Organocatalytic, Dioxirane-Mediated C-H Hydroxylation under Mild Conditions Using Oxone

Article abstract of DOI:10.1021/acs.orglett.7b02178

Dioxiranes are among the most selective and useful reagents for C(sp³)-H hydroxylation, but the development of a general dioxirane-mediated catalytic method has been an elusive goal. A trifluoromethyl ketone catalyst in combination with Oxone is shown to enable the first dioxirane-mediated catalytic hydroxylations that approximate the reactivity and selectivity of isolated dioxiranes. The mild reaction conditions allow for selective 3° hydroxylation and 2° oxidation and are tolerant of acid-sensitive functionality and electron-neutral arenes.

Full text of DOI:10.1021/acs.orglett.7b02178

Letter
pubs.acs.org/OrgLett
Organocatalytic, Dioxirane-Mediated C−H Hydroxylation under Mild
Conditions Using Oxone
William G. Shuler, Shea L. Johnson, and Michael K. Hilinski*
Department of Chemistry, University of Virginia, Charlottesville, Virginia 22904-4319, United States
*
S Supporting Information
ABSTRACT: Dioxiranes are among the most selective and useful
3
reagents for C(sp )−H hydroxylation, but the development of a
general dioxirane-mediated catalytic method has been an elusive
goal. A trifluoromethyl ketone catalyst in combination with
Oxone is shown to enable the first dioxirane-mediated catalytic
hydroxylations that approximate the reactivity and selectivity of
isolated dioxiranes. The mild reaction conditions allow for
selective 3° hydroxylation and 2° oxidation and are tolerant of
acid-sensitive functionality and electron-neutral arenes.
he appeal of precise control of selectivity in the
Scheme 1. C−H Hydroxylation by Dioxiranes: Prior Art
T
conversion of C−H bonds to more synthetically tractable
functional groups has spurred investigations into general
catalytic methods for intermolecular, site-selective aliphatic
1
,2
3
hydroxylation.
Transition-metal complexes and organo-
4
catalysts that are capable of effecting oxygen atom transfer
have been investigated for this purpose, resulting in
considerable recent advances in the ability to selectively oxidize
3
one of many unactivated C(sp )−H bonds, even on highly
complex, druglike substrates. While these developments have
opened the door to the prospect of C−H hydroxylation as a
routine synthetic transformation, significant challenges remain
in addressing issues of regioselectivity, stereoselectivity, and
functional group compatibility that necessitate the development
of new catalytic strategies.
Some of the most extensively studied reagents capable of
selective C−H hydroxylation are dioxiranes (Scheme 1a),
particularly dimethyldioxirane (DMDO) and its more reactive
5
derivative methyl(trifluoromethyl)dioxirane (TFDO). The
high degree of site selectivity they demonstrate in the oxidation
of complex molecules makes them paradigmatic reagents for
1
1
5
,6
possibility, and success in achieving organocatalytic hydrox-
4
this purpose,
but for many applications their use is
c−e
ylation with structurally related oxazridine
and oxaziridi-
impractical, given that they must be prepared in advance as
4a
7
nium oxidants, to date no intermolecular methods that are
catalytic in ketone and that replicate the broad substrate scope
and site selectivity of TFDO and DMDO have been disclosed.
This creates a substantial barrier to pursuing tunable site-
selective and enantioselective organocatalytic methods for C−
H hydroxylation.
dilute solutions under carefully controlled conditions.
