Chinese Chemical Letters  2018, Vol. 29 Issue (5): 711-715   PDF    
Tailoring thermal conductivity of bulk graphene oxide by tuning the oxidation degree
Qing-Long Menga, Hengchang Liua,b, Zhiwei Huanga,c, Shuang Konga,c, Peng Jianga, Xinhe Baoa    
a State Key Laboratory of Catalysis, CAS Center for Excellence in Nanoscience, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian 116023, China;
b School of Physical Science and Technology, ShanghaiTech University, Shanghai 200031, China;
c University of Chinese Academy of Sciences, Beijing 100039, China
Abstract: Bulk graphene oxide (GO) shows great potential in a variety of applications, such as sensors, photodetectors, supercapacitors, lithium ion batteries and catalysts. However, its thermal conductivity, one of the most important and fundamental physical properties, is still less known. Herein, we have systematically investigated the thermal conductivity of bulk GOs and find that it can be tailored by tuning their oxidation degree during preparation process. Notably, the cross-plane thermal conductivity of bulk GO, in comparison with its precursor graphite, exhibits more than 100 times decrease at room temperature. The dependence of thermal conductivity of GO on oxidation degree is attributed to the chemical and structural changes by introducing oxygen atoms and oxygen-containing functional groups, which can lead to a significant enhancement in atomic-and nano-scale phonon scattering. Furthermore, we reveal that the thermal conductivity of bulk GOs exhibits evident anisotropic behavior. These results provide fundamental understanding and valuable information on thermal transport properties of bulk GOs for various practical applications.
Key words: Graphene oxide     Thermal conductivity     Scattering center     Anisotropy     Oxidation degree    

Heat management is a crucial issue in modern electronic industry, which requires thermal conductivity as high as possible to maintain system's long-life service and reliability [1, 2]. On the other hand, thermal insulation for energy-efficient buildings [3, 4] and thermoelectrics [5-7] needs thermal conductivity to be as low as possible to achieve high performance. Therefore, the development and search for materials with tunable thermal conductivity is of great significance for multiple practical applications such as electronics, energy-efficient buildings and thermoelectrics.

Graphene oxide (GO) has attracted tremendous interest due to its unique properties. Its tunable properties like hydrophobicity/ hydrophilicity [8, 9], electrical conductivity [10-12] as well as band gap [13, 14] can be readily achieved by tuning the ratio of sp2 to sp3 bonded carbons, which makes it promising for a variety of applications such as sensors [15], photodetectors [16], supercapacitors [17], lithium ion batteries [18, 19] and catalysts [20, 21]. The performance and reliability of these applications often depend greatly on the thermal conductivity of GOs, and therefore the fundamental understanding of thermal transport properties of GOs is of great importance.

For thermal transport properties of carbon-based materials, previous studies mainly focus on the thermal transport of carbon nanotubes and graphene thin films and/or nanosheets [22-28]. Notably, two-dimensional individual graphene nanosheet shows extremely high thermal conductivity [24, 25], which is even higher than that of carbon nanotubes. The flexible graphene-carbon fiber composite paper shows very high in-plane thermal conductivity and tensile strength, which makes it a promising candidate as heat spreader for next generation portable electronics [29, 30]. This outstanding thermal conductivity of graphene is beneficial for electronic applications in thermal management. Interestingly, the thermal conductivity of graphene can be tuned by isotopically doping 13C into 2D nanosheets [23], which provide valuable information to understand the phonon transport mechanisms of graphene. Regarding the thermal transport properties of GOs, only a few theoretical and experimental studies have investigated the thermal conductivity of GO nanosheets and thin films [31-34]. However, these studies mainly focus on the thermal conductivity of 2D GO nanosheets and thin films, and studies regarding the thermal conductivity of bulk GOs are still lacking, even they show great potential in a wide range of applications in sensors [15], photodetectors [16], supercapacitors [17], lithium ion batteries [18, 19] and catalysts [20, 21].

