DOI 10.1007/s11237-014-9377-3 Theoretical and Experimental Chemistry, Vol. 50, No. 5, November, 2014 (Russian Original Vol. 50, No. 5, September-October, 2014)
EFFECT OF THE DEGREE OF PHOTOREDUCTION OF GRAPHENE OXIDE ON ITS ABILITY TO STABILIZE GRAPHITE AND CARBON NANOTUBES IN AQUEOUS COLLOIDAL SOLUTIONS N. S. Andryushina,1 A. L. Stroyuk,1 S. Ya. Kuchmy,1 N. V. Lemesh,1 and N. A. Skorik2,3
UDC 544.526, 54.148, 546.26
The effect of the degree of photochemical reduction of colloidal graphene oxide (GO) on its ability to stabilize graphite microparticles and carbon nanotubes in aqueous solutions was determined. If photoreduced GO is used as stabilizer dispersions of graphite particles with an average size of 2-3 µm and nanotubes with lengths of up to 1 µm are formed. The amount of dispersed graphite particles increases many times with increase in the degree of reduction of the GO. The relationship between the amount of dispersed nanotubes and the degree of photoreduction of the GO passes through a minimum as a consequence of photoinduced removal of the functional groups of the GO and of decrease in its ability to interact with the partially hydroxylated surface of the nanotubes.
Key words: photoreduction, carbon nanomaterials, heterostructures, electron microscopy.
Carbon materials (CMs) such as finely dispersed graphite, graphene, and carbon nanotubes find use as electroactive components of sensors [1, 2], supercapacitors [3, 4], and electrochemical [5], photoelectrochemical [6-10], and photocatalytic [11] systems, and the range of their applications is constantly expanding. An important problem during the production of composites and heterostructures based on CMs is the attainment of a uniform distribution of the carbon allotrope in the volume of the other components or on the surface of porous matrices as, for example during the formation of electrodes for solar cells [6-10]. As a rule, for this purpose the graphite, carbon nanotubes, and other carbon materials are dispersed in various solvents by ultrasonic or mechanochemical treatment. Dispersion of carbon materials in polar media such as water can only be achieved in the presence of surfactants, which prevent rapid agglomeration of the CM [12-17]. However, most of the employed surfactants are dielectrics, the presence of which in composites based on CMs reduces their functional characteristics [14]. For this reason the search for new ________ 1 L. V. Pisarzhevskii Institute of Physical Chemistry, National Academy of Sciences of Ukraine, Prospekt Nauky, 31, Kyiv 03028, Ukraine. E-mail:
[email protected]. 2 OOO “NanoMedTekh,” Vul. Hor’koho, 68, Kyiv 03150, Ukraine. E-mail:
[email protected]. 3 G. V. Kurdyumov Institute for Metal Physics, National Academy of Sciences of Ukraine, Prospekt Akademika Vernads’koho, 36, Kyiv 03680, Ukraine. ___________________________________________________________________________________________________ Translated from Teoreticheskaya i Éksperimental’naya Khimiya, Vol. 50, No. 5, pp. 280-287, September-October, 2014. Original article submitted September 29, 2014. 282
0040-5760/14/5005-0282 ©2014 Springer Science+Business Media New York
electroconductive stabilizers that would make it possible to produce stable aqueous dispersions of CMs and of graphite and multiwalled carbon nanotubes (MCNTs) in particular is an urgent task. Reports have recently appeared on the possibility of long-term stabilization of graphite [11, 18, 19] and MCNTs [15-17, 20] in aqueous colloidal solutions of graphene oxide (GO) – the product from exfoliation of graphite oxide, which is in turn produced by oxidation of graphite in strongly acidic media [19, 21]. The graphene oxide produced by the traditional methods of Hammers and Brodie is a dielectric, but as a result of reduction it acquires electric conductivity approaching the conductivity of the initial graphite [19, 21]. At the same time, despite the wide range of methods known today for the reduction of GO, particularly those using chemical agents and electrochemical and photochemical methods [11, 18, 19, 22], there is in fact no information on the effect of the degree of reduction of colloidal GO on the nature of its interaction with the CM. In this connection the aim of the present work was to establish a relation between the degree of photochemical reduction of colloidal GO and its ability to interact with the microparticles of graphite and MCNT, resulting in the dispersion and stabilization of these materials in aqueous media, and also to determine the characteristics of the obtained dispersions of the CM.
