Few Layer Reduced Graphene Oxide : Evaluation of the Best Experimental Conditions for Easy Production

This work aimed to produce graphene oxide with few graphene layers, a low number of defects, good conductivity and reasonable amount of oxygen, adequate for use as filler in polymeric composites. Two starting materials were evaluated: expanded graphite and graphite flakes. The method of oxidation used was the Staudenmaier one, which was tested over different lengths of time. No appreciable differences were found among the oxidation times and so the lowest oxidation time (24 h) was chosen as the most adequate. An investigation was also conducted into suitable temperatures for the reduction of graphite oxide. A temperature of 1000 oC gave the best results, allowing a good quality material with few defects to be obtained. The reduction was also evaluated under inert and normal atmosphere. The best results were obtained when the least modified material, e. g., graphite flakes, was used as a starting material, oxidized for 24h and reduced at 1000 oC for 30 s in a quartz ampoule under a normal atmosphere.


Introduction
Research on new materials has grown exponentially in recent decades due to the requirements of manufacturers in industry and a plethora of new uses, even in social life.Graphene-a single sheet of carbon atoms organized in a planar bi-dimensional network-is part of a special class of nanofillers with the potential to be used in composites.Graphene, or few layer graphene, can be obtained from graphite (stacked layers of graphene), a low-cost starting material that is found in nature as flakes or a powder with diverse particle sizes.The three largest producers of graphite are China, India and Brazil 1 .Andre Geim and Konstantin Novoselov made an outstanding contribution to the study of graphene's properties and received the Nobel Prize of Physics in 2010 for their innovative experiments 2,3 .They discovered that graphene is 200 times stronger than steel and conducts electricity at room temperature faster than any other known material.It is also highly transparent (97.3%) and extremely lightweight.Due to these remarkable properties, a large variety of applications seems possible, including the creation of new materials and innovative electronic products 4 .The properties of graphene are very dependent on its structure and therefore it is of great importance investigating methodologies for producing large quantities of single layer graphene.A structured material with a little number of layers of graphene is acceptable as well.In any case, for nanotechnology applications a little amount of defects is mandatory.
The routes to producing graphene include bottom-up and top-down processes.The bottom-up process is related to the synthesis of the material, while the top-down process refers to graphite exfoliation and it is performed until the desired number of graphene layers is achieved 5 .Chemical and mechanical exfoliation of graphite are the only possible methods for obtaining graphene in large quantities and at a low cost 6 .Chemical exfoliation promotes the oxidation of graphite through the introduction of functional groups such as hydroxyl, carbonyl, and epoxide.These groups decrease the interactions between the graphene layers, resulting in an increase of the interlayer distance.The larger space between the graphene sheets helps in carrying out exfoliation, leading to the formation of monolayer or few layers graphene oxide (GO).Electrical properties are influenced by the degree of oxidation and by the surface chemistry of the GO powders 7 .Some authors compared the most commonly used oxidation processes, e. g., the Staudenmaier 10 , Hofmann 11 and Hummers 12 methods, and showed that the Staudenmaier method produces thermally reduced graphene with the lowest proportion of defects.A comparison of the electrochemical performance of the graphene samples obtained from each of these three oxidation methods and a reduction at 1000ºC showed that the sample prepared using the Hummers method presented higher heterogeneous transfer rates and lower overpotentials compared with the samples obtained via the other two methods.These data indicate that the preparation method has a dramatic influence on the properties of graphene 13 .Some studies have modified the Staudenmaier process by experimenting with the oxidation time (between 1 and 7 days) and the HNO 3 :H 2 SO 4 acids ratio (1:2 and 1:3), with the aim of obtaining graphene oxide with optimum interlayer spacing.It was found that the optimal conditions were obtained with HNO 3 :H 2 SO 4 at a 1:3 volume ratio after 96 h of oxidation.The interlayer distances could be increased from 3.55 Ǻ of the original graphite to 9.13 Ǻ in modified Staudenmaier method 12 .However, the required reaction time to achieve the appropriate level of intercalation and oxidation also depends on the starting flake size 15 .Aksay et al 16 .found that 96 h of oxidation reaction was enough to completely oxidize 45 μm flakes but insufficient for 400 μm flakes.Oxygen can be eliminated partially through the thermal or chemical reduction of GO 17 , resulting in reduced graphene oxide.However, it must be considered that the reduced GO may present defects in the crystalline network that can affect the graphene's electrical properties negatively 18 .Accordingly, intensive research into improving the methodologies for reducing GO and restoring the conductive network is being undertaken 19 .Existing studies have looked at thermal reduction at high temperatures (1050 °C) 16,20 and at low temperatures (125-240 °C) 21,22 , whilst other works have examined the influence of the amount of carbon and oxygen on conductivity after thermal reductions at 900°C and 1100°C 23 .Many of these studies are difficult to compare because they use different oxidation methods and reduction conditions.Diffusion coefficient calculations suggest that there is a critical temperature of 550 ºC that must be exceeded for exfoliation to occur 23 .The present work aims to develop a methodology for obtaining large quantities of few layer reduced graphene oxide from graphite at a low cost, while presenting good quality and a reasonable amount of oxygen, which could be used as nanofiller for composites 24 .Since the starting material plays an important role, we decided to compare two materials: graphite flakes (FK) and expanded graphite (EG).The oxidation process used was the Staudenmaier 10 method.Various reaction times were tested (24, 48, 72 e 96h).Thermal reduction processes at 600, 700, and 1000 °C, in different atmospheres (normal and inert), were also investigated.

