Microwave-Assisted Cp*CoIII-Catalyzed C-H Activation/Double C-N Bond Formation Reactions to Thiadiazine 1-Oxides

Article abstract of DOI:10.1021/acs.orglett.7b00120

A microwave-assisted, Cp*Co^III-catalyzed direct C-H activation/double C-N bond formation reaction of simple NH-sulfoximines with 1,4,2-dioxazol-5-ones to produce diverse thiadiazine-1-oxides is reported. The reaction tolerates a broad range of functional groups under external oxidant-free conditions and only releases CO₂ and H₂O as the sole byproducts. The preliminary mechanistic studies revealed an electrophilic metalation pathway is likely involved in the reaction.

Full text of DOI:10.1021/acs.orglett.7b00120

Letter
pubs.acs.org/OrgLett
Microwave-Assisted Cp*CoIII-Catalyzed C−H Activation/Double C−N
Bond Formation Reactions to Thiadiazine 1‑Oxides
Jiapian Huang, Yangfei Huang, Tao Wang, Qin Huang, Zhihua Wang, and Zhiyuan Chen*
Key Laboratory of Functional Small Organic Molecules, Ministry of Education, and College of Chemistry & Chemical Engineering,
Jiangxi Normal University, 99 Ziyang Road, Nanchang, Jiangxi 330022, P. R. China
S
* Supporting Information
ABSTRACT: A microwave-assisted, Cp*CoIII-catalyzed direct
C−H activation/double C−N bond formation reaction of
simple NH-sulfoximines with 1,4,2-dioxazol-5-ones to produce
diverse thiadiazine-1-oxides is reported. The reaction tolerates a
broad range of functional groups under external oxidant-free
conditions and only releases CO2 and H2O as the sole byproducts. The preliminary mechanistic studies revealed an electrophilic
metalation pathway is likely involved in the reaction.
ulfoximines represent an extremely important class of
structural motifs in biological natural products and
anilines.2,8 A typical preparation methodology for thiadiazine 1-
oxide begins with the key conversion of N-protected sulfoxides
into sulfoximines by utilizing O-(mesitylenesulfonyl)-
hydroxylamine (MSH) as the key aminating reagent (Scheme
1, eq 1).9 However, the starting sulfoxide material must be
S
pharmaceutical substances.1 Thiadiazine 1-oxides that contain
a sulfoximine moiety within benzene-fused heterocycles are
bioactive compounds with relevance as signicant pharmaceut-
ical agents (Figure 1).2 For instance, NSC287474 is a potential
Scheme 1. Reported Access to Thiadiazine 1-Oxides and
Our Methodology
Figure 1. Selected bioactive sulfoximine and thiadiazine 1-oxides.
reverse transcriptase inhibitor with excellent inhibition of HIV
replication in lymphocytes.3 Additionally, the tricyclic fused
sulfoximine, Go4962, is a partial benzodiazepine receptor
̈
agonist with excellent anxiolytic and anticonvulsive activities.4
The functionalized thiadiazine-1-oxide A has exhibited blood
pressure lowering activity in animal tests equipotent to the
marketed drug Prazo.5
Over the past decade, many noteworthy advances have been
made toward the synthesis of acyclic sulfoximine derivatives, in
particular via seminal contributions by Bolm et al.6 and among
others.7 Nevertheless, methods for the construction of cyclic
thiadiazine-1-oxide frameworks are rather limited. Traditional
pathways focused on multiple-step transformations starting
from nitroaryl thioethers or N-protected 2-thio-substituted
prepared stepwise in advance, and MSH must be handled with
extreme caution because it can decompose with explosive
violence. Vo-Thanh developed n-BuLi-promoted C−H func-
tionalization reaction of free NH-trifluoromethyl sulfoximines.
Using this reaction, further transformations could be achieved
to prepare cyclic fluorinated thiadiazine 1-oxides (Scheme 1, eq
Received: January 15, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.7b00120
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
2).10 Bolm et al. recently reported Rh-catalyzed ortho-
amidations using free NH-sulfoximines with tert-butyl (2,4,6-
trichlorobenzoyl)oxycarbamates.11 Further, a Boc-protected
amino group could be easily introduced, and the subsequent
ring-closing reactions thereby allow the production of diverse
thiadiazine 1-oxides (Scheme 1, eq 3). While the chemo-
selective synthesis of sulfoximines has been achieved, they also
produce stoichiometric amounts of organic wastes, and further
manipulations are generally necessary. Thus, the development
of atom- and step-economic strategies are still highly sought
after.
that the reaction was easily handled and not very sensitive to
moisture- or oxygen-atmosphere conditions. Thus, our results
suggest that the method may be potentially useful for industrial
applications.
