Probing the arenium-ion (ProtonTransfer) versus the cation-radical (Electron Transfer) mechanism of scholl reaction using DDQ as oxidant

Article abstract of DOI:10.1021/jo100611k

(Figure Presented) DDQ/H⁺ system readily oxidizes a variety of electron donors with oxidation potential as high as ～1.7 V to the corresponding cation radicals. A re-examination of the controversial arenium-ion versus cation-radical mechanisms for Scholl reaction using DDQ/H⁺ together with commonly utilized FeCl₃ as oxidants led us to demonstrate that the reaction proceeds largely via a cation-radical mechanism. The critical experimental evidence in support of a cation-radical pathway for the Scholl reaction includes the following: (i) There is no reaction in Scholl precursors in a mixture of dichloromethane and various acids (10% v/v). (ii) The necessity to use powerful oxidants such as ferric chloride (FeCl₃) or DDQ/H⁺ for Scholl reactions is inconsistent with the arenium-ion mechanism in light of the fact that aromatization of the dihydro intermediates (formed via arenium-ion mechanism) can be easily accomplished with rather weak oxidants such as iodine or air. (iii) Various Scholl precursors with oxidation potentials 1.7 V vs SCE undergo ready oxidative C-C bond formation with DDQ/H⁺ as oxidant, whereas Scholl precursors with oxidation potentials greater than >1.7 V vs SCE do not react. (iv) Finally, the feasibility of the dicationic intermediate, formed by loss of two electrons, has been demonstrated by its generation from a tetraphenylene derivative using DDQ/H⁺ as an oxidant.

Full text of DOI:10.1021/jo100611k

pubs.acs.org/joc
Probing the Arenium-Ion (ProtonTransfer) versus the Cation-Radical
(Electron Transfer) Mechanism of Scholl Reaction Using DDQ as Oxidant
Linyi Zhai, Ruchi Shukla, Shriya H. Wadumethrige, and Rajendra Rathore*
Department of Chemistry, Marquette University, P.O. Box 1881, Milwaukee, Wisconsin 53201-1881
rajendra.rathore@marquette.edu
Received March 30, 2010
DDQ/Hþ system readily oxidizes a variety of electron donors with oxidation potential as high as ∼1.7 V
to the corresponding cation radicals. A re-examination of the controversial arenium-ion versus cation-
radical mechanisms for Scholl reaction using DDQ/Hþ together with commonly utilized FeCl3 as
oxidants led us to demonstrate that the reaction proceeds largely via a cation-radical mechanism. The
critical experimental evidence in support of a cation-radical pathway for the Scholl reaction includes the
following: (i) There is no reaction in Scholl precursors in a mixture of dichloromethane and various acids
(10% v/v). (ii) The necessity to use powerful oxidants such as ferric chloride (FeCl3) or DDQ/Hþ for
Scholl reactions is inconsistent with the arenium-ion mechanism in light of the fact that aromatization of
the dihydro intermediates (formed via arenium-ion mechanism) can be easily accomplished with rather
weak oxidants such as iodine or air. (iii) Various Scholl precursors with oxidation potentials e1.7 V vs
SCE undergo ready oxidative C-C bond formation with DDQ/Hþ as oxidant, whereas Scholl precursors
with oxidation potentials greater than >1.7 V vs SCE do not react. (iv) Finally, the feasibility of the
dicationic intermediate, formed by loss of two electrons, has been demonstrated by its generation from a
tetraphenylene derivative using DDQ/Hþ as an oxidant.
Introduction
as intramolecular oxidative C-C bond formation between
the two benzenoid rings to produce a biaryl linkage, i.e.,
eq 1.
The Scholl reaction1 is one of the oldest C-C bond-
forming reactions and has been extensively utilized for
intramolecular (oxidative) cyclodehydrogenation of a variety
of o-terphenyls and hexaarylbenzenes to produce the corres-
ponding planar polycyclic aromatic hydrocarbons (PAHs)
such as triphenylenes, hexa-peri-hexabenzocoronenes (HBCs),
etc.2 In its simplest form, the Scholl reaction can be represented
(1) (a) Scholl, R.; Mansfeld, J. Ber. Dtsch. Chem. Ges. 1910, 43, 1734.
(b) Kovacic, P.; Jones, M. B. Chem. Rev. 1987, 87, 357.
The Scholl reaction in eq 1 can be accomplished by using a
variety of oxidants such as FeCl3,3 CuCl2 or Cu(OTf)2 and
€
(2) (a) Berresheim, A. J.; Muller, M.; Mullen, K. Chem. Rev. 1999, 99,
1747. (b) Watson, M. D.; Fechtenkotter, A.; Mullen, K. Chem. Rev. 2001,
€
AlCl3,4 Tl(O2CCF3)3 in CF3CO2H or BF3 OEt2,5 Pb(OAc)4/
€
3
101, 1267. (c) Clar, E. Polycyclic Hydrocarbons; Academic Press: New York,
1964. (d) Harvey, R. G. Polycyclic Aromatic Hydrocarbons; Wiley-VCH:
New York, 1996.
BF3 Et2O in MeCN,6 triethyloxonium hexachloroantimo-
3
nate (Et3OþSbCl6-),7 SbCl5,8 MoCl5,9 etc. Moreover, the
4748 J. Org. Chem. 2010, 75, 4748–4760
Published on Web 06/24/2010
DOI: 10.1021/jo100611k
r
2010 American Chemical Society

