Unified crystallographic, Langmuir layer, and Langmuir-Blodgett film investigation of novel amphiphiles based on 4-nitroaniline

Article abstract of DOI:10.1021/jp0261105

Crystal structures of amphiphilic molecules should serve as a convenient basis to explore the molecular associations and intermolecular interactions in the corresponding Langmuir layers and Langmuir-Blodgett (LB) films. However, the difficulty of growing single crystals of such molecules often precludes such investigations. We present, in this paper, the crystal structure investigation and LB film study of new amphiphiles based on the 4-nitroaniline moiety of interest in quadratic nonlinear optical applications. A molecular-level explanation of the general instability of monolayers of these amphiphiles is proposed based on near neighbor interactions observed in the crystals. An approach to stabilize them through the formation of composite bilayers with the phospholipid molecule, distearoylphosphatidylcholine (DSPC), is developed. The complex pressure-area isotherms of the composites and their transformation through multiple isocycles are interpreted using the molecular structure and conformation from single crystal analysis. Electronic absorption spectroscopy of the solution, solid state, and LB films complemented by semiempirical quantum chemical computations allows us to model the intermolecular interactions in the LB film. This study demonstrates a novel approach to correlating the 3-dimensional and quasi-2-dimensional assembly of molecules in crystals and Langmuir/LB films.

Full text of DOI:10.1021/jp0261105

J. Phys. Chem. B 2003, 107, 147-156
147
Unified Crystallographic, Langmuir Layer, and Langmuir-Blodgett Film Investigation of
Novel Amphiphiles Based on 4-Nitroaniline
Sonika Sharma and T. P. Radhakrishnan*
School of Chemistry, UniVersity of Hyderabad, Hyderabad - 500 046, India
ReceiVed: May 14, 2002; In Final Form: October 4, 2002
Crystal structures of amphiphilic molecules should serve as a convenient basis to explore the molecular
associations and intermolecular interactions in the corresponding Langmuir layers and Langmuir-Blodgett
(LB) films. However, the difficulty of growing single crystals of such molecules often precludes such
investigations. We present, in this paper, the crystal structure investigation and LB film study of new
amphiphiles based on the 4-nitroaniline moiety of interest in quadratic nonlinear optical applications. A
molecular-level explanation of the general instability of monolayers of these amphiphiles is proposed based
on near neighbor interactions observed in the crystals. An approach to stabilize them through the formation
of composite bilayers with the phospholipid molecule, distearoylphosphatidylcholine (DSPC), is developed.
The complex pressure-area isotherms of the composites and their transformation through multiple isocycles
are interpreted using the molecular structure and conformation from single crystal analysis. Electronic absorption
spectroscopy of the solution, solid state, and LB films complemented by semiempirical quantum chemical
computations allows us to model the intermolecular interactions in the LB film. This study demonstrates a
novel approach to correlating the 3-dimensional and quasi-2-dimensional assembly of molecules in crystals
and Langmuir/LB films.
Introduction
tion of single crystal and LB film characteristics was presented.
This exercise demonstrates the paucity of systems of interest
One of the versatile aspects of molecular materials is the
variety of forms in which they can be assembled, ranging from
single crystals to amorphous powders, Langmuir-Blodgett (LB)
films to thermally evaporated films, and host-guest complexes
to embeddings in polymer films. Understanding the nature of
intermolecular interactions is critical for the effective design
and utilization of molecular materials. Among the different bulk
structures with varied application potential, a detailed atomic/
molecular level description of the interactions between the
building blocks is most conveniently obtained in single crystals
in which atom positions can be determined unambiguously
through X-ray diffraction analysis. However, among the different
approaches to the assembly of molecular materials, the LB
technique stands out as a simple, elegant, and efficient organiza-
tion tool.1 Molecular assembly at the air-water interface and
LB film deposition can be tailored by careful control of a variety
of parameters including subphase composition and conditions,
substrate preparation and dipping protocols. However in the
majority of cases of interest in materials applications, the very
nature of the molecules which form LB films, i.e., their
amphiphilic structure, hampers the formation of good quality
single crystals precluding a structure determination. We have
carried out a crystallographic database2 survey of amphiphilic
molecules having an n-tridecyl or longer n-alkyl chain and
bearing any kind of π-conjugated chromophore ring unit; the
cutoff of the tridecyl unit was chosen because 12-carbon and
shorter chain systems usually do not form stable monolayers.
Crystal structures have been reported for only 31 cases, and a
search of the literature indicated that of these only one system
was characterized in LB studies;3 however, no detailed correla-
in optical materials applications on which detailed crystal
structure characterization and LB studies have been concurrently
carried out. In this paper, we present a combined crystallographic
and Langmuir/LB film investigation of two amphiphilic mol-
ecules of potential interest in nonlinear optical (NLO) applica-
tions.
The prototypical “push-pull” system for molecule-based
quadratic NLO materials is the 4-nitroaniline moiety; some of
the most successful materials such as N-(4-nitrophenyl)-S-
prolinol (NPP)4 and methyl-(2,4-dinitrophenyl)-aminopropanoate
(MAP)5 are based on this framework. Despite several other
families of NLO chromophores developed over the years, this
basic structural unit continues to be the focus of fundamental
studies.6 Unfortunately, successive attempts at LB film fabrica-
tion of several amphiphiles based on the 4-nitroaniline deriva-
tives have met with limited success.7 Although this combination
of NLO chromophore and molecular assembly technique is a
prominent and unique one, little is known about the basic cause
of its failure. A molecular level insight into this problem can
be gained through a combination of structural investigation of
crystals of the amphiphile and monolayer studies together with
the examination of some physical characteristics common to
both states of assembly.
The present study addresses this question and explores
approaches to stabilize monolayers of amphiphiles based on the
4-nitroaniline moiety and to deposit their LB film. Because many
of the successful crystalline NLO molecular materials are
based on chiral molecules, we have chosen to study three
new molecules: N-(4-nitrophenyl)-3(R)-octadecyloxypyrrolidine
(NP3OP), N-(4-nitrophenyl)-O-hexadecyl-S-prolinol (NPOHP),
and N-(4-nitrophenyl)-O-octadecyl-S-prolinol (NPOOP); the
latter two are NPP derivatives. Amphiphiles such as these rarely
* To whom correspondence should be addressed. E-mail: tprsc@
uohyd.ernet.in. Fax: 91-40-3012460.
