2H-chromenes generated by an iron(III) complex-catalyzed allylic cyclization

Article abstract of DOI:10.1002/adsc.201500058

A straightforward method based on an iron(III) complex-catalyzed cyclization of 2-(1-hydroxyallyl)phenols is reported to access a large variety of 2H-chromenes. This method was applied to the total synthesis of a natural product, tephrowatsin B.

Full text of DOI:10.1002/adsc.201500058

COMMUNICATIONS
DOI: 10.1002/adsc.201500058
2H-Chromenes Generated by an Iron(III) Complex-Catalyzed
Allylic Cyclization
L. Calmus,a A. Corbu,a,* and J. Cossya,*
a
Laboratoire de Chimie Organique, Institute of Chemistry, Biology and Innovation (CBI) – UMR 8231 – ESPCI
ParisTech/CNRS/PSL Research University, 10 rue Vauquelin, 75231 Paris Cedex 05, France
E-mail: andrei.corbu@espci.fr or janine.cossy@espci.fr
Received: January 22, 2015; Published online: && &&, 0000
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201500058.
Abstract: A straightforward method based on an
iron(III) complex-catalyzed cyclization of 2-(1-hy-
droxyallyl)phenols is reported to access a large
variety of 2H-chromenes. This method was applied
to the total synthesis of a natural product, tephro-
watsin B.
One particular method is the acid-catalyzed ring-clos-
ing of 2-(1-hydroxyallyl)phenols III to the corre-
sponding pyran ring IV (Scheme 1). This reaction in-
volves either a chiral Brønsted acid,[21] a metallic
Lewis acid catalyst such as a gold[22] or a palladium
complex.[23] The cyclization of intermediate III is par-
ticularly attractive, as these compounds are readily
available in one step, starting from ortho-hydroxy-
Keywords: allylic cyclization; 2H-chromenes; iron
catalysis; synthetic methods; total synthesis
ACHUTGTNRENbNUG enzCAHTUNGTRNEaNUGN ldehydes I and vinyl organometallics II. Howev-
er, 2-(1-hydroxyallyl)phenols are sensitive to Brønsted
acids and in order to have good yields in 2H-chro-
menes, the allylic alcohols have to be highly substitut-
ed at the C-2 and/or C-4 position(s) to eventually gen-
2H-Chromenes are important structural motifs pres- erate a stabilized carbocation (Scheme 1).[21,23] On the
ent in a number of natural and synthetic products other hand, palladium and gold catalysts are expen-
that exhibit a wide range of biological activities. For sive and/or toxic, thus affordable, eco-friendly and
example, cannabichromene has analgesic, anti-inflam-
matory and antiviral properties;[1] tuberatolide B is an
efficient farnesoid X receptor antagonist;[2] iclaprim is
an antibiotic used for skin infections[3] while sclerota-
mide[4] and tephrowatsin B possess insecticidal and
antifungal activities (Figure 1).[5] The structural diver-
sity of 2H-chromenes around the benzene and pyran
rings confers to these compounds specific biological
activities that have attracted a continuous interest
from organic synthetic chemists.[6]
Various methods have been developed to attain the
benzopyran core,[7] such as: Wittig olefinations,[8] oxi-
dation of the corresponding chromanes,[9] Pd-cata-
lyzed oxidative cyclization of O-allylphenols,[10] ring-
closing metathesis,[11] oxa-Michael/aldol sequence,[12]
epoxide ring-opening,[13] direct functionalization of
simple chromenes or chromene acetals,[14] O-quinone
methide cyclization under thermal conditions,[15]
Morita–Baylis–Hillman reaction,[16] Petasis reac-
tion,[17] cobalt-catalyzed carbene radicals insertion,[18]
propargyl ether hydroarylation.[19] In addition, Pd-cat-
alyzed asymmetric allylic substitution was developed
to generate optically active chromanes that were sub-
sequently oxidized to the corresponding chromenes.[20] Figure 1. Biologically active 2H-chromenes.
