- 學位論文
金屬光柵在量子井紅外線光偵測器上的特性
Characteristics of Metallic Grating on Quantum Well Infrared Photodetector
摘要
入射電磁波在金屬光柵附近及通過之後，會有特殊且複雜的現象，目前的研究大多是穿透率實驗，討論穿透光譜特性，而偵測器距離光柵遠大於波長範圍。我們想要瞭解電磁場在距離光柵短於波長的範圍之特性，所以在QWIP上製作光柵。QWIP的作用區大約在光柵下1μm處，而我們觀察的波長範圍大約在10μm附近，小將近一個數量級。我們成功以這種設計研究金屬光柵的特性，也能對應穿透率實驗的結果。 我們利用一系列的實驗證明，來自金屬光柵的波鋒產生之位置與穿透率實驗的位置相近，和基板的介電係數與光柵週期有關，但是穿透實驗的光柵波鋒來自TE模態，而QWIP實驗的光柵波鋒來自TM模態的入射光，這是由於量測距離不同所造成。我們改變光柵週期，將光柵波鋒調整到結構波鋒處，將QWIP的響應最佳化。同時我們發現邊緣耦合提供了很大部分的響應，來自TM模態Brewster角入射的結果，正面入射空框也有響應，來自尺寸效應的結果。 我們發現增加光柵厚度可以TM模態響應減少較快，增加TM除以TE模態響應的比例，但整體來說也會減少響應。最後我們發現將金屬光柵埋入元件，也就是讓光柵靠近QWIP的作用區，可以提高響應；但另一方面，由於光柵厚度變大，穿過的電磁波衰減與也使得響應減少，而在這兩者之間可以找到平衡點。
並列摘要
Electromagnetic wave exited around metallic grating and transmitted through grating is complicated and interesting. Recent studies about subwavelength metallic structure were mainly experiments measuring transmission spectrums which photodectors were set vary far from the sample. We wanted to understand the characteristics of grating near the sample. We fabricated grating structure on QWIP and the active region of QWIP is under the surface about 1 μm. By observing spectral response of QWIP with grating, we successively found fundamental characteristics about grating and there was a good correspondence to transmission spectrums. The experiments showed that the grating peak appeared at similar position in response of QWIP and transmission spectrum of sample, dominated by period of grating and refractive index of substrate. The grating peak of QWIP experiment was excited by TM incident light, but TE incident contributed peak in transmission spectrum, due to the distance of the measurement. We tuned the grating peak to structure peak to enhance the responsivity. In the mean time, we discovered that the edge coupling provided a large part of response, which came from TM incident light in Brewster’s angle. Normal incidence of hollow frame also provided response because of finite size effect. We discovered that the grating thickness could raise the ratio of TM over TE mode, because the TE mode decayed faster than TM mode. But it is a trade off that the total responsivity was decreased. At last, we could embed grating in to QWIP, the grating was much closer to the active region of QWIP. More quantity of exited wave absorbed to enhance responsivity. On the other hand, as the thickness increase, the transmitted electromagnetic wave decays, less wave went through grating to decrease responsivity. We could find some balance between these two factors.