- Email: [email protected]
Mechanisms of adsorption of an azo-polyelectrolyte onto layer-by-layer films
Sensors and Actuators B 126 (2007) 311–317
Mechanisms of adsorption of an azo-polyelectrolyte onto layer-by-layer films Quirina Ferreira, Paulo J. Gomes, Manuel J.P. Maneira, Paulo A. Ribeiro, Maria Raposo ∗ Centro de Investiga¸ca˜ o em F´ısica Tecnol´ogica (CEFITEC), Faculdade de Ciˆencias e Tecnologia, UNL, 2829-516 Caparica, Portugal Available online 13 January 2007
Abstract The mechanisms which lead to the adsorption of the azopolyelectrolyte poly[1-[4-(3-carboxy-4-hydroxyphenylazo) benzene sulfonamido]-1,2ethanediyl, sodium salt] (PAZO) layer adsorbed onto layer-by-layer films were investigated by monitoring the adsorbed amount per unit of area as a function of time. The adsorption kinetics curves were obtained as a function of adsorption variables as PAZO concentration and drying process. The PAZO adsorption kinetics data was shown to be described by two processes, one with smaller adsorption characteristic time, of the order of seconds, which was attributed to a nucleation mechanism and a second process, with adsorption characteristic times of the order of tenth of minutes, associated to diffusion. During the first process, PAZO molecules rapidly adsorb onto the substrate surface creating an electrostatic barrier that makes difficult the adsorption of more PAZO molecules. However, positive counterions are allowed to be adsorbed and as they are the potential barrier is lowered, allowing more PAZO molecules to be adsorbed. This last mechanism is accounting for the second adsorption process. The adsorption processes are not seen to be influenced by the layer drying and only contribute for the changes in the values of adsorption parameters, i.e., adsorption characteristic times and corresponding adsorbed amounts. These parameters were related with the available number of adsorption sites which are dependent of the presence of both water and counterions adsorbed on the last adsorbed layer. © 2007 Elsevier B.V. All rights reserved. Keywords: Layer-by-layer; Adsorption; Polyelectrolyte; Water network; Azo-polyelectrolyte; PAZO
1. Introduction The main interest to study the formation of PAZO layer is due to its photoisomerization capabilities about the N N bond which makes it suitable for multipurpose device applications as holographic image storage, optical switching, optical computation, reversible optical storage, optical relief refractive and index gratings, digital video recorders, photosensitive artificial membranes and eventually molecular-machines [1]. The adsorption kinetics of polyelectrolytes onto solid surfaces, obtained by measuring the adsorbed amount with time, has been investigated for several decades by adsorbing them onto solid substrates, usually silica spheres, that where initially immersed in an aqueous solution and the polyelectrolyte injected afterwards while the solution is being stirred. Consequently, the polyelectrolyte adsorption kinetics has a characteristic time which is dependent of the polyelectrolyte diffusion process in the solvent and
∗
Corresponding author. Tel.: +351 965285266; fax: +351 212948549. E-mail address: [email protected] (M. Raposo).
0925-4005/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.snb.2007.01.007
of the polyelectrolyte interactions with the substrate, assuming that the polyelectrolyte concentration, pH and ionic strength and other factors as temperature and pressure are held constant during adsorption. With the advent of the layer-by-layer (LbL) technique [2–4], which consists of sequential adsorption cationic and anionic polyelectrolytes from aqueous solutions onto solid substrates, with the adsorption taking place at a planar substrate during a determined period of time. Both, the experimental procedure and, consequently, the adsorption kinetics are different. Although the adsorption process is expected to be also driven by electrostatic forces, different adsorption characteristic times are expected, since the solid planar substrate is immersed rapidly in the polyelectrolyte solution, allowing that the polyelectrolyte molecules are instantaneously in contact with substrate. Moreover, since during the production of LbL films the polyelectrolyte layers are often dried after adsorption other mechanisms should be involved. In fact, experimental results of adsorption kinetics dependence on the layer drying process, obtained for polyanilines and poly(p-phenylene vinylene) (PPV) precursor [5,6], revealed that polyelectrolyte adsorption onto itself is allowed which is odd in what is expected to be
312
Q. Ferreira et al. / Sensors and Actuators B 126 (2007) 311–317
an electrostatic driven process. Different kinetics curves were also obtained for the case of polyanilines and poly(p-phenylene vinylene) precursor [5,6], where it was shown that the number of immersions and the drying of the film influences the film growth kinetics. This behaviour was associated with the polyelectrolyte hydrogen bonding capabilities. However, this could not be the reason for the adsorption of all polyelectrolytes. Consequently, the experimental procedure used to obtain the true kinetics curves must be well controlled when using the LbL technique namely for taking into account the adsorption processes dependence on the surface where adsorption is taking place. As suggested [7,8], the kinetics curves should be obtained by doing only one immersion per layer using different adsorption periods of times, being the kinetics curves the adsorbed amount per layer as a function of immersion time. This was the procedure adopted in this work, to investigate the mechanisms involved on the adsorption kinetics of a PAZO layer, a high degree of charge polyelectrolyte [9], onto already deposited PAH/PAZO LbL films during different periods of time, under different adsorption conditions as solution concentration and drying process. In this article, the mechanisms involved in the adsorption of one PAZO layer onto PAH/PAZO LbL films are discussed which explain the two processes, nucleation and diffusion, which are occurring during the PAZO adsorption. 2. Experimental The poly[1-[4-(3-carboxy-4-hydroxyphenylazo) benzenesulfonamido]-1,2-ethanediyl, sodium salt] (PAZO) kinetics adsorption studies were performed adsorbing PAZO onto (poly(allylamine hydrochloride)(PAH)/PAZO)/PAH LbL films. For that, aqueous solutions of PAH (average Mw = 50,000– 65,000 g/mol) with concentration of 10−2 M and solutions of PAZO with different concentrations in a buffer solution at pH 10 were prepared. The polyelectrolyte concentrations were based on the molecular weight repeat unit. Chemicals were obtained from Sigma–Aldrich and the polyelectrolytes molecular structures are shown in Fig. 1. The pure water with a resistivity of 18.2 Mcm was supplied by a Millipore system (Milli-Q, Millipore GmbH). For pH 10 buffer solution, a 0.05 M sodium bicarbonate (NaHCO3 ) aqueous solution and a 0.1 M sodium hydroxide (NaOH) solution were used in the proportion 500:107 (v/v) [10]. The films were directly deposited on BK7 optical glass substrates for the adsorption kinetics studies. The substrates were hydrophilized with a H2 SO4 /H2 O2 (1:1) bath for 1 h, after which
the glass and silicon slides were washed exhaustively with pure water [5]. Two types of films, dry and wet, were prepared. The dried films were obtained by immersing the substrate in the PAH solution during 3 min, washing it with water, drying it using a nitrogen flux, immersing it during a determined period of time in the PAZO solution, washing it with buffer solution and finally drying it with a nitrogen flux. This procedure gives rise to a bilayer film. Films with several bilayers were obtained by repeating this procedure. The wet samples were obtained immersing the substrate in the PAH solution during 3 min, washing with water, immersing the substrate in the PAZO solution during a determined period of time and washing with the buffer solution. This procedure was repeated during five times and finally, the samples were left to dry at room conditions. The adsorbed amount of PAZO per unit of area was quantified by measuring the UV–vis absorbance spectra of PAZO solutions at different concentrations which were obtained dissolving the weighted PAZO polymer powder in the buffer solution at pH 10 and plotting the maximum absorbance at 360 nm, which is associated to the → * azo-chromophore transition, versus the solution concentration. The obtained experimental points were fitted with a straight line and using the Beer–Lambert law, an extinction coefficient (ε360 nm ) of 4.30 ± 0.07 g−1 m2 at 360 nm for PAZO was estimated. The amount of PAZO per unit of area (Γ ) adsorbed onto a solid substrate is estimated using again the Beer–Lambert law: Γ =
Abs360 nm 2ε360 nm
(1)
The UV–vis absorbance spectra of liquid and solid samples were measured using a Shimadzu UV-2101PC spectrophotometer. 3. Results and discussion 3.1. PAZO adsorption kinetics at pH 10 of dried samples The adsorption kinetics of PAZO onto (PAH/PAZO)n /PAH LbL films was obtained preparing PAH/PAZO LbL films with different PAZO adsorption times and maintaining the adsorption time of 3 min for the PAH. After each adsorption period, the samples were dried with a nitrogen flux and the absorbance spectra of LbL films having different number of bilayers were measured. The graph of Fig. 2 shows the spectra of (PAH/PAZO) LbL films with different number of bilayers obtained for PAZO adsorption time of 30 s. Plotting the maximum absorbance as a function of the number of bilayers for the LbL films prepared
Fig. 1. Polyelectrolyte molecular structures: (a) poly(allylamine hydrochloride) (PAH) and (b) poly[1-[4-(3-carboxy-4-hydroxyphenylazo) benzenesulfonamido]1,2-ethanediyl, sodium salt] (PAZO).
