Fluid dynamics: an emerging route for the scalable production of graphene in the last five years

Abstract

Bulk applications of graphene in fields such as advanced composites, conductive ink, and energy storage require cheap and scalable graphene. Fortunately, in the last decade, the liquid-phase exfoliation of graphite to obtain pristine graphene has emerged as a promising way to achieve massive production of graphene at high efficiency and low cost by utilizing a cheap and abundant graphite source and a variety of cost-effective exfoliation techniques. Though numerous exfoliation techniques are available, this article will highlight the recent progress in the fluid dynamics route, which has emerged in the last five years as a promising scalable and efficient method for graphene production. A focus is made on vortex fluidic devices and pressure- and mixer-driven fluid dynamics, with our perspectives on the latest progress, the exfoliation mechanisms, and some key issues that require further study in order to realize industrial applications.

---

Min Yi

Min Yi received his bachelor's degree in engineering from the Beihang University (BUAA) in 2010. In 2015, he obtained his PhD degree in engineering, working with Prof. Zhigang Shen's group in BUAA. Presently, he is a postdoctoral researcher at the Institute of Materials Science, TU Darmstadt, Germany. His research interests are focused on the production of graphene and its analogues by fluid dynamics and their relevant applications, and mechanics of functional materials.

Zhigang Shen

Zhigang Shen is a full professor at Beihang University (BUAA) and the director of Beijing Key Laboratory for Powder Technology Research and Development. He received his PhD degree in aerodynamics in BUAA in 1989. Since then, he has worked in BUAA as a professor. His research interests include the design, mechanical synthesis and chemical synthesis of functional nanoparticles, graphene and graphene analogues, and their applications in lubrication, energy and space protection.

---

1. Introduction

Due to its exceptional properties and intriguing applications, graphene has attracted intensive interest in the advanced science and technology fields. The last decade has witnessed numerous breakthroughs in the research into graphene. A series of novel properties and promising applications have been reported.^1–7 With the achievements made by the graphene community in recent years, it is believed by many that graphene will become the next subversive technology, substituting some of the currently used materials and resulting in new markets, and thus a scientific and technological revolution. However, the scalable and cost-effective production of graphene still remains a critical issue for realizing its commercialization and full potential. If graphene cannot be produced at low cost and high efficiency, its commercial and widespread use could be reduced or even ultimately fatally hindered.

Since graphene was discovered in 2004,⁸ significant advances in the mass production of this material have been achieved, as shown in Fig. 1. Numerous methods have been proposed to produce graphene,^9–35 offering choices suitable for specific applications. The bottom-up methods,^{10,15,18,23,25–35} such as chemical vapour deposition and epitaxial growth, can produce high-quality, large-size and thickness-controllable graphene. The resulting graphene is ideal for fabricating graphene electronics, field-effect transistors, flexible transparent electrodes, functional touchscreen panel devices, etc. However, these substrate-based techniques suffer from the limited scale and high cost and cannot meet the requirement for macroscopic quantities of graphene for applications such as advanced composites, coatings, conductive ink, and energy storage. Fortunately, the liquid-phase exfoliation (LPE) of graphite to afford graphene has been recently proven to be an effective scalable method.^{2,3,9,12,13,16,17,19–22,36–87} This method uses cheap and abundant graphite flakes as the precursor. The graphene products generated by this method can fit the requirement of scalability, reproducibility, processability and low production cost. Though the graphene quality by this method remains in question, consideration of graphene quality should be considered in relation to the needs of the specific application. For example, in applications for high catalytic activity^88,89 and high storage of capacitive charges,⁹⁰ graphene with edge defects are preferable, whereas a graphene nanomesh with a porous structure is desirable for semiconductor applications, which require a tunable bandgap.^78,91–93 Therefore, the LPE method is very promising. It should be noted that the LPE method depends on the exfoliation medium and the exfoliation technique applied. To date, the exfoliation medium, such as suitable organic solvents, surfactant/water solutions, aromatic solvents, ionic liquid, has been discussed in several state-of-art reviews.^{9,12,14,16,17,20} The most widely used exfoliation technique is sonication, which has also been reviewed recently.^13,41,94 Sonication depends on liquid cavitation for exfoliation. However, sonication-induced cavitation is a relatively harsh process that can involve high local temperatures (several thousand K), extreme pressure (several thousand atm) and rapid heating/cooling rates (several billion K s⁻¹).^95–97 However, these harsh conditions in the cavitation could result in damage to graphene. Thus, graphene produced by sonication has been verified to have much more defects, as could be expected.^74,98–100 Moreover, the distribution and intensity of the sonication-induced cavitations are highly dependent on the vessel size and shape, which often induce localized cavitation pictures.^101–104 If the position of the ultrasonic vibration source is fixed, the cavitation field in the liquid is almost static. These drawbacks are not favourable for efficient exfoliation and a large quantity of graphite flakes settle down to the bottom and remain unexfoliated. In addition, in the literature, it has been reported that sonic tips can only effectively process volumes no larger than a few 100 mL, leading to low production rates. While sonic baths can be used to process hundreds of millilitres, the power transfer from the bath to liquid is relatively poor, leading to long exfoliation times and hence low production rates. Another route for efficiently transferring mechanical energy directly into the liquid is thus desired.

Fig. 1 There are several methods for the mass production of graphene, which allow a wide choice in terms of size, quality and price for any particular application. Reproduced with permission from ref. 4. Copyright 2012 Nature Publishing Group.

Apart from sonication, fluid dynamics has emerged over the last five years as a novel exfoliation technique for the scalable production of graphene. Within the fluid dynamics, graphite flakes can move with the liquid and thus can be exfoliated repeatedly at different positions. Moreover, multiple fluid dynamic events are responsible for exfoliation. These features are intrinsically different from that of sonication, rendering fluid dynamics as a potentially efficient technique. Hence, keeping these key factors in mind, this progress report will examine three promising fluid-dynamics-based exfoliation routes, namely, vortex fluidics, pressure- and mixer-driven fluid dynamics, and highlight their recent progress and challenges.

