details of PLATON tests

 SHELXL2018 and later will automatically include the final '.res' file in the
 CIF as an embedded comment preceded by the dataname '_shelx_res_file'.
 That dataname should not be renamed into '_iucr_refine_instructions_details'.
 This embedded file contains the complete details of the refinement model
 used and can along with the embedded _shelx_hkl_file be extracted to repeat
 the final refinement in order to recreate the final '.fcf' file to be used
 for the validation of the refinement. 

 This ALERT reports the number of atomic sites that are flagged in the CIF as
 distance or angle restrained (D-Flag). The associated geometry parameter 
 values may be less informative.   

 This ALERT reports the number of non-H atoms that are flagged as handled   
 with Uiso or U(i,j) restraints (U-Flag). Their usage may hide unresolved
 disorder issues.   

 This ALERT reports on polymeric networks and their dimensionality as found in  
 the crystal structure. 
 Note: Polymeric structures can be legitimate or due to an erroneous structure
 analysis.

 No embedded 'refinement details' records were found in the CIF. Acta Cryst.
 requires the inclusion of the last .res file (in case of a SHELXL, XL or 
 OLEX2 refinement or similar for non-SHELXL refinement programs) in the CIF,   
 embedded between records with semicolons in position 1, preceded by a
 _shelx_res_file or '_iucr_refine_instructions_details' record.  
 Note: SHELXL20xy will automatically include the final .res as an embedded  
 comment with the dataname '_shelx_res_file'.   

 This ALERT reports the refinement of an extinction parameter.  
 SHELXL corrects Fobs values in the FCF for Extinction. JANA not.   
 Note: Large values of this parameter may be caused by other factors than 
 extinction (e.g. twinning, systematically under or over-estimated
 intensities and other systematic errors).

 It is standard practice to refine H-atoms on hetero atoms such as O & N as
 proof of their correct assignment. 

 No atom coordinates were detected in the CIF prior to the U(i,j) loop. The 
 atom labels in both loops should be identical in order to be associated with
 each other. 

 The supplied CIF contains a '_shelx_res_file' record but not a valid   
 associated '_shelx_res_checksum' record. A valid pair of embedded '.res' and 
 .hkl files allows the automatic creation of a .fcf file with SHELXL20xy to be  
 used for a detailed analysis of the refinement result. A file is valid when
 the calculated and reported checksums are identical. Only characters with an   
 ASCII value higher than 32 contribute to the checksum. An embedded '.res'
 file might be broken, either due to data transfer errors or to deliberate or
 accidental post-refinement editing of its content. Please re-refine after
 implementing any changes to the '.res' file.  

 The supplied CIF contains a '_shelx_hkl_file' record but not a valid   
 associated '_shelx_hkl_checksum' record. A valid pair of embedded .res and 
 .hkl files allows the automatic creation of a .fcf file with SHELXL20xy to be  
 used for a detailed analysis of the refinement result. A file is valid when
 the calculated and reported checksums are identical. Only characters with an   
 ASCII value higher than 32 contribute to the checksum. 
 An embedded '.res' file might be broken, either due to transfer errors or to  
 deliberate or accidental post-refinement editing of its content.   

 The supplied CIF contains a '_shelx_fab_file' record but not a valid   
 associated '_shelx_fab_checksum' record. A valid pair of embedded '.res' and 
 '.hkl' files allows the automatic creation of a '.fcf' file with SHELXL20xy
 to be used for a detailed analysis of the refinement result. A file is valid
 when the calculated and reported checksums are identical. Only characters
 with an ASCII value higher than 32 contribute to the checksum. An embedded
 '.res' file might be broken, either due to transfer errors or to deliberate
 or accidental post-refinement editing of its content.   

 No embedded .fab record was found in the CIF file that was created with
 SHELXL20xy (or XL). SHELXL20xy automatically includes the '.fab' file that 
 was used in the refinement (along with the final '.res' & '.hkl' files) as 
 an embedded comment with the dataname '_shelx_fab_file'. Do not change this
 dataname. Such a record is useful for archival and follow-up calculations. 

 Check for the correct scattering type assignment to this atom. 

 The values of '_diffrn_reflns_theta_max' and '_diffrn_reflns_theta_full'
 are  reported as equal. However, the associated reported values of
 '_diffrn_measured_fraction_theta_max'  and
 '_diffrn_measured_fraction_theta_full' are not equal. This is inconsistent.
 CIF Dataname Definitions:
 _diffrn_measured_fraction_theta_max:
    Fraction of unique (symmetry-independent) reflections measured out to
    '_diffrn_reflns_theta_max'.
 _diffrn_measured_fraction_theta_full:
    Fraction of unique (symmetry-independent) reflections measured out to
    '_diffrn_reflns_theta_full'.
 _diffrn_reflns_theta_max:
   Maximum theta angle in degrees for the measured intensities. The fraction
   of unique reflections measured out to this angle is given by
   '_diffrn_measured_fraction_theta_max'.
 -diffrn_reflns_theta_full:
    The theta angle (in degrees) at which the measured reflection count is
    close to complete. The fraction of unique reflections measured out to
    this angle is given by '_diffrn_measured_fraction_theta_full'.

 The reported value of '_diffrn_measured_fraction_theta_full' is less than
 the reported value of '_diffrn_measured_fraction_theta_max'. Their ratio is  
 reported when less than 1.0. When theta_full is less than theta_max such   
 a value indicates that there are relatively more reflections missing at
 lower resolution.
 The measured fraction (completeness) is expected to be higher at a lower
 theta_full value.
 Usually, a theta_full value corresponding to sin(theta)/lambda = 0.6 should
 give a satisfactory data completeness value.
 CIF Dataname Definitions:
 _diffrn_measured_fraction_theta_max:
    Fraction of unique (symmetry-independent) reflections measured out to
    '_diffrn_reflns_theta_max'.
 _diffrn_measured_fraction_theta_full:
    Fraction of unique (symmetry-independent) reflections measured out to
    '_diffrn_reflns_theta_full'.
 _diffrn_reflns_theta_max:
   Maximum theta angle in degrees for the measured intensities. The fraction
   of unique reflections measured out to this angle is given by
   '_diffrn_measured_fraction_theta_max'.
 -diffrn_reflns_theta_full:
    The theta angle (in degrees) at which the measured reflection count is
    close to complete. The fraction of unique reflections measured out to
    this angle is given by '_diffrn_measured_fraction_theta_full'.

 The value of Rint (i.e. '_diffrn_reflns_av_R_equivalents') should normally be
 considerably less than 0.12 and in the order of magnitude of the reported  
 R-values. Rint may be relatively meaningless when based on a very limited  
 number of averaged data. Higher values should be accompanied by a suitable 
 explanation in the '_publ_section_exptl_refinement' section. However,
 authors should first ensure that there are not overlooked problems 
 associated with the data or the space-group. Elevated values for   
 '_diffrn_reflns_av_R_equivalents' may be indicative of a need to recollect   
 the data from a crystal of higher quality or that there is a problem with  
 the data treatment. Consider the following 
 (a) The absorption corrections are inadequate or inappropriate.
 (b) The overall quality of the data may be poor due to the crystal quality.
 (c) The crystal is very weakly diffracting, so that a large proportion of  
     essentially "unobserved" reflections are being used in the refinement. 
     You should consider using a better crystal or a data collection at 
     low temperature and/or, if the compound is organic, using Cu radiation.
 (d) You are working in the wrong crystal system or Laue group. 
 (e) You have only a very small number of equivalent reflections, which 
     may lead to artificially high values of '_diffrn_reflns_av_R_equivalents'
 Note that if '_diffrn_reflns_av_sigmaI/netI' is also large, the quality  
 of the data should be considered to be suspect.

 The expected number of reflections corresponds to that in the asymmetric   
 unit of the Laue group. Expected ratio: less-or-equal 1 for centro symmetric   
 structures and less than 2 for non-centrosymmetric structures. 
 Reasons to exceed those numbers can be:
  1 - Systematic absencies not omitted from the observed data count.
  2 - Refinement with redundant (i.e. not merged/unique) data set.  
  3 - SHELXL HKLF 5 Refinement  

 Test for data completeness. The ratio of the reported number of unique 
 reflections and expected number of reflections for the resolution given is 
 reported. The ratio can be low due to a missing cusp of data when collected
 with a 2D-detector. Alternatively, the wrong asymmetric part of reciprocal 
 space was collected on a serial detector system.   

 Check resolution of the data set. This alert is issued when sin(theta)/lambda
 < 0.6 (i.e. theta < 25.24 degree for MoKa or 67.7 degree for CuKa radiation).
 In principle, all observed data should be included in the refinement.
 Alternatively, a sin(theta)/Lambda cutoff value can be used at a value where
 average(I/sigma(I)) < 2 in order not to refine on noise.

 Check reported h,k,l - range with calculated range based on reported 
 theta-max.

 Check whether a sufficient fraction of the unique data set is indeed above  
 the 2 * sigma(I) level.
 All reflection data are included in this test. Check whether this low ratio
 is caused by weak data beyond sin(theta)/lambda > 0.5 (i.e. low resolution).   

