Polymerization and Surface Behavior of Alkyl-Substituted Aniline

Langmuir Film Polymerization of 1,22-Bis(2-aminophenyl)docosane: A Two-Dimensional Cross-linked Polyalkylaniline. L. J. Kloeppner and R. S. Duran. Jou...
0 downloads 0 Views 218KB Size
Download PDF
6734

Langmuir 1998, 14, 6734-6742

Polymerization and Surface Behavior of Alkyl-Substituted Aniline Surfactants at the Air-Aqueous Interface: A Kinetic Study L. J. Kloeppner and R. S. Duran* The Butler Polymer Research Laboratories, Department of Chemistry, University of Florida, Gainesville, Florida 32611-7200 Received March 24, 1998. In Final Form: August 14, 1998 The surface behavior of 2-, 3-, and 4-pentadecylaniline (3a, 3b, and 3c, respectively) was studied at the air-aqueous interface of an acidic subphase. Surface dipole moments (µn) were calculated from the surface potentials (∆V) of 3a-c and were related to the orientation of the anilinium ion at the interface. Further, 3a and 3b were polymerized in a Langmuir film, and analysis of the absorption spectra showed that the resulting polymers were in different oxidative states upon completion of the polymerization. A method that monitors the change in the mean molecular area (A) of the surfactant at a constant applied surface pressure and predicts the concentrations of monomer and polymer and/or oligomer at the interface in order to determine the formation rate (Rp) of poly(alkylaniline) is presented. The Rp values of 3a and 3b are expressed in terms of the concentrations of the monomer and polymer at the interface. It was determined that the Rp of 3a was dependent on the applied surface pressure while 3b showed little dependence on surface pressure. Differences in Rp can be explained by the conformation of the anilinium group and the steric hindrance of the alkyl group at the interface. A characteristic “autoacceleration effect” was observed for both 3a and 3b but was more significant in the case of 3a.

Introduction The Langmuir film technique is a unique method by which the polymerization of amphiphilic monomers can be studied at the air-aqueous interface. The advantage of polymerization in a Langmuir film is that the molecules can be held in a well-defined orientation at an interface1 while the polymerization kinetics can be observed in real time.2 In past studies, the kinetics of Langmuir film polymerization has been shown to be sensitive to applied surface pressure and temperature,1,3-7 but the effect that the reactive group conformation has on the polymerization rate is still not well understood. In solution chemistry, aniline is normally polymerized by the oxidation of the monomer with classical oxidants such as ammonium peroxydisulfate and potassium dichromate or by electrochemical methods under acidic conditions (0-2 pH).8 The mechanism of aniline polymerization is still under debate,9 but many believe the first step is the formation of the anilinium radical cation, followed by the coupling of two cations to form aminodiphenylamine, benzidine, or N,N-diphenylhydrazine.10-16 Of the three (1) Gaines, G. L. Insoluble Monolayers at Liquid-Gas Interfaces; Interscience: New York, 1966. (2) Bodalia, R. R.; Duran, R. S. J. Am. Chem. Soc. 1993, 115, 11467. (3) Gee, G. Proc. R. Soc. London 1935, 32, 39. (4) Letts, S. A.; Fort, T.; Lando, J. B. J. Colloid Interface Sci. 1976, 56, 64. (5) Fukuda, K.; Shibasaki, Y.; Nadahara, H. J. Macromol. Sci. Chem. 1981, A15, 999. (6) Shibata, A.; Oasa, A.; Hashimura, Y.; Yamashita, S.; Ueno, S.; Yamashita, T. Langmuir 1990, 6, 217. (7) MacRitchie, F. Chemistry at Interfaces; Academic: New York, 1990; Chapter 9. (8) Genies, E. M.; Boyle, A.; Lapkowski, M.; Tsintavis, C. Synth. Met. 1990, 36, 139. (9) Lux, F. Polymer 1994, 35, 2915. (10) Mohilner, D. M.; Adams, R. N.; Argersinger, W. J. J. Am. Chem. Soc. 1962, 84, 3618. (11) Breitenbach, M.; Heckner, K.-H. J. Electroanal. Chem. 1971, 29, 309. (12) Breitenbach, M.; Heckner, K.-H. J. Electroanal. Chem. 1971, 33, 45.