EXPERIMENTAL Graphite powder (Merck), concentrated H2SO4, HNO3, and HCl, KMnO4 (Sigma-Aldrich), and aqueous 30% H2O2 solution (Khimlaborreaktiv) were used for the work. The graphite oxide was synthesized by modification of the Hammers method [18, 19] by oxidation of graphite with potassium permanganate in concentrated sulfuric acid. Colloidal graphene oxide was obtained by ultrasonic treatment of graphite oxide in distilled water for 20-25 min by means of a UZDN-A disperser (130 W, 22 kHz) [22]. The obtained colloids with a concentration of 500 mg/L of GO were diluted to 25-50 mg/L for the spectral measurements and the photochemical experiments. Photoreduction of the colloidal GO was realized in thermostated glass cuvettes with a thickness of 10.0 mm for 30-90 min with light from a DRSh-1000 high-pressure mercury lamp, from the spectrum of which the section with l = 310-390 nm was isolated (intensity I = 1.2·1017 quanta/s). The MCNTs were obtained by catalytic decomposition of ethylene on Ni nanoparticles formed on the surface of CaO by the method reported in [23]. The freshly prepared MCNTs were kept in boiling concentrated nitric acid for 1 h in order to remove the amorphous carbon and catalyst. In order to obtain the aqueous dispersions of graphite and MCNTs 0.5-1.0 g of the dispersed material was placed in 10 mL of the colloidal solution of GO or the photochemically reduced GO and treated with ultrasound for 20-25 min. During the treatment the mixture was cooled. The obtained suspension was centrifuged at 3000 rpm, and 2-3 mL of the solution was taken for the subsequent investigations. The absorption spectra of the solutions were recorded on a Specord 210 spectrometer. Scanning electron micrographs (SEM) were obtained on a Tescan Mira 3 microscope at an accelerating potential of 10-15 kV. The atomic-force photomicrographs (AFM) were recorded on a Nanoscope IIIa microscope. The surface roughness of the samples was analyzed with the ImageJ software package. Data on the hydrodynamic size of the particles in the investigated solutions were obtained on a ZetaSizer Nano instrument at 22 °C. The samples were exposed to the light from a He–Ne laser with l = 633 nm, and the scattered light was detected at an angle of 173°.
RESULTS AND DISCUSSION Characteristics of Colloidal GO. The exfoliation of graphite oxide, produced by the Hammers method, by ultrasonic treatment in water leads to the formation of transparent colloidal solutions that retain the long-term aggregation stability [22]. According to data from AFM (Fig. 1a), the colloids contain planar particles with a thickness of 1.3-1.6 nm (Fig. 1b), characteristic of one- or two-layer GO [19, 21], and with a lateral size ranging from several tens of nanometers to a micrometer. As shown by the data from SEM (Fig. 1c, inset), the GO particles have a “crumpled” or partly folded form, indicating interaction between the individual fragments of the GO planes. Polished silicon with a surface layer of SiO2 was used as substrate for the AFM measurements. Strong interaction of the functional groups of the GO with the substrate surface probably 283
Fig. 1. a) Atomic-force micrograph of GO particles. b) Thickness of the GO particles measured along the line indicated in Fig. 1a by white triangles. c) Absorption spectrum of the GO colloid before exposure (1) and after exposure for 30 (2), 60 (3), and 90 (4) min. (Inset: Scanning electron photomicrograph of the GO particles before exposure of its colloid; scale 200 nm.)