Materials
All materials were used as received.Expanded graphite Micrograph HC11 (EG) with an average size of around 50 μm and Graflake 9950 (FK) with a size of around 150 μm were provided by Nacional de Grafite Ltda.Brazil.Sulfuric acid, nitric acid, hydrochloric acid, and potassium chlorate were purchased from Merck Chemical Company (Brazil).

Graphite Oxide Synthesis
The graphite oxide (GO) was synthesized from EG and FK powders using the Staudenmaier method 10 .An acidic mixture of H 2 SO 4 (160 mL) and HNO 3 (90 mL) was stirred in a round-bottomed flask placed in an ice bath for 1 h.After this time, 10g of graphite (EG or FK) were added and stirred for 20 min.Subsequently, 110g of KClO 3 was added slowly over 15 mins, with the temperature being controlled to prevent a rise above 35°C.Small amounts of KCl were added to minimize the risk of explosion, which can be caused by the ClO 2 gas.The reaction was allowed to proceed over different intervals of time (24, 48, 72 e 96h).On completion of the reaction, an aqueous HCl solution (10% v/v) was added to the suspension in order to remove sulfate ions.Afterwards, washing with deionized water and centrifuging (4500 rpm) were carried out several times in order to reach pH 3. The resulting suspension was placed in a dialysis membrane until the pH of solution was 5, and then the resulting brown suspension was dried in an oven at 150°C.The graphite oxide was then heated at 600, 700, and 1000 o C for 30 s in an oven, using a closed quartz ampoule with either a normal or an inert atmosphere of argon.

