Three-dimensional graphene nanosheet films towards high performance solid lubricants

Three-dimensional graphene nanosheet films towards high performance solid lubricants

Applied Surface Science 467–468 (2019) 30–36 Contents lists available at ScienceDirect Applied Surface Science journal homepage: www.elsevier.com/lo...

2MB Sizes 0 Downloads 11 Views

Applied Surface Science 467–468 (2019) 30–36

Contents lists available at ScienceDirect

Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc

Full Length Article

Three-dimensional graphene nanosheet films towards high performance solid lubricants ⁎

F.X. Chena, Y.J. Maia, , Q.N. Xiaoa, G.F. Caib, L.Y. Zhanga, C.S. Liua, X.H. Jiea, a b

T



School of Materials and Energy, Guangdong University of Technology, Guangzhou 510006, China School of Materials Science and Engineering, Nanyang Technological University, 639798 Singapore, Singapore

A R T I C LE I N FO

A B S T R A C T

Keywords: Graphene Friction and wear Transfer layers

Graphene nanosheet films (GNSF) show great potential as solid lubricants. They, however, usually suffer from short lifetime and low load capability, limiting their further engineering application. In this work, we try to address these crucial challenges, for the first time, by inducing a unique three-dimensional microstructure into the GNSF. The as-prepared three-dimensional graphene nanosheet films (3D-GNSF) are composed of many crumpled and partially standing graphene nanosheets (GNS) that overlap with each other and form a threedimensional interconnected network. The morphology, composition and structure of the 3D-GNSF are investigated in detail using scanning electron microscope, X-ray diffractometer, transmission electron microscope and X-ray photoelectron spectroscopy to deduce their possible formation mechanism. Tribological tests are conducted in air as a function of contact pressure that is up to about 1GPa. The results suggest it is the unique architecture of the 3D-GNSF that quickly produces a compact and stable sliding interface consisting of GNS against GNS, enabling 3D-GNSF as promising solid lubricants with low friction, excellent anti-wear ability, good durability and high load capability.

1. Introduction Minimizing friction and wear is essential to improve the efficiency, reliability and durability of moving mechanical systems. To approach this goal, the search for new materials, coatings, and lubricants continues around the globe [1–5]. Due to the unique advantages, such as high chemical inertness, extreme strength, low shear strength and atomically smooth surface, graphene has attracted great attention for macro-scaled tribological application [6–8]. For example, graphene nanosheet films (GNSF), containing a single layer, a few layers, or multiple layers of two-dimensional graphene nanosheets (GNS), were prepared by various methods, such as electrophoretic deposition [9], self-assembly technique based on Marangoni effect [10], spray [11,12] and chemical vapor deposition (CVD) [13] and tested as solid lubricants. The results suggested GNSF were very effective for friction and wear reduction, demonstrating the feasibility of using GNSF as solid lubricants. These GNSF, however, usually suffered from short lifetime and low load capability, limiting their further engineering application. This is mainly caused by structural degradation of GNS induced by sliding and excessive removal of GNS out of the contact interface as a consequent of their weak adhesion to the substrate [14–16]. Some measures have



been explored to overcome these problems. For instance, periodic supply of GNS suspension was used to ensure the presence of GNS in the wear track throughout the test, which reduced the coefficient of friction (COF) from 0.2 to around 0.15 [12]. The CVD grown time was extended from 5 to 20 min. to increase the uniformity and quality of GNSF, which extended their durability from 2000 to 6000 cycles [15]. Hydrogen was introduced during sliding test to passivate the dangling bonds in a ruptured graphene, which significantly improved the stability and lifetime of few-layer graphene [17]. Coupling agent of (3-Aminopropyl) triethoxysilane was used to improve the bonding between graphene oxide nanosheets (GONS) and stainless steel, which resulted in 20 times enhancement in durability for the GONS film [16]. Despite these achievements, it is still a challenge to obtain good GNSF solid lubricant with low friction, excellent anti-wear ability, good durability and high load capability. Recently, the assembly of two-dimensional GNS into three-dimensional structures has attracted intensive interest, because the utilization of three-dimensional graphene materials is considered as one of the most effective ways to apply the unique properties of two-dimensional GNS in practical applications. Due to their unique mechanical characteristics, excellent electrical conductivity and large surface area, three-dimensional graphene materials and their composites have shown

