Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Apr 9.
Published in final edited form as: Adv Mater Interfaces. 2021 Mar 1;8(7):2002005. doi: 10.1002/admi.202002005

Fabrication of nanomolded Nafion thin films with tunable mechanical and electrical properties using thermal evaporation-induced capillary force lithography

Jong Seob Choi 1, Jonathan H Tsui 2, Fei Xu 3, Su Han Lee 4, Heon Joon Lee 5, Chao Wang 6, Hyung Jin Kim 7, Deok-Ho Kim 8
PMCID: PMC8115721  NIHMSID: NIHMS1681646  PMID: 33996383

Abstract

In this paper, we report a simple and facile method to fabricate nanomolded Nafion thin films with tunable mechanical, and electrical properties. To achieve this, we combine a novel thermal evaporation-induced capillary force lithography method with swelling process to obtain enhanced pattern fidelity in nanomolded Nafion films. We demonstrate that structural fidelity and mechanical properties of patterned Nafion thin films can be modulated by changing fabrication parameters such as swelling time, Nafion polymer concentration, and curing temperature. Interestingly, we also find that impedance properties of nanomolded Nafion thin films are associated with the Nafion polymer concentration and curing temperature. In particular, 20% Nafion thin films exhibit greater impedance stability and lower impedance values than 5% Nafion thin films at lower frequencies. Moreover, curing temperature-specific impedance changes are observed. These results suggest that capillary lithography can be used to fabricate Nafion nanostructures with high pattern fidelity capable of modifying mechanical and electrical properties of Nafion thin films.

Keywords: capillary force lithography, thermal evaporation, Nafion, nanopatterns, swelling

Graphical Abstract

A facile method for enhancing the fidelity of nanomolded Nafion thin films is described using a novel thermal evaporation-induced capillary force lithography. PDMS mold surfaces were plasma-treated and Nafion nanopatterned substrates could be swelled for further improvement of nanopattern fidelity. In addition, tunability of mechanical and electrical properties of nanomolded Nafion films could be achieved by controlling specific fabrication parameters.

graphic file with name nihms-1681646-f0006.jpg

1. Introduction

Nafion has been widely applied in proton exchange membranes and biointerfaces over the past couple of decades.[16] Nafion is ion-permeable and consists of a hydrophilic sulfonic group and a hydrophobic polyfluoroethylene backbone. Previous findings show that material processing and fabrication conditions play an important role in regulating the characteristics of the produced Nafion.[7,8] Specifically, changes in humidity and processing temperature can influence both the water content and the ion connectivity of the sulfonic group phase, which then affects proton conductivity.[9] Humidity and annealing temperature also have significant effects on the mechanical properties of Nafion.[1014] In addition, the modification of Nafion surface morphology through the patterning of features such as meshes,[15] pillars,[16,17] grooved lines[1820] and prism patterns[21] increases the interfacial area of Nafion membranes and has the potential to improve fuel cell performance.[18,20,22,23]

Patterning thin polymer films by molding typically involves capillary, hydrostatic, and dispersion forces. These forces need to be precisely controlled so that the polymer films can be molded into the features of a polymeric mold with high pattern fidelity and physical integrity.[24] Poor control of size and pattern fidelity can cause undesirable effects in material function. Recently, our group has demonstrated that Nafion films can be nanopatterned via a rapid prototyping and drying process, and that by integrating these nanopatterned films with electrodes, Nafion films can be used for obtaining electrophysiological recordings of excitable cells such as cardiomyocytes and neurons.[5,6] However, Nafion nanopatterns fabricated by solution casting have often encountered critical issues with nanopattern fidelity. For instance, after rapid prototyping, Nafion nanopatterns tended to shrink significantly due to solvent evaporation, and difference in the evaporation rates at the center and at the edges of the mold-substrate interface induced patterning non-uniformity over the entirety of the substrate area. Previous literature exhibited the influence of fabrication parameters such as Nafion concentration and curing temperature upon the fidelity of Nafion nanopatterns.[6]

