Push-pull arylethynyl porphyrins: New chromophores that exhibited large molecular first-order hyperpolarizabilities

Article abstract of DOI:10.1021/ja953610l

A new class of chromophores has been fabricated that features both electron-releasing and electron-withdrawing groups fused via and intervening ethynyl moiety to the carbon framework of a (porphinato)metal complex. These species possess large molecular first-order hyperpolarizabilities (β). We report herein the synthesis, optical spectroscopy, and the hyper-Rayleigh scattering data used to determine the beta values of two archetypal members of this new class of exceptional nonlinear chromophores: [5-[[4'-(dimethylamino) phenyl]ethynyl]-15-[ (4″-nitrophenyl) -ethynyl]-10,20-diphenylporphinato] copper(II) and [5-[[4'- (dimethylamino)phenyl]ethynyl]-15-[ (4″- nitrophenyl)ethynyl]-10,20-diphenylporphinato]zinc(II) . One of these compounds, [5-[[4'-(dimethylamino) phenyl]ethynyl]-15-[ (4″-nitrophenyl)ethynyl]-10,20 -diphenylporphinato] zinc(II), possesses β values near 5000 × 10^-30 esu at incident irradiation wavelength of both 830 and 1064 nm. These studies suggest that this chromophoric structural motif may find utility in the development of electrooptic devices as well as materials for efficient frequency doubling of incident irradiation (second harmonic generation). A new class of chromophores has been fabricated that features both electron-releasing and electron-withdrawing groups fused via an intervening ethynyl moiety to the carbon framework of a (porphinato)metal complex. These species possess large molecular first-order hyperpolarizabilities (β). We report herein the synthesis, optical spectroscopy, and the hyper-Rayleigh scattering data used to determine the β values of two archetypal members of this new class of exceptional nonlinear chromophores: [5-[[4'-(dimethylamino)phenyl]ethynyl]-15-[(4''-nitrophenyl)ethynyl]-1 0,20-diphenylporphinato]copper(II) and [5-[[4'-(dimethylamino)phenyl]ethynyl]-15-[(4''-nitrophenyl)ethynyl]-1 0,20-diphenyl]porphinato]zinc(II). One of these compounds, [5-[[4'-(dimethylamino)phenyl]ethynyl]-15-[(4''-nitrophenyl)ethynyl]-1 0,20-diphenylporphinato]zinc(II), possesses β values near 5000 x 10^-30 esu at incident irradiation wavelengths of both 830 and 1064 nm. These studies suggest that this chromophoric structural motif may find utility in the development of electrooptic devices as well as materials for efficient frequency doubling of incident irradiation (second harmonic generation).

Full text of DOI:10.1021/ja953610l

J. Am. Chem. Soc. 1996, 118, 1497-1503
1497
Push-Pull Arylethynyl Porphyrins: New Chromophores That
Exhibit Large Molecular First-Order Hyperpolarizabilities
Steven M. LeCours,† Hann-Wen Guan,‡ Stephen G. DiMagno,†,§
C. H. Wang,*,‡ and Michael J. Therien*,†
Contribution from the Department of Chemistry, UniVersity of PennsylVania,
Philadelphia, PennsylVania 19104-6323, and the Department of Chemistry,
UniVersity of Nebraska, Lincoln, Nebraska 68588-0304
ReceiVed October 27, 1995X
Abstract: A new class of chromophores has been fabricated that features both electron-releasing and electron-
withdrawing groups fused via an intervening ethynyl moiety to the carbon framework of a (porphinato)metal complex.
These species possess large molecular first-order hyperpolarizabilities (â). We report herein the synthesis, optical
spectroscopy, and the hyper-Rayleigh scattering data used to determine the â values of two archetypal members of
this new class of exceptional nonlinear chromophores: [5-[[4′-(dimethylamino)phenyl]ethynyl]-15-[(4′′-nitrophenyl)-
ethynyl]-10,20-diphenylporphinato]copper(II) and [5-[[4′-(dimethylamino)phenyl]ethynyl]-15-[(4′′-nitrophenyl)ethy-
nyl]-10,20-diphenylporphinato]zinc(II). One of these compounds, [5-[[4′-(dimethylamino)phenyl]ethynyl]-15-[(4′′-
nitrophenyl)ethynyl]-10,20-diphenylporphinato]zinc(II), possesses â values near 5000 × 10-30 esu at incident irradiation
wavelengths of both 830 and 1064 nm. These studies suggest that this chromophoric structural motif may find
utility in the development of electrooptic devices as well as materials for efficient frequency doubling of incident
irradiation (second harmonic generation).
Introduction
chemical systems should be examined in this regard if significant
advances are to be realized. Toward such ends, our group has
focused on new approaches to regulate chromophore photo-
physical properties.11-16 We report herein a new set of
engineering criteria for large molecular first-order NLO re-
sponses; these design notions are manifest in a porphyrin-based
NLO chromophore structural motif that not only provides for
significant molecular first hyperpolarizabilities, but also displays
exceptional thermal stability as well.
Future generations of optoelectronic devices for telecom-
munications, information storage, optical switching, and signal
processing are predicated to a large degree on the development
of materials with exceptional nonlinear optical (NLO) re-
sponses.1 Toward this end, considerable effort has been spent
to elucidate the chromophore design elements that correlate most
closely with large first-order hyperpolarizabilities; this is due
to the fact that an NLO chromophore with an exceptional
molecular first hyperpolarizability would be a highly desirable
candidate for incorporation into electric field poled-polymeric
systems and thus serve as the basis for macroscopic materials
for frequency doubling as well as electrooptical devices that
function as waveguide switches, modulators, filters, and polar-
ization transformers.1d
A tremendous number of chromophores has been evaluated
for suitability to function as components in hypothetical NLO
materials. A few of these chromophores have served as
components of functioning polymer-based optoelectronic de-
vices; the physical properties of all these prototype materials
possess one or more critical deficiencies that render com-
mercialization of these systems impractical.1-10 These facts
suggest that it may be appropriate to reexamine the accepted
tenets of NLO chromophore design and that new types of
(4) (a) Levine, B. F.; Bethea, C. G. J. Chem. Phys. 1977, 66, 1070-
1074. (b) Lalama, S. J.; Garito, A. F. Phys. ReV. A. 1979, 20, 1179-1194.
(c) Levine, B. F.; Bethea, C. G. Appl. Phys. Lett. 1974, 24, 445-447. (d)
Singer, K. D.; Garito, A. F. J. Chem. Phys. 1981, 75, 3572-3580.
(5) (a) Marder, S. R.; Gorman, C. B.; Tiemann, B. G.; Cheng, L.-T. J.
Am. Chem. Soc. 1993, 115, 3006-3007. (b) Meyers, F.; Marder, S. R.;
Pierce, B. M.; Bre´das, J. L. J. Am. Chem. Soc. 1994, 116, 10703-10714.
(c) Bourhill, G.; Bre´das, J.-L.; Cheng, L.-T.; Marder, S. R.; Meyers, F.;
Perry, J. W.; Tiemann, B. G. J. Am. Chem. Soc. 1994, 116, 2619-2620.
(6) (a) Garito, A. F.; Singer, K. D.; Teng, C. C. Appl. Phys. Lett. 1981,
39, 1-26. (b) Lalama, S. J. Appl. Phys. Lett. 1981, 39, 940-942.
(7) (a) Stiegman, A. E.; Graham, E.; Perry, K. J.; Khundkar, L. R.; Cheng,
L.-T.; Perry, J. W. J. Am. Chem. Soc. 1991, 113, 7658-7666. (b) Dehu,
C.; Meyers, F.; Bre´das, J. L. J. Am. Chem. Soc. 1993, 115, 6198-6206.
(8) Marder, S. R.; Cheng, L.-T.; Tiemann, B. G.; Friedli, A. C.;
Blanchard-Desce, M.; Perry, J. W.; Skindhøj. J. Science (Washington, D.
C.) 1994, 263, 511-514.
(9) (a) Cheng, L.-T.; Tam, W.; Stevenson, S. H.; Meredith, G. R.; Rikken,
G.; Marder, S. R. J. Phys. Chem. 1991, 95, 10631-10643. (b) Chen, C.-
T.; Marder, S. R.; Cheng, L.-T. J. Am. Chem. Soc. 1994, 116, 3117-3118.
(10) (a) Di Bella, S.; Fragala´, I. L.; Ratner, M. A.; Marks, T. J. J. Am.
Chem. Soc. 1993, 115, 682-686. (b) Kanis, D. R.; Lacroix, P. G.; Ratner,
M. A.; Marks, T. J. J. Am. Chem. Soc. 1994, 116, 10089-10102.
(11) Lin, V. S.-Y.; DiMagno, S. G.; Therien, M. J. Science (Washington,
D. C.) 1994, 264, 1105-1111.
* To whom correspondence should be addressed.
† University of Pennsylvania.
‡ University of Nebraska.
§ Present address: Department of Chemistry, University of Nebraska,
Lincoln, NE 68588-0304.
X Abstract published in AdVance ACS Abstracts, January 1, 1996.
(1) (a) Williams, D. J. Agnew. Chem., Int. Ed. Engl. 1984, 23, 690-
703. (b) Chemla, D. S.; Zyss, J. Nonlinear Optical Properties of Organic
Molecules and Crystals; Academic Press: Orlando, FL, 1987; Vols. 1 and
2. (c) Prasad, P. N.; Williams, D. J. Introduction to Nonlinear Optical Effects
in Molecules and Polymers; John Wiley and Sons: New York, 1991. (d)
Burland, D. M.; Miller, R. D.; Walsh, C. A. Chem. ReV. 1994, 94, 31-75.
(2) (a) Ward, J. F. ReV. Mod. Phys. 1965, 37, 1-18. (b) Orr, B. J.; Ward,
J. F. Mol. Phys. 1971, 20, 513-526.
(12) (a) Lin, V. S.-Y.; Therien, M. J. Chem. Eur. J. In press. (b) Lin, V.
S.-Y.; Williams, S. A.; Therien, M. J. Manuscript in preparation.
(13) LeCours, S. M.; DiMagno, S. G.; Therien, M. J. J. Am. Chem. Soc.
Submitted.
(14) LeCours, S. M.; Williams, S. A.; Therien, M. J. Manuscript in
preparation.
(15) (a) DiMagno, S. G.; Lin, V. S.-Y.; Therien, M. J. J. Am. Chem.
Soc. 1993, 115, 2513-2515. (b) DiMagno, S. G.; Lin, V. S.-Y.; Therien,
M. J. J. Org. Chem. 1993, 58, 5983-5993.
(16) LeCours, S. M.; de Paula, J. C.; Therien, M. J. J. Phys. Chem.
Submitted.
(3) (a) Oudar, J. L.; Le Person, H. Opt. Commun. 1975, 15, 258-262.
(b) Oudar, J. L.; Chemla, D. S. J. Chem. Phys. 1977, 66, 2664-2668.
0002-7863/96/1518-1497$12.00/0 © 1996 American Chemical Society

1498 J. Am. Chem. Soc., Vol. 118, No. 6, 1996
LeCours et al.
collected and evaporated to give a deep green (almost black) residue
(97.5 mg, 35%).
Experimental Section
Materials. All compounds were handled using methods and
protocols described previously.15 Elemental analyses were performed
at Robertson Microlit Laboratories (Madison, NJ).
Instrumentation. Electronic spectra were recorded on an OLIS UV/
vis/NIR spectrophotometry system that is based on the optics of a Carey
14 spectrophotometer.
[5-[[4′-(Dimethylamino)phenyl]ethynyl]-15-[(4′′-nitrophenyl)ethy-
nyl]-10,20-diphenylporphinato]zinc(II) [1(ZnII)]. [5-[[4′-(Dimethy-
lamino)phenyl]ethynyl]-15-bromo-10,20-diphenylporphinato]zinc(II) (52.0
mg, 102 µmol), (p-nitrophenyl)acetylene (35.2 mg, 240 µmol), Pd-
[PPh3]4 (15.0 mg, 13 µmol), and copper(I) iodide (6.6 mg, 35 µmol)
were placed in a 50 mL Schlenk tube under an N2 atmosphere. THF
(10 mL) and diethylamine (1 mL) were added. The reaction vessel
was shielded from light and allowed to react at room temperature. After
12 h, the solution was evaporated to dryness and the solid purified by
column chromatography (silica gel, 3:1 hexanes:THF). The green band
was collected and dried to give the desired product (57.0 mg, 97%).
1H NMR (250 MHz, CDCl3, pyridine-d5): δ 9.63 (d, 2 H, J ) 4.6),
9.51 (d, 2 H, J ) 4.6), 8.78 (d, 2 H, J ) 4.6), 8.73 (d, 2 H, J ) 4.6),
8.10 (m, 6 H), 7.83 (d, 2 H, J ) 8.7), 7.75 (d, 2 H, J ) 8.9), 7.64 (m,
6 H), 6.69 (d, 2 H, J ) 8.8), 2.93 (s, 6 H). 13C NMR (60 MHz, CDCl3,
pyridine-d5): δ 152.06, 151.46, 150.08, 149.93, 146.13, 142.40, 134.25,
132.66, 132.55, 131.94, 131.31, 131.11, 130.79, 129.84, 127.23, 126.29,
123.60, 122.73, 111.82, 110.49, 104.14, 99.67, 98.58, 97.52, 94.11,
91.23. Vis (THF): 460 (5.27), 678 (4.88). FAB MS: 812.19 (calcd
812.19). Anal. Calcd for C54H40N6O3Zn‚C4H8O(THF): C, 73.18; H,
4.55; N, 9.48. Found: C, 73.04; H, 4.48; N, 9.40.
Hyper-Rayleigh Light Scattering (HRS) Experiments. The details
of the method17,18 as well as the experimental set up at the University
of Nebraska have been presented previously.18 Normally, the first-
order nonlinear effect is absent in an isotropic medium because of the
centrosymmetric environment; removal of centrosymmetry and mea-
surement of the molecular first hyperpolarizability has most commonly
been accomplished by the electric field-induced second harmonic
generation (EFISH) technique. The primary advantage of the HRS
method lies in the fact that no external electric field is necessary for
the experiment, since thermal motion constantly distorts the local
symmetry of the medium, removing instantaneously centrosymmetry
in the liquid and allowing incoherent second harmonic light to be
observed. It is important to note that the HRS technique is very
sensitive to solution concentration; it is thus an ideal method for the
evaluation of the hyperpolarizabilities of the high extinction coefficient
porphyrin-derived chromophores reported in this paper.
[5-[[4′-(Dimethylamino)phenyl]ethynyl]-15-[(4′′-(nitrophenyl)-
ethynyl]-10,20-diphenylporphinato]copper(II) [1(CuII)]. A 20.6 mg
(25.3 µmol) amount of analytically pure 1(ZnII) was dissolved in 50
mL of a 1:1 THF:CHCl3 solution in a 125 mL Erlenmeyer flask.
Concentrated hydrochloric acid (0.5 mL) was added dropwise. After
5 min, triethylamine (5 mL) was added. The mixture was stirred an
additional 5 min before being placed into a 125 mL separatory funnel.
The organic layer was washed once with 1.0 M KOH (20 mL) and
then with distilled water. The aqueous fractions were discarded, and
the organic layer was rotovapped to dryness. The free base push-
pull (arylethynyl)porphyrin was used without further purification. Vis
(THF): 446, 624, 708.
The 5-[[4′-(dimethylamino)phenyl]ethynyl]-15-[4′′-(nitrophenyl)ethy-
nyl]-10,20-diphenylporphyrin was placed in a 200 mL round bottom
flask equipped with a stir bar. Toluene (60 mL) and cupric acetate
hydrate (100 mg, 501 mmol) were added. The resulting mixture was
heated to reflux. After 6 h, the mixture was cooled and placed in a
125 mL separatory funnel. The toluene layer was washed once with a
1 M ammonium hydroxide solution (50 mL) and then twice with water
(50 mL). The toluene layer was rotovapped to a total volume of
approximately 15 mL and then chromatographed on silica gel using
toluene as eluant. A tight, single green band was rotovapped to dryness
and further dried under high vacuum to give 1(CuII) [12.0 mg, 58%
based on 20.6 mg of compound 1(ZnII)]. Vis (THF): 452 (5.19), 656
(4.74). Low resolution FAB MS: 812 (calcd 811).
Nd:YAG (10 Hz; 3-8 ns pulse width) and Ti-sapphire lasers
(Coherent, Mira Model 9000; 76 MHz; 100 fs pulse width) were used
as excitation sources. The intensity of the incident beam from the Nd:
YAG laser, after it was filtered through a set of three long pass filters
to remove the second harmonic component at 0.53 µm, was varied via
the combination of a polarizer with a multiple-order half wave plate.
The incident beam was focused onto the sample with a f/10 lens and
the scattered light was collected by a f/1.3 lens followed by a biconvex
lens with a focal length of 30 cm. The collected signal was directed
to a photomultiplier tube equipped with an interference filter and a
sharp bandpass filter set at the second harmonic frequency attached in
the front. A boxcar integrator was used to process the signal when
the Nd:YAG laser was used as an excitation source. To take advantage
of its high repetition rate, a photon counting system was utilized to
process the light scattering signal when the Ti-sapphire laser was used
as an excitation source. All HRS experiments were carried out at room
temperature. The scattering angle was 90°.
Samples for HRS studies were dissolved in chloroform (CHCl3) and
filtered through a 0.2 µm membrane to remove dust and any
adventitious particulates. Samples differing in chromophore concentra-
tion were prepared by serial dilution of a concentrated standard. Glass
solution cells (3.5 mL) were used throughout these experiments.
(5,15-Dibromo-10,20-diphenylporphinato)zinc(II) and 4-Substi-
tuted Phenylacetylenes. [p-(Dimethylamino)phenyl]acetylene and (p-
nitrophenyl)acetylene were synthesized similarly to literature methods.19
The preparation of (5,15-dibromo-10,20-diphenylporphinato)zinc(II) has
been previously reported.15
Results and Discussion
[5-[[4′-(Dimethylamino)phenyl]ethynyl]-15-bromo-10,20-diphe-
nylporphinato]zinc(II). (5,15-Dibromo-10,20-diphenylporphinato)-
zinc(II) (252 mg, 369 µmol), CuI (10 mg, 52 µmol), Pd[PPh3]4 (35
mg, 30 µmol), diethylamine (5 mL), and [p-(dimethylamino)phenyl]-
acetylene (63.1 mg, 435 µmol) were brought together along with 30
mL of tetrahydrofuran (THF) in a 100 mL Schlenk tube under an N2
atmosphere. The resulting solution becomes intensely green as the
reaction proceeds at room temperature. At the reaction endpoint (t )
12 h), the crude product was purified by column chromatography on
silica gel using 4:1 hexanes:THF as eluant. The green band was
Design, modification, and further fine-tuning of the magnitude
of the molecular first hyperpolarizability (â) of a given
chromophore has generally been thought of in the context of
Oudar’s two-state model:2-4
µ2ge
â
(µee - µgg)
(1)
E2ge
(17) (a) Terhune, R. W.; Maker, P. D.; Savage, C. M. Phys. ReV. Lett.
1965, 14, 681-684. (b) Clays, K.; Persoons, A. Phys. ReV. Lett. 1991, 66,
2980-2983. (c) Clays, K.; Hendrickx, E.; Triest, M.; Verbiest, T.; Persoons,
A.; Dehu, C.; Bredas, J.-L. Science (Washington, D. C.) 1993, 262, 1419-
1422. (d) Laidlaw, W. M.; Denning, R. G.; Verbiest, T.; Chauchard, E.;
Persoons, A. Nature (London) 1993, 363, 58-60.
Here g and e represent the indices of the ground and charge
transfer (CT) excited states, respectively; µ is the dipole matrix
element and E is the transition energy. The dominant compo-
nent of the â tensor is thus seen to be proportional to the change
in dipole moment between the ground and excited states as well
as the square of the transition oscillator strength, while inversely
proportional to the square of the energy gap between these two
states.
(18) Pauley, M. A.; Wang, C. H.; Jen, A. K.-Y. J. Chem. Phys. 1995,
102, 6400-6405.
(19) (a) Eastmond, R.; Walton, D. R. M. Tetrahedron 1972, 28, 4591-
4599. (b) Westmijze, H.; Vermeer, P. Synthesis 1979, 390-392. (c)
Takahashi, S.; Kuroyama, Y.; Sonogashira, K.; Hagihara, N. Synthesis 1980,
627-630. (d) Mesnard, D.; Bernadou, F.; Miginiac, L. J. Chem Res. (S)
1981, 270-271. (e) Zhang, Y.; Wen, J. Synthesis 1990, 727-728.
Most NLO chromophores are composed of a donor (D) and
acceptor (A), which are the molecular entities chiefly involved

Push-Pull Arylethynyl Porphyrins
J. Am. Chem. Soc., Vol. 118, No. 6, 1996 1499
in charge redistribution, as well as a bridge, which we define
as the molecular scaffolding that links the D and A portions of
the molecule. To date, the design of chromophores with good
second-order nonlinear properties has focused primarily on
engineering: (i) the electronic nature of the D and A, and (ii)
the conjugation length of the bridge.5-10 The former controls
D-A mixing with respect to a specific bridge, while the latter
plays a role in modulating D-A electronic coupling as well as
determines the magnitude of the change in dipole moment. It
intimately on the nature of the bridge highlights the need to
explore new types of molecular scaffolding to connect D to A
in NLO chromophores.
In an effort to construct entirely new classes of NLO
chromophores with exceptional photophysical properties, we
have focused on engineering bridge electronics and topology.
Our most basic criterion is that the bridge should be much more
polarizable than the simple polyene, polyyne, polyphenylene,
and polyheteroaromatic structures that have been most com-
monly used. Moreover, the bridge must be the primary photonic
component of the molecule in the sense that a significant portion
of the oscillator strength of the charge transfer transition that
couples D to A should be bridge localized. Ideally, the bridge-
localized excited state should dramatically alter D-A electronic
coupling relative to the coupling the ground-state bridge
provides.
is important to recognize that the parameters µee - µgg, µ2
,
ge
and E2ge are all intimately related: for example, increasing the
bridge conjugation length, in general, results in an increase in
the magnitude of the ground-to-excited state change in dipole
moment, while concomitantly diminishing the square of the
dipole matrix element and augmenting the square of the CT
transition energy; the latter two effects have their genesis in
the fact that increased bridge lengths abate D-A electronic
coupling. Maximizing â thus involves a subtle interplay
between three parameters that do not necessarily simultaneously
attain their optimal value for a particular molecular structure
(D, A, bridge); thus, optimization of the first hyperpolarizability,
â, obviously involves a balancing act.20
One approach to enabling such differential ground- and
excited-state coupling is to choose a D-A bridging motif that
is capable of accessing a resonance form in its excited state
that is unattainable for the ground-state structure. Such an
excited-state resonance structure would be optimal if it produced
a large transition dipole oriented directly along the D-to-A
molecular charge transfer axis. A designed excited state with
these properties would facilitate large molecular first hyperpo-
larizabilities since the magnitude of the change in dipole moment
would not be held ransom by significant diminution of the
oscillator strength of the CT transition or an increase in the
transition energy at relatively large D-A distances, since a high
oscillator strength, bridge-centered transition would now directly
couple D to A. Presumably in such a system, if the orientation
and dipolar nature of the bridge-centered CT transition could
The overwhelming number of chromophores that have been
studied to date for their second-order nonlinear properties can
be classified as D-A systems in which the molecular bridge is
either based on ethene, phenylene, ethyne, small-ring heteroaro-
matic, or styrene building blocks, or a combination of two or
more of these simple units. Although a variety of different
organic media have been utilized as D-A bridging moieties,
comparatively little attention has been paid to how the details
of the bridge topology and electronic structure impact the
chromophore second-order NLO response, particularly when
viewed alongside the tremendous body of literature describing
how D-A electronic properties and bridge length modulate the
molecular first hyperpolarizability.5-10
be maintained over a long range, µee - µgg and µ2 would
ge
simultaneously increase with augmented bridge lengths while
concomitantly maintaining or slightly reducing E2
.
ge
Many of these desirable design elements for the bridge in
such an NLO chromophore have already been realized. We
have reported the synthesis and optical spectroscopy of a variety
of ethynyl- and butadiynyl-linked porphyrin arrays.11 A subset
of this new class of compounds is one in which ethynyl groups
are utilized to attach multiple porphyrins together at their
respective meso-carbon positions; the excited states of some of
these species, exemplified by [5,15-bis[[5′-[(10,20-diphenylpor-
phinato)zinc(II)]]ethynyl]-10,20-diphenylporphinato]zinc(II), pos-
sess substantial dipole moments.11,12 Such large transition
dipoles derive in part from the extended conjugation enabled
by the approximate coplanarity of the porphyrin macrocycles
in solution. Fast time-resolved anisotropy measurements show
that the excited state of such multichromophoric systems is
singly degenerate and oriented along the C2 molecular axis of
highest conjugation.12b
The bridge mediated donor-acceptor electronic interaction
must be large in order to maximize the strength of the transition
matrix element (associated with the oscillator strength) of the
charge-transfer transition. Yet, if this D-A bridge-mediated
mixing is too large, the D and A states will be strongly mixed
together and the molecule will lose its electronic asymmetry
(that is, the dipole moment difference between the ground and
excited states). In the extreme limit where all asymmetry is
lost, there is no dipole moment change between ground and
excited state, so â is diminished within the context of the two-
state model.2-4
Some simple models for NLO chromophores that capture the
essence of the problem are cast in a four-orbital framework:
two of these orbitals correspond to the donor- and acceptor-
localized states and two to the bridge frontier orbitals.20 Within
this four-orbital description, a control parameter ∆ can be
defined that is approximately related to the relative orbital
energies of the isolated donor (HOMO) and isolated acceptor
(LUMO) orbitals (ED and EA, respectively) in units of the
effective coupling interaction (t′) between D and A provided
by the bridge.
Porphyrin conjugation can be extended when the ethynyl
group is used to link the chromophore to an aromatic entity
other than porphyrin. We have synthesized a variety of
symmetrically-substituted [5,15-bis(arylethynyl)-10,20-diphe-
nylporphinato]metal complexes;13 like the meso-to-meso ethy-
nyl-bridged porphyrin arrays, the electronically excited states
of these species show substantial energetic splitting between x-
and y-polarized populations, with the short-time anisotropy
consistent with a singly degenerate excited state localized along
the molecular axis defined by the arylethynyl groups.14
∆ ) (ED - EA)/|t′|
(2)
The magnitudes of µee - µgg, µ2ge, and E2ge will be effected
by the dimension of the parameter ∆;20 that ∆ depends
Using a modification of the synthetic procedures developed
to assemble the conjugated porphyrin arrays and the 5,15-bis-
(arylethynyl)-10,20-diphenylporphyrins,13,15 one can fabricate
asymmetric, arylethynyl-elaborated porphyrins. Scheme 1
shows the synthesis of [5-[[4′-(dimethylamino)phenyl]ethynyl]-
(20) (a) Beratan, D. N. In New Materials for Nonlinear Optics, Hann,
R. A.; Bloor, D., Eds.; ACS Symposium Series 455, American Chemical
Society: Washington, D. C., 1991; pp 89-102. (b) Marder, S. R.; Beratan,
D. N.; Cheng, L.-T. Science (Washington, D. C.) 1991, 252, 103-106. (c)
Risser, S. M.; Beratan, D. N.; Marder, S. R. J. Am. Chem. Soc. 1993, 115,
7719-7728.

1500 J. Am. Chem. Soc., Vol. 118, No. 6, 1996
LeCours et al.
Scheme 1
15-[(4′′-nitrophenyl)ethynyl]-10,20-diphenylporphinato]zinc-
(II) [1(ZnII)], the archetypal member of a new class of D-A
chromophores, the push-pull (arylethynyl)porphyrins.
Photophysical studies of 1(ZnII) and its (porphinato)copper
analogue, 1CuII, show that these species possess a host of
unusual optical properties.14,16 For example, transient resonance
Raman studies indicate that the excited states of [5-[[4′-
(dimethylamino)phenyl]ethynyl]-15-[4′′-(nitrophenyl)ethynyl]-
10,20-diphenylporphinato]metal complexes exhibit reduced bond
order in the 4′-(dimethylamino)phenyl and 4′′-nitrophenyl aryl
rings as well as in the nitro group with respect to that seen for
these modes in their ground-state spectra; downshifting of these
vibrational frequencies is not observed when the transient
excited-state resonance Raman spectrum is compared with the
analogous ground-state spectrum for the corresponding sym-
metrically-substituted (arylethynyl)porphyrins such as [5,15-bis-
[[4′-(dimethylamino)phenyl]ethynyl]-10,20-diphenylporphinato]-
metal complexes.16 Scheme 2 summarizes this data and presents
our current picture of the excited state of 1(ZnII), 1(ZnII)*; note
that 1(ZnII)* can access a cumulenic resonance form while 1-
(ZnII) is best described as a supramolecular chromophore in
which three aromatic ring systems are linked by ordinary
carbon-carbon single and triple bonds.
Since the push-pull (arylethynyl)porphyrin structural motif
in fact displays two of the proposed elements we thought key
to the fabrication of exceptional NLO chromophores, namely a
highly-polarizable D-A bridge that (i) contains significant
oscillator strength and (ii) provides spectacular D-A electronic
coupling in its excited state with respect to its ground state, we
examined the molecular first hyperpolarizabilities of two [5-[[4′-
(dimethylamino)phenyl]ethynyl]-15-[(4′′-nitrophenyl)ethynyl]-
10,20-diphenylporphinato]metal complexes, 1(ZnII) and [5-[[4′-
(dimethylamino)phenyl]ethynyl]-15-[(4′′-nitrophenyl)ethynyl]-
10,20-diphenylporphinato]copper(II) [1(CuII)], using the hyper-
Rayleigh (light) scattering (HRS) technique.17,18
Figure 1. Electronic spectra of 1(CuII) and 1(ZnII) recorded in
chloroform. Arrows at 532 and 415 nm indicate the wavelengths
corresponding to the second harmonic of the fundamental incident 1064
and 830 nm irradiation, respectively.
The total scattered light intensity at 0.5 λ (2ω) is given by:
I(2ω) ) K(Fsâ2s + Fcâ2c)I(ω)2
(3)
where FS and FC are the number densities of solvent and
chromophore molecules, âS and âC the major hyperpolarizability
tensor of the solvent and chromophore, ω the frequency of
incident irradiation, 2ω the second harmonic frequency, I the
light intensity, and K a quantity determined only by the
scattering geometry and local field factors at low chromophore
concentrations. Figures 2 and 3 plot the hyper-Rayleigh
scattering signal at I(2ω) versus chromophore number density
for 1(ZnII) and 1(CuII) in chloroform solvent in the very dilute
concentration regime for irradiation at wavelengths λ1 and λ2;
â830 and â1064 for both chromophores can be directly evaluated
from these plots.18 Non-negative intercepts were obtained in
all cases at low chromophore concentrations. The quadratic
dependence of I(2ω) on I(ω) verified that the scattered light
intensity clearly arises from the nonlinear hyper-Rayleigh
scattering process (data not shown). The evaluated chromophore
molecular first hyperpolarizabilities were referenced internally
against the known solvent â value; the well-studied NLO
chromophore p-nitroaniline (PNA) was used as an external
reference in these experiments.21 Consistent â values were
obtained irrespective of the choice of reference.
Measurements of the molecular first hyperpolarizability for
both 1(ZnII) and 1(CuII) were carried out in chloroform solvent
for incident irradiation wavelengths of 830 (λ1) and 1064 (λ2)
nm. Figure 1 shows the electronic absorption spectra of 1(ZnII)
and 1(CuII). Note that the scattered light at the second harmonic
of 830 nm (0.5 λ1) will be partially absorbed by both chro-
mophores on the blue shoulder of the Soret transition, while
scattered light of wavelength 532 nm (0.5 λ2) corresponds to a
frequency in the optically-transparent region of the spectra of
both 1(ZnII) and 1(CuII). HRS data obtained at incident
irradiation wavelength λ1 were Beer’s Law-corrected for chro-
mophore absorption to eliminate the effective signal attenuation
prior to calculating â830 using an approach described previ-
ously.18
The â values evaluated from the data presented in Figures 2
and 3 are exceptionally large. The wavelength dependence of
the molecular first hyperpolarizability for 1(CuII) shows that
changing the incident irradiation wavelength from 830 to
1064 nm results in a decrease in the magnitude of the evaluated
â by approximately 65% (â830 ) 4374 × 10-30 esu; â1064
)
1501 × 10-30 esu). The data obtained for 1(ZnII) in chloro-
form is exceptionally intriguing: in contrast to that seen for
1(CuII), the measured â value appears to be virtually indepen-
(21) Internal reference, CHCl3 solvent: â(chloroform) ) -0.49 × 10-30
esu. External reference, p-nitroaniline (PNA) in CHCl3: â(PNA) ) 23 ×
10-30 esu at 1.06 µm (HRS); â(PNA) ) 25 × 10-30 esu at 1.06 µm (EFISH).

Push-Pull Arylethynyl Porphyrins
J. Am. Chem. Soc., Vol. 118, No. 6, 1996 1501
Figure 2. Hyper-Rayleigh scattering signal (corrected for absorption) vs chromophore number density for 1(CuII) and 1(ZnII) in chloroform;
excitation source ) Nd:YAG laser. The solid line represents the best fit to the data points as described in the text. The quantities (Fs, âs) and (Fc,
âc) denote the concentration and hyperpolarizabilty of solvent molecules and chromophores, respectively.
Figure 3. Hyper-Rayleigh scattering signal (corrected for absorption) vs chromophore number density for 1(CuII) and 1(ZnII) in chloroform;
excitation source ) Ti-Sapphire laser. The solid line represents the best fit to the data points as described in the text. The quantities (Fs, âs) and (Fc,
âc) denote the concentration and hyperpolarizabilty of solvent molecules and chromophores, respectively.
