Reduction of strained polycycles: How much strain can a pyrene anion take?

Article abstract of DOI:10.1021/ja040039u

The reduction of a series of [n](2,7)pyrenophanes (n = 7-10) with lithium or potassium metal shows that the strain in the system, controlled by the length of the tether, determines the nature of the reduction products. The reduction of [7](2,7)pyrenophane

Full text of DOI:10.1021/ja040039u

Published on Web 05/08/2004
Reduction of Strained Polycycles: How Much Strain Can a
Pyrene Anion Take?
Ivan Aprahamian,† Graham J. Bodwell,‡ Jim J. Fleming,‡ Greg P. Manning,‡
Michael R. Mannion,‡ Brad L. Merner,‡ Tuvia Sheradsky,† Rudolf J. Vermeij,‡ and
Mordecai Rabinovitz*,†
Contribution from the Department of Organic Chemistry, The Hebrew UniVersity of Jerusalem,
Jerusalem 91904, Israel, and Department of Chemistry, Memorial UniVersity of Newfoundland,
St. John's, NL, Canada A1B 3X7
Received February 3, 2004; E-mail: mordecai@vms.huji.ac.il
Abstract: The reduction of a series of [n](2,7)pyrenophanes (n ) 7-10) with lithium or potassium metal
shows that the strain in the system, controlled by the length of the tether, determines the nature of the
reduction products. The reduction of [7](2,7)pyrenophane (2) and [2]metacyclo[2](2,7)pyrenophane (3) leads
to reductive dimerization followed by novel intramolecular σ-bond formation as a means of escaping strained
anti-aromaticity. [8](2,7)Pyrenophane (4) affords only reductive dimerization, and no two-electron reduction
is observed. The reduction of [9](2,7) pyrenophane (5) and [10](2,7)pyrenophane (6) leads to reductive
dimerization, followed by the formation of a dianionic anti-aromatic species, which eventually cleaves the
solvent, THF-d8. The similarity between the reduction of the latter systems and the reduction of pyrene (1)
is discussed.
synthesized and studied in order to address this issue.6 The
degree of distortion from planarity and therefore the strain in
Introduction
The discovery of fullerenes1 and the subsequent isolation of
these systems can be controlled by the length and type of the
tether that connects the two remote ends (positions 2 and 7) of
1; the shorter the tether, the greater the strain. It was found that
it is possible to bend the pyrene unit, which is found in the
surfaces of several fullerenes and buckybowls of such systems
to a greater extent than that of the pyrene unit in C70.6b The
studies also showed that enforcing nonplanarity in the pyrene
nucleus does little to diminish the anisotropy effect of the bent
systems. This means that, despite the enforced curvature, the
pyrene system still maintains most of its aromatic character.
2
C60 have rekindled interest in the study of curved polycyclic
aromatic hydrocarbons (PAHs), especially those that can be
mapped onto the surface of C60 or other fullerenes.3 A well-
studied aspect of such curved systems is their aromatic
behavior,4 as such systems can help to clarify the nature of
aromaticity as they manifest a compromise between strain and
conjugation.3d,5
The effect of curvature on the aromatic character of PAHs
can be studied by comparing planar PAHs and their curved
analogues. Recently a variety of [n](2,7)pyrenophanes in which
pyrene (1) is strongly distorted from planarity have been
The aromaticity of 1 can be rationalized by the “peripheral
model”,7 since it has 4n + 2 conjugated π-electrons at the
periphery. This model emphasizes the contribution of peripheral
π-conjugation to the aromatic character of the system and
considers the inner bonds to be a bridging perturbation to the
annulene skeleton. It is possible to convert 1 into an anti-
aromatic entity by adding two electrons to its periphery. This
can be achieved by alkali metal reduction.8
† The Hebrew University of Jerusalem.
‡ Memorial University of Newfoundland.
(1) Kroto, H. W.; Heath, J. R.; O’Brien, S. C.; Curl, R. F.; Smalley, R. E.
Nature 1985, 318, 162.
(2) Kra¨tschmer W.; Lamb L. D.; Fostiropoulos, K.; Huffman, D. R. Nature
1990, 347, 354.
(3) For recent reviews on bowl-shaped, fullerene-related hydrocarbons, see:
(a) Siegel, J. S.; Seiders, T. J. Chem. Ber. 1995, 313. (b) Faust, R. Angew.
Chem., Int. Ed. Engl. 1995, 34, 1429. (c) Rabideau, P. W.; Sygula, A. In
AdVances in Theoretically Interesting Molecules; Thummel, R. P., Ed.; JAI
Press: Greenwich, CT, 1995; Vol. 3, p 1. (d) Rabideau, P. W.; Sygula, A.
Acc. Chem. Res. 1996, 29, 235. (e) Scott, L. T. Pure Appl. Chem. 1996,
68, 291. (f) Mehta, G.; Rao, H. S. P. Tetrahedron Report Number 470.
Tetrahedron 1998, 54, 13325.
(6) (a) Bodwell, G. J.; Bridson, J. N.; Houghton, T. J.; Kennedy, J. W. J.;
Mannion, M. R. Angew. Chem., Int. Ed. 1996, 35, 1320. (b) Bodwell, G.
J.; Bridson, J. N.; Houghton, T. J.; Kennedy, J. W. J.; Mannion, M. R.
Chem.sEur. J. 1999, 5, 1823. (c) Bodwell, G. J.; Fleming, J. J.; Mannion,
M. R.; Miller, D. O. J. Org. Chem. 2000, 65, 5360. (d) Bodwell, G. J.;
Bridson, J. N.; Cyran´ski, M. K.; Kennedy, J. W. J.; Krygowski, T. M.;
Mannion, M. R.; Miller, D. O. J. Org. Chem. 2003, 68, 2089.
(7) (a) Dewar, M. J. S. J. Am. Chem. Soc. 1952, 74, 3341. (b) Platt, J. R. J.
Chem. Phys. 1954, 22, 1448.
(8) (a) Mu¨llen, K. HelV. Chim. Acta 1978, 61, 2307. (b) Minsky, A.; Meyer,
A. Y.; Rabinovitz, M. J. Am. Chem. Soc. 1982, 104, 2475. (c) Schnieders,
C.; Mu¨llen, K.; Huber, W. Tetrahedron 1984, 40, 1701. (d) Eliasson, B.;
Lejon, T.; Edlund, U. J. Chem. Soc., Chem. Commun. 1984, 591.
(4) For example, see: (a) Ferrer, S. M.; Molina, J. M. J. Comput. Chem. 1999,
20, 1412. (b) Steiner, E.; Fowler, P. W.; Jenneskens, L. W. Angew. Chem.,
Int. Ed. 2001, 40, 362. (c) Poater, J.; Frader, X.; Duran, M.; Sola`, M.
Chem.sEur. J. 2003, 9, 1113.
(5) For some recent reviews about aromaticity and its structural aspects, see:
(a) Schleyer, P. v. R.; Jiao, H. Pure Appl. Chem. 1996, 68, 209. (b)
Krygowski, T. M.; Cyran´ski, M. K.; Czarnocki, Z.; Ha¨felinger, G.;
Katritzky, A. R. Tetrahedron Report Number 520. Tetrahedron 2000, 56,
1783. (c) Krygowski, T. M.; Cyran´ski, M. K. Chem. ReV. 2001, 101, 1385.
9
10.1021/ja040039u CCC: $27.50 © 2004 American Chemical Society
J. AM. CHEM. SOC. 2004, 126, 6765-6775
6765

A R T I C L E S
Aprahamian et al.
Scheme 1. Reduction Process of 1 with Lithium Metal
The reduction of PAHs with alkali metals9 yields negatively
charged π-conjugated molecules that may undergo structural
changes from structural distortions,10 rearrangements,11 and
aggregations12 to the formation of new chemical bonds,13 as is
the case in reductive dimerization processes.14 The reduction
of 1 can yield different products, depending on solvent, counter-
ion, and temperature. Three successive diamagnetic species can
be detected when 1 is reduced with lithium metal in THF-d8
(Scheme 1): a protonated pyrene monoanion (1a), a dianion
(1b), and finally a monoanion that incorporates the atoms of a
former solvent molecule (solvent cleavage) (1c).8c Recently we
have shown15 that the reduction of the two strained pyrene sys-
tems, [7](2,7)pyrenophane (2)6c and [2]metacyclo[2](2,7)pyreno-
phane (3),16 with lithium metal affords totally different results
than those encountered in the reduction of the planar parent
pyrene system.17 The reduction of these systems shows that the
strain in the pyrene moiety profoundly affects its reactivity.
study was conducted to understand the effect strain can have
on the reduction process of 1 and assess whether a limit exists
to the degree of curvature 1 can withstand, without under-
going skeletal changes. The investigation was carried out
by reducing a series of [n](2,7)pyrenophanes (n ) 7-10), in
which the degree of distortion from planarity of the pyrene unit
varies with tether length,6c with alkali metals. The characteriza-
tion of the resulting anionic species (structure and aromatic
behavior) was achieved using various NMR spectroscopy
methods.
Results and Discussion
Tether of Seven Carbon Atoms: [7](2,7)Pyrenophane (2)
and [2]metacyclo[2](2,7) pyrenophane (3) that have tethers of
seven carbons have been reduced with lithium15 and potassium
metals (Scheme 2). The reduction process was monitored by
NMR spectroscopy. Both 2 and 3 show similar two-step
reduction processes, which means that the isolated benzene ring
in 3 has a negligible effect on the overall reduction process.
This also indicates that the electron transfer through the
“cyclophane hub”18 is minimal. In both cases the reduction is
accompanied by symmetry changes that are observable in the
19
1H NMR and 13C NMR spectra.
A. One-Electron Reduction.15b The reduction of 2 (colorless)
and 3 (yellow) with lithium or potassium metal (Scheme 2) gives
the dimeric diamagnetic species 2a (wine-red) and 3a (brown)
that show no dynamic processes in their 1H NMR spectra over
a wide range of temperatures (165-273 K). The dimers were
identified by the high field absorption of carbon atom C1 in
species 2a and 3a (δ ) 48.3 and 48.7 ppm, respectively, Table
1) and the magnitude of its 1JC,H coupling (128.2 and 130.0 Hz
for 2a and 3a, respectively, Table 1), which are compatible with
(9) For a review about the alkali metal reduction of PAHs, see: Benshafrut,
R.; Shabtai, E.; Rabinovitz, M.; Scott, L. T. Eur. J. Org. Chem. 2000, 1091.
