S-Layer Protein for Resistive Switching and Flexible Nonvolatile

Jan 8, 2018 - Pum , D.; Toca-Herrera , J. L.; Sleytr , U. B. S-Layer Protein Self-Assembly Int. J. Mol. Sci. 2013, 14, 2484– 2501 DOI: 10.3390/ijms1...
0 downloads 0 Views 7MB Size
Research Article www.acsami.org

Cite This: ACS Appl. Mater. Interfaces 2018, 10, 4866−4873

S‑Layer Protein for Resistive Switching and Flexible Nonvolatile Memory Device Akshay Moudgil,†,§ Neeti Kalyani,‡,§ Gaurav Sinsinbar,‡ Samaresh Das,† and Prashant Mishra*,‡ †

Centre for Applied Research in Electronics and ‡Department of Biochemical Engineering and Biotechnology, Indian Institute of Technology Delhi, Hauz Khas, New Delhi 110016, India

Downloaded via WESTERN SYDNEY UNIV on January 14, 2019 at 06:04:52 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

S Supporting Information *

ABSTRACT: In this work, a flexible resistive switching memory device consisting of S-layer protein (Slp) is demonstrated for the first time. This novel device (Al/Slp/indium tin oxide/polyethylene terephthalte) based on a simple and easy fabrication method is capable of bistable switching to low resistive state (LRS) and high resistive state (HRS). This device exhibits bistable memory behavior with stability and a long retention time (>4 × 103 s), being stable up to a 500 cycle endurance test and with significant HRS/LRS ratio. The device possesses consistent switching performance for more than 100 times bending, corresponding to desired applicability for biocompatible wearable electronics. The memory mechanism is attributed to a trapping/de-trapping process in S-layer protein. These promising results of the flexible memory device could find a way in the wearable storage applications like smart bands and sports equipments’ sensors. KEYWORDS: resistive switching, nonvolatile memory, flexible electronics, bionanodevices, S-layer protein bonds in a highly organized manner, forming various patterns.18 Their molecular weight varies between 40 and 200 kDa and consists of similar monomeric proteins or glycoprotein subunits that are able to self-assemble themselves into a pattern, hence forming isoporous patterned monolayers.21 Slps form a highly porous and approximately 10 nm thick protein mesh, which has different lattice arrangements like square, hexagonal, and oblique symmetry,22 with the unit size ranging between 3 and 30 nm. The arrangements depend on the source microorganism of Slp.23 Slp is particularly suited and a favorable candidate for the patterning applications in bio-nanotechnology and biomimetics because of its inherent potential to recrystallize in a specific array in suspension on solid surfaces and on diverse lipid structures.24,25 Slp has been used as building units for the precise bottom-up fabrication of different interfaces and molecular conformations. Previous research has shown application of some biomolecules like proteins, nucleic acids, lipids, and glycans in various devices,26 but application of Slp in the nonvolatile memory devices has not been reported yet. In this work, the behavior of S-layer protein for the development of a novel resistive memory device on a flexible substrate has been examined. For this, the S-layer protein was cloned, overexpressed, and purified in a large amount. The

1. INTRODUCTION The extensive advancement in the digital electronics systems has triggered an urgent need for miniaturization of these systems and developing materials with superior performance. Therefore, new materials have been sought after as an alternative to silicon-based devices to overcome their limitations.1,2 Inspired by nature, several biomolecules are potential candidates for electronic devices due to their inherent properties like biocompatibility, light weight, simple fabrication process, and ease to transfer on flexible substrates. These systems involve discrete processes like the generation, transfer, storing, and processing of information through electronic signals via various reactions.3 Biomolecules with desired functions, structures, and self-assembly characteristics have an added advantage in miniaturization and biodegradability.4 Moreover, biological processes are dynamic, efficient, and precise and hence can be used as biomimetic alternatives in the field of molecular electronics.5,6 Previous research with biomolecules like proteins has already been reported for the development of transistors,7 memory devices,8 biosensors,9 and light-emitting diodes.10 Resistive switching devices based on azurin,11 silk fibroin,12 egg albumin,13 ferritin,14 and bacteriorhodopsin15 have also been studied. S-layer protein (Slp) has been employed in this study, which is present as an outermost layer of the cell envelope of walled bacteria and archaea.16−18 The Slps have been found in some species of Lactobacillus, such as L. acidophilus, L. gallinarum, L. casei, L. buchneri, Leptothorax bulgaricus, L. kefir, and L. brevis.19,20 These proteins are attached to the cell envelope by noncovalent © 2018 American Chemical Society

