing bridge as ring
IV. Although other systems have been introduced in recent years, we
continue to use the original Robinson nomenclature in the discussion in
this paper, see: Gulland, J. M.; Robinson, R. Mem. Proc. Manchester Lit.
Philos. Soc. 1925, 69, 79.
Our own synthesis of (()-dihydrocodeinone1 employed a
tandem radical cyclization of an easily obtained aryl cyclohex-
enyl ether and introduced an unprecedented intramolecular
hydroamination reaction carried out under reductive conditions.
This key step, which may involve a nitrogen radical, completed
the construction of the morphine ring system.
10.1021/jo0513008 CCC: $33.50 © 2006 American Chemical Society
Published on Web 11/04/2005
J. Org. Chem. 2006, 71, 449-455
449

Parker and Fokas
SCHEME 1
SCHEME 2
In light of continuing interest in morphine as a synthetic target
and given the recurring use of the reductive hydroamination
protocol in the context of the synthesis of morphinoids,5,6 we
disclose the full history and experimental procedures for this
work. In addition, we describe herein the application of our
tandem radical cyclization/reductive hydroamination approach
to the asymmetric synthesis of (-)-dihydrocodeinone. (-)-
Dihydrocodeinone has been converted to codeine, and codeine
has been demethylated to morphine. Therefore, this paper
constitutes a report of a formal asymmetric synthesis of
morphine.
Background of the Tandem Radical Cyclization Strategy.
Our strategy for the synthesis of the morphine alkaloids is based
on the cyclization of an allyloxy aryl radical to produce a 3,3-
disubstituted dihydrobenzofuran, a key structural component of
the target molecules. In studies focused on the kinetics of radical
cyclization (Scheme 1), Beckwith reported that the diazonium
salt 2, when treated with tributyltin hydride, affords a 71% yield
of the 3,3-disubstituted dihydrobenzofuran 5, presumably via
the intermediate radicals 3 and 4.7 Thus, it appeared that a key
structural feature of our target molecules might be generated in
one step from a readily available precursor.
afford the tetracyclic radical 9. Hydride trapping would then
complete the conversion of an aryl allyl ether to an advanced
morphine intermediate 10.
A preliminary study showed that the tandem radical closure
of substrates of this type, specifically 11, provide efficient
preparations of the desired cis,cis-tetracycles (12a and 12b).8
However, this and a related study9 suggested that a reaction of
this general type is unlikely to generate products in which the
C-9 center is formed with stereocontrol.
Furthermore, it seemed likely that the tandem radical cy-
clization strategy might be extended to cascades triggered by
the formation of aryl radicals. For example (Scheme 2),
cyclization of aryl radical 7 (from halo arene 6) to cyclohexyl
radical 8 might be followed by a second cyclization event to
(3) (a) Lednicer, D.; Mitscher, L. A. In The Organic Chemistry of Drug
Synthesis: Compounds Related to Morphine; John Wiley and Sons: New
York, 1980; Vol. 2, Chapter 10. (b) Lednicer, D. Central Analgesics; John
Wiley and Sons: New York, 1982; pp 137-213. (c) Lenz, G. R.; Evans, S.
M.; Walters, D. E.; Hopfinger, A. J.; Hammond, D. L. Opiates; Academic
Press: Orlando, 1986. (d) Casy, A. F.; Parfitt, R. T. Opioid Analgesics:
Chemistry and Receptors; Plenum Press: New York, 1986. (e) Zimmerman,
D. M.; Leander, J. D. J. Med. Chem. 1990, 33, 895. (f) Iversen, L. Nature
1996, 383, 759.
Therefore, we modified our approach to one in which the
cyclization would produce a functionalized but not tetrahedral
C-9, as in the closure of 13 to 14. The C-9 center might then
be introduced in a subsequent step.
(4) (a) A critical review of morphine total synthesis appeared recently;
see: Taber, D. F.; Neubert, T. D.; Schlecht. M. F. The EnantioselectiVe
Synthesis of Morphine: Strategies and Tactics in Organic Synthesis 2004,
5, 353. (b) For an entertaining history of morphine and a personal account
of phenanthrene-based approaches to the synthesis of morphine, see:
Blakemore, P. R.; White, J. D. Chem. Commun. 2002, 1159. (c) For a
comprehensive review of contributions to morphine synthesis and biosyn-
thesis for the periods 1996-99 and 2000-2004, see: Novak, B. H.;
Hudlicky, T.; Reed, J. W.; Mulzer, J.; Trauner, D. Curr. Org. Chem. 2000,
4, 343. Zezula, J.; Hudlicky, T. Synlett 2005, 388. (d) The history,
biosynthesis, and total synthesis of morphine as well as the semisynthesis
of some derivatives has been reviewed, see: Franckenpohl, J. Chem. Unserer
Z. 2000, 34, 99. (e) An earlier review that includes a discussion of the
strategic bond disconnection approach as applied to morphine has been
provided by Maier, see: Maier, M. Organic Synthesis Highlights II 1995,
357.
