Novel, terpyrrole-containing, aromatic expanded porphyrins

Article abstract of DOI:10.1016/S0040-4020(01)00243-5

The synthesis and characterization of two expanded porphyrins, namely 18,23-diphenyl-2,3,6,7,10,11-hexamethyl-[26]hexaphyrin(1.1.1.1.0.0) and 18-mesityl-2,3,6,7,10,11-hexamethyl-[22]pentaphyrin(1.1.1.0.0) is described. Their common feature is a terpyrrolic subunit that is connected to a pyrrane building block. Both compounds display aromaticity and are stable in their respective protonated and free-base forms.

Full text of DOI:10.1016/S0040-4020(01)00243-5

TETRAHEDRON
Pergamon
Tetrahedron 57 /2001) 3743±3752
Novel, terpyrrole-containing, aromatic expanded porphyrins
Jonathan L. Sessler,p Daniel Seidel, Christophe Bucher and Vincent Lynch
Department of Chemistry and Biochemistry, Institute for Cellular and Molecular Biology, The University of Texas at Austin,
Austin, TX 78712-1167, USA
Received 18 October 2000; accepted 6 December 2000
AbstractÐThe synthesis and characterization of two expanded porphyrins, namely 18,23-diphenyl-2,3,6,7,10,11-hexamethyl-[26]hexa-
phyrin/1.1.1.1.0.0) and 18-mesityl-2,3,6,7,10,11-hexamethyl-[22]pentaphyrin/1.1.1.0.0) is described. Their common feature is a terpyrrolic
subunit that is connected to a pyrrane building block. Both compounds display aromaticity and are stable in their respective protonated and
free-base forms. q 2001 Elsevier Science Ltd. All rights reserved.
One of the most intriguing aspects of expanded porphyrin
chemistry is the vast structural diversity that is seen within
this general class of compounds.1±3 While porphyrin itself is
remarkably ¯exible, given its size, aromaticity, and pyrrole-
based composition,4 there are nonetheless limits to its ability
to undergo conformational changes. On the other hand,
expanded porphyrins with their attendant larger frameworks
enjoy a greater conformational freedom. As a result, a range
of structural motifs including `®gure eight' and other
twisted forms, have been characterized in recent years.1±3
phyrin/1.0.0.0.1.0.0.0),8 does not exhibit a ®gure eight
type structure. It does, however, deviate strongly from
planarity and is, therefore, best considered as being non-
aromatic.
Recently, even larger oligopyrrolic macrocycles have been
reported with topologies that are anything but planar.9,10
However, one requirement of aromaticity /and antiaromati-
city as well) is `some' degree of planarity.11 Curiously, no
macrocyclic system possessing more than six pyrrolic units
is known that displays aromatic character. Indeed, in spite of
its planarity, [28]heptaphyrin/1.0.0.1.0.0.0) /1), the ®rst
heptaphyrin to be reported in literature,8 was found to be
stable as an antiaromatic system, rather than as an aromatic
one. In fact, many expanded porphyrins containing only ®ve
or six pyrrolic subunits are not aromatic.1±3 In many
instances this is because highly nonplanar conformations
are adopted, with rosarin 2 /systematically: [24]hexa-
phyrin/1.0.1.0.1.0)) representing an extreme example of
such behavior.12,13
Over the course of the last two decades, considerable atten-
tion has been devoted to the synthesis of new oligopyrrolic
systems. In spite of this progress, the factors that in¯uence
nonplanarity and conformational structure remain poorly
understood. Particularly prominent in this context is the
role of aromaticity. Here, one of the complicating factors
is that, in contrast to what is true for porphyrins, many
expanded porphyrins are nonaromatic or even formally anti-
aromatic. Further compounding the problem is the fact that
for all conjugated polypyrrolic macrocycles, numerous
oxidation states can be considered, the large majority of
which contain either /4n12) p-electron or 4n p-electron
peripheries. Finally, as a result of deviations from planarity,
systems with either /4n12) or 4n p-electron peripheries can
become essentially nonaromatic, as judged from their
Ph
N
N
NH
N
N
H
H
N
HN
N
N
HN
H
N
NH
H
N
1
Ph
respective H NMR spectral characteristics. This is true
for systems such as turcasarin /[40]decaphyrin-
Ph
/1.0.1.0.0.1.0.1.0.0)),5
[32]octaphyrin/1.0.1.0.1.0.1.0),6
[36]octaphyrin/2.1.0.1.2.1.0.1]7
and
[34]octaphyrin-
1
2
/1.1.1.0.1.1.1.0),7 macrocycles that have shown to exhibit
`®gure eight' structures in the solid state and in solution.
Another octapyrrolic macrocycle, namely [32]octa-
Quite recently, expanded porphyrins capable of achieving a
planar arrangement by virtue of having one or two pyrrolic
subunits `inverted' /i.e. with their NH moieties `¯ipped
out') have been reported. One of the ®rst systems of this
type to be analyzed in detail is tetraphenylsapphyrin /3).
Keywords: macrocycles; porphyrin; aromaticity.
Corresponding author. Tel.: 11-512-471-5009; fax: 11-512-471-7550;
e-mail: sessler@mail.utexas.edu
p
0040±4020/01/$ - see front matter q 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020/01)00 243-5

3744
J. L. Sessler et al. / Tetrahedron 57 -2001) 3743±3752
While inverted in the free base form, upon protonation, this
macrocycle undergoes a conformational change to adopt a
`normal' structure with all ®ve nitrogens pointing
inward.14,15 More recently, structural evidence was put
forward in support of the contention that hexaphyrin 4
exhibits a bis-inverted structure.16 Interestingly, the original
hexaphyrin /5), reported by Gossauer in 1983,17 was
suggested as adopting a different conformation wherein
the two meso-CH protons are pointing towards the
cavity. This proposal, which would not require the inversion
of pyrrolic subunits bearing large substituents
/RˆCH2CH2CO2CH3 for 5) would nonetheless provide for
an overall planar structure that displays typical aromatic
features, as is indeed seen by experiment. By contrast, inver-
sion is seen in the case of so-called heteroatom analogues of
rubyrin, [26]hexaphyrin/1.1.0.1.1.0). While rubyrin 6 itself,
synthesized with all b-pyrrolic positions substituted,
yielded a solid state structure for its stable bis-protonated
form with all six nitrogens pointing inward,18 hetero-
rubyrins /e.g. 7 with XˆS or Se, respectively) display bis-
inverted conformations under appropriate conditions.19
Another mode of inversion has also been reported very
recently.20 It involves dioxarubyrin 8, a system wherein
two pyrroles within a bipyrrolic subunit are inverted.