10.1021/jp0261105 CCC: $25.00 © 2003 American Chemical Society
Published on Web 12/10/2002

148 J. Phys. Chem. B, Vol. 107, No. 1, 2003
Sharma and Radhakrishnan
1.5-1.6 (2H, m), 1.2 (30H, br s), 0.9 (3H, t). 13C- NMR
(CDCl3): δ/ppm ) 151.8, 137.2, 126.1, 111.0, 71.8, 70.7, 58.7,
48.6, 31.9, 29.6, 29.4, 28.9, 26.1, 23.2, 22.6, 14.0. Elemental
analysis (calculated for C29H50N2O3): %C ) 73.65 (73.42), %H
) 10.62 (10.55), %N ) 5.98 (5.91).
Samples for the LB experiments were purified by multiple
recrystallizations. L-R-distearoylphosphatidylcholine (DSPC)
with 99+% purity from Sigma Chemicals was used directly.
Crystals of NP3OP and NPOHP were grown from dilute
solutions of hexane-dichloromethane and chloroform respec-
tively over a period of 2 days at ∼ 5 °C.
Crystal Structure Determination. X-ray diffraction data
were collected on an Enraf-Nonius MACH3 diffractometer. Mo
KR radiation with a graphite crystal monochromator in the
incident beam was used. Data were reduced using Xtal3.4;8
Lorentz and polarization corrections were included. All non-
hydrogen atoms were found using the direct method analysis
in SHELX-97,9 and after several cycles of refinement, the
positions of the hydrogen atoms were calculated and added to
the refinement process. Details of data collection, solution and
refinement, fractional coordinates with anisotropic thermal
parameters, and full lists of bond lengths and angles are
submitted as Supplementary Information.
Figure 1. Scheme of synthesis of NP3OP, NPOHP and NPOOP.
form single crystals suitable for X-ray analysis. We have
succeeded in growing crystals of NP3OP and NPOHP and have
investigated their molecular structure and crystal packing.
Similar to earlier observations on related systems,7 these
amphiphiles show very poor spreading behavior on water and
do not form stable monolayers. We have stabilized them on
the water surface by mixing with the phospholipid, dis-
tearoylphosphatidylcholine (DSPC). Characteristic features of
the pressure-area isotherms of the composites are investigated
in detail. Although pressure-area isotherms by themselves
provide little structural information, together with the input from
the crystal structures, we draw useful insight into the likely
structure of the Langmuir layers. Finally, we have achieved
quantitative transfer of the composite bilayers of NP3OP-DSPC
and NPOOP-DSPC onto suitable substrates, and the deposited
LB films are characterized using electronic absorption spec-
troscopy supported by semiempirical quantum chemical model-
ing. Correlation of the electronic absorption spectra of the
crystalline material and the LB film allows us to extend the
molecular level picture in the crystal to the LB film. The present
study sheds light on intermolecular interactions that possibly
impede stable monolayer formation of amphiphiles based on
4-nitroaniline and suggests routes for the fabrication of their
LB films.
LB Film Studies. Pressure-area (π-A) isotherms were
investigated on a Nima model 611M LB trough (area ) 30 ×
10 cm2) using a Wilhelmy plate for sensing surface pressure.
High purity water (Millipore MilliQ) was used as the subphase.
Chloroform (Uvasol grade, EMerck) solutions (∼1 mM) were
spread on the subphase. A 15 min wait period was allowed for
solvent evaporation. All experiments were carried out at 25 °C.
The π-A isotherms were recorded using a barrier speed of 30
cm2/min. Experiments were repeated on fresh subphases 2-3
times to confirm reproducibility; the precision of molecular areas
extracted is ∼1 Å2. Area per molecule A1 is obtained by
extrapolation to π ) 0 mN/m, of the straight regions of the
isotherms just below the plateau representing the flip-over
transition (observed above 13 mN/m in all the systems) in the
first compression. A2 is obtained similarly, from the final
collapse of the bilayer in the reproducible compressions. The
presence of sufficiently long straight regions in the isotherms
make this unambiguous in the case of NP3OP and NPOOP; in
some of the isotherms of NPOHP, the extrapolated areas would
be less accurate because of the limited straight region in the
isotherms.
Experimental and Computational Section
Synthesis and Characterization. NP3OP, NPOHP, and
NPOOP were synthesized following the reaction schemes in
Figure 1. The products were characterized as follows:
NP3OP. mp ) 82-84 °C. FT-IR (KBr): ν/cm-1 ) 1520,
Glass and quartz microscope slides with a hydrophilic surface
for the LB film deposition were prepared as follows; high purity
water was used in all operations. The substrate was cleaned with
detergent followed by several rinsings in water; it was soniccated
in fresh lots of water three times for 30 min each. The surface
was made hydrophilic by immersing in aqueous sodium
hydroxide for 12 h followed by soniccation and rinsing in water.
The LB film was coated on both sides of the substrate by the
vertical dipping procedure employing a dipping speed of 3 mm/
min with the monolayer held at a pressure of 30 mN/m. The
specular reflectance (8° incidence) spectrum of the LB film on
quartz was recorded on a Shimadzu model UV-3101 UV-
visible spectrometer using the integrating sphere (ISR 3100)
assembly. The sample as well as the reference uncoated plate
were placed with a Teflon sheet at the back. We have chosen
to use reflectance over absorption spectra for the LB films
because of the improved signal-to-noise ratio obtained. The
reflectance spectrum of the solid material was recorded using
pellets made from a mixture of the compound and KBr.
Absorption spectra of the solutions in hexane were also recorded.
1
1483, 1331, 1105. H NMR (CDCl3) : δ/ppm ) 8.1 (2H, d),
6.5 (2H, d), 4.2 (1H, broad), 3.2-3.6 (6H, m), 2.1 (2H, m),
1.5-1.6 (2H, m), 1.3 (30H, br s), 0.9 (3H, t). 13C NMR
(CDCl3): δ/ppm ) 151.9, 136.9, 126.3, 110.5, 77.8, 69.4, 53.6,
46.1, 31.9, 31.0, 29.7, 29.4, 26.2, 22.7, 14.1. Elemental analysis
(calculated for C28H48N2O3): %C ) 73.25 (73.04), %H ) 10.48
(10.43), %N ) 6.18 (6.09).