dimers of aniline that form, aminodiphenylamine is the primary product under oxidative conditions when the pH is 0-2 and is the intermediate that subsequently polymerizes. Once polymerization is initiated, it has often been observed that the polymer formation rate (Rp) undergoes an autoacceleration or autocatalytic effect.2,17-19 Wei and co-workers have shown that the increase in Rp in an electrochemical polymerization of aniline at a Pt electrode surface is directly proportional to the amount of polymer present.17 This effect was also observed when aniline was polymerized by a chemical oxidant in an aqueous medium.19 Shim and Park suggested that autoacceleration occurs when newly formed polymer or oligomers assist the monomeric aniline oxidation, therefore accelerating the propagation.18 They concluded that the polymer, which is more easily oxidized than the monomer, can then act as an oxidizing agent and oxidize additional monomer. Early work done in our group noted that the Langmuir film polymerization rate of 2-pentadecylaniline (3a) was dependent on the applied surface pressure.20,21 In this study, the polymerization of 3a at the air-aqueous interface was done in the presence of the strong chemical oxidant, ammonium peroxydisulfate, and an acid.20 The polymerization was monitored by observing the barrier movement needed to maintain a specific applied surface pressure. The area change per molecule was attributed to the replacement of van der Waals radii by covalent (13) Breitenbach, M.; Heckner, K.-H. J. Electroanal. Chem. 1973, 43, 267. (14) Hand, R. L.; Nelson, R. F. J. Am. Chem. Soc. 1974, 96, 850. (15) Hand, R. L.; Nelson, R. F. J. Electrochem. Soc. 1978, 125, 1059. (16) Genies, E. M.; Syed, A. A.; Tsintavis, C. J. Electroanal. Chem. 1985, 195, 109. (17) Wei, Y.; Sun, Y.; Tang, X. J. Phys. Chem. 1989, 93, 4878. (18) Shim, Y. B.; Park, S. M. Synth. Met. 1989, 29, E169. (19) Tzou, K.; Gregory, R. V. Synth. Met. 1992, 47, 267. (20) Duran, R. S.; Zhou, H. C. Polymer 1992, 33, 4019. (21) Zhou, H.; Duran, R. S. In Macromolecular Assemblies in Polymeric Systems; Stroeve, P., Balazs, A., Eds.; ACS Symposium Series 493; American Chemical Society: Washington, DC, 1992; Chapter 4.

10.1021/la980330l CCC: $15.00 © 1998 American Chemical Society Published on Web 10/07/1998

Behavior of Alkyl-Substituted Aniline Surfactants Scheme 1

bonds between monomer molecules and conformational differences between monomer and polymer.20,21 At higher applied surface pressures, the polymerization rate was faster than that at lower pressures, suggesting that the reactive group conformation at the interface might have a significant effect on the polymerization rate at an interface. The Rp of 3a at the air-aqueous interface was determined and a single rate constant assigned.2,20 However, this constant was assigned from the point of a maximum reaction rate, normally 10-30 min into the polymerization. While showing the effect of surface pressure on reaction rate, these studies did not address the autoacceleration rate effect on the complex polymerization kinetics. Furthermore, the polymerization was assumed first order in terms of the monomer concentration during the course of the reaction, potentially oversimplifying the kinetics compared to what is observed in the polymerization of aniline by chemical and electrochemical oxidation.17-19 Bodalia et al. furthered this work by addressing autoacceleration and assigning a second rate constant to the overall rate of the polymerization.23 However, this method was accurate only when the initial polymerization rate was orders of magnitude slower than the maximum rate, which is not always the case for aniline surfactant polymerizations.24 In this paper, a method of determining the Rp of o- and m-alkyl-substituted polyaniline in a Langmuir film is proposed. This method monitors the change in the mean molecular area of the initial surfactant at a constant applied surface pressure and predicts the concentrations of the monomer, [ANI], and polymer, [PANI], at the interface. In addition, Langmuir films of 3-pentadecylaniline (3b) were formed and polymerized, and the kinetics are reported for the first time. A quantitative study of the surface pressure and reactive group orientation on reaction rates for the two isomers is presented. Surface pressure and potential measurements are used to determine the relative effect that different surface areas have on the reactive group conformation of the monomers. The Rp values of the two monomers are correlated to these conformational and steric differences. Experimental Section Synthesis. The three isomers of pentadecylaniline were synthesized by the route indicated in Scheme 1. Modifications of reported literature procedures were used to prepare the (22) Zhou, H.; Stern, R.; Batich, C.; Duran, R. S. Makromol. Chem., Rapid Commun. 1990, 11, 409. (23) Bodalia, R.; Manzanares, J.; Reiss, H.; Duran, R. Macromolecules 1994, 27, 2002. (24) Kloeppner, L. J.; Nguyen, C. K.; Duran, R. S. Polym. Prepr. (Am. Chem. Soc., Div. Polym. Chem.) 1995, 36, 580.

