Access to dienophilic ene-triketone synthons by oxidation of diketones with an oxoammonium salt

Article abstract of DOI:10.1021/ol2030873

Here we describe the oxidation of 1,3-cyclohexanediones with 4-acetamido-2,2,6,6-tetramethylpiperidine-1-oxoammonium tetrafluoroborate (Bobbitt's salt) to generate 5-ene-1,2,4-triones in moderate-to-good (40-80%) yields. This inexpensive oxidant facilitated an unprecedented cascade of oxidation and elimination to yield novel ene-triketones. The reactivity of these products was explored in the Diels-Alder reaction and provided moderate-to-good yields of cycloaddition products. The products described in this study represent unique, densely functionalized, and versatile building blocks for the synthesis of more complex molecules.

Full text of DOI:10.1021/ol2030873

ORGANIC
LETTERS
2012
Vol. 14, No. 2
498–501
Access to Dienophilic Ene-Triketone
Synthons by Oxidation of Diketones with
an Oxoammonium Salt
Nicholas A. Eddy, Christopher B. Kelly, Michael A. Mercadante,
Nicholas E. Leadbeater, and Gabriel Fenteany*
Department of Chemistry, University of Connecticut, Storrs, Connecticut 06269,
United States
gabriel.fenteany@uconn.edu
Received November 17, 2011
ABSTRACT
Here we describe the oxidation of 1,3-cyclohexanediones with 4-acetamido-2,2,6,6-tetramethylpiperidine-1-oxoammonium tetrafluoroborate
(Bobbitt’s salt) to generate 5-ene-1,2,4-triones in moderate-to-good (40À80%) yields. This inexpensive oxidant facilitated an unprecedented
cascade of oxidation and elimination to yield novel ene-triketones. The reactivity of these products was explored in the DielsÀAlder reaction and
provided moderate-to-good yields of cycloaddition products. The products described in this study represent unique, densely functionalized, and
versatile building blocks for the synthesis of more complex molecules.
Oxoammonium salts are attractive species because they
are easily prepared, environmentally benign, and recycl-
able compounds.1,2 Similarly to other nonmetal green
oxidants,3 these oxidants are able to generate aldehydes,4
esters,5 and 1,2-diketones.6 We present here a novel reaction
of 2,2-disubstituted 1,3-cyclohexanediones through the use of
4-acetamido-2,2,6,6-tetramethylpiperidine-1-oxoammonium
tetrafluoroborate (Bobbitt’s salt, 1) to generate ene-trike-
tones, which are under-investigated but intriguing mole-
cules. This structural motif is found in certain biologically
active natural products as either the ene-triketone or a
derivative thereof.7,8 Despite the potential usefulness of
ene-triketones as densely functionalized synthons, there
are sparse accounts of their synthesis.7
(1) For reviews on oxidation with oxoammonium salts, see:(a)
Merbouh, N.; Bobbitt, J. M.; Bruckner, C. Org. Prep. Proceed. Int.
2004, 36 (1), 1–31. (b) Bobbitt, J. M.; Bruckner, C.; Merbouh, N. Org.
React. 2010, 74, 103–424.
(2) (a) Golubev, V. A.; Zhdanov, R. I.; Protsishin, I. T.; Rozantsev,
E. G. Izv. Akad. Nauk SSSR, Ser. Khim. 1970, 195, 952–955. (b) Golubev,
V. A.; Rudyk, T. S.; Sen, V. D.; Aleksandrov, A. L. Bull. Acad. Sci. USSR,
Div. Chem. Sci. 1976, 25, 744–750. (c) Bobbitt, J. M.; Guttermuth, M. C. F.;
Ma, Z.; Tang, H. Heterocycles 1990, 30, 1131–1140. (d) Weik, S.;
Nicholson, G.; Jung, G.; Rademann, J. Angew. Chem., Int. Ed. 2001,
40, 1436–1439. (e) Merbouh, N.; Bobbitt, J. M.; Bruckner, C. J.