2. Vortex fluidic device (VFD)

In order to avoid graphene defects and solvent degradation induced by cavitation in sonication, a vortex fluidic device (VFD) was developed to generate a less energy intensive shear process for exfoliating graphite. The VFD is schematized in Fig. 2a. It consists of a tube open at one end. When it is rapidly rotated, intense shear will be generated in the resulting thin films with finite sub-millilitre volumes of liquid. The shear fluidic film can be controlled by adjusting the speed and orientation of the tube and the other operating parameters.¹⁰⁵ According to fluid dynamics,¹⁰⁶ a rapidly rotating fluid can generate boundary and shear layers parallel to the axis of rotation, named as Stewartson–Ekman layers. Within this layer, the liquid flow is upwards at the internal surface of the rotating tube and downwards close to the liquid surface, as shown in Fig. 2a. Graphene dispersed in N-methyl-pyrrolidone (NMP) was successfully produced via exfoliating graphite by shear vortex fluidic films in the ‘confined mode’ of operation of the VFD (without jet feeds, as shown in Fig. 2a).⁵⁸ The graphite flakes dispersed in NMP will initially accelerate to the walls of the tube due to the large centrifugal force. Then, the partial lifting and slippage on the tube wall will cause the exfoliation mechanism, as shown in Fig. 2b and c. The slippage process can be highlighted by the “finger print” of partially stacked graphene in Fig. 2e. This slippage process requires the individual sheets to be partially lifted from the surface of the bulk material at some point to provide the necessary lateral force to start the slippage (Fig. 2b). Moreover, the graphite flakes are pushed against the tube wall by the centrifugal force and experience a shear-induced displacement along the tube, resulting in exfoliation at the tube surface (Fig. 2c).

Fig. 2 (a) Cartoon showing the confined mode thin fluid film in the rapidly rotating tube of the VFD, which has a shearing force associated with the Stewartson–Ekman layers. (b) The exfoliation process with slippage and partial lift. (c) Slippage on the inner surface of the tube. (d) Images of the resulting colloidal suspensions of graphene (top) sheets in NMP. (e) Partially stacked graphene as evidence of the slippage. (f) and (g) AFM images of graphene. Reproduced with permission from ref. 58 and 72. Copyright 2012 and 2013 The Royal Society of Chemistry.

In contrast, by using the ‘continuous flow mode’ of the VFD, graphene-based hybrid materials⁷² and functionalized graphene¹⁰⁷ can also be readily produced. In the continuous flow mode, another jet feed can deliver liquid into the rotating tube (Fig. 2a). This will generate additional shear in the thin films by the viscous drag as the liquid whirls along the tube. As shown in Fig. 3, the ‘confined mode’ is first used to exfoliate graphite into multi-layer graphene in water. Then, in the continuous flow mode, a feed jet at the base of the tube is used to deliver the recirculating liquid of graphene and microalgae mixed suspension. With this route, the multi-layer graphene sheets can be decorated on the surface of microalgal cells.

Fig. 3 Schematic of the overall hybridization process, involving the exfoliation of graphite flakes into multi-layer graphene sheets followed by the hybridization of these sheets with algal cells using a VFD. Reproduced with permission from ref. 72. Copyright 2013 The Royal Society of Chemistry.

Following the same idea, Tran et al.¹⁰⁸ applied a Taylor–Couette flow reactor to generate a vortex flow for exfoliating graphite flakes, as schematized in Fig. 4. In the reactor, a mixture of graphite and solvent is sheared between a rapidly rotating inner cylinder and a stationary outer cylinder. Thus, a vortex flow-induced high wall shear stress and pressure can be utilized to exfoliate the graphite flakes. It has been demonstrated that this method can efficiently produce few-layer graphene with a low degree of defects. However, the gap between these two cylinders is only 2.5 mm, limiting the throughput.

Fig. 4 Schematic of the exfoliation of graphite in a Taylor–Couette flow reactor. Reproduced with permission from ref. 108. Copyright 2016 The Royal Society of Chemistry.

This VFD offers an alternative and tunable low-energy source for mild exfoliation and thus high-quality graphene. However, the vortex fluidic film or rather the gap between is extremely thin (in the order of millimetres), which limits the quantity of graphite used for exfoliation and hence the graphene output.

The lateral size of graphene prepared by VFD is generally less than 1 μm, as shown in Fig. 2e–g and 5a and b. This is due to the weak shear force generated by VFD can only exfoliate smaller graphite flakes; because the collective van der Waals interaction between the layers for larger graphene flakes is much higher. The thickness of graphene prepared by VFD changes from less than 1 nm to more than 20 nm, as shown in Fig. 2f and g and 5b. However, the number of transmission electron microscopy (TEM) and atomic force microscopy (AFM) images in the literature is too small to give a statistical analysis of the distribution of sheet size. Since the graphene lateral size and thickness can determine whether graphene can be integrated into practical devices and whether its ultimate properties are attainable, it is highly recommended to obtain information on the size distribution of graphene prepared by VFD in the near future. As for the defects, it is anticipated that VFD generates a weak shear force and will not damage graphene. However, to date, there are only Raman spectra results (Fig. 5c) available to assess this aspect. A deeper study by means of different microscopic and spectral techniques is required for a full determination of the defects or oxides in graphene prepared by VFD, in order to really establish VFD as a defect-free method.

Fig. 5 (a) AFM image of exfoliated graphene sheets obtained in water using the VFD. (b) Height profiles of the selected area in (a). (c) Raman spectra of graphite flakes (a) before and (b) after VFD processing. Reproduced with permission from ref. 72. Copyright 2013 The Royal Society of Chemistry.

3. Pressure-driven fluid dynamics (PFD)

In order to realize the large-scale production of graphene, pressure-driven fluid dynamics (PFD) is utilized.^{55,56,61,69,75,78,84,86,109} Unlike VFD, which depends on a rapidly rotating tube, PFD relies on a series of flow channels for the exfoliation. The pressure difference between the inlet and the outlet can generate rich fluid dynamics events in the flow channels. Within the PFD device, graphite flakes can move with the liquid along the flow channel and thus can be exfoliated repeatedly at different position. This is totally different from the case in sonication wherein the location of the cavitation field and exfoliation events is almost static. This feature renders PFD as a much more efficient technique. The flow channel can be in the order of either micrometer^75,84,109 (Fig. 6a and 7) or millimeter^{55,56,61,69,78,86} (Fig. 6b). The number of the expansion and contraction channels in PFD devices can be adjusted during manufacturing. The fluid dynamics in PFD is characterized by cavitation, pressure release, viscous shear stress, turbulence, and collision. As illustrated in Fig. 6c, there are multiple fluid dynamics events responsible for normal- and shear-force dominated exfoliation. Cavitation and pressure release can generate normal forces for exfoliation. The velocity gradient-induced viscous shear stress, the turbulence-induced Reynolds shear stress and shear effects stemming from the turbulence and flow channel-induced collisions can generate shear forces for exfoliation, resulting in graphite self-exfoliation down to single or few layers through its lateral self-lubricating ability. All these dynamics events have fragmentation effects that also facilitate exfoliation as the collective interaction force between two adjacent layers is smaller in the smaller flakes. Hence, PFD possesses multiple ways for exfoliation, providing great advantages over sonication, which possesses only a single cavitation effect and ball milling or VFD, which possess a single shear effect, in terms of yield and efficiency.

