论文标题
应变工程2D MOS $ _ {2} $带薄膜应力封盖层
Strain Engineering 2D MoS$_{2}$ with Thin Film Stress Capping Layers
论文作者
论文摘要
我们展示了一种通过压力薄膜的沉积来封装剥落的薄片,将拉伸和压缩应变诱导到二维过渡金属二甲基(TMDC)MOS $ _ {2} $中。通过这种技术,我们可以通过改变沉积的薄膜应力来直接设计$ _ {2} $应变幅度,从而允许在片段上施加可变应变。这些薄膜应力源类似于sin $ _ {x} $基于工业CMOS工艺以增强SI迁移率的压力源,这表明我们的概念是高度可扩展的,可以应用于应变工程TMDC设备的大规模整合。我们选择光学透明的压力源,以使我们能够通过拉曼光谱探测MOS $ _ {2} $应变。将MOS $ _ {2} $中拉曼峰移动的厚度依赖性分析与原子模拟相结合,我们可以探索一层应变转移。 MOS $ _ {2} $在常规基质(Sio $ _ {2} $,MGO)上显示应变转移到多层片的前两层,由于底物粘附而受到有限的应变转移到单层。为了减轻这一限制,我们还探索了由单层(1L)MOS $ _ {2} $构建的范德华异质结构的压力源,氮化硼(H-BN)。这个概念释放了1L-mos $ _ {2} $,允许将0.85 $ \%$菌株应用于单层,而相应的应变诱导的带隙更改为75 meV。通过使用压力较高的薄膜,应变可能会更高。考虑了各种应激源和沉积方法,显示了应力材料独立的应变转移,仅取决于使用热蒸发时诱导到MOS $ _ {2} $的应激膜力的力。
We demonstrate a method to induce tensile and compressive strain into two-dimensional transition metal dichalcogenide (TMDC) MoS$_{2}$ via the deposition of stressed thin films to encapsulate exfoliated flakes. With this technique we can directly engineer MoS$_{2}$ strain magnitude by changing deposited thin film stress, therefore allowing variable strain to be applied on a flake-to-flake level. These thin film stressors are analogous to SiN$_{x}$ based stressors implemented in industrial CMOS processes to enhance Si mobility, suggesting that our concept is highly scalable and may be applied for large-scale integration of strain engineered TMDC devices. We choose optically transparent stressors to allow us to probe MoS$_{2}$ strain through Raman spectroscopy. Combining thickness dependent analyses of Raman peak shifts in MoS$_{2}$ with atomistic simulations, we can explore layer-by-layer strain transfer. MoS$_{2}$ on conventional substrates (SiO$_{2}$, MgO) show strain transfer into the top two layers of multilayer flakes with limited strain transfer to monolayers due to substrate adhesion. To mitigate this limitation, we also explore stressors on van der Waals heterostructures constructed of monolayer (1L) MoS$_{2}$ on hexagonal boron nitride (h-BN). This concept frees the 1L-MoS$_{2}$ allowing for a 0.85$\%$ strain to be applied to the monolayer with a corresponding strain induced bandgap change of 75 meV. By using thin films with higher stress, strain may be engineered to be even higher. Various stressors and deposition methods are considered, showing a stressor material independent transfer of strain that only depends on stressor film force with negligible defects induced into MoS$_{2}$ when thermal evaporation is used.