Nonlinear optical properties and excited-state dynamics of highly symmetric expanded porphyrins

Article abstract of DOI:10.1021/ja064773k

A strong correlation among calculated Nucleus-Independent Chemical Shift (NICS) values, molecular planarity, and the observed two-photon absorption (TPA) values was found for a series of closely matched expanded porphyrins. The expanded porphyrins in ques

Full text of DOI:10.1021/ja064773k

Published on Web 10/06/2006
Nonlinear Optical Properties and Excited-State Dynamics of
Highly Symmetric Expanded Porphyrins
Zin Seok Yoon,† Jung Ho Kwon,† Min-Chul Yoon,† Mi Kyoung Koh,† Su Bum Noh,†
Jonathan L. Sessler,*,‡ Jeong Tae Lee,‡ Daniel Seidel,‡ Apolonio Aguilar,‡
Soji Shimizu,§ Masaaki Suzuki,§ Atsuhiro Osuka,*,§ and Dongho Kim*,†
Contribution from the Department of Chemistry and Center for Ultrafast Optical Characteristics
Control, Yonsei UniVersity, Seoul 120-749, Korea, Department of Chemistry and Biochemistry,
1 UniVersity Station, A5300, The UniVersity of Texas at Austin, Austin, Texas 78712-0165, and
Department of Chemistry, Graduate School of Science, Kyoto UniVersity, and Core Research for
EVolutional Science and Technology (CREST) of Japan Science and Technology Agency (JST),
Kyoto 606-8502, Japan
Received July 5, 2006; E-mail: dongho@yonsei.ac.kr; sessler@mail.utexas.edu; osuka@kuchem.kyoto-u.ac.jp
Abstract: A strong correlation among calculated Nucleus-Independent Chemical Shift (NICS) values,
molecular planarity, and the observed two-photon absorption (TPA) values was found for a series of closely
matched expanded porphyrins. The expanded porphyrins in question consisted of [26]hexaphyrin, [28]-
hexaphyrin, rubyrin, amethyrin, cyclo[6]pyrrole, cyclo[7]pyrrole, and cyclo[8]pyrrole containing 22, 24, 26,
28, and 30 π-electrons. Two of the systems, [28]hexaphyrin and amethyrin, were considered to be
antiaromatic as judged from a simple application of Hu¨ckel’s [4n + 2] rule. These systems displayed positive
NICS(0) values (+43.5 and +17.1 ppm, respectively) and gave rise to TPA values of 2600 and 3100 GM,
respectively. By contrast, a set of congeners containing 22, 26, and 30 π-electrons (cyclo[n]pyrrole, n ) 6,
7, and 8, respectively) were characterized by a linear correlation between the NICS and TPA values. In
the case of the oligopyrrolic macrocycles containing 26 π-electron systems, a further correlation between
the molecular structure and various markers associated with aromaticity was seen. In particular, a decrease
in the excited state lifetimes and an increase in the TPA values were seen as the flexibility of the systems
increased. Based on the findings presented here, it is proposed that various readily measurable optical
properties, including the two-photon absorption cross-section, can provide a quantitative experimental
measure of aromaticity in macrocyclic π-conjugated systems.
Introduction
theoretical level but also on a practical one because they allow
predictions of new, potentially interesting synthetic targets to
Expanded porphyrins, including inter alia oligopyrrolic mac-
rocycles containing more than four pyrrole rings, have attracted
considerable attention recently because of their potential utility
in applications such as medicine, sensor development, and
catalysts.1-4 These systems also provide a test bed in which
issues of aromaticity in large heteroannulenes may be explored.
Such studies are of fundamental interest, not only on a
be made with greater confidence.5 Unfortunately, to date,
chemical characterizations of expanded porphyrins have been
largely limited to structural studies involving X-ray diffraction
methods, as well as UV-vis, NMR, and MCD spectroscopic
analyses. While this has provided considerable insight into the
(4) (a) Chandrasekhar, V.; Nagendran, S.; Azhakar, R.; Kumar, M. R.;
Srinivasan, A.; Ray, K.; Chandrashekar, T. K.; Madhavaiah, C.; Verma,
S.; Priyakumar, U. D.; Sastry, G. N. J. Am. Chem. Soc. 2005, 127, 2410.
(b) Anand, V. G.; Pushpan, S. K.; Venkatraman, S.; Narayanan, S. J.; Dey,
A.; Chandrashekar, T. K.; Roy, R.; Joshi, B. S.; Deepa, S.; Sastry, G. N.
J. Org. Chem. 2002, 67, 6309. (c) Srinivasan, A.; Kumar, R. M.; Pandian,
R. P.; Mahajan, S.; Pushpan, S. K.; Sridevi, B.; Narayanan, S. J.;
Chandrashekar, T. K. J. Porphyrins Phthalocyanines 1998, 2, 305. (d)
Srinivasan, A.; Reddy, V. M.; Narayanan, S. J.; Sridevi, B.; Pushpan. S.
K.; Ravikumar, M.; Chandrashekar, T. K. Angew. Chem., Int. Ed. Engl.
1997, 36, 2598.
(5) (a) Chen, Z.; Wannere, C. S.; Corminboeuf, C.; Puchta, R.; Schleyer, P. v.
R. Chem. ReV. 2005, 105, 3842. (b) Cyran˜ski, M. K. Chem. ReV. 2005,
105, 3773. (c) Raczyn´ska, E. D.; Kosin´ska, W.; Os´miałowski, B.;
Gawinecki, R. Chem. ReV. 2005, 105, 3561. (d) De Proft, F.; Geerlings, P.
Chem. ReV. 2001, 101, 1451. (e) Balaban, A. T.; Oniciu, D. C.; Katritzky,
A. R. Chem. ReV. 2004, 104, 2777. (f) Cyran˜ski, M. K.; Krygowski, T.
M.; Katritzky, A. R.; Schleyer, P. v. R. J. Org. Chem. 2002, 67, 1333. (g)
Schleyer, P. v. R. Chem. ReV. 2001, 101, 1115. (h) Gomes, J. A. N. F.;
Mallion, R. B. Chem. ReV. 2001, 101, 1349. (i) Katritzky, A. R.; Jug, K.;
Oniciu, D. C. Chem. ReV. 2001, 101, 1421. (j) Mitchell, R. H. Chem. ReV.
