Ab initio calculation of the dynamical properties of PPP and PPV

合集下载

从头计算分子动力学方法及其应用

从头计算分子动力学方法及其应用

收稿日期:2004-12-21基金项目:山东省自然科学基金资助项目(Y2003A01)和石油科技中青年创新基金(04E7038)作者简介:蓝建慧(1979-),女(汉族),山东即墨人,硕士研究生,专业方向为计算物理。

文章编号:100025870(2005)0420143204综述从头计算分子动力学方法及其应用蓝建慧,卢贵武,黄乔松,李英峰,朱 阁(中国石油大学物理科学与技术学院,山东东营257061)摘要:从头计算分子动力学方法把密度泛函理论和分子动力学方法有机地结合起来,使电子的极化效应及化学键的本质均可用计算机分子模拟方法进行研究,是目前计算机模拟实验中最先进、最重要的方法之一。

文章简述了从头计算分子动力学方法的基本原理,介绍了该方法在水、水溶液及其他氢键液体的结构与动力学研究中的应用。

关键词:从头计算;密度泛函理论;分子动力学;计算机分子模拟中图分类号:O 35 文献标识码:AMethod of ab initio molecular dynamics and its applicationsLAN Jian 2hui ,L U Gui 2wu ,HUAN G Qiao 2song ,L I Y ing 2feng ,ZHU G e(College of Physics Science and Technology in China U niversity of Pet roleum ,Dongying 257061,China )Abstract :The ab initio molecular dynamics method ,which combines the density functional theory with the molecular dy 2namics methodology ,made it convenient to study the electronic polarization effects and the nature of the chemical bonds in term of the computer molecular simulation.The method is one of the most im portant and advanced com puter simulation ex 2periment methods.The basic principle of the ab initio molecular dynamics method and its applications in structure and dy 2namics research of liquid water ,aqueous solutions and other hydrogen 2bond liquids were introduced.K ey w ords :ab initio ;density functional theory ;molecular dynamics ;computer molecular simulation 现代凝聚态理论研究应用最普遍的方法之一是分子动力学(MD )方法。

化学键【精品】

化学键【精品】

Inorganic Chemistry
43(2004)
This paper makes an important contribution to the still sparse literature on the correlation of the topology of the energy along a reaction path and the associated changes in structure determined by the topology of the electron density. --- R.F.W. Bader
键鞍点的特性
∇2ρ(r) at the BCP
Ionic bonding:
bcp
∇2ρ = u + u + 1 2
u 3 2 ∇ ρ〉0
Covalent bonding:
∇2ρ =
u + u + u3 1 2
∇ ρ〈0
2
键的椭圆度
键的弯曲度
环鞍点 (rcp)
笼鞍点(ccp)
4. 零流面
Gradient Path and Gradient field
水分子孤对电子伸展方位
(6)反应过程
计算量子化学研究室
量子拓扑分析理论方法的研究拓展
84年以来,先后在国际、国内重要理论化学 年以来,先后在国际、 年以来 刊物发表有关研究论文近百篇, 刊物发表有关研究论文近百篇,在国际上取得 了一定的影响,先后得到多项国家、 了一定的影响,先后得到多项国家、省自然科 学基金资助。 学基金资助。
化学键的新概念
R.F.W. Bader “Atoms in molecules: A quantum theory”.

δ-Pu态密度的动力学平均场理论研究

δ-Pu态密度的动力学平均场理论研究

δ-Pu态密度的动力学平均场理论研究刘以良;肖培【摘要】利用局域密度近似结合动力学平均场理论研究了δ-Pu的态密度. 借助赫伯德模型下的多体哈密顿量, 并使用赫伯德 I方法进行杂质求解. 计算结果很好的呈现出了强关联电子体系的低赫伯德带以及高赫伯德带, 并可以呈现出费米能级处的准粒子Kondo共振峰.【期刊名称】《西南民族大学学报(自然科学版)》【年(卷),期】2010(036)006【总页数】4页(P1014-1017)【关键词】态密度;赫伯德模型;赫伯德;I近似【作者】刘以良;肖培【作者单位】西南民族大学电气信息工程学院,四川成都,610041;西南民族大学电气信息工程学院,四川成都,610041【正文语种】中文【中图分类】O56钚是一种剧毒的、人造的强放射性锕系元素, 在能源、军事、航天等领域有极为广泛的应用.其在固态下与元素周期表中其他元素相比, 显示出复杂而反常的属性:钚有6种晶体结构, 低温下(273-373 K)以α相的形式存在, α相晶胞中有16个原子, 随着温度的升高, 会经历一系列的相过渡, 最终形成结构相对简单的面心立方δ相(573 –723 K)和体心立方ε相(~773 – 923 K), 温度再高就会融化; 晶体中钚原子的体积与温度的关系反常, 在从α 向δ相的过渡中原子体积扩展明显, 大约增加了25%, 而在δ相内部呈现一热膨胀的负增长, 并且从α相向更高温度的ε相过渡中原子体积大约减少了5%; 反常的电磁行为, 与其他的重费米体系类似, 钚呈现反常的电阻行为, 但是它的6个相都是没有磁性的, 磁化系数非常的小, 并且相对而言与温度无关; 反常的光谱特征, 从钚的光子发射谱可以看到在费米能级处有一个较强且窄的类Kondo峰存在, 这个现象与钚具有较大的线性比热系数一致[1-2].这些奇特的性质主要是由Pu的5f电子引起的, 5f电子同时具有定域和离域的特征.这种情况下, 电子处于强关联状态, 尽管传统的密度泛函理论(DFT)基础上的第一性原理可以成功计算很多实际材料的电子结构, 而对于强关联电子体系,例如存在未满壳层的d电子或者f电子的体系, 电子运动受到的限制明显, 轨道较窄, 电子之间的库仑相互作用较大与其带宽数值相差不大, 甚至超过电子的动能.DFT/LDA得出的结果将不再可靠, 这是因为局域密度近似(LDA)建立在弱关联电子模型基础上, 对于强关联电子体系, 电子空间密度变化剧烈, 电子局域密度是一个常数的假设不再成立, 并且电子之间相互作用明显, 单电子不能再被看成在一个静态的平均场中运动, 因此DFT通常只能处理非定域的弱关联的电子体系.例如DFT会将绝缘体CoO和La2CuO错误的计算成为导体.鉴于强关联电子体系的重要性, 其处理仍是是当今凝聚态物理发展的一个重要方向, 找到一个就像DFT解决弱关联体系一样成功的解决强关联电子体系的第一性原理方法就显的尤为重要.典型的强关联电子体系第一性原理解决方法有LDA+U、LDA++以及Local GW等的方法[3-7], 然而这些方法相对比较粗糙.作为20世纪90年代发展起来的一种非微扰多体技术, 动力学平均场理论DMFT[8-9]及其团簇扩展CDMFT[10-11]可以同时考虑电子的能带特性和类原子特性, 为研究强关联电子体系提供了新的有效途径.尤其是将DFT/LDA与DMFT结合起来,用DFT/LDA 处理模型哈密顿量的若关联部分, 用DMFT处理体系的强关联部分的LDA+DMFT方法正逐渐被越来越多的科研人员认同并发展起来[12-13].LDA+DMFT方法已经被成功的用来解决一些过渡金属氧化物, 磁过渡金属以及Ce和Pu等稀土金属的光谱、传输以及热力学等的性质[14-17].本文首先简要介绍了DMFT方法的本质、物理思想、基本条件以及对应的哈密顿量的形式.然后在653K下,利用赫伯德模型和LDA+DMFT方法计算了δ-Pu的态密度(DOS), 其中使用了赫伯德 I近似解决杂质问题, 最后将计算结果与其他的理论以及实验结果进行了比较.DMFT的本质就是用一个单位量子杂质模型替代原来的晶格模型, 这个单位量子杂质模型镶嵌在一个自洽的有效媒介中, 并且自洽条件满足平移不变性和连贯性效应, 其根本的物理思想就是对于某个确定的晶格位,其动力学可以看成是位上的自由度与一个外部“浴室”的相互作用, 这个“浴室”是由给定晶格位周围其它位所有自由度产生的.具体而言, 就是除了某个特定晶格位外, 其他的晶格位上的库仑相互作用都用自能来代替, 而这个特定位上的电子之间有相互作用并可以在整个晶格中运动, 但是电子在其他的位上传播是通过自能产生的媒介而不是通过电子间的相互作用, DMFT引入的“单位问题”与安德森杂质问题是等价的, 而且安德森杂质问题需要通过自能与格林函数的关系(Dyson方程)自洽求解.杂质模型为量子多体问题的局域动力学提供了直观的图像, 鉴于杂质问题迄今已经历四十几年的发展并得到一系列可用的解法[5][8][12][17-20], 杂质模型问题成为 DMFT方法中非常重要的一环.空间的纬数越大, DMFT方法就越精确, 近似说来, 也就是晶格的配位数越高计算结果就越精确,达到无限纬极限时最精确[8-9][13], 所以DMFT方法迄今难以进行表面电子态的计算.对于具有强关联f电子的体系, DMFT计算是将哈密顿量进行如下的分解[17]:多带周期安德森模型的非相互作用部分描述库仑相互作用的局域贡献, 用一个关联项来补充上面的公式如下:为了避免双重计算, 需要将 LDA 哈密顿量中的库仑关联减去.LDA 哈密顿量中的库仑关联可以近似用库仑相互作用能的平均值给出.新的非相互作用哈密顿量如下:右边第一项可以用LDA方法计算, 右边第二项可以用DMFT方法计算.如上所示, 关联电子的哈密顿量虽然可以精确写出, 但是非常复杂无法精确求解, 甚至高于 10个晶格位时已经不能够得出数值解, 因此需要用到近似的方法, 这里DMFT就是一种强有力的考虑了电子关联的近似方法, 在DMFT中, 库仑关联的效应由局域近似的自能算符来表示, 自能Σ(iωn)和格林函数G(iωn)在虚 Matsubara频率iωn=iπ(2n+1)/ β 下得到.如果要计算光谱性质就需要实轴上的格林函数.为了进行LDA+DMFT计算, 我们用密度泛函理论的局域密度近似结合量子杂质求解模型完成一个完整的LDA+DMFT循环.首先DFT Code 用以产生δ-Pu在k空间哈密顿量矩阵, δ-Pu为面心立方结构, 在653K下的实验晶格参数为8.759 a.u.[21], 使用LDA作为交换关联势, 电子的波函数基矢使用线性响应muffin-tin 轨道(LMTO)和原子球近似(ASA), 鉴于δ-Pu处于顺磁性态, 实际自旋的数目为1, 并且在未考虑S-O耦合的情况下, 在6×6×6四面体网格分割的k空间, 哈密顿量在spdf的LMTO下写成16×16的矩阵.并得到满足7s6p6d5f上的总电子数为14的化学势为8.875 eV.其次, 使用了赫伯德-I近似来解决杂质问题, 赫伯德-I方法相对比较近似, 但是使用的计算时间比较短, 更加趋于反应Pu的原子特性, 因此其得到的结果可以为其他的杂质求解器提供参考, 取库仑相互作用的平均值U=F0=4.0 eV, 杂质能级为9.025 eV, 温度为653 K(0.0562 eV), 考虑自旋量子数, f 轨道的兼并度取14, 结合解析四面体方法求解格林函数, 并最终使用Pade近似将Matsubara频率下的格林函数和自能转化到实频下.得出δ-Pu的DOS,如图1所示.计算的结果与文献[16]吻合的很好, 并且呈现出强关联电子体系的一个典型的特征,即低赫伯德带以及高赫伯德带的存在, 将费米能级处态密度与实验光电谱进行比较, 如图 2所示.计算结果与实验测得的光电谱[22]有一定的吻合, 可以看到在费米能级处一个较强的准粒子Kondo共振峰的存在.然而由于赫伯德 I方法比较适宜计算顺磁Mott绝缘子的光电谱, 因此计算中并未得到费米能级处的连贯峰.由于DMFT方法的相对不成熟性, 以及其不是完整意义上的ab initio计算方法,需要半经验的参数, 使得各种计算结果以及与实验的吻合度有所差异[2][14-16][22].利用局域密度近似结合动力学平均场理论研究了653K下δ-Pu的态密度, 使用赫伯德模型下的多体哈密顿量,以及使用赫伯德 I近似进行杂质模型求解.计算结果很好的呈现出了强关联电子体系的低赫伯德带以及高赫伯德带, 并可以看到在费米能级处较强的准粒子 Kondo共振峰的存在.然而由于赫伯德 I方法不够精确, 并且更适宜计算顺磁Mott绝缘子的光电谱, 所以要想得到更加符合实验结果的光电谱就需要量子Monte Carlo(QMC)方法、精确对角化方法(ED)等相对精确的杂质解决器, 而且需要进一步考虑 f轨道的电子占有数、经过 DMFT循环后电荷密度改变对哈密顿量的影响、平均库仑相互作用、能级的兼并度、自旋轨道耦合、强关联指数等因素对计算结果的影响.【相关文献】[1] SAVRASOV S Y.Spectral density functionals for electronic structure calculations[J].Phys Rev B, 2004, 69:245101-245105.[2] ZHU J X, MCMAHAN A K, JONES M D, et al.Spectral properties of δ-plutonium: Sensitivity to 5f occupancy[J].Phys Rev B, 2007,76:145118-145123.[3] SAVRASOV S Y, KOTLIAR G.Ground State Theory of δ-Pu[J].Phys Rev L, 2000, 84: 3670-3675.[4] BOUCHET J, SIBERCHICOT B, JOLLET F, et al.Equilibrium properties of δ-Pu: LDA + U calculations (LDA ≡ local density approximation)[J].J Phys: Condens Matter, 2000, 12: 1723-1727.[5] ANISIMOVY V I, ARYASETIAWANZ F, LICHTENSTEIN A I.First-principles calculations ofthe electronic structure and spectra of strongly correlated systems: the LDA +Umethod[J].J Phys Condens Matter, 1997,9: 767-772.[6] LICHTENSTEINA I, KATSNELSON M I.Ab initio calculations of quasiparticle band structure in correlated systems: LDA++approach[J].Phys Rev B, 1998, 57: 6884-6889. [7] ZEIN N E, ANTROPOV V P.Self-Consistent Green Function Approach for Calculation of electronic Structure in Transition MetalS[J].Phys Rev L, 2002, 89: 126402-126407.[8] GEORGES N, KOTLIAR G, KRARTH W, et al.Dynamical mean-field theory of strongly correlated fermion systems and the limit of infinite dimensions[J].Rev Mod Phys, 1996, 68: 13-19.[9] KOTLIAR G, SAVRASOV S Y, HAULE K, et al.Electronic structure calculations with dynamical mean-field theory[J].Rev Mod Phys, 2006, 78: 865-871.[10] MAIER T, JARRELL M, HETTLER M H.Quantum cluster theory[J].Rev Mod Phys, 2005, 77: 1027-1033.[11] TREMBLAY A-M S, KYUNG B, SÉNÉCHAL D.Pseudogap and high-temperature superconductivity from weak to strong coupling.Towards a quantitative theory (Review Article) [J].Low Temperature Physics, 2006, 32: 424-428.[12] HELD K, NEKRASOV I A, KELLER G, et al.The LDA+DMFT Approach to Materials Strong Electronic Correlations, Quantum Simulations of Complex Many-Body Systems: From Theory to Algorithms[J].Lecture Notes, NIC Series, 2002(10): 175-181.[13] HELD K, ANDERSEN O K, FELDBACHERM, YAMASAKI A, et al.Bandstructure meets many-body theory: the LDA + DMFT method[J].J Phys: Condens.Matter, 2008, 20: 064202-064207.[14] SAVRASOV S Y, KOTLIAR G, ABRAHAMS E.Correlated electrons in δ-plutonium withina dynamicalmean-field picture[J].Nature, 2001, 410: 793-797.[15] SHIM J H, HAULE K, KOTLIAR G.Fluctuating valence in a correlated solid and the anomalous properties of δ-plutonium[J].Nature, 2007, 446: 513-519.[16] ZHU J X, JONES M D.Electronic Structure Calculations with Dynamical Mean Field Theory in δ-Pu[M].CONDENSED MATTER,MATERIALS SCIENCE, 2005.[17] NEKRASOV I A, HELD K.Calculation of photoemission spectra of the doped Mott insulator La1-xSrxTiO3 using LDA+DMFT(QMC) [J].Eur Phys J B, 2000, 18: 55-61.[18] ZOLFL M B, PRUSCHKE bining density-functional and dynamical-mean-field theory for L a1ÀxSrxTiO3[J].Phys Rev B,2000, 61: 12810-12815.[19] Haule k.Quantum Monte Carlo impurity solver for cluster dynamical mean-field theory and electronic structure calculations with adjustable cluster base[J].Phys Rev B, 2007, 75: 155113-155117.[20] WERNER P, COMANAC A, MEDICI L D, et al.Continuous-Time Solver for Quantum Impurity Models[J].Phys Rev B, 2006, 97:076405-076409.[21] ARKO A J, JOYCE J J, MORALES L, et al.Electronic structure of α- and δ-Pu from photoelectron spectroscopy[J].Phys Rev B 2000, 62: 1773-1778.。