Methods for C−H hydroxylation that involve the in situ
generation of TFDO from 1,1,1,-trifluoroacetone avoid this
complication but require an excess amount of the highly volatile
8
parent ketone to achieve synthetically useful conversion. In
contrast, oxidation of more reactive functional groups can be
achieved using a catalytic amounts of ketone. In particular, the
development of dioxirane-mediated catalytic epoxidations by
Shi, Denmark, and Yang helped to usher in the advent of
We have recently reported the first examples of ketone-
4b
catalyzed C−H hydroxylation (Scheme 1b). While that study
established that catalytic turnover in a dioxirane-mediated C−H
hydroxylation could indeed be achieved, hydroxylation of
aliphatic C−H bonds was limited to highly reactive adamantane
9
modern organocatalysis two decades ago. The considerable
advantages of ketone catalysis as applied to enantioselective
1
0
epoxidations, in particular, illustrate the appeal of expanding
this platform to include catalytic enantioselective C−H
hydroxylation. Despite a longstanding interest in this
Received: July 17, 2017
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.7b02178
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
a
and decalin substrates. In addition, a large excess of 50%
Table 1. Optimization of the Reaction Conditions
aqueous H O2 (16 equiv), high temperature, and acidic
2
reaction conditions were required, likely precluding the
possibility of developing enantioselective and functional
group tolerant transformations. This has led us to seek new
strategies that would enable ketone catalysis as a widely
applicable platform for hydroxylation. We report herein
substantially milder reaction conditions using Oxone as the
terminal oxidant that allow for greatly expanded substrate
scope. Reactivity of this catalytic system is analogous to that
b
c
entry catalyst solvent adjustment
additive (5 mol %) yield of 2 (%)
5
observed using isolated TFDO and thereby reduces the need
1
2
3
4
5
6
7
8
9
1
3
4
5
6
6
6
6
6
6
6
4
49
13
61
for onerous reagent preparation and storage. Our findings also
expand the range of competent ketone catalysts to include alkyl
trifluoromethyl ketones in addition to the previously reported
4b
aryl trifluoromethyl ketones, enabling greater flexibility in
catalyst design to address current challenges in selective
hydroxylation.
18-crown-6
76
80
26
9
nBu NHSO4
4
5:1 HFIP/H O
nBu NHSO4
2
4
Although Oxone is a typical oxidant used for in situ
1:5 HFIP/H O
nBu NHSO4
2
4
12
generation of dioxiranes and has the advantages of being
inexpensive, safe, and environmentally benign, our previous
investigations were structured to deliberately avoid its use. Our
hypothesis was that the demonstrated instability of dioxiranes
1.5:1 TFE/H O
nBu NHSO4
40
25
2
4
0
1.5:1 CH CN/H O
nBu NHSO4
3
2
4
a
Reaction conditions: 0.1 mmol of 1, 0.02 mmol of catalyst, 0.1 mmol
of Oxone, 0.4 mmol of NaHCO , 0.04 M in 1.5:1 HFIP/H O at 4 °C
3
2
1
3
b
in its presence contributed to earlier unsuccessful attempts to
for 24 h unless otherwise noted. The solvent mixture noted replaced
11
achieve catalysis of intermolecular hydroxylation. Ultimately,
this led to the discovery that catalytic turnover could be
achieved when 50% aqueous H O was used as the terminal
1.5:1 HFP/H O while maintaining a concentration of 0.04 M in 1.
Corrected GC yield using dodecane as an internal standard.
2
c
2
2
4b
a
oxidant at elevated temperature (70 °C). Importantly, the use
of 1,1,1,3,3,3-hexafluoro-2-propanol (HFIP) as a cosolvent was
essential to realize the highest yields and lowest catalyst
Table 2. Portionwise Addition of Reagents Improves Yield
8
loading. In related work, Du Bois has reported that N-
sulfonyloxaziridines, which are structurally and mechanistically
related to dioxiranes, are stabilized in aqueous fluoro alcohol
4
c,8
solvent mixtures. In light of these previous observations, we
now report that the use of HFIP as a cosolvent can enable
dioxirane-mediated catalysis in the presence of Oxone.
Initial investigations into the hydroxylation of cis-decalin (1)
revealed that catalytic turnover using Oxone (1 equiv relative to
Oxone
NaHCO
(equiv)
total reaction
time (h)
yield of
3
b
(
equiv)
addition protocol
8
(%)
1
2
2
4
8
8
single portion
single portion
two portions
24
24
48
46
46
73
4
equiv of NaHCO ) could be achieved at 4 °C using 20 mol %
3
(
t = 0, 24 h)
three portions
t = 0, 24, 48 h)
catalyst loading and a 1.5:1 mixture of HFIP/H O as the
2
c
3
12
72
85
solvent (Table 1, entries 1−4). DMDO precursor acetone gave
only trace conversion, but TFDO precursor 1,1,1-trifluoroace-
tone (4) and commercially available 1,1,1-trifluorohexan-2-one
(
a
Reaction conditions unless otherwise noted: 0.1 mmol of 7, 0.02
mmol of 6, 0.04 M in 1.5:1 HFIP/H O using amounts of oxidant,
base, and addition protocol noted. Corrected GC yield using
dodecane as an internal standard unless otherwise noted. Isolated
2
(6) gave promising initial results. Interestingly trifluoroaceto-
b
c
phenones, which demonstrated superior catalytic ability using
our previously reported protocol, were inferior to the alkyl
trifluoromethyl ketone catalysts (entry 3). Ketone 6 was chosen
for further investigation due to superior performance as well as
ease of handling (boiling point of 90 °C versus 22 °C for 4).