Corresponding authors. E-mail addresses: [email protected] (Y.J. Mai), [email protected] (X.H. Jie).

https://doi.org/10.1016/j.apsusc.2018.10.125 Received 9 August 2018; Received in revised form 8 October 2018; Accepted 15 October 2018 Available online 16 October 2018 0169-4332/ © 2018 Elsevier B.V. All rights reserved.

Applied Surface Science 467–468 (2019) 30–36

F.X. Chen et al.

( )

Fig. 1. SEM images with low and high magnification of the surface morphology of brass-3D-GNSF (a & b) and brass-CC (c & d), and their corresponding XRD patterns (e).

2. Experimental

great potential in various applications including, batteries [18–20], capacitors [21,22], adsorbents [23], catalytic [24], and so on. However, there is no report on tribological application of three-dimensional graphene materials. Here, we report a one-step fabrication of threedimensional graphene nanosheet films (3D-GNSF) on the surface of a metal substrate and test them as solid lubricants. The as prepared 3DGNSF is composed of crumpled and partially standing GNS which overlap with each other and form a three-dimensional interconnected network. Those crumpled and partially standing GNS in 3D-GNSF provide ample active contact areas during sliding process, facilitating the rapid transfer of GNS to the counter-face. Moreover, the three-dimensional interconnected network in 3D-GNSF increases the adhesion of GNS against each other and substrate, suppressing the excessive removal of GNS out of the contact interface. As a consequence, the resulting 3D-GNSF yields excellent tribological performance as solid lubricants with low friction, excellent anti-wear ability, good durability and high load capability.

2.1. Preparation of 3D-GNSF on the surface of brass disk 3D-GNSF grown on the surface of brass disk were prepared by a onestep hydrothermal method. Typically, GONS synthesized according to the well-known Hummer’s method and ascorbic acid (40 mmol/L) were ultrasonically dispersed in deionized water, forming a uniform GONS suspension (0.1 mg/ml). Then, a brass disk (25 × 20 × 3 mm) was immersed in 40 ml of the above suspension in a 50-ml Teflon-lined stainless-steel autoclave. The sealed autoclave was heated at 180 °C for 6 h. After the autoclave was naturally cooled to room temperature, the treated brass disk was repeatedly rinsed with deionized water and dried. For convenience, this simple is referenced as brass-3D-GNSF. To investigate the effect of ascorbic acid on the morphology and composition of the obtained films, another brass disk was also treated with the same hydrothermal process with a suspension containing GNOS only (without ascorbic acid). For convenience, the obtained sample is noted as brass-CC. “CC” is short for composite coating. As demonstrated in the following section, a composite coating composing of ZnO and GNS is formed on the surface of brass substrate. As a control sample for the 31

Applied Surface Science 467–468 (2019) 30–36

F.X. Chen et al.