leads to the result that the GO particles acquire a predominantly planar form, which is not observed on the images obtained by the SEM method. The absorption spectrum of the colloidal GO (Fig. 1c, curve 1) represents the superimposition of three spectral components reflecting the distinctive features of the structure of GO. The bands with a maximum in the region of 230-270 nm are typical of carbon materials containing aromatic sp2 carbon – graphite, carbon nanotubes, graphene, and GO [19, 20, 24]. In the case of the highly aromatic graphite and graphene the maximum of the band lies at lmax = 270 nm [24]. Oxidation of the graphene and its conversion into GO leads to partial transition of the carbon atoms from the sp2- to the sp3-hybridized state and to bonding with various oxygen-containing functional groups (OFG). Here lmax undergoes a shortwavelength shift, which is more pronounced the more intensive the oxidation of the graphene and the smaller the fraction of residual sp2 carbon in its composition [19, 21]. The characteristic maximum for the colloidal GO examined in this work lies at lmax = 233 nm. The shoulder in the region of 290-340 nm can be attributed to electronic transitions of the np* type in the —C=O groups of the GO. Finally, the absorption in the region of l > 400 nm, which falls regularly into the longwavelength region, is due to electronic excitation of the aromatic fragments of the sp2 carbon retained during oxidation of the graphite [19, 21]. As shown in [22], excitation of GO in this spectral range leads to indirect electron transitions with minimum energy Ebg » 1.9 eV. Effect of Photoreduction on the Characteristics of GO. As we showed earlier, exposure of graphene oxide colloids to UV light leads to photoreduction of the GO particles accompanied by a series of spectral changes [22]. In particular, in the absorption spectrum of the photoreduced GO (PRGO) the characteristic maximum lmax is shifted toward the longwavelength side, the np* band disappears, and the light absorption in the visible and near-IR regions is significantly increased (Fig. 1c, curves 2-4). These changes are more pronounced the longer the exposure and, consequently, the more intensive the photoreduction of the GO. Exposure of the GO colloids to UV light for 90 min shifts lmax from 233 to 260 nm and also reduces the energy Ebg from 1.9 to 0.9 eV (Table 1) [22]. In view of the fact that the position of lmax is directly proportional to while the Ebg value is inversely proportional to the ratio of the amounts of sp2- and sp3-hybridized carbon in the GO it can be concluded that photoreduction leads to decrease in the number of OFGs and increase in the area of the aromatic fragments in the 284
TABLE 1. Effect of the Exposure Time on the Energy Ebg of the Electronic Transition, the Position of the Characteristic Maximum lmax in the Absorption Spectrum, and the Average Hydrodynamic Size dHD of the Colloidal GO and PRGO Particles and also on the Integral Intensity of the Absorption Bands of Graphite IG and MCNTs INT in the Absorption Spectra of Their Composites with GO (PRGO) Exposure time, min
Ebg, eV
lmax, nm
dHD, nm
0
1.90
233
320
0.7
6.9
30
1.30
252
520
1.7
6.4
60
1.10
257
360
2.1
7.3
90
0.90
260
300
2.3
9.4
INT, IG, rel. units rel. units
Note. The accuracy of determination of Ebg, lmax, and dHD is 0.05 eV, 1 and 10 nm respectively, and of IG and INT 0.1.
composition of the GO particles. We confirmed this conclusion in [22] by data from IR spectroscopy, according to which photoreduction of GO leads to a significant decrease of the amount of OFG in the GO particles and to an increase in the intensity of the characteristic D and G bands in the Raman spectrum due to vibrational excitation of the fragments containing sp2 carbon. We obtained further arguments in favor of the increased aromaticity of the GO particles as a result of photoreduction by X-ray fluorescence spectroscopy [25], according to which exposure of the GO colloid in the given time interval increases the amount of sp2- and sp3-hybridized carbon from 0.7 to 6. As a quantitative measure of the degree of photoreduction of GO it is possible in principle to use any of the variable spectral parameters of the colloid proportional to the duration of the photoprocess, e.g., the position of lmax or the energy Ebg. In view of the fact that the accuracy of determination of Ebg can be affected to a significant degree by scattering of light from the fraction of large GO particles in the present work the value of lmax was used as parameter characterizing the degree of reduction of GO. The hydrodynamic size of the particles in the initial GO colloids, determined by laser photocorrelation spectroscopy, varies from ~150 to ~550 nm with the center of distribution at dHD = 320 nm (Table 1). As we showed in [26], photoreduction of colloidal GO is accompanied by changes in the hydrodynamic size of its particles, which have nonmonotonic character. At the initial stage (the first 30 min) irradiation of the GO colloids leads to an increase of dHD to 520 nm. The value then decreases to 360 nm during irradiation for 60 min and after irradiation for 90 min reaches a