(5) (a) Trauner, D.; Bats, J. W.; Werner, A.; Mulzer, J. J. Org. Chem.
1998, 63, 5908. (b) Trauner, D.; Porth, S.; Opatz, T.; Bats, J. W.; Giester,
G.; Mulzer, J. Synthesis 1998, 653. (c) Nagata, H.; Miyazawa, N.;
Ogasawara, K. Chem. Commun. 2001, 1094. (d) Yamada, O.; Ogasawara,
K. Org. Lett. 2000, 2, 2785.
(6) See also the nonreductive method of Trost: (a) Trost, B. M.; Tang,
W. J. Am. Chem. Soc. 2002, 124, 14542. (b) Trost, B. M.; Tang, W. J. Am.
Chem. Soc. 2003, 125, 8744.
The remaining challenges for application of this methodology
in the total synthesis of morphine were to (1) include additional
required functionality in the cyclization substrate, (2) devise a
practical method of closing the final ring IV of the target
structure, (3) complete the morphine molecule, at least in a
formal sense, and (4) demonstrate that the synthesis could be
applied to an enantiomerically enriched intermediate in order
to ultimately afford an enantiomerically enriched product. In
this paper we are pleased to report having achieved all of these
goals.
(8) Parker, K. A.; Spero, D. M.; Van Epp, J. J. Org. Chem. 1988, 53,
4628.
(9) Parker, K. A.; Fokas, D. J. Org. Chem. 1994, 59, 3927.
(7) Beckwith, A. L. J.; Meijs, G. F. J. Chem. Soc., Chem. Commun. 1981,
136.
450 J. Org. Chem., Vol. 71, No. 2, 2006

EnantioselectiVe Synthesis of (-)-Dihydrocodeinone
SCHEME 3
SCHEME 5
SCHEME 4
alcohol 23. Cyclohexenediol 24 was prepared by regioselective
isomerization13 of epoxy alcohol 23 with Ti(OiPr)4 according
to the Sharpless protocol. Silylation of the less hindered hydroxyl
group of cis-diol 24 afforded the target monoprotected diol 25.
Preparation of the cyclization substrate required coupling of
alcohol 25 with phenol 26, which was obtained in one step from
condensation of bromoisovanillin with diethyl phenylthiometh-
ylphosphonate. The key step was accomplished by Mitsunobu
coupling, affording aryl ether 15a. Removal of the silyl
protecting group generated alcohol 15b (R ) H), which proved
to be a suitable substrate for the proposed radical-initiated
reaction. When heated with Bu3SnH (0.035 M) and AIBN in
benzene in a sealed tube (130 °C), bromoaryl ether 15b
underwent the projected tandem cyclization/elimination se-
quence (Scheme 5) to afford the tetracyclic styrene 16 in a
modest but serviceable 35% yield. Chatgilialoglu’s reagent, tris-
(trimethylsilyl)silane,14 effected the same transformation, af-
fording the desired styrene 16 in 20-30% yield.
A byproduct in the tributyltin hydride-initiated reaction,
isolated in 11% yield, proved to be ketone 21. Formation of
ketone 21 may be the result of intramolecular hydrogen
abstraction from the homoallylic position which bears the
hydroxyl group in radical r-1 (Scheme 5). The R-hydroxy radical
r-2 could then expel the adjacent phenoxide radical to produce
the conjugated dienol corresponding to enone 21.
We reasoned that we might improve the yield of the
cyclization step by using a modified substrate. We assumed that
the aryl radical derived from the cis-hydroxy ether 27, the epimer
of 15b, would undergo the tandem closure but be unable to
abstract a hydrogen from the homoallylic carbinol carbon (as
in r-1) because of geometric constraints. Attempts to prepare
alcohol 27 from alcohol 15b, under the standard Mitsunobu
conditions,15 led only to the recovery of starting material.
Results and Discussion
Preparation and Tandem Cyclization of Fully Function-
alized Substrates. To generate a tandem closure product, which
would exhibit maximum correspondence of functionality with
that of the target structures, we needed a cyclization substrate
in which the latent C-6 would bear a hydroxyl or keto group
and in which the latent C-13 would bear an aminoethyl side
chain. Furthermore, to optimize the overall efficiency of the
final scheme, we sought routes in which the nitrogen functional-
ity would be in place from the beginning of the synthesis. We
chose, therefore, to base our synthesis on the key intermediate
15. Scheme 3 shows the retrosynthetic analysis which drove
our experiments. We anticipated that the tandem closure of the
aryl radical derived from bromoarene 15 would give tetracycle
16. Conversion of this product to the known morphine inter-
mediate, dihydroisocodeine (17), would require only deprotec-
tion and formation of the C9-N bond.
On the basis of our model studies9,10 we expected to be able
to prepare substrate 15 (Scheme 4) by Mitsunobu coupling of
phenol 26 and cyclohexenol 25 and both coupling partners to
be readily available. These assumptions proved to be correct.