Zhai et al.
JOCArticle
SCHEME 1. Cation Radical (or Electron Transfer) Mechanism
for the Scholl Reaction
SCHEME 2. Arenium Ion (or Proton Transfer) Mechanism for
the Scholl Reaction
reaction in eq 1 can also be effected by electrochemical oxida-
tion.10 The involvement of cation radical intermediates (for-
med by 1-e- oxidation) in biaryl synthesis has been carefully
probed by Parker and co-workers.11 A detailed mechanism of
oxidative biaryl synthesis involving cation radical intermedi-
ates is presented in Scheme 1.
The arenium-ion mechanism, depicted in Scheme 2, de-
mands that C-C bond formation be accomplished only if the
protonation occurs at a least preferred site, i.e., meta to an
electron-releasing phenyl substituent (i.e., red arrow in
Scheme 2) and finally producing a nonaromatic cyclohexa-
diene derivative.12,13 Furthermore, the mechanism of aro-
matization of the cyclohexadiene derivative is overlooked in
these studies12,13 and supposedly is rapid and proceeds via an
ECEC mechanism (vide infra).10,11 It is further noted that no
experimental evidence is available from these studies for the
time-dependent acceleration of the Scholl reaction or the
acid-catalyzed formation of the nonaromatic cyclohexadiene
derivatives.12,13
Reactions involving proton transfers differ from electron
transfer reactions in that the reactive intermediates are
diamagnetic cations as opposed to paramagnetic cation
radicals, and such a traditional (ionic versus radical) dicho-
tomy continues to chronicle these mechanistic processes in
separate, discrete mechanistic categories. Interestingly, how-
ever, the most favored method for generating paramagnetic
cation radicals from electron-rich aromatic donors for ESR
spectroscopic studies14 involves their exposure to strong
acids such as sulfuric or trifluoroacetic acids, where the
conjugate acid (DHþ), formed by protonation of an aro-
matic donor (D), acts as an oxidant.15,16 It is important to
A recent series of papers by King and co-workers12,13 have
advocated that Scholl reaction occurs via diamagnetic are-
nium ion intermediates (see Scheme 2) owing to the presence
of adventitious acids in various oxidizing systems3-9 utilized
for Scholl reaction. It is further suggested by King and co-
workers12 that these reactions are accelerated as they proceed
because of the formation of 2 equiv of acid per C-C bond
formation. Interestingly, however, no evidence has been forth-
coming for such a time-dependent acceleration which, if true,
can be taken advantage of by added acids (see Scheme 2).
(3) (a) Sarhan, A. A. O.; Bolm, C. Chem. Soc. Rev. 2009, 38, 2730.
(b) Boden, N.; Bushby, R. J.; Headdock, G.; Lozman, O. R.; Wood, A. Liq.
Cryst. 2001, 28, 139. (c) Boden, N.; Bushby, R. J.; Cammidge, A. N.;
Duckworth, S.; Headdock, G. J. Mater. Chem. 1997, 7, 601. (d) Percec, V.;
Wang, J. H.; Okita, S. J. Polym. Sci., Polym. Chem. 1991, 29, 1789.
(4) (a) Simpson, C. D.; Mattersteig, G.; Martin, K.; Gherghel, L.; Bauer,
€
R. E.; Rader, H. J.; Muellen, K. J. Am. Chem. Soc. 2004, 126, 3139.
€
€
(b) Kubel, C.; Eckhardt, K.; Enkelmann, V.; Wegner, G.; Mullen, K.
J. Mater. Chem. 2000, 10, 879.
(5) McKillop, A.; Turrell, A. G.; Young, D. W.; Taylor, E. C. J. Am.
Chem. Soc. 1980, 102, 6504.
(6) Aylward, J. B. J. Chem. Soc. B 1967, 1268.
(7) (a) Rathore, R.; Kumar, A. S.; Lindeman, S. V.; Kochi, J. K. J. Org.
Chem. 1998, 63, 5847. (b) Mori, T.; Shinkuma, J.; Sato, M.; Saito, H.; Wada,
T.; Inoue, Y. Enantiomer 2002, 7, 115.
(8) (a) Yang, J.-S.; Swager, T. M. J. Am. Chem. Soc. 1998, 120, 5321.
(b) Yamaguchi, S.; Swager, T. M. J. Am. Chem. Soc. 2001, 123, 12087. (c) Rose,
A.; Tovar, J. D.; Yamaguchi, S.; Nesterov, E. E.; Zhu, Z.; Swager, T. M. Philos.
Trans. R. Soc. London, Ser. A 2007, 365, 1589 and references therein.
€
(9) (a) Kramer, B.; Frohlich, R.; Waldvogel, S. R. Eur. J. Org. Chem.
2003, 3549. (b) Waldvogel, S. R.; Aits, E.; Holst, C.; Frohlich, R. Chem.
Commun. 2002, 1278. (c) Kovacic, P.; Lange, R. M. J. Org. Chem. 1963, 28,
968. (d) Waldvogel, S. R. Synlett 2002, 622.
€
(14) (a) Kaiser, E. T.; Kevan, L. Radical Ions; Wiley: New York, 1968.
(b) Forbes, W. F.; Sullivan, P. D. Can. J. Chem. 1966, 44, 1501. (c) Kiseles, S.;
Konovalov, S. Russ. Chem. Rev. 1989, 58, 230. (d) Fife, T. H. Acc. Chem. Res.
1972, 8, 264. (e) Sankararaman, S.; Lau, W.; Kochi, J. K. J. Chem. Soc., Chem.
Commun. 1991, 396. (f ) Eberson, L.; Radner, F. J. Chem. Soc., Chem. Commun.
1991, 1233. (g) Davies, A. G. Chem. Soc. Rev. 1996, 25, 299. (h) Valenzuela,
J. A.; Bard, A. J. J. Phys. Chem. 1968, 72, 286. (i) Gassman, P. G.; Singleton,
D. A. J. Am. Chem. Soc. 1984, 106, 7993. ( j) Yueh, W.; Bauld, N. L. J. Chem.
Soc., Perkin Trans. 2 1995, 871. (k) Ciminale, F.; Lopez, L.; Mele, G. Tetra-
hedron 1994, 50, 12687.
(10) (a) Ronlan, A.; Hammerich, O; Parker, V. D. J. Am. Chem. Soc.
1973, 95, 7132. (b) Ronlan, A.; Parker, V. D. J. Org. Chem. 1974, 39, 1014.
(c) Rathore, R.; Kochi, J. K. J. Org. Chem. 1995, 60, 7479.
(11) Hammerich, O.; Parker, V. D. Adv. Phys. Org. Chem. 1984, 20, 55
and references therein.
(12) Rempala, P.;Kroulik, J.;King, B. T. J. Am. Chem. Soc. 2004, 126, 15002.
(13) (a) Rempala, P.; Kroulik, J.; King, B. T. J. Org. Chem. 2006, 71,
5067. (b) King, B. T.; Kroulik, J.; Robertson, C. R.; Rempala, P.; Hilton,
C. L.; Korinek, J. D.; Gortari, L. M. J. Org. Chem. 2007, 72, 2279. (c) Also
see: Dou, X.; Yang, X. Y.; Bodwell, G. J.; Wagner, M.; Enkelmann, V.;
Mullen, K. Org. Lett. 2007, 9, 2485.
(15) (a) Rathore, R.; Zhu, C.-J.; Lindeman, S. V.; Kochi, J. K. J. Chem.
Soc., Perkin Trans. 2 2000, 1837. (b) Hart, H.; Teuerstein, A.; Babin, M. A.
J. Am. Chem. Soc. 1981, 103, 903.
(16) Rathore, R.; Kochi, J. K. Acta Chem. Scand. 1998, 52, 114.
J. Org. Chem. Vol. 75, No. 14, 2010 4749

Zhai et al.
JOCArticle
SCHEME 3. Two Pathways for Radical Cation Coupling To
Form the Dicationic Intermediate
note that, generally, only a very small fraction of the donor
molecules are oxidized to the corresponding cation radical (i.e.,
,1%) even upon extended exposure to the strongly acidic
media.14,17 It is a daunting task to clearly distinguish the
arenium-ion (Hþ transfer) versus cation-radical (e- transfer)
pathways in a given reaction owing to the similar chemical
reactivities of paramagnetic cation radicals and arenium ions.
This distinction is especially difficult in the Scholl reaction
because of the coexistence and ready interchangeability of
these reactive intermediates. However, despite these diffi-
culties, ample experimental evidence has been accumulated
for the cation-radical pathway7,10,11,18,19 for the Scholl reaction
(or biaryl synthesis) as opposed to the arenium-ion mech-
anism.12,13
For example, according to theoretical calculations by
King and co-workers,12 a key contentious step in the
cation-radical pathway for the Scholl reaction (see Schemes 1
and 3) is the intermediacy of the dicationic intermediate
arising from either the coupling of two benzenoid cation
radicals or via a coupling of the cation radical with its neutral
counterpart (i.e., via a dimeric cation radical) followed by
1-e- oxidation, i.e., Scheme 3.
FIGURE 1. Representative X-ray structures of dimeric cation ra-
dicals showing the sandwich-like geometry: (A) [octamethyl-
- 22
- 24
biphenylene]2þ•SbCl6
,
(B) [perylene]2þ•PF6
,
(C) [octamethyl-
- 22
- 25
anthracene]2þ•SbCl6
,
(D) [tetramethoxynaphthalene]2þ•SbF6
,
.
(E) [naphthalene]2þ•SbF6
- 26
of crystalline dimeric cation-radical salts from a variety of
aromatic electron donors (e.g., see Figure 1).22,24-26
It is also noted that formation of the dimeric cation
radicals in Scheme 3 is exothermic by ∼200-300 mV (or
4-7 kcal/mol), and lowering of the redox potentials owing to
the stabilization of the cationic charge within a stabilized
dimer formation is observed for both intermolecular as well
as intramolecular pairs.27 Second, the feasibility of the
dicationic intermediate, which results in biaryl formation
by loss of two protons (see Scheme 3), has also been demon-
strated by isolation and X-ray crystallographic characteriza-
tion of a robust dicationic intermediate from a tetraphenyl-
ene derivative, i.e., Scheme 4.28
For example, octamethoxytetraphenylene (OMT) under-
˚
goes a rapid single C-C bond formation (C1-C5 = 1.56 A)
between the ipso-carbons C1 and C5 of opposite phenylene
rings upon a 2-electron oxidation either chemically (using
NOþBF4-, triethyloxonium hexachloroantimonate, or a
stable hindered naphthalene cation-radical salt) or electro-
chemically to produce a magenta red (λmax = 610 nm)
First, the existence of dimeric cation radicals via a spon-
taneous association of a cation radical with its neutral
counterpart in a sandwich-like geometry from a variety of
aromatic electron donors has been established not only by
spectroscopic methods by the observation of doubling of the
hyperfine lines in their ESR spectra20-22 and by the presence of
intense intervalence transitions in near-IR region of their electro-
nic spectra22,23 but also by isolation and X-ray crystallography
solution of dicationic OMTþ2 28
The structure of highly
.
robust OMTþ2 was established by X-ray crystallography
(see Scheme 4).28 It is also noted that the neutral OMT can be
regenerated quantitatively by a 2-electron reduction of
OMTþ2 (Scheme 4) using zinc dust.28
(17) The most likely explanation for formation of only a small amount of
cation radical from various aromatic donors in strong acids is that the
production of cation radical is accompanied by the formation of reduced
dihydro-aromatic compound (see Rathore et al. in ref 15a), which in turn is
reoxidized to the corresponding aromatic donor by cation radicals, thereby
maintaining a small concentration of the cation radicals suitable for EPR
spectroscopic measurments.
(18) Wadumethrige, S. H.; Rathore, R. Org. Lett. 2008, 10, 5139.
(19) Rathore, R.; Wygand, U.; Kochi, J. K. J. Org. Chem. 1996, 61, 5426.
(20) Lewis, L. C.; Singer, L. S. Chem. Phys. 1965, 43, 2712. The hyperfine
splittings in the naphthalene dimeric radical cation are as follows: 2.77 (8H)
and 1.03 (8H), whereas they are 5.86 (4H) and 1.66 (4H) in the naphthalene
monomeric radical cation. (Erickson, R.; Benetis, N. P.; Lund, A.; Lindgren,
M. J. Phys. Chem. 1997, 101, 2390).
(21) Howarth, O. W.; Fraenkel, G. K. J. Am. Chem. Soc. 1966, 88, 4514.
(22) Kochi, J. K.; Rathore, R.; Le Magueres, P. J. Org. Chem. 2000, 65, 6826.
(23) (a) Badger, B.; Brocklehurst, B. Trans. Faraday Soc. 1969, 65, 2582–
2588. (b) Badger, B.; Brocklehurst, B. Trans. Faraday Soc. 1970, 66, 2939.
(c) Badger, B.; Brocklehurst, B. Nature 1968, 219, 263. (d) Rodgers, M. A. J.
J. Chem. Soc., Faraday Trans. 1 1972, 68, 1278.
(24) Endres, H.; Keller, H. J.; Muller, B.; Schweitzer, D. Acta Crystal-
logr., Sect. C: Cryst. Struct. Commun. 1985, 41, 607.
(25) Olovsson, G.; Olovsson, I. Acta Crystallogr., Sect. B: Struct. Sci.
1991, 47, 355.
(26) (a) Fritz, H. P.; Gebauer, H.; Friedrich, P.; Ecker, P.; Artes, R.;
Schubert, U. Z. Naturforsch., B: Chem. Sci. 1978, 33, 498. (b) Le Magueres,
P.; Lindeman, S. V.; Kochi, J. K. Org. Lett. 2000, 2, 3567.
(27) (a) Rathore, R.; Chebny, V. J.; Kopatz, E. J.; Guzei, I. A. Angew.
Chem., Int. Ed. 2005, 44, 2771. (b) Rathore, R.; Kochi, J. K. Can. J. Chem.
1999, 121, 913. (c) Rathore, R.; Abdelwahed, S. H.; Guzei, I. A. J. Am. Chem.
Soc. 2003, 125, 8712. (d) Shukla, R.; Lindeman, S. V.; Rathore, R. Org. Lett.
2007, 9, 1291. (e) Rathore, R.; Burns, C. L.; Abdelwahed, S. H. Org. Lett.
2004, 6, 1689. (f) Chebny, V. J.; Shukla, R.; Rathore, R. J. Phys. Chem. A
2006, 110, 13003. (g) Chebny, V. J.; Shukla, R.; Lindeman, S. V.; Rathore, R.
Org. Lett. 2009, 11, 1939. (h) Chebny, V. J.; Navale, T. S.; Shukla, R.;
Lindeman, S. V.; Rathore, R. Org. Lett. 2009, 11, 2253.
(28) Rathore, R.; Le Magueres, P.; Lindeman, S. V.; Kochi, J. K. Angew.
Chem., Int. Ed. 2000, 39, 809.
4750 J. Org. Chem. Vol. 75, No. 14, 2010

