The Growth Kinetics of Multilayer Graphene Grown on Al2O3 Substrate by Chemical Vapour Deposition

Available online at www.sciencedirect.com

ScienceDirect Procedia Engineering 148 (2016) 1295 – 1302

4th International Conference on Process Engineering and Advanced Materials

The Growth Kinetics of Multilayer Graphene Grown on Al2O3 Substrate by Chemical Vapour Deposition May Alia,*, Suraya Abdul-Rashida,b, Mohd Nizar Hamidona,b, Faizah Md Yasina,b a

Chemical & Environmental Engineering Department, Faculty of Engineering, Universiti Putra Malaysia, 43400, Selangor, Malaysia. b Institute of Advanced Technology (ITMA), Universiti Putra Malaysia, 43400, Serdang, Selangor Darul Ehsan, Malaysia .

Abstract The growth of multilayer graphene (MLG) by chemical vapor deposition (CVD) on a dielectric Al 2O3 of (96%) substrate was studied to produce large area thin films with high carrier mobility. The growth of MLG on the deposition of Co-Ni over Al2O3 surface from ethanol decomposition at 700oC-800oC under ambient pressure, was explored using Raman spectroscopy (RS), Field Emission scanning electron microscopy (FESEM) and Transmission electron microscopy (TEM). Experimentally, it was observed that the implementation of ethanol as carbon precursors plays a crucial role in MLG growth, which offer a favourable kinetics model that determine the activation energy and the rate of deposition. Interestingly, MLG grown on dielectric substrates would enable various electrical and chemical sensor applications. © by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license ©2016 2016Published The Authors. Published by Elsevier Ltd. (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee of ICPEAM 2016. Peer-review under responsibility of the organizing committee of ICPEAM 2016 Keywords: Multilayer graphene; chemical vapor deposition; kinetics model; ethanol dehydration.

1. Introduction Graphene is composed of a single layer of sp2-bonded carbon atoms arranged in a two-dimensional (2D) honeycomb crystal lattice either in a freely suspended form or adhered to any substrate. Since 2004, graphene has attracted considerable interest due to its extraordinary physical and chemical properties which make it suitable for a large variety of applications such as transistors, nanoelectronic devices and sensors [1,2]. * Corresponding author. Tel.: +6018262529. E-mail address: [email protected]