Preparation of alcohol 25 required seven steps and was
achieved in a straightforward fashion. Birch reduction of
m-methoxyphenethylamine afforded the nonconjugated dienol
ether 19. Tosylation of the amino group followed by hydrolysis
of the enol ether afforded enone 20. N-Alkylation11 gave enone-
sulfonamide 21, and Luche reduction12 afforded allylic alcohol
22. Peracid epoxidation of cyclohexenol 22 yielded the cis-epoxy
(13) Morgans, D. J.; Sharpless, K. B.; Traynor, S. G. J. Am. Chem. Soc.
1983, 103, 462.
(10) Parker, K. A.; Fokas, D. J. Org. Chem. 1994, 59, 3933.
(11) Hendrickson, J. B.; Bergeron, R. Tetrahedron Lett. 1973, 3839.
(12) Gemal, A. L.; Luche, J.-L. J. Am. Chem. Soc. 1981, 103, 5454.
(14) Chatgilialoglu, C.; Griller, D.; Lesage, J. J. Org. Chem. 1989, 54,
2492.
(15) Mitsunobu, O. Synthesis 1981, 1.
J. Org. Chem, Vol. 71, No. 2, 2006 451

Parker and Fokas
However, Dess-Martin oxidation16 of 15b followed by im-
mediate sodium borohydride reduction of the intermediate â,γ-
enone (Scheme 5) allowed the formation of the desired
stereoisomeric alcohol 27. The tributyltin hydride cyclization
of this substrate afforded a 20-30% yield of styrene 28 with
no detectable byproduct 21. Although the cyclization reaction
of substrate 27 did not produce the fragmentation byproduct,
the yield of tetracyclic material was not improved.
SCHEME 6
Closure of Ring IV. With ready access to tosylamides 16
and 28, we were now in a position to consider the completion
of the morphine skeleton by closure of the final ring. We
postulated that the C9-N bond connection would result from
the intramolecular addition of an electrophilic aminyl radical
to the electron-rich styrene via the stable benzylic radical.
N-Centered radicals are available from N-functionalized amines.17
Therefore, we reviewed the standard methods for desulfonation
of sulfonamides, focusing first and foremost on the dissolving
metal reduction which we viewed as particularly intriguing in
the present context. It seemed possible that the nitrogen radical
(or anion), generated by reductive detosylation of intermediate
16, might add to the â-carbon of the styrene moiety, affording
dihydroisocodeine (17) directly.
Although the reductive desulfonation of olefinic sulfonamides
had not previously resulted in a cyclization event,18 treatment
of tosylamide 16 with Li/NH3 in the presence of t-BuOH (-78
°C) afforded (()-dihydroisocodeine (17) in 85% yield (see
Scheme 6). Likewise, treatment of sulfonamide 28 (Scheme 6)
with Li/NH3 at -78 °C afforded (()-dihydrocodeine (29)19 in
70% yield. This type of closure provides an efficient means of
connecting the C9-N17 bond of the morphine ring system.
Others have used it in the elaboration of morphinan intermedi-
ates.5,6,20 However, it is not representative of a general trans-
formation for the closure of piperidine rings.21
variations in this approach, the former, proceeding through
intermediate 16 and dihydroisocodeine (17), is shorter and
allows for a higher overall yield. Therefore, we applied this
synthesis in the chiral series.
Chiral Synthesis of (-)-Morphine Alkaloids. The asym-
metric synthesis of morphine alkaloids by the approach de-
scribed above requires chiral alcohol (R)-22. In principle, this
chiral starting material would be available by either a chiral
reduction26 of enone 21 or a Sharpless kinetic resolution27 of
the racemic allylic alcohol 22. Although chiral reduction of
3-substituted-2-cyclohexenones generally does not proceed with
good enantiomeric excess, Terashima’s reagent had been
reported to reduce 3-methyl-2-cyclohexenone with 90% ee.28
In our studies, however, Terashima’s reagent reduced unsatur-
ated ketone 21 to afford allylic alcohol 22 with only 5%
enantiomeric excess. We obtained better results with the Corey-
Bakshi-Shibata (CBS) (S)-oxazaborolidine-catechol borane
reagent,29 which produced the desired (R)-enantiomer of alcohol
22 with a 35-40% enantiomeric excess from reduction of
ketone 21. An attempt to use this procedure in conjunction with
a Sharpless kinetic resolution30 yielded material in which the
enantiomeric excess was improved to 44%.
Formal Syntheses of (()-Morphine. The cyclizations of
tosylamides 16 and 28 constitute formal total syntheses of
morphine. Both dihydroisocodeine (17)22 and dihydrocodeine
(29)23 have been oxidized to dihydrocodeinone (30, Scheme 6),
which has been converted to codeine (1b).24 Facile O-demeth-
ylation of codeine provides morphine (1a).25 Of the two
(16) Dess, D. B.; Martin, J. C. J. Org. Chem. 1983, 48, 4155.
