Aggregation of stilbene derivatized fatty acids and phospholipids in monolayers and vesicles

Article abstract of DOI:10.1021/ja972292i

Synthetic fatty acid and phosphatidylcholine amphiphiles incorporating a trans-stilbene (TS) chromophore in the fatty acid chain have been found to exhibit sharp changes in absorption and fluorescence spectra upon self- assembly in Langmuir-Blodgett films

Full text of DOI:10.1021/ja972292i

J. Am. Chem. Soc. 1997, 119, 12481-12491
12481
Aggregation of Stilbene Derivatized Fatty Acids and
Phospholipids in Monolayers and Vesicles1
Xuedong Song,† Cristina Geiger, Mohammad Farahat, Jerry Perlstein, and
David G. Whitten*,‡
Contribution from the Department of Chemistry and NSF Center for Photoinduced Charge
Transfer, UniVersity of Rochester, Rochester, New York 14627
ReceiVed July 10, 1997X
Abstract: Synthetic fatty acid and phosphatidylcholine amphiphiles incorporating a trans-stilbene (TS) chromophore
in the fatty acid chain have been found to exhibit sharp changes in absorption and fluorescence spectra upon self-
assembly in Langmuir-Blodgett films and aqueous dispersions. The spectral changes are readily associated with
aggregates in which there is a strong noncovalent interaction between the TS chromophores. In this paper, we
report determination of the size, structure, and properties of these “supramolecular” aggregates using both experiments
and simulations. Important findings are that the key “unit aggregate” having distinctive spectroscopic properties is
a cyclic “pinwheel” tetramer characterized by strong edge-face interactions. These tetramers may be packed together
to form an extended aggregate with only small changes in absorption or fluorescence properties. While it was
initially expected that aggregation occurred as a consequence of amphiphile self-assembly, studies of films of the
stilbene fatty acids at the air-water interface show the predominance of aggregate prior to compression. Similarly
in phospholipid dispersions aggregates persist above temperature at which chain melting occurs. Aggregation produces
strong effects on the photophysics and photochemistry of the TS chromophore. No photoisomerization occurs;
however, a slow photobleaching is observed for certain assemblies which can be attributed to formation of a photodimer.
Fluorescence from aggregated TS chromophores is attributed to extended aggregates and dimers (excimers), and the
latter is likely responsible for the photodimerization. The supramolecular aggregates observed in this study appear
quite general and closely related to those observed with a wide variety of amphiphiles incorporating aromatic
chromophores.
Introduction
Chart 1
Aggregation of trans-stilbene (TS, either simple2-4 or polar
substituted5-7 ) amphiphiles, especially, trans-stilbene fatty acid
derivatives (SFAs, see Chart 1) in monolayers and Langmuir-
Blodgett (LB) films has been observed in several investigations.
While these compounds can be dispersed in micelles or saturated
phospholipid vesicles at high dilution to give absorption and
fluorescence similar to that observed for the same compounds
in dilute organic solutions, their photophysical behavior in either
pure or mixed (with saturated fatty acids such as arachidic acid)
supported Langmuir-Blodgett (LB) films is quite different.4
In many cases, SFAs form stable monolayers at the water-air
interface which can be transferred onto solid substrates to form
LB films.2 The monolayers and LB films of SFAs exhibit blue-
shifted absorption and red-shifted fluorescence relative to the
spectra in organic solvents. This type of spectral shift has been
attributed to the formation of H-aggregates in which the
chromophores stack in a “card-pack” array.2 Since the SFAs
exhibit film-forming properties (isotherms, limiting area/
† Current address: CST-1, Los Alamos National Laboratory, Los Alamos,
NM 87545.
X Abstract published in AdVance ACS Abstracts, December 1, 1997.
(1) A portion of this research has appeared in preliminary communica-
tions. (a) Song, X.; Geiger, C.; Furman, I.; Whitten, D. G. J. Am. Chem.
Soc. 1994, 116, 4103. (b) Song, X.; Geiger, C.; Leinhos, U.; Perlstein, J.;
Whitten, D. G. J. Am. Chem. Soc. 1994, 116, 10340.
(2) Whitten, D. G. Acc. Chem. Res. 1993, 26, 502.
(3) Spooner, S. P.; Whitten, D. G. Proc. SPTE Int. Soc. Opt. Eng. 1995,
117, 7816.
(4) Mooney, W. F., III.; Brown, P. E.; Russell, J. C.; Costa, S. B.;
Pedersen, L. G.; Whitten, D. G. J. Am. Chem. Soc. 1984, 106, 5659.
(5) Furman, I.; Geiger, H. C.; Whitten, D. G.; Penner, T. L.; Ulman, A.
Langmuir 1994, 10, 837.
(6) Lippitsch, M. E.; Draxler, S.; Koller, E. Thin Solid Films 1992, 217,
162.
(7) Breton, H. L.; Letard, J. F.; Lapovyade, R.; Lecalvez, A.; Rassoul,
R. M.; Freysz, E.; Ducasse, A.; Belin, C.; Morand, J. P. J. Phys. Lett. 1995,
242, 604-616.
molecule, etc.)4 very similar to linear saturated fatty acids such
as arachidate, the formation of a card-pack aggregate was
S0002-7863(97)02292-0 CCC: $14.00 © 1997 American Chemical Society

12482 J. Am. Chem. Soc., Vol. 119, No. 51, 1997
Song et al.
choline (DMPC, 99+%) and L-R-dipalmitoylphosphatidylcholine (DPPC,
99%) were obtained from Sigma Chemical Company. All solvents
for spectroscopic studies were spectroscopic grade from Fisher or
Aldrich. Milli-Q water was obtained by passing in-house distilled water
through a Millipore-RO/UF water purification system. Deuterated
solvents were purchased from MSD Isotopes or Cambridge Isotope
laboratories. 3-(4-Methylbenzoyl)propionic acid was obtained from
Lancaster. 4-(N,N-Dimethylamino)pyridine (DMAP), recrystallized
from chloroform/diethyl ether, and trimethylacetyl chloride were
purchased from Aldrich. Rexyn I-300, mixed resin, is from Fisher
Scientific.
Melting points were taken on an El-Temp II melting point apparatus
and are uncorrected. Proton NMR spectra were recorded on a General
Electric QE300 MHz spectrometer using deuterated solvent locks or
on a 500 MHz Varian VXR-500S spectrometer; coupling constants (J)
are reported in hertz. FAB mass spectra were measured at the Midwest
Center for Mass Spectrometry. Absorption spectra were obtained on
a Hewlett-Packard 8452A diode array spectrophotometer. The circular
dichroism (CD) study was carried out on a JASCO J-710 spectropo-
larimeter. Fluorescence spectra were recorded on a SPEX Flurolog-2
spectrofluorimeter and were uncorrected. Fluorescence quantum yields
(QY) were determined by comparison with a standard trans-stilbene
in methylcyclohexane (QY ) 0.05). Differential scanning calorimetry
(DSC) measurements were carried out on a MC-2 Ultrasensitive
scanning calorimeter from MicroCal, Inc. Size extrusion experiments
for vesicles were performed with an extruder through CoStar nucleopore
polycarbonate filters. Dynamic light scattering measurements were
carried out at the Eastman Kodak Research Laboratories.11 All samples
were routinely filtered through a 1.2 µm nylon syringe filter before
data acquisition. Cryo-transmission electron microscope measurements
(Cryo-TEM) were made at the laboratory of NSF Center for Interfacial
Engineering, Department of Chemical Engineering, University of
Minnesota.12 Specific wavelengths for irradiation were obtained from
a 200 W Mercury/Xenon lamp (Oriel) through a monochromator (10
nm slit).
The general methods used for preparing Langmuir-Blodgett films
and self-assemblies are based on techniques described by Kuhn et al.13
Monolayers of SFAs or SPLs were prepared by spreading a chloroform
solution of the material to be studied onto an aqueous subphase
containing cadmium chloride (2.5 × 10-4 M) and sodium bicarbonate
(3 × 10-5 M, pH ) 6.6-6.8) on a KSV 5000 automatic film balance
at room temperature (23 °C). The monolayers were then transferred
onto quartz substrates. Reflectance spectra for monolayers at the air-
water interface were recorded with a SD1000 fiber optics spectrometer
(Ocean Optics, Inc.) equipped with optical fibers, an LS-1 miniature
tungsten halogen lamp, and a CCD detector. Bilayer vesicles were
prepared according to established protocols (all of the bilayer dispersions
were prepared by sonication).14 A cell disrupter W220F from Heat
Systems Ultrasonics, Inc. (setting 6.5, 35 w) was used for probe-
sonication. Aggregate size measurements and sample preparation for
Cryo-TEM are identical to the procedures9 reported for azobenzene
derivatized phospholipids (APLs).