2001, 101, 1301. (k) Schleyer, P. v. R.; Maerker, C.; Dransfeld, A.; Jiao,
H.; Hommes, N. J. R. v. E. J. Am. Chem. Soc. 1996, 118, 6317.
† Yonsei University.
‡ The University of Texas at Austin.
§ Kyoto University.
(1) (a) Jasat, A.; Dolphin, D. Chem. ReV. 1997, 97, 2267. (b) Lash, T. D. Angew.
Chem., Int. Ed. 2000, 39, 1763. (c) Furuta, H.; Maeda, H.; Osuka, A. Chem.
Commun. 2002, 1795. (d) Sessler, J. L.; Seidel, D. Angew. Chem., Int. Ed.
2003, 42, 5134. (e) Ghosh, A. Angew. Chem., Int. Ed. 2004, 43, 1918. (f)
Chadrashekar, T. K.; Venkatraman, S. Acc. Chem. Res. 2003, 36, 676.
(2) (a) Sessler, J. L.; Tomat, E.; Mody, T. D.; Lynch, V. M.; Veauthier, J. M.;
Mirsaidov, U.; Markert, J. T. Inorg. Chem. 2005, 44, 2125. (b) Sessler, J.
L.; Maeda, H.; Mizuno, T.; Lynch, V. M.; Furuta, H. J. Am. Chem. Soc.
2002, 124, 13474. (c) Weghorn, S. J.; Sessler, J. L.; Lynch, V.; Baumann,
T. F.; Sibert, J. W. Inorg. Chem. 1996, 35, 1089.
(3) (a) Shin, J.-Y.; Furuta, H.; Yoza, K.; Igarashi, S.; Osuka, A. J. Am. Chem.
Soc. 2001, 123, 7190. (b) Shimizu, S.; Shin, J.-Y.; Furuta, H.; Ismael, R.;
Osuka, A. Angew. Chem., Int. Ed. 2003, 42, 78. (c) Srinivasan, A.; Ishizuka,
T.; Osuka, A.; Furuta, H. J. Am. Chem. Soc. 2003, 125, 878. (d) Mori, S.;
Osuka, A. J. Am. Chem. Soc. 2005, 127, 8030. (e) Hata, H.; Kamimura,
Y.; Shinokubo, H.; Osuka, A. Org. Lett. 2006, 8, 1169. (f) Anand, V. G.;
Saito, S.; Shimizu, S.; Osuka, A. Angew. Chem., Int. Ed. 2005, 44, 7244.
9
14128
J. AM. CHEM. SOC. 2006, 128, 14128-14134
10.1021/ja064773k CCC: $33.50 © 2006 American Chemical Society

Photophysical Properties of Expanded Porphyrins
A R T I C L E S
Chart 1. Molecular Structures of Expanded Porphyrins Considered in This Study
relationship between structure and electronic character, more
sophisticated experimental and theoretical methods are needed
if the issues of aromaticity in expanded porphyrins and,
ultimately, related heteroannulene systems are to be more fully
understood. In this report, a detailed comparison of the
photophysical properties of a series of expanded porphyrins
containing six ([26]hexaphyrin (1), [28]hexaphyrin (2), rubyrin
(3), amethyrin (4), cyclo[6]pyrrole (5)), seven (cyclo[7]pyrrole,
6) and eight (cyclo[8]pyrrole, 7) pyrrole rings is presented. These
systems, chosen because of their relative conformational rigidity,
have been investigated by various time-resolved spectroscopic
techniques, as well as through femtosecond Z-scan measure-
ments. Based on these analyses, we have found that a simple,
first-order relationship exists between the excited-state dynamics
and various calculated and observed molecular features associ-
ated with (anti)aromaticity, including Nucleus-Independent
Chemical Shift (NICS) values, the number of π-electrons,
π-conjugation pathway, and structural planarity. Particularly
noteworthy was the correlation between the calculated NICS
values, obtained from GIAO/6-31G* calculations, and the two-
photon absorption (TPA) coefficients determined using the
femtosecond Z-scan method. Based on these findings, we
suggest that various photophysical properties, most notably the
TPA coefficients, can provide a quantitative experimental
measure of aromaticity in various aromatic and antiaromatic
macrocyclic π-conjugated systems.
sources and detection systems. This has permitted NIR absorp-
tion and emission studies, as well as femtosecond TPA
measurement, to be carried out in the case of several expanded
porphyrins. Indeed, two of our groups have recently applied
these optical methods to the study of a matched pair of
hexapyrrolic expanded porphyrins [26]- and [28]hexaphyrin (1
and 2). This study included determinations of both the photo-
physical and nonlinear optical (NLO) properties, as well as an
analysis of the findings in the context of such easy-to-appreciate
parameters as structural planarity, number of π-electrons, and
aromaticity.6 It also led to the tentative prediction that the TPA
values could be used as a simple experimental marker of
aromaticity. However, it was also recognized that support for
this potentially far-reaching conclusion would require more
extensive experimental and theoretical analyses. We have thus
expanded our study to include a number of oligopyrrolic systems
containing not only six but also seven and eight pyrrole rings,
as well as 22, 24, 26, 28, and 30 π-electrons.
The systems chosen for the present study are shown in Chart
1 and involve representative hexa-, hepta-, and octapyrrolic
systems recently reported in the literature.6-9 Starting from [26]-
hexaphyrin 1,3a,6 these expanded porphyrins may be classified
according to the connectivity of the pyrrole rings. Rubyrin 3
(6) (a) Ahn, T. K.; Kwon, J. H.; Kim, D. Y.; Cho, D. W.; Jeong, D. H.; Kim,
S. K.; Suzuki, M.; Shimizu, S.; Osuka, A.; Kim, D. J. Am. Chem. Soc.
2005, 127, 12856. (b) Neves, M. G. P. M. S.; Martins, R. M.; Tome´, A.