Langmuir, Vol. 14, No. 23, 1998 6735 monomers.24,25 All reagents were commercially purchased (Aldrich) and used without further purification. Pentadecyltriphenylphosphonium Bromide (1). Tetradecyl bromide (50.0 g, 0.180 mol) and triphenylphosphine (50.0 g, 0.190 mol) were dissolved in 200 mL of acetonitrile and refluxed overnight. The acetonitrile was removed under reduced pressure, and the product was allowed to solidify. The solid was recrystallized in diethyl ether and a white crystalline material collected. Yield: 80%. Mp: 89.5-90.0 °C. 1H NMR (DMSO): δ 0.85 (t, 3H); 1.22 (bs, 22H); 1.45 (bs, 2H); 3.61 (t, 2H); 7.75-7.95 (m, 15H). 13C NMR (DMSO): δ 13.81, 19.85, 20.51, 21.67, 21.98, 28.06, 28.60, 28.78, 28.92, 29.57, 29.79, 31.18, 117.95, 119.08, 130.02, 130.19, 133.43, 133.56, 134.72. Anal. Calcd for C32H44BrP: C, 71.23; H, 8.22; Br, 14.81. Found: C, 71.25; H, 8.20; Br, 14.78. General Procedure for the Synthesis of Pentadecylaniline. Compound 1 (1 equiv) was dissolved in potassium-dried THF under an argon atmosphere and cooled in an ice bath to 0 °C. One equivalent of 2.5 M n-butyllithium was added dropwise over 15 min with stirring. The dark red mixture was stirred for an additional 1 h at 0 °C, and then 1 equiv of the appropriate isomer of nitrobenzaldehyde in potassium-dried THF was added dropwise. The solution was warmed to room temperature, mixed 12 h, poured into water, extracted with three portions of methylene chloride, and dried over magnesium sulfate. The solvent was removed under reduced pressure to yield a yellow oil, 2. Compound 2 (which is a mixture of cis and trans isomers) was purified by column chromatography (silica/methylene chloride). Purity was verified by gas chromatography (GC). Compound 2 (6.1 g) was placed in 20 mL of 95% ethanol with 0.8 g of 10% Pd on carbon. The solution was mixed with a pure hydrogen gas atmosphere until no more hydrogen gas was consumed. The mixture was filtered and the solvent was removed under reduced pressure, yielding a solid, 3, which was purified by column chromatography (silica/methylene chloride). 2-Pentadecylaniline (3a). A light yellow solid melted at 34.0-34.5 °C. Yield: 75%. 1H NMR (CDCl3): δ 0.89 (t, 3H); 1.27 (bs, 24H); 1.60 (m, 2H); 2.48 (t, 2H); 3.60 (bs, 2H); 6.65-6.77 (m, 2H); 7.00-7.05 (m, 2H). 13C NMR (CDCl3): δ 14.11, 22.68, 28.74, 29.35, 29.66, 31.29, 31.91, 115.46, 118.72, 126.77, 126.96, 129.39, 143.98. FTIR (KBr, cm-1): 3433, 3308 (vNH); 1276 (vNAr). HRMS. Calcd for C21H37N (M+): 303.2926. Found: 303.2937. Anal. Calcd for C21H37N: C, 83.10; H, 12.29; N, 4.61. Found: C, 83.06; H, 12.30; N, 4.57. 3-Pentadecylaniline (3b). A white solid melted at 44.044.5 °C. Yield: 43%. 1H NMR (CDCl3): δ 0.90 (t, 3H); 1.28 (bs, 24H); 1.60 (m, 2H); 2.53 (t, 2H); 3.60 (bs, 2H); 6.50-6.64 (m, 3H); 7.08 (t, 1H). 13C NMR (CDCl3): δ 13.92, 22.49, 29.48, 31.18, 31.72, 35.78, 112.31, 144.99, 118.58, 128.87, 144.05, 146.03. FTIR (KBr, cm-1): 3409, 3336 (vNH); 1278 (vNAr). HRMS. Calcd for C21H37N (M+): 303.2926. Found: 303.2931. Anal. Calcd for C21H37N: C, 83.10; H, 12.29; N, 4.61. Found: C, 83.14; H, 12.30; N, 4.62. 4-Pentadecylaniline (3c). A light yellow solid melted at 51.0-52.0 °C. Yield: 29%. 1H NMR (CDCl3): δ 0.90 (t, 3H); 2.50 (bs, 24H); 1.55 (m, 2H); 2.50 (t, 2H); 3.50 (bs, 2H); 6.65 (d, 2H); 6.95 (d, 2H). 13C NMR (CDCl3): δ 13.92, 22.50, 29.11, 29.37, 31.64, 34.89, 114.94, 128.93, 132.92, 143.76. FTIR (KBr, cm-1): 3394, 3315 (vNH); 1272 (vNAr). HRMS. Calcd for C21H37N (M+): 303.2926. Found: 303.2934. Anal. Calcd for C21H37N: C, 83.10; H, 12.29; N, 4.61. Found: C, 83.14; H, 12.28; N, 4.62. Monolayer Characterization. All of the monolayer and polymerization studies were done on KSV LB5000 LangmuirBlodgett equipment. The surface pressure (π) measurements were obtained using the Wilhelmy plate technique. Paper plates hung from a Pt wire were used for the Wilhelmy balance measurements. Each plate was cut from P5 Fisherbrand filter paper. The surface potential (∆V) measurements were conducted using the vibrating plate method. This method uses a vibrating plate ca. 3 mm above the surface of the interface and a platinum plate, 3 cm2, just below the surface of the air-aqueous interface. Using a Teflon trough and barriers, the π and ∆V vs mean (25) Bodalia, R.; Stern, R.; Batich, C.; Duran, R. J. Polym. Sci., Polym. Chem. Ed. 1993, 31, 2123.