(17) For leading references in aminyl radicals, see: (a) Guindon, Y.;
Gue´rin, B.; Landry, S. R. Org. Lett. 2001, 3, 2293. (b) Tsuritani, T.;
Shinokubo, H.; Oshima, K. Org. Lett. 2001, 3, 2709. (c) Newcomb, M.;
Musa, O. M.; Martinez, F. N.; Horner, J. H. J. Am. Chem. Soc. 1997, 119,
4569. (d) Maxwell, B. J.; Tsanaktsidis, J. J. Am. Chem. Soc. 1996, 118,
4276. (e) Kim, S.; Joe, G. H.; Do, J. Y. J. Am. Chem. Soc. 1994, 116,
5521. (f) Newcomb, M.; Weber, K. A. J. Org. Chem. 1991, 56, 1309. (g)
Bowman, W. R.; Clark, D. N.; Marmon, R. J. Tetrahedron Lett. 1991, 32,
6441. (h) Newcomb, M.; Esker, J. L. Tetrahedron Lett. 1991, 32, 1035. (i)
Newcomb, M.; Deeb, T. M.; Marquardt, D. J. Tetrahedron 1990, 46, 2317.
(j) Newcomb, M.; Marquardt, D. J.; Deeb, T. M. Tetrahedron 1990, 46,
2329. (k) Tokuda, M.; Yamada, Y.; Takagi, T.; Suginome, H. Tetrahedron
1987, 43, 281. (l) Stella, L. Angew. Chem., Int. Ed. Engl. 1983, 22, 337.
(m) Maeda, Y.; Ingold, K. U. J. Am. Chem. Soc. 1980, 102, 328. (n) Perry,
C. A.; Chen, S. C.; Menon, B. C.; Hanaya, K.; Chow, Y. L. Can. J. Chem.
1976, 54, 2385.
(18) (a) Heathcock, C. H.; Smith, K. M.; Blumenkopf, T. A. J. Am. Chem.
Soc. 1986, 108, 5022. (b) Schultz, A. G.; McCloskey, P. J.; Court, J. J. J.
Am. Chem. Soc. 1987, 109, 6493.
(19) The NMR spectrum of (()-dihydrocodeine was identical to that of
an authentic sample of (-)-dihydrocodeine, prepared by the hydrogenation
of codeine (Rapoport, H.; Payne, G. B. J. Org. Chem. 1950, 15, 1093).
(20) Hanada, K.; Miyazawa, N.; Ogasawara, K. Org. Lett. 2002, 4, 4515.
(21) The factors required for “trapping” of a reactive N-centered species,
during the reductive detosylation, are the subject of current study in our
laboratories.
(23) For oxidation of dihydrocodeine to dihydrocodeinone under Op-
penauer conditions, see: Rapoport, H.; Naumann, R.; Bissell, E. R.; Bonner,
R. M. J. Org. Chem. 1950, 15, 1103. However, oxidation of dihydroiso-
codeine to dihydrocodeinone, under the same Oppenauer conditions, was
unsuccessful.
(24) (a) Weller, D. D.; Rapoport, H. J. Med. Chem. 1976, 19, 1171. (b)
Iijima, I.; Rice, K. C.; Silverton, J. V. Heterocycles 1977, 6, 1157.
(25) (a) Rice, K. C. J. Med. Chem. 1977, 20, 164. (b) Lawson, J. A.;
DeGraw, J. I. J. Med. Chem. 1977, 20, 165.
(26) For more information regarding asymmetric reductions of different
classes of ketones, see: (a) Itsuno, S. Org. React. 1998, 52, 395. (b) Corey,
E. J.; Helal, C. J. Angew. Chem., Int. Ed. Engl. 1998, 37, 1986. (c) Brown,
H. C.; Ramachandran, P. V. Acc. Chem. Res. 1992, 25, 16. (d) Midland,
M. M. Chem. ReV. 1989, 89, 1553. (e) Brown, H. C.; Park, W. S.; Cho, B.
T.; Ramachandran, P. V. J. Org. Chem. 1987, 52, 5406.
(27) (a) Martin, V. S.; Woodard, S. S.; Katsuki, T.; Yamada, Y.; Ikeda,
M.; Sharpless, K. B. J. Am. Chem. Soc. 1981, 103, 6237. (b) Gao, Y.;
Hanson, R. M.; Klunder, J. M.; Ko, S. Y.; Masamune, H.; Sharpless, K. B.
J. Am. Chem. Soc. 1987, 109, 5765.
(28) (a) Kawasaki, M.; Suzuki, Y.; Terashima, S. Chem. Lett. 1984, 239.
(b) Kawasaki, M.; Suzuki, Y.; Terashima, S. Chem. Pharm. Bull. 1985,
33, 52.
(29) (a) Corey, E. J.; Bakshi, R. K.; Shibata, S. J. Am. Chem. Soc. 1987,
109, 5551. (b) Corey, E. J.; Bakshi, R. K.; Shibata, S.; Chen, C.-P.; Singh,
V. K. J. Am. Chem. Soc. 1987, 109, 7925. (c) Mathre, D. J.; Jones, T. K.;
Blacklock, T. J. J. Org. Chem. 1991, 56, 751.