value of 300 nm, which is lower than in the initial unirradiated colloid (Table 1). Taking account of the fact that photoreduction of GO does not lead to aggregation of its particles into multilayer structures [26] and to loss of stability by the colloids the main factor that determines the experimentally observed changes of dHD in the course of photoreduction of the GO is probably change in the form (the degree of crumpling) of the GO particles [26]. The deviation of the colloidal particles of the initial GO from planarity (crumpling) is due, as supposed in [9, 27], to the formation of hydrogen bonds between the OFG of the various segments of the GO particles, including those with water as bridging molecule. Photoreduction of GO leads to disappearance of most of the OFG, as shown by changes in the spectra of the colloids that appear most clearly in the first 30 min of exposure [22]. Removal of the OFG is accompanied by rupture of the 285
Fig. 2. a) Absorption spectra of GO (1), PRGO (2), and composites GO-G (1¢) and PRGO-G (2¢) (concentration of GO 0.025 mg/mL, initial mass of graphite 1 mg/mL, PRGO obtained as a result of exposure for 90 min; inset, differential spectra of GO-G (1) and PRGO-G (2) colloids, obtained by subtracting 1¢ and 2¢ from 1 and 2 respectively on the main figure). b) Dependence of the integral intensity (IG) of the absorption band of graphite on the position of the characteristic maximum lmax of GO and PRGO. c) Distribution of the particles by the hydrodynamic size d in aqueous PRGO colloid (90 min exposure) (1) and the dispersion of PRGO-G (2). d) Scanning photomicrograph of PRGO-G particles (scale 10 mm; inset, distribution of PRGO-G particles by size).
bonds between the individual fragments (folds) of the PRGO particles. At the same time the small sizes of the aromatic regions at the initial stage of the photoreaction and also the residual OFGs and defects in the basal plane prevent effective pp-stacking interaction between the aromatic fragments of the PRGO particles and the formation of new folds. As a result the PRGO particles acquire more clearly defined planar structure, as reflected in the increase of the dHD value. During subsequent photoreduction there is a further decrease in the amount of OFGs and an increase in the area of the aromatic segments of the PRGO particles. Since the PRGO particles here become more hydrophobic they acquire a tendency to minimize the contact with the solvent by crumpling of the particles and reduce their effective cross section. As shown by calculations [27], the main driving force of such crumpling is the formation of pp bonds between the aromatic fragments on the various sections of the GO particles. This leads to the formation of more compact aggregates in the individual fragments of the GO particle [27] compared 286
with the initial assembly formed with the participation of hydrogen bonds. For this reason prolonged photoreduction makes it possible to obtain PRGO particles with a significantly lower dHD value compared with the initial colloid. Interaction of Colloidal GO (PRGO) with Graphite. Ultrasonic treatment of graphite powder, placed in the GO colloid, followed by centrifuging leads to a change of color from light-brown to gray as a result of the passage of part of the graphite into solution. A solution of GO containing dispersed graphite (GO-G) is stable for several months. There are no spectral indications of the presence of graphite particles in a solution obtained by ultrasonic treatment of graphite in the absence of GO. Accurate determination of the mass of dispersed graphite is difficult on account of the hygroscopicity of GO and also as a result of loss of part of the GO, which interacts with the graphite in the deposit formed during centrifuging. The presence of graphite in the solution of GO is indicated in the absorption spectrum by the appearance of a band with a maximum at 265-270 nm (Fig. 2a, curve 1¢, inset, curve 1), characteristic of graphite and graphene [24], that overlaps with the absorption bands of GO. In spite of the fact that colloidal graphite stabilized with GO does not visually scatter light the possibility of a certain contribution from light scattering to the extinction of the solution described by curve 1 (Fig. 2a, inset) cannot be ruled out entirely. In view of this the optical density of the solution at any arbitrarily selected wavelength only characterizes the amount of dispersed graphite approximately. Apparently, a more accurate quantitative indication is provided by the integral intensity IG of the graphite absorption band in the electronic spectrum calculated after removal of the absorption of the GO (PRGO) from the spectral curve. Figure 2b and Table 1 show the dependence of the parameter IG on the position of the characteristic maximum lmax in the absorption spectra of the GO and PRGO colloids used for dispersion of the graphite, i.e., essentially on the degree of reduction of the GO. As seen, the amount of dispersed graphite increases in direct proportion to lmax, indicating that the photoreduction of GO leads to a significant increase of its ability to interact with the