單軸應力對少數層硒化銦角分析二倍頻的影響

Abstract

硒化銦晶體是一種新興的凡德瓦材料，具有良好的電性與光電特性，並且已被應用於製成元件。此種層狀堆疊的材料在少數層時具有良好的彈性，使其具有被應用在彈性元件上的潛力，探討應力如何改變硒化銦的光電特性便成為一個重要的議題。然而，在進行應力與硒化銦光電特性的探討前，需要得知如何量測硒化銦之應力大小，本次研究目的便是藉由探討硒化銦二倍頻與其所受應力大小的關係，以期使二倍頻技術應用於測量硒化銦所受應力。 近年來有許多技術能直接或間接地測量材料所受之應力大小，而利用二倍頻量測也是其中之一。由於二倍頻對於材料的結構息息相關，透過測量材料不同角度的二倍頻訊號能解析材料的結構，也因此當材料受應力而產生結構改變時，會使二倍頻訊號受影響，若測量不同應力下的角分析二倍頻，能得出材料所受的應力方向以及大小。 本篇論文探討硒化銦少數層受單軸應力時，其角分析二倍頻的變化。在實驗方面，我們透過測量拉曼光譜確認晶體為硒化銦，並以光致螢光光譜得知其能隙。由於層狀結構的硒化銦晶體可以不同的堆疊方式構成，我們利用斜向入射二倍頻的技術得知硒化銦的相位為γ，並以剝離的製程方式得到少數層硒化銦，將其放置於透明且具彈性的基板上。利用自製的應力裝置擠壓基板的左右兩側，使其形變，施加單軸壓縮應力於少數層硒化銦，並利用脈衝雷射激發硒化銦產生二倍頻訊號。我們得到角分析二倍頻的方式是旋轉樣品本身，在不同的選轉角度下測量其二倍頻訊號。從實驗結果得知硒化銦的角分析二倍頻訊號隨壓縮應力增加而減小。 理論計算方面，我們利用第一原理進行硒化銦的電子結構計算，並更進一步得到不同應力下硒化銦的二階極化率，藉由數學計算二倍頻與角度的關係，結合第一原理的計算結果與數學公式，得出理論上硒化銦受單軸壓縮應變時，其角分析二倍頻的訊號亦會遞減。 儘管實驗和理論結果都顯示了二倍頻隨應力增加而減小的現象，我們使用基於第一原理結果的模型來擬合實驗數據時發現仍存在差異。我們利用計算相關係數的方式得知理論模型與實驗數據的相關性。結果顯示，隨著應力越大，相關係數的值變小，可知由第一原理計算的變化較實際值小，因而在應力較大時與實驗值的結果較不吻合。然而，在應力變化項的部分增加一修正項時，實驗數據與模型能在各個應力下得到良好的擬合結果。

Indium selenide (InSe) is an emerging van der Waals material. It exhibits well electronic and optoelectronic properties for devices. This layered material possesses well deformability when it consists of few-layers such that it applicable to flexible devices. Therefore, how strain influences InSe becomes an important issue. Recently, many techniques have been developed to measure the strain on materials, and one of them is second harmonic generation (SHG). Owing to the structure-sensitive properties, the angle-resolved SHG pattern is able to reflect the structure of the material. While the structure of the material is deformed due to the strain, the angle-resolved SHG pattern will also vary accordingly. Therefore, the angle-resolved SHG pattern can be used to measure strain on the material. In the thesis, we systematically investigated the effect of strain on angle-resolved SHG from few-layers InSe. In the experiment, we used oblique incident SHG technique to identify the phase of the InSe crystal. The few-layers InSe sample was prepared by mechanical exfoliation and placed on a flexible and transparent substrate. With the self-designed strain device, we were able to apply the strain on the sample by pushing the substrate from the two sides. To measure the SHG from InSe, we used pulse laser to induce the nonlinear effect from InSe. The way we obtain angle-resolved SHG is by rotating the sample itself and measuring its SHG at different rotation angles. The results show the SHG intensity decreases while the compressive strain level increases. For the theoretical approach, we used first-principles method based on density functional theory (DFT) and generalized gradient approximation (GGA) to calculate the electronic structure and the second-order susceptibility of InSe. Integrating with the mathematical description and the first-principles results, we obtained the relation between compressive strain and the angle-resolved SHG, which also reveals decreased intensity with increasing strain level. Despite the fact that the experimental and theoretical results exhibit similar strain effects, there is a discrepancy between them as the experimental data was fitted using a model based on the first-principles results. To quantitatively understand this difference, we calculated the coefficient of determination (r-squared) values of the fitting results for each strain case. The results showed that the r-squared value decreased with increasing strain, indicating that the first-principles method underestimated the strain effect. However, by incorporating a correction term to modify the fitting model, the first-principles results were able to agree well with the experimental results.