仿生資料庫

Database of Surface Plasmon Resonance in Au NPs(金奈米粒子表面電漿共振資料庫)

Introduction

In response to an oscillating electric field such as light illumination, the conducting electrons in a metal nanoparticle (NP) perform as a harmonic oscillator. An electric field with the opposite direction of the light’s electric field is created inside and outside of the metal NP by the redistribution of charge. Due to the displacement of electron density, a coulombic restoring force begins a series of oscillations in the NP. These coherent oscillations of charge density and electric field are called localized surface plasmons. Many metals exhibit plasmon resonances overlapping with the solar spectrum. Surface plasmon resonance is very sensitive to the physical and chemical properties of the matrix and NPs, including refractive indexes of the matrix, chemical bindings and surfactant molecules on the NPs as well as the element, size, shape, and density of the NPs, etc.

In this database, surface plasmon resonance wavelengths of gold NPs with sizes of 5-20 nm and in various environments with refractive index ranging from 1.0 to 1.5 are provided. The database is established by data mining and machine learning through the strategy of combining artificial neural networks with genetic algorithm. The learning process is up to 2,500,000 times. Root mean square (RMS) is less than 0.001. False rate is less than 0.00001.

當金屬奈米粒子照光時，其內部電子會具有諧振子的行為現象，以響應此外來的振盪電場。由於電荷的重新分佈，在金屬奈米粒子的內外會產生與入射光電場方向相反的電場。電子密度位移所誘發的庫侖恢復力，即開啟電子在金屬奈米粒子中的一系列振盪行為。這些電荷密度和電場的所產生的相干振盪即為局部表面電漿共振。許多金屬具有與太陽光譜重疊表面電漿共振頻率。而表面電漿共振對金屬奈米粒子和其存在之介質的物理和化學性質非常敏感，包括介質的折射率，奈米粒子表面的化學鍵結和表面活性劑分子，以及金屬奈米粒子的元素組成，大小，形狀和密度等。本資料庫提供不同尺度(5-20 nm)金奈米粒子於不同折射率(1.0-1.5)介質中的表面電漿共振波長。資料庫中的數據是以資料探勘和機器學習方式，經結合基因演算與人工神經網絡的深度學習所建立。學習程序為2,500,000次，均方根（RMS）小於0.001，誤率則小於0.00001。