6736 Langmuir, Vol. 14, No. 23, 1998

Figure 1. Surface pressure (π) vs mean molecular area (A) isotherms of 3a-c on a 0.10 M H2SO4 subphase at 25 °C. molecular area (A) isotherms were measured while compressing a film at a constant speed of ca. 4-6 A2 molecule-1 min-1. The solutions of monomers in chloroform were prepared (0.51.0 mg mL-1) and deposited on the air-aqueous interface by carefully adding the solution to the surface with a microsyringe and then allowing the chloroform to evaporate. The water used for the subphase was purified by a Millipore system (18 MΩ resistance). The temperature of the subphase was maintained with water jacket just below the Teflon trough connected to a circulating bath. The temperature (°C) of the subphase was determined by placing a thermometer just below the surface of the interface. The polymerization of the substituted anilines was carried out at constant applied surface pressure and subphase temperature. The aniline monomers were oxidatively polymerized on a subphase solution of 0.10 M sulfuric acid and 0.03 M ammonium peroxydisulfate unless otherwise noted. Sulfuric acid was generally chosen instead of simpler acids such as HCl to be consistent with much of the previous polyaniline synthesis literature. The film was compressed at ca. 45 A2 molecule-1 min-1 until the surface pressure was within 5 mN/m of the desired surface pressure and was then compressed at ca. 15 A2 molecule-1 min-1. Once the target surface pressure was reached, there was a brief induction period, 15-60 s, when the change in A was slower. The end of this period was designated as the beginning of the reaction, with the mean molecular area A0. This short induction period is believed to be due to film stabilization after its rapid compression. The polymerization was monitored by the change in the mean molecular area of the surfactant at a constant applied surface pressure. The polymerization was considered to be complete when the change in mean molecular area, within experimental error, was zero. The polymerized Langmuir film was removed from the airaqueous interface by passing the solution at the interface through a small glass frit, which collects the polymeric material. The polymer was washed with deionized water and dried under vacuum for a minimum of 2 h. Size-exclusion chromatography (SEC) molecular weight analysis was performed at room temperature using a Waters Associates liquid chromatography apparatus equipped with Phenomenex columns and a Kratos Analytical spectroflow 757 absorbance detector set at 254 nm; the mobile phase was THF, and the instrument was calibrated with polystyrene standards. UV-vis data were collected on a Cary 5E UV-vis-near-IR spectrophotometer. The solvent used for the UV-vis measurements was Fisher spectrophotometricgrade chloroform.

Results and Discussion Monolayer Characterization. Figure 1 shows the surface pressure (π) vs mean molecular area (A) isotherms of 3a-c on a 0.10 M H2SO4 subphase.10,24 As shown, 3c had an onset area of 27 Å2 molecule-1 and a collapse pressure of ∼50 mN m-1 at 22 Å2 molecule-1. The steep

Kloeppner and Duran

pressure increase and low onset area suggest that this symmetric amphiphilic monomer forms a condensed monolayer film with little rearrangement between the onset and collapse pressures. The surface area is considerably smaller than expected for an ionized monolayer, where electrostatic repulsion between the positively charged headgroups should expand the film.1 The lower onset and collapse areas may be explained by the divalent sulfate counterion in the subphase. Each sulfate ion at the interface would be capable of forming an ion pair with two positively charged anilinium ions in the monolayer, allowing the surfactants to pack more tightly at the interface. When the π-A isotherms of 3a and 3b were compared with that of 3c, more expanded films were observed. The larger onset and collapse areas have been previously attributed to the alkyl side chain of these unsymmetrical amphiphiles requiring larger areas at the interface as well as electrostatic repulsion between the positively charged monomers.2 To further characterize these monolayer films, surface potential measurements were performed on the three isomers of 3. A monolayer film at the air-aqueous interface can be treated as an assembly of molecular dipoles which can shift the potential across an interface.1 This potential shift, relative to a clean aqueous surface, is referred to as the surface potential of the monolayer, ∆V. Using the Helmholtz equation, ∆V can be related to an apparent or surface dipole moment, µn: 26

∆V ) µnN/0

(1)

where 0 is the permittivity of a vacuum and N is the concentration of dipoles at the interface. µn is the dipole moment normal to the interface, and hence its value is affected by the orientation of the surfactant. Davies and Rideal proposed that µn was made up of three components: µ1, reorientation of the water molecules at the interface; µ2, dipole moment from the headgroup; and µ3, dipole moment from the hydrophobic group.27,28 µn is defined by the equation

µn ) (µ1/1 + µ2/2 + µ3/3)

(2)

where 1, 2, and 3 are the local dielectric constants for each of the three layers.29 When an ionized monolayer is spread at the interface, an additional term needs to be added to the Helmholtz equation which accounts for the double-layer potential, ψ0. ψ0 can be estimated by the Gouy-Chapman equation30

ψ0 ) (2kT/e) sinh-1[Re/(5 × 10-7c0T)1/2A]

(3)

where k is the Boltzmann constant, T the absolute temperature, e the electron charge, R the degree of ionization of the monolayer, and c the concentration of monovalent ions in the subphase. The Helmholtz equation can now be written for the surface potential of an ionic monolayer:

∆V ) µnN/0 + ψ0

(4)

(26) Oliveira, O. N.; Taylor, D. M.; Lewis, T. J.; Salvagno, S.; Stirling, C. J. M. J. Chem. Soc., Faraday Trans. 1 1989, 85, 1009. (27) Davies, J. T.; Rideal, E. K. Can. J. Chem. 1955, 33, 947. (28) Davies, J. T.; Rideal, E. K. Interfacial Phenomena; Academic: New York, 1963; Chapter 2. (29) Demchak, R. J.; Tomlinson, F. J. Colloid Interface Sci. 1974, 46, 191. (30) Oliveira, O. N.; Taylor, D. M.; Morgan, H. Thin Solid Films 1992, 210/211, 76.

Behavior of Alkyl-Substituted Aniline Surfactants

Langmuir, Vol. 14, No. 23, 1998 6737

Figure 3. Surface dipole moments (µn) vs corresponding surface pressures (π) for 3a-c.

Figure 2. Surface pressure (π), surface potential (∆V), and surface dipole moment (µn) vs mean molecular area (A) isotherms of 3a (top), 3b (middle), and 3c (bottom) on a 0.20 M HCl subphase at 25 °C.

