Elsevier

Carbon

Volume 168, 30 October 2020, Pages 354-361
Carbon

Kinetic study of surfactant-free graphene exfoliation at a solvent interface

https://doi.org/10.1016/j.carbon.2020.06.043Get rights and content

Abstract

The kinetics of graphite exfoliation at a liquid-liquid interface is investigated by monitoring the increase in graphene surface area as a function of time. With only two basic assumptions, we establish a kinetic model inspired by bacteria reproduction that describes the exfoliation process of graphite to graphene. The exfoliation process is spontaneous and is driven by the spreading of graphene at a liquid-liquid interface lowering the free energy of the system. While the thermodynamics of this process are understood, the kinetics have remained elusive. Understanding the exfoliation process of pristine, unfunctionalized graphene is important for potential applications of graphene that require a scalable approach to its production. Previous kinetic studies of graphite exfoliation are rare and are either computational results or are based on systems requiring the application of mechanical energy to drive exfoliation. We find our experimental data closely fits our kinetic model and allows for the determination of rate constants and activation energies for several grades of pristine natural flake graphite exfoliation.

Introduction

Graphene, a one-atom-thick layer of carbon atoms, was first isolated from graphite in 2004 [1]. Even though the existence of graphene was known [[2], [3], [4]], its isolation had not been previously demonstrated. The excitement surrounding graphene is a result its extraordinary properties, such as electron mobility above 15000 cm2V−1s−1and a sheet resistance of ∼280 Ω/sq [5,6]. These properties make graphene an outstanding material for applications such as sensors [7], composites [8], and batteries [9]. However, the original method used for graphene’s isolation, repeatedly pealing apart graphite using adhesive tape, is not a viable approach to produce the large quantities of graphene needed for most applications.

Current methods to produce single or few-layer graphene material [10] include: mechanical exfoliation, chemical vapor deposition (CVD), the reduction of graphene oxide (GO) to reduced graphene oxide (rGO), and liquid phase exfoliation [[11], [12], [13]]. CVD produces the highest quality graphene, but does so using a high energy process that produces one sheet at a time [14,15]. A less expensive and more scalable approach is to obtain graphene by exfoliating graphite. The most common approach to exfoliating graphite is to first oxidize the graphite to GO [4], then reduce the GO to reduced graphene oxide (rGO) after exfoliation. Forming GO allows for exfoliation in water with sonication [16]. Since GO shows degraded electronic properties compared to graphene, reduction is often necessary to recover some of the original properties of pristine graphene [17,18].

Liquid phase exfoliation uses high energy sonication or mixing to exfoliate graphite and disperse graphene sheets in a solvent, often aided by added surfactants, avoiding the need for oxidation or reduction [19]. A typical approach is to use sonication to disperse graphite in solvents such as N-methyl-pyrrolidone (NMP) or dimethylformamide (DMF) [20], or in aqueous solutions with surfactants such as sodium dodecylbenzene sulphonate (SDBS) [21]. These methods are not ideal however, as high power sonication for extended periods of time can break the graphene sheets into small flakes with lateral dimensions of tens of nanometers and the high boiling organic solvents and surfactants can be difficult to remove.

Another approach is the thermodynamically driven spreading of graphite at a water-oil interface [22]. Here graphite is spontaneously exfoliated as the graphene sheets spread and stabilize the high-energy liquid-liquid interface, thus acting as surfactants. As a surfactant, these graphene sheets have been shown to stabilize water-in-oil emulsions [23,24], with dispersed water droplets surrounded and stabilized by exfoliated, overlapping graphene sheets [22,[25], [26], [27]]. This method is a fundamental departure from previous methods as it does not require the input of mechanical or chemical energy and so does not require stabilizers, sonication, or chemical modification. The thermodynamic driving force for this process has been studied both experimentally and computationally, with theory and experiment appearing to converge [22].

However, investigation of the kinetics of the exfoliation process has proven to be challenging, and the paucity of kinetic studies of graphite exfoliation in general, despite a large number of exfoliation studies, suggests the difficulty of such studies. Even papers that discuss the exfoliation of graphene oxide or reduced graphene oxide seldom investigate the kinetic mechanisms and often do not quantify the exfoliation process [28]. A review of more than 3000 papers that refer to graphene and exfoliation in their titles found that 1226 actually gave details concerning exfoliation, 86 looked at graphite rather than GO, and the kinetics of exfoliation was only discussed in 15.

Studies that do address the kinetics of graphite exfoliation include the computational work of Blankschtein who used highly-ordered pyrolytic graphite (HOPG) as the base material to study its exfoliation by the halogen intercalants ICl or IBr [29,30]. For their simulations they used several assumptions: (i) the lateral size of all graphene sheets was the same, (ii) the rate of exfoliation was independent of the number of layers, (iii) the diffusivity of graphene sheets was independent of the number of layers, since the friction factor in the Stokes-Einstein relation mainly depends on the lateral size of a graphene sheet, and (iv) the intersheet interaction potential energy was independent of the number of layers.

Coleman et al. introduced liquid exfoliation of layered materials by ultrasonication in 2010 [31] and in subsequent studies developed a kinetic model that quantitatively described the shear exfoliation process both in N-methyl-2-pyrrolidone (NMP) and in aqueous surfactant solutions with a high-shear mixer [32,33]. Those studies also made several assumptions: (i) the graphene flakes were considered to be plate-like particles coated with surfactant [21], (ii) the graphene was considered to occupy a sphere defined by the flake length [31], (iii) the exfoliation of graphene flakes could be described as a concentration-dependent Smoluchowski diffusion processes [21], and (iv) there was a concentration-independent equilibrium number density of bundles that led to a scaling relationship between bundle diameter and concentration [31] where the graphene concentration was proportional to the mass of the average flake divided by the volume of the associated sphere. In these studies their graphene exfoliation model showed the dispersed concentration scaled with exfoliation time as an exponential law, with most of their studies reporting exponents varying from 0.5 to 2 depending on the exfoliation method (a combination of mixing and sonication gave the higher value). Other parameters were also included, such as initial graphite concentration, rotor speed, rotor diameter, and liquid volume [21,[31], [32], [33]].

Another study that examined the kinetics of sonication driven exfoliation used pre-exfoliated platelets of 20–50 graphene layers with nanolatex and triblock polymer surfactants [34], and the experiment had to be performed in an ice bath to compensate for the heat from sonication. In this study the assumptions included: (i) the exfoliation process was irreversible, (ii) the location in the stack where exfoliation occurred was random, and (iii) “chopping up” the graphite flakes by sonication did not affect the kinetics. Other parameters such as stabilization concentration, temperature, mixing, and sonication power were not separately considered, rather they were treated as pseudo first-order constants and included in the respective rate constant. The study concluded that the sheet splitting process followed first-order reaction kinetics and that the thicker stacks split much more quickly than did thinner stacks [35]. All the experimental reports used liquid phase separation and the exfoliation kinetics necessarily included a stabilization mechanism. In addition, some factors typically found in kinetic studies, such as temperature and activation energy, were not presented.

In the study presented in this article, we determined the rate of graphene exfoliation as a function of temperature, concentration, initial graphite flake size, and the source of the graphite. We also estimated the values of the rate constants and activation energy of graphite from several sources without the complication of stabilizer kinetics. Our investigation was based on monitoring the increasing volume of emulsion as a function of mixing time. In our case mixing does not provide energy to drive exfoliation but simply increases the interfacial surface area between the oil and water phases at which the graphene spreads. The amount of interface stabilized by graphene is limited by the surface area of the graphene. Initially there is only graphite in the system and so the total surface area stabilized is small and no emulsion is formed. Over time, as the graphite exfoliates by spreading at the liquid interface, the total stabilized interfacial area increases and an emulsion begins to form. With increasing time, the volume of this emulsion increases until the entire volume of water in the system is dispersed in the continuous oil phase and stabilized against coalescence by a thin layer of overlapping graphene sheets. The total surface area of this layer of overlapping sheets is directly proportional to the extent of exfoliation as each exfoliation event doubles the surface area of the graphite/graphene.