Q. Ferreira et al. / Sensors and Actuators B 126 (2007) 311–317
Fig. 2. UV–vis absorbance spectra of PAH/PAZO LbL films with different number of bilayers obtained for PAZO adsorption time of 30 s.
with different adsorption times, one can observe that the films grow linearly with the number of bilayers as is usual in LbL films [9,11]. The slopes of the fitted straight lines are proportional to the PAZO adsorbed amount per bilayer and from the obtained extinction coefficient the PAZO adsorbed amount per unit of area in each bilayer can be estimated. The adsorption kinetics curves were obtained plotting the PAZO adsorbed amount per unit of area as a function of adsorption time as shown in Fig. 3 for the different PAZO concentrations, with pH 10 and having the samples dried after each adsorption period. Plots are showed in a logarithmic scale in order to emphasize that the kinetics curves are not following a first order exponential growth as reported in previous works [12,13]. Instead of that the current data can be fitted, as already suggested [5,14], with two Johnson–Mehl–Avrami equations: n t t Γ = Γn 1 − exp − + Γd 1 − exp − τn τd (2)
313
where Γ n is the maximum adsorbed amount per unit of area associated with the process which presents the smaller adsorption kinetics characteristic time τ n , Γ d the maximum adsorbed amount per unit of area associated with the process with higher adsorption characteristic time τ d , and n is a constant assuming values between 1 and 1.5. The process with smaller adsorption characteristic times has been associated to a nucleation process and the one with the higher adsorption characteristic time associated to a diffusion like process [5]. These processes are called like this because the formation of the layer starts with the random adsorption of polyelectrolyte molecules which are close to the surface-nucleation of the layer, and ends with a diffusion process associated to adsorption of molecules which migrate through the repulsive electric field and/or the double layer formed near the surface. These mechanisms will be better described ahead. The solid lines in the graph of Fig. 3 are the optimized fittings to Eq. (2). These results are somehow different from the ones obtained from adsorption of POMA onto glass and on POMA/PVS where non-first order kinetics was found for the diffusion process. As a remark, de Souza et al. [14] also showed that the kinetics adsorption of POMA onto POMA/PVS follows Eq. (2) with n = 1. The parameters of Eq. (2) obtained from the fittings are displayed in Table 1. 3.2. PAZO adsorption kinetics at pH 10 of wet samples Drying during the layer formation of LbL films has already been shown to influence the adsorbed amount, the film final structure and surface morphology [15–18]. For example, de Souza et al. [17] showed that the adsorbed amount is higher for films dried in room conditions than those dried under a nitrogen flux, while Lourenc¸o et al. [19] showed that in poly(allylamine hydrochloride)/poly(styrene sulfonate) (PAH/PSS) films the adsorbed amount was higher for samples dried in ambient air than for samples that were never dried. In order to know, if the drying process is affecting the PAZO adsorbed amount, samples of (PAH/PAZO)5 were prepared without drying during the layer production, using different PAZO adsorption times, pH 10 and different concentrations. The amount of PAZO adsorbed per bilayer, i.e., the adsorbed PAZO amount per unit of area obtained from the maximum absorbance divided by the deposited number of bilayers, plotted as function of the PAZO adsorption time is shown in the graph of Fig. 4, together with the kinetics curves of the dried samples. The samples prepared without drying after the adsorption of each layer exhibit lower PAZO adsorbed amount than those that were dried, which is Table 1 Adsorption kinetics parameters obtained from fitting the experimental data with two Johnson–Mehl–Avrami equations
Fig. 3. PAZO adsorption kinetics curves for different PAZO concentrations. After each adsorption period the samples were dried with a nitrogen gas flow. The solid lines correspond to the fits.
Sample
Concentration (M)
Γn (mg m−2 )
τn (s)
Γd (mg m−2 )
τd (s)
Wet Wet Dried Dried Dried
0.25 × 10−2 1 × 10−2 0.25 × 10−2 0.5 × 10−2 1 × 10−2
1.9 3.0 3.8 4.4 4.7
2.3 3.7 5.6 2.4 2.1
1.28 0.46 1.51 1.28 1.16
900 1000 8000 3000 800
314
Q. Ferreira et al. / Sensors and Actuators B 126 (2007) 311–317
Fig. 4. PAZO adsorption kinetics curves for different PAZO concentrations and for both wet and dried samples. The solid lines correspond to the fits.
in accordance with similar results obtained for PAH/PSS LbL films [19]. From Fig. 4 one can also conclude that the adsorption kinetics curves can be interpreted as two first order adsorption processes, so that data can be fitted to Eq. (2). The calculated parameters resulting from this fitting are displayed in Table 1. 3.3. Adsorption characteristics times From the fitting of the kinetics curves to Eq. (2), fitting parameters listed in Table 1, for both type of samples, one can see that all curves are described by two processes having characteristics times laying in two ranges, one range at lower values and at higher values, which are nucleation and diffusion processes consistent. Fig. 5 shows the calculated characteristic adsorption times versus PAZO concentration for the dried samples and reveals that the increase of the concentration is leading to the decrease of the adsorption characteristic time for both adsorption processes. However, for wet samples the characteristic adsorption time can be considered, within the experimental uncertainties, as independent of PAZO concentration.
Fig. 5. Nucleation and diffusion characteristic times versus PAZO concentration. The solid lines are eye guides.
Fig. 6. PAZO maximum adsorbed amount per unit of area for the nucleation process and PAZO adsorbed amount per unit of area versus PAZO concentration. Results obtained in wet samples.