(22) In our studies Swern oxidation of dihydroisocodeine 17 afforded
(()-dihydrocodeinone (30) in 83% yield. When applied in the chiral series
Swern oxidation of (-)-17 produced 80% of (-)-30.
452 J. Org. Chem., Vol. 71, No. 2, 2006

EnantioselectiVe Synthesis of (-)-Dihydrocodeinone
SCHEME 7
In the absence of an efficient direct method for producing
alcohol 22 in high ee we resorted to an indirect strategy (see
Scheme 7) based on the asymmetric reduction of a 2-bromocy-
clohexenone. Although 2-unsubstituted cyclohexenones give
relatively low enantiomeric excesses when subjected to the CBS
reducing system, 2-bromocyclohexenone is reported to be
reduced in a 91% ee under these conditions.29b Thus, bromi-
nation of enone 21, followed by treatment with triethylamine,
afforded bromoenone 31 in 91% yield. Reduction of 31 with
(S)-oxazaborolidine (2.4 equiv) and catechol borane afforded
alcohol (R)-32 with enantiomeric excesses ranging from 82%
to 96%.31 When reduction was performed on a 1 g scale, bromo
alcohol (R)-32 was obtained in 84% yield with an 82%
enantiomeric excess as indicated by Mosher ester analysis. (It
is important to note that reaction of 31 with a catalytic amount
(0.1 equiv) of oxazaborolidine was extremely slow; starting
material was recovered even after stirring at room temperature
for 24 h with excess catechol borane.) Attempted removal of
bromine from (R)-32 by metal halogen exchange was not
successful. When bromo alcohol (R)-32 was treated with t-BuLi
or n-BuLi at various temperatures, a mixture of starting material
and unidentified products was obtained. However, reduction by
sodium amalgam in MeOH-THF proved to be a suitable
method for debromination. Thus, bromo alcohol (R)-32 (82%
ee) was converted to allylic alcohol (R)-22 in 90% yield
(Scheme 7); this material was shown to have an 80% ee by
Mosher ester analysis. Having established access to chiral
alcohol 22 we were poised to complete an asymmetric synthesis
of (-)-morphine alkaloids.
As we had determined that there was no practical advantage
in a scheme based on the R-alcohol 27, we chose the more direct
route in which the â-alcohol 15b is the substrate for tandem
cyclization. In fact, allylic alcohol (R)-22 was converted to (-)-
dihydroisocodeine [(-)-17] by application of Schemes 4 (22
f 15b), 5 (15b f 16), and 6 (16 f 17, 17 f 30) as described
above for the racemic synthesis. Yields, optical rotations, and
enantiomeric excesses as calculated by Mosher ester analysis
or comparison with optical rotation values for authentic materials
are reported in Scheme 7 as well as in the Experimental Section.
Synthetic (-)-dihydroisocodeine [(-)-17], prepared according
to Scheme 7, had an [R]24D -75.9 in CHCl3. A negative specific
rotation ([R]28 -136 in CHCl3)32b,c has been reported for
D
dihydroisocodeine derived from natural materials.32
Swern oxidation of (-)-dihydroisocodeine (17) afforded (-)-
dihydrocodeinone (30) in 80% yield (Scheme 7). Our synthetic
(-)-dihydrocodeinone ([R]24D -119.7 in CHCl3) was produced
with approximately 75% ee, as calculated from the specific
rotation of an authentic sample prepared from natural codeine
(30) For an example of a double kinetic resolution and references to
procedures which couple a kinetic resolution with an enantioselective
reaction, see: Brown, S. M.; Davies, S. G.; de Sousa, J. A. A. Tetrahe-
dron: Asymmetry 1991, 2, 511.
(31) Reduction of 31 with LiAlH4/Darvon Alcohol (Yamaguchi, S.;
Mosher, H. S. J. Org. Chem. 1973, 38, 1870. Reich, C. J.; Sullivan, G. R.;
Mosher, H. S. Tetrahedron Lett. 1973, 1505) afforded bromoallylic alcohol
(R)-32 in 70% yield with 84% ee.
(32) For the synthesis of (-)-dihydroisocodeine from natural codeine,
see: (a) Baizer, M. M.; Loter, A.; Ellner, K. S.; Satriana, D. R. J. Org.
Chem. 1951, 16, 543. (b) Ginsburg, P.; Elad, D. J. Am. Chem. Soc. 1956,
78, 3691. (c) Okuda, S.; Tsuda, K.; Yamaguchi, S. J. Org. Chem. 1962,
27, 4121. (d) Chatterjie, N.; Umans, J. G.; Inturrisi, C. E. J. Org. Chem.
1976, 41, 3624. (e) Brine, G. A.; Boldt, K. G.; Coleman, M. L.; Bradley,
D. J.; Carroll, F. I. J. Org. Chem. 1978, 43, 1555. (f) Fuska, J.; Proska, B.;
Fuskova, A.; Khandlova, A. Acta Biotechnol. 1988, 8, 291.
