Coumarin (5,6-Benzo-2-pyrone) Trapping of an HDDA-Benzyne

Article abstract of DOI:10.1021/acs.orglett.1c00342

Although the parent 2-pyrone is known to react with simple o-benzynes to produce naphthalene derivatives, there appear to be no examples of the successful reaction of coumarin, a benzo-annulated 2-pyrone analogue, with an aryne. We report such a process here using benzynes generated by the hexadehydro-Diels-Alder reaction to produce phenanthrene derivatives (i.e., benzo-annulated naphthalenes). Density functional theory computations were used to help understand the difference in reactivity between 2-pyrone and the slower trapping agent, coumarin. Finally, the reaction of o-benzyne itself [from o-(trimethylsilyl)phenyl triflate and CsF] with coumarin was shown to be viable, although slow.

Full text of DOI:10.1021/acs.orglett.1c00342

pubs.acs.org/OrgLett
Letter
Coumarin (5,6-Benzo-2-pyrone) Trapping of an HDDA-Benzyne
Bhavani Shankar Chinta,† Daniel Lee,† and Thomas R. Hoye*
Cite This: Org. Lett. 2021, 23, 2189−2193
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Although the parent 2-pyrone is known to react
with simple o-benzynes to produce naphthalene derivatives, there
appear to be no examples of the successful reaction of coumarin, a
benzo-annulated 2-pyrone analogue, with an aryne. We report such
a process here using benzynes generated by the hexadehydro-
Diels−Alder reaction to produce phenanthrene derivatives (i.e.,
benzo-annulated naphthalenes). Density functional theory compu-
tations were used to help understand the difference in reactivity
between 2-pyrone and the slower trapping agent, coumarin. Finally, the reaction of o-benzyne itself [from o-(trimethylsilyl)phenyl
triflate and CsF] with coumarin was shown to be viable, although slow.
coumarin is a less reactive 4π-diene than pyrone toward
benzyne. It is interesting to note the difference in the extent of
asynchronicity in the two TSs. In TSpyrone, the bond lengths of
the two forming C−C bonds (labeled in blue) are nearly the
same as is the deformation of the two carbon atoms of the
pyrone moiety (7.5° and 9.4° of puckering at C3 and C6,
respectively). In contrast, in TScoumarin, the extent of bond
formation is considerably different (blue), as is the degree of
puckering at C3 (16.6°) versus C8a (5°). The reduced amount
of rehybridization at C8a is a computational validation of the
reluctance of the coumarin diene to sacrifice its benzenoid
aromatic resonance stabilization.
In our first experiment (Figure 3a), a solution containing
triyne 13 and coumarin (7a, 3 equiv) in chloroform was
warmed to 85 °C. Rate-limiting HDDA cycloisomerization
gave benzyne 14, which, following capture by 7a, lost CO2 to
produce the (red-colored) naphthofluorenone 16a-syn in 38%
yield. Further scrutiny of the NMR spectrum of the crude
product mixture suggested the possible presence of a second
isomeric product. When this reaction was then performed neat
(1:10, 13:7a), both 16a-syn and 16a-anti were isolated in 28%
and 5% yields, respectively. The constitution of the major
product was established by the clear NOEs indicated in
structures 16a-syn and 16a-anti. In addition, (i) the proton
resonance for the aromatic methyl group was significantly
deshielded in the anti isomer and (ii) the indicated aromatic
protons showed diagnostic differences that reflected their
he reaction of pyrone (4) with o-benzyne (3) to produce
Tnaphthalene (6) was first described by Wittig and
Hoffmann in 1962.1 Heating the thiadiazole 1,1-dioxide 1
gave 6 in 36% yield, following ejection of CO2 from the
presumed Diels−Alder intermediate adduct 5. Over time,
reactions of benzynes (or strained cycloalkynes) with a variety
of pyrone-containing substructures have been reported.2
Conspicuously absent from that body of work is a successful
reaction or an aryne with coumarin (7a, 2H-chromen-2-one).
́
Indeed, in 1997, Guitian and co-workers reported reactions of
substituted pyrones with 3, the latter produced from
anthranilic acid (2).2i In that study, an attempt to effect an
analogous reaction of 3 with coumarin (7a) to produce
phenanthrene (8) was unsuccessful. This was attributed to the
lower reactivity of 7a as a diene because a greater loss of
̀
aromatic resonance stabilization vis-a-vis the analogous
reaction with 4 itself would attend the formation of potential
intermediate 8.