(10) For example: Eshdat, L.; Ayalon, A.; Beust, R.; Shenhar, R.; Rabinovitz,
M. J. Am. Chem. Soc. 2000, 122, 12637.
(11) For example, see: Eshdat, L.; Berger, H.; Hopf, H.; Rabinovitz, M. J. Am.
Chem. Soc. 2002, 124, 3822.
(12) (a) Ayalon, A.; Sygula, A.; Cheng, P.-C.; Rabinovitz, M.; Rabideau, P.
W.; Scott, L. T. Science 1994, 265, 1065. (b) Shenhar, R.; Wang, H.;
Hoffman, R. E.; Frish, L.; Avram, L.; Willner, I.; Rajca, A.; Rabinovitz,
M. J. Am. Chem. Soc. 2002, 124, 4685.
(13) For example, see: Bock, H.; Havlas, Z.; Gharagozloo-Hubmann, K.; Sievert,
M. Angew. Chem., Int. Ed. 1999, 38, 2240.
(14) (a) Bergmann, E. D. Chem. ReV. 1968, 68, 41. (b) Edlund, U.; Eliasson,
B. J. Chem. Soc., Chem. Commun. 1982, 950. (c) Baumgarten, M.; Mu¨llen,
K. Top. Curr. Chem. 1994, 169, 1. (d) Aprahamian, I.; Hoffman, R. E.;
Sheradsky, T.; Preda, D. V.; Bancu, M.; Scott, L. T.; Rabinovitz, M. Angew.
Chem., Int. Ed. 2002, 10, 1712.
(15) (a) Aprahamian, I.; Bodwell, G. J.; Fleming, J. J.; Manning, G. P.; Mannion,
M. R.; Sheradsky, T.; Vermeij, R. J.; Rabinovitz, M. J. Am. Chem. Soc.
2003, 125, 1720. (b) Aprahamian, I.; Bodwell, G. J.; Fleming, J. J.;
Manning, G. P.; Mannion, M. R.; Sheradsky, T.; Vermeij, R. J.; Rabinovitz
M. Angew. Chem., Int. Ed. 2003, 42, 2547.
(16) Bodwell, G. J.; Miller, D. O.; Aprahamian, I.; Rabinovitz, M.; Sheradsky,
T.; Vermeij, R. J. In preparation.
(17) Pyrene was also reduced, under our conditions, for comparison.
(18) (a) Gerson, F.; Martin, W. B., Jr.; Wydler, C. HelV. Chim. Acta 1976, 59,
1365. (b) Shabtai, E.; Rabinovitz, M.; Ko¨nig, B.; Knieriem, B.; de Meijere,
A. J. Chem. Soc., Perkin Trans. 2 1996, 2589.
Here we report a comprehensive systematic study of the
reduction of a series of pyrene-based strained molecules. The
9
6766 J. AM. CHEM. SOC. VOL. 126, NO. 21, 2004

Reduction of Strained Polycycles: Pyrene Anion Strain
A R T I C L E S
Scheme 2. Reduction Process of 2 and 3 with Lithium or Potassium Metal
Table 1. 13C NMR and 1H NMR (in Parentheses) Chemical Shifts, in ppm, of the Pyrene Moiety of the Lithium Salt of 2a, 2b, 3a, and 3b in
a
THF-d8
C1
C3
C4
C5
C6
C8
C9
C10
(H1)
C2
(H3)
C3a
(H4)
(H5)
C5a
(H6)
C7
(H8)
C8a
(H9)
(H10)
C10a
C10b
C10c
2a 48.3b 99.4 129.6
110.6 131.9
(6.6)
118.5 125.7
(6.2)
113.1 129.8
(6.3)
118.1 125.6
(6.2)
102.7
(5.7)
101.6
(5.2)
102.6
(5.3)
101.6
(5.1)
103.1
142.2 119.3
(5.6)
145.8 107.9
(5.1)
142.4 118.3
(5.2)
145.1 108.1
(4.8)
135.5 114.8
(5.7)
141.6 107.9
(5.1)
131.7 114.1
(5.8)
140.5 108.1
(4.8)
142.7 110.0
(5.9)
145.8 101.6
(5.2)
143.0 109.3
(6.0)
145.1 101.6
(5.1)
129.7
(6.0)
125.7(6.2) 118.5 147.6 136.9
105.2 140.6 135.8
(3.2)
2b 33.9
(2.3)
(5.8)
33.9
(2.3)
29.2
3a 48.7c 95.7 129.0
129.1
(5.9)
125.6
(6.2)
103.5 140.5 135.8
118.1 146.9 136.5
(3.4)
3b 33.7
(2.4)
(5.9)
33.7
(2.4)
30.1
7d
103.1 128.1
144.5 104.3
135.2 104.3
144.7 103.1
128.1
103.1 144.7 137.6
b 1
3
c 1
3
a The 13C NMR chemical shifts of 7 are also added for comparison.
J
) 128.2 Hz, JH1,*H1 ) 10.3 Hz.
J
) 130.0 Hz, JH1,*H1 ) 10.6 Hz.
C1,H1
C1,H1
d Taken from ref 21.
sp3-hybridization. In addition, an indication that a new σ-bond
has been formed between two identical C1 carbons comes from
the heteronuclear multiple-bond correlation (HMBC) experi-
ment, which shows a correlation between carbon atom C1 and
3
hydrogen atom H1. The JH1,*H1 couplings (derived from the
proton-coupled heteronuclear single quantum coherence sensi-
tivity-improved (HSQCSI) experiments)20 were measured to be
10.3 and 10.6 Hz for 2a and 3a (Table 1), respectively, which
is appropriate for dimers with an anti conformation about the
new σ-bond.
The dimerization of 2 and 3 can be explained by means of a
radical coupling at carbon atom C1. Such radical coupling will
leave a negative charge on carbon atom C2 that can be
delocalized over the carbon atoms on the periphery of a
phenalene subunit, which explains why two sets of carbon atoms
are observed in the 13C NMR spectra (one at δ ) 99-119 and
the other at δ ) 129-143 ppm). The good agreement between
the carbon chemical shifts of the phenalene moiety of 2a and
3a and those of the 2-methylphenalenyl anion (7)21 clearly
supports this conclusion (Table 1).
B. Two-Electron Reduction.15a In the second reduction
process, the “intermolecular” σ-bond between the two carbon
C1 atoms of the dimer is cleaved, to afford the monomers 2b
(wine-red) and 3b (brown). 2a and 3b have a new intra-
molecular σ-bond that transforms one of the benzene rings into
a “cyclopropano-cyclopentano” (bicyclo[3.1.0]) ring system.22
The new σ-bond explains the high field absorption of carbon
atom C1 (33.9 and 33.7 ppm for 2a and 3a, respectively, Table
1) and its one-bond CH-coupling (1JC1,H1 ) 163.2 and 162.3
Hz for 2b and 3b, respectively). These values are consistent
with a strained sp3-hybridized carbon, as is the case in a
cyclopropane ring.23 The 3JH1,*H1 coupling constant (9.1 and 8.7
(19) For the sake of clarity, the results of the lithium reduction will be discussed
for all the molecules. Any dissimilarity, should it exist, between the lithium
and potassium reductions will be noted.
(20) (a) Palmer, A. G., III; Cavanagh, J.; Wrights, P. E.; Rance, M. J. Magn.
Reson. 1991, 93, 151. (b) Kay, L. E.; Keifer, P.; Saarinen, T. J. Am. Chem.
Soc. 1992, 114, 10663.
(21) Hempenius, M. A.; Heinen, W.; Mulder, P. P. J.; Erkelens, C.; Zuilhof,
H.; Lugtenburg, J.; Cornelisse, J. J. Phys. Org. Chem. 1994, 7, 296.
(22) A dynamic process is observed in the 1H NMR of 2b and 3b, which in
some cases further lowers the symmetry of the system. For the sake of
clarity, the chemical shifts of the systems with higher symmetry are
discussed.
(23) Gu¨nther, H. NMR Spectroscopy - An Introduction; Wiley: Chichester,
U.K., 1980.
9
J. AM. CHEM. SOC. VOL. 126, NO. 21, 2004 6767

A R T I C L E S
Aprahamian et al.
Scheme 3. Reduction Process of 4 with Lithium or Potassium Metal
Hz for 2b and 3b, respectively) and the correlation seen in the
HMBC experiments between hydrogen atom H1 and carbon
atom C2 are also explained by the formation of this new σ-bond.
In this scenario the high-field shift of carbon atom C2 (29.2
and 30.1 ppm for 2b and 3b, respectively) results from the
localization of the charge on it.
It can be seen in 2b and 3b that the phenalenyl anion is still
intact. This is evident from the high-field shift of carbon atoms
C8, C9, and C10a (and their symmetrically equivalent atoms)
and hydrogen atoms H8 and H9, which constitute a section of
the periphery of phenalene (Table 1).
oxygen quench of 3b gave an irresolvable NMR spectrum.
Attempts to quench the dianion 3b were also made with iodine
and dimethyl sulfate, and in both cases the product readily
underwent polymerization.
Tether of Eight Methylene Carbons: [8](2,7)Pyrenophane
(4)6c was reduced with lithium and potassium metals. The
reduction is accompanied by color change from a colorless to
wine-red solution. Only one reduction product was observed in
both cases (Scheme 3). The NMR data show that 4 also
undergoes a reductive dimerization to produce a dianionic dimer,
4a. Prolonged contact with either lithium or potassium metals
did not reduce the dimer any further. Oxygen quench experi-
ments conducted on 4a afforded neutral 4.
As far as the tether is concerned, the most interesting result
comes from following after the chemical shift of hydrogen atom
H14 in compound 3, and its reduction products, 3a and 3b. In
neutral 3 this phenyl hydrogen absorbs at δ ) 4.10 ppm, as a
result of a strong anisotropy effect.16 In 3a the absorption is at
δ ) 5.10 ppm, indicating the presence of diamagnetic ring
currents in the “phenalenyl anion” (what remains of the pyrene
moiety). To the best of our knowledge, this is the first time
that the anisotropy effect above the core of a phenalenyl anion
has been measured experimentally. The phenalenyl anion is still
intact in 3b, but hydrogen atom H14 absorbs at a lower field
(δ ) 6.83 ppm), characteristic of aromatic hydrogens. This may
be due to one or a combination of two reasons: (1) the
cyclopropyl ring moves the benzene ring away from the core
of the phenalenyl anion, and thus H14 is less affected by the
anisotropy effect; 2) The dynamics of the tether cancels the
effect.
Tether of Nine Methylene Carbons: [9](2,7)Pyrenophane
(5)6c (colorless) was reduced with lithium and potassium metals
to afford a paramagnetic species followed by three different
diamagnetic reduction products (Scheme 4); here as well, the
reduction was observable via the color change of the solution.19
A. One-Electron Reduction. The first reduction product of
5 is probably a radical anion (orange solution), since it does
1
not show a H NMR spectrum. The paramagnetic species is
followed by a dianionic dimer, 5a (wine-red), which is afforded
via radical coupling. It is worth noting that this dimer is less
soluble than the previous ones. Quench of 5a with oxygen gave
the neutral species 5.
B. Two-Electron Reduction. Upon further reduction, the 1H
NMR spectrum of the dimer disappears and a flat spectrum is
obtained. When the 1H NMR spectrum is measured at temper-
atures below 200 K, a new set of broad peaks appears. This
means that the new species is in singlet-triplet equilibrium,
with a higher contribution of the triplet state and, thus, can only
be observed at low temperatures. The 1H and 13C NMR spectra
of the new species 5b (brown-red when reduced with lithium
and blue with potassium) are relatively less complex than those
of the dimer. The broadness of the peaks is a result of both the
singlet-triplet equilibrium and the tether dynamics (to be
discussed). It can be seen that 5b has only two proton
absorptions belonging to the pyrene entity: δ ) 1.44 and 2.58
ppm for hydrogen atoms H1 and H10, respectively.26 It is
evident from the high-field shift of these 1H NMR absorptions
that the second reduction product is an anti-aromatic species
(5b). This explains the presence of the singlet-triplet equilib-
rium in 5b, for it is characteristic of anti-aromatic systems.27
Another strong indication of the anti-aromaticity of the new
species comes from the chemical shift of the tether hydrogens,
1
C. Dynamic NMR. H NMR spectra were recorded for 2b
and 3b at various temperatures in order to follow the dynamic
process of the tether, which affects the axis of symmetry along
it and to estimate its free energy of activation. The ∆Gq values
were calculated from the coalescence temperatures using the
Eyring equation.23
Lowering the temperature in 2b does not result in line splitting
but rather in broadening, which hinders the calculation of ∆Gq.24
In 3b, the signal of hydrogen atom H9, which appears as a
doublet at 260 K (δ ) 5.11 ppm), broadens at lower temper-
atures and is finally split into two broad doublets at 165 K (δ
) 5.85 and 4.82 ppm).25 The ∆Gq of this process is calculated
to be 9.1 ( 0.2 kcal/mol.
D. Quench Experiments. Oxygen-quench experiments con-
ducted on 2a and 3a afforded the neutral species 2 and 3. This
indicates that the carbon framework of the original polycyclic
moiety did not degrade as a result of reduction. In contrast, an
(26) The presence of a dynamic process in the compound lowers its symmetry,
and thus the signals of these two hydrogen atoms are split into two broad
peaks. For the sake of clarity, the average chemical shift of these broad
peaks is used (this should correspond to the species with higher symmetry
that cannot be observed clearly at low temperatures).
(27) Minsky, A.; Meyer, A. Y.; Poupko, R.; Rabinovitz, M. J. Am. Chem. Soc.
1983, 105, 2164.
(24) The broadening is a result of the low temperature and dynamics, so no
accurate line shape analysis is possible. Nevertheless, the ∆Gq of the process
seems to be below 5 kcal/mol (lower energetic limit of NMR spectroscopy).
(25) Unfortunately, even at 165 K hydrogen atom H1 in 2b and 3b appears as
a broad singlet, and no splitting is seen. Such a splitting might have made
the observation of the direct coupling between H1 and H3 possible.
9
6768 J. AM. CHEM. SOC. VOL. 126, NO. 21, 2004

Reduction of Strained Polycycles: Pyrene Anion Strain
A R T I C L E S
Scheme 4. Diamagnetic Reduction Products Afforded from 5
which function as probes for the anisotropy effect prevailing in
the system. The effect is most notable at hydrogen atoms H14
and H14′, where the latter absorbs at δ ) 2.58 ppm and the
former, which feels the paramagnetic currents more strongly,
absorbs at δ ) 7.98 ppm.28 To the best of our knowledge, this
is the first time that tether hydrogens have been used to probe
the anti-aromaticity above the core of a polycyclic system.29 It
should be noted that the effect is more prominent in 5b/2K+
where H14 and H14′ absorb at δ ) 9.93 and 3.22 ppm,
respectively.30
detected via the proton-coupled HSQCSI experiment. Whereas
a 3JH,*H interaction is observed in the dimer, no such interaction
exists for the new product,33 which means that the rehybrid-
ization of carbon atom C1 is not a result of dimerization. Nor
does it occur through protonation, as happens in pyrene (1a,
Scheme 1), because the integration of hydrogen atom H1 shows
the presence of only one hydrogen atom. More importantly the
vicinal hydrogens in the protonation product are not equivalent
and, thus, two doublets should be obtained, and this is not the
case.
In the reduction process of 1, the dianion (1b) cleaves the
solvent (THF-d8) upon heating to yield 1c (Scheme 1).8c This
solvent cleavage affords a monoanionic pyrene moiety, in which
the charge is concentrated on the phenalene subunit, and the
reaction takes place at the same carbon atom as dimerization.
The dynamic process of the tether, which also lowers the
symmetry along it, was monitored using 1H NMR spectroscopy.
The signal of hydrogen atom H10 in 5b that appears as a singlet
at 185 K (δ ) 2.58 ppm) broadens and is finally split into two
broad singlets when the temperature is further lowered (δ )
2.61 and 2.50 ppm at 165 K). The ∆Gq of the process was found
to be 8.6 ( 0.2 kcal/mol,31 which is similar to the ∆Gq of the
neutral system (8.8 ( 0.2 kcal/mol).32 This shows that the
reduction has a minimal effect on the tether dynamics.
This explains why only minor changes between the 1H and 13
C
NMR spectra of two such products are to be expected. Based
on this we assume that the new reduction product, 5c, is that of
solvent cleavage. Because the solvent used was deuterated, the
carbon signals of the cleaved solvent are too weak to be
observed (due to splitting of the carbons by the deuterium
atoms).
Quench experiments were not possible due to the instability
of the species.
C. Third Reduction Product. The third reduction product,
5c (wine-red when reduced with lithium and dark blue with
potassium), can be obtained by heating the sample to room
temperature. The third product has very similar 1H and 13C NMR
spectra to those of the dimer. In fact, except for some minor
changes in the values of the chemical shifts, the spectra are
almost identical. The only difference between the spectra is
Oxygen quench experiments conducted on 5c afforded neutral
5.
Tether of 10 Methylene Carbons: A. Synthesis. The
synthesis of [10](2,7)pyrenophane (6) (Scheme 5) was conducted
according to the synthetic pathway previously described for the
syntheses of [7]-, [8]-, and [9](2,7)pyrenophanes.6c Diestertriflate
86c was subjected to a Sonogashira reaction with 1,9-decadiyne
to afford diynetetraester 9 (68%), which was catalytically
hydrogenated to give tetraester 10 (99%). Reduction of 10 with
LiAlH4 followed by treatment of the crude reaction mixture with
HBr/HOAc led to the formation of tetrabromide 11 (66%).
(28) The difference in absorption is a result of a dissimilarity in the orientation
of the hydrogen atoms H14 and H14′ relative to the pyrene core.
7
(29) The Li NMR absorption of 5b/2Li+, δ ) 1.98 ppm, shows the existence
of solvent-separated ion pairs in the system and the presence of paramag-
netic currents in the anion.
(30) The difference in ∆δ (δH14 - δH14′) between the lithium and potassium
34
reduction products can be attributed to changes in the type of ion pairing
Reaction of 11 with Na2S/Al2O3 provided dithiacyclophane
in the system. Another indication of such a difference comes from the 13
C
12 (78%). This was then converted into [10](2,7)pyrenophane
by a standard series of five reactions without any purification
NMR spectrum where carbon atom C10b, C2, and one of the C1/3 carbons
have a relatively large high-field shift in the lithium salt compared with
the potassium salt. This shift can give an indication as to the location of
the lithium cations relative to the anionic moiety.
(31) Due to the limitation in the available temperature range (THF freezes at
160 K), the ∆Gq value is only an approximation (higher limit), for it is not
possible to measure the ∆ν at the point of minimum or no exchange.
(32) A dynamic process was also observed in all the neutral compounds.
Unpublished results.
(33) This experiment is of utmost importance, in addition to integration, it can
help to distinguish between a number of products with similar NMR
spectra.
(34) Bodwell, G. J.; Houghton, T. J.; Koury, H. E.; Yarlagadda, B. Synlett 1995,
751.
9
J. AM. CHEM. SOC. VOL. 126, NO. 21, 2004 6769

A R T I C L E S
Aprahamian et al.
Scheme 5. Synthesis of 6a
a (a) 1,9-decadiyne, Pd(PPh3)2Cl, CuI, DBU, benzene, rt, 48 h, 68%; (b) H2, Pd/C, EtOAc, rt, 16 h, 99%; (c) LiAlH4, THF, reflux, 16 h, then 30%
HBr/HOAc, reflux, 45 min, 66%; (d) Na2S/Al2O3, 10% EtOH/CH2Cl2, rt, 16 h, 78%; (e) (MeO)2CHBF4, CH2Cl2, rt, 3 h, then t-BuOK, THF, rt, 16 h, then
(MeO)2CHBF4, CH2Cl2, rt, 6 h, then t-BuOK, THF, rt, 6 h, then DDQ, benzene, rt, 22% from 12.
Scheme 6. Diamagnetic Reduction Products Afforded from 6
1
of intermediates. This involved the treatment of 12 with
(MeO)2CHBF4 (Borch reagent)35 and then t-BuOK to bring
about a Stevens rearrangement, followed by Borch reagent and
then t-BuOK to induce Hofmann elimination. This afforded a
ca. 3:1 mixture of pyrenophane 6 and an unidentified dihydro-
pyrene. Finally, reaction of this mixture with DDQ afforded 6
in 20% overall yield from 12.