张紫薇李增增英语文献原版

张紫薇李增增英语文献原版

Electrochimica Acta 108 (2013) 867–875Contents lists available at ScienceDirectElectrochimicaActaj o u r n a l h o m e p a g e :w w w.e l s e v i e r.c o m /l o c a t e /e l e c t a c taDiffusion mechanism of lithium ions in LiNi 0.5Mn 1.5O 4H.Seyyedhosseinzadeh,F.Mahboubi ∗,A.AzadmehrDepartment of Mining and Metallurgical Engineering,Amirkabir University of Technology,Hafez Avenue,P.O.Box 15875-4413,Tehran,Irana r t i c l ei n f oArticle history:Received 8March 2013Received in revised form 1July 2013Accepted 2July 2013Available online 16 July 2013Keywords:LiNi 0.5Mn 1.5O 4Diffusion mechanism Diffusion constant Activation energy Li-ion batterya b s t r a c tLiNi 0.5Mn 1.5O 4is suitable for electrochemical applications as an active material.This material has capa-bility for high rate application in Li-ion battery.In this research,the diffusion mechanism of Li ions in LiNi 0.5Mn 1.5O 4was studied by chronoamperometry technique and computer simulation (ab ini-tio and Fickian approach).According to the results,two bulk diffusion constants were calculated for LiNi 0.5Mn 1.5O 4during intercalation or deintercalation.At low concentration of Li ions in LiNi 0.5Mn 1.5O 4,the diffusion constant was about 10−9cm 2/s and for other Li ions concentration,it was about 10−11cm 2/s.In addition,the diffusion constant at the surface of LiNi 0.5Mn 1.5O 4was estimated 10−8cm 2/s by a semi empirical model.So,the diffusion constant of Li ions in LiNi 0.5Mn 1.5O 4has been estimated in the range of 10−8to 10−11cm 2/s.© 2013 Elsevier Ltd. All rights reserved.1.IntroductionThere are several crystal structures which can be used as active materials in Li-ion battery,yered,spinel and olivine like frame-works.The LiNi 0.5Mn 1.5O 4is spinel one and it is a suitable candidate for cathode of Li-ion battery.The LiNi 0.5Mn 1.5O 4spinel shows two different space groups (Fd 3m or P 4332)depending on Ni ordering in the lattice [1].It was confirmed by the X-ray diffraction (XRD)data that face-centered spinel (Fd 3m )transforms into primitive simple cubic (P 4332)structure by annealing process at 700◦C [2].The dif-ference of these two space groups is in distribution of Mn and Ni atoms in crystal structure of LiNi 0.5Mn 1.5O 4.The LiNi 0.5Mn 1.5O 4has a disordered distribution (Fd 3m ),where Li ions occupy the tetrahe-dral (8a)sites,Mn or Ni ions randomly occupy the octahedral (16d)sites and O ions are located at (32e)sites.In the ordered P 4332phase,Mn and Ni ions order on octahedral sites in a 3:1ratio (i.e.Ni atoms in 4b sites,Mn atoms in 12d sites,and O atoms in 8c and 24e sites according to the notations of Wyckoff position)[3–5].It was reported (by first principle calculation)that there is no intermediate phases between LiNi 0.5Mn 1.5O 4and Li 0Ni 0.5Mn 1.5O 4[6].It was also proved that LiMn 2O 4has better structural stability by Ni alloying due to the lower bond length of Mn O in the pres-ence of Ni.The Mn O and Mn(Ni)O bond length is 1.823and 1.800,respectively (by Raman scattering)[7].Rigid Spinel is better to pre-vent the insertion of the foreigner species instead of Li ions into it from electrolyte [8].In addition,it would be important to specify∗Corresponding author.Tel.:+982164542967;fax:+982166405846.E-mail address:Mahboubi@aut.ac.ir (F.Mahboubi).how the presence of Ni atoms would affect the overall movement of Li ions through the lattice of LiNi 0.5Mn 1.5O 4by affecting activation energies of diffusion.Quantum mechanics first principle calculation (ab initio)is a reliable computational method to calculate the activation energy for atomic diffusion.This computational method is successfully working for cutting edge technology/science [9],i.e.to design active materials for Li-ion batteries [10,11].In addition,other com-putational approaches (at mesoscale and marcoscale)are mostly relevant to assess how physical phenomena take place during de/intercalation of ions in Li-ions batteries.In general,the calcula-tion of flux (removal and insertion)of Li ions in active materials of Li-ion battery needs to consider different (by orders of magnitude)physical scales [12],e.g.more refine length/time scale at interface of active material and electrolyte in comparison to the bulk diffusion [13].It is not inequitable to claim that atomistic quantum based (ab initio)approach is more efficient to evaluate governing phys-ical mechanisms than the other computational methods at higher length scales.But atomistic calculations seem more difficult than other methods and they are also limited to small systems.Never-theless,performing simulation at any desired length scale would be helpful to predict and discover governing physical mechanisms.There are several coarse grain approaches based on the thermody-namic principles with attention to the de/intercalation phenomena [13–16].In addition,there are extensive ab initio quantum based models (atomistic approach)by considering a small part of Li-ion battery,i.e.active material or electrolyte [17–22].For LiNi 0.5Mn 1.5O 4,there are several published researches on total energy calculation,activation energy/barrier and diffusion coefficient of Li ions by computational and experimental methods0013-4686/$–see front matter © 2013 Elsevier Ltd. All rights reserved./10.1016/j.electacta.2013.07.034868H.Seyyedhosseinzadeh et al./Electrochimica Acta 108 (2013) 867–875Fig.1.The used algorithm of ab initio calculation for estimation of activation energy of Li ions for diffusion in the bulk of Li 8Ni 4Mn 12O 32.[6,23–26].Although there are some expressions on activation ener-gies of diffusion for accessible directions in LiNi 0.5Mn 1.5O 4,but it is still quite limited.In previous researches on activation energies of Li ions in LiNi 0.5Mn 1.5O 4,the diffusion direction with lowest acti-vation energy has been considered as the dominated one without considering the zigzag path of diffusion through octahedral and tetrahedral sites during diffusion.In this research,the diffusion mechanism of Li ions in LiNi 0.5Mn 1.5O 4was studied by chronoam-perometry and simulation (ab initio and Fickian approach)for more comprehensive description of diffusion of Li ions in LiNi 0.5Mn 1.5O 4.2.Experimental methods2.1.Fabrication of LiNi 0.5Mn 1.5O 4LiNi 0.5Mn 1.5O 4powders were synthesized via solid-state method.Appropriate amounts of MnO 2,Ni 2O 3and Li 2CO 3powders were mixed with ball milling for 10h.This planetary ball milling facility was used to activate powders.Subsequently,the mixedprecursor powders were calcinated in electric furnace at 800◦C for 5h and were cooled in air.2.2.Characterization of active materialX-ray diffraction (XRD)using CuK ␣radiation with wave length 0.154nm,Fourier Transform Infra-Red spectroscopy (FTIR)and scanning electron microscope (SEM)were used to characterize and assess the synthesized active material.Autolab PGSTAT30was used to measure the current versus time at constant voltage (chronoam-perometry)in active materials during deintercalation/charging (a two-electrode setup was used).The pure graphite was used as the reference electrode and cathode was connected to the working electrode.2.3.Assembly of batteryTo study the Li ions transportation in active material experi-mentally,0.06g of active material (LiNi 0.5Mn 1.5O 4)in powderformFig.2.Framework of multiscale model based on Fickian approach.H.Seyyedhosseinzadeh et al./Electrochimica Acta108 (2013) 867–875869Fig.3.The algorithm of multiscale model based on Fickian approach.without conductive material(i.e.carbon black)was poured in a small Al cap(current collector)as a cathode.And pure graphite was used as the anode.1M LiClO4in EC/DMC(1:1)was used as the electrolyte.The assembly of this cell was performed under Ar controlled atmosphere in a container.3.Frameworks of modelsIt was actually two separate models,first,ab initio model on cal-culating activation energy for Li ions during diffusion through bulk of LiNi0.5Mn1.5O4(and subsequent description of diffusion constant870H.Seyyedhosseinzadeh et al./Electrochimica Acta 108 (2013) 867–875Fig.4.An illustration of stochastic boundary condition at the interface of active material andelectrolyte.Fig.5.SEM images of LiNi 0.5Mn 1.5O 4.Fig.6.FTIR (a)and XRD (b)analysis of synthesized LiNi 0.5Mn 1.5O 4powders.by Arrhenius equation)and second,semi empirical model (multi-scale model:Fickian approach coupled with atomistic stochastic approach)for estimating the diffusion constant of Li ions at the surface of this activematerial.Fig.7.Hartree–Fock pseudopotential of atoms [29].3.1.Ab initio approachThis model was run in real physical space.Like any numerical methods,physical space must be discretized into some grids.The model used a cube with 0.81nm in each side as a domain of simula-tion.This domain was discretized by eight millions grids as meshes.The size of this cube is in accordance to the size of a supper cell of Li 8Ni 4Mn 12O 32.The total energy of an atomic system is composed of the kinetic energy of electrons,exchange–correlation energy (electrons quantum interaction),ion–ion interaction (Madelung energy),electron–electron coulomb interaction and pseudopoten-tial (valance electrons and core’s ions interaction plus the effect of inner electrons).Hartree–Fock pseudopotential [27]was used for capturing the effects of inner electrons and core’s ions poten-tial on valance electron cloud.In addition,local density approach (LDA)has been used to calculate the exchange–correlation energy.Despite the many successes of LDA theory,it is often claimed that this method is useless for strongly correlated materials (electron)[28].GGA +U method is proposed for strongly correlated materials [29,30].But the use of LDA or GGA has no effect on final results for calculating the activation energy [31].Finally,finite difference time domain was used to solve the Schrodinger equation [32].The overall strategy of present model has been shown in Fig.1.3.2.Fickian approachA single microscale particle of active material was considered in contact with electrolyte.The shape of the particle was estimatedH.Seyyedhosseinzadeh et al./Electrochimica Acta108 (2013) 867–875871Fig.8.Wave function for2s orbital. from SEM image of powders.To perform/run simulation,wholemodel must be discretized.The discreted model has106numbersof regular cubic meshes(with200nm length in side of each cubicmesh).Fig.2shows the framework of the model and its meshes.Thereby,there is the same mesh size for governing equations ofLi ionsflux both in electrolyte and active material;that is to say,there would be a difference between their time scales according tothe numerical stability condition.Instead of a proposed stochasticboundary condition,other used boundaries are common periodicboundary condition.The general algorithm of the model has beenshown in Fig.3.The proposed stochastic boundary condition workswith atoms,that is,this is an atomic approach.So,it has a new timescale different from the two time scales of atomic diffusion in activematerial and electrolyte.Tofind out an approximate time scale inused stochastic concept,following formula(Eq.(1))was used:D interface=1W ˇ2 →Stochastic time scale=ˇ2WD interface(1)Fig.9.Exchange–correlation energy for wave function in Fig.8.Fig.4shows an illustration of the used stochastic boundarycondition.ˇand W are related to the atomic configuration inLiNi0.5Mn1.5O4crystal.By having parameters in Eq.(1),D interface(the diffusion coefficient at the interface of active materialand Fig.10.Atomic distribution in Li7Ni4Mn16O32superlattice along with calculated pseudowave function in gray scale.The distance of the two adjacent slices is0.1nm.872H.Seyyedhosseinzadeh et al./Electrochimica Acta108 (2013) 867–875Fig.11.Spatial form of movement of lithium ion in three different paths. electrolyte)can be calculated.As the summation,the model pre-dicts the passing current by time evolution during charging.By regulating the calculated data by experimental results on charging current,stochastic time step can be determined and subsequently the introduced diffusion coefficient at the interface of active mate-rial and electrolyte would be specified.4.Results and discussionFig.5shows the SEM image of synthesized LiNi0.5Mn1.5O4parti-cles.The FTIR and XRD analysis of LiNi0.5Mn1.5O2have been shown in Fig.6.XRD and FTIR results confirmed the spinel LiNi0.5Mn1.5O4. Positions of Peaks in XRD and FTIR results are well adapted to the LiNi0.5Mn1.5O4spinel.The LiNi0.5Mn1.5O4spinel,depending on Niordering in the lattice,shows two different space groups(Fd3mFig.12.Calculated activation energies for three paths which have been shown in Fig.11.Fig.13.The chronoamperometry graph(current–time−0.5)of charging or during deintercalation of Li ions from LiNi0.5Mn1.5O4structure.or P4332).Although it is hard to detect these two space groups from each other by conventional XRD method,it was confirmed by the X-ray diffraction(XRD)data that face-centered spinel(Fd3m) transforms into primitive simple cubic(P4332)structure by anneal-ing process at700◦C[2].Hence,the simple cubic(P4332)structure was predicted for the space group of synthesized LiNi0.5Mn1.5O4 powders according to the used calcination temperature at800◦C in this research.In addition,by accepting the simple cubic(P4332) structure for synthesized LiNi0.5Mn1.5O4powders,some noisy like XRD patterns could be appointed as peaks such as(210),(211) and(220)peaks.Finally,this synthesized LiNi0.5Mn1.5O4powders were used for electrochemical tests as an active material in a cell. Along with electrochemical tests,two models were used forfinding out the diffusion mechanisms of Li ions in LiNi0.5Mn1.5O4.First model wasfirst principle calculation(ab initio model).This ab initio model uses a simple method to calculate the total energy of systems in real atomic space.For each atom(Li,O,Mn and Ni),a spherical symmetric pseudopotential was applied.Final potential is the summation of pseudopotential of all atoms in the model.Fig.7 shows the pseudopotential curve for atoms in Atomic Rydberg Unit (ARU/Ry).The same approach was used to define exchange–correlation energy.This term of energy is a function of local wave function, that is,this function is updated for every time step by new cal-culated wave function during simulation.Fig.8shows the wave function,which was defined on guesswork.This is a wave func-tion for2s orbital of Li atom.The wave function in Fig.8was used to draw the curve in Fig.9,which is shown the development of exchange–correlation potential in outward direction from the core of Liatom.Fig.14.Spiral diffusion path of Li ion though path3.H.Seyyedhosseinzadeh et al./Electrochimica Acta108 (2013) 867–875873In this model,the calculated wave function is pseudowave func-tion and it does not reflect the actual wave function of atoms.But the result is reliable on total energy.Fig.10shows the atomic arrangements in Li7Ni4Mn12O32superlattice with P4332crystal structure.According to the atomic arrangement in Li7Ni4Mn12O32 superlattice with P4332crystal structure,three directions or paths were checked for diffusing Li ions via them.These mentioned three paths have been spatially shown in Fig.11.Paths1and3have also shown in Fig.10by marking*and**.The difference between these two paths(1and3)is the connection of the nearest Ni ion to the Li ion when this Li ion is in the octahedral site during diffusion (Fig.11).In Fig.10,the movement of starred Li ion from slice6to the starred position in slice5and then to the starred vacancy posi-tion in slice4is related to the diffusion of Li ion along path1.And in thisfigure,the movement of double starred Li ion in slice2to the double starred position in slice3and then to the starred vacancy position in slice4is related to the diffusion of Li ion along path3.The activation energies have been calculated for these three mentioned paths by the proposed ab initio model.As it is obvi-ous,path2is not a proper candidate for diffusion of Li ion in Li7Mn16O32superlattice due to very high activation energy(Fig.12). Hence,the paths1and3are the most possible diffusion directions in Li7Ni4Mn12O32framework with24.18kJ/mol and39.14kJ/mol activation energies(Fig.12).So,the two possible passes for Li ion diffusion in LiNi0.5Mn1.5O4 have been confirmed by ab initio calculation.In ref.[24],it has been explained that10−9cm2/s diffusion constant would be achiev-able at low concentration of Li ions in Li X Ni0.5Mn1.5O4.It means that the diffusion mechanism of Li ions in LiNi0.5Mn1.5O4would Fig.15.Electrical potential around active material calculated from Poisson’s equa-tion.be changed during de/intercalation.This subject has been checked by chronoamperometry(Fig.13).As it was obvious,the graph (current–time−0.5;time is in second)of charging has different slope during deintercalation of Li ions from LiNi0.5Mn1.5O4structure.The overall diffusion mechanism of Li ion in this material would be described by considering the graph(Fig.13)in two separate regions. In region1,two mechanisms are controlling the produced current of the Li-ion cell(charges at the surface of active materials and dein-tercalation of Li ion by diffusion).As the curve in region2is more linear than region1,it might be true to consider the dominated overall mechanism is controlled by diffusion of Li ions for producing electrons in this region.In addition,it is obvious that thediffusion Fig.16.Concentration of Li ions in electrolyte after two seconds of charging at4.5V.874H.Seyyedhosseinzadeh et al./Electrochimica Acta 108 (2013) 867–875Fig.17.Estimated interfacial diffusion coefficient.coefficient in region 2is higher than in region 1as the slope of curve in region 2is higher than the slope of curve in region 1.To explain the diffusion mechanisms of Li ions in bulk of LiNi 0.5Mn 1.5O 4related to these two regions in Fig.13,two simple models have been considered.For region 1,the concentration of Li ions in active materials is not low,consequently all atomic posi-tions for diffusion of Li ions through path 1(the lowest activation energy of diffusion)are not always accessible.So,Li ions would dif-fuse through path 3(the path with higher activation energy than path 1)sometimes,that is,Li ions would diffuse through both paths (1and 3)according to their accessibility.For describing the over-all diffusion mechanism of Li ion in active materials in region 1,LiNi 0.5Mn 1.5O 4has been considered as a composite of these two passes.According to equation 2,the diffusion of Li ions can be esti-mated in the superlattice of Li 7Ni 4Mn 12O 32.By inserting 1012for Debye frequency (ϑ)and 0.28nm for length of Li ions hopping (a)and w =4(w is the number of octahedral sites around a Li ion in the tetrahedral site),the diffusion coefficient of paths 1and 3will be 2.89×10−9cm 2/s and 1.99×10−12cm 2/s.D =ϑa 2wexp−Q RT(2)Overall diffusion constant would be achieved by considering the Li 7Ni 4Mn 12O 32as a composite of the two paths (1and 3)for Li ions diffusion.Volume fraction of the two paths is the same (1:1).So,D composite =0.5D path 1+0.5D path 3It is good to notice that this calculated diffusion constant is related to the hopping of a Li ion from a tetrahedral site to the other nearest tetrahedral one via an octahedral site,that is,we cannot report these results as the overall diffusion coefficient of Li ions in Li 7Ni 4Mn 12O 32lattice because of neglecting the zigzag path of Li ions in Li 7Ni 4Mn 12O 32during diffusion in region 1.So,it still needs more correction for calculating the overall diffusion coefficient ofLi 7Ni 4Mn 12O 32.As the successful diffusion needs two jumps (First:from a tetrahedral site to octahedral one –Second:from an octahe-dral site to tetrahedral one),so the overall possible vacant positions for jumping are 8×4(4is the number of octahedral sites around a tetrahedral site and 8is the number of tetrahedral sites around an octahedral site).So,the geometrical probability of a successful jump of a Li ion from a tetrahedral to the nearest same one is 1/(8×4),which must be multiply to the D composite for calculating the overall diffusion of Li ion in Li 7Ni 4Mn 12O 32lattice.By this final modifica-tion,the zigzag movement during diffusion is applied to calculate the overall diffusion coefficient of Li 7Ni 4Mn 12O 32.According to mentioned subject the overall diffusion constant was calculated 4.51×10−11cm 2/s.This calculated result is consistent with pub-lished experimental results [6,26].This composite model was for region 1and the diffusion path of Li ion would be zigzag in LiNi 0.5Mn 1.5O 4.But in region 2(the low concentration of Li in LiNi 0.5Mn 1.5O 4),Li ions prefer to diffuse through lowest activation energy direction because of availability more vacant site for hoping a Li ion.Hence,Li ion diffuses through path 3and the overall diffusion coefficient in this region would be at the level of 10−9cm 2/s which was calculated 2.89×10−9cm 2/s.The diffusion path in this case would not be zigzag like region 1.In region 2,if Li ions prefer to diffuse through just path 3,the overall diffusion path would be spiral in one direction (Fig.14).So,Li ion moves forward in a direction with spiral movement.In addition,the diffusion constant at the surface of this active material was also estimated by a multiscale model for understand-ing the overall diffusion mechanism of Li ions in LiNi 0.5Mn 1.5O 4.A multi-scale model was developed to predict transportation of Li ions with main objective on interface of active materials and electrolyte.The calculated potential and Li ions distribution in elec-trolyte according to the framework in Fig.2have been shown in Figs.15and 16.These calculated results are measured after two seconds of charging.Finally,the model predicted 7.6×10−8cm 2/sH.Seyyedhosseinzadeh et al./Electrochimica Acta108 (2013) 867–875875as the diffusion constant at the surface of LiNi0.5Mn1.5O4 (Fig.17).5.ConclusionAn electrochemical test(chronoamperometry)and computer simulations(ab initio and Fickian approach)were used to estimate the overall diffusion mechanism of Li ion in LiNi0.5Mn1.5O4.The highlighted features of the present research are as follows:1.In LiNi0.5Mn1.5O4spinel with P4332crystal structure,it was con-firmed that just two directions are the possible corridors for Li ions to diffuse.The activation energies for these possible paths were calculated24.18and39.14kJ/mol.2.According to the calculated activation energies for diffusion ofLi ions in the spinel of Li7Ni4Mn12O32and by using Arrhenius form of diffusion equation,the calculated diffusion coeffi-cient for the two mentioned paths were2.89×10−9cm2/s and1.99×10−12cm2/s.3.By chronoamperometry test(at different concentration of Li ionin active material),two different mechanisms were explained for diffusion of Li ions in bulk of LiNi0.5Mn1.5O4.i.Low concentration of Li ions in LiNi0.5Mn1.5O4:Li ions preferto diffuse through lowest activation energy direction(with 10−9cm2/s diffusion constant).The diffusion path in this case would be spiral.ii.High and intermediate concentration of Li ions in LiNi0.5Mn1.5O4:By considering Li7Ni4Mn12O32as a composite of the two paths for diffusion of Li ions(with2.89×10−9cm2/s and1.99×10−12cm2/s diffusion constants),the overall dif-fusion constant was calculated4.51×10−11cm2/s,which hasa good consistency with experimental results.The path ofdiffusion would be zigzag.4.The diffusion coefficient of Li ions for jumping through theinterface of LiNi0.5Mn1.5O4and electrolyte was estimated by a multiscale model.The estimated value for this diffusion coeffi-cient was7.6×10−8cm2/s.References[1]C.M.Julien,F.Gendron,A.Amdouni,M.Massot,Lattice vibrations of materialsfor lithium rechargeable batteries.VI:ordered spinels,Materials Science and Engineering B130(2006)41.[2]S.H.Park,S.-W.Oh,S.H.Kang,I.Belharouak,K.Amine,Y.-K.Sun,Compara-tive study of different crystallographic structure of LiNi0.5Mn1.5O4−␦cathodes with wide operation voltage(2.0–5.0V),Electrochimica Acta52(2007) 7226.[3]T.-F.Yi,Y.Xie,M.-F.Ye,L.-J.Jiang,R.-S.Zhu,Y.-R.Zhu,Recent developmentsin the doping of LiNi0.5Mn1.5O4cathode material for5V lithium-ion batteries, Ionics17(2011)383.[4]G.Q.Liu,L.Wen,Y.M.Liu,Spinel,LiNi0.5Mn1.5O4and its derivatives as cath-odes for high-voltage Li-ion batteries,Journal of Solid State Electrochemistry 14(2010)2191.[5]N.Amdouni,K.Zaghib,F.Gendron,A.Mauger,C.M.Julien,Structure and inser-tion properties of disordered and ordered LiNi0.5Mn1.5O4spinels prepared by wet chemistry,Ionics12(2006)117.[6]H.Xi,Y.S.Meng,L.Lu,G.Ceder,Electrochemical properties of nonstoichiomet-ric LiNi0.5Mn1.5O4−␦thin-film electrodes prepared by pulsed laser deposition, Journal of the Electrochemical Society154(2007)A737.[7]Y.Wei,K.-B.Kim,G.Chen,Evolution of the local structure and electrochemicalproperties of spinel LiNi X Mn2−X O4(0≤X≤0.5),Electrochimica Acta51(2006) 3365.[8]J.B.Goodenough,A.Manthiram,B.Wnetrzewski,Electrodes for lithium batter-ies,Journal of Power Sources4344(1993)269.[9]A.Jain,G.Hautier,C.J.Moore,S.P.Ong,C.C.Fischer,T.Mueller,K.A.Persson,G.Ceder,A high-throughput infrastructure for density functional theory calcula-tions,Computation Materials Science50(2011)2295.[10]S.P.Ong,V.L.Chevrier,G.Ceder,Comparison of small polaron migration andphase separation in olivine LiMnPO4and LiFePO4using hybrid density func-tional theory,Physical Review B83(075112)(2011)1.[11]G.Ceder,Opportunities and challenges forfirst-principles materials design andapplications to Li battery materials,MRS Bulletin35(2010)693.[12]X.Zhang,Multiscale Modeling of Li-ion Cells:Mechanics,Heat Generation andElectrochemical Kinetics,University of Michigan,2009(Ph.D.thesis).[13]A.M.Colclasure,K.A.Smith,R.J.Kee,Modeling detailed chemistry and transportfor solid-electrolyte-interface(SEI)films in Li-ion batteries,Electrochimica Acta 58(2011)33.[14]tz,J.Zausch,Thermodynamic consistent transport theory of Li-ion batter-ies,Journal of Power Sources196(2011)3296.[15]O.V.Bushkova,O.L.Andreev,N.N.Batalov,S.N.Shkerin,M.V.Kuznetsov,A.P.Tyutyunnik,O.V.Koryakova,E.H.Song,H.J.Chung,Chemical interactions in the cathode half-cell of lithium-ion batteries:Part I.Thermodynamic simulation, Journal of Power Sources157(2006)477.[16]H.Yokokawa,N.Sakai,K.Yamaji,T.Horita,M.Ishikawa,Thermodynamic deter-mining factors of the positive electrode potential of lithium batteries,Solid State Ionics113–115(1998)1.[17]A.Van der Ven,G.Ceder,First principles calculation of the interdiffusion coef-ficient in binary alloys,Physical Review Letters94(2005)1.[18]A.Van der Ven,G.Ceder,M.Asta,P.D.Tepesch,First-principles theory of ionicdiffusion with nondilute carriers,Physical Review B64(2001)1.[19]M.V.Koudriachova,N.M.Harrison,S.W.de Leeuw,First principles predictionsfor intercalation behaviour,Solid State Ionics175(2004)829.[20]A.Van der Ven,G.Ceder,Lithium diffusion in layered Li X CoO2,Electrochemicaland Solid-State Letters3(2000)301.[21]D.Morgan,A.Van der Ven,G.Ceder,Li conductivity in Li X MPO4M=(Mn,Fe,Co,Ni)olivine materials,Electrochemical and Solid-State Letters7(2004)A30.[22]M.K.Aydinol,A.F.Kohan,G.Ceder,K.Cho,J.Joannopoulos,Ab initio studyof lithium intercalation in metal oxides and metal dichalcogenides,Physical Review B56(1997)1354.[23]T.-F.Yi,Y.-R.Zhu,R.-S.Zhu,Density functional theory study of lithium inter-calation for5V LiNi0.5Mn1.5O4cathode materials,Solid State Ionics179(2008) 2132.[24]Ma Xi,B.Kang,G.Ceder,High rate micron-sized ordered LiNi0.5Mn1.5O4,Journalof the Electrochemical Society157(2010)A925.[25]X.Fang,N.Ding,X.Y.Feng,Y.Lu,C.H.Chen,Study of LiNi0.5Mn1.5O4syn-thesized via a chloride-ammonia co-precipitation method:electrochemical performance,diffusion coefficient and capacity loss mechanism,Electrochim-ica Acta54(2009)7471.[26]M.Mohamedi,M.Makino,K.Dokko,T.Itoh,I.Uchida,Electrochemical investi-gation of LiNi0.5Mn1.5O4thinfilm intercalation electrodes,Electrochimica Acta 48(2002)79.[27]/∼mdt26/casino2pseudopotentials.html(accessed01.10.13).[28]V.I.Anisimov,J.Zaanen,O.K.Andesen,Band theory and Mott insulators:Hub-bard U instead of Stoner I,Physical Review B44(1991)493.[29]L.Wang,Th.Maxisch,G.Ceder,Oxidation energies of transition metal oxideswithin the GGA+U framework,Physical Review B73(2006)1.[30]D.Kramer,G.Ceder,Tailoring the morphology of LiCoO2:afirst principles study,Chemistry of Materials21(2009)3799.[31]A.Van der Ven,First Principle Investigation of the Thermodynamicand Kinetic Properties of Lithium Transition Metal Oxides,MIT,2000 (Ph.D.thesis).[32]M.Strickland, D.Yager-Elorriaga,A parallel algorithm for solving the3d Schrodinger equation,Journal of Computational Physics229(2010) 6015.。