 Ideally (and a requirement for publication in Acta Crystallographica), 
 the dataset should be essentially complete, as defined by  
 '_diffrn_measured_fraction_theta_full' (close to 1.0), up to 
 sin(theta)/lambda = 0.6 (i.e. 25.24 degrees MoKa). 
 The three major causes of incomplete data sets are:
 1 - A missing cusp of data due to data collection by rotation around   
     the spindle axis only (standard on some image-plate systems).  
     Cure: collect an additional data set after remounting the crystal. 
 2 - The DENZO image processing package has problems with certain strong
     reflections. They are often excluded from the data set.
     Cure: Add an additional scan at lower power setting in order to include
     strong low order reflections.  
 3 - Incomplete scans.  

 Ideally, the reported '_diffrn_measured_fraction_theta_max' value, 
 corresponding to theta-max, should be close to 1.0.

 Ideally (and a requirement for publication in Acta Crystallographica), 
 this fraction should be close to 1.0 for theta-full greater or equal   
 to sin(theta/lambda) = 0.6 (i.e. 25.24 degrees for MoKa and 67.7 degrees   
 for CuKa  radiation).  
 The three major causes of incomplete data sets are:
 1 - A missing cusp of data due to data collection by rotation around   
     the spindle axis only (standard on some image-plate systems).  
     Cure: collect an additional data set after remounting the crystal. 
 2 - The DENZO image processing package has problems with certain strong 
     reflections. They are often excluded from the data set.
     Cure: Add an additional scan at lower power setting in order to
     include strong low order reflections.  
 3 - Incomplete scans, possibly based on erroneously assumed higher than
     actual symmetry.   
 Note: The default value of '_diffrn_measured_fraction_theta_full' that   
 is automatically calculated and inserted in the CIF by SHELXL-97 might 
 generate A-level ALERTS when significant numbers of reflections are
 missing at higher theta values. In order to avoid such an ALERT,   
 substitute the values calculated with the SHELXL instruction 'ACTA 50' 
 for '_diffrn_reflns_theta_full' and '_diffrn_measured_fraction_theta_full' 
 respectively. For Mo-radiation, corresponding values of 25 degees (or  
 higher) and 0.99 (or higher) are expected. (See SHELXL manual).
 PLATON may be used to analyse the case at hand (by invoking either the 
 'FCF-VALIDATION' mode or the 'ASYM-VIEW' mode).

 The number of measured reflections should be equal or greater than the 
 number of unique reflections.  

 This test checks whether a refined extinction parameter is meaningful  
 i.e. whether its value is significantly larger than its corresponding s.u. 
 If not, this parameter should be removed from the model and the structure  
 refined without this meaningless additional parameter. 
 The current default gives a warning when its value is within 3.33 s.u. 
 SHELXL97-2 will not allow negative values leading to ill-convergence and   
 non-zero maximum shift/error values: remove the extinction parameter from the  
 refinement.

 Check the validity of the absolute structure determination.
 A high s.u. indicates that the experimental data do not support the
 determination of the absolute structure. This will generally be the case   
 with light atom MoKa data where f" is nearly zero. 
 Note: Use the TWIN & BASF 0.0 instructions in SHELXL97. The default
 FLACK parameter is not always reliable, in particular when strongly
 correlated with the position of the origin (e.g. along y in space-group P21).  
 Please refer to Flack,H.D. & Bernardinelli, G. (1999) Acta Cryst. A55, 
 908-915 and (2000) J. Appl. Cryst., 33, 1143-1148. 

 Check the relevance/validity of the absolute structure determination. Please
 refer to Flack,H.D. & Bernardinelli, G. (1999) Acta Cryst. A55, 908-915 and
 (2000) J. Appl. Cryst., 33, 1143-1148. A value of the Flack parameter that
 deviates significantly from zero  (taking into account the associated s.u.)
 might indicate that the absolute structure should be inverted in case of a
 value closer to 1.0 than to zero. A value close to 0.5 may be indicative of
 an inversion twin or a missed centre of inversion. For valid absolute
 structure assignments, abs(x) should be less than 2 * s.u., with s.u. < 0.04.
 For enantiopure compounds, s.u. should be less than 0.1.  

 No Flack parameter value given for non-centrosymmetric structure with  
 heaviest atom Z > Si. This might be intentional.   

 Options are 'rm', 'ad', 'rmad', 'syn', 'unk' or '.'
 rm   : absolute configuration established by the structure determination   
        of a compound containing a chiral reference molecule of known   
        absolute configuration. 
 ad   : absolute configuration established by anomalous dispersion effects  
        in diffraction measurements on the crystal. 
 rmad : absolute configuration established by the structure determination   
        of a compound containing a chiral reference molecule of known   
        absolute configuration and confirmed by anomalous dispersion
        effects in diffraction measurements on the crystal. 
 syn  : absolute configuration has not been established by anomalous
        dispersion effects in diffraction measurements on the crystal.  
        The enantiomer has been assigned by reference to an unchanging  
        chiral centre in the synthetic procedure.   
 unk  : absolute configuration is unknown, there being no firm chemical 
        evidence for its assignment to hand and it having not been  
        established by anomalous dispersion effects in diffraction  
        measurements on the crystal. An arbitrary choice of enantiomer  
        has been made.  
 .    : inapplicable.   

 No standard uncertainty found for the Flack parameter. When the structure
 refinement was done with SHELXL97-2, the likely reason for this is a missing
 BASF instruction. This applies in particular when the associated Flack
 parameter has the value 0.000. No valid conclusions on the absolute structure
 can be drawn in that case.  

 No information is given about the theta value for which the dataset is 
 complete, subject to the percentage given with the dataname
 _diffrn_measured_fraction_theta_full.  

 This fraction should be specified in combination with the theta value given 
 with the dataname '_diffrn_reflns_theta_full'. 

 This fraction should be specified in combination with the value for
 '_diffrn_reflns_theta_max'.  

 Alert for 'no H-atoms' in CIF. This is unusual for carbon containing
 compounds, but may be correct.

 In the ideal case, both SumFormula strings (reported and calculated) should
 be identical. If not, the reason for the difference should be clear. Examples  
 are cases where populations do not add up to integer numbers, or when solvent  
 molecules have been SQUEEZED.  
 Note: SHELXL97 reports population parameters in the CIF with two decimals  
 only. This may lead to non-integer atom counts in cases of disorder
 due to rounding.   
 Note: Alerts _041, _042 & _045 can probably be ignored when the relevant   
 values differ by the same factor.  

 In the ideal case, the MoietyFormula string as reported should be identical
 to the MoietyFormula string calculated from the data in the CIF. If not, the   
 reason should be clear. Examples are cases where there is no separating space  
 between two element names or cases where populations do not add up to  
 integer numbers or when moieties are separated by '.' instead of ','.  
 Example: NO3 should be given as N O3   
 Note: Alerts _041, _042 & _045 can probably be ignored when the relevant   
 values differ by the same factor.  

 Note: atomic weights used in the calculation of the molecular weight are   
 taken from Inorg. Chim. Acta 217 (1994) 217-218 which deviate in a few cases   
 slightly from the older values used in SHELXL97-2. 
 Note: The tabulated atomic weights that are used may deviate from the actual
 value in case of special isotopes of e.g. Uranium or Plutonium. 

 In the ideal case, both data items should be the same within a small   
 tolerance. If not, the reason should be clear. Check also whether the
 density value is reported in the CIF.

 In the ideal case, both data items (Z(calc) & Z(reported)) should be   
 the same. If not, the reason for the difference should be clear. An example
 is the situation where PLATON gives Z = 1 when the program cannot work out 
 a proper Z.
 Note: Alerts _041, _042 & _045 can probably be ignored when the relevant   
 values differ by the same factor.  

 D(calc) as calculated from the reported Z and MW is compared for consistency   
 with the reported d(calc). 

 The Sumformula, corresponding with the Moietyformula, should be given. 

 The Moiety formula (i.e. the specification of the various species in the   
 structure) should be given in the CIF. 
 Example: '(Cd 2+)3, (C6 N6Cr 3-)2, 2(H2 O)'

 The calculated density will with a few exceptions be larger than 1.0.  
 A smaller value may indicate either an incomplete model or incorrect   
 symmetry. (e.g. a missing 'bar' in P-1 etc.)   

 The linear absorption coefficient corresponding to the Sumformula should be
 given. 

 In the ideal case, both Mu(calc) and Mu(cif) values should be the same within
 a small tolerance. If not, the reason for the difference should be clear.
 PLATON/checkCIF calculates Mu(calc) based on mu/Rho values tabulated in the
 Int. Tables for wavelengths Cu(Ka), Mo(Ka) and Ag(Ka). For other wavelengths, 
 such as those for Synchrotron, Ga(Ka) and In(Ka) sources, those values are
 calculated following S. Brennan & P.L. Cowan (1992). Rev. Sci. Instr., 63,
 850-853. Those values may differ slightly from values calculated with
 alternative external tools, in particular for wavelengths near absorption
 edges. The mu/Rho values used by PLATON/checkCIF for other wavelengths can be
 listed with the PLATON instruction 'ANOM wavelength'.  

 The treatment/method of absorption(correction) should be given explicitly. 
 Set _exptl_absorpt_correction_type to 'none' when no correction is done.   
 Other recognized values are 'psi-scan', 'empirical', 'multi-scan', 
 'refdelf', 'analytical', 'numerical', 'gaussian'.  