brass-3D-GNSF and the corresponding high-resolution XPS spectra of C1s, Cu 2p and Zn 2p, respectively. The wide survey spectrum (Fig. 3a) confirms the presence of carbon, oxygen, copper and zinc, and the carbon is the main componence (about 86 at. %). There are four components in C1s XPS spectrum (Fig. 3b) that correspond to the carbon atoms in different functional groups: the nonoxygenated ring C, the C in C–O bonds, the carbonyl C, and the carboxylate carbon [26,27]. However, the peak intensities of the oxygen functionalities are greatly reduced with respect to those of GONS precursor, indicating GONS are reduced to GNS after the hydrothermal process. The Cu 2P XPS spectrum (Fig. 3c) shows two strong peaks centered at 933.4 eV and 953.2 eV, respectively. This coupled with a broad shakeup satellite peak at 943.5 eV indicates the presence of CuO [27,28]. Also, the Zn 2P XPS spectrum (Fig. 3d) shows two strong peaks centered at 1022.6 eV and 1045.9 eV, respectively, corresponding to the Zn 2p 3/2 and Zn 2p 1/2 of ZnO, respectively, These results are consistent with the HRTEM data. Based on the above observation, the formation mechanism of 3DGNSF on the surface of brass disk is deduced as follow. It was reported that GONS could be synchronously reduced and assembled on the surface of active metals when they were immersed in the GONS suspension, for GONS could harvest electron from the active metal along with the oxidation of the active metal substrates [29–31]. Therefore, GNS decorated with CuO and ZnO nanocrystallines was formed and deposited on the surface of brass disk at the beginning of the hydrothermal process. As the hydrothermal process goes on, the increased temperature makes the GONS preferentially reduced by the ascorbic acid and supercritical water to GNS with clean surface. These GNS selfassemble to a stable 3D interconnected network owing to the enhanced attraction interaction between GNS as a consequence of more conjugated domains and more hydrophobic surface [32,33]. Fig. 4a shows the typical curves of COF vs. sliding duration of brassHT, brass-CC and brass-3D-GNSF under a normal load of 6 N. The COF of brass-HT shows a high value of about 0.53 and exhibits large fluctuations during the whole test, indicating the absence of lubricating interface. The COF of brass-CC continuously increases from a begin value of about 0.17 and reaches a value of about 0.26 at the end of the test. Though no steady state friction is achieved, the value of COF is obviously reduced compared to the brass-HT, highlighting the lubricating role of GNS in the ZnO/GNS composite coating. Similar lubricating phenomenon are reported in many GNS containing composites [34–36]. By contrast, the COF curve of brass-3D-GNSF is rather stable and the COF value is as low as 0.13 after a short running-in period (about 20 s), indicating attainment of steady state friction. What’s more, the COF of brass-3D-GNSF keeps lower than 0.2, the critical value of self-lubricating materials, after a long duration of 250 min. (the inset of Fig. 4a). These results highlight the superior lubricating capability and excellent durability of the as-prepared 3DGNSF. Fig. 4b shows the evolution of average COF of the tested samples as a function of normal loads ranging from 4 to 8 N. The average COF of brass-HT slightly decreases as the increase of the loads and exhibits the highest value among the three samples. By contrast, the average COF of brass-3D-GNSF exhibits the lowest COF value, and more importantly, it keeps stable even the load is increased to 8 N, corresponding to a maximum hertzian contact pressure of about 1 GPa. Under such a high contact pressure, its COF is as low as 0.13, which is about 41% and 75% lower than that of the brass-CC and brass-HT samples, respectively. This load capability is obviously better than our previously reported electrochemically reduced graphene oxide nanosheet coatings [11], and other graphene films or coatings, such as CVD graphene film [15], solution-processed graphene coatings [12], electrophoretic deposition GONS film [9] and 3-aminopropyltriethoxysilane associated self-assembled GNS coatings [16,37]. Fig. 4c shows the corresponding specific wear rate of the tested samples. For brass-HT, the wear rate exhibits an order of

