Selective Carbonyl?C(sp3) Bond Cleavage To Construct Ynamides, Ynoates, and Ynones by Photoredox Catalysis

Article abstract of DOI:10.1002/anie.201611897

Carbon–carbon bond cleavage/functionalization is synthetically valuable, and selective carbonyl?C(sp³) bond cleavage/alkynylation presents a new perspective in constructing ynamides, ynoates, and ynones. Reported here is the first alkoxyl-radical-enabled carbonyl?C(sp³) bond cleavage/alkynylation reaction by photoredox catalysis. The use of novel cyclic iodine(III) reagents are essential for β-carbonyl alkoxyl radical generation from β-carbonyl alcohols, including alcohols with high redox potential (E^ox_P>2.2 V vs. SCE in MeCN). β-Amide, β-ester, and β-ketone alcohols yield ynamides, ynoates, and ynones, respectively, for the first time, with excellent regio- and chemoselectivity under mild reaction conditions.

Full text of DOI:10.1002/anie.201611897

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201611897
German Edition: DOI: 10.1002/ange.201611897
Photochemistry
3
À
Selective Carbonyl C(sp ) Bond Cleavage To Construct Ynamides,
Ynoates, and Ynones by Photoredox Catalysis
Kunfang Jia, Yue Pan, and Yiyun Chen*
Abstract: Carbon–carbon bond cleavage/functionalization is
by UV light irradiation to generate the acyl radical, however
3
À
À
synthetically valuable, and selective carbonyl C(sp ) bond
unselective OC alkyl bond cleavage usually occurs and no
cleavage/alkynylation presents a new perspective in construct-
further functionalization of the acyl radical has been repor-
ing ynamides, ynoates, and ynones. Reported here is the first
ted.[2a,b] Transition-metal catalysis requires very strong heat-
3
À
À
alkoxyl-radical-enabled carbonyl C(sp ) bond cleavage/alky-
ing for unstrained OC alkyl bond cleavage to generate the
nylation reaction by photoredox catalysis. The use of novel
cyclic iodine(III) reagents are essential for b-carbonyl alkoxyl
radical generation from b-carbonyl alcohols, including alco-
hols with high redox potential (Eopx > 2.2 V vs. SCE in MeCN).
b-Amide, b-ester, and b-ketone alcohols yield ynamides,
ynoates, and ynones, respectively, for the first time, with
excellent regio- and chemoselectivity under mild reaction
conditions.
acyl metal intermediate, and limits the chemoselectivity and
functional-group compatibility of the reaction.[2c–f]
The alkoxyl radical is a versatile reactive intermediate for
carbon–carbon bond-cleavage reactions.[3] However, the OC
À
alkyl bond cleavage/b-carbonyl functionalization of the
alkoxyl radical is unknown.[3a–c] Our group previously discov-
ered that the cyclic iodine(III) reagent[4] acetoylbenziodoxole
enables alkoxyl radical generation from cyclic and linear
alcohols by photoredox catalysis,[3f] and provides a unique
entry for studying unprecedented b-carbonyl alkoxyl radical
reactivity with b-carbonyl alcohols (Scheme 1b).[5] Ynones,
ynamides, and ynoates are versatile synthons for natural
product and heterocycles syntheses.[6] While the syntheses of
ynones are widely carried out using aldehydes and ketoacids
as acyl precursors, the syntheses of ynamides and ynoates are
challenging using the corresponding aldehyde and ketoacid
derivatives as carbamoyl and alkoxylcarbonyl precursors,
T
he carbonyl group is an important functional group in
organic molecules, of which selective manipulation is valuable
in organic synthesis. Compared to many synthetic efforts to
3
construct carbonyl C(sp ) (OC alkyl) bonds,[1] it is challeng-
À
À
ing to cleave the OC alkyl bond selectively (Scheme 1a).[2] A
traditional Norrish type-I reaction cleaves the OC alkyl bond
À
À
respectively.[1a,4e] Herein, we report the first selective carbon-
yl C(sp ) bond cleavage/alkynylation reactions of b-amide,
b-ester, and b-ketone alcohols to construct ynamides, ynoates,
3
À
and ynones, respectively (Scheme 1c).
