Regular Article
Influence of interface hydration on sliding of graphene and molybdenum-disulfide single-layers

https://doi.org/10.1016/j.jcis.2018.12.089Get rights and content

Abstract

Humidity influences friction in layered materials in peculiar ways. For example, while water improves the lubricating properties of graphite, it deteriorates those of molybdenum disulfide (MoS2). The reasons remain debated, not the least due to the difficulty in experimentally comparing dry and hydrated interface frictions. Here we show that the hydration of interfaces between a mica substrate and single-layers of graphene and MoS2 with a molecularly thin water layer affects strain transfer from the substrate to the 2D materials. For this, we strain the substrate and detect strain in graphene and MoS2 by changes in Raman and photoluminescence spectra, respectively. Strain relaxation in graphene changes from stick-slip in dry contact, to viscous when hydrated. In contrast, there is no viscous relaxation in MoS2 regardless of hydration. Our work provides a novel approach for better understanding the impact of hydration on friction in layered materials.

Introduction

Layered materials are common in both technology and nature, with solid lubricants and clays being prominent examples. The performance of lubricants and the probability of landslides in clay soils depend on their hydration. The governing mechanisms of how hydration influences properties of layered materials, and the related question of how confinement influences the properties of water are still disputed [1], [2], [3], [4], [5], [6]. Graphite and MoS2 are two representative layered materials. Both are broadly used as solid lubricants, and exhibit opposite dependencies of their lubrication performances on humidity [2], [7], [8], [9]. The investigation of graphite lubrication dates back to Bragg, who suggested that slipperiness of graphite is an intrinsic property of weakly bound graphene layers constituting graphite crystals [10]. Later it was recognized that graphite fails to lubricate in vacuum [7]. Certain low molecular weight vapours, most notably water under ambient conditions, are required for graphite to perform as a solid lubricant. A number of experimental evidences indicate that the vapours form physisorbed molecular layers on the graphite surface, and increase in friction is detected when the physisorbed layers evaporate [7], [8]. Interpretation of this is not straightforward owing to the difficulty to obtain detailed knowledge of the true nature of the sliding interfaces [11]. Computer simulations, as well as fast permeation of water through graphene capillaries within graphene oxide membranes indicate low friction of water flow along graphitic surfaces implying lubrication of graphene sliding by water [12], [13], [14]. MoS2 contrary to graphite, lubricates well in vacuum, while ambient humidity deteriorates its lubrication performance. It has been recently shown that water molecules form a physisorbed layer on its surface similar to graphite [2], [9]. Hence, a physisorbed water layer deteriorates lubrication by MoS2 while it enhances the one by graphite.

We used single-layer graphene and MoS2 flakes as strain sensors to measure strain transfer from a substrate to the flakes through dry and hydrated interfaces (Fig. 1). Strain in graphene and MoS2 can be detected by changes in Raman and photoluminescence (PL) spectra, respectively [15], [16], [17], [18], [19]. Thereby we obtain insight into the impact of hydration of the sliding interfaces on their frictional characteristics. As substrate we used mica, which is a hydrophilic layered crystal and its cleavage results in macroscopically large atomically flat and clean surfaces. Single-layer flakes of the two-dimensional materials were mechanically exfoliated on freshly cleaved mica in dry nitrogen. Under these conditions the mica surface remains dry, as does its interface with the two-dimensional materials’ flakes exfoliated thereon [20], [21], [22]. A molecularly thin layer of water wets the graphene- and MoS2- mica interfaces at elevated humidities (Fig. 1F and [21], [23]).

Section snippets

Materials and methods

Muscovite mica slabs (Ratan mica Exports, grade V1 (optical quality)) were cleaved in a glove box (LABmaster, M. Braun Inertgas-Systeme GmbH) filled with nitrogen and less than 10 ppm of water. Thin graphite flakes were peeled off a piece of highly oriented pyrolytic graphite (HOPG, grade ZYA, Momentive performance Inc.) and gently pressed onto and then removed from the cleaved mica surface with a pair of electrically grounded tweezers. Similarly to preparation of graphenes, freshly cleaved MoS2

Results and discussion

For graphene, the position of the 2D Raman peak is used to follow strain transfer to the graphene flakes, since it is more strain sensitive than the G peak position [28], [29], [30]. For MoS2, it is rather the PL than the Raman peak position, which is more strain sensitive. The PL spectra from a MoS2 monolayer are dominated by two peaks (“A” and “B”) (Fig. 1E) [31] and both peak positions red-shift with strain. We used the more intense A peak position to follow strain in MoS2 flakes [16].