Received: October 4, 2017 Accepted: January 8, 2018 Published: January 8, 2018 4866

DOI: 10.1021/acsami.7b15062 ACS Appl. Mater. Interfaces 2018, 10, 4866−4873

Research Article

ACS Applied Materials & Interfaces

Figure 1. (a) Secondary structure of Lactobacillus brevis Slp predicted by PSIPRED online server (http://bioinf.cs.ucl.ac.uk/psipred), and (b) threedimensional structure of Slp predicted by Phyre2 web server (http://www.sbg.bio.ic.ac.uk/phyre2).

protein was immobilized on flexible indium tin oxide (ITO) substrates, and its memory characteristics were studied.

layer protein (10 mg/mL) was immobilized by incubating the substrates with protein overnight at 4 °C. The device was fabricated using the indium tin oxide/polyethylene terephthalte (ITO/PET) flexible substrate, where Slp film was deposited on ITO. Then, Slp was etched off from the half portion of the flexible device and this portion acted as the ground terminal for electrical characterization. The aluminum electrode having the diameter 1 mm was patterned by a shadow mask to act as the top electrode and ITO acted as the bottom electrode on the Slp/ITO/PET cell structure.

2. MATERIALS AND METHODS S-layer gene from Lactobacillus brevis ATCC 8287 was cloned in fusion with 6× histidine tag in Escherichia coli. Thereafter, S-layer protein was overexpressed and purified using Ni-NTA chromatography. For the immobilization of S-layer protein, the ITO substrates were functionalized with 3-mercaptopropyltrimethoxysilane (MPTS). The S4867

DOI: 10.1021/acsami.7b15062 ACS Appl. Mater. Interfaces 2018, 10, 4866−4873

Research Article

ACS Applied Materials & Interfaces

Figure 2. Schematic representation of single-crystal growth of S-layer protein preceded by the amorphous to crystal transaction of the S-layer proteins. (1) Lyophilized S-layer proteins have random aggregates of various sizes distributed as observed using dynamic light scattering. It is schematically represented in an intensity vs size (nm) graph obtained by dynamic light scattering. (2) The aggregated or amorphous state of S-layer proteins. (3) The amorphous state restructures to initiate nucleation of S-layer protein to form ordered self-assembled monolayers in solution. (4) The nucleation sites grow in large self-assembled monolayers, and the growth takes place by the assembly of monomeric S-layer proteins at the kink sites (red). The size distribution also narrows down as amorphous to crystal transaction of the S-layer proteins takes place. (5) The self-assembled monolayers of S-layer proteins attain an equilibrium, and after a certain size, the sheets start precipitating because of their large size. The sheet sizes narrow down, as observed using DLS.

3. RESULTS AND DISCUSSION Slp was chosen for this study because of the following reasons: (i) Slp is hydrophilic (GRAVY index = −0.406) and stable (instability index = 0.88), as determined by ProtParam (http://web.expasy.org/cgi-bin/protparam), (ii) Slp is redox inactive and hence can be used for memory applications (Figure S1), (iii) Slp is majorly composed of β-sheets (Figure 1) similar to sericin protein, which has shown excellent memory behavior,27 (iv) Slp layer thickness can be monitored with incubation time (Figure 2).The light-scattering intensity from the self-assembled complex arrangement of Slp depends on both the number of Slp molecules forming the complex and the size of the complex formed.28 We used the dynamic light-scattering (DLS) data to optimize the CaCl2 and protein concentration required for the self-assembly of Slp. After 9 h, the size distribution of Slp protein crystalline domains in solution was large but with time, the size distribution narrows down to a small range. A schematic illustration (Figure 2) of the process of selfassembly of S-layer proteins in solution was deduced on the basis of the data obtained from DLS.28,29 The simple Al-protein-indium tin oxide/polyethylene terephthalte (ITO/PET) device structure, which is shown in Figure 3a,