C.; Silvestre, A. J. D.; Silva, A. M. S.; Fe´lix, V.; Drew, M. G. B.; Cavaleiro,
J. A. S. Chem. Commun. 1999, 385.
(7) (a) Shimizu, S.; Taniguchi, R.; Osuka, A. Angew. Chem. 2005, 44, 2225.
(b) Sessler, J. L.; Morishima, T.; Lynch, V. Angew. Chem. 1991, 30, 977.
(8) (a) Sessler, J. L.; Weghorn, S. J.; Hiseada, Y.; Lynch, V. Chem.sEur. J.
1995, 1, 56. (b) Hannah, S.; Seidel, D.; Sessler, J. L.; Lynch, V. Inorg.
Chim. Acta 2001, 317, 211.
(9) Ko¨hler, T.; Seidel, D.; Lynch, V.; Arp, F. O.; Ou, Z.; Kadish, K. M.; Sessler,
J. L. J. Am. Chem. Soc. 2003, 125, 6872.
Although long appreciated as being potentially informative,
detailed optical studies of expanded porphyrins have hitherto
been few in number. One of the key limitations to carrying out
such studies is that many of the systems in question absorb and
emit in the NIR (near-infrared) spectral region. However, recent
advances have led to the development of the needed light
9
J. AM. CHEM. SOC. VOL. 128, NO. 43, 2006 14129

A R T I C L E S
Yoon et al.
from the visible to the NIR region in toluene for 1-3 and in
CH2Cl2 for 4-7. As a general rule, the spectral features were
found to be significantly red-shifted compared to what is seen
in 18 π-electron porphyrins, as would be expected in light of
the enhanced π-conjugation pathways (Figure 1).10
Overall, the absorption spectra exhibit two apparent bands,
presumably originating from intense S0-S2 and relatively
weaker S0-S1 transitions in 1-3 and, conversely, weaker S0-
S2 and relatively strong S0-S1 transitions in 5-7. This leaves
4 as a notable exception; this system exhibits broad and
featureless absorption bands in the range 300-500 nm, which
we consider a diagnostic reflection of the nonaromatic nature
of the molecule. To a first approximation, the fluorescence
spectra appear as the mirror images of the corresponding
absorption spectra. Key absorption and emission features and
the relevant Stokes shifts are summarized in Table 1.
In contrast to the lack of disparity seen in their respective
absorption spectra, notable differences were seen in the fluo-
rescence emission spectra of 1 and 2. Whereas a small Stokes
shift is seen in the emission spectrum of the aromatic hexaphyrin
1, its reduced congener 2 exhibits broad features in its emission
spectrum, with a relatively large Stokes shift also being
observed. We interpret these differences in terms of the presence
of conformational heterogeneities in 2 that arise from geo-
metrical changes induced by electronic relaxation within the
excited state that are greater than those seen in the case of 1.6
Among the hexapyrrolic macrocycles (1-5), only 4 exhibits
significantly broad absorption and fluorescence bands in the
visible region. In this case, there is no clear-cut discrimination
between the S1 and S2 transitions, and a large Stokes shift, the
biggest within the present series of compounds, is seen.
The steady-state absorption and fluorescence spectra of 5-7
show a systematic red shift of the lowest energy bands into the
NIR region as the number of π-electrons increases. The spectral
bandwidths and the Stokes shifts of compounds 5-7 likewise
exhibit a systematic trend in parallel with the increasing number
of π-electrons, a finding that can be interpreted in terms of a
narrower conformational distribution and smaller structural
changes between the ground and excited states in 5 compared
to 6 and 7. The differences in steric hindrance between the
pyrrole rings induced by totally symmetric breathing modes,
coupled with the slight loss in planarity associated with an
increase in the number of pyrrole rings, likely determine the
extent of structural relaxation in 5-7.11
Figure 1. Steady-state absorption and emission spectra of 1-3 in toluene
and 4-7 in CH2Cl2, measured in UV-vis-NIR region. Emission spectra
were obtained by exciting 1-3 and 5-7 with a CW He-Cd laser at 442
nm in NIR region. The emission spectrum of 4 was obtained via excitation
at several wavelengths with commercial spectrophotometer in vis region.
(formally [26]hexaphyrin(1.1.0.1.1.0)),7 for instance, can be
considered to be an analogue of [26]hexaphyrin(1.1.1.1.1.1)
from which two meso-bridged carbons have been removed.
Likewise, amethyrin 4 (formally [24]hexaphyrin(1.0.0.1.0.0))8
can be formed by “extracting” two additional meso-bridged
carbons from rubyrin, whereas cyclo[6]pyrrole 5 (formally [22]-
hexaphyrin(0.0.0.0.0.0)),9 results from a process wherein the
last two meso-bridged carbons in amethyrin are “removed”.
Cyclo[7]pyrrole 6 (formally [26]heptaphyrin(0.0.0.0.0.0.0)) and
cyclo[8]pyrrole 7 (formally [30]octaphyrin(0.0.0.0.0.0.0.0)) are
homologues of cyclo[6]pyrrole that likewise contain no bridging
meso carbon atoms. While known for several years, to the best
of our knowledge, in no cases has the photophysics of these
systems been investigated in detail. A first goal of this work
was thus to carry out such a study, while a second was to
correlate the putative nonlinear optical (NLO) properties of these
various expanded porphyrins with such key parameters as
aromaticity/antiaromaticity and structural planarity as inferred
from the corresponding optimized molecular geometries. In
accord with what was inferred from our previous study of 1
and 2, a strong correlation between electronic structure and
excited-state optical properties was indeed found.
Geometry Optimization and Quantitative Analysis of
Aromaticity. NICS (Nucleus-Independent Chemical Shift)
values are a well-known parameter used to evaluate aromaticity
that is recognized for its simplicity and efficiency.5 Although
quantitative measures of aromaticity for many five- and six-
membered ring systems have been documented using this
approach, corresponding NICS studies for larger π-conjugation
systems, such as porphyrins, are scarce.12 Moreover, the use of
NICSs to quantify the putative aromaticity of expanded por-
phyrins has not, to the best of our knowledge, been previously
attempted. Given this, we have calculated the NICS values for
Results and Discussion
(10) Yoon, M.-C.; Jeong, D. H.; Cho, S.; Kim, D.; Rhee, H.; Joo, T. J. Chem.
Phys. 2003, 118, 164.
Steady-State Absorption and Fluorescence Spectra. The
steady-state absorption and fluorescence spectra of a series of
expanded porphyrins, specifically 1-7 (Figure 1), were recorded
(11) Cho, H. S.; Kim, S. K.; Kim, D.; Fujiwara, K.; Komatsu, K. J. Phys. Chem.
A 2000, 104, 9666.
(12) (a) Steiner, E.; Fowler, P. W. Org. Biomol. Chem. 2004, 2, 34. (b) Furuta,
H.; Maeda, H.; Osuka, A. J. Org. Chem. 2001, 66, 8563.
9
14130 J. AM. CHEM. SOC. VOL. 128, NO. 43, 2006

Photophysical Properties of Expanded Porphyrins
A R T I C L E S
Table 1. Number of π-electrons,a Band Maxima of Absorption and Fluorescence Spectra,b-d Stokes Shifts,e NICS(0) Values,f Aspect Ratios
(AR),g TPA Cross-Section Values (σ(2)),h and Singlet and Triplet Excited-State Lifetimesi,j of 1-7
b,c
e
i
j
no. of
λabs
λfld
(nm)
∆
EStokes
NICSf
(ppm)
σ(2)h
(GM)
τS
τT
a
1
molecules
π
e-
(nm)
(cm-
)
ARg
(ps)
(µs)
1
2
3
4
5
6
7
[26]hexaphyrin
[28]hexaphyrin
rubyrin
(1.1.1.1.1.1)
(1.1.1.1.1.1)
(1.1.0.1.1.0)
(1.0.0.1.0.0)
(0.0.0.0.0.0)
(0.0.0.0.0.0.0)
(0.0.0.0.0.0.0.0)
26
28
26
24
22
26
30
567,b 1018c
594,b 1009c
543,b 926c
489,b 598c
398,b 794c
430,b 940c
427,b 1155c
1036
1040
968
632
842
171
-14.3
+43.5
-15.8
+17.1
-11.4
-11.6
-12.3
1.44
1.44
1.23
1.17
1
9890s
2600s
9080h
3100i
2100j
2350k
3030k
100s
26s
1.8s
295
469
900
718
1204
1974
0.04s
560l
110m
430n
146o
p
amethyrin
-
cyclo[6]pyrrole
cyclo[7]pyrrole
cyclo[8]pyrrole
16q
4r
1060
1496
1
1
6.6n
0.7n
p
-
a Number of π-electrons. b Absorption wavelength maxima at B-like bands. c Absorption wavelength maxima at Q-like bands. d Fluorescence wavelength
maxima. e Stokes shifts, that is, energy gaps between c and d. f Calculated at the ring center, NICS(0). g Calculated from optimized geometry. h Excitation
at 1050 nm. i Excitation at 800 nm. j Excitation at 1550 nm. k Excitation at 1600 nm. l Pumped and probed at 550 and 650 nm, respectively. mMeasured by
fluorescence decay using TCSPC. Amethyrin in CH2Cl2 was excited at 410 nm, and the fluorescence was monitored at several wavelengths from 580 to 630
nm. n Pumped and probed at 400 and 570 nm, respectively. o Pumped and probed at 532 and 630 nm, respectively. p Not detected. q Pumped and probed at
800 and 550 nm, respectively. r Pumped and probed at 900 and 550 nm, respectively. s Reference 6a.
1-7 at the ring center, NICS(0), from the energy minimized
molecular structures using a Gaussian ab initio method at the
B3LYP level, employing the GIAO (Gauge-Including Atomic
Orbital) with the 6-31G* basis set. In these calculations, all
peripheral substituents were ignored. This simplified the task
of assigning the number of π-electrons participating in the
π-conjugation pathways giving rise to potentially observable
magnetic shielding/deshielding effects.
particular, it is suggested that 2 may be nonaromatic in character
as judged by the short-range shielding contributions of protons
yet nonetheless be antiaromatic when the full long-range
contribution of the delocalized π-electrons (manifest in ring
current effects) is taken into consideration.13 Although we have
not yet clarified the reason [28]hexaphyrin shows spectral
features that make it best designated as nonaromatic, such a
finding could reflect its structural flexibility. Consistent with
this conclusion is the finding that [28]hexaphyrins that are
constrained as conformationally rigid (e.g., via formation of the
corresponding Au(III) complexes) display distinctly antiaromatic
character.3d
A different feature of the NICS calculations is revealed via
a comparison between 1 and 3. Although both macrocycles
contain the same number of π-electrons (26), 3 is actually
smaller than 1 (in terms of ring size) and displays a slightly
larger NICS value. Such a finding leads to the conclusion that
the NICS value is only weakly dependent on ring size and is
better viewed as being a gauge of net aromaticity. Consistent
with this latter conclusion is the finding that the NICS parameter
for 6, also a net 26 π-electron system, is smaller than that for
3.
As shown in Table 1, the calculated NICS(0) values are
negative for the presumed-to-be-aromatic compounds (1, 3, 5,
6, and 7) and positive for the putative antiaromatic species (2
and 4). Such findings are fully consistent with the aromaticity/
antiaromaticity predictions that would be obtained by extending
Hu¨ckel’s original [4n + 2] rule to encompass large hetero-
annulene systems, such as 1-7. In this context it is to be noted
that a strict counting of the total number of π-electrons is not
always sufficient for determining the predicted Hu¨ckel aroma-
ticity of a system because often there is a superposition of
several pathways.13 On the other hand, the presence of a
conjugated [4n + 2] π-electron periphery provides a remarkably
good, albeit qualitative, predictor of aromaticity in oligopyrrolic
macrocyclic systems.
Two-Photon Absorption (TPA) Properties. Two-photon
absorption (TPA) is a nonlinear optical process, occurring under
an intense electric field usually generated by pulse lasers with
simultaneous absorption of two or more photons by a single
chromophore to produce a high-energy excited state. The TPA
cross-section thus provides a measure of a third-order nonlinear
optical process produced in response to an external electric field.
Within the context of such general considerations, the number
of π-electrons and/or the molecular geometry associated with
the static and dynamic polarizability have been established as
determining factors in controlling nonlinear optical susceptibil-
ity, at least within simple molecular systems.14
Given the defining roles of molecular electronics and structure
in regulating NLO effects, expanded porphyrins are of particular
interest since both of these key parameters can, at least in
principle, be easily controlled. Thus, with the matched set of
compounds 1-7 in hand, we performed femtosecond Z-scan
experiments to probe the nature and extent of the presumed
nonlinear optical response (Supporting Information). The excita-
The formal addition of 2 molar equiv of hydrogen into 1 (to
produce 2) leads to a reversal in the sign of the NICS values
from negative (aromaticity) to positive (antiaromaticity). This
is as expected, given the change in the number of π-electrons
in the conjugation pathway (26 and 28). However, in the case
of 2, the experimental chemical shift values, as determined by
1H NMR spectroscopy, reveal that 2 is best considered as being
nonaromatic, rather than antiaromatic. In a sense, the results
obtained from the 1H NMR spectroscopic analyses and the NICS
calculations contradict each other. However, it is important to
note that the magnetic deshielding predicted by the NICS
calculation reflects the ring current produced by the full
complement of π-electrons delocalized over the entire π-con-
jugation pathway. Conversely, the chemical shift of a given
signal within the 1H NMR spectrum depends on the local
electronic environment for the proton or protons in question.
In this context it is interesting to consider that individual protons
in 2, although ostensibly nonaromatic (as inferred from their
respective chemical shift values), could nonetheless be contained
within a system displaying a net antiaromatic ring current. In
(14) (a) Lee, S.; Thomas, K. R. J.; Thayumanavan, S.; Bardeen, C. J. J. Phys.
Chem. A 2005, 109, 9767. (b) Day, P. N.; Nguyen, K. A.; Pachter, R. J.
Phys. Chem. A 2005, 109, 1803. (c) Yang, W. J.; Kim, D. Y.; Kim, C. H.;
Jeong, M.-Y.; Lee, S. K.; Jeon, S.-J.; Cho, B. R. Org. Lett. 2004, 6, 1389.
(13) (a) Aihara, J. J. Am. Chem. Soc. 2006, 128, 2873. (b) Juse´lius, J.; Sundholm,
D. Phys. Chem. Chem. Phys. 2000, 2, 2145.
9
J. AM. CHEM. SOC. VOL. 128, NO. 43, 2006 14131

A R T I C L E S
Yoon et al.
tion wavelengths for the open aperture experiments were in the
NIR region (see the footnotes in Table 1); this is a spectral
region where the contribution from linear absorption effects is
nearly negligible. In our previous study, we compared the TPA
values of 1 and 2 and explained that the smaller TPA value
seen for 2 reflects its decreased molecular planarity, as well as
its reduced aromatic character, as determined by the number of
π-electrons (Hu¨ckel’s [4n + 2] rule) present in the system.6a
TPA values increase with increasing polarizability within a given
series of molecules, as well as the well-established linear
relationship between polarizability and softness, leads us to
interpret the larger TPA values seen for 7 relative to 6 and 5 in
terms of Koopmans’ theorem,18
ꢀL - ꢀH
η )
(2)
2
In our current studies, we found that 3 exhibits a TPA value
that is similar to 1 (both 26 π-electron systems) while compound
4 (a 24 π-electron system) is characterized by a TPA that is
much smaller than that displayed by either 1 or 3. This finding
is consistent with previous investigations involving the NLO
properties of smaragdyrin analogues by Chandrashekar et al.;
here, a larger TPA value was seen in the para-linked isomer
than in the corresponding meta-linked system, a finding that
was attributed to the lack of complete π-conjugation in the meta
isomer.15 An important ancillary conclusion that could be drawn
from this prior work is that the extent of effective π-conjugation
is strongly correlated with the molecular geometry. This then
provides a means for interpreting what is seen in 4. In particular,
we suggest that the TPA values, and hence optical nonlinearity,
should depend on the aspect ratio (A.R., long-axis/short-axis),
which in turn can be estimated from the optimized molecular
geometry. In the case of the hexapyrrolic systems being
considered in this study, consecutive synthetic “removal” of a
pair of meso-carbons on passing from 1 (A.R. ) 1.44) to 3
(A.R. ) 1.23) and then to 4 (A.R. ) 1.17) should give rise to
TPA values that decrease in sequence such that 1 > 3 > 4.
Consistent with this postulate is the finding discussed in a
previous report that the TPA values of fused zinc(II) porphyrin
arrays (TN) with “fully conjugated” π-electron systems increase
as the π-conjugation pathway becomes elongated.16 Although
this observation most likely reflects an enhanced π-conjugation
effect, the molecular geometry, as quantified in the aspect ratio,
can also be an important factor in determining TPA values. In
fact, oligoporphyrins linked within a squarelike arrangement to
provide a two-dimensional π-extended array (referred to as a
porphyrin “sheet”) invariably display a small (i.e., near-unity)
aspect ratio. As a consequence the TPA value of the porphyrin
sheet is much smaller than that of the corresponding linear fused
arrays.17
where η is the hardness, and ꢀL and ꢀH are the lowest unoccupied
and highest occupied molecular orbital energies, respectively.
As illustrated by the smaller HOMO-LUMO energy gap
observed in the steady-state absorption spectra and molecular
orbital calculations, the hardness of 7 is smaller than that of 5
or 6. We therefore suggest that the larger TPA value seen in 7
compared to 5 and 6 arises from the greater softness, which
results in an increased polarizability in the case of 7.
To summarize the above discussion, it is proposed that both
features of (1) aromaticity and (2) the specifics of the π-con-
jugation pathway, as determined by molecular shape, play
important roles in determining the TPA values.