Optimal yields were achieved with addition of 5 mol % of a
tetraalkylammonium salt phase-transfer catalyst (entry 6). As
shown by entries 7−10, both the choice of organic solvent and
its ratio to water are critical. 2,2,2-Trifluoroethanol (TFE) was
inferior to HFIP, and the use of a typical cosolvent for
yield on 0.5 mmol scale.
with portionwise addition of Oxone and NaHCO3 gave
substantially improved conversion. Ultimately, portionwise
addition of 3 equiv of Oxone and 12 equiv of NaHCO over
3
72 h was found to be optimal.
Investigations of the substrate scope of this new catalytic
method revealed a strong preference for 3° hydroxylation,
characteristic of dioxirane reactivity (Table 3). As expected,
regioselectivity is strongly influenced by substituent effects, with
products of 3° hydroxylation remote to electron-withdrawing
groups strongly preferred when multiple sites of oxidation are
10
dioxirane-mediated oxidations, acetonitrile, did not result in
catalytic turnover.
For less reactive substrates, up to 3 equiv of Oxone were
required to achieve the highest yields. This necessitated the
5
available (Table 3, entries 5−7). Ester, primary ether, and
13
development of a portionwise addition protocol (Table 2).
imide groups are tolerated (Table 3, entry 1). Cyclic substrates
can be selectively hydroxylated in up to 98% yield. The remote
oxidation of cyclic substrate 31, rather than the unusual
proximal hydroxylation we observed for our first-generation
method, supports our earlier conclusion that proximal
Hydroxylation of 3,7-dimethyloctyl acetate (7) using a single
equivalent of Oxone gave only a modest yield (46%). Doubling
the amount of oxidant and base gave no improvement in yield
after 24 h, with no additional conversion observed thereafter.
However, increasing the reaction time to 48 h in combination
4
b
hydroxylation is specific to trifluoroacetophenone catalysts.
B
DOI: 10.1021/acs.orglett.7b02178
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
a
Table 3. C−H Hydroxylation Catalyzed by Ketone 6
Using the latter method, hydroxylation proceeds in 33% yield
for 7 and 34% yield for 9, compared to 85% and 71%
respectively using the conditions outlined in Table 3.
Substantial improvement in the ratio of 3° hydroxylation at
the position remote rather than proximal to the ester were also
observed for our catalytic conditions. For substrate 7, the
stoichiometric TFDO method provides the product in a 4:1
ratio favoring product 8 over the product of hydroxylation of
the proximal 3° C−H bond, compared to a ratio of ≥28:1 using
18
our catalytic conditions. For substrate 9, product 10 was
generated in quantitative yield based on recovered starting
material, indicating complete selectivity for remote over
proximal hydroxylation (compared to 7:1 selectivity observed
by Wong). This high degree of selectivity is consistent with
selectivities observed for other hydroxylation reactions that
1
9
employ the strong hydrogen bond donor HFIP as a
4a
cosolvent, which might arise through deactivation of the
proximal C−H bond as a result of H-bond donation by HFIP
to the ester. Confirming that these high selectivities are
primarily due to the modified conditions rather than the bulkier
catalyst, the use of 4 as a TFDO precursor under our catalytic
conditions gave a ratio of ≥19:1 favoring the remote
20
hydroxylation product for substrate 7 (41% yield).
The mild conditions identified for catalytic hydroxylation
lead to advantages in functional group tolerance over other
catalytic methods. For example, under the essentially neutral
reaction conditions (pH approximately 7.5) no cleavage of a
pH-sensitive 1° TBS ether is observed (product 12). In
comparison, representative catalysts capable of site-selective
modification of complex molecules, such as Du Bois’
4c
benzoxathiazine organocatalyst and White’s nonheme iron
3g
catalyst, require acidic reaction conditions. Attempts to
hydroxylate substrate 11 using these catalysts led to a
substantial degree of silyl ether cleavage and at most only a
20
trace amount of hydroxylation product 12. Further oxidation
of the liberated primary alcohol to the corresponding aldehyde
4
a
and carboxylic acid was also observed in trace amounts.