1877-7058 © 2016 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license

(http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee of ICPEAM 2016

doi:10.1016/j.proeng.2016.06.533

1296

May Ali et al. / Procedia Engineering 148 (2016) 1295 – 1302

Graphene can be prepared by several methods like mechanical exfoliation of graphite, chemical reduction of graphene oxide, chemical vapor deposition (CVD), plasma enhanced CVD, and epitaxial graphene growth on SiC surfaces and many other substrates [4]. Among others, the CVD method has become extremely favoured by scientists, because of its ability to produce high quality and large-area graphene films, as well as the availability of inexpensive and transferrable process [5]. Thus, a great deal of interest in research and development of CVD graphene growth has proliferated all over the world [1-5]. The understanding of the growth kinetics of single-crystal graphene layers is key to develop its prospective applications [3,6]. The information about the growth rate can be extracted by observing the variation in graphene flake size over time [3]. Losurdo et al. [7] explored low pressure CVD kinetics in a real-time monitoring by using CH4–H2 as carbon sources on two polycrystalline metal substrates of copper and nickel catalysts at 900oC, with the aim to understand the role of hydrogen in differentiating graphene growth kinetics. Their study showed that hydrogen slows down the deposition kinetics of graphene on Cu by blocking sites on the Cu surface and also has a negative effect on the quality of the graphene grown on Cu. By optimizing and minimizing the hydrogen content, graphene with very minimal defects can be grown on polycrystalline Ni substrate [7]. After that, Bi et al. [8] reported the direct growth of graphene films on dielectric substrates by ambient pressure CVD (APCVD) at 1100–1200ºC using a gas mixture of CH4, H2 and Ar. Their results showed few-layered graphene films which obtained at high temperature 1200oC over BN, Si, SiO2 and AlN substrates. Multilayer graphene (MLG) is a graphene thin film of weak van der Waals interaction between its layers that show a distinguished electronic band structure which has attracted considerable attention from researchers due to its novel electronic properties along with its high potential for sensing applications [9]. However, a major step with regards to MLG growth techniques at low growth temperature, band gap and high carrier mobility has never been successfully achieved [3]. The growth rate is critical to understand the fundamental mechanisms that govern the CVD of MLG. Ethanol (C2H5OH) as a potential carbon source is a very important renewable energy source, which can be used as a neat fuel, as an oxygenate additive, or simply as a fuel extender as employed in gasohol. Despite its importance as a potential replacement for hydrocarbon fossil fuels, the kinetics and mechanism for its decomposition reaction at high temperatures are still not well characterized [10]. Herein, a significant study for high quality, large-area of epitaxial MLG growth based CVD and the growth kinetics and mechanism is reported. The deposition of bimetal Co-Ni alloy on top of the functionalized Al 2O3 substrate will facilitate the high-quality graphene growth and control the number of graphene layers [2,5]. Lowgrowth temperature range of 700oC-800oC was implemented with a fixed growth time of 2h. This work can achieve a better control on number of graphene layers even on dielectric substrates. We also reported Raman spectroscopy (RS), Field Emission scanning electron microscopy (FESEM) and Transmission electron microscopy (TEM) measurements of how hydrocarbon source along with growth temperature affected the MLG quality. The kinetics model for MLG/CVD is examined, thereby producing simple explanation of how a safe and low-cost ethanol as carbon feedstock promoted a favourable kinetics process. Nomenclature MLG multilayers graphene CVD chemical vapor deposition Al2O3 alumina RS Raman Spectroscopy FESEM Field Emission Scanning Electron microscope 2D two dimensional GC gas chromatograph LH Langmuir Hinshelwood

1297

May Ali et al. / Procedia Engineering 148 (2016) 1295 – 1302

2. Experimental details In this work, the formation of MLG films by ethanol decomposition on Co-Ni/Al2O3 substrate were performed using CVD growth technique. Firstly, Al2O3 of (1.5×1.5cm2) with ~ (0.61-0.67) g weight, was functionalized by immersing in a solution of HNO3 (50ml), H2SO4 (5ml) and H2O (500ml) in a solution shaker for 30 minutes. The Al2O3 substrate was placed inside the resulted solution for 1 week to encourage surface etching for the deposition of Co-Ni catalyst. For the second step, nitrate hexahydrate Ni(NO3)2.6H2O was mixed with cobalt nitrate Co(NO3)2.6H2O in different ratios, which were used for the catalyst preparation. The two metallic precursors were placed in 50ml of C2H5OH for 24 h, since the metallic ions in the solutions form chemical complexes which become larger in time [11]. Then, the catalyst solution was sonicated to obtain better dispersion. Afterwards, various catalyst ratios of 6ml from the solution were dropped via a syringe on the Al2O3 substrate, where the substrate was placed on a heater. Table 1 presents different catalyst ratios (Co/Ni) and their elemental weight composition that was utilized for MLG growth. It is known that the type of catalyst and its thickness determine the number of graphene layers [2]. For the graphene preparation, both substrate and catalyst (CoNi/Al2O3) were weighed according to required amount before placing them inside a CVD furnace reactor where C2H5OH was used as carbon source. Prior to the commencement of the process, the reactor was purged with Ar gas for 30 min to generate an inert atmosphere. C2H5OH was flowed for 2 h under a temperature ranged between 700 oC and 800oC, then the produced MLG films on Co-Ni/Al2O3 was allowed to cool down. However, the product gases from C2H5OH decomposition were analysed based on gas chromatograph (GC) (Agilent Technologies 7820A GC System). The obtained rate of carbon deposition was estimated from the flow of C2H5OH conversion determined as by the analysis of GC device. It was assumed that ethanol decomposed to CH4, Co, C2H4 and H2. Table 1. Various Co/Ni catalyst ratios that used for the MLG thin films preparation. No Catalyst Element O Element N Element Co Ratio(Co/Ni) Weight % Atomic % Weight % Atomic % Weight % 1 0.3/0.7 58.00 74.93 9.17 13.53 11.13 2 0.5/0.5 55.83 73.40 9.44 14.18 16.15 3 0.7/0.3 51.69 71.31 8.80 13.87 26.29 4 0.16/0.84 52.57 71.66 9.05 14.10 5.00 5 0.84/0.16 51.25 71.12 8.70 13.79 33.66