Carbohydr. Chem. 2002, 21, 65–77. (f) Zakrzewski, J.; Grodner, J.;
Bobbitt, J. M.; Karpinska, M. Synthesis 2007, 16, 2491–2494.
(3) (a) ten Brink, G.-J.; Arends, I. W. C. E.; Sheldon, R. A. Science
2000, 287, 1636–1639. (b) Bisai, A.; Chandrasekhar, M.; Singh, V. K.
Tetrahedron Lett. 2002, 43, 8355–8357. (c) Li, C.; Zheng, P.; Li, J.;
Zhang, H.; Cui, Y.; Shao, Q.; Ji, X.; Zhang, J.; Zhao, P.; Xu, Y. Angew.
Chem. 2003, 115, 5217–5220. (d) Noyori, R.; Aoki, M.; Sato, K. Chem.
Commun. 2003, 1977–1986. (e) Zhan, B.-Z.; White, M. A.; Sham, T.-K.;
Pincock, J. A.; Doucet, R. J.; Rao, K. V. R.; Robertson, K. N.;
Cameron, T. S. J. Am. Chem. Soc. 2003, 125, 2195–2199.
We investigated the R-oxidation of 2,2-dimethyl-1,
3-cyclohexanedione (2), which has been used for the syn-
thesis of bioactive molecules.9 Our attempts to oxidize
(7) For information on the ene-triketone chromophore’s UV-visible
and IR spectroscopic properties, see:Martin, H.-D.; Kummer, M.;
Martin, G.; Bartsch, J.; Bruck, D.; Heinrichs, A.; Mayer, B.; Rover,
S.; Steigel, A.; Mootz, D.; Middelhauve, B.; Scheutzow, D. Chem. Ber.
1987, 120, 1133–1149.
(8) Kong, Z.-L.; Yu, S. C.; Dai, S. A.; Tu, C.-C.; Pan, M.-H.; Liu,
Y.-C. Helv. Chim. Acta 2011, 94, 892–896.
(9) (a) Murai, A.; Tanimoto, N.; Sakamoto, N.; Masamune, T.
J. Am. Chem. Soc. 1988, 110, 1985–1986. (b) Roy, O.; Pattenden, G.;
Pryde, D. C.; Wilson, C. Tetrahedron 2003, 59, 5115–5121.
(10) (a) Gaoni, Y.; Wenkert, E. J. Org. Chem. 1966, 31, 3809–3814.
(b) Rubottom, G. M.; Juve, H. D. J. Org. Chem. 1983, 48, 422–425.
(4) (a) Bobbitt, J. M. J. Org. Chem. 1998, 63, 9367–9374. (b) Bobbitt,
J. M.; Merbouh, N. Org. Synth. 2005, 82, 80–86.
(5) Merbouh, N.; Bobbitt, J. M.; Bruckner, C. J. Org. Chem. 2004,
69, 5116–5119.
(6) (a) Golubev, V. A.; Miklyush, R. V. Zh. Org. Khim. 1972, 8, 1376–
1377. (b) Liu, Y.; Ren, T.; Guo, Q. Chin. J. Chem. 1996, 14, 252–258.
r
10.1021/ol2030873
Published on Web 12/29/2011
2011 American Chemical Society

2 with SeO2, Pb(OAc)4, and Rubottom conditions met
with failure.10 Peroxides caused the formation of the
unwanted Baeyer-Villager lactone as the major product.
In the absence of peroxides, SeO2 in refluxing dioxane-
water failed to react. Pb(OAc)4 gave low yields of over-
acetoxylated products (<10%) after 24 h of reflux in both
benzene and cyclohexane. Seeking alternatives, we turned to
the use of oxoammonium salts, which have been reported to
promote the R-oxidation of ketones.6 Using 1, the desired
oxidation was serendipitously accompanied by alkene gen-
eration, forming the ene-triketone 3 (Figure 1).
Table 2. Scope of the Oxidation for Various Substituted
Diketonesa
Figure 1. Oxidation of 2 with Bobbitt’s salt (1).