Herein, we focus on the thermal conductivity of bulk GOs. We successfully tailored the thermal conductivity by tuning the oxidation degree during preparation process. Our results suggest that the thermal conductivity of bulk GOs decreases dramatically with an increase in oxidation degree. Especially, the cross-plane thermal conductivity of GO with high oxidation degree, in comparison with its precursor graphite, shows more than 100 times decrease at room temperature. Such a significant decrease in thermal conductivity is due to the chemical and structural changes by oxidation, which lead to a significant enhancement in atomicand nano-scale phonon scattering in the thermal transport process. The ultralow thermal conductivity of bulk GO with high oxidation degree promises its applications in energy-efficient buildings and thermoelectrics. Furthermore, the basic understanding of thermal transport properties of bulk GOs provides valuable information for their various practical applications.

In this work, GOs were prepared from graphite flakes based on the Hummers method [35] without using NaNO3, followed by ultrasonication. The oxidation degree was controlled by tuning the amount of potassium permanganate (KMnO4). GOs with low and high oxidation degree, corresponding to 2 g and 4 g KMnO4 used for preparation, are denoted as GOL and GOH, respectively. The preparation details are given in our previous work [36]. The bulk GOs were fabricated with the prepared GO powders by a hand press machine. The prepared GO powders were loaded into a die with a diameter of 12.7 mm and pressed under 70.50 kN for 5 min.

X-ray photoelectron spectroscopy (XPS) results, which were obtained on a Thermo Scientific spectroscope equipped with an Escalab250xi X-ray source, are shown in Fig. S1 in Supporting information. Table 1 lists the amount of each functional group and oxygen content (O/C) of graphite and GOs derived from XPS spectra. It is evident that the oxygen-containing functional groups and oxygen content (O/C) increase with increasing the amount of KMnO4, indicating that the oxidation degree of GO is tunable by changing the KMnO4 amount.

Table 1
Relative amount of each functional group and O/C ratio of graphite and GOs

The microstructure of GOs was observed by FEI QUANTA 200 FEG scanning electron microscope (SEM) at 20 KV. Due to the low electrical conductivity of GOH, gold sputtering was used to obtain clear SEM images. Fig. 1 shows the top-view and cross-section SEM images of bulk graphite and GOs, which reveals a layered microstructure of graphite and GOs. This suggests evident anisotropy in structure. It is expected that the structural anisotropy of graphite and GOs might lead to the anisotropic properties in electrical and thermal transport.

Download:
Fig. 1. (a) Top-view image of graphite. (b) Cross-section image of graphite. (c) Topview image of GOL. (d) Cross-section image of GOL. (e) Top-view image of GOH. (f) Cross-section image of OH.

X-ray diffraction (XRD) patterns were collected by a PANalytical EMPYREAN system with Cu radiation (λ = 1.542 Å). The measurement was conducted at a 2θ range of 5°-40° at room temperature with a scanning rate of 13.5 degree/min. As can be seen in Fig. 2, graphite and GOs show a characteristic (002) peak, corresponding to a d(002) spacing of 0.34 nm between sp2 carbon layers. It is evident that GOs present a broader (002) peak width and lower peak intensity than graphite (Fig. 2). In addition, the intensity of (002) peak decreases significantly with oxidation degree. These changes reveal a lower crystal quality of GOs than graphite due to lattice distortion by oxidation. On the other hand, the formation of oxygen-containing functional groups of GOL and GOH leads to the appearance of (001) peak, thus the simultaneous appearance of (001) and (002) peaks of GOs is triggered by the heterogeneous structure of GOs, consisting of oxidized sp3 domains and unoxidized sp2 domains [37]. Obviously, an evident shift of (001) peak to lower 2θ degrees with oxidation degree can be seen in Fig. 2, corresponding to an increase in d(001) spacing. Moreover, the (001) peak intensity of GOs also increases with oxidation degree, but the (002) peak intensity shows an opposite trend. These are indicative of an increase in the amount of oxygencontaining functional groups and more sp3 domains [37], as suggested by XPS results. It is generally acknowledged that the epoxy and hydroxyl functional groups are located on the basal planes while carbonyl and carboxyl groups are at the edges [38, 39], thus we can infer that the intercalation of epoxy and hydroxyl functional groups into interlayers is responsible for the increase in d(001) spacing.

Download:
Fig. 2. XRD patterns of graphite, GOL and GOH.

