Yang Huab,
Zhuang Liac,
Hongqiang Liacd,
Yanyan Liuacd and
Shaoxian Song*acd
aSchool of Resources and Environmental Engineering, Wuhan University of Technology, Luoshi Road 122, Wuhan, Hubei 430070, China. E-mail: shaoxian@uaslp.mx; Fax: +86 2787 212127
bInstituto de Metalurgia, Universidad Autonoma de San Luis Potosi, Av. Sierra Leona 550, San Luis Potosi C.P. 78210, Mexico
cHubei Provincial Collaborative Innovation Center for High Efficient Utilization of Vanadium Resources, Wuhan University of Technology, Luoshi Road 122, Wuhan, Hubei 430070, China
dHubei Key Laboratory of Mineral Resources Processing and Environment, Wuhan University of Technology, Luoshi Road 122, Wuhan, Hubei 430070, China
First published on 22nd January 2016
The effects of the hydrocarbon chain-length of aliphatic surfactants on the preparation of graphene oxide from mildly-oxidized graphite oxide through intercalation and the intercalating mechanism have been studied in this work. This study was performed on a mildly-oxidized graphite oxide and its exfoliated products through the measurements of X-ray diffraction, ultraviolet visible spectrophotometer, atomic force microscopy, cyclic voltammetry, zeta potential, Fourier transformed infrared spectra and X-ray photoelectron spectrometer. The experimental results have shown that the in-plane graphitic order remained unaltered after intercalating surfactants into the graphite oxide. The yield and quality of prepared graphene oxide were proportional to the hydrocarbon chain-length of the aliphatic surfactants. Also, it was found that the aliphatic surfactants were intercalated into the lamellas of graphite oxide by means of hydrogen bonding and hydrophobic attractive force instead of electrostatic attraction or chemical bonding.
Although reducing GO can produce large amount of single-layer sheets, the resulting graphenes bear a substantial amount of structural defects and residual oxygen inherited from the oxidation step.11,12 To solve this problem, a method proposed was to decrease the oxidation degree of graphite oxide13 or exfoliate the graphite directly. As the yield of graphene produced through exfoliating graphite directly was low and most of the ribbons had two or more layers,14 it should be better to use the Mildly-oxidized Graphite Oxide (MGrO) as the initial raw material. Although a few of oxygen groups exist on the layers of the MGrO, the exfoliation of it with ultrasonic is still very hard.15
As reported by other researchers, ions (Mn2+, Mn3+ and SO4− etc.), surfactants (tetrabutylammonium hydroxide solution (40% water), oleylamine and octadecylamine etc.) and polymers (polyethylene oxide) were used to intercalate this MGrO to facilitate the exfoliation.16–20 Compared with other intercalators, the surfactant was most studied due to its high efficiency and ability of dispersing the suspension of reduced graphene oxide. However, most reports about the assistant exfoliation of MGrO with surfactants were concentrated on the cationic surfactants, few result about the anionic surfactants was shown. Hence, it is very valuable to study the effect of anionic surfactants on the exfoliation of MGrO. Considering the nonpolar carbon chain is the main component of anionic surfactants, the length of carbon chain should also be an important factor for the ultrasonic exfoliation of MGrO intercalated with anionic surfactants.
In this work, we attempted to investigate the effect of the length of carbon chain of short chain aliphatic surfactant aliphatic surfactants (2–18 carbon atoms, water-soluble) on the exfoliation of MGrO through the measurements of X-ray diffraction (XRD), ultraviolet visible spectrophotometer, atomic force microscopy (AFM), and cyclic voltammetry (CV). Besides, the mechanism by which the surfactants were intercalated into the lamellas of MGrO was studied through the measurement of zeta potential, Fourier transformed infrared spectra (FTIR) and X-ray photoelectron spectrometer (XPS). The objective was to obtain more understandings of the ultrasonic exfoliation of MGrO by means of intercalation of anionic surfactants.