we reasoned that assembly of an expanded porphyrin-type
macrocycle, wherein some of the pyrrolic subunits are
b-substituted while others are not, might allow this goal
to be attained. As detailed in a recent communication,21
[26]hexaphyrin/1.1.1.1.0.0) /11), prepared in accord with
such a strategy does indeed display pyrrole inversion at
one speci®c and fully anticipated site.21 Here, we wish to
report full experimental details for this new expanded
porphyrin, including further experimental support for the
proposed inverted structure. We also report the successful
synthesis of its smaller analogue, [22]pentaphyrin/1.1.1.0.0)
/14). Both, 11 and 14 contain a terpyrrolic subunit incorpo-
rated within their macrocyclic frameworks. In both cases,
these fragments are then `¯eshed out' with oligopyrrane
fragments, tripyrrane or dipyromethane, respectively, that,
in turn, were obtained from the condensation of pyrrole with
benzaldehyde under appropriate conditions.22,23
As previously communicated21 and as outlined in Scheme 1,
[26]hexaphyrin/1.1.1.1.0.0) /11) is obtained from the acid
catalyzed condensation of a 1:1 mixture of the diphenyl-
tripyrrane 9 with the diformylhexamethylterpyrrole 10.
Carrying out this condensation in ethanol under conditions
of re¯ux in the presence of air and an excess of para-
toluenesulfonic acid /4 M equiv.), followed by chromato-
graphic workup and treatment with NaOH, provides 11 in
46% yield as the sole isolable product.
C6F5
C6F5
N
H
N
Ph
Ph
NH
N
N
N
HN
C6F5
C6F5
The above success then led us to attempt the synthesis of
[22]pentaphyrin/1.1.1.0.0) /14a). This smaller analog of 11
and an isomer of smaragdyrin, has recently been reported,
but only in a form wherein all the b-pyrrolic sites of the
macrocycle are substituted with alkyl groups.24 By using a
dipyrromethane with pyrroles that lack substituents in the
b-positions, we intended to study whether this particular
pentapyrrolic macrocycle, like its hexapyrrolic congener
11, would be capable of undergoing inversion at one of its
pyrrolic sites, presumably one of the less sterically demand-
ing b-unsubstituted pyrroles. To the extent, this was found
to occur, it would support the contention that the all
b-substituted version, only characterized in the form of its
HCl salt, would not be able to undergo such an inversion
presumably as the result of [22]pentaphyrin/1.1.1.0.0) being
unable to accommodate inward pointing alkyl groups.
H
N
HN
N
Ph
Ph
N
C6F5
C6F5
3
4
R
R
N
H
N
H
N
N
N
N
H
N
NH
R
HN
R
N
H
H
N
H
N
5
6
Initial efforts to obtain [22]pentaphyrin/1.1.1.0.0) /14a)
were made using a synthetic procedure analogous to that
used to prepare 11. Speci®cally, a mixture of diformyl-
hexamethylterpyrrole, 1325 5-phenyldipyrromethane, 1226
and 4 equiv. of para-toluenesulfonic acid was heated at
re¯ux in ethanol for a period of 12 h while being left open
N
Mes
Ph
Ph
Mes
N
N
N
H
H
X
X
O
O
H
H
N
N
N
Ph
Ph
Mes
Mes
N
Ph
Ph
H
N
Ph
Ph
H
N
i)EtOH
HTos
O2
8
7
NH
HN
9
NH
N
HN
N
+
ii)NaOH
OHC
CHO
H
N
Given the fact that the large majority of macrocyclic
systems containing inverted pyrrolic subunits have been
obtained as the result of serendipity, rather than rational
design, we became intrigued by the question of whether it
would be possible to create a system that would be
`inverted' at a particular, pre-chosen pyrrolic site. Here,
NH
HN
H
N
11
10
Scheme 1.

J. L. Sessler et al. / Tetrahedron 57 -2001) 3743±3752
3745
EtOH,
HTos,
O2
NH HN
12
NH HN
+
OHC
CHO
NH
N
H
N
NH
HN
H
N
14a
13
Scheme 2.
to the air. To our surprise, however, we were unable to
isolate any of the desired products under these reaction
conditions /Scheme 2). Accordingly, the approach and
conditions used successfully to synthesize the all-b-substi-
tuted version of [22]pentaphyrin/1.1.1.0.0) were employed.
Speci®cally, as outlined in Scheme 3, 1,9-bisformyl-5-
phenyldipyrromethane /15a),27 hexamethylterpyrrole /16)
and an excess of HCl were reacted under an inert atmos-
phere using chloroform as the solvent, followed by chroma-
tographic workup. In this instance, the desired product,
phenylisosmaragdyrin 14a, could be isolated in 44% yield
in the form of its HCl-salt /H14a1´Cl2). Unfortunately,
H14a1´Cl2 proved to be only poorly soluble in most
common organic solvents, such as dichloromethane, chloro-
form, acetonitrile and methanol. This hampered efforts at
characterization. However, in d7-DMF, satisfactory 1H
NMR spectral data could be obtained for H14a1´Cl2 that
are in agreement with the proposed structure. Additionally,
single crystals of H14a1´Cl2 suitable for X-ray crystallo-
graphic characterization could be obtained by recrystalliza-
tion from DMSO. The resulting structure is shown in Fig. 1.
Figure 1. ORTEP29 views of H14a1´Cl2´2DMSO showing a partial atom
labeling scheme. The hydrogen atoms on the methyl groups have been
omitted for clarity. Of the two DMSO molecules found in the structure,
only one is shown in the side view of H14a1´Cl2´2DMSO. Dashed lines are
indicative of a hydrogen bonding interaction. Thermal ellipsoids are scaled
to the 50% probability level. Hydrogen atoms shown are drawn to an
arbitrary scale.