The reaction scope for this convenient preparation of
thiadiazine 1-oxide 3 was next studied (Scheme 2). The
a
Scheme 2. Reaction Scope Studies
Here, we report our success in the modular one-pot and one-
step construction of thiadiazine 1-oxides with good to excellent
yields (Scheme 1, eq 4). Employing commercially available free
NH-sulfoximines 1 as the starting material and 1,4,2-dioxazol-5-
ones 2 as the amination reagent, diverse thiadiazine 1-oxide
derivatives could be efficiently generated under cobalt(III)-
catalyzed and microwave-assisted conditions. 1,4,2-Dioxazol-5-
ones are versatile, easy to handle, and have been widely used as
amination reagents for the preparation of aza-heterocycles.12
Only CO2 and H2O were generated as chemical wastes during
these reactions, which indicates this method is environmentally
benign. To our knowledge, this is the first example of
transition-metal-catalyzed, robust C−H activation/double C−
N bond formation on NH-sulfoximines by means of trans-
formable directing group strategy. The method also represents
the currently “most concise” approach to generating
thiadiazine-1-oxides.
We first screened reaction conditions in the metal-catalyzed
cyclization of NH-sulfoximine (1a) with a “Ph−CN” source
(2).13 The desired product 3aa was isolated after an extensive
survey of reaction conditions when 1,4,2-dioxazol-5-one 2a was
employed as an amination substrate and [Cp*RhCl2]2/AgSbF6
was used as the catalyst, whereas the desired product was only
generated in rather low yields (Table S1, entry 1).14 It should
be noted that no reaction occurred when using “Ph−CN”
surrogates such as 5,5-dimethyl-3-phenyl-1,4,2-dioxazole15
2a_1 or 5-phenyl-1,3,2,4-dioxathiazole 2-oxide 2a_2 as reagents
under the catalysis conditions.16 No desired product was
detected when a first-row metal catalyst, [Cp*CoCl2]2 complex,
was added instead of [Rh] (Table S1, entry 2). However,
significant improvement in product formation was obtained
when a [Cp*Co(CH3CN)3(SbF6)2] complex was employed
(Table S1, entry 4). Further screening revealed that Cp*Co-
(CO)I2 was an effective catalyst, and the GC yield of 3aa could
be improved to 50% (Table S1, entry 5). The starting material,
1a, never reached full consumption under the above-mentioned
conventional oil-bath heating conditions. However, rapid
conversion was observed when microwave irradiation con-
ditions were used, and the cyclization product 3aa was formed
with 69% yield while the reaction time was shortened to 4 h
(Table S1, entry 6).
We reasoned that the anion counterpart of the silver salt
should be essential for activation of the Co-complex. Indeed,
use of AgNTf2 offered a slight improvement (Table S1, entries
7 vs 6), whereas use of AgOTf was deleterious, possibly due to
the weak acidity of AgOTf (Table S1, entry 8). Subsequent
solvent screening demonstrated that a mixture solvent of t-
AmOH/CHCl3 = 1:1 (v/v) was optimal, led an acceptance
reaction rate in only 2 h and 79% isolated yield (Table S1, entry
9). No reaction occurred when PivOK was used in place of
PivOH, indicating that the acid additive was vital for the
transformation (Table S1, entry 10). Furthermore, we found
a
Reaction conditions: NH-sulfoximine 1a (0.20 mmol), 1,4,2-dioxazol-
5-one 2 (0.40 mmol) in 2.0 mL of solvent at 110 °C under microwave
heating conditions. Isolated yield. On 1.0 mmol scale.
