Nanoscale Perforation of Graphene Oxide during Photoreduction ...

2 downloads 1 Views 3MB Size Report
Nov 21, 2016 - ABSTRACT: We present a study of the process of reduction of thin graphene oxide (GO) films consisting of flakes with lateral size of up to 100 ...

Article pubs.acs.org/JPCC

Nanoscale Perforation of Graphene Oxide during Photoreduction Process in the Argon Atmosphere Maxim K. Rabchinskii,† Vladimir V. Shnitov,*,† Artur T. Dideikin,† Aleksandr E. Aleksenskii,† Svetlana P. Vul’,† Marina V. Baidakova,†,‡ Igor I. Pronin,†,‡ Demid A. Kirilenko,†,‡ Pavel N. Brunkov,†,‡ Juliane Weise,§ and Serguei L. Molodtsov‡,§,∥ †

Ioffe Institute, 26 Politekhnicheskaya, Saint-Petersburg 194021, Russia ITMO University, 49 Kronverksky Pr., Saint-Petersburg 197101, Russia § TU Bergakademie Freiberg, Akademiestrasse 6, Freiberg 09599, Germany ∥ European XFEL GmbH, Holzkoppel 4, Hamburg, Germany ‡

ABSTRACT: We present a study of the process of reduction of thin graphene oxide (GO) films consisting of flakes with lateral size of up to 100 μm through soft ultraviolet irradiation in the argon atmosphere. It was found out that the reduction process leads to a significant decrease in the overall content of the basal-plane functional groups, namely, epoxides and hydroxyls, but with simultaneous increase in the total number of the edge-located carboxyl groups. Obtained transmission electron microscopy images showed that this effect is related to formation of nanoscale holes in the course of reduction. Based on the data obtained, we have proposed a mechanism of the observed GO structural modification in terms of photoinduced chemical reactions between the carbon network and functional groups. These reactions result in progressive growth of the initially existing and newly formed vacancies with formation of the nanoholes with size of up to 100 nm. Thus, reduced graphene oxide films with the restored conjugated network and many edges terminated with carboxyl groups can be probably obtained via the photoreduction process in the argon atmosphere and further used in several applications, such as production of gas sensors and organic light-emitting devices.

1. INTRODUCTION Since its discovery, graphene has attracted an enormous amount of attention because of its unique electronic, optical, and mechanical properties1−3 together with a great variety of its perspective applications, including supercapacitors, transparent conductive coatings, solar cells, and so on.4,5 Various methods, such as chemical vapor deposition on transition metals,6 conversion of SiC(0001) to graphene via sublimation,7 and reduction of graphene oxide8,9 (GO), have been developed for producing high-quality graphene films with a desirable structure. Among them, reduction of the graphene oxide, a chemical derivative of graphene, is one of the most promising ways for production of graphene-based materials and devices due to low cost and high yield of GO synthesis process together with the possibility to form large-scale coatings on various substrates prior to its reduction as a result of hydrophilic nature of graphene oxide. With this method, obtaining of GO flakes with large lateral size and selection of an appropriate reduction method are the key factors in obtaining high-quality conducting transparent films convenient for further applications. High-temperature annealing10,11 in a reducing environment appeared to be one of the most effective GO reduction methods resulting in nearly complete removal of oxygencontaining groups and in restoring the sp2-bond network. © 2016 American Chemical Society

However, annealing also damages the substrate thus making this reduction method incompatible with the majority of potential device fabrication technologies. On the other hand, reduction of the graphene oxide films through ultraviolet irradiation under mild conditions promises to be a rapid and facile way of producing reduced graphene oxide (rGO) films with the optimal quality of the obtained material and negligible influence on the substrate. As compared with conventional thermal and chemical routes, photoreduction of GO additionally shows such advantages as high efficiency, tunable reduction degree, and flexible patterning, making this method highly attractive for its integration in production of graphene-based microdevices.12,13 Several works were published, studying the processes of reduction graphene oxide films by ultraviolet radiation of a wide range of wavelengths, exposure times, irradiation intensities, and environment compositions, which can be based on on either photothermal or photochemical reactions that lead to removal of oxygen-containing functional groups.12−19 It was demonstrated that chemical composition and structural properReceived: August 30, 2016 Revised: November 18, 2016 Published: November 21, 2016 28261

DOI: 10.1021/acs.jpcc.6b08758 J. Phys. Chem. C 2016, 120, 28261−28269

Article

The Journal of Physical Chemistry C

Figure 1. (a) TEM image and selected area diffraction (SAED) pattern (in the inset) of an initial GO flake. (b) AFM image of several GO flakes on a silicon wafer. (c) Size distribution of GO flakes in aqueous suspension measured by the optical diffraction method.

ties of the rGO films obtained by this method depend strongly on the parameters of environment and radiation used. In particular, the photoreduction of graphene oxide films in the oxygen-containing environment16,17 leads to etching of graphene oxide flakes due to reactions between GO films and active oxygen species arising in the reduction process. Reduction of GO films by ultraviolet irradiation in vacuum results in removing surface functional groups, however, with retaining boundary groups.18 On the other hand, reduction in the presence of hydrogen19 results only in decreasing the amount of oxygen-containing species in the absence of any etching processes able to cause structural modifications of the obtained material. Hereby we present a study of the reduction of graphene oxide films composed from GO flakes with lateral size of up to 100 μm under the action of ultraviolet irradiation in the argon atmosphere. We have demonstrated that the method applied leads to nearly complete removal of basal-plane groups and restoration of the sp2-hybridizied carbon network. However, as shown below, the photoreduction process also resulted in the film modification with formation of a large number of nanoholes distributed inside the GO flakes, though there was no free oxygen in the reduction chamber during the process. We have proposed a simple model describing well the observed structural modification of GO films based on etching the preexisting and newly formed vacancies due to chemical reactions occurring between the functional groups and carbon network at the edges of those defect sites in the course of the photoreduction. These finding allowed us to reveal new aspects of the graphene oxide film reduction under ultraviolet irradiation and to indicate its dependence on the environment and radiation parameters.