Support for the first effect (aromaticity) comes from a
comparison between the aromatic system 1 and its reduced,
antiaromatic congener 2, as well as by a similar comparison
between the aromatic 3 and the antiaromatic 4. In both cases,
the less aromatic system, as judged from an analysis of the
number of π-electrons and supporting NICS calculations,
displayed the lower TPA values. While not strictly analogous
to these “pairs”, the fact that the highly antiaromatic cyclo-
octatetraene core (NICS(0) ) +61.7) found in the center of
the hitherto reported porphyrin sheet gives rise to low TPA
values further supports this generalization.
Support for the second effect (specifics of the molecular
structure) comes from an analysis of the matched set consisting
of 1, 3, and 6. These three macrocycles have the same number
of π-electrons (26), but macrocycle 1 has the highest aspect
ratio and displays the largest TPA value within the series. The
fact that a larger TPA value was seen for the fused porphyrin
linear tetramer T4, as compared to the porphyrin sheet, in
previous work, also lends credence to the conclusion that the
aspect ratio and hence structure correlate with TPA values.
Similar support comes from the work of Anderson et al., wherein
a higher TPA value was seen in the case of various π-conjugated
zinc(II) porphyrin dimers linked via the para position of an
intervening spacer, rather through a meta substituted link.19
Singlet and Triplet Excited-State Lifetimes. To explore the
excited singlet and triplet state properties of 1-7 and to obtain
insights into the corresponding energy relaxation processes,
femtosecond and nanosecond transient absorption experiments
were performed. These studies were carried out at room
temperature in toluene solution in the case of 1-3 and in CH2-
Cl2 solution for 4-7 (Supporting Information). The resulting
decays were fit to give the parameters summarized in Table 1.
In our previous study, involving an analysis of [26] and [28]-
hexaphyrins 1 and 2, we interpreted the singlet and triplet
excited-state lifetimes using conformational dynamics as re-
In the present study we found that the antiaromatic systems
2 and 4 show smaller TPA values than 1 and 3, even though
the aspect ratio of 2 is roughly the same as that of 1, and that
of 4 is slightly smaller than that of 3. We think that this is
because molecules tend to avoid antiaromatic situations (28 and
24 π-electrons for 2 and 4, respectively) by adopting, in the
present cases, distorted “tub” structures. These latter systems
then give rise to small TPA values as a result of the reduced
planarity.
Furthermore, as the overall ring size increases, i.e., upon going
from 5 to 7, the TPA values also increase. Considering that
(15) Misra, R.; Kumar, R.; Chandrashekar, T. K.; Nag, A.; Goswami, D. Org.
Lett. 2006, 8, 629.
(16) (a) Ahn, T. K.; Kim, K. S.; Kim, D. Y.; Noh, S. B.; Aratani, N.; Ikeda, C.;
Osuka, A.; Kim, D. J. Am. Chem. Soc. 2006, 128, 1700. (b) Kim, D. Y.;
Ahn, T. K.; Kwon, J. H.; Kim, D.; Ikeue, T.; Aratani, N.; Osuka, A.;
Shigeiwa, M.; Maeda, S. J. Phys. Chem. A 2005, 109, 2996.
(17) Nakamura, Y.; Aratani, N.; Shinokubo, H.; Takagi, A.; Kawai, T.;
Matsumoto, T.; Yoon, Z. S.; Kim, D. Y.; Ahn, T. K.; Kim, D.; Muranaka,
A.; Kobayashi, N.; Osuka, A. J. Am. Chem. Soc. 2006, 128, 4119.
(18) (a) Padmanabhan, J.; Parthasarathi, R.; Subramanian, V.; Chattaraj, P. K.
J. Phys. Chem. A 2005, 109, 11043. (b) Chattaraj, P. K.; Fuentealba, P.;
Gomez, B.; Contreras, R. J. Am. Chem. Soc. 2000, 122, 348.
(19) Drobizhev, M.; Stepanencko, Y.; Dzenis, Y.; Karotki, A.; Rebane, A.;
Taylor, P. N.; Anderson, H. L. J. Phys. Chem. B 2005, 109, 7223.
9
14132 J. AM. CHEM. SOC. VOL. 128, NO. 43, 2006

Photophysical Properties of Expanded Porphyrins
A R T I C L E S
vealed by temperature-dependent femtosecond transient absorp-
tion experiments.6a The shorter excited-state lifetime of 2 relative
to 1 was thus attributed to the perturbed electronic structures
induced by antiaromaticity resulting from the 4n π-electron
configuration of 2. In other words, the excited-state dynamics
could be correlated with the aromaticity/antiaromaticity and thus
related to molecular geometry.
angles” and the beta-pyrrolic substituents without undergoing
some slight structural deformation (e.g., deviation from planar-
ity).
Overall, by comparing the singlet and triplet excited-state
lifetimes in 1-7, we are able to note several systematic
tendencies that are useful in terms of rationalizing and predicting
the excited state lifetimes. First, the nature of the connectivity
is important in the case of the hexapyrrolic systems, and second,
within an analogous series of aromatic expanded porphyrins,
the number of pyrroles, as well as the way they are connected
to one another, is important. As a result, the differences between
ostensibly similar macrocycles can be profound.
In the present study, we have found that “removing” two
meso-bridged carbons from 1 (1.1.1.1.1.1) to make 3 (1.1.0.1.1.0)
leads to a significant increase in both the singlet and triplet
excited-state lifetimes, even though both compounds have the
same number of π-electrons (26) in their most favorable
conjugation pathway. We interpret this result in terms of the
rigid and stabilized molecular structure of 3, which should make
it the most conformationally favored among the hexapyrrolic
systems included in this study. In other words, the longer
excited-state lifetimes of 3 (relative to 1) support the notion
that Hu¨ckel-type [4n + 2] aromaticity is going to be manifest
to a greater extent within those systems that can adopt easily
rigid, planar structures without having to pay an excessive
energetic price in terms of steric distortion.