b
reaction of 1a with 2a could be easily scaled up to 1.0 mmol,
and the product 3aa could be isolated in 78% yield. The
cyclization reactions of various 3-aryl-1,4,2-dioxazol-5-ones 2
bearing para electron-donating and -withdrawing groups all
reacted smoothly with 1a, resulting in moderate to high yields
(3aa−ag). Synthetically useful halo substitutions including
chloro (3ai and 3ak), bromo (3ae), and iodo (3af) groups
were all tolerated without difficulty. The production of fluoro-
(3ad) and trifluoromethyl-substituted (3ag) thiadiazine 1-
oxides through this microwave-assisted Cp*CoIII-catalyzed
cyclization reaction methodology indicates a potential utility
in generating fluorine-containing compounds. An excellent
yield of 92% (3aj) was obtained for the reaction of 1a with o-
methyl substituted 3-aryl-1,4,2-dioxazol-5-one 2j. The substrate
2l, which contains a 2-naphthyl group, underwent the reaction
smoothly to give 3al in serviceable yield, whereas the 2-thienyl
heterocycle furnished the desired product 3am with only
moderate yield. Notably, the reaction displayed excellent
functional group compatibility for alkenyl, cyclohexyl, and
alkyl groups, and the corresponding products 3an−ap were
produced in good yields.
For various substitutions affiliated on the aromatic ring of
NH-sulfoximine 1, good to excellent yields of the correspond-
ing products 3 were obtained whenever the substitutions on the
B
DOI: 10.1021/acs.orglett.7b00120
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
aromatic ring of 1 had electron-donating (3bb−bp) or
-withdrawing properties (3ca and 3cc, Scheme 3). Thieno-
Scheme 4. Control Experiments
a
Scheme 3. Reaction Scope Studies
Thus, a radical process can be excluded from the trans-
formations, which are quite different from recent theoretical
DFT calculations on CoIII-catalyzed C(sp2)−H oxygenation
reactions.18 Intermolecular competition reactions between NH-
sulfoximine 1a and the differently functionalized substrates 2c
and 2g were conducted (Scheme 4b). The results suggested
that substrate 2c, which contains an electron-donating MeO
group, preferentially participated in the cyclization, thus
indicating an intermolecular electrophilic substitution-type
C−H cobaltation event likely be involved in the key C−H
activation step.19
To investigate the possible reaction intermediates, N-Me
substituted sulfoximine 4 was synthesized. The reaction of
compound 4 with 2a produced the NH-benzoyl-substituted
sulfoximine 5 in 47% yield (Scheme 4c). This indicates that a
high-valent cobalt(III)-catalyzed ortho-C−H amidation reaction
should be involved in the reaction.20 Moreover, NH-benzoyl-
substituted sulfoximine 6 was synthesized following Luisi’s
procedure.21 Treatment of this compound with 2.0 equiv of
PivOH in t-AmylOH/CHCl3 delivered product 3aa in 94%
yield (Scheme 4c).
On the basis of the aforementioned mechanistic studies, we
propose a plausible catalytic cycle for the reaction depicted in
Scheme 5. The reaction of neutral complex Cp*Co(CO)I2 with
AgNTf2 in the presence of PivOH first generates a reactive
cationic species A, which then undergoes electrophilic metal-
ation with 1a to give a five-membered cobaltacyle B upon
release of PivOH and HNTf2. Coordination of the amidating
reagent 2a with cobaltacycle B to produce a dioxazolone-bound
cobaltacyle C is assumed, followed by subsequent amido
insertion to produce amido complex species D.22 Proto-
a
Reaction conditions: NH-sulfoximine 1 (0.20 mmol), 1,4,2-dioxazol-
5-one 2 (0.40 mmol) in 2.0 mL of solvent at 110 °C under microwave
heating conditions. Isolated yield.
1,2,4-thiadiazine derivatives were found to be potent openers of
Kir 6.2/SUR1 KATP channels in the suppression of glucose-
stimulated insulin release from rat islets.17 Interestingly, the
products 3dl, 3dh, and 3dp, which contain thieno[3,2-
e][1,2,4]thiadiazine 1-oxide heterocyclic scaffolds, were isolated
in moderate to good yields in the reaction of 2,2′-
sulfonimidoyldithiophene 1d with 1,4,2-dioxazol-5-ones 2. In
the reaction of 1-methyl-4-(phenylsulfonimidoyl)benzene 1e
with 3-phenyl-1,4,2-dioxazol-5-one 2a, two isomer products 3ea
+ 3ea′ were isolated with 71% yield with molecular ratio of 1:1.