 The smallest crystal dimension should be supplied in the CIF.  
 The expected value should be a real number (i.e. not 0.35mm)   

 The medium crystal dimension should be supplied in the CIF.
 The expected value should be a real number (i.e. not 0.35mm)   

 The largest crystal dimension should be supplied in the CIF.   
 The expected value should be a real number (i.e. not 0.35mm)   

 Spherical correction for absorption is reported. The radius used is not
 supplied.  

 You have indicated that an absorption correction has not been applied. 
 (_exptl_absorpt_correction_type 'none'). However, the predicted values of  
 Tmin & Tmax, based on the crystal dimensions given in the CIF, are 
 sufficiently unequal that absorption effects appear to be significant. 
 Therefore, the application of a suitable absorption correction would   
 appear to be required. Also check that the crystal dimensions given in the 
 CIF do represent the actual crystal dimensions as closely as possible. 
 Inaccuracies here can lead to a poor prediction of Tmin & Tmax and give
 rise to these alerts. It should normally be possible to estimate the   
 crystal dimensions to 2 decimal places. Rough estimates to only 1 decimal  
 place may be too inaccurate to provide reliable estimates of Tmin & Tmax.  

 The Maximum transmission factor should be specified in the case a correction   
 for absorption was done. This is NOT the value that is calculated  
 automatically with SHELXL when a SIZE instruction is given in the SHELXL   
 instruction file. The values reported by SHELXL represent the EXPECTED 
 correction range. Some correction packages (e.g. SADABS) will provide  
 only one 'relative- correction-factor'. In such cases, Tmax should be  
 given as Tmax-expected and Tmin = relative-correction-factor * Tmax.   

 The Minimum transmission factor should be specified in case a correction   
 for absorption was done. This is NOT the value that is calculated  
 automatically with SHELXL when a SIZE instruction is given in the SHELXL   
 instruction file. The values reported by SHELXL represent the EXPECTED 
 correction range. Some correction packages (e.g. SADABS) will provide  
 only one 'relative-correction-factor'. In such cases, Tmax should be   
 given as Tmax-expected and Tmin = relative-correction-factor * Tmax.   

 see IUCR WEB-Pages 

 see IUCR WEB-Pages 

 Some (empirical) correction packages (e.g. SADABS) will provide only one   
 'relative-correction-factor'. In such cases, Tmax should be given as   
 Tmax-expected (as calculated from the crystal dimensions) and  
 Tmin = relative-correction-factor * Tmax.  

 Alert for crystals with at least one dimension probably too large for the  
 homogeneous part of the X-ray beam when used for datacollection using  
 crystal monochromated radiation. An exception will be datacollection using 
 a beta-filter and a sufficiently large collimator. 
 See also: C.H.Gorbitz (1999), Acta Cryst. B55, 1090-1098.  

 Check that the values entered under '_exptl_absorpt_correction_T_min' and
 '_exptl_absorpt_correction_T_max' have not been reversed or if there is a
 typographical error for one of these two items.

 For high mu * mid values, numerical absorption correction procedures are   
 recommended (either based on Gaussian integration or analytical) in case of
 homogeneous beam profiles and crystals small enough to fit within the  
 homogeneous part of the X-ray beam.
 In case of in-homogeneous beams, a combination of numerical crystal face   
 based correction and multi-scan correction is recommended. 

 The predicted and reported transmission ranges are found to be identical   
 which is not to be expected. CIF's generated with SHELXL97 report  
 transmission ranges based on the crystal dimensions supplied on the
 SIZE card. Those values have nothing to do with the actual corrections 
 for absorption as applied to the data: they just report the EXPECTED range.
 Some correction packages (e.g. SADABS) will provide only one   
 'relative-correction-factor'. In such cases, Tmax should be given as   
 Tmax-expected and Tmin = relative-correction-factor * Tmax.

 Minimum and Maximum dimensions are likely exchanged in the CIF. 

 In the ideal case, both data items should have the same value. If not, 
 the reason should be clear. A reason might be the output by SHELXL 
 of population parameters to the CIF with only two decimals.
 Note: SHELXL counts the number of electrons in the unit cell. The result   
 will in general be an integer. This is also the number checked for here.   
 The official definition calls for 'The effective number of electrons in the
 crystal unit cell contributing to F(000)'. It may contain dispersion   
 contributions and is calculated as:
 F(000) = [ (sum f~r~)^2^ + (sum f~i~)^2^ ]^1/2^
 f~r~ = real part of the scattering factors at theta = 0
 f~i~ = imaginary part of the scattering factors at theta = 0   

 The CIF contains duplicate labels posing interpretation problems for   
 PLATON/CHECK. Derived geometry ALERTS may have their origin in this problem.

 The CIF contains labels posing problems for PLATON/CHECK.  
 Example: label HN1 with no scattering type information supplied.   
 Validation is aborted. 

 The first parameter on the SHELXL weighting line and used in the refinement
 has an exceptionally large value. SHELXL will recommend 'WGHT 0.2 0.0' to
 be used in the next refinement when it judges that the refined weight
 parameters values to be excessive. This points to a likeable problem with the
 reflection data and/or the refinement model such as improper reflection
 s.u.'s or an unresolved problem such as missed twinning.

 The structure contains refined hydrogen atoms. However the data item   
 '_refine_ls_hydrogen_treatment' has the value 'constr'. The value 'mixed' is
 more appropriate. 

 The CIF contains an atom with occupancy less than 0.0001
 Please check whether the zero occupancy specification is intended. 

 The CIF contains an atom with Occupancy greater than 1.0.  

 The CIF contains an atom sitting on a special position with occupancy  
 specified as less than 1.0. This is often an error and the result of the   
 confusion of the notions 'occupancy' and 'population parameter'. The first 
 should be 1.0 for a fully occupied site. The latter multiplies the 
 site-symmetry with the occupancy. Thus, for a fully occupied site on a 
 mirror plane the site-symmetry will be 0.5 * 1.0 = 0.5.
 Note: a wrong occupancy number will lead to an incorrect expected chemical
 formula.  

 The unit cell contains a non-integer number of atoms of a given atom type. 
 Valid reasons include partially occupied (solvent) sites and substitutional
 disorder.  

 The structure contains no hydrogen atoms. However the data item
 _atom_sites_solution_hydrogen had the value 'geom'. This value is likely the
 SHELXL default and should be replaced by '.'.   

 The structure contains no hydrogen atoms. However the data item
 _refine_ls_hydrogen_treatment has the value 'mixed'. This value is likely the
 SHELXL default and should be replaced by '.'.   

 Convergence of the refinement is proved with a close to zero shift/error   
 value for all refined parameters. Such a convergence is easily achieved
 with a few additional refinement cycles at little cost.
 Note: Some SHELXL-97 versions do not allow for negative Flack parameter
 values. Convergence in such a case may be never reached because the Flack
 parameter value is reset to zero.

 A higher than usual R1 indicates either an insufficient model or poor  
 quality data.  

 The second parameter on the SHELXL weighting line has an exceptionally large 
 value. This may indicate either improper reflection s.u.s or an unresolved
 problem such as missed twinning. 

 wR2 will in general have a value twice of that of R1 with refinement on F**2.  
 Significantly larger values usually indicate a poor refinement model. Also
 check for unaccounted for twinning.   

 The weighting scheme is found to be left at the SHELXL default. This
 default value is recommended for the preliminary structure refinement. It is
 uncommon that this unoptimized weight gives the best final refinement result.

 The Goodness-of-Fit value S should in general be close to 1 at the end of a
 refinement with a proper weighting scheme. If not, there might be significant
 unresolved problems with the refinement model or reflection data.

 The Goodness-of-Fit value S should in general be close to 1 at the end of a
 refinement with a proper weighting scheme. If not, there might be significant
 unresolved problems with the refinement model or reflection data.

 The data/parameter ratio should in general be higher than 10 for a quality 
 structure determination. This ratio can be improved by not refining
 C-H parameters other than riding on their carrier atom.

 The data/parameter ratio should in general be higher than 7 for a quality  
 determination of a structure containing atoms with Z less than 18. This
 ratio can be improved by not refining C-H parameters other than riding on
 their carrier atom. 
 Note: The number of reflections used in this ratio is the number obtained  
 by Laue group averaging.   

 The data/parameter ratio should in general be higher than 8 for a quality  
 determination for a structure containing heavy atoms with ZMAX greater 
 than 17. This ratio can be improved by not refining C-H parameters other   
 than riding on their carrier atom.
 The number of reflections used for this ratio is the Laue averaged number
 of reflections. The validation criteria used are expected to be met with a
 dataset complete with a resolution sin(theta)/lambda .GE. 0.6 Ang-1. 

 No Wavelength specification found in the CIF.  

 The wavelength, specified in the CIF, is not CuKa = 1.5418A, GaKa = 1.3414A,   
 MoKa = 0.71073A, AgKa = 0.56086A or InKa = 0.51359A radiation within a 
 tolerance of 0.0005A.  Valid exceptions are Kb, Neutron and Synchrotron data.  

 The 'mixed' type Hydrogen atom refinement is reported (SHELXL-97 default). 
 However, no Hydrogen atoms with freely refined positions are found in the  
 CIF. Likely, the value 'constr' or 'refU' for  
 '_refine_ls_hydrogen_treatment' will be more appropriate (e.g. when all
 Hydrogen atoms have been refined in the riding mode on their carrier atom).