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
ÞÞ
These are not the final page numbers!

COMMUNICATIONS
L. Calmus et al.
Table 1. Screening of the conditions for the Lewis/Brønsted
acid-catalyzed cyclization.
Scheme 1. Acid-catalyzed intramolecular cyclization of
2-(1-hydroxyallyl)phenols.
milder alternative catalysts are needed to realize
chemical transformations in the context of green
chemistry.
Herein, we report a straightforward access to vari-
ously substituted 2H-chromenes, featuring an unpre-
cedented iron-catalyzed cyclization of phenolic allylic
alcohols. The reactivity of a model substrate 1a, was
examined and different Lewis acids as well as differ-
ent conditions were screened (Table 1). Compound 1a
was inert when treated with ZnCl2, AlCl3 or FeCl3 at
room temperature (Table 1, entries 1, 2 and 3) and
[a]
Sealed tube, microwave irradiation.
30 mol%.
[b]
was decomposed when treated with FeCl2, sidual Brønsted acid such as HCl, that decomposed
FeCl3·6H2O or Fe(OTf)3 (Table 1, entries 4, 5 and 6). most of the starting allylic alcohol 1a (Table 1,
Iron(III) salts are known to activate allylic alcohols entry 13). For practical purposes, the complex formed
and their derivatives under either acidic[24] or basic in situ was used to investigate the scope and limita-
conditions.[25] Thus, to modulate the Lewis acidity of tions of the reaction, instead of the isolated complex.
iron salts, ligands such as pyridine were added, and
With the FeCl3/2,2’-bipyridine catalytic system in
under these conditions, no conversion of 1a was ob- hand, the influence of functional groups on the aro-
served (Table 1, entry 8). Switching to a chelating matic ring of various 2-(1-hydroxyallyl)phenols was
ligand, such as 2,2’-bipyridine, the iron(III) triflate studied (Table 2). The synthesis of the 2-(1-hydroxyal-
was still too active as 1a was completely decomposed lyl)phenols 1a–1x was straightforward, by adding vi-
(Table 1, entry 7) but, interestingly, we observed for nyllithium or vinyl Grignard reagents on the corre-
the first time the conversion of 1a into the desired sponding unprotected phenolic aldehydes.[27]
2H-chromene 2a by using a preformed FeCl3/2,2’-bi-
When the cyclization conditions (FeCl3/2,2’-bypyri-
pyridine complex (15 mol%) at room temperature, dine, THF, 808C, MW, 5 h) were applied to 1a–1x,
albeit in a low 5% yield (Table 1, entry 8). At room chromenes 2a–2x were obtained in good yields for
temperature, the catalytic system was not active phenols 1 bearing electron-rich substituents such as
enough and, to circumvent the low activity of the cat- 1b, 1c, 1d and 1e, 86%, 64%, 83% and 99% yields in
alytic system FeCl3/2,2’-bipyridine, the reaction was chromenes 2b, 2c, 2d and 2e were respectively ob-
thermally activated. When 1a was heated under mi- tained, as well as for naphthol 1f that generates the
crowave (MW) irradiation (FeCl3/2,2’-bipyridine, corresponding benzochromene 2f in 92% yield. When
15 mol%, 808C, 5 h), a complete conversion of 1a was an electron-withdrawing group such as a nitro group
observed and chromene 2a was isolated in 61% yield at the C-6 position deactivates the nucleophilicity of
(Table 1, entry 9). We have to point out that the use the phenol group, a partial conversion of 1g, 51%,
of pyridine or Et3N instead of the strong chelating was observed and chromene 2g was isolated in 43%
2,2’-bipyridine, gave only partial conversion of 1a yield. When studying the influence of halogen sub-
(Table 1, entries 10 and 11). Interestingly, when using stituents on the cyclization, 6-bromo-2H-chromene 2h
the complex [FeCl2bipy2][FeCl4], obtained by mixing and 6-iodo-2H-chromene 2i were obtained in 76%
FeCl3 and 2,2’-bipyridine and after crystallization in and 81% yield, respectively.
acetonitrile, 2a was isolated in 77% yield (Table 1,
However, a detrimental effect of a chlorine atom in
entry 12).[26] This result proves that the complex was the ortho position of the phenol, like in 1j, was ob-
responsible for generating 2a, and not the possible re- served for the cyclization as only 19% of 1j was con-
2
asc.wiley-vch.de
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 0000, 000, 0 – 0
ÝÝ
These are not the final page numbers!