Zhai et al.
JOCArticle
We16 and others29 have recently shown that quinones [such
as tetrachloro-p-benzoquinone (chloranil, Ered = -0.02 V vs
SCE) or dichlorodicyano-p-benzoquinone (DDQ, Ered
SCHEME 4. X-ray Crystallographic Evidence for the Existence
of the Dicationic Intermediate Formed by a 2-Electron Oxidationa
=
þ0.60 V vs SCE)] in the presence of an added acid readily
oxidize a variety of aromatic electron donors (such as naphtha-
lene, anthracene, hexaalkylbenzenes, 1,4-dialkoxybenzenes,
biphenyl, etc.), with oxidation potentials as high as ∼1.7 V vs
SCE, to the corresponding cation radicals according to the
stoichiometry shown in eq 2.
It is important to emphasize that in the absence of an
added acid both chloranil and DDQ form vividly colored
electron donor-acceptor (EDA) complexes with various
aromatic donors,30 and the highly endothermic electron
transfer reaction in eq 2 is dramatically accelerated in the
presence of an added acid.16,29 The electron-transfer stoichio-
metry in eq 2 was independently confirmed by a quantitative
isolation of a stable hydroquinone ether cation radical accord-
ing to eq 3.16
aMethoxy groups are omitted from X-ray structures for the sake of
clarity.
CHART 1. Structure and Numbering Scheme for Various
Scholl Precursors Used in This Study
Herein, we will show that the DDQ/acid system (which
readily oxidizes a variety of aromatic donors to the corres-
ponding cation radicals) can be employed for Scholl
reactions (or biaryl synthesis) as demonstrated by the prepa-
ration of a number of triphenylenes and hexa-peri-hexaben-
zocoronenes under mild conditions.31 Moreover, we will
show that a number of highly reactive (electron-rich) Scholl
precursors including hexarylbenzenes (see Chart 1) do not
undergo cyclodehydrogenation or produce nonaromatic
cyclohexadiene derivatives (Scheme 2) when exposed to
strong acids (such as HCl, CF3CO2H, CH3SO3H, HBF4)
or Lewis acids (such as BF3 OEt2 or AlCl3) alone for
3
extended periods (1 to 24 h). Interestingly, however, various
Scholl precursors in Chart 1 do undergo efficient oxidative
cyclodehydrogenation within minutes if the above mixtures
are treated with one equiv. of DDQ per C-C bond forma-
tion. Herein, the details of these findings are discussed in the
context of arenium-ion versus cation-radical mechanisms of
the Scholl reaction (or oxidative biaryl synthesis) by utilizing
both DDQ and commonly utilized ferric chloride (FeCl3) as
oxidants and a variety of Scholl precursors including hexa-
arylbenzenes.
Results and Discussion
Synthesis of various Scholl precursors. The Scholl precur-
sors (Chart 1), employed in this study were synthesized using
standard synthetic procedures.
(29) (a) Handoo, K. L.; Gadru, K. Curr. Sci. 1986, 55, 920–922. and
references therein. (b) Also see: Eberson, L.; Hartshorn, M. P.; Persson, O.
J. Chem. Soc., Perkin Trans. 2 1997, 195. (c) Fukuzumi, S.; Ishikawa, K.;
Hironaka, K.; Toshio, T. J. Chem. Soc., Perkin Trans 2 1987, 751 and
references therein.
For example, the o-terphenyl derivatives (1a-1e and 1k)
were obtained by standard Suzuki reactions32 while the
(30) Rathore, R.; Lindeman, S. V.; Kochi, J. K. J. Am. Chem. Soc. 1997,
119, 9393 and references therein.
(31) Zhai, L.; Shukla, R.; Rathore, R. Org. Lett. 2009, 11, 3474.
(32) (a) Suzuki, A. Chem. Commun. 2005, 4759. (b) Hassan, J.; Sevignon,
M.; Gozzi, C.; Schulz, E.; Lemaire, M. Chem. Rev. 2002, 102, 1359.
J. Org. Chem. Vol. 75, No. 14, 2010 4751

Zhai et al.
JOCArticle
diarylmethane derivatives 1f and 1g were prepared by an
acid-catalyzed condensation of 1,2-dimethoxybenzene with
acetone and fluorenone,33 respectively. The hexaaryl-
benzenes 1l-1n,34 the diarylpropane 1h35 and other precur-
sors (i.e., 1i and 1j)36 were available from published literature
procedures from our laboratory. The synthetic details for
various Scholl precursors in Chart 1 and the spectral data
together with the 1H/13C NMR spectra are presented in the
Supporting Information Section.
Reaction of Various Scholl Precursors with Various Protic/
Lewis Acids. When a 0.01 M solution of o-terphenyl 1a
in a mixture of dichloromethane and methanesulfonic
acid (10% v/v) was stirred under an argon atmosphere at
22 °C, it turned pale yellow. The reaction was monitored
during a course of 24 h by 1H NMR spectroscopy as
follows. Thus, periodically an aliquot was removed and
treated with solid sodium carbonate to neutralize the acid
followed by filtration and removal of the solvent in vacuo.
The 1H NMR spectroscopic and thin-layer chromato-
graphic analysis of the resulting residue showed that 1a
remained completely unchanged during a course of 24 h,
i.e. eq 4.
FIGURE 2. Comparison of the UV-vis absorption spectra of
tetramethoxytriphenylene cation radical 2aþ• generated using
-
DDQ/Hþ (black line) and MAþ•SbCl6 (red line) in dichloro-
methane. The green spectrum was obtained from an aliquot re-
moved after 30 s from the reaction in eq 5.
sodium bicarbonate (20 mL) followed by a standard aqueous
workup, which afforded the corresponding triphenylene 2a in
quanttative yield (i.e., eq 5). Note that the reduced hydro-
quinone (DDQ-H2) dissolves well in the aqueous sodium
bicarbonate layer and can be recovered quantitatively by an
acidification followed by extraction with ether. The identity
of recovered DDQ-H2 was confirmed by comparison of the
13C NMR spectrum and melting point with an authentic
sample.37
Similarly, o-terphenyl 1a did not undergo any reaction
when it was exposed to other protic acids such as CF3CO2H,
HBF4, a saturated solution of gaseous HCl in dichloro-
methane or Lewis acids such as BF3 OEt2 and anhydrous
3
aluminum trichloride (AlCl3) in dichloromethane for up to
24 h at ambient temperatures. Furthermore, all Scholl pre-
cursors (i.e., 1a-1n) in Chart 1 similarly did not undergo any
reaction when treated with a 1:9 mixture of methanesulfonic
acid in dichloromethane according to the eq 4 or with a
saturated solution of HCl in dichloromethane for up to 24 h.
It is also important to emphasize that hexakis(4-isoalkyl-
phenyl)benzene derivatives 1l-1n, precursors to planar
HBCs, remained unchanged even when reacted with either
a protic acid or a Lewis acid for up to 72 h.
Oxidative Cyclodehydrogenation of 1a with Various Protic/
Lewis Acids and DDQ. Interestingly, when a 0.01 M solution
of o-terphenyl 1a in a mixture of dichloromethane and
methanesulfonic acid (10%, v/v) was treated with 1 equiv
of dichlorodicyano-p-benzoquinone (DDQ) under an argon
atmosphere at ∼0 °C, the solution immediately turned dark
green. Upon stirring for 5 min at 0 °C, the reaction mix-
ture took on a dark brown coloration. The reaction was
quenched by the addition of a saturated aqueous solution of
The instantaneous appearance of green coloration in the
reaction in eq 5 was determined to arise owing to the
formation of the cation radical of tetramethoxytriphenylene
2a. For example, when a solution of 2a in a 9:1 mixture of
dichloromethane-methanesulfonic acid was exposed to 0.5
equiv of DDQ, the solution immediately took on the char-
acteristic green coloration. The UV-vis spectral analysis of
the green solution (see Figure 2) together with a spectral
comparison of an authentic spectrum of the 2aþ• [generated
using 9,10-dimethoxy-octahydro-[1,4:5,8]-dimethanoan-
-
thracenium hexachloroantimonate,38 MAþ•SbCl6 (Ered
=
1.14 V vs SCE), as a robust 1-electron oxidant] confirmed the
formation of the tetramethoxytriphenylene cation radical 2aþ•
(λmax = 410, 613, 684, and 762 nm), i.e., Scheme 5.
(33) Apel, S.; Nitsche, S.; Beketov, K.; Seichter, W.; Seidel, J.; Weber, E.
J. Chem. Soc., Perkin Trans. 2 2001, 1212.
(34) Chebny, V. J.; Gwengo, C.; Gardinier, J. R.; Rathore, R. Tetra-
hedron Lett. 2008, 49, 4869.
(35) Shukla, R.; Rathore, R. Synthesis 2008, 3769.
(36) Sun, D.; Lindeman, S. V.; Rathore, R.; Kochi, J. K. J. Chem. Soc.,
Perkin Trans. 2 2001, 1585.
(37) Hageman, L.; McNelis, E. J. Org. Chem. 1975, 40, 3300. Note that
DDQ can be regenerated quantitatively from DDQH2 by oxidation using
dinitrogen tetraoxide (Brook, A. G. J. Chem. Soc. 1952, 5040) or conc nitric
acid (Newman, M. S.; Khanna, V. K. Org. Prep. Proc. Int. 1985, 17, 422).
(38) (a) Rathore, R.; Burns, C. L.; Deselnicu, M. I. Org. Synth. 2005,
82, 1. (b) Rathore, R.; Kochi, J. K. J. Org. Chem. 1995, 60, 4399.
4752 J. Org. Chem. Vol. 75, No. 14, 2010

Zhai et al.
JOCArticle
SCHEME 5. Generation of Tetramethoxytriphenylene Cation Radical (2aþ•) Using DDQ/Hþ and MAþ• in CH2Cl2 at 22 °C
SCHEME 6. Formation of Tetramethoxytriphenylene 2a Using a Combination of DDQ and Various Protic Acids and Lewis Acids in
CH2Cl2 at 22 °C
As summarized in Scheme 6, the oxidative cyclodehy-
drogenation of o-terphenyl 1a can be accomplished using
DDQ as an oxidant in the presence of either a protic acid
(such as methanesulfonic acid, trifluoroacetic acid, or
tetrafluoroboric acid) or a Lewis acid in quantitative yields.
Note that DDQ or acid alone is not sufficient to carry
out the transformation in Scheme 6. It is further empha-
sized that a combination of both DDQ and an acid is
necessary to accomplish the transformation in Scheme 6
(also see eq 5).
aqueous workup afforded pure HBC 2l in a quantitative
yield (eq 6).
Oxidative Cyclodehydrogenation of Various Scholl Precur-
sors Using CH3SO3H/DDQ. Using the protocol developed
in eq 5 (or Scheme 6), a variety of Scholl precursors (Chart 1)
were subjected to the reaction with DDQ in the presence of
readily available and cheap methanesulfonic acid, under the
standard conditions described in eq 5, to afford the corres-
ponding cyclodehydrogenated products in excellent yields
(see Table 1). It is noteworthy that the Scholl reactions using
DDQ/acid oxidation system were equally effective with
substrates undergoing both intramolecular and intermole-
cular aryl-aryl C-C bond formations (see entries 1f and 1i
in Table 1).
The versatility of the DDQ/acid system for the oxida-
tive cyclodehydrogenation is further demonstrated by the
preparation of soluble hexa-peri-hexabenzocoronenes
(HBCs) from hexakis(4-isoalkylphenyl)benzenes 1l-1n,
where 6 new C-C bonds are formed in one step, in excellent
yields. Thus, a reaction of a 0.01 M solution of hexaar-
ylbenzene 1l in a mixture of dichloromethane and metha-
nesulfonic acid (10%, v/v) was treated with 6 equiv of
DDQ under an argon atmosphere at ∼0 °C. The resulting
dark-colored mixture was stirred for 2 h, and a standard
The quantitative transformations of 1l-1n to the corres-
ponding HBC derivatives 2l-2n require the formation of six
C-C bonds and thus 6 equiv of DDQ is necessary for the
complete conversion (eq 6). However, when the oxidation of
a hexaphenylbenzene derivative (such as 1m) was carried out
with substoichiometric amounts of DDQ (e.g., 2 equiv), it
afforded 33% of HBC 2m and 66% unreacted 1m with a
quantitative mass balance, as determined by 1H NMR
spectroscopy (see Figure 3) as well as by the separation of
1m and 2m by chromatography. As such, this finding simply
confirms that partially cyclized (planarized) intermediates
of hexaarylbenzenes have relatively lower oxidation poten-
tials as compared to the corresponding hexaarylbenzenes
and therefore are far more reactive toward additional C-C
bond formation as compared to the corresponding hexaaryl-
benzenes.39
Arenium Ion versus Cation Radical Mechanism of the
Scholl Reaction. In order to facilitate the discussion of the
arenium-ion versus cation-radical mechanism of Scholl
J. Org. Chem. Vol. 75, No. 14, 2010 4753