 The ratio of the maximum and minimum residual density excursions is
 unusual. This might indicate unaccounted for twinning or missing atoms 
 (e.g. associated with disordered solvent). 

 No residual electron density maximum given in CIF. 

 No residual electron density minimum given in CIF. 

 A residual density maximum larger than expected is reported. This might be 
 caused by residual absorption artefacts, unaccounted for twinning, disorder,   
 wrongly assigned atom types, missing hydrogen atoms and other model errors.

 A residual density minimum larger than expected is reported. This might be 
 caused by residual absorption artefacts, disorder, wrongly assigned atom   
 types and other model errors.  

 Likely interchanged maximum and minimum values. Alternatively, the minimum
 residual density has the (unlikely) value zero.  

 The norm is integer. The CIF-definition for _cell_formula_units_Z 
 specifies Z values as 1 -> infinity. SHELXL may report in certain cases a
 real number (E.g. Z = 0.67). In such a case, check whether really more
 decimals are intended and edit accordingly, i.e. 0.67 => 0.66667.

 Check the reported crystal system against the reported space group.
 Alternatively, no crystal system was reported. 

    
 The wavelengths reported in CIF format and on the CELL record in the embedded
 RES file should be identical. 

 Inversion twinning is meaningless in centrosymmetric spacegroup.   

 The SHELXL TWIN instruction matrix corresponds to a symmetry operation of the  
 non-centrosymmetric Laue group of the crystal structure. This is not a valid   
 twin operation.

 The SHELXL TWIN instruction matrix corresponds to an inverted 
 non-centrosymmetric pointgroup operation belonging to the corresponding 
 centrosymmetric Laue group. The reported twinning factor is actually the
 Flack x value. The close to zero CIF reported Flack parameter value is 
 meaningless.  Please re-refine with the TWIN instruction replaced by 'TWIN'
 (i.e. without numerical data).

 Tests for missed symmetry are done with ADDSYM, an extended MISSYM (C) 
 clone. These tests warn for missed or possible higher (pseudo) symmetry in 
 the structural model (i.e. based on the coordinate data).  
 Close examination of the situation at hand is indicated in order to
 prove/disprove the issue (usually in combination with the reflection data).
 Report on potential (pseudo/real) lattice centering or cell halving.   
 Note: H-atoms and disordered atoms are not taken into account in the tests.

 Tests for missed symmetry are done with ADDSYM, an extended MISSYM (C) 
 clone. These tests warn for missed or possible higher (pseudo) symmetry in 
 the structural model (i.e. based on the coordinate data). Close examination
 of the situation at hand with PLATON/ADDSYM is indicated in order to prove/
 disprove the issue (usually in combination with the reflection data). 
 This ALERT reports on a potential additional (pseudo/real) inversion centre.   
 A pseudo-centre may be incompatible with existing symmetry elements.   
 Chiral molecules are incompatible with an inversion centre.
 Note: H-atoms and disordered atoms are not taken into account in the test. 

 Tests for missed symmetry are done with ADDSYM, an extended MISSYM (C) 
 clone. These tests warn for missed or possible higher (pseudo) symmetry in 
 the structural model (i.e. based on the coordinate data). Close examination
 of the situation at hand is indicated in order to prove/disprove the issue 
 (usually in combination with the reflection data). This ALERT reports on
 potential additional (pseudo/real) rotation axes and mirrors.   
 In addition, (pseudo/real) lattice centering/translations are reported as  
 A, B, C, I, X, Y, Z, S. (Here S stands for special and not covered by the  
 preceding types). Full details on the situation at hand should be gleaned  
 from an actual PLATON/ADDSYM run.  
 Note: Atom types are treated in this test as EQUAL for structures with less
 than 250 atoms in the asymmetric unit in order to detect cases of misassigned
 atom types.  
 Chiral molecules are incompatible with an inversion centre or (glide)planes.   
 Note: H-atoms and disordered atoms are not taken into account in the tests.

 Tests for missed symmetry are done with ADDSYM, an extended MISSYM (C) 
 clone. These tests warn for missed or possible higher (pseudo) symmetry in 
 the structural model (i.e. based on the coordinate data). Close examination
 of the situation at hand is indicated in order to prove/disprove the issue 
 (usually in combination with the reflection data). 
 Chiral molecules are incompatible with an inversion centre or (glide)planes.   
 For an example of reported pseudo-symmetry see I.A.Guzei et al, (2002).
 Acta Cryst. C58, m141-m143.
 Note: H-atoms and disordered atoms (i.e. atoms with population less
 than 1.0) are not taken into account in the tests. This may artificially   
 lead to a symmetry higher than the actual one. 
 Note: Atoms are treated as having the same atom type in order to catch 
 certain types of disorder or incorrect atom type assignment.   

 ADDSYM has problems to reconstruct a space group from the symmetry operation
 found in the symmetry expanded coordinate set. The reason being either
 intricate additionally detected pseudo-symmetry or serious errors in the data
 set.   

 Tests for missed symmetry are done with ADDSYM, an expanded MISSYM (C) 
 clone. This ALERT reports on local inversion symmetry, not compatible  
 with the reported space-group symmetry.
 Note: H-atoms and disordered atoms are not taken into account in the test. 

 
 A (Pseudo) Lattice translation was detected and implemented before the 
 current ADDSYM analysis. 

 A symmetry operation should be specified in the CIF either without 
 spaces or between quotes.  

 Space group symmetry should be provided in the CIF both explicitly with a  
 '_symmetry_equiv_pos_as_xyz' loop and implicitly with
 '_symmetry_space_group_name_H-M'.
 An unusual (non-standard) choice of origin may also raise this ALERT.  
 Please check and Explain.  

 Symmetry in the CIF should be provided both explicitly with a  
 '_symmetry_equiv_pos_as_xyz' loop and implicitly with
 '_symmetry_space_group_name_H-M'.
 Test for valid '_symmetry_space_group_name_H-M' symbol.  

 Symmetry in the CIF should be provided both explicitly with a  
 _symmetry_equiv_pos_as_xyz loop and implicitly with
 _symmetry_space_group_name_H-M.
 Test for missing (i.e. ?) '_symmetry_space_group_name_H-M' symbol.   

 Symmetry in the CIF should be provided in the CIF both explicitly with a   
 '_symmetry_equiv_pos_as_xyz' loop and implicitly with
 '_symmetry_space_group_name_H-M' or '_symmetry_space_group_name_H-M_alt'.  
 Test for uninterpretable or inconsistent Space group information.  

 Symmetry in the CIF should be provided in the CIF both explicitly with a   
 '_symmetry_equiv_pos_as_xyz' loop and implicitly with
 '_symmetry_space_group_name_H-M'.
 Test for uninterpretable or absent explicit symmetry records.  

 Optionally specify the Hall symbol. The Hall symbol provides an
 unambiguous definition of the space-group symmetry where the Hermann-  
 Mauguin symbol leaves room for alternative choices of the origin.  
 E.g. for space-group P21, the screw axis is in general taken to coincide   
 with the b-axis. However, sometimes it is chosen to be shifted by 1/4  
 in the c-axis direction to bring out the relation with P21/c. The Hall 
 symbols will be 'P 2yb' and 'P 2ybc' respectively. 
 Refer to: S.R.Hall, Space Group Notation with an Explicit Origin;  
           Acta Cryst. (1981), A37, 517-525.
       or: http://www.kristall.ethz.ch/LFK/software/sginfo/hall_symbols.html

 The reported Hall-symbol is found to be in error or uninterpretable.   
 Refer to: S.R.Hall, Space Group Notation with an Explicit Origin;  
           Acta Cryst. (1981), A37, 517-525.
       or: http://cci.lbl.gov/sginfo/hall_symbols.html

 The reported Hall-symbol is not identical to the one internally tabulated
 for the space group as derived from the explicitly supplied symmetry
 operations. As an example, the Hall symbols '-P 2yn' and '-P 2yabc' both 
 correspond with space group 'P21/n'. Alternatively, no Hall-symbol could be
 derived by PLATON for the explicit set of symmetry operations. This may be
 the case when an unusual origin is chosen. 
 Refer to: S.R.Hall, Space Group Notation with an Explicit Origin;  
           Acta Cryst. (1981), A37, 517-525.
       or: http://cci.lbl.gov/sginfo/hall_symbols.html

 P21/n and I2/a etc. settings are often preferred over P21/c and C2/c when  
 that leads to a closer to 90 degrees beta angle.   
 Standard choice of origin is to be preferred. A non-standard origin choice 
 might be useful for for the comparison of related structures.  

 The reported space-group name is unusual.  

 Symmetry constraints on cell dimensions are checked.   

 Symmetry constraints on cell dimensions are checked.   

 Symmetry constraints on cell dimensions are checked.   

 Symmetry constraints on cell dimensions are checked.   

 Symmetry constraints on cell dimensions are checked.   

 Symmetry constraints on cell dimensions are checked.   

 Symmetry constraints on cell dimensions are checked.   

 Symmetry constraints on cell dimensions are checked.   

 Symmetry constraints on cell dimensions are checked.   

 Symmetry constraints on cell dimensions are checked.   

 Symmetry constraints on cell dimensions are checked.   