COMMUNICATIONS
2H-Chromenes Generated by an Iron(III)-Catalyzed Allylic Cyclization
Table 2. Synthesis of a diversity of functionalized 2H-chro-
menes.
as in 1o and 1p, the ortho-electron-withdrawing effect
prevails and, as in the case of phenol 1o, chromene 2o
could not even be isolated. It seems that the electron-
withdrawing group in the ortho position has a detri-
mental effect on the cyclization. Once more, the reac-
tivity was restored for the 2,2-dimethyl analog, as 2p
was obtained in 60% yield. When the allylic alcohol
was substituted at C-4 by a methyl or a phenyl group,
good to excellent yields in 2H-chromenes were ob-
tained (2q–2t, 59–98%). In addition, the C-2 and C-3
substituted 2H-chromenes 2u and 2v, along with 2,2-
dimethylchromenes 2w and 2x, were obtained in good
yields (72–91%).
According to the obtained results, it seems that the
low nucleophilicity of a phenolic ring can be compen-
sated by an increase of the electrophilicity of the al-
lylic alcohol as the cyclized products 2k, 2l, 2n, 2p
were formed in good to moderate yields. As a general
trend, the reactivity of 2-(1-hydroxyallyl)phenols can
be tuned in a two way fashion: either by enhancing
the nucleophilicity of the phenolic ring by electron-
rich substituents, or by substitution of the allylic alco-
hol at C-2 and/or C-4 by an alkyl or a phenyl group.
Taking into account the previous results, the FeCl3/
2,2’-bipyridine complex was applied to electron-rich
phenols possessing a substituted allylic alcohol moiety
(Table 3).[27] Substrates 3a–c were very reactive as
5 mol% of the FeCl3/2,2’-bipyridine catalyst were suf-
ficient and the reaction could be run at room temper-
Table 3. Formation of electron-rich 2H-chromenes at room
temperature.
[a]
Conversion of the starting material.
Starting material recovered.
[b]
verted, nonetheless 2j was isolated in 19% yield. In
order to increase the yield in chromenes, the allylic al-
cohol was substituted by two alkyl groups at C-2.
When phenols 1k and 1l were treated with FeCl3/2,2’-
bipyridine, 2k and 2l were isolated in 98% and 58%
yield, respectively. The substitution of the allylic alco-
hol is important for an efficient synthesis of 2H-chro-
menes, more than the detrimental inductive effect of
the substituents on the phenol ring. When the phenol
ring was highly deactivated by two fluorine atoms,
chromene 2m was not even detected, but its 2,2-di-
methyl analog, 2n, was isolated in 54% yield. When
a para-electron-donating and an ortho-electron-with-
drawing group were present on the phenol rings such
[a]
Substrate used as a crude product in the cyclization.
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3
ÞÞ
These are not the final page numbers!

COMMUNICATIONS
L. Calmus et al.
Table 4. Isomeric allylic alcohols as substrates for the iron(III)-catalyzed cyclization.
ature. In the case of the allylic alcohol 3a, the cycliza-
As 4a and 4b were obtained in good yields, the
tion generated the corresponding chromene 4a in total synthesis of a natural product, tephrowatsin B,
45% yield, the reactivity of the allylic alcohol being was envisioned (Scheme 2). Tephrowatsin B was iso-
amplified by the phenyl substituent at C-2. For sub- lated from Tephrosia watsoniana, a plant which is
strate 3b, the nucleophilic character of the phenol was known for its insecticidal activities and was described
not enhanced by the two electron-rich methoxy sub- as a racemic mixture.[5] The synthesis started from the
stituents in the meta position, but they increased the
reactivity of the allylic alcohol, already activated by
the bis-alkyl substitution at C-2, leading to chromene
4b in a good yield of 91%. The same substitution pat-
tern of the allylic alcohol was used in 3c, but in this
case the presence of three ether groups on the aro-
matic ring were activating the phenol as well as the
allylic system, and chromene 4c was generated in an
excellent yield of 96%.