We chose the b-amide alcohol 1 as the model substrate, as
3
À
the amide C(sp ) bond-cleavage reaction is not applicable
through either Norrish type-I or transition-metal-catalysis
reactions (Table 1).[7] We initially tested acetoylbenziodoxole
(BI-OAc; 4) with an alkynyl benziodoxole (BI-alkyne; 2) as
the radical acceptor.[8] Under blue LED (lmax = 468 Æ 25 nm)
irradiation with [Ru(bpy)3](PF6)2 as the photocatalyst, the
À
OC alkyl bond cleavage/alkynylation adduct 3 was obtained
in 34% yield (entry 1). Unfortunately, with extensive screen-
ing of solvents and additives, we could not further improve the
reaction.[9] We then turned to CIR derivatives which have not
been investigated in photoredox catalysis before.[10] Using the
F-BI-OAc derivatives 5 and 6, with an electron-withdrawing
fluoro substituent, we found an improved yield of 3 (entries 2
and 3). When 3,4-F-BI-OAc (7), having two fluoro substitu-
ents, was used, the yield increased to 61% (entry 4). With
2,3,4,5-F-BI-OH (8), bearing four fluoro substituents, the
reaction yield was satisfyingly optimized to 74% (entry 5). In
comparison, using 3,4-MeO-BI-OAc (9), having two electron-
rich methoxy groups, a decreased yield was observed
(entry 6). The noncyclic iodine(III) reagents PhI(OAc)2 and
PhI(OCOCF3)2 were ineffective (entries 7 and 8). The photo-
catalyst and light irradiation were both critical for the
reaction (entries 9 and 10).
3
À
Scheme 1. Carbonyl C(sp ) bond cleavage/functionalization for yna-
mide, ynoate, and ynone formation. TM=transition metal.
[*] K. Jia, Y. Pan, Prof. Dr. Y. Chen
State Key Laboratory of Bioorganic and Natural Products Chemistry
Shanghai Institute of Organic Chemistry
University of Chinese Academy of Sciences
Chinese Academy of Sciences
345 Lingling Road, Shanghai 200032 (China)
E-mail: yiyunchen@sioc.ac.cn
Y. Pan
Department of Chemistry, Shanghai University
99 Shangda Road, Shanghai, 200444 (China)
Supporting information for this article can be found under:
http://dx.doi.org/10.1002/anie.201611897.
Angew. Chem. Int. Ed. 2017, 56, 1 – 5
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

Angewandte
Communications
Chemie
Table 1: Optimization of the b-amide alcohol alkynylation.
Entry
Reaction conditions[a]
Conversion [%][b]
Yield [%][b]
1
2
3
4
5
6
7
8
9
BI-OAc (4)
54
69
63
78
>95
41
<5
<5
<5
<5
34
46
48
61
74 (71)
19
<5
<5
<5
<5
2-F-BI-OAc (5)
3-F-BI-OAc (6)
3,4-F-BI-OAc (7)
2,3,4,5-F-BI-OH (8)
3,4-OMe-BI-OAc (9)
PhI(OAc)2
PhI(OCOCF3)2
entry 5, no [Ru]
entry 5, no light
10
[a] Reaction conditions: 1 (0.20 mmol, 2.0 equiv), 2 (0.10 mmol,
1 equiv), Ru(bpy)3(PF6)2 (0.002 mmol, 0.02 equiv), and additives
(0.20 mmol, 2.0 equiv) in 2.0 mL dichloromethane under nitrogen with
4 W blue LED irradiation at 258C for 24 h, unless otherwise noted.
1
[b] Conversion and yields were determined by H NMR analysis, and
yields of isolated products are given within parentheses. Ar=
4-MeOC6H4. BI=benziodoxole.
We next tested different b-amide alcohols using the CIR/
photoredox catalytic system of entry 5 in Table 1. Aniline-
Scheme 2. Substrate scope with respect to b-amide, b-ester, and
b-ketone alcohols. Reaction conditions a: entry 5 in Table 1, 24 h.
Reaction condition b: entry 5 in Table 1, 5 h. Reaction condition c:
entry 1 in Table 1, 24 h.[11] Yields of isolated products were reported.