Characterization
X-ray diffraction (XRD) measurements for the powdered samples were performed on a diffractometer (Rigaku, DMAX 2200) equipped with a Cu tube (λ = 0.15418 nm) and a secondary monochromator.The goniometer used was a Siemens D500 and the detector was the scintillator (NaI and Tl).XRD samples were prepared by coating the sample holder with a uniform layer of the powdered samples at room temperature.From the XRD parameters, the interlayer distances (d 002 ) in the graphene were estimated using Bragg's Law 25 .
The grain size (C) was calculated using the Scherrer equation 26,27 : Here, β is the line width at half height in the radians and θ is the diffraction angle.
The number of graphene layers was calculated by dividing the crystal size (C) by the interlayer distance (d) added to the thickness of one graphene sheet (0.1 nm).
Raman analyses were performed at room temperature using an Olympus microscope and an iH320 Jobin Yvon Spectrometer with a CCD (charged coupled device) detector, cooled by liquid nitrogen.The excitation source was a 10 mW HeNe laser (632.8 nm) and the acquisition time was 20 s.
GO samples were analyzed at 25 °C as pellets diluted in KBr by transmission FT-IR spectroscopy, using a Varian FTIR spectrophotometer (640-IR), which accumulated 32 scans at a 4 cm -1 resolution.
The fractions of C, H, and N in the graphite samples were analyzed using a Perkin-Elmer MCHNSO/2400 analyzer.A sample mass of 2 mg was employed in each analysis.
Scanning Electron Microscopy (SEM) was performed with a Phillips microscope (model XL30) operating at 20 kV, using aluminum stubs and gold metallization.
Transmission Electron Microscopy (TEM) images of the reduced graphene oxide were obtained using a JEOL 2010 transmission electron microscope operated at 200 kV.All the samples were prepared by depositing an acetone suspension drop on a copper grid (300 mesh) covered with amorphous carbon.
Electrical impedance spectroscopy was performed to measure the electrical conductivity of the samples.A sine wave with 1.0 Vpp was applied to the sample and the frequency was varied from 1 to 10 6 Hz.The current and the potential differences between the samples' faces were measured.The phase difference between the voltage and current in the sample allows the determination of the real and imaginary parts of electrical impedance.The real part is associated with conductivity.A numerical fitting algorithm was applied to the data, which provided the conductivity values.

Results and Discussion
In order to evaluate the need for pre-treatment in the oxidation process, two different types of graphite were employed as starting materials: expanded graphite (EG) and graphite flake (FK).Figure 1 shows the XRD patterns of EG and FK.It is possible to observe the interlayer spacing along the c-axis-(d 002 ) at 26.5º and (d 004 ) at ~55º-as sharp peaks for FK.In Table 1, it can be seen that both graphites present a very close spacing between the graphene layers but the crystal size of EG is smaller, which indicates a smaller graphene stack.The inexistence of peak at 55º in EG suggests it has a lower aspect ratio than FK.
Raman spectroscopy is commonly considered to be the most important technique in characterizing the structure of graphitic materials 28 .The main Raman features of graphene and graphite are the G band (~1580 cm -1 ), the D band (~1330 cm -1 ) and the G´ band (~2670 cm -1 ) for an excitation wavelength of 632.8 nm.The G band originates from the  in-plane vibration of sp 2 carbon atoms.The D band is related to crystal disorder due to sp 3 defects in the sp 2 lattice and the G´ band is an overtone of the disorder-induced D-band.
For single layer graphene, the G´ band appears as a sharp and symmetrical peak, whereas it becomes broader when the thickness increases (for example, in few layer graphene) 29 .
Figure 2 shows the Raman spectra of the two types of graphitic structures used in this work, namely, FK and EG.It can be observed that both structures present defects (peak D at 1330 cm -1 ).The larger and less intense G´ band suggests that the EG sheets contain more defects compared to the FK sheets.