tribology tests, another brass disk was treated with the same hydrothermal process with deionized water as hydrothermal solution. For convenience, the obtained sample is noted as brass-HT. 2.2. Tribological test Tribological tests were carried out on a rotating ball-on-disk tribometer (WTM-2E, China). The AISI 52100 bearing steel ball with a diameter of 6 mm was used as the counterpart. The applied loads were changed from 4 N to 8 N. The rotating speed was 200 rpm with a rotating radius of 5 mm (0.105 m/s). All the tests were carried out in air with a relative humidity of 55–70% at room temperature. The sliding time was set as 15 min. The COF was recorded continuously during the test. The specific wear rate was calculated by the wear formula of W = V/(L⋅D), where W, V, L and D were specific wear rate (mm3⋅N−1⋅m−1), wear volume (mm3), normal load (N) and sliding distance (m), respectively. 2.3. Characterizations The morphology, composition and microstructure of 3D-GNSF were investigated using scanning electron microscope (SEM, SU8010), X-ray diffractometer (XRD, Ultima-Ⅳ), transmission electron microscope (TEM, Tecnai G2 F20) and X-ray photoelectron spectroscopy (XPS, 250Xi). The wear tracks of tested samples and the transfer layer formed on the steel ball counterpart were analyzed using SEM, optical microscope (OM, DMi8C) and Raman microscopy (LabRAM Aramis). The wear volume of tested sample was determined by a 3D laser measuring microscope (OLS4000). 3. Results and discussion Fig. 1 compares the surface morphology of brass disk after hydrothermal treatment using GONS suspension with/without ascorbic acid. With the present of ascorbic acid, the whole surface of brass disk is covered with a uniform coating (Fig. 1a). Higher magnification image further reveals it is composed of many crumpled and partially standing GNS, which overlap with each other and form a 3D interconnected network (Fig. 1b). Obviously, it is different from the previous GNSF obtained via the spray coating method and the layer-by-layer deposition technique, for they were characterized with the well-packed layer structure [11,25]. On the other hands, the surface of brass-CC were covered with a composite coating, where aggregated GNS stack disorderly with rod-like ZnO (evidenced by the following XRD result) (Fig. 1c and d). Fig. 1e compares the XRD patterns obtained from the surface of brass-3D-GNSF and brass-CC. For the brass-3D-GNSF, only a broad peak centered at around 24°, which can be assigned to GNS, is observed except those diffraction peaks assigned to the brass substrate. For the brass-CC, a series of sharp peaks associated with the ZnO are additionally observed, which is consisted with the above SEM results. These observations suggest the ascorbic acid affects not only the morphology but also the composition of the obtained films, highlighting its essential role in inducing the 3D interconnect structure of the GNSF. TEM investigation further reveals 3D-GNSF are made of two kinds of GNS which own different surface morphology. One’s surface is decorated with many nanoparticles (Fig. 2a) and the other one is as clean as the precursor, GONS (Fig. 2c). HRTEM reveals that the lattice spacing of some decorated nanoparticles is about 0.25 nm and some of them is about 0.21 nm, corresponding to the (0 1 0) plane of ZnO and (2 0 0) plane of CuO, respectively (Fig. 2b). This result is consistent with the SAED (the inset of Fig. 2a), further confirming those decorated nanoparticles include CuO and ZnO nanocrystallines. Given that no diffraction peaks belonging to CuO or ZnO are observed in the XRD pattern of brass-3D-GNSF, it is reasonable to believe that their content is small. Fig. 3 shows the wide survey spectrum collected from the surface of 32

Applied Surface Science 467–468 (2019) 30–36

F.X. Chen et al.

Fig. 2. TEM (a) and HRTEM (b) images of GNS decorated with nanoparticles and TEM image of GNS decorated without nanoparticles (c). The inset of Fig. 2a is SAED collected from the pink circle area marked in Fig. 2a.

10−3 mm3 N−1 m−1. It is reduced to 10−5 mm3 N−1 m−1 for brass-CC and further reduced to 10−6 mm3 N−1 m−1 for brass-3D-GNSF. This trend almost follows the trend of average COF, indicating the reduced friction significantly contributes to the increased wear resistance of the brass-3D-GNSF. Similar effect of wear reduction is reported for various

(a)

substrates including steel, copper, silicon and titanium after they were covered with GNSF [9–12,37]. The above tribological tests suggest 3D-GNSF are excellent solid lubricants with low friction, excellent anti-wear ability, good durability and high load capability. To investigate the reasons to the improved

(b) 80.0k

C1s

500k

Intensity / a.u.

Intensity / a.u.

600k

400k Zn 2p Cu 2p

O1s

300k 200k 100k

60.0k 40.0k C-O 20.0k

C=O C(O)O

0.0

0 250

500

750

282

1000

286

288

290

(d)

21.0k

Cu 2p 3/2

Cu 2p

Intensity / a.u.

18.0k

te

te

lli

Cu

15.0k

pe

2+

ak

Cu 2p 1/2

Zn 2p

Zn 2p 3/2

24.0k

sa

Intensity / a.u.

