Load-Dependent Friction Hysteresis on Graphene - ACS Nano (ACS

Subscriber access provided by TULANE UNIVERSITY

Article

Load-Dependent Friction Hysteresis on Graphene Zhijiang Ye, Philip Egberts, Gang Hee Han, A.T. Charlie Johnson, Robert W Carpick, and Ashlie Martini ACS Nano, Just Accepted Manuscript • DOI: 10.1021/acsnano.6b00639 • Publication Date (Web): 25 Apr 2016 Downloaded from http://pubs.acs.org on April 26, 2016

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 Nano 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 25

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 Nano

ToC graphic 39x35mm (300 x 300 DPI)

ACS Paragon Plus Environment

ACS Nano

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

Load-Dependent Friction Hysteresis on Graphene Zhijiang Ye,† Philip Egberts,‡ Gang Hee Han,¶ A. T. Charlie Johnson,§ Robert W. Carpick,∥ and Ashlie Martini∗,† School of Engineering, University of California Merced, 5200 N. Lake Road, Merced, CA, 95343, USA, Department of Mechanical and Manufacturing Engineering, University of Calgary, 40 Research Place NW, Calgary, Alberta T2L 1Y6, Canada, Physics Division, Center for Integrated Nanostructure Physics (CINAP), Sungkyunkwan University (SKKU), Suwon, South Korea, Department of Physics and Astronomy, University of Pennsylvania, 209 S. 33rd Street, Philadelphia, Pennsylvania, 19104, USA, and Department of Mechanical Engineering and Applied Mechanics, University of Pennsylvania, 220 S. 33rd Street, Philadelphia, Pennsylvania 19104, USA E-mail: [email protected]

Abstract Nanoscale friction often exhibits hysteresis when load is increased (loading) and then decreased (unloading) and is manifested as larger friction measured during unloading compared to loading for a given load. In this work, the origins of load-dependent friction hysteresis were ∗ To

whom correspondence should be addressed of Engineering, University of California Merced, 5200 N. Lake Road, Merced, CA, 95343, USA ‡ Department of Mechanical and Manufacturing Engineering, University of Calgary, 40 Research Place NW, Calgary, Alberta T2L 1Y6, Canada ¶ Physics Division, Center for Integrated Nanostructure Physics (CINAP), Sungkyunkwan University (SKKU), Suwon, South Korea § Department of Physics and Astronomy, University of Pennsylvania, 209 S. 33rd Street, Philadelphia, Pennsylvania, 19104, USA ∥ Department of Mechanical Engineering and Applied Mechanics, University of Pennsylvania, 220 S. 33rd Street, Philadelphia, Pennsylvania 19104, USA † School

1 ACS Paragon Plus Environment

Page 2 of 25

Page 3 of 25

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 Nano

explored through atomic force microscopy (AFM) experiments of a silicon tip sliding on chemical vapor deposited graphene in air and molecular dynamics simulations of a model AFM tip on graphene mimicking both vacuum and humid air environmental conditions. It was found that only simulations with water at the tip-graphene contact reproduced the experimentallyobserved hysteresis. The mechanisms underlying this friction hysteresis were then investigated in the simulations by varying the graphene-water interaction strength. The size of the watergraphene interface exhibited hysteresis trends consistent with the friction, while measures of other previously-proposed mechanisms, such as out-of-plane deformation of the graphene film and irreversible reorganization of the water molecules at the shearing interface, were less correlated to the friction hysteresis. The relationship between the size of the sliding interface and friction observed in the simulations was explained in terms of the varying contact angles in front of and behind the sliding tip, which were larger during loading than unloading.

Keywords: atomic force microscope (AFM), graphene, copper, nanotribology, chemical vapor deposition (CVD), molecular dynamics simulations

New insight into the fundamental mechanisms of friction has been achieved through the advent of nanoscience and the associated characterization and simulation tools that are now available. Understanding these mechanisms is the first step towards development of a truly predictive model of sliding friction, which would have significant implications for a variety of science and engineering fields. However, the complexity of the mechanical, physical, and chemical processes that occur when two bodies slide against one another has hindered progress towards this goal. One example of the complexity of friction is the sliding-history dependence of friction. This phenomenon can manifest itself in a number of ways, but here we will focus on the behavior observed when friction is measured using an atomic force microscope (AFM) as the normal load is increased and then decreased for a two-dimensional material sample, graphene. Friction experiments have shown that, at a given normal load during a load-unload cycle, the friction force measured during unloading is significantly larger than the friction force that was measured during loading, 1–7 and the behavior is reproducible (i.e., not due to irreversible wear). This behavior, referred to as load-dependent fric2 ACS Paragon Plus Environment