As illustrated in the top portion of Fig. 1, the protonated
macrocycle H14a1´Cl2 interacts with the chloride counter-
anion that is located above the mean pentaaza plane. There
are three hydrogen bonding interactions, namely between
Ê
Ê
N5H±Cl /2.33/2) A), N4H±Cl /2.28/2) A) and N2H±Cl
Ê
/2.46/2) A). Two molecules of dimethylsulfoxide per
On the basis of its proposed structure, H14a1´Cl2 is
expected to be a /4n12) p-electron system and hence poten-
tially aromatic. Nonetheless, the angles across the meso
positions deviate substantially from 1208; they amount to
133.1 /C17±C18±C19), 131.4 /C22±C23±C1) and 132.18
/C4±C5±C6), respectively, indicating that the molecule
experiences a severe strain. In spite of this, the overall mole-
cule is rather planar as would be expected for an aromatic
system. The main deviation from planarity involves the
middle pyrrole of the terpyrrolic subunit, as indicated by
torsion angles of 20.9 and 22.38 for N2±C9±C10±N3 and
N3±C13±C14±N4, respectively. On the other hand, the
dipyrromethane subunit is rather planar, with torsion angels
of 3.2 /N5±C22±C23±C1) and 12.88 /C22±C23±C1±N1)
being observed.
macrocyclic unit are incorporated into the structure, only
one of which is hydrogen bound to the macrocycle /cf.
side view given in Fig. 1). There are two hydrogen bonding
interactions involving this latter DMSO molecule, namely
Ê
Ê
N1H±O /2.15/2) A) and N3H±O /2.04/3) A).
The structure of H14a1´Cl2 can be compared to that of
H21121´2Cl2 which was previously reported.21 In this latter
instance, pyrrole inversion, something not seen for
H14a1´Cl2, is observed. This leads to the conclusion that
removal of alkyl groups in [22]pentaphyrin/1.1.1.0.0) is not,
in and of itself, suf®cient to induce pyrrole inversion, at least
Scheme 3.

3746
J. L. Sessler et al. / Tetrahedron 57 -2001) 3743±3752
Table 1. Solid state structural parameters for [26]hexaphyrin/1.1.1.1.0.0)
11 and its bis HCl salt, H21121´2Cl2
11
H21121´2Cl2 a
Angles -8)
C22±C23±C24
C27±C28±C1
C4±C5±C6
126.0133.4
122.2
121.7
126.4
120.9
119.8
132.7
C9±C10±C11
Torsion angles -8)
N3±C14±C15±N4
N4±C18±C19±N5
N6±C27±C1±C2
C27±C28±C1±C2
C3±C4±C5±C6
46.041.3
46.5
19.5
5.1
9.6
17.5
34.1
16.8
16.8
15.5
20.6
C4±C5±C6±N2
a
Data taken from Ref. 21.
In order to obtain a more soluble version of [22]penta-
phyrin/1.1.1.0.0) /H14a1´Cl2), we decided to replace the
meso-phenyl group by a meso-mesityl group; it was hoped
that such a replacement would serve to break up the various
p-stacking interactions thought to account for the low
solubility of the macrocycle. The requisite 1,9-bisformyl-
5-mesityldipyrromethane /15b) was obtained from
5-mesityldipyrromethane using the common Vilsmeier-
type conditions employed recently in the synthesis of
1,9-bisformyl-5-phenyldipyrromethane /15a).27 With 15b
in hand, mesityl-isosmaragdyrin /H14b1´Cl2) could readily
be obtained in 38% yield, using the same procedure used to
prepare H14a1´Cl2. As expected, the solubility of
H14b1´Cl2 improved as compared to H14a1´Cl2, with
moderate solubility being observed in chloroform and
dichloromethane. On the other hand, the solubility in
DMSO was found to be reduced. The solubilities are similar
for both the HCl-salt /H14b1´Cl2) and the free base form
/14b), the latter being obtained upon treatment of
H14b1´Cl2 with aqueous NaOH.
Figure 2. ORTEP29 views of 11 showing a partial atom labeling scheme.
The hydrogen atoms on the methyl groups have been omitted for clarity.
Thermal ellipsoids are scaled to the 50% probability level. Hydrogen atoms
shown are drawn to an arbitrary scale.
not at this protonation state and under these solid phase
conditions. Accordingly, other protonation states were
studied, not just in the case of 14a but also 11. In fact, in
the course of the present study, we explicitly succeeded in
growing X-ray diffraction quality single crystals of 11. This
aspect of the chemistry is discussed immediately below.
The UV±Vis spectra of H14b1´Cl2 and 14b are shown in
Fig. 3. The free base 14b displays three strong absorptions,
namely at 391, 458 and 473 nm, that are considered to be
The free base form of [26]hexaphyrin/1.1.1.1.0.0) /11) was
obtained by washing a chloroform solution of H21121´2Cl2
with aqueous NaOH. Single crystals were obtained from
dichloromethane/hexanes. The resulting X-ray structure is
depicted in Fig. 2. One feature of this structure is that 11
crystallizes under these conditions without any additional
solvent being incorporated into the structure, in contradic-
tion to what is commonly seen for similar systems /includ-
ing H21121´2Cl2) that readily incorporate hydrogen bound
and/or `gap-®lling' solvent molecules into their crystal
lattices. The structure of 11 further deviates from that of
H21121´2Cl2 in that a greater degree of distortion for the
middle pyrrole of the terpyrrole subunit is seen in the free
base form /cf. Table 1 for a detailed comparison of struc-
tural parameters). This latter subunit is tilted out of the plane
in both instances; however, this distortion is greater for the
free base form. This might explain the reduced ring current
effect previously observed for 11 as compared to
H21121´2Cl2.21 On the other hand, the overall deviation
from planarity for the tripyrrane subunit is less in the case
of 11 than it was in the case of H21121´2Cl2.
Figure 3. UV±Vis spectra of H14b1´Cl2 /Ð) and 14b /- - -) recorded in
dichloromethane.

J. L. Sessler et al. / Tetrahedron 57 -2001) 3743±3752
3747
indicative of an aromatic ring current effect. Unfortunately,
in this instance, no resonances could be observed for another
set of marker signals whose chemical shift values can
provide evidence of aromaticity or antiaromaticity, namely
the pyrrolic NH protons. Presumably, the inability to
observe these latter signals re¯ects a fast equilibrium
between different putative tautomers of the macrocycle.
Consistent with such a supposition, it is interesting to note
that the two ortho-methyl groups of the mesityl substituents
are magnetically equivalent on the NMR time scale and
therefore cannot be distinguished. They resonate as one
singlet at 1.84 ppm. Since it appears reasonable to assume
that the mesityl substituent is unable to rotate at room
temperature, the observation of a singlet implies either a
very ¯at overall macrocyclic structure or substantial confor-
mational ¯exibility, with the result that the ortho-methyl
groups of the mesityl substituent become chemically
²
equivalent in a time average sense.
Figure 4. UV±Vis spectra of H21121´2Cl2 /Ð) and 11 /- - -) recorded in
dichloromethane.