Raman spectroscopy was performed on a LABRAM HR 800 apparatus using 532.08 nm laser as the exciting source. For each sample, five different spots were selected to obtain Raman spectra. Fig. 3a presents the Raman spectra for graphite, GOL and GOH. The characteristic peaks near 1350 cm-1, 1580 cm-1, 2700 cm-1 and 2900 cm-1 correspond to D, G, 2D and D + D' bands, respectively [40]. The prominent features of Raman spectra of graphite and GOs are the difference in the intensity of D band and G band. The G band indicates the existence of sp2 carbon networks, while the D band is due to the structural defects like grain boundaries or the attachment of hetero-atoms [33, 41]. Our XPS results have confirmed the formation of large amount of oxygen-containing functional groups for as-prepared GOs, which is also consistent with the observation by scanning tunnelling microscopy and highresolution transmission electron microscopy [42-44]. The amount of domain boundaries is related to the average sp2 cluster size, which can be obtained byestimating the distance between defects, LD (nm) [40, 45], using the following equation [46],

Download:
Fig. 3. Raman spectra (a), LD (b), the G peak position and FWHM (c) of graphite and GOs.

(1)

where EL is the excitation laser energy; ID and IG are the D band intensity and G band intensity of Raman spectra, respectively. The LD shown in Fig. 3b decreases with oxidation degree, indicative of a drop in sp2 cluster size and an increase of sp2/sp3 domain boundaries. Furthermore, the peak position and HWFM of G peak are used to qualitatively evaluate the sp3 carbon fraction [37]. As show in Fig. 3c, the G peak position and HWFM of GOs increase significantly as compared to those of graphite, corresponding to an evident increase in sp3 fraction, in line with XPS and XRD results. Overall, the Raman spectra of graphite and GOs results indicate a transition from crystalline graphite to an amorphous state with an increase in oxidation degree.

On the basis of XPS, XRD and Raman analysis, we can conclude that significant chemical and structural changes occur after oxidation with prominent features as follows: 1) the increase in the amount of oxygen atoms and oxygen-containing functional groups; 2) the increase in interlayer distance due to intercalation of oxygen-containing functional groups; 3) the drop in sp2 cluster size and growth in sp2/sp3 domain boundaries.

Next, we investigated the thermal conductivities by measuring the thermal diffusivity (D) and specific heat capacity (Cp) using a commercial Netzch LFA-457 instrument. The specific heat capacity (Cp) can be given as Eq. (2)

(2)

where Q, ΔT and m are the absorbed energy for temperature rise, the temperature rise after energy absorption and the mass, respectively. Based on equation (2), the ratio of Cpsam/Cpstd can be given as Eq. (3)

(3)

where the lower cases of "sam" and "std" represent the sample and the standard sample (Pyroceram 9606), respectively. By assuming that Qsam is equal to Qstd, we can getthe specific heat capacity of the sample as Eq. (4)

(4)

To confirm the validity of the calculated specific heat capacity by Eq. (4), we compared the specific heat capacity of graphite obtained from LFA measurement with standard values of graphite and the uncertainty is within ±5% (Fig. S2 in Supporting information). All the measurements for each sample were repeated twice, and for each repeated measurement, four data points were collected at each temperature (Fig. S3 in Supporting information).

Fig. 4 presents the cross-plane and in-plane thermal conductivities of bulk graphite and GOs as a function of temperature. Both the cross-plane and in-plane thermal conductivities of graphite decrease with temperature (Figs. 4a and b), indicating that the thermal conductivity is limited by Umklapp phonon scattering [33, 47]. The thermal conductivity of bulk graphite exhibits evident anisotropic behavior and the in-plane thermal conductivity is approximately five times higher than the cross-plane thermal conductivity, which is attributed to the structural anisotropy as suggested by SEM. The in-plane thermal conductivity of graphite in this work (~57 W m-1 K-1) is lower than highly oriented pyrolytic graphite (HOPG) (~2000 W m-1 K-1) [47], which suggests that graphite used in this work has a lower quality than HOPG [47].

Download:
Fig. 4. Cross-planethermal conductivity(K) of graphite (a), GOL (c) and GOH (e) as a function of temperature. In-plane thermal conductivityof graphite (b), GOL (d) and GOH (f) as a function of temperature.