Table 1. Optimization of the R-Oxidation Reactiona
entry solvent salt equivalents temperature (°C) yield (%)b
c
1
2
3
4
5
6
7
8
9
10
DCM
1
24
;
;
;
9d
6d
c
c
EtOAc
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
DMF
1
24
1
24
1
50
1
reflux
50
2
14
53
65
44
51
3
50
3.8
4.2
3.8
50
50
50
a Reactions were performed at 0.2 M on a 1 mmol scale. b Refers to
isolated yield. c No observable reaction. d Yield determined by NMR.
a Reactions were performed in anhydrous acetonitrile at 0.2 M and
50 °C on a 2 mmol scale. b Refers to isolated yield. c No ene-triketone
formed. d 63% recovered starting material.
We varied solvent, temperature, and oxidant stoichiom-
etry to maximize the yield of 3 (Table 1). No reaction was
observed in DCM or EtOAc (Table 1, entries 1 and 2), but
the reaction reached completion in 4 h when using anhy-
drous DMF (Table 1, entry 10) as the solvent.11 Com-
monly, silica was added to the mixture at the end of the
reaction to allow for facile separation of the desired
product from the spent oxidant. Purification of the
product from the DMF-silica slurry created complications
in successful isolation. In contrast, the use of anhydrous
MeCN (Table 1, entries 3À9) for these reactions resulted
in successful and facile purification. The optimal reaction
temperature was found to be around 50 °C, (Table 1,
entries 4 and 6À9).12 Decreased yields were obtained
both at room temperature (Table 1, entry 3) and reflux
(Table 1, entry 5). Finally, the effect of varying the
stoichiometric equivalents of oxoammonium salt was
explored. At oxidant loadings above 4 equiv, yields began
to decrease. In contrast, when using 3 equiv or less, the
reaction failed to reach completion and gave poor yields.
With 3.8 equiv of the oxoammonium salt, the reaction
reached completion and gave optimal yield (Table 1,
entry 8).13
(11) Determined by TLC and starch paper; a purple coloration to the
starch paper is indicative of the presence of oxidant.
(12) The synthesis of the ene-triketones was complished as follows:
oven-dried round-bottom flask was cooled under nitrogen, then charged
with the 1,3-cyclohexanedione (2 mmol) and anhydrous MeCN (8 mL).
Bobbitt’s salt (3.8 equiv) was added to the solution and placed into a
50 °C oil bath. The mixture was stirred for 16 h, then allowed to cool.
Silica (1À2 weight equiv) was added, and the solvent was removed in
vacuo. The material was loaded on a short pad of silica gel and eluted
with 5À10 volumes of 35% EtOAc/hexanes.
Org. Lett., Vol. 14, No. 2, 2012
499

Figure 2. Proposed mechanism for the oxidation of cyclic 1,3-diketones with Bobbitt’s salt.
Purity of the oxoammonium salt played a considerable
role in the outcome of the reaction. When 1 was allowed
to air-dry to constant weight before use, severely dimin-
ished yields of 3 (20À30%) were observed. Vacuum-
desiccated salt gave fair yields of 3 (40À50%). The best
results were obtained from recrystallized and rigorously
dried salt (60À80%).14 These results are in contrast to
oxidations of alcohols, where the oxoammonium salt needs
no further purification.4 We attribute the diminished yields
observed in our case to residual water in the salt.15
parable to that for the corresponding diketone (Table 2,
entry 1) under the same conditions. With 2j (Table 2,
entry 10), the reaction did not reach completion (ca. 95%
by NMR), but gave good yields. Decreasing the ring from
cyclohexanedione to cyclopentanedione (Table 2, entry 12)
gave excellent yields of the dehydrogenated product. This
result is noteworthy in that it shows that olefin formation
seemingly occurs faster than the subsequent oxidation.