1
The room temperature H NMR spectrum /CDCl3) of the
protonated macrocycle H14b1´Cl2 was also recorded and
again found to be in accord with not only the proposed
structure but also its /4n12; nˆ5) p-electron aromatic
formulation. For instance, the pyrrolic methyl groups were
found to resonate between 3.75 and 3.84 ppm, while the
expected doublets of the b-pyrrole protons of the dipyrro-
methane subunit were observed as two broad, unresolved
singlets at 8.85 and 9.54 ppm, respectively. Another singlet
at 10.53 ppm was also observed; it is ascribed to the two
meso-like protons. Finally, in contrast to what proved true
for the free base form of the macrocycle 14b, all the
expected NH resonances could be observed. Speci®cally,
three singlets in the ratio of 1:2:2 and integrating to ®ve
protons were seen at 25.42, 22.84 and 22.11 ppm, respec-
tively. Such up®eld shifts for these signals are in accord
with the proposed aromatic formulation and its attendant
ring current effect.
Soret-type bands. Three weaker Q-like bands at 638, 659
and 720nm are also observed. The UV±Vis spectrum of
H14b1´Cl2 is less complex and is characterized by a
molar absorbtivity that is approximately twice as large as
that of 14b. Speci®cally, one sharp Soret-type band is seen
at 447 nm and a single, less intense Q-type band is seen at
868 nm.
To facilitate comparisons, the UV±Vis spectra of 11 and
H21121´2Cl2 are displayed in Fig. 4. The spectra are in
accord with what would be expected for a higher homologue
of 14b. For instance, analogous bands with similar extinc-
tion coef®cients are seen in both series, with the bands for
the larger hexapyrrolic system being red shifted as
compared to the smaller macrocycle. For instance, a split
Soret-type band at 502 and at 531 nm is seen for 11 /vs 391,
458 and 473 nm for 14; vide supra). A single broad Q-type
band at 782 nm is also observed for 11 /vs 638, 659 and
720nm for 14b). The UV±Vis spectrum of H21121´2Cl2 is
also characterized by a split Soret-like band at lmax 533 and
556 nm and two Q-type transitions at 733 and 877 nm,
respectively. Again, these are red shifted as compared to
what is seen for H14b1´Cl2. Such ®ndings, taken in concert,
are consistent with the proposal that all four species, 14, 11,
H14b1´Cl2 and H21121´2Cl2 are aromatic and that the
conjugation pathways in 11 and H21121´2Cl2 are larger
than in their pentapyrrolic congeners.
Besides the observation of NH signals, one noteworthy
difference between the 1H NMR spectra of H14b1´Cl2
and 14b is that in the case of H14b1´Cl2, the two ortho-
methyl groups of the mesityl substituents give rise to two
separate singlets at 1.59 and 2.14 ppm, respectively. This
lack of equivalence provides support for the contention that
not only is the mesityl group unable to rotate but also that
the macrocyle as a whole is locked into a single conforma-
tion. Presumably, this is a consequence of interacting with,
and binding to, the chloride counteranion; this, it is
suggested, serves to stabilize in solution a structure
analogous to that seen in the solid state /cf. Fig. 1), at
least on the NMR time scale.
Further evidence that the new macrocyclic system 14b is
best considered as being aromatic came from H NMR
1
spectral studies. For instance, in the case of the free base
14b, the 1H NMR /CDCl3) spectrum reveals signals at 3.71,
3.79 and 3.90ppm, respectively, that are ascribed to the
three different methyl groups attached to the b-pyrrolic
positions. The down®eld shifts of these resonances are
typical of what is seen for alkyl groups attached to an
aromatic periphery. Two doublets, corresponding to the
b-pyrrole protons of the dipyrromethane subunit, are also
seen at 8.78 and 9.49 ppm, respectively. The meso-like
protons are found to resonate as one singlet at 10.38 ppm,
in accord with what is expected based on symmetry con-
siderations. The down®eld position of this signal is
To date, in contrast to what is seen for the [26]hexa-
phyrin/1.1.1.1.0.0), system 11, there has been no indication
²
In an effort to `visualize' the NH protons of 14b, the spectrum was
recorded in deuterated DMSO. Unfortunately, the very low solubility in
this medium required acquisition times of two hours or longer in order to
obtain a reasonable spectrum. Even under these conditions, a broadening
of all signals was seen. In particular, no signals that could be ascribed to
NH protons were observed, a ®nding consistent with strong interactions
with solvent molecules. Increasing the temperature up to 808C did not
result in any signi®cant sharpening of the various peaks and did not
permit observation of the NH signals.

3748
J. L. Sessler et al. / Tetrahedron 57 -2001) 3743±3752
that either form of 14 /i.e. H14b1´Cl2 or 14b) displays any
tendency to adopt a conformation in which one pyrrolic
subunit is inverted. On the other hand, very interesting
results were obtained when tri¯uoroacetic acid /TFA) was
added stepwise to a solution of free base [22]penta-
phyrin/1.1.1.0.0) /14b) in CDCl3, including possible indica-
tions of pyrrole inversion at very high TFA concentrations.
and 6.85 ppm. One of these latter /that at 6.41 ppm) was not
resolved but was assigned as such based on a COSY spectral
analysis. In any event, the chemical shift of these signals
supports the contention that they do not experience any type
of macrocyclic ring current effect.
In addition to the above, one singlet is found at 6.55 ppm. It
integrates to one proton and is assigned to a meso-like CH
resonance. Another singlet integrating to two protons and
ascribed to a `meso'-CH2-signal is also seen at 4.23 ppm.
The presence of this CH2-group is the result of a second
protonation at the meso position, and accounts in large
measure for the structure of H214b21´2[TFA2] being
assigned as depicted in Scheme 4. Still, in spite of this
meso protonation, compound H214b21´2[TFA2] appears
¯at /or time-averaged ¯at) as judged from the fact that the
two ortho-methyl groups of the mesityl substituent are now
magnetically equivalent; speci®cally, two singlets in the
ratio of 2:1 are observed at 1.92 and 2.36 ppm, respectively.
As expected, these signals show a crosspeak in the COSY
spectrum.
At ®rst, when less than one equivalent of tri¯uoroacetic acid
is added to the solution, the H NMR spectrum re¯ects the
1
coexistence of both the free base and the protonated forms
/14b and H14b1´TFA2). The addition of one full equivalent
of tri¯uoroacetic acid then leads, as expected, to a dis-
appearance of the signals ascribable to the free base form,
and a growing in of spectral features ascribable to
H14b1´TFA2. These latter, importantly, are virtually
identical to those observed for H14b1´Cl2. Next, when a
little bit more than one total equivalent of tri¯uoroacetic
acid is added, extreme line broadening is observed. This is
rationalized in terms of a fast exchange process involving
binding and release of the TFA2 counteranion. This effect
reaches its maximum at the point where a total of two
equivalents of tri¯uoroacetic acid are added to the initial
sample of 14b. Indeed, at this juncture, practically no
signals are observed. The addition of one further equivalent
of tri¯uoroacetic acid /3 equiv. total) resulted in the obser-
vation of some peaks, albeit ones that remained extremely
broad.