Both the cross-plane and in-plane thermal conductivities of GOs decrease dramatically with increasing oxidation degree (Figs. 4c-f). Especially, the cross-plane thermal conductivity of GOH shows more than 100 times decrease compared to its precursor graphite at room temperature. The cross-plane thermal conductivity of GOs shows a more significant decrease than inplane thermal conductivity. In contrast with graphite, both the cross-plane and in-plane thermal conductivities of GOs monotonically increase with temperature, suggesting that the hopping of localized excitations dominates the heat-conduction mechanism [47, 48]. This conclusion is corroborated by XPS, XRD and Raman data, which reveals that GOs are disordered and amorphous solid materials with a structure of isolated sp2 (graphitic) clusters embedding within sp3 (amorphous) matrix.

It is well-known that heat conduction is dominated by phonons in graphite, graphene, carbon nanotubes and their derivatives [23, 28, 33, 47, 49, 50], unlike metals dominated by electrons/holes [47]. Moreover, the estimated electronic thermal conductivities of graphite, GOL and GOH (Section S2 in Supporting information) are negligible compared to their total thermal conductivities. Therefore, the heat conduction is dominated by phonons in graphite and GOs. For graphene and GOs, both molecular dynamics simulations and experimental results suggest that the chemical (oxygen atoms and oxygen-containing functional groups) and structural (interface and grain/cluster/domain boundary) changes by oxidation could enhance phonon scattering and reduce thermal conductivity [31-33]. Herein, the reduced GOH after oxygen removal (annealing at 1500 ℃), as characterized by Raman (Fig. S4 in Supporting information), shows about 25 times higher thermal conductivity than GOH, but about 3.5 times lower thermal conductivity than graphite at room temperature (Fig. S5 in Supporting information). These results suggest that the significant reduction in thermal conductivity of GOs is due to the enhancement in phonon scattering by both the chemical and structural changes, and the chemical changes seem to play a major role due to the evident increase in thermal conductivity of reduced GOH, which is in good line with previous molecular dynamics simulation results [31, 32].

Fig. 5 shows the cross-plane and in-plane thermal conductivities of graphite and GOs at room temperature as a function of oxygen content, interlayer distance and 1/LD, respectively. An almost linear dependence of the logarithm of thermal conductivity on oxygen content, interlayer distance and 1/LD is observed, thus it can be believed that the significant decrease in thermal conductivity is triggered by these chemical and structural changes.

Download:
Fig. 5. Cross-plane (K) and in-plane (K//) thermal conductivities of graphite and GOs at room temperature versus O/C ratio (a), versus d spacing (b) and versus 1/LD (c). Here, solid circles and open circles denote the cross-plane and in-plane thermal conductivities, respectively

According to the kinetic theory of gases, the phonon thermal conductivity (Kp) is quantitatively defined as [26, 47]

(5)

where Cp is the specific heat capacity, υ is the phonon group velocity, and Λ is the phonon mean free path (MFP) standing for the length between scattering events. The difference in specific heat capacity (Cp) between graphite and GOs is almost negligible (Fig. S6 in Supporting information) compared with the difference in thermal conductivity, which cannot be responsible for the significant decrease in thermal conductivity. Additionally, the phonon group velocity is usually approximated by the sound velocity [47], and there is no significant change in phonon group velocity with oxidation degree as suggested by molecular dynamic simulation [31, 32]. According to the relaxation-time approximation, the phonon MFP can be quantitatively defined as Λ=τυ, where τ is the phonon relaxation time [47]. The reciprocal of τ (1/τ), namely the phonon scattering rate, which limits the phonon MFP, is used to characterize scattering mechanisms, being expressed as [47]

(6)

where i is the scattering processes, including phonon-phonon scattering, phonon-defect scattering, phonon-impurity scattering, phonon-interface scattering, phonon-boundary scattering and so on. When the phonon MFP is comparable to the dimensions of scattering centers, such as impurity atoms, interfaces and domain boundaries, the scattering of phonons takes place [49]. Consequently, these scatterings lead to a decrease in thermal conductivity. In typical solid materials, the distribution of MFP covers a wide range from atomic-to nano-length scales [51].