Likewise, the reaction of 1e with 2l produced 3el + 3el′ in 66%
yield.
Limitations became clear when NH-sulfoximines 1 contained
a nonaromatic moiety (Scheme 3). No reaction occurred for
the methyl- (1f) or benzyl-functionalized (1g) NH-sulfoximines
with 2a. The vinyl-substituted NH-sulfoximine 1h was found to
decompose to fragment species in the cyclization reaction.
We conducted several control experiments to understand the
possible mechanism of this microwave-assisted, Cp*CoIII-
catalyzed C−H activation/double C−N bond formation
reaction (Scheme 4). Moderate to good yields were observed
for the reaction of 1a with 2a while in the presence of typical
radical scavengers (TEMPO, BHT and styrene) (Scheme 4a).
C
DOI: 10.1021/acs.orglett.7b00120
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 5. Proposed Mechanism
REFERENCES
■
(1) Lucking, U. Angew. Chem., Int. Ed. 2013, 52, 9399.
̈
(2) Wagner, A.; Reinohl, G. Liebigs Ann. Chem. 1964, 675, 189.
̈
(3) Buckhelt, R. W.; Fliakas-Boltz, V.; Decker, W.; Roberson, J.; Pyle,
C.; White; Bowdon, B.; McMahon, J.; Michael, R.; John, P.; David, G.;
Hubert, B.; Tammy, K. Antiviral Res. 1994, 25, 43.
(4) Bartoszyk, G.; Dooley, D.; Barth, H.; Hartenstein, J.; Satzinger, G.
J. Pharm. Pharmacol. 1987, 39, 407.
(5) Dillard, R.; Yen, T.; Stark, P.; Pavey, D. J. Med. Chem. 1980, 23,
717.
(6) (a) Cheng, Y.; Bolm, C. Angew. Chem., Int. Ed. 2015, 54, 12349.
(b) Wang, L.; Huang, H.; Priebbenow, D.; Pan, F.; Bolm, C. Angew.
Chem., Int. Ed. 2013, 52, 3478. (d) Pirwerdjan, R.; Becker, P.; Bolm, C.
Org. Lett. 2016, 18, 3307. (e) Teng, F.; Cheng, J.; Bolm, C. Org. Lett.
2015, 17, 3166. (f) Pirwerdjan, R.; Becker, P.; Bolm, C. Org. Lett.
2015, 17, 5008. (g) Pirwerdjan, R.; Priebbenow, D.; Becker, P.;
Lamers, P.; Bolm, C. Org. Lett. 2013, 15, 5397. (h) Chen, X.; Wang,
L.; Frings, M.; Bolm, C. Org. Lett. 2014, 16, 3796. (i) Priebbenow, D.;
Becker, P.; Bolm, C. Org. Lett. 2013, 15, 6155. (j) Wang, J.; Frings, M.;
Bolm, C. Chem. - Eur. J. 2014, 20, 966.
(7) (a) Chen, Z.; Huang, J.; Wang, Z. J. Org. Chem. 2016, 81, 9308.
(b) Zenzola, M.; Doran, R.; Degennaro, L.; Luisi, R.; Bull, J. A. Angew.
Chem., Int. Ed. 2016, 55, 7203. (c) Shen, X.; Zhang, W.; Ni, C.; Gu, Y.;
Hu, J. J. Am. Chem. Soc. 2012, 134, 16999. (d) Rit, R.; Yadav, M.;
Ghosh, K.; Shankar, M.; Sahoo, A. Org. Lett. 2014, 16, 5258. (e) Miao,
J.; Richards, N.; Ge, H. Chem. Commun. 2014, 50, 9687. (f) Zou, Y.;
Xiao, J.; Peng, Z.; Dong, W.; An, D. Chem. Commun. 2015, 51, 14889.
(8) Williams, T.; Cram, D. J. Am. Chem. Soc. 1971, 93, 7333.
(9) Dillard, R.; Yen, T.; Stark, P.; Pavey, D. J. Med. Chem. 1980, 23,
717.
decobaltation of D delivers the key intermediate 6, along with
the regeneration of the reactive metal catalyst A to complete
the catalytic cycle. Finally, dehydration and cyclization reaction
of compound 6 which is facilitated by PivOH would produce
the product 3aa.
