EAPSI: Probing Photosynthetic Processes with Computers

尽管为绿色能源生产制造更高效、更环保的太阳能机器已经取得了巨大的进步，但设计既高效又廉价的太阳能电池仍然是一个挑战。这就是研究在植物和细菌中发现的自然发生的光合复合体的主要原因：这些生物含有进化优化的结构，能够产生接近100%的效率，将阳光转化为潜在的可行能源。由于许多关键的光合作用过程发生得非常快，只能通过实验获得有限的信息。这使得计算研究特别有用，并促使我们发展基于量子和经典物理的精确可靠的数学模型来模拟亚纳米和飞秒化学。在与韩国浦项科技大学的Young Min Rhee博士的合作中，这项研究将结合两种计算方法来开发一种准确的表示，即阳光如何在自然发生的光合作用复合体中转化为化学能。反应中心是光系统II色素-蛋白复合物的一个亚基，在此发生激子到电荷跃迁流形的量子跃迁。几个小组已经研究了这个复合物，然而，仍然不清楚为什么一些电荷转移状态产生而另一些没有。本研究的重点是建立一个统一的反应中心模型。蛋白质环境可以看作是激子流形的经典点电荷。然而，对于电荷分离的流形，带电的发色团将显著极化局部环境，静态电荷描述将是不够的。因此，本研究将为QM/MM和MD计算构建蛋白质的极化表示。整体的能量景观将被实施到量子动力学算法中，以模拟能量的传递和转换。美国国家科学基金会EAPSI奖与韩国国家研究基金会合作资助。

Although great advances are being made to build more efficient and environment-friendly solar machines for green energy production, it still remains a challenge to design solar cells that are both efficient and inexpensive. This is the main reason to study naturally-occurring photosynthetic complexes found in plants and bacteria: these organisms contain evolutionarily optimized architecture capable of yielding near 100% efficiency in converting sunlight to potentially viable energy. Since many of the key photosynthetic processes occur extremely fast, limited information can be obtained experimentally. This makes computational studies especially useful and motivates our development of accurate and reliable mathematical models based on quantum and classical physics to simulate the sub-nanometer and femtosecond chemistry. In collaboration with Dr. Young Min Rhee at Pohang University of Science and Technology in Korea, this research will combine two computational methodologies to develop an accurate representation of how sunlight is converted to chemical energy in naturally occurring photosynthetic complexes.The reaction center is a subunit of the Photosystem II pigment-protein complex where exciton to charge-transition manifold quantum transition occurs. Several groups have studied this complex and yet, it still remains unclear why some charge-transfer states are created while others are not. This study will focus on developing a unifying model for the reaction center. The protein environment can be treated as classical point charges for the exciton manifold. However, for the charge-separated manifold, the charged chromophores will significantly polarize the local environment and a static charge description will not be sufficient. Thus, this study will build a polarizable representation of the protein for the QM/MM and MD calculations. The overall energy landscape will be implemented into quantum dynamics algorithm to simulate energy transfer and conversion. This NSF EAPSI award is funded in collaboration with the National Research Foundation of Korea.