B. Reduction. The reduction of 6 (colorless) with lithium or
potassium metal affords three diamagnetic reduction compounds
(Scheme 6): (1) A dimer (6a) (wine-red); (2) an anti-aromatic
species (6b) (brown-red when reduced with lithium and blue
with potassium); and in the end, (3) a solvent cleavage product
(6c) (wine-red when reduced with lithium and violet with
potassium).
“humps” are seen in the appropriate region of the H NMR
spectrum.
D. Two-Electron Reduction. As is the case with 5, the two-
electron reduction of 6 yields an anti-aromatic species 6b. The
anti-aromaticity of the system manifests itself most notably at
hydrogen atoms H14 and H14′, which absorb at δ ) 6.95 and
2.38 ppm, respectively.28 Again the effect is more noticeable
in 6b/2K+, where H14 and H14′ absorb at δ ) 8.68 and 3.10
ppm, respectively.30
The ∆Gq of the tether dynamics, which lowers the symmetry
along it, was measured for both 6 and 6b by monitoring the
temperature dependence of the absorption of hydrogen atom
H1 in 6 and hydrogen atom H2 in 6b. It was found that ∆Gq is
8.4 ( 0.2 and 8.9 ( 0.2 kcal/mol for 6 and 6b,31 respectively.
This again shows that the reduction has a minimal effect on
the dynamic process.
C. One-Electron Reduction. The first reduction compound
1
is possibly a radical anion (orange solution with no H NMR
E. Third-Reduction Product. The compound 6b is not stable
and eventually cleaves the solvent to afford 6c. Quenching 6c
with oxygen gives neutral 6.
Systems at Large: The reduction of the series of [n](2,7)-
pyrenophanes (n ) 7-10) shows the processes involved are
not alkali metal-dependent. It is possible to generalize and state
that as far as the single-electron reduction is concerned, the
spectrum) that with time affords the insoluble dimer 6a: The
reduction product with lithium (6a/2Li+) produces a very weak
1
1
and partial H NMR spectrum, assigned according to the H
NMR spectrum of 6c (no 13C NMR is available). When the
compound is reduced with potassium (6a/2K+), only small
(35) Borch, R. F. J. Org. Chem. 1969, 34, 627.
9
6770 J. AM. CHEM. SOC. VOL. 126, NO. 21, 2004

Reduction of Strained Polycycles: Pyrene Anion Strain
A R T I C L E S
degree of strain in the system plays no crucial role. However,
the product of the two-electron reduction is seen to depend on
the strain in the system.
product depends on the length of the tether, i.e., strain.
Compounds 5 and 6, which have relatively long tethers, behave
like 1. Therefore, their two-electron reduction yields anti-
aromatic species, 5b and 6b.
A. Dimerization. The first reduction process, reductive
dimerization, is common to all the [n](2,7)pyrenophanes studied.
To explain why dimerization occurs, one should refer to the
radical anion of 1. It has been shown for 1 that position C1
(and its symmetrical counterparts) has the highest spin density
for the odd electron in the monoanion radical.36 This explains
the high reactivity of this site and the likelihood of it undergoing
radical coupling.37
Why is it then that 1 itself does not dimerize but is rather
protonated? The difference between the two sets of systems is
the alkane tether. This has the dual function of enforcing
nonplanarity on the polycyclic aromatic portion of the systems
and increasing their solubility, which is evident from the
comparatively high solubility of all the neutral pyrenophanes
studied.6c It seems that curvature does not significantly alter
the spin distribution in the radical anion of the pyrene moiety
because dimerization in the [n](2,7)pyrenophanes and protona-
tion in 1 take place at the same carbon atom.
The surprising finding is that the strain in 2 and 3 brings
about the formation of an intramolecular σ-bond as a means of
avoiding an unfavorable strained dianionic anti-aromatic state.
The new σ-bond has a dual function: First, it releases some of
the strain in the pyrene moiety by producing a “cyclopropyl”
anion, which because of its geometrical requirements introduces
a pronounced fold into what was the pyrene moiety, leaving
behind a relatively flat aromatic “phenalenyl” anion.40 Second,
it provides a means of separating the two charges, thus allowing
the system to avoid acquiring anti-aromatic character by
sacrificing the benzene ring.
The threshold is a tether of eight carbon atoms where no two-
electron reduction occurs. It seems that the strain in 4 is large
enough that it does not get into an anti-aromatic state. Moreover,
it seems that the formation of an intramolecular σ-bond is not
favored either, because insufficient stabilization is gained in this
way (the amount of strain is not large enough).
With regard to solubility, the effect of the tether on the
systems is profound.38 In our experiments some precipitate was
seen along with the protonation product of 1. The solubility of
the precipitate increased with temperature, but no diamagnetic
species other than the protonation product was observed in the
NMR spectra. Mu¨llen et al. previously reported that, at low
temperatures, the radical anion of 1 might be in equilibrium
with a diamagnetic dimer, which undergoes cleavage at higher
temperatures.8c In view of this, one may assume that a dimeric
product might also be produced from 1 but is not observed by
NMR spectroscopy due to low solubility.39
It seems that although the tether contributes to the solubility
of the neutral systems, its contribution to the solubility of the
dimers is limited; the solubility of the dimer decreases as a
function of tether length, i.e., as the pyrene system becomes
less distorted.
Whereas 2 and 3 yield two reduction products, 5 and 6 have
an additional step in the reduction process. The solvent cleavage
of 5b and 6b is a result of the nature of the anti-aromatic species.
The solvent cleavage affords a monoanionic pyrene entity, in
which the phenalenyl anion motif can be easily recognized (see
5c and 6c, Schemes 4 and 5). Thus, by attacking the solvent,
the system transforms from an anti-aromatic to an aromatic
species. Again it should be noted that the attack is at carbon
atom C1, as this carbon has the highest density of charge on it
in the dianionic state.8c
The similarity in the behavior of 1, 5, and 6 upon reduction
strengthens the hypothesis that dimerization may indeed occur
in pyrene itself.
Conclusions
The reduction with alkali metals of a series of [n](2,7)-
pyrenophanes (n ) 7-10), in which the strain in the system is
controlled by the length of the tether, shows that the process
can be divided into two steps. The first step, which is common
to all the compounds, is a reductive dimerization process via
electron coupling. The dimers formed in this process contain
an aromatic phenalenyl anion subunit. In the second step, the
reduction product depends on the length of the tether. The
threshold is a tether of eight carbons, where no two-electron
reduction occurs. Molecules with longer tethers behave like
pyrene and produce anti-aromatic anions that can cleave the
solvent at room temperature. Pyrenophanes with shorter tethers
tend to form an intramolecular σ-bond to avoid a state of strained
anti-aromaticity. The new σ-bond has two functions: lowering
the relative strain in the system and separating the aromatic
phenalenyl anion and the second added electron.
The nonplanarity of the pyrene unit, induced by the tether,
imparts an element of asymmetry to each half of the respective
dimers. This leads to the possibility of generating diastereomeric
reduced species. Even if one assumes that the new σ-bond forms
on the less hindered exo faces of the pyrene moieties, two
possible products still remain, a meso diastereomer and a dl
pair. From the data available, only one diastereoisomer appears
to dominate. Unfortunately, it is not possible to fully assign
the relative stereochemistry of the dimers, although it is apparent
from the JH,*H coupling constants that they all adopt an anti
conformation about the new σ-bond.
B. Two-Electron Reduction. Whereas in the first reduction
process the length of the tether shows no influence on the nature
of the reduction product, in the second reduction step, the
3
(36) Hoijtink, G. J.; Townsend, J.; Weissman, S. I. J. Chem. Phys. 1961, 34,
507.
(37) Further support comes from the calclulation of the atomic-orbital coefficients
of the LUMO of 2 that contains the most strained pyrene moiety. DFT
calculations (B3LYP/6-311G**) predict a high spin density for the odd
electron in the monoanion radical of 2 at carbon atom C1 and its
symmetrical counterparts.
The fact that compounds 5 and 6 behave like pyrene when
reduced with alkali metals and the presence of solubility
problems in dimers with large tethers suggest that a dimerization
process might also be possible in pyrene. The difficulty in
observing such a dimer may lie in its low solubility.
(38) The contribution of nonplanarity to the solubility of the systems is not
well understood. Because nonplanarity is a function of tether length, the
discussion on solubility will only focus on the latter, which indirectly
contains the effect of the former.
(40) Sethson, I.; Johnels, D.; Edlund, U.; Sygula, A. J. Chem. Soc., Perkin Trans.
2 1990, 1339.
(39) Such a reduction process has been observed in azulene. See ref 14b.
9
J. AM. CHEM. SOC. VOL. 126, NO. 21, 2004 6771

A R T I C L E S
Aprahamian et al.
0.62 (m, 1H, H16′), 0.35 (m, 1H, H15), 0.21 (m, 1H, H14), 0.08 (dd,
1H, H12′), -0.12 (t, 1H, H13 and H15′), -0.39 (dd, 1H, H13′), -0.60
(dd, 1H, H14′) ppm. δC (THF-d8, 200 K) 142.66 (C8a), 142.19 (C5a),
140.56 (C10b), 138.85 (C10c), 135.48 (C7), 131.87 (C4), 129.75 (C10),
129.56 (C3), 119.28 (C6), 114.78 (C8), 110.65 (C3a), 110.04 (C9),
105.25 (C10a), 102.75 (C5), 99.41 (C2), 48.28 (C1), 40.74 (C11), 37.27
(C17), 34.55 (C14), 32.53 (C16), 31.50 (C12), 27.84 (C13), 25.09 (C15)
ppm. δLi (THF-d8, 200 K) -0.63 ppm.
The use of the proton-coupled HSQCSI method was crucial
in this study as it helped differentiate between the diverse
reduction products.
Experimental Section
The 1D and 2D NMR spectra were recorded on a Bruker DRX-400
pulsed FT spectrometer operating at 400.13, 100.62, and 155.51 MHz
for 1H, 13C, and 7Li NMR, respectively. Chemical shifts were measured
relative to the most downfield solvent peak (THF-d8 3.57 ppm relative
to TMS). Coupling constants are reported whenever they are essential.
The temperature was calibrated with methanol.