 The s.u. on the a-axis is small or missing. The presence of s.u.'s (where  
 required) and value are checked. S.u.'s as given by the diffractometer 
 software are often much smaller than realistic.

 The s.u. on the b-axis is small or missing. The presence of s.u.'s (where  
 required) and value are checked. Su's as given by the diffractometer software
 are often much smaller than realistic.

 The s.u. on the c-axis is small or missing. The presence of s.u.'s (where  
 required) and value are checked. Su's as given by the diffractometer software
 are often much smaller than realistic.

 The s.u. on alpha is small or missing. The presence of s.u.'s 
 (where required) and value are checked. Su's as given by the diffractometer
 software are often much smaller than realistic. 

 The s.u. on beta is small or missing. The presence of s.u.'s (where required)  
 and value are checked. Su's as given by the diffractometer software are often
 much smaller than realistic. 

 The s.u. on gamma is small or missing. The presence of s.u.'s
 (where required) and value are checked. Su's as given by the diffractometer
 software are often much smaller than realistic. 

 There should be no s.u. on symmetry constrained cell angles.   
 Example: No s.u. on alpha, beta and gamma for orthorhombic symmetry.   

 The s.u. on the reported -axis is unexpectedly large.  

 The s.u. on the reported angle is too large.   

 An ALERT is issued when the reported unit cell volume differs significantly
 from the volume calculated on the basis of the supplied cell dimensions.   

 Missing s.u. on cell volume.   

 Some software packages calculate Volume s.u.'s incorrectly. The correct
 formula  for triclinic, monoclinic and orthorhombic systems
 (based on the propagation of error expression) may be found in:
 M. Nardelli, Computer & Chemistry, (1983), 7, 95-98.   
 or 
 C. Giacovazzo ed. in 'Fundamentals of Crystallography', Second Edition,
 Oxford University Press, 2003,  p135.  
 Also note that several cell parameters for higher symmetry cells are no
 longer independent. S.u. calculations need special treatment in those cases.   

 The reported cell axes s.u.'s are reported equal. Please check whether 
 this is correct or a software default value.   

 The reported cell angle s.u.'s are reported equal. Please check whether
 this is correct or a software default value.   

 Unless for special reasons related to the structure/content, a unit cell and   
 structure is best reported with reference to the Niggli Reduced Cell. This
 ALERT may originate also from a failure to order the axes from small to large
 dimension.

 The axial order should be from small to large in the triclinic cell.   

 By convention, the Monoclinic beta angle is always chosen to be larger than
 90.0 Degrees. A trivial transformation (1 0 0/0 -1 0/0 0 -1) should be 
 applied to the data.   

 Unless for special reasons related to the structure/content, a unit cell and   
 structure is best reported with reference to the Niggli Reduced Cell.  

 Missing or Zero s.u. (esd) on x-coordinate.
 Positional parameters for all non-hydrogen atoms in general positions are  
 checked for the presence of a non-zero s.u. on them. This includes 
 parameters fixed to fix the origin in polar space-groups which is no longer
 necessary when refinement is done with modern programs (e.g. SHELXL, XTAL).

 Missing or Zero s.u. (esd) on y-coordinate.
 Positional parameters for all non-hydrogen atoms in general positions are  
 checked for the presence of a non-zero s.u. on them. This includes 
 parameters fixed to fix the origin in polar space-groups (e.g. P21) which  
 is no longer necessary when refinement is done with modern programs
 (e.g. SHELXL, XTAL).   

 Missing or Zero s.u. (esd) on z-coordinate.
 Positional parameters for all non-hydrogen atoms in general positions are  
 checked for the presence of a non-zero s.u. on them. This includes 
 parameters fixed to fix the origin in polar space-groups (e.g. P41) which  
 is no longer necessary when refinement is done with modern programs
 (e.g. SHELXL, XTAL).   

 Warning: Refined C-H H-atoms in heavy-atom structure (i.e. containing an   
 element beyond element #18). Such H-atoms are in general better refined
 at calculated positions riding on the atoms they are attached to. A
 better data over parameter ratio will be achieved. 

 Report on restrained (riding) Non-Hydrogen atoms. Note: This may lead to   
 non meaningful bond and angle s.u.'s (ALERTS _751, _752). 
 R-flagged atoms may arise unintentional being caused by an "AFIX 0" line   
 being missing in a shelxl.ins file (SHELXL-97 refinement). 
 Alternatively, the number of refined parameters may have been limited  
 deliberately (e.g. by refinement of C-F with fixed known geometry, 
 similar to C-H) in order to keep the data/parameter ratio acceptable.  

 Calc-flagged atoms are not supposed to carry s.u.'s on their coordinates.  

 SHELXL applies a default effective standard deviation value of 0.04 Ang for
 the DANG restraints. The use of a much stronger restraint, with a lower
 value of those parameters, may hide serious problems with a structure.

 The use of EXYZ record(s) in the SHELXL .res file should be documented in the  
 experimental section of the associated publication.

 The use of 'AFIX 1' record(s) in the SHELXL '.res' file should be documented  
 in the experimental section of the associated publication.

 Insufficient data encountered in the coordinate loop. A possible cause might
 be the use of a SHELX style '=' continuation line.  

 The use of EADP record(s) in the SHELXL '.res' file should be documented in  
 the experimental section of the associated publication.

 The use of DFIX record(s) should be documented in the experimental section 
 of the associated publication. 

 The use of DANG record(s) should be documented in the experimental section 
 of the associated publication. 

 The use of FLAT record(s) should be documented in the experimental section 
 of the associated publication. 

 The use of SAME record(s) in the SHELXL '.res' file should be documented in
 the experimental section of the associated publication.

 The use of SADI record(s) in the SHELXL '.res' file should be documented in 
 the  experimental section of the associated publication.

 The use of DELU record(s) in the SHELXL .res file should be documented in the  
 experimental section of the associated publication.

 The use of SIMU record(s) in the SHELXL '.res' file should be documented in
 the experimental section of the associated publication.

 The use of CHIV record(s) in the SHELXL '.res' file should be documented in
 the experimental section of the associated publication.

 It is unusual that more cell parameters end with a zero and the s.u. is 10.
 This problem might be caused by the specification of insufficient  
 'meaningful' digits as compared to the reported s.u.   
 See also: W.Clegg, Acta Cryst. (2003) E59, e2-e5.  

 One of the angles in a monoclinic cell is expected to be not exactly 90
 degrees.   

 One angle should have an s.u. greater than zero in a monoclinic cell.  

 Please supply the value for '_cell_measurement_reflns_used'. 

 Please supply the value for '_cell_measurement_theta_min'.   

 Please supply the value for _cell_measurement_theta_max.   

 The use of ISOR record(s) in the SHELXL '.res' file should be documented in
 the experimental section of the associated publication.

 The use of RIGU record(s) in the SHELXL '.res' file should be documented in
 the experimental section of the associated publication.

 SHELXL applies a default effective standard deviation value of 0.04 for the
 SIMU restraint. The use of a much stronger restraint with a lower value of 
 that parameter may hide serious problems with a structure such as wrong atom
 type assignments.   

 SHELXL applies a default effective standard deviation value of 0.02 Ang for
 the SAME 1,2 or 1,3 bond restraints. The use of a much stronger restraint,
 with a lower value of those parameters, may hide serious problems with a
 structure.

 SHELXL applies a default effective standard deviation value of 0.004 for the
 RIGU restraint. The use of a much stronger restraint with a lower value of
 that parameter may hide serious problems with a structure such as wrong atom
 type assignments.

 SHELXL applies a default effective standard deviation value of 0.02 Ang for
 the SADI bond restraints. The use of a much stronger restraint, with a lower
 value of those parameters, may hide serious problems with a structure.

 SHELXL applies a default effective standard deviation value of 0.01 Ang for
 the DELU restraints. The use of a much stronger restraint, with a lower
 value of those parameters, may hide serious problems with a structure.

 The reported '_cell_measurement_temperature' deviates from the reported  
 '_diffrn_ambient_temperature value'. The relevant set of cell parameter
 values are those at the datacollection temperature (i.e. ambient temperature
 at the crystal) because the derived geometry parameter values will be correct
 and meaningful only when those cell parameter values are used. When relevant,
 cell parameter values at temperatures differing from the
 '_diffrn_ambient_temperature' are best archived as comment value under 
 _cell_special_details.   

 Non-default DEFS parameter values in CIF-Embedded .res file. 
 See SHELXL manual.

 SHELXL applies a default effective standard deviation value of 0.02 Ang for
 the DFIX restraints. The use of a much stronger restraint, with a lower
 value of those parameters, may hide serious problems with a structure.

 SHELXL places Hydrogen atoms at calculated positions with default distance
 values to their parent atoms depending on the temperature reported in the
 TEMP record. When the TEMP record is absent in the embedded '.res' file a 
 default temperature of 293K is assumed. An ALERT is issued when the CIF 
 reported measurement temperature differs from that value.

 Please specify the temperature (kelvin) at which the unit cell parameters  
 were determined. That temperature should be generally identical to the 
 temperature value specified with '_diffrn_ambient_temperature' (the
 temperature of the crystal during data-collection) in order to make sense
 when used in the calculation of derived geometry parameter values (i.e. bond
 distances, bond angles etc.).

 Please specify the temperature (kelvin) at which the intensity data were   
 collected. The reported cell parameter values should be those at this 
 temperature to make sense of the derived geometry parameter values. 

