Synthesis of angularly fused aromatic compounds from alkenyl enediynes by a tandem radical cyclization process

Article abstract of DOI:10.1002/anie.201103318

Let's get radical: A general synthetic route toward angularly ortho-fused polyaromatic [4]helicenes starting from aryl alkenyl N-substituted cyclic enediynes is described (see scheme; DMSO=dimethyl sulfoxide, Ns=4-nitrobenzenesulfonyl). The process involved a Bergman cyclization (BC) as the key step of an unprecedented tandem radical reaction.

Full text of DOI:10.1002/anie.201103318

Communications
DOI: 10.1002/anie.201103318
Radical Reactions
Synthesis of Angularly Fused Aromatic Compounds from Alkenyl
Enediynes by a Tandem Radical Cyclization Process**
Snigdha Roy, Anakuthil Anoop, Kumar Biradha, and Amit Basak*
The synthetic potential of the Bergman cyclization (BC)[1] has
not been greatly explored despite remarkable progress in the
understanding of the reaction mechanism[2] and the mode of
biological action of enediynes.[3] The radicals within the 1,4-
diradical generated during the BC are distally oriented, thus
ring[10] positioned above one of the alkynes of an enediyne
framework would have on the BC. We did observe some
interesting variations in reactivity depending on the R
substituent present on the aromatic ring. However, the
more important aspect is the synthesis of the angularly
fused polyaromatic compounds [4]helicenes (C)[11] in high
yields. The process, which involves the BC as the key step of
an unprecedented tandem radical reaction, offers a general
route to these compounds and also expands the synthetic
potential of the BC.
The aryl enediynes 1 required for our study were prepared
from o-iodo propargyl amine 2[12] by using a six-step protocol.
A Sonogashira coupling[13] with trimethylsilyl acetylene
followed by a desilylation produced the enediyne 4. Another
coupling reaction with the Z-iodo alkene (obtained by a halo-
Scheme 1. Synthesis of [4]helicenes.
preventing self-quenching, which can lead to a
highly strained bicyclo system, and are spin-
paired by an external quencher.[4] The 1,4-
diradical can undergo polymerization,[5] which
causes the usual low yield of cyclized products.
BC and related reactions involve the formation
of benzenoid frameworks and, hence, are
attractive for the synthesis of polyaromatic
and benzannulated compounds. John and
Tour[6] reported the formation of polypheny-
lenes using BC, while Grissom and Calkins[7]
demonstrated the tandem radical cyclization of
enediynes with diverse alkenyl acceptors. Di-
radicals generated in a porphyrin network have
been trapped with a neighboring aromatic ring
(Smith and co-workers,[8a] and Zaleski and co-
workers[8b]). Taking a cue from these results and
Scheme 2. Synthesis of target enediynes 1a–i. Reagents and conditions: a) TMS-
acetylene, Pd(0), CuI, Et3N, THF, RT, 8 h, 75%; b) KF, CH3OH, RT, 1 h, 80%; c) 5,
Pd(0), CuI, Et3N, THF, 12 h, 45–56%; d) PPTs, EtOH, RT, 6 h, 82–85%; e) MsCl, Et3N,
CH2Cl2, 08C, 5 min, 55–62%; f) K2CO3, DMF, RT, 4 h, 54–70%. DMF=N,N’-dimethyl-
formamide, Ms=methanesulfonyl, Ns=4-nitrobenzenesulfonyl, PPTs=pyridinium p-
toluenesulfonate, THF=tetrahydrofuran, THP=tetrahydropyranyl, TMS=trimethylsilyl.
the ortho effect reported by Alabugin and co-
workers,[9] we studied the reactivity of aryl
exomethylene N-substituted cyclic enediynes
of type A (Scheme 1). The initial intention was
to explore the effect the p cloud of an aromatic
Wittig reaction)[14] followed by THP removal furnished the
acyclic enediyne 7. This was converted into the mesylate 8,
which on treatment with K2CO3 in anhydrous DMF[15]
produced the cyclic enediyne 1 (Scheme 2). NOESY spectra
(see the Supporting Information) confirmed the positioning
of the aryl ring to be above the enediyne alkyne.
As a test study, enediyne 1 was dissolved in [D6]DMSO
and kept at 908C (Scheme 3). The reaction was monitored by
recording the 1H NMR spectra at different times. There was a
gradual decrease in the signals for the substrate accompanied
[*] Dr. S. Roy, Prof. A. Anoop, Prof. K. Biradha, Prof. A. Basak
Department of Chemistry, Indian Institute of Technology
Kharagpur 721 302 (India)
E-mail: absk@chem.iitkgp.ernet.in
[**] DST is acknowledged for an SERC grant to A.B. and A.A. which
supported this research. S.R. is grateful to CSIR, Government of
India, for a research fellowship. DST is also thanked for the NMR
and X-ray facility under the IRPHA and FIST programme, respec-
tively. A.A. acknowledges IIT Kharagpur for an ISIRD grant.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201103318.
8316
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 8316 –8319