Here, we report a new thermal evaporation-induced capillary force lithographic method for enhancing nanopattern fidelity in Nafion thin films. Specifically, we investigated the synergistic effects of combining polydimethylsiloxane (PDMS) mold surface plasma treatment and the inherent swelling properties of Nafion to optimize structural fidelity by altering various fabrication parameters such as swelling time, the polymer concentration, and curing temperature. We characterized structural pattern fidelity and mechanical properties of differently fabricated Nafion thin films using nanoindentation, scanning electron microscope (SEM), and atomic force microscopy (AFM). Chemical and electrical properties of nanomolded Nafion thin films were analyzed using X-ray photoelectron spectroscopy (XPS) and electrochemical impedance spectroscopy (EIS).

2. Results and discussion

2.1. Enhanced nanopattern fidelity by thermal evaporation-induced solidification

The fabrication of nanopatterned PUA and PDMS molds, as well as of Nafion thin films, was done using capillary force lithography (CFL) (Figure 1A). Structural fidelity was well maintained when transferring the pattern from an initial Si master mold to the subsequent PUA and PDMS molds (Figure 1B, C). The ridges were 765 ± 12 nm and 739 ± 36 nm, the grooves were 761 ± 18 nm and 734 ± 26 nm, and the depths were 595 ± 11 nm and 584 ± 23 nm, for the PUA and PDMS molds, respectively. Although the feature dimensions of the PUA and PDMS molds were similar to that of the initial Si molds, the dimensional fidelity of the nanopatterns was not preserved for the Nafion thin films (Figure 1D). The ridges produced were 270 ± 15 nm for curing at room temperature (RT), 472 ± 25 nm at 65°C and 532 ± 23 nm at 105°C. Meanwhile, the grooves were 1291 ± 17 nm at RT, 1077 ± 25 nm at 65°C and 1008 ± 19 nm at 105°C, the depths were 276 ± 10 nm at RT, 358 ± 5 nm at 65°C and 447 ± 16 nm at 105°C, respectively (Figure 1E). Curing at more than 105°C caused several challenging issues related to the copy fidelity between the PDMS mold and produced Nafion films. Curing at 145°C was also attempted, but the high temperature caused deformation of the PDMS mold, causing poor feature uniformity in resulting Nafion films. In addition, we tried to use PUA, perfluoropolyether (PFPE) and Norland optical adhesive (NOA) instead of PDMS master molds, which are well-established materials for direct nanopatterning. However, these materials also exhibited some technical limitations, including poor Nafion compatibility and poor film uniformity. Since the curing (or solidification) of Nafion is based on thermal evaporation-induced capillary force lithography, the evaporation rate and the wettability are some of the most important factors that determine fabrication quality. To achieve improved fidelity, the PDMS molds were oxygen plasma-treated before the CFL process to control and increase the surface hydrophilicity. A simple model is presented in order to understand the effects of enhanced wettability on structural fidelity of the nanopatterns fabricated via CFL. The structure of the nanopatterns is determined by a balance between the pressure of trapped air and the atmospheric pressure, described as:

Pa=P0HHh

Figure 1. Fabrication of nanomolded Nafion thin films with high pattern fidelity using thermal evaporation-induced capillary force lithography.

Figure 1.

(A) Schematic illustration of fabrication process integrated with thermal evaporation-induced solidification to generate the Nafion nanopatterns. The representative SEM images of (B) the UV-cured nanopatterned PUA and (C) the nanopatterned PDMS molds generated from the PUA master molds. (D) The representative AFM images of the nanomolded Nafion thin films. (E) Changes in the dimensions of the ridge and the groove widths, and of the groove depths in the Nafion nanopatterns before and after the plasma treatment of the PDMS molds at the various Nafion curing temperatures. In presented data, *p < 0.05, **p < 0.0001. Data were analyzed by the two-way ANOVA.