In the

Conclusions

We investigate strain relaxation in single layers of graphene and MoS2 resting on dry muscovite mica and with an intercalated molecularly thin water layer. Graphene on dry mica exhibits “stick-and-slip” strain relaxation with the frictional forces per area of up to about 100 kPa. Strain in pieces of graphene with a hydrated interface relaxes nearly exponentially. We show that the exponential relaxation can be explained by assuming viscous friction at the interface. The estimated viscous

Acknowledgments

H.L. and A.R. gratefully acknowledge support through the International Max Planck Research School on Multiscale Biosystems and the German Science Foundation (DFG) through IRTG 1524, respectively. We acknowledge Donau Carbon GmbH for their gifts of activated carbon.

References (38)

  • B. Rezania et al.

    J. Colloid Interface Sci.

    (2013)
  • F. Memarian et al.

    Superlattices Microstruct.

    (2015)
  • M.J. Ikari et al.

    J. Geophys. Res. Solid Earth

    (2007)
  • G. Levita et al.

    ChemPhysChem

    (2017)
  • A. Schlaich et al.

    Nano Lett.

    (2017)
  • J. Klein

    Friction

    (2013)
  • G. Algara-Siller et al.

    Nature

    (2015)
  • Y. Zhu et al.

    Phys. Rev. Lett.

    (2001)
  • R.H. Savage

    J. Appl. Phys.

    (1948)
  • J.K. Lancaster et al.

    J. Phys. D Appl. Phys.

    (1981)
  • H.S. Khare et al.

    Tribol. Lett.

    (2014)
  • W.L. Bragg

    Introduction to Crystal Analysis

    (1928)
  • I.C. Roselman et al.

    J. Phys. D Appl. Phys.

    (1976)
  • R.R. Nair et al.

    Science

    (2012)
  • G. Tocci et al.

    Nano Lett.

    (2014)
  • W. Chen et al.

    Phys. Rev. Lett.

    (2015)
  • T.M.G. Mohiuddin et al.

    Phys. Rev. B – Condensed Matter Mater. Phys.

    (2009)
  • A. McCreary et al.

    ACS Nano

    (2016)
  • H.J. Conley et al.

    Nano Lett.

    (2013)
  • Cited by (18)

    • Two-dimensional nanoporous and lamellar membranes for water purification: Reality or a myth?

      2022, Chemical Engineering Journal
      Citation Excerpt :

      The free-spacing of the MoS2 membrane was observed to significantly reduce from 0.9 nm to 0.3 nm following 2 h heating at 60 °C [135,137]. MD simulations have been carried out to understand the interaction between water molecules and MoS2 layers and revealed that water molecules form a multi-layer structure within the free-spacing of MoS2 membrane, which increases the free-spacing in MoS2 membranes [140,141]. Therefore, for efficient transport of water, MoS2 membranes should always be kept hydrated.

    • MoS<inf>2</inf>-based membranes in water treatment and purification

      2021, Chemical Engineering Journal
      Citation Excerpt :

      Wang et al., discovered MoS2 membranes assembled via pressure-assisted filtration almost entirely lost its water permeability after oven-dried at 60 °C for 2 h, the free spacing was changed from ~ 0.9 nm to ~ 0.3 nm [15], as shown in Fig. 3a. Another study also demonstrated the similar experimental results (Fig. 3b) [8]. Molecular dynamics investigation has been conducted to try to explain this phenomenon in few past years, and these theoretical investigations revealed that water molecules could form a planar multi-layered structure between two MoS2 layers [68-70], leading to the increasement of the interlayer distance of stacking layers. Additionally, these findings demonstrated water intercalation did not change the arrangement of S-Mo atoms on the planar surface [68].

    View all citing articles on Scopus
    View full text