Scheme 3. Possible BC products. DMSO=dimethylsulfoxide.
by the appearance of new signals corresponding to product.
The reaction conversion was mostly free from unwanted side
reactions (see the NMR studies in the Supporting Informa-
tion) and the products were isolated and purified by column
chromatography on silica gel. The 1H NMR spectrum showed
two sets of doublets, one for H1 and H3 each coupled to H2,
which gave a multiplet at d 7.71, and the other for H9 and H11
each coupled to H10, which gave a multiplet at d 7.69
(Figure 1a). This finding indicated the presence of two 1,2,3-
trisubstituted benzene rings as shown in structure N and thus
ruled out the normal BC product M as well as the alternate
structure O (Scheme 3). The mass spectra also supported the
presence of two deuterium atoms in the product. Carrying out
the reaction in DMSO furnished the fully protiated product,
from which two new signals appeared in the 1H NMR
spectrum, one at d 7.72 (s, H8) and the other (H12) obscured
in the region d 7.64–7.72; the combined integration for this
region amounted to six protons (Figure 1b). Final confirma-
tion of the structure came from single-crystal X-ray analysis[16]
of the product 9a (Figure 2). The reaction was found to be
very general, and a large array of differently substituted
[4]helicenes were obtained (Scheme 4). Considering the
concurrent triple annulation, the yields can be considered to
be impressive.
Figure 1. COSY spectra of 9 f performed in: a) deuteriated DMSO,
b) nondeuteriated DMSO.
Regarding the mechanism, we propose the following. The
tandem cyclization was initiated by the BC of 1 to furnish two
new rings, B and C (Scheme 5). The second step involved
addition of the aryl radical 10a to the proximally placed
aromatic ring E resulting in the simultaneous formation of
ring D and a new radical on ring E.[17] This radical, upon
hydrogen abstraction from ring A,[18] furnished another new
radical intermediate 10c. Abstraction of deuterium (hydro-
gen in the case of nondeuteriated DMSO) with a subsequent
oxidation produced the helicenes. The diradical 10a is
possibly quite shielded by the molecular framework, thus
making it almost inaccessible for polymerization or external
quenching.
Figure 2. ORTEP diagram of 9a with the thermal elipsoids shown at
30% probability.
substituents on ring E as well as their orientations (Table 1).
For different ortho substituents on ring E, the measured rate
constants varied in the range of 6 ꢀ 10À6 to 29 ꢀ 10À6 sÀ1. The
cyclization rates of compounds with nitro substituents at o, m,
and p positions follow a decreasing trend (1 f–h). To assess the
NMR-based kinetic studies were performed with sub-
strates 1a–i at 908C in [D6]DMSO to determine the effect of
the substituents on the rate of the annulations. A considerable
rate perturbation was observed upon variation of the
Angew. Chem. Int. Ed. 2011, 50, 8316 –8319
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
8317