Where Pa is the pressure of air trapped between the PDMS mold and Nafion, P0 is the initial air pressure during which the capillary rise takes (~1 bar), H is the height of the PDMS mold, and h is the height of the Nafion nanopatterns.[2527] Herein, the only pressure within the trapped air interacting with atmospheric pressure is the Laplace pressure by capillarity PL=(2γcosθ/L), where γ is the surface tension of the Nafion/air interface, θ is the contact angle of the Nafion/air interface, and L is the length of the mold channel. Hence, the above equation (1) may be expressed as:

h=H(1P0L2γcosθ)

According to well established literature,[28] the maximum aspect ratio (AR: ratio between height and width; h/w) before the collapse of patterns is given by:

AR=hw=EL8γsinθ3

Where E is the elastic modulus of the material. Thus, the width of the patterns is expressed as:

w=H(1P0L2γcosθ)EL/8γsinθ3

This is in accordance with the proposed model, which displays increase in the width of the Nafion nanopattern ridges with decrease in the value of θ. With oxygen plasma treatment, the fabricated nanomolded Nafion thin films exhibited wider ridges and thinner grooves, resulting in a significant improvement in the fidelity of these features; however, groove depth was slightly diminished (Figure 1E). We investigated the effects of various curing temperatures since the PDMS molds were observed to swell when exposed to Nafion solvent. By minimizing curing time through an increase in temperature, the duration of the PDMS exposure to the solvent could be reduced. Consequently, the widths of the ridges increased from 270 ± 15 nm to 445 ± 31 nm at room temperature (RT). Accordingly, the 472 ± 25 nm-wide ridges were increased to 560 ± 22 nm at 65°C and the 532 ± 23 nm-wide ridges to 632 ± 14 nm at 105°C. For the grooves, the 1291 ± 18 nm-wide grooves were reduced to 1136 ± 32 nm at RT, the 1077 ± 25 nm-wide grooves to 1028 ± 27 nm at 65°C, and the 1008 ± 19 nm-wide grooves to 961 ± 13 nm at 105°C. Fast evaporation at higher curing temperature reduces the exposure time between a PDMS mold and Nafion solvent, resulting in minimizing PDMS mold deformation and increased Nafion copy fidelity. Meanwhile, pattern depth was reduced under all the temperature conditions.

2.2. Further enhancement of Nafion nanopattern fidelity induced by swelling

The swelling phenomenon of Nafion thin films occurs when solvent molecules diffuse in between Nafion polymer chains and cause an increase in the spacing between the chains, thus causing an overall expansion of volume.[29] We observed that the swelling process increased the cross-sectional area of the nanomolded features for both 5% and 20% Nafion thin films, but with different expansion ratios. Nafion swelling appeared to peak at 24 h, as demonstrated by the invariant nanopattern dimensions after 24 h (Figure 2A). Additionally, AFM analysis showed that the 20% Nafion nanomolded thin films had better nanopattern fidelities than the 5% Nafion thin films after the 24 h swelling process. In contrast, the swelling ratio of 5% Nafion nanopatterns was relatively low, and thus the final feature dimensions were smaller than the initial mold dimensions. Moreover, the structures fabricated with 5% Nafion tended to collapse after 24 hours, resulting in undesirable feature dimensions. In dry Nafion thin films (before immersion in PBS), we observed that “rabbit ears” appeared on the top of the nanopatterned ridges (Figure 2B). These shapes were likely formed by irregular capillary rise of polymer solution along the mold walls during evaporation. However, these shapes disappeared after the swelling process occurs in PBS. Therefore, the inherent swelling property of Nafion could be harnessed to further improve nanopattern fidelity in these thin films.

Figure 2. Enhanced nanopattern fidelity with swelling process.

Figure 2.

(A) Changes in the Nafion nanopattern feature dimensions with swelling in the PBS at the different time points. The dotted line indicates the initial mold dimensions. (B) The representative AFM images of the Nafion nanopatterns immersed in the PBS for 0 h and 24 h, respectively. In presented data, **p < 0.0001, Data were analyzed by the one-way ANOVA.

2.3. Tunable mechanical properties of nanomolded Nafion thin films by curing temperature

We investigated the mechanical properties of nanomolded Nafion thin films using nanoindentation method. The 5% Nafion thin films possess significantly higher modulus and hardness than the 20% Nafion thin films for all curing temperatures (Figure 3AC). Furthermore, an increase in curing temperature leads to a corresponding increase in the modulus and the hardness for both 5% and 20% Nafion thin films. To find a link between cross-sectional morphology and mechanical properties, SEM images were taken (Figure 3D) and XPS analysis (Figure 4) was conducted for the fabricated thin films. As seen in the SEM images, the cross-sectional morphology of the Nafion thin films were dramatically changed by fabrication parameters such as concentration and curing temperature. For the 5% Nafion cured at RT, the main two spectrum peaks (CF2-O-CF2 and SO3-) of O1s are detected at ~532.66 eV and 534.89 eV, respectively. By curing the Nafion at 65°C and 105°C, the binding energy peaks (CF2-O-CF2) are shifted to the left, to 535.37 eV and 535.73 eV, respectively. Similar trends were also observable for the 20% Nafion. This shift indicates an increase in binding energies between the molecules, which would make the Nafion stiffer and harder. Additionally, we analyzed the ratio of peak area for CF2-O-CF2 and SO3- corresponding to specific curing temperatures to show that the ratio of area for the two peaks are closely related to the curing temperatures. Specifications of these results are summarized in Table 1. -SO3- peak area for 5% and 20% Nafion gradually increased with curing temperature. This may be attributed to -SO3- functional groups in Nafion molecules tightly interacting with one another in thermal conditions via hydrogen bonding. In addition, according to previous literature,[6] 20% Nafion shows broad amorphous regions in XRD spectrum, implying that 5% Nafion is more crystalized and hard relative to the 20% Nafion. As a result, the mechanical properties of the Nafion nanopatterns can be controlled via fabrication parameters for further applications.

Figure 3. Mechanical property characterization of nanomolded Nafion thin films.

Figure 3.

(A) Reduced modulus, (B) Young’s modulus, and (C) surface hardness measured by nanoindentation. Unpaired t test *p<0.0001 two-tailed. (D) Representative SEM images of cross-sections of different Nafion membranes that were fabricated by changing curing temperatures (RT, 65°C and 105 °C) and concentration (5% and 20%).

Figure 4. Deconvolution of O1s XPS spectra for 5% and 20% Nafion thin films prepared at different curing temperatures.

Figure 4.

The ratio of CF2-O-CF2 and -SO3- peaks is closely related to curing temperature. (A) 5% (B) 20%, (i) 25°C, (ii) 65°C and (iii) 105°C

Table 1.

With increasing curing temperatures, -SO3- peak is more generated for both 5% and 20% Nafion

Nafion
Concentration
Curing
Temperature
CF2-O-CF2 : -SO3-
5% 25°C 1 : 0.92
65°C 1 : 1.23
105°C 1 : 1.78
20% 25°C 1 : 0.85
65°C 1 : 0.90
105°C 1 : 2.94