Atomic % 3.90 5.76 9.85 1.85 12.68

Element Ni Weight % 21.70 18.21 13.21 33.39 6.39

Atomic % 7.64 6.66 4.97 12.40 2.42

3. Results and Discussion RS was used to evaluate the graphene quality. The MLG thin films which synthesised onto Co-Ni/Al2O3 were characterized using RS (Jobin Yvon Horiba HR800UV) model. Fig. 1 shows the Raman spectra of the samples grown at 700°C, 750°C and 800°C, respectively. The main features of the Raman spectrum are the D, G and 2D peaks commonly seen for graphene type structures. The G peak which is assigned to the tangential mode of the graphene sheet, can be seen centred at 1537-1585 cm−1 in the spectrum. While, the 2D-peak at 2682-2701 cm−1 indicated the number of graphene layers, and the D-band at 1343-1351 cm−1 is also observed, indicating the presence of amorphous carbon or defects in the wall structure [5]. The main peaks with their corresponding temperature are shown in Table 2. Moreover, the qualitative information of the CVD graphene films that obtained from the ratios of I(D)/I(G) and I(2D)/I(G) which measures the graphene defects and the number of graphene layers, respectively [5]. From Table 2, the I(2D)/I(G) ratio =1.74 at 800°C is presented higher number of graphene layers than that of 700°C with less D peak values. While the lowest I(D)/I(G) value of 0.85 at 700°C is clearly indicated the growth of carbon nanotubes and graphene crystals as shown in Fig. 2 a.

1298

May Ali et al. / Procedia Engineering 148 (2016) 1295 – 1302

Fig. 1. RS image of graphene grown on Co-Ni/Al2O3 at versus growth temperature.

Table 2. Presents the final result from RS analysis and the intensity ratios for the MLG films. T (oC)

D band

G band

2D band

ID/IG

I2D /IG

700

1351.73

1585.79

2701.33

0.85

1.70

750

1345.76

1571.85

2696.31

0.86

1.72

800

1343.78

1537.86

2682.41

0.87

1. 74

The synthesized graphene at different reaction temperatures were analysed using FESEM (JSM-7800F) micrograph, as displayed in Fig. 2(a-c). Careful observation of the FESEM micrographs reveal that the morphology of MLG at 700°C have crystal shapes with high degree of defects. Moreover, there was a sign of agglomeration of the MLG thin films at 700°C and 750°C (Fig. 2 a,b). However, as the reaction temperature increases to 800°C (Fig. 2 c), the defects were noticed to be reduced. Moreover, it was observed that at higher magnification for the 800°C reaction temperature, the graphitized graphene were gradually grew to form a MLG thin films stacked nearby each other which indicate a high level of graphene crystallinity (see Fig. 2c). Herein, the reaction of growth temperature (700800°C) is lower than the common CVD temperature range (~1000oC) [7,8], which is significantly affected the large area and high quality of MLG films.

a

b

May Ali et al. / Procedia Engineering 148 (2016) 1295 – 1302

1299

c

Fig. 2. FESEM imaging of MLG/CVD grown on a Co-Ni/Al2O3 surface at various temperatures, a) 700oC, b) 750oC and c) 800oC.

