Atom economical, one-pot, three-reaction cascade to novel tricyclic 2,4-dihydro-1 H-benzo[ f ]isochromenes

Article abstract of DOI:10.1021/ol401600j

Reaction of 6-methyl-1-phenylhept-3-yne-2,6-diol with various aldehydes under Lewis acid conditions provides an atom economical, two-component cascade reaction sequence to novel 2,4-dihydro-1H-benzo[f]isochromene compounds. Aliphatic aldehydes as well as

Full text of DOI:10.1021/ol401600j

ORGANIC
LETTERS
2013
Vol. 15, No. 16
4070–4073
Atom Economical, One-Pot,
Three-Reaction Cascade to Novel Tricyclic
2,4-Dihydro‑1H‑benzo[f]isochromenes
Robert J. Hinkle* and Shane E. Lewis
Department of Chemistry, The College of William & Mary, P.O. Box 8795,
Williamsburg, Virginia 23187-8795, United States
rjhink@wm.edu
Received June 6, 2013
ABSTRACT
Reaction of 6-methyl-1-phenylhept-3-yne-2,6-diol with various aldehydes under Lewis acid conditions provides an atom economical, two-
component cascade reaction sequence to novel 2,4-dihydro-1H-benzo[f]isochromene compounds. Aliphatic aldehydes as well as electron-
deficient and -rich aromatic aldehydes can be used.
In efforts to increase atom economy,1 reduce harmful
environmental effects, and increase the overall efficiency
of experimental sequences, one-pot multicomponent reac-
tions2 and cascade3 or domino reactions4 have become much
more common in organic synthesis.5,6 We have recently
described a three-component additionÀcyclization protocol7
as well as a mechanistic investigation of a Bi(III)-initiated
silyl-Prins cyclization toward dihydropyrans.8 Herein, we
report a cascade sequence involving an alkynyl-Prins
cyclization, FriedelÀCrafts arylation, and dehydration/
aromatization to afford unique tricyclic benzo[f]isochromene
derivatives9 in which only H2O is generated as a byproduct
from the two reaction components. A minimum of 72%
average yields are observed for each of the reactions
involved in the cascade.
The Prins cyclization10,11 and related reactions12,13 are
versatile methods for the synthesis of dihydropyran and
tetrahydropyran rings as well as numerous derivatives.
Recently, a tandem intermolecular Prins/FriedelÀCrafts
reaction wasreported tofurnish 4-aryldihydropyrans from
homopropargylic alcohols and 4-aryltetrahydropyrans
from homoallylic alcohols (Scheme 1) when the aromatic
compound was used as solvent.14 An intramolecular
(1) (a) Sheldon, R. A. Chem. Soc. Rev. 2012, 41, 1437–1451. (b) Trost,
B. M. Handbook of Green Chemistry 2012, 7, 1–33. (c) Trost, B. M.
Angew. Chem., Int. Ed. Engl. 1995, 34, 259–281.
ꢀ
(2) Toure, B. B.; Hall, D. G. Chem. Rev. 2009, 109, 4439–4486.
(3) (a) Kleinke, A. S.; Webb, D.; Jamison, T. F. Tetrahedron 2012, 68,
6999–7018. (b) Nicolaou, K. C.; Chen, J. S. Chem. Soc. Rev. 2009, 38,
2993–3009.
(4) Tietze, L.-F. Domino Reactions in Organic Synthesis; Wiley-VCH:
Weinheim, 2006.
(5) For reactions described in the literature as cascade reactions, see: (a)
Horning, B. D.; MacMillan, D. W. C. J. Am. Chem. Soc. 2013, 135, 6442–
6445. (b) Ohno, H. Asian J. Org. Chem. 2013, 2, 18–28. (c) Pintado-Sierra,
ꢀ
M.; Rasero-Almansa, A. M.; Corma, A.; Iglesias, M.; Sanchez, F. J. Catal.
2013, 299, 137–145. (d) Surendra, K.; Qiu, W.; Corey, E. J. J. Am. Chem.
Soc. 2011, 133, 9724–9726. (e) Li, H.; Loh, T.-P. Org. Lett. 2010, 12, 2679–
2681. (f) Dalgard, J. E.; Rychnovsky, S. D. Org. Lett. 2005, 7, 1589–1591.
(6) For reactions described as domino sequences, see: (a) Liu, X.-L.;
Han, W.-Y.; Zhang, X.-M.; Yuan, W.-C. Org. Lett. 2013, 15, 1246–
1249. (b) Long, Y.; She, Z.; Liu, X.; Chen, Y. J. Org. Chem. 2013, 78,
2579–2588. (c) Lu, B. Z.; Wei, H.-X.; Zhang, Y.; Zhao, W.; Dufour, M.;
Li, G.; Farina, V.; Senanayake, C. H. J. Org. Chem. 2013, 78, 4558–4562.