Our approach requires only two assumptions: (i) the sphere size of the dispersed water droplets is constant during the exfoliation process, and (ii) the graphene coverage of the spheres is constant over the timescale of the experiment (the spheres contain equal graphene surface area). The size of the droplets has been studied with both optical microscopy and acoustic spectroscopy as described in the SI and shown to be nearly constant irrespective of the emulsion volume. The graphene surface area that stabilizes each sphere should be the minimum value required for stabilization against coalescence. Spheres with less coverage would coalesce to form larger spheres, and spheres with more the minimum surface might be expected to form additional spheres during mixing. The reasonably narrow dispersity of sphere sizes shown in our acoustic spectroscopy studies suggests that while some variation in coverage likely exists, it does not significantly affect our results.

Section snippets

Experiment and method

The graphite used for the majority of this study was obtained from Asbury Carbon. Heptane was purchased from MilliporeSigma (grade 99%) and used as provided. A typical graphene emulsion was prepared by adding 150 ml of deionized water and 50 ml of heptane into a 300 ml beaker. The beaker was then covered and put into a constant temperature water bath. The temperature was monitored and recorded by a thermocouple (Omega TSS series, type K) and data collector. When the solution reached the desired

Effect of graphite concentration on rate

In order to analyze the rate of the graphene exfoliation, we monitored the emulsion volume increase with time using increasing amounts of graphite. The results are shown in Fig. 1. The emulsion volume was measured and converted to the volume fraction of the emulsion relative to the volume of the original solution. The time required to achieve the maximum emulsion volume decreased with increasing amounts of initial graphite. Additionally, the lower graphite concentration samples showed a lag

Conclusion

The kinetics of graphite’s exfoliation to graphene were investigated using pristine graphite at a water-oil interface. Graphite spontaneously spreads and exfoliates at the high energy liquid interface, driven by a lowering of interfacial energy between aqueous and organic solvents. In this way the graphene acts as a surfactant and stabilizes water-in-oil emulsions with thin layer of overlapping graphene sheets. The total internal surface area of the emulsions is directly related to the surface

CRediT authorship contribution statement

Taoran Hui: Formal analysis. Douglas H. Adamson: Formal analysis.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

The authors wish to thank the National Science Foundation DMREF program for funding through grant # DMR-1535412.

References (39)

  • X. Wang et al.

    Transparent, conductive graphene electrodes for dye-sensitized solar cells

    Nano Lett.

    (2008)
  • W. Choi et al.

    Synthesis of graphene and its applications: a review

    Crit. Rev. Solid State Mater. Sci.

    (2010)
  • C. Backes et al.

    Guidelines for exfoliation, characterization and processing of layered materials produced by liquid exfoliation

    Chem. Mater.

    (2017)
  • J. Plutnar et al.

    The chemistry of CVD graphene

    J. Mater. Chem. C.

    (2018)
  • R. Muñoz et al.

    Review of CVD synthesis of graphene

    Chem. Vap. Depos.

    (2013)
  • S. Park et al.

    Chemical methods for the production of graphenes

    Nat. Nanotechnol.

    (2009)
  • C. Gómez-Navarro et al.

    Electronic transport properties of individual chemically reduced graphene oxide sheets

    Nano Lett.

    (2007)
  • D. Li et al.

    Processable aqueous dispersions of graphene nanosheets

    Nat. Nanotechnol.

    (2008)
  • Y. Hernandez et al.

    High-yield production of graphene by liquid-phase exfoliation of graphite

    Nat. Nanotechnol.

    (2008)
  • Cited by (0)

    View full text