Sodium acetate (SA), sodium laurate (SL), sodium stearate (SS), potassium permanganate, phosphoric acid (85% wt) and hydrogen peroxide aqueous solution (30% wt) from the Sinopharm Chemical Reagent Co. Ltd (China), sulfuric acid (98% wt) and hydrochloric acid (36% wt) from the Xinyang chemical reagent (China) were employed in this work. All the chemical reagents were of analytical purity. The deionized water used in this work was produced using a Millipore Milli-Q Direct 8/16 water purification system. The residual conductivity of the water was less than 1 μS cm−1.
The GO was prepared in the following procedure. First, the purified MGrO powder was mixed with water to obtain a 0.05 wt% dispersion, and the pH of this dispersion was adjust to 10. Then a certain dosage of aliphatic surfactant was added in and mixed with the MGrO for a certain time at 300 rpm. After that, an ultrasonic treatment was carried out for a given time by using a Cole Parmer ultrasonic processor (750 W and 20 kHz) with 20% amplitude. This treatment aimed to exfoliate the graphite oxide. Then, the dispersion was applied to a centrifugation at 3000 rpm and rotor radius of 5.2 cm for 30 min with a Thermo Scientific Sorvall ST 16 centrifuge, in order to remove the unexfoliated particles.
Ultraviolet-visible (UV-vis) absorption spectra was recorded using a 1 cm path length quartz cuvette with a Thermo Scientific Orion AquaMate 8000 UV-VIS spectrophotometer. The wavelength of 230 nm was used to determine the absorbance of GO solution, because the light absorption at this wavelength was due to the π → π* transition of CC. Based on the Lambert Beer's law, the absorbance was expressed as follows:
A = εCl | (1) |
Fourier transformed infrared spectra was used to investigate the chemical structure of MGrO samples. The FTIR patterns were recorded in the range of 500–4000 cm−1 using the KBr pressed-disk technique, with a Thermo/Nicolet Nexus Fourier transform infrared spectroscopy.
Atomic force microscope images for GO were obtained with a Bruker MultiMode 8 AFM under peak force tapping Mode. To prepare the samples for AFM measurement, the aqueous GO colloidal solution was coated onto the freshly cleaved mica flakes at 800 rpm with a spin coater, then they were transferred into an automatic thermostatic blast air oven and dried at 60 °C for 2 h. The detailed morphology of GO edges was observed using a Japanese JEM-2100F high-resolution transmission electron microscopy (HRTEM).
The X-ray photoelectron spectrometer analysis was performed using a VG Multilab 2000 X-ray photoelectron spectrometer and an Al Kα X-ray source with a solution of 0.47 eV was used in this test.
Cyclic voltammograms were measured using a Princeton Applied Research Versa STAT 4 electrochemical workstation at the ambient temperature in a three-electrode configuration, in which a slice of platinum, a saturated calomel electrode and the GO-modified glassy carbon electrode were used as auxiliary electrode, reference electrode and working electrode, respectively. 1 mol L−1 of H2SO4 aqueous solution served as the electrolyte and the CV tests were performed in the potential range from −1.0 to 1.0 V (vs. SCE) at a scan rate of 50 mV s−1.
A Malvern zetasizer nano-ZS90 with electrophoresis technology was used to determine the zeta potentials of the MGrO particles in aqueous solutions. Aqueous MGrO suspension with 0.05 wt% of solid concentration was first conditioned at 300 rpm for 30 min to disperse the MGrO particles well. After that, a few milliliter of the slurry was withdrawn to be put into a disposable folded capillary call for the measurement. Acidic and alkaline titrations with 0.1 mol L−1 HCl and 0.1 mol L−1 NaOH were added automatically to adjust the slurry pH. The temperature was kept at 25 ± 2 °C throughout the measurement.