Scheme 2
dent of irradiation wavelength (â830 ) 5142 × 10-30 esu; â1064
) 4933 × 10-30 esu).
abilities of the push-pull (arylethynyl)porphyrin complexes,22
they do not account for the observed frequency dependence of
â for these compounds.23 These studies estimate that â1064 for
While theoretical studies based on INDO/SCI calculations
within a sum over states formalism are consistent with the very
large, experimentally determined molecular first hyperpolariz-
(22) Priyadarshy, S.; Therien, M. J.; Beratan, D. N. J. Am. Chem. Soc.
1996, 118, 1504-1510.

1502 J. Am. Chem. Soc., Vol. 118, No. 6, 1996
LeCours et al.
Table 1. Comparative Nonlinear Optical Data for 1(ZnII) and Selected Benchmark Organic Chromophores
1(ZnII) should be approximately three- to six-fold enhanced with
respect to â(0), the molecular first hyperpolarizability at zero
frequency. Such a diminution of â(0) with respect to â1064 may
be overestimated, given that theory predicts â830 to be much
greater than â1064, and that the mode of resonant enhancement
predicted to be important for 1(ZnII) in these calculations is
not expected to be experimentally significant.24,25 Nevertheless,
a conserVatiVely estimated minimal magnitude of â(0) for this
chromophore of ∼800 x 10-30 esu compares well with the
extrapolated â(0)’s for the very best organic chromophores for
second harmonic generation reported to date: these molecules
possess â(0)’s not significantly in excess of 1000 × 10-30 esu
(Table 1).26 As such, 1(ZnII) represents the first metal-
containing chromophore with a molecular first hyperpolariz-
ability in this range.
Given that 1(ZnII) is simply the prototype molecule based
on our new set of design criteria for D-bridge-A compounds
that possess large molecular first hyperpolarizabilities, we
emphasize that (i) it is a virtual certainty that chemical
modifications to the push-pull [(arylethynyl)porphinato]metal
blueprint will lead to systems possessing significantly larger
â(0)’s, and (ii) the concepts for chromophore engineering
discussed herein should find wide application in the field of
nonlinear optics. Moreover, because 1(ZnII) possesses a
significant dipole moment (∼12.3 D),22 the technologically
relevant parameter µâ(0) will also be correspondingly large.
Table 1 compares the molecular first hyperpolarizabilities, µâ(0)
values, as well as the quantity µâ(0) normalized by the
chromophore molecular weight of 1(ZnII) with some of the more
impressive all-organic NLO chromophores reported to date and
the 4-(dimethylamino)-4′-nitro-trans-stilbene (DANS) standard;
1(ZnII) ranks favorably with respect to these systems regardless
of the gauge utilized for chromophore comparison. It is also
worthy of note that this extrapolated â(0) value for 1(ZnII) is
more than 20-fold greater than the molecular first hyperpolar-
izabilities that have been evaluated for the simple diarylethynes
(23) These studies22 predict that â830 should be resonantly enhanced 1
order of magnitude more than â1064 for 1(ZnII), which is not borne out by
experiment.
(24) These calculations22 predict that resonant enhancement of â830 and
â1064 for 1(ZnII) derives from a B (Soret) band two-photon absorption (λ
) 920 nm) process. Such a predicted degree of resonant enhancement for
â1064 with respect to â(0) is inconsistent with experiment: the observed
quadratic dependence of I(2ω) on I(ω) (the input power) rules out any
significant contribution of two-photon-derived enhancements of the HRS
signal. Furthermore, two-photon resonant enhancement requires at least some
oscillator strength in the transition to the virtual state that lies intermediate
in energy between the ground and excited states; any significant absorption
cross section for these processes would result in the production of the highly
fluorescent S1 excited state of these porphyrin chromophores. No stimulated
emission can be detected under the conditions of these HRS experiments.
(25) A reviewer has pointed out that it is not possible to completely rule
out a contribution of two-photon absorbance fluorescence photons to the
signal intensity data plotted in Figures 2 and 3 (see ref 24); such two-
photon absorption fluorescence has been previously shown to be a major
source of photons in HRS SHG measurements (see, for example: Clays,
K.; Persoons, A. ReV. Sci. Instrum. 1994, 65, 2190-2194). Two-photon
fluorescence may account for the fact that theory does not model the
observed frequency dependence of â for 1(ZnII) (see ref 22); we wish to
point out however that recent EFISH measurements carried out on 1(ZnII)
at 1906 nm irradiation gives a value of µâ(0) consistent with that reported
in Table 1 (Cai, Y.-M.; LeCours, S. M.; Therien, M. J.; Jen, K.-Y.,
manuscript in preparation), which is based on the HRS data reported herein.
(26) (a) Dhenaut, C.; Ledoux, I.; Samuel, I. D. W.; Zyss, J.; Bourgault,
M.; Le Bozec, H. Nature 1995, 374, 339-342. (b) Gilmour, S.; Montgom-
ery, R. A.; Marder, S. R.; Cheng, L.-T.; Jen, A. K.-Y.; Cai, Y.; Perry, J.
W.; Dalton, L. R. Chem. Mater. 1994, 6, 1603-1604. (c) Blanchard-Desce,
M.; Lehn, J.-M.; Barzoukas, M.; Ledoux, I.; Zyss, J. Chem. Phys. 1994,
181, 281-289. (d) Rao, V. P.; Jen, A. K.-Y.; Wong, K. Y.; Drost, K. J. J.
Chem. Soc., Chem. Commun. 1993, 1118-1120.

Push-Pull Arylethynyl Porphyrins
J. Am. Chem. Soc., Vol. 118, No. 6, 1996 1503
(eq 4),7 demonstrating rather eloquently the degree of enhance-
to stress that such systems offer a number of obvious advantages
over the vast landscape of compounds and materials for second
harmonic generation that have employed more typical unsatur-
ated organic structures to bridge an electron donor and acceptor.
