uncorrected proof

Ultrasonics - Sonochemistry xxx (2018) xxx-xxx

Contents lists available at ScienceDirect

Ultrasonics - Sonochemistry

PR OO F

journal homepage: www.elsevier.com

Graphene nanosheets preparation using magnetic nanoparticle assisted liquid phase exfoliation of graphite: The coupled effect of ultrasound and wedging nanoparticles Alireza Hadia⁠ , Jafar Zahirifara⁠ , Javad Karimi-Sabetb⁠ ,⁠ ⁎⁠ , Abolfazl Dastbaza⁠ a b

Department of Chemical Engineering, Faculty of Engineering, University of Tehran, Tehran, Iran Material and Nuclear Fuel Research School (MNFRS), Nuclear Science and Technology Research Institute, Tehran, Iran

ABSTRACT

Keywords: Graphite Graphene Liquid phase exfoliation Fe3⁠ O4⁠ nanoparticles

This study aims to investigate a novel technique to improve the yield of liquid phase exfoliation of graphite to graphene sheets. The method is based on the utilization of magnetic Fe3⁠ O4⁠ nanoparticles as “particle wedge” to facilitate delamination of graphitic layers. Strong shear forces resulted from the collision of Fe3⁠ O4⁠ particles with graphite particles, and intense ultrasonic waves lead to enhanced exfoliation of graphite. High quality of graphene sheets along with the ease of Fe3⁠ O4⁠ particle separation from graphene solution which arises from the magnetic nature of Fe3⁠ O4⁠ nanoparticles are the unique features of this approach. Initial graphite flakes and produced graphene sheets were characterized by various methods including field emission scanning electron microscopy (FE-SEM), transmission electron microscopy (TEM), Raman spectroscopy, atomic force microscopy (AFM), Fourier transform infrared spectroscopy (FTIR), X-ray diffraction (XRD), and Zeta potential analysis. Moreover, the effect of process factors comprising initial graphite concentration, Fe3⁠ O4⁠ nanoparticles concentration, sonication time, and sonication power were investigated. Results revealed that graphene preparation yield and the number of layers could be manipulated by the presence of magnetic nanoparticles.

UN CO RR EC TE D

ARTICLE INFO

1. Introduction

Graphene has become a hot research field because of its fabulous properties. Some intrinsic features of graphene are superior electrical conductivity with charge carrier mobility of 2 × 105⁠ cm2⁠ /V.S [1], high thermal conductivity of about 5000 w/m k [2], and extremely large young modulus of 1 TPa [3]. These intriguing properties of graphene have resulted in diverse applications in a broad range. Various methods have been used to prepare graphene including chemical exfoliation [4,5], liquid phase exfoliation [6–10], electrochemical technique [11–13], chemical vapor deposition (CVD) [14,15], supercritical fluid exfoliation [16–19], thermal exfoliation [20], etc. But the major challenge is how to prepare high quality graphene in a large scale. In fact, there is an inverse relationship between the productivity and quality of prepared graphene in its synthesis methods. Researchers have been trying to overcome this limitation in graphene preparation since it has been discovered in 2004. To date, an effective approach which can completely solve the problem was not reported. It means that some methods like micromechanical cleavage lead to high quality graphene, the yield of synthesis is low, though. On the other hand, some routs such as chemical exfoliation result in high yield graphene, but the product has high defect densities, disorders and impurities which ruin the inherent properties of graphene.

⁎

Liquid phase exfoliation (LPE) is one of the most important methods among top down graphene production techniques. This procedure includes direct exfoliation of graphite to graphene layers by shear forces through ultrasonication. Liquid phase exfoliation has been used for exfoliation of other 2D materials such as h-BN, and metal dichalcogenides like MoS2⁠ [21–24]. LPE has some important advantages over other common graphene production methods like chemical vapor deposition (CVD) and, chemical exfoliation (oxidation, exfoliation, and reduction). This is a simple, cheap, and environmental friendly approach without using hazardous materials such as strong acids and hydrazine derivatives. Although, this method does not contain any chemical modification of graphite, but long term sonication might introduces some defects to the structure of prepared graphene. Another limitation of LPE is the low exfoliation yield which makes it inappropriate method for graphene synthesis in the commercial scale. Some research has been conducted in order to solve the above-mentioned problems. Rational selection of solvent as a media for graphite exfoliation is the main criterion that has been studied the most. It has been found that solvents whose solubility parameters are close to that of graphene are suitable for LPE process [25]. In this regard, mixed solvent systems have been shown to be effective for graphite exfoliation through altering the solubility parameters of the main solvent. For instance, Tasis et al. [26] reported the exfoliation of graphite in binary mixtures. They concluded that mix

Corresponding author. Email address: [email protected], [email protected] (J. Karimi-Sabet)

https://doi.org/10.1016/j.ultsonch.2018.02.028 Received 27 November 2017; Received in revised form 7 January 2018; Accepted 15 February 2018 Available online xxx 1350-4177/ © 2017.