shielding gas due to its inert nature and high atomic weight (39.9 Da) exceeding molecular weights of both gaseous oxygen and most of the oxygen-containing species that can be formed during reduction process (e.g., water, carbon monoxide, etc.), thus ensuring their complete removal from the films surface before and during the process. Prior to starting the reduction process, oxygen was removed from the reduction chamber by purging it for 10 min with a 150 sccm flow of pure argon gas; immediately after that the UV source was switched on, and the process of reduction was launched at the same gas flow rate. The constant flow of argon through the reduction chamber also provided cooling of the irradiated films, minimizing their heating and excluding photothermal reduction of graphene oxide. Temperature of the sample was monitored by means of thermocouple device and was measured to be at the range of 40−50 °C during the irradiation process. The reduced graphene oxide films produced by high-temperature annealing at 800 °C for 1 h in the hydrogen environment were used as reference samples for comparing the reduction degrees and chemical compositions of the obtained material. Samples fabricated by ultraviolet irradiation and high-temperature annealing are designated hereinafter as rGO-UV and rGOHT, respectively. To perform transmission electron microscopy (TEM; Jeol JEM-2100F, 200 kV, point-to-point resolution of 0.19 nm), one drop of the aqueous GO suspension with the concentration of 7·10−4 wt % was deposited on a copper and nickel grids. Surface morphology and lateral size of the GO flakes were analyzed using atomic force microscopy (AFM; Veeco Dimension 3100) in the tapping mode by using RTESP probes. Monolayer films for AFM imaging were prepared by the Langmuir−Blodgett method, according to the procedures published elsewhere.11,21 The electrical conductivity of the GO and rGO multilayer films deposited on SiO2/Si substrate were measured with TUNA AFM mode using DDESP-FM-10 diamond probe as the first contact and tungsten tip mounted on the manipulator as the second one. In order to obtain size distribution of GO flakes in aqueous solution, optical diffraction measurements (OD; Mastersizer 2000) were carried out. The GO and rGO films were primarily analyzed by ultraviolet− visible absorption spectroscopy (UV−vis; Shimadzu-2450). Fourier transform infrared spectroscopy was further performed on the Infralum-08 FTIR spectrometer equipped with the attenuation total reflectance attachment in order to study chemical composition before and after reduction processes. Xray photoelectron spectroscopy (XPS) measurements were

2. EXPERIMENTAL SECTION Graphene oxide was synthesized by a modified Hummers method.20 In the process, sonication was excluded in order to prevent damaging of graphene oxide flakes and obtain suspensions with the utmost size of GO flakes. To prepare GO films, 200 μL of GO suspension 0.07 wt % in concentration was drop-casted on a silicon/quartz wafer and dried overnight at room temperature (20 °C). The GO reduction by ultraviolet irradiation was carried out using a 30 W deuterium lamp with a quartz window at wavelengths ranging from 186 to 360 nm (3.6−6 eV) with the maximum intensity at 220 nm (5.63 eV). Samples were placed 5 mm behind the lamp quartz window in a quartz cylinder at room temperature. Argon was used as a 28262

DOI: 10.1021/acs.jpcc.6b08758 J. Phys. Chem. C 2016, 120, 28261−28269

Article

The Journal of Physical Chemistry C carried out on a Thermo Fisher ESCALAB 250Xi XPS system with a monochromatic Al Kα X-ray source (1486.6 eV). To compensate possible negative consequences of GO films, surface-charging XPS spectra obtained for low-conducting GO and rGO-UV films were aligned with respect to the spectra obtained for well-conducting rGO-HT film. Position of the latter one was corrected according to the position of the Au 4f7/2 line (84.0 eV). The C 1s spectra curve fittings were performed by using Raymond Kwok’s free XPS Peak 4.1 software.

3. RESULTS AND DISCUSSION 3.1. Morphology of GO Flakes. Figure 1a presents a TEM image of an initial GO flake, demonstrating the flake structure continuity and absence of any observable structural defects, such as rips or holes with lateral size of more than tens of nanometers, distributed within GO flakes. Sharpness of the obtained diffraction spots and their intensities ratio also manifest the monolayer character of the GO flake and suggest its well-preserved crystalline structure with the long-range order of minimum of several tens of nanometers. Figure 1b presents a typical AFM image of several GO flakes with the lateral size of up to 100 μm which corresponds to the highest values reported previously.21,22 Obtaining of single-layer GO flakes with such a lateral size is crucial for the formation and subsequent reduction of GO films due to critical dependence of their physical properties23 and chemical composition, in particular the number of formed during reduction process highly stable boundary chemical groups11,24,25 such as phenols, ketones, and ethers, on the GO flake average size. In order to further evaluate the size distribution of the obtained GO flakes, optical diffraction (OD) measurements of the GO aqueous dispersion were carried out. Figure 1c presents the numerical size distribution of GO flakes and demonstrates that, though the suspension contains observed GO flakes with the lateral size of more than 100 μm, their numerical fraction is lower (∼2.5%) than that of smaller flakes whose sizes range from 10 to 20 μm (10.8%). This can be explained by the fact that large-scale GO flakes (up to 100 μm) tend to be softer than smaller ones, thus being less stable and having a higher chance to crack resulting in formation of GO flakes with lateral size of several tens of micrometers. On the other hand, no flakes with the lateral size below 5−10 μm were observed in the obtained suspension. Collectively, these results suggest that obtained GO films consist of monolayer flakes with lateral size of 10−100 μm with the absence of any initially presented observable defects. 3.2. UV−Vis and IR Spectra. To evaluate the degree of reduction of the rGO-UV and rGO-HT films, we first measured UV−vis spectra that are represented in Figure 2. As it can be seen, graphene oxide exhibits two distinctive features, the main absorption peak at 230 nm that can be attributed to π−π* transitions of CC in amorphous carbon system and a broad shoulder with center at ∼300 nm that is commonly attributed to n−π* transitions in CO bonds.26 However, this absorption band can be also assigned to optical transitions between π and π* states in the finite-sized molecular sp2 domains and the nanometer-size sp2 clusters.27,28 On the completion of the GO film reduction, the 230 nm peak redshifts to higher wavelengths (∼270 nm), while the shoulder at 300 nm disappears, and overall absorption in the range up to near-infrared region rises significantly due to removing of functional groups from the GO basal plane and restoring its conjugated structure.29 As one can see, the rGO-UV and rGO-

Figure 2. UV−vis absorption spectra of the initial GO, rGO-UV, and rGO-HT films.

HT films exhibit almost similar absorption spectra, which suggests comparable degree of removal of the basal-plane functional groups performed by both reduction methods. Observed modification of the UV−vis spectra during the reduction process is also accompanied by drastic enhancement of the films’ conductivity. Initial GO films show an insulating nature with a sheet resistance value of about 1012 Ω/sq or higher. However, sheet resistance for films after both hightemperature annealing and photoreduction process become significantly lower and measured to be 106 Ω/sq for rGO-UV and 105 Ω/sq for rGO-HT films. These results coincide well with the UV−vis spectroscopy results, confirming suggestion that both methods lead to nearly complete restoration of sp2conjugated graphene network. Further characterization of the GO and rGO films was performed by using FTIR spectroscopy. Figure 3 (red curve)

Figure 3. FTIR spectra of the initial GO and rGO films, obtained by the photoreduction process and high-temperature annealing.

presents the IR spectrum of the initial graphite oxide. The spectrum indicates the presence of oxygen-containing functional groups together with interlayer and adsorbed water, which complies with the published data.24,30,31 In particular, a set of overlapping absorption bands in the 3000−3700 cm−1 region are commonly attributed to O−H stretching in hydroxyl, carboxyl groups, and water molecules. The existence of the 28263

DOI: 10.1021/acs.jpcc.6b08758 J. Phys. Chem. C 2016, 120, 28261−28269

Article

The Journal of Physical Chemistry C

Figure 4. (a) Survey and (b) C 1s XPS spectra obtained for initial GO and GO reduced by UV irradiation (rGO-UV) and thermal treatment (rGOHT). For clarity, subtracted from experimental C 1s spectra Shirley backgrounds are not shown, while the spectra and their fits are vertically offset from the respective fitting components.

same GO and rGO films were further carefully studied by using the XPS technique. The survey and high-resolution C 1s core level XPS spectra obtained in this study are shown in Figure 4a and 4b, respectively. The analysis of the survey spectra enabled us to determine elemental composition of the films and confirmed expectedly low content of impurities (mostly sulfur) which appeared to be less than 2.0 at. %. More detailed data on GO and rGO chemical composition, including a relative content of differently bonded carbon atoms and different oxygen-containing functional groups, were obtained from highresolution C 1s spectra with the help of curve-fitting technique. The spectra were fitted using the set of eight Gaussian− Lorentzian sum peaks (Gaussian by 90% and Lorentzian by 10%) consisting of seven symmetric and one asymmetric peak (peak CC). Half of the peaks positioned, respectively, at the binding energies (BE) of 283.9, 284.6, 285.1, and 291 eV were attributed to the following pure carbon C 1s components: the vacancy defects of graphene lattice (peak C−V),32 sp2-bonded carbons located within domains of prefect graphene lattice (peak CC), sp3-bonded carbons within domains of graphene lattice distorted by attachment of oxygen-containing groups (peak C−C), and shakeup satellite of the peak CC (peak π−π*).33,34 Asymmetry of the CC peak is probably related to screening of the core holes left by the photoelectrons, which, as it was demonstrated, reveals itself not only in C 1s spectra of graphite and graphene but also in the similar spectra of highly reduced GO films (Figure 4b,4c)35,36 Other four peaks positioned, respectively, at BE of 286.1, 286.8, 287.7, and 288.8 eV were interpreted as oxygen-related C 1s components and assigned to the following functional groups: phenol (C−OH(p)), hydroxyl and epoxide (C−OH and O−C−O), ketone (CO), and carboxyl (OC− OH).33,34,37 The assignment of almost all above-listed C 1s spectra components is quite common, only the assignment of the low-energy C−V component for the first time suggested in

interlayer and adsorbed water can be also confirmed by the presence of the absorption band at 1620 cm−1. At the same time, the absorption band at 1720 cm−1 we assign to CO stretching in carboxyl groups and ketones which are adjacent to benzene rings, whereas the absorption band at 1040 cm−1 possibly corresponds to the edge phenols. A set of the overlapping bands at 1365 and 1415 cm−1 is attributed to deformations of the basal-plane hydroxyl groups and carboxyl groups, respectively, while the high-intensity sharp peak at 1220 cm−1 is due to vibration of the epoxy groups. The absorption band at 970 cm−1 corresponds to peroxides or five-membered ring lactols30 and absorption band at 1280 cm−1 can be attributed to ethers.25 After the UV photoreduction, the FTIR spectrum significantly changes, reflecting the elimination of several oxygen-containing functional groups. Absorption at 3000−3700 cm−1 noticeably decreases with diminishing of intensity of the 1620 cm−1 absorption band due to almost complete removal of the interlayer and adsorbed water. The absorption band at 1220 cm−1 also disappears thus confirming that epoxy groups have been eliminated during the reduction. The new 1580 cm−1 absorption band corresponding to CC vibrations arises thus suggesting recuperation of the conjugated aromatic structure. However, distinguishable absorption bands that correspond to the phenol and carboxyl/ketone groups (1040, 1415, and 1720 cm−1) still remain and become dominant. At the same time, the IR spectrum of rGO-HT does not contain any absorption bands corresponding to either basal-plane functional groups or edge ones. Thus, the IR spectroscopy data show that ultraviolet irradiation in an argon atmosphere results in substantial decrease in the amounts of the basal-plane groups and interlayer water, while a decrease in the number of the edge groups, carboxyl and ketone, appeared to be considerably less prominent. 3.3. XPS Characterization of GO and rGO. To verify the conclusions made on the base of FTIR spectroscopy data, the 28264