2.4. Electrical property characterization of nanomolded Nafion thin films

Electrochemical impedance spectroscopy has been used extensively to gain insight into the electrochemical activity of capacitors, batteries, and fuel cells.[30] To measure and characterize the electrochemical activity of nanomolded Nafion thin films that were fabricated using a variety of parameters (concentration, curing temperature), electrodes devices were overlaid and integrated with nanomolded Nafion. Total impedance is influenced in the high frequency regimes by Ohmic behavior, while the impedance at lower frequencies is dominated by ion conductance (Figure 5B). 20% Nafion thin films exhibit greater impedance stability and lower impedance values than 5% Nafion thin films at lower frequencies. Moreover, cure temperature specific impedance changes were observed. Increasing curing temperature induces lower impedance values at lower frequencies for both 5% and 20% Nafion thin films. The Nyquist plots (Figure 5C and D) represent the interfacial characteristic responses of the Nafion nanopatterned layers, and the inclined and straight lines of the spectra at all frequencies are most likely caused by the ion diffusion process between two separated electrodes via the Nafion films.[31] The charge transfer resistance of Nafion layer is proportional to the diameter of the semicircle or slope of Nyquist plots.[3133] 5% and 20% Nafion thin films fabricated at higher curing temperature showed lower slopes, which indicates a stronger attraction and penetration of ions within PBS into the Nafion thin films cured at higher temperatures, resulting in faster ion diffusion through the Nafion. On the other hand, higher slopes for Nafion thin films fabricated at lower temperatures seemed to indicate electric double layer formation of ions on the electrode/electrolyte interface.[32,34] At lower frequencies and lower curing temperature, the generation of electric double layers directly onto the electrodes impeded the ion diffusion/movements between two separated electrodes, which resulted in increased measurement errors and higher impedance values. Furthermore, 5% Nafion exhibited higher imaginary and real impedance values at lower frequency ranges than 20% Nafion layer (Figure 5C), which also could be due to the electric double layer formation at the nanopattern surface/electrolyte interface.[32] This data illustrates different ion permeable/conductive properties at different fabrication parameters of Nafion nanostructures.

Figure 5. Electrochemical impedance spectroscopy curves of nanomolded Nafion thin films with respect to curing temperature and concentration.

Figure 5.

(A) Configuration and fabrication of interdigitated electrode devices with Nafion nanointerface. Photolithography was used to define the interdigitated electrodes, a PDMS chamber was designed to cover the electrode area, Nafion solution was deposited inside the PDMS chamber and pressed by a PDMS nanopatterned stamp followed by curing until completely dry. (B) Total impedance behaviors vs. frequency. (C) The Nyquist plots (imaginary Z vs. real Z) for different Nafion layers. (D) Zoom-in images of high frequency ranges in (C).

3. Conclusion

In this study, we describe a facile method for enhancing the fidelity of nanomolded Nafion thin films using a novel thermal evaporation-induced capillary force lithography. This approach is based on the well-controlled polymer solidification after solvent evaporation. The Nafion nanopattern fidelity is partially improved with the controlled evaporation rates. Plasma treatment of the PDMS mold surfaces was proven to be useful to obtain more hydrophilic surfaces, which increases the wettability and thus may improve the capability to hold the Nafion polymer chains onto the PDMS surfaces. The improved Nafion nanopatterns are further swelled, additionally enhancing structural fidelity. In addition to structural fidelity, we demonstrated that the mechanical properties can be modulated by changing fabrication parameters such as swelling time, the Nafion polymer concentration and curing temperature. Furthermore, we found that impedance properties of nanomolded Nafion thin films can be controlled by the fabrication parameters such as cure temperature and polymer concentration. Enhanced nanopattern fidelity, tunable mechanical and electrical properties of ion-permeable Nafion membranes could be beneficial in controlling interfacial area of the membrane, which may improve the performances of fuel cell or electronic systems.

4. Experimental Section/Methods

Fabrication of nanomolded Nafion thin films:

Nanomolded Nafion thin films were prepared from polyurethane acrylate (PUA; MINS-301RM, Minuta Technology Company)-based master molds. Nanopatterned PUA molds were prepared from a silicon (Si) master mold with an anisotropic arrangement of ridges and grooves, where each ridge was 800 nm wide and each groove was 600 nm deep. The Si master mold was prepared by National Nanofab Center (South Korea). Due to residual PDMS and Nafion remaining in between the nanopatterns of the Si master after peel-off, a secondary PUA master mold was added to the fabrication process. PUA solution was deposited onto the Si mold and pressed using a polyethylene terephthalate (PET) film to homogeneously spread the solution. After 2 min of UV irradiation, the PUA film was cured and then peeled off of the Si master mold. At least 4 h of additional UV irradiation was required for the complete crosslinking of the PUA film. Using this secondary PUA mold, polydimethylsiloxane (PDMS; Sylgard 184, Dow) molds were then prepared. PDMS mixtures with a 10:1 base:catalyst weight ratio were poured onto the PUA secondary mold, degassed, and cured in an 65°C oven for at least 4 h. Nafion solution (Sigma Aldrich, 5 wt.% and 20 wt.% in mixture of lower aliphatic alcohols and water) was deposited onto a PET substrate and pressed with the nanopatterned PDMS mold. After curing in an oven for 24 hrs, the PDMS master mold was carefully removed from the PET substrate. The nanomolded Nafion on PET was then immersed in phosphate buffered saline (PBS) for 24 hrs to swell.

Surface characterization of nanomolded Nafion thin films:

Scanning electron microscopy (SEM, ZEOL, JSM-7610F) was used for analyzing the cross-sectional morphologies of Nafion thin films. SEM samples were prepared by depositing drops of Nafion solution onto a glass coverslip and cured in an oven for 24 hrs before breaking the coverslip and attached Nafion thin film in half. Nafion samples were sputter-coated with Au/Pt. Atomic force microscopy (AFM; Bruker ICON) was used with a silicon tip at a 1 Hz scan rate to measure the Nafion nanopattern dimensions. The measured area was 10 μm × 10 μm. X-ray photoelectron spectroscopy (XPS; ESCALAB 250, Thermo Scientific) equipped with a micro-focused Al K-α source was used for chemical analysis. The base pressure was 2 ~ 3 × 10−8 mbar.

Nanoindention testing:

Hardness and modulus data were collected on a Bruker Hysitron TI-980 triboindenter. A Berkovich geometry diamond tip attached to a standard load transducer was used. The tip was calibrated for size using a standard fused quartz sample. A maximum load of 100 μN, which also yielded a displacement between 100 nm and 200 nm, was used for all data. The subsequent data and reduced modulus (Er) of each sample were analyzed and measured using the Hysitron TriboScan analysis software to get the averages and the standard deviations for the measured values. Young’s modulus Esample was recalculated using the equation (5) below for the standard diamond indenter probe Eindenter ~ 1140 GPa, the indenter Poisson’s ratio vindenter ~ 0.07, and the sample Poisson’s ratio vsample ~ 0.4 for Nafion.[35,36]

1Er=(1vsample2Esample)+(1vindenter2Eindenter)

Impedance analysis of nanomolded Nafion thin films:

Impedance properties of nanomolded Nafion layers were measured on an Autolab PGSTAT302N potentiostat equipped with an electrochemical impedance spectroscopy (EIS) module. The frequency interval employed for the measurements ranged from 1 MHz to 100 Hz with an AC amplitude of 10 mV. Before the experiments, electrodes were prepared by conventional photolithography. First, a negative photoresist was spin-coated and patterned under mask aligner to define the electrode patterns and then 20 nm of Cr and 200 nm of Au were deposited in sequence by a thermal evaporator. The negative photoresist was lifted off by soaking in acetone for 1h. In order to confine the Nafion and electrolyte solution on the electrodes device, PDMS chambers were fabricated to enclose each device. A PDMS sheet 5 mm in thickness was prepared by mixing PDMS prepolymer and curing agent (10:1 weight ratio), degassed to remove bubbles, and placed in an 65°C oven to cure overnight. Chambers were then created by using a 6 mm hole puncher on the cured PDMS sheet and attached to each electrodes device. To add nanomolded Nafion thin films onto the electrodes, Nafion solution was dropped inside each chamber and pressed with a prepatterned PDMS mold. After curing at R.T., 65 °C, and 105 °C in an oven for 3 days, the PDMS mold was carefully removed from the surface of electrodes device, leaving behind a nanomolded Nafion membrances. The PDMS chamber was filled with 1 M PBS (Sigma Aldrich) to submerge the Nafion layer.