graphite. On the assumption that the absorption of light by the graphite obeys the Lambert–Beer law it can be concluded that the solution produced with PRGO with the maximum exposure in the investigated range (90 min) contains ~5 times more graphite than the colloid based on unexposed GO. The stabilization of the hydrophobic particles of graphite in an aqueous medium evidently results from pp-stacking interaction between the aromatic regions of the GO (PRGO) and the p-electronic system of the outer graphene sheets of the multilayer graphite particles during simultaneous electrostatic repulsion between the ionized carboxyl groups of the GO (PRGO), thereby preventing aggregation of the individual graphite particles. It is known [2, 19, 22] that the photoreduction of GO leads to significant enlargement of the aromatic regions and to decrease of the portion of sp3-hybridized carbon in the composition of its particles but at the same time has practically no effect on the carboxyl groups bonded to the GO framework. It is obvious that such changes in the structure of the GO must promote its pp* interaction with the graphite without loss of the hydrophilic interactions with the dispersion medium, and this is observed experimentally in the form of the relationship presented in Fig. 2b. Thus, in spite of the fact that the GO (PRGO) is not a surfactant in the classical sense (does not change the surface tension of the aqueous solutions and does not form micelles [28]) it can be regarded as a selective surfactant toward graphite and likewise potentially toward any hydrophobic micro/nano subjects and molecules that have a developed conjugated p-electronic system. Here the activity of GO as a surfactant can be varied by controlled photoreduction taking place under mild conditions without the use of additional reducing agents and stabilizers. The PRGO particles with lmax = 260 nm obtained as a result of exposure for 90 min are characterized by hydrodynamic sizes in the range of 200-500 nm (Fig. 2c, curve 1) with a distribution maximum at dHD = 300 nm (Table 1). Here larger micron-sized formations, which could indicate aggregation of the PRGO particles during the photoreaction, are not present in the solution. Interaction of the PRGO colloid with graphite in solution results in the appearance of PRGO-G particles whose hydrodynamic size amounts to 2-4 mm (Fig. 2c, curve 2). Here the amount of particles of reduced graphite oxide with dHD = 300 nm is reduced as a result of its partial entry into the composition of the heterostructure of the PRGO-G. On the whole the results of laser photocorrelation spectroscopy agree with the results of scanning electron micrography of the PRGO-G colloid (Fig. 2d), according to which most of the dispersed graphite particles have a lateral dimension of 2-5 mm. It should be noted that according to optical spectroscopy the initial graphite powder is extremely polydisperse and contains particles ranging from 1 to 200 mm and more. It can thus be concluded that the fraction of the smallest graphite particles from the initial assembly undergoes dispersion during interaction with PRGO.
287
Fig. 3. a, c, d) Scanning electron micrographs of nanotubes dispersed in aqueous PRGO colloids (90 min exposure) (inset in Fig. 3, a, SEM images of initial carbon nanotubes). b) Relationship between the integral intensity INT of the absorption band of the MCNTs and the position of the maximum lmax of the characteristic absorption band of the GO (PRGO) (inset, normalized absorption spectra of the aqueous colloid GO (1) and GO-MCNT (2)).
Interaction between GO (PRGO) and MCNTs. According to data from transmission electron microscopy [27] and also the data from scanning electron microscopy obtained in the present work, the investigated samples of MCNTs represent aggregates of closely intertwined nanotubes with external diameters of 20-25 nm and lengths of tens of micrometers. Ultrasonic treatment of such MCNTs in distilled water does not give stable dispersions, but the MCNTs that pass temporarily into the liquid phase are characterized by a high degree of aggregation (Fig. 3a, inset). At the same time ultrasonic treatment in colloidal GOs gives dispersions with an absorption band characteristic of MCNTs [20] with a maximum at 255 nm (Fig. 3b, inset, curve 2). The GO-MCNT colloids are stable with respect to sedimentation for 2-3 months. As seen from the photomicrographs presented in Fig. 3a, c, d, the dispersions contain mostly individual MCNTs, and here the fraction with the shortest nanotubes with lengths of up to 1 mm passes into the liquid phase. As in the case of graphite, the integral intensity INT of the absorption band of the MCNTs, calculated after subtraction of the absorption spectrum of the initial GO (PRGO) from the spectrum of the GO-MCNT (PRGO-MCNT) dispersion, was used as a quantitative parameter characterizing the content of MCNTs in the dispersion. From the relationship between INT and the position of the characteristic maximum in the absorption spectrum of the GO (PRGO) lmax presented in Fig. 3b it is seen that the photochemical reduction of the GO has a substantial effect on its interaction with the MCNTs. At the initial stage of 288
