以氧化鋅奈米結構達成室溫波思—愛恩斯坦凝結及高效能太陽能轉換

Abstract

在高經濟發展和全球暖化的兩難下，開發能提供高亮度低能耗或高太陽能轉換效率 的靳新光電元件和/或系統，是一件極為廹切的事情。雖然半導體雷射具有高電子-電洞復 合率(EHR)和輸出效率，但需操作在某個閾值電流上，這樣就不保証能節能。然而，擁有 共振腔LED，雖操作在閾值電流下，但因具有微共振腔增強電子-光子作用，增加了EHR 速率和修正輻射場形。假使將半導體材料或結構激發產生夠高的激子密度，使得激子之平 均間距已小於它們的波函數的範圍，則激子為一種准玻色子，因為重疊而無法分辨彼此， 因此將可能一起處於最低能態。這個狀態稱為激子玻思-愛恩斯坦凝聚(BEC)。其結果是只 要有任一個激子復合，就會觸發所有的凝結激子同時復合發出相同頻率、極化、和方向的 光。這個像雷射般的同調輻射具有高內部和取出效率，即相當小的能耗；最近已在19K 下的CdTe 微腔結構，經低於雷射50 倍閾值下光泵浦被實現。另外，Gratzel 提出染敏電 池(DSCs)結構可將光吸收的染料和作為電極的半導材料（TiO2)界面，所產生的電子電洞 對，在它們復合之前快速分離，以製成高效率太陽能電池。 類似GaN 和TiO2(for DSCs with h =11%)，氧化鋅是一種綠色或對環境無害的寬能隙 半導體(bandgap ~ 3.37 eV) 具有很大的激子束縛能(~60 meV)，遠大於室溫(RT)熱能的26 meV；因此，室溫下最主要為激子放光。另外，由於ZnO 具高Mott 密度(3.7 x 1019 cm-3) 和 大Rabi splitting 約120 meV，理論上已証實在ZnO 微腔中無法觀察到Rabi 振盪，相反地， 它反倒是觀察Polariton BEC 的最佳選擇。另一方面，ZnO 具高電子遷移率約155 cm2/Vs， 比TiO2 還大許多，使得ZnO 極有可能取代TiO2 作為DSC 的電極。目前，我們已利用ZnO 奈米結構將DSC 推展至紀錄的h > 5.0%。以我們過去在成長高晶體品質的ZnO 奈米結 構、磊晶薄膜、和ZnO/MgZnO 多重量子井結構的能力，我們預期有能力利用ZnO 微共 振腔結構實現室溫polariton BEC。實現室溫polariton BEC 不但是物理學研究上的重要基 礎課題，也包含其它非線性物理課題，如超流現象、渦流、糾纏極化子態、和渾沌約瑟遜 振盪應用於高達50Gb/s 之渾沌通訊。如達成本計畫預期將可帶動低耗能LED 躍進式的 技術開發能量。同時，如能將ZnO 奈米結構的DSC 效率提奸升至10%，就可商品化； 如此將對近年來每年成長25%至30%的能源產值的太陽能光電產業作出貢獻。

In the dilemma of high economic growth and global warming, it is urgent to develop new devices and/or systems that provide either bright light sources with low energy consumption or efficient photon harvest from solar energy. Although a laser diode possesses both high electron-hole recombination (HER) rate and excellent extraction efficiency, it requires operating beyond a threshold pumping that does not guarantee low energy consumption. The resonant cavity LED however has been operated below the lasing threshold through enhancing electron-photon coupling to increase EHR rate and modifying the emission profile. If high density of excitons are excited in semiconductor materials or structures such that the average separation of excitons is shorter than the extend of their wavefunctions, then the excitons as quasi-Bosons are intrinsically indistinguishable and will all condense to the lowest energy state. It is called the exciton Bose-Einstein condensates (BEC). As a result, if any one of the excitons recombines, it will stimulate all the condensate excitons to simultaneously emit photons at the same frequency, polarization, and direction. This coherent emission as a laser does with high internal and extraction efficiencies has extreme low energy consumption recently realized in CdTe microcavity at 19K under optical excitation 50 times below the lasing threshold. For obtaining high efficient solar cells, Gratzel has proposed fast separation of electrons and holes in dye sensitized solar cells (DSC) prior to being recombined at the junction of photon harvesting dye and semiconductor as an electrode. Similar to GaN and TiO2 (for DSCs with h =12%), ZnO is a “green” or environmental-friendly wide direct bandgap semiconductor (bandgap ~ 3.37 eV) having large exciton binding energy (~60 meV), which is much larger than the thermal energy at RT (26 meV). Therefore, dominant exciton emission can be constantly observed in ZnO even at RT. Due to its high Mott density (3.7 x 1019 cm-3) and large Rabi splitting of 120 meV, it has been theoretically shown that Rabi oscillations should not be observed in ZnO microcavities, which nevertheless remain good candidates for exciton BEC involving the lower polariton branch. On the other hand, ZnO has very high electron mobility of 155 cm2/Vs as compared with TiO2 that makes ZnO a potential candidate for DSCs. Recently, we have made ZnO nanostructure based DSC achieve the record h > 5.0%. With the capability of fabricating high crystal quality ZnO nanostructures, epitaxial films, and ZnO/ZnMgO multiple quantum-well structures, I anticipate the realization of RT polariton BEC in the optical MC embedded ZnO MQWs under NSC support. Accomplishing RT polariton BEC is a common goal in Physics with other interesting effects, including superfluidity, vortex formation, quantum entanglement, and chaotic Josephson oscillations applicable for chaos communication at rates up to 50 Gb/s, and generation of entangled polariton states. It not only is a fundamentally important scientific research but also provides a quantum jump to the technology development of LEDs under low injection current---an extremely effective energy usage. On the other hand, achieving energy conversion efficiency of DSCs up to 10% would put DSCs into the photovoltaic market, which is expected to grow by 25 to 30% of energy production yearly in the coming decades.