Herein, the chemical and structural changes of GOs by oxidation provide new scattering centers ranging from atomic-to nanolength scales. Hence, these changes are believed to be responsible for the change in thermal transport properties. First, the introduction of atomic-scale oxygen atoms and oxygen-containing functional groups enhances the atomic-scale scattering for phonons with short wavelength (phonon-impurity scattering, Fig. 6), closely coinciding with a significant decrease in thermal conductivity of graphene only by isotopically doping [23, 52] or by hydrogen functionalization [53, 54]. Second, the introduction of oxygen atoms makes interlayer distance increase from 0.34 nm to 0.72 nm, which could enhance the nano-scale interface scattering for phonons with medium wavelength (phonon-interface scattering, Fig. 6). Such a decrease in thermal conductivity by phononinterface scattering is analogous to the multilayer graphene on an amorphous support [55] because adjacent layers of GOs can be considered as interfaces due to the non-uniform distribution of sp2/sp3 domains in c-axis direction. Third, the increase in oxygen atoms creates more sp2/sp3 domain boundaries, which may enhance the nano-scale domain-boundary scattering for phonons with long wavelength (phonon-boundary scattering, Fig. 6). Overall, the introduction of oxygen atoms and oxygen-containing functional groups into GOs can lead to the formation of all lengthscale scattering centers to effectively scatter phonons with a wide range of MFP. Therefore, the thermal conductivity of GOs decreases dramatically with an increase in oxidation degree.

Download:
Fig. 6. Illustration of scattering events of graphene oxide

In summary, we have systematically investigated the tunable thermal conductivity of bulk GOs. We found that the cross-plane and in-plane thermal conductivities of bulk GOs could be tuned readily by manipulating oxidation degree and decrease dramatically with an increase in oxidation degree. Especially, the crossplane thermal conductivity of GOH shows more than 100 times decrease at room temperature compared to its precursor graphite. Furthermore, both graphite and GOs exhibit a strong anisotropy in thermal conductivity due to the structural anisotropy. The increase in oxidation degree of GOs leads to the following changes: 1) the increase in the amount of oxygen atoms and oxygen-containing functional groups; 2) the increase in interlayer distance; 3) the drop in sp2 cluster size and growth in sp2/sp3 domain boundaries. Consequently, these changes lead to effective atomic-and nanoscale scattering for phonons with short, medium and long wavelength, thus the thermal conductivities of bulk GOs decrease dramatically with increasing oxidation degree.

Acknowledgments

This work is financially supported by the National Natural Science Foundation of China (Nos. 21273228 and 51290272) and 100 Talents Program of Chinese Academy of Sciences.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.cclet.2017.10.028.

References
[1]
Y. Zhang, H.X. Han, N. Wang, et al., Adv. Funct. Mater. 25 (2015) 4430-4435. DOI:10.1002/adfm.v25.28
[2]
Z. Yan, G.X. Liu, J.M. Khan, A.A. Balandin, Nat. Commun. 3 (2012) 827-834. DOI:10.1038/ncomms1828
[3]
B. Wicklein, A. Kocjan, G. Salazar-Alvarez, et al., Nat. Nanotechnol. 10 (2015) 277-283. DOI:10.1038/nnano.2014.248
[4]
A. Covaci, A.C. Gerecke, R.J. Law, et al., Environ. Sci. Technol. 40 (2006) 3679-3688. DOI:10.1021/es0602492
[5]
F. Bonaccorso, L. Colombo, G.H. Yu, et al., Science 347 (2015) 1246501. DOI:10.1126/science.1246501
[6]
Y.K. Koh, C.J. Vineis, S.D. Calawa, M.P. Walsh, D.G. Cahill, Appl. Phys. Lett. 94 (2009) 153101. DOI:10.1063/1.3117228
[7]
G.J. Snyder, E.S. Toberer, Nat. Mater. 7 (2008) 105-114. DOI:10.1038/nmat2090
[8]
X.B. Hu, Y. Yu, W.M. Hou, J.E. Zhou, L.X. Song, Appl. Surf. Sci. 273 (2013) 118-121. DOI:10.1016/j.apsusc.2013.01.201
[9]
B.M. Ganesh, A.M. Isloor, A.F. Ismail, Desalination 313 (2013) 199-207. DOI:10.1016/j.desal.2012.11.037
[10]
K. Hatakeyama, H. Tateishi, T. Taniguchi, et al., Chem. Mater. 26 (2014) 5598-5604. DOI:10.1021/cm502098e
[11]
I. Jung, D.A. Dikin, R.D. Piner, R.S. Ruoff, Nano Lett. 8 (2008) 4283-4287. DOI:10.1021/nl8019938
[12]
C.M. Chen, J.Q. Huang, Q. Zhang, et al., Carbon 50 (2012) 659-667. DOI:10.1016/j.carbon.2011.09.022
[13]
G. Eda, Y.Y. Lin, C. Mattevi, et al., Adv. Mater. 22 (2010) 505-509. DOI:10.1002/adma.v22:4
[14]
G. Eda, C. Mattevi, H. Yamaguchi, H. Kim, M. Chhowalla, J. Phys. Chem. C 113 (2009) 15768-15771. DOI:10.1021/jp9051402
[15]
Y.J. Yun, W.G. Hong, N.J. Choi, et al., Nanoscale 6 (2014) 6511-6514. DOI:10.1039/C4NR00332B
[16]
Y. Ito, W.F. Zhang, J.H. Li, et al., Adv. Funct. Mater. 26 (2016) 1271-1277. DOI:10.1002/adfm.v26.8
[17]
X.C. Dong, H. Xu, X.W. Wang, et al., ACS Nano 6 (2012) 3206-3213. DOI:10.1021/nn300097q
[18]
G.J. Hu, C. Xu, Z.H. Sun, et al., Adv. Mater. 28 (2016) 1603-1609. DOI:10.1002/adma.201504765
[19]
R. Rajagopalan, L. Zhang, S.X. Dou, H.K. Liu, Adv. Energy Mater. 6 (2016) 1501760. DOI:10.1002/aenm.201501760
[20]
W. Yan, F. He, S.L. Gai, et al., J. Mater. Chem. A 2 (2014) 3605-3612. DOI:10.1039/C3TA14718E
[21]
Y. Wang, S.L. Gai, N. Niu, F. He, P.P. Yang, J. Mater. Chem. A 1 (2013) 9083-9091. DOI:10.1039/c3ta11161j
[22]
H. Malekpour, K.H. Chang, J.C. Chen, et al., Nano Lett. 14 (2014) 5155-5161. DOI:10.1021/nl501996v
[23]
S.S. Chen, Q.Z. Wu, C. Mishra, et al., Nat. Mater. 11 (2012) 203-207. DOI:10.1038/nmat3207
[24]
J.H. Seol, I. Jo, A.L. Moore, et al., Science 328 (2010) 213-216. DOI:10.1126/science.1184014
[25]
A.A. Balandin, S. Ghosh, W.Z. Bao, I. Calizo, Nano Lett. 8 (2008) 902-907. DOI:10.1021/nl0731872
[26]
S. Berber, Y.K. Kwon, D. Tomanek, Phys. Rev. Lett. 84 (2000) 4613-4616. DOI:10.1103/PhysRevLett.84.4613
[27]
J.W. Che, T. Cagin, W.A. Goddard, Nanotechnology 11 (2000) 65-69. DOI:10.1088/0957-4484/11/2/305
[28]
J. Hone, M. Whitney, C. Piskoti, A. Zettl, Phys. Rev. B 59 (1999) R2514-R2516. DOI:10.1103/PhysRevB.59.R2514