![]() | ||
Fig. 1 Absorbance column pattern of GO prepared without or with 10−3 mol L−1 SA, SL and SS. The pre-intercalation time is 0. |
It is known that the layer number is a very important property of graphene as just a single-, double- and few- (3 to <10) layer graphene can be distinguished as a 2D crystal.23 As the precursor of chemical converted graphene, the thickness of GO is a key property to evaluate its quality. In Fig. 3, the AFM images of GO–SA, GO–SL and GO–SS (GO prepared in the presence of SA, SL and SS) at 30 min were presented. We can find that the morphologies of GO sheets (Fig. 3a, d and g) obtained with the help of surfactants were plate-like sheets with lateral dimensions between 100 nm and 1.2 μm, and these flakes were stretched on the mica substrate well due to their hydrophilicity. Line profiles taken from the AFM images (Fig. 3b, e and h) revealed the flakes of GO–SA to have an apparent thickness of 1.24–1.96 nm (e.g., Fig. 3b), the flakes of GO–SL to have a thickness of 1.47–2.05 nm (e.g., Fig. 3e) and the flakes of GO–SS to have a thickness of 1.09–1.3 nm (e.g., Fig. 3h). According to literature, the thickness of a single-layer GO should be about 1 nm.24 The thickness results showed that the particles of GO prepared with the help of surfactants should be 1 or 2 layers, indicating a good quality of them. Similarly, the HRTEM images showed in Fig. 4a–c also presented the plate-like particles of GO–SA, GO–SL and GO–SS, the enlarged images of their edges indicated that these particles were single layer. In addition, the statistical distribution histograms of GO–SA, GO–SL and GO–S were presented in Fig. 3c, f and i. It was shown that the normalized frequency of flake thickness distributed at 1.0–1.5 nm increased as the chain length of surfactants increased, indicating that the surfactant with longer chain length will help to prepare better (more single- or few-layer) GO.
![]() | ||
Fig. 4 (a–c) HRTEM images of GO–SA, GO–SL and GO–SS at 30 min. The edges of these GO particles were magnified and showed at the top left corner. |
When surfactants were added into the suspension of MGrO, it was believed that the effect of surfactants on the basal spacing of MGrO also would be a very important factor of the exfoliation. The variation of basal spacing of MGrO with respect to the change of surfactant was shown in Fig. 5. As shown in the XRD pattern of pristine MGrO, the characteristic diffraction peak located at 2θ = 26.14° should belong to the graphitized part in mildly-oxidized graphite, which was very close to the peak (2θ = 26.24°) of pure graphite. Another apparent characteristic peak in the XRD curve of MGrO located at 2θ = 11.86°, indicating a low oxidation degree of the oxidized area. After surfactants were added, the peaks of the graphitized area and oxidized area both moved leftward gradually. It might be attributed to that the intercalation of surfactants into the interlayer of MGrO has broken the π–π van der Waals bond between the lamellas further, leading to the increase of the basal space of the interlayers. This enlargement of the sp3 C–O structure will make it easier for the surfactants to enter into the mildly or unoxidized area in MGrO further during the exfoliation. In this case, it will be easier to obtain the GO sheets with low oxidization degree from MGrO after surfactants was added. These GO sheets can be used to prepare reduced graphene oxide through thermal or chemical reduction, which contain a few defects on the surface of it.
![]() | ||
Fig. 5 XRD patterns of MGrO–SA, MGrO–SL and MGrO–SS. MGrO–SA, MGrO–SL and MGrO–SS corresponding to the MGrO mixed with 10−3 mol L−1 SA, SL and SS for 24 h at 300 rpm, respectively. |
Moreover, the two patterns in Fig. 5 both show 100 and 110 reflections, indicating the maintenance of in-plane graphite oxide order. These reflections and hence the in-plane graphitic order remain unaltered after intercalating surfactants into MGrO. This phenomenon also was reported in the works of Mauro et al.25 In particular, a clear 002 reflections also was observed on the XRD pattern of MGrO–SS, which correspond to increases of spacing between MGrO layers from 0.696 to 0.839 nm, as a consequence of the good intercalation ability of SS. In addition, it was also found that the rank order of the basal spacing of GO–SA, GO–SL and GO–SS should be GO–SA < GO–SL < GO–SS. This phenomenon indicates that the longer the carbon chain of surfactant, the bigger the effect of it on the basal spacing of MGrO.