Ar=4-MeOC6H4.
derived amide alcohols with either a free NH or N-alkyl-
3
À
substituents underwent carbonyl C(sp ) bond cleavage to
deliver carbamoyl radicals smoothly, and subsequently
yielded the ynamides 10 and 11 (Scheme 2).[11] Dialkyl-
substituted amide alcohols led to the ynamides 12 and 13 in
79% and 89% yield, respectively. BI-alkynes substituted with
electron-rich methoxy and electron-deficient chloride groups
reacted well to give 14 and 15, respectively. For b-ester
alcohols which are resistant to both Norrish type-I and
transition-metal catalysis reactions,[12] the ynoates 16 and 17
were obtained, via alkoxylcarbonyl radical intermediates,
uneventfully in 70% and 73% yield, respectively. For a
b-ester alcohol derived from tert-butyl alcohol, the tert-butyl
alkyne 18 was obtained in 73% yield after both ketone and
carbon dioxide elimination.[13]
To gain mechanistic insights into the reactivity of the
b-carbonyl substrate under the CIR/photoredox catalytic
system, we mixed the b-ketone cyclohexanol 41 (peaks
marked with circles; Scheme 3a) with 4 and observed a new
set of 1H NMR signals (triangles; Scheme 3a). These charac-
teristic peaks belong to the unprecedented b-ketone cyclo-
hexanol/benziodoxole complex 42, which is an air-stable
white solid.[15] X-ray crystallography of 42 shows a long
iodine–oxygen bond within the benziodoxole moiety (I-O3 =
2.18 ꢀ), and a short iodine–oxygen bond (I-O1 = 2.01 ꢀ)
connecting the benziodoxole and the cyclohexanol moie-
The b-ketone alcohols with various electron-deficient or
electron-rich aryl substitutions were then tested
ty.[3f,4d] We next measured the oxidation potential of 42 (Eopx
=
3
(Scheme 2).[11] The carbonyl C(sp ) bond cleavage/alkynyla-
1.06 V vs. SCE in MeCN) by cyclic voltammetry and found
a significant decrease compared to that of 41 (Eopx = 1.69 V vs.
SCE in MeCN), and it explains the benziodoxole activation of
the b-carbonyl alcohols (Scheme 3b). We then tested the
CIR-alcohol complexes between different CIR derivatives
and 41. The oxidation potential of the complex 44, with two
electron-withdrawing fluoro substituents on the benziodoxole
moiety, further decreased (Eopx = 0.93 V vs. SCE in MeCN),[16]
while the oxidation potential of the complex 45, with two
electron-rich methoxy substituents, decreased to a lesser
extent (Eopx = 1.21 V vs. SCE in MeCN). The complex
between 41 and 8 was too unstable to isolate, but the optimal
À
tion adducts 19–27 were obtained selectively in 72–91%
yields without carbonyl C(sp ) bond cleavage.[14] The hetero-
2
À
cyclic N-methylindole reacted without indole oxidation to
give 28 in 67% yield. For alkyl-substituted ketone derivatives,
3
À
the carbonyl C(sp ) bond b to the alcohol was selectively
cleaved to give 29 and 30. Various aryl substituents on BI-
alkynes were compatible, including electron-rich phenyl and
methoxy groups, as well as electron-deficient fluorides,
chlorides, ketones, or nitriles (31–38). n-Hexyl- or silyl-
substituted BI-alkynes reacted smoothly to give the corre-
sponding ynones 39 and 40 in 64% and 77% yield,
respectively.
2
www.angewandte.org
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 5
These are not the final page numbers!

Angewandte
Communications
Chemie
Scheme 4. Mechanistic investigations and proposals.
above, we propose that the reaction is initiated by oxidation
3+
of the photoexcited Ru(bpy)32+* to Ru(bpy)3 by either the
CIR or its decomposing CIR radical adduct (Scheme 4c).[8f]
3+
The Ru(bpy)3 then oxidizes the b-carbonyl alcohol/CIR
complex formed in situ and releases the CIR cation (CIR+)
for a new CIR catalytic cycle. The alkoxyl radical undergoes
3
À
a b-carbonyl C(sp ) bond-cleavage reaction to yield either
the carbamoyl, alkoxylcarbonyl, or acyl radical, which does
a radical addition to the alkynyl benziodoxole to form
ynamides, ynoates, or ynones, respectively.