Many triynes such as 10 will cycloisomerize to benzynes 11
in a process now commonly called the hexadehydro-Diels−
Alder (HDDA) reaction.3 We describe here a variety of
reactions between HDDA-benzynes 11 and coumarins. To the
best of our knowledge, these are the first examples of trapping
reactions of arynes using coumarin or substituted coumarins
(Figure 1).2
As a prelude to introducing our experimental observations,
we show in Figure 2 the results of DFT calculations of the
reactions of the parent o-benzyne (3) with pyrone (4, panel a)
as well as with coumarin (7a, panel b). As expected intuitively,
the reaction to form the initial bicyclic adduct 5 is more
exergonic than that leading to 8 because of the aforementioned
increased loss of aromaticity that attends the addition to
coumarin. Accordingly, the activation barrier through tran-
sition structure TScoumarin is also larger than that through
TSpyrone. These data support the earlier assessment2i that
Received: January 28, 2021
Published: March 2, 2021
https://dx.doi.org/10.1021/acs.orglett.1c00342
© 2021 American Chemical Society
Org. Lett. 2021, 23, 2189−2193
2189

Organic Letters
pubs.acs.org/OrgLett
Letter
Figure 1. (a) Reaction of o-benzyne (3) with pyrone (4) to produce
naphthalene (6), following loss of CO2 from initial adduct 5. (b)
Reactions of HDDA-benzynes 11 with coumarin derivatives 7a−e to
give phenanthrenes 12.
Figure 3. (a) Reaction of triyne 13 with coumarin gives isomeric
naphthofluorenones 16a-syn (major) and 16a-anti (minor). (b)
Competition experiment showing that trapping by furan (2 equiv) is
considerably faster than that by coumarin (7a, 10 equiv) (see the
Supporting Information for the NMR spectrum of the crude product
mixture in which 16a-syn was detected).
relative extent of embeddedness in the bay region of the
pentacycle (see the Supporting Information for details).
To gain an understanding of the sense of regioselectivity
shown by the reaction of benzyne 14 with coumarin (7a), we
identified the optimized transition structures of the species
leading to TSsyn versus TSanti by DFT calculations. We used
Figure 2. DFT calculations [SMD(CHCl3)/B3LYP/6-311G+(d,p) at 298 K] of the reaction of o-benzyne (3) with (a) pyrone (4) and (b)
coumarin (7a). Gibbs energies in kilocalories per mole.
2190
https://dx.doi.org/10.1021/acs.orglett.1c00342
Org. Lett. 2021, 23, 2189−2193

Organic Letters
pubs.acs.org/OrgLett
Letter
Figure 4. DFT calculations [SMD(CHCl3)/B3LYP/6-311G+(d,p) at 358 K] for the reaction of benzyne 14 (the truncated nor-methoxy analogue
of the benzyne from triyne 13) with coumarin (7a). The small differences in TS energies are consistent with, in fact, remarkably close to, that
reflected by the 16a-syn:16a-anti product ratio (6:1).
the same functional, basis set, and solvation model that were
used for the reactions with benzyne itself (Figure 2). The
results are summarized in Figure 4. The computed activation
barrier leading to adduct 15-syn* is 1.5 kcal mol−1 lower than
that proceeding to 15-anti* in the competing pathway. This is
remarkably consistent with the observed 16a-syn:16a-anti
versus pyrone with o-benzyne (Figure 2). A competition
experiment (Figure 3b) was performed in which 13 was heated
in CDCl3 in the presence of the excellent trapping reagent
furan (2 equiv) and coumarin (10 equiv). Direct NMR analysis
of this reaction mixture after 14 h showed quite clean
conversion to furan adduct 17 and that ∼0.05% of 16a-syn was
present (see the Supporting Information for details). We
conclude that coumarin reacts >1000 times more slowly with
the benzyne than does furan.
Having established the ability of a HDDA-benzyne to
engage coumarin itself, we explored (a) several coumarin
derivatives (7b−e) as well as (b) several aryl-substituted triyne
substrates (18a−e) to establish some of the generality of the
process. The results are shown in Figure 5. In each instance,
only the major syn isomer of products 16 or 19 was isolated
and characterized (although when the crude product mixture
was analyzed, a second minor isomer was present).