(B) Fluorescence Lifetime Measurements. Time-correlated single
photon counting experiments were carried out on an instrument
consisting of a mode-locked Nd:YLF laser (Quantronix) operating at
76 MHz as the primary laser source. The second harmonic (KTP
crystal) of the Nd:YLF laser was used to synchronously pump a dye
laser (Coherent 700) circulating Rhodamine 6G in ethylene glycol as
the gain medium. The pulse width of the dye laser was typically 8 ps,
as determined by autocorrelation, and was cavity dumped at a rate of
1.9 MHz. The desired excitation wavelength (295 nm) was obtained
by frequency doubling of the dye laser output in a BBQ crystal.
Emission from the sample was collected by two convex lenses and
initially assumed to be a consequence of forcing the chro-
mophores together in the process of forming the solid LB films
at the air-water interface. A surprising finding was that the
aggregates of SFAs on monolayers and LB films appeared
resistant to dilution with saturated fatty acids such as arachidic
acid.8 The SFAs remain aggregated even when a relatively high
ratio of saturated fatty acids was co-spread on the water surface
such that on a statistical basis little aggregate should be
expected.2,8 These results suggest that the aggregation process
for the SFAs must be energetically favorable in order to
overcome the decrease in entropy caused by aggregation. This
observation is further supported by the finding that mixtures of
different SFAs in monolayers and LB films form H-aggregates
similar to those for pure SFA monolayers and LB films with
no evidence of the occurrence of any mixed aggregates.8 All
of these observations suggested that aggregation of the SFAs
in LB films is a more complex process than initially suspected
and that the aggregates might be more stable and more complex
than a simple sandwich or card-pack as might be proposed from
simple packing considerations.2 In LB films, an apparent phase
separation occurs in which SFAs form small stable aggregate
clusters separated by matrices of the saturated fatty acid. Since
the hydrophobicity of SFAs and saturated fatty acids used for
the dilution are similar, the energy stabilizing the clusters is
attributed to noncovalent interactions between stilbenes, which
must be stronger than simple hydrophobic interaction between
hydrophobic chains. This phenomenon is also observed for
related aromatic compounds such as 1,4-diphenyl-1,3-butadi-
ene,8 1,6-diphenyl-1,3,5-hexatriene,8 azobenzene,9 and squaraine
derivatives.10
From the results obtained from supported multilayers and LB
films of SFAs at the air-water interfaces, important questions
related to chromophore aggregation emerge: what is the
aggregate structure on a molecular level and what is the nature
of the stabilization of these aggregates? Since direct probing
of aggregate structures is extremely difficult, we have applied
several indirect approaches including spectroscopies, conven-
tional solution techniques, and Mount Carlo simulations to
obtain information about the aggregate structures. The difficulty
in manipulating LB films and monolayers, led us to study the
aggregation in aqueous dispersions because of the experimental
convenience and expected structural similarities among bilayer
vesicles, monolayers, and LB films. We have designed and
synthesized several stilbene-derivatized phospholipids (SPLs)
which were anticipated to form bilayer assemblies in aqueous
solution. It was found that the aggregates formed in aqueous
SPL assemblies show similar spectroscopic behavior to those
formed in the monolayers and LB films of SFAs. In this paper,
we discuss the determination of the size and structure of the
aggregates formed from the SPLs as well as their photophysics.
Results of our study based on experiments and simulation
indicate that the key stilbene “unit aggregates” are small and
relatively stable, chiral “pinwheel” structures stabilized by strong
noncovalent interactions.
Experimental Section
(A) Materials and General Techniques. Synthetic reagents were
purchased from Aldrich Chemical Company and used as received unless
otherwise stated. R-, â-, and γ-Cyclodextrins (CDNs, 99+%) were
purchased from Aldrich. Rattlesnake venom, L-R-glycero-3-phospho-
rylcholine as the cadmium chloride complex, lipophilic Sephadex LH-
20 (25-100 µm bead), Sephadex-50G, L-R-dimyristoylphosphatidyl-
(11) Light scattering studies were performed by Dr. Thomas Whitesides
of the Eastman Kodak Research Laboratories.
(12) The cryo-TEM measurements were carried out in the Department
of Chemical Engineering, University of Minnesota, with assistance of Dr.
Michael Bench.
(13) Kuhn, H.; Moˆbius, D.; Bo¨cher, H. In Physical Methods of Chemistry;
Weissberger, A., Rossiter, B. W., Eds.; Wiley: New York, 1972; Vol. 1, p
577.
(8) Spooner, S. P. Ph.D. Dissertation, University of Rochester, 1993.
(9) (a) Song, X.; Perlstein, J.; Whitten, D. G. J. Am. Chem. Soc. 1995,
117, 7816. (b) Song, X.; Perlstein, J.; Whitten, D. G. J. Am. Chem. Soc.
1997, 119, 9144.
(10) (a) Chen, H.; Law, K. Y.; Whitten, D. G. J. Am. Chem. Soc. 1995,
117, 7257. (b) Chen, H.; Farahat, M. S.; Law, K. Y.; Perlstein, J.; Whitten,
D. G. J. Am. Chem. Soc. 1996, 118, 2586.
(14) (a) Hope, M. J.; Bally, M. B.; Webb, G.; Cullis, P. R. Biophys.
Acta 1985, 55, 812. (b) Saunders, L.; Perrin, J.; Gammock, D. B. J. Pharm.
Pharmacol. 1962, 14, 567.

Aggregation of Stilbene-DeriVatized Fatty Acids
J. Am. Chem. Soc., Vol. 119, No. 51, 1997 12483
focused at the entrance slit of a Spex 1681 monochromator (0.22m)
and was detected by a multichannel plate (MCP) detector (Hamamtsu
R3809U-01). The single photon pulses from the MCP detector were
amplified and used as the stop signal for a time to amplitude converter
(TCA, EG&G Ortec) while the signal from a photodiode, detecting a
small fraction of the dye laser output, was used as the starting signal
for the TCA. The starting and stop signals for the TCA were
conditioned before entering the TCA by passing through two separate
channels of a constant fraction discriminator (CFD, Tennelec). The
output of the TCA was connected to a multichannel analyzer (MCA)
interface board (Norland 5000) installed inside a 486DX2 personal
computer. The MCA was controlled by software from Edinburgh
Instruments (Edinburgh, U.K.). The same software was used to carry
out the deconvolution of the data and exponential fitting using a
nonlinear least-squares method. Measurements were made with air-
saturated samples.
3.34 (s, 9H), 3.80-4.40 (m, 8H), 5.20 (brs, 1H), 7.00-7.50 (m, 22H).
FAB: m/z calcd for [M + 1]+ ) 810.4, found 810.4.
mix-4S6,S10EPC. bis-4S6EPC (100 mg, 0.11 mmol) was dissolved
in 20 mL of CH2Cl2/MeOH (99:1, v/v), followed by addition of 10
mL of an aqueous solution containing 20 mM Tris-HCl (pH 8.0), 40
mM CaCl2, and 2 mg of rattlesnake venom (Crotalus adamanteous).
The reaction vessel was stirred vigorously in the dark for 3 days. When
the reaction was complete, the organic solvent was removed by a stream
of nitrogen and the lyso product was extracted by the Bligh-Dyer
procedure.15d The pure product was obtained by passing a Sephadex
LH-20 column using chloroform as eluent. mono-4S6EPC was obtained
in 65% yield. 1H NMR (CDCl3) δ: 0.86 (t, 3H), 1.32 (m, 4H), 1.56
(m, 6H), 2.28 (t, 2H), 2.55 (t, 4H), 3.14 (s, 9H), 3.46-4.28 (m, 10H).
6.95-7.4 (m, 10H).
mono-4S6EPC (39 mg, 1.0 equiv) of obtained above and DMAP (16
mg, 2.0 equiv) DMAP were dissolved in 3 mL of CH2Cl2, followed by
addition of 66 mg of S10-Piv (2.0 equiv) in 3 mL of CH2Cl2 prepared
from S10A and pivaloyl chloride (Piv-Cl). The reaction mixture was
stirred at room temperature for 48 h. The solvent was then removed,
and the product was extracted by the Bligh-Dyer procedure substituting
1 M HCl for water to remove the DMAP. The product was purified
using a Sephadex LH-20 column with chloroform as eluent. Yield:
40%; mp ) 164-182 °C. 1H NMR (CDCl3) δSPCLN 0.95 (t, 3H),
1.30 (m, 14H), 1.62 (m, 10H), 2.30 (t, 4H), 2.60 (m, 6H), 3.38 (s, 9H),
3.80-4.60 (m, 8H), 5.22, (m, 1H), 7.0-7.52 (m, 21H). FAB: m/z
calcd for [M + 1]+ ) 922.5, found 922.5.
(C) Synthesis. The nomenclature and synthetic schemes used for
preparation of the SPLs in this study are shown in Chart 1. The general
synthetic approach is based on methods previously described in the
literature.15 These SPLs spontaneously form bilayer assemblies in
aqueous solutions as do their natural counterparts such as DPPC and
DMPC. For comparison, mono-SPLs were designed to prevent the
formation of extended chromophore aggregates. The mixed bis-SPLs
were originally expected to facilitate the formation of J-aggregates
through the mismatch of the stilbenes in the two fatty acid chains.