To gain some insight into the possible mechanism,
the reactivity of allylic alcohols 1w, (Z)-6 and (E)-6
was examined (Table 4). Allylic alcohol 1w gave the
best result, as 2,2-dimethyl-chromene 2w was isolated
in 72% yield, without any diene 5 as the side product,
whereas (Z)-6 and its (E)-isomer afforded a lower
yield in chromene 2w (60% and 54%, respectively)
and the diene by-product 5 was observed (2w/5 4:1
and 2:1, respectively). Indeed, if the allylic benzylic
carbocation V were an intermediate, all three cycliza-
tion pathways should produce similar results. It is
clear that for the allylic alcohol (E)-6, carbocation V
is involved in the process in order to explain the con-
version of the (E) double bond to the (Z) double
bond necessary for the ring closure. According to
these results, it seems that an SN2’-type reaction of
the 2-(1-hydoxyallyl)phenol 1w, going through transi-
tion state VI, is more likely the preferred mechanism,
without excluding carbocation V, that seems to be the
intermediate in the case of the cyclization of isomers
(Z)-6 and (E)-6 (Table 4).
Scheme 2. Total synthesis of tephrowatsin B.
4
asc.wiley-vch.de
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 0000, 000, 0 – 0
ÝÝ
These are not the final page numbers!

COMMUNICATIONS
2H-Chromenes Generated by an Iron(III)-Catalyzed Allylic Cyclization
completion of the reaction. The solvent was then removed
under reduced pressure and the residue was purified by
flash chromatography on silica gel using an appropriate mix-
ture of petroleum ether and EtOAc as an eluent to afford
the 2H-chromene.
commercially available 2,4,6-trimethoxybenzaldehyde
7 which was selectively monodemethylated using tri-
bromoborane and the resulting phenol 8 was obtained
in 45% yield and subsequently C-allylated at C-3 by
a two-step procedure, implying an O-allylation fol-
lowed by a thermal Claisen rearrangement. The allyl-
phenol 9 was isolated in an overall yield of 34%.
Phenol 9 was then transformed to the corresponding
prenyl derivative 10 by using a cross-metathesis (89%
yield). After transformation of hydroxybenzaldehyde
10 to intermediate 11, by addition of b-styryllithium,
the direct treatment of this crude mixture with
iron(III) chloride/2,2’-bipyridine at room temperature,
produced (Æ)-tephrowatsin B in 54% yield, the NMR
spectral data matching those of the published isolated
tephrowatsin B (Scheme 2).[5]
References
[1] B. Romano, F. Borrelli, I. Fasolino, R. Capasso, F. Pisci-
telli, M. G. Cascio, R. G. Pertwee, D. Coppola, L. Vas-
sallo, P. Orlando, V. Di Marzo, A. A. Izzo, Br. J. Phar-
macol. 2013, 169, 213–229.
[2] H. Choi, H. Hwang, J. Chin, E. Kim, J. Lee, S.-J. Nam,
B. C. Lee, B. J. Rho, H. Kang, J. Nat. Prod. 2011, 74,
90–94.
In conclusion, we have developed a practical and
highly straightforward method to synthesize a variety
of diversely substituted 2H-chromenes featuring an
unprecedented iron(III)-catalyzed allylic cyclization.