To simplify the calculation of the double-layer potential (ψ0), a 0.20 M HCl subphase was used instead of a 0.10 M H2SO4 subphase. The resulting monovalent anion, Cl-, allows straightforward use of eq 3 to evaluate ψ0. The degree of ionization, R, is assumed to be 1 in all three cases. This is a good assumption since the pKa values of 3a31 and 4-octadecylaniline32 at the air-aqueous interface have previously been estimated at 4.6 and 3.6, respectively. Figure 2 shows ∆V, µn, and π vs A for 3a, 3b, and 3c, respectively. The π-A isotherms of 3a and 3b on 0.20 M (31) Herod, T. Ph.D. Thesis, University of Florida, Gainesville, FL, Dec 1997. (32) Zhao, X.; Subrahmanyan, S.; Eisenthal, K. B. Chem. Phys. Lett. 1990, 171, 558.

HCl were similar to those observed on 0.10 M H2SO4, while that of 3c was much more expanded. The π-A isotherm of 3c was similar to one previously reported for 4-hexadecylaniline on a 1.0 M HCl subphase, in which the onset pressure was ca. 60 A2 molecule-1.32 Ion-pairing effects may contribute to the more condensed isotherm observed on 0.10 M H2SO4. Upon compression, the value of µn increased dramatically prior to the onset pressures for all three isomers and reached a maximum at ca. the same A as the onset pressure. As the film was compressed further, µn slowly decreased. When the values of µn for all three isomers at the corresponding pressure are compared, the largest decrease in µn was observed for 3a as the surface pressure was increased (Figure 3). The contribution of µ3 has been estimated to be positive when organized in a Langmuir film and relatively constant over the region measured.26 The anilinium headgroup (Car-NH3+) contribution, µ2, is predicted to be positive and largest when the C-N bond is ca. perpendicular to the interface. This prediction is based on surface potential measurements of an aliphatic ammonium ion, Cal-NH3+, which also gives a large positive contribution to µn.33 As the dipole contribution from the anilinium ion is rotated away from a perpendicular orientation, the value of µ2 should approach zero as the dipole arranges parallel to the interface. The value of µ1 is also affected by the headgroup, but because of the similar chemical nature of 3a-c, the differences in µ1 should be minimal. The slopes of the lines in Figure 3, which are a measure of (δµ/δπ)T, are -4.41, -2.68, and -1.05 mD m mN-1 for 3a, 3b, and 3c, respectively. The similarity of the zero pressure intercept for the three isomers (ca. 410 mD) suggests that the orientation of the headgroup at the interface is similar at very low surface pressures but possesses different conformations at higher pressures. The different surface behavior of the three isomers can be attributed to structural differences that affect their arrangement at the aqueous surface. These differences suggest that the conformation of the anilinium group at the air-aqueous interface is affected by the amount of surface area that confines the surfactants. In the case of 3a, the alkyl chain has more space to occupy at larger areas, but as the area decreases, the alkyl group will be pushed up and away from the interface, thus reorienting the anilinium group. The surface potential and pressure measurements of 3b also suggest that the anilinium group (33) Vogel, V.; Mo¨bius, D. J. Colloid Interface Sci. 1988, 126, 408.

6738 Langmuir, Vol. 14, No. 23, 1998

Kloeppner and Duran

Figure 5. Absorption spectra of poly(3a) just after completion of the polymerization reaction (A) and after the polymer is allowed to remain at the interface for an additional 4 h (B). Absorption spectrum of poly(3b) just after completion of the polymerization reaction (C). Peaks 1, 2, 3, and 4 are 290, 405, 550, and 740 nm, respectively.

Figure 4. Predicted surface behavior of the anilinium headgroups of 3a-c at changing surface pressures. The arrows show the predicted conformational change of the headgroup when going from low to higher surface pressures.

is reorienting at the surface of the air-aqueous interface as the amount of surface area that confines the surfactant changes, but not to the extent 3a does. Surface pressure and potential measurements of 3c suggest the anilinium group undergoes little conformational change as the film is compressed to smaller surface areas. The observed surface behavior of these isomers may be summarized by the drawing in Figure 4. Polymerization of Pentadecylaniline. In theory, the polymer products of 3a and 3b form the same product since alkyl substituents on the monomers are equivalent on the polyaniline backbone, making comparison of 3a and 3b particularly useful. It should be noted that 3c did not polymerize under these conditions, so further studies on this isomer were not undertaken. Analysis of the polymerization product of 3a at the air-aqueous interface has been previously reported.10 Because of the small amount of polymer produced during one polymerization (about 50 µg), size-exclusion chromatography (SEC) and UV-vis spectroscopy are the most useful techniques for the analysis of the final polymer. SEC analysis of the polymerization product of 3b, poly(3b), produced two peaks of approximately equivalent intensity: one peak corresponded to high molecular weight polymer, and a second peak indicated lower weight oligomers. The first peak had a weight-average molecular