J. Org. Chem, Vol. 71, No. 2, 2006 453

Parker and Fokas
([R]24 -159.1 in CHCl3).33,34 The synthesis of (-)-dihydro-
codeinone completes a formal total synthesis of (-)-codeine
(1b) and (-)-morphine (1a).
with 2 mL of 10% HF. The reaction mixture was stirred at room
temperature for 24 h, quenched with water, and extracted with CH2-
Cl2 (3 × 30 mL). The combined organic solution was washed with
brine, dried over Na2SO4, and concentrated to produce the crude
product. Flash chromatography with EtOAc-Hex-acetone (1:3:
1) yielded 503 mg (98%) of a foamy solid, mp 56-59 °C. IR
D
Conclusions
(CDCl3) 3515, 3044, 1580, 1474, 1330, 1285, 1156, 1026 cm-1
.
This convergent synthesis of the morphine alkaloids by radical
cyclization illustrates the versatility of these processes for
construction of multifunctional polycyclic compounds. In
particular, it demonstrates the power of this methodology for
“stitching” rings together to build convex ring systems. This
synthesis is completely stereospecific. A single stereocenter, the
first to be introduced in the synthetic scheme, controls the
remaining four stereocenters in the target structure.
The key steps in this synthetic scheme come in sequence at
the end of the synthesis. They are (1) the convergent step, the
efficient Mitsunobu coupling of a readily available cyclohexene
cis-diol derivative with a phenol to form the cyclization
substrate, (2) the tandem radical cyclization of an aryl cyclo-
hexenyl ether to produce a suitably functionalized tetracyclic
styrene, and (3) completion of the morphine ring system by the
reductive deprotection/hydroamination reaction.
400 MHz 1H NMR (CDCl3) δ 7.65 (d, 2 H, J ) 8.2 Hz), 7.42 (dd,
2 H, J ) 7.5, 1.4 Hz), 7.35 (t, 2 H, J ) 7.6 Hz), 7.28 (m, 3 H),
7.20 (d, 1 H, J ) 8.7 Hz), 7.02 (d, 1 H, J ) 15.4 Hz), 6.85 (d, 1
H, J ) 8.7 Hz), 6.71 (d, 1 H, J ) 15.3 Hz), 5.82 (s, 1 H), 4.74 (d,
1 H, J ) 3.6 Hz), 4.08 (br s, 1 H), 3.86 (s, 3 H), 3.32 (m, 1 H),
3.06 (m, 1 H), 2.73 (s, 3 H), 2.61 (m, 1 H), 2.41 (s, 4 H), 2.24-
2.14 (m, 3 H), 1.97 (d, 1 H, J ) 4.5 Hz), 1.77 (m, 1 H). 100.6
MHz 13C NMR (CDCl3) δ 152.1, 144.7, 143.1, 134.92, 134.90,
131.2, 130.3, 130.2, 130.1, 129.9, 129.5, 129.1, 127.2, 127.0, 124.7,
121.4, 118.9, 111.4, 80.2, 68.5, 55.8, 49.4, 34.7, 32.2, 25.4, 21.4.
HRMS (FAB/NBA) for C31H34O579BrNS2: calcd, 643.1061; found,
643.1054.
(-)-15b. The above procedure was applied to aryl ether (-)-
15a to afford a white solid, mp 52-54 °C, in 95% yield. This
material had [R]24 -51.5 (c ) 0.08 in CHCl3), corresponding to
D
a 78% ee by Mosher ester analysis.
Cyclization of Alcohol (()-15b. Tetracyclic Sulfonamide (()-
16 and Enone 21. A solution of 120 mg (0.187 mmol) of the
alcohol (()-15b, 75 µL (0.280 mmol) of Bu3SnH, and a catalytic
amount of AIBN (0.1-0.2 equiv) in 8 mL of benzene was heated
in a sealed tube at 130 °C. A small amount of AIBN was added
every 8 h in order to maintain the radical chain. After 35 h the
reaction mixture was concentrated, and the residue was dissolved
in Et2O and then stirred vigorously with 10% KF for 2 h. The ether
phase was separated, washed with brine, dried over Na2SO4, and
concentrated to produce a yellow residue. Preparative TLC on silica
gel with EtOAc-Hex-acetone (1:3:1) afforded 30 mg (35%) of
sulfonamide (()-16 as a yellow oil. IR (CDCl3) 3500, 3057, 1498,
1446, 1333, 1153, 1086 cm-1. 250 MHz 1H NMR (CDCl3) δ 7.58
(d, 2 H, J ) 8.2 Hz), 7.28 (d, 2 H, J ) 8.4 Hz), 6.70 (d, 1 H, J )
8.1 Hz), 6.64 (d, 1 H, J ) 8.1 Hz), 6.37 (d, 1 H, J ) 9.5 Hz), 5.83
(dd, 1 H, J ) 9.5, 5.7 Hz), 4.54 (d, 1 H, J ) 7.2 Hz), 3.89 (s, 3 H),
3.46 (m, 1 H), 3.13 (m, 1 H), 2.84 (m, 1 H), 2.61 (s, 3 H), 2.48 (m,
1 H), 2.42 (s, 5 H), 1.88-1.65 (m, 2 H), 1.35 (m, 1 H), 0.93 (m,
1 H). HRMS (EI) for C25H29O5NS: calcd, 455.1766; found,
455.1756. Also recovered from the preparative TLC plate was 7
mg (11%) of enone 21.