was used to examine the nonvolatile memory characteristics in Slp. Figure 3b shows the typical atomic force microscopy (AFM) image of Slp immobilized on ITO/PET, where roughness of around 6.3 nm was observed. Figure 3c shows the cross-sectional scanning electron microscopy (SEM) image of the device, where a distinctive layer-by-layer configuration was confirmed. The optical image shown in Figure 3d corresponds to the bending capability of the device. Figure 4a illustrates the typical I−V characterization of the Al/ Slp/ITO/PET device under sweeping mode, where the Al/Slp/ ITO electrode acts as the top terminal and Al/ITO electrode as the ground terminal. When the voltage sweeps from 0 V to the set voltage (Vset), the device switched from a high resistive state (HRS) to low resistive state (LRS). In the next sweep from 8 to 0 V, the memory device stays in the low resistive state. During the sweep from 0 to −8 V, the low resistive state switches back to the high resistive state of the device. The I−V characteristics repeatability was also investigated for more than 500 cycles, which is comparable to that reported earlier.30 Figure 4b shows the I−V characteristics for 10 cycles, with a higher degree of consistency. The endurance parameters were extracted at a read voltage of 5 V. Figure 4c shows the assessment of memory performance of the novel Slp-based flexible device at room 4868

DOI: 10.1021/acsami.7b15062 ACS Appl. Mater. Interfaces 2018, 10, 4866−4873

Research Article

ACS Applied Materials & Interfaces

Figure 3. (a) Schematic of the fabricated flexible memory device with a symmetric electrode configuration. (b) AFM image of Slp, having 6.3 nm roughness. (c) Scanning electron microscopy image of the cross section of the device structure. (d) Optical image of Slp memory device on a flexible substrate.

trapping process within the switching materials.27,31 To probe the switching mechanism of the Slp device for redox reaction properties, electrochemical analysis was carried out using cyclic voltammetry. However, no oxidation/reduction peaks were observed in the cyclic voltammogram (Figure S1). Second, the conception that facile oxidation is accountable for electrochemical memory effect in Al/Slp/ITO configuration is ruled out by assessing the device. The aluminum electrode symmetry also possesses resistive switching memory behavior. The Slp acts as a semiconductor/ insulator inherently (observed from the cyclic voltammogram), and the absence of a permanent memory without applied voltage rules out the possibility of metal filament formation.32 Therefore, the trapping/de-trapping process in the Slp layer could be attributed to the resistive switching mechanism. The output characteristics is confined by the space charge limited current (SCLC) model.32 The I−V characteristics in the log−log scale could be partitioned into two regions, as shown in Figure 4a, inset. The first partition could be where the current is in the linear scale and the second could be after nonlinearity in the current curve, which signify the Ohmic and trap free space charge limit current, respectively. An equation model of the trap-affected current in the Slp layer can be given by following equations,31 where the transport current is J, free charge concentration is n, free carrier concentration of the trapped charge is nt, dielectric constant of the material is ε, free carrier mobility is μ, applied voltage is V, and layer thickness is L.

temperature. The retention characteristics were measured under 5 V read voltage. The device exhibited very stable LRS and HRS states over the 500 cycle measurements, with less than ±5% variation, as shown in Figure 4d. This Slp-based device exhibits bipolar memory, as it sustains only two states for positive as well as negative sweeps. The resistance of both states was almost constant for more than 4 × 103 s and did not show resistance variations for more than ±0.3 kΩ. A significant LRS to HRS switching ratio of around 6.2 was obtained in the Slp memory device. However, the device can be operated in a lower read voltage with two repeatable stable states, as shown in Figure 4b. For example, the device shows 4.05 and 5.9 LRS to HRS ratio at 2 and 4 V read voltage, respectively. The major advantage for any flexible device is their bending capability up to an extent that does not affect the device performance, so to examine such an important aspect, the device was bent more than 100 times at a radius of curvature = 28.5 mm. Figure 4e shows the steady bending characteristics with very consistent switching states at 5 V read voltage. As the bending number goes up, there is a slight decrement in the LRS and HRS. To further investigate the bending performance, the flexible device was bent for various bending radius of curvatures. Figure 4f demonstrates the LRS/ HRS variation with the radius of curvature (flat to 9.5 mm). The variation in the LRS and HRS could be observed for different radius of curvatures. Because of the higher bending stress, the maximum variation is observed for the 9.5 mm radius of curvature. The prominent results obtained in this study suggest that the Slp-based novel memory device can be conceivable for nonvolatile memory applications. The resistive switching mechanism in memory devices occurs due to three proposed phenomena as follows: (1) redox reaction of materials, (2) metal filament formation, and (3) trapping/de-