Communications
role of the substituents on ring E (electronic or steric), the
activation energy for the BC step was computed using density
functional theory (for computational details see the Support-
ing Information). The simple phenyl-sustituted enediyne 1a
was taken as the reference. The activation free energy for the
BC step for 1a was calculated to be 34.06 kcalmolÀ1. The
phenyl group in 1a is involved in through-bond conjugation of
the p system extended over alkene and alkyne carbon atoms,
all of which lie nearly in the same plane. Substitution at the
ortho position forces the aryl ring to be tilted out of this plane
(Scheme 6), thus causing reduced conjugation and destabili-
Scheme 4. Synthesized [4]helicenes.
Scheme 6. a) Schematic drawing showing the steric interaction
between R and H for the structure with the phenyl ring in plane with
the olefinic and acetylenic carbon atoms, b) and c) G.S and T.S.
structures in which steric repulsion is avoided by having the phenyl
ring tilted, d) optimized geometry of 1 f in which the phenyl ring is
tilted as shown in (b).
zation of the system. In the transition-state geometries for all
the cases studied (1° a–d and 1° f–h), the phenyl group is also
tilted (Scheme 6c) and therefore the destabilization resulting
from the bulky substituent is relatively less in the transtition
state compared to that in the adducts. This greater destabi-
lization of 1 f as compared to 1a, is expected to reduce the
relative activation energy, thus causing a rate enhancement.[19]
Experimentally, a fivefold increase in the reaction rate
was observed when there was a nitro substituent in the ortho
position, thus giving a predicted difference in energy barrier
of about 1 kcalmolÀ1. The activation energy difference from
B3LYP calculation (1.31 kcalmolÀ1) is in reasonable agree-
ment with the experimental results. Thus, it can be concluded
that increasing the steric bulk of R will create more tilting of
the aryl ring and will thus cause a greater destabilization of
the starting enediyne and hence greater reactivity. The
calculated order, o-NO2 (1 f) > o-Me (1b) > o-OMe (1d) >
o-F (1c) > H (1a) > m-NO2 (1g) > p-NO2 (1h) roughly fol-
lows the steric bulk of R (Table 1).[20] The extent of tilting (F,
Table 1) also follows the same order. Except for the methoxy-
substituted phenyl (1d), which reacted at the slowest rate, the
predicted order is in agreement with the experimental
observation. Because of the rotational flexibility,[21] it is
difficult to rank the methoxy group according to its steric
bulk. Changing the position of the nitro substitution on the
phenyl group increases the activation free energy in the order
Scheme 5. Mechanism of tandem cyclization.
Table 1: Calculated and experimental results. F represents the tilting of
the substituted phenyl ring (Scheme 6b) measured as the torsion angle
made of C1-C2-C3-C4.
Enediyne
BP86
DEact DGact DEact
24.03 23.69 34.40 34.06
23.26 22.97 33.43 33.14 15.43
23.84 23.32 34.15 33.63
23.51 23.29 33.75 33.52
B3LYP [kcal
F [8]
Experimental rate
molÀ1
]
constant [ꢀ106 secÀ1
]
DGact
1a
1b
1c
1d
1 f
5.02
7.83
17.58
10.61
6.86
29.17
6.06
6.75
9.27
22.99 22.41 33.33 32.75 23.88
1g
1h
24.10 23.85 34.45 34.20
23.95 24.23 34.47 34.75
4.80
4.53
5.89
8318
www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 8316 –8319