We also examined the reaction of a tricyclic coumarin
derivative, namely, benzocoumarin 20a and its brominated
analogue, 20b. These were reacted with triyne 13 to give
chrysene derivatives 21a and 21b, respectively, by processes
that closely paralleled those with coumarin itself. Bromo
analogue 21b readily afforded phenyl-substituted chrysene 21c.
The molecular skeleton of compound 21b was established by a
single-crystal X-ray diffraction analysis, which also revealed the
twisted4 nature of the polycyclic indenochrysenone skeleton
(Figure 6).
Finally, we briefly re-examined the reaction of coumarin with
o-benzyne (3) itself,2i here generated via the Kobayashi
method.5 Most informative was an experiment performed in
CD3CN using silylated phenyl triflate 22 and CsF in the
presence of coumarin (7a, 3 equiv). Direct monitoring of the
reaction by 1H NMR spectroscopy (see the Supporting
1
product ratio (5:1, H NMR of the crude product mixture for
the neat reaction at 85 °C).
This preference for the regioselective addition of the
unsymmetrical benzyne dienophile to coumarin can be
explained by a careful examination of the two TS geometries.
As with o-benzyne itself, there is a significantly advanced
degree of bond formation at C3 versus C8a for each of the
transition structures TSsyn and TSanti, indicating substantially
asynchronous reactions. The shorter distance of 2.0 Å is
identical in both; curiously, the second partial bond is shorter
in TSanti (2.7 Å vs 3.0 Å in TSsyn) even though TSanti is slightly
higher in energy. Subtle remote steric compressions are likely
responsible for this seeming anomaly. The difference in
puckering angle at the benzenoid C8a in each is, again,
informative; the slightly larger deformation in TSanti (8.1°)
(and 17.2° at C3) versus that in TSsyn (5.5°) (and 17.5° at C3)
reflects a greater degree of sacrifice in aromaticity. Finally, we
note that an FMO analysis of this cycloaddition is not a
meaningful approach for rationalizing the sense of regiose-
lectivity. That is, the π-type orbital coefficients at C3 versus
C8a in the HOMO of coumarin are computed to be virtually
identical (see the Supporting Information for details).
The imperfect yield of this transformation implied that the
rate of the trapping by coumarin was slow (although
serviceable), again consistent with the earlier conclusion of
2i
́
Guitian et al. To further evaluate that point, consider the
computed activation barriers for the reactions of coumarin
2191
https://dx.doi.org/10.1021/acs.orglett.1c00342
Org. Lett. 2021, 23, 2189−2193

Organic Letters
pubs.acs.org/OrgLett
Letter
though the steady-state concentration of 3 is, of course, quite
small.7
In conclusion, we have described a new mode of aryne
reactivity with coumarins. Namely, coumarins and o-benzynes
undergo a [4+2] cycloaddition, albeit slowly, followed by a
cheletropic ejection of CO2 to afford conjugated polyaromatic
scaffolds. DFT computations have provided additional
mechanistic understanding of some of the elementary steps
involved in this class of transformation.
ASSOCIATED CONTENT
* Supporting Information
■
sı
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.orglett.1c00342.
Experimental procedures for the preparation of and
characterization data for all previously unknown
1
compounds, computational details, and copies of H
and 13C NMR spectra (PDF)
Accession Codes
Figure 5. (a) Reactions of triyne 13 with coumarin derivatives (7a−
e) to produce products 16a−e, respectively. (b) Reactions of triynes
CCDC 2057879 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.
a
18 with coumarin (7a) to produce products 19a−e. In the case of
19e, the reaction was performed on a 1 mmol scale and the minor anti
isomer was also isolated and characterized (see the Supporting
Information for details).
AUTHOR INFORMATION
Corresponding Author
■
Thomas R. Hoye − Department of Chemistry, University of
Minnesota, Minneapolis, Minnesota 55455, United States;
orcid.org/0000-0001-9318-1477; Email: hoye@
umn.edu
Authors
Bhavani Shankar Chinta − Department of Chemistry,
University of Minnesota, Minneapolis, Minnesota 55455,
United States
Daniel Lee − Department of Chemistry, University of
Figure 6. Reactions of triyne 13 with benzocoumarins 20a and 20b to
produce indenochrysenones 21a and 21b, respectively.