DOI: 10.1021/acs.jpcc.6b08758 J. Phys. Chem. C 2016, 120, 28261−28269

Article

The Journal of Physical Chemistry C

caused by almost complete removal of respective basal-plane groups and emergence of edge phenol groups (peak C− OH(p)) (see Table 1), is quite similar for both processes and well corresponds with the current literature data on GO reduction. To our knowledge, the above-mentioned substantial growth of the amount of carboxyl groups has never been observed yet in the case of UV photoreduction performed in inert gas atmosphere. After Dimiev et al.40 we consider that the majority of the edge OC−OH groups are located within the graphene basal planes or, more exactly, on the various internal edges of GO flakes including edges of nanosized holes and even vacancy defects. Therefore, it is very likely that the significant increase in the relative content of carboxyl groups observed in the C 1s spectrum of rGO-UV film is actually caused by the UV irradiation induced increase in the number and size of nanoholes with their intensive functionalization by carboxyl groups. Interestingly, this assumption can be quite unexpectedly confirmed by the evolution of the pure carbon C−V peak, which, as was stated above, can be ascribed to the nonterminated edge sites of the vacancy defects. As Figure 4b and Table 1 show, in the case of rGO-HT film the relative area of this peak appears to be approximately three times greater than that in the case of rGO-UV. At first sight, such a finding seems to be rather odd since 800 °C vacuum annealing of GO film (accompanied by an obvious increase in the mobility of single carbon atoms) must provide much more effective healing/ removal of such defects than UV irradiation performed at room temperature. Nevertheless, both effects, fourfold decrease in the relative area of C−V peak and threefold increase in the relative area of OC−OH peak, can be explained within the frames of the very simple model based on the following idea. The observed substantial decrease in the relative area of C−V peak is caused not by the healing of graphene lattice vacancy defects but by their etching, induced by UV irradiation, which gradually converts them into sufficiently large nanoholes and provides simultaneous termination of the edges of these nanoholes by carboxyl and phenol functional groups. 3.4. TEM Studies. To confirm this suggestion, we performed TEM imaging of graphene oxide films treated by the ultraviolet irradiation with different exposure times. Figure 5a demonstrates a TEM image of the GO film after 1 min UV exposure, which manifests retention of its continuous structure but with formation of a distinguishable hole (noted on the image by the arrow) with the lateral size of ∼5 nm. However, 5 min irradiation of the GO film results in formation of a set of random-shaped holes 5−25 nm in lateral size located close to each other (Figure 5b). Further increase in the exposure time leads to a considerable increase in the lateral size of the observed holes due to their progressive growth and merging into larger ones along with formation of new holes. This results in formation of the highly perforated structure of the film which in particular can explain significant difference in conductivity of rGO-UV and rGO-HT films observed, although these samples have nearly equal degrees of reduction of basal plane functional groups as is demonstrated by the results of FTIR and XPS measurements. At the same time, the observed structural modification almost fully stops after 20 min of ultraviolet irradiation of monolayer samples (Figure 5c), thus suggesting a time-limited nature of the perforation process. Thereafter, no significant changes in the number and lateral size of the holes are observed. Note that for multilayer and monolayer films

work 32 may be considered as a somewhat tentative one. However, it is quite reasonable due to a large amount of such type defects in the structure of GO flakes as a result of the synthesis process.38,39 The results obtained by quantitative analysis as shown in Figure 4 XPS spectra are presented in Tables 1 and 2. It is also worth noting that the shape of the Table 1. Relative Fractionsa of C 1s Spectra Components Obtained for Initial and Reduced GO Films C−V CC C−C C−OH(p) C−OH and O−C−O CO OC−OH π−π*

GO

rGO-UV

rGO-HT

0.079 0.317 0.109 ∼0 0.420 0.045 0.028 0.002

0.019 0.702 0.091 0.011 0.043 0.010 0.101 0.021

0.061 0.830 0.028 0.018 0.010 0.016 0.018 0.042

a

Relative fraction of each component was determined via dividing its area (AX) by the total area of all C 1s components (ATot)

Table 2. Oxidation Degree (O/C)a Obtained from Survey and C 1s Spectra survey spectra C 1s spectra

GO

rGO-UV

rGO-HT

0.425 0.493

0.310 0.165

0.062 0.062

a

In the case of survey spectra, O/C values were obtained using the areas of respective O 1s and C 1s lines determined after the Shirley background subtraction and corrected in accordance with respective sensitivity factors. In the case of high-resolution C 1s spectra, these values were obtained via dividing the common area of four oxygenrelated components by the total area of the all C 1s components.

asymmetric CC peak was chosen so as to provide complete matching between both values of O/C ratio obtained for rGOHT film (see Table 2). Tables 1 and 2 demonstrate that initial GO films had sufficiently high degree of oxidation mostly provided by a large fraction of such basal-plane functional groups as hydroxyl and epoxide. Obvious discrepancy between values of O/C ratio obtained from survey and C 1s spectra of GO and rGO films is likely to arise from the presence in multilayer GO films of a large amount of intercalated water that is not chemically bonded with carbon atoms and, therefore, can reveal themselves only in survey XPS spectra. On the other hand, values of O/C obtained from C 1s spectra may be also significantly exaggerated due to presence in GO content of a large fraction of epoxide groups (C−O−C) each of which can oxidize two carbon atoms. Careful analysis of C 1s spectra and obtained on their base quantitative data shows that, despite close similarity between the spectra obtained for rGO-UV and rGO-HT films, some of their components demonstrate quite different evolution which strongly depends on the type of GO reduction process. In particular, UV photoreduction leads to more than threefold increase of carboxyl fraction (peak OC−OH), while thermal reduction causes its 40% decrease. As well, GO photoredution is accompanied by the decrease in the area of assigned to graphene vacancy defects peak C−V which appears to be much more pronounced than the decrease of the respective peak in the case of thermally reduced GO. Evolution of other C 1s components, such as elimination of peak C−OH and C−O−C 28265