weight (Mw) of ca. 740 000 and a number-average molecular weight (Mn) of ca. 250 000. The polydispersity of poly(3b) was calculated to be 2.9. The previous analysis of poly(2-pentadecylaniline), poly(3a), found that Mw was 300 000 and that the polydispersity was 2.0. Because very small amounts of impurities can have a large effect on the degree of reaction at an interface, little can be concluded from comparison of the differences in the SEC data of the two isomers other than that polymerization occurred and that a high molecular weight polymer was the result. Figure 5 shows the absorption spectra of 3a and 3b after polymerization. Poly(3a) was collected just after the reaction was completed, 50 min (Figure 5, spectrum A). The resulting dark blue-green material had absorption maxima at 740 and 290 nm, which are similar to those previously reported.10 Further, the spectrum also showed very weak absorption maxima at 550 and 405 nm. Poly(3b), which was collected after the reaction was completed (4 h), was a violet-blue material which had absorption maxima at 550 and 405 nm (Figure 5, spectrum C). The spectrum is similar to that previously reported by Lee et al. for PANI in the completely oxidized pernigraniline base form. Lee reported peaks at 296, 433, and 542 nm. In the spectrum shown in Figure 5, the peak at 296 nm is believed to be hidden in the shoulder of the adjacent peak. From these spectra, it is believed that poly(3b) is in the pernigraniline base form at the end of the reaction and the strong absorption at 740 nm indicates that poly(3a) is probably in a lower oxidative state such as the emeraldine or nigraniline state.34 3a normally reacts completely in less than 1 h, while the polymerization of 3b is much slower. When the polymerization product of 3a was allowed to remain under reactive conditions for 4 h, the absorption spectrum was similar to the spectrum of poly(3b), suggesting that they are both in the perngraniline state after 4 h (Figure 5, spectrum B). No significant area change was observed when oxidizing poly(3a) from the lower oxidative state to the perngraniline state. Kinetics of Polymerization. The initial concentration of aniline monomer in the monolayer film at the (34) Gee, G. Trans. Faraday Soc. 1936, 32, 187.

Behavior of Alkyl-Substituted Aniline Surfactants

Langmuir, Vol. 14, No. 23, 1998 6739

interface can be expressed by the following equation:

[ANI0] ) 1/A0

(5)

where A0 is the mean molecular area (Å2 molecule-1) of the surfactant at the beginning of the polymerization. In a similar manner, the concentration of the polymer expressed in repeat units at the end of the polymerization can be calculated (repeat units/Å2):

[PANI∞] ) 1/A∞

(6)

where A∞ is the mean molecular area at the end of the reaction. As is commonly assumed in monolayer film kinetics, the change in the mean molecular area of the monomer (A) is proportional to the fraction of surfactant that has reacted:3,34,35

ξ ) (A0 - A)/(A0 - A∞)

(7)

Figure 6. ξ and dξ/dt vs time for the polymerization of 3a at an applied surface pressure of 10 mN m-1 and a temperature of 25 °C.

where ξ is defined as the extent of reaction. When eqs 5-7 are combined, it is easily shown that the concentrations of the monomer and polymer at the interface can be expressed in terms of ξ:

[ANI] ) (1/A)(1 - ξ) ) (1/A)(A - A∞)/(A0 - A∞)

(8)

[PANI] ) (1/A)(ξ) ) (1/A)(A0 - A)/(A0 - A∞) (9) Equations 8 and 9 now allow the calculation of the monomer and polymer concentrations at any point in the course of the reaction if A, A0, and A∞ are known. Two assumptions were made in order for eqs 8 and 9 to be valid. The first is that all of the monomer was consumed and formed polymer at the interface with no solubilization into the subphase. The second assumption made was that the mean molecular areas of the polymer and monomer surfactants are additive. When the rate of a reaction in solution chemistry is calculated, the volume of the reaction vessel is considered constant, greatly simplifying the data interpretation. When there is a significant change in the volume during a reaction, the rate expression needs to be written to account for this change. Therefore, when the total area of a monolayer changes during the course of a reaction, this change must be accounted for when calculating the rate of the reaction in terms of the concentration of either the reactants or products. Most previous studies that determined rates of reactions from the change in A in a Langmuir film at constant surface pressure and temperature have not accounted for this change on the concentration of the surfactant at the interface but have derived their rate expression from the change in area over time, dA/dt. Changes in the total area of a Langmuir film during the polymerization of 3a are considerable, ca. 40%. Therefore, the rate of polymerization, Rp, at the interface can be defined as

Rp ) (1/β) (dnp/dt)

(10)

where dnp/dt is the rate of polymer formation in terms of repeat units vs time and β is defined as the total area of the interface at any point in the reaction or, more simply stated, 1/β is a correction factor for the change in the area of the film during the course of the polymerization. β can be calculated from the total initial area of the monolayer (β0) times the fraction of change (A/A0) during the course (35) Gee, G. Proc. R. Soc. London A 1936, CLIII, 129.

Figure 7. A and Rp vs time for the polymerization of 3a at an applied surface pressure of 10 mN m-1 and a temperature of 25 °C.

of the reaction. The amount of polymer, np, in area β can then be expressed as

np ) [PANI](β) ) β0/A0(ξ)

(11)

When eq 11 is substituted into eq 10, the polymer formation rate can be written as

Rp ) (1/A) (dξ/dt) ) (1/A) (d((A0 - A)/(A0 - A∞))/dt) (12) Figure 6 shows the extent of reaction (ξ) and dξ/dt plotted against time; dξ/dt is calculated from the derivative of ξ against time. When the values of dξ/dt are substituted into eq 12, the overall rate of the reaction can be calculated. Figure 7 shows plots of A and Rp vs time. Like monomers 3a and 3b, the ∆V of their polymeric films can be affected by µ1, µ2, µ3, and ψ0. Because of the complexity of the polymeric film’s counterion and subphase, a more convenient way of expressing the surface potential results is the surface potential per molecule, m, determined from ∆V divided by the number of molecules, n, per centimeters squared:36

m ) ∆V/n

(13)

It is predicted that as the monomer becomes incorporated (36) Harkins, W. D.; Fischer, E. K. J. Chem. Phys. 1933, 1, 852.