When an even larger excess of tri¯uoroacetic acid is added
/i.e. 100 M equiv. or more), the 1H NMR spectrum changes
again. Now, nine different methyl groups are observed,
some of which are shifted to lower ®elds as compared to
their presumed non-aromatic precursor H214b21´2[TFA2].
The signal at 4.23 ppm that was ascribed to the `meso' CH2
group in H214b21´2[TFA2] is no longer present. On the
other hand, one singlet integrating to one proton only, is
observed at 8.2 ppm; it is assigned to a meso-CH proton.
Four doublets, ascribed to the b-pyrrolic protons, are also
observed. These latter signals are shifted down®eld as
compared to H214b21´2[TFA2]. The best rationalization
of these ®ndings involves the generation of a species with
an inverted `meso-carbon' such as 17. For a species like this,
one would expect to observe a signal for the inner meso CH
proton that is shifted very far up®eld as compared to the
outer meso-CH proton. Unfortunately, however, a signal of
this type could not be detected with any degree of certainty.
Nor could signals corresponding to the NH protons be seen.
While these signals may be hidden in the alkyl region, an
inability to detect them unambiguously means that the struc-
ture assigned as 17 must be considered tentative as present.
Once ®ve or more equivalents /up to approximately
14 equiv.) of tri¯uoroacetic acid are added, a spectrum
consistent with the formation of a new species is observed.
On the basis of its NMR spectral features, this new species
has been assigned the structure H214b21´2[TFA2] /Scheme
1
4). Consistent with this assignment, all H NMR chemical
shifts are in accord with a nonaromatic system with little or
no internal symmetry. For instance, six pyrrolic methyl
signals are observed between 2.11 and 2.45 ppm, a region
that implies that they are not experiencing any aromatic ring
current effect. Five singlets between 10and 13 ppm, whose
chemical shift varies slightly depending on the precise
amount of tri¯uoroacetic acid added, are also observed.
These signals, ascribed to NH protons, resonate in a region
that is very typical for `normal' NH protons. Thus, this
®nding also supports strongly the proposed nonaromatic
structure for H214b21´2[TFA2]. Four doublets correspond-
ing to the b-pyrrolic protons are also seen at 6.41, 6.66, 6.68
The spectrum of 14b1TFA gains further complexity when
only slightly more tri¯uoroacetic acid is added to the
F
F
TFA
(5-14 equiv.)
TFA
(1 equiv.)
F
F
F
F
NH HN
NH HN
NH HN
H
NH HN
O
O
O
O
H
NH
N
H
N
2
NH
HN
H
H
N
N
H214b2+ 2[TFA-]
H14b+ TFA-
14b
Scheme 4.

J. L. Sessler et al. / Tetrahedron 57 -2001) 3743±3752
3749
solution in which 17 is considered to be the dominant
species. This added complexity makes it extremely dif®cult
to assign signals in the methyl region. Further, while the
four doublets for the b-pyrrolic protons of 17 are seen to
decrease in intensity, new signals are also seen to appear.
These latter consist of two doublets at around 9.5 ppm and
two doublets at approximately 5 ppm, all equal in intensity.
While highly speculative at this stage of investigation, the
presence of two sets of doublets, one shifted to higher ®eld
and one to lower ®eld, is consistent with a structure such as
18 that has an inverted pyrrolic subunit. Given the rather
small Dd of the two sets of doublets, however, one would
expect 18 to be highly distorted from planarity.
Figure 5. UV±Vis spectra showing the spectral changes observed upon the
addition of TFA to solutions of 14b in dichloromethane. These are consid-
ered to re¯ect the stepwise conversion of 14b into H14b1´TFA2.
To shed further light on the changes seen upon varying the
concentration of tri¯uoroacetic acid, we performed a UV±
Vis titration starting with the free base 14b. The results of
this titration are summarized in Figs. 5 and 6, respectively.
Fig. 5 shows the spectral changes presumed to occur during
stepwise conversion of the free base 14b to the monoproto-
nated species H14b1´TFA2, observed upon the addition of
increasing quantities of TFA. Supporting the assumption of
two species coexisting at equilibrium, an isosbestic point is
observed over the course of this titration. Again, the ®nal
spectrum, assigned to H14b1´TFA2, is virtually identical to
that of H14b1´Cl2.
Figure 6. UV±Vis spectra showing spectral changes observed upon the
addition of TFA to solutions of H14b1´TFA2 in dichloromethane. These
are considered to re¯ect the stepwise conversion of H14b1´TFA2 into
H214b21´2[TFA2].
The solution of H14b1´TFA2 produced as the result of the
above titration also provides the starting point for the addi-
tion of further tri¯uoroacetic acid, as shown in Fig. 6. Under
the conditions of this second titration, the spectrum corre-
sponding to H14b1´TFA2 is seen to evolve gradually into a
different one, speci®cally one considered to re¯ect the
formation of the non-aromatic species H214b21´2[TFA2].
In support of this latter conclusion is the ®nding that
under these conditions, Q-type bands are no longer
observed. Likewise, the intensity of the maximal Soret-
like absorption is considerably decreased as compared to
H14b1´TFA2. This latter band is also considerably
broadened, a ®nding that is very much consistent with the
presence of a nonaromatic species.
Fig. 7 shows the spectrum of 14b measured in pure tri¯uoro-
acetic acid. This spectrum is clearly different from those
assigned to H14b1´TFA2 and H214b21´2[TFA2]. In par-
ticular, it displays a small band around 700 nm that some-
what resembles a Q-type band. Such a band is, of course,
expected were, as is being suggested, an aromatic species
being formed at higher TFA concentrations. Unfortunately,
these UV±Vis experiments, at least at their present level of
Figure 7. UV±Vis spectrum of 14b in pure tri¯uoroacetic acid.

3750
J. L. Sessler et al. / Tetrahedron 57 -2001) 3743±3752
analysis, do not allow an assignment of whether pyrrole
inversion has or has not occurred. On the other hand, they
certainly do not rule it out.