284

Binding Energy / eV

Binding Energy / eV

(c)

C-C

C 1s

Zn 2p 1/2

21.0k

18.0k

12.0k 930

940

950

15.0k

960

1020

1030

1040

1050

Binding Energy / eV

Binding Energy / eV

Fig. 3. Wide survey spectrum (a) collected from the surface of brass-3D-GNSF and the corresponding high-resolution XPS spectra of C1s (b), Cu 2p (c) and Zn 2p (d). 33

Applied Surface Science 467–468 (2019) 30–36

F.X. Chen et al.

(a)

(b) 0.60

brass-HT

0.6

Average COF

0.45 COF

COF

brass-HT 0.30 0.25

long cycling @ 6N

0.20 0.15

0.30

0.10

0

60

120

180

Time / min.

240

brass-CC

0.4 brass-CC 0.2

brass-3D-GNSF

0.15

brass-3D-GNSF

0

3

6

9

Time / min. (c)

12

15

4

5

6

7

8

Load / N

0.01

Wear rate / mm3N-1m-1

0.0

Brass-HT Brass-CC Brass-3D-GNSF

1E-3 1E-4 1E-5 1E-6

4N

6N

8N

Fig. 4. Typical curves of COF vs. sliding duration (a) of brass-HT, brass-CC and brass-3D-GNSF sliding against steel ball under a normal load of 6 N (the inset is the long cycling performance of brass-3D-GNSF sample with a duration of 250 min.), the average COF (b) and the corresponding specific wear rate (c) as a function of normal loads ranging from 4 to 8 N.

in 3D-GNSF provide ample active contact areas, facilitating the quick formation of high quality transfer layers on the counter-face. Second, the 3D interconnected network in 3D-GNSF increases the adhesion of GNS against each other and substrate, suppressing the excessive removal of GNS out of the contact interface. As a result, a compact and stable sliding interface consisting of GNS against GNS would be formed as quickly as possible. It is this unique interface that prevents the direct metal-to-metal contact, provides easy shearing and decreases shear stress transmitted from the surface to the subsurface during sliding, enabling 3D-GNSF as promising solid lubricants with low friction, excellent anti-wear ability, good durability and high load capability.

tribological performance, the wear track of brass-3D-GNSF and the contact surface of steel ball counterpart after wear test under a load of 6 N are studied by various analysis technologies. Fig. 5a shows the Raman spectra obtained from two representative positions on the wear track (point 2 & point 3). Both of them show two broad bands appearing at ∼1330 and 1598 cm−1, which are the characteristics for the D and G bands of graphitic carbon and are similar to those obtained from original 3D-GNSF before wear test (point 1), confirming the presence of GNS on the surface of wear track. Meanwhile, the contact surface of stainless steel ball is covered by stable transfer layers which are optically smooth, uniform and well adhered to the surface (the inset of Fig. 5b). This coupled with a thick transfer layers appearing on the marginal of the contact surface suggests the GNS is easy to be transferred from brass-3D-GNSF to the ball counterpart. Raman characterization is used to verify the chemical nature of the transfer layers on the steel ball (Fig. 5b). Detection positions include two representative points on the contact surface (point 5 & 6) and one point locating at the marginal of the contact surface (point 4). For all the detection positions, two broad bands appear at ∼1330 and 1598 cm−1, which are the characteristics for the D and G bands of graphitic carbon, confirming the GNS have been transferred to the surface of steel ball. Furthermore, the SEM image coupled with 3D rough topography shows the wear track of brass-3D-GNSF is rather smooth with only a few of microploughs (Fig. 5c & e), which is very different from that of brass-HT, exhibiting many wear debris, fracturing, deep plough, and plastic deformation (Fig. 5d) along with much larger width and roughness (Fig. 5f). Thus, the wear mechanism is dominated by the mildly abrasive wear for brass-3D-GNSF, whereas adhesive wear and abrasive wear for brass-HT. Basing on the above observation, it is safe to conclude that the excellent tribological performance of 3D-GNSF come from their unique microstructure. First, those individual, crumpled and free-standing GNS