According to the discussing above, the surfactants facilitate the exfoliation through intercalating into the interlayer. The intercalation of surfactants might be realized in two periods during the exfoliation process (Fig. 6): (1) the surfactants were intercalated into lamellas as they were mixed with MGrO suspension, this period was the pre-intercalation process; (2) the surfactants were intercalated when ultrasonic was exerted on the MGrO particles to help open the lamellas, this period was the exfoliation-intercalation process. Fig. 7 presented the absorbance of GO–SA, GO–SL and GO–SS prepared with 0, 2, 12 and 24 h pre-intercalation. It was shown that the absorbance of GO–SA increased about 0.3 as the pre-intercalation time increased from 0 to 24 h, and this increment of absorbance is the largest in these three samples. Whereas, compared with the increase of absorbance which caused by changing the exfoliation time, this 0.3 increment was really small. In addition, it was noteworthy that the increment of the absorbance of GO–SS was just about 0.06, indicating that the effect of pre-intercalation on the exfoliation with SS was very small. Above all, it can be concluded that the exfoliation-intercalation played a major role in the exfoliation process.
![]() | ||
Fig. 6 Schematic representation for the pre-intercalation process and exfoliation-intercalation process. |
Besides, the height of the three white columns in Fig. 7 were nearly same, meaning that a long pre-intercalation time may eliminate the effect of chain length on the exfoliation. However, after a 2 min centrifugation was carried out on the GO samples with 24 h pre-intercalation, the absorbance of GO–SA, GO–SL and GO–SS decreased from about 1.07 to 0.85, 0.91 and 1.02, respectively. This slightly decrease of absorbance may derive from the existence of thick particles in GO suspension. The shorter the length of the carbon chain in surfactants, the more the thick particles in the GO suspension prepared, which corresponding well with the distribution results of the thickness of GOs in Fig. 3c, f and i.
As surfactant effect the exfoliation of MGrO through intercalating into the lamellas, it is very important to understand how does it been moved into the interlayer and also how does it been kept inside the lamellas of MGrO. Fig. 8 illustrated the zeta potential of pristine MGrO with respect to the pH. It was shown that the zeta potential of MGrO particles decreased as pH increased from 2 to 9, then it increased a little with the continual increase of pH. In our work, the particles of MGrO were negatively charged as the pH was adjust to 10, and also the surfactant was negatively charged. In this case, there shouldn't exist electrostatic attractive force between the MGrO particles and aliphatic surfactants. Hence, the surfactants added will not be intercalated into lamellas by electrostatic attraction.
It is known that the adsorption species can be divided into physical adsorption and chemical adsorption. As there are abundant of oxygen groups exist on the structure of MGrO, it is wonder that SA, SL and SS may react with the oxygen groups to be fixed on MGrO. Fig. 9 illustrated the IR results of MGrO which was mixed with surfactants for 12 h and then followed with water washing. The peaks at 3417 and 1622 cm−1 were attributed to OH stretching vibration and OH bending vibration, respectively, indicating the existence of physisorbed water molecules and structural OH groups.26 The band at 1730 cm−1 might refer to the CO stretching motions in COOH groups located at the edges and vacancy defects of MGrO lamellae, and also that of the quinone and ketone groups.27 Besides, a band at 1224 cm−1 might attributed to C–O–C stretching vibration, indicating the existence of ethers and/or epoxides.28 The band at 1054 cm−1 was caused by the C–O stretching vibration in C–OH bond. In Fig. 9, the location of the oxygen groups described above didn't change after SA, SL and SS were added, indicating there was no chemical bond connection between the oxygen groups and the surfactants.
![]() | ||
Fig. 9 IR results of MGrO which was mixed with surfactants for 12 h and then followed with water washing. |
In addition, it is known that the typical peak of CO in unionized carboxylic acid is at 1710–1713 cm−1, and the sodium carboxylate stretch generally appear at 1557–1560 cm−1.29,30 As no peaks appeared at these two areas in Fig. 9, it can be concluded that the FTIR results should belong to the vibration of oxygen groups in MGrO but not in aliphatic acid. This result suggest that the surfactant adsorbed on MGrO may be moved through repeated water washing, the adsorption type should belong to a physical adsorption or hydrogen bonding.