7.90/bs, 4H, phenyl- H), 7.98 /d, JHHˆ4.2 Hz, 2H, b-pyrrole
CH), 8.53 /s, 2H, meso-H), 9.76 /s, 1H, NH); 13C NMR
/125 MHz, d6-DMSO) d [ppm] 10.48, 11.55, 11.86, 110.34,
114.6, 124.16, 125.48, 125.88, 126.66, 127.22, 128.55,
129.77, 132.27, 132.92, 134.37, 138.13, 138.97, 140.23,
143.16, 135.49; UV±Vis /CH2Cl2) lmax [nm] /e in
mol21 L21) 502 /61100), 531 /73400), 782 /20200); HRMS
/CI): m/z 677.3399 /HM1), calcd for C46H41N6 677.3393.
H21121´2Cl2 was obtained by washing a chloroform solution
of 11 with 1 M aqueous HCl /2£10mL). 1H NMR
/500 MHz, d6-DMSO) d [ppm] 21.99 /s, 2H, b-pyrrole
CH), 20.75 /s, 1H, NH), 20.14 /s, 2H, NH), 0.05 /s, 2H,
NH), 3.35 /s, 6H, CH3), 3.73 /s, 6H, CH3), 3.79 /s, 6H, CH3),
7.89±7.92 /m, 2H, phenyl-H), 8.14±8.17 /m, 4H, phenyl-H),
8.93 /bs, 4H, phenyl-H), 9.17 /d, JHHˆ4.3 Hz, 2H, b-pyrrole
CH), 9.84 /d, JHHˆ4.3 Hz, 2H, b-pyrrole CH), 11.35 /s, 2H,
meso-H), 15.03 /s, 1H, NH); 13C NMR /125 MHz, d6-
DMSO) d [ppm] 12.16, 14.04, 14.77, 111.31, 112.34,
118.18, 123.37, 124.92, 125.46, 127.18, 128.52, 128.74,
129.31, 131.43, 131.50, 133.11, 135.43, 137.79, 139.73,
141.33, 144.54; UV±Vis /CH2Cl2) lmax [nm] /e in
mol21 L21) 533 /173000), 556 /156000), 733 /10400), 877
/30200).
In conclusion, the synthesis and characterization of the
terpyrrole-derived pentapyrrolic and hexapyrrolic macro-
cycles 11 and 14, containing aryl-substituted dipyrro-
methane and diaryl substituted tripyrrane subunits,
respectively, have been described. The larger species
display structures with an inverted pyrrole in both its free-
base and diprotonated forms. By contrast, the smaller penta-
pyrrolic species 14 does not appear to contain an inverted
pyrrole, at least in its well characterized free-base and
monoprotonated forms. On the other hand, in the presence
of excess tri¯uoroacetic acid, species are observed that,
depending on the exact concentration of TFA, are either
/i) non-conjugated /H214b21´2[TFA2]) or /ii) aromatic
with inverted `meso' carbons or /iii) pyrrolic subunits /17
and 18, respectively). While de®nite proof for the existence
of these latter two structural variants /i.e. 17 and 18) will
require further study, the available data serves, at the very
least, to underscore the fact that the protonation and confor-
mational behavior of initially aromatic species such as 14b
is very complex.
1.1.2. 1,9-Bisformyl-5-mesityldipyrromethane ,15b). In a
50mL round bottom ¯ask equipped with a re¯ux condenser
and a septum, dry DMF /5 mL) was cooled with stirring for
10min in an external ice water bath under argon. Freshly
distilled POCl3 /1 mL) was then added slowly via syringe.
The ice bath was removed and the reaction mixture was
stirred at room temperature for 20min. The resulting yellow
solution was cooled using an external ice water bath before
addition of 1 g /3.78 mmol) of meso-/mesityl)dipyrro-
methane28 dissolved in 10mL of dry DMF. After addition,
the reaction mixture was heated to 808C for an hour. The
resulting reddish mixture was then allowed to cool to room
temperature at which point saturated aqueous NaOAc
/30mL) was added. The reaction mixture was reheated to
808C for a period of 15 min and subsequently added to
300 mL of ice water. The solid, that precipitated out as
the result of this addition was isolated by ®ltration and
washed with water until the pH of the washings were
1. Experimental
1.1. General
1H and 13C NMR spectra were measured at 258C on a Varian
Unity Plus spectrometer at 300 MHz, on a Varian Unity
Innova at 500MHz, or on a Bruker AC at 250MHz. UV±
Vis spectra were recorded on a Beckman DU-650spectro-
photometer. Low-resolution CI mass spectra were obtained
on a Finnigan MATTSQ 700 mass spectrometer. High-
resolution CI mass spectra were obtained on a VGZAB2-E
mass spectrometer. All solvents and chemicals were
obtained commercially and used as received.
1.1.1. 18,23-Diphenyl-2,3,6,7,10,11-hexamethyl-[26]hexa-
/56 mg
phyrin,1.1.1.1.0.0) ,11). The tripyrrane
9
1
neutral. Yield of 15b: 0.9 g /74%). H NMR /250MHz,
148 mmol) together with 50mg /148 mmol) of the di-
formyl-terpyrrole 10 were suspended in 230mL of ethanol.
The resulting mixture was heated until complete dissolution
of both components occurred. At this juncture, 120mg /ca.
4 M equiv.) of para-toluenesulfonic acid was added, and the
solution heated under re¯ux for an additional 12 h, open to
air. After allowing the solution to cool to room temperature,
the solvent was evaporated off under reduced pressure. The
solid residue was dissolved in dichloromethane and subject to
column chromatography, using silica gel as the stationary
phase and dichloromethane /containing 5% methanol) as
the eluent. The pink-red band was collected and the solvent
removed under reduced pressure. The remaining solid was
redissolved in 50mL chloroform and washed with 1 M
aqueous NaOH /2£10mL). After drying the organic phase
over anhydrous Na2SO4, removal of solvent in vacuo yielded
46 mg /46% yield) of 11 in the form of a dark microcrystal-
CDCl3): d [ppm] 2.04 /s, 6H), 2.26 /s, 3H), 5.93 /s, 1H),
6.09 /m, 2H), 6.86 /s, 2H), 6.88 /m, 2H), 9.23 /s, 2H), 9.55
/bs, 2H); 13C NMR /62.5 MHz, CDCl3): d [ppm] 20.5, 20.7,
38.9, 110.9, 122.0, 130.4, 130.9, 132.0, 137.2, 137.5, 140.3,
178.1; HRMS /CI): m/z 321.1606, calcd for C20H20N2O2
/M11)1: 321.1603.
1.1.3.