Instead, H-aggregates are exclusively formed.
bis-4S6EPC. 4S6A (0.44 mmol, 1.0 equiv) was dissolved in meth-
ylene chloride (20 mg/mL) and dry triethylamine (2.0 equiv) was added
via syringe, followed by trimethylacetyl chloride (Piv-Cl, 4.0 equiv).
TLC (solvent B, CHCl3/methanol/water ) 65/25/4) showed that the
reaction was complete after ca. 5 h of stirring. After the solvent was
removed with a stream of nitrogen, the residue was dried in a vaccum
dessicator for 2 h and then dissolved in methylene chloride (20 mg/
mL), followed by addition of a suspension of L-R-glycero-3-phospho-
rylcholine as the cadmium chloride complex (GPC‚CdCl2, 1.0 equiv)
and 4-(N,N-dimethylamino)pyridine (DMAP, 2.0 equiv) in 2 mL of
methylene chloride. The reaction mixture was stirred at room tem-
perature for ca. 48 h. The solvent was removed by rotary evaporation,
and the residue was dissolved in solvent A (methanol/chloroform/water
) 5/4/1). The solution was filtered and passed through an ion exchange
column, Rexyn-I 300, to remove the CdCl2 and DMAP using solvent
B as eluent. After the solvent was removed, the crude product was
purified by passing it through a sephadex LH-20 column eluted with
chloroform followed by recrystallization with chloroform/ether.
Yield: 45%; mp ) 228-234 °C. 1H NMR (CDCl3) δSPCLN 0.92 (t,
6H), 1.36 (m, 8H), 1.61 (m, 12H), 2.30 (m, 4H), 2.61 (m, 8H), 3.36 (s,
9H), 3.73-4.48 (m, 8H), 5.23 (m, 1H), 7.0-7.5 (m, 20H). FAB: m/z
calcd for [M + H]+ ) 922.5, found 922.5.
bis-S4EPC. The preparation procedure is identical to that for bis-
4S6EPC. Yield: 51%. 1H NMR (500 MHz, CDCl3) δ: 7.49 (d, J )
7.5, 4H), 7.42 (dd, J ) 7.5, 2.0, 4H), 7.35 (t, J ) 7.5, 4H), 7.27 (overlap
with peak of CDCl3, 4H), 7.15 (dd, J ) 7.0, 4.5, 4H), 7.06 (s, 4H),
5.25 (m, 1H), 4.38 (m, 3H), 4.16 (dd, J ) 7.5, 5.0, 1H), 4.04 (t, J )
6.5, 2H), 3.85 (m, 2H), 3.32 (s, 9H), 2.62 (m, 4H), 2.34 (m, 4H), 1.92
(m, 4H). FAB: m/z calcd for [M + H]+ ) 754.3, found 754.3.
bis-4S4EPC. The synthesis of this compound is identical to that for
bis-4S6EPC. Yield: 46%; mp ) 230-237 °C. 1H NMR (CDCl3)
δSPCLN 0.95 (t, 6H), 1.38 (m, 4H), 1.61 (m, 4H), 1.96 (m, 4H), 2.34
(m, 4H), 2.64 (t, 8H), 3.30 (s, 9H), 3.64-4.50 (m, 8H), 5.26 (m, 1H),
7.0-7.5 (m, 20H). FAB: m/z calcd for [M + 1]+ 866.4, found 866.4.
mix-4S6,S12EPC. The synthesis is identical to that for mix-4S6,S10-
EPC. Yield: 66%; mp ) 260-268 °C. 1H-NMR (CDCl3) δ: 0.94 (t,
3H), 1.26 (s, 18H), 1.60 (m, 10H), 2.32 (s, 6H), 2.61 (m, 6H), 3.38 (s,
9H), 3.82-4.56 (m, 8H), 5.22 (m, 1H), 7.10-7.55 (m, 21H). FAB:
m/z calcd for [M + H]+ ) 949.5, found 952.5.
mix-4S6EPPC. The preparation procedure is similar to that for mix-
4S6,S10EPC. Yield: 44%; mp ) 184-188 °C. 1H NMR (CDCl3)
δSPCLN 0.86-0.99 (m, 6H), 1.26 (s, 24H), 1.34-1.72 (m, 12H), 2.25-
2.66 (m, 8H), 3.43 (s, 9H), 4.0-4.60 (m, 8H), 5.24 (m, 1H), 7.0-7.5
(m, 10H). FAB: m/z calcd for [M + H]+ ) 828.5, found: 828.4.
mix-S4EMPC. The synthesis is similar to that for mix-4S6,S10EPC.
Yield: 51%. 1H NMR (300 MHz, CDCl3) δ: 0.90 (t, 3H), 1.24 (m,
18H), 1.56 (m, 2H), 1.93 (m, 2H), 2.28 (t, 2H), 2.36 (t, 2H), 2.65 (t,
2H), 3.32 (s, 9H), 3.68-4.44 (m, 8H), 5.23 (m, 1H), 7.06-7.54 (m,
11H). FAB: m/z calcd for [M + H]+) 716.4, found 716.3.
Results and Discussion
All of the trans-stilbene (TS)-derivatized phospholipids
(SPLs) including bis-SPLs, mono-SPLs, and mix-SPLs are
readily soluble in organic solvents such as methylene chloride
and chloroform to give solutions with absorption λmax at 315
nm and fluorescence λmax at 356 nm, which are characteristic
of the TS monomer. The behavior is similar to that observed
previously for SFAs in organic solvents.2 The presence of
normal “monomeric” absorption and fluorescence spectra for
the SPLs suggests there is little or no tendency for the TS
chromophores to associate in those solvents in either the ground
or excited state, despite the close proximity of the chromophores
in the phospholipids containing two TS units. The SPLs can
be dispersed in water by probe sonication to form stable, clear
aqueous solutions. In contrast to the “monomeric” spectra
observed in organic solvents, the aqueous dispersions of SPLs
show blue-shifted absorption and red-shifted fluorescence
spectra which are attributed to the formation of TS H-aggregates
similar to those observed in LB films2 of SFAs and SPLs. To
a first approximation the spectral shifts suggest the TS chro-
mophores in the H-aggregates pack in a “card-pack” array where
a new exciton band results from excitonic interaction between
the transition dipole moments of the TS chromophores.16 The
only allowed transition for a “perfect” H-aggregate card-pack
is to the highest energy level of the exciton band from the
ground state, which results in a blue-shift in absorption spectra
relative to the corresponding monomer. The presence of red-
shifted fluorescence could be ascribed to emission from the
bis-S10EPC. The preparation procedure is identical to that for bis-
4S6EPC. Yield: 84%; mp ) 184-190 °C. 1H NMR (CDCl3) δ: 1.30
(m, 20H), 1.62 (m, 8H), 2.32 (t, 4H), 2.63 (t, 4H), 3.38 (s, 9H), 3.70-
4.50 (m, 8H), 5.23 (m, 1H), 7.0-7.6 (m, 22H). FAB: m/z calcd for
[M + 1]+ ) 922.5, found 922.5.
bis-S6EPC. The sythesis of this compound is identical to that for
bis-4S6EPC. Yield: 42%; 1H NMR (CDCl3) δSPCLN 1.30-1.36 (m,
4H), 1.55-1.64 (m, 8H), 2.24-2.30 (t, 4H), 2.53-2.60 (t, 4H), 3.24-
(15) (a) Kates, M. In Methods in Membrane Biology Korn, E. D., Ed.;
Plenum: New York, 1977; Vol. 8, p 219. (b) Khorana, H. G.; Chakrabarti,
P. Biochemistry 1975, 14, 5021. (c) Bligh, E. G.; Dyer, W. J. Can. J.
Biochem. Physiol. 1959, 37, 911.

12484 J. Am. Chem. Soc., Vol. 119, No. 51, 1997
Song et al.
Table 1. Absorption and Fluorescence Data for SPLs in Different
Media
chloroform (nm)
λabs λem
water (nm)
shift (nm)d
λabs λem ∆λabs ∆λem (wn)a
max
FWHMb
SPLs
4S6A
max
max
max
315
356
268c
406c
47
33
33
39
45
43
41
41
44
37
50
9
S4EMPC
4S6EPPC
S4EPC
315
315
315
315
316
320
320
318
286
356
356
356
356
358
360
366
365
366
282
282
276
270
273
274
274
272
278
365
391
423
395
389
417
400
418
420
6340
6350
4830
4210
4640
4180
3780
6800
6690
35
67
39
33
61
44
62
64
S6EPC
S10EPC
4S4EPC
4S6EPC
4S6, S10EPC
4S6, S12EPC
a Wavenumber. b Full width at half maximum. c Data in LB film on
quartz. d Relative to S6A in chloroform.