 The cell determination temperature is set in the CIF by default by SHELXL97
 to 293 K when the TEMP instruction is not used. The actual temperature is  
 likely either slightly or significantly (in case of a low temperature data 
 collection)  different. A likely erroneous temperature of 273K (0C) is 
 also flagged.  

 The data collection temperature is set in the CIF by default by SHELXL97 to
 293 K if the TEMP instruction is not used. A likely erroneous temperature of   
 273K (0C) is also flagged. The actual temperature is likely either slightly
 or significantly (for a low temperature data collection) different.

 This test reports on non-hydrogen atoms that were refined with isotropic   
 displacement parameters only in the main residue. Such a practice is unusual   
 by modern standards and only needed for minor disorder modelling.  

 This test reports on isotropically refined atoms in small moieties (usually
 anions or solvent).

 Isotropic U(iso) values are expected to have a positive value. 

 No anisotropically refined atoms in CIF ?  

 This test reports on non-positive definite (i.e. complex and unrealistic)  
 anisotropic displacement parameters in the main residue.   

 This test reports on non-positive definite (i.e. complex and   
 unrealistic) anisotropic displacement parameters in an anion or solvent
 residue.   

 The main axes values of the ADP(S) of the main residue(s) are determined and   
 ordered: U1 < U2 < U3. The value of SQRT(U3/U1) main axis ADP ratio (Angstrom  
 Units) is tested for the main residue(s). Large values may indicate unresolved 
 disorder. Oblate criterium: U3 - U2 < U2 - U1. Prolate otherwise.  

 The main axes values of the ADP(S) of the minor residue(s) are determined and  
 ordered: U1 < U2 < U3. The value of SQRT(U3/U1) main axis ADP ratio (Angstrom  
 Units) is tested for the main residue(s). Large values may indicate
 unresolved disorder. Oblate criterium: U3 - U2 < U2 - U1. Prolate otherwise.  

 The maximum and minimum main axis ADP ratio (Angstrom Units) is tested for 
 the main residue. Large values may indicate unresolved disorder.   

 The maximum and minimum main axis ADP ratio (Angstrom Units) is tested for 
 the minor residue(s). Large values may indicate unresolved disorder.   

 Check & Correct U(aniso) data for completeness etc.
 Do not use SHELX style '=' continuation line.  

 U(i,j) components have been constrained (i.e. without s.u.) in the refinement

 This test reports on a larger than usual U(eq) range for the specified 
 element type in the non-solvent/anion part of the structure.   
 Too high or too low Ueq's may be an indication for incorrectly 
 identified atomic species (i.e. O versus N).   

 This test reports on a larger than usual U(eq) range for the non-hydrogen  
 atoms solvent/anion. Too high or too low Ueq's may be an indication for
 incorrectly identified atomic species (i.e. Br versus Ag). 

 This test reports on a larger than usual range of U(eq) values for hydrogen
 atoms in the non-solvent/anion part of the structure.  
 Possible causes are:   
 1 - disorder, e.g. in t-butyl moieties.
 2 - poor data, not adequate for the refinement of individual displacement  
     parameters.
 3 - Misplaced hydrogen atoms (i.e. there is no density at the position 
     where one of the H-atoms is positioned).   

 This test reports on large ranges in displacement parameters for hydrogen  
 atoms in the solvent/anion part of the structure.  

 This test reports on a large difference between Ueq in the CIF and the Ueq 
 calculated from the 6 reported U(i,j) values.                            .

 The components of the anisotropic displacement parameters along chemical   
 bonds are assumed to be equal in magnitude. Large differences might
 indicate contamination of these parameters with other (unresolved) effects 
 such as (substitutional) disorder, model or data errors and/or 
 over-refinement. Atomic sites assigned the wrong scattering type (e.g. Ag  
 versus Br) should generate 'problem signals' with this test. Data sets 
 corrected for absorption effects with DELREF techniques (e.g. DIFABS, SHELXA,  
 XABS2) often show large DELU values for bonds involving the heaviest atom. 
 Note: The original 'Hirshfeld-test' was defined in absolute terms (see 
 F.L.Hirshfeld, Acta Cryst. (1976). A32, 239-244). The current test is with 
 reference to the associated standard uncertainty.  

 The components of the anisotropic displacement parameters along chemical   
 bonds are assumed to be equal in magnitude. Large differences might
 indicate the contamination of these parameters with other (unresolved)
 effects such as (substitutional) disorder, model or data errors and/or 
 over-refinement. Atomic sites assigned the wrong scattering type (e.g. Ag  
 versus Br) should generate 'problem signals' with this test. Data sets 
 corrected for absorption effects with DELREF techniques (e.g. DIFABS, SHELXA,  
 XABS2) often show large DELU values for bonds involving the heaviest atom. 
 Note: The original 'Hirshfeld-test' was defined in absolute terms (see 
 F.L.Hirshfeld, Acta Cryst. (1976). A32, 239-244). The current test is with 
 reference to the associated standard uncertainty.  

 The components of the anisotropic displacement parameters along chemical   
 bonds are assumed to be equal in magnitude. Large differences might
 indicate contamination of these parameters with other (unresolved) effects 
 such as (substitutional) disorder, model or data errors and/or 
 over-refinement. Atomic sites assigned the wrong scattering type (e.g. Ag  
 versus Br) should generate 'problem signals' with this test. Data sets 
 corrected for absorption effects with DELREF techniques (e.g. DIFABS, SHELXA,  
 XABS2) often show large DELU values for bonds involving the heaviest atom. 
 A special case are M-C=O type of systems that generally show significant   
 differences for the M-C bond. See D.Braga & T.F. Koetzle (1988), Acta Cryst.   
 B44, 151-155). 
 Note: The original 'Hirshfeld-test' was defined in absolute terms (see 
 F.L.Hirshfeld, Acta Cryst. (1976). A32, 239-244). The current test is with 
 reference to the associated standard uncertainty.  
 Note: The 'Hirshfeld-test' ALERTS are suppressed for polymeric or disordered 
 structures.

 The components of the anisotropic displacement parameters along chemical   
 bonds are assumed to be equal in magnitude. Large differences might
 indicate contamination of these parameters with other (unresolved) effects 
 such as (substitutional) disorder, model or data errors and/or 
 over-refinement. Atomic sites assigned the wrong scattering type (e.g. Ag  
 versus Br) should generate 'problem signals' with this test. Data sets 
 corrected for absorption effects with DELREF techniques (e.g. DIFABS, SHELXA,  
 XABS2) often show large DELU values for bonds involving the heaviest atom. 
 Note: The original 'Hirshfeld-test' was defined in absolute terms (see 
 F.L.Hirshfeld, Acta Cryst. (1976). A32, 239-244). The current test is with 
 reference to the associated standard uncertainty.  

 The components of the anisotropic displacement parameters along chemical   
 bonds are assumed to be equal in magnitude. Large differences might
 indicate contamination of these parameters with other (unresolved) effects 
 such as (substitutional) disorder, model or data errors and/or 
 over-refinement. Atomic sites assigned the wrong scattering type (e.g. Ag  
 versus Br) should generate 'problem signals' with this test. Data sets 
 corrected for absorption effects with DELREF techniques (e.g. DIFABS, SHELXA,  
 XABS2) often show large DELU values for bonds involving the heaviest atom. 
 Note: The original 'Hirshfeld-test' was defined in absolute terms (see 
 F.L. Hirshfeld, Acta Cryst. (1976). A32, 239-244). 

 The U(eq) value of an atom is compared with the average U(eq) for to   
 non-hydrogen atoms bonded to it. Large differences may indicate that the   
 wrong atom type was assigned (e.g. N instead of O).

 The U(eq) value of an atom is compared with the average U(eq) for  
 non-hydrogen atoms bonded to it. Large differences may indicate that the   
 wrong atom type was assigned (e.g. N instead of O). False alarms may   
 occur for terminal groups such as the t-butyl moiety.  

 The U(eq) value of an atom in the solvent or ion is compared with the average  
 U(eq) for non-hydrogen atoms bonded to it. Large differences may indicate  
 that the wrong atom type was assigned (e.g. N instead of O).   

 The U(eq) value of an atom in the solvent or ion is compared with the average  
 U(eq) for non-hydrogen atoms bonded to it. Large differences may indicate  
 that the wrong atom type was assigned (e.g. N instead of O). False alarms may  
 occur for terminal groups such as the t-butyl moiety.  

 U(iso) of a hydrogen atom is generally expected to be greater than the 
 U(eq) of the non-hydrogen atom it is attached to.  

 An average value of the U(i,j) tensor of the asymmetric unit of a residue  
 is calculated. An ALERT is generated when the corresponding main axis U3/U1
 ratio deviates significantly from 1.0. Large values of this ratio should be
 taken as an indication of possible systematic errors in the data or errors
 in the model. Visual inspection of an ORTEP plot will show that many  
 displacement ellipsoids have their major axis pointing in the same direction.  

 A large average Ueq value for a residue may indicate refinement with a too 
 high (possibly fixed) population parameter value.  

 This site is expected to be fully occupied but has been constrained (e.g.  
 with a SHELXL FVAR variable) or fixed at a value less than 1.0. Please check   
 for incomplete (substitutional) disorder handling. 

