Polymorph or not

[Jsn]

Hi everybody,

I need your opinion and expertise for confirming, the crystal form of a particular compound that I have with me is a polymorph or is it a mere structural re-determination.


Neutron diffraction data (1976)

a = 10.523(4) b = 11.222(4) c = 8.068(3)
α = 101.22(1) β = 112.15(1) γ = 112.65(1)

Triclinic P-1

Single Crystal X-ray diffraction data (2012)
a = 8.021(1) b = 10.530(1) c = 10.596(1)
α = 107.0 β = 101.32(1) γ = 112.10

Triclinic P-1


Would also like to know, the procedure to check whether a compound is polymorph or not using their unit cell parameters in WingX / Platon

Thank you
Jsn

[rwwh]

I asked Toine Schreurs from Utrecht for help. His "2view" program that he developed as part of the eval14/eval15 suite of integration software comes up with the following congruence:

Code: Select all

Cell from first.rmat
Cell from second.rmat
             first                                    second
  0.1071235   0.0085253   0.0549863        0.1298009   0.0046707   0.0206438
  0.0514638   0.1048826   0.0536935        0.0609509   0.1119026   0.0415699
  0.0000000   0.0000000   0.1239465       -0.0000000  -0.0000000   0.0943752
Determinant:  0.1338207E-02                            0.1343938E-02
           a       b       c      alpha    beta   gamma    volume
first  :  10.523  11.222   8.068  101.22  112.15  112.65      747.3
second :   8.021  10.530  10.596  107.00  101.32  112.10      744.1
Volume ratio = 1.004 Trying 160 solutions
Nr Rotangle       Rotvec(xyz)           RotVec(hkl)    ( angle)     RotVec(uvw)    ( angle)     Fom
 1   98.223  0.4182 -0.7318 -0.5382  10.90-13.00 -6.99 (  0.38)   2.15-22.04-25.00 (  0.23)   2.720
Selected Solution 1
           a       b       c      alpha    beta   gamma    volume
first  :  10.523  11.222   8.068  101.22  112.15  112.65      747.3 <
second :   8.021  10.530  10.596  107.00  101.32  112.10      744.1 <
second*:  10.530  11.195   8.021  100.18  112.10  113.22      744.1 <
       second==> second*                        second*==> second
  0.0000000  -1.0000000   0.0000000        0.0000000   0.0000000  -1.0000000
  1.0000000   1.0000000   1.0000000       -1.0000000   0.0000000  -0.0000000
 -1.0000000   0.0000000   0.0000000        1.0000000   1.0000000   1.0000000
Determinant:   1.000000                                 1.000000
H'= -0.011*H -0.920*K +0.128*L
K'= +0.835*H +0.952*K +1.405*L
L'= -0.825*H -0.033*K -0.219*L
Nr Rotangle       Rotvec(xyz)           RotVec(hkl)    ( angle)     RotVec(uvw)    ( angle)     Fom
 1   98.223  0.4182 -0.7318 -0.5382  10.90-13.00 -6.99 (  0.38)   2.15-22.04-25.00 (  0.23)   2.720 <


[Rudi]

Maybe you will also find helpful the COMPSTRU tool at the Bilbao Crystallographic Server:

http://www.cryst.ehu.es/cryst/compstru.html

[Zamu]

Hi,

I think that in this case, with just the cell parameters and no atomic coordinates, CELLTRAN (http://www.cryst.ehu.es/cryst/celltran.html) would be more useful, since it will calculate the reduced cell, which are very similar:

Code: Select all

Neutron 8.0680 10.5230 10.5620 107.410 100.460 112.150 V=747.3
X-ray   8.0210 10.5300 10.5960 107.000 101.320 112.100 V=744.1

[Jsn]

Hi rwwh ,

thanks for the reply, so you mean to say that both the structures are same right.

[Zamu]

Hi,

Based on Rob's use of "2view" program (which shows that both unit cells are equivalent) and the fact that the reduced cells are so similar, I would say that the XR data is a redetermination of the same crystal form. But a better proof would be a comparison of the crystal structures themselves (either using atomic coordinates or indirectly through a comparison of the powder diffractograms), since it woukdn't be the first case of two different crystal forms with very similar unit cell parameters.

Pablo

[Jsn]

Hi Rudi,

I tried using COMPSTRU, but i don't know what to look for further. Since there is a minor deviations in cell parameters, I just want to know, whether the structures are same or different.

jsn

[johnewarren]

This is also a function in CCTBX something like:

Code: Select all

phenix.explore_metric_symmetry  --unit_cell="UNIT CELL OF 1"  --space_group=P1 --other_unit_cell="UNIT CELL OF THE OTHER"  --other_space_group=SPACEGROUP --no_point_group_graph

But unlike the above CCTBX is a bit of a beast just to do this requires you to download ~100MB I think. So unless you already have it I would probably not bother.

So I ran the code for you:

Code: Select all

phenix.explore_metric_symmetry  --unit_cell="10.523 11.222 8.068 101.22 112.15 112.65"  --space_group=P1 --other_unit_cell="8.021 10.530 10.596 107.0 101.32 112.10"  --other_space_group=P1 --no_point_group_graphA second unit cell has been specified.


Unit cell comparison
--------------------

The unit cells will be compared. The smallest niggli cell,
will be used as a (semi-flexible) lego-block to see if it
can construct the larger Niggli cell.



Crystal symmetries in supplied setting

Target crystal symmetry:
    Unit cell: (10.523, 11.222, 8.068, 101.22, 112.15, 112.65)
    Space group: P 1 (No. 1)
Building block crystal symmetry:
    Unit cell: (8.021, 10.53, 10.596, 107, 101.32, 112.1)
    Space group: P 1 (No. 1)

Crystal symmetries in Niggli setting

Target crystal symmetry:
    Unit cell: (8.068, 10.523, 10.5619, 107.406, 100.457, 112.15)
    Space group: P 1 (No. 1)
Building block (lego cell) crystal symmetry:
    Unit cell: (8.021, 10.53, 10.596, 107, 101.32, 112.1)
    Space group: P 1 (No. 1)

Volume ratio between target and lego cell:  1.00

Cartesian basis (column) vectors of lego cell:
  /   8.0  -4.0  -2.1 \ 
  |   0.0   9.8  -4.2 | 
  \   0.0   0.0   9.5 / 

Cartesian basis (column) vectors of target cell:
  /   8.1  -4.0  -1.9 \ 
  |   0.0   9.7  -4.2 | 
  \   0.0   0.0   9.5 / 

A total of    8 matrices in the hermite normal form have been generated.
The volume changes they cause lie between    2 and    1.

Trying all matrices

   1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0
   * . . . . . . .
 Listing all possible solutions

Solution    1
--------------------------------------------------------------
Target unit cell :       8.1  10.5  10.6 107.4 100.5 112.2
Lego cell :              8.0  10.5  10.6 107.0 101.3 112.1

               /   1    0    0  \ 
matrix :  M =  |   0    1    0  | 
               \   0    0    1  / 

Additional Niggli transform:      x,y,z
Additional similarity transform:  x,y,z
Resulting unit cell :    8.0  10.5  10.6 107.0 101.3 112.1
Deviations :             0.6  -0.1  -0.3   0.4  -0.9   0.1
Deviations for unit cell lengths are listed in %.
Angular deviations are listed in degrees.

 
--------------------------------------------------------------


[Jsn]

So John,

what do you say, is it a polymorph or not.

[Zamu]

Hi,

As I said before, it seems to be a redetermination of the same crystal structure, not a polymorph.

Cheers, Pablo

[Jsn]

Hi,

Thank you so much.

Jsn