Zhai et al.
JOCArticle
TABLE 1. Scholl Reaction and Products Obtained under the Standard
Set of Conditions Using DDQ/MeSO3H in Dichloromethane at ∼0 °C
1
FIGURE 3. (A) H NMR spectrum of hexakis(4-isobutylphenyl)-
benzene (1m) in CDCl3. (B) 1H NMR spectrum of the crude product
obtained after a reaction of 1m with 6 equiv of DDQ in CH2Cl2 in
the presence of CH3SO3H showing the quantitative formation of the
corresponding HBC 2m. (C) 1H NMR spectrum of the crude
product obtained after a reaction of 1m with 2 equiv of DDQ in
CH2Cl2 in the presence of CH3SO3H showing that only one-third of
1m is transformed to corresponding HBC 2m.
TABLE 2. Electrochemical Oxidation Potentials of Various Scholl
Precursors (SP) and the Corresponding Cyclodehydrogenated Products
(CD)a
Scholl
precursor
Eox(SP)
V
cyclized
donor
Eox(CD)
V
Eox(SP)
- Eox(CD)
vs SCEb
vs SCEc
1a
1b
1c
1d
1e
1f
1g
1h
1i
1j
1k
1l
1m
1n
1.20
1.11
1.20
1.68
1.56
1.26
1.27
1.32
1.35
1.27
1.33
∼1.6
∼1.6
∼1.6
2a
2b
2c
2d
2e
2f
2g
2h
2i
2j
2k
2l
2m
2n
1.15
1.08
0.89
1.18
1.28
0.85
0.86
1.11
1.24
1.11
1.41
1.00
1.00
1.00
0.05
0.03
0.31
0.50
0.28
0.41
0.41
0.21
0.11
0.16
-0.08
0.48
0.48
0.48
aExperimental conditions: a 1.0 mM solution of the substrate and
0.1 M nBu4NPF6 (as the supporting electrolyte) in dichloromethane at
a scan rate of 200 mV s-1 and at 22 °C. bIrreversible cyclic voltammo-
grams, Eox(SP) were determined from square-wave voltammograms.
cReversible cyclic voltammograms.
and therefore the oxidation potentials of 1a-1n were
determined by square-wave voltammetry40 and are listed
in Table 2. In contrast, the cyclized polyaromatic hydro-
carbons 2a-2n showed reversible cyclic voltammograms
(CV) at scan rates between 25 and 400 mV s-1 with anodic/
cathodic peak current ratios of ia/ic = 1 (theoretical) at
ambient temperatures. The calibration of the CV peaks
of 2a-2n with ferrocene as an internal standard provided
the reversible oxidation potentials that are compiled in
Table 2.
Before examining the viability of the arenium-ion verses
cation-radical mechanism for the Scholl reaction, it is im-
portant to add the caveat that diamagnetic arenium ions and
paramagnetic cation radicals can coexist under the experi-
mental conditions employed for Scholl reactions.15,16,9 With
this fact in mind, let us first consider the arenium-ion path-
way for the Scholl reaction in light of the experimental facts
described above. The arenium-ion pathway requires the
protonation of a benzenoid ring of the Scholl precursor
(SP) at a carbon meta to an electron-releasing group (see
aPercent conversions were determined by 1H NMR, and the yields
refer to isolated products. bApproximately 30-40% of an uniden-
tified compound was also formed. cNMR of crude reaction mixture
showed that it contained 2:1 ratio of 2d and 1d. d1.5 equiv of DDQ was
used.
reaction, the redox potentials of various Scholl precursors
(1a-1n) and the corresponding cyclodehydrogenated poly-
aromatic hydrocarbons (2a-2n) were determined by cyclic
voltammetry. Electrochemical oxidations were performed
at a platinum electrode as 1ꢀ10-3 M solutions in dichloro-
methane containing 0.1 M tetra-n-butylammonium hexa-
fluorophosphate as the supporting electrolyte. Various
Scholl precursors showed irreversible cyclic voltammograms,
(39) The calculated structures (DFT at B3LYP/6-31G* level) of various
polyaromatic intermediates and their energies during the transformation of a
representative hexaarylbenzene to the corresponding HBC clearly show that
introduction of and increasing number of C-C bonds in a hexaarylbenzene
leads to planarized polycyclic aromatic hydrocarbons with increasing aro-
maticity and thus decreased redox potentials and in turn leads to their
increased reactivity towards the Scholl reaction.
(40) Osteryoung, J. G.; Osteryoung, R. A. Anal. Chem. 1985, 57, 101A.
4754 J. Org. Chem. Vol. 75, No. 14, 2010

Zhai et al.
JOCArticle
SCHEME 7. Arenium Ion Mechanism for the Scholl Reaction
SCHEME 8. Cation Radical Mechanism for the Scholl Reaction
then undergoes an intramolecular C-C bond formation to
yield a distonic radical cation (CR2).42 It is the distonic
nature of this rearranged radical cation that facilitates the
removal of a second electron at a much lower potential43 and
leads to the formation of a dicationic intermediate (CR3)
which upon the loss of two protons completes the Scholl syn-
thesis, i.e., Scheme 8.
The following are the important considerations based on
the experimental facts in support of the cation-radical me-
chanism in Scheme 8 for the Scholl reaction:
(1) The first evidence for the cation-radical mechanism for
the Scholl reaction in Scheme 8 is the simple fact that most
commonly employed oxidants including FeCl3 for Scholl
synthesis are powerful enough oxidants to effect the 1-electron
oxidation of various Scholl precursors including various
hexaarylbenzene derivatives to the corresponding cation
radicals.44 The DDQ/acid system described above as another
efficient oxidant for the Scholl reaction is known to effect the
1-electron oxidation of a variety of electron donors, with
oxidation potentials as high as ∼1.7 V vs SCE, to the
corresponding cation radicals.16,29 Furthermore, various
Scholl precursors in Table 1 as well as most of the hexaaryl-
benzenes excluding parent hexaphenylbenzene have oxida-
tion potentials lower than 1.7 V vs SCE (see Table 2).
In this context it is important to note that highly electron-
poor hexakis(4-bromophenyl)benzene45 (Eox >2.1 V vs SCE)
fails to undergo oxidative cyclodehydrogenation even upon
prolonged exposure (1 week) to a large excess of FeCl3.46
As such this observation simply suggests that the inability of
FeCl3 to oxidize electron donors to their cation radicals with
oxidation potentials beyond ∼2 V prevents it from carrying
out the oxidative cyclodehydrogenation of electron-poor
hexakis(4-bromophenyl)benzene (Scheme 9). In order to
further corroborate the thesis that 1-electron oxidation is a
Scheme 2) to form an arenium ion intermediate (AI1), which
then undergoes an intramolecular Friedel-Crafts-type C-C
bond formation (AI2) followed by a rapid proton loss to
produce a nonaromatic cyclohexadiene derivative (AI3),12,13
i.e., Scheme 7. The role of various oxidants [such as FeCl3,3
CuCl2 or Cu(OTf)2 and AlCl3,4 Tl(O2CCF3)3 in CF3CO2H
or BF3 OEt2,5 Pb(OAc)4/BF3-Et2O in MeCN,6 triethylox-
3
9
onium hexchloroantimonate (Et3OþSbCl6-),7 MoCl5 ] in
Scholl reactions, according to the arenium-ion mechanism,
is limited to aromatization of the dihydroaromatic inter-
mediate AI3 to complete the Scholl synthesis.
Interestingly, the necessity to use powerful oxidants such
as ferric chloride (FeCl3), the most commonly utilized oxi-
dant for the Scholl reactions, would be unnecessary owing to
the fact that aromatization of the dihydro intermediates of
type AI3 can be easily accomplished with rather weak
oxidants such as iodine (I2) or oxygen, e.g., eqs 7 and 8.41
(42) The distonic cation radicals (Kiminkinen, L. K. M.; Stirk, K. G.;
Kenttamaa, H. I. J. Am. Chem. Soc. 1992, 114, 2027 and Bauld, N. L.; Gao,
D. J. Chem. Soc., Perkin 2 2000, 931) have ample literature precedent in
the syntheses of a variety of biaryls, see: Hammerich, O.; Parker, V. D.
in ref 11.
(43) Cyclohexadienyl-type radicals are known to undergo oxidation at
∼0 V (vs SCE); see: Haddon, R. C.; Wudl, F.; Kaplan, M. L.; Marshall, J. H.;
Cais, R. E.; Bramwell, F. B. J. Am. Chem. Soc. 1978, 100, 7629.
(44) FeCl3 is known to affect oxidation of variety of electron donors to
their cation radicals; see: Connelly, N. G.; Geiger, W. E. Chem. Rev. 1996,
96, 877 and references therein.
The cation-radical pathway for the Scholl reaction re-
quires an initial 1-electron oxidation of the Scholl precursor
to its cation radical (i.e., Scheme 8). The cation radical (CR1)
(41) (a) Tang, X.-Q.; Harvey, R. G. J. Org. Chem. 1995, 60, 3568.
(b) Blackburn, E. V.; Timmons, T. J. Quart. Rev. Chem. Soc. 1969, 23,
482. (c) Liu, L.; Yang, B.; Katz, T. J.; Poindexter, M. K. J. Org. Chem. 1991,
56, 3769. (d) Laarhoven, W. H. In Organic Photochemistry; Padwa, A., Ed;
Marcel Dekker: New York, 1989; Vol. 10, p 163. (e) Floyd, A. J.; Dyke, S. F.;
Ward, S. E. Chem. Rev. 1976, 76, 509. (f) Sudhakar, A.; Katz, T. J.; Yang, B. W.
J. Am. Chem. Soc. 1986, 108, 2790.
(45) Rathore, R.; Burns, C. L. Org. Synth. 2005, 82, 30.
(46) (a) Wu, J.; Fechtenktter, A.; Gauss, J.; Watson, M. D.; Kastler, M.;
Fechtenktter, C.; Wagner, M.; Muellen, K. J. Am. Chem. Soc. 2004, 126,
11311. (b) Wu, J.; Watson, M. D.; Zhang, L.; Wang, Z.; Muellen, K. J. Am.
Chem. Soc. 2004, 126, 177 and references therein.
J. Org. Chem. Vol. 75, No. 14, 2010 4755

Zhai et al.
JOCArticle
SCHEME 9. Effect of Redox Potentials on the Oxidative Cyclodehydrogenation of Various Hexaarylbenzenes
SCHEME 10. Formation of Tetramethoxytriphenylene 2a Using Varying Equivalents of FeCl3 as Oxidant in CH2Cl2 at ∼0 °C
necessary step in Scholl reaction, we carried out the following
tell-tale experiment.
(2) In many reported procedures,48-50 it is suggested that
yields of Scholl reactions can be dramatically improved if a
large excess of FeCl3 (3-6 equiv per C-C bond formation) is
employed, despite the fact that only 2 equiv of FeCl3 is
needed per C-C bond formation. For example, a 0.01 M
solution of o-terphenyl 1a in dichloromethane at ∼0 °C was
treated with varying equivalents of FeCl3 to immediately
produce a characteristic dark-green solution (vide infra).
After 30 min, the reactions were quenched with excess
dry methanol, and the products were isolated by a stan-
dard aqueous workup. The results are summarized in
Scheme 10.
The incomplete transformation of 1a to 2a with 2 equiv
of FeCl3 in Scheme 10 can be reconciled by the fact that
the cyclized product 2a undergoes an efficient oxida-
tion to its green-colored cation radical (2aþ•) according to
the stoichiometry in eq 9. The identity of 2aþ• in eq 9 was
confirmed by UV-vis spectral comparison with the spec-
trum of authentic 2aþ• generated using either DDQ/acid
system or CRETþ•SbCl6- as an oxidant (see Figure 2, vide
supra).
As reiterated above, the DDQ/Hþ system fails to oxidize
donors to their cation radicals with oxidation potentials
beyond ∼1.7 V vs SCE. Furthermore, parent hexaph-
enylbenzene51 with Eox >1.8 V vs SCE is known to undergo
oxidative cyclodehydrogenation using FeCl3 as an oxidant.12
Expectedly, however, a reaction of hexaphenylbenzene (1q)
with DDQ/Hþ system for 72 h did not afford the correspond-
ing parent HBC (2q) but allowed a quantitative recovery of
both hexaphenylbenzene and DDQ (Scheme 9). Furthermore,
only a slightly electron-rich derivative of hexaphenyl-
benzene 1o47 undergoes a quantitative conversion to the
corresponding HBC 2o when reacted with DDQ/Hþ
(Scheme 9) or FeCl3.34
The experimental facts presented in Scheme 9 are not
readily reconciled on the basis of the arenium-ion mechanism
presented in Scheme 7 but are completely consistent with the
cation-radical pathway depicted in Scheme 8.
(47) Hexakis(4-tert-butylphenyl)benzene (1o) can also be prepared by
Co2(CO)8-catalyzed trimerization of the bis-4,4(di-tert-butylphenyl)-
acetylene48 or by FeCl3-catalyzed tert-butylation of hexaphenylbenzene.49
(48) Herwig, P. T.; Enkelmann, V.; Schmelz, O.; Muellen, K. Chem.;
Eur. J. 2000, 6, 1834.
(49) Rathore, R.; Burns, C. L. J. Org. Chem. 2003, 68, 4071–4074.
(50) (a) Zhou, Y.; Liu, W.-J.; Zhang, W.; Cao, X.-Y.; Zhou, Q.-F.; Ma,
Y.; Pei, J. J. Org. Chem. 2006, 71, 6822–6828. and references therein. (b) Liu,
W.-J.; Zhou, Y.; Zhou, Q.-F.; Ma, Y.; Pei, J. Org. Lett. 2008, 10, 2123.
(c) Zhou, Y.; Liu, W.-J.; Zhang, W.; Cao, X.-Y.; Zhou, Q.-F.; Ma, Y.; Pei, J.
J. Org. Chem. 2006, 71, 6822.
(51) (a) Bell, F. A.; Ledwith, A.; Sherrington, D. C. J. Chem. Soc. C 1969,
13, 2719. (b) Bell, F. A.; Ledwith, A.; Sherrington, D. C. J. Org. Chem. 1976,
13, 155. (c) We also thank one of the reviewer for the suggestion to utilize magic
blue as an oxidant to provide additional credence for the cation-radical mechanism
for Scholl reaction.
In addition, FeCl3 also forms a complex with hydro-
chloric acid, generated during the course of C-C bond
4756 J. Org. Chem. Vol. 75, No. 14, 2010

Zhai et al.
JOCArticle
formation, and thereby renders it to be a less efficient
oxidant, i.e., eq 10.
according to the stoichiometry in eq 12 (compare eqs 2 and 3)
and Figure 4.
FeCl3 þ HCl
HFeCl4
ð10Þ
CH2Cl2
The cation-radical mechanism demands that Scholl pre-
cursor 1a undergo oxidation to its cation radical. Unfortu-
nately, the oxidation of 1a to 1aþ• in Scheme 10 is hampered
-
owing to the lowered oxidizing abilities of both Hþ FeCl4
and 2aþ• FeCl4- (both of which are formed during the course
of the reaction in Scheme 10) in comparison to the FeCl3.
Note that if 1a was allowed to react with 2 equiv of FeCl3 for
an extended period (8 h) with a slow flow of argon to remove
gaseous hydrochloric acid, it affords a nearly quantitative
yield of 2a (see Scheme 10, last entry). Interestingly, this
observation serves to contradict the conjecture by King and
co-workers12 that Scholl reactions are accelerated as they
proceed because of the formation of 2 equiv of acid per C-C
bond formation.
(3) It is apparent from the examination of Tables 1 and 2
that the Scholl precursors with oxidation potentials as high
as ∼1.7 V (entry 1d in Table 2) can be successfully subjected
to the oxidative cyclodehydrogenation using the DDQ/
acid system (eqs 5 and 6 and Scheme 6). Although, most
Scholl precursors (see Table 1) produced nearly quantita-
tive yields of the cyclodehydrogenated products in the
presence of 1 equiv DDQ (per C-C bond formation),
interestingly, however, 1,2-bis(3-methoxyphenyl)benzene
(1d) afforded only ∼66% of the cyclized 2,7-dimethoxy-
triphenylene (2d) together with ∼33% of unreacted 1d.
However, if precursor 1d was subjected to the same reac-
tion conditions with 1.5 equiv of DDQ, it afforded a
quantitative yield of methoxytriphenylene 2d after aqu-
eous workup, i.e., eq 11.
As summarized in Scheme 11, the formation of 2d from 1d in
critical yield of 66% in eq 11 is readily reconciled by the
fact that 1 equiv of DDQ is required to convert 1 equiv of 1d
to 2d, whereas the oxidation of relatively electron-rich 2d
to its cation radical, under same conditions, requires only
0.5 equiv of DDQ (i.e., eq 12). A relatively low oxidation
potential of 2d (Eox = 1.18 V vs SCE) as compared to 1d
(Eox ∼ 1.68 V vs SCE) ensures that it undergoes relatively fast
oxidation to its cation radical as compared to 1d. Further-
more, a highly endothermic (ΔEox ∼ þ500 mV) electron
transfer from 1d (Eox ∼ 1.68 V vs SCE) to 2dþ• (Eox=1.18 V
vs SCE) prevents the remaining 1d (i.e., 0.33 equiv) from
undergoing the Scholl cyclodehydrogenation (Scheme 11).
Also, note that, owing to the high oxidation potential of 1d
(Eox ∼ 1.68 V vs SCE), its oxidation to cation radical by 2dþ•
SCHEME 11. Rationalization of the Observation of 2:1 Stoi-
chiometry in Equations 11 and 12
Moreover, it was independently confirmed that dimethox-
ytriphenylene 2d (Eox =1.18 V vs SCE) undergoes a ready
oxidation to its cation radical 2dþ• (λmax = 620, 687, and
760 nm) in the presence of DDQ and methanesulfonic acid
FIGURE 4. Comparison of the UV-vis absorption spectra of
dimethoxytriphenylene cation radical 2dþ• generated using DDQ/
Hþ (red line) and NOþSbCl6- (blue line) in dichloromethane.
J. Org. Chem. Vol. 75, No. 14, 2010 4757