A. Hadi et al.


ture of 90% v/v of 1,2-dichlorobenzene (1,2-DCB) and 10% v/v of isobutanol has better exfoliation yield than neat solvents. Yi et al. [27] studied the exfoliation of graphite in water- acetone mixtures. They reported that the optimum composition of mixture occurs at an acetone mass fraction of ∼75% which is related to tailoring the Hansen solubility parameters of solvent. In addition, some studies carried out in order to enhance the graphene exfoliation yield by combining LPE method with exceptional properties of supercritical fluids. In this method, ultrasonic waves in supercritical fluid media with low surface tension and high diffusivity make intercalation and exfoliation processes easier [17,28,29]. Another route for improving the exfoliation yield of LPE process is to use different additives in the graphite dispersion. Xu et al. [30] used naphthalene as a molecular wedge for improving the yield of liquid phase exfoliation in organic solvents. Results showed that graphene dispersion with concentrations as high as 0.15 mg/mL could be obtained by this method. Yang et al. [31] utilized 1-pyrenesulfonic acid sodium salt in order to exfoliate graphite in water based liquid phase exfoliation. There are very few studies on the addition of particles for improving the efficiency of LPE process. for example, Carrasco et al. reported the preparation of graphene sheets by using cellulose nanocrystals through the exfoliation of graphite in aqueous solutions [32]. The results of this study showed that this additive could enhance the stabilization of graphene dispersion and exfoliation yield. Castarlenas et al. [33] prepared graphene from liquid phase exfoliation of graphite in methanol. They used titanosilicate JDF-L1 as an additive and analysis methods revealed the production of defect-free multi-layer graphene. Mendoza-Sanchez et al. have reported the co-exfoliation of graphite and manganese oxide to form a hybrid material using LPE process in isopropanol solvent [34]. In another work, Sun et al. have utilized Ag nanoparticles to prepare highly stable dispersion of graphene in isopropanol solvent [35]. They stated that uniform dispersion of Ag nanoparticles on the surface of graphene sheets could improve stability of flakes in low boiling point solvents. Although graphene-based composite produced in these studies are applicable in many cases, but prepared graphene sheets contain a noticeable amount of additives which is unfavorable for those applications demanding pure graphene. In this research, novel nanoparticle assisted liquid phase exfoliation method has been introduced for preparation of graphene sheets. The effect of process parameters especially the presence of magnetic nanoparticles on the exfoliation yield and number of graphene layers is presented. Prepared graphene sheets were characterized by different methods in order to evaluate the graphene properties and the exfoliation extent. The objective of this study is to improve the efficiency of LPE process by facilitating the delamination of graphitic layers for preparing high quality graphene. Particle induced shear forces would lead to better exfoliation which could results in higher yield in shorter sonication time.

2. Experimental 2.1. Materials

PR OO F

Graphite powder with particle size of less than 50 µm and N-methyl-2-pyrrolidone (NMP) solvent (>98%) were supplied by Merck. Fe3⁠ O4⁠ nanoparticles (particle size < 50 nm) was purchased from US Research Nanomaterials. All reagents were used without any pretreatments. 2.2. Methods

UN CO RR EC TE D

2.2.1. LPE process In a typical experiment, 100 ml of graphite dispersion with the concentration of 0.5 mg/ml was prepared by 5 minutes sonication (the sonication power of this stage was similar to that of main ultrasonication step listed in Table 1). Then predetermined amount of iron particles were added to the dispersion. Hereafter, the dispersion was subjected to ultrasonication by a probe sonicator (maximum output 1200W). Sonication power and the time of process were adjusted for each experiment. Afterward, iron oxide nanoparticles were completely omitted from the dispersion by magnetic separation for five hours. Final dispersion was centrifuged at 500 rpm for 30 min in order to remove unexfoliated graphite flakes. The supernatant was centrifuged at 11000 rpm and resultant materials were washed repeatedly with ethanol. The products were dried at 50 °C overnight to obtain pure graphene powder. Table 1 represents experimental conditions for producing graphene sheets using nanoparticle assisted liquid phase exfoliation. It is worth mentioning that the actual power dissipated to the solution has been measured using calorimetric method introduced by Lorimer et al. [36]. The yield of graphene exfoliation in different conditions was obtained using Beer- Lambert law. After experimentally finding the ab⁠ 1 m− ⁠ 1) of graphene dispersion sorbance coefficient (α = 825/22 ml mg− in NMP solvent (Fig. 1), the absorbance of resultant dispersions of different synthesis conditions was measured at 660 nm. Using Beer- Lambert law, the concentration of graphene could be obtained, and finally the yield of process is calculated using Eq. (1). (1)

where Y is the graphene production yield, C is graphene concentration in final dispersion, and C0⁠ is initial graphite concentration.

Table 1 Experimental condition for preparation of graphene using Fe3⁠ O4⁠ nanoparticles assisted LPE method. Experiment no. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Graphite concentration (mg/ml)

Fe3⁠ O4⁠ concentration (mg/ml)

ultrasonication time (min)

Ultrasonication power (W)

Actual power dissipated (W)

