Exo-dig radical cascades of skipped enediynes: Building a naphthalene moiety within a polycyclic framework

Article abstract of DOI:10.1002/chem.201304092

Cascade radical transformations of acyclic precursors open efficient, convenient and atom-economical access to functionalized compounds of increased structural complexity. This report describes a selective sequence of 5-exo-dig and 6-exo-dig cyclizations followed by attack at a pendant aromatic moiety and rearomatization.

Full text of DOI:10.1002/chem.201304092

DOI: 10.1002/chem.201304092
Communication
&
Polycyclic Aromatics
Exo-Dig Radical Cascades of Skipped Enediynes: Building
a Naphthalene Moiety within a Polycyclic Framework
Kamalkishore Pati, Audrey M. Hughes, Hoa Phan, and Igor V. Alabugin*[a]
Abstract: Cascade radical transformations of acyclic pre-
cursors open efficient, convenient and atom-economical
access to functionalized compounds of increased structur-
al complexity. This report describes a selective sequence
of 5-exo-dig and 6-exo-dig cyclizations followed by attack
at a pendant aromatic moiety and rearomatization.
The overall transformation is a new approach for building
a naphthalene moiety within a polycyclic framework. Further-
more, the high efficiency for the key 6-exo step of the cascade
paves the way for the preparation of defect-free graphene
nanoribbons. Radical cascades are valuable tools for the con-
struction of complex polycyclic frameworks.[1] At their best,
these reactions impart striking efficiency to synthetic strat-
egies.[2] The advantage of radical reagents over their ionic
counterparts is in the relatively broad functional group toler-
ance, mild reaction conditions, and the combination of high re-
activity with controllable selectivity.[3] Alkynes are attractive
precursors for the rapid construction of carbon-rich polycyclic
frameworks and materials due to the high carbon content,[4]
the possibility of modular assembly via reliable cross-coupling
chemistry, and controllable reactivity. We recently utilized
these features for the preparation of polyaromatic ribbons
from oligoalkynes through selective radical[5] and metal-cata-
lyzed cascade cyclizations,[6] each of which correspond to con-
trolled “polymerization” of alkyne moieties sandwiched be-
tween the two rows of aromatic rings (Scheme 1).
Scheme 1. Variations on radical cascade cyclizations of oligoalkynes for con-
jugated (left) and “skipped” substrates (right).
with conjugated oligoalkynes, the first cyclization that involves
two alkynes is the 5-exo cyclization. As a consequence, the
“polyacetylene ribbon” formed from the oligoalkyne always
starts with a pentagon. One of the questions that we wanted
to address in this work was whether the presence of this pen-
tagon can be avoided, that is, whether the first reaction be-
tween two alkynes could be a 6-exo-dig closure. A potential
limitation of this approach is that the rate of 6-exo-dig cycliza-
tion is expected to be approximately 50-fold slower than the
5-exo-dig process.[7]
In order to test the alternative design, we changed the oli-
goalkyne reactant from conjugated to “skipped” by adding
one extra carbon. In order to initiate the regioselective forma-
tion of the vinyl radical, we took advantage of our earlier find-
ing that CÀI bonds can be selectively activated in the presence
of several alkyne moieties by the Bu3SnH/AIBN system. The
requisite starting materials 1a–k are readily prepared from the
respective 2-bromobenzaldehydes by via the combination of
Sonogashira cross-coupling and nucleophilic addition of acety-
lide anions to the aldehyde[8] (Figure 1).
The radical version of such approaches to graphene-like
nanoribbons relied on a selective initial attack at the central
alkyne of the oligoalkyne precursor. Only when the central
alkyne is the initial target all alkyne moieties in the precursor
are fully converted into expanded polyaromatic framework
through a sequence of exo-dig cyclizations. For the oligoalkyne
with three triple bonds, such selectivity is achievable by the
use of selective intermolecular Bu3Sn radical attack at the cen-
tral alkyne.[5a] For the systems with four alkyne moieties, the se-
lectivity was achieved through covalent attachment of the ini-
tiating group near the central alkyne, so the first attack at the
triple bond is intramolecular.[5b] Because both approaches start
Sonogashira cross-coupling of 2-bromobenzaldehydes with
alkynes proceeded in good yields to produce the library of 2-
(phenylethynyl)benzaldehydes in 70–84% yields (Scheme 2).