Conclusions
In this work, we have explored the nonlinear optical properties
and excited-state dynamics of various expanded porphyrins as
an extension of our previous comparative study of [26] and [28]-
hexaphyrins. Based on the steady-state absorption and fluores-
cence spectral features observed in the UV-NIR region and
the results of quantitative analyses of aromaticity using NICS
calculations, the excited-state dynamics of expanded porphyrins
3-7 can be predicted qualitatively from the corresponding
molecular structures as defined by the number of bridging meso-
carbon atoms. Using femtosecond Z-scan methods, the nonlinear
optical properties of this matched set of congeners was also
investigated, from which strategies to enhance TPA values were
inferred. In particular, it is predicted that enhanced nonlinear
optical properties will be seen when (i) the effective π-conjuga-
tion is accompanied by a larger number of π-electrons as long
as Hu¨ckel’s generalized [4n + 2] rule for aromaticity is satisfied,
(ii) the specific molecules in question are flat and rigid, and
(iii) the molecular structures are best described as rectangular,
rather than square or circular, so as to increase the polarizability
along the long molecular axis.
Within the assumptions inherent in Hu¨ckel’s [4n + 2] rule,
NICS calculations can be used to determine aromaticity or
antiaromaticity in a quantitative manner according to the number
of π-electrons. However, such assessments are strictly theoreti-
cal. On the other hand, the combination of experimental
femtosecond Z-scan measurements coupled with NICS calcula-
tions leads us to suggest that, at least within an appropriately
matched set of molecular structures, the experimentally derived
TPA values can be regarded as an empirical measure of
aromaticity. To the extent this is true, the parameter could
emerge as one that could complement others, such as ring
current effects and UV-vis spectroscopic features (λmax, ꢀ),
which are currently being used for this purpose. Accordingly,
further investigations into the relationship between NLO proper-
ties and aromaticity are now under way.
The relative stability of 3 becomes further apparent when
comparisons are made to its congener 4 (a 24 π-electron system)
from which two meso-bridge carbon atoms have been elimi-
nated. The resulting system compared to 3 exhibits a shorter S1
state lifetime, a result that is most readily interpreted in terms
of structural deformation induced by the antiaromaticity present
in 4. Consistent with this was the finding that we were unable
to detect a T-T absorption signal in the case of 4; presumably,
the structural deformation induced so as to mitigate the effects
of Hu¨ckel-type antiaromaticity effects gives rise to a shorter
excited-state lifetime. This putative structural deformation was
also reflected in the absorption and fluorescence spectra, wherein
broad spectral features and a larger Stokes shift are observed,
along with a positive NICS value.
Interestingly, as two additional meso-bridging carbon atoms
are removed (i.e., on passing from 4 to the [22]hexaphyrin-
(0.0.0.0.0.0) 5), longer singlet and triplet state lifetimes are once
again observed; this is rationalized in terms of the “reappear-
ance” of aromaticity (5 is a 22 π-electron system) and the
associated structural stabilization.
As the number of pyrrole rings increases, as it does within
the matched set 5, 6, and 7, the singlet and triplet excited-state
lifetimes decrease. Decreasing band gap energies between the
HOMO and LUMO levels are seen. Likewise, increasing Stokes
shifts are observed. Furthermore, the singlet excited state of 7
proved exceedingly short. In fact, it proved too short-lived (τs
) 0.7 ps) to populate the triplet excited state. Thus, no triplet
state decay processes could be monitored in the case of 7. Taken
in concert, these findings are thought to reflect the increased
conformational flexibility expected upon going from 5 to 7, as
well as the presence of extended π-conjugation pathways. In
this context, it is interesting to note that the compounds 1, 3,
and 6 have the same number of π-electrons (26), resulting in
similar band gap energies. Nonetheless, the singlet excited-state
lifetime of 6 is considerably shorter than those of 1 and 3. Again,
this is interpreted in terms of steric constraints within a system
that is “too small” to accommodate both the intraring “bite
Experimental Section
Steady-State Absorption and Emission. Steady-state absorption
spectra were acquired using a UV-vis-NIR spectrometer (Varian,
Cary5000). Steady-state fluorescence spectra were recorded on a
fluorescence spectrometer (Hitachi, FL2500). For the observation of
near-infrared (NIR) emission spectra, a photomultiplier (Hamamatsu,
R5108) and a lock-in amplifier (EG&G, 5210), combined with a
chopper after laser excitation at 442 nm from a CW He-Cd laser
(Melles Griot, OmNichrome 74) were used.
Time-Correlated Single Photon Counting. Time-resolved fluo-
rescence was detected using a time-correlated single photon counting
9
J. AM. CHEM. SOC. VOL. 128, NO. 43, 2006 14133

A R T I C L E S
Yoon et al.
(TCSPC) technique. A homemade cavity dumped Ti:Sapphire oscillator
pumped by a CW Nd-YVO4 laser (Coherent, Verdi) was used as the
excitation light source; this provided ultrashort pulses (100 fs at full
width half-maximum) and allowed for a high repetition rate (200-
400 kHz). The output pulse of the oscillator was frequency-doubled
with a second harmonic crystal. The TCSPC detection system consisted
of a multichannel plate photomultiplier (Hamamatsu, R3809U-51) with
a cooler (Hamamatsu, C4878), a TAC (time-to-amplitude converter)
(EG&G Ortec, 457), two discriminators (EG&G Ortec, 584 (signal)
and Canberra, 2126 (trigger)), and two wideband amplifiers (Philip
Scientific (signal) and a Mini Circuit (trigger)). A personal computer
with a multichannel analyzer (Canberra, PCA3) was used for data
storage and processing. The overall instrumental response function was
about 60 ps (fwhm). A sheet polarizer, set at an angle complementary
to the magic angle (54.7°), was placed in the fluorescence collection
system. The decay fittings were made by using a least-squares
deconvolution fitting process (LIFETIME program with an iterative
nonlinear least-squares deconvolution procedure developed at the
University of Pennsylvania).20
were made to remove oxygen rigorously by repeated freeze-pump-
thaw cycles. To ensure reproducibility and reliability, the triplet state
dynamics of a known standard, Zn(II)TPP (TPP ) tetraphenylporphy-
rin), were first recorded in toluene under unaerobic conditions, giving
a lifetime of 1 ms at room temperature. Because the concentration of
molecules also affects significantly the excited triplet state lifetime (as
the result of, e.g., triplet-triplet annihilation processes), efforts were
also made to keep the concentration of the samples at or below 10-5
M. A relatively low photoexcitation density at 532 nm (produced by
the second harmonic output of a Q-switched Nd:YAG laser) was also
used.