Fig. 3 presents the TEM (H-7100 Electron Microscope) image of the MLG wall structure. Where, high quality MLG thin films of sharp edges and less defects were densely grown on Co-Ni/Al2O3 surface parallel to each other.

Fig. 3. TEM characterization of MLG/CVD grown on a Co-Ni/Al2O3 surface at 800oC.

3.1. Kinetics studies It is known that the activation energy indicates the type of reaction occurs, either physical or chemical reaction that happens at the catalyst. This work implemented a simple growth kinetics model that can explain the observed MLG growth. Under combustion conditions, C2H5OH may decompose via many energetically accessible product channels, such as

Adsorption: Surface reaction: Surface reaction: Desorption:

C2 H 5OH ( g )  x o C2 H 5OH . x C2 H 5OH .x o (CH 2 ) 2 O. x  2 H . x 1 2(CH 2 ) 2 O. x o C2 H 4 . x  CH 4 . x  2CO n  H 2 n 2 C2 H 4 . x o C 2 H 4  x

(1) (2) (3) (4)

1300

May Ali et al. / Procedia Engineering 148 (2016) 1295 – 1302

Desorption: CH 4 . x o CH 4  x

(5)

2H . x o H 2 n  2 x

(6)

Desorption:

To understand the mechanism of C2H5OH carbon sources formation occurs on Co-Ni/Al2O3 surface, we assumed a schematic diagram (Scheme 1 below) that summarized all the dehydrogenation steps. The first block on the left, showed the adsorption step. And the other two blocks in the same row presented the surface reaction steps. While Eq. 3 above, represents the MLG growth which is clearly identified by the forth block. CH3 C2H5OH

CH3

CH2

Co-Ni

C2H5OHCoNi

Al2O3

H

CH2

H

O

O

Co-Ni

Co-Ni

Al2O3

Al2O3

Co H

H

C2H4 G Co-Ni CH4

G

CH3CHOCoNi

Al2O3

Scheme 1. The mechanism for MLG nucleation form the ethanol dehydration over Co-Ni/Al2O3.

The kinetics expression for the rate of ethanol consumed denoted by ( rC2 H5OH ), is concluded from the mechanism presented in Scheme 1 and calculated by Eq. 7 below which is known as Langmuir Hinshelwood (LH) Equation [12] rC2 H 5OH

krxn PC2 H 5OH PH 2 (1  kC2 H 5OH PC2 H 5OH  k H 2 PH 2 )3

(7)

Where krxn is the kinetics parameter that represents the constant reaction rate, and PC2 H5OH and PH 2 are the partial pressures of ethanol and hydrogen, respectively. Table 3 shows the values of kinetic parameters ( krxn , kC2 H5OH and kH2 ) which were estimated from Eq. 7 by using the Polymath software. Whereas, for each temperature value, rC2 H5OH , PC2 H5OH and PH 2 were inputted and then the program runs after predicting the initial model parameters of

krxn , kC2 H5OH and kH2 , which indicated the accurate krxn value.

1301

May Ali et al. / Procedia Engineering 148 (2016) 1295 – 1302 Table 3. The estimation data of kinetics parameters.

krxn

kC2 H5OH

kH 2

700

0.0028

9.31643

0.0299

750

0.00504

0.0587

1.324

800

0.00905

1.575

2.011

T (oC)

Then from the experimental data in Table 3 and in order to estimate the activation energy E (J/mol), and the frequency factor A, a plot of ln(krxn ) versus 1/T was made in Fig. 4(a) based on the Arrhenius Equation (Eq. 8), where the reaction barrier is derived as a logarithm of the reaction constant at any temperature [12] k A (T )

Ae  E / RT

(8)

where R= 8.314 J/mol is the gas constant and T is the absolute temperature (K). The R2 value of 0.975 shows a strong fitting of the LH model to the kinetics data. The activation energy value of 13.72 kJ/mol obtained signifies that the mechanism is by chemical adsorption, where the ethanol bonds are broke down and form a new molecules of MLG films. While the pre-exponential factor (A) value of 5.5×104/s showed that the reactant species collide at a faster rate forming the desired product. Furthermore, Fig. 4(b) shows a comparison performance between the observed and predicted rC2 H5OH which is successfully displayed a good agreement to the kinetics data.

Fig. 4. a) Calculation of the activation energy from an Arrhenius plot, and b) Comparison of the predicted and the observed rate of reaction.

4. Conclusions MLG/CVD with high crystalline quality was found to cover the entire of Co-Ni/Al2O3 catalyst substrate at (700800oC) under ambient presser, where ethanol was used as a cheap and safe carbon precursors to produce high quality, large-area MLG films. The morphology of MLG was systematically characterized by using RS, FESEM and TEM analysis. Based on the results obtained, we concluded that the main effects of the high temperature along with the increment in hydrocarbon flow were significant factors in MLG growth. Furthermore, the kinetics studies showed that the CVD reaction rate is controlled and can be described by Langmuir-Hinshelwood kinetics model.

1302

May Ali et al. / Procedia Engineering 148 (2016) 1295 – 1302

This work provides a clear understanding of mechanism and growth kinetics of MLG by CVD in the presence of ethanol under low growth temperature.

Acknowledgements The authors appreciate the financial contribution from the FRGS grant (Vote. No 5524471) in UPM.

References [1] W.W. Liu, S.P. Chai, A.R. Mohamed, U. Hashim, Synthesis and characterization of graphene and carbon nanotubes: A review on the past and recent developments, J. Indus. Eng. Chem. 20 (2014) 1171-1185. [2] C.M. Seah, S.P. Chai, A.R. Mohamed, Mechanisms of graphene growth by chemical vapour deposition on transition metals, Carbon 70(2014) 1-21. [3] H. Kim, E. Saiz, M. Chhowalla, C. Mattevi, Modeling of the self-limited growth in catalytic chemical vapor deposition of graphene, New Journal of Physics 15 (2013) 053012. [4] X. Chen, L. Zhang, S. Chen, Large area CVD growth of graphene. Synthetic Metals 210 (2015) 95-108. [5] S.P. Chai, K.Y. Lee, S. Ichikawa, A.R. Mohamed, Synthesis of carbon nanotubes by methane decomposition over Co– Mo/Al2O3: Process study and optimization using response surface methodology. App. Catalyst. A: General 396 (2011) 52-58. [6] K.F. McCarty, P.J. Feibelman, E. Loginova, N.C. Bartelt, Kinetics and thermodynamics of carbon segregation and graphene growth on Ru(0001), Carbon 47 (2009) 1806-1813. [7] M. Losurdo, M.M. Giangregorio, P. Capezzuto, G. Bruno, Graphene CVD growth on copper and nickel: role of hydrogen in kinetics and structure, Physical Chemistry Chemical Physics. 13 (2011) 20836-20843. [8] H. Bi, S. Sun, F. Huang, X. Xie, M. Jiang, Direct growth of few-layer graphene films on SiO2 substrates and their photovoltaic applications. Journal of Materials Chemistry 22 (2012) 411-416. [9] Katsnelson, Mikhail I., and Mikhail Iosifovich Kat︠ ︡snelʹson. Graphene: carbon in two dimensions. Cambridge University Press, 2012. [10] J. Park, R.S. Zhu, M.C. Lin, Thermal Decomposition of Ethanol. 1. Ab Initio MO/RRKM Prediction of Rate Constant and Product Branching Ratios, J. Chem. Phys. 117 (2002) 3224-31. [11] M. Li, L. Liu, Y. Xiong, X. Liu, A. Nsabimana, X. Bo, L. Guo, Bimetallic MCo (M= Cu, Fe, Ni, and Mn) nanoparticles doped-carbon nanofibers synthetized by electrospinning for nonenzymatic glucose detection. Sensors and Actuators B: Chemical. 207 (2015) 614-622. [12] H.S. Fogler, Essentials of Chemical Reaction Engineering, Pearson Education. 2010. G

G