物理化学学报 >> 2019, Vol. 35 >> Issue (10): 1119-1127.doi: 10.3866/PKU.WHXB201812023

所属专题: 二维材料及器件

论文 上一篇    下一篇

单层和双层二硫化钼化学气相沉积生长的动力学蒙特卡罗模拟研究

陈帅,高峻峰,SRINIVASAN Bharathi M.,张永伟*()   

  • 收稿日期:2018-11-12 发布日期:2019-01-17
  • 通讯作者: 张永伟 E-mail:zhangyw@ihpc.a-star.edu.sg
  • 基金资助:
    the Science and Engineering Research Council through Grant(152-70-00017);Use of Computing Resources at the A*STAR Computational Resource Centre and National Supercomputer Centre, Singapore

A Kinetic Monte Carlo Study for Mono- and Bi-layer Growth of MoS2 during Chemical Vapor Deposition

Shuai CHEN,Junfeng GAO,Bharathi M. SRINIVASAN,Yong-Wei ZHANG*()   

  • Received:2018-11-12 Published:2019-01-17
  • Contact: Yong-Wei ZHANG E-mail:zhangyw@ihpc.a-star.edu.sg
  • Supported by:
    the Science and Engineering Research Council through Grant(152-70-00017);Use of Computing Resources at the A*STAR Computational Resource Centre and National Supercomputer Centre, Singapore

摘要:

通过化学气相沉积方法,可控合成所需层数的二硫化钼仍然是一个挑战。因此,建立一个能够定量预测单层和多层二硫化钼生长的理论模型,并为实验制备提供指导,是十分必要的。在本文中,我们建立了一个动力学蒙特卡罗模型,来预测单层和双层二硫化钼的化学气相沉积生长。首先,我们提出了第一层和第二层的生长速率受吸附原子浓度分布的控制,以及紧凑三角形二硫化钼的生长过程为扭结成核和传播。其中,原子浓度是由吸附原子流量,吸附原子的有效寿命,生长温度,边的单位长度能量,单层和双层的单位面积结合能,成核准则决定的。扭结成核和传播是由锯齿边和扶手边附加原子所需的能量势垒决定的。然后,我们采用热力学理论准则对这些参数进行了标定。通过标定的动力学蒙特卡罗模型,我们发现第二层的生长速率与第一层的尺寸有很强的依赖性。随着第一层尺寸增加,第二层的生长速率呈单调递减趋势,甚至在第一层达到某个尺寸时,第二层的生长会被抑制。此外,我们还分析了不同生长温度和吸附原子流量下,双层二硫化钼的尺寸和形貌演化。在双层二硫化钼的整个生长过程中,第一层和第二层的形貌保持紧凑三角形,验证了扭结成核和传播模型的正确性。模拟结果表明,生长温度的升高或吸附原子流量的降低,促进了双层二硫化钼的生长,这与已报导的实验结果相吻合。生长温度升高使得第二层二硫化钼边缘的吸附原子浓度,随着远离第二层边缘的吸附原子浓度降低而相应降低,促进了双层二硫化钼的生长。同样,吸附原子流量降低减小了基体上的吸附原子浓度,降低了第一层远离边缘和靠近边缘的吸附原子浓度差,从而减缓了第一层的生长。第一层的生长减慢,减缓了第二层远离边缘和靠近边缘的吸附原子浓度差减小到零,从而促进双层二硫化钼的生长。为了更好地指导实验,我们进一步构建了双层二硫化钼生长的相图,可通过控制生长温度和吸附原子流量来实现或阻止双层二硫化钼的生长。因此,本工作不仅揭示了单层和双层二硫化钼生长所需的实验条件,而且为可控合成所需层数的二硫化钼提供了详细指导。

关键词: 二硫化钼, 单层生长, 双层生长, 化学气相沉积, 动力学蒙特卡罗

Abstract:

Controllable synthesis of MoS2 with desired number of layers via chemical vapor deposition (CVD) remains challenging. Hence, it is highly desirable to develop a theoretical model that can be used to predict the single- and multilayer growth of MoS2 quantitatively, and provide guidelines for experimental fabrication. Herein we have established a kinetic Monte Carlo (kMC) model to predict the CVD growth of mono- and bilayer MoS2. First, we proposed that the growth rates of layer 1 and layer 2 were governed by the distribution of the adatom concentration, and the growth kinetics of compact triangular MoS2 followed the kink nucleation-propagation mechanism. The adatom concentration was formulated in terms of adatom flux, effective lifetime of adatoms, growth temperature, binding energies, edge energies, and nucleation criterion. The kink nucleation and propagation were determined by energy barriers of the adatom attachments to the zigzag and armchair edges. We then employed an analytic thermodynamic criterion to extract these parameters. Using the calibrated model, we found that the growth rate of layer 2 strongly depended on the size of layer 1 and decreased monotonically with increasing size of layer 1, and might even become prohibited at the maximum size of layer 1. Furthermore, we analyzed the size and morphology evolutions of bilayer MoS2 at different growth temperatures and adatom fluxes. Throughout the growth processes of bilayer MoS2, the morphologies of layers 1 and 2 maintained triangular shapes with compact edges, consistent with the kink nucleation-propagation growth mechanism. Our simulations revealed that the growth of bilayer MoS2 was promoted by increasing the growth temperature or decreasing the adatom flux, which corroborated the experimental observations. The increase in growth temperature led to reduced adatom concentration at the edge of layer 2 in accordance with the adatom concentration far from the edge of layer 2, resulting in a consistent difference in the adatom concentration to promote the growth of bilayer MoS2. Similarly, the decrease in adatom flux lowered the difference between the adatom concentrations far from the edge and at the edge of layer 1, decelerating the growth of layer 1. The decelerated growth of layer 1 reduced the difference between the adatom concentrations far from the edge and at the edge of layer 2 to zero, permitting the growth of bilayer MoS2. To guide the experimental synthesis, we constructed a phase diagram to delineate the permitted or prohibited growth of bilayer MoS2 at different growth temperatures and adatom fluxes. Hence, this work not only unveils the conditions for the growth of mono- and bi-layer MoS2, but also provides guidelines for controllable synthesis of MoS2 with the desired number of layers.

10.3866/PKU.WHXB201809013.F009  

Key words: MoS2, Monolayer growth, Bilayer growth, Chemical vapor deposition, Kinetic Monte Carlo

MSC2000: 

  • O641