By tuning the Lewis acidity of FeCl3, we have shown
that the use of a chelating ligand such as 2,2’-bipyri-
dine was decisive. Non-activated phenols have to be
heated under microwave irradiation, while electron-
rich phenols could easily cyclize at room temperature
to produce the desired 2H-chromenes. In addition
chromenes possessing electron-withdrawing groups on
the aromatic ring can be generated if the allylic alco-
hol precursors are substituted at C-2. Finally, these
conditions were successfully used to synthesize teph-
rowatsin B.
[3] P. Schneider, S. Hawser, K. Islam, Bioorg. Med. Chem.
Lett. 2003, 13, 4217–4221.
[4] A. C. Whyte, J. B. Gloer, D. T. Wicklow, P. F. Dowd, J.
Nat. Prod. 1996, 59, 1093–1095.
[5] F. Gꢁmez, L. Quijano, J. S. Calderꢁn, C. Rodrꢂquez, T.
Rꢂos, Phytochemistry 1985, 24, 1057–1059.
[6] K. C. Nicolaou, J. A. Pfefferkorn, A. J. Roecker, G. Q.
Cao, S. Barluenga, H. J. Mitchell, J. Am. Chem. Soc.
2000, 122, 9939–9953.
˝
[7] A. Lꢃvai, T. Tꢂmꢄr, P. Sebok, T. Eszenyi, Heterocycles
2000, 53, 1193–1203.
[8] a) B. Begasse, M. Le Corre, Tetrahedron 1980, 36,
3409–3412; b) E. E. Schweizer, A. T. Wehman, D. M.
Nycz, J. Org. Chem. 1973, 38, 1583–1588.
[9] Y. Noda, M. Yasuda, Helv. Chim. Acta 2012, 95, 1946–
1952.
[10] R. C. Larock, L. Wei, T. R. Hightower, Synlett 2000,
522–524.
[11] a) S. Chang, R. Grubbs, J. Org. Chem. 1998, 63, 864–
866; b) W. A. L. van Otterlo, E. L. Ngidi, S. Kuzvidza,
G. L. Morgans, S. S. Moleele, C. B. de Koning, Tetrahe-
dron 2005, 61, 9996–10006.
Experimental Section
General Procedure for Thermal Cyclization (Table 2)
[12] a) D. Dauzonne, R. Royer, Synthesis 1984, 348–349;
b) S.-P. Luo, Z.-B. Li, L.-P. Wang, Y. Guo, A.-B. Xia,
D.-Q. Xu, Org. Biomol. Chem. 2009, 7, 4539–4546;
c) D. Lanari, O. Rosati, M. Curini, Tetrahedron Lett.
2014, 55, 1752–1755.
To a flask containing FeCl3 (12 mg, 0.075 mmol, 0.15 equiv.)
and 2,2’-bipyridine (12 mg, 0.075 mmol, 0.15 equiv.) was
added dry THF (2 mL). The solution was stirred for 15 min
at room temperature. During the stirring, the color changed
from yellow to orange/red. This solution was added to a solu-
tion of allylic benzylic alcohol (0.5 mmol, 1 equiv., in 3 mL
of THF). After 5 h stirring at 808C under microwave irradi-
ation, the solvent was then removed under reduced pressure
and the residue was purified by flash chromatography on
silica gel using an appropriate mixture of petroleum ether
and EtOAc as an eluent to afford the 2H-chromene.
[13] J.-Y. Goujon, F. Zammattio, J.-M. Chrꢃtien, I. Beaudet,
Tetrahedron 2004, 60, 4037–4049.
[14] a) D. J. Clausen, P. E. Floreancig, J. Org. Chem. 2012,
77, 6574–6582; b) X. Li, J. Zhao, Z. Wang, J. Han, J.
Zhao, S. Zhu, Tetrahedron 2012, 68, 8011–8017; c) P. N.
Moquist, T. Kodama, S. E. Schaus, Angew. Chem. 2010,
122, 7250–7254; Angew. Chem. Int. Ed. 2010, 49, 7096–
7100; d) T. J. A. Graham, A. G. Doyle, Org. Lett. 2012,
14, 1616–1619; e) M. Rueping, C. M. R. Volla, I. Ato-
diresei, Org. Lett. 2012, 14, 4642–4645.