Scheme 3. Discovery of the cyclic iodine(III) reagent/b-ketone alcohol
complex.
In conclusion, we have developed the first regio- and
3
À
chemoselective carbonyl C(sp ) bond-cleavage/alkynylation
77% yield (74% isolated) of 43 suggested its superior
b-carbonyl alcohol activation.
reaction of b-amide, b-ester, and b-ketone alcohols[19] under
mild photoredox catalysis conditions. Novel cyclic iodine(III)
reagents are essential for alkoxyl radical generation from
b-carbonyl alcohols, including alcohols with high redox
potentials (Eopx > 2.2 V vs. SCE in MeCN). We envision this
novel photoredox catalytic system with CIRs, bearing differ-
ent electronic properties, will find various applications in
chemoselective synthesis and biomolecule studies.
With these mechanistic insights, we tested b-carbonyl
alcohols without electron-rich p-methoxyphenyl groups, as
such compounds would significantly expand the substrate
scope to alcohols with higher redox potentials (Eopx > 2 V vs.
SCE in MeCN; Scheme 3c).[3f,9] For phenyl- (48) and tolyl-
substituted b-carbonyls (49) (Eopx ꢀ 2 V vs. SCE in MeCN),
À
the use of 8 improved the OC alkyl bond-cleavage/alkyny-
lation reaction to 71–74% yield, respectively. For the alcohols
50 and 51, with very high redox potentials (Eopx ꢀ 2.2 V vs.
SCE in MeCN), yields of 51 and 57%, respectively, were
obtained when using 8.
Acknowledgments
Financial support was provided by the National Basic
Research Program of China 2014CB910304, Strategic Priority
Research Program of the Chinese Academy of Sciences
XDB20020200, and National Natural Science Foundation of
China 21472230, 21622207. We thank the SIOC X-ray
crystallography facility for assistance.
We next carried out the TEMPO radical quenching
experiment with the b-ketone alcohol 22a (Scheme 4a). The
alkynylation was inhibited and only the acyl-TEMPO adduct
52 was obtained in 21% yield.[17] When potassium persulfate
was used to replace the cyclic iodine(III) reagent, the desired
product 22 was not obtained (Scheme 4b). The cyclic iodine-
(III) reagent is catalytic in the reaction as 0.3 equivalents of 8
is effective for obtaining 22 in 77% yield. These results
collectively suggest that the b-carbonyl alcohol/CIR complex
is the key reaction intermediate, which is oxidized in the
photocatalytic system to generate the b-carbonyl alkoxyl
radical.[18] Based on the mechanistic investigations described
Conflict of interest
The authors declare no conflict of interest.
Angew. Chem. Int. Ed. 2017, 56, 1 – 5
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

Angewandte
Communications
Chemie
6547 – 6551; e) X. S. Liu, Z. T. Wang, X. M. Cheng, C. Z. Li, J.
Am. Chem. Soc. 2012, 134, 14330 – 14333; f) H. C. Huang, G. J.
Zhang, L. Gong, S. Y. Zhang, Y. Y. Chen, J. Am. Chem. Soc.
2014, 136, 2280 – 2283; g) F. Chen, A. S. K. Hashmi, Org. Lett.
2016, 18, 2880 – 2882; h) A. P. Schaffner, V. Darmency, P.
Renaud, Angew. Chem. Int. Ed. 2006, 45, 5847 – 5849; Angew.
Chem. 2006, 118, 5979 – 5981; i) J. Xie, S. Shi, T. Zhang, N.
Mehrkens, M. Rudolph, A. S. K. Hashmi, Angew. Chem. Int. Ed.
2015, 54, 6046 – 6050; Angew. Chem. 2015, 127, 6144 – 6148.
[9] See the Supporting Information for details.
[10] a) Y. F. Li, D. P. Hari, M. V. Vita, J. Waser, Angew. Chem. Int.