 This site is expected to be fully occupied but has been constrained (e.g.  
 with a SHELXL FVAR variable) or fixed at a value less than 1.0. Please check   
 for incomplete (substitutional) disorder handling. 

 Atom sites that are not fully occupied are counted. A large fraction of
 disordered atoms may be both a signal for serious structure analysis   
 problems or less reliable/interesting results. 

 Atom sites that are not fully occupied are counted.

 Hydrogen atoms are generally connected to only one other atom. A hydrogen  
 atom between two oxygen atoms is a special case. Investigate whether this  
 hydrogen atom is better described with a disorder model with two partially 
 occupied sites. A difference map might show a double-well density. 

 A non-integer number of atoms has been found in a residue. 

 This test reports on hydrogen atoms that are not on bonding distance to any
 atom. This ALERT may indicate that the hydrogen atom refined to a  
 non-bonding position or needs a symmetry operation to bring it to bonding  
 distance. It also may indicate a problem with incompatible population  
 parameters (e.g. C - H with population 0.8 and 0.9 respectively).  

 This test reports on oxygen atoms that are not within bonding distance to  
 any other atom in the structure. A common reason may be that no hydrogen   
 atoms are given for a water molecule. Attempts should be made to locate
 those hydrogen atoms from a difference map.

 This test reports on metal atoms that are not bonded or at coordination
 distance of other atoms. Isolated ions are very unusual (or non-existent ?)

 This test reports on single bonded (coordinated) metal atoms/ions. This
 represents a very unusual situation. There are literature examples where such  
 a 'single bonded metal' was shown to be a halogen. 

 Single bonded Oxygen with C-O > 1.3 Angstrom. Missing H-Atom ? Check   

 This test identifies (very) short contacts between atoms that only becomes 
 apparent after the application of symmetry on the primary coordinate set.  

 This test reports on oxygen atoms (not full weight) that are not within
 bonding distance to any other atom in the structure.   
 A common reason may be that no hydrogen atoms are given for a water molecule.  

 Strange C-O-H geometry with C-O  <   1.25 Angstrom detected. Misplaced 
 H-Atom ?   

 Oxygen atom with three covalent bonds detected. Check for correct atom type
 assignment (e.g. N rather than O)  
 Note: Exceptions are H3O+ (Oximium or Hydroxonium) and 
                      H5O2+ (Hydronium or aqua-hydroxonium) species.

 A water molecule coordinated to a metal is detected with an unusually small 
 value of the Metal-Oxygen-Hydrogen Angle.

 Check for missing H-atoms. 

 An sp3 hybridized C was detected as part of a C=N moiety. Only one 
 attached H atom in sp2 configuration is expected and not two.  
 In SHELXL terms this corresponds with an erroneous AFIX 23 rather than an  
 AFIX 43 type of H atom position generation and refinement. 

 An sp3 hybridized C was detected as part of a C=N moiety. Only one 
 attached H atom in sp2 configuration is expected and not two.  
 In SHELXL terms this corresponds with an erroneous AFIX 23 rather than an  
 AFIX 43 type of H atom position generation and refinement. 

 The test attempts to assign one of three hybridisations to N atoms in main 
 residue: sp, sp2 or sp3 on the basis of the angles around N.   
 This ALERT may indicate a mis-assigned H atom position (e.g. an atom   
 placed in a sp2 position instead of sp3).  

 The test attempts to assign one of three hybridisations to N atoms in main 
 residue: sp, sp2 or sp3 on the basis of the angles around N.   
 This ALERT may indicate a mis-assigned H atom position (e.g. an atom   
 placed in a sp2 position instead of sp3).  

 The test attempts to assign one of three hybridisations to C atoms in main 
 residue: sp, sp2 or sp3 on the basis of the angles around C. In this way,  
 missing H atoms or too many H-atoms on a carbon atom should be detected.   

 The test attempts to assign one of three hybridisations to C atoms in  
 solven/anion: sp, sp2 or sp3 on the basis of the angles around C. In this  
 way missing H atoms or too many H-atoms on a carbon atom should be detected.   

 The test attempts to assign one of three hybridisations to a non-C atom in 
 the main residue: sp, sp2 or sp3 on the basis of the angles around the non-C
 atom. In this way, missing H atoms or too many H-atoms should be detected. 

 The test attempts to assign one of three hybridisations to a non-C atom in 
 the solvent/anion: sp, sp2 or sp3 on the basis of the angles around the
 non-C atom. In this way, missing H atoms or too many H-atoms should be 
 detected.  

 Check for possibly missing Hydrogen atom on Nitrogen coordinating to a metal
 in the main residue.   

 Check for possibly missing Hydrogen atom on Nitrogen coordinating to a metal
 in the solvent/anion.  

 Check for possibly missing Hydrogen atom on Carbon with sp3-like geometry  
 in the main residue.   

 Check for possibly missing Hydrogen atom on Carbon with sp3-like geometry  
 in the solvent/anion.  

 Check for a possibly missing Hydrogen atom on Phosphorus with sp3-like 
 geometry.  

 The hybridisation of this calbon atom could not be established from the
 number of substituents and their geometry. Check for missing hydrogen atoms
 and/or poor geometry.  

 The standard average C-C bond distance in a phenyl ring is 1.395 Angstrom. 
 The actual average ring distance may be larger than expected due to
 systematic errors in the unit cell dimensions (e.g. the use of an incorrect
 wavelength value for the determination of the unit cell parameters).

 The standard average C-C bond distance in a phenyl ring is 1.395 Angstrom. 
 The average ring distance may be smaller due to large thermal motion or
 incorrect unit cell dimensions (e.g. the use of an incorrect wavelength value
 for the determination of the unit cell parameters). 

 The standard average C-C in a phenyl ring is 1.395 Angstrom. Bond distances
 in the ring are expected to vary only slightly due to thermal motion or
 substituent effects. Large deviations are likely due to data or model errors.  

 The standard average C-C bond distance in a flat six carbon atom containing
 aromatic ring is 1.395 Angstrom.   
 The actual average ring distance may be larger than expected due to
 substituents such as '=O', single bonds or systematic errors in the cell   
 dimensions (E.g. when the wrong wavelength is used in the derivation of
 the cell parameters).  

 The standard average C-C bond distance in a phenyl ring is 1.395 Angstrom.
 The average ring distance may be smaller due to large thermal motion,  
 substituents such as '=O' or incorrect cell dimensions.

 The standard average C-C bond distance in a benzene ring is 1.395 Angstrom.
 Bond distances in the ring are expected to vary only slightly when due to  
 substituent effects (exceptions include =O substituents). Large deviations 
 may indicate data or model errors. 

 Check this bond distance.  

 Cyclohexane moieties should have be significantly puckered as measured by  
 the average torsion angle tau. Unresolved disorder generally results in
 flattened rings and elongated displacement ellipsoids. A disorder model
 should be included in the calculations.

 The average s.u. for X-Y bonds is tested (named bond-precision). X-Y will  
 generally be C-C bonds, unless there are none. In the last case the s.u.'s of  
 the lowest element numbers are considered (excluding hydrogen). There are  
 three test ranges: one for structures with the largest element Z < 20, one 
 for the largest Z in the range 20 to 39 and one for structures with Z(max) 40  
 or higher (_340, _341 and _342 respectively).  

 The average s.u. for X-Y bonds is tested (named bond-precision). X-Y will  
 generally be C-C bonds, unless there are none. In the last case the s.u.'s of  
 the lowest element numbers are considered (excluding hydrogen). There are  
 three test ranges: one for structures with the largest element Z < 20, one 
 for the largest Z in the range 20 to 39 and one for structures with Z(max) 40  
 or higher (_340, _341 and _342 respectively).  

 The average s.u. for X-Y bonds is tested (named bond-precision). X-Y will  
 generally be C-C bonds, unless there are none. In the last case the s.u.'s of  
 the lowest element numbers are considered (excluding hydrogen). There are  
 three test ranges: one for structures with the largest element Z < 20, one 
 for the largest Z in the range 20 to 39 and one for structures with Z(max) 40  
 or higher (_340, _341 and _342 respectively).  

 The angle range is larger than usual for the tentatively assigned  
 hybridisation of the reported atom in the main residue.

 The angle range is larger than usual for the tentatively assigned  
 hybridisation of the reported atom in the solven/anion.

 Large deviations from generally accepted values may indicate model problems,   
 over refinement etc.  Default C-H = 0.96 Ang. (X-Ray) value from SHELXL.   

 Large deviations from generally accepted values may indicate model problems,   
 over refinement etc.  Default C-H = 0.96 Ang. (X-Ray) value from SHELXL.   

 Large deviations from generally accepted values may indicate model problems,   
 over refinement etc.  Default N-H = 0.87 Ang. (X-Ray) value from SHELXL.   

 Large deviations from generally accepted values may indicate model problems,   
 over refinement etc.  Default N-H = 0.87 Ang. (X-Ray) value from SHELXL.   

 Large deviations from generally accepted values may indicate model problems,   
 over refinement etc.  Default O-H = 0.82 Ang. (X-Ray) value from SHELXL.   

 Large deviations from generally accepted values may indicate model problems,   
 over refinement etc.  Default O-H = 0.82 Ang. (X-Ray) value from SHELXL.   

