24小时热门版块排行榜    

北京石油化工学院2026年研究生招生接收调剂公告
查看: 2034  |  回复: 33

xiaowandouer

木虫 (正式写手)

mscic能给我们这些没用过G4的虫子扫扫盲么?

以下是自己搜的关于G4的介绍,贴出来跟大家一起学习:

Gaussian-4 (G4) Theory

Gaussian-4 theory1 is the fourth in a series of Gx methods for calculation of molecular energies. It is a composite technique in which a sequence of well-defined ab initio molecular orbital calculations is performed to arrive at a total energy of a given molecular species. Geometries are determined using B3LYP density functional theory with the 6-31G(2df,p) basis set [6-31G(2fg) for third-row nontransition metal elements). Zero-point energies are also calculated with this basis set. Correlation level calculations are done using Moller-Plesset perturbation theory up to fourth-order and with coupled cluster theory. Large basis sets, including multiple sets of polarization functions, are used in the correlation calculations. Three basis sets that are used in G4 theory are available on this web site. These are the G3LargeXP basis used in the MP2(FU) step and two basis sets (quadrupole-zeta and quintuple-zeta) used for extrapolation to the Hartree-Fock limit. The latter two are derived from correlation consistent basis sets.

The G4 theory has been assessed on the G3/05 test set2. It has 454 energies including enthalpies of formation, ionization potentials, electron affinities, proton affinities, and hydrogen bond energies. The comprehensive set includes the energies in the original G2 test set, the G2/97 test set, and the G3/99 test set. contains 302 entries.

Basis sets for single point energy calculations: G3LargeXP, HFQZ, HF5Z

Basis set  for K,Ca,Ga-Kr for B3LYP geometry optimizations: 6-31G(2fg)

G5/03 test set: G4 total energies

G5/03 test set: G4 deviations with experiment

[ Last edited by xiaowandouer on 2010-12-31 at 11:41 ]
31楼2010-12-31 11:36:35
已阅   回复此楼   关注TA 给TA发消息 送TA红花 TA的回帖

hairan

木虫 (著名写手)

瞌睡木虫


小木虫(金币+0.5):给个红包,谢谢回帖交流
引用回帖:
Originally posted by mscic at 2010-12-31 10:59:46:



不需要直接用CCSD(T)/aug-cc-pVTZ算单点,不然就跟你说的一样,不用那么费事了。

%chk=h2.chk
# BMK/6-31+G(2df,p) Opt

Hydrogen molecule G4(MP2)-6X calculation

0 1
H 0.000000 0.000000 0.3 ...

G4是CCSD(T)+大基组计算不动时不得已而为之,CCSD(T)+大基组能算的动的时候当然要直接算
另外,用G4不如用CBS-APNO,后者计算量只是稍微大一点,精度要高一大截
据权威报纸调查,84%的青少年犯罪者有玩电子游戏的经历,故应该禁止经营电子游戏。据我们所知,100%的青少年犯罪者有穿鞋的经历,故制鞋厂应该停业。
32楼2010-12-31 11:49:52
已阅   回复此楼   关注TA 给TA发消息 送TA红花 TA的回帖

xiaowandouer

木虫 (正式写手)

引用回帖:
Originally posted by hairan at 2010-12-30 00:28:39:


这样的配置跑600多基组的MP2应该没问题。
单点的MP2计算,组里的学生跑过800多将近900的(闭壳层)
我算的最大的到480左右,但同时计算了opt和freq
配置是:Q6600 CPU/8GB 内存/500GB 硬盘
想不通你的为什 ...

还是一步都没有走下去,这是输出文件的结尾部分:

(Enter /public1/software/gaussian09/g09/l906.exe)
Frozen-core derivative calculation, NFC=  29 NFV=   0.
FulOut=F Deriv=T AODrv=T NAtomX=    40
   MMem=           0  MDisk=          65 MDiskD=          65
  W3Min=    72596576 MinDsk=    88981564 NBas6D=         632
NBas2D=      200898    NTT=      200028    LW2=     3000000
    MDV=  2142038402 MDiskM=       47029 NBas2p=      166500
Disk-based method using ON**2 memory for 65 occupieds at a time.
Permanent disk used for amplitudes=  3497027292 words.
Estimated scratch disk usage= 28997742868 words.
IMap=   1   2   3   4   5   6   7   8   9  10  11  12  13  14  15  16  17  18  19  20
IMap=  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40
IMap=  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60
IMap=  61  62  63  64  65  66  67  68  69  70  71  72  73  74  75  76  77  78  79  80
IMap=  81  82  83  84  85  86  87  88  89  90  91  92  93  94
Actual    scratch disk usage= 27260645652 words.
JobTyp=2 Pass  1:  I=  30 to  94 NPSUse= 16 ParTrn=T ParDer=T DoDerP=T.