[15] a) J. J. Talley, Synthesis 1983, 845–847; b) S. B. Ferreira,
F. de C. da Silva, A. C. Pinto, D. T. G. Gonzaga, V. F.
Ferreira, J. Heterocycl. Chem. 2009, 46, 1080–1097.
[16] G.-N. Ma, J.-J. Jiang, M. Shi, Y. Wei, Chem. Commun.
2009, 5496.
General Procedure for Room Temperature
Cyclization (Table 3)
To a flask containing FeCl3 (4 mg, 0.025 mmol, 0.05 equiv.)
and 2,2’-bipyridine (4 mg, 0.025 mmol, 0.05 equiv.) was
added dry THF (1 mL). The solution was stirred for 15 min
at room temperature. During the stirring the color changed
from yellow to orange/red. This solution was added to a solu-
tion of allylic benzylic alcohol (0.5 mmol, 1 equiv., in 2 mL
of THF). The solution was stirred at room temperature until
[17] a) Q. Wang, M. G. Finn, Org. Lett. 2000, 2, 4063–4065;
b) N. A. Petasis, A. N. Butkevich, J. Organomet. Chem.
2009, 694, 1747–1753.
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
5
ÞÞ
These are not the final page numbers!

COMMUNICATIONS
L. Calmus et al.
[18] N. D. Paul, S. Mandal, M. Otte, X. Cui, X. P. Zhang, B.
de Bruin, J. Am. Chem. Soc. 2014, 136, 1090–1096.
[19] a) S. J. Pastine, S. W. Youn, D. Sames, Org. Lett. 2003,
5, 1055–1058; b) R. S. Menon, A. D. Findlay, A. C. Bis-
sember, M. G. Banwell, J. Org. Chem. 2009, 74, 8901–
8903; c) W.-W. Qiu, K. Surendra, L. Yin, E. J. Corey,
Org. Lett. 2011, 13, 5893–5895; d) P. Morꢄn-Poladura,
E. Rubio, J. M. Gonzꢄlez, Beilstein J. Org. Chem. 2013,
9, 2120–2128.
[20] B. M. Trost, H. C. Shen, L. Dong, J.-P. Surivet, C. Syl-
vain, J. Am. Chem. Soc. 2004, 126, 11966–11983.
[21] M. Rueping, U. Uria, M.-Y. Lin, I. Atodiresei, J. Am.
Chem. Soc. 2011, 133, 3732–3735.
[23] a) B.-S. Zeng, X. Yu, P. W. Siu, K. A. Scheidt, Chem.
Sci. 2014, 5, 2277; b) Y. Xia, Y. Xia, Y. Zhang, J. Wang,
Org. Biomol. Chem. 2014, 12, 9333–9336.
[24] A. Guꢃrinot, A. Serra-Muns, C. Gnamm, C. Bensous-
san, S. Reymond, J. Cossy, Org. Lett. 2010, 12, 1808–
1811.
[25] G. Kabalka, M. Yao, S. Borella, L. Goins, Organometal-
lics 2007, 26, 4112–4114.
[26] a) B. N. Figgis, P. A. Reynolds, N. Lehner, Acta Crystal-
logr. Sect. B 1983, 39, 711–717; b) J. G. Vos, Inorg.
Chim. Acta 1985, 99, 117–120.
[27] The stable starting substrates are fully described in the
Supporting Information.
[22] A. Aponick, B. Biannic, M. R. Jong, Chem. Commun.
2010, 46, 6849.
6
asc.wiley-vch.de
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 0000, 000, 0 – 0
ÝÝ
These are not the final page numbers!

COMMUNICATIONS
2H-Chromenes Generated by an Iron(III) Complex-
7
Catalyzed Allylic Cyclization
Adv. Synth. Catal. 2015, 357, 1 – 7
L. Calmus, A. Corbu,* J. Cossy*
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
7
ÞÞ
These are not the final page numbers!