3.4. Adsorbed amount during nucleation For both type of samples, the kinetics curves are at short periods of adsorption times described by a process with similar characteristic time values. However, the parameters obtained from fittings, see Table 1, revealed to be concentration dependent. In Fig. 6, the maximum adsorbed amount due to the nucleation process, Γ n , and the adsorbed amount per unit of area during the first 15 s of adsorption are plotted as a function of the PAZO monomer concentration. These results were obtained from the wet samples kinetics curves. It can be seen that the adsorbed amount due to the nucleation is concentration dependent and follows the adsorbed amount behaviour for short times. This behaviour is consistent with the rapid adsorption of the PAZO molecules when the substrate is immersed in the PAZO solution. Only at very small concentrations, i.e., 0.001 M, the adsorbed amount for short times can be considered proportional to the concentration. A proportional behaviour has also been observed in polyaniline adsorption studies for short adsorption periods and small concentrations [8]. In the present case, for PAZO concentrations higher than 0.003 M, a saturation of the adsorbed amount is observed indicating that an adsorbed amount of about 1.7 mg/m2 is sufficient to compensate the substrate electrical charges where PAZO is being adsorbed. It should be remarked here that PAZO at pH 10 presents a high degree of charging [9] which suggest that electrostatics interactions are mainly accounting for the adsorption. The PAZO adsorbed molecules create a repulsing potential barrier, as predicted by Bo¨ohmer et al. [20], avoiding that more PAZO molecules are being adsorbed in this regime. Similar results were found from the adsorption kinetics of the dried samples. Fig. 7 shows the adsorbed amounts during the nucleation regime for both wet and dried samples. From the graphs of this figure one can clearly see that the PAZO adsorbed amount due to the nucleation process is dependent of drying. Since the PAZO molecules in solution are in the same conditions for both types of samples, this behaviour could be accounted by the increase of surface
Q. Ferreira et al. / Sensors and Actuators B 126 (2007) 311–317
Fig. 7. PAZO maximum adsorbed amount per unit of area for the nucleation process for both wet and dried samples.
roughness as shown by Ribeiro et al. [21]. However, the topographies obtained from atomic force microscopy measurements, for both wet and dried samples with the same number of bilayers, revealed that the dried samples present smaller roughness values. Consequently, the increase of adsorbed amount in the dried samples is consistent with the removal of water molecules and consequent crystallization of adsorbed counterions within the last layer. However, the formed salt crystals will be dissolved when the sample is newly immersed in the next polyelectrolyte solution, since the salt concentration in the polyelectrolyte solution is below of the saturation value, leading to the increase of the number of adsorption sites, in such a way that more polyelectrolyte can be adsorbed. In the case of wet samples, the presence of water and ionic networks diminishes the number of adsorption sites and consequently the amount of adsorbed polyelectrolyte. This explanation is corroborated by a high content of counterions in layer-by-layers films when prepared without drying between adsorption of layer in comparison with films which are dried [19].
315
models have been proposed [22–25]. However, the models do not take into account the process of build-up the polyelectrolyte layer. The model proposed here for the adsorption of a PAZO layer takes that into account and is consistent with the experimental results where it was observed that the adsorption of PAZO molecules onto solid substrates is governed by two processes: nucleation and diffusion. The nucleation process can be explained by the following mechanism: as the positive substrate is immersed in the PAZO solution, the PAZO molecules which are near the substrate could be rapidly adsorbed on the substrate surface, since the free energy associated with the adsorption is smaller than the free energy of the polymer chain in the solution. The two regimes equation used to explain this process, Eq. (2), could in fact be obtained if one considers that adsorbed chains or nucleus are deposited in preferential sites and PAZO molecules in solution are repelled by the already adsorbed ones. The number of vacant sites per unit of area is seen to decrease exponentially leading to an exponential increase to saturation of the adsorbed amount per unit of area. The adsorbed amount being independent of PAZO solution concentration also corroborates this mechanism. The number of sites will be dependent of substrate and PAZO electrical charge and size of PAZO molecules. The presence of PAZO molecules adsorbed on the substrate creates a potential barrier which makes difficult the adsorption of more PAZO molecules. However, at this stage, the sodium counterions are being attracted by the PAZO adsorbed molecules which lead to a decrease of the potential barrier, allowing more PAZO molecules to be adsorbed and giving rise to the second adsorption regime. This behaviour is supported by previous results where the presence of sodium atoms was detected in PAH/PAZO LbL films obtained with a PAZO adsorption time of 3 min [11]. At this point, it is worth to mention that this explanation does not exclude the possibility of reconformation of the last adsorbed macromolecules. However, the effect of diffusion of PAZO molecules into the layer-by-layer is less plausible taking into account the studies about the internal structure of layer-bylayer films [26]. Since these processes are occurring in both types
3.5. Adsorbed amount at long adsorption times Looking at the total PAZO adsorbed amount during the two processes as a function of PAZO concentration, i.e., the adsorption isotherm shown in the plot of Fig. 8, it can seen that in the measured concentration range the total adsorbed amount is practically concentration independent for both wet and dried samples. In fact, the adsorbed amount increases rapidly with the concentration to a saturation value, indicating that the adsorption process is effective even for very small concentrations. This behaviour definitely indicates that the electrostatic interactions are governing the PAZO adsorption process. 3.6. PAZO adsorption model The processes that lead to the adsorption of polyelectrolytes onto solid substrates have been extensively studied and several
Fig. 8. PAZO adsorption isotherms for wet and dried samples. The solid lines are eye guides.
316
Q. Ferreira et al. / Sensors and Actuators B 126 (2007) 311–317
of samples so that they are independent of the drying process, how can the obtained differences in the kinetics parameters for both types of samples be explained? It follows that if the PAZO adsorbed layer is not dried, the adsorbed PAZO molecules will be covered with a network of water molecules and ions [27] and as the film is newly immersed in the next polyelectrolyte solution the presence of the network makes difficult the adsorption of more polyelectrolyte. This is consistent with the smaller adsorbed amounts and higher characteristic times obtained for the same polyelectrolyte concentration—counterions must be replaced by the polyelectrolyte for adsorption to occur. If the adsorbed layer is dried, the water molecules will be removed, counterions crystallizes [19,27] and when the layer is newly immersed in the polyelectrolyte solution, crystals dissolve leaving more sites available for adsorption. 4. Conclusions The adsorption of a PAZO layer onto an already adsorbed PAH/PAZO bilayer is explained by two processes. One at early stages of adsorption corresponds to the instantaneous adsorption of the first PAZO molecules in the surface and is giving rise to a nucleation like process. These first adsorbed PAZO molecules avoid that more PAZO are adsorbed as a result of an electrostatic barrier formation. This effect is somehow attenuated since as PAZO molecules are being adsorbed, positive counterions from solution are also adsorbed, which leads to a decrease of the potential barrier and allows that more PAZO molecules are adsorbed. It has also been demonstrated that the effect of drying does not influences the adsorption mechanisms but only changes adsorption parameters, adsorption characteristic times and adsorbed amounts. The adsorption parameters are proved to be dependent on the number of available adsorption sites and on the presence of both water and counterions adsorbed on the last layer. Acknowledgments This work was supported by the “Plurianual” financial contribution of “Fundac¸a˜ o para a Ciˆencia e Tecnologia” (Portugal) and by the project POCTI/FAT/47529/2002 (Portugal). References [1] O.N. Oliveira Jr., J.-A. He, V. Zucolotto, S. Balasubramanian, L. Li, H.S. Nalwa, J. Kumar, S.K. Tripathy, Layer-by-layer polyelectrolyte-based thin films for electronic and photonic applications, in: S.K. Tripathy, J. Kumar, H.S. Nalwa (Eds.), Handbook of Polyelectrolytes and their Applications, vol. 1, American Scientific Publishers, 2002, p. 1. [2] G. Decher, J.D. Hong, J. Schmitt, Buildup of ultrathin multilayer films by a self-assembly process. 3. Consecutively alternating adsorption of anionic and cationic polyelectrolytes on charged surfaces, Thin Solid Films 210 (1992) 831–835. [3] G. Decher, Fuzzy nanoassemblies: toward layered polymeric multicomposites, Science 277 (1997) 1232–1237. [4] O.N. Oliveira Jr., M. Raposo, A. Dhanabalan, Langmuir–Blodgett (LB) and self-assembled (SA) polymeric films, in: H.S. Nalwa (Ed.), Handbook of Surfaces and Interfaces of Materials, vol. 4, Academic Press, New York, 2001, pp. 1–63.