0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 1 1.5

0.5 0.5 0.5 0.5 0.5 0.5 0.5 0 0.1 0.25 0.5 1 0.5 0.5 0.5

30 60 90 120 60 60 60 60 60 60 60 60 60 60 60

600 600 600 600 240 420 600 600 600 600 600 600 600 600 600

282 282 282 282 175 218 282 282 282 282 282 282 282 282 282

2

A. Hadi et al.


3. Results and discussion 3.1. Mechanism of particle assisted LPE

PR OO F

Fig. 2 represents a schematic illustration of Fe3⁠ O4⁠ nanoparticle assisted liquid phase exfoliation of graphene. Generally, LPE process using Fe3⁠ O4⁠ particles could be explained by a multi-step mechanism. At the first stage, graphite flakes are separated from agglomerated and stacked structures of initial graphite powder. These flakes are broken down to smaller separated graphitic particles. Due to hitting Fe3⁠ O4⁠ nanoparticles to the graphite flakes, this step is carried out more quickly in the presence of nanoparticles. At the second step, ultrasonic waves induce exfoliation of graphite in different extent depending on LPE condition such as ultrasonic time and power. The principle of LPE is delamination of layered materials by using shear forces generated by ultrasonic waves and cavitation which produces intense shear forces near the adjacent layers of a layered structure. In presence of Fe3⁠ O4⁠ nanoparticles, the ultrasonic energy results in movement of nanoparticles in random directions. This phenomenon affects graphite exfoliation into graphene sheets. The considerable part of nanoparticles collides to graphite flakes perpendicular to their C-axis direction. At first stages of process, these collisions lead to production of multilayer graphene, and continuing the process results in the exfoliation of few-layered graphene to single layer sheets through “particle wedge” role of Fe3⁠ O4⁠ nanoparticles (Fig. 2).

Fig. 1. Linear relationship between absorbance per cell length (660 nm) and concentration of graphene dispersion in NMP solvent

UN CO RR EC TE D

2.2.2. Characterization UV-vis spectroscopy (Mecasys, OPTIZEN POP, Korea) was used in order to measure the yield of graphite exfoliation in different conditions. Obeying Beer-Lambert law, the yield of graphite conversion was simply obtained by measuring the absorbance of graphene dispersion at 660 nm. X-ray diffraction of graphite and exfoliated samples were conducted by a X’Pert PRO MPD diffractometer (PANalytical Company) using Cu Kα radiation (λ = 1.5406 Å) that was operated at 40 KV and 40 mA. SEM micrograph of graphite particles was captured using a scanning electron microscope (Seron Technology, AIS2100, Korea). FE-SEM images were recorded using a field emission scanning electron microscope (Hitachi S4160). SEM and FE-SEM samples were prepared by casting droplets of graphite and graphene dispersions on the aluminum foil and drying. Raman spectra of graphite and graphene powder were taken by a Raman microscope (Teksan company, model: Takram) with the laser wavelength of 532 nm. Atomic force microscopy (AFM) was carried out using an atomic force microscope (Park scientific Instruments CP-Research (VEECO), USA) on the tapping mode. Samples for AFM imaging were prepared by dropping a small amount of graphene dispersion on the clean surface of delaminated mica substrate followed by air drying. Transmission electron microscopy (TEM) images were captured by means of a transmission electron microscope (Zeiss – EM10C – 80 KV, Germany). Some droplets of graphene dispersion were casted on lacey carbon-coated grid (300 mesh, Cu) for preparation of TEM samples. Graphite and exfoliated graphene powder were utilized for FTIR analysis using a Bruker Tensor 27 spectrometer. Zeta potential analysis of graphene dispersion in ethanol solvent was performed using a Zetasizer (Nano ZS (red badge) ZEN 3600).

3.2. Characterization Delamination of graphite particles into graphene layers were investigated by different characterization methods. Besides the verification of graphite exfoliation, these techniques help us to identify the exfoliated graphene properties. It is worth mentioning that all characterization methods have been applied on the products obtained at graphite concentration of 0.5 mg/ml, Fe3⁠ O4⁠ nanoparticles concentration of 0.5 mg/ ml, sonication time of 60 min, and sonication power of 600 W. 3.2.1. Field emission scanning electron microscopy (FE-SEM) Fig. 3 displays SEM image of graphite particles dispersed in NMP solvent just before LPE process, and FE-SEM micrographs of exfoliated graphene sheets obtained from ultrasonication in presence of Fe3⁠ O4⁠ nanoparticles. Well-defined edges of graphene sheets make it possible to measure the size of exfoliated sheets in any directions. As can be seen from Fig. 3, lateral size of graphene sheets have been reduced significantly after exfoliation. Two explanations could be considered for smaller lateral size of graphene sheets in comparison to initial graphite particles. The first one is breaking or cutting of graphite particles from their weak points (or defects) as a result of receiving the

Fig. 2. Schematic illustration of nanoparticle assisted LPE method 3


UN CO RR EC TE D

PR OO F

A. Hadi et al.

Fig. 3. a) SEM image of graphite particles, b–d) FE-SEM images of exfoliated graphene sheets prepared by Fe3⁠ O4⁠ nanoparticle assisted LPE method

intense sonication waves. The second presumption is related to exfoliation mechanism. Generally, due to the less extended inter-layer forces, peeling the layers of smaller graphite particles seems to be easier comparing with the larger ones. In such cases, complete exfoliation is achieved by receiving less energy during ultrasonication. Therefore, large graphite particles are separated as partially exfoliated, un-exfoliated or aggregated particles through centrifugation stage. Consequently, the final product of LPE process is a dispersion consisting of graphene sheets with lateral sizes (∼0.5–5 µm) considerably smaller than that of initial graphite (∼2–30 µm). Moreover, Fig. 3 could be used for investigation of the morphology of graphene sheets. The wrinkle-less surface of graphene sheets is the main point that could be seen from the FE-SEM images. This feature could guarantee the intrinsic properties of graphene produced using this method.