Ed. 2016, 55, 4436 – 4454; Angew. Chem. 2016, 128, 4512 – 4531;
b) R. D. Richardson, J. M. Zayed, S. Altermann, D. Smith, T.
Wirth, Angew. Chem. Int. Ed. 2007, 46, 6529 – 6532; Angew.
Chem. 2007, 119, 6649 – 6652.
Keywords: acyl compounds · hypervalent compounds ·
photochemistry · radicals · reaction mechanisms
[1] a) C. Chatgilialoglu, D. Crich, M. Komatsu, I. Ryu, Chem. Rev.
1999, 99, 1991 – 2069; b) G. Sartori, R. Maggi, Chem. Rev. 2006,
106, 1077 – 1104; c) M. C. Willis, Chem. Rev. 2010, 110, 725 – 748.
[2] a) R. G. W. Norrish, C. H. Bamford, Nature 1936, 138, 1016;
b) V. Ramamurthy, Tetrahedron 1986, 42, 5753 – 5839; c) K.
Ruhland, Eur. J. Org. Chem. 2012, 2683 – 2706; d) A. Dermenci,
J. W. Coe, G. B. Dong, Org. Chem. Front. 2014, 1, 567 – 581; e) F.
Chen, T. Wang, N. Jiao, Chem. Rev. 2014, 114, 8613 – 8661; f) L.
Souillart, N. Cramer, Chem. Rev. 2015, 115, 9410 – 9464.
[11] For b-ester and b-amide alcohols with free NH groups, the
reaction yields are significantly improved when using 8 for
a reaction time of 24 h (conditions a), and much lower yields are
obtained with 4 within 24 h (conditions c). For b-ketone and
b-amide alcohols without a free NH group, the reaction yields
are similar when using either 8 or 4. However, the reactions are
complete within 5 h with 8 (conditions b), and 24 h with 4
(conditions c).
[12] J. A. Pincock, Acc. Chem. Res. 1997, 30, 43 – 49.
[13] M. F. Saraiva, M. R. C. Couri, M. Le Hyaric, M. V. de Almeida,
Tetrahedron 2009, 65, 3563 – 3572.
[14] M. Tobisu in Cleavage of Carbon-Carbon Single Bonds by
Transition Metals (Eds.: M. Murakami, N. Chatani), Wiley-VCH,
Weinheim, 2016, Chapter 6.
[3] a) G. J. Rowlands, Tetrahedron 2010, 66, 1593 – 1636; b) J.
Hartung, Eur. J. Org. Chem. 2001, 619 – 632; c) J. Hartung, T.
Gottwald, K. Spehar, Synthesis 2002, 1469 – 1498; d) J. C. Zhang,
Y. , Acta Chim. Sin. 2016, DOI: 10.6023/A16080416; e) J. Zhang,
Y. Li, F. Y. Zhang, C. C. Hu, Y. Y. Chen, Angew. Chem. Int. Ed.
2016, 55, 1872 – 1875; Angew. Chem. 2016, 128, 1904 – 1907;
f) K. F. Jia, F. Y. Zhang, H. C. Huang, Y. Y. Chen, J. Am. Chem.
Soc. 2016, 138, 1514 – 1517; g) H. G. Yayla, H. Wang, K. T.
Tarantino, H. S. Orbe, R. R. Knowles, J. Am. Chem. Soc. 2016,
138, 10794 – 10797; h) C. Wang, K. Harms, E. Meggers, Angew.
Chem. Int. Ed. 2016, 55, 13495 – 13498; Angew. Chem. 2016, 128,
13693 – 13696; i) J. J. Guo, A. H. Hu, Y. L. Chen, J. F. Sun, H. M.
Tang, Z. W. Zuo, Angew. Chem. Int. Ed. 2016, 55, 15319 – 15322;
Angew. Chem. 2016, 128, 15545 – 15548.
[15] CCDC 1513742 (42) and 1520184 (44) contain the supplemen-
tary crystallographic data for this paper. These data are provided
free of charge by The Cambridge Crystallographic Data Centre.