4. Conclusion Novel three-dimensional graphene nanosheet films with outstanding tribological performance have prepared on the metal surface, for the first time, by a facile one-step hydrothermal method. Taking the merits of plentiful active contact areas that facilitate the rapid transfer of GNS to the counter-face and three-dimensional interconnected network which increases the adhesion of GNS against each other and substrate, thus, quick formation of a compact and stable sliding interface consisting of GNS against GNS, the prepared three-dimensional graphene nanosheet films show low COF of ∼0.13, excellent anti-wear ability and good durability even when the maximum hertzian contact pressure is up to 1 GPa. This work unravels the new insights on the microstructure modulation of graphene nanosheet films in enhancing tribological performance and is open up a window of opportunity for designing graphene solid lubricant with low friction, excellent antiwear ability, good durability and high load capability.

34

Applied Surface Science 467–468 (2019) 30–36

F.X. Chen et al.

(b)

16.0k 1 2 3

Intensity / a. u.

12.0k

4 5 6

50k

Intensity / a. u.

(a)

8.0k 4.0k 0.0

40k

30k

20k 1250

1500

1750

1250

2000

1500

1750

2000

Raman shift / cm-1

Raman shift / cm-1

Fig. 5. Raman spectra obtained from wear track of brass-3D-GNSF (a) and the surface of stainless steel counterpart (b) after sliding test under a load of 6 N, the insets are their corresponding optical images where Raman measured regions are marked; SEM images (c & d) and 3D rough topographies (e, & f) of wear track of brass-3DGNSF and brass-HT after sliding test under a load of 6 N.

Acknowledgments

[12] D. Berman, A. Erdemir, A.V. Sumant, Carbon 54 (2013) 454–459. [13] S. Bhowmick, A. Banerji, A.T. Alpas, Carbon 87 (2015) 374–384. [14] Y. Huang, Q. Yao, Y. Qi, Y. Cheng, H. Wang, Q. Li, Y. Meng, Carbon 115 (2017) 600–607. [15] M.-S. Won, O.V. Penkov, D.-E. Kim, Carbon 54 (2013) 472–481. [16] S. Qi, X. Li, H. Dong, Mater. Lett. 209 (2017) 15–18. [17] D. Berman, S.A. Deshmukh, S. Sankaranarayanan, A. Erdemir, A.V. Sumant, Adv. Funct. Mater. 24 (2014) 6640–6646. [18] J. Zhou, Y. Huang, X. Cao, B. Ouyang, W. Sun, C. Tan, Y. Zhang, Q. Ma, S. Liang, Q. Yan, H. Zhang, Nanoscale 7 (2015) 7035–7039. [19] Y. Cai, F. Liu, Z. Luo, G. Fang, J. Zhou, A. Pan, S. Liang, Energy Storage Mater. 13 (2018) 168–174. [20] Y. Cai, X. Cao, Z. Luo, G. Fang, F. Liu, J. Zhou, A. Pan, S. Liang, Adv. Sci. 5 (2018) 1800680. [21] S. Liu, J. Wu, J. Zhou, G. Fang, S. Liang, Electrochim. Acta 176 (2015) 1–9. [22] X. Cao, Z. Yin, H. Zhang, Energy Environ. Sci. 7 (2014) 1850–1865. [23] H.-P. Cong, X.-C. Ren, P. Wang, S.-H. Yu, ACS Nano 6 (2012) 2693–2703. [24] Y. Ito, H.-J. Qiu, T. Fujita, Y. Tanabe, K. Tanigaki, M. Chen, Adv. Mater. 26 (2014) 4145–4150. [25] P. Saravanan, R. Selyanchyn, H. Tanaka, D. Darekar, A. Staykov, S. Fujikawa, S.M. Lyth, J. Sugimura, ACS Appl. Mater. Interfaces 8 (2016) 27179–27187. [26] S. Stankovich, D.A. Dikin, R.D. Piner, K.A. Kohlhaas, A. Kleinhammes, Y. Jia, Y. Wu, S.T. Nguyen, R.S. Ruoff, Carbon 45 (2007) 1558–1565. [27] Y.J. Mai, M.P. Zhou, H.J. Ling, F.X. Chen, W.Q. Lian, X.H. Jie, Appl. Surf. Sci. 433 (2018) 232–239. [28] I. Platzman, R. Brener, H. Haick, R. Tannenbaum, J. Phys. Chem. C 112 (2008) 1101–1108. [29] C. Hu, X. Zhai, L. Liu, Y. Zhao, L. Jiang, L. Qu, Sci. Rep. 3 (2013) 2065. [30] X. Cao, D. Qi, S. Yin, J. Bu, F. Li, C.F. Goh, S. Zhang, X. Chen, Adv. Mater. 25 (2013) 2957–2962. [31] Z.-J. Fan, W. Kai, J. Yan, T. Wei, L.-J. Zhi, J. Feng, Y.-M. Ren, L.-P. Song, F. Wei, ACS Nano 5 (2011) 191–198.

