24小时热门版块排行榜

返回列表

当前主题已经存档。

tutu2000

金虫 (正式写手)

应助: 0 (幼儿园)
金币: 499.8
帖子: 351
在线: 95.2小时
虫号: 345082
注册: 2007-04-14
专业: 物理有机化学

[交流] 【讨论】什么是第一性原理计算(What are first principles calculations?)

最近总有朋友问我这个问题，我对他们说就是基于量子力学的量子化学电子结构计算，但他们总是不満足这个回答，因为他们对量子化学所知较少，我就在网上搜了下，有个较详细的说法如下，不知道大家对这个有没有更易理解的回答。

1。What are first principles calculations?

A material is simply a collection of atoms that are bound by chemical reactions. Chemical reactions, in turn, are simply interactions between electrons. These interactions are described by the laws of quantum physics. This means that all material properties (chemical, mechanical, electrical, magnetic, optical, thermal,…) can, in principle, be predicted from nothing more than the atomic number and mass of the atomic species involved, with the aid of quantum physics. This is precisely what first principles calculations attempt to do.

One of the joys of first principles calculations is that a few atomic numbers make abstract quantum concepts come to life in the form of quantitatively accurate, experimentally verifiable predictions - for quantities ranging from the Young modulus of diamond to the absorption spectra of conjugated polymers.

2。How do we do first principles calculations?

Solving the Schrödinger equation by brute mathematical force is extremely demanding computationally and not practical for all but tiny systems. Instead, we use a combination of two physical approximations:

1. We use density functional theory (DFT), for which Prof. Walter Kohn was awarded the 1998 Nobel prize in chemistry. DFT maps the original many-electron problem into an equivalent single-electron problem. It does so by lumping all the many-body quantum phenomena (such as Pauli’s exclusion principle and electron correlation) into a single additive “exchange-correlation” potential, which is a functional of the charge density alone. In principle, this mapping is exact. In practice, the exact functional is unknown and people use approximate forms for the functional, usually (but not always) derived from properties of a uniform electron gas.

2. We use first principles pseudopotential theory. The periodic table tells us that chemical reactivity is governed by valence electrons, with core electrons being chemically inert. Pseudopotentials make use of this basic fact by replacing the inert core electrons with an effective potential. This reduces, sometimes drastically, the number of electrons we need to solve for. Even more importantly, this results in much smoother wave functions for the remaining valence electrons, making the problem much easier to solve numerically.

For the calculation of excited state properties (notably optical ones), we have to move beyond DFT, which is a ground state theory. Our favorite tool is time-dependent DFT –an extension of the original method.

Doesn’t the use of physical approximations mean that we’re not really doing first principles work? No! Our approximations are systematic. All “hidden parameters” within them are determined objectively from theoretical results for properties of the electron gas (DFT) or the isolated atom (pseudopotentials), and not by fitting experiments.

[ Last edited by csfn on 2008-1-31 at 10:28 ]

回复此楼