ARTICLE pubs.acs.org/Langmuir
Electrochemical Impedance Spectroscopy Studies of Organic-SolventInduced Permeability Changes in Nanoporous Films Derived from a Cylinder-Forming Diblock Copolymer D. M. Neluni T. Perera, Bipin Pandey, and Takashi Ito* Department of Chemistry, Kansas State University, 213 CBC Building, Manhattan, Kansas 66506-0401, United States
bS Supporting Information ABSTRACT: In this paper we report electrochemical investigations of the influence of organic solvents dissolved in aqueous solution on the permeability of nanoporous films derived from a cylinder-forming polystyrenepoly(methyl methacrylate) diblock copolymer (CF-PS-b-PMMA). The nanoporous films (ca. 30 nm in pore diameter) were prepared on planar gold electrodes via UV-based degradation of the cylindrical PMMA domains of annealed CF-PS-b-PMMA films (3045 nm thick). The permeability of the electrode-supported nanoporous films was assessed using cyclic voltammetry and electrochemical impedance spectroscopy (EIS). The faradic current of Fe(CN)63/4 decreased upon immersion in aqueous solutions saturated with toluene or methylene chloride (5.8 mM and 0.20 M, respectively). EIS data indicated that the decrease in faradic current mainly reflected an increase in the pore resistance (Rpore). In contrast, Rpore did not change in a saturated n-heptane solution, 0.17 M ethanol, or 5.8 mM aqueous solutions of methylene chloride, diethyl ether, methyl ethyl ketone, or ethanol. Atomic force microscopy images of a nanoporous film in aqueous solution with and without 5.8 mM toluene showed a reversible change in the surface morphology, which was consistent with a toluene-induced change in Rpore. The solvent-induced increase in Rpore was attributed to the swelling of the nanoporous films by the organic solvents, which decreased the effective pore diameter. The reversible permeability changes suggest that the surface of CF-PS-b-PMMA-derived nanoporous films can be functionalized in organic environments without destroying the nanoporous structure. In addition, the solvent-induced swelling may provide a simple means for controlling the permeability of such nanoporous films.
’ INTRODUCTION Nanoporous films derived from cylinder-forming block copolymers have attracted considerable attention because of their structural characteristics such as uniform and controllable nanoscale pore sizes, cylindrical pore shapes, and organized pore distribution.13 In addition, these films can be prepared using a relatively simple procedure: cylindrical nanoscale domain formation via selfassembly and subsequent etching of the domains using chemical,46 thermal,7 or photochemical8,9 processes. Due to these unique features, they have been applied as lithographic masks for fabricating nanoscale structures,13,10 templates for nanomaterials synthesis,1,2,1113 and membranes for chemical separations.2,3,1422 Many such applications require knowledge about the influence of organic solvents on the structural and physicochemical properties of these polymer-based nanoporous films. For example, the solvent resistance of a nanoporous film must be carefully considered in choosing an appropriate chemical environment for nanopore surface functionalization,4,16,2325 nanomaterials synthesis,12,13 and chemical separations.17 The swelling of a nanoporous polymer film due to the uptake of an organic solvent will change film properties such as nanopore size and permeability. r 2011 American Chemical Society
Swelling-based changes in pore size can also be a useful approach to switch size-based selectivity in chemical and biomolecular separations, which was demonstrated by nanoporous membranes modified with stimuli-responsive layers.26,27 Permeability changes due to the solvent-induced swelling and shrinking of a nanoporous film will permit designing electrochemical sensors for solvents, as electrochemical sensing was demonstrated on the basis of nanoporous polymer films responsive to other various stimuli.28 The resistance of nanoporous films derived from cylinder-forming block copolymers was investigated by comparing their atomic force microscopy (AFM) images before and after exposure to pure organic solvents.17 However, the influence of organic solvents on the properties of such nanoporous films have not been assessed quantitatively. Polymer swelling was previously investigated by measuring changes in the mass or volume of a polymer.2931 The extent of polymer swelling by an organic solvent is known to reflect the Received: May 30, 2011 Revised: July 18, 2011 Published: July 20, 2011 11111
dx.doi.org/10.1021/la202005n | Langmuir 2011, 27, 11111–11117
Langmuir relative swelling power of the solvent, which is related to the solubility parameters of the solvent and polymer32,33 and the cross-link density of the polymer.30 Swelling-induced changes in the volume and optical properties of a polymer film were employed to develop interferometric sensors for organic solvent vapors.31,34 In addition, the swelling of hydrophobic polymer resins in aqueous solutions containing organic solvents35 is known to influence their performance as stationary phases in chromatography36 and their efficiency as adsorbents for water treatment.37,38 In this study, the influence of organic solvents on the permeability of a nanoporous thin film derived from a cylinder-forming polystyrenepoly(methyl methacrylate) diblock copolymer (CF-PS-b-PMMA) was investigated using cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS).39,40 Nanoporous films were obtained by spin-casting from toluene solution of CF-PS-b-PMMA (0.3 PMMA volume fraction, 82 000 molecular weight) on planar gold electrodes. Upon annealing of the materials and subsequent UV-based degradation of the PMMA domains, thin films (3045 nm thick) comprised of cylindrical nanopores (ca. 30 nm in diameter) oriented perpendicular to the electrode surface were obtained.8 Pore orientation perpendicular to the underlying electrode facilitated measurements of changes in nanopore permeability as electrochemical signals.16,25,4143 The organic solvents examined in this study ranged from those that strongly swell PS (toluene and methylene chloride) to those that have little or no effect (n-heptane and ethanol).32 Pore resistance values (Rpore) were obtained by fitting EIS data to an appropriate equivalent circuit to assess solvent-induced changes in the film permeability. EIS measurements were also carried out on thin PS homopolymer layers immobilized on gold electrodes to understand the importance of the nanoporous structure of CF-PS-b-PMMA-derived films on solvent-induced changes in film permeability. Finally, AFM was employed to image the surface structure of a nanoporous film in toluene-free and toluene-saturated aqueous solutions. This study will clarify the properties of the nanoporous films in the presence of organic solvents that may be employed for chemical functionalization of the nanopore surface and for the swelling-induced control of the permeability of such films.