NPOHP. mp ) 73-76 °C. FT-IR (KBr): ν/cm-1 ) 1514,
1479, 1336, 1118. 1H NMR (CDCl3): δ/ppm ) 8.1 (2H, d), 6.6
(2H, d), 4.0 (1H, broad), 3.2-3.6 (6H, m), 2.1 (4H, m), 1.5-
1.6 (2H, m), 1.2 (26H, br s), 0.9 (3H, t). 13C NMR (CDCl3):
δ/ppm ) 151.8, 137.0, 126.2, 110.7, 71.7, 70.5, 58.7, 48.6, 31.9,
29.7, 29.4, 28.9, 26.1, 23.2, 22.7, 14.1. Elemental analysis
(calculated for C27H46N2O3): %C ) 72.76 (72.65), %H ) 10.38
(10.31), %N ) 6.32 (6.28).
NPOOP. mp ) 78-80 °C. FT-IR (KBr): ν/cm-1 ) 1514,
1
1483, 1336, 1118. H NMR (CDCl3) : δ/ppm ) 8.1 (2H, d),
6.6 (2H, d), 4.0 (1H, broad), 3.2-3.6 (6H, m), 2.1 (4H, m),

Investigation of Amphiphiles
J. Phys. Chem. B, Vol. 107, No. 1, 2003 149
Computational Studies. Semiempirical quantum chemical
studies were carried out using the AM110 method in the
MOPAC93 program package.11 The microenvironment of the
molecule in the condensed phases and the solution phase was
simulated using the solvation model COSMO;12 the utility of
this approach has been demonstrated in our earlier studies.13
The absorption energies and the corresponding transition dipoles
were computed from a configuration interaction (CI) calculation
on geometries of the molecule and supramolecular assemblies
extracted from the crystal structure; the hydrocarbon chains were
replaced by methyl groups, and all H atom positions were
optimized. The CI calculations invoked the full set of singlet
configurations (400) involving 6 molecular orbitals bracketing
the HOMO and the LUMO.
Results and Discussion
As noted above regarding several 4-nitroaniline based am-
phiphiles,7 NP3OP, NPOHP, and NPOOP do not spread as
stable monolayers on the water surface but form visible
aggregates. We have therefore explored equimolar mixtures of
these systems with cadmium arachidate. In the case of NP3OP,
a large hysteresis is observed in the compression-expansion
isotherm (isocycle) indicating strong collapse. NPOHP and
NPOOP mixtures showed isotherms with a plateau similar to
the one observed by Munn and Szczur7f for another NPP-based
amphiphile mixed with stearic acid. In all of the three cases,
the second isocycle was shifted to lower areas. The molecular
areas could not be satisfactorily explained on the basis of
formation of composite monolayers. Because no consistent
model for the mixed monolayers or bilayers could be developed
from these observations, we have explored mixtures of these
novel amphiphiles with DSPC. Phospholipids form stable
monolayers on water and are known to discourage aggregation
of guest molecules.14 Our earlier study of fatty acid-DSPC
composites has also suggested its utility as a host matrix.15
Equimolar Compositions of NP3OP, NPOHP, and NPOOP
with DSPC. We first examined equimolar mixtures of NP3OP,
NPOHP, and NPOOP with DSPC on the water surface. The
isotherms are presented in Figure 2 parts a-c respectively for
NP3OP-DSPC, NPOHP-DSPC, and NPOOP-DSPC; the
values plotted on the x axis are the mean molecular areas based
on the total number of molecules spread on the water surface.
The monolayers of the composite systems display complex
behavior. In each case, the first isocycle shows plateau formation
and hysteresis; subsequent compressions and expansions lead
to isocycles which are highly reproducible over extended periods
of time but shifted with respect to the first isocycle. Alhough
the general patterns are similar, there are significant differences
in the details between the three systems. For NP3OP, the second
and subsequent isocycles are close to the expansion curve of
the first one. In NPOHP-DSPC and NPOOP-DSPC, these are
shifted considerably from the first isocycle; reproducibility is
achieved from the third and second isocycles, respectively, in
the two systems. The shape change between the first and final
isocycle is moderate in the NPOHP system but quite dramatic
in the case of NP3OP and NPOOP. The values of area/molecule,
A1, at the onset of the plateau (see the Experimental Section
for details) are similar in the case of NP3OP-DSPC and
NPOOP-DSPC but larger in NPOHP-DSPC. The latter
suggests a different orientation of the amphiphile headgroups
on the water surface compared to the first two. The extrapolated
areas, A2, from the final collapse of the reproducible compression
isotherms in all of the three cases correspond closely to that of
a DSPC monolayer as described later. Plateaus in pressure-
Figure 2. π-A isocycles of 1:1 mixtures of (a) NP3OP-DSPC, (b)
NPOHP-DSPC, and (c) NPOOP-DSPC; values on x axis are the mean
molecular areas based on the total number of molecules spread.
area isotherms arise as a result of molecular reorientation or
chain alignment associated with phase transitions16 or multilayer
formation.17 The extrapolated areas and the evolution of the
isotherm shapes in Figure 2 strongly suggest a “flipping over”
of the amphiphiles onto the DSPC monolayer; this would be a
natural consequence of the superior stability of DSPC films
compared to that of the other amphiphiles. The structure that
results is best described as a bilayer as shown by several

150 J. Phys. Chem. B, Vol. 107, No. 1, 2003
Sharma and Radhakrishnan
Figure 3. Molecular structure of (a) NP3OP and (b) NPOHP from single-crystal X-ray analysis showing 10% probability thermal ellipsoids.
TABLE 1: Crystallographic Data for NP3OP and NPOHP
observations described below. We develop this model and probe
the nature of intermolecular interactions in the monolayer and
composite bilayer through a unique combination of crystal
structure investigations, detailed pressure-area isotherm studies
of the mixtures, and electronic absorption spectroscopy of the
LB film.
Crystal Structure of NP3OP and NPOHP. We have succeeded
in growing single crystals of NP3OP and NPOHP; crystallization
of NPOOP was not successful, but the salient features of
NPOHP are expected to carry over to NPOOP. The crystal-
lographic data are summarized in Table 1. NP3OP crystallizes
in the space group P21 with one molecule in the asymmetric
unit (Figure 3a); the alkyl chain has all-trans conformation. The
dihedral angle, τ (C14-C13-O15-C16), that determines the
orientation of the alkyl chain with respect to the headgroup is
75.3°. The chain is oriented approximately perpendicular to the
dipole axis of the headgroup; the angle between the vector N2-
N1 and the vector O15-C33 is 78.6°. Based on the unit cell
parameters, the molecular area ({ab sin γ}/2) of NP3OP is
NP3OP
NPOHP
molecular formula
formula weight
crystal system
space group
a, Å
b, Å
c, Å
C28H48N2O3
460.68
monoclinic
P21
8.2829(10)
5.306(3)
31.269(5)
91.897(11)
2
1.114
0.71
4570
1874
C27H46N2O3
446.66
orthorhombic
P212121
5.1217(19)
8.7292(17)
59.463(9)
90
â, deg.
Z
4
F
calc., g cm-3
1.116
0.72
4515
4154
293
0.964
µ, cm-1
number of unique reflections
number of reflections with I g σI
number of parameters
GOF
299
0.983
R (for I g 2σI)
0.0463
0.1306
0.0598
0.2336
wR2
calculated to be 22.0 Å2; the factor of 2 arises as a result of
two molecules sharing the area of the ab face (see packing

Investigation of Amphiphiles
J. Phys. Chem. B, Vol. 107, No. 1, 2003 151
a critical influence on their crystal packing (Figure 5); this would
also have strong implications for their behavior at the air-water
interface18 as discussed later. The chains are interdigitated in
NPOHP. The molecular packing in both crystals takes advantage
of the dispersion interaction between the alkyl chains as well
as dipole-dipole interactions between the 4-nitroaniline head-
groups. The dipole vectors (N2-N1) of neighboring molecules
are oriented nearly antiparallel, the angle being 178.5° and
170.7° in NP3OP and NPOHP, respectively. Interestingly, the
molecular planes of the headgroups of the near neighbors in
both the crystals adopt a nearly orthogonal orientation; the angle
between the mean benzene planes of adjacent molecules in
NP3OP and NPOHP are respectively 79.6° and 83.1°. Such near
neighbor orientations have been observed in crystals of alkyl
substituted dipolar chromophores.19 In the 4-nitroaniline based
amphiphiles, this orthogonal orientation appears to be facilitated
by a multitude of weak interactions20 between the nitro group
atoms and several atoms on the neighboring molecules. The
shortest of the intermolecular contacts observed in NP3OP are
O4...C16′ (3.348 Å), O3...C12′′ (3.537 Å), N2...C11′′′ (3.566
Å), and O3...C11′′′ (3.584 Å) and in NPOHP are O3...C11′
(3.246 Å), O3 ...C11′′ (3.360 Å), O4...C15′′′ (3.394 Å),
N2...C11′′ (3.402 Å), and N1...O3′′′′(3.481 Å) (the primes
denote different neighbors of the unprimed molecule). The O....C
and N...C represent weak H-bond interactions and the N...O
represents the electrostatic interaction. We believe that the
push-pull nature of the headgroup enhances the electron density
on the nitro group oxygen atoms, improving the strength of these
interactions. Such an orthogonal orientation of the push-pull
chromophore headgroups which could arise even when the
molecular dipoles are oriented parallel (as when spread on a
water surface and compressed) would hinder efficient headgroup
packing and thwart the formation of stable monolayers of
4-nitroaniline based amphiphiles.
Variable Compositions of NP3OP, NPOHP, and NPOOP
with DSPC. In Figure 2a, it is seen that the area, A1, of the
equimolar NP3OP-DSPC system is 38.6 Å2/molecule and the
area A2 is 28.0 Å2/molecule. The molecular area of DSPC is
∼54 Å2.15,21 Using the molecular area of NP3OP determined
from the crystal structure analysis, the mean molecular area for
the equimolar composite works out to be 38 Å2, in good accord
with the observed value of A1. The value of A2 corresponds
closely to the area/molecule of a monolayer of pure DSPC; it
should be noted that, because the number of molecules of DSPC
is half of the total, the value on the x axis should be doubled to
calculate the area/molecule with respect to DSPC. The values
of A1 and A2 are consistent with the proposed model of formation
of a composite monolayer of NP3OP and DSPC from which
the NP3OP molecules flip over onto the DSPC monolayer at
the plateau transition. Such bilayer formations have been
reported in earlier studies of composite systems at the air-water
interface.22 The novel and interesting feature in the present case
is that on expansion and further compression the isotherm shape
changes completely and appears like that of a pure DSPC
monolayer exhibiting long-term stability and reproducibility.
This indicates that the NP3OP molecules which have flipped
over onto the DSPC molecules are retained in some complexed
form during expansion and further isocycles. To examine further
this model of bilayer formation, we have recorded the π-A
isotherms of different compositions of NP3OP-DSPC (Figure
6a). The composition of 90% and 10% DSPC are extreme cases,
the former behaving nearly like pure DSPC and the latter
showing no stable monolayer formation. The extrapolated areas,
A1 and A2, for the different compositions are presented in Table
Figure 4. AM1/COSMO calculated enthalpy of formation versus
dihedral angle, (a) τ (C14-C13-O15-C16) in NP3MP and (b) τ (N1-
C14-C15-O16) in NPOMP for different values of the dielectric
constant, ꢀ.
diagram later). NPOHP crystallizes in the space group P212121
with one molecule in the asymmetric unit (Figure 3b); the alkyl
chain has an all-trans conformation. The dihedral angle, τ (N1-
C14-C15-O16), is 179.7°. This leads to an improved align-
ment of the alkyl chain with respect to the dipole axis compared
to that for NP3OP; the angle between the vector N2-N1 and
the vector C15-C32 is 45.0°. The molecular area is estimated
to be 22.4 Å2.
To rationalize the observed τ angles in NP3OP and NPOHP
and to estimate the energetics associated with chain reorienta-
tions, we have carried out semiempirical AM1 calculations on
the model headgroups, NP3MP and NPOMP, in which the alkyl
chains of the original molecules are replaced by methyl groups.
