1,3-Diol synthesis via controlled, radical-mediated C-H functionalization

Article abstract of DOI:10.1021/ja802491q

The invention of a method for the synthesis of 1,3-diols from the corresponding alcohols is described, via controlled, radical-mediated C-H functionalization. The sequence described herein entails near quantitative conversion to the corresponding trifluoroethyl carbamate, followed by a variant of the Hofmann-Loffler-Freytag reaction, cyclization, and hydrolysis to provide the 1,3-diols. In addition to the 10 examples presented herein, the syntheses of four natural products are facilitated by this directed oxyfunctionalization. This methodology is demonstrated to be orthogonal to other known C-H oxidations. Finally, this sequence is efficient, practical, inexpensive, and scalable. Copyright

Full text of DOI:10.1021/ja802491q

Published on Web 05/16/2008
1,3-Diol Synthesis via Controlled, Radical-Mediated C-H Functionalization
Ke Chen, Jeremy M. Richter, and Phil S. Baran*
Department of Chemistry, The Scripps Research Institute, 10550 North Torrey Pines Road,
La Jolla, California 92037
Received April 9, 2008; E-mail: pbaran@scripps.edu
Nature’s ability to functionalize C-H bonds has inspired and
motivated organic chemists for decades.1 From the vantage point
of complex molecule synthesis, such methods hold the potential to
dramatically transform the practice of retrosynthetic analysis.2 Yet,
a consistent and profound challenge for C-H oxidation remains
in achieving complete control of chemo-, regio-, and stereospeci-
ficity. This communication presents a classical, yet powerful,
approach to C-H functionalization inspired by the venerable
Hofmann-Lo¨ffler-Freytag (HLF) reaction.3 The described process
accomplishes the conversion of an alcohol into a 1,3-diol and has
been demonstrated in multiple contexts and utilized in rapid
syntheses of four natural products.
Several conventional methods are available to access 1,3-diols,
such as aldol/reduction, conjugate addition/reduction, or various
manipulations of allyl alcohols (Figure 1A). To the best of our
knowledge, no methods for the conversion of alcohols to 1,3-diols
exist, despite the intrinsic value of this hypothetical transform. Such
a conversion has the potential to expedite or at least provide a
valuable alternative route to numerous natural products and
medicinally important compounds.
The HLF reaction, well-known for its ability to intramolecularly
form C-N bonds, has been a powerful method for the direct
halogenation of C-H bonds since its inception in the 1880s.3
Indeed, the HLF reaction might be considered one of the first
directed C-H activation reactions ever reported.1 Figure 1B
Figure 1. 1,3-Diol synthesis using a modified HLF reaction.
illustrates how the principles of the HLF reaction were incorporated
into our reaction design. The proposed conversion of a carbamate
(A) into a 1,3-diol (F) via alkyl bromide C is accompanied by two
primary challenges. First, the formation of C from B requires a
1,6-hydrogen atom transfer (seven-membered transition state), while
the HLF reaction has been predominately used for the synthesis of
five-membered rings via 1,5-hydrogen atom transfer (Ha vs Hb
abstraction). Second, even if C were successfully formed, cycliza-
tion at the oxygen rather than the nitrogen of the carbamate would
be required (to form iminocarbonate D instead of cyclic carbamate
E).4,5
The first challenge was addressed by studying the effect of
carbamate structure in the conversion of B to C (the C-H activation
step). Of the N-bromocarbamates evaluated (Figure 2), most gave
very low (<30%) conversion to the corresponding alkyl bromide
(C) and returned only debrominated starting material (A) after
photolysis with a 100 W flood lamp.6 In order to generate a more
reactive N-centered radical,7,8 the trifluoroethyl carbamate was
investigated. In the end, the use of this moiety (never before
employed in an HLF reaction) was found to be essential in rendering
this process synthetically useful. Furthermore, a simple procedure
was developed for the near quantitative conversion of alcohols to
the corresponding carbamate using 1 equiv of CF3CH2NCO (see
Supporting Information for details).
Figure 2. Relative efficiency of selected N-bromocarbamates.
D. Hydrolysis with acetic acid and K2CO3 provided F in short
order.9 All intermediates in this process were isolated and fully
characterized (see Supporting Information for a detailed optimiza-
tion table). The overall transformation is shown in Scheme 1 for
substrate 1.