Minnesota, Minneapolis, Minnesota 55455, United States
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.orglett.1c00342
Information for details) clearly showed the formation of,
principally, phenanthrene (9) along with a smaller amount of
the known benzyne dimer, biphenylene6 (23), in a ratio of
4.8:1 (Figure 7). Thus, 7a is capable of trapping o-benzyne (3)
itself, but the reaction rate is relatively slow, because
dimerization of two molecules of 3 is competitive, even
Author Contributions
†B.S.C. and D.L. contributed equally to this work.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The contributions of Professor R. W. Hoffmann (University of
Marburg, Marburg, Germany) through both his Ph.D. thesis
research in the laboratory of G. Wittig (e.g., ref 1) and his 1967
classic monograph on “dehydrobenzene” (ref 8) are acknowl-
edged for providing a foundation that has guided an immense
amount of ensuing advances in aryne chemistry. Support for
this research was provided by the National Institutes of
General Medical Sciences (R35 GM127097) and the National
Science Foundation (CHE 1665389). Some of the NMR
spectral data were recorded using instrumentation funded
through the National Institutes of Health Shared Instrumenta-
tion Grant program (S10OD011952). Some of the mass
Figure 7. Reaction of coumarin (7a) and o-benzyne (3) affords
phenanthrene (9) and biphenylene (23). The isolated yield of co-
eluting hydrocarbon products was 19%; analysis of the reaction
1
mixture directly by H NMR spectroscopy indicates that 9 and 23
were, by far, the major products produced (see the Supporting
Information).
2192
https://dx.doi.org/10.1021/acs.orglett.1c00342
Org. Lett. 2021, 23, 2189−2193

Organic Letters
pubs.acs.org/OrgLett
Letter
(6) (a) First observation of biphenylene (23) from o-benzyne:
spectrometry data were collected at the Analytical Biochem-
istry Shared Resource in the Masonic Cancer Center at the
University of Minnesota with instrumentation funded in part
by a Cancer Center Support Grant (CA-77598). X-ray data
were collected on a diffractometer purchased with funds from a
grant from the National Science Foundation (MRI 1229400).
̈
̈
Wittig, G.; Pohmer, L. Uber das intermediare Auftreten von
Dehydrobenzol. Chem. Ber. 1956, 89, 1334−1351. Biphenylene
(23) as a byproduct from (o-trimethylsilyl)phenyl triflate:
(b) Okuma, K.; Nojima, A.; Nakamura, Y.; Matsunaga, N.;
Nagahora, N.; Shioji, K. Reaction of benzyne with formamides and
acetylimidazole. Bull. Chem. Soc. Jpn. 2011, 84, 328−332.
(7) For a more general discussion of biphenylenes from benzynes,
see: Takano, H.; Ito, T.; Kanyiva, K. S.; Shibata, T. Recent advances
of biphenylene: Synthesis, reactions and uses. Eur. J. Org. Chem. 2019,
2019, 2871−2883.
REFERENCES
■
(1) Wittig, G.; Hoffmann, R. W. Dehydrobenzol aus 1.2.3-
Benzothiadiazol-1.l-dioxyd. Chem. Ber. 1962, 95, 2718−2728.
(8) Hoffmann, R. W. Dehydrobenzene and Cycloalkynes; Organic
Chemistry, a Series of Monographs 11; Academic Press: New York,
1967.
(2) (a) Meier, H.; Molz, T.; Merkle, U.; Echter, T.; Lorch, M.
Aromatensynthese aus 1,2,3-Selenadiazolen und α-Pyronen. Liebigs
Ann. Chem. 1982, 1982, 914−923. (b) Moody, C. J. Diels-Alder
reactivity of pyrano < 3,4-b > indol-3-ones, stable, synthetic
equivalents of indole-2,3-quinodimethanes. J. Chem. Soc., Chem.
Commun. 1984, 925−926. (c) Moody, C. J. Diels-Alder reactivity of
pyrano < 3,4-b > indol-3-ones, stable analogues of indole-2,3-
quinodimethanes. J. Chem. Soc., Perkin Trans. 1 1985, 2505−2508.
(d) Jackson, P. M.; Moody, C. J. Diels-Alder reactions of 2,3-
dimethylenepyrrole analogues; a new synthesis of indoles. J. Chem.