ACS Nano

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

tion hysteresis, suggests important contributions to friction have not been accounted for in many of the proposed friction models. Note that, while surface wear can result in friction hysteresis, 2 such behavior can leave observable damage, and the subsequent friction measurements will be different. Wear can often be avoided in AFM studies, and indeed none of the studies discussed above reported wear. This shows that repeatable hysteresis can occur without irreversible damage or material displacement. Load-dependent friction hysteresis has been observed on a variety of materials, including graphene, 1,3 polymer layers, 4,5,8–10 silica and muscovite mica. 7 All of these studies reported higher friction during unloading than loading, but the measurements have also been found to be sensitive to various experimental parameters and conditions. For example, the magnitude of the hysteresis on graphene was observed to increase with maximum applied load. 1 It was also shown that temperature can affect friction hysteresis on poly(N-isopropylacrylamide) surfaces. 5 However, hysteresis was found to be independent of humidity, loading rates, and material contrast on polymethyl-methacrylate (PMMA) surfaces. 4 These finding suggest that friction hysteresis can arise from different, complex physical mechanism. So far, several different mechanisms have been considered to explain load-dependent hysteresis. For graphene and graphite, friction hysteresis was explained by the relaxation of the outof-plane deformation of the topmost, locally-delaminated graphene layer upon unloading. 1,3 In methylcellulose films, the hysteresis was attributed to an irreversible reorganization of the topmost polymer layer. This reorganization was suggested to have arisen from the combination of high load and shear stress, 8 which resulted in the lifting of this topmost polymer layer from its substrate as load was decreased. A comparable mechanism has been proposed for the hysteric friction behavior of graphene 1 and graphite. 3 Friction hysteresis on polymers has also been explained in terms of the temperature-dependent interaction of the polymer with water present on the surface, 5 and by viscoelasticity leading to contact area hysteresis between loading and unloading. 9–11 Another hypothesis, applicable to multiasperity contacts, is that microscopic motion, induced by Poisson contraction or expansion, produces a strong memory dependence of contact area as a function of


Page 4 of 25

Page 5 of 25

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 Nano

the loading history. 4 Finally, the water meniscus that forms between tip and substrate in humid air has been shown to lead to friction hysteresis at a silica-mica contact. 7 Friction hysteresis therefore represents a complex aspect of frictional sliding that is not well understood, sensitive to multiple experimental parameters, and attributable to a variety of different mechanisms. In this paper, we probe the origins of load-dependent friction hysteresis on graphene using AFM measurements and molecular dynamics (MD) simulations. The experiments, conducted under ambient conditions (21◦ C and 40% relative humidity), provide direct evidence of the behavior, and the simulations of atomic friction in vacuum and humid air conditions are used to explore the mechanisms that underlie the experimental observations. The simulations enable investigation of the roles of three previously proposed friction hysteresis mechanisms – out-ofplane deformation, irreversible structural reorganization, and the presence of water in the sliding interface – and their possible effects on frictional behavior. Our study provides atomic insight into the mechanisms underlying load-dependent friction hysteresis on graphene and potentially other materials that exhibit a similar behavior.

Results AFM friction experiments, summarized in Fig. 1 (a), were conducted on chemical vapor deposited (CVD) graphene samples having less than one monolayer coverage, using silicon tips as the counter surface. These CVD-graphene samples, which were examined on the original polycrystalline copper substrates, exhibited folded regions within the graphene islands, as described in Ref. 1 and shown in topographic and lateral force images such as Fig. 1 (b) and (c), respectively. As depicted in Fig. 1 (d), the folded regions correspond to three layers of graphene covering the copper substrate, compared with the remaining area of the island that corresponded to one layer of graphene covering the copper substrate and can be identified by the lower friction, where the darker region enclosed in Fig. 1 (c) indicates the 3-layer region of graphene. Fig. 1 (a) shows the variation of the friction force with normal force on this folded, three-layer graphene region. The data was acquired


ACS Nano

(a)

(b)

(c) 24 nm

7

0.13 nN

Loading Unloading

6 5 Friction (nN)

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 25

4

80 nm

80 nm 0 nm

3

0 nN

(d)

Graphene

2 1 -30

-20

-10

0

10

20

30

40

Copper Substrate

Load (nN)

Figure 1: (a) An AFM measurement of load-dependent friction hysteresis on graphene. Black squares show the friction measured while the load was increased to a pre-determined value (loading) and red circles show the friction measured while the load was decreased from the predetermined set point until the tip was pulled out of contact with the sample. Error bars represent the standard deviation in the mean value of friction calculated. (b) Topographic image of the graphene covered copper foil showing the topographic variation from one layer, over a folded region of three layers, and then back to one layer. (c) Flattened lateral force image (forward scan direction) used to identify the three layer region acquired at an applied load of 0 nN. Darker regions correspond to lower values of the lateral force. (d) Schematic of the surface structure. The region containing three layers of graphene is enclosed within a blue dashed box, which is the region used to generate the load dependence curve in (a).


Page 7 of 25

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 Nano

by measuring friction at a normal force of approximately 0 nN applied normal force, increasing the applied force to ∼35 nN, and then decreasing the normal force until the tip was pulled out of contact with the surface. Fig. 1 (a) clearly shows significant friction hysteresis, well outside of measurement uncertainty. Examination of the surface after the measurement did not show any alteration of the surface (wear), which has been proposed as one of the mechanisms for friction hysteresis during load dependence measurements, 2 but does not appear to be contributing to the hysteresis here. Additionally, no viscoelastic effects were observed as repeated measurements on single layer graphene in the same area showed that the friction force returned to the same value at applied normal forces during loading, which also suggests that wear is not a mechanism for the observed hysteresis. 1

Figure 2: Snapshots of the MD simulations of an amorphous SiO2 AFM tip apex sliding on a multilayer graphene surface with (right) and without (left) water molecules in the interface. Additional simulation details can be found in supplementary materials. 12 MD simulations of sliding friction during loading and unloading were conducted for two possible cases, vacuum (Fig. 2 (left)) and with water present in the tip-sample contact (Fig. 2 (right)), in an attempt to reproduce the experiments most accurately. First, we used a hemispherical silicon dioxide tip sliding on three layers of graphene, matching the number of layers of graphene in the experiments, and in the absence of an atmosphere (vacuum) as illustrated in Fig. 2 (left). The simulated data showed that the friction force increased during loading and decreased during unloading, as shown in Fig. 3. However, no friction hysteresis was observed. Thus, the inability for the MD simulations to reproduce the experimental friction observations indicates a difference 6 ACS Paragon Plus Environment