With optimized conditions in hand for the efficient and practical
preparation of 1,3-diols, the scope of this reaction was investigated
(Table 1). This methodology proved amenable for the directed
oxidation of a variety of simple tertiary centers (4, 6, and 10) to
provide the diols in good yields. Esters (9) and epoxides (12) are
The second challenge (conversion of C to D rather than E) was
addressed using Ag2CO3 to promote cyclization to iminocarbonate
9
10.1021/ja802491q CCC: $40.75
2008 American Chemical Society
J. AM. CHEM. SOC. 2008, 130, 7247–7249 7247

C O M M U N I C A T I O N S
Scheme 1. Synthesis of 1,3-Diol 4 from Carbamate 1a
Table 1. Scope of Directed C-H Oxidation
a Reagents and conditions: (a) CF3CH2NCO (1.0 equiv), DCM, Pyr (1.0
equiv), 23 °C, 2 h, 97%; (b) CH3CO2Br (1.0 equiv), DCM, 0 °C, 5 min;
(c) PhCF3 (0.05 M), CBr4 (1.0 equiv), 23 °C, hν, 7 min; (d) Ag2CO3 (1.25
equiv), DCM, 23 °C, 1 h; then AcOH, 15 min; (e) K2CO3 (5.0 equiv),
MeOH, 23 °C, 2 h, 69% overall. DCM ) dichloromethane, Pyr ) pyridine.
tolerated in this sequence, even though alternate side reactions can
be envisioned under these conditions. It is noteworthy that the 1,3-
diols 8, 13, and 20 (isopulegol hydrate)10,11 are obtained selectively,
despite the presence of additional tertiary centers that can participate
in the HLF reaction, a point which will be returned to shortly (vide
infra). The reaction also proved effective for the generation of 1,3-
benzylic diols (5, 7, 11, and 1812,13) in excellent yields and moderate
to good diastereoselectivities. Due to the ability of sp3 carbon
radicals to trigonalize, no diastereoselection is observed during
C-bromination; any diastereomeric ratios observed in the 1,3-diol
products arise during the silver-promoted cyclization. As a dem-
onstration of this effect, racemic compound 8 is obtained when
either racemic or enantiopure tetrahydrogeraniol is utilized in this
oxidation protocol. When a competition experiment is performed
between tertiary and benzylic centers, the benzylic 1,3-diol (14) is
the sole product obtained. Finally, in simple acyclic cases, where
both 1,5- and 1,6-hydrogen transfer can occur, no selectivity is
observed and the products of both 1,5- (15) and 1,6- (16) hydrogen
transfer are obtained, although the former cannot cyclize under these
conditions.
Certain limitations have been identified for this method: (1) only
benzylic and tertiary C-H bonds are oxidized in synthetically useful
yields, and (2) olefins,14 free carboxylic acids, amines, amides,
unprotected alcohols, and azides are not tolerated (as with most
C-H oxidations). Aside from the use of stoichiometric Ag2CO3, a
reagent used in many C-H functionalization protocols,15 this
method uses readily available reagents and a simple, scalable
protocol. The typical reaction sequence, from alcohol to 1,3-diol,
only requires 6-8 h. It should also be noted that this reaction can
easily be performed on a gram scale (10, 56% yield).
a Isolated yield. b Yield brsm. c CBr4 is not necessary. d A 56%
isolated yield on gram scale; 88% yield brsm.
been prepared by non-selective microbial oxidation (Cephalo-
sporium aphidicola) of menthol in only 4.8% isolated yield.16
Finally, triols rengyol and isorengyol (24 and 25)17,18 were
synthesized utilizing this procedure, beginning with installation of
the carbamate on the commercially available alcohol 21. Reduction
of the ketone (22) and in situ acetylation generated 23, which could
be used directly in the oxygenation reaction to give 24 and 25 in
42% yield as a 3:2 mixture of diastereomers. This synthesis
proceeds in only three steps and 36% overall yield from 21.
One of the most useful features of the transformation discussed
herein is the unique chemo- and regioselectivity19 observed during
the course of the reaction, specifically in cases when multiple
tertiary C-H bonds can be activated. Figure 3 presents a compari-
son between this method and those of Curci20 and White.2a For
example, carbamate 26 leads to diol 13 as a single regioisomer in
55% isolated yield under our conditions (via activation of C-Hc).