(7) Hinkle, R. J.; Lian, Y.; Speight, L. C.; Stevenson, H. E.; Sprachman,
M. M.; Katkish, L. A.; Mattern, M. C. Tetrahedron 2009, 65, 6834–6839.
(8) Lambert, R. F.; Hinkle, R. J.; Ammann, S. E.; Lian, Y.; Liu, J.;
Lewis, S. E.; Pike, R. D. J. Org. Chem. 2011, 76, 9269–9277.
(9) Fused pyrans have been investigated as antioxidants and for anti-
platelet and antiproliferative properties: (a) Reddy, K. R.; Longato, G. B.; de
Carvalho, J. E.; Ruiz, A. L.; Silva, L. F., Jr. Molecules 2012, 17, 9621–9630.
(b) Rafinejad, A.; Fallah-Tafti, A.; Tiwari, R.; Shirazi, A. N.; Mandal, D.;
Shafiee, A.; Parang, K.; Foroumadi, A.; Akbarzadeh, T. DARU J. Pharm.
Sci. 2012, 20, 100. (c) Guiso, M.; Marra, C.; Arcos, R. R. Nat. Prod. Res.
2008, 22, 1403–1409. (d) Lorenz, P.; Zeh, M.; Martens-Lobenhoffer, J.;
Schmidt, H.; Wolf, G.; Horn, T. F. W. Free Radical Res. 2005, 39, 535–545.
(10) For recent reviews on the Prins reaction and Prins cyclization in
organic synthesis, see: (a) Han, X.; Peh, G.; Floreancig, P. E. Eur. J. Org.
Chem. 2013, 1193–1208. (b) Pastor, I. M.; Yus, M. Curr. Org. Chem.
2012, 16, 1277–1312.
r
10.1021/ol401600j
Published on Web 08/01/2013
2013 American Chemical Society

might be accessible utilizing a cascade sequence involving
an alkynyl-Prins reaction17,18 and FriedelÀCrafts cycliza-
tion19 from an unsymmetrical alkynediol and an aldehyde
(RÀCHO) under Lewis acid mediated conditions (Scheme 2).
Scheme 1. Intermolecular FriedelÀCrafts
Scheme 2. Retrosynthetic Analysis
variation of this reaction with homoallylic alcohols pro-
vided a number of tricyclic compounds via an intermediate
secondary alkyl carbocation.15 While a number of fused
polycyclic pyran derivatives exist in the literature, this
represented a novel approach to their construction, as
many other examples use substrates with a preformed ring
system.16 In contrast to these results, the reaction we
describe occurs via a putative, high-energy alkenyl cation
and includes a subsequent dehydrative aromatization to
afford 2,4-dihydro-1H-benzo[f]isochromenes, which also
represent a novel class of heterotricycles.
The requisite unprotected alkynediol 2a was synthesized
from the known propargylic alcohol 1a20 by conversion to
the dianion with n-BuLi followed by addition to isobutyl-
ene oxide in the presence of BF3 OEt2 (eq 1).21
3
Based on our previous work, we envisioned that a tricyclic
2,4,5,6-tetrahydro-1H-benzo[f]isochromen-5-ol product
(11) For mechanistic studies on the Prins reaction, see: (a) Ogawa, Y.;
Painter, P. P.; Tantillo, D. J.; Wender, P. A. J. Org. Chem. 2013, 78, 104–
115. (b) Zheng, K.; Liu, X.; Qin, S.; Xie, M.; Lin, L.; Hu, C.; Feng, X.
J. Am. Chem. Soc. 2012, 134, 17564–17573. (c) Chio, F. K.; Warne, J.;
Gough, D.; Penny, M.; Green, S.; Coles, S. J.; Hursthouse, M. B.; Jones,
P.; Hassall, L.; McGuire, T. M.; Dobbs, A. P. Tetrahedron 2011, 67,
5107–5124. (d) Gesinski, M. R.; Tadpetch, K.; Rychnovsky, S. D. Org.
Lett. 2009, 11, 5342–5345. (e) Gesinski, M. R.; Van, O. L. J.; Rychnovsky,
S. D. Synlett 2008, 363–366. (f) Tadpetch, K.; Rychnovsky, S. D. Org. Lett.
2008, 10, 4839–4842. (g) Jasti, R.; Rychnovsky, S. D. J. Am. Chem. Soc.