The radical cyclization approach to morphine has proven
amenable to asymmetric synthesis, affording (-)-dihydroiso-
codeine with 75% ee in a total of 13 steps and in an overall
yield of 4.2% from the commercially available m-methoxyphen-
ethylamine.
Experimental Section
(()-Aryl Ether 15a. A solution of 0.25 mL (1.61 mmol) of
diethyl azodicarboxylate in 5 mL of THF at 0 °C was treated
dropwise with 0.4 mL (1.61 mmol) of tri-n-butylphosphine. The
resulting solution was stirred at 0 °C for 10 min and then added
dropwise at 0 °C to a THF solution (50 mL) containing 589 mg
(1.34 mmol) of alcohol (()-25 and 495 mg (1.47 mmol) of phenol
26. The reaction mixture was stirred in an ice bath for 4 h and
then concentrated to give a yellow residue. Flash chromatography
with EtOAc-Hex (1:4) afforded 842 mg (83%) of a colorless oil.
IR (CDCl3) 3040, 1596, 1585, 1474, 1333, 1286, 1204, 1086, 1033,
1
1004 cm-1. 400 MHz H NMR (CDCl3) δ 7.65 (d, 2 H, J ) 8.2
Hz), 7.43 (dd, 2 H, J ) 7.5, 1.5 Hz), 7.35 (t, 2 H, J ) 7.6 Hz),
7.28 (d, 3 H, J ) 8.2 Hz), 7.21 (d, 1 H, J ) 8.7 Hz), 7.05 (d, 1 H,
J ) 15.3 Hz), 6.84 (d, 1 H, J ) 8.7 Hz), 6.72 (d, 1 H, J ) 15.3
Hz), 5.81 (d, 1 H, J ) 4.8 Hz), 4.52 (s, 1 H), 3.95 (s, 1 H), 3.87
(s, 3 H), 3.35 (m, 1 H), 2.97 (m, 1 H), 2.72 (s, 3 H), 2.54 (m, 1 H),
2.44 (s, 4 H), 2.20 (d, 2 H, J ) 10.0 Hz), 2.00 (m, 1 H), 1.67 (br
s, 1 H), 0.76 (s, 9 H), -0.14 (s, 3 H), -0.19 (s, 3 H). 100.6 MHz
13C NMR (CDCl3) δ, 152.4, 144.3, 143.0, 135.1, 134.7, 130.6,
130.4, 130.3, 129.9, 129.5, 129.1, 127.4, 127.0, 124.6, 121.4, 119.4,
111.3, 79.2, 67.3, 55.8, 50.0, 35.1, 33.4, 25.8, 25.7, 25.1, 21.5, 20.8,
18.0, -5.1, -5.2. HRMS (FAB/NBA) for C37H48O5N79BrS2Si:
calcd, 757.1925; found, 757.1984.
(+)-Tetracyclic Sulfonamide 16. Application of the above
procedure to alcohol (-)-15b afforded a yellow oil, [R]24 +10.3
D
(c ) 0.007 in CHCl3), in 30% yield.
(()-Dihydroisocodeine (17). A 10 mg (1.44 mmol) amount of
lithium metal was added to a solution of THF-anhydrous ammonia
(10 mL) containing 20 µL (0.21 mmol) of t-BuOH at -78 °C. The
resulting dark blue solution was stirred at -78 °C for 5 min, and
10 mg (0.022 mmol) of tetracyclic styrene (()-16 in THF (0.5 mL)
was then added dropwise. The initial blue color was discharged
after the addition of (()-16, and an additional 5 mg (0.72 mmol)
of lithium metal was added until the blue color became persistent.
After stirring at -78 °C for 10 min the reaction was quenched
with a MeOH-aqueous NH4Cl solution, and the resulting mixture
was extracted with CH2Cl2 (20 mL). After the aqueous phase was
washed with CH2Cl2 (3 × 15 mL) the combined organic solution
was dried over Na2SO4 and concentrated. Purification by standard
acid-base extraction afforded 5.6 mg (85%) of a white solid, mp
134-136 °C. IR (CH2Cl2) 3590 (s), 3370 (br), 3051, 1633, 1603,
1504, 1443, 1087, 1057 cm-1. 400 MHz 1H NMR (CDCl3) δ 6.73
(d, 1 H, J ) 8.2 Hz), 6.64 (d, 1 H, J ) 8.2 Hz), 4.36 (d, 1 H, J )
6.6 Hz), 3.87 (s, 3 H), 3.44 (m, 1 H), 3.14 (br s, 1 H), 3.01 (d, 1
H, J ) 18.4 Hz), 2.54 (d, 1 H, J ) 12.0 Hz), 2.43 (s, 3 H), 2.38
(dd, 1 H, J ) 17.9, 4.6 Hz), 2.26-2.17 (m, 2 H), 1.91-1.80 (m,
2 H), 1.71 (dd, 1 H, J ) 12.5, 2.1 Hz), 1.59 (m, 1 H), 1.37 (q, 1
H, J ) 12.7 Hz), 0.98 (q, 1 H, J ) 13.0 Hz). 100.6 MHz 13C NMR
(-)-Aryl Ether (15a). The above procedure was applied to (+)-
25 to afford a colorless oil, [R]24 -52.9 (c ) 0.07 in CHCl3), in
D
83% yield.