In comparison, the catalyst system developed by White2a selectively
activates C-Ha in a 3:1 ratio with C-Hb (42% combined yield)
and no C-Hd activation observed. Curci conditions20 (1 equiv of
TFDO, -15 °C) indiscriminately provides a complex mixture of
oxidized products in 91% combined yield. As a further example
of the orthogonal nature of this transformation, menthol carbamate
19 was investigated under the three sets of conditions. Our
conditions selectively activated C-Hc (20, 42% isolated yield), the
White conditions selectively activated C-Ha (51% isolated yield),
and TFDO (1 equiv, -15 °C) provided a complex mixture of
oxidized products in 86% combined yield. It is notable that the
current system undergoes regioselective 1,6-activation (Hc), even
To further demonstrate the utility of this method, the simple
natural products 18, 20, 24, and 25 were synthesized (Scheme 2).
Upon exposure to the conditions developed herein, carbamate 17
cleanly provided the natural product (18) in high yield and good
diastereoselectivity. Furthermore, carbamate 19 provided isopulegol
hydrate (20) efficiently in one step. It is noteworthy that 20 has
9
7248 J. AM. CHEM. SOC. VOL. 130, NO. 23, 2008

C O M M U N I C A T I O N S
Scheme 2. Total Synthesis of 15, 17, 21, and 22a
Acknowledgment. We thank Prof. M. Christina White for a
generous sample of the iron catalyst used in ref 2a. We thank Dr.
D. H. Huang and Dr. L. Pasternack for NMR spectroscopic
assistance, and Dr. G. Siuzdak for mass spectrometric assistance.
We are grateful to Bristol-Myers Squibb for a predoctoral fellowship
(J.M.R.). Financial support for this work was provided by Bristol-
Myers Squibb.
Supporting Information Available: Detailed experimental proce-
dures, copies of all spectral data, and full characterization. This material
is available free of charge via the Internet at http://pubs.acs.org.
References
(1) (a) Dyker, G., Ed. Handbook of C-H Transformations; Wiley-VCH:
Weinheim, Germany, 2005; Vols. 1-3. (b) Labinger, J. A.; Bercaw, J. E.
Nature 2002, 417, 507–514. (c) Davies, H. M. L.; Manning, J. R. Nature
2008, 451, 417–424.
(2) For insightful discussion and demonstration, see: (a) Chen, M. S.; White,
M. C. Science 2007, 318, 783–787. For related oxidations, see: (b) Kim,
C.; Chen, K.; Kim, J.; Que, L., Jr J. Am. Chem. Soc. 1997, 119, 5964–
5965.
(3) (a) Hofmann, A. W. Ber. 1883, 16, 558–560. (b) Hofmann, A. W. Ber.
1885, 18, 5–23. (c) Hofmann, A. W. Ber. 1885, 18, 109–131. (d) Wolff,
M. E. Chem. ReV. 1963, 63, 55–64. (e) Mackiewicz, P.; Furstoss, R.
Tetrahedron 1978, 34, 3241–3260. (f) For classic application of the HLF,
see: Corey, E. J.; Hertler, W. R. J. Am. Chem. Soc. 1960, 82, 1657–1668.
(4) Alanine, A. I. D.; Fishwick, C. W. G.; Szantay, C. Tetrahedron Lett. 1989,
30, 6571–6572.
(5) For the closest example of a carbamate being used in an HLF to generate
nitrogen-containing heterocycles, see: Dicks, P. F.; Glover, S. A.; Goosen,
A.; McCleland, C. W. Tetrahedron 1987, 43, 923–934.
(6) Using menthol carbamate (19), the photolysis reaction can be replaced by
thermal conditions (PhCF3, 0.1 equiv of AIBN, 70 °C, 5 min) to provide
a similar yield of 1,3-diol, without requiring the use of CBr4.
(7) For HLF-based 1,5-hydrogen abstraction using a trifluoroacetamide, see:
Reddy, L. R.; Reddy, B. V. S.; Corey, E. J. Org. Lett. 2006, 8, 2819–
2821.
a Reagents and conditions: (a) CH3CO2Br (1 equiv), DCM, 0 °C, 5 min;
then PhCF3, 23 °C, hν, 25 min; then Ag2CO3 (1.25 equiv), DCM, 23 °C,
4 h; then AcOH, 23 °C, 15 min; then K2CO3 (5.0 equiv), MeOH, 18 h,
42%; (b) Pyr (1.0 equiv), CF3CH2NCO (1.0 equiv), DCM, 0 to 23 °C, 2 h,
94%; (c) L-Selectride (1.0 equiv), THF, -78 °C, 1 h; then Pyr (10 equiv),
AcCl (5.0 equiv), DMAP (cat.), DCM, 0 °C, 30 min, 92%. DCM )
dichloromethane, THF ) tetrahydrofuran, Pyr ) pyridine.