Firstly, push-pull [(arylethynyl)porphinato]metal structures en-
able the alignment of the massive porphyrin transition dipole
along the molecular charge-transfer axis; furthermore, as
indicated in eq 5, the magnitude of this transition dipole can be
readily modified.
ment in â(0) values that can be achieved when the bridge
connecting D to A is polarizable and manifests a large transition
moment with considerable CT character (Scheme 2).
Summary and Conclusion
Due to their exceptionally large â830 and â1064 values, this
new class of chromophoric compounds along with those yet to
be built based on the design criteria described herein, hold
considerable promise for materials for second harmonic genera-
tion as well as for electrooptic devices based on the Pockels
effect.1d Prior to their incorporation into any material or device,
the origin of the different wavelength dependences of the
molecular first hyperpolarizability for both 1(ZnII) and 1(CuII)
will have to be explored, as well as its potential correlation with
metal electronic structure. This aside, preliminary evaluation
of the thermal stability of 1(ZnII), however, does indicate that
this new chromophoric archetype appears to be suitable for
eventual incorporation into the types of polymer matrices
thought to be ideal for such electrooptical applications.27
Most importantly, in addition to demonstrating how the push-
pull [(arylethynyl)porphinato]metal structural motif can be used
as a critical design element in the fabrication of compounds
that possess large molecular first hyperpolarizabilities, we wish
Secondly, such systems feature relatively small transition
energies between the ground and excited states; moreover, when
the molecular bridge linking the electron-releasing and electron-
accepting groups is based on porphyrin or similar chromophores,
this parameter can be readily fine tuned. Thirdly, relative to
traditional materials for second harmonic generation based
on donor-acceptor polyenes or low band gap conducting
polymers,5,8-10,28 in which the alteration of the contribution of
the charge-separated form to the ground state structure has relied
chiefly on D-A modification, assemblies such as these allow
(i) exploration of the role played by metal electronic structure
(transition metal size, spin multiplicity, oxidation state) in
determining the dimension of the molecular first hyperpolariz-
ability; (ii) modulation of chromophore electronic properties
through axial ligand variation at the (porphinato)metal center,
potentially providing precise and subtle regulation of molecular
optoelectronic properties as well as a new mode of covalent
attachment of chromophores to a polymer backbone; and (iii)
amplification of chromophore polarizability through the engi-
neering of metal-to-metal or macrocycle-to-macrocycle charge
transfer character along the D-to-A molecular axis. Such ease
of electronic modification of an extremely potent NLO chro-
mophore for second harmonic generation is unusual as well as
advantageous; it strongly suggests that the present large mo-
lecular first hyperpolarizability exhibited by 1(ZnII) will soon
be surpassed by other compounds and materials based on the
push-pull [(arylethynyl)porphinato]metal structural motif.
(27) (a) Preliminary TGA studies of 1(ZnII) indicates that 1% of the
mass is lost over a 160-300 °C temperature range. Mass spectral analysis
of the vapor shows fragments of 72 and 44 atomic mass units (amu). While
the peak at 72 amu may correspond to loss of residual tetrahydrofuran, the
44 amu peak may derive from dealkylation of the amino functionality. As
Moylan has pointed out (J. Am. Chem. Soc. 1993, 115, 12599-12600), the
electron-releasing dialkylamino moieties are generally the least stable entities
in push-pull chromophores that feature these groups; replacing the
dialkylamino functionality with a diphenylamino group typically increases
the Td of an NLO chromophore in excess of 50 °C without any appreciable
alteration of the molecule’s first hyperpolarizability; such a simple change
of functionality would move our chromophores well within the >200 °C
thermal stability range thought to be crucial for optoelectronic device
applications. In any event, the 1% mass loss is too small to correlate with
any uniform decomposition process and may in fact derive from chemical
reactions occurring at the surface of the material. The next decomposition
process occurs over a 340-420 °C temperature domain and coincides with
an additional 6% mass loss and shows a number of fragments of varying
molecular weight in the mass spectrum. (b) More relevant to the suitability
of these new chromophoric systems to the development of nonlinear
materials is their stability in poled organic films. An optical quality, guest-
host NLO chromophore-polyimide thin film was prepared by imidization
of a 12 weight % solution of 1(ZnII) in polyamic acid [Ultradel 4212
(AMOCO)] at 220 °C using standard methods. Isothermal heating of this
film at 250 ˚C for 30 min showed no changes in the optical spectrum of
the chromophore guest; further heating of the film at 275 °C for an identical
period of time showed a minor (<5%) decrease in the B- and Q-band
absorption intensities of the 1(ZnII) guest. While it is clear that 1(ZnII)
exhibits remarkable thermal stability in a polyimide host, it is an open
question at this time whether this small loss in optical intensity observed at
275 °C derives from a decomposition process or sublimation of the
chromophore. (Cai, Y.-M.; LeCours, S. M.; Therien, M. J.; Jen, K.-Y.,
manuscript in preparation.)
Acknowledgment. M.J.T. is indebted to the Searle Scholars
Program (Chicago Community Trust), the Arnold and Mabel
Beckman Foundation, E. I. du Pont de Nemours, and the
National Science Foundation for Young Investigator Awards,
as well as the Alfred P. Sloan Foundation, the National Institutes
of Health (GM 48130-01A1.4), and the U. S. Department of
Energy (DE-FGO2-94ER14494) for their generous financial
support. C.H.W. thanks the Office of Naval Research and the
National Science Foundation (DMR 912933) for their continued
funding.
JA953610L
(28) (a) Bredas, J. L. Synth. Methods 1987, 17, 115-121. (b) Bre´das, J.
L.; Dory, M.; The´mans, B.; Delhalle, J.; Andre´, J. M. Synth. Methods 1989,
28, D533-D542.