The authors are grateful to the National Natural Science Foundation of China (51605097) and Training Programs of Innovation and Entrepreneurship for Undergraduates (201811845148) for financial support. References [1] Y.H. Liu, B.J. Yu, Z.Y. Cao, P.F. Shi, N.N. Zhou, B. Zhang, J.Y. Zhang, L.M. Qian, Appl. Surf. Sci. 439 (2018) 976–982. [2] S.S. Xu, J.Y. Sun, L.J. Weng, Y. Hua, W.M. Liu, A. Neville, M. Hu, X.M. Gao, Appl. Surf. Sci. 447 (2018) 368–373. [3] G.L. Hou, Y.L. An, X.Q. Zhao, H.D. Zhou, J.M. Chen, S.J. Li, X. Liu, W. Deng, Appl. Surf. Sci. 411 (2017) 53–66. [4] D.G. Liu, J.P. Tu, C.F. Hong, C.D. Gu, Y.J. Mai, R. Chen, Appl. Surf. Sci. 257 (2010) 487–494. [5] Y.J. Mai, Y.G. Li, S.L. Li, L.Y. Zhang, C.S. Liu, X.H. Jie, J. Alloy. Compd. 770 (2019) 1–5. [6] D. Berman, S.A. Deshmukh, S.K.R.S. Sankaranarayanan, A. Erdemir, A.V. Sumant, Science 348 (2015) 1118–1122. [7] W. Zhai, N. Srikanth, L.B. Kong, K. Zhou, Carbon 119 (2017) 150–171. [8] S.C. Tjong, Mat. Sci. Eng. R. 74 (2013) 281–350. [9] H. Liang, Y. Bu, J. Zhang, Z. Cao, A. Liang, ACS Appl. Mater. Interfaces 5 (2013) 6369–6375. [10] P. Wu, X. Li, C. Zhang, X. Chen, S. Lin, H. Sun, C.-T. Lin, H. Zhu, J. Luo, ACS Appl. Mater. Interfaces 9 (2017) 21554–21562. [11] Y. Mai, H. Ling, F. Chen, C. Liu, L. Zhang, X. Jie, Mater. Res. Bull. 102 (2018) 324–329.

35

Applied Surface Science 467–468 (2019) 30–36

F.X. Chen et al.

[35] Z. Jia, P. Zhao, J. Ni, X. Shao, L. Zhao, B. Huang, B. Ge, C. Ban, J. Mater. Eng. Perform. 26 (2017) 4434–4441. [36] F. Chen, J. Ying, Y. Wang, S. Du, Z. Liu, Q. Huang, Carbon 96 (2016) 836–842. [37] P.F. Li, H. Zhou, X.-H. Cheng, Appl. Surf. Sci. 285P (2013) 937–944.

[32] L. Chun, S. Gaoquan, Adv. Mater. 26 (2014) 3992–4012. [33] Y. Xu, K. Sheng, C. Li, G. Shi, ACS Nano 4 (2010) 4324–4330. [34] Y.J. Mai, F.X. Chen, W.Q. Lian, L.Y. Zhang, C.S. Liu, X.H. Jie, J. Alloy. Compd. 756 (2018) 1–7.

36