力学生物的英文

力学生物的英文

力学生物的英文Biomechanics is a fascinating field that combines the principles of physics and engineering with the study of living organisms. It explores the mechanical properties and functions of biological systems, providing insights into how living beings move, interact with their environment, and adapt to various challenges.At its core, biomechanics examines the forces and stresses that act on the body, and how the body responds to these forces. This includes understanding the mechanics of muscle contraction, bone and joint structure, and the dynamics of movement. By applying the laws of mechanics, biomechanists can analyze the efficiency and effectiveness of biological processes, from the microscopic level of cells and tissues to the macroscopic level of entire organisms.One of the primary areas of biomechanics is the study of human movement. Researchers in this field investigate the biomechanics of walking, running, jumping, and other physical activities, with the goal of improving athletic performance, preventing injuries, and enhancing the quality of life for individuals with physical disabilities or impairments. By understanding the biomechanical principles that govern human movement, scientists can develop better trainingmethods, design more effective prosthetic devices, and optimize the design of sports equipment.Beyond human movement, biomechanics also plays a crucial role in understanding the structure and function of other living organisms. For example, biomechanists study the locomotion of animals, such as the swimming of fish or the flight of birds, to gain insights into the evolutionary adaptations that have enabled these creatures to thrive in their respective environments. Similarly, biomechanics is essential for understanding the mechanical properties of plant tissues, which are essential for their growth, survival, and reproduction.In the field of medicine, biomechanics has numerous applications. Orthopedic surgeons use biomechanical principles to design and develop more effective prosthetic limbs, joint replacements, and other medical devices. Biomechanics also contributes to the understanding of injury mechanisms, allowing for the development of better protective equipment and rehabilitation strategies. Additionally, biomechanics plays a role in the design of medical imaging techniques, such as magnetic resonance imaging (MRI) and computed tomography (CT) scans, which provide valuable information about the structure and function of the human body.The applications of biomechanics extend beyond the realm of human health and performance. In the field of engineering, biomechanics isincreasingly being used to inspire the design of innovative technologies. Researchers are studying the remarkable abilities of various organisms, such as the adhesive properties of gecko feet or the efficient flight patterns of birds, to develop biomimetic solutions for a wide range of engineering challenges. These bioinspired designs have the potential to revolutionize fields like robotics, materials science, and energy production.As the field of biomechanics continues to evolve, it is becoming increasingly interdisciplinary, drawing on expertise from fields such as biology, physics, computer science, and mathematics. This collaborative approach allows for the development of more comprehensive and sophisticated models of biological systems, leading to groundbreaking discoveries and advancements in our understanding of the natural world.In conclusion, biomechanics is a dynamic and rapidly advancing field that has far-reaching implications for our understanding of living organisms and our ability to engineer innovative solutions to complex problems. By unraveling the mechanical principles that govern biological systems, biomechanists are paving the way for new breakthroughs in fields as diverse as medicine, sports science, and engineering. As we continue to explore the fascinating world of biomechanics, we can expect to witness even more remarkablediscoveries and applications that will shape the future of our understanding of the natural world.。

第五章 晶体中电子能带理论

第五章 晶体中电子能带理论

ˆ 具有晶格周期性。 因此晶体中单电子哈密顿量 H
ˆ ˆ ˆ ˆ ˆ T ( Rn ) H (r ) (r ) H (r Rn ) (r Rn ) H (r )T ( Rn ) (r )
ˆ, H ˆ ] HT ˆ ˆ TH ˆˆ 0 [T

Байду номын сангаас

n1 n2 n3 ˆ ˆ ˆ ˆ 可得到 T ( Rn ) (r ) T (a1 ) T (a2 ) T (a3 ) (r ) n1 n2 n3 (a1 ) (a2 ) (a3 ) (r ) ( Rn ) (r ) n1 n2 n3 即 ( Rn ) (a1 ) (a 2 ) (a 3 ) (a1 )、 (a2 )、 (a3 ) ? 设晶体在 a1、a 2、a3方向各有 N 1、N 2、N 3个原胞 ,
第五章 晶体中电子 能带理论
能带理论