2006, 128, 13640–13648. (h) Jasti, R.; Rychnovsky, S. D. Org. Lett. 2006, 8,
2175–2178. (i) Jasti, R.; Anderson, C. D.; Rychnovsky, S. D. J. Am. Chem.
Soc. 2005, 127, 9939–9945.
(12) The “silyl-Prins” reaction has been studied extensively: (a)
Hemery, T.; Wibbeling, B.; Froehlich, R.; Hoppe, D. Synthesis 2010,
329–342. (b) Meilert, K.; Brimble, M. A. Org. Biomol. Chem. 2006, 4,
2184–2192. (c) Dobbs, A. P.; Guesne, S. J. J. Synlett 2005, 2101–2103. (d)
Semeyn, C.; Blaauw, R. H.; Hiemstra, H.; Speckamp, W. N. J. Org.
Chem. 1997, 62, 3426–3427.
Several different Lewis acid conditions were examined
with alkynediol 2a and propionaldehyde or isobutyralde-
hyde as electrophiles (Table 1). When using propionalde-
hyde (entry 1) relatively low yields were obtained with
1.1 equiv of BF3 OEt2 (average of 72% per reaction).
3
Reactions with the silylated derivative 2b (eq 1) and
stoichiometric Bi(OTf)3 made isolation of 3a more difficult
(entry 2). We also attempted cyclizations with the mono-
TBDPS protected analog 2c, but recovered starting mate-
rials. Reactionsusing isobutyraldehyde weremoreefficient
(entries 3À6), although stoichiometric amounts of BiBr3
and Bi(OTf)3 provided 3b in lower yields (entries 4 and 5).
Halving the concentration of the reaction had no effect on
the isolated yield (entry 6).
(13) The intramolecular silyl-mediated Sakuria reaction (ISMS) in-
volves the same intermediate cation as the silyl-Prins reaction: (a)
ꢀ
Marko, I. E.; Plancher, J.-M. Tetrahedron Lett. 1999, 40, 5259–5262.
ꢀ
(b) Marko, I. E.; Bayston, D. J. Tetrahedron 1994, 50, 7141–7156.
(14) Reddy, U. C.; Saikia, A. K. Synlett 2010, 1027–1032.
(15) (a) Subba Reddy, B. V.; Kumar, H.; Borkar, P.; Yadav, J. S.;
Sridhar, B. Eur. J. Org. Chem. 2013, 1993–1999. (b) Reddy, B. V. S.;
Jalal, S.; Borkar, P.; Yadav, J. S.; Reddy, P. P.; Kunwar, A. C.; Sridhar,
B. Org. Biomol. Chem. 2012, 10, 6562–6568.
As a result of our success with catalytic quantities of
Bi(III) compounds in the synthesis of dihydropyrans,
catalytic quantities (10À20%) of BiBr3 and Bi(OTf)3 were
screened as potential initiators,7,8 but starting materials
(16) (a) Guo, K.;Chen, Y. Mater. Chem. Phys. 2013, 137, 1062–1066. (b)
ꢀ
Schevenels, F.; Tinant, B.; Declercq, J.-P.; Marko, I. E. Chem.;Eur. J.
2013, 19, 4335–4343. (c) Yuan, F.-Q.; Han, F.-S. Adv. Synth. Catal. 2013,
355, 537–547. (d) Shekhar, A. C.; Kumar, A. R.; Sathaiah, G.; Raju, K.;
Rao, P. S.; Sridhar, M.; Narsaiah, B.; Srinivas, P. V. S. S.; Sridhar, B. Helv.
Chim. Acta 2012, 95, 502–508. (e) Sousa, C. M.; Berthet, J.; Delbaere, S.;
Coelho, P. J. J. Org. Chem. 2012, 77, 3959–3968. (f) Thirunavukkarasu,
V. S.; Donati, M.; Ackermann, L. Org. Lett. 2012, 14, 3416–3419. (g) Silva,
L. F., Jr.; Quintiliano, S. A. Tetrahedron Lett. 2009, 50, 2256–2260.
(17) (a) Gharpure, S. J.; Prasath, V. J. Chem. Sci. 2011, 123, 943–949.
(b) Zheng, D.-P.; Gong, W.-C.; Ma, Z.-D.; Ma, B.-Z.; Zhao, X.-L.; Xie,
Z.-X.; Li, Y. Tetrahedron Lett. 2011, 52, 314–317. (c) Kim, C.; Bae, H. J.;
Lee, J. H.; Jeong, W.; Kim, H.; Sampath, V.; Rhee, Y. H. J. Am. Chem.