18-Phenyl-2,3,6,7,10,11-hexamethyl-[22]penta-
phyrin,1.1.1.0.0) ,14a). In a 500 mL round bottom ¯ask,
1,9-bisformyl-5-phenyldipyrromethane /15a)27 /50mg,
0.18 mmol) and hexamethylterpyrrole 16 /50.5 mg,
0.18 mmol) were dissolved in 300 mL of chloroform.
Argon was bubbled through the resulting solution for
about 15 min. Two drops of concentrated HCl in 2 mL of
ethanol were then added and the reaction mixture stirred
for 10h while maintaining it under a positive argon
atmosphere. The reaction vessel was then opened to the
air and stirring was continued for an additional two hours.
The solvent was then evaporated off under reduced
pressure and the remaining was subjected to puri®cation
1
line powder. H NMR /500 MHz, d6-DMSO) d [ppm] 2.11
/s, 6H, CH3), 2.70/s, 6H, C H3), 2.84 /s, 6H, CH3), 3.52 /d,
JHHˆ1.8 Hz, b-pyrrole CH), 7.26 /d, JHHˆ4.2 Hz, 2H, b-pyr-
role CH), 7.48 /m, 2H, phenyl-H), 7.67 /m, 4H, phenyl-H),

J. L. Sessler et al. / Tetrahedron 57 -2001) 3743±3752
3751
via column chromatography /silica gel; 2% methanol in
dichloromethane eluent). The green fractions were
combined and the volume of solvent was reduced to
approximately 20mL, using a rotary evaporator. The result-
ing solution was placed in a fridge overnight, and the result-
ing green microcrystalline material was ®ltered off and
washed with a small amount of dichloromethane. Drying
under vacuum yielded 44 mg /44% yield) of H14a1´Cl2.
1H NMR /500 MHz, d7-DMF) d [ppm] 22.32 /bs, 3H,
NH), 21.87 /bs, 2H, NH), 3.72 /s, 6H, CH3), 3.78 /bs,
6H, CH3), 3.87 /s, 6H, CH3), 7.93 /bs, 3H, phenyl-H),
8.15±8.5 /bm, 2H, phenyl-H), 8.68 /bs, 2H, b-pyrrole-H),
9.75 /bs, 2H, b-pyrrole-H), 10.79 /bs, 2H, meso-H).
/s, 3H, mesityl-CH3), 2.42 /s, 3H, pyrrole-CH3), 2.45 /s, 3H,
pyrrole-CH3), 4.23 /s, 2H, `meso'-CH2) 6.41 /bs, 1H, b-pyr-
role-CH), 6.55 /s, 1H, meso-H), 6.66 /d, JHHˆ3.7 Hz, 1H,
b-pyrrole-CH), 6.68 /d, 1H, JHHˆ2.8 Hz, b-pyrrole-CH),
6.85 /d, JHHˆ3.7 Hz, 1H, b-pyrrole-CH), 6.95 /s, 2H,
mesityl-H), 10.05 /s, 1H, NH), 10.22 /s, 1H, NH), 10.67
/s, 1H, NH), 11.59 /s, 1H, NH), 12.11 /s, 1H, NH); 13C
NMR /125 MHz, CDCl3, extracted from hmbc and hmqc
projection) d [ppm] 9.09, 9.84, 11.35, 12.28, 13.04, 19.05,
20.93, 25.81, 103.21, 119.37, 120.49, 121.62, 122.00,
124.63, 125.56, 126.50, 127.26, 128.38, 129.89, 130.07,
130.82, 131.58, 132.89, 133.83, 135.14, 137.96, 139.84,
143.22, 144.16, 145.85, 146.42, 147.73.
1.1.4.
18-Mesityl-2,3,6,7,10,11-hexamethyl-[22]penta-
phyrin,1.1.1.0.0) ,14b). Using a procedure identical to
that described above, 1,9-bisformyl-5-mesityldipyrro-
methane /15b) /190mg, 0.59 mmol) was condensed with
hexamethylterpyrrole 16 /167 mg, 0.59 mmol) to produce
14b. In this instance, chromatographic workup /silica gel;
0.5% methanol in dichloromethane, eluent) yielded the
product as a green band. Collection and recrystallization
from dichloromethane/hexanes then provided 135 mg
/38% yield) of a green microcrystalline solid /H14b1´Cl2).
1H NMR NMR /300 MHz, CDCl3) d [ppm] 25.42 /bs, 1H,
NH), 22.84 /bs, 2H, NH), 22.11 /bs, 2H, NH), 1.59 /s, 3H,
mesityl-CH3), 2.14 /s, 3H, mesityl-CH3), 2.73 /s, 3H, mesityl-
CH3), 3.75 /s, 12H, pyrrole-CH3), 3.84 /s, 6H, pyrrole-CH3),
7.37 /s, 1H, mesityl-H), 7.47 /s, 1H, mesityl-H), 8.85 /bs, 2H,
b-pyrrole-CH), 9.54 /bs, 2H, b-pyrrole-CH), 10.53 /bs, 2H,
meso-CH); 13C NMR /75 MHz, CDCl3) d [ppm] 12.15,
14.07, 14.41, 20.25, 21.53, 99.74, 102.86, 120.66, 121.75,
123.23, 124.65, 125.97, 126.86, 128.39, 130.21, 132.13,
137.49, 138.34, 141.31; HRMS /CI): m/z 601.2977 /M1),
calcd for C38H40N5Cl1 601.2972, m/z 566.3282 /M12Cl),
calcd for C38H40N5 566.3284; UV±Vis /CH2Cl2) lmax [nm]
/e in mol21 L21) 447 /150000), 868 /12000). The free base
form, 14b, was obtained by washing a CH2Cl2 solution of the
corresponding HCl-salt H14b1´Cl2 with 1 M aqueous NaOH
/3£). Drying of the resulting solution over anhydrous Na2SO4
and concentration of the solution to a minimal volume on the
rotary evaporator led to crystallization of the free base 14b in
the form of blue crystals, which were collected via vacuum
®ltration. 1H NMR NMR /300 MHz, CDCl3) d [ppm] 1.84 /s,
6H, mesityl-CH3), 2.70/s, 3H, mesityl-C H3), 3.71 /s, 6H,
pyrrole-CH3), 3.79 /s, 6H, pyrrole-CH3), 3.90/s, 6H, pyr-
role-CH3), 7.38 /s, 2H, mesityl-H), 8.78 /d, JHHˆ4.5 Hz,
2H, b-pyrrole-CH), 9.49 /d, JHHˆ4.5 Hz, 2H, b-pyrrole-
CH), 10.38 /s, 2H, meso-CH); 13C NMR /75 MHz, CDCl3)
d [ppm] 12.12, 14.64, 21.03, 22.11, 30.93, 97.16, 101.27,
120.63, 121.49, 122.52, 124.60, 126.00, 126.58, 127.95,
129.82, 131.57, 131.99, 137.08, 138.21, 140.00; HRMS
/CI): m/z 566.3286 /M1), calcd for C38H40N5 566.3284;
UV±Vis /CH2Cl2) lmax [nm] /e in mol21 L21) 391
/50600), 458 /71600), 473 /75300), 638 /17500), 659
/15600), 720 /31600).