J= 4869

9nμεV 2 (with traps) 8ntL3

(1) DOI: 10.1021/acsami.7b15062 ACS Appl. Mater. Interfaces 2018, 10, 4866−4873

Research Article

ACS Applied Materials & Interfaces

Figure 4. (a) Current vs voltage (I−V) curves of the Al/Slp/ITO device in voltage sweep mode at ambient condition (log (I)−log (V) plot in the inset), (b) repeatability test for current vs voltage (I−V) characteristics, (c) retention characteristics under 5 V read voltage, (d) endurance characteristics under 5 V read voltage. (e) bending characteristics of the flexible device for both resistive states at 5 V read voltage, and (f) the variation of HRS/LRS with bending radius of curvature (flat to 9.5 mm).

J=

9μεV 2 (filled traps) 8L3

injected from the electrode, which follows eq 1. Hence, the free charge concentration “n” will start to increase. Then, as the voltage increases, at a certain voltage, free charge will be accommodated by trap sites and get filled. These filled sites with charge carriers, as described by eq 2, will form a conductive path, and this mechanism is illustrated in Figure 5 for Slp film. For the higher bias, the transport mechanism can be quantified by the Fowler−Nordeim tunneling expression given by eq 3,33 where E is the electric field, charge is q, electron mass is m*

(2)

Initially, with a small applied voltage, the number of the injected carriers is smaller than thermally injected carriers due to very small electric field application. In the nonlinear region, more charge carriers were injected from the electrode due to increase in the applied voltage. The excess charge carrier generation could not be accommodated by the bulk material, and accordingly, space charges are formed in the vicinity of the interface of the electrode, leading to the SCLC conduction mechanism.32 The LRS mechanism in the Slp memory device may correspond to the hopping conduction, as available trap centers are already filled by the injected carriers. In HRS, Slp film shows near to nonconducting property, as there may be few defects prompted at the Slp and aluminum electrode interface.27 When the voltage is applied, at first, empty sites will be filled by charge carriers

J=

⎡ −4 2m* (qϕ)3/2 E2 ⎤ ⎥ ⎢ exp ⎥⎦ ⎢⎣ 3ℏqE 16π 2ϕℏ q 2E2

(3)

where Φ is the energy barrier and ℏ is the reduced Planck constant. For the tunneling process, ln(J/V2) vs 1/V (V−1) is plotted, where linearity in the curve for more than 6 V (0.125− 0.166 V−1) applied voltage is observed, as shown in Figure 6. At 4870

DOI: 10.1021/acsami.7b15062 ACS Appl. Mater. Interfaces 2018, 10, 4866−4873

Research Article

ACS Applied Materials & Interfaces

Figure 5. Schematic diagram for the conduction mechanism in Slp-based memory, where the hollow circle in Slp represents the unfilled charge site, whereas the filled circle portrays the filled charge site. (a) Initial high resistance state without external bias. (b) Applied voltage injects charge from the electrode to fill few empty charge sites. (c) After applying set voltage, most of the empty site is filled with charge and hopping of charge from one filled site to other is initiated, which corresponds to the low resistance state.(d) Applied reset voltage corresponds to breakage of the conductive path and few sites are left fully filled, and the device switches to HRS.

higher bias, current shoots up due to F−N tunneling, as shown in Figure 6, inset.

Figure 6. ln (J/V2) vs 1/V (V−1) curve for the LRS, inset shows the ln (J/V2) vs 1/V (V−1) Fowler−Nordeim tunneling at higher bias.

An environmental stability test was performed for the longterm performance of the Slp-based flexible memory device. The device was tested under ambient condition for a consecutive span of 4 weeks. Figure 7a shows the stability of both resistive states, where no obvious change was observed for the entire test span. The longevity feature of this device emulates its feasibility to be employed in various biocompatible applications. The device-todevice set voltage and reset voltage distribution was studied, as shown in Figure 7b.Ten devices were tested at ambient conditions for set and reset voltage consistency, where the average swing of voltage distribution was found to be ±0.15 V. This minimal swing entitles the device performance as optimal for the memory operation. The device-to-device resistance of both resistive states at 5 V read voltage was investigated.

Figure 7. (a) Stability test of both resistive states over the 4 week span of the Slp-based flexible memory device (b) Device-to-device distribution of Vset and Vreset, where the results are extracted data from 10 devices. (c) Device-to-device distribution of resistance of both resistive states at 5 V read voltage for 10 devices.

Remarkably, highly stable and consistent HRS and LRS were obtained, as shown in Figure 7c. Overall, for flexible memory applications, this device shows effective usability with high stability. The performance of the memory device can be improved in terms of the on−off current ratio by changing active materials switching properties. 4871

DOI: 10.1021/acsami.7b15062 ACS Appl. Mater. Interfaces 2018, 10, 4866−4873

Research Article

ACS Applied Materials & Interfaces

(3) Lee, T.; El-Said, W. A.; Min, J.; Choi, J. W. Multifunctional DNABased Biomemory Device Consisting of ssDNA/Cu Heterolayers. Biosens. Bioelectron. 2011, 26, 2304−2310. (4) Chung, Y.-H. Redox State Control of a Metalloprotein to Generate Multi-Level Signals for Applications to Bioelectronic Devices. BioChip J. 2015, 9, 215−221. (5) Mas-Torrent, M.; Rovira, C.; Veciana, J. Surface-Confined Electroactive Molecules for Multistate Charge Storage Information. Adv. Mater. 2013, 25, 462−468. (6) Chen, Q.; Yoo, S.-Y.; Lee, T.; Kim, S.-U.; Nam, E. S.; Min, J.; Choi, J.-W. DNA-Recombinant Azurin Conjugation as a Biomemory Platform with Enhanced Sensitivity. J. Nanosci. Nanotechnol. 2016, 16, 11857− 11861. (7) Kalyani, N.; Moudgil, A.; Das, S.; Mishra, P. Metalloprotein Based Scalable Field Effect Transistor with Enhanced Switching Behaviour. Sens. Actuators, B 2017, 246, 363−369. (8) Ko, Y.; Ryu, S. W.; Cho, J. Biomolecule Nanoparticle-Induced Nanocomposites with Resistive Switching Nonvolatile Memory Properties. Appl. Surf. Sci. 2016, 368, 36−43. (9) Saha, S.; Arya, S. K.; Singh, S. P.; Sreenivas, K.; Malhotra, B. D.; Gupta, V. Zinc Oxide-Potassium Ferricyanide Composite Thin Film Matrix for Biosensing Applications. Anal. Chim. Acta 2009, 653, 212− 216. (10) Marcello, A.; Sblattero, D.; Cioarec, C.; Maiuri, P.; Melpignano, P. A Deep-Blue OLED-Based Biochip for Protein Microarray Fluorescence Detection. Biosens. Bioelectron. 2013, 46, 44−47. (11) Yagati, A. K.; Kim, S.-U.; Lee, T.; Min, J.; Choi, J.-W. Recombinant Azurin-CdSe/ZnS Hybrid Structures for Nanoscale Resistive Random Access Memory Device. Biosens. Bioelectron. 2017, 90, 23−30. (12) Wang, H.; Zhu, B.; Wang, H.; Ma, X.; Hao, Y.; Chen, X. UltraLightweight Resistive Switching Memory Devices Based on Silk Fibroin. Small 2016, 12, 3360−3365. (13) He, X.; Zhang, J.; Wang, W.; Xuan, W.; Wang, X.; Zhang, Q.; Smith, C. G.; Luo, J. Transient Resistive Switching Devices Made from Egg Albumen Dielectrics and Dissolvable Electrodes. ACS Appl. Mater. Interfaces 2016, 8, 10954−10960. (14) Meng, F.; Sana, B.; Li, Y.; Liu, Y.; Lim, S.; Chen, X. Bioengineered Tunable Memristor Based on Protein Nanocage. Small 2014, 10, 277− 283. (15) Ashwini, R.; Vijayanand, S.; Hemapriya, J. Photonic Potential of Haloarchaeal Pigment Bacteriorhodopsin for Future Electronics: A Review. Curr. Microbiol. 2017, 74, 996−1002. (16) Debabov, V. G. Bacterial and Archaeal S-Layers as a Subject of Nanobiotechnology. Mol. Biol. 2004, 38, 482−493. (17) Howorka, S.; Sára, M.; Wang, Y.; Kuen, B.; Sleytr, U. B.; Lubitz, W.; Bayley, H. Surface-Accessible Residues in the Monomeric and Assembled Forms of a Bacterial Surface Layer Protein. J. Biol. Chem. 2000, 275, 37876−37886. (18) Breitwieser, A.; Iturri, J.; Toca-Herrera, J.-L.; Sleytr, U.; Pum, D. In Vitro Characterization of the Two-Stage Non-Classical Reassembly Pathway of S-Layers. Int. J. Mol. Sci. 2017, 18, 1−12. (19) Hollmann, A.; Delfederico, L.; Glikmann, G.; De Antoni, G.; Semorile, L.; Disalvo, E. A. Characterization of Liposomes Coated with S-Layer Proteins from Lactobacilli. Biochim. Biophys. Acta, Biomembr. 2007, 1768, 393−400. (20) Khaleghi, M.; Kermanshahi, R. K.; Yaghoobi, M. M.; ZarkeshEsfahani, S. H.; Baghizadeh, A. Assessment of Bile Salt Effects on SLayer Production, Slp Gene Expression And, Some Physicochemical Properties of Lactobacillus acidophilus ATCC 4356. J. Microbiol. Biotechnol. 2010, 20, 749−756. (21) Iturri, J.; Vianna, A. C.; Moreno-Cencerrado, A.; Pum, D.; Sleytr, U. B.; Toca-Herrera, J. L. Impact of Surface Wettability on S-Layer Recrystallization: A Real-Time Characterization by QCM-D. Beilstein J. Nanotechnol. 2017, 8, 91−98. (22) Sára, M.; Sleytr, U. B. S-Layer Proteins. J. Bacteriol. 2000, 182, 859−868. (23) Pum, D.; Toca-Herrera, J. L.; Sleytr, U. B. S-Layer Protein SelfAssembly. Int. J. Mol. Sci. 2013, 14, 2484−2501.

4. CONCLUSIONS Nonvolatile resistive switching characteristics of the Slp flexible device have been demonstrated for the first time. This device exhibits a bistable memory behavior with stability and long retention time (>4 × 103 s), is stable up to 500 cycle endurance test and has a significant HRS/LRS ratio. The device possesses a consistent switching performance for more than 100 times of bending, corresponding to desired applicability for biocompatible wearable electronics. The device-to-device distributions of Vset and Vreset and HRS and LRS turn out to be ±0.15 V and ±0.1 kΩ, respectively. Remarkable environment stability was observed over a span of 4 weeks, with no obvious fluctuations in both resistive states. The trapping/de-trapping of charge carriers and conduction via trapped charge carrier path seems to be responsible for the conduction mechanism of the Slp-based memory device. This nonvolatile memory can be used for active memory storage in a powered off state of electronic chips. These low-cost biocompatible flexible Slp-based memory devices could find a way in the wearable storage applications like smart band, etc.



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.7b15062. Cyclic voltammogram of Slp (PDF)



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: (+91) 01126591015. ORCID

Prashant Mishra: 0000-0002-9947-0496 Author Contributions §

A.M. and N.K. contributed equally.

Author Contributions

The manuscript was written through the contributions of all authors. All authors have given approval to the final version of the manuscript. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS The work was partially supported by the Ministry of Electronics and Information Technology (MeitY), Government of India. Akshay Moudgil received financial support from the Ministry of Electronics and Information Technology, New Delhi. Neeti Kalyani acknowledges financial support from the Council of Scientific and Industrial Research, New Delhi.



ABBREVIATIONS Slp, S-layer protein; ITO, indium tin oxide; PET, polyethylene terephthalate; MPTS, 3-mercaptopropyltrimethoxysilane; LRS, low resistive state; HRS, high resistive state



REFERENCES

(1) Kim, S.-U.; Lee, T.; Lee, J. H.; Yagati, A. K.; Min, J.; Choi, J. W. Nanoscale Film Formation of Ferritin and Its Application to Biomemory Device. Ultramicroscopy 2009, 109, 974−979. (2) Yuan, Y.; Zhou, S.; Yang, G.; Yu, Z. Electrochemical Biomemory Devices Based on Self-Assembled Graphene-Shewanella oneidensis Composite Biofilms. RSC Adv. 2013, 3, 18844−18848. 4872

DOI: 10.1021/acsami.7b15062 ACS Appl. Mater. Interfaces 2018, 10, 4866−4873

Research Article

ACS Applied Materials & Interfaces (24) Sleytr, U. B.; Huber, C.; Ilk, N.; Pum, D.; Schuster, B.; Egelseer, E. M. S-Layers as a Tool Kit for Nanobiotechnological Applications. FEMS Microbiol. Lett. 2007, 267, 131−144. (25) Park, T. J.; Lee, S. J.; Park, J. P.; Yang, M. H.; Choi, J. H.; Lee, S. Y. Characterization of a Bacterial Self-Assembly Surface Layer Protein and Its Application as an Electrical Nanobiosensor. J. Nanosci. Nanotechnol. 2011, 11, 402−407. (26) Moreno-Cencerrado, A.; Iturri, J.; Pum, D.; Sleytr, U. B.; TocaHerrera, J. L. Influencing Bacterial S-Layer Protein Recrystallization on Polymer Brushes through Surface Charge and Accessible Volume: A Combined AFM and QCMD Analysis. Polymer 2016, 102, 379−385. (27) Wang, H.; Meng, F.; Cai, Y.; Zheng, L.; Li, Y.; Liu, Y.; Jiang, Y.; Wang, X.; Chen, X. Sericin for Resistance Switching Device with Multilevel Nonvolatile Memory. Adv. Mater. 2013, 25, 5498−5503. (28) Rad, B.; Haxton, T. K.; Shon, A.; Shin, S.-H. H.; Whitelam, S.; AjoFranklin, C. M. Ion-Specific Control of the Self-Assembly Dynamics of a Nanostructured Protein Lattice. ACS Nano 2015, 9, 180−190. (29) Chung, S.; Shin, S. S.-H.; Bertozzi, C. R.; De Yoreo, J. J. SelfCatalyzed Growth of S Layers via an Amorphous-to-Crystalline Transition Limited by Folding Kinetics. Proc. Natl. Acad. Sci. U.S.A. 2010, 107, 16536−16541. (30) Kim, R. H.; Kim, H. J.; Bae, I.; Hwang, S. K.; Velusamy, D. B.; Cho, S. M.; Takaishi, K.; Muto, T.; Hashizume, D.; Uchiyama, M.; André, P.; Mathevet, F.; Heinrich, B.; Aoyama, T.; Kim, D. E.; Lee, H.; Ribierre, J. C.; Park, C. Non-Volatile Organic Memory with Sub-Millimetre Bending Radius. Nat. Commun. 2014, 5, No. 3583. (31) Pan, F.; Gao, S.; Chen, C.; Song, C.; Zeng, F. Recent Progress in Resistive Random Access Memories: Materials, Switching Mechanisms, and Performance. Mater. Sci. Eng., R 2014, 83, 1−59. (32) Lin, H. T.; Pei, Z.; Chan, Y. J. Carrier Transport Mechanism in a Nanoparticle-Incorporated Organic Bistable Memory Device. IEEE Electron Device Lett. 2007, 28, 569−571. (33) Chang, Y.-F.; Fowler, B.; Chen, Y.-C.; Chen, Y.-T.; Wang, Y.; Xue, F.; Zhou, F.; Lee, J. C. Intrinsic SiOx-Based Unipolar Resistive Switching Memory. II. Thermal Effects on Charge Transport and Characterization of Multilevel Programing. J. Appl. Phys. 2014, 116, No. 043709.

4873

DOI: 10.1021/acsami.7b15062 ACS Appl. Mater. Interfaces 2018, 10, 4866−4873