’ EXPERIMENTAL SECTION Chemicals and Materials. All aqueous solutions were prepared from water having a resistivity of 18 MΩ cm or higher (Barnstead Nanopure Systems). Films were prepared from 82K CF-PS-b-PMMA (PS, Mn = 57 000 g/mol; PMMA, Mn = 25 000 g/mol, Mw/Mn = 1.07) and 115K PS (Mn = 115 000 g/mol, Mw/Mn = 1.04). Both polymers were purchased from Polymer Source and used as received. Potassium dihydrogen phosphate (Fisher Chemical), potassium hydroxide (Fisher Chemical), potassium nitrate (Fisher Chemical), potassium ferricyanide (Acros Organics), and potassium ferrocyanide (Acros Organics) were all of reagent-grade quality or better and used without further purification. Toluene (Fisher Chemical), methylene chloride (CH2Cl2; Fisher Chemical), methyl ethyl ketone (MEK; Fisher Chemical), diethyl ether (Et2O; Fisher Chemical), ethanol (Decon Laboratories), and n-heptane (Fisher Chemicals) were all used as received. To prepare aqueous solutions saturated with the individual solvents, the mixture of aqueous solution and a small volume of organic solvent was shaken in a stoppered glass container. Gold-coated silicon substrates, which were prepared by sputtering 10 nm of Ti followed by 200 nm of Au onto Si(100) wafers, were purchased from LGA Thin Films (Foster City, CA).
ARTICLE
Figure 1. Schematic illustration of a planar gold electrode coated with a CF-PS-b-PMMA-derived nanoporous film and modified Randles equivalent circuit employed to analyze EIS data measured on polymer-coated electrodes. The nanoporous films were made of PS as a result of the selective etching of the cylindrical PMMA domains in annealed CF-PS-b-PMMA films via UV irradiation and subsequent acetic acid treatment.
Electrode Preparation. CF-PS-b-PMMA-derived nanoporous films on gold electrodes were prepared according to procedures reported previously.42,43 Briefly, thin films of CF-PS-b-PMMA were prepared by spin-casting from toluene solution on clean, gold-coated silicon electrodes. The films were subsequently annealed at 180 °C in vacuum (∼0.3 Torr) for 60 h. The film thickness was measured to be 3046 nm by spectroscopic ellipsometry (J.A. Woollam Alpha-SE). The cylindrical PMMA domains in the film were degraded via UV irradiation (254 nm, 20 mW/cm2, 40 min) under an Ar atmosphere and subsequent sonication in glacial acetic acid. PS homopolymer-coated electrodes were prepared by heating gold electrodes coated with PS cast films at 190 °C for 24 h. These films were then sonicated in toluene for 2 min to remove excess polymer, irradiated by UV (254 nm, 40 min) under an Ar atmosphere, and finally sonicated in glacial acetic acid. The ellipsometric thickness of the PS homopolymer layers was 5.3 ( 1.0 nm. Electrochemical Measurements. Polymer-coated electrodes were immobilized at the bottom of the electrochemical cell as reported previously.16,25,42,43 The film area exposed to the solution was defined by an O-ring (0.66 cm in diameter). CV and EIS measurements were performed in a three-electrode cell containing a Ag/AgCl (3 M KCl) reference electrode and a Pt counter electrode. A CH Instruments model 618B electrochemical analyzer was employed. Prior to CV and EIS measurements, the film-coated electrodes were immersed in aqueous solutions containing 3 mM K3Fe(CN)6, 3 mM K4Fe(CN)6, and 0.1 M KH2PO4K2HPO4 buffer (pH 7) for several hours until the faradic current of Fe(CN)63/4 or EIS data were stable. EIS data were measured at the open circuit potential (0.220.23 V) with a 5 mV ac potential in the frequency range of 105 to 0.001 Hz. The measured complex impedance plots were fitted with an appropriate equivalent circuit model (Figure 1)40,44,45 using software provided by CH Instruments. The fitting error was defined as follows: vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi !2 u i u N Xfiti Xexptl t error ¼
∑i
i Xexptl
N
ð1Þ
In this equation, Xfit and Xexp are fitting and experimental data, respectively, and N is the total number of data points. The fitting error was found to be