[7] J. W. Grissom, T. L. Calkins, J. Org. Chem. 1993, 58, 5422.
[8] a) H. Aihara, L. Jaquinod, D. J. Nurco, K. M. Smith, Angew.
Chem. 2001, 113, 3547; Angew. Chem. Int. Ed. 2001, 40, 3439;
b) M. Nath, J. C. Huffman, J. M. Zaleski, Chem. Commun. 2003,
858.
ortho > meta > para. Thus, the steric effects are more pro-
nounced for the nitro group, because the electronic effect on
the ortho and para substitution is similar but there is a
difference in the energy barrier of 1.45 kcalmolÀ1. This comes
from a larger energy difference between the para- and ortho-
enediyne substrates (1g is more stable than 1 f by 4.46 kcal
molÀ1) than from their corresponding transition state
(3.01 kcalmolÀ1). Substitution at the meta and para positions
does not have any steric influence and therefore has similar
activation barriers (difference of ca. 0.5 kcalmolÀ1). This
small difference may be due to the electronic effects coming
from the meta and para positions. A similar trend of reactivity
was also obtained in BP86-based calculations.
[9] T. A. Zeidan, S. V. Kovalenko, M. Manoharan, I. V. Alabugin, J.
Org. Chem. 2006, 71, 962.
[10] A. Basak, S. S. Bag, A. K. Das, Eur. J. Org. Chem. 2005, 1239.
[11] a) I. Stary, I. G. Starꢁ, Z. Alexandrovꢁ, P. Sehnal, F. Teply, D.
Saman, L. Rulꢂsek, Pure Appl. Chem. 2006, 78, 495; b) A.
Urbano, Angew. Chem. 2003, 115, 4116; Angew. Chem. Int. Ed.
2003, 42, 3986; c) H. Hopf, Classics in Hydrocarbon Chemistry,
Wiley-VCH, Weinheim, 2000, p. 54; H. Aihara, L. Jaquinod, D. J.
Nurco, K. M. Smith, Angew. Chem. 2001, 113, 3547; Angew.
Chem. Int. Ed. 2001, 40, 3439; d) Ref. [8b]; Syntheses, Concepts,
Perspectives, Wiley-VCH, Weinheim, 2000, p. 323; e) T. J. Katz,
Angew. Chem. 2000, 112, 1997; Angew. Chem. Int. Ed. 2000, 39,
1921.
[12] This was prepared following the literature procedure, reported
earlier by our group, for the synthesis of similar compounds: A.
Basak, S. Mandal, A. K. Das, V. Bertolasi, Bioorg. Med. Chem.
Lett. 2002, 12, 873.
[13] K. Sonogashira, Y. Tohda, N. Hagihara, Tetrahedron Lett. 1975,
16, 4467.
In conclusion we have been successful in developing a BC-
mediated tandem radical route to [4]helicenes. The proposed
mechanism is well supported by the results in deuteriated and
nondeuteriated solvents. Presently we are working on extend-
ing the method to the synthesis of chiral helicenes.
Received: May 15, 2011
Published online: July 21, 2011
[14] X. Zhang, P. Zhong, F. Chen, Synth. Commun. 2004, 34, 1729.
[15] A. Basak, J. C. Shain, U. K. Khamrai, K. R. Rudra, A. Basak, J.
Chem. Soc. Perkin Trans. 1 2000, 1955.
[16] CCDC 823859 contains the supplementary crystallographic data
for this paper. These data can be obtained free of charge from
The Cambridge Crystallographic Data Centre via www.ccdc.
cam.ac.uk/data_request/cif.
Keywords: cyclizations · density functional calculations ·
enediyne · radicals · tandem reactions
.
[1] a) R. P. Jones, R. G. Bergman, J. Am. Chem. Soc. 1972, 94, 660;
for the scope of BC see: b) A. Basak, S. Mandal, S. S. Bag, Chem.
Rev. 2003, 103, 4077; M. Kar, A. Basak, Chem. Rev. 2007, 107,
2861.
[2] a) B. Kçnig, Angew. Chem. 1996, 198, 177; Angew. Chem. Int.
Ed. Engl. 1996, 35, 165; b) R. Lindh, T. J. Lee, A. Bernhardsson,
B. J. Persson, G. Karlstrom, J. Am. Chem. Soc. 1995, 117, 7186;
c) M. D. Lee, G. A. Ellestad, D. B. Borders, Acc. Chem. Res.
1991, 24, 235.
[3] K. C. Nicolaou, A. L. Smith, Acc. Chem. Res. 1992, 25, 497 – 503.
[4] J. J. Hangeland, J. J. De Voss, J. A. Heath, C. A. Townsend, W.-d.
Ding, J. S. Ashcroft, G. A. Ellestad, J. Am. Chem. Soc. 1992, 114,
9200 – 9202.
[5] J. D. Rule, J. S. Moore, Macromolecules 2005, 38, 7266.
[6] J. A. John, J. M. Tour, Tetrahedron 1997, 53, 15515.
[17] Zaleski and co-workers[8b] observed a rate acceleration in the
Bergman cyclization for porphyrinic enediynes from such an
addition of a radical to a suitably placed phenyl ring.
[18] Arene participation to radicals have been reported earlier: M. F.
Semmelhack, T. Neu, F. Foubelo, J. Org. Chem. 1994, 59, 5038.
[19] On the same grounds, the exomethylene-substituted enediyne
with no phenyl conjugation has a lower activation barrier of
32.02 kcalmolÀ1
.
[20] E. L. Eliel, S. H. Wilen, Stereochemistry of Organic Compounds,
Wiley, New York, 1993, p. 696.
[21] F. C. Pickard IV, R. L. Shepherd, A. E. Gillis, M. E. Dunn, S.
Feldgus, K. N. Kirschner, G. C. Shields, M. Manoharan, I. V.
Alabugin, J. Phys. Chem. A 2006, 110, 2517.
Angew. Chem. Int. Ed. 2011, 50, 8316 –8319
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
8319