24小时热门版块排行榜

返回列表

【奖励】本帖被评价21次，作者xi2004增加金币 19.3 个

当前只显示满足指定条件的回帖，点击这里查看本话题的所有回帖

xi2004

至尊木虫 (职业作家)

CMEI: 10
应助: 125 (高中生)
金币: 16769.9
帖子: 4161
在线: 436.9小时
虫号: 354550

[资源] 关于绝对构型和Friedel对的数目的探讨

这个问题很多人问.
其实checkcif的说明里已经比较明白. 但有些虫友看英文可能不太方便.
所以我归纳一下, 供大家参考.
也请高手指正.

1. Friedel pairs 数目的确定

在.ins中加merg 3,精修后在CIF中的total reflection number
如 _reflns_number_total             2627
再把merg 3换成MERG 2,精修后在CIF中的total reflection number
如 _reflns_number_total             3762

两者相减, 得到的就是Friedel pairs 的数目.

当你利用反常散射确定绝对构型时, 在CIF中要加
  _refine_ls_abs_structure_details 'Flack (1983), XXXX Friedel pairs'

这里的XXXX 用 3762-2627 = 1035来代替.

另外,  在CIF中填加
_chemical_absolute_configuration  ad
以表明绝对构型是用反常散射确定的.

参考英文:
[An easy way to determine the number of Friedel pairs is to look at the difference between the number of unique reflections used in SHELXL when a MERG2 and MERG 3 instruction is used (MERG 3 forces Friedel pairs to be merged before use).]

2.  如何确定绝对构型

利用反常散射是很常用的一个手段.

如果你的结构中没有重于Si的元素, 又用的是Mo靶, 那么,这个方法不可靠.
如果结构中的重原子仅是  S, P or Cl,  那强烈建议用与独立点对应的所有的Friedel pair 来精修.
重原子越重, 需要的Friedel pairs的数目百分比就越少; 含中等重量的原子的结构, 建议要>50%, 这里的百分比是 friedel pairs: unique reflections.

所对应的英文是
If the structure is non-centrosymmetric with atoms HEAVIER than silicon then it is expected that Friedel pairs will be used in the refinement and that the absolute structure will be determined experimentally, even if one or more chiral centres in the molecule are already unambiguously known from the chemistry or synthesis of the compound. If the heaviest element present is S, P or Cl, then it is strongly recommended that the Friedel opposites of all symmetry-unique reflections be included in the data set. The proportion of Friedel related reflections required decreases with the increasing atomic weight of the heaviest element that is present, but for medium weight elements, it is recommended that at least 50% of the potential Friedel related reflections have been recorded. The more Friedel pairs that are present in the data set, the smaller will be the s.u. for the absolute structure parameter (Flack's x).

如果绝对构型确定了,
那么请在CIF中加以下几行

_chemical_absolute_configuration  ad
_refine_ls_abs_structure_details
'Flack (1983), 1035 Friedel pairs'
_refine_ls_abs_structure_Flack 0.092(18)

3. 如果绝对构型确定不了怎么办?

有时候, 由于没有>Si的重原子, 或Friedel pairs的数目不够多, 或者是racemic twin, 结果flack 参数没有靠近0,而且偏差大, 比如 0.49(13). 这时候, 绝对构型无法靠反常散射来确定, 那么, Friedel pairs要被合并掉, 具体如下.

1). 在.ins中加 merg 3, 精修到收敛

2).  在CIF中加
_chemical_absolute_configuration  unk

  在CIF中删去

_refine_ls_abs_structure_details  'Flack (1983), 1283 Friedel pairs'
_refine_ls_abs_structure_Flack ?

3). 要在refine special一栏说明绝对构型不确定的原因.

具体说明请看
http://journals.iucr.org/services/cif/checking/REFLT_03.html

不过要加一句, 一切以编辑的意见为准. 已经发现有的编辑在绝对构型无法确定的情况下还要求你提供flack parameter和number of firedel pairs的.

TEST_3 : Type_4
For noncentrosymmetric structures, test if the _reflns_number_total count includes any Friedel-related reflections, and estimate how may and what fraction they are of the symmetry-unique count NREF. The test is divided into "heavy atom" structures (heavier atoms than Si are present) and "light atom" structures. The light-atom test includes the radiation used in order to establish if anomalous scattering is sufficient to permit the reliable determination of the absolute structure.