troducing some impurities to graphene; for example remained oxygen functional groups on reduced graphene oxide prepared by chemical exfoliation or remained surfactant molecules in additive assisted exfoliation. Low amount of Fe3⁠ O4⁠ nanoparticles could be assumed as impurities which associated with graphene produced through any exfoliation techniques. Therefore, other characterization methods were performed to explore the degree of purity of exfoliated graphene sheets. In addition, Energy-dispersive X-ray spectroscopy (EDS) analysis of prepared graphene has been presented in the Supporting information which proves the high purity of the products. 3.2.3. X-ray diffraction (XRD) X-ray diffraction was used to investigate the phase change of graphite after exfoliation, and check whether Fe3⁠ O4⁠ nanoparticles as impurities are present in the final graphene samples. According to Fig. 6 (a), peak position of patterns of graphite and exfoliated samples did not change and both specimens have a diffraction peak around 26.5°. The peak intensity has weakened from graphite to graphene pattern, however. This observation showed that in contrast to chemical exfoliation methods in which the peak position of exfoliated samples is changed significantly indicating the introduction of oxygen-containing functional groups, graphene sheets with same carbon lattice as graphite is prepared by nanoparticle assisted LPE method. The tremendously declined intensity of graphene diffraction peak in comparison to that of graphite is related to reduction in number of layers which proves the successful exfoliation. The weakened intensity of XRD peak through exfoliation of graphite to graphene has been reported in the literature [29]. The important point of XRD pattern is that Fe3⁠ O4⁠ peaks which are shown in Fig. 6 (b) are not observed in graphene pattern. This shows that magnetic separation is a practical route for preparation of pure graphene sheets.

3.2.2. Transmission electron microscopy (TEM) Transmission electron microscopy (TEM) is the most effective method to ascertain the peeling of graphite particles into graphene sheets as well as observation the morphology of products. Fig. 4 shows TEM images of initial graphite particles and exfoliated graphene sheets with different number of layers. In comparison to graphene sheets (Fig. 4(b–i)), TEM image of initial graphite particle illustrated in Fig. 4(a) is completely opaque to electron beam which indicates thick structure of graphite particle. Therefore, the transparency of graphene in TEM images confirms the successful exfoliation of graphite through ultrasonication in NMP solvent. Unlike reduced graphene oxide which is produced by chemical oxidation and subsequent reduction of graphite, graphene samples prepared in this study does not have lots of wrinkles. The relatively smooth surface of graphene sheets is related to utilization of neat and pristine graphite (without any chemical reactions) as precursor. Generally, folding is the most observed morphological characteristic of exfoliated graphene sheets prepared by Fe3⁠ O4⁠ assisted LPE technique. Although the majority of produced graphene sheets did not have considerable amount of Fe3⁠ O4⁠ nanoparticles, but Fe3⁠ O4⁠ nanoparticles loaded graphene sheets were also observed in TEM samples that could be reduced by longer magnetic separation duration. Such cases are depicted in the following figures. Generally, improving the exfoliation yield by any methods is at the cost of in

3.2.4. Fourier transform infrared (FTIR) Fourier transform infrared (FTIR) spectroscopy is widely used for identifying the different functional groups in chemical structures. Herein, FTIR was performed in order to characterize graphite powder and graphene sample exfoliated by Fe3⁠ O4⁠ assisted LPE method. Fig. 7 outlines the FTIR spectra of initial graphite and exfoliated graphene samples. Due to the absence of special functional groups in pure graphite and graphene, there should not be any peaks in

4


UN CO RR EC TE D

PR OO F

A. Hadi et al.

Fig. 4. TEM images of a) initial graphite particle, b and c) exfoliated multilayer graphene, d–g) exfoliated few layer graphene, and h–i) exfoliated single to three-layer graphene sheets

Fig. 6. XRD patterns of a) initial graphite powder and exfoliated graphene sheets using LPE method in presence of Fe3⁠ O4⁠ nanoparticles, and b) Fe3⁠ O4⁠ nanoparticles

its FTIR spectra, inherently. However, because of the adsorption of solvent molecules and moisture some peaks are usually seen in FTIR spectra of graphite and graphene [37]. According to Fig. 7, the most intense peak of both graphite and graphene spectra is located around ⁠ 1 3413 cm− which could be assigned to stretching vibrations of O ⁠ 1 H bond (peaks between 3300 and 3600 cm− ) which is related to adsorbed moisture or presence of hydroxyl groups in initial graphite. An⁠ 1 other peak locating at wavenumber of 1616 cm− arises from sp2⁠ carbon atoms from C C bonds in graphite and graphene honeycomb lattice. ⁠ 1 A relatively strong peak could be seen at wavenumber of 620 cm− in both graphite and exfoliated sample that could be arising from impurities associated with graphite precursor. As shown in Fig. 7, characteristic peaks of Fe3⁠ O4⁠ are not present, and FTIR spectra of graphite and graphene are completely similar in both intensity and peak positions which indicates that no impurity (especially Fe3⁠ O4⁠ ) introduces to graphene structure during exfoliation process. This observation further confirms the XRD results and shows the purity of exfoliated graphene.

3.2.5. Atomic force microscopy (AFM) AFM is used as a common analysis for identifying flake thickness of layered 2D materials and thereby the number of layers after any exfoliation processes. This nondestructive technique is based on measuring the step heights from the bare surface of smooth substrates (especially mica and SiO2⁠ ) to the top of deposited graphene sheets. Figs. 8 and 9 show the AFM images of exfoliated graphene sheets which prepared using LPE technique without and in the presence of Fe3⁠ O4⁠ nanoparticles, respectively. Height profiles shown in Fig. 8 describe the thickness of two different graphene sheets which are located across lines A and B in AFM image. The thicknesses of about 0.77 and 3.8 nm have been measured for these selected sheets. Considering the theoretical thickness of 0.34 nm for each graphene layer, one can specify the number of layers per flake. however, due to different reasons like instrumental offset and morphological properties of graphene sheets (folding and wrinkles), step heights less than 1 nm are usually considered as single layer graphene in AFM studies. Therefore, for more accuracy, we assumed that the flakes with heights of shorter than 1 nm are less than three-layer graphene. Thus, abovementioned