3
Incompatibility of catalytic C(sp )−H hydroxylation conditions
with electron-rich or electron-neutral arenes has also been
noted. Using catalyst 6, benzylic oxidation of substrate 33 is
preferred over arene oxidation, illustrating an additional
advantage of this approach over previously reported catalytic
1
,4c
a
Reaction conditions: 0.5 mmol of substrate, 1.5 mmol of Oxone, and
21
6
mmol of sodium bicarbonate added in three portions, 0.025 mmol of
hydroxylations.
b
nBu HSO , 0.04 M in 1.5:1 HFIP/H O, 4 °C. Isolated yield after
chromatography unless otherwise noted. 0.75 mL of CH Cl added to
improve substrate solubility. Corrected GC yield. Isolated yield 41%
due to product volatility. Corrected GC yield.
A highly valued characteristic of dioxirane reactivity is the
retention of a substantial degree of site selectivity when used for
the late-stage oxidation of highly complex, biologically relevant
4
4
2
c
2
2
d
e
6
substrates. To evaluate whether this characteristic is
maintained using this new catalytic protocol, we attempted
the hydroxylation of bile acid derivative 39 and observed
selective 3° hydroxylation at the A/B cis ring junction that
Importantly, retention of configuration was observed, suggest-
ing no deviation from the current mechanistic understanding of
8
,14
22
dioxirane hydroxylations.
A strong preference for benzylic
occurred with retention of configuration (Scheme 2). No
rather than aliphatic hydroxylation was revealed by substrate
other products were observed in more than trace amounts. This
steroidal substrate bearing 36 aliphatic C−H bonds (including
5
3
3. Consistent with the reactivity of isolated TFDO, oxidation
of unactivated 2° C−H bonds is also possible, as demonstrated
for both norbornane and cyclohexane (Table 3, entries 10 and
Scheme 2. Application to Late-Stage Hydroxylation
1
1). Overall, oxidations using this catalytic method generate
15
products consistent with dioxirane reactivity and provide
isolated products in moderate to excellent yields with a
16
substrate scope that approximates that of TFDO.
Comparison of results observed for substrates 7 and 9 reveal
advantages of this catalytic method over the use of
stoichiometric amounts of TFDO generated in situ from
17
1,1,1-trifluoroacetone as reported by Wong and co-workers.
C
DOI: 10.1021/acs.orglett.7b02178
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
six 3° centers) illustrates that substrate- and catalyst-controlled
selectivity can combine to make this a useful method for late-
stage functionalization.
2014, 16, 6504−6507. (c) Adams, A. M.; Du Bois, J. Chem. Sci. 2014,
5
, 656. (d) Litvinas, N. D.; Brodsky, B. H.; Du Bois, J. Angew. Chem.,
Int. Ed. 2009, 48, 4513−4516. (e) Brodsky, B. H.; Du Bois, J. J. Am.
Chem. Soc. 2005, 127, 15391−15393.
In summary, we have demonstrated that ketone-catalyzed
C−H hydroxylation can be achieved using a combination of
Oxone and an aqueous fluoro alcohol solvent system. The
substrate scope achieved using this catalytic protocol rivals that
observed using typical stoichiometric or superstoichiometric
dioxirane protocols, which can require large excesses of
difficult-to-handle reagents. Compared to our previously
reported method, the inclusion of acyclic substrates and
realization of 2° C−H bond oxidation greatly expand the
range of substrates amenable to ketone-catalyzed hydroxylation.
In addition, the mild reaction conditions at essentially neutral
pH and decreased temperature led to initial observations of
compatibility with sensitive functionality. More importantly,
this catalytic protocol overcomes a significant limitation in
dioxirane chemistry that had previously prevented the develop-
ment of ketone catalysts to address issues of stereoselectivity,
regioselectivity, and chemoselectivity in intermolecular C−H
oxidation. Future efforts will focus on catalyst design to address
these limitations and to further expand the utility of
organocatalytic aliphatic oxidations in synthesis.
(
(
5) Curci, R.; D’Accolti, L.; Fusco, C. Acc. Chem. Res. 2006, 39, 1−9.
6) For selected recent examples, see: (a) Kawamura, S.; Chu, H.;
Felding, J.; Baran, P. S. Nature 2016, 532, 90−93. (b) Michaudel, Q.;
Journot, G.; Regueiro-Ren, A.; Goswami, A.; Guo, Z.; Tully, T. P.;
Zou, L.; Ramabhadran, R. O.; Houk, K. N.; Baran, P. S. Angew. Chem.,
Int. Ed. 2014, 53, 12091−12096. (c) Wender, P. A.; Hilinski, M. K.;
Mayweg, A. V. W. Org. Lett. 2005, 7, 79−82.