photoreduction the ability of the GO to interact with the MCNTs is somewhat reduced, as demonstrated by the decrease of the INT value (Fig. 3b), but increases again with increase of the exposure time and with shift of the position of lmax toward the longwavelength side as a result of increase in the degree of aromaticity of the PRGO particles. Thus, according to the spectral data in the PRGO-MCNT dispersion produced using the PRGO with the maximum value of lmax = 260 nm (90 min exposure) in the investigated range, INT is 40%-50% higher than the intensity of the band of the MCNT in the GO-MCNT colloid (Fig. 3b). Analysis of the photomicrographs of the carbon nanotubes dispersed in PRGO colloids made it possible to suggest two possible alternatives for stabilization. In the first of them, which is mostly realized in the case of large particles of GO (PRGO) with size in the order of 1 mm, the MCNTs are combined with the GO particles, being situated on the planes of the GO particles by analogy with the composites obtained in [17, 20]. (Examples of such geometry in the PRGO-MCNT particles are indicated by the arrows in Fig. 3c.) In the second alternative, which is more likely for PRGO particles of smaller size that form the bulk of the particles in the assembly [22], the PRGO particles are twisted around the MCNTs into one or more layers (see Fig. 3d). The sections of the MCNTs on which the GO particles are located are characterized by greater thickness and are therefore observed on the photomicrographs as regions of enhanced contrast. According to data in [23], at the stage of purification of the MCNTs by treatment with hot HNO3 the surface is functionalized by hydroxyl groups. Thus, in contrast to graphite the MCNTs can interact with the GO (PRGO) in two ways – by pp stacking interactions between the aromatic fragments of the GO (PRGO) particles and the surface of the MCNTs [15, 17, 20] and by the formation of hydrogen bonds involving the OFGs of the nanotubes and the GO (PRGO) particles. For this reason in both alternatives of the geometry of the GO(PRGO)-MCNT heterostructures it must be expected that the relationship between the amount of dispersed MCNTs and lmax will be more complicated in nature than in the case of graphite. As seen from Fig. 3b, the INT–lmax relationship is in fact nonlinear in nature and passes through a minimum. Attention is drawn to the relationship between the hydrodynamic size of the GO (PRGO) particles and the intensity of the absorption band of the dispersed MCNTs (Table 1): the higher the value of dHD characterizing the GO (PRGO) particles, the smaller the amount of MCNTs that can be dispersed and stabilized in such a solution of the GO (PRGO). Since, as discussed above, increase of the dHD value indicates increase in the planarity of the GO particles and decrease in their ability to form folds it can be expected that this tendency will also prevent twisting of the GO (PRGO) particles around the MCNT framework. On the other hand, removal of the OFG from the basal plane of the graphene oxide at the initial stage of photoreduction reduces the ability of the GO particles to form hydrogen bonds with the MCNTs, and this decrease is obviously not fully compensated by the enlargement of the aromatic subsystem of the GO. In the course of further photoreduction the proportion of sp2-hybridized carbon in the GO increases, and this leads to an increase in the probability of pp-stacking interaction between the PRGO and MCNTs and also between the individual aromatic fragments in the composition of the PRGO particles. The last factor promotes return of the PRGO particles to the “crumpled” form with a smaller dHD and also their twisting around the MCNTs, which as a whole leads to an increase in the amount of nanotubes dispersed in the solution. Thus, in this work the effect of the degree of photochemical reduction of colloidal graphene oxide on its ability to disperse and stabilize microparticles of graphite and carbon nanotubes in aqueous solutions was established. It was shown that the amount and size of the dispersed graphite particles increase many times with increase in the degree of reduction of the graphene oxide particles. In the case of carbon nanotubes the relationship between the amount of dispersed nanotubes and the length of photoreduction of graphene oxide passes through a minimum. This results from photoinduced removal of the oxygen-containing functional groups of the graphene oxide and a decrease of its tendency to interact with the partially hydroxylated surface of the nanotubes and also from a decrease in the ability of the small particles of graphene oxide to twist around the nanotubes. If photoreduced GO is used as stabilizer dispersion of graphite particles with average size amounting to 2-3 mm and nanotubes with lengths of up to 1 mm, which maintain long-term stability and sedimentation, are formed. The work was carried out with support from the State Foundation for Basic Research of Ukraine (project No. F41.2/005) and a grant from the President of Ukraine for Doctors of Sciences (project No. F47/20).