In summary, we have developed a novel, microwave-assisted,
Cp*CoIII-catalyzed C−H activation/double C−N bond for-
mation reaction of free NH-sulfoximines with 1,4,2-dioxazol-5-
ones to produce diverse thiadiazine 1-oxides in moderate to
excellent yields. No external oxidants were required, and only
stoichiometric CO2 and H2O were generated as the major
chemical wastes. This novel methodology thus represents an
environmentally benign and straightforward avenue to
synthesize medicinally significant thiadiazine 1-oxides. The
reaction mechanism has been proposed based on control
experiments. Further applications of this air- and moisture-
tolerant reaction and more detailed mechanistic investigations
are currently underway.
́
(10) Le, T.; Diter, P.; Pegot, B.; Bournaud, C.; Toffano, M.; Guillot,
R.; Vo-Thanh, G.; Magnier, E. Org. Lett. 2016, 18, 5102.
(11) Cheng, Y.; Dong, W.; Wang, H.; Bolm, C. Chem. - Eur. J. 2016,
22, 10821.
(12) (a) Wang, H.; Lorion, M.; Ackermann, L. Angew. Chem., Int. Ed.
2016, 55, 10386. (b) Wang, X.; Lerchen, A.; Glorius, F. Org. Lett.
2016, 18, 2090. (c) Wang, F.; Wang, H.; Wang, Q.; Yu, S.; Li, X. Org.
Lett. 2016, 18, 1306. (d) Wang, J.; Zha, S.; Chen, K.; Zhang, F.; Song,
C.; Zhu, J. Org. Lett. 2016, 18, 2062.
(13) See the Supporting Information for details.
(14) CCDC 1517010 (3aa) contains the supplementary crystallo-
graphic data for this paper. These data can be obtained free of charge
from the Cambridge Crystallographic Data Centre via www.ccdc.cam.
ac.uk/data_request/cif.
(15) Couturier, M.; Tucker, J.; Proulx, C.; Boucher, G.; Dube, P.;
Andresen, B.; Ghosh, A. J. Org. Chem. 2002, 67, 4833.
(16) Chen, M.; Sun, N.; Chen, H.; Liu, Y. Chem. Commun. 2016, 52,
6324.
(17) Nielsen, F.; Ebdrup, S.; Jensen, A.; Ynddal, L.; Bodvarsdottir, T.;
Stidsen, C.; Worsaae, A.; Boonen, H.; Arkhammar, P.; Fremming, T.;
Wahl, P.; Kornø, H.; Hansen, J. J. Med. Chem. 2006, 49, 4127.
(18) (a) Guo, X.; Zhang, L.; Wei, D.; Niu, J. Chem. Sci. 2015, 6, 7059.
(b) For a recent review on Co-catalyzed C−H functionalization based
on CMD and SET mechanisms, see: Wei, D.; Zhu, X.; Niu, J.-L.; Song,
M.-P. ChemCatChem 2016, 8, 1242.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.7b00120.
1
Experimental details, characterization data, and H and
13C NMR spectra for new compounds (PDF)
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: zchen@jxnu.edu.cn.
ORCID
(19) Ma, W.; Mei, R.; Tenti, G.; Ackermann, L. Chem. - Eur. J. 2014,
20, 15248.
(20) Mei, R.; Loup, J.; Ackermann, L. ACS Catal. 2016, 6, 793.
(21) Zenzola, M.; Doran, R.; Degennaro, L.; Luisi, R.; Bull, J. Angew.
Chem., Int. Ed. 2016, 55, 7203.
(22) (a) Park, J.; Chang, S. Angew. Chem., Int. Ed. 2015, 54, 14103.
(b) Wang, H.; Tang, G.; Li, X. Angew. Chem., Int. Ed. 2015, 54, 13049.
(c) Bizet, V.; Buglioni, L.; Bolm, C. Angew. Chem., Int. Ed. 2014, 53,
5639.
Zhiyuan Chen: 0000-0002-5159-7287
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the National Natural Science Foundation of China
(21462022, 21562026, 21672085) and the Natural Science
Foundation of Jiangxi Province (20161BAB203084) for
financial support.
D
DOI: 10.1021/acs.orglett.7b00120
Org. Lett. XXXX, XXX, XXX−XXX