Acknowledgements
We are indebted to Dr Ben Shoulders, Steve Sorey and the
staff at the NMR facility at the University of Texas for their
assistance. This work was supported by the National
Science Foundation /grant CHE-9725399 to J. L. S.).
References
1. Sessler, J. L.; Weghorn, S. J. Expanded Contracted and
Isomeric Porphyrins, Vol. 15; Pergamon: New York, 1997.
2. Jasat, A.; Dolphin, D. Chem. Rev. 1997, 97, 2267±2340.
3. Sessler, J. L.; Gebauer, A.; Weghorn, S. J. Expanded
Porphyrins. In The Porphyrin Handbook; Kadish, K. M.,
Smith, K. M., Guilard, R., Eds.; Academic: San Diego,
2000; vol. 2, pp 55±124.
4. Senge, M. O. Highly Substituted Porphyrins. In The Porphyrin
Handbook; Kadish, K. M., Smith, K. M., Guilard, R., Eds.;
Academic: San Diego, 2000; vol. 1, pp 239±347.
5. Sessler, J. L.; Weghorn, S. J.; Lynch, V.; Johnson, M. R.
Angew. Chem., Int. Ed. Engl. 1994, 33, 1509±1512.
È
6. Broring, M.; Jendrny, J.; Zander, L.; Schmickler, H.; Lex, J.;
Wu, Y.-D.; Nendel, M.; Chen, J.; Plattner, D. A.; Houk, K. N.;
Vogel, E. Angew. Chem., Int. Ed. Engl. 1995, 34, 2515±2517.
È
7. Vogel, E.; Broring, M.; Fink, J.; Rosen, D.; Schmickler, H.;
Lex, J.; Chan, K. W. K.; Wu, Y.-D.; Plattner, D. A.; Nendel,
M.; Houk, K. N. Angew. Chem., Int. Ed. Engl. 1995, 34, 2511±
2514.
8. Sessler, J. L.; Seidel, D.; Lynch, V. J. Am. Chem. Soc. 1999,
121, 11257±11258.
9. Setsune, J.-i.; Katakami, Y.; Iizuna, N. J. Am. Chem. Soc.
1999, 121, 8957±8958.
10. Lash, T. D. Angew. Chem., Int. Ed. Engl. 2000, 39, 1763±
1767.
11. Minkin, V. I.; Glukhovtsev, M. N.; Simkin, B. Y. Aromaticity
and Antiaromaticity, Wiley: New York, 1994.
12. Sessler, J. L.; Weghorn, S. J.; Morishima, T.; Rosingana, M.;
Lynch, V.; Lee, V. J. Am. Chem. Soc. 1992, 114, 8306±8307.
13. Lament, B.; Dobkowski, J.; Sessler, J. L.; Weghorn, S. J.;
Waluk, J. Chem. Eur. J. 1999, 5, 3039±3045.
14. Chmielewski, P. J.; Latos-Grazynski, L.; Rachlewicz, K.
Chem. Eur. J. 1995, 1, 68±73.
1.2. Nonaromatic species tentatively assigned as
H214b21´2[TFA2]
15. Rachlewicz, K.; Sprutta, N.; Latos-Grazynski, L.;
Chmielewski, P. J.; Szterenberg, L. J. Chem. Soc., Perkin
Trans. 2 1998, 959±967.
1H NMR /500 MHz, CDCl3) d [ppm] 1.92 /s, 6H, mesityl-
CH3), 2.11 /s, 3H, pyrrole-CH3), 2.27 /s, 3H, pyrrole-CH3),
2.30/s, 3H, pyrrole-C H3), 2.31 /s, 3H, pyrrole-CH3), 2.36
Â
16. Neves, M. G. P. M. S.; Martins, R. M.; Tome, A. C.; Silvestre,

3752
J. L. Sessler et al. / Tetrahedron 57 -2001) 3743±3752
Â
A. J.; Silva, A. M. S.; Felix, V.; Drew, M. G. B.; Cavaleiro,
J. A. S. Chem. Commun. 1999, 385±386.
17. Gossauer, A. Chimia 1983, 37, 341±342.
Â
23. Bucher, C.; Seidel, D.; Lynch, V.; Kral, V.; Sessler, J. L. Org.
Lett. 2000, 2, 3103±3106.
24. Sessler, J. L.; Davis, J. M.; Lynch, V. J. Org. Chem. 1998, 63,
7062±7065.
18. Sessler, J. L.; Morishima, T.; Lynch, V. Angew. Chem., Int.
Ed. Engl. 1991, 30, 977±980.
25. Meyer, S.; Andrioletti, B.; Sessler, J. L.; Lynch, V. J. Org.
Chem. 1998, 63, 6752±6756.
19. Narayanan, S. J.; Sridevi, B.; Chandrashekar, T. K.; Vij, A.;
Roy, R. J. Am. Chem. Soc. 1999, 121, 9053±9068.
20. Narayanan, S. J.; Srinivasan, A.; Sridevi, B.; Chandrashekar,
T. K.; Senge, M. O.; Sugiura, K.-i.; Sakata, Y. Eur. J. Org.
Chem. 2000, 2357±2360.
26. Lee, C.-H.; Lindsey, J. S. Tetrahedron 1994, 50, 11427±
11440.
27. Clarke, O. J.; Boyle, R. W. Tetrahedron Lett. 1998, 39, 7167±
7168.
21. Sessler, J. L.; Seidel, D.; Bucher, C.; Lynch, V. Chem.
Commun. 2000, 1473±1474.
22. BruÈckner, C.; Sternberg, E. D.; Boyle, R. W.; Dolphin, D.
Chem. Commun. 1997, 1689±1690.
28. Littler, B. J.; Miller, M. A.; Hung, C.-H.; Wagner, R. W.;
O'Shea, D. F.; Boyle, P. D.; Lindsey, J. S. J. Org. Chem.
1999, 64, 1391±1396.
29. Farrugia, L. J. J. Appl. Crystallogr. 1997, 30, 565.