[16] We also performed luminescence quenching experiments and
found electron-deficient cyclic iodine(III) reagents quenched
Ru(bpy)32+* luminescence more effectively; see the Supporting
Information for details. Both the decreased oxidation potential
of the CIR-alcohol complex and increased oxidative quenching
efficiency for photoexcited RuII* are responsible for the
improved reaction kinetics and yields when using 8.
[4] a) P. J. Stang, Chem. Rev. 1996, 96, 1123 – 1178; b) V. V. Zhdan-
kin, P. J. Stang, Chem. Rev. 2008, 108, 5299 – 5358; c) A.
Yoshimura, V. V. Zhdankin, Chem. Rev. 2016, 116, 3328 – 3435;
d) H. C. Huang, K. F. Jia, Y. Y. Chen, Angew. Chem. Int. Ed.
2015, 54, 1881 – 1884; Angew. Chem. 2015, 127, 1901 – 1904;
e) H. C. Huang, G. J. Zhang, Y. Y. Chen, Angew. Chem. Int. Ed.
2015, 54, 7872 – 7876; Angew. Chem. 2015, 127, 7983 – 7987; f) H.
Tan, H. J. Li, W. Q. Ji, L. Wang, Angew. Chem. Int. Ed. 2015, 54,
8374 – 8377; Angew. Chem. 2015, 127, 8494 – 8497.
[5] a) W. G. Bentrude, K. R. Darnall, J. Am. Chem. Soc. 1968, 90,
3588; b) Y. Sawaki, Y. Ogata, J. Am. Chem. Soc. 1976, 98, 7324 –
7327; c) A. S. K. Tsang, A. Kapat, F. Schoenebeck, J. Am. Chem.
Soc. 2016, 138, 518 – 526.
[17] T. Vogler, A. Studer, Synthesis 2008, 1979 – 1993.
[18] The reduction of the CIR-alcohol complex by the photoexcited
RuII* to generate an alkoxyl radical and iodobenzoate is
unlikely, as the CIR is catalytic in the reaction and the
conversion of the CIR into the iodobenzoate is irreversible.
[19] The preparations of b-amide, b-ester, and b-ketone alcohols are
straightforward and proceeds in two steps from simple starting
materials; see the Supporting Information for details.
[6] a) M. C. Bagley, C. Glover, E. A. Merritt, Synlett 2007, 2459 –
2482; b) T. J. J. Mꢁller, Top. Heterocycl. Chem. 2010, 25, 25 – 94.
[7] a) V. P. Litvinov, Advances in Heterocyclic Chemistry, Vol. 91,
Academic Press, London, 2006; b) S. Ferrini, F. Ponticelli, M.
Taddei, J. Org. Chem. 2006, 71, 9217 – 9220.
[8] a) V. V. Zhdankin, P. J. Stang, Tetrahedron 1998, 54, 10927 –
10966; b) J. P. Brand, J. Waser, Chem. Soc. Rev. 2012, 41,
4165 – 4179; c) M. Ochiai, Y. Masaki, M. Shiro, J. Org. Chem.
1991, 56, 5511 – 5513; d) V. V. Zhdankin, C. J. Kuehl, A. P.
Krasutsky, J. T. Bolz, A. J. Simonsen, J. Org. Chem. 1996, 61,
Manuscript received: December 7, 2016
Revised: January 5, 2017
Final Article published: && &&, &&&&
4
www.angewandte.org
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 5
These are not the final page numbers!

Angewandte
Communications
Chemie
Communications
Photochemistry
Aye, aye CIR: Novel cyclic iodine reagents
(CIRs) are essential for b-carbonyl alkoxyl
K. Jia, Y. Pan, Y. Chen*
&&& — &&&
radical generation from b-carbonyl alco-
3
À
hols in carbonyl C(sp ) bond cleavage/
3
À
Selective Carbonyl C(sp ) Bond Cleavage
To Construct Ynamides, Ynoates, and
Ynones by Photoredox Catalysis
alkynylation reactions under photoredox
catalysis. b-Amide, b-ester, and b-ketone
alcohols yield ynamides, ynoates, and
ynones, respectively, with excellent regio-
and chemoselectivity.
Angew. Chem. Int. Ed. 2017, 56, 1 – 5
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!