5

A. Hadi et al.


PR OO F

AFM images in Fig. 8 are related to single and multilayer graphene. From Fig. 9, it can be seen that thinner graphene sheets have been obtained in presence of Fe3⁠ O4⁠ nanoparticles. Two graphene sheets with thicknesses of 0.83 and 1.67 nm are shown in the height profiles which are related to single and few-layer graphene. For better describing the effect of magnetic nanoparticles on the exfoliation process, flake thickness distribution was measured by analyzing more than 40 graphene sheets in AFM images. The results have been shown in Fig. 10. It could be seen that in both cases the majority of exfoliated graphene sheets are less than three-layer. Results indicated that the percentage of graphene sheets with thicknesses less than 1 nm are about 63 and 54% for LPE process without and with magnetic nanoparticles, respectively. Although the presence of nanoparticles has not changed the exfoliation of few layer graphene to less than three-layer sheets, but delamination of partially exfoliated flakes are greatly affected by the presence of Fe3⁠ O4⁠ nanoparticles. By comparison two flake thickness distributions, it could be clearly observed that unlike simple LPE method, when magnetic nanoparticles were used as the exfoliating agent, graphene sheets with thicknesses more than 3 nm (containing several graphene layers) were not present in the products. This finding indicates that the exfoliating role of nanoparticles during LPE process is more pronounced through the delamination of partially exfoliated flakes to few layer graphene sheets.

UN CO RR EC TE D

Fig. 7. FTIR spectra of graphite precursor and exfoliated graphene

Fig. 8. AFM image and height profiles of exfoliated graphene sheets using LPE method without Fe3⁠ O4⁠ nanoparticles 6


PR OO F

A. Hadi et al.

UN CO RR EC TE D

Fig. 9. AFM image and height profiles of exfoliated graphene sheets using LPE method in presence of Fe3⁠ O4⁠ nanoparticles

Fig. 10. Flake thickness distribution measured using AFM analysis of exfoliated graphene sheets prepared by LPE method a) with, and b) without Fe3⁠ O4⁠ nanoparticles

removed by long-term and strong magnetic separation. This results show that high quality graphene sheets could be obtained by magnetic nanoparticle assisted LPE method.

3.2.6. Raman spectroscopy Describing defects density, doping or functionalization extent, and determining the number of layers make Raman spectroscopy as one of the most useful characterization methods for graphene samples. Three well-known characteristic peaks are present in Raman spectra of ⁠ 1 graphene and graphite. The first one is D band at around 1350 cm− arising from defects and disorders in graphene structure which cause the activation of breathing modes of six-member rings. The G peak lo⁠ 1 which is originates from in-plane vibrations cates at around 1580 cm− ⁠ 1 asof sp2⁠ carbon atoms. The final peak is 2D band at around 2700 cm− cribes to breathing-like mode of graphene in-plane lattice [38,39]. The shape and position of 2D peak is a measure to determine the number of graphene layers per each flake. By decreasing the number of layers in graphitic materials, 2D peak becomes more symmetric and shifts to lower wavenumbers as well. These changes could be seen in 2D peaks ⁠ 1 in of Fig. 11. After exfoliation, 2D peak has a shift from 2696 cm− − ⁠ 1 graphite spectra to 2689 and 2680 cm for exfoliated graphene samples which approves the successful exfoliation of graphite to few layer graphene. Generally, D band in raman spectra of liquid phase exfoliated graphene has a considerable intensity (for example 0.49 [30] and 0.73 [40]) which arises from impurities remained in exfoliated graphene sheets, and defects introducing by intense ultrasonic waves. As presented in Fig. 11, defect density (ID⁠ /IG⁠ ) of exfoliated graphene samples is noticeably low in comparison to other studies, and ID⁠ /IG⁠ of graphene spectra is even less than that of initial graphite. Low intensity ratio of ID⁠ /IG⁠ of exfoliated graphene samples confirms again that Fe3⁠ O4⁠ nanoparticles and Fe3⁠ O4⁠ loaded graphene sheets have been effectively

3.2.7. Stability evaluation of prepared graphene dispersion Stability is one of the most important obstacles toward utilization of graphene in different applications. Zeta potential analysis could present practical information about the stability of various dispersed particles in common solvents based on the surface charge characteristics. Fig. 12 shows photographs of re-dispersed prepared graphene powder in NMP solvent which depicts stability of graphene sheets after different time periods from a day to a month. Following this figure, graphene dispersion in NMP solvent is highly stable for over a week. Stability of graphene in ethanol as one of the most practical solvents across different industries is evaluated by Zeta potential analysis. In comparison to NMP, there is a larger Hansen solubility parameter distance between ethanol and pure graphene. Therefore, stability of graphene sheets is naturally poorer in ethanol, but the level of stability could be obtained by Zeta potential. Fig. 13 displays Zeta potential distribution graph of graphene dispersion in ethanol. Due to the lack of oxygen functional groups on the surface, direct exfoliation of graphite to graphene results in positively charged graphene sheets. Zeta potential and electrical conductivity of synthesized graphene in ethanol was measured to be 14 mV and 0.0157 mS/cm. Therefore, Zeta potential analysis showed that graphene sheets have a moderate stability in ethanol. Relatively stable dispersion of graphene in ethanol (being stable for several hours) improves its processability and could be suitable for many potential applications.

