报告摘要：(Sub)millimeter disk polarization is an exciting new field of research that has been revolutionized by ALMA. In contrast to its canonical picture, the polarization on the disk scale revealed does not come from magnetic fields but mostly from scattering. This scattering-induced polarization is very sensitive to the sizes of dust grains and local radiation fields. It is a powerful tool for studying grain growth and dust settling. I will discuss how the polarization signatures depends on the dust scale height and present detailed modeling on protoplanetary disks and constraints on the dust settling. Theoretically, the large (mm-sized) grains in protoplanetary disks cannot be aligned by magnetic fields due to their long Larmor precision timescale and short gas-damping timescale, which agrees with ALMA’s non-detection of magnetic fields. The magnetic alignment conditions are much better in the disk atmosphere, inhabited by micron-sized dust grains. We propose a brand-new method to probe magnetic fields using the near-IR polarimetry. Last but not the least, I will introduce our recent development of a Neural-Network aided GPU MCRT code to solve scattering by large aligned dust grains.
（亚）毫米波盘状结构的偏振是一个令人兴奋的新研究领域，阿塔卡马大型毫米 / 亚毫米波阵（ALMA）使其发生了革命性变化。与传统认知不同，盘尺度上的偏振并非来自与磁场，而主要来自尘埃的自散射。这种散射引起的偏振对尘埃颗粒的大小和局部辐射场非常敏感。它是研究颗粒生长和尘埃沉降的有力工具。我将讨论尘埃的标高如何影响偏振特征，进而通过对原行星盘的详细建模对尘埃沉降给出限制。从理论上讲，原行星盘中的大尺寸（毫米级）尘埃颗粒由于其长拉莫尔进动时标和短气体阻尼时标而无法被磁场对齐，这与 ALMA 未探测到磁场的结果一致。在由微米级尘埃颗粒占据的原行星盘表面大气中，磁场对齐条件要好得多。基于此，我们指出近红外散射光可能追踪盘状大气中的磁场，为原行星盘内磁场探测指出了新方向。在报告的最后，我将介绍我们最近开发的神经网络赋能的基于GPU的蒙特卡洛光线追踪（MCRT）代码，以解决大尺寸不规则对齐尘埃颗粒的散射问题。
个人简介：杨海峰，浙江大学百人计划研究员。本科毕业于北京大学，博士毕业于美国弗吉尼亚大学。2018年-2021年以杨振宁青年学者身份在清华大学高等研究院工作，2021年-2024年以博雅博士后在北京大学科维理天文与天体物理中心工作。杨海峰的工作专注于原行星盘和系外行星形成，研究方法包括（亚）毫米波段偏振观测、辐射转移模拟和磁流体力学模拟等。