Zhai et al.
JOCArticle
(Eox = 1.18 V vs SCE) is further hampered by the presence of
DDQ-H2, which effectively competes with 1d for electron
transfer equilibrium with 2dþ•
Importantly, the rationale in Scheme 11 further disfavors
the arenium-ion pathway as the predominant mechanism for
the Scholl reaction. It is re-emphasized that 2dþ• (Ered =1.18
V vs SCE) could have easily served as an effective oxidant for
the aromatization of the dihydro-aromatic intermediate
(AI3) in an arenium-ion pathway and thus would have led
to a quantitative conversion of 1d to 2d in eq 11.
hexacloroantimonate NOþ (Ered = 1.46 V vs SCE),52 under
the reaction conditions depicted in eq 13, afforded a quanti-
tative yield of 2b. The successful usage of clean 1-e- oxi-
dants (Chart 1) for the Scholl reaction in eq 13 further
provides the credence to the cation-radical pathway for
Scholl reaction.
(5) The feasibility of a dicationic intermediate (CR3) in the
cation-radical mechanism (see Scheme 8) for the Scholl
reaction has been questioned by King and co-workers.12
Herein, we provide conclusive evidence, using DDQ/Hþ as
oxidant, that dicationic intermediate CR3 is a viable inter-
mediate in the cation-radical pathway of the Scholl reaction
as follows. A dichloromethane solution of ocatmethoxy-
tetraphenylene (OMT), which forms a stable dicationic salt
upon 2-electron oxidation akin to the dicationic intermediate
CR3 (see Scheme 8),28 was treated with 1 equiv of DDQ in
the presence of methanesulfonic acid to afford a character-
istic magenta-red solution of OMTþ2 in quantitative yield
(see Figure 5) according to the stoichiometry depicted in
Scheme 12.
(4) Although we have earlier demonstrated16,29 that DDQ
in the presence of either a protic acid or a Lewis acid can
affect the oxidation of a variety of aromatic donors to the
corresponding cation radicals quantitatively (e.g., see eq 3),
the need of an acid in the usage of DDQ as an oxidant in the
Scholl reaction remains a source of concern as to the validity
of the electron-transfer mechanism of Scholl reaction. Ac-
cordingly, herein we utilize three well-known 1-e- oxidants
(see Chart 2) for the oxidative cyclodehydrogenation of an
electron-rich precursor 1b as follows.
CHART 2. Structures of Various Well-Known 1-e- Oxidants
for the Oxidative Cyclodehydrogenation of Electron-Rich
Precursor 1b
FIGURE 5. Comparison of the UV-vis absorption spectra of
dicationic OMTþ2 obtained by addition of 1 equiv of OMT to a
3 mL solution of 2.5 ꢀ 10-5 M DDQ in 9:1 mixture of dichlor-
Thus, a 0.01 M solution of 1b in dichloromethane was
treated with 2 equiv of MAþ• (Ered = 1.11 V vs SCE),38 under
an argon atmosphere at ∼0 °C, which immediately afforded
a characteristic green-colored solution identical to that
obtained with DDQ/Hþ. The resulting mixture was stirred
for 5 min and was subjected to standard aqueous workup to
afford 2b together with neutral hydroquinone ether MA in a
quantitative yield (eq 13).
omethane and CH3SO3H (red line) or 5.0 ꢀ 10-5 M NOþSbCl6
-
(black line)28 in dichloromethane.
The minor differences in the absorption spectra of OMTþ2
when generated using DDQ/Hþ [λmax = 510 and 561
(shoulder) nm] and NOþ [λmax = 521 and 576 (shoulder)
nm]28 are attributed to different counteranions, i.e., CH3SOh3
and SbClh6, which differ significantly in their nucleophili-
cities.
It is noted that in above tetraphenylene OMT, the forma-
tion of dicationic intermediate OMTþ2 is possible because of
the fact that that the cyclized cation radical intermediate
(such as CR2) in the case of OMT does not have a proton that
can be lost and therefore undergoes a second oxidation to
produce the dicationic intermediate. However, at this junc-
ture, one cannot easily distinguish whether a Scholl reaction
proceeds via the pathway of a dicationic intermediate (see
Scheme 8).
Summary and Conclusions
In summary, we have demonstrated that the DDQ/Hþ
oxidation system (which is known to oxidize a variety of
electron donors with oxidation potential as high as ∼1.7 V to
the corresponding cation radicals) can be effectively em-
ployed for the preparation of a variety of oxidative cyclode-
hydrogenation reactions. The procedure described herein
with DDQ/Hþ rather than most commonly utilized oxidants
such as FeCl3, MoCl5, or SbCl5 for the Scholl reaction
Similarly, a reaction of Scholl precursor 1b with either
magic blue MBþ• (Ered = 1.15 V vs SCE)51 or nitrosonium
(52) Rathore, R.; Abdelwahed, S. H.; Guzei, I. A. J. Am. Chem. Soc.
2004, 126, 13582 and references therein.
4758 J. Org. Chem. Vol. 75, No. 14, 2010

Zhai et al.
JOCArticle
SCHEME 12. Oxidation of OMT to OMTþ2 Using DDQ/Hþ in CH2Cl2
obviates problems such as chlorination of the polyaromatic
products and the usage of a large excess of oxidants.
hexaarylbenzene derivatives (1l-1n) were synthesized according
to a previously known literature procedure.34 Bichromophoric
2,2-bis(3,4-dimethoxyphenyl)propane (1f ),33 9,9-bis(3,4-dime-
thoxyphenyl)fluorene (1g),33 1,3-bis(3,4-dimethoxyphenyl)pro-
pane (1h),35 1o,47 1p,45 1q,53 and octamethoxytetraphenylene28
were prepared by literature procedures. The details of the synthe-
sis and spectral data for various Scholl precursors, utilized in
this study, are described in Supporting Information.
The arenium-ion versus cation-radical pathways for
the Scholl reaction involves diamagnetic carbocations and
paramagnetic cation radicals as different types of reactive
intermediates. Unfortunately, however, both of these inter-
mediates are known to coexist and can undergo facile inter-
change under the conditions of the Scholl reaction. Herein
we list critical experimental evidence in support of the cation-
radical pathway for the Scholl reaction: (i) The necessity to
use powerful oxidants such as ferric chloride (FeCl3) for
Scholl reactions is inconsistent with the arenium-ion me-
chanism. Note that the role of oxidant in the arenium-ion
mechanism is to aromatize the dihydro intermediates (AI3),
which can be easily accomplished with rather weak oxidants
such as iodine or air. (ii) Various Scholl precursors in Table 1
as well as most hexaarylbenzenes have oxidation potentials
lower than 1.7 V vs SCE and thus can be readily oxidized to
the corresponding cation radicals using either DDQ/Hþ or
FeCl3 as oxidants. Moreover, relatively electron-poor hexakis-
(4-bromophenyl)benzene (Eox>2.0 V vs SCE) has an oxida-
tion potential higher than 1.7 and does not undergo cyclode-
hydrogenation with FeCl3 or DDQ/Hþ. (iii) The observation
that Scholl reaction using FeCl3 canbedriventocompletionby
removal of gaseous hydrochloric acid, by bubbling nitrogen or
argon through the reaction mixture, clearly contradicts the
conjecture of King and co-workers12 that Scholl reactions
are accelerated as they proceed because of the production of
2 equiv of acid per C-C bond formation. (iv) The experi-
mental observation of 66% conversion of an o-terphenyl 1d to
2d in Scheme 11 is readily reconciled by the cation-radical
pathway and helps to disfavor the arenium-ion pathway as the
predominant mechanism for the Scholl reaction.
General Procedure for the Oxidative Cyclodehydrogenation
with DDQ and Methanesulfonic Acid. Under an argon atmo-
sphere, a protic acid (such as CH3SO3H, CF3CO2H, or HBF4;
10% v/v, 1 mL) or a Lewis acid (such as BF3 Et2O or AlCl3;
3
5-10 equiv) was added to a solution of o-terphenyl 1a (0.1
mmol) in dichloromethane (9 mL) at 22 °C. The reaction was
stirred under an argon atmosphere for 30 min to 24 h. Under the
argon flow, a 1-mL aliquot of the reaction mixture was removed
and poured onto solid Na2CO3 to neutralize the added acid. The
solution was then filtered and evaporated in vacuo to afford
the starting o-terphenyl 1a quantitatively, as judged for 1H
NMR spectroscopy and thin-layer chromatography. It is noted
that various Scholl precursors in Chart 1 did not produce any
cyclized products when exposed to protic acids or Lewis acids
for extended periods of time (i.e., ∼24 h) as described above.
This conclusion was confirmed by a quantitative recovery of the
starting materials in each case.
When a solution of o-terphenyl 1a (0.1 mmol) in dichloro-
methane (10 mL) containing protic acid (10% v/v) or Lewis acid
(∼10 equiv) at ∼0 °C was exposed to DDQ (1 equiv per C-C
bond formation, i.e., 0.1 mmol), the solution immediately took
on a dark-green coloration. [Note that the solution colors varied
depending on the Scholl precursors used; see Chart 1.] The
progress of the reaction was monitored by TLC and 1H NMR
spectroscopy. After the completion of the reaction (∼5 min), it
was quenched with a saturated aqueous solution of NaHCO3
(20 mL). The dichloromethane layer was separated, washed
with water and brine solution, dried over anhydrous MgSO4,
and filtered. Removal of the solvent in vacuo afforded the
The experimental evidence presented herein together with
the electrochemical studies by Parker and co-workers11
strongly favor the cation-radical pathway for the Scholl
reaction. However, a caveat is added that the involvement
of a parallel arenium-ion mechanism for the Scholl reaction
(however minor it may be) cannot be ruled out owing to the
coexistence and interchangeability of the arenium ions and
cation radicals under conditions of the Scholl reaction.
1
crude product, which by H/13C NMR analysis was shown to
be 2,3,10,11-tetramethoxytriphenylene (2a). Various Scholl pre-
cursors were subjected to similar reaction conditions, and the
characterization data for the oxidative cyclodehydrogenation
products are summarized in Supporting Information.
General Procedure for the Oxidative Cyclodehydrogenation
Using FeCl3. To a solution of o-terphenyl 1a (0.1 mmol) in 10 mL
of dichloromethane at 0 °C was added FeCl3 (2-6 equiv) under
an argon atmosphere. The resulting mixture was stirred at 0 °C
for 30 min. The reaction was quenched by addition of dry
Experimental Section
The Suzuki coupling32 provided a general route for the pre-
paration of various substituted o-terphenyls (1a-e and 1k). The
(53) Fieser, L. F.; Ramey, C. E.; Boekelheide, V. Organic Synthesis;
Wiley: New York, 1973; Collect. Vol. V, p 604; 1966, 46, 44.
J. Org. Chem. Vol. 75, No. 14, 2010 4759

