The squeeze tool is already around for 25 years

The SQUEEZE Tool is already around for 25 years

• The SQUEEZE Tool is already around for 25 years

• It is a method to take the scattering of a disordered solvent into account in L.S. as an alternative for a discrete disorder model.

• There are believers and dis-believers in the method, some rightly and some ill-informed.

• Its use is reported in the CSD at least 10000 x

• A ‘work-around’ was needed to run SQUEEZE along with a SHELX76/L97 based refinement.

• This issue is now addressed with SHELXL2013

• Handling solvent disorder + twinning possible

Identify the disordered solvent volume

• Identify the disordered solvent volume

• Calculate by difference map iteration the contribution of the electron density found in the voids to the calculated structure factors.

• Subtract the solvent contribution from the observed data and continue SHELX refinement.

• The R-value improved from 9.6 to 4.5 %

• P. van der Sluis & A.L.Spek (1990). Acta Cryst. C46, 883-886 and A46, 194-201 (BYPASS)

A new and distributable version (now called SQUEEZE) was made part of the PLATON program and to work with SHELXL97 (LINUX/MAC-OSX/MS-WINDOWS versions)

• A new and distributable version (now called SQUEEZE) was made part of the PLATON program and to work with SHELXL97 (LINUX/MAC-OSX/MS-WINDOWS versions)

• Input to SQUEEZE: ‘shelxl.res’ + ‘shelxl.hkl’

• Output: a new solvent removed ‘shelxl.hkl’ suitable to continue with the final L.S. refinement with SHELXL97.

• A proper FCF file with original Fo2 is created afterwards based on the information hidden beyond column 80.

• An often stated Point of Critic: The SHELXL97 refinement against modified observed data. In practice: no real difference with CRYSTALS refinement + observed data.

Example: Organometallic compound + THF with two test refinements:

• Example: Organometallic compound + THF with two test refinements:

• A first calculation was based on a disorder model for the disordered solvent (THF)

• The second calculation invoked SQUEEZE with the shelxl.res and shelxl.hkl without the THF parameters in the shelxl.res model.

• The results of both calculations are very similar as shown with their ORTEP’s

A typical crystal structure has only in the order of 65% of the available space filled.

• A typical crystal structure has only in the order of 65% of the available space filled.

• The remainder volume is in voids (cusps) in-between atoms (too small to accommodate an H-atom)

• Solvent accessible voids can be defined as regions in the structure that can accommodate at least a sphere with radius 1.2 Angstrom without intersecting with any of the van der Waals spheres assigned to each atom in the structure.

• Next Slide: Void Algorithm: Cartoon Style 

Calculate the VOID-map

• Calculate the VOID-map

• Calculate a difference density Fourier map (FFT)

• Use the VOID-map as a mask on the FFT-map to set all density outside the VOID’s to zero.

• FFT-1 this masked Difference map -> contribution of the disordered solvent to the structure factors (A(cal) & B(cal)

• Calculate an improved difference map with F(obs) phases based on the vectorial sum of F(calc) + F(solv) and F(calc) without the solvent contribution.

• Recycle to 2 until convergence.

‘Observed Data’ were calculated from the published coordinates.

• ‘Observed Data’ were calculated from the published coordinates.

• The ether molecule was subsequently removed

• SQUEEZE was tested to see whether the method recovers the ether contribution to the structure factors.

The SQUEEZE procedure is NOT refinement: it just determines the disordered solvent contribution to the structure factors. The discrete model parameters remain fixed.

• The SQUEEZE procedure is NOT refinement: it just determines the disordered solvent contribution to the structure factors. The discrete model parameters remain fixed.

• The final SHELXL97 refinement only optimizes the discrete parameter model

• Residual Issues: #1:The estetically not nice but technically needed subtraction of the solvent contribution from I(obs) and #2: the handling of cases with additional Twinning.

SHELXL2013 elegantly solved issue #1 by now accepting a fixed solvent contribution to F(calc) in the refinement against F(obs)2.

• SHELXL2013 elegantly solved issue #1 by now accepting a fixed solvent contribution to F(calc) in the refinement against F(obs)2.

• The input to SQUEEZE is now a .cif and .fcf and the output a (.lis), .ins, .hkl & .fab for the final SHELXL2013 refinement (SQUEEZE info in .fab)

• New: Handling of issue #2: SQUEEZE + Twinning: Based on the detwinning option in SHELXL2013 via a LIST 8 style FCF (as opposed to a LIST 4 FCF)

The current SQUEEZE tool can either be run old-style with name.res and name.hkl as input files (Mode 1) or with the pair name.cif and name.fcf (Mode 2). Mode 2 is preferred in combination with SHELXL2013

• The current SQUEEZE tool can either be run old-style with name.res and name.hkl as input files (Mode 1) or with the pair name.cif and name.fcf (Mode 2). Mode 2 is preferred in combination with SHELXL2013

• Output files are either named ‘name_sq.ext’ or ‘name_sqd.ext’ (sq for sum and sqd for difference)

• The use of the ‘sq’ files (i.e. F(obs)2 unmodified) is to be preferred in combination with SHELXL2013

• In case of twinning, the name.fcf should be of the SHELXL2013 ‘LIST 8’ type

SQUEEZE can be run either from the PLATON menu (going through the various stages) or the terminal window command line as ‘PLATON –q name.cif’. The latter without any further interaction (i.e. a filter)

• SQUEEZE can be run either from the PLATON menu (going through the various stages) or the terminal window command line as ‘PLATON –q name.cif’. The latter without any further interaction (i.e. a filter)

• There are options to visualize the voids

Step 1: SHELXL refinement based a twin.ins (that should include ‘LIST 8’, ‘ACTA’, ‘BASF’ and ‘HKLF5’ records) and a twin.hkl file [merohedral: BASF/TWIN]

• Step 1: SHELXL refinement based a twin.ins (that should include ‘LIST 8’, ‘ACTA’, ‘BASF’ and ‘HKLF5’ records) and a twin.hkl file [merohedral: BASF/TWIN]

• Step 2: Run PLATON with the twin.cif and twin.fcf files produced in Step 1

• Step 3: Continue SHELXL refinement with the files twin_sq.ins, twin_sq.hkl and twin_sq.fab produced by PLATON in step 2.

• Note: The twin5_sq.fab file contains the solvent contribution to the SF and the details of SQUEEZE

The ‘.res’, ‘.hkl’ ,‘.fab’ and checksum information that is embedded in the SHELXL2013 CIF should be left untouched.

• The ‘.res’, ‘.hkl’ ,‘.fab’ and checksum information that is embedded in the SHELXL2013 CIF should be left untouched.

• The embedded ‘.fab’ contains info about the use of SQUEEZE and will suppress certain irrelevant ALERTS related to the SQUEEZE use.

• It is probably best not to include the SQUEEZED solvent info in the formula and related items such as MolWeight and density .

• Report the details in the experimental section.

There should be no residual unresolved density in the discrete model region of the structure because of its impact in the solvent region.

• There should be no residual unresolved density in the discrete model region of the structure because of its impact in the solvent region.

• The data set should be reasonably complete and with sufficient resolution [i.e. sin(theta)/lambda >0.6]. Low temperature data help a lot.

• There should be no unresolved charge balance issues that might effect the chemistry involved (e.g. The valency of a metal in the ordered part of the structure)

• The SQUEEZE technique can not handle cases where the solvent region is not constant but varies due to (coupled) disorder in the discrete part of the model.

• The solvent region is assumed not to contain significant anomalous scatterers (Friedels averaged)

• Designed for ‘small molecule structures’

One of the issues that needs to be addressed is the number of additional parameters to be added in the calculation associated with SQUEEZE on the L.S. Command. The default value is set to: (E x n) / (Z x m) where

• One of the issues that needs to be addressed is the number of additional parameters to be added in the calculation associated with SQUEEZE on the L.S. Command. The default value is set to: (E x n) / (Z x m) where

• E = the number of recovered electrons in the unit cell

• Z = the number of asymmetric units

• m = the number of electrons in a CH2 fragment (=8)

• n  = the number of parameters usually refined for a CH2 fragment (=9)

• This formula has the nice property that it vanishes when there is no residual density in the void

The original BYPASS paper described a procedure involving an inner and an outer loop. In general only the inner loop (SQUEEZE) is done.

• The original BYPASS paper described a procedure involving an inner and an outer loop. In general only the inner loop (SQUEEZE) is done.

• Under certain conditions (large disorder contribution to the SF or twinning) the full BYPASS procedure including the outer loop (i.e. L.S./SQUEEZE/L.S./SQUEEZE/. ..) can be attempted. [INVOKED as BYPASS with name.cif and name.fcf]. This routine assumes SHELXL2013.

The use of SQUEEZE is recommended in cases of severe solvent disorder (being often of unknown composition or a mixture). Otherwise a disorder model should (at least) be attempted.

• The use of SQUEEZE is recommended in cases of severe solvent disorder (being often of unknown composition or a mixture). Otherwise a disorder model should (at least) be attempted.

• The peaklist file ‘name.sqz’ that is produced from the optimized difference map as part of the SQUEEZE calculation may suggest a solvent model afterall.

• The use of the SQUEEZE procedure should always be detailed in the experimental section of a paper that is based on its use.

• Further development of BYPASS/SQUEEZE is ongoing.