The crystalline environment was simulated using the solvation
model, COSMO. The variation of the enthalpy of formation
with τ is shown in Figure 4. In NP3MP, minima are observed
near τ ) 60° and 180°. Inclusion of the solvation model with
increasing dielectric constants leads to a consistent global
minimum at 70°; this is close to the angle observed in the crystal.
It is also seen that the barrier to rotation along the τ coordinate
is ∼3.5 kcal/mol. Similar computations on NPOMP gives the
global minimum at 160°, again in good agreement with the
experimental value. The rotational barrier is <1 kcal/mol.
The different orientations of the alkyl chains with respect to
the dipole axis in the two molecules (Figure 3) appear to exert

152 J. Phys. Chem. B, Vol. 107, No. 1, 2003
Sharma and Radhakrishnan
Figure 5. Molecular packing in the crystals of (a) NP3OP and (b) NPOHP.
TABLE 2: Estimateda and Observed (Figure 6) Area/
Molecule Corresponding to the “Flip-Over” Transition (A1)
and Final Collapse (A2) for Different Compositions of
NP3OP-DSPC, NPOHP-DSPC, and NPOOP-DSPC
compositions (Table 2) lends strong support to the “flip-over”
model and bilayer formation.
The equimolar NPOHP-DSPC mixture shows an elongated
plateau during the first compression (Figure 2b); the value of
A1 is 47.2 Å2. Because the molecular area of NPOHP determined
from the crystal structure is similar to that of NP3OP, the large
A1 suggests that the NPOHP headgroups are tilted toward a
horizontal orientation at the air-water interface. This results
from a more pronounced hydrophilic headgroup-water interac-
tion because of the shorter length of the alkyl chain of NPOHP
and the low energy barrier (Figure 4b) facilitating chain
reorientation. The area estimated for the equimolar mixture
assuming a horizontal face-on orientation of the headgroup of
NPOHP is given in Table 2. Because the observed value is
smaller, a tilted orientation of the headgroup is likely. The value
of A2 from the final reproducible isotherms is 28.5 Å2, closely
agreeing with the value for a DSPC monolayer as in the case
of NP3OP-DSPC. The plateau of the first compression
represents a headgroup reorientation toward the vertical, together
with part of the NPOHP molecules flipping over, and the second
and third compressions complete the flipping process. The
flipping over of the NPOHP molecules appears to be more
difficult than that of NP3OP. This is borne out also by the
similarity of the shapes between the first and subsequent
isocycles of NPOHP-DSPC which suggests partial reversal of
the flip-over. A physical interpretation of this behavior is that
NPOHP shows a relatively higher affinity to remain on the water
surface than NP3OP. The mechanism of chain reorientation on
the water surface and relaxation back to the tilted orientation
when flipped over onto DSPC promotes the bilayer formation
in NP3OP but is altogether absent in the case of NPOHP. The
π-A curves for the different compositions of NPOHP-DSPC
are presented in Figure 6b, and the A1 and A2 values are collected
in Table 2. In the case of mixtures with less than 50% DSPC,
the A1 values suggest a more horizontal orientation of the
headgroups of NPOHP. The A2 values agree with the estimates
assuming two molecules of NPOHP flipping over onto each
area/molecule (Å2)
NP3OP-DSPC NPOHP-DSPC NPOOP-DSPC
mol %
A1
of DSPC or A2 estimate obsd estimate obsd estimate obsd
30
40
50
60
70
90
A1
A2
A1
A2
A1
A2
A1
A2
A1
A2
A1
A2
31.6
25.0
34.8
26.0
38.0
27.0
41.2
32.4
44.4
37.8
50.8
48.6
28.8
22.6
33.1
25.0
38.6
28.0
42.1
33.1
44.1
37.4
b
54.8
18.4
54.7
21.6
54.6
27.0
54.5
32.4
54.4
37.8
54.1
48.6
56.2
19.5
59.2
21.9
47.2
28.5
45.0
36.3
b
31.9
18.4
35.0
21.6
38.2
27.0
41.4
32.4
44.5
37.8
50.8
48.6
34.3
19.1
37.3
22.4
40.0
26.6
40.8
31.7
b
38.4
b
47.8
40.3
b
47.5
48.3
a A1 ) nDSPCADSPC + nNPANP. A2 ) nDSPCADSPC + (nNP - xnDSPC)ANP
{for nNP > xnDSPC} ) nDSPCADSPC {for nNP < xnDSPC}. ADSPC ) 54 Å2;
A
NP, the molecular area for NP3OP, NPOHP, or NPOOP are estimated
from crystal structures: NP3OP (22 Å2), NPOHP (55.2 Å2 for horizontal
‘face-on’ orientation and 22.4 Å2 for vertical orientation, for A1 and
A2 calculation respectively), NPOOP (22.4 Å2 from NPOHP); n ) mol
fraction; x ) 1 (NP3OP), 2 (NPOHP, NPOOP) (Figure 7). b No clear
“flip-over” transition observed.
2. The values estimated using the model of NP3OP flipping
over (Figure 7a) are also presented in the table. We assume
that, under compression conditions, the alkyl chains of NP3OP
are reoriented to align with the headgroup axis and that, on
flipping over, these molecules relax their chains back to the
tilted orientation found in their crystals. This makes them
compatible in size with the DSPC molecules, supporting a 1:1
complexation in the bilayer. The A2 values of the 30% and 40%
DSPC mixtures provide critical evidence for this stoichiometry
in the bilayer by accounting for the area due to the DSPC layer
as well as the unflipped NP3OP. The good agreement between
the experimental values and the estimates across the different

Investigation of Amphiphiles
J. Phys. Chem. B, Vol. 107, No. 1, 2003 153
The equimolar NPOOP-DSPC mixture shows a shorter
plateau than the one in the case of NPOHP-DSPC during the
first compression (Figure 2c); a smooth reorientation of the
headgroup is suggested by a minor inflection in the isotherm
near 10 mN/m. The value of A1 is 40.0 Å2, in good agreement
with the value estimated for a monolayer of NPOOP and DSPC.