DOI: 10.1021/acs.jpcc.6b08758 J. Phys. Chem. C 2016, 120, 28261−28269

Article

The Journal of Physical Chemistry C

Figure 5. TEM images of the initial GO and rGO-HT films after ultraviolet irradiation. (a−c) Monolayer GO films after 1, 5, and 20 min of UV irradiation, respectively. (d) rGO-HT film after 20 min of UV irradiation. Arrow indicates that the hole appeared after 1 min of the UV photoreduction.

only a low number of functional groups remains after hightemperature annealing (as demonstrated by both the FTIR and XPS measurements), no significant structural modification of the film is expected to be observed after ultraviolet irradiation. Indeed, as shown in Figure 5d, the rGO-HT film after UV irradiation contains a large number of holes less than 5 nm in diameter, which drastically differs from the case when several large holes, grouped together, are observed in the rGO-UV films after the same treatment. Note that the presence of holes in rGO-HT coincides well with the previously reported results25,42 and is related to the elimination of oxygencontaining functional groups during the annealing process. Consequently, the obtained results strongly confirm the existence of a relationship between the observed structural modification and overall content of the oxygen-containing functional groups on the treated film. This idea is also supported by the absence of any etching and structural modification processes in monolayer and multilayer films, exposed to UV irradiation for 20 and 120 min, respectively, when all the functional groups that might be involved in the etching process are eliminated.

different duration of irradiation is required to transfer observed structural modification into the stage of saturation. This observation can be explained by the fact that UV radiation is mainly absorbed by the first layer of the GO films and intensity of the UV radiation reaching the bottom layers is significantly decreased thus slowing structural modification process in these layers. Generally, etching of the graphene oxide flakes and formation of nanoholes were already observed earlier during the photoreduction process in oxygen-containing atmosphere16,17 or in aqueous GO solutions.41 These processes are generally based on formation of either ozone or hydroxyl radicals under UV irradiation and their subsequent reaction with GO film or flakes that leads to their etching and formation of nanoholes. However, in our case there are no sources of reactive oxygen in the reducing environment; thus, the observed structural modification could be based only on chemical reactions between functional groups that cover the surface and boundaries of GO flakes. To verify this assumption, further TEM imaging was performed in order to analyze the morphology of the rGO-HT film additionally treated by ultraviolet irradiation for 20 min in the argon atmosphere. As 28266

DOI: 10.1021/acs.jpcc.6b08758 J. Phys. Chem. C 2016, 120, 28261−28269

Article

The Journal of Physical Chemistry C

Scheme 1. Progressive Etching of the Vacancy Due to Chemical Reactions between Basal-Plane and at the Edge OxygenContaining Functional Groups

3.5. The Model of the Graphene Oxide Perforation during Photoreduction Process. Generally, photoreduction of graphene oxide can proceed due to both photochemical and photothermal processes. Since the temperature of the sample during the photoreduction process had not exceeded 50 °C, which is not enough for thermal deoxygenation of graphene oxide, we assume that the photothermal effect is negligible in our case and the photochemical mechanism is mainly involved in reduction of the films and observed formation of nanoscale holes. Consequently, on the basis of the above-discussed data, we have proposed a simple model which relies on photochemical reactions occurring between functional groups at the edges of such structural defects as mono- or bivacancies and holes with the lateral sizes of 1−2 nm, which initially exist in each GO flake due to the synthesis process (Scheme 1). Edges of these defects are chemically active and terminated mainly with ketone and phenol groups,25,31,40 while the surrounding basal plane is covered with hydroxyl and epoxy groups (structure 1). During the reduction process, ultraviolet irradiation triggers photoinduced condensation of two neighboring hydroxyl groups resulting in formation of a water molecule and ionized hydroxyl group that then commonly transforms into an epoxy group. However, if the condensation process proceeds near the edge site terminated by the ketone group, an intermediate structure can be formed consisting of ketone, ionized hydroxyl group, and that water molecule (structure 2). Subsequent redistribution of the electron density in this structure leads to rearrangement of chemical bonds and, hence, to a cleavage of the C−C bond with simultaneous formation of phenol and carboxyl groups concurrently with the hole enlargement (structure 3). The obtained structure is essentially stable. But, it can be further modified, if an additional hydroxyl group has been initially located next to

the carboxyl group formed in the transition from structure 2 to structure 3 or shifted to it due to photoinduced and thermodriven motion of hydroxyl groups to the edges of the nanoholes and GO flake26,43,44 (structure 4), which is caused by electrostatic repulsion between basal-plane functional groups. Further interaction between carboxyl and hydroxyl groups (structure 5) gives rise to CO and H2O molecules production and, as a result, formation of a new ketone group, and then the composition and distribution of the groups return to the initial state (structure 6). Thus, provided the amount of the basalplane hydroxyl groups is sufficient and they gradually migrate to the edge sites, carboxyl groups can alternately arise and disappear, promoting progressive growth of the hole. At the same time, migration of epoxy groups to the edge of the growing nanoholes leads to formation of new additional ketones9,44 which at once are getting involved in the process. The net result is that overall number of these carboxyl groups increases along with the enlargement of the nanoholes. Note that the transition from structure 2 to structure 3 also can be caused by the epoxy group ionization directly at the ketone. But, extra water molecules are required in this case, for instance, those that arose due to either the carboxyl group removal (transition from structure 5 to structure 6) or as a result of hydroxyl group condensation (transition from structure 1 to structure 2). In addition, enlargement of the holes according to the represented scheme is accompanied by their merging into larger ones and by formation of additional holes, resulting in perforation of GO flakes. Those additional holes arise due to etching the new vacancies formed in disruption of the carbon network caused by straightforward photoreduction of the basalplane groups with formation of CO molecules, as was demonstrated previously.45 However, the decrease in the 28267