4
of spectral shifts depends upon the structure of the individual
molecule as shown in Table 1. Absorption spectra of bis-SPLs
in water are always sharper and more blue-shifted than those
for mono-SPLs, which have relatively broad spectra as indicated
by their large full width at half maximum (FWHM). The
broader, less blue-shifted absorption spectra of mono-SPLs are
attributed to the perturbation of the saturated fatty acid chain,
which should inhibit the formation of highly extended ag-
gregates. The absorption spectra of mix-SPLs vary with their
molecular structures. The absorption spectrum of mix-4S6,S10-
EPC, in which the two TS chromophores have a four-methylene
offset, is particularly sharp with a large blue-shift in water
relative to the monomer, while mix-4S6,S12EPC, which has a
six methylene group mismatch for the two TS chromophores,
shows a broad absorption spectrum with a relatively small blue-
shift. While the excitation spectrum of the monomer in an
organic solution of SPL is identical to the corresponding
absorption spectrum, excitation spectra of SPL assemblies in
water are slightly different from their absorption spectra and
show some emission wavelength dependence, particularly when
monitored at the red edge of the fluorescence band. This is
attributed to the microheterogeous nature of the bilayer system.
The SPLs, moderately or highly diluted in DPPC or DMPC
vesicles, display excitation spectra similar to their corresponding
absorption spectra. As expected, they are also wavelength-
dependent. SFAs and mono-SPLs in the presence of excess
DMPC or DPPC show identical monomeric absorption and
excitation spectra. The bis-SPLs in DPPC vesicles with a low
dilution ratio of [DPPC]/[bis-SPL] (for example, 5) show
excitation spectra intermediate between those of pure bis-SPL
aggregates and TS monomer and exhibit emission wavelength-
dependence. They can actually be fit into a summation of a
mixture of dimers (or monomers in case of mono-SPLs) and
H-aggregates.
The absorption and fluorescence spectra for bis-S4EPC and
bis-S6EPC assemblies in water persist over a broad range of
temperatures from 5-80 °C with the exception of a gradual
decrease in fluorescence intensity. This implies that the TS
chromophores in the aggregates experience little change in
packing geometry even at temperatures above Tc values at which
the polymethylene chains melt. The persistence of the ag-
gregates above chain-melting reinforces the idea of a relatively
high stability for the TS aggregates and is supported by the
observation that the H-aggregates in the monolayers and LB
films of SFAs resist relatively high dilution with saturated fatty
acids.16
Fluorescence. The fluorescence spectra associated with SPL
assemblies in water are relatively complicated and can be
assigned as due primarily to two different species: an aggregate
and an excimer.24 The aggregate has a higher fluorescence
quantum yield (QY) and a lifetime of 3-10 ns, while the
excimer has a lower fluorescence QY and a lifetime that is
Figure 1. Absorption and fluorescence spectra of bis-S10EPC in water,
methylene chloride, and DMPC vesicles (DMPC/bis-S10EPC ) 100).
forbidden low-energy exciton band of the H-aggregates.16 The
SPLs can also be codispersed with an excess of the saturated
phospholipids such as DPPC or DMPC in water to form mixed
vesicles which exhibit absorption and fluorescence spectra
similar to those of the monomers in organic solvents, but broader
and less structured. The broader absorption spectra are assigned
to a TS dimer on the bassis of the study of SFA and SPL
complexes with cyclodextrins45(CDNs, see below) as well as
the results reported by others.17,18 The broadening of the
absorption spectrum can be ascribed to a small excitonic splitting
for an oblique dimer.35b Although the absorption spectra of
SPLs, in the presence of excess saturated phospholipid and SFA
or SPLs in γ-CDN are similar, the fluorescence spectra are quite
different with the latter showing a broad “excimer-like”
fluorescence. Evidently, the restricted environment of DMPC
and DPPC vesicles inhibits the formation of an TS excimer for
SPLs while the two TS chromophores of SFAs or SPLs in
γ-CDN are probably in a sufficiently flexible environment to
adopt a geometry required for one excited TS to interact with
another TS in the ground state to give the red-shifted excimer
fluorescence. Compared to the broad absorption spectra of bis-
SPLs in DPPC or DMPC, the absorption spectra of the mono-
SPLs in the matrix of DPPC or DMPC vesicles are sharper,
slightly red-shifted (4 nm) and with clear vibrational peaks
similar to those of SFAs solubilized in DPPC and DMPC
vesicles. The sharp fine-structured absorption spectra are
associated with the TS monomer. The absorption and fluores-
cence spectra of bis-S10EPC in water, CH2Cl2, and DMPC are
shown in Figure 1, which are assigned to H-aggregate,
monomer, and dimer, respectively. Since the spectrum in water
is unaffected by dilution up to several-fold with DMPC, we
attribute this spectrum to pure aggregate; similarly the spectrum
in chloroform is identical to “monomer” alkylstilbene and thus
attributed to pure monomer. The “dimer” spectrum in DMPC/
water probably consists of a mixture of monomer, aggregate,
and dimer, but the dimer predominates.
Although all SPL assemblies investigated in this paper show
blue-shifted absorption and red-shifted fluorescence in water
relative to the monomer in organic solvents, the precise extent
(16) (a) Kasha, M. Radiat. Res. 1963, 20, 55. (b) Kasha, M.; Rawls, H.
R.; El-Bayoumi, M. Pure Appl. Chem. 1965, 11, 371. (c) Hochstrasser, R.
M.; Kasha, M. Photochem. Photobiol. 1964, 3, 317.
(17) Lewis, F. D.; Wu, T.; Burch, E. L.; Bassani, D. M.; Yang, J.;
Schneider, S.; Jager, W.; Letsinger, R. L. J. Am. Chem. Soc. 1995, 117,
8785-8792.
(18) Anger, I.; Sandros, K.; Sundahl, M.; Wennerstrom, O. J. Phys. Chem.
1993, 97, 1920-1923.

Aggregation of Stilbene-DeriVatized Fatty Acids
J. Am. Chem. Soc., Vol. 119, No. 51, 1997 12485
considering the two following possible mechanisms: (1) An
energy transfer process from the major species, the aggregates,
to residual dimers occurs in the systems. The small portion of
dimers, which is not evident in the absorption spectra, can be,
in some cases, a very efficient energy trap for the excitation
created in the aggregates. Similar cases have been observed
previously for both H- and J-aggregate systems19-21 in organized
assemblies. In fact, there is evidence for the formation of an
excimer state for aromatic clusters that shows weak excitonic
interaction in the ground state22 (for this reason, the terms, dimer
and excimer, are used interchangably in this paper). It appears
that the structure of the SPL is responsible for the observed
photophysical and photochemical (see photolysis section) varia-
tions. Both hydrophobic interaction of fatty acid chains and
chromophore-chromophore interaction of TS play important
roles in determining the aggregate structures. In a molecule
such as bis-S10EPC, where TS is situated relatively far away
from the head group, the formation of highly ordered and
uniform TS aggregates is favored probably due to the long fatty
acid chains which are flexible enough to allow optimal geometry
for the chromophore-chromophore interaction. Thus, the
fluorescence originates mainly from the radiative decay of the
excited states of the H-aggregates (unit aggregates). In contrast,
the population of the dimer is expected to be relatively large
for a molecule such as bis-S4EPC in which the TS is close to
the rigid head group. In this case, efficient energy transfer from
the H-aggregates to the excimers occurs, and the fluorescence
is generated directly from the TS excimer. Such efficient energy
transfer from H-aggregates to excimers has also been observed
for styrylthiophene23 derivatives as well as for other systems.19-21
In the cases of bis-4S6EPC, bis-S6EPC, and mix-S4EMPC, some
energy transfer process occurs, which results in the intermediate
fluorescence spectra. The dependence of the H-aggregation
upon molecular structures has also been observed for azoben-
zene9 and styrylthiophene derivatives.23 (2) A variation in
geometry after excitation to the excitonic state of the aggregate
compared to the ground state geometry is responsible for the
excimer fluorescence. In the case of bis-S4EPC, due to the
relatively small aggregate size, a small change in geometry is
sufficient to allow excimer formation or even dimerization. The
latter process is expected to be an inefficient process due to
more demanding structural requirements for [2 + 2] cycload-
dition in a restricted environment.43 For bis-S10EPC assemblies,
the large, highly ordered and rigid aggregate appears to enhance
the structural integrity of the TS aggregate to the point that even
a small excited state structural change becomes unfeasible;
therefore, excimer formation and photodimerization processes
become inefficient.
Figure 2. Comparism of fluorescence spectra of SPL assemblies in
water, excited at 270 nm.
Table 2. Fluorescence Lifetimes and Quantum Yields for SPLs in
Water
phospholipid
τ3 (ns)a
τ4 (ns)a
φf
mix-S4EMPC
bis-S4EPC
2.7 (91%)b
3.4 (13%)
2.7 (35%)
3.2 (74%)
3.4 (10%)
4.1 (52%)
23.3 (9%)
22.6 (87%)
18.5 (64%)
17.8 (26%)
18.1 (90%)
15.9 (42%)
0.10
0.74
bis-S6EPC
bis-S10EPC
DPPC/bis-S10EPCc
bis-4S6EPC
0.09
0.43
0.10
0.10
mix-4S6EPPC
mix-4S6,S10EPC
bis-4S4EPC
a The data were obtained from time resolved fluorescence decay
curves at 500 nm emission wavelength. b Values in parenthesis are
contributions of the various decays to the total fluorescence intensity.
c DPPC/S10EPC ) 110.
longer than 10 ns. The lifetimes for both species are longer
than the radiative lifetime for TS (1.7 ns43). The variations in
the excited state properties for the present systems originate
from the differences in the structures of the molecules despite
the presence of TS in all molecules as the photophysically active
chromophore in the spectral region of interest. Figure 2
compares the fluorescence spectra of SPL assemblies in water.