7


PR OO F

A. Hadi et al.

Fig. 11. Raman spectra of initial graphite and exfoliated graphene powder prepared by magnetic nanoparticle assisted LPE technique (all spectra have been normalized to the most intense G peak)

min leads to yield enhancement from 17.21 to 20.84%. Generally, LPE process has some steps: in the earlier stages, large and thick graphite particles delaminated to smaller and thinner ones. Then, by applying further energy during sonication, these particles are exfoliated to graphene sheets with a distribution of number of layers. In this period, the yield of graphene preparation increases with sonication time. Following Fig. 16, by increasing the sonication time to 120 min, exfoliation yield dips to just less than 20%. This observation could be explained by aggregation of the exfoliated graphene sheets. Mainly, two factors govern the efficiency of LPE process which are the exfoliation and aggregation rates. Before reaching to maximum yield, when the suspension mostly contains the un-exfolated or partially exfoliated flakes, exfoliation rate is higher than aggregation. After a specific sonication time (which depends on other conditions) when the exfoliation yield exceeds the maximum, the aggregation rate becomes more pronounced in comparison with exfoliation. Because after optimum sonication time, the number of exfoliated graphene sheets per volume unit of solvents is high, and consequently the number of collisions between dispersed graphene sheets is more than exfoliation of remained graphite particles which leads to reduction of efficiency. Another reason of declining the exfoliation yield at sonication time longer than 90 min could be the oxidative degradation of solvent, as reported in the literature [41].

UN CO RR EC TE D

3.3. Effect of process factors on the exfoliation yield

3.3.1. Effect of initial graphite concentration The impact of initial graphite concentration has been shown in Fig. 14. Graphite content in NMP solvent was varied from 0.5 to 1.5 mg/ ml at three stages. According to Fig. 14, the exfoliation yield decreases by increasing graphite concentration in the initial dispersion. Obviously, when the graphite concentration rises, the energy of ultrasonication would be absorbed by a large number of graphite flakes and as a result, each graphitic particle receives less energy. Besides, high concentration of graphite particle results in higher extent of aggregation of graphene sheets which is unfavorable for LPE process. Therefore, in high graphite concentrations the exfoliation yield is declined, and partially exfoliated flakes are the main portion of the products.

3.3.2. Effect of Fe3⁠ O4⁠ nanoparticles concentration Fig. 15 illustrates the influence of Fe3⁠ O4⁠ nanoparticles in five stages on the exfoliation yield. As could be seen in this figure, increment of Fe3⁠ O4⁠ concentration from 0 to 0.5 mg/ml leads to a relatively steep climb in exfoliation yield. The ascending trend could be related to the peeling effect of magnetic nanoparticles through collision with partially exfoliated flakes. This phenomenon provides larger galleries between graphene layers and facilitates the exfoliation process. By further increasing the concentration of Fe3⁠ O4⁠ to 1 mg/ml, the exfoliation yield shows a fall of about 4.5%. At high levels of Fe3⁠ O4⁠ nanoparticles, the mean free path of nanoparticles becomes shorter so that a considerable portion of ultrasonic energy is consumed through useless collisions between Fe3⁠ O4⁠ nanoparticles. Therefore, the number of effective collisions between Fe3⁠ O4⁠ nanoparticles and graphite particles (and partially exfoliated flakes) would be reduced. Moreover, at the same ultrasonic power when the concentration of Fe3⁠ O4⁠ nanoparticles is higher, the energy carried by each nanoparticle is lower resulting in weaker shear forces and less effective exfoliation. In addition, when the concentration of Fe3⁠ O4⁠ nanoparticles rises, the portion of graphene sheets coated by magnetic nanoparticles (as shown in Fig. 5) increases proportionally. These kind of exfoliated sheets are removed from final suspension by magnetic separation which causes yield reduction. 3.3.3. Effect of sonication time Another important parameter is the time duration of ultrasonic treatment which has been varied in the range of 30–120 min. Fig. 16 illustrates the effect of sonication time on the exfoliation yield while other parameters remained constant. This figure shows that increasing the ultrasonic time from 30 to 90

3.3.4. Effect of sonication power Sonication power is one of the crucial factors which efficiency of LPE process is significantly affected by. In order to explore the influence of sonication power on the exfoliation yield, ultrasonic treatment with three levels of 240, 420, and 600 W have been subjected. Fig. 17 demonstrates the impact of ultrasonic power on the exfoliation yield. As shown in this Figure, exfoliation yield grows from 8.07% at sonication power of 240 W to 19.63% at treatment power of 600 W. Since the principle of LPE method is to overcome interlayer van der waals forces of graphite by shear forces and cavitation of micro-jets in the liquid, at higher sonication power, graphite particles absorb more energy which results in the more efficient exfoliation. Extremely high power sonication will reduce the quality of prepared graphene through making the sheets smaller, though. Therefore, sonication power higher than 600 W was not examined in this study. 4. Conclusion In this study single to few layer graphene produced by a novel magnetic nanoparticle assisted liquid phase exfoliation method. Characterization techniques confirmed the preparation of high quality graphene layers. Besides the


UN CO RR EC TE D

PR OO F

A. Hadi et al.

Fig. 12. Photographs of graphene dispersion in NMP solvent at different time periods after exfoliation

ultrasound induced cavitation, Fe3⁠ O4⁠ naoparticles generate strong shear forces that enhances the exfoliation of graphite flakes. Results showed that both the exfoliation yield and number of graphene layers are affected in the presence of Fe3⁠ O4⁠ nanoparticles. The magnetic nature of Fe3⁠ O4⁠ particles allows facile separation of pure graphene sheets. Although, some extent of Fe3⁠ O4⁠ loading on the graphene surface was observed in TEM images, but other characterization methods including XRD, FTIR, and Raman spectroscopy showed that products have little amount of impurities. Therefore, facilitated exfoliation using particle wedges would be an applicable technique in order to reduce ultrasonication time and suppress structural defects arising from long term sonication. More studies could be helpful to address the effect of wedge particles with different surface chemistry, morphology, and particle size (from nano to micro scale) in order to improve the efficiency of LPE method.