Subsequent treatment with arylethynyl lithium produced the
“skipped” acetylenic alcohols. The hydroxyl group in these
compounds serves as a convenient point for the attachment of
the pendant radical initiator. This was accomplished by treat-
[a] Dr. K. Pati, A. M. Hughes, H. Phan, Prof. I. V. Alabugin
Department of Chemistry and Biochemistry, Florida State University
Chemistry Sciences Building, Tallahassee, Fl 32306 (USA)
E-mail: alabugin@chem.fsu.edu
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201304092.
Figure 1. Skipped diynes are readily available.
Chem. Eur. J. 2014, 20, 390 – 393
390
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
whereas the over-reduced products were observed with
2 equiv of Bu3SnH. Overall, Bu3SnH/AIBN (1.5 equiv/0.3 equiv)
in toluene (1108C, 14 h) was found to be the optimal choice,
providing 2a in 80% yield (entry 4).
In order to evaluate the generality of this approach to fused
naphthalenes, we tested the reactions of additional diynes
1b–1k under the optimized conditions (Table 2). The cascade
is fully compatible with acceptor (ester, cyano and trifluoro-
methyl) and donor (methoxy, methyl, methylenedioxy) substi-
tution at either one of the two arylalkynyl termini. In the reac-
tion of a b-naphthyl-substituted bis-alkyne (1b), capable of
Scheme 2. Synthesis of 1,(2-alkynylphenyl)propargyl iodoethyl ethers. Reac-
tion yields are given for R=R1 =Ph (see the Supporting Information for addi-
tional details).
Table 2. Scope of substrates for radical cascades.
ment with ethyl bromoacetate to produce the propargyl
esters[9] in 60–70% overall yields. Reduction of the esters pro-
duces alcohols converted into the requisite iodo precursors
1 for the radical cascade by the Appel reaction.
Entry
Substrate
Product
Yield [%]
Table 1 shows screening of reagents and initiators for the
key transformation of 1,(2-alkynylphenyl)propargyl iodoethyl
ether (1a) into dihydro-1H-benzo[1,2]fluoreno[3,4-b]furan (2a).
1
85
Table 1. Screening against different reagents, initiators and solvents.
2
3
70
58
Entry
Reagent/initiator[a]
Conditions
Yield[b] [%]
1
2
3
4
5
6
7
8
Bu3SnH/AIBN
Et3SiH/AIBN
Bu3SnH/AIBN
Bu3SnH/AIBN
Et3SiH/AIBN
Ph3SnH/AIBN
Bu3SnH/ABCN
Bu3SnH/DTBPB
benzene, 16 h, 808C
benzene, 16 h, 808C
MeCN, 15 h, 808C
toluene, 14 h, 1108C
toluene, 14 h, 1108C
toluene, 14 h, 1108C
toluene, 14 h, 1108C
toluene, 14 h, 1108C
65
42
45
80
70
[c]
–
65
[c]
–
4
5
72
65
[a] Reagent: 1.5 equiv; initiator: 0.3 equiv. [b] Product yields are reported
after purification from silica gel column. [c] Mixture of products.
ABCN=azobiscyclohexanenitrile; DTBPB=di-tert-butyl peroxide.
a
Although Bu3SnH/AIBN and Et3SiH/AIBN in refluxing benzene
led to complete consumption of starting iodide 1a, the first
two entries gave only moderate yields (65 and 42%) of the de-
sired 2a. CH3CN as solvent decreases the yield of 2a to 45%,
but switch to toluene with the concomitant increase in the re-
action temperature had the opposite effect (78 and 70%, re-
spectively, for Bu3SnH and Et3SiH, entries 3 and 4). The use
Ph3SnH/AIBN gave a mixture of products in less than 10–15%
(entry 6). Entries 7 and 8 describe the effect of different initia-
tors. Whereas Bu3SnH/ABCN (1.5/0.3 equiv) showed good selec-
tivity with 65% yield, the combination of Bu3SnH/DTBPB gives
a complicated mixture of products. Use of 1 equiv of Bu3SnH
reagent led to incomplete consumption of starting iodide 1a
6
7
70
72
Chem. Eur. J. 2014, 20, 390 – 393
www.chemeurj.org
391
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
The two possible mechanisms for this cascade transforma-
tion are shown in Scheme 3. Both paths start with the same
chemoselective formation of an alkyl radical A from the inter-
molecular attack of the Bu3Sn radical at the CÀI bond of the re-
Table 2. (Continued)
Entry
Substrate
Product
Yield [%]
68
8
9
60
70
80
10
11
[a] Isolated yields reported after silica column chromatography.
giving a mixture of products in the final step of the cascade
(vide infra), formation of a single regioisomer 2b was observed
in 85% yield. These results highlight the broad scope and se-
lectivity of this method.
Scheme 3. The two mechanistic alternatives for the proposed cascade.
The structures of the cascade products were confirmed with
the combination of HMBC and HSQC NMR spectroscopy tech-
niques and, in the case of compound 2b and 2h, with X-ray
crystallography[10] (Figure 2). The flat conjugated core of the
molecules (dihedral angles between the naphthyl and fused
actant 1a. The chemoselectivity is noteworthy since Bu3Sn rad-
ical is well known to attack similarly substituted alkynes.[11] The
intermediate A undergoes 5-exo-dig cyclization to form the
first cycle and the first vinyl radical intermediate B. The latter
undergoes 6-exo-dig cyclization at the remaining triple bond
with the formation of a second vinyl radical C. The two possi-
ble mechanisms diverge at this point.
In the first scenario, the vinyl radical attacks the p system of
the neighboring phenyl group. In the second scenario, the
same radical abstracts a hydrogen from an aromatic CÀH
bond. Although the Csp2ÀH bonds are relatively strong, such
path would avoid transient loss of aromaticity. The “translocat-
ed” aryl radical F produced via such H abstraction can “come
back” at the alkene moiety via a 5-endo-trig cyclization.[12] Alter-
natively, the first path produces delocalized radical D capable
of rearomatization via a fast 1,5-hydrogen shift. Such shifts
were shown in our previous work to have low activation barri-
ers.[13] At this point, the two mechanisms converge to give the
penultimate intermediate G, which provides the final dihydro-
1H-benzo[1,2]fluoreno[3,4-b]furan product (2a) after the final
H-abstraction step.[14] The difference between the two mecha-
nisms is important in those nonsymmetrically substituted al-
kynes when the final ring formation can provide a mixture of
products. Formation of a single regioisomer in the cyclization
of naphthyl-substituted substrate 1b provides a key insight
that allows us to differentiate between the two pathways. The
Figure 2. The ORTEP for 2b and 2h. Probability level 50%.
aryl systems in 2b and 2h are <68) facilitates conjugation be-
tween the alkoxysubstituted naphthalene and the annealed ar-
omatic rings. Significant changes observed in the UV spectra
as a function of substitution (see the Supporting Information)
suggest that electronic communication between the fused
rings is efficient.
Chem. Eur. J. 2014, 20, 390 – 393
www.chemeurj.org
392
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
96, 207; d) A. Gansꢁuer, H. Bluhm, Chem. Rev. 2000, 100, 2771; e) M. P.
Sibi, S. Manyem, J. Zimmerman, Chem. Rev. 2003, 103, 3263; f) U. Wille,
Chem. Rev. 2013, 113, 813.
[3] a) A. L. J. Beckwith, Tetrahedron 1981, 37, 3073; b) A. L. J. Beckwith, C. H.
Schiesser, Tetrahedron 1985, 41, 3925; general discussion of selectivity
of alkyne cyclizations: c) K. Gilmore, I. V. Alabugin, Chem. Rev. 2011, 111,
6513; I. V. Alabugin, K. Gilmore, M. Manoharan, J. Am. Chem. Soc. 2011,
133, 12608; 5-exo/6-endo competition in conjugated systems: d) I. V.
Alabugin, M. Manoharan, J. Am. Chem. Soc. 2005, 127, 12583.
[4] Reviews: Fragments of Fullerenes and Carbon Nanotubes: Designed Syn-
thesis Unusual Reactions, and Coordination Chemistry (Eds.: M. A. Petru-
khina, L. T. Scott), Wiley, Hoboken, 2012; E. T. Chernick, R. R. Tykwinski,
J. Phys. Org. Chem. 2013, 26, 742. Selected examples: M. B. Goldfinger,
K. B. Crawford, T. M. Swager, J. Am. Chem. Soc. 1997, 119, 4578; X. Feng,
W. Pisula, K. Mꢂllen, Pure Appl. Chem. 2009, 81, 2203; L. T. Scott, E. A.
Jackson, Q. Zhang, B. D. Steinberg, M. Bancu, B. Li, J. Am. Chem. Soc.
2012, 134, 107; L. Luo, D. Resch, C. Wilhelm, C. N. Young, G. P. Halada,
R. J. Gambino, C. P. Grey, N. S. Goroff, J. Am. Chem. Soc. 2011, 133,
19274.
[5] Intermolecular initiation: a) I. V. Alabugin, K. Gilmore, S. Patil, M. Mano-
haran, S. V. Kovalenko, R. J. Clark, I. Ghiviriga, J. Am. Chem. Soc. 2008,
130, 11535; b) Intramolecular initiation: P. M. Byers, I. V. Alabugin, J. Am.
Chem. Soc. 2012, 134, 9609.
[6] Au-catalyzed endo-dig cyclizations: P. M. Byers, J. I. Rashid, R. K. Mo-
hamed, I. V. Alabugin, Org. Lett. 2012, 14, 6032.
[7] D. P. Curran, in Comprehensive Organic Synthesis, Vol. 4 (Eds.: B. M. Trost,
I. Fleming), Pergamon, Oxford, 1992, p.715; b) S. Kim, K. S. Yoon, Y. S.
Kim, Tetrahedron 1997, 53, 73.
Scheme 4. Regioselectivity of radical attack agrees with the radical addition
to the terminal aromatic system in the final cyclization step.
observed isomer corresponds to the more favorable direction
of radical attack at the naphthalene moiety (confirmed by the
X-ray diffraction study; Scheme 4). We had shown earlier that
a radical attack at the a position is more favorable than a b at-
tack and provided a theoretical rationale to this selectivity.[15]
Absence of product formation through b attack clearly shows
that the product is formed via a path that involves the inter-
mediate D (Scheme 3).
In summary, we have developed a new approach to substi-
tuted benzo[1,2]fluoreno[3,4-b]furan derivatives from skipped
enediynes through intramolecular radical cascade cyclization
which involves formation of three new cycles via sequence of
5-exo-dig, 6-exo-dig ring closures and attack at the aromatic
ring. Subsequent aromatization furnishes the fused naphtha-
lene products in high yields. This radical cascade opens a new
avenue for the preparation of polycyclic frameworks. Future
work includes expansion of this radical cascade toward longer
and wider polycyclics.
[8] Z. Chen, M. Zeng, Q. Yang, Y. Peng, Org. Lett. 2012, 14, 3588.
[9] Y. Chen, M. Chen, Y. Liu, Angew. Chem. 2012, 124, 6599; Angew. Chem.
Int. Ed. 2012, 51, 6493.
[10] CCDC 954366 (2b) and 958993 (2h), contain the supplementary crystal-
lographic 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.
[11] a) S. Peabody, B. Breiner, S. V. Kovalenko, S. Patil, I. V. Alabugin, Org.
Biomol. Chem. 2005, 3, 218; b) S. V. Kovalenko, S. Peabody, M. Manohar-
an, R. J. Clark, I. V. Alabugin, Org. Lett. 2004, 6, 2457.
[12] This sequence is similar to “radical translocations” suggested by Curran:
a) D. P. Curran, K. Dooseop, L. Hong Tao, S. Wang, J. Am. Chem. Soc.
1988, 110, 5900, for an expansion of these ideas, see: b) T. Zeidan, S. V.
Kovalenko, M. Manoharan, R. J. Clark, I. Ghiviriga, I. V. Alabugin, J. Am.
Chem. Soc. 2005, 127, 4270; I. V. Alabugin, B. Gold, J. Org. Chem. 2013,
78, 7777; for the mechanistic analysis of 5-endo-trig cyclizations, see:
c) C. Chatgilialoglu, C. Ferreri, M. Guerra, V. Timokhin, G. Froudakis, T. Gi-
misis, J. Am. Chem. Soc. 2002, 124, 10765; d) K. Gilmore, M. Manoharan,
J. Wu, P. v. R. Schleyer, I. V. Alabugin, J. Am. Chem. Soc. 2012, 134, 10584.
[13] I. V. Alabugin, M. Manoharan, B. Breiner, F. Lewis, J. Am. Chem. Soc.
2003, 125, 9329.
[14] The most recent CCSD(T*)-F12 calculations data of Santos and co-work-
ers suggest that BDE for this CÀH bond should be less than 30 kcal
molÀ1. R. C. Santosa, F. Agapitoa, E. M. GonÅalvesa, J. A. Martinho Simꢃe-
sa, R. M. Borges dos Santos, J. Chem. Thermodyn. 2013, 61, 83. See
ref. [15] for the B3LYP data.
Acknowledgements
The support of the National Science Foundation (CHE-
1213578).
Keywords: 6-exo-dig · cyclization · naphthalene · polycyclic
aromatics · radical cascade
[1] a) K. C. Nicolaou, D. J. Edmonds, P. G. Bulger, Angew. Chem. 2006, 118,
7292; Angew. Chem. Int. Ed. 2006, 45, 7134; b) D. P. Curran, N. A. Porter,
B. Giese, Stereochemistry of Radical Reactions: Concepts, Guidelines and
Synthetic Applications, Wiley-VCH, Weinheim, 1996; c) H. Miyabe, Y.
Takemoto, Chem. Eur. J. 2007, 13, 7280; d) Radicals in Organic Synthesis
Vols. 1 and 2 (Eds.: P. Renaud, M. P. Sibi), Wiley-VCH, Weinheim, 2001;
e) A. J. McCarroll, J. C. Walton, Angew. Chem. 2001, 113, 2282; Angew.
Chem. Int. Ed. 2001, 40, 2224; f) E. Godineau, Y. Landais, Chem. Eur. J.
2009, 15, 3044; g) G. J. Rowlands, Tetrahedron 2009, 65, 8603; h) G. J.
Rowlands, Tetrahedron 2010, 66, 1593.
[15] a) A. Baroudi, J. Alicia, P. Flack, J. Kirincich, I. V. Alabugin, J. Org. Chem.
2011, 76, 1521; b) A. Baroudi, P. Flack, I. V. Alabugin, Chem. Eur. J. 2010,
16, 12316; c) A. Baroudi, J. Alicia, I. V. Alabugin, Chem. Eur. J. 2010, 16,
7683.
Received: October 22, 2013
[2] a) D. P. Curran, Synthesis 1988, 417, 489; b) C. P. Jasperse, D. P. Curran,
T. L. Fevig, Chem. Rev. 1991, 91, 1237; c) K. K. Wang, Chem. Rev. 1996,
Published online on November 22, 2013
Chem. Eur. J. 2014, 20, 390 – 393
www.chemeurj.org
393
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim