Floating films of a nonamphiphilic porphyrazine at the air-water interface and LS multilayer construction and optical characterization

Article abstract of DOI:10.1021/jp036695i

2,3,7,8,12,13,17,18-Octakis(methylthio)-5,10,15,20-21H,23H-porphyrazine, H₂OMTPz, was used for the preparation of thin films by the Langmuir-Shaì?fer technique. The related analysis of the floating layer on the water surface has revealed the existence of large and randomly oriented three-dimensional aggregates, investigated by Brewster angle microscopy and by UV-vis reflection spectroscopy. The transfer of the floating layer onto the solid hydrophobized substrates (quartz and glass) originated films that on the contrary have exhibited, by UV-vis spectroscopic analysis with polarized light, an average preferential edge-on arrangement of porphyrazine molecules with respect to the support surface. Finally, spectroscopic ellipsometry allowed us to measure the film thickness (consistent with previous data) and optical parameters.

Full text of DOI:10.1021/jp036695i

7854
J. Phys. Chem. B 2004, 108, 7854-7861
Floating Films of a Nonamphiphilic Porphyrazine at the Air-Water Interface and LS
Multilayer Construction and Optical Characterization
Giampaolo Ricciardi,† Sandra Belviso,† Gabriele Giancane,‡ Raffaele Tafuro,‡
Thomas Wagner,§ and Ludovico Valli*,‡
Dipartimento di Chimica, UniVersita` della Basilicata, Via N. Sauro, 85, I-85100 Potenza, Italy, Dipartimento
di Ingegneria dell’InnoVazione, UniVersita` degli Studi di Lecce, Via Monteroni, I-73100 Lecce, Italy, and
L.O.T.sOriel GmbH & Co., Im Tiefen See 58, D-64293 Darmstadt, Germany
ReceiVed: September 9, 2003; In Final Form: April 13, 2004
2,3,7,8,12,13,17,18-Octakis(methylthio)-5,10,15,20-21H,23H-porphyrazine, H2OMTPz, was used for the
preparation of thin films by the Langmuir-Sha¨fer technique. The related analysis of the floating layer on the
water surface has revealed the existence of large and randomly oriented three-dimensional aggregates,
investigated by Brewster angle microscopy and by UV-vis reflection spectroscopy. The transfer of the floating
layer onto the solid hydrophobized substrates (quartz and glass) originated films that on the contrary have
exhibited, by UV-vis spectroscopic analysis with polarized light, an average preferential edge-on arrangement
of porphyrazine molecules with respect to the support surface. Finally, spectroscopic ellipsometry allowed us
to measure the film thickness (consistent with previous data) and optical parameters.
Introduction
groups in Pcs could exert a sort of sterical hindrance in
comparison with the case of Pzs.3
The preparation of new well-designed and -performing
substances acting as building blocks for complex supramolecular
structures and films, operating as active layers in functioning
devices, is one of the most investigated subjects in the
interdisciplinary field of materials science. In this light, the high
self-organizing aptitude of porphyrazines (also denoted as
tetraazaporphyrins) (Pzs) in the solid state, ascribable above all
to the effective attracting interactions among their delocalized
electron densities, the well-known π-π stacking forces, offers
the opportunity to construct organized supramolecular structures.
It is notorious the pronounced resemblance between the two
macrocyclic systems of phthalocyanines (Pcs) and Pzs above
all for the absorption spectra in the UV-vis region.1 The
nitrogen atoms in the meso positions induce supplementary
electronic transitions with respect to other macrocycles such as
porphyrins. Electronic absorption data accumulated so far
indicate that the absorption coefficients of Pzs are generally
smaller than those of Pcs having the same central metal and
peripheral substituents.2 Of course, also from a chemical point
of view, Pcs and Pzs are analogous two-dimensional π-conju-
gated systems; for example, both macrocycles may coordinate
many metal ions and have different pendant groups on the
periphery. But, to the best of our knowledge, Pzs have rarely
been studied in the field of materials science so far, especially
in comparison with Pcs; therefore, this paper represents a
contribution to weigh up the potential of Pzs as film-forming
materials and, consequently, as possible active layers in
functioning devices for future applications (for example, NLO
and chemical sensors). The main rationale of this choice is that,
since the central ring has the most important task in many
phenomena, such as light absorption or charge transport, benzo
The first tetraazaporphyrin, magnesium octaphenyltetraaza-
porphyrin, was prepared by Cook and Linstead during the
1930s.4 Moreover, it is long well-known that substituted Pzs
are important catalysts5 or pigments,6 are suitable for the
preparation of modified electrodes,7 and, in the form of thin
films, are of growing interest as active layers for nonlinear
optics8-10 or chemical sensors,11 as organic semiconducting
materials,12 and as model compounds for the analysis of
chemical reactivity in monolayers at the air-water interface.13-15
To carry out the construction of ultrathin films containing
Pz derivatives onto solid supports, several techniques such as
self-assembly and Langmuir-Blodgett (LB) methods (in some
films Pzs have been deposited in mixture with film promoting
substances, such as fatty acids or long chain alcohols)3,16-18
and spin-coating9 have been utilized. Over the past decades,
the Langmuir-Blodgett (LB) method has been widely used to
construct ordered molecular assemblies of different macrocycles
with well-controlled composition, thickness, and architecture.19
Thermal stability, rigidity, crystallinity, anisotropy, metal
complex formation, gas adsorption, and possible polymerization
are some desirable properties of such multilayers. The conju-
gated aromatic systems of the Pzs are good examples satisfying
these requirements and, in addition, they evidence other
interesting properties of semiconductors and photoconductors
and demonstrate photovoltaic and nonlinear optical effects.
Moreover, it has already been reported that usually such Pz
assemblies contain three-dimensional structures rather than
monodimensional molecular piles. Such aggregates are probably
ascribable to electron hopping involving the delocalized electron
density in the macrocycles rather than other weak physical
forces.3
All these considerations prompted us to use a Pz derivative,
2,3,7,8,12,13,17,18-octakis(methylthio)porphyrazine,20,21
H2OMTPz, in the construction of multilayered films and to
check the structure of the floating film at the air-water interface
not only by the registration of Langmuir isotherm, but above
* Corresponding author. E-mail: ludovico.valli@unile.it. Tel.:
+39.0832.297325. Fax: +39.0832.297525.
† Universita` della Basilicata.
‡ Universita` degli Studi di Lecce.
§ L.O.T.sOriel GmbH & Co.
10.1021/jp036695i CCC: $27.50 © 2004 American Chemical Society
Published on Web 05/18/2004

Floating Films of a Nonamphiphilic Porphyrazine
J. Phys. Chem. B, Vol. 108, No. 23, 2004 7855
1
all through utilization of two innovative techniques, Brewster
angle microscopy (BAM) and reflection spectroscopy in the
UV-vis wavelength range. Also the film structure onto solid
substrates was explored by transmission UV-vis spectroscopy
and ellipsometry. Future developments will include also the
utilization of this derivative in resistive chemical sensors for
gases and vapors. Similar octaalkylthio-substituted Pzs have
been reported to evidence semiconductor properties, the specific
conductivities ranging from 10-11 to 10-9 S/cm.21
C, 42.20; H, 3.84; N, 16.41; S, 37.55. H NMR (499.6 MHz,
CDCl3, 297 K), δ/ppm: 3.42 (s, 24 H, S-CH3), -3.26 (s, 2
H). UV-vis (CHCl3), λmax/nm (log ꢀ): 356 (4.50) Soret; 504
(4.16), 621 (4.24), 700 (4.34) Q-bands.
Langmuir Experiments. Chloroform (Fluka, HPLC grade)
was used in making up the spreading solution: 2.22 mg of
H2OMTPz was completely dissolved in 10 mL of chloroform
(concentration of 3.3 × 10-4 M). Langmuir experiments were
carried out using a KSV5000 System3 apparatus (subphase
surface of 850 cm2). Ultrapure water (resistivity greater than
18 MΩ cm) from a Milli-Q/Elix3 Millipore system was used
as the subphase (pH ≈ 5.9). It was thermostated at 293 K by a
Haake GH-D8 apparatus. A 200 µL aliquot of the spreading
solution was spread onto the subphase. After solvent evapora-
tion, the floating film was compressed at a speed of 7 Å2
molecule-1 min-1. During the depositions, the transfer surface
pressure was fixed at 20 mN m-1. Glass and quartz substrates
were all rendered hydrophobic before deposition by storing
overnight in a desiccator in contact with vapors of 1,1,1,6,6,6-
hexamethydisilazane. The transfer onto all substrates was
performed by the horizontal lifting method (Langmuir-Scha¨fer
technique).
Reflection Spectroscopy and Brewster Angle Microscopy.
Reflection spectroscopy and Brewster Angle Microscopy analy-
sis were carried out using a NIMA 601BAM apparatus. Also
in this case, a compression speed of 7 Å2 molecule-1 min-1
was utilized, after evaporation of chloroform spreading solvent.
The reflection data (∆R) here reported were obtained by an NFT
RefSpec instrument. They were acquired under normal incidence
of radiation according to the description given in ref 24 and
correspond to the difference between the reflectivities of the
floating film-liquid interface and the clean air-liquid interface.
All reflection spectra were carried out at 293 K. Concerning
BAM measurements, they were obtained by a NFT BAM2plus
system with a lateral resolution of 2 µm.
UV-vis Spectroscopy. Solution electronic spectra in 0.1 cm
path length quartz cells in the region 250-800 nm were
performed on a UV-vis-NIR 05E Cary spectrophotometer
equipped with a polarized spectroscopy setup. A Glan-Taylor
polarizer was set between the lamp and sample to obtain the
desired polarized UV-vis irradiation.
Ellipsometry. The data were measured on a variable angle
spectroscopic ellipsometer (VASE) manufactured by the J.
A.Woollam Co. It was equipped with an AutoRetarder, which
utilizes patented technology to achieve maximum measurement
accuracy. VASE measurements for this report were taken from
280 to 2300 nm (0.54-4.43 eV) and at multiple angles of
incidence (60°-75° by 5°). The beam diameter was fixed at
1.5 mm. All data were acquired and analyzed (WVASE version
3.422). In addition, transmission intensity spectroscopic data
had been acquired and fit together with the ellipsometric data.
Experimental Details
Synthesis. All chemicals and solvents (Aldrich Chemicals
Ltd.) were of reagent grade and used in the syntheses as supplied
without further purification. Silica gel used for chromatography
was Merck Kieselgel 60 (70-230 mesh). Solvents used in
physical measurements were of spectroscopic or HPLC grade.
1
The proton H NMR spectrum was recorded on a 500 MHz
INOVA Varian spectrometer with SiMe4 as internal standard.
H2OMTPz: 2,3,7,8,12,13,17,18-Octakis(methylthio)-5,10,-
15,20-porphyrazine. The free-base porphyrazine was prepared
according to a modified literature method.20,22,23 This method
involved the synthesis of cis-1,2-dicyano-1,2-bis(methylthio)-
ethylene by reaction of disodium cis-1,2-dicyano-1,2-ethylene-
dithiolate with iodomethane in MeOH and its subsequent
template condensation in a suspension of magnesium isopro-
poxide to form MgOMTPz. Dissolution of the Mg derivative
in cold, concentrated CF3COOH followed by careful neutraliza-
tion produced H2OMTPz in high yield.
cis-1,2-Dicyano-1,2-bis(methylthio)ethylene (1). Iodomethane
(6.09 g, 43.0 mmol), dissolved in 5.0 mL of methanol, were
added dropwise to a vigorously stirred suspension of disodium
cis-1,2-dicyano-1,2-ethylenedithiolate (4.0 g, 21.5 mmol) in
methanol (15.0 mL) cooled by water-ice bath (0 °C). The
temperature was increased to 25 °C and the solution was stirred
continuously for 24 h in the dark. After removal of the solvent
under vacuum, the compound was dissolved in CHCl3, washed
by water, dried over Na2SO4, and filtered. The compound, freed
of the solvent using a rotary evaporator, appeared as a very
malleable yellow-brown solid. The crude product was carefully
purified by flash chromatography on silica gel using (1:1)
(v/v) CH2Cl2/n-hexane as eluent. The yield of the reaction was
>70%.
[2,3,7,8,12,13,17,18-Octakis(methylthio)-5,10,15,20-por-
phyrazinato]magnesium(II), MgOMTPz (2). Mg powder (0.3
g), previously washed with diethyl ether and dried, was refluxed
for 8 h in n-propanol (50 mL). Compound 1 was added to the
suspension under stirring and the solution was refluxed for 36
h. Subsequent addition of water to the cooled reaction mixture
led to a dark green solid that was collected by filtration. The
crude product was carefully purified by flash chromatography
on silica gel using (1:1) (v/v) CH2Cl2/n-hexane as eluent (first
band) (yield 70%).
Results and Discussion
[2,3,7,8,12,13,17,18-Octakis(methylthio)-5,10,15,20-21H,23H-
porphyrazine], H2OMTPz. Pure solid 2 was dissolved in a
small amount of concentrated CF3COOH and carefully trans-
ferred on ice water. The solution was then washed with a NH3
solution (30%) until the washing water was fully neutralized.
The dark product was collected with CHCl3 using a separating
funnel, dried over sodium sulfate, and filtered. After removal
of the solvent, the crude product was carefully purified by flash
chromatography on silica gel (first band) using (7:3) (v/v) CH2-
Cl2/n-hexane as eluent. The free-base porphyrazine was obtained
in a yield of ca. 65% with respect to MgOMTPz. Anal. Found:
C, 42.15; H, 3.70; N, 16.25; S, 37.80. Calcd for C24H26N8S8:
Langmuir Experiments. The reproducible Langmuir iso-
therm of floating films of H2OMTPz is illustrated in Figure 1.
The Π vs A curve does not evidence a peculiar feature other
than the smooth transition to the condensate state, as already
observed in a previous study by the other researchers for a
different tetraazaporphyrin.25 The limiting area per molecule,
AΠf0 (obtained by extrapolation of the steepest portion of the
curve to 0 mN/m pressure), is 16 Å2. Density functional
calculations26 have suggested that the area per molecule of H2-
OMTPz is at least 200 Å2 when the molecule is prone on the
water surface or about 70 Å2 when it assumes an edge-on

7856 J. Phys. Chem. B, Vol. 108, No. 23, 2004
Ricciardi et al.
the H2OMTPz molecules out of the interface. We have also
tried to improve the floating film homogeneity waiting for long
periods after spreading (at least 2 h) in order to allow unspread
fractions to diffuse to the water surface. But also in these cases,
no variations in the behavior of the floating films (through Π
vs A curves and BAM analysis) was monitored.
The collapse pressure of the monolayer is high (about 55 mN/
m), thus suggesting the high stability of H2OMTPz floating film
on the water surface. The film can sustain a surface pressure of
40 mN/m for at least 12 h with negligible area loss.
BAM investigations were carried out during the compression
of the floating layer and gave a further confirmation about the
presence of large aggregates on the water surface. The con-
temporaneous subsistence of three-dimensional aggregates and
clean water surface was observed even just after solvent
evaporation and at low surface pressures and low area densities.
It is possible to distinguish bright regions of 3D aggregates and
dark areas of pure water surface. Aggregates of H2OMTPz of
different sizes are clearly evidenced at 0.5 mN/m (area per
molecule of about 22 Å2) onto the water subphase and are
illustrated in Figure 2a. During compression such domains
coalesce and broaden; they begin to evidence more clearly facets
and different shades of gray, suggesting the presence of
aggregates with different thickness and arrangements. This is
illustrated in Figure 2b, taken at a surface pressure of 2 mN/m
(area per repeat unit of about 19 Å2). Further compression at
higher surface pressures permits generation of progressively
larger clusters on the average, even though for Π > 15 mN/m
the morphology of the floating film appeared less sensitive to
surface pressure variations than for smaller surface pressures.