Reduced graphene oxide obtained from expanded graphite
The strong oxidation of EG using Staudenmaier methodology over 96 h resulted in graphite oxide (GO96h).The formation of GO is indicated by the increase of the interlayer spacing along the c-axis (d 002 ) due to the insertion of various oxygen-containing functional groups in the graphite structure 2 .Figure 3 shows the XRD patterns of GO prepared from EG oxidation.
The data listed in Table 2 refers to the parameters of XRD and the percentage of carbon and oxygen obtained from CHN analysis for EG, GO96h and reduced GO at different temperatures (600, 700 or 1000°C).
The formation of GO96h is confirmed by the absence of the (002) diffraction line at 26.52º (d-space = 0.33 nm) that is typical of EG and the appearance of the diffraction peak at 11.23° (d-space = 0.78 nm) (Figure 3).Moreover, a marked decrease in the crystal size (8 nm) compared to the starting graphite (14 nm) was also observed.During the oxidation process, the insertion of various oxygen-containing functional groups that are bonded to the edges of basal planes of the graphite structure occurs.The ratio between the carbon and oxygen obtained by CHN analysis shows that oxidation was almost complete after 96 h (C/O , 1).A high percentage of oxygen is very important for the expansion and exfoliation of graphene sheets, since the percentage represents the number of molecules that will decompose in gases during thermal treatment 15 .Aksay et al. 16 determined the composition of gases that cause the rapid expansion of graphite-namely, CO 2 and H 2 O-demonstrating they originate from the exothermic decomposition of the hydroxyl and epoxide groups of GO.
When EG is oxidized to become GO, the insertion of oxygen functional groups decreases electrical conductivity due to a disruption in the planar sp 2 carbon sheets because of the introduction of sp 3 -hybridized carbons 5 .In order to restore the π-network, the GO was submitted to thermal reduction and the influence of the reduction temperature on the reduced graphene oxide was evaluated.The thermal treatments were carried out in an oven using a quartz ampoule.
Figure 4 shows the XRD patterns of GO96h samples that were heated at 600, 700, and 1000 °C (GO96h6, GO96h7, and GO96h10, respectively).All the reduced GO samples showed a broad peak that corresponded to 0.78 nm GO interlayer separation, indicating that the reduction process was not complete.A very large peak at 2θ , 26.5° reappeared for the samples reduced at 600 and 700°C, showing that the graphitic structure was partially restored.However, the peak broadens even more for higher temperatures, presumably due to the corrugated structure of the reduced GO sheets.Although the absence of the (002) diffraction line at 2θ = 26.5° in the XRD spectrum of GO96h10 suggests that all stacking was lost, it may also indicate that any remaining stacking was disordered 16 .2, the data from the elemental analysis is consistent with the results obtained from XRD.An increase in the C/O ratios in the reduced samples can be seen when compared with the starting GO96h (1.0), evidencing a successful thermal reduction.Although oxygen functional groups were still retained in the structure at all the evaluated temperatures, it was observed that the C/O ratio increased with increasing of the reduction temperature.
The oxygen-functionalized graphene sheets present opportunities for several applications, such as the preparation of composites in which those groups can have an important role in the interaction between graphene sheets and the polymeric matrix 7 .Covalent functionalization of graphene with the polymer can improve the dispersion in nanocomposites 8 .For example, some hydroxyl functional groups on the surface of graphene were used to support the catalyst system, in the in situ polymerization of ethylene, improving the graphite exfoliation in the polymeric matrix and consequently the nanocomposite properties 9 .Table 2 also shows that the differences in the amount of oxygen in the samples reduced at temperatures of 600 and 700°C are not significant.Even at the highest reduction temperature, 1000ºC, roughly 19% of oxygen atoms remained.
Figure 5 shows the Raman spectra for thermally reduced GO96h.According to the Raman spectra, part of the sp 2 carbon network was restored, though with many defects.
Considering that 96 h of oxidation reaction seems to damage the graphitic structure of EG excessively, shorter reaction times (24, 48, and 72 h) were evaluated with the purpose of reducing the defects in the graphene sheets obtained after thermal reduction.However, after thermal reduction of the GO samples at 1000°C, the Raman spectra showed no significant differences, when compared with using an oxidation time of 96 h (see Figure 6).
In fact, these Raman spectra show that a shorter EG oxidation time neither reduces the prevalence of defects, nor affects the stacking ordering.The sp 2 structure, therefore, is not recovered completely during the reduction process.Considering the EG exfoliation, it is possible to conclude that a high degree of oxidation was obtained even after the lowest period of oxidation (24 h), giving C/O ratios close to 1, reducing the crystal size, and increasing interlayer distances.However, even at the highest temperature (1000ºC), the reduction was insufficient for recuperating the sp 2 network, thus leading to the production of few layer reduced graphene oxide with many defects.Materials Research

Reduced graphene oxide from graphite flake
Using the same experimental approach as for EG, GO samples were prepared from graphite flakes (FK).Different oxidation times (96, 72, 48, and 24 h) were evaluated.Figure 7 shows some of the typical XRD patterns for GO obtained from FK.In contrast to what was observed with EG, FK was only partially oxidized over all the evaluated reaction times.The existence of fewer defects in the FK structure relative to EG may affect the oxidative breakup and a time longer than 96 hours may be required for complete oxidation to occur 16 .Elemental analysis was performed on the graphene oxides in order to evaluate the degree of oxidation (Table 3).According to Table 3, the degree of oxidation was practically independent of the reaction time.It is possible to conclude, therefore, that a 24-hour period of oxidation is long enough to prepare oxidized graphite flake with about 30% oxygen.
The FTIR spectrum of graphite oxide FK24h (presented in Figure 8) shows the stretching vibrations of O-H and C-O at 3398 and 1083cm -1 respectively.The stretching C=C double bonds of the condensed rings of graphene appear at 1621cm -1 .This band has a shoulder at around 1700cm -1 that can be attributed to the presence of some carbonyl groups (C=O) in small amounts.The presence of some stretching vibrations from C-H bonds (C sp 3 ) is also evident between 2800 and 3000cm -1 .
The thermal reduction of the flake oxidized over 24 h was tested at 600, 700, and 1000 ºC.A very simple method of reduction was evaluated using an open quartz capsule placed in the oven.However, this procedure resulted on the loss of the most exfoliated fractions, especially at the highest reduction temperatures.Thus, for the reduction reactions, a closed quartz ampoule was chosen instead, which was tested using normal and inert atmospheres.Figure 9 shows the XRD patterns for the samples obtained after FK was oxidized over 24 h and reduced at 600, 700, and 1000 °C in a closed quartz ampoule, both in an inert atmosphere (FK24h6APIN, FK24h7APIN, FK24h10APIN) and a normal atmosphere (FK24h6AP, FK24h7AP, FK24h10AP).Table 4 gives the XRD patterns for the data obtained.
According to Figure 9, the reduction of GO in an inert atmosphere at temperatures of 600 and 700 °C presents large diffraction peaks, suggesting heterogeneity in the stacking distances.At a reaction temperature of 1000°C, the peak is thinner, even though the average interlayer distance is the same, namely 0.35 nm, which is higher than in the starting graphite (0.33 nm in FK).In all cases, there was significant exfoliation from about 40 graphenes per crystal in FK to 5-11 graphenes per crystal in the reduced graphene oxides.The reduction performed in a closed quartz ampoule in a normal atmosphere was more efficient at a temperature of 1000°C because at 600 and 700°C, there is still a broad diffraction band at 11.29°, which is characteristic of GO.It can thus be seen that the reduction is less efficient at lower temperatures.At a reduction temperature of 1000ºC, the exfoliation is the same in an inert or a normal atmosphere (11 graphenes per crystal).
The Raman spectra in Figure 10 show that the oxidized graphite (FK24h) reduced at 1000°C in a normal atmosphere gives the best recuperation of the sp 2 network, compared with the sample reduced at lower temperatures or in an inert atmosphere.The ratio between the intensities of the D and G peaks is normally used as a quantitative parameter of defect Few Layer Reduced Graphene Oxide: Evaluation of the Best Experimental Conditions for Easy Production prevalence in graphite.In Figure 10, the I D /I G ratios for normal and inert atmospheres are 0.8 and 1.1 respectively, confirming the presence of fewer defects in the reduced oxide obtained at 1000ºC in a normal atmosphere, compared with other samples.
According to the results of the elemental analysis (Table 5), an increase in the reaction temperature from 600 to 1000°C raises the C/O ratio, as expected.The use of an inert or a normal atmosphere does not have a significant influence on the amount of oxygen that occurs in the reduced graphite.After reduction at the highest temperature (1000°C), there is still a reasonable amount of oxygen present (13%), which can be used in compatibility reactions for the preparation of composites.In any case, the amount of oxygen is inferior to that obtained when EG is used as a starting material.The HRTEM image of the FK24h10AP sample shows a few layer graphene sheet, which indicates that an effective exfoliation of graphite oxide has taken place (Figure 11).
Table 6 shows the conductivity values for the samples, measured by electrical impedance spectroscopy prior to and after reduction treatment.One can observe that the oxidation of the FK decreased the electric conductivity of sample FK24h slightly; however, the reduction process, independent of the temperature, was very efficient in restoring the sp 2 network, which is responsible for conductivity.The oxidation process and the reduction resulted in a material composed of few layer graphene, which is roughly one order of magnitude more conductive than the raw material.