2a/2K+: δH (THF-d8, 260 K, dark-red) 6.73 (d, 1H, H4), 6.00 (d,
1H, H9), 5.94 (d, 1H, H10), 5.88 (s, 1H, H3), 5.87 (d, 1H, H5), 5.81
(s, 1H, H8), 5.73 (s, 1H, H6), 3.60 (s, 1H, JC,H ) 130.52 Hz, JH,*H
)
11.52 Hz, H1), 2.48 (dd, 1H, H17), 2.15 (d, 1H, H11), 1.77 (m, H17′),
1.61 (t, 1H, H11′), 1.42 (b, 1H, H16), 0.85 (b, 1H, H12), 0.67 (b, 1H,
H16′), 0.42 (b, 1H, H15), 0.23 (b, 1H, H14), 0.16 (dd, 1H, H12′), -0.10
(b, 1H, H13), -0.20 (t, 1H, H15′) -0.48 (m, 1H, H13′), -0.57 (m,
1H, H14′) ppm. δC (THF-d8, 260 K) 143.71 (C8a), 142.09 (C5a), 140.93
(C10b), 136.25(C7), 135.85 (C10c), 132.71 (C4), 131.07 (C3), 129.11
(C10), 120.53 (C6), 115.83 (C8), 111.56 (C3a), 110.04 (C9), 104.55
(C5), 104.10 (C10a), 100.61 (C2), 50.68 (C1), 40.53 (C11), 37.18 (C17),
34.39 (C14), 32.28 (C16), 31.94 (C12), 27.89 (C13), 25.35 (C15) ppm.
2b/2Li+: δH (THF-d8, 273 K, wine-red) 6.18 (d, 1H, H10), 5.22 (d,
1H, H9), 5.06 (s, 1H, H8), 2.31 (s, 1H, JC,H ) 163.16 Hz, JH,*H ) 9.06
Hz, H1), 2.00 (m, 1H, H17), 1.42 (m, 1H, H16), 1.21 (m, 1H, H11),
1.03 (m, 1H, H15), 0.82 (m, 1H, H12), 0.80 (m, 1H, H14), 0.71 (m,
1H, H13) ppm. δC (THF-d8, 273 K) 147.61 (C10b), 145.76 (C8a),
141.60 (C7), 136.94 (C10c), 125.66 (C10), 118.53 (C10a), 107.93 (C8),
101.60 (C9), 38.84 (C17), 33.92 (C1 and C13), 33.95 (C16), 32.76
(C11), 30.85 (C14), 29.26 (C15), 29.16 (C2), 27.99 (C12) ppm. δLi
(THF-d8, 273 K) 0.16 ppm.
2b/2K+: δH (THF-d8, 260 K, brown) 6.20 (d, 1H, H10), 5.24 (d,
1H, H9), 5.05 (s, 1H, H8), 2.28 (s, 1H, JC,H ) 162.44 Hz, JH,*H ) 7.89
Hz, H1), 1.98 (t, 1H, H17), 1.40 (b, 1H, H16), 1.17 (b, 1H, H11), 1.01
(b, 1H, H15), 0.80 (m, 1H, H12), 0.77 (m, 1H, H14), 0.71 (m, 1H,
H13) ppm. δC (THF-d8, 273 K) 147.39 (C10b), 145.71 (C8a), 143.01
(C7), 136.52 (C10c), 126.29 (C10), 119.12 (C10a), 107.86 (C8), 102.13
(C9), 38.70 (C17), 33.84 (C1), 33.66 (C13), 32.82 (C16), 32.56 (C11),
30.77 (C14), 29.23 (C15 and C2), 27.85 (C12) ppm.
The assignment of the tether hydrogen atoms is conducted as
follows: the tether hydrogen showing a NOESY (Nuclear Overhauser
Effect Spectroscopy) interaction with hydrogen atom H1, which belongs
to the pyrene moiety, is assigned as H11. This hydrogen atom will
show a COSY interaction with hydrogen atom H11′ (if these atoms
are not equivalent) and a secondary interaction with hydrogen atom
H12. The rest of the hydrogen atoms are assigned following this
sequence, using the COSY interactions. In some cases no integration
data is given, because integration of the tether hydrogens was not
possible due to overlapping or to their lying under the solvent peaks.
For the synthetic work, reactions were performed under nitrogen
unless otherwise indicated. Solvents designated as dry were purified
and dried using standard procedures. Spectroscopic grade benzene was
degassed under reduced pressure immediately prior to use. All other
solvents were used as received. Chromatographic separations were
performed using 230-400 mesh silica gel. TLC plates were visualized
using a short wave (254 nm) UV lamp. Melting points are uncorrected.
1H and 13C NMR were acquired using CDCl3 solutions at 500 MHz
and 125 MHz, respectively. Reported multiplicities are apparent. Low-
and high-resolution EI mass spectroscopic data were obtained at 70
eV.
Reduction of the Samples. All [n](2,7)pyrenophanes were reduced
in 5 mm NMR glass tubes equipped with an upper reduction chamber.
The pyrenophanes (3-5 mg) were introduced into the lower chamber
of the tube under argon atmosphere. The alkali metal (kept in paraffin
oil, cleansed from the oxidized layer and rinsed in petroleum ether
40-60 °C) was introduced under argon to the reduction chamber as a
lithium wire or a piece of potassium. The tube was then placed under
high vacuum and dried by flame. In the case of potassium, the metal
was sublimed several times, creating a potassium mirror on the reduction
chamber. Approximately 1 mL of anhydrous THF-d8 (dried over a
sodium/potassium alloy under high vacuum) was vacuum-transferred
to the NMR tube and was degassed several times. Finally the tube was
flame-sealed under high vacuum.
3a/2Li+: δH (THF-d8, 240 K, wine-red) 6.57 (m, 1H, H19), 6.56
(m, 1H, H18/H20), 6.29 (m, 1H, H18/H20), 6.26 (d, 1H, H4), 6.02 (d,
1H, H9), 5.91 (d, 1H, H10), 5.87 (s, 1H, H3), 5.82 (s, 1H, H8), 5.26
(d, 1H, H5), 5.24 (s, 1H, H6), 5.10 (s, 1H, H14), 3.41 (s, 1H, JC,H
)
130.03 Hz, JH,*H ) 10.60 Hz, H1), 2.66 (q, 1H, H17), 2.47 (t, 1H,
H11), 2.25 (m, 1H, H16′), 2.23 (m, 1H, H17′), 2.21 (m, 1H, H11′),
1.95 (d, 1H, H12′), 1.85 (m, 1H, H16), 1.45 (t, 1H, H12) ppm. δC
(THF-d8, 240 K) 143.02 (C8a), 142.37 (C5a), 140.52 (C10b), 139.55
(C13/C15), 139.33 (C13/C15), 135.79 (C10c), 131.74 (C7), 129.80 (C4
and C14), 129.10 (C10), 129.00 (C3), 125.71 (C18/C20), 124.81 (C19),
124.21 (C18/C20), 118.28 (C6), 114.13 (C8), 113.08 (C3a), 109.34
(C9), 103.48 (C10a), 102.56 (C5), 95.74 (C2), 48.68 (C1), 40.22 (C16),
38.05 (C11), 37.95 (C17), 35.79 (C12) ppm. δLi (THF-d8, 240 K) -0.50
ppm.
3a/2K+: δH (THF-d8, 260 K, brown) 6.58 (t, 1H, H19), 6.51 (d,
2H, H18 and H20), 6.36 (d, 2H, H18 and H20), 6.51 (d, 1H, H4), 6.03
(d, 1H, H9), 5.94 (s, 1H, H3), 5.90 (s, 1H, H8), 5.88 (d, 1H, H10),
5.40 (d, 1H, H5), 5.36 (s, 1H, H6), 5.15 (s, 1H, H14), 3.67 (s, 1H, JC,H
) 130.98 Hz, JH,*H ) 10.80 Hz, H1), 2.69 (q, 1H, H17), 2.55 (t, 1H,
H11), 2.37 (dd, 1H, H16), 2.30 (m, 1H, H11′), 2.23 (m, 1H, H17),
1.99 (d, 1H, H12), 1.85 (m, 1H, H16), 1.49 (t, 1H, H12′) ppm. δC
(THF-d8, 260 K) 143.97 (C8a), 142.08 (C5a), 140.58 (C10b), 139.39
(C13/C15), 139.32 (C13/C15), 135.48 (C10c), 134.66 (C7), 132.37
(C4), 132.02 (C14), 130.35 (C3), 128.42 (C10), 125.89 (C18/C20),
124.93 (C19), 124.26 (C18/C20), 119.42 (C6), 115.21 (C8), 113.89
(C3a), 109.11 (C9), 104.67 (C5), 102.46 (C10a), 96.79 (C2), 51.29
(C1), 39.98 (C16), 37.88 (C17), 37.74 (C11), 36.17 (C12) ppm.
3b/2Li+: The ∆Gq for the dynamic process was measured to be
9.1 ( 0.2 kcal/mol.
Controlled Reduction Process. The reduction takes place when the
THF-d8 solution is brought into contact with the metal by inverting
the sample in solid dry ice. Reduction is stopped by returning the sample
to the upright position separating the metal from the solution. The
formation of the anions was detected visually by the change in the
1
color of the solution and by H NMR spectroscopy.
Quench Reactions. Quenching with oxygen was performed under
a nitrogen funnel. The sample tube was broken, and the reduction
chamber containing the metal was removed. A mild stream of oxygen
gas was bubbled through a syringe into the cooled solution until the
color changed. Reaction with iodine and dimethyl sulfate was performed
by breaking the sample tube (ca. 10 mg of substrate) in a glovebox
and pouring the solution into a vial containing an appropriate amount
of the reagent. The product was then extracted with dichloromethane
and dried over magnesium sulfate, and the solvent was evaporated.