ACS Nano

6

Vacuum:

Loading

Unloading

Hydrophobic:

Loading

Unloading

Hydrophilic:

Loading

Unloading

5

Friction (nN)

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 8 of 25

4

3

2

1

0 -20

-10

0

10

20

30

40

50

60

Load (nN)

Figure 3: Load-dependent friction from simulations run in vacuum (loading and unloading represented by black squares and purple circles, respectively) and humid air. Blue triangles (solid and hollow for loading and unloading, respectively) are the friction for the standard water-graphene interaction strength (hydrophobic, 112◦ contact angle) and red diamonds (solid and hollow for loading and unloading, respectively) correspond to friction from simulations with the artificially strengthened interactions (hydrophilic, 40◦ contact angle). Simulations with different size of water droplets were also performed, results can be found in supplementary materials. 12


Page 9 of 25

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 Nano

between the conditions in the simulations and experiments that must be addressed. The experimental data were obtained at 40% humidity, and thus water (or other adsorbates) was likely present in the contact. Previous experiments on mica and silica surfaces suggested that humidity can affect load-dependent friction hysteresis due to a water meniscus present between the tip and substrate. 7 Other previous studies showed that water and small hydrocarbon molecules, adsorb on any surface exposed to air and can affect friction. 13,14 To explore the influence of a water meniscus on the friction properties measured friction force, water molecules were placed between the tip and the substrate, as illustrated in Fig. 2 (right). Since the wettability of water on graphite has been shown to vary over time with environmental exposure, 15 we modeled two cases: hydrophilic and hydrophobic. The hydrophilic case was modeled using artificially strengthened graphene-water interactions, corresponding to a contact angle of 40◦ , whereas the hydrophobic case was modeled with the weaker interactions, corresponding to a contact angle of 112◦ . Fig. 3 shows that friction hysteresis was observed when water molecules were present in the simulation, with either contact angle. Furthermore, Fig. 3 shows that the magnitude of the hysteresis, or the difference between unloading and loading friction, is dependent on the hydrophobicity of the graphene surface: increasing the hydrophobicity of the surface increased the degree of hysteresis observed in the friction force. Finally, for both hydrophobic and hydrophilic surfaces, the average friction force during loading and unloading was larger than that calculated from the vacuum simulations. In the next section, the friction mechanisms attributing to the origins of the measured friction hysteresis in both experiment and simulation, as well as the offset between the friction force measured in the three simulations, will be discussed.

Discussion To understand the observed friction hysteresis, we investigated previously-proposed mechanisms, searching for evidence of their contributions in our simulations. We first explored the potential role of out-of-plane deformation of the graphene. 1,3 Out-of-plane deformation, also referred to as


ACS Nano

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

puckering, occurs when the upper graphene layer conforms to the tip such that there is a wrinkle in front of the tip that can resist sliding, as well as increase the area of the tip the graphene covers during sliding, and therefore increase friction. This deformation has been proposed to be the source of variations in friction with number of graphene layers. 16,17 We quantified deformation in these simulations as the time average of the difference between the vertical position of the graphene wrinkle in front of the tip and the vertical position of the graphene in the contact below the tip. In all cases, with and without water, as well as with varying contact angle, the difference between the deformation during loading and unloading at a given load was statistically insignificant. This suggests that out-of-plane graphene deformation is unlikely to be contributing in the load-dependent hysteresis observed in the simulations. Another hypothesis to explain load-dependent friction hysteresis is irreversible reorganization of the adsorbates in the contact interface, which has been proposed as an explanation for friction hysteresis in polymer layers 6,9,10 as well as adhesion hysteresis in various surfactant monolayers. 18 In this case, we looked for structural changes in the water in the contact area. Structure or organization of the water molecules within the contact was characterized using the radial distribution function (RDF) of the oxygen atoms in the water molecules, which has previously been used to quantify structure of water in both experiment and simulation. 19 Fig. 4 shows the RDF for four cases that are derived from the time and load averaged structure of the water molecules in the contact during the loading and unloading processes for both the hydrophobic and hydrophilic surfaces. Note that these are calculated only using the oxygen atoms in the contact area, i.e. the layer of water atoms adjacent to the graphene, and so do not necessarily exhibit the features expected for bulk water. Regardless, the height of the RDF peak (particularly the first peak) does provide a tool for characterizing structure. We first observe that the RDF peak is sharper (greater maximum value) for the hydrophilic case than the hydrophobic case, during both loading and unloading. Sharper RDF peaks correspond to more well-defined structure in the contact, which may partially explain the larger friction observed in Fig. 3 for the hydrophilic graphene. We also observe that, for either contact angle, the RDFs during loading exhibit slightly sharper peaks than during unloading. This


Page 10 of 25

Page 11 of 25

suggests that the water adopts a more ordered structure during loading which is not completely lost on unloading. However, we also compare the structural hysteresis, i.e., the difference between the heights of the first RDF peak during loading and unloading, of the two contact angle cases (see inset of Fig. 4). Both contact angle cases exhibit structural hysteresis with approximately the same magnitude. This is inconsistent with the large difference in friction hysteresis between the hydrophobic and hydrophilic cases in Fig. 3 which indicates that, while irreversible reorganization may be a contributing factor, there is at least one other mechanism involved in the process.

4

Hydrophobic:

Loading

Unloading

Hydrophilic:

Loading

Unloading

4

Irreversible Reorganization

3 3

RDF

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 Nano

2

2 0.32

0.34

0.36

0.38

1

0 0.2

0.4

0.6

0.8

1.0

Distance (nm)

Figure 4: Radial distribution function of the O-O pairs in the water in the contact calculated during loading and unloading. Inset is a close up of the first RDF peak indicating there is some irreversible reorganization between loading and unloading, but the magnitude of this hysteresis (difference between RDF peak height during loading and unloading) is not dependant on contact angle. We next explore the possible contribution of the water meniscus. In previous work, it was 10 ACS Paragon Plus Environment

ACS Nano

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

suggested that friction hysteresis may be attributed to hysteresis in the size of the contact area caused by water around the AFM tip which built up during loading and did not evaporate during unloading. 7 There is no evaporation in the simulations, but this gives us a clue as to another possible way the water could be affecting hysteresis. We quantified the size of the contact as the number of water molecules in contact with the graphene, where contact was determined based on an empirical distance criteria. Specifically, a water molecule was assumed to be in contact if the minimum distance between its oxygen atom and any carbon atom in the substrate was less than 0.35 nm, where 0.35 nm was chosen to be a reasonable estimate of contact based on visual observation of the simulation. Note that this is distinct from the common notion of contact area used for dry contacts, where the solid-solid contact area is usually considered. With water present in the simulation, the shearing interface during sliding becomes the water-graphene interface, rather than the silicon tip-graphene interface, as the silicon tip did not penetrate through the last layer of water molecules at any of the loads examined. We therefore will apply the term “contact area” in the remainder of the discussion to denote the size of this shearing interface, despite the term’s typical restriction to solid-solid sliding interface. Furthermore, the tip and the water molecules form a single body in the simulations that slides together along the graphene surface, resulting in a necessary redefinition of the contact area as stated earlier. Thus, we consider the interaction area between the meniscus and the graphene for the present analysis. Fig. 5 shows the number of water molecules in the contact as a function of load during loading and unloading (symbols) for both hydrophobic and hydrophilic graphene. The contact area of the water without the tip present is shown as dashed lines. Hydrophilic graphene corresponds to larger water contact areas as expected. This is demonstrated by the offset between the blue dashed line/data points and the red dashed line/data points in Fig. 5. Furthermore, the presence of the tip increases the contact area at the shearing interface, as demonstrated by the offset in the contact area between the dashed line and the data points for both cases shown in Fig. 5; while the effect is larger for hydrophobic graphene (13-61% increased contact area), the effect for hydrophilic graphene is still significant (5-15% increased contact area).


Page 12 of 25

Page 13 of 25

More importantly, the contact areas in Fig. 5 exhibit load-dependent hysteresis, consistent with the hysteretic friction behavior in Fig. 3. Similarly, the degree of hydrophobicity influences the size of the contact at the sliding interface, again with the same trend as the friction. More hydrophilicity corresponds to a bigger water-graphene contact area and higher friction. Therefore, the friction hysteresis appears to be a direct result of the hysteresis in the contact area of the shearing interface, i.e. between water and graphene, which is only observed if water is present in the interface. Hydrophobic:

Loading

Unloading

Hydrophilic:

Loading

Unloading

130 # of water molecules

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 Nano

120 110 100 90 80 70 -20

-10

0

10

20

30

40

50

60

Load (nN)

Figure 5: Number of water molecules in the contact (a measure of contact area) with the substrate during loading and unloading process. Blue triangles (solid and hollow for loading and unloading, respectively) correspond to the standard water-graphene interaction strength (hydrophobic graphene, 112◦ contact angle) and red diamonds (solid and hollow for loading and unloading, respectively) correspond to the artificially strengthened interactions (hydrophilic graphene, 40◦ contact angle). Dashed lines represent the contact area of the water on the graphene without the tip present.


ACS Nano

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

This suggests that the water-graphene contact area, as opposed to puckering or structure of the water molecules in the contact, is the dominant friction mechanism of the three investigated. However, the origin of the contact area hysteresis itself requires further discussion, since it cannot be explained by evaporation as proposed previously. 7 One possibility is suggested by the difference between the hysteresis (in both friction and area) in the hydrophobic and hydrophilic simulations. It is well known that contact angle hysteresis can be observed when water is added to or removed from a droplet, or when a droplet moves on a tilted surface due to gravity, due to the pining of the contact line at impurities, defects, or other heterogeneities on the surface. 20 Therefore, it is possible that a meniscus is present at the tip-sample contact in the experiments, and contact angle hysteresis occurs during loading and unloading as depicted in Fig. 6. For the sliding case, there is a distribution of contact angles around the contact line, with the largest difference occurring between the front and the back side of the tip. This is because the front angle is always larger as the water is pushed forward by the lateral motion of tip. The load-dependent hysteresis applies to the contact angle at all locations around the contact line such that, during loading, all of the contact angles will decrease as the tip pushes down towards the surface; the contact area will increase with increasing load correspondingly. Then, during unloading, the contact angles will increase as the tip pulls the water up as load is decreased, but the hysteresis in all the contact angles (and the contact area) will prevent them from returning to the same magnitude at the same load. We quantified the contact angle during loading and unloading in the simulations, by linearly fitting the positions of the bottom 0.7 nm oxygen atoms at the perimeter of the water at the locations of the extreme contact anlge values, i.e., in front and back of the tip, shown in Fig. 6. The results, time averaged over loading and unloading for each equilibrium contact angle, are summarized in Table 1. We observe that both the front and back contact angles are clearly larger during loading than unloading. Furthermore, the difference between front and back angles is much greater for the hydrophobic case, consistent with the fact that the magnitude of contact angle hysteresis due to adding or removing water to a droplet tends to be directly related to the equilibrium contact angle. 20 Most importantly, the load-dependent contact angle hysteresis we observe is consistent with the


Page 14 of 25

Page 15 of 25

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 Nano

Loading

θfront

Unloading

θback

θfront

θback

Figure 6: Snapshots of the tip/water sliding (from right to left) along the surface at 0 nN during loading (left) and unloading (right). Only the oxygen atoms at the perimeter of the water droplet are shown in the snapshot from the MD simulation to simplify the image. The contact angles have been determined using the linear fits (thick dashed lines) of the outer edges of the water ball and are denoted θ f ront and θback for each case. friction hysteresis in Fig. 3 and contact area hysteresis in Fig. 5. This can be explained as follows: during loading, the water in the contact area is pushed outward and spreads out on the surface. The contact area increases while the contact angle decreases; then, during unloading, the contact line was pined to the surface, while the contact area decreases and the contact angle increases, they do not return to their initial values. Therefore, the friction hysteresis in the simulations, and possibly in the experiments, may be explained by contact area/angle hysteresis. Table 1: Contact angles in front of and in back of the tip during loading and unloading for each equilibrium contact angle. θ0 is the equilibrium contact angle and θL − θU is the contact angle hysteresis, the difference between the loading and unloading angles.

θ0 Hydrophilic(40◦ )

θ f ront Loading 60.8◦ Unloading 52.1◦ θL − θU 8.7◦ Hydrophobic(112◦ ) Loading 129.9◦ Unloading 97.6◦ θL − θU 32.2◦


θback 41.8◦ 35.2◦ 6.6◦ 63.1◦ 51.5◦ 11.6◦

ACS Nano

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

Conclusion In summary, MD simulations were used to investigate the origins of load-dependent friction hysteresis measured in AFM experiments. Only simulations with water in the contact area were able to reproduce the experimentally-observed friction hysteresis. We then used the simulations with two different water-graphene interaction strengths to explore the origins of the hysteresis. First, the out-of-plane deformation of graphene was calculated from the atomic positions of the carbon atoms and found not to exhibit hysteresis, demonstrating that this mechanism does not explain the hysteresis observed in the friction. Based on the necessity of having water in the simulations to capture friction hysteresis, we next focused on mechanisms related to the water in the contact. Quantifying the structure of the water in the contact using a radial distribution function, we found that there was some hysteresis due to irreversible reorganization of the water molecules in the contact, but that hydrophobicity of the graphene surface did not affect the magnitude of the hysteresis, which was inconsistent with the friction trends. Given the small effect of hydrophobicity on irreversible reorganization, this mechanism could not fully explain the observed friction hysteresis. Lastly, we calculated the water-graphene contact area and found trends in the effect of loading/unloading and the effect of hydrophobicity that were consistent with those observed in the friction measurements in both experiments and simulations. The relationship between contact area and friction was explored by calculating the range of contact angles around the tip during sliding in the simulations. All contact angles around the tip exhibited hysteresis during loading vs. unloading, consistent with the friction measurements. Therefore, the simulations support the following mechanism to explain the load-dependent friction hysteresis: during loading, water is pushed outward as the contact area increased, and then during unloading, the water is pulled away from the surface decreasing the contact area, but not to its original value due to pinning of the contact line. The effect is much stronger for more hydrophobic graphene, where hydrophobicity was artificially controlled by changing the graphene-water interaction strength in the MD potential. This suggests an intrinsic energy dissipation mechanism related to the motion of water molecules across the graphene surface that is strongly dependent on humidity and the hydrophobicity of the 15 ACS Paragon Plus Environment

Page 16 of 25

Page 17 of 25

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 Nano

material on which the friction is measured. The mechanism leading to the hysteresis itself appears to be an intrinsic aspect of the water-graphene interaction: the simulated graphene sample does not have any impurities or defects. Rather, an intrinsic pinning and dissipative interaction between water molecules and the graphene surface is seen to occur as forces act the move the water along the graphene surface. This effect could be further increased by features such as impurity or vacancy defects, steps, grain boundaries, or contaminants, all of which may be present in the experiment.

Methods Graphene Synthesis Polycrystalline copper foils (25 µ m thick, Alfa Aesar Inc.) were cut into square pieces of approximately 5 cm × 5 cm. The copper foils were sonicated in acetic acid for 5 min at 12 Watt to remove any surface contamination, including the native oxide, from the foil, and subsequently rinsed with de-ionized water. The rinsed foils were then placed in an alumina plate and inserted into a vacuum furnace. The grain size was increased and the surface roughness was decreased 21 through annealing the foil at 1050◦ C for 1 hr, which also reduced the amount of residual strain in the copper. During this annealing stage, the furnace was evacuated and subsequently re-filled to a pressure of 1 atm with 5 ccm of argon and 200 ccm of hydrogen gas mixture, preventing the re-formation of a native oxide and further reduce the surface roughness 22 of the copper foil. The copper foil was then cooled back to room temperature and etched by wiping a kimwipe saturated with the CE100 etchant (Transene Company, Inc.) for 5 s, and then rinsed with de-ionized water. Graphene was grown on the cleaned copper foils under atmospheric pressure in the furnace that was evacuated and then re-filled with a mixture of Ar (500 sccm) and H2 (50 sccm) and heated to a temperature of 1057◦ C. Upon reaching this temperature, the gas mixture was altered to 35 sccm H2 and 2 sccm of CH4 , with the Ar concentration remaining unchanged and held for 30 min. After this time, the furnace heating elements were turned off and the CH4 gas flow was terminated after another 10 min, and the sample was removed from the furnace. The quality and 16 ACS Paragon Plus Environment

ACS Nano

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

thickness of the graphene/copper samples produced in this manner was characterized using Raman spectroscopy. 21,23 Finally, friction measurements were conducted within 1-4 days of depositing the graphene on the copper foil.

AFM Experiments A beam deflection AFM (MFP 3D, Asylum Research) was used to measure friction under ambient conditions (∼ 21◦ C, 40% relative humidity). Samples were secured to the AFM by attaching the outer 0.5 mm of the copper foils to glass slides with double-sided tape. While attaching the copper foils to the glass slides, specific attention was paid to ensure that the area examined in friction measurements were not deformed. Commercial silicon cantilevers (PPP-CONT, Nanosensors) were used in all friction experiments. These cantilevers, which have an integrated tip typically having a radius of 5-15 nm in radius, are terminated with a native silicon oxide. To determine the normal and lateral spring constants of the individual cantilever used, the thermal resonance in both normal bending and lateral twisting was measured. Fits of the thermal resonance data were performed to extract the specific resonance frequencies and the Q-factor in each case. Using the Sader method 24,25 and the fit data, the cantilever spring constants were determined to be between 0.05-0.1 N/m and 30-100 N/m in normal bending and lateral twisting, respectively. The measured voltage signal corresponding to the cantilever displacement in normal bending from the photodetector was converted to a distance by multiplying it with the fitted slope of the linear portion of a normal force versus sample displacement curve, referred to as the deflection sensitivity. The lateral sensitivity of the photodetector was assumed to be the same as the normal sensitivity. An applied force of zero nanoNewtons is defined as the deflection signal measured far from the surface, corresponding to when the cantilever was unbent. In the data presented in this manuscript, compressive forces then correspond to positive normal forces and adhesive or tensile forces correspond to negative normal forces. Adhesive forces for the tips used in these experiments ranged from -2 nN to -20 nN. While sliding the tip along the surface, the twisting signal of the cantilever was measured. This twisting signal will be referred 17 ACS Paragon Plus Environment

Page 18 of 25

Page 19 of 25

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 Nano

to as the measured lateral force. To determine the friction force, the difference between the lateral forces measured while sliding forward and backwards was calculated and then averaged. The experimental results presented within this manuscript have been repeated with at least three different tips on various regions of the sample, to ensure consistent results. All measurements of the dependence of the friction force with normal force was determined using the following procedure. Before measuring friction, a normal force versus distance curve was acquired, allowing for a measurement of the zero deflection value. Subsequently, the tip was pressed into the surface at a load corresponding to zero applied load. Friction forces were then measured for a single frame at each load, with increasing applied load between each frame until a predetermined maximum normal force was reached. Friction measured from the zero applied load to the predetermined maximum value will be referred to as friction measured during loading. Upon reaching the maximum normal force, the normal force was then incrementally decreased at the same value until the tip did not remain in contact with the surface. This second segment of friction measurement will be referred to as friction measured during unloading. Before moving to a new location, a force versus distance curve measured the change in the zero deflection value over the course of the experiment. The difference between these two values was used to adjust the zero position of the normal force at every point along the measurement, assuming a linear change in the zero value of the normal force. The sliding speed during these measurements was between 10 and 20 µ m/s. Load dependent friction measurements were conducted on three-layer graphene supported by the copper substrate. To locate these regions, large-scale images of the surface were acquired until a region covered by graphene was identified. Upon finding a graphene island, a region of less than 500 × 500 nm2 was examined. The local roughness in this area was determined through analysis of topographic images to be 4.1 nm RMS within a 20 × 20 nm2 area. This area typically had both folded (3-layer) graphene and flat (1-layer) graphene. Load dependent measurements were acquired on this area by acquiring one image frame at each normal force set point. The data for 3-layer graphene were separated from 1-layer graphene by graphically selecting one folded region


ACS Nano

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

individually from each image acquired using customized software. Thus, a single data point on the a friction force versus normal force plot represents the average value from the selected region, and the error is the standard deviation of the averaged data points associated with this region. Surface topographic and friction images at a scale larger than 500 × 500 nm2 were acquired before and after load dependent friction measurements. Images were compared to ensure that there was no evidence of surface wear or degradation of image quality due to tip wear. In some instances, repeated load dependence measurements also verified the absence of any evidence of tip wear, as the loading portions of the curve had nearly identical friction forces at the same normal force value 1 Measurements were repeated with at least three different tips on several different graphene islands on multiple CVD samples.

MD Simulations The fully atomistic model consisted of the apex of an amorphous SiO2 AFM tip placed on a multilayer graphene surface with and without water molecules near the contact as shown in Fig. 2. The graphene substrate had in-plane dimensions of 16×10 nm2 ; the positions of the atoms in the bottom graphene layer were fixed. The amorphous SiO2 tip apex was obtained by heating and quenching a block of crystalline SiO2 and then cutting the hemispherical shape (2.5 nm radius) from the block. The topmost atoms in the tip were treated as a rigid body that was subject to a range of external normal loads (-18∼60 nN) and connected by a harmonic spring to a support that moved laterally at a constant speed of 1 m/s. The spring had stiffness of 8 N/m in the horizontal directions, but did not resist motion in the vertical direction (normal to the graphene surface). A Langevin thermostat was applied to the free atoms maintained a temperature of 300 K. The sliding distance in all simulations was 5 nm. The simulations were performed using LAMMPS simulation software. 26 The inter-atomic interactions within the tip, water and substrate were described via the Tersoff, 27 TIP4P 28 and Adaptive Intermolecular Reactive Empirical Bond Order (AIREBO) 29 potentials, respectively, and the long range interactions between tip, water and substrate were modeled 19 ACS Paragon Plus Environment

Page 20 of 25

Page 21 of 25

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 Nano

using the Lennard-Jones (LJ) potential, similar to Refs. 30,31 We initially used LJ parameters calculated using the standard mixing rules for estimating interactions between dissimilar atoms. However, to ensure the model was sufficiently accurate for this study, we compared the contact angle between a water droplet and surface calculated from the simulation to the range of experimentally measured values. For both the graphene and SiO2 , we created a water droplet by placing an array of water molecules on the surface of the material and then running dynamics. The water molecules were attracted to one another and formed a droplet; these simulations were run to reach equilibrium status until the potential energy of the system was stable, i.e. fluctuated around a constant value. Then we measured the water contact angle by linearly fitting the positions of the bottom 0.7 nm oxygen atoms at the perimeter of the water at the locations of the extreme contact angle values, i.e., in front and back of the tip. The water contact angle on SiO2 was found to be 96◦ , much larger than the range of experimentallymeasured values (31-66◦ ). 32,33 These smaller values are well-established to be due to the presence of hydrophilic silanol (-OH) groups on the surface. The mimic the effect of these groups being present without explicitly modelling them (due to limitations of our potential), we fine-tuned the LJ interaction strength ε between the Si and the O atoms (εSi−O =0.00909 eV) until a contact angle of 36◦ , within the experimental range, was achieved. The water contact angle on graphene was calculated to be 112◦ , which was within the range of values (40◦ ∼ 127◦ ) found in the literature. 15,34–36 Therefore, no parameter changes were required for this system. However, to explore the effect of wettability, we also ran simulations with an artificially strong water-graphene interaction strength (εC−O =0.00683 eV) which yielded an unphysically-small water contact angle of 40◦ .

Acknowledgement This work was funded by the U.S. National Science Foundation under CMMI-1068552 and CMMI1068741. P. E. would like to acknowledge financial support from the Natural Sciences and Engineering Research Council (NSERC) of Canada and the University of Calgary. This work used 20 ACS Paragon Plus Environment

ACS Nano

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

the Extreme Science and Engineering Discovery Environment (XSEDE), which was supported by National Science Foundation Grant No. ACI-1053575.

References 1. Egberts, P.; Han, G. H.; Liu, X. Z.; Johnson, A. C.; Carpick, R. W. Frictional Behavior of Atomically Thin Sheets: Hexagonal-shaped Graphene Islands Grown on Copper by Chemical Vapor Deposition. ACS Nano 2014, 8, 5010–5021. 2. Gosvami, N. N.; Egberts, P.; Bennewitz, R. Molecular Order and Disorder in The Frictional Response of Alkanethiol Self-assembled Monolayers. J. Phys. Chem. A 2011, 115, 6942– 6947. 3. Deng, Z.; Smolyanitsky, A.; Li, Q.; Feng, X.-Q.; Cannara, R. J. Adhesion-Dependent Negative Friction Coefficient on Chemically Modified Graphite at The Nanoscale. Nat. Mater. 2012, 11, 1032–1037. 4. Rubinstein, S.; Cohen, G.; Fineberg, J. Contact Area Measurements Reveal Loading-history Dependence of Static Friction. Phys. Rev. Lett. 2006, 96, 256103. 5. Dedinaite, A.; Thormann, E.; Olanya, G.; Claesson, P. M.; Nyström, B.; Kjøniksen, A.-L.; Zhu, K. Friction in Aqueous Media Tuned by Temperature-responsive Polymer Layers. Soft Matter 2010, 6, 2489–2498. 6. Thormann, E. Negative Friction Coefficients. Nat. Mater. 2013, 12, 468–468. 7. Feiler, A. A.; Stiernstedt, J.; Theander, K.; Jenkins, P.; Rutland, M. W. Effect of Capillary Condensation on Friction Force and Adhesion. Langmuir 2007, 23, 517–522. 8. Bodvik, R.; Thormann, E.; Karlson, L.; Claesson, P. M. Temperature Responsive Surface Layers of Modified Celluloses. Phys. Chem. Chem. Phys. 2011, 13, 4260–4268.


Page 22 of 25

Page 23 of 25

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 Nano

9. Wahl, K.; Stepnowski, S.; Unertl, W. Viscoelastic Effects in Nanometer-scale Contacts Under Shear. Tribol. Lett. 1998, 5, 103–107. 10. Wahl, K.; Asif, S.; Greenwood, J.; Johnson, K. Oscillating Adhesive Contacts Between Micron-scale Tips and Compliant Polymers. J. Colloid Interface Sci. 2006, 296, 178–188. 11. Nalam, P. C.; Gosvami, N. N.; Caporizzo, M. A.; Composto, R. J.; Carpick, R. W. Nanorheology of Hydrogels Using Direct Drive Force Modulation Atomic Force Microscopy. Soft Matter 2015, 11, 8165–8178. 12. Supplementary Materials. 13. He, G.; Müser, M. H.; Robbins, M. O. Adsorbed Layers and The Origin of Static Friction. Science 1999, 284, 1650–1652. 14. Cheng, S.; Luan, B.; Robbins, M. O. Contact and Friction of Nanoasperities: Effects of Adsorbed Monolayers. Phys. Rev. E 2010, 81, 016102. 15. Li, Z.; Wang, Y.; Kozbial, A.; Shenoy, G.; Zhou, F.; McGinley, R.; Ireland, P.; Morganstein, B.; Kunkel, A.; Surwade, S. P. et al. Effect of Airborne Contaminants on The Wettability of Supported Graphene and Graphite. Nat. Mater. 2013, 12, 925–931. 16. Lee, C.; Li, Q.; Kalb, W.; Liu, X.-Z.; Berger, H.; Carpick, R. W.; Hone, J. Frictional Characteristics of Atomically Thin Sheets. Science 2010, 328, 76–80. 17. Ye, Z.; Tang, C.; Dong, Y.; Martini, A. Role of Wrinkle Height in Friction Variation With Number of Graphene Layers. J. Appl. Phys. 2012, 112, 116102. 18. Briscoe, W. H.; Klein, J. Friction and Adhesion Hysteresis Between Surfactant Monolayers in Water. J. Adhes. 2007, 83, 705–722. 19. Soper, A. The Radial Distribution Functions of Water and Ice From 220 to 673 K and at Pressures Up to 400 MPa. Chem. Phys. 2000, 258, 121–137. 22 ACS Paragon Plus Environment

ACS Nano

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

20. Eral, H. B.; ’t Mannetje, D. J. C. M.; Oh, J. M. Contact Angle Hysteresis: A Review of Fundamentals and Applications. Colloid Polym. Sci. 2013, 291, 247–260. 21. Luo, Z.; Kim, S.; Kawamoto, N.; Rappe, A. M.; Johnson, A. T. C. Growth Mechanism of Hexagonal-shape Graphene Flakes With Zigzag Edges. ACS Nano 2011, 5, 9154–9160. 22. Kim, S. M.; Hsu, A.; Lee, Y.-H.; Dresselhaus, M.; Palacios, T.; Kim, K. K.; Kong, J. The Effect of Copper Pre-cleaning on Graphene Synthesis. Nanotechnology 2013, 24, 365602. 23. Ferrari, A. C.; Meyer, J. C.; Scardaci, V.; Casiraghi, C.; Lazzeri, M.; Mauri, F.; Piscanec, S.; Jiang, D.; Novoselov, K. S.; Roth, S. et al. Raman Spectrum of Graphene and Graphene Layers. Phys. Rev. Lett. 2006, 97, 187401. 24. Sader, J. E.; Chon, J. W. M.; Mulvaney, P. Calibration of Rectangular Atomic Force Microscope Cantilevers. Rev. Sci. Instrum. 1999, 70, 3967. 25. Green, C. P.; Lioe, H.; Cleveland, J.; Proksch, R.; Mulvaney, P.; Sader, J. E. Normal and Torsional Spring Constants of Atomic Force Microscope Cantilevers. Rev. Sci. Instrum. 2004, 75, 1988. 26. Plimpton, S. Fast Parallel Algorithms for Short-range Molecular Dynamics. J. Comput. Phys. 1995, 117, 1–19. 27. Tersoff, J. Empirical Interatomic Potential for Silicon With Improved Elastic Properties. Phys. Rev. B 1988, 38, 9902. 28. Jorgensen, W. L.; Chandrasekhar, J.; Madura, J. D.; Impey, R. W.; Klein, M. L. Comparison of Simple Potential Functions for Simulating Liquid Water. J. Chem. Phys. 1983, 79, 926–935. 29. Stuart, S. J.; Tutein, A. B.; Harrison, J. A. A Reactive Potential for Hydrocarbons with Intermolecular Interactions. J. Chem. Phys. 2000, 112, 6472. 30. Ramos-Alvarado, B.; Kumar, S.; Peterson, G. P. Hydrodynamic Slip in Silicon Nanochannels. Phys. Rev. E 2016, 93, 033117. 23 ACS Paragon Plus Environment

Page 24 of 25

Page 25 of 25

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 Nano

31. Vilhena, J. G.; Pimentel, C.; Pedraz, P.; Luo, F.; Serena, P. A.; Pina, C. M.; Gnecco, E.; Perez, R. Atomic-Scale Sliding Friction on Graphene in Water. ACS Nano 2016, acsnano.5b07825. 32. Williams, R.; Goodman, A. M. Wetting of Thin Layers of SiO2 by Water. Appl. Phys. Lett. 1974, 25, 531–532. 33. Thomas, R. R.; Kaufman, F. B.; Kirleis, J. T.; Belsky, R. A. Wettability of Polished Silicon Oxide Surfaces. J. Electrochem. Soc. 1996, 143, 643–648. 34. Taherian, F.; Marcon, V.; van der Vegt, N. F.; Leroy, F. What Is The Contact Angle of Water on Graphene? Langmuir 2013, 29, 1457–1465. 35. Shih, C.-J.; Strano, M. S.; Blankschtein, D. Wetting Translucency of Graphene. Nat. Mater. 2013, 12, 866–869. 36. Shih, C.-J.; Wang, Q. H.; Lin, S.; Park, K.-C.; Jin, Z.; Strano, M. S.; Blankschtein, D. Breakdown in The Wetting Transparency of Graphene. Phys. Rev. Lett. 2012, 109, 176101.


Load-Dependent Friction Hysteresis on Graphene - ACS Nano (ACS

Recommend Documents