Such a trend is illustrated in Figure 2c-f, which cover the range
from Π ) 5 to 40 mN/m. This behavior is also consistent with
the pattern shown in the Langmuir isotherm, whose slope is
practically constant above 15 mN/m. In some cases, wide
Figure 1. Langmuir isotherm of H2OMTPz at 293 K. Spreading
solvent, CHCl3; concentration, 3.3 × 10-4 M; spread volume, 200 µL;
barrier speed, 7 Å2 molecule-1 min-1. In the inset, one of the possible
conformations accessible to the H2OMTPz molecule is illustrated (see
text).
configuration at the air-water interface. On the contrary, the
analysis of the Langmuir isotherm evidences that the observed
value of AΠf0 (16 Å2) is too small in comparison with both
values corresponding to the two previous limit orientations. A
plausible rationale of such a small value of AΠf0 is that, upon
compression of the floating layer, a significant fraction of
molecules are forced out of the interface in different arrange-
ments. It is also probable that a small percentage of molecules
are effectively at the air-water interface, while the rest
essentially extends far from the contact with the water subphase
because the hydrophobic nature of the whole macrocycle forces
Figure 2. BAM images of the floating layer of H2OMTPz at different surface pressures and average areas per repeat unit: (a) Π ) 0.5 mN/m and
A ) 22 Å2; (b) Π ) 2 mN/m and A)19 Å2; (c) Π ) 5 mN/m and A ) 17 Å2; (d) Π ) 15 mN/m and A ) 14 Å2; (e) Π ) 30 mN/m and A ) 11.5
Å2; (f) Π ) 40 mN/m and A ) 10 Å2. The field of view along the x axis is 430 µm.

Floating Films of a Nonamphiphilic Porphyrazine
J. Phys. Chem. B, Vol. 108, No. 23, 2004 7857
domains at least as large as the field of view of the BAM
instrument (430 µm) were evidenced (Figure 2d).
Such a rigid floating film is not transferable by the usual
vertical dipping method, but multilayers have been fabricated
by the horizontal lifting (Langmuir-Scha¨fer) technique onto
different substrates. Films containing up to 160 layers were
deposited, giving dark green multilayers.
The UV-vis absorption spectrum of H2OMTPz reported in
ref 20 for a solution in chlorobenzene contains four maxima at
709, 637, 515, and 367 nm, with molar extinction coefficients
of 35.0 × 103, 25.5 × 103, 20.0 × 103, and 42.3 × 103 L cm-1
mol-1, respectively (comparable values were obtained by us in
different solvents, vide infra). The molecular organization of
H2OMTPz floating layers at the air-water interface was
investigated by spectroscopic measurements through reflection
of light. We have studied reflection of light under normal
incidence at the water surface covered with the tetraazaporphyrin
derivative layer. The reflection method constitutes a powerful
investigation technique on the chromophore behavior on the
water surface and was first introduced by Kuhn and Mo¨bius.27
It is particularly tailored for this purpose, since only chro-
mophores at the interface contribute to the enhanced reflection.
The difference ∆R in reflectivity from the chromophore floating
layer on the subphase and reflectivity from the bare subphase
surface was monitored as a function of wavelength. The
corresponding reflection spectra from H2OMTPz on the water
surface at different fixed surface pressures after reaching
equilibrium (i.e. no variation in surface pressure and enhanced
reflectivity) are shown in Figure 3a. It is apparent that, upon
compression, the reflection enhances because on the average
the surface density grows, while its profile does not change.
The reflection spectra were then normalized to the same surface
density of H2OMTPz by multiplying ∆R by the surface area,
i.e., ∆Rnorm ) ∆RA, where A (nm2 per repeat unit) is taken from
the Langmuir isotherm at the relative value of surface pressure.
The normalized spectra are illustrated in Figure 3b. The
continuous increase in ∆Rnorm, together with the constancy of
the maximum wavelength, suggest that, after initial aggregation
after solution spreading, the self-organized and associated
clusters are dragged on the water surface and coalesce, while
consequently the surface density of the floating layer enhances.
As can be detected, the normalized reflection spectra of
H2OMTPz undergo continuous and monotonic enhancement
during the compression process. This could be also rationalized
considering that, after initial preaggregation, the 3D domains
of aggregated molecules are constrained together but significant
reorganization on the water surface does not take place. This
spectral behavior is also consistent with the markedly hydro-
phobic character of the polymer (from the surface pressure vs
area per repeat unit curve) and the BAM images.
Figure 3. Absolute reflection spectra (a) and normalized reflection
spectra (b) from the floating film on the water surface at different fixed
surface pressures after equilibrium.
dense packing of 2,7,12,17-tetrakis(tert-butyl)-5,10,15,20-por-
phyrazine, determined by intermolecular charge transfer, was
generated.
At last, two shoulders at about 345 and 470 nm are apparent
in the case of the floating film at the air-water interface
(irrespective of the surface pressure). These shoulders propose
probably the existence of more than a single aggregation state
and organization of H2OMTPz molecules. Moreover, a depen-
dence of the relative intensities of the main peak and the
shoulders on the surface tension seems to be absent, thus
suggesting that correspondingly the number of H2OMTPz
macromolecules in the two different associated states does not
vary significantly with Π pressure.
In the ∆R spectra of H2OMTPz molecules on the water
surface, two peaks are apparent at 375 and 506 nm for all
analyzed surface pressures. Their position is not sensitive to
surface tension variations, thus suggesting that the average
arrangement of molecules is not varying while increasing Π
pressure. A large plateau is evidenced in the range between 600
and 700 nm; a possible rationale could be the presence of
another hidden minor peak in this region as suggested also by
Blinov et al.,28 who in the absorption spectra of 2,7,12,17-
tetrakis(tert-butyl)-5,10,15,20-porphyrazine in Langmuir-
Blodgett films detected a Stark effect (relative change in
transmittance). A new band appeared at 600-650 nm in the
Langmuir-Blodgett film; the proposed rationale was that a
UV-vis Spectroscopic Characterization of H2OMTPz in
Solution and in the Langmuir-Sha1fer Film. Solution Spectra.
Before discussing the polarized absorption UV-vis spectra of
the LS films, we examine the salient features of the solution
spectra of the constituent molecule.
Figure 4 shows the electronic absorption spectra of H2OMTPz
in n-hexane, chloroform, and toluene at the same concentration
(10-6 M, a value which is sufficiently low to prevent significant
aggregation phenomena to occur) and experimental conditions.
As inferred from Figure 4, the ground-state absorption spectrum
of H2OMTPz is quite solvent dependent, even in noncoordi-
nating media. In general, the main features of the spectrum are
shifted to the red in the aromatic solvent compared to chloroform

7858 J. Phys. Chem. B, Vol. 108, No. 23, 2004
Ricciardi et al.
Figure 4. Ground-state absorption spectra of H2OMTPz in n-hexane
(continuum line), chloroform (dashed line), and toluene (dotted line).
and, particularly, to the purely aliphatic medium. Additionally,
the peak extinction coefficient of the strong near-UV “Soret”
band is, either in chloroform and in toluene, significantly smaller
than in n-hexane. Similarly, the broad Q-bands and the
prominent absorption with large sulfur lone pair to porphyrazine
ring π* charge-transfer character (SPCT) lying between the
Q-region and the “Soret” band have, either in chloroform and
in toluene, reduced peak intensity. Interestingly, in n-hexane
the Qy(0,0) absorption is partially resolved and shows up as a
shoulder at about 629 nm.
The breadth of the absorption bands, which is large, regardless
of the nature of the solvent, can be ascribed in large measure
to inhomogeneous broadening due to the presence of multiple
accessible conformations in solution and, in the second place,
to the large number of excited states related to the presence of
the heteroatoms at the periphery of the macrocycle. As for the
broadening effects caused by the presence of multiple accessible
conformations in solution, these most likely include rotating/
tilting of the multiple peripheral methylthio groups and their
steric/electronic interactions with each other and with the
prominent lone pair orbitals of the macrocycle bridging nitro-
gens. The differential stabilization of the accessible conforma-
tions on going from aliphatic to aromatic media is, most likely,
at the origin of the observed sensitivity of the energy and
intensity of the B, SPCT, and Q-bands to the nature of the
solvent.
UV-Vis Spectroscopy of the Langmuir-Sha¨fer Film. The LS
films were characterized by UV-vis spectroscopy with both
isotropic and polarized light. The spectrum of a LS film (62
runs/layers) of H2OMTPz is characterized by the Soret band at
355 nm and a SPCT band at 492 nm, at the same wavelengths
as in chloroform, toluene, and n-hexane, respectively. These
bands are slightly broader then in solution, which is typical of
quasisolid spectra. A more significant difference between
solution and solid spectra is observed in the 600-750 nm range,
where the typical Q-bands of alkyl(sulfanyl)porphyrazine
systems are located. The Qx(0,0) absorption that occurs at 693,
700, and 703 nm in n-hexane, chloroform, and toluene,
respectively, shifts to 728 nm, as is usually observed for
tetrapyrrolic systems having a transition from fluid to solid
phase. So for the Qy(0,0) absorption that is observed at 647 nm
in the solid state, a value ∼20 nm lower than in n-hexane is
clearly discernible. Moreover, both the x- and y-polarized
Q-bands are rather broadened (although to an extent not
significantly larger than in solution). Both these spectral features,
i.e., the red shift and broadening of the bands, are indicative of
the fact that non-negligible interactions are operative between
the molecules in their “frozen” conformations in the solid state.
Figure 5. Limiting orientations of the porphyrazine molecules on the
quartz plane: (a) flat orientation; dichroic ratio ) 1; (b) edge-on
orientation to the substrate plane with the normal n of the macrocycle
plane parallel to the dipping direction, R < 1; (c) edge-on orientation
to the substrate plane with the normal n of the macrocycle plane
perpendicular to the dipping direction, R > 1.
(We notice, in passing, that in view of the modest peripheral
crowding, the planarity of the porphyrazine core should be
largely preserved in the methylthioporphyrazine in the con-
densed phase, whatever the relative orientation of the peripheral
methyl groups is. Accordingly, a substantially D2h symmetry
can be assumed for the porphyrazine ring.) When the film is
investigated with polarized light by using the light polarization
plane, E, parallel (Ep) or perpendicular (Es) to the incidence
plane of radiation at an incidence angle I ) 0° of the light beam
with respect to the plane normal, the main absorption features
show a marked dependence on the direction of the electric field
vector with respect to the dipping direction (see Figure 5).