Fig. 6 (a) Schematic of the apparatus with one constriction channel for producing graphene. High pressure (P_h) is exerted by a plunger pump, and P_o denotes ambient pressure. (b) Schematic of the apparatus with four constriction channels. (c) Schematic of the exfoliation mechanism of pressure-driven fluid dynamics. SEM images of (d) graphite particles and (e) graphene flakes produced by the apparatus in (b). (f) AFM image of the graphene sheets prepared by the apparatus in (a). Reproduced with permission from ref. 69 and 84. Copyright 2013 and 2014 Springer.

Fig. 7 Illustration of graphite delamination by a high-pressure homogenizer. The suspension is pumped through a nozzle and released into an expansion chamber. Reproduced with permission from ref. 109. Copyright 2015 The Royal Society of Chemistry.

The produced graphene can be verified by different characterizations such as scanning electron microscopy (SEM) and AFM (Fig. 6d–f). By capturing a large number of AFM images, the lateral size (or flake area) and thickness distribution of graphene prepared by PFD can be obtained. Moreover, by adjusting the pressure and treating time, the graphene concentration, thickness distribution and flake area distribution can be controlled, as shown in Fig. 8.⁸⁶ Depending on the pressure and treatment time, the flake area and thickness distribution can be changed. As the treatment time is increased, the thickness distribution becomes narrower and shifts to lower values Fig. 8a. For example, in the 0.5 h sample, flakes with a thickness <3 nm (less than 10 layers) occupy ∼80%. The percentages of thin flakes with a thickness <1.5 nm (less than 5 layers) for the 0.5 h, 4 h and 8 h treatment time samples are 29%, 63% and 79%, respectively. In contrast, the flake area sharply moves to small values. Fig. 8b shows that over 85% of the flakes occur in an area less than 10⁵ nm² in the 8 h sample and the mean value of the flake area decreased by an order of magnitude in comparison with that in the 0.5 h sample. These results of the size distribution establish PFD as a controllable method for preparing graphene with a specific size. As for the defects, only Raman results on the graphene-based films are available^86,109 and indicate low-level basal plane defects. Nevertheless, microscopic study of the individual flakes by scanning tunneling microscopy (STM) and X-ray photoelectron spectroscopy (XPS) is still required to obtain more detailed information on the atomic structure and chemical components of the basal plane of graphene prepared by PFD.

Fig. 8 The distribution of graphene (a) thickness and (b) flake area for different treatment times of 0.5 h, 4 h, and 8 h under a pressure of 15 MPa. (c) Graphene concentration as a function of treatment time and pressure. (d) Graphene thickness and flake area distribution under a pressure of 20 MPa and a treatment time of 4 h. Reproduced with permission from ref. 86. Copyright 2015 American Scientific Publishers.

Fig. 8c shows that higher pressure leads to a higher yield of graphene. These results are useful for scaling up this technique from 10 L per pot in a laboratory to several hundred liters in an industry. Most interestingly, if pressure is increased to higher values (e.g. 30 MPa), this PFD technique can be used to produce a graphene nanomesh,⁷⁸ which has recently emerged as a novel graphene nanostructure with a bandgap large enough for room-temperature transistor operation.^91,92 The mechanism for this is a combination of exfoliation and perforation of the graphene sheets (Fig. 9a). The obtained graphene nanomesh is shown in Fig. 9b and c. It is estimated that the total area of the pores within a 1 μm² nanomesh is ∼0.15 μm² and the pore density is ∼22 μm⁻². This provides a novel route for the large-scale production of graphene nanomeshes.

Fig. 9 (a) Schematic of the pressure-driven fluid dynamics for preparing a graphene nanomesh. Typical (b) AFM and (c) TEM images of the as-produced graphene nanomesh. Reproduced with permission from ref. 78. Copyright 2014 The Royal Society of Chemistry.

4. Mixer-driven fluid dynamics (MFD)

Another recently emerging method involves mixer-driven fluid dynamics (MFD). The device for realizing this method is relatively simple and easily available. A commercially available rotor/stator mixer can be used for graphene production,^79,80 as shown in Fig. 10. The head of the mixer constitutes a rotor and a stator as the critical component for the exfoliation. The rotor diameters (Fig. 10b and c) can be adjusted. By using NMP as the solvent, the mixer can result in graphene–NMP dispersions (Fig. 10d) in which graphene flakes have a lateral size of several hundred nanometers, as shown in the TEM image in Fig. 10e and f. The shear exfoliation mechanism can be further revealed in terms of the rotor diameter and the mixer-induced fluid dynamics. It was found that even when the Reynolds number Re_Mix of the flow field is less than 10⁴, which corresponds to a not fully developed turbulent flow, well-exfoliated graphene can still be obtained, as shown in the region below the Re_Mix line in Fig. 10g. However, when the shear rate [small gamma, Greek, dot above] is lower than 10⁴ s⁻¹, the graphite flakes are poorly exfoliated, as shown in the region below the [small gamma, Greek, dot above] = 10⁴ s⁻¹ line in Fig. 10g. In the case of the mixer at a number of different combinations of rotating speed and rotator diameter, the minimum shear rate [small gamma, Greek, dot above]_min is also around 10⁴ s⁻¹ (Fig. 10h). This suggests that any mixer that can achieve a shear rate above 10⁴ s⁻¹ can be used to produce graphene. The exfoliation mechanism can also be qualitatively explained in terms of the fluid dynamics events,⁷⁹ as illustrated in Fig. 11. As with ball milling and VFD, this is a shear-force dominated method but the cavitation and collision effects also favour an efficient exfoliation (Fig. 11). However, in the rotor–stator mixer (Fig. 10b and c), very high shear rates are mainly localized in the gap between the rotor and stator and in the holes in the stator. This implies a well-defined localized region of high shear rate, indicating that most of the exfoliation events are localized in the vicinity of the rotor/stator.

Fig. 10 (a) A Silverson model L5M high-shear mixer with mixing head in a 5 L beaker of graphene dispersion. (b) and (c) Mixing head with a rotor and a stator. (d) Graphene–NMP dispersions. (e) TEM images of an individual nanosheet. (f) Multi-layer graphene (bottom left) and monolayer graphene (right) as evidenced by its electron diffraction pattern (inset). (g) Phase diagram of rotor speed, N, versus the mixing head diameter, D, for dispersions showing good exfoliation. The region above the black line represents fully developed turbulence, i.e., Re_Mixer > 10⁴, whereas the region above the red line represents [small gamma, Greek, dot above]_min > 10⁴ s⁻¹. (h) Concentration of graphene as a function of shear rate for rotors with diameters of 32, 16 and 12 mm (mixing time 1 min). All the three data sets are consistent with the same minimum shear rate. (i) TEM image of a partially exfoliated BN flake, consistent with exfoliation by shear sliding. Reproduced with permission from ref. 80. Copyright 2014 Nature Publishing Group.