Soc., Perkin Trans. 1 1990, 2156−2158. (e) Pérez, D.; Guitián, E.;
Castedo, L. A new approach to the synthesis of antitumor
benzophenanthridine alkaloids. Formal synthesis of nitidine. J. Org.
Chem. 1992, 57, 5911−5917. (f) Jackson, P. M.; Moody, C. J. Diels-
Alder reactions of 1,5-dihydropyrano[3,4-b]pyrrol-5(1H)-ones, pyr-
role-2,3-quinodimethane analogues; a new synthesis of indoles.
Tetrahedron 1992, 48, 7447−7466. (g) Martín, G.; Guitián, E.;
Castedo, L.; Saá, J. M. Intermolecular benzyne cycloaddition (IBC), a
versatile approach to benzophenanthridine antitumor alkaloids.
Formal synthesis of nitidine and chelerythrine. J. Org. Chem. 1992,
57, 5907−5911. (h) Andrews, J. F. P.; Jackson, P. M.; Moody, C. J.
Pyrrole-2,3-quinodimethane analogues in the synthesis of indoles.
Part 2.1 Synthesis and Diels-Alder reactions of 1,6-dihydropyrano-
[4,3-b]pyrrol6(1H)-ones. Tetrahedron 1993, 49, 7353−7372. (i) Es-
cudero, S.; Pérez, D.; Guitián, E.; Castedo, L. [4 + 2] Cycloadditions
between 2-pyrones and benzyne. Application to the synthesis of
binaphthyls. Tetrahedron Lett. 1997, 38, 5375−5378. (j) Escudero, S.;
Pérez, D.; Guitián, E.; Castedo, L. A new convergent approach to the
polycyclic framework of dynemicin A. J. Org. Chem. 1997, 62, 3028−
3029. (k) Palacios, F.; Ochoa de Retana, A. M.; Martînez de
Marigorta, E.; Manuel de los Santos, J. 2H-Azirines as synthetic tools
in organic chemistry. Eur. J. Org. Chem. 2001, 13, 2401−2414.
(l) Shah, T. K.; Medina, J. M.; Garg, N. K. Expanding the strained
alkyne toolbox: Generation and utility of oxygen-containing strained
alkynes. J. Am. Chem. Soc. 2016, 138, 4948−4954. (m) Darzi, E. R.;
Barber, J. S.; Garg, N. K. Cyclic alkyne approach to heteroatom-
containing polycyclic aromatic hydrocarbon scaffolds. Angew. Chem.,
Int. Ed. 2019, 58, 9419−9424. (n) Meguro, T.; Chen, S.; Kanemoto,
K.; Yoshida, S.; Hosoya, T. Modular synthesis of unsymmetrical
doubly-ring-fused benzene derivatives based on a sequential ring
construction strategy using oxadiazinones as a platform molecule.
Chem. Lett. 2019, 48, 582−585.
(3) (a) For a recent, comprehensive review of HDDA benzyne
chemistry, see: Fluegel, L. L.; Hoye, T. R. Hexadehydro-Diels-Alder
reaction: Benzyne generation via cycloisomerization of tethered
triynes. Chem. Rev. 2021, 121, 2413−2444. (b) For a recent,
comprehensive review of Kobayashi benzayne chemistry see: Shi, J.;
Li, L.; Li, Y. o-Silylaryl triflates: A journey of Kobayashi aryne
precursors. Chem. Rev. 2021, DOI: 10.1021/acs.chemrev.0c01011.
(4) For reviews of helicenes, see: (a) Shen, Y.; Chen, C.-F.
Helicenes: Synthesis and applications. Chem. Rev. 2012, 112 (3),
1463−1535. (b) Wang, X.-Y.; Yao, X.; Mullen, K. Polycyclic aromatic
̈
hydrocarbons in the graphene era. Sci. China: Chem. 2019, 62, 1099−
1144. (c) Mori, T. Chiroptical properties of symmetric double, triple,
and multiple helicenes. Chem. Rev. 2021, 121, 2373.
(5) Himeshima, Y.; Sonoda, T.; Kobayashi, H. Fluoride-induced 1,2-
elimination of o-(trimethylsilyl)phenyl triflate to benzyne under mild
conditions. Chem. Lett. 1983, 12, 1211−1214.
2193
https://dx.doi.org/10.1021/acs.orglett.1c00342
Org. Lett. 2021, 23, 2189−2193