能带论是目前研究固体中的电子状态,说明固体性质最重
要的理论基础。

能带理论是用量子力学的方法研究固体内部电子运动的理 论。它曾经定性地阐明了晶体运动的普遍特点,并进而说 明了导体与绝缘体、半导体的区别所在,解释了晶体中电 子的平均自由程问题。

能带论的基本出发点是认为固体中的电子不再是完全被束


, 2 e
ik a2
, 3 e
ik a3

( Rn ) e ---布洛赫定理 ik Rn (r Rn ) e (r )
ik Rn
( Rn ) e ik Rn (r Rn ) e (r ) ---布洛赫定理

常用量子化学词汇

常用量子化学词汇

Average,期望值,ab initio, 从头计算approximate,近似accurate, 精确atomiticity, 粒子性active, 活性的adiabatic, 绝热的,非常缓慢的anti-symmetry principle 反对称原理Basis,基组bra, 左矢,左矢空间,右矢空间的对偶空间boundary,边界条件Born-Oppenheimer 波恩奥本海默近似,绝热近似退耦后的进一步近似Configuration, 组态,电子排布correlation, 电子的相关作用commutation, 对易子coordinate, 坐标conjugate, 共轭core, 原子实convergence, 收敛,级数或积分收敛coupling, 耦合Coulomb’s Law, 库仑定律,麦克斯韦场方程的点电荷近似correspondence principle, 对应原理complete, 完备的complete active space (CAS), 完备的活性空间closed-shell, 闭壳层closed system, 封闭体系configuration state function (CSF)组态波函数Diagonalization,对角化Diagonal, 对角阵,对角元DFT, 密度泛函理论density,电子密度D-CI, double CIdynamical, 动力学的deterministic, 行列式的diabatic 未对角化的,非自身表象的,透热的Effective Hamiltonian, 有效哈密顿electron, 电子eigenvalue, 本征值eigenvector, 本征矢,无限维Hilbert空间中的态矢量external, 外加的energy, 能量excitation, 激发态excited, 被激发的exclusion principle不相容原理Functional, 泛函数function, 函数Fock space, Fock空间force, 力.,field场Gradient,梯度Gaussian, 高斯程序,高斯函数generic, 普适的Gauge 规范Hamiltonian, 哈密顿,Hessian, 二阶导数Hermitian, 厄米的Hartree 原子能量单位Integral, 积分internal, 内部的(内部自由度的)interaction, 相互作用independent, 不独立的invariant, 不变的iteration, 叠代interpretation, (几率)诠释interpolation,inactive不活动的J-integral, j积分jj-coupling jj耦合K-integral, k积分ket,右矢,右矢空间Linear algebra, 线性代数,linear combination of atomic orbitals (LCAO),原子轨函线性组合(法)local, 定域的locality, 定域物理量linear scaling, 线性标度low order,低对称性,有序度较低的情形Matrix, 矩阵,metric,矩阵的momentum, momenta,动量many-body theory,多体理论mechanics,力学,机理,机制multiconfiguration self-consistent field (MCSCF),多组态自洽场multireference (MR),多参考态方法minimization,最小化Normalization,归一化normal order, 正常序norm,已归一化的(波函数),N-electron, N电子体系nondynamic,非动力学的nonadiabtaic 非绝热的,有交换作用的,非渐变的Orbit,轨道orbital,轨道波函数,轨函observable, 可观测的(物理量)operator, 算符optimization, 优化one-electron,单电子,orthogonal, 正交的orthonormal, 正交归一的,open-shell,开壳层open system,开放体系,order-N第N阶(近似,导数)Principle,原理,原则property,性质particle, 粒子probability, 几率probabilistic, 几率性的potential,势PES, 势能面pseudo-, 赝的,pseudo vector赝矢量,pseudopotential,赝势perturbation theory,微扰理论Quantum, quanta, 量子quantized, 使量子化quantization,量子化的过程quotient,商quantity,数量,物理量Relativity,相对论,relativistic, 相对论性的representation,表示,表象Reference,参考系,参考态Spin, 自旋S-matrix, s矩阵,线性变换矩阵,散射矩阵symmetry, 对称性SCF, 自洽场stability,稳定性state,态scale,标度,测量shell,电子壳层spin-orbit coupling,自旋轨道耦合static,静态的space,空间,坐标空间的,banach/Hilbert Space,巴拉赫,希尔伯特空间spatial,空间的similarity transformation, 相似变换self-consistent field (SCF), 自洽场secondary ,二阶的,二级的,second quantization,二次量子化Transition state,过渡态time-dependent,含时的,对时间依赖的trace,矩阵的迹.Transformation,变换Universal,统一的,全同的。

Ab20initio20calc_高斯计算

Ab20initio20calc_高斯计算

Project out rotation and translation
• Generate transformation Matrix D: - D1 to D3 correspond to translation, for example, water (mH=1, mO=16):
D1 (1,0,0,4,0,0,1,0,0)t D2 (0,1,0,0,4,0,0,1,0)t D3 (0,0,1,0,0,4,0,0,1)t - D4 to D6 correspond to rotational motion:
• Truncate to the Second order:
1 V (r ) V (re ) Fr (r re ) 2 2
• Gradient (negative force) and Hessian (force constant): 2V V F 2 G 0, r r re r r re
HCN -> HNC transition state 0 C N X H 1 1 NC 2 QN 1 90.0 3 HQ 2 90.0 1 0.0
H4
X3
C1
N2
NC=1.18 QN=1.15 HQ=0.8 D
Estimating the Hessian
• Taylor expansion at equilibrium position:
D4 j ,i (( Py ) i X j ,3 ( Pz ) i X j , 2 ) / mi D5 j ,i (( Pz ) i X j ,1 ( Px ) X j ,3 ) / mi D6 j ,i (( Px ) i X j , 2 ( Py ) i X j ,1 ) / mi j=x, y, z; i is over all atoms; P is the dot product of R and X - Schmidt Orthogonalization is used to generate the remaining vectors from D7 to D3N

计算凝聚态物理研究

计算凝聚态物理研究

where
VH [n(r)]
n(r)n(r') d 3rd 3r' | r r'|
History of Density Functional Theory
Variational principle: E = min{ E[n(r)] }
We have single electron equation (Kohn-Sham):
Projects in my group
Doping of graphene hydrogen fuel cell Pt catalyst carbon alloy to replace Pt which kind of structure is the key how to realize it
About the VASP code
1. Written by Georg Kresse, Martijn Marsman, and Jurgen Furthmuller Computational Physics, Faculty of Physics Universitat Wien Sensengasse 8, A-1130 Wien, Austria
1 2
n(r )n(r ' )drdr ' | r r'|
Approximation for Exc[n(r)]
The local density approximation (LDA)
ExLcDA[n(r)] exc[n(r)]n(r)dr
For a uniform electron gas:
exc (n)
0.458 rs
rs
0.44 0.78
Simple but the results are surprisingly useful

