How Ligands Affect Resistive Switching in Solution-Processed HfO2

Jan 17, 2018 - As flash memory approaches its scaling limits, recent studies have focused on the development of emerging nonvolatile memory devices su...
3 downloads 10 Views 2MB Size
Subscriber access provided by LUNDS UNIV

Article

How Ligands Affect Resistive Switching in Solution-Processed HfO Nanoparticle Assemblies 2

Jiaying Wang, Satyan Choudhary, Jonathan De Roo, Katrien De Keukeleere, Isabel Van Driessche, Alfred J Crosby, and Stephen S. Nonnenmann ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.7b17376 • Publication Date (Web): 17 Jan 2018 Downloaded from http://pubs.acs.org on January 17, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Applied Materials & Interfaces is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

How Ligands Affect Resistive Switching in Solution-Processed HfO2 Nanoparticle Assemblies Jiaying Wang1, Satyan Choudhary2, Jonathan De Roo3, Katrien De Keukeleere3, Isabel Van Driessche,3 Alfred J. Crosby2, and Stephen S. Nonnenmann1,* 1. J. Wang, Prof. S.S. Nonnenmann Department of Mechanical and Industrial Engineering, University of MassachusettsAmherst, 160 Governors Drive, 219 Engineering Laboratory I, Amherst, MA, 01003, USA, *e-mail: [email protected] 2. S. Choudhary, Prof. A.J. Crosby Polymer Science and Engineering Department, University of Massachusetts Amherst, Silvio O. Conte National Center for Polymer Research, 120 Governors Drive, Amherst, MA 01003, USA 3. Jonathan De Roo, Katrien De Keukeleere, Isabel Van Driessche Department of Chemistry, Ghent University, Krijgslaan 281- building S3, B-9000 Gent, Belgium Keywords: (solution-processed, resistive switching, ligands, nanoparticles, hafnium oxide) Abstract Advancement of resistive random access memory (ReRAM) requires fully understanding the various complex, defect-mediated transport mechanisms to further improve performance. While thin film oxide materials have been extensively studied, the 1 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 21

switching properties of nanoparticle assemblies remain underexplored due to difficulties in fabricating ordered structures. Here we employ a simple flow coating method for the facile deposition of highly ordered HfO2 nanoparticle nanoribbon assemblies. The resistive switching character of nanoribbons was determined to correlate directly with the organic capping layer length of their constituting HfO2 nanoparticles, using oleic acid, dodecanoic acid, and undecenoic acid as model nanoparticle ligands. Through a systematic comparison of the forming process, operating set/reset voltages, and resistance states, we demonstrate a tunable resistive switching response by varying the ligand type, thus providing a base correlation for solution-processed ReRAM device fabrication. Introduction As flash memory approaches its scaling limits, recent studies have focused on the development of emerging non-volatile memory devices such as resistive random access memory (ReRAM),1-3 phase change memory (PCM)4 and ferroelectric random access memory (FeRAM)5. ReRAM exhibits fast switching speeds, excellent reliability, and low power consumption, thus motivating its use as a next generation memory platform.1-2 Resistive switching is defined by a reversible change between a high resistance state (HRS) and low resistance state (LRS) dependent upon the history of an applied electric field.6 While the fabrication and characterization of resistive switching oxide thin films,7-10 nanoparticles,11 and nanowires12 have received considerable attention, the development of rapid and low-cost solution-processed alternatives to complicated multi-step lithographic approaches remains stunted. Many studies of charge transport within solution-processed semiconductors such as colloidal CdSe quantum dots and PbS photovoltaic nanoparticles exist,13,14 however few studies demonstrate resistive switching between nanoparticles, where the ligand interparticle interface profoundly affects the response.15-16

Studies of the memristive character of individual, sol-gel 2

ACS Paragon Plus Environment

Page 3 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

produced TiO2 nanoparticles revealed both complementary and bipolar switching character that exhibited high sensitivity to annealing and partial redox processes involving the substoichiometric core.11 Ag/Ag2Se nanoparticle film/Au17 and Ag/Ag2Se NP/MnO NP/Au bilayer devices18 exhibited low power consumption, high retention, and high endurance properties that were retained upon the application of ± 0.4% stress states. Most studies of resistive switching within colloidal systems observe the electrical characteristics of spin-coated films comprising binary oxide19 or perovskite complex oxide20 nanoparticles, dip-coated thin films,21-23or compact pellets.24 These assemblies are annealed, forming dense structures free of ligands. Other approaches use electrostatic accelerating voltage during film growth to induce interparticle bonding.25 These studies remove the ligands between the nanoparticles, effectively transforming the low-dimensional nanostructures into thin films. To improve the performance of solution-processed memristive systems the correlation between changes in ligand chemistry and properties such as operating voltages, device stability, and switching type must be systematically evaluated. We previously demonstrated memristive behavior in nanoribbons comprising perovskite strontium titanate (SrTiO3) nanocubes capped with oleic acid that either retained or altered its switching mechanism upon transfer from the original substrate to a second, arbitrary substrate.26 This study extends our previous work to systematically compare the dependence of resistive switching behavior on ligand length within individual HfO2 nanoribbons capped with oleic, dodecanoic, and undecenoic acid. Hafnia, HfO2, is a prototypical resistive switching binary oxide that exhibits excellent scalability, reliability, and CMOS compatibility in both amorphous and crystalline form.27 We use a facile, low-cost method to prepare highly ordered nanoribbon arrays comprising single-crystalline HfO2 nanocrystals via convective self-assembly. This approach addresses typical issues facing nanocrystal assembly (i.e. scalability and periodicity) to produce ordered nanostructure arrays without the complexity of fabrication templates or

3 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 21

lithographic patterning.28 Here the HfO2 nanoribbons exhibited both threshold switching (TS) and bipolar resistive switching (BRS) induced by manipulating the conductive filament morphology, similar to recent studies of NiO thin films.29 Both the forming voltage and SET voltage scale with ligand length, suggesting that the interparticle tunnel distance heavily influences the carrier transport within HfO2 nanoribbons. Motivated by these results, we finally demonstrate the fabrication and characterization of a solution-processed hafnia thin film device that exhibits lower operating voltages, higher uniformity, and larger ON/OFF ratios than its ALDprepared counterpart. Results and discussion HfO2 nanocrystals (NCs) were synthesized from hafnium chloride and benzyl alcohol in a recently developed solvothermal process.30-31 After synthesis, the nanocrystals were washed with diethyl ether and functionalized with a carboxylic acid (oleic acid, dodecanoic acid, or undecenoic acid). Deposition of the HfO2 NCs into nanoribbon assemblies required the implementation of the ‘stop-and-go’ flow coating method, previously demonstrated for CdSe quantum dots28,32 and Au nanoparticles.33-34 For further details regarding nanocrystal synthesis and nanoribbon deposition we direct the reader to the methods section. Nanoribbons comprising HfO2 NCs capped with three types of ligands; oleic acid, dodecanoic acid, and undecenoic acid, labeled hereafter as HfO2-O, HfO2-D and HfO2-U, were deposited onto silicon substrate coated with 5 nm Ti / 30 nm Pt as the bottom electrode. Figure 1 shows the nanocrystalline morphology of the HfO2 nanoparticles and schematic illustration of the ‘stopand-go’ flow coating method. The transmission electron microscopy (TEM) image of HfO2-D in Figure 1a shows the as-prepared nanocrystal structure, while the inset shows a high-resolution TEM image that indicates NC dimensions on the order of 5 nm. The x-ray diffraction (XRD) in Figure 1b reveals the phase pure, monoclinic structure of HfO2 NCs. The stop-and-go flow coating method, as illustrated in Figure 1c, deposits nanoparticles onto a hard substrate by 4 ACS Paragon Plus Environment

Page 5 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

flowing a NC solution under the rigid silicon blade. The height and the width of the ribbon can be precisely controlled from nanoscale to bulk dimensions by varying the stop time of the stop shift, thus enabling the direct fabrication of nanostructures ranging from individual nanoribbons to thin films. The optical microscopy image in Figure 1d shows highly ordered arrays of individual HfO2 nanoribbons prepared by ‘stop-and-go’ flow coating. Figure 1e shows the three-dimensional topographic profile of an individual HfO2 nanoribbon collected via atomic force microscopy (AFM), indicating that nanoribbons possess a wedge-like cross sectional profile, with an average width of 8 µm and an average height (maximum; top of the wedge) of 100 nm. Conductive AFM (c-AFM) is a powerful tool for studying RS behavior with the ability to spatially resolve local heterogeneities within the electronic response, and probe the current distribution simultaneously.11, 35-36 Figure 2a illustrates the general c-AFM setup, where a Pt-Ir coated silicon tip is placed in direct contact with the surface of individual HfO2 nanoribbons, creating a Pt-Ir/HfO2 NCs/Pt test structure. Here the bias is applied to the bottom Pt electrode, with the current flowing through the ribbon vertically, which is then read via a grounded, conductive tip rastering along the top surface of the nanoribbon. No current is measured across the nanoribbons before a forming process is induced under a large voltage applied to the middle of the ribbon. The large applied field creates and initiates the migration of oxygen vacancies, manifesting as a bright spot corresponding to a highly conductive region, as shown in Figure 2b on the HfO2-U system. Current imaging was conducted by scanning the nanoribbon with an applied voltage of +2.5 V (Figure 2b), suggesting the conducting path formed locally during forming/SET process and its nonvolatile property. The LRS is then recovered in the followed scan with -2 V applied bias (Figure 2c). The electroforming process induced filament formation within HfO2 nanoribbons capped with the three ligand systems is shown in Figure 3a, where the HfO2-U, HfO2-D and HfO2-O systems displayed forming voltages of 6.1 V, 8.2 V and 10 V, respectively. As 10 V represents 5 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 21

the limit of the internal bias source of the AFM, electroforming HfO2-O required applying the bias of 10 V for longer time periods (between 30 – 60 sec). Figure 3b shows the semi-log I-V response of the three ligand capped HfO2 nanoribbons. They all display typical bipolar switching character, where the SET and RESET process occur under positive and negative voltage, respectively, and exhibit a variation in operating voltage, current level and the resistance ratio between them. Here we found the HfO2-D displays a smaller set voltage of 1.2 V while HfO2-O requires a larger set voltage of 2 V. A standard 1 MΩ resistor was used to transition between operative bipolar resistive switching and threshold switching modes in the nanoribbons. The current-voltage (I-V) character observed in Figure 3c represents the threshold behavior under voltage sweeping (± 2 V, 1 Hz) when the resistor is connected in series with the HfO2-U system. Unlike the bipolar switching, the threshold switching shows a symmetric I-V curve. Moreover, it shows a high resistance (109 Ω) initially followed by an abrupt increase in current level upon reaching a threshold voltage of approximately +1 V. The ribbon maintains LRS and recovers to the HRS suddenly at a hold voltage of +0.25 V. This threshold behavior can be tuned with the external resistor for each sample, exhibiting similar operating voltage ranges and stability. Figure 3d shows 50 switching cycles within a single HfO2-U nanoribbon, suggesting the TS behavior remains stable, with a selectivity of 103. When the resistor is removed, the typical bipolar resistive switching hysteresis is observed. To evaluate the effect of ligand length on memristive behavior, we estimated the interparticle spacing of the three samples based on the covalent chain length: HfO2-U (1.44 nm), HfO2-D (1.57 nm), HfO2-O (2.2 nm). Figure 4a plots the SET and RESET voltage distribution for bipolar switching as a function of the ligand length. The SET voltage clearly scales with increasing ligand length, while the RESET voltage exhibits a smaller dependence on length. An increase in the VSET distribution can also be observed in the statistical plot with the increased ligand length, while for VRESET, however, no clear correlation is found. Figure 4b presents the

6 ACS Paragon Plus Environment

Page 7 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

cumulative distribution function of the LRS and HRS of the three ligand types. The results were collected using DC sweeping mode at a read voltage of 0.5 V. The resistance in LRS exhibits a much smaller variation compared to the large variations observed for HRS for all three types. The HfO2-O system shows a large HRS fluctuation range of about 104 while the HfO2-U and HfO2-D systems show improved uniformity of about 103 and 102, respectively. Previous studies of PbSe NCs showed that the carrier mobility depends on nanoparticle size and the ligand length, the latter of which determines the interparticle tunnel distance.15 Using the same nanoparticle size for all samples, we posit the variation in the ligand length causes the observed difference in RS behavior. The charge transport mechanism involves sequential electron hopping through nanoparticles, as described by a tunneling model where the mobility strongly depends on the interparticle distance16 Here both the forming (Figure 3a) and the subsequent SET process (Figure 4a) display a ligand length dependence, as both VF and VSET increase with increasing ligand length. During the charge transfer process the ligand facilitates electron tunneling by effectively serving as an insulating spacer.37 Using the c-AFM tip as the top electrode forms a Pt-Ir tip-HfO2 NCs-bridgemetal structure, where the term bridge indicates the capping ligand separating neighboring NCs and the metal bottom electrode with the bulk HfO2 NCs. When a positive voltage is applied to the bottom electrode during the electroforming process, the high electric field enables the creation of oxygen vacancies within the NCs and induces migration toward the tip along the grain boundary to form a low resistance conductive path.38 The electrons injected from the tip under the negative voltage thus pass through the conductive channel and tunnel between NPs, inducing a transition from HRS to LRS. When a negative voltage is applied to the bottom electrode, the oxygen vacancies migrate towards the counter electrode as driven by the electric field of opposite sign, thus severing the conductive channel. The mechanism driving the formation and rupture of conducting filaments in HfO2-based ReRAM is typically attributed to 7 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

oxygen vacancy migration.27,

38-40

Page 8 of 21

For the inorganic:organic capping ligand system presented

here we cannot fully exclude the possibility of more complicated switching mechanisms involving ligand-mediated vacancy transport, however as the ligands comprise long hydrocarbon backbones and possess limited available oxygen sites compared to the oxide nanoparticles, this process is highly unlikely to occur. The coexistence of tunable TS and BRS character within HfO2 and NiO thin films29, 40-42 has been attributed to current-controlled morphology changes of the dominant conductive filament. Joule heating also contributes to thermally-assisted processes and thus facilitates the formation and rupture of the conducting path.41 Limiting the current flow with an external resistor significantly reduces the current and subsequently induces an instability within the filament, ultimately resulting in filament rupture after the applied voltage is removed. Conversely, the current increases when the external resistor is removed such that a stronger filament is formed by continuous growth, thus producing an active conducting path even after the removal of external voltage. The significant ligand length dependent RS behavior is observed in both electroforming and subsequent switching processes. In general, variations in VRESET (Fig. 4a) and the LRS (Fig. 4b) result from either changes in the filament diameter or the number of conducting filaments formed during the SET process, while variations in VSET (Fig. 4a) and the HRS (Fig. 4b) highly depend on variations in the tunneling gap.27 HfO2-O nanoribbons exhibit a larger distribution in VSET and HRS compared to the other two ligand systems, suggesting that the large variation in the reset process occurs due to the longer ligand length of oleic acid. Also, HfO2-U possesses a lower HRS resistance compared to HfO2-D since it possesses the shortest ligand length for electron tunneling at HRS. To demonstrate a solution-processed memristive platform, we fabricated a crossbar device with a flow-coated HfO2-U thin film sandwiched between Ti top and Pt bottom electrode. 8 ACS Paragon Plus Environment

Page 9 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

The schematic of the cross-point HfO2-U test structure is shown in Figure 5a. Using the same solution concentration, the width of the ribbon is simply extended by varying the stopping distance and substrate velocity. Using the longer stopping distance, we fabricated a 5x5 µm2 crossbar type device comprising a 40 nm thick HfO2-U thin film sandwiched between 30 nm thick Pt electrodes. The bipolar resistive switching I-V response was measured by sweeping a DC voltage from -0.5 V to +1.5 V after a forming process using a compliance current of 100 µA (Figure 5b). An ALD-deposited 10 nm HfO2 ReRAM device with the same area was fabricated for comparison. Here the solution-processed HfO2-U device clearly exhibits smaller operating voltages, narrow distribution/increased uniformity, and larger ON/OFF ratio as shown in Figure 5c, d. The average SET and RESET voltages were 0.97 V and -0.36 V, respectively for HfO2-U; the ALD-deposited HfO2 displayed values of 3.2 V and -2.5 V, respectively. The ON/OFF ratio improved from 10 for ALD HfO2 to 105 for the HfO2-U at the expense of a small decrease in the uniformity of the resistance. Thus, the solution-processed device exhibits both stable switching behavior and small operating voltages that are competitive with more commonly employed ALDdeposited HfO2 thin film-based devices.

Conclusion In summary, we introduced a facile, inexpensive and versatile fabrication method to produce memristive nanoribbon structures comprising solution-processed, binary hafnium oxide nanoparticles. We demonstrated that ligand length direct affects the resistive switching behavior by comparing three commonly employed ligand chemistries. Our c-AFM results showed that individual HfO2 nanoribbons demonstrated both threshold switching and bipolar resistive switching modes, dependent upon an external resistance applied to the system, which limits current and further changes the filament morphology. Oxygen vacancies formed during hydrothermal process and under high applied voltage drive the switching behavior, while the operating parameters and stability correlate directly with ligand length. Finally, a solution-based

9 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 21

device displays promising advantages, including low cost, lower operating voltages, increased uniformity, and larger ON/OFF ratio compared to the typical ALD thin film HfO2 memory device. The chosen ligands for this study are frequently used for nanoparticle stabilization, and thus should immediately apply to optimizing resistive switching behavior in other solution-processed, complex oxide nanoparticle systems. Methods. Hafnia Nanocrystal Synthesis: HfO2 nanocrystals (0.4 mmol HfO2) were synthesized from HfCl4 and benzyl alcohol according to De Roo et al.30-31 After solvothermal synthesis and washing with diethyl ether, the nanocrystals are redispersed in chloroform (4mL). 0.2 mmol of fatty acid (oleic acid, dodecanoic acid, or 10-undecenoic acid) was added to the milky suspension. In case of 10-undecenoic acid, 5 % of dodecanoic was also added to ensure colloidal stability. Under stirring or ultrasonication, oleylamine (0.15 mmol or 50 µL) was added until a transparent and colorless suspension was obtained. Finally, the particles are purified three consecutive times by adding acetone to induce precipitation, followed by centrifugation and resuspension in chloroform. After the last purification, the nanocrystals were dispersed in toluene. HfO2 Nanoribbon ‘Stop-and-Go’ Assembly: The custom-built flow coating instrument uses a silicon wafer (University Wafer Inc.) cut to an edge length of 15 mm as a fixed blade. This blade is attached to a tilt-translation stage with pitch, roll and height control capability. With visual guidance, the blade edge is aligned parallel to the substrate at the approximate height 150-300 µm for all coating procedure. The substrate is attached to a programmable nanopositioner (Burleigh Inchworm controller 8200), which performs a series of stop-and-go steps. Here the primary variables that control the nanoribbon geometry (height, width) are solution concentration, stop time (td), spacing (d) and velocity between stops (v). The HfO2 nanoribbons in this study were prepared using the following parameters: v = 1500 µm/s, solution concentration = 1 mg/ml, d = 200 µm, and a td between 1000 to 6000 ms. Different stop times were used to vary the 10 ACS Paragon Plus Environment

Page 11 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

nanoribbon height (h) and width (w), td = 1000 ms (h = 80 nm, w = 8 µm), td = 3000 ms (h = 160 nm, w = 13 µm), and td = 6000 ms (h = 230 nm, w = 17 µm). A fixed volume of 1 mg/ml HfO2 NC solution (10 µl) is injected between the fixed blade and the substrate, where capillary forces confine the solution to the blade edge. The HfO2 NC nanoribbons were deposited directly on silicon wafer substrates (undoped , University Wafer Inc.) coated with 5 nm Ti / 30 nm Pt thin films. Before deposition all substrates and blades were rinsed with isopropyl alcohol and toluene then dried after each step with a filtered stream of N2 gas. Associated Content Supporting Information The Supporting Information is available free of charge on the ACS Publications website at DOI: Topographic

AFM

scans,

(wedge)

thickness-dependent

current-voltage

(I-V)

spectroscopy, thickness-dependent first voltage sweep electroforming I-V response, thicknessdependent BRS I-V response; cycling endurance plots for BRS of nanoribbons capped with the three different ligands. Acknowledgements This research was partially supported by University of Massachusetts-Amherst start-up funding and the UMass Center for Hierarchical Manufacturing (CHM), a NSF Nanoscale Science and Engineering Center (CMMI-1025020). A.J.C. and S.C. gratefully acknowledge financial support from the Army Research Office (W911NF-14-1-0185). The authors also acknowledge use of the facilities at the CHM Conte Nanotechnology Cleanroom for thin film deposition and plasma clean processes. J.D.R. acknowledges support from the Belgian American Education Foundation (BAEF), Fulbright and the European Union's Framework Program for Research and Innovation Horizon 2020 (2014-2020) under the Marie SkłodowskaCurie Grant Agreement COMPASS No. 691185. J.W. performed all substrate/electrode preparation, c-AFM measurements, transport analysis, and wrote the manuscript. S.C. 11 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 21

performed all nanoribbon fabrication and transfer to arbitrary substrates. J.D.R. and K.D.K. performed hydrothermal synthesis and structural characterization of HfO2 NCs. A.J.C. co-wrote the manuscript and supervised all nanoribbon formation. S.S.N. lead the overall project, supervised general fabrication, local probe, and data analysis studies and wrote the manuscript.

12 ACS Paragon Plus Environment

Page 13 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Figures and Figure Captions

Figure 1 a, TEM image shows the crystalline structure of HfO2 nanoparticles with the scale bar denotes 50 nm. Inset: HRTEM image of NCs indicates the lattice fringe and size of 5 nm. b, Xray diffraction patterns of monoclinic HfO2. c, An illustration of the “stop-and-go” flow coating process. Inset shows the assembly of the NCs driven by solvent evaporation. The shape and profile of the ribbon are confirmed by d, three-dimensional AFM image. e, An optical micrograph (scale bar 200 µm) showing highly ordered ribbons comprising HfO2 nanoparticles. The width and height of the wedge are approximately 8 µm and 100 nm, respectively.

13 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 21

Figure 2 a, An illustration of the c-AFM measurement setup across an individual HfO2 nanoribbon. b, Current mapping of individual HfO2-D ribbon with an applied voltage of +2.5 V, displaying a yellow conducting spot formed after a forming voltage of +6 V. c, Corresponding current mapping of the area with an applied voltage of -2 V. No observed conducting spot indicates the locally reset process occurred.

14 ACS Paragon Plus Environment

Page 15 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Figure 3 a, I-V curves for the forming process of HfO2-U, HfO2-D and HfO2-O, respectively. b, Representative I-V response of BRS for three samples. c, I-V curve of the TS behavior of HfO2U sample. d, 50 cycles endurance of TS by sweeping at a read voltage of 0.5 V, showing a selectivity of 103 between the HRS (light purple; open symbols) and LRS (dark purple; closed symbols).

15 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 21

Figure 4 a, The statistical distribution plot of switching voltages as a function of ligand length. Both VSET and its distribution scale with increasing ligand length. b, The cumulative probability for the LRS (filled symbols) and HRS (open symbols) of HfO2-U (purple; square), HfO2-D (green; circle), and HfO2-O (orange; triangle), respectively.

16 ACS Paragon Plus Environment

Page 17 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Figure 5 a, Schematic of a single Pt/HfO2-U/Pt ReRAM device. b, The representative I-V characteristics of Pt/HfO2-U/Pt ReRAM device. c, Comparative statistic cumulative probability of the operating voltage for HfO2-U and ALD-prepared HfO2. d, The cumulative probability of the resistances in each state of HfO2-U and ALD-prepared HfO2.

17 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 21

References 1. Waser, R.; Aono, M., Nanoionics-Based Resistive Switching Memories. Nat. Mater. 2007, 6 (11), 833-40. 2. Sawa, A., Resistive Switching in Transition Metal Oxides. Mater. Today 2008, 11 (6), 2836. 3. Ilia, V.; Rainer, W.; John, R. J.; Michael, N. K., Electrochemical Metallization Memories—Fundamentals, Applications, Prospects. Nanotechnology 2011, 22 (25), 254003. 4. Ielmini, D.; Lacaita, A. L., Phase Change Materials in Non-Volatile Storage. Mater. Today 2011, 14 (12), 600-607. 5. Nonnenmann, S. S.; Gallo, E. M.; Spanier, J. E., Redox-Based Resistive Switching in Ferroelectric Perovskite Nanotubes. Appl. Phys. Lett. 2010, 97 (10), 102904. 6. Yang, J. J.; Pickett, M. D.; Li, X.; OhlbergDouglas, A. A.; Stewart, D. R.; Williams, R. S., Memristive Switching Mechanism for Metal//Oxide//Metal Nanodevices. Nat. Nanotechnol. 2008, 3 (7), 429-433. 7. Waser, R.; Dittmann, R.; Staikov, G.; Szot, K., Redox-Based Resistive Switching Memories – Nanoionic Mechanisms, Prospects, and Challenges. Adv. Mater. 2009, 21 (25-26), 2632-2663. 8. Valov, I., Redox-Based Resistive Switching Memories (ReRAMs): Electrochemical Systems at the Atomic Scale. ChemElectroChem 2014, 1 (1), 26-36. 9. Chang, T.-C.; Chang, K.-C.; Tsai, T.-M.; Chu, T.-J.; Sze, S. M., Resistance Random Access Memory. Mater. Today 2016, 19 (5), 254-264. 10. Kyung Min, K.; Doo Seok, J.; Cheol Seong, H., Nanofilamentary Resistive Switching in Binary Oxide System; a Review on the Present Status and Outlook. Nanotechnology 2011, 22 (25), 254002. 11. Schmidt, D. O.; Hoffmann-Eifert, S.; Zhang, H.; La Torre, C.; Besmehn, A.; Noyong, M.; Waser, R.; Simon, U., Resistive Switching of Individual, Chemically Synthesized TiO2 Nanoparticles. Small 2015, 11 (48), 6444-6456. 12. Nagashima, K.; Yanagida, T.; Oka, K.; Taniguchi, M.; Kawai, T.; Kim, J.-S.; Park, B. H., Resistive Switching Multistate Nonvolatile Memory Effects in a Single Cobalt Oxide Nanowire. Nano Lett. 2010, 10 (4), 1359-1363. 13. Voznyy, O.; Sutherland, B. R.; Ip, A. H.; Zhitomirsky, D.; Sargent, E. H., Engineering Charge Transport by Heterostructuring Solution-Processed Semiconductors. Nat. Rev. Mater. 2017, 2, 17026. 14. Niu, G.; Wang, L.; Gao, R.; Ma, B.; Dong, H.; Qiu, Y., Inorganic Iodide Ligands in Ex Situ Pbs Quantum Dot Sensitized Solar Cells with I-/I3- Electrolytes. J. Mater. Chem. 2012, 22 (33), 16914-16919. 15. Liu, Y.; Gibbs, M.; Puthussery, J.; Gaik, S.; Ihly, R.; Hillhouse, H. W.; Law, M., Dependence of Carrier Mobility on Nanocrystal Size and Ligand Length in Pbse Nanocrystal Solids. Nano Lett. 2010, 10 (5), 1960-1969. 16. Yu, X.; Malvankar, N.; Landis, R.; Eymur, S.; Miranda, O. R.; Rotello, V. M., Impedance Spectroscopy of Ionic Ligand-Modulated Charge Transport of Gold Nanoparticle Films. Small 2015, 11 (31), 3814-3821. 17. Jang, J.; Pan, F.; Braam, K.; Subramanian, V., Resistance Switching Characteristics of Solid Electrolyte Chalcogenide Ag2Se Nanoparticles for Flexible Nonvolatile Memory Applications. Adv. Mater. 2012, 24 (26), 3573-3576. 18. Hu, Q.; Lee, T. S.; Lee, N. J.; Kang, T. S.; Park, M. R.; Yoon, T.-S.; Lee, H. H.; Kang, C. J., Resistive Switching Characteristics in MnO Nanoparticle Assembly and Ag2Se Thin Film Devices. J. Nanosci. Nanotechnol. 2017, 17 (10), 7189-7193.

18 ACS Paragon Plus Environment

Page 19 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

19. Verbakel, F.; Meskers, S. C. J.; de Leeuw, D. M.; Janssen, R. A. J., Resistive Switching in Organic Memories with a Spin-Coated Metal Oxide Nanoparticle Layer. J. Phys. Chem. C 2008, 112 (14), 5254-5257. 20. Shirolkar, M. M.; Li, J.; Dong, X.; Li, M.; Wang, H., Controlling the Ferroelectric and Resistive Switching Properties of a BiFeO3 Thin Film Prepared Using Sub-5 nm Dimension Nanoparticles. Phys. Chem. Chem. Phys. 2017, 19 (38), 26085-26097. 21. Quanli, H.; Sung Mok, J.; Hyun Ho, L.; Yong-Sang, K.; Young Jin, C.; Dae-Hwan, K.; KiBum, K.; Tae-Sik, Y., Resistive Switching Characteristics of Maghemite Nanoparticle Assembly. J. Phys. D Appl. Phys. 2011, 44 (8), 085403. 22. Jae Woo, Y.; Quanli, H.; Yoon-Jae, B.; Young Jin, C.; Chi Jung, K.; Hyun Ho, L.; DoJoong, L.; Hyun-Mi, K.; Ki-Bum, K.; Tae-Sik, Y., Resistive Switching Characteristics of Maghemite Nanoparticle Assembly on Al and Pt Electrodes on a Flexible Substrate. J. Phys. D Appl. Phys. 2012, 45 (22), 225304. 23. Kim, J.-D.; Baek, Y.-J.; Choi, Y. J.; Kang, C. J.; Lee, H. H.; Kim, H.-M.; Kim, K.-B.; Yoon, T.-S., Investigation of Analog Memristive Switching of Iron Oxide Nanoparticle Assembly between Pt Electrodes. J. Appl. Phys. 2013, 114 (22), 224505. 24. Kim, T. H.; Jang, E. Y.; Lee, N. J.; Choi, D. J.; Lee, K.-J.; Jang, J.-t.; Choi, J.-s.; Moon, S. H.; Cheon, J., Nanoparticle Assemblies as Memristors. Nano Lett. 2009, 9 (6), 2229-2233. 25. Verrelli, E.; Tsoukalas, D.; Normand, P.; Kean, A. H.; Boukos, N., Forming-Free Resistive Switching Memories Based on Titanium-Oxide Nanoparticles Fabricated at Room Temperature. Appl. Phys. Lett. 2013, 102 (2), 022909. 26. Wang, J.; Choudhary, S.; Harrigan, W. L.; Crosby, A. J.; Kittilstved, K. R.; Nonnenmann, S. S., Transferable Memristive Nanoribbons Comprising Solution-Processed Strontium Titanate Nanocubes. ACS Appl. Mater. Interfaces 2017, 9 (12), 10847-10854. 27. Wong, H. S. P.; Lee, H. Y.; Yu, S.; Chen, Y. S.; Wu, Y.; Chen, P. S.; Lee, B.; Chen, F. T.; Tsai, M. J., Metal–Oxide RRAM. Proc. IEEE 2012, 100 (6), 1951-1970. 28. Kim, H. S.; Lee, C. H.; Sudeep, P. K.; Emrick, T.; Crosby, A. J., Nanoparticle Stripes, Grids, and Ribbons Produced by Flow Coating. Adv. Mater. 2010, 22 (41), 4600-4604. 29. Peng, H. Y.; Li, Y. F.; Lin, W. N.; Wang, Y. Z.; Gao, X. Y.; Wu, T., Deterministic Conversion between Memory and Threshold Resistive Switching Via Tuning the Strong Electron Correlation. Sci. Rep. 2012, 2, 442. 30. De Roo, J.; Van den Broeck, F.; De Keukeleere, K.; Martins, J. C.; Van Driessche, I.; Hens, Z., Unravelling the Surface Chemistry of Metal Oxide Nanocrystals, the Role of Acids and Bases. J. Am. Chem. Soc. 2014, 136 (27), 9650-9657. 31. De Roo, J.; Van Driessche, I.; Martins, J. C.; Hens, Z., Colloidal Metal Oxide Nanocrystal Catalysis by Sustained Chemically Driven Ligand Displacement. Nat. Mater. 2016, 15 (5), 517521. 32. Lee, D. Y.; Pham, J. T.; Lawrence, J.; Lee, C. H.; Parkos, C.; Emrick, T.; Crosby, A. J., Macroscopic Nanoparticle Ribbons and Fabrics. Adv. Mater. 2013, 25 (9), 1248-1253. 33. Pham, J. T.; Lawrence, J.; Lee, D. Y.; Grason, G. M.; Emrick, T.; Crosby, A. J., Highly Stretchable Nanoparticle Helices through Geometric Asymmetry and Surface Forces. Adv. Mater. 2013, 25 (46), 6703-6708. 34. Lawrence, J.; Pham, J. T.; Lee, D. Y.; Liu, Y.; Crosby, A. J.; Emrick, T., Highly Conductive Ribbons Prepared by Stick–Slip Assembly of Organosoluble Gold Nanoparticles. ACS Nano 2014, 8 (2), 1173-1179. 35. Lee, M. H.; Hwang, C. S., Resistive Switching Memory: Observations with Scanning Probe Microscopy. Nanoscale 2011, 3 (2), 490-502. 36. Lee, H.; Kim, H.; Van, T. N.; Kim, D.-W.; Park, J. Y., Nanoscale Resistive Switching Schottky Contacts on Self-Assembled Pt Nanodots on SrTiO3. ACS Appl. Mater. Interfaces 2013, 5 (22), 11668-11672.

19 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 21

37. Adams, D. M.; Brus, L.; Chidsey, C. E. D.; Creager, S.; Creutz, C.; Kagan, C. R.; Kamat, P. V.; Lieberman, M.; Lindsay, S.; Marcus, R. A.; Metzger, R. M.; Michel-Beyerle, M. E.; Miller, J. R.; Newton, M. D.; Rolison, D. R.; Sankey, O.; Schanze, K. S.; Yardley, J.; Zhu, X., Charge Transfer on the Nanoscale:  Current Status. J. Phys. Chem. B 2003, 107 (28), 6668-6697. 38. Lanza, M.; Bersuker, G.; Porti, M.; Miranda, E.; Nafría, M.; Aymerich, X., Resistive Switching in Hafnium Dioxide Layers: Local Phenomenon at Grain Boundaries. Appl. Phys. Lett. 2012, 101 (19), 193502. 39. Xavier, S.; Enrique, M.; David, J.; Shibing, L.; Ming, L.; Joan Marc, R.; Francesca, C.; Jordi, S., Threshold Switching and Conductance Quantization in Al/HfO2/Si(P) Structures. Jpn. J. Appl. Phys. 2013, 52 (4S), 04CD06. 40. Sharath, S. U.; Vogel, S.; Molina-Luna, L.; Hildebrandt, E.; Wenger, C.; Kurian, J.; Duerrschnabel, M.; Niermann, T.; Niu, G.; Calka, P.; Lehmann, M.; Kleebe, H.-J.; Schroeder, T.; Alff, L., Control of Switching Modes and Conductance Quantization in Oxygen Engineered HfOx Based Memristive Devices. Adv. Func. Mater. 2017, 27 (32), 1700432-n/a. 41. Chang, S. H.; Lee, J. S.; Chae, S. C.; Lee, S. B.; Liu, C.; Kahng, B.; Kim, D. W.; Noh, T. W., Occurrence of Both Unipolar Memory and Threshold Resistance Switching in a NiO Film. Phys. Rev. Lett. 2009, 102 (2), 026801. 42. Lin, C.-Y.; Chen, P.-H.; Chang, T.-C.; Chang, K.-C.; Zhang, S.-D.; Tsai, T.-M.; Pan, C.H.; Chen, M.-C.; Su, Y.-T.; Tseng, Y.-T.; Chang, Y.-F.; Chen, Y.-C.; Huang, H.-C.; Sze, S. M., Attaining Resistive Switching Characteristics and Selector Properties by Varying Forming Polarities in a Single HfO2-Based RRAM Device with a Vanadium Electrode. Nanoscale 2017, 9 (25), 8586-8590.

20 ACS Paragon Plus Environment

Page 21 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

TOC Image

21 ACS Paragon Plus Environment