DOI: 10.1021/acs.jpcc.6b08758 J. Phys. Chem. C 2016, 120, 28261−28269

The Journal of Physical Chemistry C



overall content of basal-plane functional groups makes elimination of carboxyl groups via reaction with incoming hydroxyl groups less probable than respective transformation of ketones into carboxyl groups. This leads to preservation of the carboxyl groups and their accumulation together with termination of the GO sheet perforation process, resulting in formation of reduced GO films both with strongly perforated structure and with substantial increase in the number of edge carboxyl groups. Since the above-mentioned reactions involve water molecules originating from condensation of hydroxyl groups as well as from elimination of carboxyl groups, the water desorption rate should be as low as possible for reaction to proceed. As the desorption rate in vacuum is significantly higher than in argon atmosphere, etching of GO films in vacuum is hardly possible which correlates with results obtained by Shulga et al.18 We also suppose that the presented mechanism of the structural modification might require UV radiation with wavelengths, lying within the range of the graphene oxide main absorption band centered at 230 nm. Irradiation of GO films with UV of such wavelengths leads to the optical transitions between disorder-induced localized states of the carbon atoms in distorted network46 and thus intensifies the motion of hydroxyl and epoxy groups to the edge sites and activating etching process. Radiation of such wavelengths might be obtained with a deuterium lamp, whereas high-pressure mercury lamps cannot ensure the respective spectrum range.

Article

AUTHOR INFORMATION

Corresponding Author

*E-mail address:[email protected]ffe.ru . Tel. : +7 812 2927142. ORCID

Maxim K. Rabchinskii: 0000-0003-4264-7147 Vladimir V. Shnitov: 0000-0001-7605-2884 Aleksandr E. Aleksenskii: 0000-0002-5004-6993 Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS



REFERENCES

The presented work was partially financially supported by the Russian Foundation for Basic Research. Participation of A.T.D., M.K.R., and P.N.B. was supported by grant no. 15-02-05153, while participation of D.A.K. by grant no. 16-32-60165. The AFM and TEM studies were carried out at the Joint Research Center “Material science and characterization in advanced technology”. M.V.B., I.I.P., D.A.K., P.N.B., and S.L.M. are grateful to ITMO University for support of the work within the frame of UniFEL Center. Authors are also grateful to Professor A. Ya. Vul′ for the fruitful discussions.

(1) Sarma, D. S.; Adam, S.; Hwang, E. H.; Rossi, E. Electronic Transport in Two-Dimensional Graphene. Rev. Mod. Phys. 2011, 83, 407−470. (2) Bonaccorso, F.; Sun, Z.; Hasan, T.; Ferrari, A. C. Graphene Photonics and Optoelectronics. Nat. Photonics 2010, 4, 611−622. (3) Scarpa, F.; Adhikari, S.; Srikantha Phani, A. Effective Elastic Mechanical Properties of Single Layer Graphene Sheets. Nanotechnology 2009, 20, 065709. (4) Huang, Y.; Liang, J. J.; Chen, Y. S. An Overview of the Applications of Graphene-Based Materials in Supercapacitors. Small 2012, 8, 1805−1834. (5) Novoselov, K. S.; Fal’ko, V. I.; Colombo, L.; Gellert, P. R.; Schwab, M. G.; Kim, K. A Roadmap for Graphene. Nature 2012, 490, 192−200. (6) Mattevi, C.; Kim, H.; Chhowalla, M. A Review of Chemical Vapour Deposition of Graphene on Copper. J. Mater. Chem. 2011, 21, 3324−3334. (7) De Heer, W. A.; Berger, C.; Wu, X.; First, P. N.; Conrad, E. H.; Li, X.; Li, T.; Sprinkle, M.; Hass, J.; Sadovski, M. L.; Potemski, M.; Martinez, G.; et al. Epitaxial Graphene Electronic Structure and Transport. Solid State Commun. 2007, 143, 92−100. (8) Dreyer, D. R.; Park, S.; Bielawski, C. W.; Ruoff, R. S. The Chemistry of Graphene Oxide. Chem. Soc. Rev. 2010, 39, 228−240. (9) Gao, W.; Alemany, L. B.; Ci, L.; Ajayan, P. M. New Insights into The Structure and Reduction of Graphite Oxide. Nat. Chem. 2009, 1, 403−408. (10) Ganguly, A.; Sharma, S.; Papakonstantinou, P.; Hamilton, J. Probing the Thermal Deoxygenation of Graphene Oxide Using HighResolution in Situ X-ray-Based Spectroscopies. J. Phys. Chem. C 2011, 115, 17009−17019. (11) Aleksenskii, A. E.; Brunkov, P. N.; Dideikin, A. T.; Kirilenko, D. A.; Kudashova, Yu. V.; Sakseev, D. A.; Sevryuk, V. A.; Shestakov, M. S. Single Layer Graphene Oxide Films on a Silicon Surface. Tech. Phys. 2013, 58, 1614−1618. (12) Zhang, Y.-L.; Guo, L.; Xia, H.; Chen, Q.-D.; Feng, J.; Sun, H.-B. Photoreduction of Graphene Oxides: Methods, Properties, and Applications. Adv. Opt. Mater. 2014, 2, 10−28. (13) Li, H.; Bubeck, C. Photoreduction Processes of Graphene Oxide and Related Applications. Macromol. Res. 2013, 21, 290−297.