Ab initio calculation of zero-field

Ab initio calculation of zero-field

J. Chem. Soc ., Perkin Trans. 2, 1999, 2299–23032299This journal is © The Royal Society of Chemistry 1999Ab initio calculation of zero-field splitting and spin-orbit coupling in ground and excited triplets of m -xylyleneZdeneˇk Havlas a and Josef Michl *b aInstitute of Organic Chemistry and Biochemistry, Academy of Science of the Czech Republic,16610 Prague 6, Czech Republic bDepartment of Chemistry and Biochemistry, University of Colorado, Boulder,CO 80309-0215, USAReceived (in Cambridge, UK) 16th August 1999, Accepted 28th September 1999We report CASSCF(6,6)/cc-pVDZ optimized geometries, energies (also single-point CASPT2(6,6)/cc-pVDZ),electron spin–spin dipolar interaction (D ,E ) tensor, and spin-orbit coupling (SOC) for m -xylylene in the lowest triplet T 1 (13B 2), in the next triplet 13A 1, and in the slightly higher 23B 2. The zero-field splitting (zfs) parameters computed for T 1 (D /hc =0.013 cm Ϫ1, E /hc =Ϫ0.003 cm Ϫ1) agree well with the observed values |D /hc |=0.011 cm Ϫ1, |E /hc |<0.001cm Ϫ1. If 3A 1 is the T 2 state as calculated, its computed D /hc (Ϫ0.040 cm Ϫ1) and E /hc (0.001 cm Ϫ1) agree with the value |D /hc |=0.04 ± 0.01 cm Ϫ1 deduced from experiment assuming E =0. If 23B 2 is the T 2 state, the experimental data need to be reevaluated, since its computed E /hc value (Ϫ0.012 cm Ϫ1) is not negligible relative to D /hc (0.038 cm Ϫ1). The SOC matrix elements of T 1–T 3 with the lowest and the ππ* excited singlets are small (~0.01–0.1 cm Ϫ1), while those with representative 1σπ* states are large (~10 cm Ϫ1). The former lack one-center terms and therefore are much smaller than expected from the standard one-electron approximation. Computed SOC a ffects D and E slightly, and supports the proposed vibronic mechanism of intersystem crossing from T 2.1. Introductionm -Xylylene (1), a highly reactive biradical with no classicalKekulé structure, has found considerable use as a building block in magnetic organic materials.1 I t was first observed spectroscopically in 1975 by Migirdicyan and Baudet 2 in n -alkane polycrystalline matrices at 77 K. They determined the first electronic transition energy from the position of the fluorescence 0,0 band (22730 cm Ϫ1). According to calcu-lations,2–8 m -xylylene has a 3B 2 triplet ground state T 1 and two low-lying excited triplet states of di fferent symmetries, 13A 1 and 23B 2 (here and in the following, symmetry labels refer to the space part of the electronic wave function). The computed energy separation between these excited triplets (T 2 and T 3) is very small. Most of the calculations place the 13A 1 state a little below the 23B 2 state, but some methods of calculation yield the reverse order.5,7,8 The existence of these two close-lying triplets has not been con firmed by direct observation, but the absence of mirror-image symmetry in laser-induced fluorescence excit-ation spectra suggests strongly that both calculated 13A 1 and 23B 2 states are indeed present and vibronically coupled.6,9 The energy order of the zero-order electronic states has not been established by experiment. The same general pattern of triplet state energies is found in methyl substituted m -xylylenes.8Wright and Platz 10 found that the EPR signal intensity of m -xylylene follows the Curie law, in agreement with the assign-ment of the ground state as a triplet (T 1). They found |D/hc |=0.011 cm Ϫ1, |E/hc |≤0.001 cm Ϫ1 for its zero-field-splitting (zfs)parameters, and this was later con firmed by Goodman and Berson,11 who synthesized m -xylylene by an independent route.Fluorescence decay of methylated m -xylylene biradicals isbiexponential, and this was attributed to emission from non-equilibrating triplet sublevels.7,12 From the dependence of the lifetime on magnetic field strength the zfs parameter |D/hc |=0.04 ± 0.01 cm Ϫ1 of the excited T 2 electronic state was deduced,13 assuming the E parameter to be negligible. This was the first zfs parameter measurement for an upper triplet state of an organic molecule; results for upper triplets of other molecules have become available more recently.14The first semiempirical calculation 15 of the D parameter of the T 1 state preceded the experimental measurement 10 and overestimated the D value by a factor of three (|D/hc |=0.032cm Ϫ1). This was followed by a more complete semiempirical calculation 16 that produced D values of 0.025, 0.049 and 0.053cm Ϫ1 for the 13B 2 (T 1), 13A 1 (T 2) and 23B 2 (T 3) states, respect-ively. Both sets of calculations used only the spin–spin dipolar operator and assumed negligible spin-orbit coupling e ffects.They also made no attempt to calculate the E parameter. Even in the later calculation,16 the D value for the T 1 state was over-estimated considerably, but its increase upon excitation to the T 2 state was reproduced qualitatively. This is true regardless of whether the latter is of 3A 1 or 3B 2 symmetry, since similar D values were calculated for the 13A 1 and 23B 2 states, calculated to be closely spaced.The purpose of the present study is threefold. First, we wish to establish whether present-day ab initio methods are capable of reproducing the zfs parameters of the T 1 and T 2 states more accurately. We are not aware of prior ab initio zfs constant calculations for higher triplet states of molecules of this size.Second, we wish to test the degree of validity of the common belief that zfs of the triplet states of planar π-electron hydro-carbons is dictated by electron spin–spin dipolar coupling and is not a ffected much by spin-orbit coupling. Third, we wish to find out whether the previously proposed 7 interpretation of the qualitative features of intersystem crossing from the T 2 level to low-lying singlets is supported by actually computed values of the relevant spin-orbit coupling matrix elements.2. Method of calculationThe calculations were performed with CASSCF(6,6) wave func-P u b l i s h e d o n 01 J a n u a r y 1999. D o w n l o a d e d o n 18/11/2015 18:32:05.View Article Online / Journal Homepage / Table of Contents for this issue2300J. Chem. Soc ., Perkin Trans. 2, 1999, 2299–2303tions using the cc-pVDZ basis set.17 The molecular geometry of the three lowest triplet states was optimized under constraint to C 2v symmetry and to C s symmetry with state-speci fic CAS orbitals. The six active orbitals were of π symmetry (three a 2and three b 1 MOs; these six π orbitals are energetically well separated from the others). The energies of the triplet states were recalculated at the CASPT2(6,6)/cc-pVDZ level at the CASSCF optimized geometries.Geometry optimization and CASSCF calculations were done with the GAUSSIAN98 program,18 CASPT2 calculations with the MOLCAS 4.1 program,19 and zfs and spin-orbit coupling calculations with the ab initio program suite SOSS,20 which has been presently modi fied to calculate the expectation values of the spin –spin Breit –Pauli operator for excited state wave func-tions. This modi fication allows the calculation of electronic spin –spin dipolar tensor and zfs parameters for any type of CI wave function. Electron spin –spin dipolar coupling and spin-orbit coupling were computed using CASSCF(6,6)/cc-pVDZ wave functions determined for each state at its optimized CASSCF(6,6) C 2v geometry. Both the one-electron and the two-electron parts of the spin-orbit coupling Breit –Pauli Hamiltonian were included.I n the calculation of spin-orbit coupling matrix elements between the T 1–T 3 states and 1σπ* states, the six active orbitals (two σ and four π or π*) were those that appeared in the most important con figurations in a CASSCF(10,10) calculation in which the usual space of six π orbitals was extended by adding the two σ and two σ* orbitals that were the closest to the Fermi level. Among the lowest 25 singlet states, there were none with a signi ficant weight of any πσ* con figurations.3. Results and discussionCalculated triplet geometries and energiesCASSCF(6,6) optimized geometries of the three lowest triplet states subject to a C 2v symmetry constraint are shown in Fig. 1.The C –C bond lengths of the T 1 (13B 2) state are all close to ordinary aromatic bond lengths. In the T 2 (13A 1) state, the exo-cyclic C –C bonds are 3 pm shorter, the C 1–C 2 and C 2–C 3 bonds remain almost unchanged, and the other bonds are longer (C 4–C 5 and C 5–C 6 by 8 pm). The C 1–C 2–C 3 angle increases by 2.4Њ.Compared with the T 1 state, in the T 3 (23B 2) state the exocyclic bonds are all 3 pm longer than the bonds between atoms C 6, C 1,C 2, C 3, and C 4. The C –H bond lengths are constant (1.079 to 1.082 Å), as are the valence angles.A CASSCF(6,6) optimization subject to a C s symmetry con-straint, and starting from a C s trial geometry, led to the same C 2v structure for the T 1 state. However, for the T 2 and T 3 states,it ended in di fferent C s stationary points. CASSCF calculations for the T 2 state predict the best C s geometry to lie 12.5 kcal mol Ϫ1 below the best C 2v geometry. However, single-point PT2corrections to the T 2 CAS wave functions reverse this order and place the T 2 energy at the C 2v geometry 1.3 kcal mol Ϫ1 below that at the C s geometry. In the case of the T 3 state the energy at the best CASSCF C s geometry is 3.6 kcal mol Ϫ1 above that at the best CASSCF C 2v geometry, and the di fference increases to Fig. 1The CASSCF(6,6)/cc-pVDZ optimized geometry of the first three triplet states of 1 (bond distances in Å and bond angles indegrees).9.4 kcal mol Ϫ1 after single-point CASPT2 corrections. A reli-able determination of the minimum geometry of the T 2 state will require a rather expensive unconstrained geometry opti-mization at the CASPT2 level, preferably with a larger basis set,and a vibrational frequency analysis. It is quite possible that T 2and T 3 are two branches of a double cone surface (conical intersection). We believe that the best T 2 and T 3 C 2v geometries found presently are adequate for the our purposes, but realize that they need not correspond exactly to true minima in these surfaces.Comparison with other calculations and with observed spectra Bond lengths of the T 1 state are systematically shorter than those obtained in the pioneering UHF/STO-3G calculation of Kato et al .;4 the largest di fference is 3.6 pm for r (C 4–C 5). Negri and Orlandi 6 calculated geometries for the three lowest triplet states by a modi fied QCFF/PI method. Their bond lengths for the T 1 state di ffer from the CAS values at most by 2 pm, but the trends observed upon going to higher excited triplet states are di fferent. The T 1 distances calculated by Hrovat et al .21 at the CASSCF(8,8)/6-31G* level di ffer from the present results by less then 1 pm.According to our CASPT2 calculations the energy of the T 2(13A 1) state is 67.2 kcal mol Ϫ1 above that of the T 1 (13B 2) ground state and the T 3 (23B 2) state lies 75.0 kcal mol Ϫ1 above the T 1state. The observed T 2→T 1 electronic transition energy deter-mined from the fluorescence 0–0 band 2 is 64.8 kcal mol Ϫ1 and thus agrees quite well with the calculated value. The energy of the T 3 state has not yet been experimentally determined, but it cannot be much above T 2. Semiempirical π-SCF-MO-CI calcu-lations 2 provided excitation energies of 66.9 (13A 1) and 79.3(23B 2) kcal mol Ϫ1, but the results were strongly dependent on the MOs used. Later, the transition energies were recalculated 8for a series of methyl substituted m -xylylenes with adjusted semiempirical parameters. For the parent biradical they were 64.9 (23B 2) and 67.2 (13A 1) kcal mol Ϫ1.Calculated spin-dependent properties(i) Spin –spin dipolar coupling. The zfs parameters calculated without inclusion of spin-orbit coupling are listed in Table 1(Fig. 2). They follow the general pattern of the prior semi-empirical results,16 but are numerically smaller. Among the three lowest triplet states, the T 1 state has the smallest |D /hc |value because the unpaired electrons reside far apart, primarily one on each of the exocyclic carbon atoms C 7 and C 8 (Fig. 3).As was proposed earlier 16 on the basis of semiempiricalFig. 2CASPT2(6,6) triplet energies of 1. Left, C 2v state labels for standard axis labels. Right, triplet sublevels and orientation ofmagnetic axes in EPR notation.P u b l i s h e d o n 01 J a n u a r y 1999. D o w n l o a d e d o n 18/11/2015 18:32:05.J. Chem. Soc ., Perkin Trans. 2, 1999, 2299–23032301Table 1Calculated [CASSCF(6,6)] and experimental zfs parameters for the three lowest triplet states of 1 (in cm Ϫ1)Calcd.Exp.State D/hcE/hc State |D /hc ||E/hc |13B 2 (T 1)13A 1 (T 2)b 23B 2 (T 3)b0.013Ϫ0.043 (Ϫ0.040)c 0.038Ϫ0.0030.001Ϫ0.012T 1T 20.011a0.04±0.01d —<0.001a 0e —aRef. 10. b The calculated order of the 13A 1 and the 23B 2 states could be incorrect since they are very close in energy. c With correction for T 2–S 12spin-orbit coupling. The other results are not a ffected by spin-orbit coupling with the singlet states considered (Table 2). d Ref. 13. e Assumed.Table 2Non-zero 1-electron, 2-electron, and total spin-orbit matrix elements between the lowest triplet and the most stable and the lowest ππ* and σπ* excited singlet states of 1 (in cm Ϫ1)a11A 1 (S 0)b11B 2 (S 1)c21B 2 (S 2)c21A 1 (S 3)c 11A 2 (S 12)d 11B 1 (S 14)d 21B 1 (S 21)d T 1 (13B 2)T 2 (3A 1)T 3 (23B 2)1-electron 2-electron total1-electron 2-electron total1-electron 2-electron total0.0138Ϫ0.0130.0008 (x )0.1926Ϫ0.1993Ϫ0.0067 (x ) 1.4541Ϫ1.4816Ϫ0.0275 (x )0.4695Ϫ0.4803Ϫ0.0108 (x )0.849Ϫ1.0423Ϫ0.1933 (x )0.4031Ϫ0.515Ϫ0.1119 (x )9.9368Ϫ5.27454.6623 (y )27.3672Ϫ13.957313.4099 (z )17.9282Ϫ9.51598.5159 (y )17.0195Ϫ8.89988.1197 (z )23.2832Ϫ12.518910.6643 (y )12.3315Ϫ6.55055.7810 (z )18.3706Ϫ9.53278.8379 (z )13.2179Ϫ6.72796.4900 (y )18.9564Ϫ10.23658.7199 (z )aThe label of the coupling triplet sublevel is shown in parentheses. b The most stable singlet. c ππ* excited singlet. d σπ* excited singlet.calculations, in the T 2 state the unpaired electrons are located in the ring, mostly on the vicinal carbons C 4, C 5, and C 6, and their proximity causes |D /hc | to increase. The T x and T y sublevels of the T 2 state are very close, and the |E /hc | parameter is only 0.001cm Ϫ1. In the T 3 state, the unpaired electrons are distributed over a chain of carbon atoms (C 7–C 1–C 2–C 3–C 8), which also results in a large splitting of sublevels. Compared to the T 2 state, in the T 3 state the unpaired spins are distributed in a less linear fashion and |E /hc | is much larger.(ii) Spin-orbit coupling. No prior results are available, and the spin-orbit interaction of the three lowest triplet states, all of which have an even number of π electrons, and the four lowest singlet states, all of which also have an even number of π elec-trons, was investigated first. By El-Sayed ’s rules,22 spin-orbit coupling among such states should be weak, since one-center terms vanish. The four lowest singlet states are 11A 1 (S 0), 11B 2(S 1), 21B 2 (S 2), and 21A 1 (S 3), with calculated energies 13.2, 47.2,104.1, and 110.3 kcal mol Ϫ1, respectively, relative to the T 1 state.The calculated energy of the S 0 state agrees well with the pub-lished values 21 of 12.9 and 11.7 kcal mol Ϫ1, calculated at the CASSCF(8,8) and CASPT2N levels, respectively. I n the C 2v symmetry group the singlet A 1 states have non-zero spin-orbit matrix elements with the T x sublevel of triplet B 2 states and the singlet B 2 states have non-zero elements with the T x sublevel of the triplet A 1 state, while other matrix elements vanish. Table 2shows that even the non-zero elements are all extremely small,as expected. One-center contributions vanish by symmetry and two-center contributions are generally small, both because of the r Ϫ3 dependence on the distance from the atomic nucleus,and because the one- and two-electron contributions nearly cancel.Fig. 3CASSCF(6,6) spin populations (%) for the three lowest triplet states of 1.The reason for this cancellation is readily seen when NBO analysis 20 is performed. For instance, the contributions to the 〈13B 2|H SO |21A 1〉x element are largely due to a pair of π-symmetry AOs located on C 6 and C 7 (C 4 and C 5) interacting either with the positive charge of the nucleus at C 1 (C 3) (one-electron contribution), or with the negative charge due to elec-trons at C 1 (C 3) (two-electron contribution). Since C 1 (C 3) is relatively distant, the two types of interaction approximately cancel.Clearly, the approximate proportionality between the total,the one-electron, and the two-electron contributions to the spin-orbit coupling operators, which is the basis for the usual one-electron approximation to this operator that invokes empir-ical atomic constants, cannot hold for the two-center contribu-tions. In this case, as the distance from the nucleus increases, the total spin-orbit coupling contribution goes to zero much faster than the one-electron and the two-electron parts individually, at least if the nuclear charge is approximately balanced by the electrons in its vicinity. Thus, in systems such as m -xylylene, in which one-center contributions vanish by symmetry, the usual empirical one-electron approximation cannot be used for spin-orbit coupling between states with equal numbers of π electrons, nor can the usual computer programs such as GAUSSIAN98. However, it is also true that in such cases spin-orbit coupling is small and can often be neglected altogether,for instance when evaluating zfs parameters. In the present case,we find that the e ffect of spin-orbit coupling of the T 1, T 2, and T 3 states with low-energy singlet states on the values of |D /hc |and |E /hc | is entirely negligible, less then 10Ϫ5 cm Ϫ1.El-Sayed ’s rules suggest that the matrix elements for the spin-orbit mixing of the low-lying triplet states with high-energy excited singlet states with an odd number of electrons, which are dominated by πσ* and σπ* con figurations, will be larger.We have therefore attempted to identify a few representative states of this kind, both singlets and triplets, using CASCSF-(10,10)/cc-pVDZ calculations. I n these, the (6,6) active space was extended by adding two σ (a 1 and b 2) and two σ* (a 1 and b 2)orbitals. Among the resulting 25 lowest triplet states there was only one state of this type (T 16, 3A 2, σπ*), located 270 kcal mol Ϫ1 above the T 1 state. Among the resulting 25 lowest singlet states, four states of the desired kind were found, all σπ* (ener-gies in kcal mol Ϫ1): S 12 (1A 2, 279), S 14 (1B 1, 289), S 21 (1B 1, 334),P u b l i s h e d o n 01 J a n u a r y 1999. D o w n l o a d e d o n 18/11/2015 18:32:05.2302J. Chem. Soc ., Perkin Trans. 2, 1999, 2299–2303and S 23 (1A 2, 345). Due to the intrinsic limitations of our spin-orbit program we had to restrict the spin-orbit calculations to CASSCF(6,6) wave functions by deleting those orbitals occur-ring in the (10,10) active space that were either doubly occupied or unoccupied in all of the most important con figurations of the interacting states. Although these four states are just a few of a large number, one can hope that their properties are repre-sentative and that they give us a glimpse of what we might expect if all of them were included.Table 2 lists the non-zero spin-orbit coupling matrix elements between the T 1, T 2, and T 3 states and the S 12, S 14, and S 21 states.Note that for these one-center dominated terms the usual approximate proportionality of the one-electron, two-electron,and total contributions holds well.Only the T y sublevel of T 1 and T 3, which are of 3B 2 symmetry,can couple with S 12 (1A 2), and only their T z sublevel can couple with S 14 and S 21 (1B 1). Similarly, only the T z sublevel of T 2couples with S 12 and only its T y sublevel couples with S 14 and S 21. These matrix elements are large (~10 cm Ϫ1), comparable to that in carbene.20 However, since these singlet states are so high in energy, the e ffect of this spin-orbit coupling on the |D /hc | and |E /hc | values of T 1, T 2, and T 3 is below 0.001 cm Ϫ1 and neg-ligible. The only exception is the value of |D /hc | in T 2, which is reduced by 0.003 cm Ϫ1 by spin-orbit interaction with S 12. Still,this is merely a ~10% e ffect, beyond the present computational accuracy. However, it will not be negligible if a high-accuracy calculation is attempted in the future. With this single excep-tion, the spin –spin dipolar |D /hc | and |E /hc | values given in Table 1 also represent the total presently computed values.We realize that the cumulative e ffect of the many other σπ*(and probably also πσ*) states located at higher energies on |D /hc | and |E /hc | could be larger in the absence of fortuitous compensation. Much compensation can however be expected,since there will be roughly as many 1A 2 states as 1B 1 states, and their matrix elements will be of comparable magnitude as long as the σ skeleton is approximately isotropic in the molecular plane. Then, the T y and the T z sublevels of low-energy triplets will be stabilized to a comparable degree. However, there is no analogous mechanism for stabilizing the T x sublevels, and it seems that the e ffect of spin-orbit coupling with σπ* and πσ* states on the zfs parameters cannot be safely ignored in molecules such as 1 when high accuracy is required. An improvement of the present calculation of D and E in the direction of a larger basis set and a better description of electron correlation would thus make little sense unless spin-orbit coupling with the high-energy σπ* and πσ* states is fully included also.Comparison with experiment(i) Zero-field splitting. There is an element of uncertainty in the computed D and E values. First, the absolute 20 values of the calculated zfs parameters are known to be too large and to converge slowly with the quality of the basis set and with the amount of correlation introduced into the wave function. At the present level of calculation, we estimate the absolute values to be about 10% too high. Second, spin-orbit coupling with high-energy σπ* and πσ* states introduces many small correc-tions that are likely to mutually cancel to a considerable degree,but not perfectly, the net e ffect being a stabilization of the T y and T z sublevels relative to T x , and a small change in the separ-ation of T y and T z . The magnitude of the few contributions of this kind that we did calculate suggests that the neglect of the rest introduces an error of unknown direction in the calculation of D and E , most likely again ~10% in magnitude. I t is encouraging and perhaps somewhat fortuitous that the values of zfs parameters calculated for the T 1 (13B 2) state are in excel-lent agreement with the experimental values and exceed them only by 0.002 cm Ϫ1, and that the D /hc parameter of the T 2(13A 1) state was calculated within the experimental error (Table 1). The assumption E =0 that was made in the interpretation of the experimental data for T 2 seems well justi fied.The calculated value of the |D /hc | parameter for the 23B 2 state is close to that of the 13A 1 state and thus does not help to assign the observed T 2 triplet as either 3A 1 or 3B 2. The situation would change dramatically if the E parameter of T 2 could be meas-ured, since the values calculated for 23B 2 and 13A 1 di ffer by an order of magnitude. If the symmetry of T 2 is not 3A 1 as calcu-lated here, but instead is 3B 2, it would not be advisable to assume E =0 in the analysis of the experimental results.(ii) T 2 to S intersystem crossing. The observed biexponential T 2→T 1 fluorescence decays were interpreted 16 under the assumption that T 2 is of 3B 2 symmetry, but the conclusions for 3A 1 symmetry, assumed here, would be the same. I t was pro-posed that one of the three triplet sublevels, T x , decays slowly since its intersystem crossing to lower-energy singlets is ine ffi-cient (the spin-orbit matrix elements with the low-lying singlets are non-zero but very small). The primary decay mechanism was postulated to be T 2→T 1 internal conversion. The other two triplet sublevels, T y and T z , decay faster and at comparable rates, and this was attributed to vibronic intersystem crossing induced by a 2 and b 1 vibrations that mix σπ* (or πσ*) character into low-energy singlet or triplet wave functions.Our numerical results are in perfect agreement with the pro-posed interpretation, in that the matrix elements for purely elec-tronic coupling of the T x level with low-energy singlets located within ~20–40 kcal mol Ϫ1 are three orders of magnitude smaller than the matrix elements for coupling with σπ* singlets located one order of magnitude (~200–400 kcal mol Ϫ1) higher in energy (recall that a square of the matrix element enters the rate expression), and in that there is no pronounced anisotropy di fferentiating the y and z directions.Our results also leave little doubt that the lowest excited states with an odd number of π electrons are σπ* and not πσ*in character.4. ConclusionsThe following conclusions can be drawn:(i) Ab initio calculations at the CASSCF(6,6)/cc-pVDZ level yield very satisfactory results for the zfs parameters of the T 1and T 2 states of m -xylylene, and the method can be expected to be useful for triplet states of other organic molecules of this size.(ii) As expected from symmetry arguments and in agreement with a long-held belief, spin-orbit coupling in m -xylylene has a very small e ffect on the zfs parameters of low-lying ππ* triplets,and they are dominated by the spin-dipolar term. However, the e ffect of spin-orbit coupling with high-energy σπ* and πσ*states is not entirely negligible and we estimate that it intro-duces a ~10% uncertainty into the computed D and E values.This result is likely to be general for molecules of this type.(iii) The non-zero spin-orbit coupling elements between the T x sublevel of triplets with an even number of π electrons and singlets with the same number of π electrons, which do not contain one-center terms for symmetry reasons and are small as expected from El-Sayed ’s rule, do not follow the usual pro-portionality rule between the one-electron and two-electron parts and the total e ffect, and the cancellation of the two parts is much more pronounced than usual. The reasons for this are easily understood in qualitative terms, and we conclude that two-center terms are generally likely to be overestimated in the usual one-electron aproximation. This provides an additional argument in favor of their neglect in the simple model for spin-orbit coupling in biradicals proposed earlier.23(iv) I n accord with El-Sayed ’s rules, spin-orbit coupling matrix elements between the T y and/or T z sublevels of triplets with an even number of π electrons and singlets with an odd number of π electrons (σπ* excited) are large and comparableP u b l i s h e d o n 01 J a n u a r y 1999. D o w n l o a d e d o n 18/11/2015 18:32:05.View Article OnlineJ. Chem. Soc ., Perkin Trans. 2, 1999, 2299–23032303to that in carbene. They are important for vibronically induced intersystem crossing induced by out-of-plane vibrations. This result is also likely to be general.AcknowledgementsThis work was supported by the National Science Foundation (CHE-9819179 and CHE-9709195).References1A. Rajca, in Modular Chemistry , ed. J. Michl, Kluwer, Dordrecht,The Netherlands, 1997, p. 193.2E. Migirdicyan and J. Baudet, J. Am. Chem. Soc., 1975, 97, 7400.3J. Baudet, J. Chim. Phys.-Chim. Biol., 1971, 68, 191.4S. Kato, K. Morokuma, D. Feller, E. R. Davidson and W . T.Borden, J. Am. Chem. Soc., 1983, 105, 1791.5M. Gisin and J. Wirz, Helv. Chim. Acta , 1983, 66, 1556.6F . Negri and G. Orlandi, J. Phys. Chem., 1989, 93, 4470.7A. Despres, V . Lejeune, E. Migirdicyan and W . Siebrand, J. Phys.Chem., 1988, 92, 6914.8V . Lejeune, A. Despres, E. Migirdicyan, J. Baudet and G. Berthier,J. Am. Chem. Soc., 1986, 108, 1853.9V . Lejeune, A. Despres and E. Migirdicyan, J. Phys. Chem., 1984,88, 2719.10B. B. Wright and M. S. Platz, J. Am. Chem. Soc., 1983, 105, 628.11J. L. Goodman and J. A. Berson, J. Am. Chem. Soc., 1985, 107, 5409.12V . Lejeune, A. Despres, B. Fourmann, O. B. d ’Azy and E.Migirdicyan, J. Phys. Chem., 1987, 91, 6620.13V . Lejeune, A. Despres and E. Migirdicyan, J. Phys. Chem., 1990,94, 8861.14E. Migirdicyan, B. Kozankiewicz and M. S. Platz, in Advanc es in Carbene Chemistry , JAI Press, 1998, vol. 2, p. 97.15M. Rule, A. R. Matlin, E. F . Hilinski, D. A. Dougherty and J. A. Berson, J. Am. Chem. Soc., 1979, 101, 5098.16V . Lejeune, G. Berthier, A. Despres and E. Migirdicyan, J. Phys.Chem., 1991, 95, 3895.17T. H. Dunning, Jr., J. Chem. Phys., 1989, 90, 1007; R. A. Kendall,T. H. Dunning, Jr. and R. J. Harrison, J. Chem. Phys., 1992, 96,6796.18M. J. Frisch, G. W . Trucks, H. B. Schlegel, G. E. Scuseria, M. A.Robb, J. R. Cheeseman, V . G. Zakrzewski, J. A. Montgomery, Jr., R.E. Stratmann, J. C. Burant, S. Dapprich, J. M. Millam, A. D.Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J. Tomasi, V . Barone,M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C. Adamo,S. Cli fford, J. Ochterski, G. A. Petersson, P . Y . Ayala, Q. Cui,K. Morokuma, D. K. Malick, A. D. Rabuck, K. Raghavachari,J. B. Foresman, J. Cioslowski, J. V . Ortiz, B. B. Stefanov, G. Liu,A. Liashenko, P . Piskorz, I. Komaromi, R. Gomperts, R. L. Martin,D. J. Fox, T. Keith, M. A. Al-Laham, C. Y . Peng, A. Nanayakkara,C. Gonzalez, M. Challacombe, P . M. W . Gill, B. Johnson, W . Chen,M. W . Wong, J. L. Andres, C. Gonzalez, M. Head-Gordon, E. S.Replogle and J. A. Pople, GAUSSIAN98, Revision A.6, Gaussian,Inc., Pittsburgh, PA, 1998.19K. Andersson, M. R. A. Blomberg, M. P . F ülscher, G. Karlstr öm,R. Lindh, P .-Å. Malmquist, P . Neogr ády, J. Olsen, B. O. Roos, A. J.Sadlej, M. Sch ütz, L. Seijo, L. Serrano-Andr és, P . E. M. Siegbahn and P .-O. Widmark, MOLCAS, Version 4.1, Lund University,Sweden, 1997.20Z. Havlas, J. W . Downing and J. Michl, J. Phys. Chem. A , 1998, 102,5681.21D. A. Hrovat, M. A. Murcko, P . M. Lahti and W . T. Borden,J. Chem. Soc., Perkin Trans. 2, 1998, 1037.22M. El-Sayed, J. Chem. Phys., 1963, 38, 2834.23J. Michl, J. Am. Chem. Soc., 1996, 118, 3568.Paper 9/06648IP u b l i s h e d o n 01 J a n u a r y 1999. D o w n l o a d e d o n 18/11/2015 18:32:05.View Article Online。

基泰尔 固体物理导论 英文版 第八版 introduction

基泰尔 固体物理导论 英文版 第八版 introduction

基泰尔固体物理导论英文版第八版introductionIntroductionSolid-state physics is a critical field of study that delves into the fundamental properties and behaviors of materials in their solid form. The understanding of solid-state phenomena has been instrumental in the development of numerous technological advancements, from semiconductor devices to superconducting materials. The eighth edition of "Bataile's Introduction to Solid-State Physics" provides a comprehensive and up-to-date exploration of this dynamic and ever-evolving discipline.At the heart of solid-state physics lies the study of the crystalline structure of materials and the ways in which atoms and molecules are arranged within these structures. This knowledge is essential for understanding the physical, chemical, and electrical properties of solids, as well as their response to various external stimuli, such as temperature, pressure, and electromagnetic fields.One of the key topics covered in this textbook is the concept oflattice structures. Lattices are the underlying frameworks that define the spatial arrangement of atoms or molecules in a solid material. By understanding the symmetry and periodicity of these lattice structures, researchers can gain valuable insights into the behavior of electrons, phonons (vibrations of the crystal lattice), and other fundamental particles within the material.The book also delves into the electronic properties of solids, exploring the behavior of electrons in the presence of a crystalline structure. This includes the study of energy bands, which describe the allowed energy levels for electrons in a solid, as well as the concept of semiconductors and their applications in modern electronics.Another crucial aspect of solid-state physics is the study of magnetic materials. The textbook examines the various types of magnetic ordering, such as diamagnetism, paramagnetism, ferromagnetism, and antiferromagnetism, and how these properties are influenced by the atomic structure and composition of the material.In addition to these core topics, the eighth edition of "Bataile's Introduction to Solid-State Physics" also covers more advanced concepts, such as superconductivity, the quantum Hall effect, and the behavior of materials under extreme conditions, such as high pressure or intense magnetic fields.One of the strengths of this textbook is its clear and concise explanations of complex theoretical concepts, accompanied by numerous illustrations and examples to aid in the reader's understanding. The authors have also included a wealth of problem sets and exercises at the end of each chapter, allowing students to apply the knowledge they have gained and deepen their understanding of the subject matter.Furthermore, the textbook is regularly updated to reflect the latest advancements in the field of solid-state physics, ensuring that readers are exposed to cutting-edge research and emerging technologies. This commitment to staying current with the rapidly evolving field of solid-state physics is a testament to the dedication and expertise of the authors and the publishers.In conclusion, the eighth edition of "Bataile's Introduction to Solid-State Physics" is an invaluable resource for students, researchers, and professionals working in the field of materials science, condensed matter physics, and related disciplines. Its comprehensive coverage, clear explanations, and practical applications make it an essential tool for anyone seeking to deepen their understanding of the fascinating world of solid-state physics.。

关于收敛问题

关于收敛问题

关于收敛问题(L502, L508, L9999)对于一个优化计算,它的过程是先做一个SCF计算,得到这个构型下的能量,然后优化构型,再做SCF,然后再优化构型。

因此,会有两种不收敛的情况:一是在某一步的SCF 不收敛(L502错误),或者构型优化没有找到最后结果(L9999错误)。

预备知识:计算时保存chk文件,可以在后续计算中使用guess=read读初始猜测.对于SCF不收敛,通常有以下的解决方法:1. 使用小基组,或低级算法计算,得到scf收敛的波函数,用guess=read读初始波函数。

2. 使用scf=qc,这个计算会慢,而且需要用stable关键字来测试结果是否波函数稳定。

如果这个还不收敛,会提示L508错误。

3. 改变键长,一般是缩小一点,有时会有用。

4. 计算相同体系的其他电子态,比如相应的阴离子、阳离子体系或单重态体系,得到的收敛波函数作为初始猜测进行计算。

5. 待补充.对于优化不收敛,即L9999错误,实际上是在规定的步数内没有完成优化,即还没有找到极小值点。

(或者对于过渡态优化,还没有找到过渡态)这有几种可能性:1. 看一下能量的收敛的情况,可能正在单调减小,眼看有收敛的趋势,这样的情况下,只要加大循环的步数(opt(maxcycle=200)),可能就可以解决问题了。

2. 加大循环步数还不能解决的(循环步数有人说超过200再不收敛,再加也不会有用了,这虽然不一定绝对正确,但200步应该也差不多了),有两种可能。

一是查看能量,发现能量在振荡了,且变化已经很小了,这时可能重新算一下,或者构型稍微变一下,继续优化,就可以得到收敛的结果(当然也有麻烦的,看运气和经验了);二是构型变化太大,和你预计的差别过大,这很可能是你的初始构型太差了,优化不知道到哪里去了,这时最好检查一下初始构型,再从头优化。

3. 对于L9999快达到收敛时,考虑减小优化步长有时对于能量振荡的情况也是有用的,opt(maxstep=1).(flyingheart )一个建议是,对于大体系,难收敛体系,先用小基组,低精度算法优化一下,以得到较好的初始构型,再用高精度的计算接着算。

First-principles Calculations of the Electronic Structure and Spectra of Strongly Correlate

First-principles Calculations of the Electronic Structure and Spectra of Strongly Correlate

a rX iv:c ond-ma t/974231v1[c ond-m at.str-el]28A pr1997First-principles calculations of the electronic structure and spectra of strongly correlated systems:dynamical mean-field theory V.I.Anisimov,A.I.Poteryaev,M.A.Korotin,A.O.Anokhin Institute of Metal Physics,Ekaterinburg,GSP-170,Russia G.Kotliar Serin Physics Laboratory,Rutgers University,Piscataway,New Jersey 08854,USA Abstract A recently developed dynamical mean-field theory in the iterated per-turbation theory approximation was used as a basis for construction of the ”first principles”calculation scheme for investigating electronic struc-ture of strongly correlated electron systems.This scheme is based on Local Density Approximation (LDA)in the framework of the Linearized Muffin-Tin-Orbitals (LMTO)method.The classical example of the doped Mott-insulator La 1−x Sr x TiO 3was studied by the new method and the results showed qualitative improvement in agreement with experimental photoemission spectra.1Introduction The accurate calculation of the electronic structure of materials starting from first principles is a challenging problem in condensed matter science since un-fortunately,except for small molecules,it is impossible to solve many-electron problem without severe approximations.For materials where the kinetic energy of the electrons is more important than the Coulomb interactions,the most successful first principles method is the Density Functional theory (DFT)within the Local (Spin-)Density Ap-proximation (L(S)DA)[1],where the many-body problem is mapped into a non-interacting system with a one-electron exchange-correlation potential approxi-mated by that of the homogeneous electron gas.It is by now,generally accepted that the spin density functional theory in the local approximation is a reliable starting point for first principle calculations1of material properties of weakly correlated solids(For a review see[2]).The situation is very different when we consider more strongly correlated materials, (systems containing f and d electrons).In a very simplified view LDA can be regarded as a Hartree-Fock approximation with orbital-independent(averaged) one-electron potential.This approximation is very crude for strongly correlated systems,where the on-cite Coulomb interaction between d-(or f-)electrons of transition metal(or rare-earth metal)ions(Coulomb parameter U)is strong enough to overcome kinetic energy which is of the order of band width W.In the result LDA gives qualitatively wrong answer even for such simple systems as Mott insulators with integer number of electrons per cite(so-called”undoped Mott insulators”).For example insulators CoO and La2CuO4are predicted to be metallic by LDA.The search for a”first principle”computational scheme of physical proper-ties of strongly correlated electron systems which is as successful as the LDA in weakly correlated systems,is highly desirable given the considerable impor-tance of this class of materials and is a subject of intensive current research. Notable examples offirst principle schemes that have been applied to srongly correlated electron systems are the LDA+U method[3]which is akin to orbital-spin-unrestricted Hartree-Fock method using a basis of LDA wave functions,ab initio unrestricted Hartree Fock calculations[4]and the use of constrained LDA to derive model parameters of model hamiltonians which are then treated by exact diagonalization of small clusters or other approximations[5].Many interesting effects,such as orbital and charge ordering in transition metal compounds were successfully described by LDA+U method[6].However for strongly correlated metals Hartree-Fock approximation is too crude and more sophisticated approaches are needed.Recently the dynamical mean-field theory was developed[7]which is based on the mapping of lattice models onto quantum impurity models subject to a self-consistency condition.The resulting impurity model can be solved by var-ious approaches(Quantum Monte Carlo,exact diagonalization)but the most promising for the possible use in”realistic”calculation scheme is Iterated Per-turbation Theory(IPT)approximation,which was proved to give results in a good agreement with more rigorous methods.This paper is thefirst in a series where we plan to integrate recent devel-ompements of the dynamical meanfield approach with state of the art band structure calculation techniques to generate an”ab initio”scheme for the cal-culation of the electronic structure of correlated solids.For a review of the historical development of the dynamical meanfield approach in its various im-plementations see ref[7].In this paper we implement the dynamical mean-field theory in the iterated perturbation theory approximation,and carry out the band structure calculations using a LMTO basis.The calculational scheme is described in section2.We present results obtained applying this method to La1−x Sr x TiO3which is a classical example of strongly correlated metal.22The calculation schemeIn order to be able to implement the achievements of Hubbard model theory to LDA one needs the method which could be mapped on tight-binding model.The Linearized Muffin-Tin Orbitals(LMTO)method in orthogonal approximation[8]can be naturally presented as tight-binding calculation scheme (in real space representation):H LMT O= ilm,jl′m′,σ(δilm,jl′m′ǫil n ilmσ+t ilm,jl′m′ c†ilmσ c jl′m′σ)(1)(i-site index,lm-orbital indexes).As we have mentioned above,LDA one-electron potential is orbital-inde-pendent and Coulomb interaction between d-electrons is taken into account in this potential in an averaged way.In order to generalize this Hamiltonian by including Coulomb correlations,one must add interaction term:1H int=Un d(n d−1)(3)2(n d= mσn mσtotal number of d-electrons).3In LDA-Hamiltonianǫd has a meaning of the LDA-one-electron eigenvalue for d-orbitals.It is known that in LDA eigenvalue is the derivative of the total energy over the occupancy of the orbital:ǫd=ddn d (E LDA−E Coul)=ǫd−U(n d−12)(7)(q is an index of the atom in the elementary unit cell).In the dynamical mean-field theory the effect of Coulomb correlation is de-scribed by self-energy operator in local approximation.The Green function is:G qlm,q′l′m′(iω)=1The chemical potential of the effective medium µis varied to satisfy Luttinger theorem condition:1d(iωn)Σ(iωn)=0(11)In iterated perturbation theory approximation the anzatz for the self-energy is based on the second order perturbation theory term calculated with”bath”Green function G0:Σ0(iωs)=−(N−1)U21kT,Matsubara frequenciesωs=(2s+1)πβ;s,n integer numbers.The termΣ0is renormalized to insure correct atomic limit:Σ(iω)=Un(N−1)+AΣ0(iω)β iωn e iωn0+G(iωn)),B=U[1−(N−1)n]−µ+ µn0(1−n0)(15)n0=1iω+µ−∆(iω)+δµ+n(N−1)β iωn e iωn0+G CP A(iωn)(18) D[n]=n iωn e iωn0+1energy to time variables and back:G0(τ)=1V Bd k[z−H(k,z)]−1(24)After diagonalization,H(k,z)matrix can be expressed through diagonal matrix of its eigenvalues D(k,z)and eigenvectors matrix U(k,z):H(k,z)=U(k,z)D(k,z)U−1(k,z)(25) and Green function:G(z)=1V Bd k U in(k,z)U−1nj(k,z)V Bvd kU in(k,z)U−1nj(k,z)V B(28)6v is tetrahedron volumer n i=(z−D n(k i,z))2k(=j)(D n(k k,z)−D n(k j,z))ln[(z−D n(k j,z))/(z−D n(k i,z)]1+a2(z−z2)1(30)where the coefficients a i are to be determined so that:C M(z i)=u i,i=1,...,M(31) The coefficients a i are then given by the recursion:a i=g i(z i),g1(z i)=u i,i=1,...,M(32)g p(z)=g p−1(z p−1)−g p−1(z)3ResultsWe have applied the above described calculation scheme to the doped Mott insulator La1−x Sr x TiO3is a Pauli-paramagnetic metal at room tem-perature and below T N=125K antiferromagnetic insulator with a very small gap value(0.2eV).Doping by a very small value of Sr(few percent)leads to the transition to paramagnetic metal with a large effective mass.As photoemission spectra of this system also show strong deviation from the noninteracting elec-trons picture,La1−x Sr x TiO3is regarded as an example of strongly correlated metal.The crystal structure of LaTiO3is slightly distorted cubic perovskite.The Ti ions have octahedral coordination of oxygen ions and t2g-e g crystalfield splitting of d-shell is strong enough to survive in solid.On Fig.1the total and partial DOS of paramagnetic LaTiO3are presented as obtained in LDA calculations (LMTO method).On3eV above O2p-band there is Ti-3d-band splitted on t2g and e g subbands which are well separated from each other.Ti4+-ions have d1 configuration and t2g band is1/6filled.As only t2g band is partiallyfilled and e g band is completely empty,it is reasonable to consider Coulomb correlations between t2g−electrons only and degeneracy factor N in Eq.(12)is equal6.The value of Coulomb parameter U was calculated by the supercell procedure[9]regarding only t2g−electrons as localized ones and allowing e g−electrons participate in the screening.This cal-culation resulted in a value3eV.As the localization must lead to the energy gap between electrons with the same spin,the effective Coulomb interaction will be reduced by the value of exchange parameter J=1eV.So we have used effective Coulomb parameter U eff=2eV.The results of the calculation for x=0.06(dop-ing by Sr was immitated by the decreasing on x the total number of electrons) are presented in the form of the t2g-DOS on Fig.2.Its general form is the same as for model calculations:strong quasiparticle peak on the Fermi energy and incoherent subbands below and above it corresponding to the lower and upper Hubbard bands.The appearance of the incoherent lower Hubbard band in our DOS leads to qualitatively better agreement with photoemission spectra.On Fig.3the exper-imental XPS for La1−x Sr x TiO3(x=0.06)[12]is presented with non-interacting (LDA)and interacting(IPT)occupied DOS broadened to imitate experimental resolution.The main correlation effect:simultaneous presence of coherent and incoherent band in XPS is successfully reproduced in IPT calculation.However, as one can see,IPT overestimates the strength of the coherent subband.4ConclusionsIn this publication we described how one can interface methods for realistic band structure calculations with the recently developed dynamical meanfield8technique to obtain a fully”ab initio”method for calculating the electronic spectra of solids.With respect to earlier calculations,this work introduces several method-ological advances:the dynamical meanfield equations are incorporated into a realistic electronic structure calculation scheme,with parameters obtained from afirst principle calculation and with the realistic orbital degeneracy of the compound.To check our method we applied to doped titanates for which a large body of model calculation studies using dynamical meanfield theory exists.There results are very encouraging considering the experimental uncertainties of the analysis of the photoemission spectra of these compounds.We have used two relative accurate(but still approximate)methods for the solution of the band structure aspect and the correlation aspects of this problem:the LMTO in the ASA approximation and the IPT approximation. In principle,one can use other techniques for handling these two aspects of the problem and further application to more complicated materials are necessary to determine the degree of quantitative accuracy of the method.9References[1]Hohenberg P.and Kohn W.,Phys.Rev.B136,864(1964);Kohn W.andSham L.J.,ibid.140,A1133(1965)[2]R.O.Jones,O.Gunnarsson,Reviews of Modern Physics,v61,689(1989)[3]Anisimov V.I.,Zaanen J.and Andersen O.K.,Phys.Rev.B44,943(1991)[4]S.Massida,M.Posternak, A.Baldareschi,Phys.Rev.B46,11705(1992);M.D.Towler,N.L.Allan,N.M.Harrison,V.R.Sunders,W.C.Mackrodt,E.Apra,Phys.Rev.B50,5041(1994);[5]M.S.Hybertsen,M.Schlueter,N.Christensen,Phys.Rev.B39,9028(1989);[6]Anisimov V.I.,Aryasetiawan F.and Lichtenstein A.I.,J.Phys.:Condens.Matter9,767(1997)[7]Georges A.,Kotliar G.,Krauth W.and Rozenberg M.J.,Reviews of ModernPhysics,v68,n.1,13(1996)[8]O.K.Andersen,Phys.Rev.B12,3060(1975);Gunnarsson O.,Jepsen O.andAndersen O.K.,Phys.Rev.B27,7144(1983)[9]Anisimov V.I.and Gunnarsson O.,Phys.Rev.B43,7570(1991)[10]Lambin Ph.and Vigneron J.P.,Phys.Rev.B29,3430(1984)[11]Vidberg H.J.and Serene J.W.,Journal of Low Temperature Physics,v29,179(1977)[12]A.Fujimori,I.Hase,H.Namatame,Y.Fujishima,Y.Tokura,H.Eisaki,S.Uchida,K.Takegahara,F.M.F de Groot,Phys.Rev.Lett.69,1796(1992).(Actually in this article the chemical formula of the sample was LaTiO3.03, but the excess of oxygen produce6%holes which is equivalent to doping of 6%Sr).105Figure captionsFig.1.Noninteracting(U=0)total and partial density of states(DOS)for LaTiO3.Fig.2.Partial(t2g)DOS obtained in IPT calculations in comparison with noninteracting DOS.Fig.3.Experimental and theoretical photoemission spectra of La1−x Sr x TiO3 (x=0.06).11)LJ 7L G H J '26 V W D W H H 9 D W R P (QHUJ\ H97L G W J7RWDO /D7L2 '26 V W D W H H 9 F H O O3HUWXUEDWHG)LJ'26 V W D W H V H 9 (QHUJ\ H98QSHUWXUEDWHG,Q W H Q V L W \ H 9(QHUJ\ H9。

Introduction to Computational Chemistry (2)

Introduction to Computational Chemistry (2)

• What can we predict with modern Ab Initio methods?
– Geometry of a molecule – Dipole moment – Energy of reaction – Reaction barrier height – Vibrational frequencies – IR spectra – NMR spectra – Reaction rate – Partition function – Free energy – Any physical observable of a small molecule
Born-Oppenheimer Approximation
• The potential surface is a Born-Oppenheimer potentials surface, where the potential energy is a function of geometry. Motion of the nuclei is assumed to be independent of the motion of the electrons
– There is an enormous toolbox of theoretical methods available, and it will take skill and creativity to solve real-world problems.
Electronic Structure Theory
! $ =
c e"#ir2 i
i
Electronic Structure Theory
• A plane-wave basis set is a common choice for predicting properties of a crystal

Ab-Initio Calculations of Materials Properties:材料性质的从头计算

Ab-Initio Calculations of Materials Properties:材料性质的从头计算

Ab-Initio Calculations of Materials PropertiesRLE Groupab-initio Physics GroupAcademic and Research StaffProfessor John D. JoannopoulosProfessor Marin SoljacicGraduate StudentsK.C. HuangMatthew EvansPostdoctoral FellowsEvan ReedTechnical and Support StaffMargaret O’MearaIntroductionPredicting and understanding the properties and behavior of real materials systems is of great importance both from technological and academic points of view. The theoretical problems associated with these systems are, of course, quite complex. However, we are currently at the forefront of beginning to overcome many of these problems. Our research is devoted to creating a realistic microscopic quantum mechanical description of the properties of real material systems. In the past, theoretical attempts to deduce microscopic electronic and geometric structure have been generally based on optimizing a geometry to fit known experimental data. Our approach is more fundamental: predicting geometric, electronic, and dynamical structure, ab-initio — that is, given only the atomic numbers of the constituent atoms as experimental input. Briefly, our method makes it possible to accurately and efficiently calculate the total energy of a solid by the use of density functional theory, pseudopotential theory and a conjugate gradients iterative minimization technique for relaxing the electronic and nuclear coordinates. Ab-initio investigations are invaluable because they make possible theoretical calculations or simulations that can stand on their own. They may complement experimental observations but need not be guided by experimental interpretations. Our objective is to obtain a fundamental, microscopic understanding of various physical and chemical phenomena of real materials systems.The Nature of Boron NanotubesSponsorsU.S. Department of EnergyDE-FG02-99ER45778Project StaffMatthew Evans and J. JoannopoulosBoron, carbon’s first-row neighbor, has only three valence electrons. Its natural crystalline structure is a rhombohedral lattice with 12-atom icosahedral clusters at each lattice site.[1] Nevertheless, there are some intriguing similarities with carbon. Boron’s three electrons could in principle form sp2 hybrid orbitals that might lead to planar and tubular structures similar to thoseformed by carbon. Since carbon nanotubes and fullerenes[2] are metastable structures, formed only under kinetically constrained conditions,[3] one might envision analogous boron structures. Indeed, initial resuts by Boustani et al.[4,5] have demonstrated the possibility of such metastable structures with relatively low energy cost. Crystalline[6] and amorphous[7] boron nanowires with diameters as small as 20 nm have recently been fabricated, suggesting that boron nanotubes may already be within the range of experimental possibility.Figure 1. The (4,4) boron nanotube. (a) Ball and stick structural model. (b) A valence electron density isosurface of 0.67 electrons/A3. The ion cores are shown as dark spheres.There is an intriguing and potentially significant difference between carbon and boron however. Boron has only three valence electrons, so that in sp2-bonded planar or tubular boron structures the relative occupations of the sp2- and the π-bonded bands depend on the energetic positions and dispersions of the two bands, perhaps opening up a broader range of possibilities.Using an ab-initio total-energy density functional approach we have examined in detail the electronic structure and relative stabilities of planar and tubular boron structures. We find that boron does form a stable sp2-bonded hexagonal graphene-like sheet, but a planar triangular lattice has an even larger cohesive energy, though still smaller than that of the bulk α-rhombohedral structure. The triangular planar structure has an unusual property. It is essentially a homogeneous electron gas system with a threefold-degenerate ground state. This degeneracy makes the flat triangular plane unstable with respect to buckling, which breaks the symmetry and introduces a preferred direction defined by strong σ bonds. When rolled into a tube, this preferred direction, which is not present in carbon nanotubes, defines the chirality and controls the electron density, cohesive energy, and elastic response of boron nanotubes. The properties of the (n,0) tubes arise from the flat plane and are very similar to carbon nanotubes. The properties of the (n,n) boron nanotubes are derived from the buckled plane and contrast sharply with carbon nanotube structures. As a result of the buckling, the curvature energies of (n,n) tubes are lower than those of (n,0) tubes and show a nonmonotonic plateau structure as a function of n. The contrast in the electron densities of the buckled and flat triangular planes also explains the differing elastic responses of the (n,n) and (n,0) tubes.We define the chirality of a boron nanotube based on a triangular lattice. An (m,n) tube is constructed by rolling a triangular plane such that the head of the lattice vector m a + n b meets its tail; a and b are the primitive vectors of the triangular lattice.Consider first the case of an (n,n) nanotube, shown in Figure 1(a). Here, the σ bond direction can be chosen to lie along the length of the tube. Given this possibility, the boron atoms will form σbonds running along the nanotube, as the electron density isosurface in Figure 1(b) shows. The strong longitudinal bonds allow the tube to buckle laterally, emulating the buckled plane structure.Instead of the circular cross section seen in carbon nanotubes, the (8,8) tube has a square cross section. The sides of the square are sections of the buckled plane, and the corners have only a slight distortion. In contrast, the cross section of the (6,6) tube shows no buckling. With only four atoms on each side, it is not possible to buckle the sides without distorting the topology of the corners. A “buckled” structure of the (6,6) tube breaks the mirror symmetries of the actual (6,6) and (8,8) structures.Table I. Summary of the cohesive energy E coh , the curvature energy with respect to the buckled plane E curv , the equilibrium diameter d, the modified Young’s modulus Y s (see Equation 1), and the Poisson ratio σ for boron nanotubes.Chirality E coh (ev) E curv (eV) D(Å) Y s (Tpa nm) σ (4,4) 6.71 0.08 4.34 0.29 0.5(6,6) 6.65 0.14 5.65 0.15 0.4(8,8) 6.76 0.03 8.48 0.22 0.1(7,0) 6.36 0.43 3.99 0.49 0.2(8,0) 6.39 0.40 4.62 0.49 0.1Larger diameter tubes sharing the favorable buckled structure of the (8,8) tube can be constructed by adding atoms to the sides of the square in pairs. This implies that among the (n,n) boron nanotubes, (4n,4n) tubes should have lower curvature energies and be more stable. This trend is suggested by the cohesive and curvature energies summarized in Table I. In the table, the curvature energy is defined as the difference in cohesive energy between the tube and the plane: E curve ≡E coh tube −E coh planeAn (n,0) nanotube cannot align σ bonds longitudinally (see Figure 2(a)). Although buckling and selecting a σ bond direction proves energetically favorable in the (n,n) tubes, it is not required by symmetry. Buckling is necessary to break the symmetry of the flat plane, but rolling up the plane into an achiral tube breaks the degeneracy automatically. The threefold planar degeneracy reduces to a twofold degeneracy (spiraling σ bonds related by chiral symmetry) and a nondegenerate state (σ bonds running longitudinally or laterally). It is possible for σ bonds to run along the circumference of an (n,0) tube, but the electron density of an (8,0) nanotube, shown in Figure 2(b), shows no such bonds. Instead, the density of (n,0) boron nanotubes is nearly uniform, exhibiting the free electron character of the flat triangular plane. The energetic consequences are apparent. The curvature energies of (n,0) boron nanotubes lie 0.25 – 0.4 eV above those of the (n,n) tubes; this is roughly the same as the 0.26 eV cohesive energy difference between the flat and buckled triangular planes. Between the achiral limits, there may be a critical chiral angle at which boron nanotubes switch from the σ bond dominated electronic structure of the buckled plane to the free-electron-like structure of the flat plane. This could have important consequences for the behavior of boron nanotubes under torsion.Elastic properties of boron nanotubes exhibit a strong chirality dependence as well. Among the most important characteristics of a cylindrical object is the Young’s modulus Y s , defined for single walled tubes as:Y s =1S 0∂2E ∂ε2⎛ ⎝ ⎜ ⎜ ⎞ ⎠ ⎟ ⎟ ε=0 (1)where S0 is the equilibrium surface area, E is the total energy, and ε is the longitudinal strain. Since the walls of our boron nanotube are only a single atom thick, it is not possible to define the tube volume, and we must use this modified Young’s modulus. For the (n,n) tubes with square cross sections, we define the diameter as the diagonal of the square, and we ignore buckling in calculating the surface area.Figure 2. The (8,0) boron nanotube. (a) Ball and stick structural model. (b) A valence electron densityisosurface of 0.67 electrons/A3. The ion cores are shown as dark spheresAnother important elastic property of a tube is the Poisson ratio σ:d−d eqd eq=−σε (2)where d is the tube diameter at strain ε, and d eq is the equilibrium tube diameter. The Poissonratio measures the change in the tube’s radius as it is strained longitudinally. The combination ofthe Young’s modulus and the Poisson ratio provides information about the strength of both thelongitudinal and lateral bonds of the nanotube.We have calculated the Young’s modulus Y s and Poisson ratio σ for the (n,n) and (n,0) boronnanotubes discussed above. As with the binding energy, chirality plays a crucial role indetermining these properties. As is clear from the electron density isosurface in Figure 1b, thestrength of the (n,n) boron nanotubes arises from the bonds running along the length of the tube.Since the lateral bonds are comparatively weak, we would expect that an (n,n) nanotube will beable to expand and contract circumferentially to relieve stress, leading to a high Poisson ratio anda low Young’s modulus. This is indeed the case for the small-diameter (4,4) and (6,6) tubes.For the (4,4) tube, Y s = 0.29 TPa nm and σ = 0.5. The Young’s modulus is comparable to boronnitride nanotubes and roughly half that of carbon nanotubes, while the Poisson ratio is nearlytwice as large as the value for either carbon or boron nitride nanotubes.For the (8,8) boron nanotube, however, structural dynamics play a more important role. Usingthe previous definition of the Poisson ratio, we find that σ = 0.1, considerably smaller than thePoisson ratio of the (4,4) and (6,6) tubes. The diameter of the (8,8) tube does not changesignificantly with strain, but there is considerable lateral relaxation that results in a low Young’smodulus. The simple square cross sections of the (4,4) and (6,6) tubes, as well as the circularcross sections of the (n,0) tubes, permit only uniform lateral dilation and contraction if thesymmetry of the structure is to be maintained. The buckled sides of the (8,8) (and larger (4n,4n)tubes, as discussed previously) permit tube walls to relax without changing the overall squarestructure.Although the (8,8) boron nanotube does not change its diameter under longitudinal strain, the comparatively weak lateral bonds do allow for relaxation of the buckled tube sides. Buckling introduces a third dimension into the planar structure, enabling strain to be relaxed in an internal degree of freedom that does not break the planar symmetry. As the buckled plane is stretched, atoms can move perpendicular to the plane to relieve stress, an option that is forbidden by symmetry in flat planar structures. This is analogous to the fundamental difference between the phonon modes of a monatomic Bravais lattice and those of a lattice with a basis: in the latter, optical modes are present that allow relative motion without a net translation of the crystal.In contrast to the (n,n) nanotubes, the (n,0) tubes have no dominant bonding direction. Straining the tube stresses bonds both laterally and longitudinally, making it difficult for the tube to expand or contract circumferentially. An increased Young’s modulus and a decreased Poisson ratio reflect this cost. For the (8,0) boron nanotube, Y s = 0.49 Tpa nm and σ = 0.1. Although the radius of the (8,0) tube is only 6% larger than that of the (4,4) tube, the Young’s modulus is 68% larger. This is in sharp contrast to carbon nanotubes, where tubes of similar radius have Young’s moduli that differ by only a few percent.References1. A.E. Newkirk, in Boron, Metallo-Boron Comounds and Boranes, edited by R.M. Adams(Wiley, New York, 1964).2. M.S. Dresselhaus, G. Dresselhaus, and P.C. Eklund, Science of Fullerenes and CarbonNanotubes, (Academic, San Diego, 1996).3. T.W. Ebbesen and P.M. Ajayan, Nature 358, 220 (1992); A. Thess, R. Lee, P. Nikolaev, H.Dai, P. Petit, J. Robert, C. Xu, Y.H. Lee, S.G. Kim, A.G. Rinzler, D.T. Colbert, G.E. Scuseria,D. Tománek, J.E. Fischer, and R.E. Smalley, “Crystalline Ropes of Metallic CarbonNanotubes,” Science 273, 483-487 (1996).4. I. Boustani, A. Quandt, and A. Rubio, “Boron Quasicrystals and Boron Nanotubes: Ab InitioStudy of Various B96 Isomers,” J. Solid State Chem. 154, 269-274 (2000).5. I. Boustani, A. Quandt, E. Hernández, and A. Rubio, “New Boron Based NanostructuredMaterials,” J. Chem. Phys. 110, 3176-3185 (1999).6. C.J. Otten, O.R. Lourie, M.-F. Yu, J.M. Cowley, M.J. Dyer, R.S. Ruoff, and W.E. Buhro,“Crystalline Boron Nanowires,” J.Am. Chem. Soc. 124, 4564-4565 (2002).7. J. Cao, J. Liu, C. Gao, Y. Li, Y.Q. Wang, Z. Zhang, Q. Cui, G. Zou, L. Sun, and W. Wang,“Synthesis of Well-aligned Boron Nanowires and Their Structural Stability under High Pressure,” J.Phys. Condens. Matter 14, 11017-11021 (2002); Y.Q. Wang and X.F. Duan, “Amorphous Feather-like Boron Nanowires,” Chem. Phys. Lett. 367, 495-499 (2003); X.M.Meng, J.Q. Hu, Y. Jiang, C.S. Lee, and S.T. Lee, “Boron Nanowires Synthesized by Laser Ablation at High Temperature,” Chem. Phys. Lett. 370, 825-828 (2003).PublicationsJournal Articles PublishedPeter Bermel, J.D. Joannopoulos, Yoel Fink, Paul A. Lane, and Charles Tapalian, “Properties of Radiating Pointlike Sources in Cylindrical Omnidirectionally Reflecting Waveguides,”Phys. Rev. B 69, 035316-1–035316-7, (2004).S.G. Johnson and J.D. Joannopoulos, “Electromagnetic Theory of Photomic Crystals,”Encyclopedia of Modern Optics , ed. R. D. Guenther. D. G. Steel and L. Bayvel, Elsevier, Oxford (2004).S.G. Johnson, M. Sojacic and, J.D. Joannopoulos, “Photonic Crystals for Enhancemnet of Nonlinear Effects,” Encyclopedia of Nonlinear Science, ed. Alwyn Scott. New York and London: Routledge, 2004.Chiyan Luo, Marin Soljacic, and J.D. Joannopoulos, “Superprism Effect Based on Phase Velocities,” Optics Letters 29, 745-747 (2004).Elefterios Lidorikis, Marin Soljacic, Mihai Ibanescu, Yoel Fink, and J.D. Joannopoulos, “Cutoff Solitons in Axially Uniform Systems,” Optics Letters 29, 851-853 (2004).Ian Applebaum, Tairan Wang, J.D. Joannopoulos, and V. Narayanamurti, “Ballistic Hot Electron Transport in Nanoscale Semiconductor Heterostructures: Exact Self-Energy of Three-dimensional Periodic Tight-Binding Hamiltonian,” Phys. Rev. B 69, 165301-1–165301-6(2004).Kerwyn Casey Huang, Elefterios Lidorikis, Xunya Jiang, John D. Joannopoulos, Keith A.Nelson, Peter Bienstman, and Shanhui Fan, “Nature of Lossy Bloch States in Polaritonic Photonic Crystals,” Phys. Rev. B 69, 195111-1–195111-10 (2004).Marin Soljacic and J.D. Joannopoulos, “Enhancement of Non-linear Effects using Photonic Crystals,” Nature Materials 3, 211-219 (2004).M. Ibanescu, S.G. Johnson, D. Roundy, C. Luo, Y. Fink, and J.D. Joannopoulos, “Anomalous Dispersion Relations by Symmetry Breaking in Axially Uniform Waveguides,” Phys. Rev.Lett. 92, 063903-1–063903-4 (2004).M.L. Povinelli, Steven G. Johnson, Elefterios Lidorikis, J.D. Joannopoulos, and Marin Soljacic, “Effect of a Photonic-band Gap on Scattering from Waveguide Disorder,” Appl. Phys.Lett. 84, 3639-3641 (2004).David Roundy, Elefterios Lidorikis, and J.D. Joannopoulos, “Polarization-selective Waveguide Bends in a Photonic Crystal Structure with Layered Square Symmetry,” J. Appl. Phys. 96, 7750-7752 (2004).Kerwyn Casey Huang, M.L. Povinelli, and John D. Joannopoulos, “Negative Effective Permeability in Polaritonic Crystals,” Appl. Phys. Lett. 85, 543-545 (2004).X. Jiang, S. Feng, C.M. Soukoulis, J. Zi, J.D. Joannopoulos, and H. Cao “Coupling, Competition, and Stability of Modes in Random Lasers,” Phys. Rev. B 69, 104202-1– 104202-7 (2004).Marin Soljacic, Elefterios Lidorikis, Mihai Ibanescu, Steven G. Johnson, and J.D.Joannopoulos, "Optical Bistability and Cutoff Solitons in Photonic Bandgap Fibers,"Optics Express 12, 1518-1527 (2004).E. Reed, M. Soljacic, M. Ibanescu and J. Joannopoulos, “Comment on Observation of theInverse Doppler Effect,” Science 305, 778 (2004).M. Qi, E. Lidorikis, P. Rakich, S.G. Johnson, J.D. Joannopoulos, E.P. Ippen and H.I. Smith, “A 3D Optical Photonic Crystal with Designed Point Defects,” Nature 429, 538-542 (2004).M. Bayindir, F. Sorin, A.F. Abouraddy, J. Viens, S. Hart, J.D. Joannopoulos and Y. Fink “Metal-Insulator-Semiconductor Optoelectronic Fibers,” Nature 431, 826-829 (2004).K. Kuriki, O. Shapira, S. Hart, G. Benoit, M. Bayindir, J. Joannopoulos and Y. Fink, “Hollow Multilayer Photonic Bandgap Fibers for Near IR Applications,” Optics Express 12, 1510-1517 (2004).Chiyan Luo, Arvind Narayanaswamy, Gang Chen and J.D. Joannopoulos, “Thermal Radiation for Photonic Crystals: A Direct Calculation,” Phys. Rev. Lett. 93, 213905-1–213905-4 (2004).S. Assefa, P.T. Rakich, P. Bienstman, S.G. Johnson, G.S. Petrich, J.D. Joannopoulos, L. A.Kolodziejski, E. P. Ippen, and H. I. Smith " Guiding 1.5µm Light in Photonic Crystals Based on Dielectric Rods," Appl. Phys. Lett. 85, 6110-6112 (2004).。