IF Z > Si .and. NFDRAT < 0.5 issue a General ALERT
"WARNING: Large fraction of Friedel related reflns needed to determine absolute structure."
If the structure is non-centrosymmetric with atoms HEAVIER than silicon then it is expected that Friedel pairs will be used in the refinement and that the absolute structure will be determined experimentally, even if one or more chiral centres in the molecule are already unambiguously known from the chemistry or synthesis of the compound. If the heaviest element present is S, P or Cl, then it is strongly recommended that the Friedel opposites of all symmetry-unique reflections be included in the data set. The proportion of Friedel related reflections required decreases with the increasing atomic weight of the heaviest element that is present, but for medium weight elements, it is recommended that at least 50% of the potential Friedel related reflections have been recorded. The more Friedel pairs that are present in the data set, the smaller will be the s.u. for the absolute structure parameter (Flack's x).
If the structure is non-centrosymmetric with atoms HEAVIER than Silicon the following two line items must be present in the CIF:

  _refine_ls_abs_structure_details
  _refine_ls_abs_structure_Flack

Add the number of Friedel pairs used in the refinement to the _refine_ls_abs_structure_details line so that it looks like:

  _refine_ls_abs_structure_details 'Flack (1983), XXXX Friedel pairs'

and replace the XXXX with the actual number of Friedel pairs used in the refinement. [An easy way to determine the number of Friedel pairs is to look at the difference between the number of unique reflections used in SHELXL when a MERG2 and MERG 3 instruction is used (MERG 3 forces Friedel pairs to be merged before use).]

IF Z > Si .and. NFDRAT > 0.5 issue a General ALERT
"Please check that the estimate of the number of Friedel pairs is correct. If it is not, please give the correct count in the _publ_section_exptl_refinement section of the submitted CIF."
If the structure is non-centrosymmetric with atoms HEAVIER than silicon the following two line items must be present in the CIF:
  _refine_ls_abs_structure_details
  _refine_ls_abs_structure_Flack

We have estimated the number of Friedel related reflections in your data set from a comparison of your value of _reflns_number_total with the theoretical number calculated for the symmetry-unique portion of reciprocal space out to your stated theta(max). We ask you to check our estimate and correct as necessary. Then add the number of Friedel pairs used in the refinement to the _refine_ls_abs_structure_details line so that it looks like:

  _refine_ls_abs_structure_details 'Flack (1983), XXXX Friedel pairs'

and replace the XXXX with the actual number of Friedel pairs used in the refinement. [An easy way to determine the number of Friedel pairs is to look at the difference between the number of unique reflections used in SHELXL when a MERG2 and MERG 3 instruction is used (MERG 3 forces Friedel pairs to be merged before use).]

IF Z < Si .and. radiation is CuKa .and. NFDRAT < 0.5 issue a General ALERT
"WARNING: CuKa measured Friedel data can be used to determine absolute structure in a light-atom study only if the Friedel fraction is large."
As you only have a light atom structure (heaviest element lighter than silicon), anomalous dispersion effects are very small, even with Cu radiation. The reliability of the absolute structure determination will be improved if as many Friedel pairs of reflections as possible are present in the data set. Your current data set appears to contain less than 50% of the total potential Friedel related reflections that are possible up to your stated theta(max) and the absolute structure determination is probably unreliable (i.e. the value of the absolute structure parameter is meaningless because of its large s.u. value). If you are attempting to draw any conclusions about the absolute structure from the crystallographic data, then it is strongly recommended that the Friedel opposites of all symmetry-unique reflections be included in the data set. A recollection of your data is therefore probably warranted.
If the structure is non-centrosymmetric with only atoms LIGHTER than silicon, but you believe that the crystallographic experiment has successfully determined the absolute structure, the following two line items must be present in the CIF:

  _refine_ls_abs_structure_details
  _refine_ls_abs_structure_Flack

We have estimated the number of Friedel related reflections in your data set from a comparison of your value of _reflns_number_total with the theoretical number calculated for the symmetry-unique portion of reciprocal space out to your stated theta(max). We ask you to check our estimate and correct as necessary. Then add the number of Friedel pairs used in the refinement to the _refine_ls_abs_structure_details line so that it looks like:

  _refine_ls_abs_structure_details 'Flack (1983), XXXX Friedel pairs'

and replace the XXXX with the actual number of Friedel pairs used in the refinement. [An easy way to determine the number of Friedel pairs is to look at the difference between the number of unique reflections used in SHELXL when a MERG2 and MERG 3 instruction is used (MERG 3 forces Friedel pairs to be merged before use).]

If no useful absolute structure parameter can be refined (i.e. the value of the absolute structure parameter is meaningless because of its large s.u. value), authors should consider merging Friedel-pair reflections before final refinement, and stating this in the _publ_section_exptl_refinement section of the CIF, along with the meaningless absolute structure parameter value (and s.u. value) obtained from any refinement with Friedel pairs, as justification of the merging of Friedel-pair data. Users of SHELXL-97 can merge Friedel-pair data with the MERG 3 instruction.

IF Z < Si .and. radiation is MoKa .and. NFDRAT > 0.05 issue a General ALERT
"ALERT: MoKa measured Friedel data cannot be used to determine absolute structure in a light-atom study EXCEPT under VERY special conditions."
With non-centrosymmetric structures and Mo radiation, if no atoms heavier than Si are present, the DELTA-f'' terms in the scattering factor expression are very small. In such cases, if no useful absolute structure parameter can be refined (i.e. the value of the absolute structure parameter is meaningless because of its large s.u. value), authors should consider merging Friedel-pair reflections before final refinement, and stating this in the _publ_section_exptl_refinement section of the CIF, along with the meaningless absolute structure parameter value (and s.u. value) obtained from any refinement with Friedel pairs, as justification of the merging of Friedel-pair data. Users of SHELXL-97 can merge Friedel-pair data with the MERG 3 instruction.

另外, 关于Merg的含义, 请参考下面这段
另有说明在shelx说明书的 2.4页,
2.5  Initial processing of reflection data 一节

MERG n [2]

--------------------------------------------------------------------------------
If n is equal to 2 the reflections are sorted and merged before refinement; if the structure is non-centrosymmetric the Friedel opposites are not combined before refinement (necessary distinction from SHELXS). If n is 1 the indices are converted to a 'standard setting' in which l is maximized first, followed by k, and then h; if n is zero, the data are neither sorted nor converted to a standard setting. n = 3 is the same as n = 2 except that Friedel opposites are also merged (this introduces small systematic errors and should only be used for good reason, e.g. to speed up the early stages of a refinement of a light atom structure before performing the final stages with MERG 2). Note that the reflections are always merged, and Friedel opposites combined, before performing Fourier calculations in SHELXL_93 so that the (difference) electron density is correctly scaled. Even with n = 0 the program will change the reflection order within each data block to optimize the vectorization of the structure factor calculations (it is shuffled back into the MERG order for LIST 4 output). Note that MERG may not be used in conjunction with TWIN or HKLF 5 or 6. In SHELX_76, MERG 3 had a totally different meaning, namely the determination of inter-batch scale factors; in SHELXL_93, these may be included in the refinement using the BASF instruction.

其他,

如果偏差太大, flack参数没有意义.

先提供一个链接
http://www.iucr.org/__data/iucr/lists/coredmg/msg00003.html

再看这个

The correct absolute structure has been defined by the atomic coordinates if _refine_ls_abs_structure_flack is close to 0.0 (and the s.u. is sufficiently small). If _refine_ls_abs_structure_flack is close to 1.0, the incorrect enantiomer is being modelled and the atomic coordinates should be inverted and refined again.
In cases of intermediate values of _refine_ls_abs_structure_flack, and you believe that a merohedral twin or a partial mix of enantiomers is present, choose the configuration that gives the lowest value for _refine_ls_abs_structure_flack.

Note that if the s.u. of the Flack parameter is large, e.g. greater than or equal to 0.3, one cannot confidently derive the absolute structure from the data, because, within the 3sigma confidence limits, the full range of possible values of the Flack parameter are plausible. In such cases, even if the Flack parameter itself is close to 0.0, no conclusions about the absolute structure are justified. This will usually be the case if the compound is a weak anomalous scatterer (i.e. no atom heavier than Si is present). If a heavy atom is present, but the s.u. is still large, you should ensure that a significant fraction (ideally all) of the Friedel opposites of the symmetry unique reflections have been included in the data set.

[ Last edited by xi2004 on 2009-7-14 at 09:11 ]

回复此楼

» 收录本帖的淘帖专辑推荐

晶体生长合成专辑	单晶解析	晶体解析相关	晶体好贴
晶体	与晶体相关的好帖子	都是牛贴~	crystal
测试	科研工具	单晶解析问题集锦	晶体实例