To understand the effect mechanism of surfactant on MGrO more, a XPS measurement was carried out on MGrO–SS and the results were presented in Fig. 10. Analysis of the XPS data was carried out using Thermo Scientific Avantage software, the peaks assigned to carbon atoms with sp2 hybridized orbitals (C–C) were all calibrated to 284.60 eV.31 The XPS data of C 1s of MGrO (Fig. 10a) exhibits the peak of C–O and CO bonds, located at 286.33 eV and 287.56 eV, respectively. The peaks at 532.46 eV and 534.36 eV in Fig. 10b were assigned to C–O and C
O, respectively. According to literature,32 the carbon–oxygen double bond in this MGrO should belong to ketone in which the oxygen connected with aromatic nucleus but not belong to carboxyl group. This result corresponded well with the report which indicated that the after SS was added into the suspension of MGrO–SS, the XPS spectrums of C 1s and O 1s both changed a little compared with that of pristine MGrO. The position of the peaks of C–C, C–O and C
O bonds were located at 284.60 eV, 286.59 eV and 288.73 eV, respectively. For the XPS results of O 1s, the binding energies of C–O and C
O were located at 532.48 eV and 535.23 eV (Fig. 10c), and a new peak appeared at 530.73 eV (Fig. 10d) which indicated the existence of the O–C
O in SS.
![]() | ||
Fig. 10 (a) C 1s XPS spectra of MGrO; (b) O 1s XPS spectra of MGrO; (c) O 1s XPS spectra of MGrO–SS and (d) O 1s XPS spectra of MGrO–SS. |
It is known that the binding energy will shift if the chemical environment around the atom changed. According to the report of Stevens et al., the formation of a new hydrogen bond will result in a high binding energy shift for the acceptor atom (O atom).33 As the FTIR results have proven that there is no new bond generated after SS was added, then it can be suggested that the red shift of the binding energy of C–O and CO should be attributed to the hydrogen bonding between MGrO and the hydrolysate (stearic acid) of SS (Fig. 11). Especially, the shift value of the peak of C
O corresponded well with the results of Flamia et al., in which the binding energy of C 1s in C
O changed from 287.6 eV to 288.8 eV as C
O was transformed into C
O⋯H.34
Considering the nonpolar carbon chains in anionic surfactants and the CC aromatic structure in mildly-oxidized graphite, it can be assumed that the surfactant and MGrO should be attracted into the interlayer through hydrophobic force (Fig. 11). The longer the carbon chain, the stronger the hydrophobic force. The stronger hydrophobic force makes it easier for the surfactant to be intercalated into the lamellas of MGrO, meaning better exfoliation efficiency. This mechanism corresponds well with the variation of the absorbance of GO–SA, GO–SL and GO–SS. According to the studies on the intercalation of cationic surfactants into graphite oxide,19,25 it was very hard to remove the cationic surfactants from GO prepared as a strong electrostatic attraction was formed between them. Compared with the cationic surfactants, the anionic surfactants used in our work can be moved through repeated water washing as the interaction between the surfactants and GO was not very strong, which was beneficial to the further application of GO prepared.
![]() | ||
Fig. 11 Schematic representation for the hydrogen bonding and hydrophobic connection between MGrO and SS. |
The CV curves of GO–SA, GO–SL and GO–SS were shown in Fig. 12, it can be observed that the faradaic peak decreased with the increase of the length of carbon chain (in the red rectangle). This variation indicates that the content of the oxygen groups on GO surface should also decrease with the increase of the length of carbon chain. After GO was reduced into reduced graphene oxide (RGO), the less the oxygen groups, the less the defects in RGO. Hence, it can be concluded that the aliphatic surfactants (contain 2–18 carbon atoms) with longer chain length may help to produce more and better GO through exfoliating mildly-oxidized graphite.
2. The intercalation of the aliphatic surfactants into graphite oxide was implemented through hydrogen bonding and hydrophobic attractive force instead of electrostatic attraction or a chemical bonding. In this case, the surfactants adsorbed on graphite oxide may be moved through repeated water washing, which is beneficial to the further application of graphene oxide prepared.
This journal is © The Royal Society of Chemistry 2016 |