Fig. 11 3D sectional drawing of the high-shear mixer, and schematic of the mechanism for preparing graphene by shear force, collision and cavitation. Reproduced with permission from ref. 79. Copyright 2014 The Royal Society of Chemistry.

Seeing that the high shear rates are mainly localized around the rotor/stator structure, a mixer equipped with rotating blades was proposed to induce fully developed turbulence to generate a high shear rate all over the flow field. The simplest way to realize this is using a kitchen blender,^71,81,83,110 as shown in Fig. 12a and 13a. In the kitchen blender, if the rotating speed of the blades is sufficiently high, the high-shear region is not localized in any single portion of the holder. Though the shear rate decreases with the increasing distance from the blade, the high shear rate can cover the entire region of the holder if a turbulence is fully developed. Therefore, the turbulence is mainly responsible for the full-field high shear rate and thus the exfoliation mechanism, as shown in Fig. 12b. In terms of the characteristics of the turbulent flow in the kitchen blender, it was demonstrated that four fluid dynamics events are responsible for the exfoliation and fragmentation: (I) a velocity gradient, which can induce viscous shear stress; (II) intensive velocity fluctuations in the turbulence, which can induce Reynolds shear stress; (III) in the turbulence, the Reynolds number is very large, and thus the inertial forces dominate over viscous forces to enhance graphite–graphite collisions; (IV) it is possible that turbulent-pressure-fluctuation-induced pressure differences can also exfoliate graphite in a normal-force style. The mechanism can be verified by the TEM observations. The slipped configuration with a lateral relative displacement of translation (Fig. 12c) or rotation (Fig. 12d) indicates that lateral exfoliation really occurs, and there coexist two ways by which this can occur, i.e. translation and rotation. The exfoliation efficiency is much higher than that in the standard sonication or ball milling exfoliation methods.

Fig. 12 (a) Schematic of a kitchen blender for preparing graphene flakes with DMF as the solvent. (b) Illustration of the exfoliation mechanism. Deliberately captured partially exfoliated graphene flakes with translational (c) and rotational (d) lateral exfoliation. Reproduced with permission from ref. 83. Copyright 2014 Elsevier.

Fig. 13 (a) The Kenwood BL370 series kitchen blender used in this work photographed during a mix. The blender is mixing graphite powder in an aqueous surfactant solution. The surfactant is the household detergent: Fairy Liquid. Note the presence of significant amounts of foam. (b) An image of the rotating blade supplied with this blender. (c) TEM image of graphene nanosheets. (d) Concentration of mixer dispersed graphene versus the blade speed N. (e) Length (L) and (f) layer number (N) distributions of flakes as measured by AFM. Reproduced with permission from ref. 81. Copyright 2014 The Royal Society of Chemistry.

By using a kitchen blender, even household detergent can be used as a surfactant for graphene production,⁸¹ as shown by the presence of significant amounts of foam in Fig. 13a. The impeller is equipped with four blades (Fig. 13b). Most of the as-produced graphene flakes are folded, as shown in the TEM image in Fig. 13c. This is different from the case of sonication in which graphene flakes are sometimes folded. The larger fraction of folded flakes in mixer-exfoliated graphene relative to sonication-exfoliated graphene reflects the differences in the fluid dynamics of the two systems. By monitoring the graphene concentration under different blade rotating speeds, a critical blade speed of around 2k rpm from this special kitchen blender in Fig. 13a could be determined, as shown in Fig. 13d. This knowledge is important for designing the rotating-blade mixer for large-scale graphene production. The use of a kitchen blender and a household detergent makes the MFD route extremely simple and cost effective.

For the large-scale production of graphene for biological applications, Pattammattel and Kumar¹¹⁰ applied a kitchen blender to exfoliate graphite in protein solutions, as schematized in Fig. 14. Dependent on the charge of the protein used, the exfoliation efficiency varied greatly. Among the five proteins, namely, BSA (bovine serum), β-lactoglobulin (bovine milk), lysozyme (egg white), ovalbumin (egg white), and haemoglobin (bovine blood), the strongly negatively charged BSA gave the highest efficiency. Using the BSA aqueous solutions, the kitchen blender could achieve an exfoliation efficiency of more than 4 mg mL⁻¹ h⁻¹ and a maximum concentration of 7 mg mL⁻¹. The BSA-coated graphene with controllable surface charge was shown to be stable under pH values of 3–11 and temperatures of 5–50 °C. The combination of a kitchen blender and proteins makes MFD an effective tool for the scalable and biological production of graphene in water.

Fig. 14 Illustration of exfoliating graphite in aqueous solutions of proteins by a kitchen blender. 8 L solutions are processed for demonstrating the scalability of the method. Reproduced with permission from ref. 110. Copyright 2015 Wiley.

As for the size distribution of graphene flakes prepared by MFD, Varrla et al.⁸¹ and Yi et al.⁸³ used AFM to perform statistical analysis. Varrla et al.⁸¹ adopted a rotation speed of 18k rpm and a treatment time of 60 min to prepare graphene. The AFM-based statistical distributions of length (L) and layer number (N) are shown in Fig. 13e and f, respectively. The average length was found to be around 320 nm. The average layer number was around 6.⁸¹ In contrast, Yi et al. adopted a rotation speed of 5k rpm and investigated the effect of preparation time on the size distribution of the resultant graphene flakes.⁸³ Fig. 15 shows the AFM-based statistical results for the flakes' dimensions. The area was chosen rather than the length or width because most graphene flakes are irregularly shaped and thus measuring their length or width is difficult. As shown in Fig. 15a, the number fraction of ≤1.5 nm-thick flakes exceeds 80% for all the preparation times, reaching a high value of ∼92% at 3 h. Additionally, the number fraction of ≤1 nm-thick graphene flakes approximately remains constant between 14.6% and 20% for all the preparation times. The flake area notably decreases with preparation time, resulting in a shift in the area distribution towards lower values, as shown in Fig. 15b. Based on these statistical data, the average thickness per flake, 〈t〉, and the average area per flake, 〈A〉, can be calculated, as shown in Fig. 15c. It can be seen that 〈t〉 hardly varies at all and maintains at ∼1.5 nm, corresponding to an average layer number of <5. Nevertheless, the thickness distribution shifts towards lower values as the preparation time increases, as illustrated in Fig. 15a. In contrast, 〈A〉 decreases with preparation time, falling from ∼2.4 μm² at 0.5 h to ∼0.1 μm² at 8 h. By fitting 〈A〉 as a function of preparation time, an inversely proportional relationship appears, as shown in Fig. 15d. For biographene produced in water/proteins solutions by MFD in the kitchen blender, Pattammattel and Kumar¹¹⁰ calculated the average layer number and lateral size as a function of exfoliation time, as shown in Fig. 16b and c. The average layer number was estimated to be ∼3.6 despite the exfoliation time (Fig. 16b). The average size was ∼0.5 μm and appeared to be highly uniform, which is different from the results of Varrla et al.⁸¹ and Yi et al.⁸³ The reason for this can be attributed to the method. Pattammattel and Kumar¹¹⁰ used Raman data to obtain the size information by an empirical equation. In contrast, Varrla et al.⁸¹ and Yi et al.⁸³ directly measured the size by AFM or TEM.