We propose the following mechanism for the formation of
the ene-triketone products with Bobbitt’s salt (Figure 2). An
ene-like reaction17 between the enol form of the diketone
and 1provides intermediate A. This mechanism is supported
by the data from the solvent screen, in which lower polarity
solvents (e.g., DCM and EtOAc) resulted in no reaction,
whereas increasing solvent polarity (e.g., MeCN and DMF)
facilitated the reaction. This is likely due to the hydrogen-
bonding nature of these solvents, which allows for increased
concentration of the enol tautomer. A undergoes a second
oxidation to the ketone B concomitantly with the formation
of piperidine C. The mechanism of the oxidation of A to B is
unclear, although Golubev posited that the counterion can
act as a base to facilitate displacement of the piperidinium
moiety.2b In the present case, this possibility is unlikely due
to the nonbasic nature of the tetrafluoroborate anion. Thus,
there are two potential pathways through which the reaction
could occur: (1) enolization of A and rearrangement to B; (2)
oxidation by a second equivalent of Bobbitt’s salt to generate
the hydroxyamine. Our optimal conditions necessitated the
amount of salt to be greater than 3 equiv, suggesting that, in
the case of 1, the second pathway predominates in the
oxidation of A to B. Tautomerization of B to its more stable
enol form allows for a second ene-like R-oxidation providing
D that then undergoes deprotonation by C. This occurs
through either an E2 or E1cb pathway to generate the final
product, which, as formation of 3l suggests, is more thermo-
dynamically favored than a subsequent R-oxidation.
With reaction conditions optimized, we screened a
series of 1,3-cyclohexanediones to define the scope of
the reaction (Table 2). Oxidation of allylic and benzylic
methylenes was not observed (Table 2, entries 2À5, 11,
and 12), and yields for the dibenzyl diketone (Table 2,
entry 4) were reduced. The low yield of this substrate can
be attributed to steric shielding caused by one of the
benzyl moieties. This is corroborated by the observation
1
of Correa and Mainero of the H chemical shifts of the
parent compound.16 Substitution by a propargyl group
led to markedly diminished yields (Table 2, entry 7).
Disubstitution with methyl groups in the 5 position of
the cyclohexanedione (Table 2, entry 8) resulted in no
oxidation reaction, likely because of steric hindrance
disfavoring the oxidation. To examine the effect of steric
hindrance, we exposed the trimethyl (Table 2, entry 9) and
dimethylphenyl (Table 2, entry 10) analogs. In the case of
2i (Table 2, entry 9), we obtained a yield that was com-
(13) The oxoammonium salt is available commercially, and provides
the same yields as the synthesized, unpurified, and undried salt.
(14) The recrystalization of the oxoammonium salt was accom-
plished as follows: 1 was dissolved in 1.5 parts of boiling water
(solution turns black), plunged in ice and vacuum filtered. The salt
was then placed into an Abderhalden and dried under vacuum over
KOH with ethanol.
(15) The nitroxide was also considered as a likely impurity. However,
when purified 1 was poisoned with 10% (w/w) of the nitroxide, no
change in yield was observed and the solution turned black. No
observable reaction takes place between the nitroxide and 2 in MeCN,
nor is there a reaction between the nitroxide and MeCN. Only trace
reaction was observed between 1, MeCN, and the nitroxide.
(16) Correa, J.; Mainero, R. M. J. Org. Chem. 1969, 34, 2192–2195.
(17) Pradhan, P. P.; Bobbitt, J. M.; Bailey, W. F. Org. Lett. 2006, 8,
5485–5487.
500
Org. Lett., Vol. 14, No. 2, 2012

yields were good with excellent selectivity, whereas 3g
(Table 3, entry 4) resulted in both a poor yield and poor
selectivity. Substitution at the 5 position of the cyclohex-
anedione moiety dramatically affected the reaction in
terms of rate. Compound 3i did not react with cyclopen-
tadiene, but, when exposed to 20 mol % TiCl4, the
reaction progressed to provide 54% yield of the desired
adduct.19 Under Lewis acid catalysis, 3j cleanly afforded
the desired adduct. It should be noted that compounds 4b,
c, d, and g contain an additional stereogenic center
however, their NMR spectra exhibited a simple mixture
of endo and exo adducts, rather than a more complex
mixture of all possible products. We believe that the
presence of the sterically hindering groups allows for
epimeric selection of the dienophile by shielding a single
face of the ene-triketone. Shielding of this nature has been
shown previously for other chiral substrates.20
Table 3. DielsÀAlder Reaction Between Ene-Triketones and
Cyclopentadiene
In summary, we have developed a unique metal-free
oxidation of various 1,3-cyclohexanediones with Bob-
bitt’s oxoammonium salt. The reaction tolerates a
number of alkyl functionalities in the 2 position of 1,
3-cyclohexanedione but does not proceed when the 5
position is disubstituted. The proposed reaction mechan-
ism depends on enol formation and is supported by the
solvent screen data. Due to the inherent stability of the
1,2-diketone enol tautomer, further oxidation occurs;
however, activated methylene groups tend to favor elim-
ination to generate the olefin. These novel ene-triketones
reacted well with cyclopentadiene, providing DielsÀ
Alder adducts in fair-to-good yields with modest-to-good
selectivity for the endoexo diastereomers. Further study
of these ene-triketones is underway with the hope of
developing new synthetic methodologies or accessing
intermediates for natural product synthesis.
Acknowledgment. We thank Jay Richardson (University
of Connecticut) and Trevor Hamlin (University of Connect-
icut) for technical assistance, Prof. James M. Bobbitt
(University of Connecticut) for helpful discussions, and
Evonik-Degussa Corp. and TCI-America for the generous
gifts of 4-acetamido-2,2,6,6-tetramethylpiperidine-1-oxyl and
4-amino-2,2,6,6-tetramethylpiperidine, respectively. This
work was supported by the National Institutes of Health
(GM077622 to G.F.) and the National Science Founda-
tion (CAREER award CHE-0847262 to N.E.L.).
a Reaction performed in DCM at room temperature and 0.5 M on a
1 mmol scale. b Refers to isolated yield. c Ratio determined by HPLC.
d Reaction performed in toluene at 22 °C with 20 mol % TiCl4. e Reac-
tion performed in toluene at 110 °C. f Determined by NMR integrations.
To explore the reactions of the ene-triketone products,
the isolated material was reacted with cyclopentadiene.
The hope was that the ene-triketones would exhibit
selectivity similar to that of analogous benzoquinones.18
This was indeed found to be the case, as shown in Table 3.
The reactions were generally rapid, reaching completion
within 3 h in DCM at room temperature. Yields varied
based on the substituent, as did the level of diastereo-
selectivity. No change to the ratios was noted under longer
reaction times, suggesting that the kinetic endo product is
irreversible. In the case of 3a and 3b (Table 3, entries 1 and 2)
Supporting Information Available. Detailed experimen-
tal procedures and compound spectra. This material is
available free of charge via the Internet at http://pubs.acs.
org.
(20) (a) McDougal, P. G.; Rico, J. G.; VanDerveer, D. J. Org. Chem.
1986, 51, 4492–4494. (b) Tripathy, R.; Franck, R. W.; Onan, K. D. J.
Am. Chem. Soc. 1988, 110, 3257–3262. (c) Fisher, M. J.; Hehre, W. J.;
Kahn, S. D.; Overman, L. E. J. Am. Chem. Soc. 1988, 110, 4625–4633.
(d) Tripathy, R.; Carroll, P. J.; Thornton, E. R. J. Am. Chem. Soc. 1990,
112, 6743–6744.
(18) Beyler, R. E.; Sarett, L. H. J. Am. Chem. Soc. 1952, 74, 1397–
1401.
(19) (a) Pindur, U.; Lutz, G.; Otto, C. Chem. Rev. 1993, 93, 741–761.
(b) Lautens, M.; Klute, W.; Tam, W. Chem. Rev. 1996, 96, 49–92.
Org. Lett., Vol. 14, No. 2, 2012
501