从头算法

从头算法
假定1: 有效作用势近似
假定2: 周期性边界条件
(Periodical Boudary Condition)
• 困难
— 欲重现实际体系的统计行为,模拟体 系应有足够数量的粒子
1dm3水 31027个H2O 计算机只能处理 有限 个粒子!
• 解决办法 — 赝无限大近似
取较小的模拟体系作中心原胞,令其在空间 重复排列
因处理实际分子在数学上的困难。Dirac本 人对量子力学在化学上的应用前景十分悲观
•பைடு நூலகம்•
1952年H. Schull等三人用手摇计算机花两年
才完成一个N2分子的从头算。
有人断言:用尽世界上的纸张恐亦无法完成
一个Fe原子的计算
50年代末,大型计算机的浮点运算速度为 0.01Mflops,仅及PIII 的 1/5000 !
Solvation: 60 Å box of waters 7209 waters
+ 15 Na+ ions 21468 atoms
*
Sawaya and Kraut 1997.
例1
(1,2,4)-4-(1,1-e 二 甲 乙 基 )-2- 烃 基 环戊羰基酰胺晶体
实验测定与 MD 模拟结果的比较
3
rQj
RPQ
4
• 粒子间存在着相互
作用
A
Q
B
定态Shrö dinger方程:
二、物理模型的三个基本近似
⒈ 非相对论近似: i = 0 ⒉ Born-Oppenheimer 近似: 电子与核运动分离 电子 哈密顿: 原子单位:三个基本物理常数 ⒊ 单电子近似 — 每个电子行为视为独立,用单电子波函 数 i (ri) 描述 (MO)

第一节第一性原理计算方法.

第一节第一性原理计算方法.

第一性原理计算的理论方法随着科技的发展,计算机性能也得到了飞速的提高,人们对物理理论的认识也更加的深入,利用计算机模拟对材料进行设计已经成为现代科学研究不可缺少的研究手段。

这主要是因为在许多情况下计算机模拟比实验更快、更省,还得意于计算机模拟可以预测一些当前实验水平难以达到的情况。

然而在众多的模拟方法中,第一性原理计算凭借其独特的精度和无需经验参数而得到众多研究人员的青睐,成为计算材料学的重要基础和核心计算。

本章将介绍第一性原理计算的理论基础,研究方法和ABINIT软件包。

1.1 第一性原理第一性原理计算( 简称从头计算,the abinitio calculation),指从所要研究的材料的原子组分出发,运用量子力学及其它物理规律,通过自洽计算来确定指定材料的几何结构、电子结构、热力学性质和光学性质等材料物性的方法。

基本思想是将多原子构成的实际体系理解成为只有电子和原子核组成的多粒子系统,运用量子力学等最基本的物理原理最大限度的对问题进行”非经验”处理。

【1】第一性原理计算就只需要用到五个最基本的物理常量即( m o.e.h.c.k b ) 和元素周期表中各组分元素的电子结构,就可以合理地预测材料的许多物理性质。

用第一性原理计算的晶胞大小和实验值相比误差只有几个百分点,其他性质也和实验结果比较吻合,体现了该理论的正确性。

第一性原理计算按照如下三个基本假设把问题简化:1.利用Born-Oppenheimer 绝热近似把包含原子核和电子的多粒子问题转化为多电子问题。

2.利用密度泛函理论的单电子近似把多电子薛定谔方程简化为比较容易求解的单电子方程。

3.利用自洽迭代法求解单电子方程得到系统基态和其他性质。

以下我将简单介绍这些第一性原理计算的理论基础和实现方法:绝热近似、密度泛函理论、局域密度近似(LDA)和广义梯度近似(GGA)、平面波及赝势方法、密度泛函的微扰理论、热力学计算方法和第一性原理计算程序包ABINIT。

  1. 1、下载文档前请自行甄别文档内容的完整性,平台不提供额外的编辑、内容补充、找答案等附加服务。
  2. 2、"仅部分预览"的文档,不可在线预览部分如存在完整性等问题,可反馈申请退款(可完整预览的文档不适用该条件!)。
  3. 3、如文档侵犯您的权益,请联系客服反馈,我们会尽快为您处理(人工客服工作时间:9:00-18:30)。

I. Introduction
The interest in the electronic and optical properties of semiconducting π-conjugated polymers, such as the polyparaphenylene (PPP) and the poly-paraphenylene-vinylene (PPV) has considerably increased due to the necessity on the improvement of the efficiency of the new polymerbased optoelectronic devices, such as LEDs and flat plane displays. This efficiency is determined by the fraction of injected electrons and holes that recombine to form emissive spin-singlet states from the excitons rather than non-emissive spin-triplet states [1]. In conventional semiconductors such as Si, the Wannier excitons are well defined as weakly bound electron-hole pairs. In organic crystals such as anthracene, on the contrary, the Frenkel exciton is essentially confined to a single molecule leading to high binding energies. However, in the π-conjugated polymers, these excitations were in the intermediate case, current subject of debate in the academic community, because it is very hard to describe the vibronic states in the theoretical treatment of the excitations of these systems [2]. Besides that, accurate experimental data for the phonon modes on these materials available in the literature are very scarce: only the strongest and the medium active modes were assigned by infrared and Raman experiments [3-8]. Moreover, on the theoretical side, only simplified models, based on the linear chain approaches, were used to explain the observed experimental data [9-11]. In order to supply the missing information on the vibrational properties of PPP, in both the Pbam and Pnnm structures, and of PPV in the P21/c crystalline symmetry,
In this work, we have calculated the vibrational modes and frequencies of the crystalline PPP (in both the Pbam and Pnnm symmetries) and PPV (in the P21/c symmetry). Our results are in good agreement with the available experimental data. Also, we have calculated the temperature dependence of their specific heats at constant volume, and of their vibrational entropies. Based on our results, at high temperatures, the PPP is more stable in the Pnnm structure than in the Pbam one.
Pnnm crystalline symmetries, respectively, at selected points of the Brillouin zone. The main difference among these two structures (Pbam and Pnnm) is a shift of one molecular chain, related to the other, along the chain direction by an amount of c/2. From our calculated structural parameters, there is almost no remarkable difference between these two modifications [14]. Moreover, despite the fact that our results show that the Pnnm is 0.54 meV more stable than the Pbam, from the accuracy of our calculations, 1 meV, we can only infer that both structures can coexist at low temperatures.

we have calculated, in this work, by using the Density Functional Theory (DFT) within the Local Density Approximation (LDA), plane-wave description of the wave functions and the pseudopotential method (ABINIT code) [12], their frequencies and the correspondent vibrational modes. In our calculations, we have used the TroullierMartins pseudopotentials (evaluated by the fhi98PP code [13]), and the phonons were obtained by means of the Density-Functional Perturbation Theory. Details about the structural properties of these systems, as well as the vibrational modes, obtained by simplified models, are described in our previous work [14, 15]. It is well known that while bonding among the atoms within the polymer chains mainly is of covalent character, interactions between the polymer chains are governed by weak van der Waals (vdW) forces. In this case, the common LDA and GGA fail to describe this weak nonlocal force, mainly in noble gases [16]. However, correlation functionals, like LDA and GGA, contract the density towards the bond and valence region thus taking negative charge out of the vdW region. Moreover, for large interchain separation d, the vdW interaction energy between chains behaves as d-4, while LDA, or GGA, predicts an exponential falloff. In polymers, as also observed in fullerenes, it behaves as d-6 [17]. So, the LDA is a good choice to describe these systems, as already shown in previous works for both PPP and PPV [18, 19].
structures. However, after a careful analysis of both evaluated DOS, we have noted that there are shifts in some peaks as follows: i) the peak at 1500 cm-1 in the Pbam DOS appears at 1300 cm-1 in the Pnnm one; ii) the peak at 1050 cm-1 in the Pbam DOS appears at 800 cm-1 in the Pnnm one; and finally, iii) there is a rearrangement of the low frequency modes below 100 cm-1 between both calculated DOS. Although the differences between the two evaluated DOS are relatively small, they should be sufficient for an interesting difference in their derived thermodynamic properties. Concerning the vibrational modes, the narrow flat band, located between 3054 and 3077 cm-1 in both dispersions, corresponds to C-H stretching and bending vibrations. We have noted that the modes located at the frequency region from 1035 to 1642 cm-1 are related to the C-C stretching and bending vibrations. Also, the modes located at the frequency region from 726 to 1035 cm-1 are C-C stretching vibrations only, while the modes with frequency lower than 726 cm-1 are C-C bending vibrations.
相关文档
最新文档