(7) (a) Mikula, H.; Svatunek, D.; Lumpi, D.; Glo
Hametner, C.; Frohlich, J. Org. Process Res. Dev. 2013, 17, 313−316.
b) Adam, W.; van Barneveld, C.; Golsch, D. Tetrahedron 1996, 52,
377−2384.
8) Zou, L.; Paton, R. S.; Eschenmoser, A.; Newhouse, T. R.; Baran,
̈
cklhofer, F.;
̈
(
2
(
P. S.; Houk, K. N. J. Org. Chem. 2013, 78, 4037−4048.
(9) MacMillan, D. W. C. Nature 2008, 455, 304−308.
(10) Wong, O. A.; Shi, Y. Chem. Rev. 2008, 108, 3958−3987.
(11) Yang, D.; Wong, M. K.; Wang, X. C. J. Am. Chem. Soc. 1998,
120, 6611−6612.
(12) Yang, D.; Wong, M.-K.; Yip, Y.-C. J. Org. Chem. 1995, 60,
3887−3889.
(13) Denmark, S. E.; Forbes, D. C.; Hays, D. S.; DePue, J. S.; Wilde,
R. G. J. Org. Chem. 1995, 60, 1391−1407.
(
(
14) Du, X.; Houk, K. N. J. Org. Chem. 1998, 63, 6480−6483.
15) Mello, R.; Fiorentino, M.; Fusco, C.; Curci, R. J. Am. Chem. Soc.
ASSOCIATED CONTENT
Supporting Information
■
1
(
989, 111, 6749−6757.
16) In most cases, yield is limited by conversion and yields based on
starting material are high. See the Supporting Information for details.
17) Fung, Y.-S.; Yan, S.-C.; Wong, M.-K. Org. Biomol. Chem. 2012,
0, 3122−3129.
18) The product of remote hydroxylation was generated in only
*
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.7b02178.
(
1
(
Experimental details as well as spectroscopic and analytic
data for all hydroxylation products (PDF)
trace amounts. The reported ratio is a calculated minimum selectivity
based on isolated amounts of product 8 and unreacted starting
material.
(
19) Colomer, I.; Batchelor-McAuley, C.; Odell, B.; Donohoe, T. J. J.
Am. Chem. Soc. 2016, 138, 8855−8861.
20) See the Supporting Information for details.
(21) Attempted hydroxylation of 33 using either benzoxathiazine or
nonheme iron catalysts leads to a complex mixture of products. See the
Supporting Information for details.
AUTHOR INFORMATION
Corresponding Author
■
(
*
E-mail: hilinski@virginia.edu.
ORCID
Michael K. Hilinski: 0000-0003-2861-7099
Notes
(22) Recycling recovered starting material one time allowed for an
increase in the initial yield of 33% to 44%.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Acknowledgment is made to the donors of the American
Chemical Society Petroleum Research Fund for support of this
research.
REFERENCES
■
(
(
1) White, M. C. Science 2012, 335, 807−809.
2) Newhouse, T.; Baran, P. S. Angew. Chem., Int. Ed. 2011, 50,
3
(
362−3374.
3) For selected recent reports, see: (a) Milan, M.; Bietti, M.; Costas,
M. ACS Cent. Sci. 2017, 3, 196−204. (b) Font, D.; Canta, M.; Milan,
M.; Cusso, O.; Ribas, X.; Klein Gebbink, R. J. M.; Costas, M. Angew.
́
Chem., Int. Ed. 2016, 55, 5776−5779. (c) Howell, J. M.; Feng, K.;
Clark, J. R.; Trzepkowski, L. J.; White, M. C. J. Am. Chem. Soc. 2015,
1
2
2
37, 14590−14593. (d) Lee, M.; Sanford, M. S. J. Am. Chem. Soc.
015, 137, 12796−12799. (e) Oloo, W. N.; Que, L. Acc. Chem. Res.
015, 48, 2612−2621. (f) Shen, D.; Miao, C.; Wang, S.; Xia, C.; Sun,
W. Org. Lett. 2014, 16, 1108−1111. (g) Gormisky, P. E.; White, M. C.
J. Am. Chem. Soc. 2013, 135, 14052−14055. (h) McNeill, E.; Du Bois,
J. Chem. Sci. 2012, 3, 1810−1813.
(
4) (a) Wang, D.; Shuler, W. G.; Pierce, C. J.; Hilinski, M. K. Org.
Lett. 2016, 18, 3826−3829. (b) Pierce, C. J.; Hilinski, M. K. Org. Lett.
D
DOI: 10.1021/acs.orglett.7b02178
Org. Lett. XXXX, XXX, XXX−XXX