The area of NPOOP is assumed to be 22.4 Å2, the same as that
of NPOHP in the vertical orientation. Unlike NPOHP, the
headgroup of NPOOP is oriented vertically at the point of
flipping over; this is compatible with the longer alkyl chain of
NPOOP. From the second compression onward, the shifted
isotherms resemble that of DSPC and are highly reproducible.
The value of A2 is 26.6 Å2, agreeing very well with the value
for a DSPC monolayer. The behavior of NPOOP-DSPC
mixtures shows that the flip-over of NPOOP molecules is easier
than that of NPOHP but less facile than that of NP3OP. The
reasoning for the latter would follow the same logic as in the
case of NPOHP, and the former can be attributed to the fact
that the longer alkyl chain of NPOOP makes the affinity to water
less pronounced than in the case of NPOHP. The π-A curves
for the different compositions of NPOOP-DSPC are presented
in Figure 6c, and the A1 and A2 in Table 2 show good agreement
with the estimated values; the estimated values of A2 assume a
2:1 complexation in the bilayer. The behavior largely follows
that of NPOHP, but the variations with composition are more
systematic because the headgroup reorientation is smooth. At
all compositions below 70% DSPC, the final isotherms are very
much DSPC-like, indicating irreversible complexation of NPOOP
with DSPC. The 70% and 90% DSPC mixtures do not show a
clear flip-over transition, but only the inflection due to head-
group reorientation. The flip-over process and bilayer formation
required to explain the plateau and area values are schematically
represented in Figure 7c.
We have shown that several features of NP3OP, NPOHP,
and NPOOP and their composites with DSPC on the water
surface can be explained using inputs from the crystal structure
of NP3OP and NPOHP. The molecular areas of the composites
extracted from the π-A isotherms show good correlation with
simple additive estimates based on individual molecular areas.
Although this could be interpreted as indicative of an immiscible
system,23 the varying plateau pressures across the different
compositions (with a clear minimum/maximum in the NP3OP
and NPOOP systems, but less systematic in the NPOHP) rule
out such a possibility and signify interacting components in the
composite monolayer. Further, the collapse pressures are found
to vary across the compositions (again more systematic in
NP3OP and NPOOP) indicating intermolecular interaction
between the nitroaniline amphiphile and DSPC in the bilayer.
The flip-over transition demonstrates the dominance of the
intermolecular interactions in the bilayer over that in the
monolayer. The less well-behaved features of NPOHP compared
to NP3OP and NPOOP in all of the cases may be attributed to
its shorter chain length. However, its crystal structure provides
useful insights into the behavior of NPOOP and explains the
critical differences in the behavior of NP3OP with respect to
NPOHP and NPOOP in terms of the different chain orientations
observed in the crystals. We show further that the molecular
level interactions in the LB films of the composite bilayers can
be visualized using parallel spectroscopic investigations of the
LB films and the crystalline phase.
Figure 6. π-A isotherms of different compositions of (a) NP3OP-
DSPC, (b) NPOHP-DSPC, and (c) NPOOP-DSPC; values on x axis
are the mean molecular areas based on the total number of molecules
spread.
DSPC to give a 2:1 complex (see Figure 7b). This picture is
realistic in view of the better alignment of the chain with the
headgroup of NPOHP observed in its crystals. Once again the
values of A2 of the 30% and 40% DSPC systems provide crucial
evidence for the stoichiometry. The 30% DSPC system shows
a strong change of shape after the first isocycle; the reproduc-
ibility observed from the third isocycle onward and its shape
suggests that the NPOHP molecules are irreversibly complexed
with DSPC and smoothly form the bilayer on compression.
LB Films. The composite bilayers of equimolar NP3OP-
DSPC and NPOOP-DSPC formed after the first isocycle
showed good stability. LB films containing up to three layers
could be deposited on hydrophilic glass/quartz surfaces. The

154 J. Phys. Chem. B, Vol. 107, No. 1, 2003
Sharma and Radhakrishnan
Figure 7. Schematic model of the “flip-over” process and bilayer formation in (a) NP3OP-DSPC, (b) NPOHP-DSPC, and (c) NPOOP-DSPC.
Note the horizontal orientation of the headgroup in the case of NPOHP.
transfer ratios were in the range 0.8-1.0. This is noteworthy in
view of the fact that phospholipids are generally not amenable
to multilayer LB film deposition.24 The contact angles of water
drops on the LB films of the composites were in the vicinity of
70°; this contrasts with the 90° angle observed on monolayer
LB films of DSPC. The decreased hydrophobicity of the LB
films of NP3OP-DSPC and NPOOP-DSPC is compatible with
the picture of bilayer formation presented above (Figure 7). The
coated films were characterized using UV-visible reflectance
spectra. The spectrum of the NP3OP-DSPC LB films is shown
in Figure 8 along with the reflection spectrum of solid NP3OP
diluted with KBr and the absorption spectrum of the hexane
solution of NP3OP. Because very similar spectra are observed
in the case of NPOOP as well, we present a discussion of only
the NP3OP system. The additional peaks seen in the LB film
and the solid-state spectra compared to those of the solution
Figure 8. Reflection spectra of LB films of 1:1 NP3OP-DSPC on
can be attributed to the electronic excitations of supramolecular
assemblies. The relation between the solid-state spectrum and
the LB film spectrum supports our earlier interpretation of the
monolayer properties using crystal structure information. It also
suggests that molecular level interactions in the crystalline
material and the LB films are similar. However, some conspicu-
ous differences can be noticed in the broader peaks and
additional shoulders in the solid-state spectrum compared to the
LB film spectrum. We have carried out semiempirical computa-
tions to probe these aspects further.
quartz and the solid pellet of NP3OP (with KBr) and the absorption
spectrum of NP3OP in hexane solution.