[5] M. Raposo, R.S. Pontes, L.H.C. Mattoso, O.N. Oliveira Jr., Kinetics of adsorption of poly(o-methoxyaniline) self-assembled films, Macromolecules 30 (1997) 6095–6101. [6] R.S. Pontes, M. Raposo, C.S. Camilo, A. Dhanabalan, O.N. Oliveira Jr., Non-equilibrium adsorbed polymer layer via hidrogen bonding, Phys. Status Solidi A 173 (1999) 41–50. [7] M. Raposo, O.N. Oliveira Jr., Energies of adsorption of poly(o-methoxyaniline) layer-by-layer films, Langmuir 16 (2000) 2839–2844. [8] M. Raposo, O.N. Oliveira Jr., Adsorption of poly(o-methoxyaniline) in layer-by-layer films, Langmuir 18 (2002) 6866–6874. [9] Q. Ferreira, P.J. Gomes, M. Raposo, J.A. Giacometti, O.N. Oliveira Jr., P.A. Ribeiro, Influence of ionic interactions on the photoinduced birefringence of poly[1-[4-(3-carboxy-4 hydroxyphenylazo) benzene sulfonamido]-1,2ethanediyl, sodium salt] films, J. Nanosci. Nanotechnol., in press. [10] R.A. Robinson, R.H. Stokes, Electrolyte Solutions, second ed. (revised), Butterworths, London, 1968. [11] Q. Ferreira, P.J. Gomes, Y. Nunes, M. Maneira, P.A. Ribeiro, M. Raposo, Atomic force microscope characterization of PAH/PAZO multilayers, Microelectr. Eng., in press. [12] O. Mermut, C.J. Barret, Effects of charge density and counterions on the assembly of polyelectrolyte multilayers, J. Phys. Chem. B 107 (2003) 2525–2530. [13] H.H. Rmaile, J.B. Schlenoff, Internal pK’s in polyelectrolyte multilayers: coupling protons and salt, Langmuir 18 (2002) 8263–8265. [14] N.C. de Souza, J.R. Silva, R. Di Thommazo, M. Raposo, D.T. Balogh, J.A. Giacometti, O.N. Oliveira Jr., The influence of solution treatment on the adsorption and morphology of poly(o-methoxyaniline) layer-by-layer films, J. Phys. Chem. B 108 (2004) 13599–13606. [15] G. Decher, Y. Lvov, J. Schmitt, Proof of multilayer structural organization in self-assembled polycation polyanion molecular films, Thin Solid Films 244 (1994) 772–777. [16] Y. Lvov, K. Ariga, M. Onda, I. Ichinose, T. Kunitake, A careful examination of the adsorption step in the alternate layer-by-layer assembly of linear polyanion and polycation, Colloids Surf. A: Physicochem. Eng. Aspects 146 (1999) 337–346. [17] N.C. de Souza, J.R. Silva, M.A. Pereira-da-Silva, M. Raposo, R.M. Faria, J.A. Giacometti, O.N. Oliveira Jr., Dynamic scale theory for characterizing surface morphology of layer-by-layer films of poly(o-methoxyaniline), J. Nanosci. Nanotechnol. 4 (5) (2004) 548–552. [18] J.F. Quinn, F. Caruso, Thermoresponsive nanoassemblies: layer-by-layer assembly of hydrophilic–hydrophobic alternating copolymers, Macromolecules 38 (2005) 3414–3419. [19] J.M.C. Lourenc¸o, P.A. Ribeiro, A.M. Botelho-do-Rego, M. Raposo, Counterions in layer-by-layer films—influence of the drying process, submitted for publication. [20] M.R. Bo¨ohmer, O.A. Evers, J.M.H.M. Scheutjens, Weak polyelectrolytes between two surfaces-adsorption and stabilization, Macromolecules 23 (1990) 2288–2301. [21] P.A. Ribeiro, R. Steitz, I. Estrela-Lopis, H. Haas, N.C. Souza, O.N. Oliveira Jr., Maria Raposo, Thermal stability of poly(o-methoxyaniline) layer-bylayer films investigated by neutron reflectivity and UV–vis spectroscopy, J. Nanosci. Nanotechnol. 6 (2006) 1396–1404. [22] G. Pan, P.S. Liss, Metastable-equilibrium adsorption theory. I. Theoretical, J. Colloid Interface Sci. 201 (1998) 71–76. [23] P. Viot, Tackling out-of-equilibrium systems by computer simulation: models of irreversible and reversible adsorption, Eur. J. Phys. 26 (2005) S39–S46. [24] Z. Adamczyk, K. Jaszczolt, A. Michna, B. Siwek, L. Szyk-Warszynska, M. Zembala, Irreversible adsorption of particles on heterogeneous surfaces, Adv. Colloid Interface Sci. 118 (2005) 25–42. [25] C. Holm, J.F. Joanny, K. Kremer, R.R. Netz, P. Reineker, C. Seidel, T.A. Vilgis, R.G. Winkler, Polyelectrolyte theory, polyelectrolytes with defined molecular architecture II, Adv. Polym. Sci. 166 (2004) 67–111. [26] M. Losche, J. Schmitt, G. Decher, W.G. Bouwman, K. Kjaer, Detailed structure of molecularly thin polyelectrolyte multilayer films on solid substrates as revealed by neutron reflectometry, Macromolecules 31 (1998) 8893–8906.
Q. Ferreira et al. / Sensors and Actuators B 126 (2007) 311–317 [27] J.M.C. Lourenc¸o, P.A. Ribeiro, A.M. Botelho do Rego, F.M. Braz Fernandes, A.M.C. Moutinho, M. Raposo, Counterions in poly(allyamine hydrochloride) and poly(styrene sulfonate) layer-by-layer films, Langmuir 20 (2004) 8103–8109.
Biographies Quirina Ferreira received the degree in chemistry engineering from the New University of Lisbon in 2004. Currently is a PhD student in physics engineering from Faculty of Science and Technology, New University of Lisbon. Research activities has been focused on analysis of macromolecular functional surfaces by atomic force microscopy and production of molecular self-assembly systems for potential applications in nano-electronics and photonic devices. Paulo J. Gomes is currently a PhD student in the Faculty of Science and Technology, New University of Lisbon and received graduation degree in physics and chemistry in 2006 from the New University of Lisbon. Current research interest include ultra-thin films of polymers and biomolecules, functional interfaces and nano-technology, molecular architectures for electronics, photonics and magnetism, biomimetic membranes and radiation effect in biological systems. Manuel J.P. Maneira, graduated in physics (BS degree) in the University of Lisbon in 1971. As assistant researcher he studied experimentally, ion pair formation in atom molecule collisions (Na, K, Cs against CCl4 and SnCl4 ). He got his PhD in experimental molecular physics in the New University of Lisbon in
317
1984 and joined the New University of Lisbon as “Professor Auxiliar”. He is now “Professor Associado” of the Physics Department of the New University of Lisbon. His current research interests include applied plasma physics (topic of magnetron sputtering plasma experiments and modeling and coatings), thin film physics (alloy films, adhesion, reflectivity, nano-topography and nano-corrosion with AFM/Electrochemical Cell) and ion pair formation in collisions between alkaline atoms and macromolecules. He is now head of the CEFITEC (Center of Physics and Technological Research of the Faculty of Science and Technology of the new University of Lisbon). Paulo A. Ribeiro is assistant professor in physics at the New University of Lisbon, Portugal, received degree in physics and materials engineering in 1989 at the New University of Lisbon and PhD in science and materials engineering in 1999 from the Sao Paulo University, Brazil. Current research interest include physics of surfaces and interfaces, optics and non-linear optics, optical spectroscopy, ultra-thin films of polymers and biomaterials, optical spectroscopy, nano-technology and molecular architectures for electronics, photonics, magnetism and sensing devices. Maria Raposo is assistant professor in physics at the New University of Lisbon, Portugal, received degree in physics and materials engineering in 1989 at the New University of Lisbon and PhD in science and materials engineering in 1999 from the Sao Paulo University, Brazil. Current research interest include ultra-thin films of polymers and biomolecules, interfaces and nano-technology, colloids, molecular architectures for electronics, photonics and magnetism, biomimetic membranes and radiation effect in biological systems.
Mechanisms of adsorption of an azo-polyelectrolyte onto layer-by-layer films Quirina Ferreira, Paulo J. Gomes, Manuel J.P. Maneira, Paulo A. Ribeiro, Maria Raposo ∗ Centro de Investiga¸ca˜ o em F´ısica Tecnol´ogica (CEFITEC), Faculdade de Ciˆencias e Tecnologia, UNL, 2829-516 Caparica, Portugal Available online 13 January 2007
Abstract The mechanisms which lead to the adsorption of the azopolyelectrolyte poly[1-[4-(3-carboxy-4-hydroxyphenylazo) benzene sulfonamido]-1,2ethanediyl, sodium salt] (PAZO) layer adsorbed onto layer-by-layer films were investigated by monitoring the adsorbed amount per unit of area as a function of time. The adsorption kinetics curves were obtained as a function of adsorption variables as PAZO concentration and drying process. The PAZO adsorption kinetics data was shown to be described by two processes, one with smaller adsorption characteristic time, of the order of seconds, which was attributed to a nucleation mechanism and a second process, with adsorption characteristic times of the order of tenth of minutes, associated to diffusion. During the first process, PAZO molecules rapidly adsorb onto the substrate surface creating an electrostatic barrier that makes difficult the adsorption of more PAZO molecules. However, positive counterions are allowed to be adsorbed and as they are the potential barrier is lowered, allowing more PAZO molecules to be adsorbed. This last mechanism is accounting for the second adsorption process. The adsorption processes are not seen to be influenced by the layer drying and only contribute for the changes in the values of adsorption parameters, i.e., adsorption characteristic times and corresponding adsorbed amounts. These parameters were related with the available number of adsorption sites which are dependent of the presence of both water and counterions adsorbed on the last adsorbed layer. © 2007 Elsevier B.V. All rights reserved. Keywords: Layer-by-layer; Adsorption; Polyelectrolyte; Water network; Azo-polyelectrolyte; PAZO
1. Introduction The main interest to study the formation of PAZO layer is due to its photoisomerization capabilities about the N N bond which makes it suitable for multipurpose device applications as holographic image storage, optical switching, optical computation, reversible optical storage, optical relief refractive and index gratings, digital video recorders, photosensitive artificial membranes and eventually molecular-machines [1]. The adsorption kinetics of polyelectrolytes onto solid surfaces, obtained by measuring the adsorbed amount with time, has been investigated for several decades by adsorbing them onto solid substrates, usually silica spheres, that where initially immersed in an aqueous solution and the polyelectrolyte injected afterwards while the solution is being stirred. Consequently, the polyelectrolyte adsorption kinetics has a characteristic time which is dependent of the polyelectrolyte diffusion process in the solvent and
∗
Corresponding author. Tel.: +351 965285266; fax: +351 212948549. E-mail address: [email protected] (M. Raposo).
0925-4005/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.snb.2007.01.007
of the polyelectrolyte interactions with the substrate, assuming that the polyelectrolyte concentration, pH and ionic strength and other factors as temperature and pressure are held constant during adsorption. With the advent of the layer-by-layer (LbL) technique [2–4], which consists of sequential adsorption cationic and anionic polyelectrolytes from aqueous solutions onto solid substrates, with the adsorption taking place at a planar substrate during a determined period of time. Both, the experimental procedure and, consequently, the adsorption kinetics are different. Although the adsorption process is expected to be also driven by electrostatic forces, different adsorption characteristic times are expected, since the solid planar substrate is immersed rapidly in the polyelectrolyte solution, allowing that the polyelectrolyte molecules are instantaneously in contact with substrate. Moreover, since during the production of LbL films the polyelectrolyte layers are often dried after adsorption other mechanisms should be involved. In fact, experimental results of adsorption kinetics dependence on the layer drying process, obtained for polyanilines and poly(p-phenylene vinylene) (PPV) precursor [5,6], revealed that polyelectrolyte adsorption onto itself is allowed which is odd in what is expected to be
312
Q. Ferreira et al. / Sensors and Actuators B 126 (2007) 311–317
an electrostatic driven process. Different kinetics curves were also obtained for the case of polyanilines and poly(p-phenylene vinylene) precursor [5,6], where it was shown that the number of immersions and the drying of the film influences the film growth kinetics. This behaviour was associated with the polyelectrolyte hydrogen bonding capabilities. However, this could not be the reason for the adsorption of all polyelectrolytes. Consequently, the experimental procedure used to obtain the true kinetics curves must be well controlled when using the LbL technique namely for taking into account the adsorption processes dependence on the surface where adsorption is taking place. As suggested [7,8], the kinetics curves should be obtained by doing only one immersion per layer using different adsorption periods of times, being the kinetics curves the adsorbed amount per layer as a function of immersion time. This was the procedure adopted in this work, to investigate the mechanisms involved on the adsorption kinetics of a PAZO layer, a high degree of charge polyelectrolyte [9], onto already deposited PAH/PAZO LbL films during different periods of time, under different adsorption conditions as solution concentration and drying process. In this article, the mechanisms involved in the adsorption of one PAZO layer onto PAH/PAZO LbL films are discussed which explain the two processes, nucleation and diffusion, which are occurring during the PAZO adsorption. 2. Experimental The poly[1-[4-(3-carboxy-4-hydroxyphenylazo) benzenesulfonamido]-1,2-ethanediyl, sodium salt] (PAZO) kinetics adsorption studies were performed adsorbing PAZO onto (poly(allylamine hydrochloride)(PAH)/PAZO)/PAH LbL films. For that, aqueous solutions of PAH (average Mw = 50,000– 65,000 g/mol) with concentration of 10−2 M and solutions of PAZO with different concentrations in a buffer solution at pH 10 were prepared. The polyelectrolyte concentrations were based on the molecular weight repeat unit. Chemicals were obtained from Sigma–Aldrich and the polyelectrolytes molecular structures are shown in Fig. 1. The pure water with a resistivity of 18.2 Mcm was supplied by a Millipore system (Milli-Q, Millipore GmbH). For pH 10 buffer solution, a 0.05 M sodium bicarbonate (NaHCO3 ) aqueous solution and a 0.1 M sodium hydroxide (NaOH) solution were used in the proportion 500:107 (v/v) [10]. The films were directly deposited on BK7 optical glass substrates for the adsorption kinetics studies. The substrates were hydrophilized with a H2 SO4 /H2 O2 (1:1) bath for 1 h, after which
the glass and silicon slides were washed exhaustively with pure water [5]. Two types of films, dry and wet, were prepared. The dried films were obtained by immersing the substrate in the PAH solution during 3 min, washing it with water, drying it using a nitrogen flux, immersing it during a determined period of time in the PAZO solution, washing it with buffer solution and finally drying it with a nitrogen flux. This procedure gives rise to a bilayer film. Films with several bilayers were obtained by repeating this procedure. The wet samples were obtained immersing the substrate in the PAH solution during 3 min, washing with water, immersing the substrate in the PAZO solution during a determined period of time and washing with the buffer solution. This procedure was repeated during five times and finally, the samples were left to dry at room conditions. The adsorbed amount of PAZO per unit of area was quantified by measuring the UV–vis absorbance spectra of PAZO solutions at different concentrations which were obtained dissolving the weighted PAZO polymer powder in the buffer solution at pH 10 and plotting the maximum absorbance at 360 nm, which is associated to the → * azo-chromophore transition, versus the solution concentration. The obtained experimental points were fitted with a straight line and using the Beer–Lambert law, an extinction coefficient (ε360 nm ) of 4.30 ± 0.07 g−1 m2 at 360 nm for PAZO was estimated. The amount of PAZO per unit of area (Γ ) adsorbed onto a solid substrate is estimated using again the Beer–Lambert law: Γ =
Abs360 nm 2ε360 nm
(1)
The UV–vis absorbance spectra of liquid and solid samples were measured using a Shimadzu UV-2101PC spectrophotometer. 3. Results and discussion 3.1. PAZO adsorption kinetics at pH 10 of dried samples The adsorption kinetics of PAZO onto (PAH/PAZO)n /PAH LbL films was obtained preparing PAH/PAZO LbL films with different PAZO adsorption times and maintaining the adsorption time of 3 min for the PAH. After each adsorption period, the samples were dried with a nitrogen flux and the absorbance spectra of LbL films having different number of bilayers were measured. The graph of Fig. 2 shows the spectra of (PAH/PAZO) LbL films with different number of bilayers obtained for PAZO adsorption time of 30 s. Plotting the maximum absorbance as a function of the number of bilayers for the LbL films prepared
Fig. 1. Polyelectrolyte molecular structures: (a) poly(allylamine hydrochloride) (PAH) and (b) poly[1-[4-(3-carboxy-4-hydroxyphenylazo) benzenesulfonamido]1,2-ethanediyl, sodium salt] (PAZO).
Q. Ferreira et al. / Sensors and Actuators B 126 (2007) 311–317
Fig. 2. UV–vis absorbance spectra of PAH/PAZO LbL films with different number of bilayers obtained for PAZO adsorption time of 30 s.
with different adsorption times, one can observe that the films grow linearly with the number of bilayers as is usual in LbL films [9,11]. The slopes of the fitted straight lines are proportional to the PAZO adsorbed amount per bilayer and from the obtained extinction coefficient the PAZO adsorbed amount per unit of area in each bilayer can be estimated. The adsorption kinetics curves were obtained plotting the PAZO adsorbed amount per unit of area as a function of adsorption time as shown in Fig. 3 for the different PAZO concentrations, with pH 10 and having the samples dried after each adsorption period. Plots are showed in a logarithmic scale in order to emphasize that the kinetics curves are not following a first order exponential growth as reported in previous works [12,13]. Instead of that the current data can be fitted, as already suggested [5,14], with two Johnson–Mehl–Avrami equations: n t t Γ = Γn 1 − exp − + Γd 1 − exp − τn τd (2)
313
where Γ n is the maximum adsorbed amount per unit of area associated with the process which presents the smaller adsorption kinetics characteristic time τ n , Γ d the maximum adsorbed amount per unit of area associated with the process with higher adsorption characteristic time τ d , and n is a constant assuming values between 1 and 1.5. The process with smaller adsorption characteristic times has been associated to a nucleation process and the one with the higher adsorption characteristic time associated to a diffusion like process [5]. These processes are called like this because the formation of the layer starts with the random adsorption of polyelectrolyte molecules which are close to the surface-nucleation of the layer, and ends with a diffusion process associated to adsorption of molecules which migrate through the repulsive electric field and/or the double layer formed near the surface. These mechanisms will be better described ahead. The solid lines in the graph of Fig. 3 are the optimized fittings to Eq. (2). These results are somehow different from the ones obtained from adsorption of POMA onto glass and on POMA/PVS where non-first order kinetics was found for the diffusion process. As a remark, de Souza et al. [14] also showed that the kinetics adsorption of POMA onto POMA/PVS follows Eq. (2) with n = 1. The parameters of Eq. (2) obtained from the fittings are displayed in Table 1. 3.2. PAZO adsorption kinetics at pH 10 of wet samples Drying during the layer formation of LbL films has already been shown to influence the adsorbed amount, the film final structure and surface morphology [15–18]. For example, de Souza et al. [17] showed that the adsorbed amount is higher for films dried in room conditions than those dried under a nitrogen flux, while Lourenc¸o et al. [19] showed that in poly(allylamine hydrochloride)/poly(styrene sulfonate) (PAH/PSS) films the adsorbed amount was higher for samples dried in ambient air than for samples that were never dried. In order to know, if the drying process is affecting the PAZO adsorbed amount, samples of (PAH/PAZO)5 were prepared without drying during the layer production, using different PAZO adsorption times, pH 10 and different concentrations. The amount of PAZO adsorbed per bilayer, i.e., the adsorbed PAZO amount per unit of area obtained from the maximum absorbance divided by the deposited number of bilayers, plotted as function of the PAZO adsorption time is shown in the graph of Fig. 4, together with the kinetics curves of the dried samples. The samples prepared without drying after the adsorption of each layer exhibit lower PAZO adsorbed amount than those that were dried, which is Table 1 Adsorption kinetics parameters obtained from fitting the experimental data with two Johnson–Mehl–Avrami equations
Fig. 3. PAZO adsorption kinetics curves for different PAZO concentrations. After each adsorption period the samples were dried with a nitrogen gas flow. The solid lines correspond to the fits.
Sample
Concentration (M)
Γn (mg m−2 )
τn (s)
Γd (mg m−2 )
τd (s)
Wet Wet Dried Dried Dried
0.25 × 10−2 1 × 10−2 0.25 × 10−2 0.5 × 10−2 1 × 10−2
1.9 3.0 3.8 4.4 4.7
2.3 3.7 5.6 2.4 2.1
1.28 0.46 1.51 1.28 1.16
900 1000 8000 3000 800
314
Q. Ferreira et al. / Sensors and Actuators B 126 (2007) 311–317
Fig. 4. PAZO adsorption kinetics curves for different PAZO concentrations and for both wet and dried samples. The solid lines correspond to the fits.
in accordance with similar results obtained for PAH/PSS LbL films [19]. From Fig. 4 one can also conclude that the adsorption kinetics curves can be interpreted as two first order adsorption processes, so that data can be fitted to Eq. (2). The calculated parameters resulting from this fitting are displayed in Table 1. 3.3. Adsorption characteristics times From the fitting of the kinetics curves to Eq. (2), fitting parameters listed in Table 1, for both type of samples, one can see that all curves are described by two processes having characteristics times laying in two ranges, one range at lower values and at higher values, which are nucleation and diffusion processes consistent. Fig. 5 shows the calculated characteristic adsorption times versus PAZO concentration for the dried samples and reveals that the increase of the concentration is leading to the decrease of the adsorption characteristic time for both adsorption processes. However, for wet samples the characteristic adsorption time can be considered, within the experimental uncertainties, as independent of PAZO concentration.
Fig. 5. Nucleation and diffusion characteristic times versus PAZO concentration. The solid lines are eye guides.
Fig. 6. PAZO maximum adsorbed amount per unit of area for the nucleation process and PAZO adsorbed amount per unit of area versus PAZO concentration. Results obtained in wet samples.
3.4. Adsorbed amount during nucleation For both type of samples, the kinetics curves are at short periods of adsorption times described by a process with similar characteristic time values. However, the parameters obtained from fittings, see Table 1, revealed to be concentration dependent. In Fig. 6, the maximum adsorbed amount due to the nucleation process, Γ n , and the adsorbed amount per unit of area during the first 15 s of adsorption are plotted as a function of the PAZO monomer concentration. These results were obtained from the wet samples kinetics curves. It can be seen that the adsorbed amount due to the nucleation is concentration dependent and follows the adsorbed amount behaviour for short times. This behaviour is consistent with the rapid adsorption of the PAZO molecules when the substrate is immersed in the PAZO solution. Only at very small concentrations, i.e., 0.001 M, the adsorbed amount for short times can be considered proportional to the concentration. A proportional behaviour has also been observed in polyaniline adsorption studies for short adsorption periods and small concentrations [8]. In the present case, for PAZO concentrations higher than 0.003 M, a saturation of the adsorbed amount is observed indicating that an adsorbed amount of about 1.7 mg/m2 is sufficient to compensate the substrate electrical charges where PAZO is being adsorbed. It should be remarked here that PAZO at pH 10 presents a high degree of charging [9] which suggest that electrostatics interactions are mainly accounting for the adsorption. The PAZO adsorbed molecules create a repulsing potential barrier, as predicted by Bo¨ohmer et al. [20], avoiding that more PAZO molecules are being adsorbed in this regime. Similar results were found from the adsorption kinetics of the dried samples. Fig. 7 shows the adsorbed amounts during the nucleation regime for both wet and dried samples. From the graphs of this figure one can clearly see that the PAZO adsorbed amount due to the nucleation process is dependent of drying. Since the PAZO molecules in solution are in the same conditions for both types of samples, this behaviour could be accounted by the increase of surface
Q. Ferreira et al. / Sensors and Actuators B 126 (2007) 311–317
Fig. 7. PAZO maximum adsorbed amount per unit of area for the nucleation process for both wet and dried samples.
roughness as shown by Ribeiro et al. [21]. However, the topographies obtained from atomic force microscopy measurements, for both wet and dried samples with the same number of bilayers, revealed that the dried samples present smaller roughness values. Consequently, the increase of adsorbed amount in the dried samples is consistent with the removal of water molecules and consequent crystallization of adsorbed counterions within the last layer. However, the formed salt crystals will be dissolved when the sample is newly immersed in the next polyelectrolyte solution, since the salt concentration in the polyelectrolyte solution is below of the saturation value, leading to the increase of the number of adsorption sites, in such a way that more polyelectrolyte can be adsorbed. In the case of wet samples, the presence of water and ionic networks diminishes the number of adsorption sites and consequently the amount of adsorbed polyelectrolyte. This explanation is corroborated by a high content of counterions in layer-by-layers films when prepared without drying between adsorption of layer in comparison with films which are dried [19].
315
models have been proposed [22–25]. However, the models do not take into account the process of build-up the polyelectrolyte layer. The model proposed here for the adsorption of a PAZO layer takes that into account and is consistent with the experimental results where it was observed that the adsorption of PAZO molecules onto solid substrates is governed by two processes: nucleation and diffusion. The nucleation process can be explained by the following mechanism: as the positive substrate is immersed in the PAZO solution, the PAZO molecules which are near the substrate could be rapidly adsorbed on the substrate surface, since the free energy associated with the adsorption is smaller than the free energy of the polymer chain in the solution. The two regimes equation used to explain this process, Eq. (2), could in fact be obtained if one considers that adsorbed chains or nucleus are deposited in preferential sites and PAZO molecules in solution are repelled by the already adsorbed ones. The number of vacant sites per unit of area is seen to decrease exponentially leading to an exponential increase to saturation of the adsorbed amount per unit of area. The adsorbed amount being independent of PAZO solution concentration also corroborates this mechanism. The number of sites will be dependent of substrate and PAZO electrical charge and size of PAZO molecules. The presence of PAZO molecules adsorbed on the substrate creates a potential barrier which makes difficult the adsorption of more PAZO molecules. However, at this stage, the sodium counterions are being attracted by the PAZO adsorbed molecules which lead to a decrease of the potential barrier, allowing more PAZO molecules to be adsorbed and giving rise to the second adsorption regime. This behaviour is supported by previous results where the presence of sodium atoms was detected in PAH/PAZO LbL films obtained with a PAZO adsorption time of 3 min [11]. At this point, it is worth to mention that this explanation does not exclude the possibility of reconformation of the last adsorbed macromolecules. However, the effect of diffusion of PAZO molecules into the layer-by-layer is less plausible taking into account the studies about the internal structure of layer-bylayer films [26]. Since these processes are occurring in both types
3.5. Adsorbed amount at long adsorption times Looking at the total PAZO adsorbed amount during the two processes as a function of PAZO concentration, i.e., the adsorption isotherm shown in the plot of Fig. 8, it can seen that in the measured concentration range the total adsorbed amount is practically concentration independent for both wet and dried samples. In fact, the adsorbed amount increases rapidly with the concentration to a saturation value, indicating that the adsorption process is effective even for very small concentrations. This behaviour definitely indicates that the electrostatic interactions are governing the PAZO adsorption process. 3.6. PAZO adsorption model The processes that lead to the adsorption of polyelectrolytes onto solid substrates have been extensively studied and several
Fig. 8. PAZO adsorption isotherms for wet and dried samples. The solid lines are eye guides.
316
Q. Ferreira et al. / Sensors and Actuators B 126 (2007) 311–317
of samples so that they are independent of the drying process, how can the obtained differences in the kinetics parameters for both types of samples be explained? It follows that if the PAZO adsorbed layer is not dried, the adsorbed PAZO molecules will be covered with a network of water molecules and ions [27] and as the film is newly immersed in the next polyelectrolyte solution the presence of the network makes difficult the adsorption of more polyelectrolyte. This is consistent with the smaller adsorbed amounts and higher characteristic times obtained for the same polyelectrolyte concentration—counterions must be replaced by the polyelectrolyte for adsorption to occur. If the adsorbed layer is dried, the water molecules will be removed, counterions crystallizes [19,27] and when the layer is newly immersed in the polyelectrolyte solution, crystals dissolve leaving more sites available for adsorption. 4. Conclusions The adsorption of a PAZO layer onto an already adsorbed PAH/PAZO bilayer is explained by two processes. One at early stages of adsorption corresponds to the instantaneous adsorption of the first PAZO molecules in the surface and is giving rise to a nucleation like process. These first adsorbed PAZO molecules avoid that more PAZO are adsorbed as a result of an electrostatic barrier formation. This effect is somehow attenuated since as PAZO molecules are being adsorbed, positive counterions from solution are also adsorbed, which leads to a decrease of the potential barrier and allows that more PAZO molecules are adsorbed. It has also been demonstrated that the effect of drying does not influences the adsorption mechanisms but only changes adsorption parameters, adsorption characteristic times and adsorbed amounts. The adsorption parameters are proved to be dependent on the number of available adsorption sites and on the presence of both water and counterions adsorbed on the last layer. Acknowledgments This work was supported by the “Plurianual” financial contribution of “Fundac¸a˜ o para a Ciˆencia e Tecnologia” (Portugal) and by the project POCTI/FAT/47529/2002 (Portugal). References [1] O.N. Oliveira Jr., J.-A. He, V. Zucolotto, S. Balasubramanian, L. Li, H.S. Nalwa, J. Kumar, S.K. Tripathy, Layer-by-layer polyelectrolyte-based thin films for electronic and photonic applications, in: S.K. Tripathy, J. Kumar, H.S. Nalwa (Eds.), Handbook of Polyelectrolytes and their Applications, vol. 1, American Scientific Publishers, 2002, p. 1. [2] G. Decher, J.D. Hong, J. Schmitt, Buildup of ultrathin multilayer films by a self-assembly process. 3. Consecutively alternating adsorption of anionic and cationic polyelectrolytes on charged surfaces, Thin Solid Films 210 (1992) 831–835. [3] G. Decher, Fuzzy nanoassemblies: toward layered polymeric multicomposites, Science 277 (1997) 1232–1237. [4] O.N. Oliveira Jr., M. Raposo, A. Dhanabalan, Langmuir–Blodgett (LB) and self-assembled (SA) polymeric films, in: H.S. Nalwa (Ed.), Handbook of Surfaces and Interfaces of Materials, vol. 4, Academic Press, New York, 2001, pp. 1–63.