The fluorescence spectrum of bis-S4EPC, as well as that of mix-
4S6,S10EPC, is broad and structureless with a large red-shift and
low fluorescence QY (Table 2), which is attributed to an excimer
type fluorescence, consistent with its structureless appearance
and the low fluorescence efficiency. In contrast, bis-S10EPC
and bis-4S6EPPC assemblies give sharper, fine-structured, and
relatively less red-shifted fluorescence with relatively high
fluorescence QY, which is assigned to the aggregate. These
two extreme cases of fluorescence spectra show insignificant
excitation wavelength dependence. Several other SPLs such
as bis-4S6EPC, bis-S6EPC, and mix-S4EMPC show intermediate
fluorescence characteristic
Fluorescence Lifetime.24 Due to the microheterogeous
nature in SPL assemblies, a distribution of fluorescence lifetimes
instead of discrete lifetimes are found.24 The distribution fi was
characterized by its center lifetime τi, standard deviation σi and
amplitude of the center lifetime Ai. The contribution of each
lifetime distribution to the total fluorescence intensity at a given
fluorescence wavelength can be estimated via eq 1.
f ) τ σ A / σ τ A
i i i
(1)
∑
i
i
i
i
The significant contribution of an excimer to the total
fluorescence for several SPL assemblies can be rationalized by
i
The fluorescence lifetime (0.25 ns) and quantum yield (0.09)
for bis-S4EPC in methylene chloride are found to be similar to
those for SFAs and simple TS in the same organic solvent.2
However, the fluorescence decays of SPL assemblies in water
are rather complicated and, in most of cases, can be fitted with
four different lifetime distributions as shown in Figure 3 for
bis-S4EPC. The two short lifetimes for all SPLs are almost
identical and center at τ1 ) 0.27 ns and τ2 ) 1.4 ns. Their
(19) (a) Evans, C. E.; Bohn, P. W. J. Am. Chem. Soc. 1993, 115, 3306.
(b) Furman, I.; Whitten, D. G.; Penner, T. L.; Ulman, A. Langmuir 1994,
10, 837.
(20) Farahat, C. W.; Penner, T. P.; Ulman, A.; Whitten, D. G. J. Phys.
Chem. 1996, 100, 12616.
(21) Kuhn, H. J. Photochem. 1979, 10, 111.
(22) Saigusa, H.; Lim, E. C. J. Phys. Chem. 1995, 99, 15738-15747.
(23) Song, X.; Perlstein, J.; Whitten, D. G. Submitted for publication.
(24) Whitten, D. G.; Farahat, M. S.; Gaillard, E. R. Photochem. Photobiol.
1997, 55 (1), 23.

12486 J. Am. Chem. Soc., Vol. 119, No. 51, 1997
Song et al.
Figure 3. Fluorescence decay curve at 450 nm emission wavelength and lifetime distribution for bis-S4EPC assemblies in water (excited at 300
nm).
fractional contributions to the total fluorescence are negligible
(<5%). The shortest lifetime τ1 is assigned to a nonconstrained
TS monomer on the basis of its similar lifetime to that of 4S6A
in CH2Cl2 (τ ) 0.24 ns), and the negligible second shortest
one τ2 is attributed to a constrained TS monomer in the
the aggregate and the relative contribution of the aggregats and
excimer. For bis-S4EPC assemblies where small aggregates are
formed, the high contribution of excimer decay seems to support
an efficient energy transfer from the H-aggregates to the
excimers, while the fluorescence of bis-S10EPC assemblies,
where more uniform aggregates are expected, shows a relatively
low contribution of the excimer.
assemblies which is comparable to that of S6A in DPPC (1.8
4
ns) and in R-CDN (1.4 ns). The two long components, τ3 and
τ4 are assigned to a TS H-aggregate and a TS excimer,
respectively. The assignment of τ4 to the TS excimer is based
on the reported results for TS excimers (11 ns)17 and our studies
(see below) of SFAs in γ-CDN (5.3-17.6 ns).18 The increase
in lifetime for the H-aggregate relative to the monomer in
organic solution is attributed mostly to the forbidden nature
(symmetry)16a of the radiative transition from the lowest energy
level of exciton band to ground state of H-aggregates as well
as the restrictive matrices of vesicles which provide a large
barrier for a competitive, nonradiative decay, namely, trans-
cis isomerization. The contribution of the four emitting species
to the total fluorescence intensity is fluorescence wavelength
dependent for all SPLs. An increasing contribution of τ1 can
be detected at the blue-edge of the fluorescence whereas the
major contributor at the peak of fluorescence is excimer. The
lifetime data for other SPLs are summarized in Table 2 , which
clearly indicate that both the total length of the aliphatic chains
and the position of the TS chromophores in the aliphatic chains
have effects on the characteristics of the fluorescence efficiency
and lifetimes. The evidence for the above lifetime assignments
also emerges from the dependence of the contribution of each
lifetime on dilution by DPPC vesicles. While the contribution
of two fast components change little with different dilution, the
excimer component increases at the expense of the aggregate
component when bis-S10EPC is diluted in DPPC vesicles
(DPPC/bis-S10EPC ) 110).
Complexes with Cyclodextrins (CDNs). Solubility of
several SFAs in water improves significantly in the presence
of R-, â-, and γ-CDNs due to the formation of host-guest
complexes.25 The absorption spectrum (see the Supporting
Information) of 4S6A complex with R-CDN is sharper than that
with â-CDN, which exhibits an absorption spectrum similar to
that for 4S6A in methylene chloride. The fluorescence spectra
in R- and â-CDNs and methylene chloride are identical and
are associated with TS monomer. The sharper spectrum in
R-CDN suggests that the TS chromophore is probably close or
even inside the hydrophobic cavity of R-CDN. In contrast, a
broad absorption spectrum similar to that for SPLs highly diluted
in DMPC or DPPC vesicles and a structureless, red-shifted
fluorescence spectrum relative to the monomer are observed
for S6A in γ-CDN, which are assigned to a TS dimer (or
4
excimer). The absence of a red-shift for the fluorescence of
SPLs in excess of DPPC or DMPC vesicles is probably due to
the restricted environment, which prevents the formation of an
excimer.19 In most of cases, the fluorescence quantum yields
(see the Supporting Information) of TSA’s in R-, â- and γ-CDNs
are higher than in CH2Cl2, due to the formation of the complexes
which to some extent inhibits the photoisomerization of TS.
The fluorescence decay of SFAs in R-CDN follows simple
monoexponential decay while the fluorescence decays in â- and
γ-CDNs are quite complicated and can be fit into either two or
three components (see the Supporting Information). This
suggests that in the latter cases, there exist more than one
It is interesting to compare the contributions of the H-
aggregate and excimer decays to the total fluorescence intensity
for four bis-SPLs: bis-S4EPC, bis-S6EPC, bis-4S6EPC and bis-
S10EPC. There exists a clear consistency between the order of
(25) (a) Herkstroeter, W. G.; Martic, P. A.; Evans, T. R.; Farid, S. J.
Am. Chem. Soc. 1986, 108, 3275. (b) Herkstroeter, W. G.; Martic, P. A.;
Farid, S. J. Am. Chem. Soc. 1990, 112, 3583-3589.

Aggregation of Stilbene-DeriVatized Fatty Acids
J. Am. Chem. Soc., Vol. 119, No. 51, 1997 12487
emitting species in which the TS chromophores probably adopt
different positions in the cavities of CDNs. In the case of
γ-CDN, there may be both the monomer and the dimer. We
suggest that an excimer or dimer is responsible for the long
fluorescence lifetime in γ-CDNs, proposed to be an excimer.
The long fluorescence lifetime for TS excimer has been reported
previously by Lewis and co-workers.17 The formation of the
monomeric complexes of TSA’s with R-, â-CDNs and a dimer
in γ-CDN are further supported by the observation of moderate
monophasic ICD spectra in R- and â-CDNs and biphasic ICD
spectra in γ-CDN, which is characteristic of an oblique dimer
(see the Supporting Information). The asymmetric feature of
ICD in γ-CDN can be attributed to a mixture of both monomer
and dimer in the solution as indicated by the broad fluorescence
spectrum. Interestingly, the ICD signals for SFAs in all three
CDNs disappear upon trans-cis photoisomerization of the TS,
indicative of exclusion of the cis-stilbenes from the cavities of
CDNs.
The tendency of self-aggregation for SPLs are much higher
than the formation of inclusion complexes in aqueous solutions
of R-, â-, and γ-CDNs. Addition of acetonitrile to the aqueous
solutions favors the complexation over self-aggregation. The
mixture of the aggregate, dimer, and monomer in a mixed
solvent of acetonitrile/water (1/2) for SPLs can be converted
into a monomer-dominating solution by addition of either R-
or â-CDNs and a dimer-prevailing solution by addition of
γ-CDN.
Characterization of SPL Assemblies. Three techniques
dynamic light scattering, membrane extrusion, and cryo-
transmission electron microscopy (Cryo-TEM), have been used
to characterize SPL assemblies in water. Light-scattering data
for aqueous dispersions of SPLs show a relatively narrow
distribution of particle sizes (based on assumption of spherical
particles), which are much larger than those formed from
saturated phospholipids such as DPPC (bis-4S4EPC: 521 nm;
bis-S4EPC: 198 nm; bis-4S4EPC: 243 nm). The particle size
for aqueous dispersions of bis-S4EPC changes little from 10 to
55 °C. Incorporation of bis-4S4EPC in DPPC vesicles slightly
increases the size of DPPC vesicles. Consistent with the results
obtained from light scattering measurements, the self-assemblies
of SPLs in water can not be extruded through a 100 nm
polycarbonate membrane but can completely pass through a 2
µm membrane. The Cryo-TEM micrograph (Figure 4) of bis-
S4EPC assemblies in water shows rather interesting structures
with two major morphologies: large open tubules covered with
smaller pieces or fragments and small open-ended tubules. These
are assumed to be bilayer structures. As expected from such
open structures, the pure S4EPC assemblies in water failed to
trap any carboxyfluorescein (CF).14,26
Figure 4. Cryo-TEM micrograph of bis-S4EPC assemblies in water
(3 days old, 2 × 10-4 M).
water were found to show sharp decrease at certain temperatures
(23 and 65 °C for bis-S4EPC and bis-S6EPC, respectively) at
which phase transitions presumably occur. The light scattering
intensity for bis-S4EPC also shows a discontinuity at 23 °C,
which is believed to be associated with a phase transition. DSC
studies for the mixtures of SPL/DPPC show good solubilization
of SPLs in DPPC vesicle hosts. The endothermic maxima for
the mixtures of several SPLs with DPPC (SPL/DPPC ) 1/100)
are shifted to lower temperatures (1-2 °C) and are slightly
broadened, compared to those for to pure DPPC vesicles (Tc )
42 °C). Such a decrease in Tc of DPPC vesicles is expected
from the small perturbation caused by the incorporation of the
foreign SPLs.
Aggregate Size. SPL assemblies in water undergo rapid
mixing with DMPC or DPPC vesicles at or above Tc values of
both microphases but only very slowly below Tc. We have
reported previously1 that a clean, simple conversion from the
aggregates to dimers (monomers in the cases of mono-SPLs)
occurs when SPL aggregates are mixed with DMPC or DPPC
vesicles above Tc.1 We also demonstrated that the aggregates
and dimers or monomers in cases of mono-SPLs are two major
species in the equilibrium solutions of SPLs moderately diluted
in DPPC or DMPC vesicles. The equilibrium between the
aggregates and dimers or monomers can be expressed by eq 2.
Further evidence for the equilibrium emerges from the ICD
spectra for the mixtures of DPPC or DMPC/SPLs with low
dilution ratio from 5 to 35, which are identical to the ICD spectra
of pure SPL assemblies in water (see below). A modified
Benisi-Hildebrand approach28 identical to the treatment of
APLs (azobenzene-derivatized phospholipids)9 and TPLs (styryl-
thiophene-derivatized phospholipids)23 is used to formulate the
equilibrium, and eq 5 is obtained in terms of absorbance of the
aggregates and dimers or monomers. The plot of ln Aagg versus
ln[Aoagg - Aagg] should give a straight line while the slope allows
determination of the aggregate size, n, which is the number of
SPL molecules per aggregate.
Differential scaning calorimetry (DSC), dynamic light scat-
tering, and fluorescence techniques were applied to investigate
phase transition behaviors of SPL assemblies in water. No
phase transition for bis-SPLs could be detected by DSC, possibly
due to their low solubility in water (less than 1.5 × 10-4 M).
Fluorescence spectroscopy provides a useful tool to study phase
transitions of organized systems. The fluorescence spectra of
SPL aggregates were found to persist within a wide range of
temperatures from 5-80 °C with the exception of a decrease
in intensity. The fluorescence intensity for SPL assemblies in
(26) (a) Weinstein, J. N.; Yoshikami, S.; Henkart, P.; Blumenthal, R.;
Hagins, W. A. Science, 1977, 195, 489. (b) Liu, Y.; Regen, S. L. J. Am.
Chem. Soc. 1993, 115, 708. (c) Nagawa, Y.; Regen, S. L. J. Am. Chem.
Soc. 1992, 114, 1668.
(27) (a) Georgecauld, D.; Desmasez, J. P.; Lapouyade, R.; Babeau, A.;
Richard, H.; Winnik, M. Photochem. Photobiol. 1980, 31, 539-545. (b)
Martin, F. J.; MacDonald, R. C. Biochemistry 1976, 15, 321. (c) Kremer,
J. M. H.; Kops-Werkhoven, M. M.; Pathmamanoharan, C.; Gijzemen, O.
L. G.; Wiersema, P. H. Biochim. Biophys. Acta 1977, 471, 177-188.
(28) Benesi, H. A.; Hildebrand, J. H. J. Am. Chem. Soc. 1949, 71, 2703.

12488 J. Am. Chem. Soc., Vol. 119, No. 51, 1997
Aggn ) nDimer (or nMonomer for mono-SPLs) (2)
Song et al.
K ) [Di]ndmpc/[Agg]dmpc
(3)
(4)
ln[An /Aagg] ) (n - 1)ln[DMPC] + C
di
In the case of constant [DMPC],
ln Aagg ) n ln[Aoaggn - Aaggn] + C′
(5)
The Aoaggn, Aaggn, and Adi are absorbances of the aggregate added
initially, aggregate, and dimer in equilibrium solution, respec-
tively, and they can be obtained simply by spectral subtraction.
C and C′are constants.
For four SPLs which have relatively low Tc values and are
easily accessible to the above method, the plots according to
eq 5 give reasonable straight lines as shown in Figure 5. The
aggregate sizes, determined by the slopes (n) are 3, 4, 4, and 7
for bis-S4EPC, bis-S4EMPC, bis-4S6EPPC, and bis-S6EPC,
respectively, in terms of SPL molecules or 6, 4, 4, and 14,
respectively, in terms of numbers of TS per aggregate. For bis-
S4EPC and bis-S6EPC, aggregate size measurements were also
carried out at different temperatures: at 40 and 45 °C for bis-
S6EPC, and 22 and 30 °C for bis-S4EPC. Almost identical
results were obtained. Interestingly, the measured aggregate
sizes are small although a relatively long range of order is
expected in the SPL assemblies. Small aggregate sizes have
also been found in other chromophore systems for both
H-aggregates (squaraine29 and semicyanine30 ) and J-aggregates
(cyanine dyes31). There appears a clear correlation between the
extent of the blue-shifts in the absorption spectra of SPL
aggregates relative to monomer and the measured aggregate
sizes as suggested by the 33, 39, and 45 nm blue-shifts for the
tetramer of mono-SPLs, the hexamer of bis-S4EPC and the
14mer of bis-S6EPC, respectively. A relatively extended
aggregate is anticipated in the monolayers of SFAs and SPLs,
which give the largest blue-shifts (47 nm) in absorption. The
validity of the aggregate size estimation is strongly supported
by a linear relationship of ln(Adi3/Aagg) with [DMPC] according
to eq 4 with a slope of 2 for a series of equilibrium samples
with the same total concentration of bis-S4EPC vesicles and
different concentration of DMPC. The estimated equilibrium
constants for the equilibrium between the aggregates and dimers
for several SPLs together with the rate constants have been
reported previously.1 In the calculation, the specific partial
volume Vm of DMPC was assumed to be identical to that of
egg yolk phosphatidycholine vesicles (0.6677 L/mol).32 The
enthalpy and entropy for the deaggregation process over the
temperature range of 27-42 °C for bis-S4EPC were found to
be ca. 30 kcal/mol and ca. 90 cal/(kmol), respectively. The
data below 25 °C for the enthalpy and entropy measurements
show a large deviation from the linear plot due to an involve-
ment of the phase transition of DMPC vesicles. Clearly, the
aggregation process is enthalpically favored but entropically
disfavored. The interaction change involved in the aggregation
can be divided into two major components: the hydrophobic
interaction and the chromophore-chromophore interaction.
Since the “effective” fatty acid chain lengths of bis-S4EPC and
DMPC are similar, the hydrophobic interactions in pure bis-
S4EPC assemblies, DMPC vesicles, and mixed vesicles of bis-
Figure 5. Benesi-Hildebrand plots of ln Aagg as a function of ln(Ao
agg
- Aagg) for SPLs in DMPC vesicles.
S4EPC and DMPC are expected to show small differences and
thus contribute little to the overall change for the aggregation.
The primary difference between the aggregate and the dimer in
dilute DMPC vesicles is the chromophore-chromophore inter-
action. One mole of TS in the aggregate is enthalpically 5 kcal
more stable than in the dimer. This extra stability can be
attributed to the apolar association between the TS chro-
mophores.33,34 Similar results have also been obtained for
azobenzene,9 squaraine,29b and styrylthiophene23 derivatives. The
relatively high stability of these “unit aggregates” compared to
the dimers thus suggests that there are attractive noncovalent
interactions beyond those in the dimer.
The deaggregation processes for bis-S4EPC, bis-S4EMPC, and
bis-S6EPC assemblies and possibly other SPLs with DMPC or
DPPC vesicles at or above Tc values show first-order kinetics
for SPLs in each case. For the case of S6EPC and possibly
other SPLs, the deaggregation rates are independent of DMPC
concentration. This suggests that the aggregates dissociate
slowly (or are extruded into bulk solution) to form dimers or
monomers followed by rapid trapping by the matrix of DMPC
vesicles.
Induced Circular Dichroism (ICD) Spectra. No CD
signals are detected for SPL monomers in organic solvents such
as chloroform even though the SPLs are chiral. The reason is
apparent since the achiral TS chromophores are relatively far
away from the chiral center and the two fatty chains are flexible.
In contrast, fresh aqueous dispersions of SPLs prepared by
sonication exhibit strong biphasic ICD spectra (Figure 6) with
a positive Cotton effect in the long wavelength region and a
negative Cotton effect in the short wavelength region for all
SPLs (with the exception of S4EPC which, surprisingly, has a
reverse biphasic ICD spectrum). The two strong Cotton effects
are associated with two allowed optical transitions for the
aggregates. The energy difference between the two transitions
is too small to be resolved by absorption spectroscopy. The
occurrence of the zero-cross points of the ICD for all SPLs at
the λmax of the blue-shifted absorption spectra suggests that the
ICD are associated with the exciton band of the H-aggregates
(see the Supporting Information). These strongly enhanced ICD
are attibuted to the chiral structures of the H-aggregates.35
Although the blue-shifted absorption spectra of SPL assemblies
are independent of preparation techniques and conditions, the
(29) (a) Chen, H.; Law, K. Y.; Whitten, D. G. J. Am. Chem. Soc. 1995,
117, 7257. (b) Chen, H.; Farahat, M. S.; Law, K. Y.; Perlstein, J.; Whitten,
D. G. J. Am. Chem. Soc. 1996, 118, 2586.
(33) Smithrud, D. B.; Sanford, E. M.; Chao, I.; Ferguson, S. B.;
Carcanague, D. R.; Evanseck, J. D.; Houk, K. N.; Diederich, F. Pure Appl.
Chem. 1990, 12, 2227-2236.
(34) Dewey, T. G.; Wilson, P. S.; Turner, D. H. J. Am. Chem. Soc. 1978,
100, 4550.
(30) Bohn, P. W. Private communication.
(31) Samha, H.; Whitten, D. G. Unpublished results.
(32) Gennis, R. B. Biomembrane, Molecular Structure and Function;
Springer-Verlag: Berlin, 1989.

Aggregation of Stilbene-DeriVatized Fatty Acids
J. Am. Chem. Soc., Vol. 119, No. 51, 1997 12489
Figure 6. ICD (top) and absorption (bottom) spectra of bis-S4EPC
and bis-S6EPC assemblies in water.
Figure 7. (a) ICD spectra of bis-S4EPC in DMPC vesicles. All of the
samples contain identical concentration of DMPC and different
concentration of bis-S4EPC. (b) ICD intensity normalized by aggregate-
absorbance versus the dilution ratio of [DMPC]/[bis-S4EPC].
apparent molecular ellipticity for the ICD is subject to a change
under different preparation conditions, indicative of coexistence
of two interconvertible enantioners in the solution. It is
reasonable to suggest that both the chirality of the head group
and the alignment of stilbenes in the aggregates determine the
ICD spectra. The dimer for SPLs and monomer for SFAs or
mono-SPLs in an excess of chiral hosts of DPPC or DMPC
vesicles exhibit no detectible ICD signals. The mixtures of
aggregates and dimers (or monomers) in SPLs moderately
diluted with DMPC or DPPC give ICDs identical to those for
pure SPL aqueous dispersions. Figure 7 shows the ICD spectra
and the aggregate-absorbance-normalized ICD intensity as a
function of dilution ratio for S4EPC solubilized in DMPC
vesicles. The common zero-cross points for ICD spectra for
all of the samples with different dilution ratio and the indepen-
dence of the absorbance-corrected ellipticity on the dilution ratio
clearly indicate that the aggregates in DPPC or DMPC vesicles
are the same as those in pure SPL assemblies and only the
aggregates and dimers (or monomers) in the DPPC or DMPC
vesicles are involved in the solution. The temperature-induced
phase transition of SPL assemblies in water has little effect on
the ICD spectra except a decrease in intensity. This is consistent
with small aggregates of relatively high stability, which remain
unperturbed through the melting process of the fatty acid chains.
Steady State Photolysis. We have observed that the ag-
gregation in the monolayers2 and bilayer assemblies44 of TS
derivatives shuts down the trans-cis photoisomerization ob-
served for isolated TS in solution. Irradiation of pure aqueous
solutions of a number of trans-SPLs results in a slow, but
detectable photoreaction, which is monitored by UV-vis
absorption, fluorescence, and ICD spectra at regular intervals
during the photolysis. Similar results are obtained for 280 and
320 nm excitation wavelengths. Bleaching of the main ag-
gregate absorption band (at or near 270 nm) was observed in
all cases. However, the extent of the bleaching, as well as the
shape of the spectra after photolysis, was sample dependent.
ICD spectra showed only a decrease in intensity consistent with
disappearance of the optically active chromophore in the samples
as shown in Figure 8. The shape of the ICD spectra remained
constant throughout the experiments. Isosbestic points are
observed for both absorption and ICD spectra, which suggests
a clear conversion from the aggregates to photoproduct(s). The
length of time required to achieve certain extent of bleaching
for the aggregate absorption band at ca. 270 nm was used to
provide a qualitative measure of relative bleaching efficiencies
(180, 32, 20 and 4.5 min of irradiation are required for 25%
decrease in the peak absorbance for bis-S10EPC, bis-S4EPC, bis-
4S4EPC and bis-4S6EPC, respectively, keeping all experimental
conditions constant). True bleaching efficiencies could not be
obtained due to the appearance of a new light absorbing species
in the spectral region associated with the starting material. The
processes causing the observed spectral changes for SPLs is
far less efficient than the trans-cis photoisomerization process
for TS in fluid solutions. The extremely low bleaching
efficiency of bis-S10EPC assemblies is consistent with a stable
and uniform aggregate. Some evidence from both spectroscopic
and 1H NMR data of one major product resulting from
irradiation at 280 nm of SPL assemblies in water suggests
occurrence of a photodimerization.36 Such a photodimerization
of TS as a major process has also been observed for bilayer
asssemblies of other TS-derivatized amphiphiles in water, where
similar aggregates were formed.44 In another case, a crystalline
TS derivative has been found to display excimer emission and
undergoes photocyclization to generate a mirror symmetric
dimer.43 The unambiguous identification of the photoproduct-
(s) and the involved mechanism for the aggregates is a subject
of further investigation.
(35) (a) Cassim, J. Y. Biophys. J. 1992, 63, 1432-1442. (b) Person, R.
V.; Peterson, B. R.; Lighter, D. A. J. Am. Chem. Soc. 1994, 116, 42-59.
(c) Wu, S.; El-Sayed, M. A. Biophys. J. 1991, 60, 190-197. (d) Ebrey, T.
G.; Becker, B.; Mao, B.; Kilbride, P. J. Mol. Biol. 1977, 112, 377-397.
(36) Farahat, M. Unpublished results.

12490 J. Am. Chem. Soc., Vol. 119, No. 51, 1997
Song et al.
Figure 9. Reflectance spectra of S4A monolayer at the air-water
4
interface at difference surface pressures. Inset: Surface pressure-area
isotherm of S4A monolayer and hysteresis of S12A monolayer.
4
the molecular area is estimated to be ca. 22 Å2, consistent with
the reported data1 for the first compression of 4S4A. In contrast,
the π-area isotherm for the first compression is broad for S12A
monolayer, but sharp for the first decompression process. In
contrast with the monomeric absorption spectrum of TS in
solution, the reflectance spectrum of the 4S4A monolayer at the
water-air interface shows blue-shifts similar to the absorption
spectra of supported LB films of SFAs and bilayer assemblies
of SPLs. The reflectance spectra change little except in intensity
from very low pressures prior to compression to relatively high
pressure, indicating that the aggregates are preformed even in
the expanded liquid or gaseous state. Dilution of S4A mono-
4
layer with stearic acid (SA) up to SA/4S4A ) 4/1 shows little
effect on H-aggregate formation as indicated by the persistence
of similar H-aggregate reflectance spectra. Similar behavior is
observed for the 4S6A monolayer. The fact that the H-
aggregates are the only observed aggregation form for TS and
other diphenyl polyene derivatives in supported films and
monolayers at the air/water interface indicates the high stability
of H-aggregated forms. Related studies for some squaraine
derivatives37 showed similar results in that the H-aggregates of
squaraine also preform without compression at the water-air
interface.
Figure 8. Absorption (top) and ICD (bottom) spectra of bis-S4EPC
assemblies in water at different stages of irradiation at 280 nm.
The fluorescence spectra for the irradiated samples showed
more variations in shape and intensity between different samples
compared to the corresponding absorption spectra. The major
effect of photolysis was the decrease in intensity of the
fluorescing species. The extent of this reduction was highly
dependent on the structure of the SPLs used. Extended
photolysis of SPL assemblies in water leads to appearance of a
new structured fluorescence band near 350 nm that resembles
that of derivatized TS in organic solution (see the Supporting
Information). This fluorescence is attributed to unaggregated
TS chromophores that remain after a large population of their
neighboring moieties have undergone photodimerization and
cease to contribute to excitonic interactions. Due to the rigidity
of their environment, these isolated TS chromophores are not
expected to undergo trans-cis isomerization and to show a high
fluorescence efficiency.
Reflectance spectra of 4S4A and 4S6A monolayers consist of
two major peaks at 268 and 240 nm, which correspond to two
transition dipoles, one along the long axis of TS and the other
perpendicular to the long axis, repectively.37 Strikingly different
reflectance spectra were obtained for S12A and S10A monolayers,
which shows only one peak at 240 nm. The absence of the
peak at 268 nm for S12A and S10A suggests that the chro-
mophores are oriented perpendicular to the water surface, since
the incident light used for the reflectance measurement is normal
to the water surface. The presence of both absorption peaks
for S4A and S6A indicates that the stilbene chromophore is
4
4
Monolayers. We have examined an array of amphiphilic
TS fatty acid derivatives and trans,trans-1,4-diphenyl-1,3-
butadiene and trans,trans,trans-1,6-diphenyl-1,3,5-hexatriene
derivatives in supported assemblies2,8 where H-aggregation with
a characteristic blue-shifted absorption and a red-shifted fluo-
rescence is observed. In several cases, only H-aggregates can
be detected, even for LB films of mixtures of the aromatics
diluted with a large excess of fatty acid host. Reflectance
spectroscopic study of SFA monolayers at the water-air
interface provides some new insights into the H-aggregation
behaviors of these aromatics. As shown in Figure 9 (inset), a
tilted to some extent from the water surface normal.
Aggregate Structure. Monolayer and bilayer assemblies are
highly ordered and can be considered to be two-dimensional
lattices with some long-range order. Monte Carlo cooling
methods have previously been used to determine the apparent
global and nearby local minima structures for monolayers.38
Applying these procedures,38 we have found1b that the lowest
energy structures, which show closest agreement with the
(37) Chen, H.; Liang, K. L. Song, X.; Samha, H.; Law, K. Y.; Perlstein,
J.; Whitten, D. G. In Micelles, Microemulsions, and Monolayers. Science
and Technology; Shah, D., Ed.; Marcel Dekker, Inc. New York, 1995, in
press.
sharp π-area isotherm for S4A monolayer was obtained, and
4

Aggregation of Stilbene-DeriVatized Fatty Acids
J. Am. Chem. Soc., Vol. 119, No. 51, 1997 12491
limiting area/molecule and the predicted exciton splitting, are
the global minimum glide or herringbone arrangement for 4S6A
and the first local minimum for S4A. While we have not yet
been able to calculate structures (monolayer or bilayer) for the
phospholipids containing two SFA units, the similarity of spectra
and photophysics2 suggests similar structures for the aggregates
of both SFA's and phospholipids in the monolayer films and
bilayer assemblies. Thus, we have proposed that a glide layer
structure is the extended arrangement for the TS chromophores
in the aggregates of the SPLs in water. The strong biphasic
ICD for the aggregates of the SPLs indicates that the aggregate
structures are chiral with two different TS chromophores per
unit cell in the two-dimensional lattice of the glide layer. While
the infinite glide or herringbone lattice is not chiral, the smaller
units within the layer structure, most notably the “pinwheel”
tetramer or other cyclic structures, are chiral (see ref 1b). Taking
together the likelihood of a glide structure as the arrangement
of an extended aggregate and the measured aggregation number,
we propose a “pinwheel” structure as the most likely arrange-
ment for the smallest “unit” aggregates. These structures, with
two different stilbenes in each “unit” aggregate, are chiral and
account nicely for the observed biphasic ICD spectra associated
with the aggregates. These structures are also attractive in that
they maximize the number of “T” interactions (face-to-edge or
σ-π interaction) which have been demonstrated to stablize the
dimers of benzene39 and related aromatics.40 The favored
structure for the benzene dimer from both theory and experiment
is a “T” arrangement with stabilizing edge-face interaction.
Our studies for other aromatic systems including azobenzene,
styrylthiophene, and tolan as well as squaraine also suggest that
the “pinwheel” units may be a general supramolecular energy
minimum for a variety of aromatics and conjugated compounds
such as diphenylbutadienes and diphenylhexatrienes, since they
give very similar aggregation under similar conditions. An
X-ray diffraction study of several weakly amphiphilic TS
deirvatives provides support for the pinwheel or unit aggregate
as a key component in the formation of crystals or extended
aggregates.41
observed for SPL assemblies in water in contrast to normal
“monomeric” absorption and fluorescence spectra for SFAs and
SPLs in organic solvents. The blue-shifted spectra are attributed
to the formation of H-aggregates, in which the transition dipole
moments of the TS chromophores are aligned in a general head-
to-head arrangement. The complicated fluorescence of the SPLs
in water is believed to result from the heterogeneity of the
assemblies and possible energy transfer between aggregates and
excimers (or a geometrical change in excited state of the
aggregates which results in the formation of the excimer).
Evaluation of the equilibrium between the aggregates and dimers
(or monomers) of SPLs in the matrix of DPPC or DMPC
vesicles allowed determination of aggregate numbers and
thermodynamic parameters associated with the aggregation
process. The measured aggregate sizes show a good correlation
with the observed blue-shifts in absorption for several SPLs.
The strong biphasic ICD spectra associated with the aggregates
indicated the chiral nature of the aggregates in which there are
two different stilbenes in the unit aggregate. Monte Carlo
cooling simulation results1 indicate that a glide layer or
herringbone structure is the most probable structures for an
extended array of SFAs in a monolayers. Taking into account
the measured aggregate sizes, the chiral nature of the aggregates,
and the spectral similarity of SPL assemblies in water to SFA,
monolayers, as well as the simulated structures, small cyclic
chiral “pinwheel” structures are proposed as likely structures
of the H-aggregates. Extended aggregates can be fairly
described as a mosaic of these small aggregate units. The
structures are reasonable in view of the likely maximization of
weakly σ-π (face-edge) attractive forces and minimization
of π-π repulsions between the closely packed stilbene chro-
mophores.42
Acknowledgment. We are grateful to the National Science
Foundation (Grant CHE-9211586) for support of this research.
Supporting Information Available: Various spectra (6
pages). See any current masthead page for ordering and Internet
access instructions.
Summary
JA972292I
Blue-shifted absorption and red-shifted fluorescence spectra
similar to those in monolayers and LB films of SFAs were
(42) (a) Linse, P. J. Am. Chem. Soc. 1993, 115, 8793. (b) Hunter, C. A.;
Saunders, J. K. M. J. Am. Chem. Soc. 1990, 112, 5525-5534. (c) Jorgenson,
W. L.; Severance, D. L. J. Am. Chem. Soc. 1990, 112, 4768-4774. (d)
Shi, X.; Bartell, L. S. J. Phys. Chem. 1988, 92, 5667-5673. (e) Pawliszyn,
J.; Szczesniak, M. M.; Scheiner, S. J. Phys. Chem. 1984, 88, 1726-1730.
(43) Cohen, M. D.; Green, B. S.; Ludmer, Z.; Schmidt, G. M. Chem.
Phys. Lett. 1970, 7, 486.
(44) Shimomura, M.; Hashimoto, H.; Kunitake, T. Langmuir 1989, 5,
174.
(45) Agbaria, R. A.; Roberts, E.; Warner, I. M. J. Phys. Chem. 1995,
99, 10056.
(38) (a) Chen, H.; Law, K. Y.; Perlstein, J.; Whitten, D. G. J. Am. Chem.
Soc. 1995, 117, 7257. (b) Perlstein, J. J. Am. Chem. Soc. 1994, 116, 455.
(c) Perlstein, J. J. Am. Chem. Soc. 1994, 116, 11420.
(39) (a) Arunan, E.; Gutowsky, H. S. J. Chem. Phys. 1993, 98, 4294.
(b) Hall, D.; Williams, D. F. Acta Crystallogr. 1975, A31, 56.
(40) (a) Schweitzer, B. A.; Kool, E. T. J. Am. Chem. Soc. 1995, 117,
1863. (b) Winnik, F. M. Chem. ReV. 1993, 93, 587.
(41) Vaday, S.; Geiger, H. C.; Perlstein, J.; Whitten, D. G. J. Phys. Chem.
1997, 101, 321.