Acknowledgements

This work was supported by National Institutes of Health grants R01 HL135143 and R01 HL146436 (to D.-H.K.). This work was also supported by the Ministry of Health & Welfare, Republic of Korea (HI19C0642) (to D.-H.K.). This research was financially supported by the Ministry of Trade, Industry and Energy and Korea Institute for Advancement of Technology (KIAT) through the International Cooperative R&D program. (Project No. P0004638)

Footnotes

Conflict of Interest

The authors declare the following competing financial interest(s): Deok-Ho Kim is a scientific founder and equity holder of Curi Bio.

Contributor Information

Jong Seob Choi, Department of Biomedical Engineering, Johns Hopkins University, Baltimore, MD, 21205, United States.

Jonathan H. Tsui, Department of Biomedical Engineering, Johns Hopkins University, Baltimore, MD, 21205, United States

Fei Xu, Department of Materials Science and Engineering, Johns Hopkins University, Baltimore, MD, 21205, United States.

Su Han Lee, Digital Healthcare Research Center, Gumi Electronics and Information Technology Research Institute (GERI), 350-27, Gumidaero, Gumi, Gyeongbuk 39253, Republic of Korea.

Heon Joon Lee, Department of Biomedical Engineering, Johns Hopkins University, Baltimore, MD, 21205, United States.

Chao Wang, Department of Materials Science and Engineering, Johns Hopkins University, Baltimore, MD, 21205, United States.

Hyung Jin Kim, Digital Healthcare Research Center, Gumi Electronics and Information Technology Research Institute (GERI), 350-27, Gumidaero, Gumi, Gyeongbuk 39253, Republic of Korea.

Deok-Ho Kim, Department of Biomedical Engineering, Johns Hopkins University, Baltimore, MD, 21205, United States; Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, MD, 21205, United States; Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, MD, 21205, United States.