6740 Langmuir, Vol. 14, No. 23, 1998

Kloeppner and Duran

Figure 8. m vs π for 3a on a 0.10 M H2SO4 subphase and poly(3a) on 0.10 M H2SO4 and 0.03 M (NH4)2S2O8 subphases at 25 °C.

into the polymer backbone the value of µ2 will decrease as compared to that of the monomer, therefore decreasing the value of m. Figure 8 shows a graph of m vs π for the monomer and polymer. In addition to the lower m values of the polymer, the change in m at different surface pressures is more significant for the monomer (slope ) -2.043 × 10-17m π-1) than for the polymer (slope ) -0.4593 × 10-17m π-1), which is attributed to the larger orientational change on the monomer headgroup relative to that of the more rigid polymer. From studies of the chemical oxidation polymerization of aniline in solution, it was determined that Rp is first order in [ANI] at the initial point in the reaction19 and can be expressed in terms of the concentration of monomer and oxidant:

Rp ) kapp[ANI][OX]

(13a)

where kapp is the apparent first-order rate constant and [OX] is the concentration of oxidant. The concentration of the ammonium peroxydisulfate in a solution polymerization normally ranges from 0.5 to 2 equiv/mol of aniline.8 In a Langmuir film polymerization, the amount of monomer in a Langmuir film is 0.1-0.5 µmol; therefore, to have a concentration of ammonium peroxydisulfate in the subphase that is high enough to polymerize the film, a very large excess is needed. The ratio of ammonium peroxydisulfate to aniline surfactant is ca. 150 000:1. Because of the large excess of oxidant, its concentration is considered constant during the course of the polymerization. Therefore, eq 13a can be simplified and rewritten as

Rp ) kapp[ANI]

(13b)

Upon further examination of Figure 7, it can be determined that Rp increases from its initial rate until it reaches a maximum value and then decreases as the remaining monomer is consumed. The values of kapp are plotted against the amount of poly(3a) formed, as calculated by eq 9, to quantitatively evaluate the effect that polymer formation has on the rate constant kapp. Figure 9 shows the linear correlation that occurs between the increases in kapp and polymer concentration. kapp can now be rewritten in terms of two rate constants (k and k′) and [PANI]:

kapp ) k[PANI] + k′

(14)

Figure 9. kapp vs poly(3a) concentration at the air-aqueous interface. The plot shows polymerization up to 80% completion. k was determined from the slope of the linear plot, and k′ was determined from the y-intercept.

where k′ is the reaction rate constant in the absence of polymer and is determined from the intercept of the line from Figure 9 and k is the observed autoacceleration reaction rate constant when polymer is present and is determined from the slope of this line. When eqs 13b and 14 are combined, the polymerization rate law can be rewritten as

Rp ) k[ANI][PANI] + k′[ANI]

(15)

Similar rate expressions have been derived for the oxidative polymerization of aniline by chemical19 and electrochemical17 means. Figure 10A is a plot of kapp vs polymer concentration for the polymerization of 3b. This figure shows that, after an initial linear increase, kapp follows a nonlinear correlation with the concentration of polymer at the interface. To calculate the values of k′ and k at different surface pressures in the case of 3b, only the linear portion at low polymer concentrations was used to calculate the values of k′ and k. Figure 10B shows an example of a linear fit of kapp at low polymer concentrations. Table 1 shows the values of k′ and k for the interfacial polymerizations of 3a and 3b at surface pressures of 5, 10, 15, 20, 25, and 30 mN/m. Figure 11 shows a plot of ln k′ vs ln π, which indicates that at low pressures (∼5 mN/m) the rate constants for both 3a and 3b are similar. Because both isomers had a similar rate constant (k′) at 5 mN/m, the initial rates of reaction (Rp at time ) 0) at this applied surface pressure were also similar. As the applied surface pressure was increased, 3a showed a linear increase in ln k′, while 3b showed essentially little or no increase. This observation may be explained by the behavior of the monomers on acidic subphases. Not only was the concentration of the surfactant at the air-aqueous interface made larger by increasing the applied surface pressure but a conformational change of the anilinium moiety is probably induced and may contribute to the effect that surface pressure has on k′. In Figure 3, µn values for 3a and 3b are similar at high surface areas and deviate as the monolayer is compressed; the same effect is shown in Figure 11 for k′. As illustrated in Figure 4, the compression of the film of 3a will have the effect of orienting the anilinium group so that the amine group and the para position on an adjacent surfactant molecule are better aligned for polymerization. This conformational change on the reactive group of the surfactant at higher applied surface pressures should make the reaction of the surfactant with a neighboring monomer more thermo-

Behavior of Alkyl-Substituted Aniline Surfactants

Langmuir, Vol. 14, No. 23, 1998 6741

Figure 11. ln k′ vs ln π for the formation of poly(3a) and poly(3b) at 25 °C.

Figure 10. kapp vs poly(3b) concentration at the air-aqueous interface. Plot A shows polymerization up to 80% completion. Plot B shows polymerization up to 15% completion. k was determined from the slope of plot B and k′ was determined from the y-intercept of plot B. Table 1. Values of k and k′ for the Polymerization of 3a and 3b at Applied Surface Pressures between 5 and 30 mN m-1 2-pentadecylaniline (3a)

3-pentadecylaniline (3b)

π (mN m-1)

k′ (min-1)

k (repeat units nm-2 min-1)

k′ (min-1)

k (repeat units nm-2 min-1)

5 10 15 20 25 30

0.0119 0.0239 0.0319 0.0499 0.0618 0.0741

0.0421 0.0364 0.0358 0.0374 0.0399 0.0407

0.0130 0.0154 0.0166 0.0144

0.0132 0.0152 0.0158 0.0140

dynamically favorable, as it is assumed that the polymer backbone is also oriented parallel to the surface of the interface. In the case of 3b, the effect of surface pressure on the rate constant k′ is not as dramatic, which is probably due to a less significant change in the conformation of the reactive group upon compression and, hence, little effect on the polymerization rate. As shown in Table 1, k remains nearly constant for both 3a and 3b as the surface pressure is varied. The value of k in the case of 3a was about 3 times greater than that of 3b. Wei and co-workers have shown that the rates of polymer formation during electrochemical polymerization of 2- and 3-toluidine at a Pt electrode surface are approximately the same.37 Therefore, electronic differ(37) Wei, Y.; Jang, G.; Chan, C.; Hsueh, K. F.; Hariharan, R.; Patel, S. A.; Whitecar, C. K. J. Phys. Chem. 1990, 94, 7716.

ences of the two monomers can be ruled out as a reason for the differences in the rates. Furthermore, if the orientation of the anilinium ion at the interface had an effect on the value of k, a difference in k at different applied surface pressures for the ortho-substituted aniline would also be observed. Different conformations of the headgroup of 3a and 3b are also ruled out as a reason. The different values of k for the two monomers could possibly be explained by steric hindrance. The long alkyl group substituted on the aniline monomer could hinder the approach of the monomer and the reactive polymer, but it is unlikely to be significant enough to explain such a large difference. The values of k′ at low surface pressure and the values of k at low and high pressures were much smaller in the case of 3b than in the case of 3a, suggesting that the total amount of time to complete the polymerization was much longer. Because of the long reaction time and the very large excess of the ammonium peroxydisulfate in the subphase, overoxidation needs to be considered when evaluating kinetic data for this polymerization. It was found that if the polymer was allowed to remain on the subphase containing the ammonium peroxydisulfate for more than 24 h, the polymer completely decomposed. Overoxidation of the polymer is most likely the reason for polymer decomposition at long reaction times. The same reason could explain why kapp (Figure 10A) is not linear at higher polymer concentrations. As the reaction time increases, polymer decomposition will become more significant. The activation energies were calculated for the polymerization of 3a and 3b at 15 mN m-1 and temperatures of 15, 20, 25, 30, and 35 °C using the Arrhenius equation:

ln k ) ln A - (Ea/RT) where k is the rate constant, A the collision frequency factor, Ea the activation energy, R the gas constant, and T the absolute temperature. Ea values calculated for k′ are 88 and 74 kJ mol-1 and those for k 45 and 34 kJ mol-1 for 3a and 3b, respectively. For both k and k′, the activation energy is slightly larger for 3a than for 3b. One possible reason for the differences in Ea for the two isomers is the steric effect of the alkyl group, which is the same reason that 3a occupies more surface area than 3b at the same surface pressure. At greater surface pressures, k′ is larger in the case of 3a, because of the favorable conformation of the reactive group. At lower surface

6742 Langmuir, Vol. 14, No. 23, 1998

pressures (5 mN m-1), k′ is smaller, which can be explained by the similar conformation of the anilinium ion at the interface and the differences in Ea. Conclusion Two-dimensional polymerization of alkyl-substituted aniline at the air-aqueous interface has been presented, and the rates of polymer formation were determined by monitoring the mean molecular area change at a constant applied surface pressure. Polymerization of the m-alkyl derivative, which is reported here for the first time, was compared to that of the o-alkyl derivative. In both cases, the initial reaction rate was dependent on the applied surface pressure. For 3a, the initial rate of polymerization increased as the applied surface pressure was increased, but for 3b, the effect was less significant. This observed effect is due to the differences in the conformation of the reactive group at the interface and activation energies. As the pressure was increased, the conformational change at the interface favored polymerization initiation. The propagation rate constant of polymerization (k) stayed

Kloeppner and Duran

relatively constant as the applied surface pressure changed but was about 3 times smaller in the case of 3b. It is this slower rate of polymer formation in the case of 3b that is used to explain why the overall rate of the reaction falls off significantly as the reaction proceeds. Acknowledgment. We are grateful for the financial support of this research provided in part from the Office of Naval Research (ONR). Additional support was supplied by the National Science Foundation (NSF); KSV Instruments Ltd., Helsinki, Finland; Microcal Software, Inc., Northhampton, MA; and EM Separations. We also acknowledge Mr. Chau Nguyen for his assistance in the synthesis and characterization of the aniline monomers used in this study. Supporting Information Available: Rp values for 3a and 3b vs time at different applied surface pressures between 5 and 30 mN m-1 (1 page). Ordering information is given on any current masthead page. LA980330L

Due to a processing error during production of this article, there are differences between the version posted ASAP on October 7 and the version in this issue (both printed and on the Web). ACS Publications regrets the error.