Conclusions
Using expanded graphite (50 μm, crystal size 14 nm) as a starting material to obtain reduced graphite oxide produced few layer graphene with many defects.The raw untreated graphite (flake, 150 μm, crystal size 17 nm) was found to be the starting material most appropriate for producing reduced graphene oxide through the modified Staudenmaier method.Reduced graphene oxide containing roughly 13% oxygen, a low density of defects and a low number of graphene layers was obtained from oxidation of the graphite flake over a 24 h period, followed by thermal reduction at 1000 ºC in a quartz ampoule using a normal atmosphere.The use of an inert atmosphere during the reduction process did not improve the structure or electrical properties of the resulting reduced graphene oxide.The electrical conductivity of the few layer graphene obtained was ten times superior to that of the starting graphite.
The few layer graphene material with a low defect density that was obtained is adequate for many applications, such as use in polymeric composites, for example.

Figure 1 :
Figure 1: XRD patterns of starting graphite materials

Figure 2 :
Figure 2: Raman spectra of starting graphite materials

Figure 4 :
Figure 4: XRD patterns of GO and reduced GO96h at different temperatures According to Table2, the data from the elemental analysis is consistent with the results obtained from XRD.An increase in the C/O ratios in the reduced samples can be seen when compared with the starting GO96h (1.0), evidencing a successful thermal reduction.Although oxygen functional groups were still retained in the structure at all the evaluated temperatures, it was observed that the C/O ratio increased with increasing of the reduction temperature.The oxygen-functionalized graphene sheets present opportunities for several applications, such as the preparation of composites in which those groups can have an important role in the interaction between graphene sheets and the polymeric matrix7 .Covalent functionalization of graphene with the polymer can improve the dispersion in nanocomposites8 .For example, some hydroxyl functional groups on the surface of graphene were used to support the catalyst system, in the in situ polymerization of ethylene, improving the graphite exfoliation in the polymeric matrix and consequently the nanocomposite properties9 .Table2also shows that the differences in the amount of oxygen in the samples reduced at temperatures of 600 and 700°C are not significant.Even at the highest reduction temperature, 1000ºC, roughly 19% of oxygen atoms remained.Figure5shows the Raman spectra for thermally reduced GO96h.According to the Raman spectra, part of the sp 2 carbon network was restored, though with many defects.Considering that 96 h of oxidation reaction seems to damage the graphitic structure of EG excessively, shorter reaction times (24, 48, and 72 h) were evaluated with the purpose of reducing the defects in the graphene sheets obtained after thermal reduction.However, after thermal reduction of the GO samples at 1000°C, the Raman spectra showed no significant differences, when compared with using an oxidation time of 96 h (see Figure6).In fact, these Raman spectra show that a shorter EG oxidation time neither reduces the prevalence of defects, nor affects the stacking ordering.The sp 2 structure, therefore, is not recovered completely during the reduction process.

Figure 9 : 4 :FKFigure 10 :
Figure 9: XRD patterns for GO samples reduced in a closed quartz ampoule in (a) an inert (APIN) and (b) a normal (AP) atmosphere

Figure 11 :
Figure 11: HRTEM micrographic image of oxide graphene oxidized over 24h and thermally reduced at 1000°C in a closed quartz ampoule in a normal atmosphere.

Table 1 :
Data obtained from the XRD patterns of EG and FK

Table 2 :
Data obtained from the XRD patterns and CHN of EG, GO96h and reduced GO96h at different temperatures.
*GO96h6: correspond to GO 96h that was reduced at 600°C.The same nomenclature was used for the other temperatures.

Table 3 :
Elemental analysis of GO obtained from different lengths of oxidation time.

Table 5 :
Elemental analysis of reduced GO obtained in a closed ampoule.

Table 6 :
Electrical conductivity of raw materials oxidized and reduced graphite.