solvent effect was modeled using the COSMO routine with the
dielectric constant value of 2, similar to that of hexane. The
strongest peaks computed at 339.2 and 209.7 nm (Table 3) when
scaled up by 20 nm match very well with the prominent solution
spectral peaks at 360.5 and 226.5 nm. We have computed also
the excitation energies and transition dipoles of the monomer
and several dimeric, trimeric, tetrameric, pentameric, and
heptameric supramolecular fragments of the crystal structure
(Figure 9) setting the value of the dielectric constant in the
Energies and transition dipoles were computed for a molecule
of NP3MP using the geometry from the crystal structure. The

Investigation of Amphiphiles
J. Phys. Chem. B, Vol. 107, No. 1, 2003 155
TABLE 3: Calculated (AM1/CI Including COSMO with E )
80) Absorption Peak Positions and Transition Dipoles for
Various Supramolecular Fragments of the Crystals of
NP3OP Represented in Figure 9
cluster
peak position (nm) [transition dipole (eÅ)]
368.2 247.8 210.8
[1.09] [0.71] [1.01]
A
339.2a
[1.03]
335.0
[1.56]
327.6
[0.82]
327.0
[1.84]
356.0
[0.30]
349.6
[0.91]
333.3
[0.33]
347.9
[0.41]
311.4
[1.22]
341.4
[0.25]
291.6
[1.14]
235.9a
[0.81]
312.1
[0.30]
261.5
[0.41]
290.9
[0.62]
326.8
[0.67]
341.7
[0.25]
315.5
[1.79]
335.1
[1.24]
241.5
[0.59]
337.3
[1.83]
209.7a
[1.01]
236.3
[0.74]
205.5
[0.62]
249.6
[0.52]
313.8
[2.08]
312.9
[1.98]
243.4
[0.31]
319.2
[1.82]
B
C
D
E
F
G
H
I
217.7
[1.05]
182.0
[0.25]
J
K
a These values correspond to ꢀ ) 2 in the COSMO model.
Figure 10. Absorption spectrum (Kubelka-Munk transformation of
the reflectance spectrum in Figure 8) of (a) solid NP3OP and (b) 3-layer
LB film of 1:1 NP3OP-DSPC. The vertical lines represent the AM1/
CI calculated absorption peaks for the clusters shown in Figure 9; the
intensities are proportional to the product of the excitation energy and
the square of the transition dipole or the oscillator strength.
observed in this region of the solid-state spectrum appears to
result from some kind of a J-aggregate structure which may
not be a part of the periodic crystal lattice. The Kubelka-Munk
transformed spectrum of the three-layer LB film is shown in
Figure 10b. All of the prominent features in the spectrum are
accounted for using selected clusters (B, C, D, E) from the set
in Figure 9. Clearly, a subset of small fragments picked from
the 3-dimensional crystal lattice, represent satisfactorily the
interacting molecules in the quasi-2-dimensional LB film.
Taking into account the possibility that LB films often exhibit
microheterogeneous morphology, the present study shows that
the crystal structure provides a useful basis to analyze the
average nature of molecular associations and interactions in the
LB film.
Figure 9. Schematic diagram of molecular clusters of NP3OP used
in the AM1/CI calculations. The inset shows the meaning of the symbols
used.
solvation model at 80 to mimic the polar environment due to
surrounding molecules in the crystal. The computed excitations
with wavelengths greater than 180 nm and transition dipoles of
at least 0.25 are collected in Table 3. The peaks scaled up by
20 nm are represented schematically in Figure 10a superposed
on the absorption spectrum of solid NP3OP obtained using the
Kubelka-Munk transformation of the reflectance spectrum in
Figure 8. The calculated peaks satisfactorily explain a major
part of the observed spectrum and indicate that the observed
peaks arise because of the electronic excitations in various
groups of molecules that can be envisioned in the crystal lattice.
Despite exploring a wide range of further clusters, no calculated
excitations were found in the region above 400 nm. The peak
Conclusion
We have presented in this paper a combination of crystal-
lographic investigations and Langmuir/LB film studies of new
amphiphiles based on the 4-nitroaniline moiety of interest in
quadratic NLO applications. The crystal structure provided
insight into molecular level interactions that possibly hamper
stable monolayer formation of 4-nitroaniline-based amphiphiles
in general. The orthogonal orientation of the 4-nitroaniline
groups observed in the crystals is likely to make it difficult for
the headgroups of the amphiphiles to arrange in eclipsed or

156 J. Phys. Chem. B, Vol. 107, No. 1, 2003
Sharma and Radhakrishnan
(6) (a) Moran, A. M.; Kelley, A. M. J. Chem. Phys. 2001, 115, 912.
(b) Mansour, G.; Creedon, W.; Dorrestein, P. C.; Maxka, J.; MacDonald,
J. C.; Helburn, R. J. Org. Chem. 2001, 66, 4050. (c) Gangopadhyay, P.;
Radhakrishnan, T. P. Angew. Chem., Int. Ed. 2001, 40, 2451.
(7) (a) Inoue, T.; Yase, K.; Okada, M.; Okada, S.; Matsuda, H.;
Nakanishi, H.; Kato, M. Jpn. J. Appl. Phys. 1988, 27, 1635. (b) Miyamoto,
Y.; Kaifu, K.; Saito, M.; Kato, M.; Kawamura, K. Thin Solid Films 1989,
178, 493. (c) Decher, G.; Tieke, B.; Bosshard, C.; Gu¨nter, P. Ferroelectrics
1989, 91, 193. (d) Shin, D.; Park, M.; Lim, S. Thin Solid Films 1998, 327-
329, 607. (e) Nalwa, H. S.; Watanabe, T.; Ogino, K.; Sato, H.; Miyata, S.
J. Mater. Sci. 1998, 33, 3699. (f) Munn, R. W.; Szczur, O. Mol. Cryst. Liq.
Cryst. 2001, 355, 305.
slipped parallel π stacks to develop into extended monolayers.
We have demonstrated the utility of DSPC monolayers in
accommodating these amphiphiles in a composite bilayer motif.
The crystallographic characterization of the amphiphilic mol-
ecules enabled us to develop a viable model for the complex
pressure-area isotherms of the composite monolayers contain-
ing the amphiphiles. The electronic absorption spectrum together
with the computational modeling leads to a picture of the
average intermolecular interactions between the push-pull
chromophore units in the LB films in terms of supramolecular
fragments of the crystal.
(8) Xtal3.4; Hall, S. R., King, G. S. D., Stewart J. M., Eds.; University
of Western Australia: Perth, Australia, 1995.
(9) SHELX-97, Sheldrick, G. M., University of Go¨ttingen: Go¨ttingen,
Germany, 1997.
(10) Dewar, M. J. S.; Zoebisch, E. G.; Healy, E. F.; Stewart, J. J. P. J.