The dichroic ratios R ) A(Ep)/A(Es) (A ) absorbance)
calculated for the 359, 647, and 728 nm absorptions are 0.85,
0.58, and 0.74, respectively. These values indicate that a
significant molecular order is present within the film. To get
an idea of the type of molecular ordering, it is instructive to
consider the polarization dependence expected for three limiting
orientations of the free base porphyrazine with respect to the
lifting direction, as shown in Figure 6.
With molecules flat (face-on) on the quartz substrate (case
a), no polarization dependence of the light absorption would
be expected, because the main absorption bands are polarized
in the xy plane (A(Ep) ) A(Es)). Dichroic ratios smaller than 1
are expected when edge-on-oriented molecules have the normal
to the macrocycle plane parallel to the lifting direction (A(Ep)
< A(Es), case b). Finally, dichroic ratios larger than 1 are
expected when edge-on-oriented molecules have the normal to
the macrocycle plane perpendicular to the lifting direction
((A(Ep) > A(Es), case c). Our results point to case b. The dichroic
ratio calculated for the Qy(0,0) absorption is, furthermore,
significantly smaller than that calculated for the Qx(0,0) absorp-

Floating Films of a Nonamphiphilic Porphyrazine
J. Phys. Chem. B, Vol. 108, No. 23, 2004 7859
highly sensitive method for the investigation of the optical
properties of Langmuir-Blodgett films.32,33 Therefore, a film
consisting of 62 layers, all transferred by the horizontal lifting
(LS) method, was deposited onto a hydrophobized flat quartz
substrate. A polarized light beam is reflected at oblique incidence
from a surface; the reflection at the surface changes the
polarization state of the light, which is measured. The measured
polarization parameters are the relative phase change, ∆, and
the relative amplitude change, Ψ, introduced by reflection from
the surface. By using Fresnel’s relation and, by assuming a
proper model, we can get information about the optical constants
and the thickness of the material.34
∆ and Ψ were measured for each possible wavelength (λ)/
angle (φ) combination (the values for λ were taken in the range
from 280 to 2300 nm and for φ from 60° to 75° by 5°).
Measurements at multiple angles improve confidence, as light
travels on different paths through the film. After an ellipsometric
measurement, data were analyzed to determine the optical
constants and layer thickness of the thin film, the analysis
consisting of multiple steps, as illustrated in the flowchart
herewith reported.
Figure 6. Electronic spectrum in polarized light of the Langmuir-
Sha¨fer film (62 layers) fabricated with H2OMTPz: (dashed line) electric
field vector parallel to the incidence plane; (solid line) electric field
vector perpendicular to the incidence plane.
tion, suggesting that the porphyrazine molecules should face
the substrate preferentially with their the y-edge.
Such variation in the absorption behavior constitutes a further
confirmation that spectral features are determined not only by
the structure of the molecule, but also by the overall organization
of the macrocycles in the film. In particular, as far as the LS
films, and the solid phase in general, are concerned, the mutual
organization and orientation of the macromolecules influence
significantly the characteristics of the spectrum. The optical
properties of such macrocycles could therefore be modulated
not only by ad hoc synthetic procedures but even by the
organization of molecules.
The spectra evidence some noteworthy peculiarities. First of
all, the band is significantly broader and less resolved in the
case of the ∆R spectrum on the water surface and of the LS
film absorbance spectrum in comparison with the solution
absorption. This is typically ascribed to the generation of
aggregates in the floating films on the water surface and in the
multilayers on solid substrates.
The information obtained by the spectroscopic investigations
of the LS films suggests that the hydrophobic surface of the
substrates have induced a profound molecular organization. In
fact, the visible portion of the spectra of the LS films does not
show the huge loss of resolution exhibited by the spectra of the
floating film on the water surface. This completely different
behavior could be ascribed to the pronounced difference
experienced by H2OMTPz molecules when transferring from
the polar water surface to the hydrophobized substrate. This is
not the first time we have experienced such a strong effect
induced by the substrate; in fact, a strong influence of the
substrate on molecular organization was already found for a
symmetrically functionalized phthalocyanine transferred onto
a Mylar substrate.29
The analysis by polarized light suggests that on the average
the preferential orientation of molecules of H2OMTPz is the
so-called edge-on or playing-card arrangement, typical of similar
macrocycles such as phthalocyanines. Therefore, even though
the LS film evidences a high roughness from ellipsometric data
(reported in the following), an average privileged disposition
of H2OMTPz molecules is apparent, also in this case according
to similar data already found for the strictly similar phthalo-
cyanine macrocycles.30,31
Finally, experimental data were compared with theoretical
ones using the mean squared error (MSE).35-37 MSE was
minimized by adjusting each fit parameter, using a Levenberg-
Marquardt fit algorithm,35,37 the lowest MSE suggesting the best
agreement between theory and experiment within the boundary
of the current model.
A graded profile and a thin surface roughness layer were
required in the best model. The gradient described a decrease
in optical constants from bottom to the top. The surface
roughness layer was modeled using a Bruggeman effective
medium approximation,38,39 composed of 50% LS material and
50% void.
In the first step we reduced the spectral range to 900-2300
nm, where the layer is transparent, and applied a Cauchy
dispersion function to fit for the layer thickness parameters and
the Cauchy parameters. The Cauchy dispersion equation40 for
transparent layers is
Bn Cn
λ2 λ4
n(λ) ) An +
+
where An, Bn, and Cn are Cauchy parameters and λ is the
wavelength in microns. Additionally, an Urbach absorption tail
was taken into account.
Ellipsometry. The estimation of film thickness and refractive
index have been performed by ellipsometry, a common and

7860 J. Phys. Chem. B, Vol. 108, No. 23, 2004
Ricciardi et al.
Figure 7. Graphs for Ψ, ∆, and transmission vs wavelength, measured by spectroscopic ellipsometry, and refractive index and extinction coefficient
vs distance from the substrate surface, as well as optical constants, according to the best model.
In the second step, we extended the spectral range further
into the visible and UV by fixing the layer thickness and fitting
n and k for each wavelength point-by-point. The optical
constants had been checked afterward to be Kramers-Kronig
consistent. The graphs for Ψ, ∆, and transmission vs wave-
length, and refractive index and extinction coefficient vs distance
from the substrate surface, as well as the optical dispersion, are
reported in Figure 7.
The main results are herewith reported: roughness thickness
) 10.8 ( 0.33 nm; layer thickness ) 438.9 ( 0.57 nm.
The average film thickness is 438 nm, which implies an
average thickness per LS layer of 7.1 nm. This value is markedly
larger than the hypothetical value of about 1.4 nm for a
homogeneous arrangement of molecules with an edge-on
orientation with respect to the substrate plane. On the other hand,
such a value is in accordance with the morphology of the
floating film at the air-water interface: large 3D-aggregates
are preferentially transferred from the water surface to the solid
support, thus accounting for also the high roughness obtained
by both models. In conclusion, ellipsometric data are consistent
with the information at the air-water interface got by BAM,
reflection microscopy, and Langmuir isotherm. The ratio
between the experimental value for each layer in the LS film
(71 Å) and the theoretical one for the thickness of a uniform
monolayer of H2OMTPz molecules (14 Å) is about 5; on the
other hand, the ratio between the expected value of the limiting
area per molecule for a homogeneous monolayer with an edge-
on arrangement of molecules (70 Å2) and the experimental one
(15 Å2) is 4.7, thus suggesting that the floating layer is almost
five monolayer equivalents thick. It is singular to observe that
the values of the two ratios are in good accordance. This is
probably a further suggestion that the floating layer on the water
surface consists of large 3D aggregates randomly and isotro-
pically oriented.
Conclusions
The main feature emerging from this research is the pro-
foundly different molecular organization at the air-water
interface and onto hydrophobic quartz and glass substrate. The
rational probably is based on the poor amphiphilic character of

Floating Films of a Nonamphiphilic Porphyrazine
J. Phys. Chem. B, Vol. 108, No. 23, 2004 7861
(9) Nalwa, H. S.; Engel, M. K.; Hanack, M.; Pawlowski, G. Appl. Phys.
Lett. 1997, 71, 2070.
H2OMTPz molecules that precludes the formation of a real
homogeneous Langmuir film on the water surface. This also
justifies the presence on the polar subphase of large three-
dimensional aggregates evidenced by BAM images and by the
Langmuir isotherm; correspondingly, the ∆R spectra on the
water surface evidence very broad bands above all in the visible
region of the spectrum. On the contrary, probably the so-called
hydrophobic-hydrophobic interaction promotes a remarkable
change in the disposition of molecules onto solid substrates; in
fact, the analysis using polarized light displays the existence of
a preferred molecular orientation with macrocycles having an
edge-on orientation with respect to the substrate surface. At the
same time, the loss of resolution in the UV-vis absorption
spectra suggests that in the LS films H2OMTPz molecules
thoroughly interact with each other. Finally, also the data from
ellipsometry confirm, through a thickness much larger than the
one expected for a sequential monolayer transfer onto the solid
and a high roughness, the existence of large three-dimensional
aggregates.
(10) Ferna´ndez-La´zaro, F.; D´ıaz-Garc´ıa, M. A.; Sastre, AÄ .; Delhae`s, P.;
Mingotaud, C.; Agullo´-Lo´pez, F.; Torres, T. Synth. Met. 1998, 93, 213.
(11) Snow, A. W.; Barger, W. R. In Phthalocyanines, properties and
applications; Leznoff, C. C.; Lever, A. B. P., Eds.; VCH: New York, 1989.
(12) Simon, J.; Andre, J.-J. Molecular Semiconductors; Springer-
Verlag: Berlin, 1985.
(13) Palacin, S.; A. Ruaudel-Teixier, A.; Barraud, A. J. Phys. Chem.
1986, 90, 6237.
(14) Palacin, S.; Lesieur, P.; Stefanelli, I.; Barraud, A. Thin Solid Films
1988, 159, 83.
(15) Bourgoin, J. P.; Palacin, S.; Vandevyver, M.; Barraud, A. Thin Solid
Films 1989, 178, 499.
(16) van Nostrum, C. F.; Nolte, R. J. M. J. Chem. Soc., Chem. Commun,
1996, 2385.
(17) Valkova, L.; Borovkov, N.; Maccioni, E.; Pisani, M.; Rustichelli,
F.; Erokhin, V.; Paternolli, C.; Nicolini, C. Colloids Surf. A 2002, 198-
200, 891.
(18) Feng, X.-S.; Yu, M.-X.; Liu, H.-G.; Qian, D.-J.; Mu, J. Langmuir
2000, 16, 9385.
(19) Motschmann, H.; Mohwald, H. Handbook of Applied Surface and
Colloid Chemistry 2002, 2, 79.
(20) Schramm, C. J.; Hoffman, B. M. Inorg. Chem. 1980, 19, 383;
Velazquez, C. S.; Fox, G. A.; Broderick, W. E.; Andersen, K. A.; Anderson,
O. P.; Barrett, A. G. M.; Hoffman, B. M. J. Am. Chem. Soc. 1992, 114,
7416.
Future developments of this research will concern the
utilization of such films as active layers in resistive chemical
gas sensors, above all in the light of the high specific
conductivities evidenced by other octaalkylthio-substituted
porphyrazines. Probably the high roughness of the LS films
offers the presence of a large number of active sites for the
interaction with the dopant gases or vapors.
(21) Shaposhnikov, G. P.; Kulinich, V. P.; Osipov, Yu. M.; Smirnov,
R. P. Chem. Heterocycl. Comput. 1986, 1036.
(22) Belviso, S.; Ricciardi, G.; Lelj, F. J. Mater. Chem. 2000, 10, 297.
(23) Lelj, F.; Morelli, G.; Ricciardi, G.; Roviello A.; Sirigu, A. Mol.
Cryst., Liq. Cryst. 1992, 12, 941.
(24) Gru¨niger, H.; Mo¨bius, D.; Meyer, H. J. Chem. Phys. 1983, 79, 3701.
(25) Souto, J.; Aroca, R.; DeSaja, J. A. J. Raman Spectrosc. 1994, 25,
435.
Acknowledgment. This work was supported by Italian
MIUR (Programmi di Ricerca Scientifica di Rilevante Interesse
Nazionale, Anno 2002, Prot. 2002033817_004). The authors
are grateful to Mr. Luigi Dimo for his technical assistance during
all Langmuir experiments. This work is dedicated to Prof. M.
Della Monica; a profoundly heartfelt and due recollection of
Prof. M. Della Monica, master in science and life, constitutes
for L.V. a further impulse in the continuation of his studies.
(26) Ricciardi, G., to be published.
(27) Kuhn, H.; Mo¨bius, D. Monolayer assemblies, in Physical Methods
of Chemistry, Vol. IXB, InVestigations of Surfaces and Interfaces; Rossiter,
B. W.; Baetzold, R. C., Eds.; Wiley-Interscience, New York, 1993, Part B,
pp 375-542.
(28) Blinov, L. M.; Kopranenkov, V. N.; Palto, S. P.; Yudin, S. G. Opt.
Spectrosc. 1987, 62, 631.
(29) Pasimeni, L.; Segre, U.; Toffoletti, A.; Valli, L.; Marigo, A. Thin
Solid Films 1996, 285, 656.
(30) Manno, D.; Rella, R.; Troisi, L.; Valli, L. Thin Solid Films 1996,
280, 249.
(31) Yoneyama, M.; Sugi, M.; Saito, M.; Ikegami, K.; Kuroda, S.; Iizima,
References and Notes
S. Jpn. J. Appl. Phys. 1986, 25, 961.
(1) Lee, L. K.; Sabelli, N. H.; LeBreton, P. R. J. Phys. Chem. 1982,
86, 3926.
(2) Kobayashi, N. In Phthalocyanines, properties and applications;
Leznoff, C. C.; Lever, A. B. P., Eds.;VCH: New York, 1989; Vol. 2, pp
97-161.
(3) Valkova, L.; Borovkov, N.; Kopranenkov, V.; Pisani, M.; Bossi,
M.; Rustichelli, F. Mater. Sci. Eng. C 2002, 22, 167.
(4) Cook, A. H.; Linstead, R. P. J. Chem. Soc. 1937, 929.
(5) Kaneko, M.; Georg, M. Nissan Chemical Industries, Ltd., Japan
Kokai 75 132 098, 1975.
(6) France, H.; Jones, Wm. O. Imperial Chemical Industries Ltd., British
Patent 689387, 1953.
(7) Toyama, M. M.; Demets, G. J. F.; Araki, K.; Toma, H. E.
Electrochem. Commun. 2000, 2, 749.
(8) de la Torre, G.; Nicolau, M.; Torres, T. In Supramolecular
PhotosensitiVe and ElectroactiVe Materials; Nalwa, H. S., Ed.; Academic
Press: New York, 2001.
(32) Ulman, A. An introduction to Ultrathin Organic Films from
Langmuir-Blodgett to Self-Assembly; Academic Press: San Diego, 1991;
pp 2-5.
(33) Gonella, G.; Cavalleri, O.; Emilianov, I.; Mattera, L.; Canepa,. M.;
Rolandi, R. Mater. Sci. Eng. C 2002, 22, 359.
(34) Azzam, R. M. A.; Bashara, N. M. Ellipsometry and Polarized Light;
North-Holland: Amsterdam, 1977.
(35) Tompkins, H. G.; McGahan, W. A. Spectroscopic Ellipsometry and
Reflectometry; John Wiley & Sons: New York, 1999.
(36) Jellison, G. E., Jr. Appl. Opt. 1991, 30, 3354.
(37) Press: W. H.; Flannery, B. P.; Teukolsky, S. A.; Vetterling, W. T.
Numerical Recipes, Cambridge University Press: Cambridge, 1988.
(38) Bruggeman, D. A. G. Ann. Phys. (Leip.) 1935, 24, 636.
(39) Aspnes, D. E.; Theeten, J. B.; Hottier, F. Phys. ReV. B 1979, 20,
3292.
(40) Cauchy, L. Bull. Des. Sci. Math. 1839, 14, 9.