Two-Photon Absorption Cross-Section (σ(2)). The TPA experiments
were performed using the open-aperture Z-scan method with 130 fs
pulses from an optical parametric amplifier (Light Conversion, TOPAS)
operating at a 5 kHz repetition rate using a Ti:sapphire regenerative
amplifier system (Spectra-Physics, Hurricane). The laser beam was
divided into two parts. One was monitored by a Ge/PN photodiode
(New Focus) as the intensity reference, and the other was used for the
transmittance studies. After passing through an f ) 10 cm lens, the
laser beam was focused and passed through a quartz cell. The position
of the sample cell could be varied along the laser-beam direction (z-
axis), so the local power density within the sample cell could be changed
under a constant laser power level. The thickness of the cell is 1 mm.
The transmitted laser beam from the sample cell was then probed using
the same photodiode as used for reference monitoring. The on-axis
peak intensity of the incident pulses at the focal point, I0, ranged from
40 to 60 GW/cm. Assuming a Gaussian beam profile, the nonlinear
absorption coefficient â can be obtained by curve fitting to the observed
open aperture traces with the following equation:
Femtosecond Transient Absorption. Measurements were made
using a setup involving a dual-beam femtosecond time-resolved
transient absorption spectrometer consisted of two independently tunable
homemade optical parametric amplifiers (OPAs) pumped by a Ti:
sapphire regenerative amplifier system (Spectra-Physics, Hurricane X)
operating at 5 kHz repetition rate and an optical detection system. The
OPAs were based on noncollinearly phase-matching geometry and
easily color-tuned by controlling the delay between the white light
continuum pulses and pump pulses. The generated visible OPA pulses
had a pulse width of ∼35 fs and an average power of 10 mW at a 5
kHz repetition rate in the range 500-700 nm. The probe beam was
split into two parts. One part of the probe beam was overlapped with
the pump beam at the sample to monitor the transient (signal), while
the other part of the probe beam was passed through the sample without
overlapping the pump beam to compensate the fluctuation of the probe
beam (reference). The time delay between pump and probe beams was
carefully controlled by making the pump beam travel along a variable
optical delay (Newport, ILS250). To obtain the time-resolved transient
absorption difference signal at a specific wavelength, the monitoring
wavelength was selected by using an interference filter. By chopping
the pump pulses at 47 Hz, the modulated probe pulses as well as the
reference pulses were detected by two separate photodiodes. The
modulated signals of the probe pulses were measured by a gated
integrator (SRS, SR250) and a lock-in amplifier (EG&G, DSP7265)
and stored on a personal computer for further signal processing. The
polarization angle between the pump and probe beams was set to the
magic angle (54.7°) in order to obviate polarization-dependent signals.
Nanosecond Transient Absorption. Nanosecond transient absorp-
tion spectra were obtained using nanosecond flash photolysis methods.6a
Specifically, an excitation pulse of 532 nm was generated from the
second harmonic output of a Q-switched Nd:YAG laser (Continuum,
Surelite). The duration of the excitation pulse was ca. 6 ns, and the
pulse energy was ca. 2 mJ/pulse. A CW Xe lamp (150 W) was used as
a probe light source for the transient absorption measurement. After
passing through the sample, the probe light was collimated and then
spectrally resolved using a 15 cm monochromator (Acton Research,
SP150) equipped with a 600 groove/mm grating. The spectral resolution
was about 3 nm for a typical transient absorption experiment. The light
signal was detected via a photomultiplier tube (Hamamatsu, R928).
For the temporal profile measurements, the output signal from the PMT
was recorded using a 500 MHz digital storage oscilloscope (Tektronix,
TDS3052). Because the triplet state dynamics of molecules in solution
are strongly dependent on the solution oxygen concentration, efforts
âI0(1 - e-R l
)
0
T(z) ) 1 -
2R0(1 + (z/z0)2)
where R0 is the linear absorption coefficient, l, the sample length, and
z0, the diffraction length of the incident beam. After obtaining the
nonlinear absorption coefficient â, the TPA cross-section σ(2) (in units
of 1 GM ) 10-50 cm4‚s/photon‚molecule) of a single solute molecule
sample can be determined by using the following relationship:
σ(2)NAd × 10-3
â )
hν
where NA is the Avogadro constant, d is the concentration of the TPA
compound in solution, h is Planck’s constant, and υ is the frequency
of the incident laser beam. So as to satisfy the condition of R0l , 1,
which allows the pure TPA σ(2) values to be determined using a
simulation procedure, the TPA cross-section value of AF-50 was
measured as a reference compound; this control was found to exhibit
a TPA value of 50 GM at 800 nm.
Acknowledgment. The work at Yonsei University was
financially supported by the Star Faculty Program of the
Ministry of Education and Human Resources. The work at
Kyoto University was supported by the CREST of JST. The
work at the University of Texas at Austin was supported in
part by the U.S. National Science Foundation (CHE 0515670
to J.L.S.).
Supporting Information Available: Preparation of 1-7;
molecular extinction coefficient of 1-7; schematic conjugation
pathways and NICS(0) values of 1-7; S1-state decay profiles
of 1-7; T1-state decay profiles of 3, 5, and 6; open-aperture
femtosecond Z-scan traces of 3-7. This material is available
free of charge via the Internet at http://pubs.acs.org.
(20) Ahn, T. K.; Yoon, Z. S.; Hwang, I.-W.; Lim, J. K.; Rhee, H.; Joo, T.; Sim,
E.; Kim, S. K.; Aratani, N.; Osuka, A.; Kim, D. J. Phys. Chem. B 2005,
109, 11223.
JA064773K
9
14134 J. AM. CHEM. SOC. VOL. 128, NO. 43, 2006