l906: semi-direct MP2


请教:能看出出问题的原因么?
33楼2011-01-03 11:01:17
已阅   回复此楼   关注TA 给TA发消息 送TA红花 TA的回帖

xiaowandouer

木虫 (正式写手)

引用回帖:
Originally posted by hq2010 at 2010-12-29 12:31:43:
有时候你用命令查看的时候,命令首先中断了gaussian进程
是你杀死了进程,而不是进程断了


bjobs这样的命令也会把进程给中断么?你都是用什么命令查看进程呢?

普通的dft优化也没见因为查看个状态而把进程给终止的啊,mp2这么脆弱么?
34楼2011-01-03 11:04:07
已阅   回复此楼   关注TA 给TA发消息 送TA红花 TA的回帖
相关版块跳转 我要订阅楼主 xiaowandouer 的主题更新
普通表情 高级回复 (可上传附件)
最具人气热帖推荐 [查看全部] 作者 回/看 最后发表
[考研] 304求调剂(085602,过四级,一志愿985) +17 化工人999 2026-04-04 17/850 2026-04-07 02:12 by BruceLiu320
[考研] 302分求调剂 一志愿安徽大学085601 +12 zyx上岸! 2026-04-04 12/600 2026-04-07 02:09 by BruceLiu320
[考研] 283求调剂 +11 A child 2026-04-04 11/550 2026-04-06 23:05 by chenzhimin
[考研] 0703总分331求调剂 +12 ZY-05 2026-04-04 16/800 2026-04-06 17:54 by lijunpoly
[考研] 求调剂 +4 wos666 2026-04-03 5/250 2026-04-06 15:22 by wos666
[考研] 319求调剂 +3 handrui 2026-04-05 3/150 2026-04-06 09:33 by jp9609
[考研] 一志愿211,化学学硕,310分,本科重点双非,求调剂 +13 努力奋斗112 2026-04-04 13/650 2026-04-06 07:13 by jj987
[考研] 359求调剂22408 +3 123456789qw 2026-03-31 3/150 2026-04-05 10:09 by zhq0425
[考研] 311分 22408 求调剂 +3 bing_bot 2026-04-03 3/150 2026-04-05 00:43 by chongya
[考研] 一志愿上海大学生物学346 +3 上海大学346调剂 2026-04-03 3/150 2026-04-04 20:20 by dongzh2009
[考研] 一志愿北京科技大学材料工程085601,求调剂 +17 cdyw 2026-04-02 18/900 2026-04-04 11:14 by w_xuqing
[考研] 293求调剂 +5 末未mm 2026-04-02 6/300 2026-04-03 15:20 by 王保杰33
[考研] 085801 总分275 本科新能源 求调剂 +19 bradoner 2026-04-01 23/1150 2026-04-03 10:07 by linyelide
[考研] 一志愿深大085601材料工程专业(专硕)300分可以调剂去哪 +8 10160315 2026-04-02 8/400 2026-04-03 09:36 by hypershenger
[考研] 081200-11408-276学硕求调剂 +3 崔wj 2026-04-02 3/150 2026-04-02 15:06 by cal0306
[考研] 材料工程322分 +8 哈哈哈吼吼吼哈 2026-04-01 8/400 2026-04-02 11:53 by 3041
[考研] 材料调剂 +12 一样YWY 2026-04-01 12/600 2026-04-02 00:21 by 百秒光年
[考研] 材料调剂 +11 一样YWY 2026-03-31 11/550 2026-04-01 22:25 by zhouyuwinner
[考研] 复试调剂 +7 双马尾痞老板2 2026-03-31 7/350 2026-03-31 19:49 by Dyhoer
[考研] 080500-315分复试调剂 +9 上岸3821 2026-03-31 9/450 2026-03-31 17:29 by 唐沐儿
信息提示
请填处理意见