Fig. 13. Zeta potential distribution of graphene dispersion in ethanol solvent

9

A. Hadi et al.


Appendix A. Supplementary data

PR OO F

Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.ultsonch.2018.02.028.

UN CO RR EC TE D

Fig. 14. Effect of initial graphite concentration on graphene exfoliation yield

Fig. 15. Effect of Fe3⁠ O4⁠ nanoparticle concentration on graphene exfoliation yield

Fig. 5. Examples of TEM images of Fe3⁠ O4⁠ nanoparticle loaded graphene sheets

10

A. Hadi et al.


PR OO F

[10] V. Štengl, J. Henych, J. Bludská, P. Ecorchard, M. Kormunda, A green method of graphene preparation in an alkaline environment, Ultrason. Sonochem. 24 (2015) 65–71. [11] S.T. Hossaina, R. Wang, Electrochemical exfoliation of graphite: effect of temperature and hydrogen peroxide addition, Electrochim. Acta 216 (2016) 253–260. [12] P. Yu, G.P. Simon, Y.L. Zhong, Electrochemical exfoliation of graphite and production of functional graphene, Curr. Opin. Colloid Interface Sci. 20 (2015) 329–338. [13] M.H. Chakrabarti, N.S.A. Manan, N.P. Brandon, R.C. Maher, F.S. Mjalli, I.M. AlNashef, et al., One-pot electrochemical gram-scale synthesis of graphene using deep eutectic solvents and acetonitrile, Chem. Eng. J. 274 (2015) 213–223. [14] A.T. Murdock, C.D.V. Engers, J. Britton, V. Babenko, S.S. Meysami, H. Bishop, et al., Targeted removal of copper foil surface impurities for improved synthesis of CVD graphene, Carbon 122 (2017) 207–216. [15] C.L.C. Rodriguez, F. Kessler, N. Dubey, V. Rosa, G.J.M. Fechine, CVD graphene transfer procedure to the surface of stainless steel for stem cell proliferation, Surf. Coat. Technol. 311 (2017) 10–18. [16] A. Hadi, J. Karimi-sabet, S.M.A. Moosavian, S. Ghorbanian, Optimization of graphene production by exfoliation of graphite in supercritical ethanol: A response surface methodology approach, J. Supercrit. Fluids 107 (2016) 92–105. [17] H. Gao, K. Zhu, G. Hu, C. Xue, Large-scale graphene production by ultrasound-assisted exfoliation of natural graphite in supercritical CO2⁠ /H2⁠ O medium, Chem. Eng. J. 308 (2017) 872–879. [18] N. Song, J. Jia, W. Wang, Y. Gao, Y. Zhao, Y. Chen, Green production of pristine graphene using fluid dynamic force in supercritical CO2⁠ , Chem. Eng. J. 298 (2016) 198–205. [19] H. Gao, C. Xue, G. Hu, K. Zhu, Production of graphene quantum dots by ultrasound-assisted exfoliation in supercritical CO2⁠ /H2⁠ O medium, Ultrason. Sonochem. 37 (2017) 120–127. [20] J.S.Y. Chia, M.T.T. Tana, P.S. Khiew, J.K. Chin, H. Lee, D.C.S. Bien, et al., Facile synthesis of few-layer graphene by mild solvent thermal exfoliation of highly oriented pyrolytic graphite, Chem. Eng. J. 231 (2013) 1–11. [21] E.D. Grayfer, M.N. Kozlova, V.E. Fedorov, Colloidal 2D nanosheets of MoS2 and other transition metal dichalcogenides through liquid-phase exfoliation, Adv. Colloid Interface Sci. 245 (2017) 40–61. [22] J. Shen, Y. He, J. Wu, C. Gao, K. Keyshar, X. Zhang, et al., Liquid phase exfoliation of two-dimensional materials by directly probing and matching surface tension components, Nano Lett. 15 (2015) 5449–5454. [23] A. Jawaid, D. Nepal, K. Park, M. Jespersen, A. Qualley, P. Mirau, et al., Mechanism for liquid phase exfoliation of MoS2, Chem. Mater. 28 (2016) 337–348. [24] Y. Wang, Z. Shi, J. Yin, Boron nitride nanosheets: large-scale exfoliation in methanesulfonic acid and their composites with polybenzimidazole, J. Mater. Chem. 21 (2011) 11371–11377. [25] Y. Hernandez, M. Lotya, D. Rickard, S.D. Bergin, J.N. Coleman, Measurement of multicomponent solubility parameters for graphene facilitates solvent discovery, Langmuir 26 (2010) 3208–3213. [26] D. Tasis, K. Papagelis, P. Spiliopoulos, C. Galiotis, Efficient exfoliation of graphene sheets in binary solvents, Mater. Lett. 94 (2013) 47–50. [27] M. Yi, Z. Shen, X. Zhang, S. Ma, Achieving concentrated graphenedispersions in water/acetone mixtures by the strategy of tailoring Hansen solubility parameters, J. Phys. D Appl. Phys. 46 (2013) 025301. [28] L. Lia, J. Xu, G. Li, X. Jia, Y. Li, F. Yang, et al., Preparation of graphene nanosheets by shear-assisted supercritical CO2 exfoliation, Chem. Eng. J. 284 (2016) 78–84. [29] W. Wang, Y. Wang, Y. Gao, Y. Zhao, Control of number of graphene layers using ultrasound in supercritical CO2 and their application in lithium-ion batteries, J. Supercritical Fluids 85 (2014) 95–101. [30] J. Xu, D.K. Dang, V.T. Tran, X. Liu, J.S. Chung, S.H. Hur, et al., Liquid-phase exfoliation of graphene in organic solvents with addition of naphthalene, J. Colloid Interface Sci. 418 (2014) 37–42. [31] H. Yang, Y. Hernandez, A. Schlierf, A. Felten, A. Eckmann, S. Johal, et al., A simple method for graphene production based on exfoliation of graphite in water using 1-pyrenesulfonic acid sodium salt, Carbon 53 (2013) 357–365. [32] P.M. Carrasco, S. Montes, I. Garcia, M. Borghei, H. Jiang, I. Odriozola, et al., High-concentration aqueous dispersions of graphene produced by exfoliation of graphite using cellulose nanocrystals, Carbon 70 (2014) 157–163. [33] S. Castarlenas, C. Rubio, A. Mayoral, C. Tellez, J. Coronas, Few-layer graphene by assisted-exfoliation of graphite with layered silicate, Carbon 73 (2014) 99–105. [34] B. Mendoza-Sánchez, J. Coelho, A. Pokle, V. Nicolosi, A 2D graphene-manganese oxide nanosheet hybrid synthesized by a single step liquid-phase co-exfoliation method for supercapacitor applications, Electrochim. Acta 174 (2015) 696–705. [35] Z. Sun, N. Dong, K. Wang, D. König, T.C. Nagaiah, M.D. Sánchez, et al., Ag-stabilized few-layer graphene dispersions in low boiling point solvents for versatile nonlinear optical applications, Carbon 62 (2013) 182–192. [36] J.P. Lorimer, T.J. Mason, K. Fiddy, Enhancement of chemical reactivity by power ultrasound: An alternative interpretation of the hot spot, Ultrasonics 29 (1991) 338–343. [37] V.S. Tucureanu, A. Matei, A.M. Avram, FTIR spectroscopy for carbon family study, Crit. Rev. Anal. Chem. 46 (2016) 502–520. [38] A.C. Ferrari, D.M. Basko, Raman spectroscopy as a versatile tool for studying the properties of graphene, Nat. Nanotechnol. 8 (2013) 235–246. [39] R. Beams, L.G. Cançado, L. Novotny, Raman characterization of defects and dopants in graphene, J. Phys.: Condens. Matter 27 (2015) 1–26. [40] A. Viinikanoja, J. Kauppila, P. Damlin, E. Mäkilä, J. Leiro, T. Ääritalo, et al., Interactions between graphene sheets and ionic molecules used for the shear-assisted exfoliation of natural graphite, Carbon 68 (2014) 195–209. [41] U. Khan, H. Porwal, A. O’Neill, K. Nawaz, P. May, J.N. Coleman, Solvent-exfoliated graphene at extremely high concentration, Langmuir 27 (2011) 9077–9082.