[5] M. Raposo, R.S. Pontes, L.H.C. Mattoso, O.N. Oliveira Jr., Kinetics of adsorption of poly(o-methoxyaniline) self-assembled films, Macromolecules 30 (1997) 6095–6101. [6] R.S. Pontes, M. Raposo, C.S. Camilo, A. Dhanabalan, O.N. Oliveira Jr., Non-equilibrium adsorbed polymer layer via hidrogen bonding, Phys. Status Solidi A 173 (1999) 41–50. [7] M. Raposo, O.N. Oliveira Jr., Energies of adsorption of poly(o-methoxyaniline) layer-by-layer films, Langmuir 16 (2000) 2839–2844. [8] M. Raposo, O.N. Oliveira Jr., Adsorption of poly(o-methoxyaniline) in layer-by-layer films, Langmuir 18 (2002) 6866–6874. [9] Q. Ferreira, P.J. Gomes, M. Raposo, J.A. Giacometti, O.N. Oliveira Jr., P.A. Ribeiro, Influence of ionic interactions on the photoinduced birefringence of poly[1-[4-(3-carboxy-4 hydroxyphenylazo) benzene sulfonamido]-1,2ethanediyl, sodium salt] films, J. Nanosci. Nanotechnol., in press. [10] R.A. Robinson, R.H. Stokes, Electrolyte Solutions, second ed. (revised), Butterworths, London, 1968. [11] Q. Ferreira, P.J. Gomes, Y. Nunes, M. Maneira, P.A. Ribeiro, M. Raposo, Atomic force microscope characterization of PAH/PAZO multilayers, Microelectr. Eng., in press. [12] O. Mermut, C.J. Barret, Effects of charge density and counterions on the assembly of polyelectrolyte multilayers, J. Phys. Chem. B 107 (2003) 2525–2530. [13] H.H. Rmaile, J.B. Schlenoff, Internal pK’s in polyelectrolyte multilayers: coupling protons and salt, Langmuir 18 (2002) 8263–8265. [14] N.C. de Souza, J.R. Silva, R. Di Thommazo, M. Raposo, D.T. Balogh, J.A. Giacometti, O.N. Oliveira Jr., The influence of solution treatment on the adsorption and morphology of poly(o-methoxyaniline) layer-by-layer films, J. Phys. Chem. B 108 (2004) 13599–13606. [15] G. Decher, Y. Lvov, J. Schmitt, Proof of multilayer structural organization in self-assembled polycation polyanion molecular films, Thin Solid Films 244 (1994) 772–777. [16] Y. Lvov, K. Ariga, M. Onda, I. Ichinose, T. Kunitake, A careful examination of the adsorption step in the alternate layer-by-layer assembly of linear polyanion and polycation, Colloids Surf. A: Physicochem. Eng. Aspects 146 (1999) 337–346. [17] N.C. de Souza, J.R. Silva, M.A. Pereira-da-Silva, M. Raposo, R.M. Faria, J.A. Giacometti, O.N. Oliveira Jr., Dynamic scale theory for characterizing surface morphology of layer-by-layer films of poly(o-methoxyaniline), J. Nanosci. Nanotechnol. 4 (5) (2004) 548–552. [18] J.F. Quinn, F. Caruso, Thermoresponsive nanoassemblies: layer-by-layer assembly of hydrophilic–hydrophobic alternating copolymers, Macromolecules 38 (2005) 3414–3419. [19] J.M.C. Lourenc¸o, P.A. Ribeiro, A.M. Botelho-do-Rego, M. Raposo, Counterions in layer-by-layer films—influence of the drying process, submitted for publication. [20] M.R. Bo¨ohmer, O.A. Evers, J.M.H.M. Scheutjens, Weak polyelectrolytes between two surfaces-adsorption and stabilization, Macromolecules 23 (1990) 2288–2301. [21] P.A. Ribeiro, R. Steitz, I. Estrela-Lopis, H. Haas, N.C. Souza, O.N. Oliveira Jr., Maria Raposo, Thermal stability of poly(o-methoxyaniline) layer-bylayer films investigated by neutron reflectivity and UV–vis spectroscopy, J. Nanosci. Nanotechnol. 6 (2006) 1396–1404. [22] G. Pan, P.S. Liss, Metastable-equilibrium adsorption theory. I. Theoretical, J. Colloid Interface Sci. 201 (1998) 71–76. [23] P. Viot, Tackling out-of-equilibrium systems by computer simulation: models of irreversible and reversible adsorption, Eur. J. Phys. 26 (2005) S39–S46. [24] Z. Adamczyk, K. Jaszczolt, A. Michna, B. Siwek, L. Szyk-Warszynska, M. Zembala, Irreversible adsorption of particles on heterogeneous surfaces, Adv. Colloid Interface Sci. 118 (2005) 25–42. [25] C. Holm, J.F. Joanny, K. Kremer, R.R. Netz, P. Reineker, C. Seidel, T.A. Vilgis, R.G. Winkler, Polyelectrolyte theory, polyelectrolytes with defined molecular architecture II, Adv. Polym. Sci. 166 (2004) 67–111. [26] M. Losche, J. Schmitt, G. Decher, W.G. Bouwman, K. Kjaer, Detailed structure of molecularly thin polyelectrolyte multilayer films on solid substrates as revealed by neutron reflectometry, Macromolecules 31 (1998) 8893–8906.
Q. Ferreira et al. / Sensors and Actuators B 126 (2007) 311–317 [27] J.M.C. Lourenc¸o, P.A. Ribeiro, A.M. Botelho do Rego, F.M. Braz Fernandes, A.M.C. Moutinho, M. Raposo, Counterions in poly(allyamine hydrochloride) and poly(styrene sulfonate) layer-by-layer films, Langmuir 20 (2004) 8103–8109.
Biographies Quirina Ferreira received the degree in chemistry engineering from the New University of Lisbon in 2004. Currently is a PhD student in physics engineering from Faculty of Science and Technology, New University of Lisbon. Research activities has been focused on analysis of macromolecular functional surfaces by atomic force microscopy and production of molecular self-assembly systems for potential applications in nano-electronics and photonic devices. Paulo J. Gomes is currently a PhD student in the Faculty of Science and Technology, New University of Lisbon and received graduation degree in physics and chemistry in 2006 from the New University of Lisbon. Current research interest include ultra-thin films of polymers and biomolecules, functional interfaces and nano-technology, molecular architectures for electronics, photonics and magnetism, biomimetic membranes and radiation effect in biological systems. Manuel J.P. Maneira, graduated in physics (BS degree) in the University of Lisbon in 1971. As assistant researcher he studied experimentally, ion pair formation in atom molecule collisions (Na, K, Cs against CCl4 and SnCl4 ). He got his PhD in experimental molecular physics in the New University of Lisbon in
317
1984 and joined the New University of Lisbon as “Professor Auxiliar”. He is now “Professor Associado” of the Physics Department of the New University of Lisbon. His current research interests include applied plasma physics (topic of magnetron sputtering plasma experiments and modeling and coatings), thin film physics (alloy films, adhesion, reflectivity, nano-topography and nano-corrosion with AFM/Electrochemical Cell) and ion pair formation in collisions between alkaline atoms and macromolecules. He is now head of the CEFITEC (Center of Physics and Technological Research of the Faculty of Science and Technology of the new University of Lisbon). Paulo A. Ribeiro is assistant professor in physics at the New University of Lisbon, Portugal, received degree in physics and materials engineering in 1989 at the New University of Lisbon and PhD in science and materials engineering in 1999 from the Sao Paulo University, Brazil. Current research interest include physics of surfaces and interfaces, optics and non-linear optics, optical spectroscopy, ultra-thin films of polymers and biomaterials, optical spectroscopy, nano-technology and molecular architectures for electronics, photonics, magnetism and sensing devices. Maria Raposo is assistant professor in physics at the New University of Lisbon, Portugal, received degree in physics and materials engineering in 1989 at the New University of Lisbon and PhD in science and materials engineering in 1999 from the Sao Paulo University, Brazil. Current research interest include ultra-thin films of polymers and biomolecules, interfaces and nano-technology, colloids, molecular architectures for electronics, photonics and magnetism, biomimetic membranes and radiation effect in biological systems.