4. CONCLUSIONS The reduction of large-scale GO flakes by ultraviolet irradiation in the argon atmosphere has been studied by a number of different experimental techniques. It has been shown that such type of graphene oxide UV photoreduction ensures nearly complete removal of the basal plane functional groups and, thus, promotes considerable restoration of the aromatic structure of graphene flakes. However, this process is also accompanied by the formation of a large number of nanoholes gradually emerging in the structure of GO flakes and manifesting themselves not only in respective TEM images, but, as well, in corresponding increase in the relative content of carboxyl groups, which is clearly visible in XPS and IR spectra of rGO-UV films. As a result, the rGO-UV flakes with two distinctive features, substantially enlarged portion of sp2 πconjugated graphene domains and considerably extended length of the flakes overall boundaries, are formed. Because the majority of edge sites appear to be decorated with mainly carboxyl groups, the resulting material can be easily chemically modified by a great variety of reagents.8,47 Due to these reasons, UV-rGO flakes with perforated structure promise to become highly attractive for production of organic light-emitting diodes5,48 or high-sensitivity gas sensors.49 Further, based on the obtained data, we have proposed a possible mechanism of GO structural modifications occurring during their photoreduction. The core point of this mechanism is a gradual growth of numerous initial and newly formed vacancy defects caused by photoinduced interactions between hydroxyl and carboxyl/ketone chemical groups that take place at the edges of these defect sites. The model proposed complies rather well with the observed time dependence of structural modification process and explains the effect of graphene lattice etching even in the absence of any active oxygen species in the reduction chamber, although specific details of the observed process still remain unclear. 28268

DOI: 10.1021/acs.jpcc.6b08758 J. Phys. Chem. C 2016, 120, 28261−28269

Article

The Journal of Physical Chemistry C

(33) Fan, X. B.; Peng, W. C.; Li, Y.; Li, X. Y.; Wang, S. L.; Zhang, G. L.; Zhang, F. B. Deoxygenation of Exfoliated Graphite Oxide under Alkaline Conditions: A Green Route to Graphene Preparation. Adv. Mater. 2008, 20, 4490−4493. (34) Li, Y.; Chen, H.; Voo, L. U.; Ji, J.; Zhang, G.; Zhang, G.; Zhang, F.; Fan, X. Synthesis of Partially Hydrogenated Graphene and Brominated Graphene. J. Mater. Chem. 2012, 22, 15021−15024. (35) Darmstadt, H.; Roy, C. Surface Spectroscopic Study of Basic Sites on Carbon Blacks. Carbon 2003, 41, 2662−2665. (36) Cheung, T. T. P. X-ray Photoemission of Carbon: Lineshape Analysis and Application to Studies of Coals. J. Appl. Phys. 1982, 53, 6857−6862. (37) Perera, S. D.; Mariano, R. G.; Nijem, N.; Chabal, Y.; Ferraris, J. P.; Balkus, K. J., Jr. Alkaline Deoxygenated Graphene Oxide for Supercapacitor Applications: An Effective Green Alternative for Chemically Reduced Graphene. J. Power Sources 2012, 215, 1−10. (38) Gómez-Navarro, C.; Meyer, J. C.; Sundaram, R. S.; Chuvilin, A.; Kurasch, S.; Burghard, M.; Kern, K.; Kaiser, U. Atomic Structure of Reduced Graphene Oxide. Nano Lett. 2010, 10, 1144−1148. (39) Eigler, S.; Dotzer, C.; Hirsch, A. Visualization of Defect Densities in Reduced Graphene Oxide. Carbon 2012, 50, 3666−3673. (40) Dimiev, A.; Kosynkin, D. V.; Alemany, L. B.; Chaguine, P.; Tour, J. M. Pristine Graphite Oxide. J. Am. Chem. Soc. 2012, 134, 2815−2822. (41) Matsumoto, Y.; Koinuma, M.; Ida, S.; Hayami, S.; Taniguchi, T.; Hatakeyama, K.; Tateishi, H.; Watanabe, Y.; Amano, S. Photoreaction of Graphene Oxide Nanosheets in Water. J. Phys. Chem. C 2011, 115, 19280−19286. (42) Mao, S.; Pu, H.; Chen, J. Graphene Oxide and Its Reduction: Modeling and Experimental Progress. RSC Adv. 2012, 2, 2643−2662. (43) Sun, P.; Wang, Y.; Liu, H.; Wang, K.; Wu, D.; Xu, Z.; Zhu, H. Structure Evolution of Graphene Oxide during Thermally Driven Phase Transformation: Is the Oxygen Content Really Preserved? PLoS One 2014, 9, e111908 (accessed Nov 5, 2014). (44) Kim, S.; Zhou, S.; Hu, Y.; Acik, M.; Chabal, Y. J.; Berger, C.; De Heer, W.; Bongiorno, A.; Riedo, E. Room-Temperature Metastability of Multilayer Graphene Oxide Films. Nat. Mater. 2012, 11, 544−549. (45) Shulga, Y. M.; Martynenko, V. M.; Muradyan, V. E.; Baskakov, S. A.; Smirnov, V. A.; Gutsev, G. L. Gaseous Products of Thermo- and Photo-Reduction of Graphite Oxide. Chem. Phys. Lett. 2010, 498, 287−291. (46) Chien, C.-T.; Li, S.-S.; Lai, W.-J.; Yeh, Y.-C.; Chen, H.-A.; Chen, I.-S.; Chen, L.-C.; Chen, K.-H.; Nemoto, T.; Isoda, S.; Chen, M.; Fujita, T.; Eda, G.; Yamaguchi, H.; Chhowalla, M.; Chen, C.-W. Tunable Photoluminescence from Graphene Oxide. Angew. Chem., Int. Ed. 2012, 51, 6662−6666. (47) Potts, J. R.; Dreyer, D. R.; Bielawski, C. W.; Ruoff, R. S. Graphene-Based Polymer Nanocomposites. Polymer 2011, 52, 5−25. (48) Loh, K. P.; Bao, Q.; Eda, G.; Chhowalla, M. Graphene Oxide as a Chemically Tunable Platform for Optical Applications. Nat. Chem. 2010, 2, 1015−1024. (49) Prezioso, S.; Perrozzi, F.; Giancaterini, L.; Cantalini, C.; Treossi, E.; Palermo, V.; Nardone, M.; Santucci, S.; Ottaviano, L. Graphene Oxide as a Practical Solution to High Sensitivity Gas Sensing. J. Phys. Chem. C 2013, 117, 10683−10690.