(18) The descriptor “Prins cyclization” has been used inconsistently
in the literature. However, many authors now interpret the typical Prins
cyclization to involve formation of two bonds in the tetrahydropyran
product, and we are using this definition. See ref 10b for a discussion of
the Prins cyclization.
(19) (a) Reddy, B. V. S.; Reddy, P. S.; Yadav, J. S.; Sridhar, B.
Tetrahedron Lett. 2013, 54, 1392–1396. (b) Suzuki, Y.; Niwa, T.; Yasui,
E.; Mizukami, M.; Miyashita, M.; Nagumo, S. Tetrahedron Lett. 2012,
53, 1337–1340. (c) Basavaiah, D.; Reddy, K. R. Org. Lett. 2006, 9, 57–60.
(20) Xu, C.-F.; Xu, M.; Yang, L.-Q.; Li, C.-Y. J. Org. Chem. 2012,
77, 3010–3016.
Soc. 2009, 131, 14660–14661. (d) Miranda, P O.; Carballo, R. n. M.;
ꢀ
Martı
´
n, V. c. S.; Padrοn, J. I. Org. Lett. 2008, 11, 357–360.
(21) Yamaguchi, M.; Hirao, I. Tetrahedron Lett. 1983, 24, 391–394.
Org. Lett., Vol. 15, No. 16, 2013
4071

Table 1. Brief Survey of Lewis Acid Activators
Table 2. One-Pot Cascade Synthesis of Benzo[f]isochromenes
entry
product
aldehyde (R)
yield (%)a
1
2
3a
3b
3c
3d
3e
3f
EtÀ
38
47
52
44
43
58
58
43
46
58
i-PrÀ
3
sec-BuÀ
4
pentan-3-ylÀ
PhCH2CH2À
PhÀ
5
6
7
3g
3h
3i
p-Br-PhÀ
p-CF3-PhÀ
p-NO2-PhÀ
p-MeO-PhÀ
8
9
10
3j
a Isolated yields of compounds after silica gel column chromatography.
a Reactions were run under argon at ∼0.25 M in 2a using 1.2 equiv of
aldehyde and 1.1 equiv of corresponding Lewis acid. b Isolated yields of
compounds after silica gel chromatography. c The bis-TES protected
analog 2b was used. d Reaction conducted at ∼0.125 M.
Scheme 3. Air Oxidation of 3a and ORTEP of Lactone 4
were largely recovered. Although the reaction did proceed
with stoichiometric amounts of both reagents (entries 2, 4,
and 5), the high molecular weight and associated cost
limited the appeal of each. Other, less expensive Lewis
acids such as TiCl4 and SnCl4 were examined in catalytic
and stoichiometric quantities, but each was more hazard-
ous to dispense and led to more complex mixtures as
well as decomposition of starting materials. Overall, a
both electron-withdrawing and -donating substituents. These
aryl-substituted benzo[f]isochromenes were significantly
more stable than the corresponding alkyl-substituted pro-
ducts 3aÀ3e. This trend was also verified by GC MS analysis
in which the molecular ions of the aryl-substituted substrates
3fÀ3j were much more evident relative to fragments.
10 mol % excess of BF3 OEt2 over diol 2a provided the
least complex reaction mixtures (entries 1 and 3).
3
Alkynediol 2a was then reacted with various other
aldehydes in the presence of 1.1 equiv of BF3 OEt2 at rt
3
in CH2Cl2 (Table 2). Compounds 3aÀ3j were all isolated
as extremely viscous oils after column chromatography
We believe that the reaction progresses through a se-
quence of steps outlined in Scheme 4. Either of the two
hydroxyls in 2a could attack the putative Lewis acid
activated aldehyde generating intermediate I or IV via
Path A or B, respectively (Scheme 4). Subsequent forma-
tion of oxocarbenium ion II or V then allows for progres-
sion of the cascade. Intramolecular 6-endo attack of the
alkyne via Path A would generate III, the alkenyl cation of
a dihydropyran, whereas intramolecular attack of the
alkyne via Path B would generate VI, the vinyl cation of
a dihydrofuran. Classic studies conducted by Mayr et al.
established that cyclic, five-membered alkenyl cations are
approximately 10 kcal/mol higher in energy than cyclic,
six-membered analogs, which are in turn similar to open-
chain alkenyl cations.24 In accord with this difference in
1
and were characterized by IR, H, 13C APT, and 13C
DEPT NMR spectroscopy as well as combustion or mass
spectral analysis.
Alkyl-substituted products, 3aÀ3e, were generally less
stable than the corresponding aryl-substituted analogs,
3fÀ3j, and decomposed at rt when exposed to air. As
reported by Guiso et al., analogs with only one aromatic
ring (isochromans) are easily oxidized to lactones.22 Based
on X-ray crystallographic analysis of a sample left at rt in
air, the alkyl 2,4-dihydro-1H- benzo[f]isochromenes 3aÀ3f
herein appear to be prone to oxidation via a similar path-
way (Scheme 3).23
Reactions using benzaldehyde and substituted benzalde-
hydes afforded benzo[f]isochromenes (3fÀ3j) containing
(22) Guiso, M.; Marra, C.; Piccioni, F. Nat. Prod. Res. 2010, 24, 331–
340.
(23) See the Supporting Information for details of the X-ray analysis.
(24) Mayr, H.; Schneider, R.; Wilhelm, D.; Schleyer, P. V. R. J. Org.
Chem. 1981, 46, 5336–5340.
4072
Org. Lett., Vol. 15, No. 16, 2013

Scheme 4. Mechanistic Rationale for Product Formation
energies, solvolyses of cyclopentenyl nonaflates occur by
SÀO bond cleavage rather than alkenyl cation forma-
tion.25 As predicted by these energetic differences, we do
not observe products from cyclization by the higher energy
manifold (Path B); however, some complex polymeriza-
tion products are generated, and we cannot completely
discount this intermediate in the formation of those mate-
rials. In Path A, a subsequent intramolecular FriedelÀ
Crafts reaction would occur at the ortho-position of the
pendant aromatic ring. After rearomatization of the ori-
ginal phenyl moiety, the central ring is formed and acid-
catalyzed dehydration/aromatization26 occurs to generate
the observed benzo[f]isochromenes 3aÀ3j. In the case of
protected alkynediol 2c (vide supra), the large TBDPS
group likely prevented cyclization due to severe steric
interactions, and the initial intermolecular addition to I is
reversible (Scheme 4, Path A).
contain an alkenyl cation that would be destabilized by the
adjacent hydroxyl group. Therefore, the products formed
proceed through Path A.
In summary, 2,4-dihydro-1H-benzo[f]isochromenes3aÀ3j
are novel tricyclic compounds that are obtained in four
chemical steps from commercially available trimethylsilyl-
acetylene. The only ring present in the initial compound is
the phenyl moiety shown in compounds 1aÀ1c and 2aÀ2c.
Conversion of 2a to the tricyclic products is achieved by an
atom economical cascade involving an alkynyl-Prins cycli-
zation, FriedelÀCrafts arylation, and dehydrative aroma-
tization. Although overall yields appear modest, three
separate reactions are part of the cascade, and the average
yields for eachofthesereactionsinvolvedinthecascade are
a minimum of 72%. Further studies involving a wider
variety of substrates and further optimization are under-
way and will be reported in due course.
It is somewhat suprising that tetrahydrofuran products
are not observed. Rychnovsky and co-workers27 as well as
Cho and co-workers28 reported that internal alkynes af-
ford 5-exo products rather than the 6-endo adducts we
observe. However, a 5-exo reaction of intermediate II
would produce VII (Scheme 4, lower right), which would
Acknowledgment. We thank the AREA Program of the
National Institutes of Health (RGM072525-01A2) as well
as the National Science Foundation (CHE-1012305) for
support of this work. We also thank Robert D. Pike, PhD,
for X-ray structure analysis of lactone 4.
Supporting Information Available. X-ray crystallo-
graphic tables of lactone 4 and characterization data,
including 1H and 13C APT, and 13C DEPT NMR spectra
for 2aÀ2c and 3aÀ3j; GC MS traces also included for
3aÀ3j. This material is available free of charge via the
Internet at http://pubs.acs.org.
(25) Apeloig, Y.; Muller, T. Theory and calculations for vinyl
cations. In Dicoordinated Carbocations; Rappoport, Z., Stang, P. J., Eds.;
Wiley: 1997; pp 9À104.
(26) For an example of a FriedelÀCrafts reaction followed by a
dehydration/aromatization, see: Kuninobu, Y.; Tatsuzaki, T.; Matsuki,
T.; Takai, K. J. Org. Chem. 2011, 76, 7005–7009.
(27) Jaber, J. J.; Mitsui, K.; Rychnovsky, S. D. J. Org. Chem. 2001,
66, 4679–4686.
(28) Chavre, S. N.; Choo, H.; Cha, J. H.; Pae, A. N.; Choi, K. I.; Cho,
Y. S. Org. Lett. 2006, 8, 3617–3619.
The authors declare no competing financial interest.
Org. Lett., Vol. 15, No. 16, 2013
4073