Zhai et al.
JOCArticle
methanol (5 mL). The reaction mixture was washed with water
(40 mL), and the aqueous layer was extracted with dichloro-
methane (2 ꢀ 10 mL). The combined organic layers were washed
with brine solution, dried over anhydrous MgSO4, and filtered.
Removal of the solvent in vacuo afforded the crude product,
which was further purified through recrystallization or column
chromatography. Various Scholl precursors were subjected to
similar reaction conditions, and the characterization data for
the oxidative cyclodehydrogenation products are summarized
here:
Hz, 2H), 7.26 (d, J = 5 Hz, 2H), 7.38 (t, J = 7.5 Hz, 2H), 7.85 (d,
J = 7.6 Hz, 2H); 13C NMR (CDCl3) δ 56.2, 56.3, 65.9, 102.3,
107.0, 120.0, 124.2, 127.8, 128.0, 134.7, 141.0, 141.7, 148.6,
149.1, 149.3. 2h: white solid, yield 100%; mp 152-153 °C
1
(lit.10b mp 158-159 °C); H NMR (CDCl3) δ 2.11-2.20 (m,
2H), 2.43 (t, J = 7 Hz, 4H), 3.92 (s, 12H), 6.77 (s, 2H), 6.89 (s,
2H); 13C NMR (CDCl3) δ 31.2, 34.0, 56.1, 56.3, 111.7, 112.0,
132.2, 133.1, 147.6, 147.9. 2i: white solid, yield 100%; mp
1
113-114 °C; H NMR (CDCl3) δ 2.02 (s, 6H), 3.83 (s, 6H),
3.91 (s, 6H), 6.65 (s, 2H), 6.77 (s, 2H); 13C NMR (CDCl3) δ 19.5,
56.0, 56.2, 112.9, 113.1, 128.36, 133.5, 146.7, 147.9. 2j: white
2a: yield quantitative; mp 215-216 °C; 1H NMR (CDCl3) δ
4.11 (s, 12H), 7.59 (q, 2H), 7.71 (s, 2H), 7.94 (s, 2H), 8.46 (q, 2H);
13C NMR (CDCl3) δ 56.1, 56.2, 104.1, 104.6, 123.0, 123.5, 123.9,
126.2, 128.9, 148.8, 149.4. 2b: white crystals, yield quantitative;
mp 282-283 °C; 1H NMR (CDCl3) δ 2.51 (s, 6H), 4.09 (s, 6H),
4.11 (s, 6H), 7.68 (s, 2H), 7.88 (s, 2H), 8.14 (s, 2H); 13C NMR
(CDCl3) δ 20.5, 56.1, 104.2, 104.5, 123.4, 123.5, 123.6, 127.2,
135.1, 148.8, 149.0. 2c: white solid, yield 50%; mp 193-194 °C;
1H NMR (CDCl3) δ 2.49 (s, 6H), 3.97 (s, 6H), 4.01 (s, 6H), 6.68
(s, 2H), 7.51 (s, 2H), 8.16 (s, 2H); 13C NMR (CDCl3) δ 20.5, 55.8,
56.0, 97.0, 98.1, 112.9, 124.3, 128.7, 132.3, 136.4, 158.8, 158.9.
Note that the triphenylene 2c undergoes decomposition under
the reaction conditions, and thus the pure 2c was isolated by
column chromatography on silica gel using 5% ethyl acetate in
1
solid, yield 100%; mp 129-130 °C; H NMR (CDCl3) δ 2.29
(s, 6H), 3.75 (s, 6H), 3.81 (s, 6H), 6.79 (s, 2H), 6.83 (s, 2H); 13C
NMR (CDCl3) δ 16.6, 56.2, 56.8, 114.0, 115.1, 125.7, 126.7,
150.9, 151.9. 2k: pale yellow crystals, quantitative yield, mp
303-306 °C; 1H NMR (CDCl3) δ 2.51 (s, 18H), 8.33 (s, 6H); 13
C
NMR (CDCl3) δ 20.5, 123.8, 127.8, 135.6. HBC 2l: yellow/
orange solid, yield qunatitative; mp >300 °C; 1H NMR
(CDCl3) δ 1.73 (d, 36H), 3.58 (m, 6H), 8.9 (s, 12H); 13C NMR
(CDCl3) δ 25.0, 35.5, 120.0, 120.3, 124.2, 130.5, 146.6. HBC 2m:
yellow/orange solid, yield quantitatve; mp >300 °C; 1H NMR
(CDCl3) δ 1.11 (s, 18H), 1.69 (s, 18H), 2.07 (s, 12H), 3.29 (s, 6H),
9.03 (s, 12H); 13C NMR (CDCl3) δ 12.1, 22.4, 31.3, 44.4, 120.1,
120.1, 124.9, 131.1, 146.8. HBC 2n: yellow/orange solid, yield
quantitatve; mp 110-112 °C; 1H NMR (CDCl3) δ 0.84 (t, 18H),
1.06 (d, 18H), 1.33 (m, 60H), 2.38 (m, 6H), 6.59 (s, 12H), 6.67 (d,
12H); 13C NMR (CDCl3) δ 14.3, 23.0, 28.3, 29.8, 32.0, 32.1, 39.2,
41.3, 120.7, 120.8, 124.9, 130.6, 146.3. HBC 2o: yellow/orange
1
hexanes as eluent. 2d: yield 99%, mp 208-211 °C; H NMR
(CDCl3) δ 2.52 (s, 6H), 4.02 (s, 6H), 7.21 (dd, J1 = 8.91 Hz, J2 =
2.58 Hz, 2H), 7.99 (d, J = 2.58 Hz, 2H), 8.28 (s, 2H), 8.44 (d, J =
8.91 Hz, 2H); 13C NMR (CDCl3) δ 20.5, 55.7, 105.7, 115.5,
123.9, 124.2, 124.5, 128.2, 130.3, 136.4, 158.3. 2e: yield 60%; mp
214-215 °C; 1H NMR (CDCl3) δ 2.49 (s, 6H), 4.02 (s, 6H), 7.25
(dd, J1 = 8.83 Hz, J2 = 2.56 Hz, 2H), 7.94 (d, J = 2.56 Hz, 2H),
8.26 (s, 2H), 8.53 (d, J = 8.83 Hz, 2H); 13C NMR (CDCl3) δ
20.5, 55.7, 106.3, 115.7, 123.5, 124.4, 125.0, 127.2, 130.8, 135.4,
1
solid, yield quantitatve; mp >320 °C (lit.34 mp >300 °C) ; H
NMR (CDCl3) δ 1.83 (s, 54H), 9.32 (s, 12H); 13C NMR (CDCl3)
δ 32.3, 36.0, 119.1, 120.7, 124.2, 130.7, 149.2.
1
Acknowledgment. We thank the National Science Foun-
dation (Chemistry NSF) for financial support.
158.6. 2f: white solid, yield 100%; mp 178-180 °C; H NMR
(CDCl3) δ 1.65 (s, 6H), 3.77 (s, 6H), 3.86 (s, 6H), 6.69 (d, 2H, J =
2 Hz), 6.76-6.84 (m, 4H); 13C NMR (CDCl3) δ 27.4, 46.7, 56.2,
56.3, 102.5, 106.1, 131.8, 146.1, 148.2, 148.7. 2g: white solid,
yield 100%; mp 242-244 °C; 1H NMR (CDCl3) δ 3.61 (s, 6H),
4.03 (s, 6H), 6.18 (s, 2H), 6.75 (d, J = 7.6 Hz, 2H), 7.14 (t, J = 7
Supporting Information Available: Synthetic details and
1H and 13C NMR data of various compounds. This material is
available free of charge via the Internet at http://pubs.acs.org.
4760 J. Org. Chem. Vol. 75, No. 14, 2010