NMR Data. 2a/2Li+: δH (THF-d8, 200 K, wine-red) 6.63 (d, 1H,
H4), 5.98 (d, 1H, H10), 5.95 (d, 1H, H9), 5.78 (s, 1H, H3), 5.71 (d,
1H, H5), 5.71 (s, 1H, H8), 5.62 (s, 1H, H6), 3.22 (s, 1H, JC,H ) 128.23
Hz, JH,*H ) 10.26 Hz, H1), 2.45 (d, 1H, H17), 2.06 (d, 1H, H11), 1.80
(m, H17′), 1.57 (m, 1H, H11′), 1.40 (m, 1H, H16), 0.76 (m, 1H, H12),
9
6772 J. AM. CHEM. SOC. VOL. 126, NO. 21, 2004

Reduction of Strained Polycycles: Pyrene Anion Strain
A R T I C L E S
δH (THF-d8, 260 K, brown-red) 6.83 (s, 1H, H14), 6.71 (t, 1H, H19),
6.60 (d, 1H, H18), 6.51 (d, 1H, H20), 6.17 (d, 2H, H10), 5.11 (d, 2H,
H9), 4.82 (s, 2H, H8), 2.57 (t, 1H, H16), 2.39 (s, 2H, JC,H ) 162.33
Hz, JH,*H ) 8.66 Hz, H1), 2.27 (t, 1H, H11), 2.20 (t, 1H, H17), 1.54 (t,
1H, H12) ppm. δC (THF-d8, 260 K) 146.95 (C10b), 146.09 (C13),
145.13 (C8a), 140.64 (C15), 140.52 (C7), 136.47 (C10c), 133.46 (C14),
126.23 (C19), 125.77 (C18), 125.58 (C10), 125.43 (C20), 118.07
(C10a), 108.47 (C8), 101.63 (C9), 39.92 (C16), 39.59 (C17), 33.71
(C1), 33.12 (C12), 32.97 (C11), 30.10 (C2). δLi (THF-d8, 260 K) -0.20
ppm.
1H, H12), 1.08 (m, 1H, H18′), 0.89 (m, 1H, H17′/H12′), 0.88 (m, 1H,
H17), 0.87 (m, 1H, H12′/H17′), 0.74 (m, 1H, H13′), 0.61 (m, 1H, H13),
0.28 (m, 2H, H15′ and 16), 0.10 (m, 3H, H14, H15, and H16′), -0.11
(m, 1H, H14′) ppm. δC (THF-d8, 220 K) 142.56 (C5a), 141.31 (C8a),
136.95 (C7), 136.75 (C10b), 134.98 (C10c), 131.27 (C10), 129.42 (C4),
126.30 (C3), 112.20 (C6), 112.20 (C3a), 109.64 (C8), 108.94 (C9),
108.31 (C10a), 108.31 (C2), 102.43 (C5), 52.98 (C1), 40.63 (C11),
37.57 (C19), 34.06 (C14), 33.60 (C16), 32.41 (C15), 30.66 (C18), 29.87
(C12), 27.02 (C17), 26.75 (C13) ppm.
5b/2Li+: The ∆Gq for the dynamic process was measured to be
8.6 ( 0.2 kcal/mol.
3b/2K+: The ∆Gq for the dynamic process was measured to be 8.9
( 0.2 kcal/mol.
δH (THF-d8, 165 K, brown-red) 7.98 (b, 2H, H14), 4.61 (b, 1H,
H15′), 3.55 (H13′), 2.61 (s, 2H, H4/H10), 2.58 (b, H14′), 2.50 (s, 2H,
H4/H10), 2.43 (b, 1H, H15), 2.14 (b, 4H, H12′ and H13), 1.95 (b, 2H,
H12), 1.57 (s, 2H, H1/H3), 1.29 (s, 2H, H1/H3), 0.96 (b, 2H, H11),
0.83 (b, 2H, H11′) ppm. δC (THF-d8, 165 K) 151.91 (C10b), 144.28
(C3a/C10a), 143.40 (C3a/C10a), 137.46 (C2), 108.53 (C3/C4), 107.68
(C3/C4), 98.04 (C1/C3), 83.74 (C1/C3), 38.12 (C11), 37.89 (C14),
36.20 (C15), 32.77 (C12), 29.39 (C13) ppm. δLi (THF-d8, 165 K) 1.98
ppm. Carbon atoms C10b and C2 were assigned according to the 13C
NMR of 1b. The differentiation between hydrogen and carbon atoms
C1 and C3 (also C4 and C10) is not possible.
δH (THF-d8, 273 K, brown) 6.85 (s, 1H, H14), 6.76 (t, 1H, H19),
6.64 (d, 1H, H18), 6.55 (d, 1H, H20), 6.25 (d, 1H, H10), 5.18 (d, 1H,
H9), 4.86 (s, 1H, H8), 2.60 (t, 1H, H16), 2.42 (s, 1H, JC,H ) 163.64
Hz, JH,*H ) 8.12 Hz, H1), 2.33 (t, 1H, H11), 2.22 (t, 1H, H17), 1.60 (t,
1H, H12) ppm. δC (THF-d8, 273 K) 146.78 (C10b), 145.90 (C13),
145.13 (C8a), 142.29 (C7), 140.37 (C15), 136.03 (C10c), 133.13 (C14),
126.59 (C19), 126.26 (C10), 125.63 (C18), 125.60 (C20), 118.98
(C10a), 108.27 (C8), 102.14 (C9), 39.72 (C16), 39.50 (C17), 33.80
(C1), 33.18 (C12), 32.85 (C11), 30.30 (C2) ppm.
4a/2Li+: δH (THF-d8, 273 K, wine-red) 6.58 (d, 1H, H4), 5.94 (dd,
2H, H9 and H10), 5.84 (s, 1H, H3), 5.76 (s, 1H, H8), 5.69 (s, 1H, H6),
5.67 (d, 1H, H5), 3.24 (s, 1H, H1, JC,H ) 130.92 Hz, JH,*H ) 10.61
Hz), 2.45 (dd, 1H, H18), 2.22 (d, 1H, H11), 1.83 (dt, 1H, H18′), 1.60
(m, 2H, H11′ and H17′), 1.03 (dd, 1H, H12), 0.86 (m, 2H, H16 and
H17′), 0.63 (dd, 1H, H16′), 0.38 (dd, 1H, H12′), 0.23 (dd, 1H, H14),
-0.01 (m, 2H, H13′ and H15′), -0.26 (dd, 1H, H13), -0.50 (dd, 1H,
H14′), -0.82 (dd, 1H, H15) ppm. δC (THF-d8, 273 K) 143.39 (C5a),
139.78 (C8a), 138.59 (C10b), 135.69 (C7), 135.31 (C10c), 131.15 (C4),
129.77 (C10), 129.13 (C3), 114.21 (C8), 111.57 (C3a), 110.16 (C6),
109.52 (C9), 108.51 (C10a), 103.31 (C2), 102.76 (C5), 49.07 (C1),
40.84 (C11), 38.37 (C18), 33.40 (C15), 32.23 (C12), 32.08 (C17), 31.91
(C14), 27.81 (C16), 27.13 (C13) ppm. δLi (THF-d8, 273 K) -0.70 ppm.
4a/2K+: δH (THF-d8, 240 K, wine-red) 6.66 (d, 1H, H4), 5.98 (d,
1H, H9), 5.88 (d, 1H, H10), 5.87 (s, 1H, H3), 5.83 (s, 1H, H8), 5.82
5b/2K+: The ∆Gq for the dynamic process was measured to be 8.6
( 0.2 kcal/mol.
δH (THF-d8, 170 K, blue) 9.93 (b, 2H, H14), 8.65 (b, 1H, H15′),
4.01 (b, 2H, H13′), 3.22 (b, 2H, H14′), 2.89 (b, 1H, H15), 2.41 (b, 2H,
H13), 2.34 (b, 2H, H12′), 2.04 (b, 2H, H12), 1.89 (s, 2H, H4/H10),
1.83 (s, 2H, H4/H10), 0.99 (s, 2H, H1/H3), 0.89 (s, 2H, H1/H3), 0.64
(b, 2H, H11), 0.56 (b, 2H, H11′) ppm. δC (THF-d8, 170 K) 156.66
(C10b), 145.02 (C3a/C10a), 144.60 (C3a/C10a), 141.70 (C2), 107.71
(C3/C4), 106.89 (C3/C4), 98.47 (C1/C3), 89.71 (C1/C3), 39.17 (C14),
37.89 (C11), 37.34 (C15), 33.77 (C13), 30.60 (C12) ppm. Carbon atoms
C10b and C2 were assigned according to the 13C NMR of 1b. The
differentiation between hydrogen and carbon atoms C1 and C3 (also
C4 and C10) is not possible.
5c/2Li+: δH (THF-d8, 273 K, dark-blue) 6.33 (d, 1H, H4), 6.24 (d,
1H, H10), 6.05 (d, 1H, H9), 5.78 (s, 1H, H3), 5.77 (s, 1H, H8), 5.59
(s, 1H, H6), 5.58 (d, 1H, H5), 2.83 (s, 1H, JC,H ) 122.36 Hz, H1),
2.38 (m, 1H, H19), 2.08 (m, 1H, H11), 1.98 (m, 1H, H19′), 1.90 (m,
1H, H11′), 1.42 (m, 1H, H18′), 1.27 (m, 1H, H12), 1.14 (m, 1H, H18),
0.89 (m, 1H, H17′), 0.86 (m, 3H, H12, H17 and H17′), 0.68 (m, 1H,
H13), 0.53 (m, 1H, H13′), 0.26 (m, 1H, H15), 0.21 (m, 1H, H14′),
0.18 (m, 1H, H16), 0.07 (m, 2H, H15′ and H16′), 0.03 (m, 1H, H14)
ppm. δC (THF-d8, 273 K) 142.25 (C5a), 141.40 (C8a), 136.23 (C10b),
136.08 (C7), 135.15 (C10c), 129.76 (C4), 128.03 (C10), 126.99 (C3),
112.99 (C6), 111.26 (C3a), 110.45 (C8), 110.16 (C9), 108.33 (C10a),
104.05 (C2), 103.55 (C5), 48.95 (C1), 39.54 (C11), 37.54 (C19), 33.80
(C14), 33.51 (C16), 32.50 (C15), 30.80 (C18), 30.62 (C12), 27.15
(C17), 26.61 (C13) ppm. δLi (THF-d8, 273 K) 0.05 ppm.
(s, 1H, H6), 5.81 (d, 1H, H5), 3.49 (s, 1H, JC,H ) 130.58 Hz, JH,*H
)
11.14 Hz, H1), 2.46 (dd, 1H, H18), 2.28 (d, 1H, H11), 1.86 (m, 1H,
H18′), 1.62 (d, 1H, H11′), 1.57 (d, 1H, H17′), 1.03 (d, 1H, H12), 0.85
(m, 1H, H15), 0.81 (m, 1H, H17′), 0.62 (t, 1H, H15′), 0.41 (dd, 1H,
H12′), 0.27 (d, 1H, H13), -0.03 (m, 1H, H14 and H16), -0.38 (d,
1H, H16′), -0.44 (d, 1H, H13′), -1.04 (dd, 1H, H14′) ppm. δC (THF-
d8, 240 K) 142.87 (C5a), 140.17 (C8a), 138.85 (C10b), 136.09 (C7),
134.97 (C10c), 131.54 (C4), 129.85 (C3), 129.36 (C10), 114.89 (C8),
111.74 (C6), 111.24 (C3a), 109.14 (C9), 107.15 (C10a), 104.51 (C5),
103.57 (C2), 49.85 (C1), 40.69 (C11), 38.17 (C18), 32.48 (C14), 32.11
(C12), 31.96 (C17), 31.51 (C13), 27.76 (C15), 27.12 (C16) ppm.
5a/2Li+: δH (THF-d8, 240 K, wine-red) 6.33 (d, 1H, H4), 5.84 (s,
1H, H3), 5.81 (d, 1H, H10), 5.76 (d, 1H, H9), 5.58 (s, 1H, H8), 5.51
(d, 1H, H5), 5.45 (s, 1H, H6), 3.27 (s, 1H, JC,H ) 130.64 Hz, JH,*H
)
5c/2K+: δH (THF-d8, 298 K, dark-blue) 6.40 (d, 1H, H4), 6.35 (d,
1H, H10), 6.14 (d, 1H, H9), 5.87 (s, 1H, H3), 5.79 (s, 1H, H8), 5.63
(d, 1H, H5), 5.61 (s, 1H, H6), 3.17 (s, 1H, JC,H ) 122.96 Hz, H1),
2.38 (m, H19), 2.14 (m, H11), 2.03 (m, H19′), 1.93 (m, H11′), 1.47
(m, H18′), 1.30 (m, H12), 1.15 (m, H 18), 0.97 (m, H17′), 0.90 (m,
H12′), 0.85 (m, H17), 0.77 (m, H13′), 0.58 (m, H13), 0.36 (m, H15′),
0.24 (m, H16′), 0.20 (m, H14′), 0.16 (m, H15), 0.14 (m, H16), -0.02
(m, H14) ppm. δC (THF-d8, 298 K) 142.64 (C5a), 141.98 (C8a), 138.30
(C7), 136.31 (C10b), 135.03 (C10c), 130.25 (C4), 129.16 (C10), 126.84
(C3), 112.47 (C6), 112.05 (C3a), 110.79 (C9), 110.00 (C8), 109.60
(C10a), 107.34 (C2), 103.07 (C5), 50.50 (C1), 39.42 (C11), 37.58 (C19),
33.93 (C14), 33.64 (C16), 32.54 (C15), 30.89 (C18), 30.35 (C12), 27.27
(C17), 26.84 (C13) ppm.
10.44 Hz, H1), 2.39 (m, 1H, H11), 2.30 (m, 1H, H19), 1.95 (m, 1H,
H19′), 1.91 (m, 1H, H11′), 1.42 (m, 1H, H18), 1.14 (m, 1H, H12),
1.08 (m, 1H, H18′), 0.87 (m, 3H, H12′, H17, and H17′), 0.69 (m, 1H,
H13′), 0.60 (m, 1H, H13), 0.27 (m, 2H, H15′ and 16), 0.08 (m, 3H,
H14, H15, and H16′), -0.14 (1H, H14′) ppm. δC (THF-d8, 240 K)
142.59 (C5a), 140.94 (C8a), 136.58 (C10b), 135.89 (C7), 135.21
(C10c), 131.07 (C10), 128.89 (C4), 126.43 (C3), 112.85 (C10a), 112.53
(C3a), 111.75 (C6), 108.73 (C8 and C9), 106.84 (C2), 102.18 (C5),
52.39 (C1), 40.85 (C11), 37.65 (C19), 34.13 (C14), 33.73 (C16), 32.42
(C15), 30.84 (C18), 30.75 (C12), 27.01 (C17), 26.66 (C13) ppm. δLi
(THF-d8, 240 K) -0.69 ppm.
5a/2K+: δH (THF-d8, 220 K, wine-red) 6.40 (d, 1H, H4), 5.88 (s,
1H, H3), 5.84 (d, 1H, H9), 5.74 (d, 1H, H10), 5.65 (s, 1H, H8), 5.60
(d, 1H, H5), 5.54 (s, 1H, H6), 3.36 (s, 1H, H1, JC,H ) 132.05 Hz, JH,*H
) 10.15 Hz), 2.44 (m, 1H, H11), 2.33 (m, 1H, H19), 1.95 (m, 1H,
H19′/H11′), 1.89 (m, 1H, H11′/H19′), 1.43 (m, 1H, H18), 1.21 (m,
6: Synthesis. A. 1,10-Bis(3,5-bis(methoxycarbonyl)phenyl)-1,9-
decadiyne (9). A mixture of 3,5-bis(methoxycarbonyl)phenyl triflate
86c (8.18 g, 23.9 mmol), DBU (4.93 mL, 9.03 g, 59.4 mmol), 1,9-
decadiyne (2.14 mL, 1.67 g 11.4 mmol), CuI (0.45 g, 2.4 mmol), and
9
J. AM. CHEM. SOC. VOL. 126, NO. 21, 2004 6773

A R T I C L E S
Aprahamian et al.
Pd(PPh3)2Cl2 (0.35 g, 0.50 mmol) in benzene (250 mL) was stirred at
room temperature for 48 h. The reaction mixture was washed with 1
M aqueous HCl solution, water, and saturated aqueous NaCl solution.
The organic layer was dried over MgSO4, filtered, and concentrated
under reduced pressure to afford an oily brown solid. Column
chromatography (CH2Cl2) afforded 9 as a light brown solid (4.21 g,
68%). Mp 127.5-129 °C (95% ethanol); δH (CDCl3) 8.55 (s, 2H), 8.22
(s, 4H), 3.94 (s, 12H), 2.46-2.44 (m, 4H), 1.67-1.65 (m, 8H); δC
(CDCl3) 166.2, 136.9, 131.2, 129.8, 125.5, 100.0, 93.0, 79.4, 52.9, 28.8,
19.8; EI-MS m/z (%) 518 (100, M+), 490 (33), 475 (14), 464 (10), 427
(14), 399 (9), 395 (4), 357 (3), 311 (8), 228 (17), 200 (8), 155 (9), 115
(13). HRMS: calcd for C30H30O8 518.1941, found 518.1948.
E. [10](2,7)Pyrenophane (6). To a solution of dithiacyclophane 12
(1.34 g, 3.26 mmol) in dry dichloromethane (30 mL) was added
dropwise Borch reagent ((MeO)2CHBF4) (1.10 g, 6.85 mmol) over a
period of 15 min. The reaction was then stirred for a further 3 h. The
reaction mixture was concentrated under reduced pressure, and the
resulting residue was stirred for 10 min with ethyl acetate (10 mL).
The solvent was removed by decantation, and the resulting gummy
pink solid was dried in vacuo until a fine pink powder was obtained.
This was slurried in dry THF (30 mL), and to this stirred mixture was
added t-BuOK (1.68 g, 14.8 mmol). The resulting brown mixture was
stirred for 16 h. Concentration of the reaction mixture gave a brown
oil, which was taken up in dichloromethane and passed through a short
column of silica gel to afford a yellow oil. This oil was dissolved in
dry dichloromethane (30 mL), and Borch reagent (0.915 g, 5.72 mmol)
was added dropwise over a period of 15 min. The mixture was stirred
for 6 h, and the solvent was removed under reduced pressure and dried
in vacuo to yield a sticky brown solid. To the resulting residue was
added dry THF (30 mL) and then t-BuOK (1.55 g, 13.7 mmol), and
the resulting mixture was stirred for 6 h. The reaction mixture was
concentrated under reduced pressure to afford a brown solid. Column
chromatography (25% CH2Cl2/hexanes) afforded a ca. 3:1 mixture of
[10](2,7) pyrenophane and an unidentified [10](2,7)dihydropyrenophane.
This mixture was dissolved in distilled benzene (25 mL), and to the
resulting solution was added DDQ (0.056 g, 0.33 mmol). The resulting
mixture was heated at reflux for 3 h, cooled to room temperature, and
quenched by the addition of a few crystals of 1,4-hydroquinone. Column
chromatography (hexanes) afforded pyrenophane 6 (0.225 g, 20% from
12) as a white solid. Mp 253-255 °C (hexanes); δH (CDCl3) 7.96 (s,
4H), 7.87 (s, 4H) 2.99-2.96 (m, 4H), 1.54-1.40 (m, 4H), 0.46-0.44
(m, 4H), -1.03, -1.04 (m, 4H), -1.70 (br s, 4H); δC (CDCl3) 137.4,
131.1, 129.0, 127.6, 126.8, 35.9, 31.5, 31.4, 25.7, 23.4; EI-MS m/z
(%) 340 (100, M+), 279 (2), 228 (36), 215, (22), 202, (3), 149 (18).
HRMS: calcd for C26H28 340.2191, found 340.2189.
B. 1,10-Bis(3,5-bis(methoxycarbonyl)phenyl)decane (10). A mix-
ture of diynetetraester 9 (2.05 g, 3.95 mmol) in EtOAc (100 mL) was
purged with nitrogen gas for 20 min, and then 10% Pd/C (0.050 g)
was added. The resulting mixture was stirred under an atmosphere of
hydrogen (balloon) for 16 h. The reaction mixture was filtered through
a plug of Celite, and the filtrate was concentrated under reduced pressure
to yield 10 as a white solid (2.05 g, 99%). Mp 99-100 °C (95%
ethanol); δH (CDCl3) 8.50 (s, 2H), 8.05 (s, 4H), 3.94 (s, 12H), 2.70 (t,
J ) 8.1 Hz, 4H) 1.66-1.63 (m, 4H), 1.31-1.25 (br s, 12H); δC (CDCl3)
166.9 144.2, 134.2, 130.9, 128.6, 52.7, 36.0, 31.6, 29.9, 29.8, 29.6;
EI-MS m/z (%) 526 (32, M+), 494 (69), 462 (100), 430 (46), 401 (36),
398 (3), 342 (3), 232 (20), 207 (42), 189 (40), 149 (27), 119 (18), 105
(11). HRMS: calcd for C30H38O8 526.2567, found 526.2560.
C. 1,10-Bis(3,5-bis(bromomethyl)phenyl)decane (11). To a stirred
slurry of LiAlH4 (4.61 g, 122 mmol) in THF (150 mL) was added by
syringe over 1 h a solution tetraester 10 (4.36 g, 8.28 mmol) in THF
(100 mL). The resulting gray slurry was brought to a gentle reflux for
16 h. The reaction mixture was cooled on an ice bath and then quenched
by the careful addition of EtOAc (200 mL) and then 1 M aqueous HCl
solution (100 mL). The resulting mixture was stirred for a further 1 h.
The layers were separated, and the aqueous layer was extracted with
EtOAc. The combined organic layers were washed with saturated
aqueous NaCl solution, dried over MgSO4, filtered, and concentrated
under reduced pressure to yield a white solid. To this was added a
mixture of glacial acetic acid (125 mL) and 30% HBr/HOAc (7.04
mL, 33.2 mmol HBr), and the resulting brown mixture was heated at
reflux for 45 min. The reaction mixture was cooled and poured into
water (50 mL) and extracted with CH2Cl2. The combined organic layers
were washed with saturated aqueous NaHCO3 solution, water, and
saturated aqueous NaCl solution. The organic layer was dried over
MgSO4, filtered, and concentrated under reduced pressure to yield a
brown oil. Column chromatography (35% CH2Cl2/hexanes) afforded
tetrabromide 11 (3.25 g, 66%) as a white solid. Mp 97-98 °C (toluene/
heptane); δH (CDCl3) 7.22 (s, 2H), 7.13 (s, 4H), 4.40 (s, 8H), 2.55 (t,
J ) 8.0 Hz, 4H), 1.62-1.56 (m, 4H), 1.27 (br s, 12H); δC (CDCl3)
144.6, 138.5, 129.5, 127.2, 35.9, 33.4, 29.7, 29.5, 26.3; EI-MS m/z
(%) M+ not observed. Satisfactory HRMS for this compound was not
obtained.
6: The ∆Gq for the dynamic process was measured to be 8.3 ( 0.2
kcal/mol. δH (THF-d8, 273 K, colorless) 7.98 (s, 1H, H10), 7.93 (s,
1H, H1), 2.97 (t, 1H, H11), 1.40 (dt, 1H, H12), 0.46 (m, 1H, H13),
-1.07 (b, 1H, H15), -1.13 (b, 1H, H14) ppm. δC (THF-d8, 298 K)
138.58 (C2), 132.09 (C10a), 128.24 (C1), 127.76 (C10), 125.63 (C10b),
36.87 (C11), 31.98 (C15), 31.34 (C14), 30.51 (12), 26.32 (C13) ppm.
6a/2Li+: δH (THF-d8, 220 K, wine-red) 6.02 (d, 1H, H4), 5.85 (d,
1H, H10), 5.56 (s, 1H, H9), 5.49 (s, 1H, H3), 5.42 (s, 1H, H8), 5.31
(d, 1H, H6), 3.45 (s, 1H, H5), 2.40 (b, 1H, H20), 2.26 (b, 1H, H11)
1
ppm. Assigned according to H NMR spectrum of 6c.
6b/2Li+: The ∆Gq for the dynamic process was measured to be
8.9 ( 0.2 cal/mol. δH (THF-d8, 165 K, brown-red) 6.95 (b, 1H, H14),
4.99 (b, 1H, H15′), 2.91 (b, 1H, H13′), 2.70 (b, 1H, H4/H10), 2.55 (b,
1H, H4/H10), 2.38 (b, 2H, H14′ and H15), 1.92 (H13), 1.77 (H1/H3),
1.67 (H12′), 1.57 (H12), 1.11 (b, 1H, H1/H3), 0.68 (b, 1H, H11), 0.54
(b, 1H, H11′) ppm. δC (THF-d8, 165 K) 151.02 (C10b), 149.27 (C3a/
C10a), 139.17 (C3a/C10a), 138.56 (C2), 110.79 (C4/C10), 106.73 (C4/
C10), 101.08 (C1/C3), 79.83 (C1/C3), 37.55 (C11), 36.54 (C14 and
C15), 30.30 (C12), 28.61 (C13) ppm. δLi (THF-d8, 165 K) 1.16 ppm.
The differentiation between hydrogen and carbon atoms C1 and C3
(also C4 and C10) is not possible. Carbon atoms C10b and C2 were
assigned according to the 13C NMR of 1b.
6b/2K+: The ∆Gq for the dynamic process was measured to be 8.4
( 0.2 kcal/mol. δH (THF-d8, 165 K, blue) 8.68 (b, 1H, H14), 6.60 (b,
1H, H15′), 3.54 (b, 1H, H13′), 3.10 (b, 1H, H14′), 2.97 (b, 1H, H15),
2.14 (b, 1H, H13), 1.81 (b, 1H, H4/H10), 1.75 (b, 1H, H12′), 1.61 (b,
1H, H4/H10), 1.53 (b, 1H, H12), 0.87 (b, 1H, H1/H3), 0.64 (b, 1H,
H1/H3), 0.24 (b, 1H, H11), 0.18 (b, 1H, H11′) ppm. δC (THF-d8, 298
K) 156.02 (C10b), 148.92 (C3a/C10a), 142.33 (C2 and C10a/C3a),
109.06 (C4/C10), 106.63 (C4/C10), 99.19 (C1/C3), 88.65 (C1/C3),
38.05 (C14), 37.79 (C15), 36.60 (C11), 31.06 (C12), 28.95 (C13) ppm.
The differentiation between hydrogen and carbon atoms C1 and C3
(also C4 and C10) is not possible.
D. 18,27-Dithia[10.3.3](1,3,5)cyclophane (12). To an efficiently
magnetically stirred solution of tetrabromide 11 (2.78 g, 4.17 mmol)
in 10% EtOH/CH2Cl2 (700 mL) under air was added Na2S/Al2O3 (5.56
g, 13.6 mmol) in four roughly equal portions over 45 min. The resulting
milky mixture was stirred for a further 16 h. The resulting mixture
was gravity filtered, and the filtrate was concentrated under reduced
pressure to afford a pale yellow solid. Flash chromatography (CH2Cl2)
of the residue afforded dithiacyclophane 12 (1.34 g, 78%) as a white
solid. Mp 175-177 °C (toluene/heptane); δH (CDCl3) 7.13 (s, 2H),
6.63 (s, 4H), 3.84 (d, J ) 14.8, 4H), 3.78, (d, J ) 14.9, 4 H), 2.33-
2.30 (m, 4H), 1.54-1.47 (m, 4H), 1.39-1.29 (m 12 H); δC (CDCl3)
143.1, 137.6, 129.2, 127.1, 39.4, 35.9, 29.9, 28.9, 27.0, 26.7. EI-MS
m/z (%) 411 (24), 410 (100, M+), 346 (39), 158 (62), 145 (31), 119
(54), 105 (33), 91 (41). HRMS: calcd for C26H34S2 410.2102, found
410.2111.
9
6774 J. AM. CHEM. SOC. VOL. 126, NO. 21, 2004

Reduction of Strained Polycycles: Pyrene Anion Strain
A R T I C L E S
6c/2Li+: δH (THF-d8, 273 K, wine-red) 6.19 (d, 1H, H4), 6.17 (d,
1H, H10), 5.98 (d, 1H, H9), 5.75 (s, 1H, H3), 5.66 (s, 1H, H8), 5.58
(s, 1H, H6), 5.49 (d, 1H, H5), 3.24 (s, 1H, JC,H ) 122.63 Hz, H1),
2.44 (m, 1H, H20), 2.37 (m, 1H, H11), 1.87 (m, 1H, H20′), 1.74 (1H,
H11′), 1.58 (H19), 1.50 (H12), 1.17 (H13), 1.17 (H19′), 1.02 (H12′),
0.99 (17), 0.94 (13′), 0.64 (1H, H17′), 0.60-0.40 (m, 6H, H14, H15,
H16, H16′, H18, and H18′), 0.32 (1H, H14′), 0.06 (m, 1H, H15′) ppm.
δC (THF-d8, 298 K) 142.98 (C5a), 140.40 (C8a), 136.77 (C7), 135.53
(C10b), 134.86 (C10c), 128.34 (C4/C10), 128.27 (C4/C10), 126.27
(C3), 111.00 (C10a and C8), 110.47 (C3a), 109.23 (C2), 109.11 (C9),
107.60 (C6), 103.32 (C5), 46.49 (C1), 39.47 (C11), 38.02 (C20), 33.26
(C13), 32.78 (C14), 32.67 (C15), 32.23 (C18), 30.81 (C19), 28.16
(C12), 27.57 (C13), 27.04 (C17) ppm. δLi (THF-d8, 273 K) 0.04 ppm.
6c/2K+: δH (THF-d8, 273 K, violet) 6.26 (d, 1H, H10), 6.24 (d,
1H, H4), 6.05 (d, 1H, H9), 5.80 (s, 1H, H3), 5.66 (s, 1H, H8), 5.58 (s,
1H, H6), 5.51 (d, 1H, H5), 3.21 (s, 1H, JC,H ) 124.44 Hz, H1), 2.40
(m, 1H, H20), 2.39 (m, 1H, H11), 1.89 (m, 1H, H20′), 1.81 (m, 1H,
H11′), 1.61 (m, 1H, H19), 1.48 (m, 1H, H12), 1.19 (m, 1H, H13), 1.16
(m, 1H, H19′), 1.07 (m, 1H, H12′), 1.04 (m, 1H, H17), 0.94 (m, 1H,
H13′), 0.65 (m, 1H, H17′), 0.60-0.40 (m, 6H, H14, H15, H16, H16′,
H18 and H18′), 0.39 (m, 1H, H14′), 0.11 (m, 1H, H15′) ppm. δC (THF-
d8, 273 K) 143.34 (C5a), 140.80 (C8a), 138.59 (C7), 135.67 (C10b),
134.79 (C10c), 129.44 (C10), 128.80 (C4), 126.25 (C3), 112.18 (C2),
111.87 (C10a), 111.31 (C3a), 110.68 (C8), 110.04 (C9), 107.12 (C6),
102.89 (C5), 47.97 (C1), 39.38 (C11), 37.97(C20), 33.44 (C16), 33.03
(C14), 32.81 (C15), 32.40 (C18), 30.83 (C19), 27.96 (C12), 27.62
(C13), 26.20 (C17) ppm.
Acknowledgment. We are grateful to the United States-Israel
Binational Science Foundation (BSF) and the Natural Sciences
and Engineering Research Council (NSERC) of Canada for
financial support.
JA040039U
9
J. AM. CHEM. SOC. VOL. 126, NO. 21, 2004 6775