物理
衍射
光学
极化(电化学)
波前
基点
数学分析
基质(化学分析)
投影(关系代数)
联轴节(管道)
领域(数学)
几何光学
角谱法
奇异值分解
平面波
物理光学
平面(几何)
斯托克斯参量
近场和远场
光场
傅里叶变换
惠更斯-菲涅耳原理
几何学
作者
HongQi Liu,yang zhao,Guanghai Zhang,Jian Bai,Cuifang Kuang,Xu Liu,Qing Yang
出处
期刊:
[Figshare (United Kingdom)]
日期:2026-04-24
标识
DOI:10.6084/m9.figshare.c.8257552
摘要
Vectorial diffraction calculation is fundamental to modern photonics, yet existing models suffer from mutually exclusive limitations: the Vectorial Debye integral (VDB), while accurate for high-NA focusing, is geometrically restricted to focal configurations and fails entirely at arbitrary propagation distances; the Vectorial Rayleigh-Sommerfeld integral (VRS), though applicable to arbitrary plane, systematically underestimates cross-polarized field components due to its scalar-like treatment of transverse propagation. Here we present a Spherical Vectorial Diffraction (SVD) framework that resolves the longstanding challenge of vectorial field calculation between arbitrary parallel planes in free space. Through systematic derivation from the Franz formula, we obtain a unified transfer matrix [H] for spherical wave propagation and demonstrate that VDB's geometric projection matrix M is the special case of [H] in the focal limit, while [H] extends seamlessly to arbitrary parellel planes with full vectorial fidelity. Crucially, SVD achieves VDB-level accuracy in high-NA focusing (mean deviation approx 1*10^-3) without additional computational cost, maintaining the efficiency of FFT-based angular spectrum methods. At arbitrary plane, SVD eliminates the order-of magnitude error amplification in longitudinal fields that plagues VRS's divergence-based reconstruction. Pinhole diffraction experiments confirm that only SVD correctly predicts cross-polarized field distributions—the signature of vectorial coupling inaccessible to either classical model. This unification reframes polarization coupling as an intrinsic diffractive phenomenon rather than a high-NA correction, establishing a general foundation for vectorial wavefront engineering.
科研通智能强力驱动
Strongly Powered by AbleSci AI