[29]
Q.Q. Kong, Z. Liu, J.G. Gao, et al., Adv. Funct. Mater. 24 (2014) 4222-4228. DOI:10.1002/adfm.v24.27
[30]
N.J. Song, C.M. Chen, C.X. Lu, et al., J. Mater. Chem. A 2 (2014) 16563-16568. DOI:10.1039/C4TA02693D
[31]
X. Mu, X.F. Wu, T. Zhang, D.B. Go, T.F. Luo, Sci. Rep. 4 (2014) 3909-3917.
[32]
H.J. Zhang, A.F. Fonseca, K. Cho, J. Phys. Chem. C 118 (2014) 1436-1442.
[33]
J.D. Renteria, S. Ramirez, H. Malekpour, et al., Adv. Funct. Mater. 25 (2015) 4664-4672. DOI:10.1002/adfm.201501429
[34]
T. Schwamb, B.R. Burg, N.C. Schirmer, D. Poulikakos, Nanotechnology 20 (2009) 405704-405708. DOI:10.1088/0957-4484/20/40/405704
[35]
W.S. Hummers, R.E. Offeman, J. Am. Chem. Soc 80 (1958) 1339-1339. DOI:10.1021/ja01539a017
[36]
Q.L. Meng, H.C. Liu, Z. Huang, et al., Carbon 101 (2016) 338-344. DOI:10.1016/j.carbon.2016.01.087
[37]
K. Krishnamoorthy, M. Veerapandian, K. Yun, S.J. Kim, Carbon 53 (2013) 38-49. DOI:10.1016/j.carbon.2012.10.013
[38]
D.R. Dreyer, S. Park, C.W. Bielawski, R.S. Ruoff, Chem. Soc. Rev. 39 (2010) 228-240. DOI:10.1039/B917103G
[39]
D.A. Dikin, S. Stankovich, E.J. Zimney, et al., Nature 448 (2007) 457-460. DOI:10.1038/nature06016
[40]
A.C. Ferrari, D.M. Basko, Nat. Nanotechnol. 8 (2013) 235-246. DOI:10.1038/nnano.2013.46
[41]
M.A. Pimenta, G. Dresselhaus, M.S. Dresselhaus, et al., Phys. Chem. Chem. Phys. 9 (2007) 1276-1291. DOI:10.1039/B613962K
[42]
K.P. Loh, Q.L. Bao, G. Eda, M. Chhowalla, Nat. Chem. 2 (2010) 1015-1024. DOI:10.1038/nchem.907
[43]
J.I. Paredes, S. Villar-Rodil, P. Solis-Fernandez, A. Martinez-Alonso, J.M.D. Tascon, Langmuir 25 (2009) 5957-5968. DOI:10.1021/la804216z
[44]
N.R. Wilson, P.A. Pandey, R. Beanland, et al., ACS Nano 3 (2009) 2547-2556. DOI:10.1021/nn900694t
[45]
L.G. Cancado, A. Jorio, E.H.M. Ferreira, et al., Nano Lett. 11 (2011) 3190-3196. DOI:10.1021/nl201432g
[46]
P. Mallet-Ladeira, P. Puech, C. Toulouse, et al., Carbon 80 (2014) 629-639. DOI:10.1016/j.carbon.2014.09.006
[47]
A.A. Balandin, Nat. Mater. 10 (2011) 569-581. DOI:10.1038/nmat3064
[48]
D.G. Cahill, R.O. Pohl, Solid State Commun. 70 (1989) 927-930. DOI:10.1016/0038-1098(89)90630-3
[49]
D.L. Nika, E.P. Pokatilov, A.S. Askerov, A.A. Balandin, Phys. Rev. B 79 (2009) 155413. DOI:10.1103/PhysRevB.79.155413
[50]
D.L. Nika, A.A. Balandin, J. Phys. Condens. Matter 24 (2012) 233203. DOI:10.1088/0953-8984/24/23/233203
[51]
L.D. Zhao, V.P. Dravid, M.G. Kanatzidis, Energy Environ. Sci. 7 (2014) 251-268. DOI:10.1039/C3EE43099E
[52]
Q.X. Pei, Y.W. Zhang, Z.D. Sha, V.B. Shenoy, Appl. Phys. Lett. 100 (2012) 101901. DOI:10.1063/1.3692173
[53]
J.Y. Kim, J.H. Lee, J.C. Grossman, ACS Nano 6 (2012) 9050-9057. DOI:10.1021/nn3031595
[54]
Q.X. Pei, Z.D. Sha, Y.W. Zhang, Carbon 49 (2011) 4752-4759. DOI:10.1016/j.carbon.2011.06.083
[55]
M.M. Sadeghi, I. Jo, L. Shi, Proc. Natl. Acad. Sci. U. S. A. 110 (2013) 16321-16326. DOI:10.1073/pnas.1306175110