Am. Chem. Soc. 1985, 107, 3902.
This study provides a systematic demonstration of the
correlation between 2-dimensional or quasi-2-dimensional Lang-
muir/LB films and 3-dimensional crystal lattice. We have carried
out similar crystallographic and LB film studies on N-octadecyl-
2,4-dinitroaniline and N-octadecyl-N-(4-nitrophenyl)-4-nitro-
aniline which shows that the approach presented in this paper
can be extended to more systems. Further characterizations of
the LB films and nonlinear optical studies are under way.
(11) MOPAC93; Fujitsu Inc.
(12) Klamt, A.; Shu¨u¨rmann, G. Perkin Trans. 1993, 799.
(13) (a) Sharma, S.; Radhakrishnan, T. P.; J. Phys. Chem. B 2000, 104,
10191. (b) Jayanty, S.; Radhakrishnan, T. P.; Chem. Mater. 2001, 13, 2460l.
(c) Sharma, S.; Sharat Chandra, M.; Radhakrishnan, T. P. Langmuir 2001,
17, 8118.
(14) Gust, D.; Moore, T. A.; Moore, A. L.; Luttrull, D. K.; DeGraziano,
J. M.; Boldt, N. J.; Van der Auwerar, M.; De Schryver, F. C. Langmuir
1991, 7, 1483.
Acknowledgment. Financial support from the Department
of Science and Technology, New Delhi (Swarnajayanti
Fellowship), the use of the National Single Crystal Diffracto-
meter Facility funded by the DST at the School of Chemistry,
University of Hyderabad and technical support from Nima
Technology, U.K. are gratefully acknowledged. S.S. thanks the
CSIR for a senior research fellowship. We thank Dr. N. M. Rao
(Center for Cellular and Molecular Biology, Hyderabad) and
Dr. M. J. Swamy for fruitful discussions and valuable sugges-
tions.
(15) Sharma, S.; Radhakrishnan, T. P. Thin Solid Films, 2001, 382, 246.
(16) (a) Vogel, V.; Mo¨bius, D. Thin Solid Films 1985, 132, 205. (b)
McConnell, H. M. Annu. ReV. Phys. Chem. 1991, 42, 171. (c) Hwang, J.;
Tamm, L. K.; Bo¨hm, C.; Ramalingam, T. S.; Betzig, E.; Edidin, M. Science
1995, 270, 610. (d) Schwartz, D. K. Surf. Sci. Rep. 1997, 27, 241. (e)
Ashwell, G. J.; Leeson, P.; Bahra, G. S.; Brown, C. R. J. Opt. Soc. Am. B
1998, 15, 484. (f) Overs, M.; Fix, M.; Jacobi, S.; Chi, L. F.; Sieber, M.;
Scha¨fer, H. F.; Fuchs, H.; Galla, H. J. Langmuir, 2000, 16, 1141.
(17) (a) Barraud, A.; Flo¨rsheimer, M.; Mo¨hwald, H.; Richard, J.;
Ruaudel-Teixier, A.; Vandevyver, M. J. Colloid Interface Sci. 1988, 121,
491. (b) Lavigne, P.; Tancre`de, P.; Lamarche, F.; Grandbois, M.; Salesse,
C. Thin Solid Films 1994, 242, 229. (c) Goldenberg, L. M.; Becker, J. Y.;
Levi, O. P.; Khodorkovsky, V. Y.; Shapiro, L. M.; Bryce, M. R.; Cresswell,
J. P.; Petty, M. C. J. Mater. Chem. 1997, 7, 901.
(18) Okuyama, K.; Watanabe, H.; Shimomura, M.; Hirabayashi, K.;
Kunitake, T.; Kajiyama, T.; Yasuoka, N. Bull. Chem. Soc. Jpn. 1986, 59,
3351.
(19) (a) Gangopadhyay, P.; Sharma, S.; Rao, A. J.; Rao, D. N.; Cohen,
S.; Agranat, I.; Radhakrishnan, T. P. Chem. Mater. 1999, 11, 466. (b)
Gangopadhyay, P.; Rao, S. V.; Rao, D. N.; Radhakrishnan, T. P. J. Mater.
Chem. 1999, 9, 1699. (c) Sharma, S.; Radhakrishnan, T. P. Mol. Cryst.
Liq. Cryst. 2000, 338, 257.
Supporting Information Available: Synthesis details of
NP3OP, NPOHP, and NPOOP, full crystallographic data on
NP3OP and NPOHP, computational results on the molecular
clusters, and summary of crystallographic database search
(18 pages). This material is available free of charge via the
Internet at http://pubs.acs.org.
References and Notes
(20) Desiraju, G. R. Acc. Chem. Res. 1996, 29, 441.
(1) (a) Roberts, G. Langmuir- Blodgett Films; Plenum Press: New
York, 1990. (b) Tredgold, R. H. Order in Thin Organic films; Cambridge
University Press: Cambridge, 1994. (c) Petty, M. C. Langmuir- Blodgett
Films - An Introduction; Cambridge University Press: Cambridge, 1996.
(2) Cambridge Structural Database, version 5.21.
(3) Yam, M. W.; Lau, V. C.; Wang, K.; Cheung, K.; Huang, C. J.
Mater. Chem. 1998, 8, 89.
(21) Bos, M. A.; Nylander, T. Langmuir 1996, 12, 2791.
(22) (a)Kawabata, Y.; Sekiguchi, T.; Tanaka, M.; Nakamura, T.; Komizu,
H.; Honda, K.; Manda, E. J. Am. Chem. Soc. 1985, 107, 5270. (b) Kurata,
T.; Tsumura, A.; Fuchigami, H.; Koezuka, H. J. Phys. Chem. 1991, 95,
8831.
(23) Dynarowicz-Ła¸tka, P.; Kita, K. AdV. Colloid Interface Sci. 1999,
79, 1.
(24) Girard-Egrot, A. P.; More´lis, R. M.; Coulet, P. R. Langmuir 1996,
12, 778.
(4) Zyss, J.; Nicoud, J. F.; Coquillay, M. J. Chem. Phys. 1984, 81,
4160.
(5) Oudar, J. L.; Hierle, R. J. Appl. Phys. 1977, 48, 2699.