Cyclization Substrate (()-15b. A solution of 605 mg (0.8
mmol) of the silyl ether (()-15a in 10 mL of CH3CN was treated
(33) We prepared (-)-dihydrocodeinone in two steps from natural
codeine. Codeine was first hydrogenated to (-)-dihydrocodeine (Rapoport,
H.; Payne, G. B. J. Org. Chem. 1950, 15, 1093), followed by a Swern
oxidation (DMSO/(COCl)2) to yield dihydrocodeinone.
(34) Recrystallized (-)-dihydroisocodeine has an [R]24D -136, see refs
32b,c. Enantiopure (-)-dihydrocodeinone is crystalline, mp 193.5-194.5
°C, and has an [R]24D -203, see: Hong, C. Y.; Kado, N.; Overman, L. E.
J. Am. Chem. Soc. 1993, 115, 11028.
454 J. Org. Chem., Vol. 71, No. 2, 2006

EnantioselectiVe Synthesis of (-)-Dihydrocodeinone
(CDCl3) δ 144.1, 143.6, 130.3, 126.2, 119.0, 113.4, 97.1, 73.2, 59.5,
56.4, 47.0, 43.0, 42.7, 42.6, 35.3, 30.0, 23.6, 20.2. HRMS (CI) for
C18H23O3N (M+): calcd, 301.1677; found, 301.1690.
3 H), 2.45 (s, 5 H), 2.30 (d, 1 H, J ) 4.5 Hz), 2.18 (m, 2 H),
1.92-1.60 (m, 4 H). 100.6 MHz 13C NMR (CDCl3) δ 143.4, 137.4,
134.8, 129.7, 127.3, 125.0, 71.0, 47.6, 35.6, 35.0, 32.0, 31.9, 21.5,
18.1. MS (CI, NH3) m/e: 386 [M+ - H] for 79Br.
(-)-Dihydroisocodeine (17). The above procedure was applied
to sulfonamide (+)-16 to afford a 70-80% yield of (-)-dihy-
droisocodeine as a yellow oil, [R]24D -75.9 (c ) 0.003 in CHCl3).
Alcohol (R)-32. A 0.70 g (1.82 mmol) amount of bromoenone
31 in toluene (5 mL) was added to a solution of 4.36 mmol of
freshly prepared (S)-oxazaborolidine29c in 35 mL of toluene at -78
°C. The resulting mixture was stirred at -78 °C for 10 min, and
4.40 mL (4.40 mmol) of catechol borane 1.0 M in THF was then
added. The reaction mixture was warmed gradually to room
temperature overnight, under stirring. Water was added to quench
the hydride excess, and the organic phase was separated. The
organic phase was first washed with 10% HCl and then 10% NaOH,
dried over Na2SO4, and concentrated to produce the crude alcohol.
Purification by flash chromatography on silica gel with EtOAc-
Hex-acetone (1:3:1) afforded 590 mg (84%) of a colorless viscous
Acknowledgment. This research was supported by the
National Science Foundation and the National Institutes of
Health. NMR spectra were acquired with a spectrometer
purchased with funds from the National Science Foundation.
Mass spectra were acquired with a spectrometer purchased with
funds from the Division of Research Resources of the National
Institutes of Health. For part of the period during which this
work was carried out, K.A.P. held an NSF Career Advancement
Award.
Supporting Information Available: General methods, experi-
mental details for compounds 19-31, and a general procedure for
the preparation of MTPA esters; copies of 1H and 13C NMR spectra
(PDF) of all new compounds. This material is available free of
charge via the Internet at http://pubs.acs.org.
oil (82% ee by Mosher ester analysis), [R]24 +32.2 (c ) 0.03 in
D
CHCl3). IR (CH2Cl2) 3589, 1649 (w), 1596, 1091 cm-1. 250 MHz
1H NMR (CDCl3) δ 7.68 (d, 2 Η, J ) 8.3 Hz), 7.33 (d, 2 H, J )
8.0 Hz), 4.25 (d, 1 H, J ) 4.2 Hz), 3.25-2.98 (m, 2 H), 2.78 (s,
JO0513008
J. Org. Chem, Vol. 71, No. 2, 2006 455