(8) For alternate methods to increase the electrophicity of N-centered radicals,
see: (a) Johnson, R. A.; Greene, F. D. J. Org. Chem. 1975, 40, 2192–
2196. (b) Chen, Q.; Shen, M.; Tang, Y.; Li, C. Org. Lett. 2005, 7, 1625–
´
1627. (c) Francisco, C. G.; Herrera, A. J.; Mart´ın, A.; Pe´rez-Mart´ın, I.;
Sua´rez, E. Tetrahedron Lett. 2007, 48, 6384–6388.
(9) For benzylic cases, NaOMe rather than K2CO3 was required.
(10) Isolation of isopulegol hydrate: Nishimura, H.; Kaku, K.; Nakamura, T.;
Fukazawa, Y.; Mizutani, J. Agric. Biol. Chem. 1982, 46, 319–320.
(11) Isopulegol hydrate is commercially available; however, it has been
synthesized many times in from 1 to 5 steps and 4.8 to 35% yield as
mixtures of regioisomers. See Supporting Information for a full reference
list.
(12) Isolation of 1,5-diphenyl-1,3-pentanediol: Niwa, M.; Jiang, P.-F.; Hirata,
Y. Chem. Pharm. Bull. 1987, 35, 108–111.
(13) For previous syntheses of 1,5-diphenyl-1,3-pentanediol, see ref 12 and:
McDougal, P. G.; Rico, J. G. Tetrahedron Lett. 1984, 25, 5977–5980.
(14) Olefins react with acetyl hypobromite before bromination of the carbamate
can occur.
(15) Chen, X.; Goodhue, C. E.; Yu, J.-Q. J. Am. Chem. Soc. 2006, 128, 12634–
12635.
(16) Rahman, A.-u.; Yaqoob, M.; Farooq, A.; Anjum, S.; Asif, F.; Choudhary,
M. I. J. Nat. Prod. 1998, 61, 1340–1342.
(17) Isolation of rengyol (24) and isorengyol (25): Endo, K.; Hikino, H. Can.
J. Chem. 1984, 62, 2011–2014.
(18) Previous syntheses of rengyol (24) and isorengyol (25): (a) Honzumi, M.;
Kamikubo, T.; Ogasawara, K. Synlett 1998, 1001–1003. (b) Kobler, C.;
Effenberger, F. Tetrahedron 2006, 62, 4823–4828.
(19) For other examples of highly selective intramolecular C-H oxyfunction-
alization reactions, see: (a) Yang, D.; Wong, M.-K.; Want, X.-C.; Tang,
Y.-C. J. Am. Chem. Soc. 1998, 120, 6611–6612. (b) Wong, M.-K.; Chung,
N.-W.; He, L.; Wang, X.-C.; Yan, Z.; Tang, Y. C.; Yang, D. J. Org. Chem.
2003, 68, 6321–6328. (c) Francisco, C. G.; Freire, R.; Herrera, A. J.; Pe´rez-
Mart´ın, I.; Sua´rez, E. Tetrahedron 2007, 63, 8910–8920.
Figure 3. Comparison of selectivity of tertiary C-H bond activation.
though Hd (1,5-activation) is readily accessible. The observed
selectivity in this case is likely due to the conformational bias of
this system rather than a preference of the carbamoyl radical for
1,6-hydrogen transfer (c.f., 16, Table 1).
Inspired by the age-old HLF reaction and enabled by the use of
a unique trifluoroethyl-substituted carbamate, a practical method
using simple reagents has been developed for the net conversion
of an alcohol to a 1,3-diol. Mechanistic studies and efforts to expand
the scope of this transformation are underway.
(20) (a) Mello, R.; Fiorentino, M.; Fusco, C.; Curci, R. J. Am. Chem. Soc. 1989,
111, 6749–6757. For related examples of oxaziridine mediated C-H
oxidation, see: (b) DesMarteau, D. D.; Donadelli, A.; Montanari, V.; Petrov,
V. A.; Resnati, G. J. Am. Chem. Soc. 1993, 115, 4897–4898. (c) Brodsky,
B. H.; Du Bois, J. J. Am. Chem. Soc. 2005, 127, 15391–15393.
JA802491Q
9
J. AM. CHEM. SOC. VOL. 130, NO. 23, 2008 7249