Fig. 15 Size distribution of graphene prepared in DMF by MFD in a kitchen blender. Statistical histogram derived from many graphene flakes showing the thickness (a) and area (b) distribution. Calculated (c) average thickness, 〈t〉, and (d) average area, 〈A〉, as a function of treatment time. Reproduced with permission from ref. 83. Copyright 2014 Elsevier.

Fig. 16 Size and defect analysis of graphene produced in water/proteins solutions by MFD in a kitchen blender. (a) Raman spectra of graphene (solid line) and graphite (dotted line). (b) Layer number and (c) lateral size as a function of exfoliation time. (d) Statistic analysis of Raman results of graphene showing the defect information. Reproduced with permission.¹¹⁰ Copyright 2015 Wiley.

The defects and oxides of graphene prepared by MFD were studied by XPS and Raman mapping, as shown in Fig. 16 (ref. 110) and 17.⁸³ By statistical analysis of the Raman intensity ratio, Pattammattel and Kumar¹¹⁰ pointed that biographene produce by MFD only has minor edge defects (Fig. 16d). The XPS results in Fig. 17b show the same bonds and a similar composition in the pristine graphite and graphene-based film, indicating that the low level of oxides in graphene are caused not by residual solvent or oxidation but by water, CO₂ or oxygen from the atmosphere. These prove that MFD does not chemically functionalize the graphene flakes. By using the Raman mapping technique, an individual graphene flake can be captured and its Raman spectrum can be obtained, as shown in Fig. 17a and c. In Fig. 17c, the 2D bands in the Raman spectra of flakes #1 and #2 reflect the graphene nature.^22,111,112 However, there are no D bands and the G bands are not remarkably widened. This indicates that the flakes are almost free of basal plane defects. For the precise examination of the local defects or atomic structure, STM and high-resolution TEM characterizations are further required.

Fig. 17 Defect analysis of graphene produced in DMF by MFD in a kitchen blender. (a) A Raman mapping image. The Raman map plots the intensity integral of the spectra between 2600 and 2800 cm⁻¹. The excitation wavelength is 532 nm. (b) Carbon 1s core-level XPS spectra of the pristine graphite and graphene-based film. (c) Raman spectra for bulk graphite, individual flake #1, individual flake #2 and the filtered film. Reproduced with permission from ref. 83. Copyright 2014 Elsevier.

5. Conclusions and perspectives

Since the second half of 2011 when the virgin idea of producing graphene by utilizing the rich flow events in fluid dynamics was first initiated,^55,56 huge progress has been made over the proceeding five years. Various methods for generating fluid dynamics have been proposed in order to explore an efficient and scalable route for graphene production such as vortex fluidics, mixer, blender, high pressure. Compared to the widely used cavitation-dominated sonication for producing graphene, fluid dynamics possess multiple exfoliation effects originating from the shear, cavitation, collision and pressure release. Therefore, the fluid dynamics route is far more efficient than sonication routes and shows great technological potential in the near future. Considering the main factors for the industrialization of graphene production, i.e. production efficiency, production cost, scalability, reproducibility, processability, the recently emerging fluid dynamics route is very promising. With the continuous effort expanded in fluid dynamics research for graphene production, numerous exciting results and new methods have so far been reported and several technologies are currently envisioned. We believe that the recently emerging fluid dynamics route provides a significant step in the direction of realizing the commercial availability of large quantities of high-quality graphene. However, in order to proceed from discovery to a commercialized technology, numerous issues remain to be further explored and overcome; a partial list of which includes the following:

(1) How can we control and optimize the exfoliation effects in fluid dynamics so that the harsh and violent effects can be lowered to a minimum level? For example, though cavitation can exfoliate graphite into graphene flakes, it induces extremely high local temperature and pressure,^95–97 which can result in defective graphene. In contrast, relying on the lateral exfoliation mechanism, exfoliation by shear force is much milder. In PFD and the local region near the rotating blade in MFD, high-speed fluids can generate cavitation. A deep understanding and precise design of the flow field in PFD and MFD are critical for eliminating the cavitation region and achieving high shear rates throughout the flow field.

(2) How to achieve monolayer dominated and large-size graphene products by fluid dynamics still remains challenging. Exfoliation in fluid dynamics is always accompanied with fragmentation, which is not desired for producing large-size graphene. How to minimize the fragmentation effects should be considered. The average layer number and lateral size of graphene produced by fluid dynamics are 3–5 and several hundred nanometers, respectively. This indicates the case for few-layer graphene. The control of graphene size may be possible by combining fluid dynamic methods and a specified centrifugation strategy.^59,113–116

(3) The nature of defects induced by fluid dynamics requires detailed investigations. Currently, a consensus has been reached on the conclusion of edge-dominated defects in graphene produced by fluid dynamics. However, the conclusion is almost totally based on the Raman spectra of filtered graphene films and not on a single graphene flake. In the filtered film, the Raman signal is a superposition of contributions from numerous single- and few-layer graphene flakes. It is suggested that a microscopic study on the individual flakes by STM and XPS should be conducted in order to gain more detailed information on the atomic structure and chemical components of the basal plane of graphene prepared by fluid dynamics.

(4) Other simple routes for generating fluid dynamics should be explored. As the schematics show in Fig. 18, random shake and liquid spray may be another two possible routes. It is anticipated that the fluid dynamics events involved in the process of random shake and liquid spray can generate viscous shear, turbulence, collision and pressure release to mildly exfoliate graphite into graphene flakes.

Fig. 18 Schematic of another two possible or presumptive routes for generating fluid dynamics for graphene production: random shake and liquid spray.

(5) Are the current fluid dynamic methods ready for industrial scale-up or just a ‘blip on the oscilloscope’? The VFD route with weak shear force can afford high-quality graphene, but the small throughput limits the scalable production. Though with high efficiency, the PFD device depends on high pressure and small flow channels, which increase the complexity and cost of the device. The MFD route is much simpler and the device for MFD is much more easily available. As the rotating blade or rotor in the mixer can transfer mechanical energy directly into the liquid, a large volume can be processed and high production efficiency can be achieved. We recommend that bench-scale experiments and pilot-scale production should be tried with PFD and MFD based on the lab-level experiences. For PFD, a commercial high-pressure homogenizer is recommended. For MFD, industrial rotating-blade stirred tank reactors may be a good choice.

Acknowledgements

This study was supported by the Fundamental Research Funds for the Central Universities and the Special Funds for Co-construction Project of Beijing Municipal Commission of Education.

Notes and references

1. M. J. Allen, V. C. Tung and R. B. Kaner, Chem. Rev., 2010, 110, 132–145.
2. E. P. Randviir, D. A. C. Brownson and C. E. Banks, Mater. Today, 2014, 17, 426–432.
3. A. C. Ferrari, F. Bonaccorso, V. Falko, K. S. Novoselov, S. Roche and P. Bøggild, et al., Nanoscale, 2015, 7, 4598–4810.
4. K. S. Novoselov, V. I. Fal'ko, L. Colombo, P. R. Gellert, M. G. Schwab and K. Kim, Nature, 2012, 490, 192–200.
5. A. K. Geim and K. S. Novoselov, Nat. Mater., 2007, 6, 183–191.
6. A. K. Geim, Science, 2009, 324, 1530–1534.
7. M. S. Dresselhaus and P. T. Araujo, ACS Nano, 2010, 4, 6297–6302.
8. K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, Y. Zhang, S. V. Dubonos, I. V. Grigorieva and A. A. Firsov, Science, 2004, 306, 666–669.
9. J. N. Coleman, Acc. Chem. Res., 2013, 46, 14–22.
10. Y. Zhang, L. Zhang and C. Zhou, Acc. Chem. Res., 2013, 46, 2329–2339.
11. C. N. Rao, H. S. Matte and K. S. Subrahmanyam, Acc. Chem. Res., 2013, 46, 149–159.
12. J. N. Coleman, Adv. Funct. Mater., 2009, 19, 3680–3695.
13. A. Ciesielski and P. Samori, Chem. Soc. Rev., 2014, 43, 381–398.
14. D. R. Dreyer, S. Park, C. W. Bielawski and R. S. Ruoff, Chem. Soc. Rev., 2010, 39, 228–240.
15. L. Wang, L.-H. Tian, G.-D. Wei, F.-M. Gao, J.-J. Zheng and W.-Y. Yang, J. Inorg. Mater., 2011, 26, 1009–1019.
16. M. Cai, D. Thorpe, D. H. Adamson and H. C. Schniepp, J. Mater. Chem., 2012, 22, 24992.
17. K. P. Loh, Q. Bao, P. K. Ang and J. Yang, J. Mater. Chem., 2010, 20, 2277–2289.
18. C. Mattevi, H. Kim and M. Chhowalla, J. Mater. Chem., 2011, 21, 3324.
19. Y. Yao, Z. Lin, Z. Li, X. Song, K.-S. Moon and C.-p. Wong, J. Mater. Chem., 2012, 22, 13494.
20. X. Cui, C. Zhang, R. Hao and Y. Hou, Nanoscale, 2011, 3, 2118–2126.
21. S. Park and R. S. Ruoff, Nat. Nanotechnol., 2009, 4, 217–224.
22. Y. Hernandez, V. Nicolosi, M. Lotya, F. M. Blighe, Z. Sun, S. De, I. T. McGovern, B. Holland, M. Byrne, Y. K. Gun'Ko, J. J. Boland, P. Niraj, G. Duesberg, S. Krishnamurthy, R. Goodhue, J. Hutchison, V. Scardaci, A. C. Ferrari and J. N. Coleman, Nat. Nanotechnol., 2008, 3, 563–568.
23. L. G. De Arco, Z. Yi, A. Kumar and Z. Chongwu, IEEE Trans. Nanotechnol., 2009, 8, 135–138.
24. K. S. Kim, Y. Zhao, H. Jang, S. Y. Lee, J. M. Kim, J. H. Ahn, P. Kim, J. Y. Choi and B. H. Hong, Nature, 2009, 457, 706–710.
25. X. Li, W. Cai, J. An, S. Kim, J. Nah, D. Yang, R. Piner, A. Velamakanni, I. Jung, E. Tutuc, S. K. Banerjee, L. Colombo and R. S. Ruoff, Science, 2009, 324, 1312–1314.
26. S. Bae, H. Kim, Y. Lee, X. Xu, J. S. Park, Y. Zheng, J. Balakrishnan, T. Lei, H. R. Kim, Y. I. Song, Y. J. Kim, K. S. Kim, B. Ozyilmaz, J. H. Ahn, B. H. Hong and S. Iijima, Nat. Nanotechnol., 2010, 5, 574–578.
27. K. V. Emtsev, A. Bostwick, K. Horn, J. Jobst, G. L. Kellogg, L. Ley, J. L. McChesney, T. Ohta, S. A. Reshanov, J. Rohrl, E. Rotenberg, A. K. Schmid, D. Waldmann, H. B. Weber and T. Seyller, Nat. Mater., 2009, 8, 203–207.
28. P. W. Sutter, J. I. Flege and E. A. Sutter, Nat. Mater., 2008, 7, 406–411.
29. T. Ohta, A. Bostwick, T. Seyller, K. Horn and E. Rotenberg, Science, 2006, 313, 951–954.
30. C. Berger, Z. Song, X. Li, X. Wu, N. Brown, C. Naud, D. Mayou, T. Li, J. Hass, A. N. Marchenkov, E. H. Conrad, P. N. First and W. A. de Heer, Science, 2006, 312, 1191–1196.
31. W. Strupinski, K. Grodecki, A. Wysmolek, R. Stepniewski, T. Szkopek, P. E. Gaskell, A. Gruneis, D. Haberer, R. Bozek, J. Krupka and J. M. Baranowski, Nano Lett., 2011, 11, 1786–1791.
32. H. Ago, Y. Ito, N. Mizuta, K. Yoshida, B. Hu, C. M. Orofeo, M. Tsuji, K. Ikeda and S. Mizuno, ACS Nano, 2010, 4, 7407–7414.
33. H. K. Yu, K. Balasubramanian, K. Kim, J. L. Lee, M. Maiti, C. Ropers, J. Krieg, K. Kern and A. M. Wodtke, ACS Nano, 2014, 8, 8636–8643.
34. B. Hu, H. Ago, Y. Ito, K. Kawahara, M. Tsuji, E. Magome, K. Sumitani, N. Mizuta, K.-i. Ikeda and S. Mizuno, Carbon, 2012, 50, 57–65.
35. C. Vo-Van, A. Kimouche, A. Reserbat-Plantey, O. Fruchart, P. Bayle-Guillemaud, N. Bendiab and J. Coraux, Appl. Phys. Lett., 2011, 98, 181903.
36. D. Li, M. B. Muller, S. Gilje, R. B. Kaner and G. G. Wallace, Nat. Nanotechnol., 2008, 3, 101–105.
37. A. B. Bourlinos, V. Georgakilas, R. Zboril, T. A. Steriotis and A. K. Stubos, Small, 2009, 5, 1841–1845.
38. A. B. Bourlinos, V. Georgakilas, R. Zboril, T. A. Steriotis, A. K. Stubos and C. Trapalis, Solid State Commun., 2009, 149, 2172–2176.
39. M. Lotya, Y. Hernandez, P. J. King, R. J. Smith, V. Nicolosi, L. S. Karlsson, F. M. Blighe, S. De, Z. Wang, I. T. McGovern, G. S. Duesberg and J. N. Coleman, J. Am. Chem. Soc., 2009, 131, 3611–3620.
40. S. Vadukumpully, J. Paul and S. Valiyaveettil, Carbon, 2009, 47, 3288–3294.
41. G. Cravotto and P. Cintas, Chem.–Eur. J., 2010, 16, 5246–5259.
42. S. De, P. J. King, M. Lotya, A. O'Neill, E. M. Doherty, Y. Hernandez, G. S. Duesberg and J. N. Coleman, Small, 2010, 6, 458–464.
43. U. Khan, A. O'Neill, M. Lotya, S. De and J. N. Coleman, Small, 2010, 6, 864–871.
44. C. Knieke, A. Berger, M. Voigt, R. N. K. Taylor, J. Röhrl and W. Peukert, Carbon, 2010, 48, 3196–3204.
45. M. Lotya, P. J. King, U. Khan, S. De and J. N. Coleman, ACS Nano, 2010, 4, 3155–3162.
46. W. Zhao, M. Fang, F. Wu, H. Wu, L. Wang and G. Chen, J. Mater. Chem., 2010, 20, 5817.
47. W. Zhao, F. Wu, H. Wu and G. Chen, J. Nanomater., 2010, 2010, 1–5.
48. Y. Zhu, S. Murali, W. Cai, X. Li, J. W. Suk, J. R. Potts and R. S. Ruoff, Adv. Mater., 2010, 22, 3906–3924.
49. L. Guardia, M. J. Fernández-Merino, J. I. Paredes, P. Solís-Fernández, S. Villar-Rodil, A. Martínez-Alonso and J. M. D. Tascón, Carbon, 2011, 49, 1653–1662.
50. U. Khan, H. Porwal, A. O'Neill, K. Nawaz, P. May and J. N. Coleman, Langmuir, 2011, 27, 9077–9082.
51. Y.-T. Liu, X.-M. Xie and X.-Y. Ye, Carbon, 2011, 49, 3529–3537.
52. D. Nuvoli, L. Valentini, V. Alzari, S. Scognamillo, S. B. Bon, M. Piccinini, J. Illescas and A. Mariani, J. Mater. Chem., 2011, 21, 3428.
53. A. O'Neill, U. Khan, P. N. Nirmalraj, J. Boland and J. N. Coleman, J. Phys. Chem. C, 2011, 115, 5422–5428.
54. J.-W. T. Seo, A. A. Green, A. L. Antaris and M. C. Hersam, J. Phys. Chem. Lett., 2011, 2, 1004–1008.
55. Z. Shen, J. Li, M. Yi, X. Zhang and S. Ma, Nanotechnology, 2011, 22, 365306.
56. M. Yi, J. Li, Z. Shen, X. Zhang and S. Ma, Appl. Phys. Lett., 2011, 99, 123112.
57. F. Bonaccorso, A. Lombardo, T. Hasan, Z. Sun, L. Colombo and A. C. Ferrari, Mater. Today, 2012, 15, 564–589.
58. X. Chen, J. F. Dobson and C. L. Raston, Chem. Commun., 2012, 48, 3703–3705.
59. U. Khan, A. O'Neill, H. Porwal, P. May, K. Nawaz and J. N. Coleman, Carbon, 2012, 50, 470–475.
60. J. Li, F. Ye, S. Vaziri, M. Muhammed, M. C. Lemme and M. Östling, Carbon, 2012, 50, 3113–3116.
61. J. Li, M. Yi, Z. Shen, S. Ma, X. Zhang and Y. Xing, Sci. China: Technol. Sci., 2012, 55, 2815–2819.
62. S. M. Notley, Langmuir, 2012, 28, 14110–14113.
63. A. S. Wajid, S. Das, F. Irin, H. S. T. Ahmed, J. L. Shelburne, D. Parviz, R. J. Fullerton, A. F. Jankowski, R. C. Hedden and M. J. Green, Carbon, 2012, 50, 526–534.
64. M. Yi, Z. Shen, S. Ma and X. Zhang, J. Nanopart. Res., 2012, 14, 1003.
65. M. Yi, Z. Shen, X. Zhang and S. Ma, J. Mater. Sci., 2012, 47, 8234–8244.
66. X. Zheng, Q. Xu, J. Li, L. Li and J. Wei, RSC Adv., 2012, 2, 10632.
67. R. Aparna, N. Sivakumar, A. Balakrishnan, A. Sreekumar Nair, S. V. Nair and K. R. V. Subramanian, J. Renewable Sustainable Energy, 2013, 5, 033123.
68. W. Du, X. Jiang and L. Zhu, J. Mater. Chem. A, 2013, 1, 10592.
69. L. Liu, Z. Shen, S. Liang, M. Yi, X. Zhang and S. Ma, J. Mater. Sci., 2013, 49, 321–328.
70. M. Lotya, A. Rakovich, J. F. Donegan and J. N. Coleman, Nanotechnology, 2013, 24, 265703.
71. Z. Shen, M. Yi, X. Zhang and S. Ma, Turbulence method for preparing high-quality graphene, China Pat., CN103350995A, 2013.
72. M. H. Wahid, E. Eroglu, X. Chen, S. M. Smith and C. L. Raston, Green Chem., 2013, 15, 650.
73. L. Xu, J.-W. McGraw, F. Gao, M. Grundy, Z. Ye, Z. Gu and J. L. Shepherd, J. Phys. Chem. C, 2013, 117, 10730–10742.
74. M. Yi, Z. Shen, S. Liang, L. Liu, X. Zhang and S. Ma, Chem. Commun., 2013, 49, 11059–11061.
75. M. Yi, Z. Shen, W. Zhang, J. Zhu, L. Liu, S. Liang, X. Zhang and S. Ma, Nanoscale, 2013, 5, 10660–10667.
76. M. Yi, Z. Shen, X. Zhang and S. Ma, J. Phys. D: Appl. Phys., 2013, 46, 025301.
77. J. T. Han, J. I. Jang, H. Kim, J. Y. Hwang, H. K. Yoo, J. S. Woo, S. Choi, H. Y. Kim, H. J. Jeong, S. Y. Jeong, K. J. Baeg, K. Cho and G. W. Lee, Sci. Rep., 2014, 4, 5133.
78. S. Liang, M. Yi, Z. Shen, L. Liu, X. Zhang and S. Ma, RSC Adv., 2014, 4, 16127.
79. L. Liu, Z. Shen, M. Yi, X. Zhang and S. Ma, RSC Adv., 2014, 4, 36464.
80. K. R. Paton, E. Varrla, C. Backes, R. J. Smith, U. Khan, A. O'Neill, C. Boland, M. Lotya, O. M. Istrate, P. King, T. Higgins, S. Barwich, P. May, P. Puczkarski, I. Ahmed, M. Moebius, H. Pettersson, E. Long, J. Coelho, S. E. O'Brien, E. K. McGuire, B. M. Sanchez, G. S. Duesberg, N. McEvoy, T. J. Pennycook, C. Downing, A. Crossley, V. Nicolosi and J. N. Coleman, Nat. Mater., 2014, 13, 624–630.
81. E. Varrla, K. R. Paton, C. Backes, A. Harvey, R. J. Smith, J. McCauley and J. N. Coleman, Nanoscale, 2014, 6, 11810–11819.
82. S. Wang, M. Yi, Z. Shen, X. Zhang and S. Ma, RSC Adv., 2014, 4, 25374.
83. M. Yi and Z. Shen, Carbon, 2014, 78, 622–626.
84. M. Yi, Z. Shen and J. Zhu, Chin. Sci. Bull., 2014, 59, 1794–1799.
85. Y. L. Zhong, Z. Tian, G. P. Simon and D. Li, Mater. Today, 2015, 18(2), 73–78.
86. S. Liang, Z. Shen, M. Yi, L. Liu, X. Zhang, C. Cai and S. Ma, J. Nanosci. Nanotechnol., 2015, 15, 2686–2694.
87. M. Yi and Z. Shen, J. Mater. Chem. A, 2015, 3, 11700–11715.
88. W. Yuan, Y. Zhou, Y. Li, C. Li, H. Peng, J. Zhang, Z. Liu, L. Dai and G. Shi, Sci. Rep., 2013, 3, 2248.
89. D. A. C. Brownson, L. J. Munro, D. K. Kampouris and C. E. Banks, RSC Adv., 2011, 1, 978.
90. C. Su, M. Acik, K. Takai, J. Lu, S. J. Hao, Y. Zheng, P. Wu, Q. Bao, T. Enoki, Y. J. Chabal and K. P. Loh, Nat. Commun., 2012, 3, 1298.
91. L. Jiang and Z. Fan, Nanoscale, 2014, 6, 1922–1945.
92. J. Bai, X. Zhong, S. Jiang, Y. Huang and X. Duan, Nat. Nanotechnol., 2010, 5, 190–194.
93. J. Yang, M. Ma, L. Li, Y. Zhang, W. Huang and X. Dong, Nanoscale, 2014, 6, 13301–13313.
94. S. E. Skrabalak, Phys. Chem. Chem. Phys., 2009, 11, 4930.
95. E. B. Flint and K. S. Suslick, Science, 1991, 253, 1397–1399.
96. K. S. Suslick, W. B. McNamara and Y. T. Didenko, Nature, 1999, 401, 772–775.
97. K. S. Suslick and D. J. Flannigan, Annu. Rev. Phys. Chem., 2008, 59, 659–683.
98. M. V. Bracamonte, G. I. Lacconi, S. E. Urreta and L. E. F. Foa Torres, J. Phys. Chem. C, 2014, 118, 15455–15459.
99. T. Skaltsas, X. Ke, C. Bittencourt and N. Tagmatarchis, J. Phys. Chem. C, 2013, 117, 23272–23278.
100. E. Y. Polyakova Stolyarova, K. T. Rim, D. Eom, K. Douglass, R. L. Opila, T. F. Heinz, A. V. Teplyakov and G. W. Flynn, ACS Nano, 2011, 5, 6102–6108.
101. V. S. Sutkar and P. R. Gogate, Chem. Eng. J., 2009, 155, 26–36.
102. B. Nanzai, K. Okitsu, N. Takenaka, H. Bandow, N. Tajima and Y. Maeda, Ultrason. Sonochem., 2009, 16, 163–168.
103. Y. Kojima, S. Koda and H. Nomura, Jpn. J. Appl. Phys., 1998, 37, 2992–2995.
104. Y. Asakura, T. Nishida, T. Matsuoka and S. Koda, Ultrason. Sonochem., 2008, 15, 244–250.
105. L. Yasmin, X. Chen, K. A. Stubbs and C. L. Raston, Sci. Rep., 2013, 3, 2282.
106. D. A. Bennetts and L. M. Hocking, Proc. R. Soc. A, 1973, 333, 469–489.
107. X. Chen, F. M. Yasin, P. K. Eggers, R. A. Boulos, X. Duan, R. N. Lamb, K. S. Iyer and C. L. Raston, RSC Adv., 2013, 3, 3213.
108. T. S. Tran, S. J. Park, S. S. Yoo, T.-R. Lee and T. Kim, RSC Adv., 2016, 6, 12003–12008.
109. T. J. Nacken, C. Damm, J. Walter, A. Rüger and W. Peukert, RSC Adv., 2015, 5, 57328–57338.
110. A. Pattammattel and C. V. Kumar, Adv. Funct. Mater., 2015, 25, 7088–7098.
111. A. C. Ferrari, J. C. Meyer, V. Scardaci, C. Casiraghi, M. Lazzeri, F. Mauri, S. Piscanec, D. Jiang, K. S. Novoselov, S. Roth and A. K. Geim, Phys. Rev. Lett., 2006, 97, 187401.
112. L. M. Malard, M. A. Pimenta, G. Dresselhaus and M. S. Dresselhaus, Phys. Rep., 2009, 473, 51–87.
113. A. A. Green and M. C. Hersam, Nano Lett., 2009, 9, 4031–4036.
114. A. A. Green and M. C. Hersam, J. Phys. Chem. Lett., 2010, 1, 544–549.
115. C. Backes, B. M. Szydlowska, A. Harvey, S. Yuan, V. Vega-Mayoral, B. R. Davies, P. L. Zhao, D. Hanlon, E. J. Santos, M. I. Katsnelson, W. J. Blau, C. Gadermaier and J. N. Coleman, ACS Nano, 2016, 10, 1589–1601.
116. X. Sun, D. Luo, J. Liu and D. G. Evans, ACS Nano, 2010, 4, 3381–3389.

---