289
REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28.
290
J. J. Gooding, Electrochim. Acta, 50, No. 15, 3049-3060 (2005). Y. Shao, J. Wang, H. Wu, et al., Electroanalysis, 22, No. 10, 1027-1036 (2010). H. Sun, X. You, J. Deng, et al., Adv. Mater., 26, No. 18, 2868-2873 (2010). G. Wang, L. Zhang, and J. Zhang, Chem. Soc. Rev., 41, No. 2, 797-828 (2012). D. Chen, L. Tang, and J. Li, Chem. Soc. Rev., 39, No. 8, 3157-3180 (2010). D. W. Chang, H.-J. Choi, A. Filer, and J.-B. Baek, J. Mater. Chem. A, 2, No. 31, 12136-12149 (2014). P. V. Kamat, K. Tvrdy, D. R. Baker, and J. G. Radich, Chem. Rev., 110, No. 11, 6664-6688 (2010). J. Chen, B. Li, J. Zheng, et al., J. Phys. Chem. C, 116, No. 28, 14848-14856 (2012). Y. H. Jang, X. Xin, M. Byun, et al., Nano Lett., 12, No. 1, 479-485 (2012). J. Kim, V. C. Tung, and J. Huang, Adv. Energy Mater., 1, No. 6, 1052-1057 (2011). X. An and J. C. Yu, RSC Adv., 1, No. 8, 1426-1434 (2011). A. A. Green and M. C. Hersam, J. Phys. Chem. Lett., 1, No. 2, 544-549 (2010). A. B. Bourlinos, V. Georgakilas, R. Zboril, et al., Small, 5, No. 16, 1841-1845 (2009). E. E. Tkalya, M. Ghislandi, G. de With, and C. E. Koning, Curr. Opin. Colloid Interface Sci., 17, No. 4, 225-232 (2012). X. An, T. Simmons, R. Shah, et al., Nano Lett., 10, No. 11, 4295-4301 (2010). L. Tian, P. Anilkumar, L. Cao, et al., ACS Nano, 5, No. 4, 3052-3058 (2011). L. J. Cote, J. Kim, V. C. Tung, et al., Pure Appl. Chem., 83, No. 1, 95-110 (2011). D. Wei and Liu Y., Adv. Mater., 22, No. 30, 3225-3241 (2010). D. R. Dreyer, S. Park, C. W. Bielawski, and R. S. Ruoff, Chem. Soc. Rev., 39, No. 1, 228-240 (2010). C. Zhang, L. Ren, X. Wang, and T. Liu, J. Phys. Chem. C, 114, No. 26, 11435-11440 (2010). Y. Zhu, S. Murali, W. Cai, et al., Adv. Mater., 22, No. 35, 3906-3924 (2010). A. L. Stroyuk, N. S. Andryushina, N. D. Shcherban’, et al., Teor. Éksp. Khim., 48, No. 1, 1-11 (2012). [Theor. Exp. Chem., 48, No. 1, 2-13 (2012) (English translation).] N. V. Lemesh, A. I. Tripol’skii, I. E. Kotenko, et al., Teor. Éksp. Khim., 46, No. 5, 288-293 (2010). [Theor. Exp. Chem., 46, No. 5, 296-301 (2010) (English translation).] F. Bonaccorso, Z. Sun, T. Hasan, and A. C. Ferrari, Nature Photonics, 4, No. 9, 611-622 (2010). N. S. Andryushina and O. L. Stroyuk, Appl. Catal. B, 148/149, 543-549 (2014). N. S. Andryushina, O. L. Stroyuk, I. B. Yanchuk, and A. V. Yefanov, Colloid Polym. Sci., 292, No. 2, 539-546 (2014). R. L. D. Whitby, V. M. Gun’ko, A. Korobeynik, et al., ACS Nano, 6, No. 5, 3967-3973 (2012). C.-J. Shih, S. Lin, R. Sharma, et al., Langmuir, 28, No. 1, 235-241 (2012).