UN CO RR EC TE D

Fig. 16. Effect of sonication time on graphene exfoliation yield

Fig. 17. Effect of sonication power on graphene exfoliation yield

References

[1] K.I. Bolotin, K.J. Sikes, Z. Jiang, M. Klima, G. Fudenberg, J. Hone, et al., Ultrahigh electron mobility in suspended graphene, Solid State Commun. 146 (2008) 351–355. [2] A.A. Balandin, S. Ghosh, W. Bao, I. Calizo, D. Teweldebrhan, F. Miao, et al., Superior thermal conductivity of single-layer graphene, Nano Lett. 8 (2008) 902–907. [3] C. Lee, X. Wei, J.W. Kysar, J. Hone, Measurement of the elastic properties and intrinsic strength of monolayer graphene, Science 321 (2008) 385–388. [4] D.R. Dreyer, S. Murali, Y. Zhu, R.S. Ruoff, C.W. Bielawski, Reduction of graphite oxide using alcohols, J. Mater. Chem. 10 (2011). [5] S. Park, R.S. Ruoff, Chemical methods for the production of graphenes, Nat. Nanotechnol. 4 (2009) 217–224. [6] M. Monajjemi, Liquid-phase exfoliation (LPE) of graphite towards graphene: An ab initio study, J. Mol. Liq. 230 (2017) 461–472. [7] W. Du, X. Jiang, L. Zhu, From graphite to graphene: direct liquid-phase exfoliation of graphite to produce single- and few-layered pristine graphene, J. Mater. Chem. A 36 (2013). [8] Y. Arao, F. Mori, M. Kubouchi, Efficient solvent systems for improving production of few-layer graphene in liquid phase exfoliation, Carbon 118 (2017) 18–24. [9] K. Muthoosamy, S. Manickam, State of the art and recent advances in the ultrasound-assisted synthesis, exfoliation and functionalization of graphene derivatives, Ultrason. Sonochem. 39 (2017) 478–493.

11