References

  • [1].Mauritz KA, Moore RB, Chem. Rev. 2004, 104, 4535. [DOI] [PubMed] [Google Scholar]
  • [2].Gierke TD, Munn GE, Wilson FC, J. Polym. Sci. Polym. Phys. Ed. 1981, 19, 1687. [Google Scholar]
  • [3].Gebel G, Lambard J, Macromolecules 1997, 30, 7914. [Google Scholar]
  • [4].Schmidt-Rohr K, Chen Q, Nat. Mater. 2008, 7, 75. [DOI] [PubMed] [Google Scholar]
  • [5].Smith AST, Choi E, Gray K, Macadangdang J, Ahn EH, Clark EC, Laflamme MA, Wu JC, Murry CE, Tung L, Kim D-H, Nano Lett. 2020, 20, 1561. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [6].Choi JS, Smith AST, Williams NP, Matsubara T, Choi M, Kim J, Kim HJ, Choi S, Kim D, Adv. Funct. Mater. 2020, 1910660. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [7].Yazdanpour M, Esmaeilifar A, Rowshanzamir S, Int. J. Hydrogen Energy 2012, 37, 11290. [Google Scholar]
  • [8].Cossari P, Simari C, Cannavale A, Gigli G, Nicotera I, Solid State Ionics 2018, 317, 46. [Google Scholar]
  • [9].Frieberg BR, Page KA, Graybill JR, Walker ML, Stafford CM, Stafford GR, Soles CL, ACS Appl. Mater. Interfaces 2016, 8, 33240. [DOI] [PubMed] [Google Scholar]
  • [10].Benziger J, Bocarsly A, Cheah MJ, Majsztrik P, Satterfield B, Zhao Q, In Structure and Bonding; 2011; pp. 85–113. [Google Scholar]
  • [11].Tang Y, Karlsson AM, Santare MH, Gilbert M, Cleghorn S, Johnson WB, Mater. Sci. Eng. A 2006, 425, 297. [Google Scholar]
  • [12].Kusoglu A, Karlsson AM, Santare MH, Polymer (Guildf). 2010, 51, 1457. [Google Scholar]
  • [13].Kusoglu A, Tang Y, Lugo M, Karlsson AM, Santare MH, Cleghorn S, Johnson WB, J. Power Sources 2010, 195, 483. [Google Scholar]
  • [14].Kusoglu A, Karlsson AM, Santare MH, Cleghorn S, Johnson WB, J. Power Sources 2006, 161, 987. [Google Scholar]
  • [15].Kim H-T, Reshentenko TV, Kweon H-J, J. Electrochem. Soc. 2007, 154, B1034. [Google Scholar]
  • [16].Cuynet S, Caillard A, Bigarré J, Buvat P, Int. J. Hydrogen Energy 2017, 42, 7974. [Google Scholar]
  • [17].Hiroshiba N, Yano W, Okumura R, Ichikawa Y, IEICE Trans. Electron. 2015, E98.C, 133. [Google Scholar]
  • [18].Bae JW, Cho Y-H, Sung Y-E, Shin K, Jho JY, J. Ind. Eng. Chem. 2012, 18, 876. [Google Scholar]
  • [19].Taylor AD, Lucas BD, Guo LJ, Thompson LT, Power Sources J 2007, 171, 218. [Google Scholar]
  • [20].Yildirim MH, te Braake J, Aran HC, Stamatialis DF, Wessling M, J. Memb. Sci. 2010, 349, 231. [Google Scholar]
  • [21].Kim SM, Kang YS, Ahn C, Jang S, Kim M, Sung Y-E, Yoo SJ, Choi M, Power Sources J 2016, 317, 19. [Google Scholar]
  • [22].Aizawa M, Gyoten H, Salah A, Liu X, J. Electrochem. Soc. 2010, 157, B1844. [Google Scholar]
  • [23].Jeon Y, Kim DJ, Koh JK, Ji Y, Kim JH, Shul Y-G, Sci. Rep. 2015, 5, 16394. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [24].Kim P, Kim DH, Kim B, Choi SK, Lee SH, Khademhosseini A, Langer R, Suh KY, Nanotechnology 2005, 16, 2420. [DOI] [PubMed] [Google Scholar]
  • [25].Suh K-Y, Park MC, Kim P, Adv. Funct. Mater. 2009, 19, 2699. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [26].Yoon H, Kim T, Choi S, Suh KY, Kim MJ, Lee HH, Appl. Phys. Lett. 2006, 88, 254104. [Google Scholar]
  • [27].Suh KY, Jeong HE, Kim D-H, Singh RA, Yoon E-S, J. Appl. Phys. 2006, 100, 034303. [Google Scholar]
  • [28].Chini SF, Amirfazli A, Langmuir 2010, 26, 13707. [DOI] [PubMed] [Google Scholar]
  • [29].Feng C, Li Y, Qu K, Zhang Z, He P, RSC Adv. 2019, 9, 9594. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [30].Wang X, Habte BT, Zhang S, Yang H, Zhao J, Jiang F, He Q, Anal. Chem. 2019, 91, 11678. [DOI] [PubMed] [Google Scholar]
  • [31].Chen Q, Ai S, Ma Q, Yin H, J. Appl. Electrochem. 2010, 40, 1379. [Google Scholar]
  • [32].Mei B-A, Munteshari O, Lau J, Dunn B, Pilon L, J. Phys. Chem. C 2018, 122, 194. [Google Scholar]
  • [33].Kumar R, Xu C, Scott K, RSC Adv. 2012, 2, 8777. [Google Scholar]
  • [34].Ibrahim M, Claudel J, Kourtiche D, Nadi M, J. Electr. Bioimpedance 2019, 4, 13. [Google Scholar]
  • [35].Kirsten W, Int. J. Electrochem. Sci. 2017, 2573. [Google Scholar]
  • [36].Xia R, Zhou H, Wu R, Wu W-P, Polymers (Basel). 2016, 8, 344. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES