1000-Fold Lifetime Extension of a Nickel Electromechanical Contact

Subscriber access provided by READING UNIV

Article

>1000-Fold Lifetime Extension of Nickel Electromechanical Contact Device via Graphene Min-Ho Seo, Jae-Hyeon Ko, Jeong Oen Lee, Seung-Deok Ko, Jeong Hun Mun, Byung Jin Cho, Yong-Hyun Kim, and Jun-Bo Yoon ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.7b15772 • Publication Date (Web): 20 Feb 2018 Downloaded from http://pubs.acs.org on February 21, 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 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 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 31 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

>1000-Fold Lifetime Extension of Nickel Electromechanical Contact Device via Graphene Min-Ho Seo†, Jae-Hyeon Ko‡, Jeong Oen Lee†, §, Seung-Deok Ko†,∥, Jeong Hun Mun†, Byung Jin Cho†, Yong-Hyun Kim‡*, and Jun-Bo Yoon†* †

School of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), 291 Daehak-ro, Yuseong-gu, Daejeon 34141, Republic of Korea

‡

Graduate School of Nanoscience and Technology, Korea Advanced Institute of Science and Technology (KAIST), 291 Daehak-ro, Yuseong-gu, Daejeon 34141, Republic of Korea

§

Present address: Department of Electrical Engineering, California Institute of Technology, Pasadena, CA 91125, USA. ∥

Present address: Department of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA 30308, USA.

*Correspondence to: [email protected] (Tel: +82-42-350-3476, Fax: +82-42-350-8565) and [email protected] (Tel: +82-42-350-1111, Fax: +82-42-350-1110)

Keywords: Graphene, MEMS/NEMS, mechanical switch, lubrication, reliability

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

Abstract Micro/nano-electromechanical (M/NEM) switches have received significant attention as promising switching devices for a wide range of applications such as computing, radiofrequency communication, and power gating devices. However, M/NEM switches still suffer from unacceptably low reliability because of irreversible degradation at the contacting interfaces, hindering adoption in practical applications and further development. Here, we evaluate and verify graphene as a contact material for reliability-enhanced M/NEM switching devices. Atomic force microscopy experiments and quantum-mechanics calculations reveal that energy-efficient mechanical contact–separation characteristics are achieved when a few layers of graphene are used as a contact material on a nickel-surface, reducing the energy dissipation by 96.6 % relative to that of a bare nickel surface. Importantly, graphene displays almost elastic contact–separation, indicating that little atomic-scale wear, including plastic deformation, fracture, and atomic attrition, is generated. We also develop a feasible fabrication method to demonstrate a MEM switch, which has high-quality graphene as the contact material, and verify that the devices with graphene show mechanically stable and elastic-like contact properties, consistent with our nanoscale contact experiment. The graphene coating extends the switch lifetime >103 times under hot-switching conditions.


Page 2 of 31

Page 3 of 31 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


1. Introduction Micro/nano-electromechanical (M/NEM) contact devices have received considerable attention as desirable switching devices for a wide range of applications such as zero-leakage computing, reconfigurable radio-frequency, and energy-efficient power gating devices.

1-5

Owing to controlled mechanical contact operation, ideal switching characteristics in terms of quasi-zero leakage current, abrupt on/off, and low power consumption have been achieved. 6-7 However, M/NEM switching devices still suffer from unacceptably low reliability, hindering adoption in practical applications and further development. 8-10 Various solid contact materials have been investigated to improve the reliability, including silicon, metals and their alloys, diamond, carbon nanotubes (CNTs), and amorphous carbon.

10-17

Regardless of the contact

material, however, these solid-to-solid mechanical contacts suffer from various failure mechanisms such as stiction by adhesion, electrical short due to welding and melting, and inevitable physical degradation by mechanical wear and abrasion.

8,10,18-21

The low reliability

and durability of contacts are thus major challenges for micro/nano-mechanical switching devices. Graphene, a sheet of carbon atoms tightly packed into a two-dimensional honeycomb lattice, has remarkable mechanical, thermal, and electrical properties that are desirable for reliable contact materials, including a high Young`s modulus and mechanical strength, high thermal conductivity and electrical mobility, and high wear- and fracture-resistance.

22-26

In

addition, recent advances in chemical vapour deposition (CVD) have made it possible to produce high quality graphene with finely controlled morphology and structure, and graphene thereby has been applied to a variety of micro and nano-devices.

27-29

However, there is very

few research to evaluate graphene as the contact material of mechanical contact devices or used in such devices. Although researches on graphene as a coating and structure materials have been done for atomic force microscopy (AFM) and electromechanical switching 3 ACS Paragon Plus Environment


applications, 30-34 and they reported enhanced performances such as reliability or significantly reduced operating voltage, they still lack of sufficient understanding about the nanoscale contact phenomena of graphene. One of the major reasons for the lack of research on graphene contact material switches is the difficulty in analysing the mechanical contact of M/NEM switches. In general, the actual operation of M/NEM switching devices consists of ‘contact-separation’ of a large number of atomic- and nano-asperities at the contacting interface.

25,35

These contacting spots are too

tiny to perform a specific analysis of nanoscale contact phenomena such as wear. The difficulty of manufacturing is another reason. As aforementioned, switches using graphene as a structure can be fabricated, but these devices showed poor reliability by structural breakage due to the thin beam thickness.

36,37

The coating of graphene contact material on the contact

part of the switch-structure is expected to be a reliable switching device; however, it is challenging to use high-temperature graphene synthesis processes (>1,000 °C) to fabricate graphene coated three-dimensional suspended micromechanical devices because of high thermal stress. Here, we evaluate graphene as the contact material of a micromechanical contact device, and verify that the lifetime of the MEM switch is dramatically improved by the graphene contact material. To understand the atomic level contact characteristics of graphene, we develop a method for precisely investigating nanoscale contact degradation, such as wear, by combining atomic force microscopy (AFM) force-distance (F-d) experiments, and density function theory (DFT). The results reveal that elastic-like contact properties can be achieved by coating nickel with multilayer graphene. We also develop a reliable method for manufacturing graphene-coated nickel-based micromechanical switching devices. The developed self-aligning process provides high-quality graphene on surfaces that mechanically contact the switch without any structural deformations. Finally, we observed markedly 4 ACS Paragon Plus Environment

Page 4 of 31

Page 5 of 31 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


improved lifetime of graphene as a contact material for micromechanical devices under hot switching conditions compared to bare metal switches.

2. Results and Discussion 2.1. Nanoscale contact-separation experiment using AFM Figure 1 shows the nanoscale contact–separation experiment performed on graphene using the AFM F-d method. A major advantage of using AFM in contact–separation studies is that the physical interaction between the nanoscale tip and sample can be measured and analysed through precisely controlled approach and retraction of the tip-mounted cantilever (Figure 1a). 38-41

Figure 1b depicts the nanoscale contact–separation process involving the approach and

retraction of the tip to and from the sample. In the approach process (left panel in Figure 1b), the cantilever, which is initially apart from the sample, moves downward (I. Initial). If the tip– sample distance is close enough, the markedly increased attraction force between the tip and sample pulls them together to form a mechanical contact (II. Contact). Then, the continuous downward movement of the cantilever produces a stable contact (III. Indentation). In the retraction process, the cantilever moves upward and tries to detach the tip from the sample (right panel in Figure 1b). However, despite the upward movement of the cantilever, the tip– sample contact is retained because of the adhesion between them. Eventually, the tip and sample are separated by the increasing cantilever restoring force (IV. Separation). Finally, the cantilever and sample return to the initial state (V. Final). During this contact–separation process, the cantilever reads the tip–sample interaction through the forces imposed on the cantilever (FCanti); negative and positive forces indicate the adhesive and repulsive interactions, respectively, between the tip and sample. Thus, nanoscale contact phenomena can be understood and predicted by analysing measured FCanti changes. To measure the contact–separation characteristics of graphene, we prepared a 1×1 cm piece 5 ACS Paragon Plus Environment


Page 6 of 31

of nickel (Ni), and synthesized high-quality graphene (Gr) on it using CVD. Multilayer graphene was synthesized on nickel because it was expected to show the greatest durability and intrinsic contact properties of graphene samples, hindering the high surface energy of nickel (Supporting Information Figure S1). The contact–separation experiment was performed using a silicon AFM tip (R ≈ 30 nm) coated with gold (Au), which can show clear wear phenomena because of its high ductility. As a control, we also performed the same experiment on bare nickel. The 5 µm-thick nickel sample was prepared using electroplating deposition on a silicon wafer and thermal annealing (3 min at 900 °C in Ar ambient) under vacuum ambient was performed for the surface roughness control. Then, it is cleaned using 50:1 diluted hydrogen fluoride solution for 30 sec (See Supporting Information Figure S2). The morphology and Raman spectrum of the fabricated nickel and graphene is illustrated in the insets of Figure 1c and d, respectively. Other specific conditions for sample preparation and experiments are provided in Supporting Information Figure S3 and Experimental Section, respectively. Figure 1c and d show the measured F-d curves for nickel and graphene-coated nickel, respectively. Obvious contact and separation phenomena are observed for both samples, but their FCanti variations by the tip-sample distance (d) are quite different. Specifically, a large Fadhesion (~22 nN) was measured on the bare nickel surface in the retraction process, whereas a much lower Fadhesion of ~7 nN was measured on the graphene-coated one. We explain this weaker adhesion by the weaker van der Waals (vdW) interaction between the graphene surface and tip compared with that between the nickel surface and tip. The graphene coating forms energetically stable pi-bonds on the nickel surface, hiding the nickel surface so that it is only weakly attracted to the tip in the weak vdW regime;

42

the decrease of surface energy

caused by graphene was also observed in the approach process as a smooth change of the negative Fcanti, indicated weak vdW interactions. 38 6 ACS Paragon Plus Environment

Page 7 of 31 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


In each curve where the tip and sample are approached and retracted, the graphene not only shows a low adhesive characteristic, but also a remarkably different hysteretic behaviour is seen by comparing the two curves. After the adhesion point during retraction (Fadhesion in Figure 1b), FCanti measured on nickel decreased gradually throughout the continuous retraction, while a sudden FCanti change was observed during the approach, generating large hysteresis. Conversely, smooth FCanti changes were found for both approach and retraction processes on graphene, resulting in little hysteresis. To analyse this hysteresis behaviour, we extracted the dissipation energy (Edissipation) in the tip–sample adhesive range, which indicates the energy by tip–sample separation.

43

Edissipation generated by the tip–sample adhesive

interaction was obtained by subtraction of Econtact from Eseparation in the FCanti-d curves, as shown as the shaded areas in Figure 1b. Figure 1e presents the obtained Econtact, Eseparation and Edissipation. During approach and retraction, the reduced mean Econtact and Eseparation were 59.74 ± 13.24 and 84.78 ± 17.68 eV, respectively, for the graphene-coated sample, and 224.12 ± 75.10 and 960.74 ± 174.30 eV, respectively, for bare nickel. As a result, Edissipation of the graphene-coated sample (25.04 ± 13.31 eV) is only 3.4% of that measured for bare nickel (736.62 ± 174.30 eV). The value obtained for the graphene-coated sample is sufficiently close to zero that it can be considered that a drastically wear-resistant and mechanically durable contact formed. These favourable nanoscale contact–separation characteristics of the graphene-coating surface are due to the graphene reducing irreversible wear such as plastic deformation, fracture and atomic attrition at the contacting surface.

2.2. Atomic-scale contact-separation analysis using DFT calculations To understand the observed nanoscale contact–separation phenomena, we performed firstprinciples DFT calculations for a gold tip on bare nickel and graphene-covered nickel surfaces (Figure 2). Such first-principles DFT calculations allows reliable and useful analyse of the 7 ACS Paragon Plus Environment


atomic-scale physical/chemical interaction between a tip and sample based on the quantum mechanical approach.

44, 45

To mimic the experimental conditions accurately, we used a gold

(111) nanowire as the tip and two types of sample surfaces: flat, bare nickel (111) and multilayer graphene. Then, we calculated total energies and atomic forces of the tip–sample systems while varying the tip–sample distance z. It should be noted that we used only three layers of graphene without a nickel substrate to model the graphene-covered nickel substrate, because the effect of the underlying nickel substrate is negligible (Supporting Information Figure S1). Figure 2 shows DFT-optimized atomic structures during contact–separation on nickel and graphene substrates, and their respective F-z curves. The calculated atomic force of the gold tip, Ftip, was determined by summing all the atomic forces of gold atoms. In the approach process, the strong adhesion forces cause a sudden atomic contact to form between the gold tip and nickel surface when their closest distance is smaller than 5.3 Å. The atomic contact is accompanied with elongation of the gold tip near the contact because of the high ductility of the gold wire. 40 As z decreases, more gold atoms migrate to the nickel surface (Approach in Figure 2a). This gold migration energetically stabilizes the nickel surface because the surface energy of nickel is higher than that of gold. 46 Gold atoms are stacked on the nickel surface and finally fully cover the nickel surface. During the rearrangement of gold atoms, stress compression and relaxation are repeatedly generated in the tip, as indicated by the severe Ftip fluctuation in the left panel of Figure 2b. 40 In the retraction process, we observe gold necking (formation of an atomic gold wire) because of the strong adhesive interaction between the gold tip and nickel surface. As the tip is retracted, the connecting “neck” is maintained as long as possible through continual rearrangement of gold atoms (Retraction in Figure 2a). When the tip position is 4 Å above its original position, the gold atomic wire finally fractures. During the retraction process, Ftip gradually decreases because of the presence of the 8 ACS Paragon Plus Environment

Page 8 of 31

Page 9 of 31 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


connecting “neck” (right panel in Figure 2b). In contrast, almost no atomic-scale wear of the gold tip is generated during the contact– separation process on the graphene surface. As the gold tip approaches the graphene surface, negligible adhesive contact is generated with no rearrangement of gold atoms, and then graphene is slightly indented in the compressive region (Approach in Figure 2c). This is because the surface energy of graphene is extremely low compared with those of gold and nickel. Similarly, no wear is observed during the retraction process of the gold–graphene system. As the gold tip returns to its original position, the compressed graphene and gold tip almost recover to their initial states without any notable atomic changes. As a result, Ftip changes slightly and smoothly with negligible hysteresis during the contact–separation cycle (Figure 2d). Finally, we understood the results of nanoscale contact-separation experiments through the simulation results. Whereas the sudden contact in the tip approach was apparently appeared on the bare nickel in the experiment and simulation at d=2.5 nm and △z=5.3 Å, respectively, it was not occurred on the graphene coated nickel. The significantly reduced adhesion force (Fadhesion) in the retraction was also obtained on the graphene in the simulation and the experiment. In particular, the sharp feature after the Fadhesion in F-d curve, originated from the necking (elongation) of Au tip,

40

were also reproduced experimentally and theoretically on

the nickel. Then, the hysteresis-less contact-separation characteristics were confirmed on the graphene, while the large hysteresis is generated on the bare nickel; the calculated energy dissipation from the nickel and graphene surface were 17.2 and 1.8 eV, respectively, and the drastically reduced experimental energy dissipation was reproduced remarkably well in the simulation. The calculated F-d curves, adhesion energy and energy dissipation values exhibit similar trends to the experimental results, confirming that the wear resistance of graphene is observed in real contact–separation situations. 9 ACS Paragon Plus Environment


2.3. Graphene coated micromechanical switch To extend and confirm the high durability of graphene observed above to long-term reliability of practical applications, which is difficult to verify using the AFM measurement due to its drift, we developed graphene-coated micromechanical switches. The switch fabrication method involved three steps (left panel in Figure 3a). First, a nickel structure, which also acted as a seed for graphene synthesis, was fabricated on a bottom insulator (silicon dioxide). Graphene was then synthesized by CVD. Because CVD was conducted before structure release, the bottom insulator fastened all the structures during the hightemperature CVD process, so graphene formed on the nickel surface without any deformation. Finally, the structure was released by isotropic wet etching of the bottom insulator, which completely removed thin regions and left the actuating part on the bottom-layer as a ‘selfaligned post’. The fabrication process and conditions are shown in Experimental Section and Supporting Information Figure S4 in detail. Each switch had a three-terminal configuration of gate (G), drain (D) and source (S) (right panel in Figure 3a). The source beam was laterally pulled toward the drain, eventually making a contact through the electrostatic force generated between the source beam and biased gate (See movie in Supporting Information). Scanning electron microscopy (SEM) images of a graphene-coated micromechanical device are displayed in Figure 3b, and the presence of fewlayer graphene on the contacting sidewall (cross section along the red dashed line in the inset of Figure 3c) was confirmed by transmission electron microscopy (TEM) (Figure 3c). The current–voltage (I-V) characteristics of the switch were measured to verify the mechanical operating stability of the switch (Icompliance between D-S = 10 nA). Figure 3d reveals that the selected device shows obvious mechanical contact with ideal switching performance (IOn/IOff >106). Furthermore, the on-voltage changes of devices with different design (different length of beam of 800, 1000, and 1200 µm, but same beam thickness of 5 µm, width of 5 µm, and 10 ACS Paragon Plus Environment

Page 10 of 31

Page 11 of 31 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


gap between gate and source h of 4 µm) highly corresponded with the theoretical lumped model, which is calculated using 40 MPa residual stress (See Supporting Information Figure S5),

47

as illustrated in Figure 3e. The transient switching responses of the device were also

monitored (Figure 3f); mechanical contact was confirmed by a typical second-order underdamped response. The electromechanical contact and reliability characteristics of the switches were investigated. Specific mechanical contact phenomena like contact area change induced by deformation can be understood by evaluating the I-V curves without current compliance between drain and source.

8

For fair comparison, a bare Ni-contacting device was also

fabricated and tested. In spite that MEMS switches normally operate in a nitrogen ambient at an atmospheric pressure, this measurement was performed under vacuum conditions (~3 mTorr) to minimize the effects of chemical oxidation and degradation on the bare-nickel device.

48

Figure 4a and b present the I-V results obtained for the bare nickel- and graphene-

contacting devices, respectively. From the measured I-V curves with the VDS=0.05 V, the initial on-resistances of both devices were confirmed. When the nickel device became the onstate at the VG=96 V, its resistance was about 250 Ω. In case of the graphene coated device, the higher resistance of 4 kΩ was measured when the device was turned on at the VG=85 V. The higher resistance of the graphene coated device is understood with the anisotropic electrical resistivity in out-of-plane direction of graphene.

49

Graphene is well known as the

excellent in-plane electrical mobility, but when made into “few layer” graphene, it shows 1000 to 10000 times higher resistivity along the out-of-plane direction due to van der Waals bonding between graphene sheets. Using the out-of-plane resistivity of graphene and our fabricated device design, we calculated the electro-mechanical on-resistance of the device based on a conventional electro-mechanical contact modeling.

47

The calculated resistance

was 3.23 kΩ, which is highly corresponded with the experimental value. From this result, we 11 ACS Paragon Plus Environment


confirmed that the fabricated device apparently operated by mechanical contact of graphene, acting as the electrical ohmic resistor at the contacting interfaces. Owing to the graphene contact material, the device shows a remarkable reliability in the contact characteristic. During on/off cycling operation, the hysteresis observed in the I-V curve of the switch with graphene was much smaller than that of the switch with bare Ni. This result is attributed to the weaker adhesion of graphene than bare nickel in the contacting switches, consistent with the AFM F-d results (Figure 1 and 2). Most importantly, the switch with graphene also displays exceptional contact properties. The right panel of Figure 4b shows the magnified on-state I-V curve of the switch with graphene. By increasing the gate voltage, a strong compressive force is generated between the source beam and the drain to widen the contacting area. [50,51] Generally, if the contacting area is changed once, it is difficult to recover to the original surface because of the strong adhesive interaction in this area. Thus, the mechanical contact is maintained despite retraction of the contacting force; this is observed as the static on current plotted against the gate-voltage retraction (right panel in Figure 4a). In contrast, the switch with graphene shows immediate on-current changes as the gate voltage is modulated, indicating the surface recovers between contact operations. This result is also explained by the nanoscale contact–separation characteristics of graphene; although graphene is compressed and the contacting area is widened by the additional contacting force, the out-of-plane elasticity and durability of graphene allow recovery of the original surface during force retraction without plastic deformation or fracture. It should be noted that although the fabricated devices were designed with a microscale-contacting surface, the real contact occurs through the nanoscale contacting spots. The elastic and durable contact characteristics of graphene lead to the switch displaying highly reliable electromechanical operations. While considerable on-current fluctuation was measured for the switch with bare nickel (Figure 4c), the graphene device showed almost the 12 ACS Paragon Plus Environment

Page 12 of 31

Page 13 of 31 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


same switching characteristics during 15 operations (Figure 4e). Finally, we tested the long-term reliability of the devices under hot-switching conditions (live voltage applied between the source beam and drain during switching) in 3 mTorr vacuum ambient. An electrical current of 1 µA and voltage of 1 V were applied to the drain– source during the cycling operations (1 kHz) and the device I-V characteristics and onresistance were measured periodically using a semiconductor parameter-analyzer (4156C, Agilent technologies). It is worthy to note that the graphene contact device did not show I-V characteristic degradations by the operation of VDS=1 V (Supporting Information Figure S6), indicating that the graphene contact material still acts as the ohmic-resistor without significant electrical breakdown.

16

The switch with graphene showed reliable electromechanical

operation (IOn/Off >108) even after 600,000 cycles, whereas the switch without graphene displayed severe contact degradation after 1,000 cycles, as illustrated in Figure 4d and f, respectively. The switch with graphene was also compared with a reference 100 nm-thick gold-contacting switch, which is known to form a reliable electromechanical contact.22, 39 The initial on-resistance of the Au coated device was about 40 Ω and the Au coated device also showed significant I-V hysteresis during on/off cycle, indicating large amount of mechanical adhesion similar to the Ni device. The detailed I-V characteristics of the Au-coated device are provided in Supporting Information Figure S7. Figure 5 compares the on-resistance changes of these switches. Compared with the bare Ni- and Au-contacting switches under the same hot-switching conditions, the switch with graphene showed extremely high reliability (>1,000,000 cycles), which is more than 1,000-fold improvement than bare-Ni switch. In particular, the graphene coated devices showed higher reliability than the Au coated device; the resistance of the Au coated device became more than 10 kΩ after tens of thousands cycles, exceeding the resistance of graphene device. We understand the high reliability of the graphene coated device with the superior wear-resistant characteristics of graphene, originated 13 ACS Paragon Plus Environment


from mechanical durability and low adhesion, and it is well explained with the performed nanoscale contact-separation analysis. Despite their impressive long-term stability, the switch with graphene did show a sudden on-resistance increase after ~3 × 106 cycles. This unexpected degradation is attributed to a chemical oxidation of Ni at the grain boundaries of graphene induced by the inevitable joule heating;

52

the high temperature at the contacting

surface associated with the intrinsically high electrical resistance of graphene device was induced by joule heating and the accumulated heat stress finally generated the device failure. In this work, our graphene device showed high resistance so that relatively lower reliability than the state-of-the art results,

53, 54

however, there is still high potential to lower the

resistance with recently reported technologies such as adjusting domain-size of graphene and device-packaging to suppress the Ni oxidation. 55

3. Conclusion In summary, we have developed a method that is able to rigorously analyse the mechanical vertical contact-separation characteristics of graphene and offer feasible fabrication method applicable to micro/nano devices on a wafer scale process. Theoretical analysis and AFM experiments revealed that high wear resistance including nearly elastic deformation and low atomic attrition could be achieved on graphene. These nanoscale properties of graphene increased the long-term reliability of graphene-contacting micro-switches more than 1,000fold compared with that of switches without graphene. This approach will significantly expand the use of graphene in numerous applications such as nano-scale data storage, nanofabrication and N/MEMS in which the utilization of graphene is very difficult.


Page 14 of 31

Page 15 of 31 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


4. Experimental Section AFM F-D Experiment: Nanoscale vertical contact tests were performed on a conventional AFM system (Park, XE-100) in air ambient (temperature of 25 °C and humidity of 23 %). A commercial AFM tip (Tap 300G, Budget Sensors, resonant frequency: ~300 kHz, force constant: ~40 N/m) was used for the nanoscale contact test. To analyse contact phenomena such as nanoscale necking and fracture, the silicon tip was coated with 70 nm-thickness of gold, which has relatively low Young’s modulus (~79 GPa) and high ductility. Vertical contact and separation processes were executed using a piezo-electrically moved tip-mounted stage (5 nm/s). Four units of AFM tips were used to measure each of the twenty nickel and graphene samples, and each tip was used for ten times.

DFT Simulation: All DFT calculations were performed with the Perdew-Burke-Ernzerhof (PBE) exchange-correlation functional [56] as implemented in the Vienna ab initio simulation package (VASP) [57,58]. We employed a plane-wave basis set with a kinetic cut-off energy of 400 eV and projector-augmented wave (PAW) potentials [59]. van der Waals interactions are correctly included by using Grimme dispersion corrected DFT-D3 method [60]. We performed the Brillouin zone integration using the Γ point only. All atomic forces were fully relaxed to less than 0.025 eV/Å. For the F-D curve simulation, we used periodic slab models, consisting of an Au (111) nanowire model and a three-layer (9 × 9) nickel (111) substrate or three-layer graphene. The gold nanowire model has nine atomic layers along the [111] direction with diameter of 1.4 nm. The bottom layer of the Ni and graphene substrates is fixed during simulation. Then, the top two layers of the gold nanowire are fixed in a single-shot F-d curve simulation, and gradually approaching by 0.5 nm to the substrate. Except the fixed layers, all the atoms in the gold nanowire and substrates are fully relaxed during the quasistatic F-d curve simulation. The atomic force of the Au tip was calculated by summing up the vertical forces of the fixed atoms on the gold nanowire. 15 ACS Paragon Plus Environment


Device Fabrication process: The detailed schematics of the device fabrication can be seen in Supporting Information Figure S4. The fabrication started with a 4-inch silicon wafer which has a 1 µm-thick thermally grown silicon di-oxide layer (SiO2) (Figure S4. I). Additional plasma enhanced chemical vapor deposited (PECVD) silicon di-oxide (SiO2) layer is deposited by 3 µm (Figure S4. II). Then, 5 µm-thick nickel structure is electroplated (pulsereverse plating mode at the current of 200 mA and temperature of 40 °C for 150 min, deposition rate=0.5 nm/s) within patterned photo-resist dam (8 µm-thick AZ-9260) on a 10 nm-thick of chrome (Cr) and 100 nm-thick of gold (Au) deposited wafer (Figure S4. III-IV.). After the electroplating step, unnecessary photo-resist (PR), gold (Au), and chrome (Cr) are removed using wet chemical solution (PR: acetone, Au: Au etchant, CNCTECH Co., Cr: CE905N, Transene Company, Inc.). The CVD graphene synthesis is then performed. The sample was first annealed at 800 °C in a 10 % H2 diluted in an Ar ambient (2000 sccm) for 8 min to remove native oxide on the nickel surface. This was followed by annealing at 900 °C in a 10 % CH4 diluted in an Ar ambient (100 sccm) for 3 min at for carbon dissolution into nickel. Then the sample was naturally cooled down (Figure S4. VI``). Finally, the release process, generating the suspended 3D switch structure by the self-aligned post is performed by immersion in a buffered oxide etchant (BOE) solution for 40 min (VII``). The Ni- and Aucontact devices are demonstrated using the same process, however, the thermal treatment at 900 °C is performed only in Ar ambient without methane (CH4) gas to avoid the unexpected graphene formation. Finally, the additional the Au electro-less plating (50 °C for 30 min, deposition rate=3.5 nm/min ) is performed to demonstrate 100 nm-thick Au-coating the Aucontact device. (Figure S4. VII`).


Page 16 of 31

Page 17 of 31 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


Acknowledgements This research was supported by the Center for Integrated Smart Sensors funded by the Ministry of Science, ICT & Future Planning as “Global Frontier Project” (No. CISS2012054187). J.-H.K. and Y.-H.K were supported by the National Research Foundation of Korea (2015R1A2A2A05027766) and Science Research Center (2016R1A5A1008184) programs.

Supporting Information Figure S1. Binding energy calculations of gold and nickel/graphene. (PDF) Figure S2. EDS (energy dispersive X-ray spectroscope) analysis on nickel sample. (PDF) Figure S3. Experimental details of AFM F-d measurement and graphene. (PDF) Figure S4. Schematic illustration of device fabrication. (PDF) Figure S5. Device on-voltage analysis considering residual stress. (PDF) Figure S6. I-V curve comparison of graphene devices, operated with different VDS. (PDF) Figure S7. I–V characteristics of a gold-contacting device. (PDF) Movie. Real time movie clip of mechanically operating graphene contact microswitch. (AVI)

Corresponding Author *E-mail: [email protected] and [email protected]

Author Contribution The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. 17 ACS Paragon Plus Environment


References (1)

Rueckes, T. Carbon Nanotube-Based Nonvolatile Random Access Memory for

Molecular Computing. Science 2000, 289, 94–97. (2)

Lu, W; Lieber, C. M.; Nanoelectronics from the Bottom up Electronics. Nat. Mater.

2007, 6, 841-850. (3)

Gammel, P.; Fischer, G. Bouchaud, J. RF MEMS and NEMS Technology, Devices,

and Applications. Bell Syst. Tech. J. 2005, 10, 29–59. (4)

Kim, H.; Shin, D. H.; McAllister, K.; Seo, M.; Lee, S.; Kang, I.-S.; Park, B. H.;

Campbell, E. E. B.; Lee, S. W. Accurate and Precise Determination of Mechanical Properties of Silicon Nitride Beam Nanoelectromechanical Devices. ACS Appl. Mater. Interfaces 2017, 9, 7282-7287. (5)

Yoon, Y.-H.; Song, Y.-H.; Ko, S.-D.; Han, C.-H.; Yun, G.-S.; Seo, M.-H.; Yoon, J.-B.

Highly Reliable MEMS Relay with Two-Step Spring System and Heat Sink Insulator for Power Applications. J. Microelectromechanical Syst. 2016, 25, 217-226. (6)

Liao, M.; Hishita, S.; Watanabe, E.; Koizumi, S.; Koide, Y. Suspended Single-Crystal

Diamond Nanowires for High-Performance Nanoelectromechanical Switches. Adv. Mater. 2010, 22, 5393–5397. (7)

Lee, J. O.; Song, Y.-H.; Kim, M.-W.; Kang, M.-H.; Oh, J.-S.; Yang, H.-H.; Yoon, J.-B.

A Sub-1-Volt Nanoelectromechanical Switching Device. Nat. Nanotechnol. 2013, 8, 36–40. (8)

Song, Y. H.; Kim, M. W.; Seo, M. H.; Yoon, J. B. A Complementary Dual-Contact

MEMS Switch Using a “Zipping” Technique. J. Microelectromechanical Syst. 2014, 23, 710– 718. (9)

Loh, O. Y.; Espinosa, H. D. Nanoelectromechanical Contact Switches. Nat.

Nanotechnol. 2012, 7, 283–295.


Page 18 of 31

Page 19 of 31 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


(10)

Choi, J.; Lee, J. I.; Eun, Y.; Kim, M. O.; Kim, J. Aligned Carbon Nanotube Arrays for

Degradation-Resistant, Intimate Contact in Micromechanical Devices. Adv. Mater. 2011, 23, 2231–2236. (11)

Penkov, O. V.; Devizenko, A. Y.; Khadem, M.; Zubarev, E. N.; Kondratenko, V. V.;

Kim, D. E. Toward Zero Micro/Macro-Scale Wear Using Periodic Nano-Layered Coatings. ACS Appl. Mater. Interfaces 2015, 7, 18136–18144. (12)

Toler, B. F.; Coutu, R. A.; McBride, J. W. A Review of Micro-Contact Physics for

Microelectromechanical Systems (MEMS) Metal Contact Switches. J. Micromech. Microeng. 2013, 23, 103001. (13)

Song, K.; Chen, D.; Polak, R.; Rubner, M. F.; Cohen, R. E.; Askar, K. A. Enhanced

Wear Resistance of Transparent Epoxy Composite Coatings with Vertically Aligned Halloysite Nanotubes. ACS Appl. Mater. Interfaces 2016, 8, 35552–35564. (14)

Caron, A.; Louzguine-Luzguin, D. V.; Bennewitz, R. Structure vs Chemistry: Friction

and Wear of Pt-Based Metallic Surfaces. ACS Appl. Mater. Interfaces 2013, 5, 11341–11347. (15)

Loh, O.; Wei, X.; Sullivan, J.; Ocola, L. E.; Divan, R.; Espinosa, H. D. Carbon-Carbon

Contacts for Robust Nanoelectromechanical Switches. Adv. Mater. 2012, 24, 2463–2468. (16)

Grogg, D.; Ayala, C.; Drechsler, U.; Sebastian, A.; Koelmans, W.; Bleiker, S.;

Fernandez-Bolanos, M.; Hagleitner C. Amorphous Carbon Active Contact Layer for Reliable Nanoelectromechanical switches. 2014 IEEE 27th international conference on micro electro mechanical systems (MEMS). 2014, 143–146. (17)

Ayala, C. L.; Grogg, D.; Bazigos, A.; Bleiker, S.; Fernandez-Bolanos, M.; Niklaus, F.;

Hagleitner C. Nanoelectromechanical Digital Logic Circuits Using Curved Cantilever Switches with Amorphous-Carbon-Coated Contacts. Solid State Electron. 2015, 113, 157-166. (18)

Hennessy, R. P.; Basu, A.; Adams, G. G.; McGruer, N. E. Hot-Switched Lifetime and

Damage Characteristics of MEMS Switch Contacts. J. Micromech. Microeng. 2013, 23, 055003. 19 ACS Paragon Plus Environment


(19)

Van Spengen, W. M. MEMS Reliability from a Failure Mechanisms Perspective.

Microelectron. Reliab. 2003, 43, 1049–1060. (20)

Beerschwinger, U.; Albrecht, T.; Mathieson, D.; Reuben, R. L.; Yang, S. J.;

Taghizadeh, M. Wear at Microscopic Scales and Light Loads for MEMS Applications. Wear 1995, 181–183, 426–435. (21) Li, Y.; Zhang, L.; Gong, H. Dynamic thermal degradation studies on amorphous carbon thin films. J. Therm. Anal. Calorim. 2005, 79, 677−683. (22)

Lee, C.; Wei, X.; Kysar, J. W.; Hone, J. Measurement of the Elastic Properties and

Intrinsic Strength of Monolayer Graphene. Science 2008, 321, 385–388 (23)

Bolotin, K. I.; Sikes, K. J.; Jiang, Z.; Klima, M.; Fudenberg, G.; Hone, J.; Kim, P.;

Stormer, H. L. Ultrahigh Electron Mobility in Suspended Graphene. Solid State Commun. 2008, 146, 351–355. (24)

Tripathi, K.; Gyawali, G.; Lee, S. W. Graphene Coating via Chemical Vapor

Deposition for Improving Friction and Wear of Gray Cast Iron at Interfaces. ACS Appl. Mater. Interfaces 2017, 9, 32336-32351. (25)

Li, S.; Li, Q.; Carpick, R. W.; Gumbsch, P.; Liu, X. Z.; Ding, X.; Sun, J.; Li, J. The

Evolving Quality of Frictional Contact with Graphene. Nature 2016, 539, 541–545. (26)

Lee, W.; Kihm, K. D.; Kim, H. G.; Shin, S.; Lee, C.; Park, J. S.; Cheon, S.; Kwon, O.

M.; Lim, G.; Lee, W. In-Plane Thermal Conductivity of Polycrystalline Chemical Vapor Deposition Graphene with Controlled Grain Sizes. Nano Lett. 2017, 17, 2361–2366. (27)

Kim, K. S.; Zhao, Y.; Jang, H.; Lee, S. Y.; Kim, J. M.; Kim, K. S.; Ahn, J.-H.; Kim,

P.; Choi, J.-Y.; Hong, B. H. Large-Scale Pattern Growth of Graphene Films for Stretchable Transparent Electrodes. Nature 2009, 457, 706–710. (28)

Li, X.; Cai, W.; An, J.; Kim, S.; Nah, J.; Yang, D.; Piner, R.; Velamakanni, A.; Jung,

I.; Tutuc, E.; Banerjee, S. K.; Colombo, L.; Ruoff, R. S. Large-Area Synthesis of HighQuality and Uniform Graphene Films on Copper Foils. Science 2009, 324, 1312–1314. 20 ACS Paragon Plus Environment

Page 20 of 31

Page 21 of 31 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


(29)

Yang, Y.; Yang, X.; Zou, X.; Wu, S.; Wan, D.; Cao, A.; Liao, L.; Yuan, Q.; Duan, X.

Ultrafine Graphene Nanomesh with Large On/Off Ratio for High-Performance Flexible Biosensors. Adv. Funct. Mater. 2017, 27, 1–9. (30)

Kim, S. M.; Song, E. B.; Lee, S.; Seo, S.; Seo, D. H.; Hwang, Y.; Candler, R.; Wang,

K. L. Suspended Few-Layer Graphene Beam Electromechanical Switch with Abrupt on-off Characteristics and Minimal Leakage Current. Appl. Phys. Lett. 2011, 99, 023103. (31)

Shi, Z.; Lu, H.; Zhang, L.; Yang, R.; Wang, Y.; Liu, D.; Guo, H.; Shi, D.; Gao, H.;

Wang, E.; Zhang, G. Studies of Graphene-Based Nanoelectromechanical Switches. Nano Res. 2012, 5, 82–87. (32)

Liu, X.; Suk, J. W.; Boddeti, N. G.; Cantley, L.; Wang, L.; Gray, J. M.; Hall, H. J.;

Bright, V. M.; Rogers, C. T.; Dunn, M. L.; Ruoff, R. S.; Bunch, J. S. Large Arrays and Properties of 3-Terminal Graphene Nanoelectromechanical Switches. Adv. Mater. 2014, 26, 1571–1576. (33)

Zang, X.; Zhou, Q.; Chang, J.; Liu, Y.; Lin, L. Graphene and Carbon Nanotube (CNT)

in MEMS/NEMS Applications. Microelectron. Eng. 2015, 132, 192–206. (34)

Martin-Olmos, C.; Rasool, H. I.; Weiller, B. H.; Gimzewski J. K. Graphene MEMS:

AFM Probe Performance Improvement. ACS Nano, 2013, 7, 4164-4170. (35)

Song, Y. H.; Ko, S. D.; Yoon, J. B. A Highly Reliable Two-Axis MEMS Relay

Demonstrating a Novel Contact Refresh Method. J. Microelectromechanical Syst. 2015, 24, 1495–1502. (36)

Milaninia, K. M.; Baldo, M. A.; Reina, A.; Kong, J. All Graphene Electromechanical

Switch Fabricated by Chemical Vapor Deposition. Appl. Phys. Lett. 2009, 95, 183105. (37)

Li, P.; You, Z.; Cui, T. Graphene Cantilever Beams for Nano Switches. Appl. Phys.

Lett. 2012, 101, 093111. (38)

Cappella, B.; Dietler, G. Force-Distance Curves by Atomic Force Microscopy. Surf.

Sci. Rep. 1999, 34, 1–104. 21 ACS Paragon Plus Environment


(39)

Bhushan, B.; Israelachvill, J. N.; Landman, U. Nanotribology: Friction, Wear and

Lubrication at the Atomic Scale. Nature 1995, 374, 607–616. (40)

Agrait, N.; Rubio, G.; Vieira, S. Plastic Deformation of Nanometer-Scale Gold

Connective Necks. Phys. Rev. Lett. 1995, 74, 3995–3999. (41)

Dupres, V.; Menozzi, F. D.; Locht, C.; Clare, B. H.; Abbott, N. L.; Cuenot, S.;

Bompard, C.; Raze, D.; Dufrêne, Y. F. Nanoscale Mapping and Functional Analysis of Individual Adhesins on Living Bacteria. Nat. Methods 2005, 2, 515–520. (42)

Kim, K.; Lee, H.; Lee, C.; Lee, S.; Jang, H. Chemical Vapor Deposition-Grown

Graphene: The Thinnest Solid Lubricant. ACS Nano 2011, 5, 5107–5114. (43)

Pfreundschuh, M.; Martinez-Martin, D.; Mulvihill, E.; Wegmann, S.; Muller, D. J.

Multiparametric High-Resolution Imaging of Native Proteins by Force-Distance Curve-Based AFM. Nat. Protoc. 2014, 9, 1113–1130. (44)

Parga, D.; Miranda, R.; Martı, F.; Alcamı, M.; Glatzel, T.; Meyer, E. Elastic Response

of Graphene Nanodomes. ACS Nano 2013, 7, 2927-2934. (45)

Choi, H.; Ko, J.; Kim, Y.; Jeong, S. Steric Hindrance Driven Shape Transition in PbS

Quantum Dots : Understanding Size-Dependent Stability. J. Am. Chem. Soc. 2013, 135, 52785281. (46)

Tyson, W.; Miller, W. A. Surface Free Energies of Solid Metals Estimation from

Liquid Surface Tension Measurements. Surf. Sci. 1977, 62, 267-276. (47)

Rebeiz, G. M. RF MEMS: Theory, design, and technology; John Wiley & Sons: New

York, 2003. (48)

Margomenos, A.; Katehi, L. P. B. Fabrication and Accelerated Hermeticity Testing of

an on-Wafer Package for RF MEMS. IEEE Trans. Microw. Theory Tech. 2004, 52, 1626– 1636. (49)

Sun, K.; Stroscio, M. A.; Dutta, M. Graphite C-Axis Thermal Conductivity.

Superlattices Microstruct. 2009, 45, 60–64. 22 ACS Paragon Plus Environment

Page 22 of 31

Page 23 of 31 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


(50)

Gregori, G.; Clarke, D. R. Mechanical Creep as a Life-Limiting Factor of Radio

Frequency Microswitches. Appl. Phys. Lett. 2005, 87, 154101. (51)

Gregori, G.; Clarke, D. R. The Interrelation between Adhesion, Contact Creep, and

Roughness on the Life of Gold Contacts in Radio-Frequency Microswitches. J. Appl. Phys. 2006, 100, 094904. (52)

Duong, D. L.; Han, G. H.; Lee, S. M.; Gunes, F.; Kim, E. S.; Kim, S. T.; Kim, H.; Ta,

Q. H.; So, K. P.; Yoon, S. J.; Chae, S. J.; Jo, Y. W.; Park, M. H.; Chae, S. H.; Lim, S. C.; Choi, J. Y.; Lee, Y. H. Probing Graphene Grain Boundaries with Optical Microscopy. Nature 2012, 490, 235–239. (53)

Parsa, R; Lee, W. S; Shavezipur, M.; Provine, J.; Maboudian R.; Mitra, S.; Wong, H. -

S. P.; Howe, R. T. Laterally Actuated Platinum-Coated Polysilicon NEM Relays. J. Microelectromechanical Syst. 2013, 22, 768-778. (54)

Newman, H. S.; Ebel, J. L.; Judy, D.; Maciel, J. Lifetime Measurement on a High-

Reliability RF-MEMS Contact Switch. IEEE Microwave Compon. Lett. 2008, 18, 100-102. (55)

Li, X.; Magnuson, C. W.; Venugopal, A.; An, J.; Suk, J. W.; Han, B.; Borysiak, M.;

Cai, W.; Velamakanni, A.; Zhu, Y.; Fu, L.; Vogel, E. M.; Voelkl, E.; Colombo, L.; Ruoff, R. S. Graphene Films with Large Domain Size by a Two-Step Chemical Vapor Deposition Process. Nano Lett. 2010, 10, 4328–4334. (56)

Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized Gradient Approximation Made

Simple. Phys. Rev. Lett. 1996, 77, 3865–3868. (57)

Kresse, G.; Joubert, D. From Ultrasoft Pseudopotentials to the Projector Augmented-

Wave Method. Phys. Rev. B 1999, 59, 1758–1775. (58)

Kresse, G.; Furthmüller, J. Efficient Iterative Schemes for Ab Initio Total-Energy

Calculations Using a Plane-Wave Basis Set. Phys. Rev. B 1996, 54, 11169–11186. (59)

Blöchl, P. E. Projector Augmented-Wave Method. Phys. Rev. B 1994, 50, 17953–

17979. 23 ACS Paragon Plus Environment


(60)

Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. A Consistent and Accurate Ab Initio

Parametrization of Density Functional Dispersion Correction (DFT-D) for the 94 Elements HPu. J. Chem. Phys. 2010, 132, 154104.


Page 24 of 31

Page 25 of 31 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


Figure 1. Nanoscale contact–separation experiment using an atomic force microscopy (AFM) force–distance (F-d) method. a) Schematic illustration of the setup for AFM F-d measurement. b) Schematics of typical F-d curves. Blue and red solid lines indicate the curves for tip approach and retraction, respectively, and the corresponding tip–sample interactions are depicted in the insets. The energy consumption originating from tip–sample adhesion is shaded on the approach and retraction curves. Measured AFM F-d curves on (c) bare nickel (Ni) and (d) graphene (Gr)-coated nickel. Schematics and morphologies of the samples are provided in the insets. The Raman spectrum of the synthesized graphene is shown in the inset of (d). e) Energy dissipation on nickel (blue circles) and graphene-coated nickel (orange circles) during the contact–separation process. The standard elastic line (Edissipation = 0) is indicated by the red dashed line. Average values of Edissipation on graphene and nickel are 25.04 and 736.62 eV, respectively.



Figure 2. Atomic-scale contact–separation simulation using density functional theory (DFT) calculations. a) DFT-optimized atomic structures of the gold (Au) tip and nickel (Ni) surface. Because of the strong Au–Ni interaction, ductile Au atoms are elongated and migrate to the Ni surface during approach (1–3), and wear including necking and fracture is sequentially generated during retraction (4–6). b) Calculated atomic force of the Au tip, Ftip, during approach (left) and retraction (right). The original tip position is set to zero in the z coordinate. Ftip fluctuates considerably and gradually recovers during approach and retraction, respectively, because of the rearrangement of Au atoms, resulting in significant hysteresis. c) 26 ACS Paragon Plus Environment

Page 26 of 31

Page 27 of 31 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


DFT-optimized atomic structures of the Au tip and graphene (Gr)-coated surface. Instead of the atomic deformation of the Au tip, graphene is elastically compressed in the contact state without generating any atomic-level wear of either the tip or graphene during the contact– separation process. d) Calculated Ftip during approach (left) and retraction (right). Almost elastic-like contact–separation characteristics without hysteresis are observed in the approach and retraction curves for the Gr-coated surface.



Figure 3. Demonstration and mechanical operation of a graphene-contacting microelectromechanical (MEM) switch. a) Schematic of the fabrication process of the CVD-grown graphene-contacting MEM switch. b) SEM image of the switch. c) Crosssectional TEM images of the few layers of graphene on the contacting sidewall of the structure. Inset is a top-view of the source and drain parts. TEM is performed on cross-section part of the red dashed line). d) Current–voltage (I-V) characteristics of the fabricated device (VDS=0.05 V, Icompliance=10 nA). Inset is a magnified I-V curve of the ‘On’ state. e) On-voltage (Von) (electrostatic operating voltage) as a function of beam length (l). The black solid line was obtained from the simplified lumped model (inset). Error bars show the standard error between individual samples. f) The transient responses of the MEM device from the ‘Off’ to the ‘On’ state.


Page 28 of 31

Page 29 of 31 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


Figure 4. Device contact–separation characteristics. Measured current–voltage (I-V) characteristics of (a) nickel- and (b) graphene-contacting microelectromechanical (MEM) switches. Contact–separation characteristics were analysed by cyclic sweeping of gate voltage without current compliance. Marked hysteresis was measured for the nickel switch (right panel in (a)), whereas the graphene-contacting device did not display hysteresis (right panel in (b)). c-f) Short- and long-term reliability tests of the switches. Fifteen cycles were performed without current compliance for the (c) nickel- and (e) graphene-contacting MEM switches. IV curve changes of (d) nickel- and (f) graphene-contacting MEM switches during long-term cyclic operation under hot-switching conditions (VDS = 1 V, IDS = 1 µA). The graphenecontacting MEM device shows extremely high reliability. On–Off ratio stabilities are also shown in the insets in (d) and (f).



Figure 5. On-resistance transition with respect to the cyclic hot-switching operations. Comparison of the on-resistance (R) variation of nickel (Ni)-, gold (Au)-, and graphene (Gr)coating devices during cyclic hot-switching (1 V / 1 µA) operations. The graphene-contacting devices show the relatively high initial resistance but highest reliability. Error bars show the standard error between individual samples.


Page 30 of 31

Page 31 of 31 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


ToC figure


1000-Fold Lifetime Extension of a Nickel Electromechanical Contact

Recommend Documents