Visible-Light-Induced C-O Bond Formation for the Construction of Five- and Six-Membered Cyclic Ethers and Lactones

Article abstract of DOI:10.1021/acs.orglett.8b03166

Visible-light-induced intramolecular C-O bond formation was developed using 2,4,6-triphenylpyrylium tetrafluoroborate (TPT), which allows the regiocontrolled construction of cyclic ethers and lactones. The reaction is likely to proceed through the single-electron oxidation of the phenyl group, followed by the formation of a benzylic radical, thus preventing a competing 1,5-hydrogen abstraction pathway. Detailed mechanistic studies suggest that molecular oxygen is used to trap the radical intermediate to form benzyl alcohol, which undergoes cyclization. This new approach serves as a powerful platform by providing efficient access to valuable five- and six-membered cyclic ethers and lactones with a unified protocol.

Full text of DOI:10.1021/acs.orglett.8b03166

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
Visible-Light-Induced C−O Bond Formation for the Construction of
Five- and Six-Membered Cyclic Ethers and Lactones
Honggu Im,†,‡ Dahye Kang,†,‡ Soyeon Choi,†,‡ Sanghoon Shin,†,‡ and Sungwoo Hong*,‡,†
†Department of Chemistry, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, 34141, Korea
‡Center for Catalytic Hydrocarbon Functionalizations, Institute for Basic Science (IBS), Daejeon, 34141, Korea
S
* Supporting Information
ABSTRACT: Visible-light-induced intramolecular C−O bond
formation was developed using 2,4,6-triphenylpyrylium tetra-
fluoroborate (TPT), which allows the regiocontrolled con-
struction of cyclic ethers and lactones. The reaction is likely to
proceed through the single-electron oxidation of the phenyl
group, followed by the formation of a benzylic radical, thus
preventing a competing 1,5-hydrogen abstraction pathway. Detailed mechanistic studies suggest that molecular oxygen is used
to trap the radical intermediate to form benzyl alcohol, which undergoes cyclization. This new approach serves as a powerful
platform by providing efficient access to valuable five- and six-membered cyclic ethers and lactones with a unified protocol.
Scheme 1. Regioselective Photocatalytic Construction of
Lactones and Cyclic Ethers
yclic ethers and lactones are the basic structural unit of
Cnumerous naturally occurring compounds and privileged
motifs in the areas of pharmaceuticals, agrichemicals, materials,
and polymers.1,2 Accordingly, a number of synthetic strategies
have been investigated to enable the rapid construction of
these skeletons.3 One such convenient and attractive route for
the synthesis of these scaffolds is direct intramolecular C−O
bond-forming reactions starting from the corresponding
carboxylic acids or alcohols.4
In Hofmann−Loffler-type reactions, the O−I bond under-
goes light-induced homolysis to form the O-centered radicals,
which enable a subsequent 1,5-hydrogen atom transfer (1,5-
HAT) event. In the process, the selectivity is governed by
kinetically controlled C−H functionalization at the δ position,
which would result in the construction of five-membered ether
rings. For substrates containing benzylic sites, the formation of
benzylic radicals can compete with this kinetically preferred
1,5-HAT, eventually providing a mixture of five- and six-
membered rings (Scheme 1a).5 Moreover, for substrates
bearing acidic functional groups, the generated acyloxy radical
species can induce nonproductive decarboxylation.6 Driven by
the need for an efficient and unified synthetic route to five- and
six-membered cyclic ethers and lactones, we were intrigued by
the possibility of visible-light-promoted, site-selective C−O
bond formation occurring through direct benzylic C−H
activation. Herein, we report an efficient photocatalytic system
for the selective construction of benzyl cyclic ethers and
lactones.
the most effective in this evaluation (>90%) and was chosen
for further exploration.
By carrying out a series of competitive luminescence
quenching experiments,8 we observed that 5-phenylpentan-1-
ol (1a) acts as an efficient reductive quencher of excited TPT
(see the Supporting Information (SI) for details). Based on
these results, we investigated the proposed transformation
using 1a and TPT under irradiation with visible light from blue
LEDs (Table 2). A substantial amount of the hemiacetal
product 4 was obtained (entries 1−3). Interestingly, when
TPT was used without any additives, 4 was formed in 10%
yield along with large amounts of decomposition products. We
rationalized that, upon single-electron oxidation of 1a by
excited TPT*, the resulting benzylic radical would react with
molecular oxygen to generate benzylic hydroperoxide. There-
To test the feasibility of our envisioned approach, we initially
aimed to identify a suitable photocatalyst (PC) by screening
the oxidative quenching abilities of widely employed PCs.7−9
For this purpose, we carried out quenching experiments by
measuring the luminescence spectra of nine different PCs
(PC1−9) with 1a (Table 1). Photocatalysts PC2 (TPT) and
PC3 were greater than 50% in quenching fractions; TPT was
Received: October 4, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.8b03166
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
a
optimized reaction conditions, tetrahydropyran product 2a was
formed in 74% yield without producing Hock rearrangement
product 4 (entry 10). Subsequent control experiments
confirmed the essential roles of visible light and TPT in this
reaction (see the SI for details). Furthermore, a significant
turnover was observed when the reaction was attempted under
an atmosphere of N2 instead (entry 11), indicating that
molecular oxygen serves as an indispensable reagent. Next, we
carried out stoichiometric control experiments under reaction
conditions similar to those of the NIS-mediated Hofmann−
Loffler-type reaction.13 As a result, the five-membered cyclized
product was obtained as a major product (Table S1 and
Scheme S1d in the SI) via a 1,5-HAT process.
Table 1. Quenching Experiment of 1a with Photocatalysts
As illustrated in Scheme 2, we next investigated the substrate
scope with respect to phenyl alcohol substrates to extend the
a
Scheme 2. Substrate Scope of Cyclic Etherification
a
Fluorescence was measured under 415 nm irradiation (0.001 mmol
of PC and 0.2 mmol of 1a in 20 mL of MeCN), F = 100 (1 − I/I0).
a
Table 2. Optimization of the Reaction Conditions
b
% yield
entry
additive (equiv)
solvent
2a
3
4
1
2
3
4
5
6
7
8
9
10
11
−
MeCN
MeCN
MeCN
HFIP
HFIP
CH2Cl2
CH2Cl2/HFIP (1:1)
CH2Cl2/HFIP (1:1)
CH2Cl2/HFIP (1:1)
CH2Cl2/HFIP (1:1)
CH2Cl2/HFIP (1:1)
0
0
0
0
0
0
0
29
23
34
8
5
4
10
34
50
12
17
2
4
0
0
0
I2 (0.1)
ZnI2 (0.1)
I2 (0.1)
ZnI2 (0.1)
ZnI2 (0.1)
ZnI2 (0.1)
ZnI2 (0.1)
ZnI2 (0.2)
ZnI2 (0.3)
ZnI2 (0.3)
21
27
11
31
57
61
74
10
c
c
c
cd
,
−
−
a
Reactions were performed with mixtures of 1 (0.1 mmol), additives,
and TPT (5 mol %) in solvent (2 mL) at rt under irradiation by a 15
b
W blue LED for 24 h under an O2 atmosphere. Yield was analyzed
c
d
by 1H NMR. 1 equiv of AcOH was added. Under a N2 atmosphere.
a
fore, it is conceivable that 4 is derived from the Hock
rearrangement of benzylic hydroperoxide.10 Remarkably, if the
same reaction was carried out in the presence of a catalytic
amount of I2 in hexafluoroisopropanol (HFIP), tetrahydropyr-
an 2a was, indeed, formed along with 4, indicating that the
desired reaction is operative and competes with the Hock
rearrangement (entry 4). More promising results were
obtained using ZnI2, presumably because the coordination of
Zn(II) with the hydroxyl group promoted the intramolecular
C−O cyclization (entry 5). Notably, 1-phenylpentane-1,5-diol
(3), which is thought to be a potential reaction intermediate
for an intramolecular C−O cyclization, was isolated in a
substantial yield.11 After extensive screening of various
Reactions were performed with mixtures of 1 (0.1 mmol), ZnI2 (0.3
equiv), AcOH (1.0 equiv), and TPT (5 mol %) in CH2Cl2 + HFIP
(1:1) at rt under irradiation by a 15 W blue LED for 24 h under O2
atmosphere. Isolated yields of products. Under air atmosphere. The
reaction was conducted at 34 W Kessil LED under an air atmosphere.
b
c
utility and generality of this method. The reaction was tolerant
of halogen-substituted substrates including fluoro, chloro, and
bromo substituents at the para-position, and products 2c, 2d,
and 2e were formed, thus enabling further synthetic
functionalizations at this position. Pentanols bearing meta-
substituents on the phenyl rings also underwent the intra-
molecular C−O cyclization reaction to provide the corre-
sponding cyclic ethers 2f and 2g. The reactions of substrates
possessing ortho-substituents worked well and provided 2h and
12
solvents, the use of a mixture of CH2Cl2 and HFIP in the
presence of AcOH gave the best performance. Under the
B
DOI: 10.1021/acs.orglett.8b03166
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
a
2i. Substrates containing no substituent at the α-carbon
performed well (2j and 2k), which indicates that the Thorpe−
Ingold effect is not a prerequisite for the cyclization. Tricyclic
indanofuran 2l was successfully synthesized from the
corresponding alcohol precursor in good yield. In addition,
the scope could be expanded to a secondary alcohol, which
afforded the desired cyclic ether product 2m. Next, we turned
our attention to the reactivity of 5-phenylpentanol substrates
to construct tetrahydropyran scaffolds. The employment of
substrates with or without dimethyl groups were efficiently
converted to desired products 2n and 2a. The present C−O
bond formation with substituted pentanols containing methyl
(2o), fluoro (2p, 2r, and 2s), and chloro (2q) groups
proceeded smoothly. It is worth noting that substrates bearing
cyclic motifs, as exemplified by cyclohexyl, cyclopentyl, and
triflate-protected piperidine groups were well tolerated in the
reaction and provided oxaspirocycles 2t, 2u, and 2v in good
yields. The utility of the present method was further
demonstrated by conveniently accessing prominent isochro-
mane scaffold 2z. In addition, 6-phenylhexan-1-ol was tested,
and seven-membered cyclic ether 2aa was formed, albeit in a
low 17% yield. When the reaction was carried out with a tosyl-
protected amine substrate, C−N bond formation proceeded to
afford the corresponding six-membered piperidine 2ab.14
Importantly, the scope of the C−H etherification reaction
could be expanded to the selective formation of γ- and δ-
lactones (Scheme 3). First, the scope of this transformation
was studied with respect to butyric acids for the synthesis of γ-
lactones in a slightly modified solvent system (CH2Cl2/HFIP =
6:1). Phenylbutanoic acid with geminal methyl groups at the
α-carbon underwent C−O bond formation to provide the
corresponding γ-lactone product 6a in good yield. An array of
butyric acids bearing methyl, tert-butyl, fluoro, chloro, bromo,
or trimethylsilyl groups on the phenyl ring were readily
transformed into the cyclized products under the reaction
conditions (6b−6i, 6l, and 6m). 4-Phenylbutyric acid
containing no substituent at the α-carbon also performed
well (6j, 77% yield). Expanding the scope to a 2-benzylbenzoic
acid was also possible, and isobenzofuran 6n was generated in
excellent yield. We subsequently assessed the applicability of
our method with respect to indanyl carboxylic acid 5o. Indeed,
the utility of the present method was further demonstrated by
providing convenient access to prominent structural motifs
featuring lactone units fused to 2,3-dihydro-1H-indene (6o).
Further exploration demonstrated that heterocyclic acids that
are typically incompatible with strong oxidative conditions, as
exemplified by thiophene, can undergo the present reaction to
afford 5p. Next, we explored the substrate scope of six-
membered lactone formation. A series of 5-phenylpentanoic
acid substrates possessing different substitution patterns were
subjected to the reaction, and the benzylic positions were
exclusively functionalized to afford the corresponding δ-
lactones 6q−6ad. Substrates bearing sterically hindered groups
at the α-position generally led to reasonable yields (6w, 63%
and 6x, 61%). We applied the present reaction to 2-
phenethylbenzoic acid and isochromanone derivatives. Diverse
isochromanone and isohydrocoumarin scaffolds15 were
successfully constructed (6z−6ad), highlighting the attractive-
ness of this method and its potential applications in the
construction of compound libraries.
Scheme 3. Substrate Scope of Lactonization
a
Reactions were performed with mixtures of 1 (0.1 mmol), ZnI2 (0.3
equiv), AcOH (1.0 equiv), and TPT (5 mol %) in CH2Cl2 + HFIP
(6:1) at rt under irradiation by a 15 W blue LED for 72 h under an O2
atmosphere. Isolated yields of products. The reaction performed in
CH2Cl2 + HFIP (1:1). The reaction was conducted for 24 h at 45 °C
b
c
and 34 W Kessil LED under an air atmosphere.
3 and benzyl iodide 8 were employed for the intramolecular
cyclization reaction (Scheme S2a). In both cases, desired cyclic
ether 2a was produced. Next, to clarify the role of a hydroxyl
group in the reaction, methyl ether 9 was subjected to the
standard reaction conditions. In this case, we were able to
detect three benzylic-functionalized products bearing hydroxyl,
hexafluoroisopropyl, or acetoxyl groups (Scheme S2b).16
Intriguingly, the use of ZnI2 under the irradiation of visible
light is critical for the cyclization of 3, whereas 8 was readily
cyclized without requiring the above-mentioned reagents.
Because benzylic C−H activation could be achieved in the
absence of hydroxyl groups in the substrate, the involvement of
an alkoxy radical-engaging 1,5 HAT pathway could be
excluded. These results are in good agreement with our
mechanistic hypothesis that a benzylic radical intermediate is
generated by the direct benzylic C−H oxidation.
To elucidate the mechanism, we conducted a series of
control experiments (see the SI for details). First, in an attempt
to identify the plausible reaction intermediates, benzyl alcohol
C
DOI: 10.1021/acs.orglett.8b03166
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
To understand the iodine-catalyzed cyclization of benzyl
alcohol 3, we carried out a DFT study (Figure 1).
controllable and selective manner. TPT enables the generation
of benzylic radicals and subsequent intramolecular C−O
cyclization through the intermediacy of benzyl alcohols. The
direct substrate oxidation that was rationalized by control
experiments offers predictable control over product selectivity
by avoiding complications from the 1,5-HAT. A variety of
value-added five- and six-membered cyclic ether and lactone
derivatives were successfully synthesized with the present
protocol, which provides a straightforward route to the
bioactive scaffolds.
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.8b03166.
Experimental procedures and characterization of new
compounds (1H and 13C NMR spectra) (PDF)
Accession Codes
Figure 1. Free energy profile for the intramolecular cyclization.
CCDC 1871380 contains the supplementary crystallographic
data for this paper. These data can be obtained free of charge
via www.ccdc.cam.ac.uk/data_request/cif, or by emailing
data_request@ccdc.cam.ac.uk, or by contacting The Cam-
bridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
Unsurprisingly, coordination of I2* to the benzylic hydroxyl
group is an energetically downhill process by −7.8 kcal/mol
(Int1). Interestingly, the bond length of I2 in the excited state
is considerably longer (3.155 Å) than in the ground state
(2.660 Å). Furthermore, we found that the bond length of I2 in
Int1 is extended to 3.304 Å, and the I−I bond can be cleaved
to generate the corresponding cationic intermediate Int2 with
a barrier of only 4.9 kcal/mol. Next, intramolecular cyclization
traversing the transition state Int2-TS at 17.0 kcal/mol
liberates IOH and affords Int3, which readily undergoes
nearly barrierless acetate-mediated deprotonation to form the
desired product 2a at −44.7 kcal kcal/mol. In the presence of
ZnI2 as a Lewis acid (Int4), the energy barrier for
intramolecular cyclization from Int4 is only 7.9 kcal/mol.17
The alternative reaction pathway associated with direct
activation and cyclization by I2* demands a much higher
energy barrier of 40.0 kcal/mol (Int2-TS′), as shown in blue.18
Based on these observations, a proposed mechanistic pathway
is shown in Scheme 4 (see the SI for details and discussions).
In summary, we have developed a visible-light-induced
photocatalytic reaction for the expedient synthesis of cyclic
ethers and lactones from alcohols and carboxylic acids in a
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: hongorg@kaist.ac.kr.
ORCID
Honggu Im: 0000-0003-0911-5982
Sungwoo Hong: 0000-0001-9371-1730
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
This research was supported financially by the Institute for
Basic Science (IBS-R010-G1).
■
REFERENCES
■
(1) (a) Clarke, P. A.; Santos, S. Eur. J. Org. Chem. 2006, 2006, 2045.
(b) Lorente, A.; Lamariano-Merketegi, J.; Albericio, F.; Alvarez, M.
́
Chem. Rev. 2013, 113, 4567.
Scheme 4. Proposed Mechanistic Pathway
(2) (a) Fay, F.; Linossier, I.; Langlois, V.; Renard, E.; Vallee-Rehel,
K. Biomacromolecules 2006, 7, 851. (b) Pal, S.; Chatare, V.; Pal, M.
Curr. Org. Chem. 2011, 15, 782. (c) Jiao, P.; Gloer, J. B.; Campbell, J.;
Shearer, C. A. J. Nat. Prod. 2006, 69, 612.
(3) (a) Elliott, M. C.; Williams, E. J. Chem. Soc., Perkin Trans. 2001,
̈
̈
1, 2303. (b) Crespi, S.; Jager, S.; Konig, B.; Fagnoni, M. Eur. J. Org.
Chem. 2017, 2017, 2147. (c) Zheng, Z.; Zhang-Negrerie, D.; Du, Y.;
Zhao, K. Sci. China: Chem. 2014, 57, 189.
(4) (a) Sathyamoorthi, S.; Du Bois, J. Org. Lett. 2016, 18, 6308.
(b) Zhang, B.; Han, L.; Li, T.; Yan, J.; Yang, Z. Synth. Commun. 2014,
44, 1608. (c) Dohi, T.; Takenaga, N.; Goto, A.; Maruyama, A.; Kita,
Y. Org. Lett. 2007, 9, 3129. (d) Beebe, T. R.; Adkins, R. L.; Bogardus,
C. C.; Champney, B.; Hii, P. S.; Reinking, P.; Shadday, J.;
Weatherford, W. D., III; Webb, M. W.; Yates, S. W. J. Org. Chem.
1983, 48, 3126. (e) Pandey, G.; Pal, S.; Laha, R. Angew. Chem., Int. Ed.
2013, 52, 5146. (f) Nakatsuji, H.; Sawamura, Y.; Sakakura, A.;
Ishihara, K. Angew. Chem., Int. Ed. 2014, 53, 6974. (g) Duhamel, T.;
Stein, C. J.; Martínez, C.; Reiher, M.; Mun
3918.
̃
iz, K. ACS Catal. 2018, 8,
D
DOI: 10.1021/acs.orglett.8b03166
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(5) (5) Teduka, T.; Togo, H. Synlett 2005, 2005, 0923.
(6) (a) Kochi, J. K. J. Org. Chem. 1965, 30, 3265. (b) Kochi, J. K.;
Bacha, J. D. J. Org. Chem. 1968, 33, 2746. (c) Sheldon, R. A. Org.
React. 1972, 19, 279.
(7) (a) Koike, T.; Akita, M. Inorg. Chem. Front. 2014, 1, 562.
(b) Romero, N. A.; Nicewicz, D. A. Chem. Rev. 2016, 116, 10075.
́
̈
(c) Kuijpers, K. P.L.; Bottecchia Cambie, C. D.; Drummen, K.; Konig,
̈
N. J.; Noel, T. Angew. Chem., Int. Ed. 2018, 57, 11278.
(8) For selected reviews, see: (a) Narayanam, J. M. R.; Stephenson,
C. R. J. Chem. Soc. Rev. 2011, 40, 102. (b) Tucker, J. W.; Stephenson,
C. R. J. J. Org. Chem. 2012, 77, 1617. (c) Prier, C. K.; Rankic, D. A.;
MacMillan, D. W. C. Chem. Rev. 2013, 113, 5322. (d) Skubi, K. L.;
Blum, T. R.; Yoon, T. P. Chem. Rev. 2016, 116, 10035. (e) Shaw, M.
H.; Twilton, J.; MacMillan, D. W. C. J. Org. Chem. 2016, 81, 6898.
(f) Arias-Rotondo, D. M.; McCusker, J. K. Chem. Soc. Rev. 2016, 45,
5803. (g) Chatterjee, T.; Iqbal, N.; You, Y.; Cho, E. J. Acc. Chem. Res.
2016, 49, 2284.
(9) Roth, H. G.; Romero, N. A.; Nicewicz, D. A. Synlett 2016, 27,
714.
(10) (a) Hock, H.; Schrader, O. Angew. Chem. 1936, 49, 595.
(b) Gardner, H. W.; Weisleder, D.; Nelson, E. C. J. Org. Chem. 1984,
49, 508. (c) Frimer, A. A. J. Photochem. 1984, 25, 211. (d) Yin, H.;
Havrilla, C. M.; Morrow, J. D.; Porter, N. A. J. Am. Chem. Soc. 2002,
124, 7745.
(11) Siddaraju, Y.; Prabhu, K. R. J. Org. Chem. 2018, 83, 11145.
(12) Ohkubo, K.; Fukuzumi, S. Org. Lett. 2000, 2, 3647.
(13) (a) Beebe, T. R.; Lin, A. L.; Miller, R. D. J. Org. Chem. 1974,
39, 722. (b) McDonald, C. E.; Beebe, T. R.; Beard, M.; McMillen, D.;
Selski, D. Tetrahedron Lett. 1989, 30, 4791. (c) Liu, T.; Mei, T.; Yu, J.
́
Q. J. Am. Chem. Soc. 2015, 137, 5871. (d) O’Broin, C. Q.; Fernandez,
P.; Martínez, C.; Muniz, K. Org. Lett. 2016, 18, 436.
(14) Zhang, H.; Muniz, K. ACS Catal. 2017, 7, 4122.
̃
̃
(15) (a) Watanabe, K.; Taniguchi, E. J. J. Pestic. Sci. 1994, 19, 209.
(b) Shinkaruk, S.; Bennetau, B.; Babin, P.; Schmitter, J. M.; Lamothe,
V.; Bennetau-Pelissero, C.; Urdaci, M. C. Bioorg. Med. Chem. 2008,
16, 9383. (c) Liu, J.; Nakamura, S.; Zhuang, Y.; Yoshikawa, M.;
Hussein, G. M. E.; Matsuo, K.; Matsuda, H. Chem. Pharm. Bull. 2013,
61, 655. (d) van der Walt, M. M.; Terre’Blanche, G. Bioorg. Chem.
2018, 77, 136.
(16) We observed that 10b and 10c could be obtained from 10a,
which indicates that 10a could be an initial reaction intermediate (see
the SI).
(17) Int4 was redefined as a standard point, 0.00 kcal/mol.
(18) The formation of benzyl cation species via direct dissociation of
IOH from Int2 appears to be less feasible, and the nucleophilic attack
by the hydroxyl group is required for the C−O bond cleavage (see the
SI for details).
E
DOI: 10.1021/acs.orglett.8b03166
Org. Lett. XXXX, XXX, XXX−XXX