(14) Xue, Y.; Zhu, L.; Chen, H.; Qu, J.; Dai, L. Multiscale Patterning of Graphene Oxide and Reduced Graphene Oxide For Flexible Supercapacitors. Carbon 2015, 92, 305−310. (15) Aleksenskii, A. E.; Vul’, S. P.; Dideikin, A. T.; Sakharov, V. I.; Serenkov, I. T.; Rabchinskii, M. K.; Afrosimov, V. V. Etching of Wrinkled Graphene Oxide Films in Noble Gas Atmosphere under UV Irradiation. Nanosyst.: Phys., Chem., Math. 2016, 7, 81−86. (16) Tu, Y.; Utsunomiya, T.; Ichii, T.; Sugimura, H. VacuumUltraviolet Promoted Oxidative Microphotoetching of Graphene Oxide. ACS Appl. Mater. Interfaces 2016, 8, 10627−10635. (17) Koinuma, M.; Ogata, C.; Kamei, Y.; Hatakeyama, K.; Tateishi, H.; Watanabe, Y.; Taniguchi, T.; Gezuhara, K.; Hayami, S.; Funatsu, A.; et al. Photochemical Engineering of Graphene Oxide Nanosheets. J. Phys. Chem. C 2012, 116, 19822−19827. (18) Shul’ga, Yu. M.; Vasilets, V. N.; Baskakov, S. A.; Muradyan, V. E.; Skryleva, E. A.; Parkhomenko, Yu. N. Photoreduction of Graphite Oxide Nanosheets with Vacuum Ultraviolet Radiation. High Energy Chem. 2012, 46, 117−121. (19) Matsumoto, Y.; Koinuma, M.; Kim, S. Y.; Watanabe, Y.; Taniguchi, T.; Hatakeyama, K.; Tateishi, H.; Ida, S. Simple Photoreduction of Graphene Oxide Nanosheet under Mild Conditions. ACS Appl. Mater. Interfaces 2010, 2, 3461−3466. (20) Hummers, W. S.; Offeman, R. E. Preparation of Graphitic Oxide. J. Am. Chem. Soc. 1958, 80, 1339−1339. (21) Lee, K. E.; Kim, J. E.; Maiti, U. N.; Lim, J.; Hwang, J. O.; Shim, J.; Oh, J. J.; Yun, T.; Kim, S. O. Liquid Crystal Size Selection of LargeSize Graphene Oxide for Size-Dependent N-Doping and Oxygen Reduction Catalysis. ACS Nano 2014, 8 (9), 9073−9080. (22) Zheng, Q.; Ip, W. H.; Lin, X.; Yousefi, N.; Yeung, K. K.; Li, Z.; Kim, J.-K. Transparent Conductive Films Consisting of Ultralarge Graphene Sheets Produced by Langmuir−Blodgett Assembly. ACS Nano 2011, 5, 6039−6051. (23) Su, C.-Y.; Xu, Y.; Zhang, W.; Zhao, J.; Tang, X.; Tsai, C.-H.; Li, L.-J. Electrical and Spectroscopic Characterizations of Ultra-Large Reduced Graphene Oxide Monolayers. Chem. Mater. 2009, 21, 5674− 5680. (24) Acik, M.; Lee, G.; Mattevi, C.; Chhowalla, M.; Cho, K.; Chabal, Y. J. Unusual Infrared-Absorption Mechanism in Thermally Reduced Graphene Oxide. Nat. Mater. 2010, 9, 840−845. (25) Bagri, A.; Mattevi, C.; Acik, M.; Chabal, Y. J.; Chhowalla, M.; Shenoy, V. B. Structural Evolution during the Reduction of Chemically Derived Graphene Oxide. Nat. Chem. 2010, 2, 581−587. (26) Kumar, P. V.; Bardhan, N. M.; Tongay, S.; Wu, J.; Belcher, A. M.; Grossman, J. C. Scalable Enhancement of Graphene Oxide Properties by Thermally Driven Phase Transformation. Nat. Chem. 2013, 6, 151−158. (27) Eda, G.; Lin, Y.-Y.; Mattevi, C.; Yamaguchi, H.; Chen, H.-A.; Chen, I.-S.; Chen, C.-W.; Chhowalla, M. Blue Photoluminescence from Chemically Derived Graphene Oxide. Adv. Mater. 2010, 22, 505−509. (28) Yan, J.-A.; Xian, L.; Chou, M. Y. Structural and Electronic Properties of Oxidized Graphene. Phys. Rev. Lett. 2009, 103, 086802. (29) Li, D.; Müller, M. B.; Gilje, S.; Kaner, R. B.; Wallace, G. G. Processable Aqueous Dispersions of Graphene Nanosheets. Nat. Nanotechnol. 2008, 3, 101−105. (30) Acik, M.; Lee, G.; Mattevi, C.; Pirkle, A.; Wallace, R. M.; Chhowalla, M.; Cho, K.; Chabal, Y. The Role of Oxygen during Thermal Reduction of Graphene Oxide Studied by Infrared Absorption Spectroscopy. J. Phys. Chem. C 2011, 115, 19761−19781. (31) Lee, D. W.; De Los Santos, L. V.; Seo, J. W.; Felix, L. L.; Bustamante, A. D.; Cole, J. M.; Barnes, C. H. W. The Structure of Graphite Oxide: Investigation of its Surface Chemical Groups. J. Phys. Chem. B 2010, 114, 5723−5728. (32) Blume, R.; Kidambi, P. R.; Bayer, B. C.; Weatherup, R. S.; Wang, Z.-J.; Weinberg, G.; Willinger, M.-G.; Greiner, M.; Hofmann, S.; Knop-Gericke, A.; et al. The Influence of Intercalated Oxygen on the Properties of Graphene on Polycrystalline Cu under Various Environmental Conditions. Phys. Chem. Chem. Phys. 2014, 16, 25989−26003. 28269

DOI: 10.1021/acs.jpcc.6b08758 J. Phys. Chem. C 2016, 120, 28261−28269