 A short B-H distance may be due to fixing that distance to a wrong value,  
 e.g. to that of the C-H distance in a CH3 group.   

 Check the long reported B-H bond distance (B-H is expected to have a value 
 around 1.16 Ang in a X-BH3 group)

 A short B-H distance may be due to fixing that distance to a wrong value.,  
 e.g. to that of the C-H distance.   

 Check the long reported B-H bond distance (B-H is expected to have a value 
 around 1.06 Ang in a (X,Y,Z)-B-H moiety).

 Large deviations from generally accepted values may indicate model problems,   
 over refinement etc. Default C4-C4 = 1.54 Ang. (X-Ray) value from  
 Ladd & Palmer, Structure Determination by Xray Crystallography (1985). 
 Note:  
  - C4-C4 indicates a bond between atoms with 4 bonds each. 

 Large deviations from generally accepted values may indicate model problems,   
 over refinement etc. Default C4-C4 = 1.54 Ang. (X-Ray) value from  
 Ladd & Palmer, Structure Determination by Xray Crystallography (1985). 
 Note:  
  - C4-C4 indicates a bond between atoms with 4 bonds each. 

 Large deviations from generally accepted values may indicate model problems,   
 over refinement etc. Default C4-C3 = 1.52 Ang. (X-Ray) value from  
 Ladd & Palmer, Structure Determination by Xray Crystallography (1985). 
 Note:  
  - C4-C3 indicates a bond between an atom with 4 bonds and one with 3 bonds.   
  - Conjugated systems may cause some 'false alarm' messages.   

 Large deviations from generally accepted values may indicate model problems,   
 over refinement etc. Default C4-C3 = 1.52 Ang. (X-Ray) value from  
 Ladd & Palmer, Structure Determination by Xray Crystallography (1985). 
 Note:  
  - C4-C3 indicates a bond between an atom with 4 bonds and one with 3 bonds.   
  - Conjugated systems may cause some 'false alarm' messages.   

 Large deviations from generally accepted values may indicate model problems,   
 over refinement etc. Default C4-C2 = 1.46 Ang. (X-Ray) value from  
 Ladd & Palmer, Structure Determination by Xray Crystallography (1985). 
 Note:  
  - C4-C2 indicates a bond between an atom with 4 bonds and one with 2 bonds.   
  - Conjugated systems may cause some 'false alarm' messages.   

 Large deviations from generally accepted values may indicate model problems,   
 over refinement etc. Default C4-C2 = 1.46 Ang. (X-Ray) value from  
 Ladd & Palmer, Structure Determination by Xray Crystallography (1985). 
 Note:  
  - C4-C2 indicates a bond between an atom with 4 bonds and one with 2 bonds.   
  - Conjugated systems may cause some 'false alarm' messages.   

 The hybridisation of at least one carbon atom is not recognized.   
 Large deviations from generally accepted values for a C-C bond may indicate
 model problems, unresolved disorder, over-refinement etc. The C-C bond is  
 tested to be not shorter than 1.15 Angstrom.   

 Large deviations from generally accepted values may indicate model problems,   
 over refinement etc. Default C?-C? = 1.50 Ang. (X-Ray) value from  
 Ladd & Palmer, Structure Determination by Xray Crystallography (1985). 

 Large deviations from generally accepted values may indicate model problems,   
 over refinement etc. Default C3-C3 = 1.34 Ang. (X-Ray) value from  
 Ladd & Palmer, Structure Determination by Xray Crystallography (1985). 
 Note:  
  - C3-C3 indicates a bond between atoms with 3 bonds each. 
  - Conjugated systems may cause some 'false alarm' messages.   

 Large deviations from generally accepted values may indicate model problems,   
 over refinement etc.  Default C3-C3 = 1.34 Ang. (X-Ray) value from 
 Ladd & Palmer, Structure Determination by Xray Crystallography (1985). 
 Note:  
  - C3-C3 indicates a bond between atoms with 4 with 3 bonds each.  
  - Conjugated systems may cause some 'false alarm' messages.   
  - A notable exception is the C-C bond in -C(=O)-C(=O)- systems with an
    observed mean value of 1.54 Angstrom.   

 Large deviations from generally accepted values may indicate model problems,   
 over refinement etc. Default C3-C2 = 1.31 Ang. (X-Ray) value from  
 Ladd & Palmer, Structure Determination by Xray Crystallography (1985). 
 Note:  
  - C3-C2 indicates a bond between an atom with 3 bonds and one with 2 bonds.   
  - Conjugated systems may cause some 'false alarm' messages.   

 Large deviations from generally accepted values may indicate model problems,   
 over refinement etc. Default C3-C2 = 1.31 Ang. (X-Ray) value from  
 Ladd & Palmer, Structure Determination by Xray Crystallography (1985). 
 Note:  
  - C3-C2 indicates a bond between an atom with 3 bonds and one with 2 bonds.   
  - Conjugated systems may cause some 'false alarm' messages.   

 Large deviations from generally accepted values may indicate model problems,   
 over refinement etc. Default C2-C2 = 1.25 Ang. (X-Ray) value from  
 Ladd & Palmer, Structure Determination by Xray Crystallography (1985). 
 Note:  
  - C2-C2 indicates a bond between atoms with 2 bonds each. 
  - Conjugated systems may cause some 'false alarm' messages.   

 Large deviations from generally accepted values may indicate model problems,   
 over refinement etc. Default C2-C2 = 1.25 Ang. (X-Ray) value from  
 Ladd & Palmer, Structure Determination by Xray Crystallography (1985). 
 Note:  
  - C2-C2 indicates a bond between atoms with 2 bonds each. 
  - Conjugated systems may cause some 'false alarm' messages.   

 Large deviations from generally observed bond distances may indicate model 
 problems, over-refinement etc. Check for wrong atom-type assignments.  
 For an example see: Acta Cryst. (2003) E59, m710-m712. 

 Strange C-O-H geometry with C-O  >   1.45 Angstrom detected.   
 Wrong Atom Type ?  

 This test alerts for possible incorrectly oriented CH3 moieties.   
 (E.g. AFIX 33 instead of AFIX 137 etc. within the SHELXL realm).

 Unusual Methyl Moiety X-C-H Angle (Ideally 109 Degrees for 4-bonded C).

 Unusual Methyl Moiety H-C-H Angle (ideally 109 Degrees).   

 The X-O-Y angle value differs from the common ~120.0 Degrees.  

 The Si-O-Si angle value differs from the common ~150.0 Degrees.

 The B-O-B angle differs significantly from 120.0 degrees.  
 This is just a notice: B-O-B angles vary widely depending on what is bonded
 to the Boron atoms.

 The C-O-C angle value differs from the common ~120.0 Degrees.  

 Short intramolecular contacts may arise when H-atoms are in (false)
 calculated positions. Short intramolecular contacts may also be a sign for a   
 false structure with the molecule sitting on a site with improper site 
 symmetry (e.g. '2' instead of '-1') which may happen when a lattice
 translation is missed. Short contacts are defined using a van der Waals
 radius of 1.2 Angstrom. For intramolecular contacts alerts are generated   
 for contacts less than 2.0 Angstrom.   

 Short intermolecular H..H contacts may indicate incorrectly determined 
 structures (i.e. wrong symmetry, missed translation symmetry, wrong position   
 with reference to the symmetry elements, hydrogen atoms on atoms where there   
 should not be any  etc..)  
 Short intermolecular contacts may be indicative for inconsistent symmetry  
 data (e.g. coordinates for space-group P43 and symmetry specified as P41 or
 P21/n & P21/c confusions). Short contacts are defined using a van der Waals
 radius of 1.2 Angstrom. For intermolecular contacts, an alert is generated 
 for contacts less than 2.4 Angstrom.   

 Short intramolecular contacts may arise when H-atoms are in (false)
 calculated positions. Short intramolecular contacts may also be a sign for a   
 false structure with the molecule sitting on a site with improper site 
 symmetry (e.g. '2' instead of '-1') which may happen when a lattice
 translation is missed. Short contacts are defined using a van der Waals
 radius of 1.2 Angstrom. Short H .. H contact involving CH3 H-atoms are often   
 hampered by the fact that they involve H atoms at not optimal calculated   
 positions. 

 Short intermolecular H..H contacts may indicate incorrectly determined 
 structures (i.e. wrong symmetry, missed translation symmetry, wrong position   
 with reference to the symmetry elements, hydrogen atoms on atoms where there   
 should not be any  etc..). Short intermolecular contacts may be indicative 
 for inconsistent symmetry data (e.g. coordinates for space-group P43 and   
 symmetry specified as P41 or P21/n & P21/c confusions). Short contacts are 
 defined using a van der Waals radius of 1.2 Angstrom. Short H .. H contact 
 involving CH3 H-atoms are often hampered by the fact that they involve 
 H atoms at not optimal calculated positions.   

 Short intra D-H..H-X contact. This might be a di-hydrogen bond of the type 
 B-H..H-O.  See Crabtree et al. Acc. Chem. Res. 1996, 29, 348-354.  

 Short inter D-H..H-X contact. This might be a di-hydrogen bond of the type 
 B-H..H-O. See: Crabtree et al. Acc. Chem. Res. 1996, 29, 348-354.  

 Short non-bonding intra D-H..H-D contacts may be related to disordered or  
 misplaced H-atoms. 

 Short non-bonding inter D-H..H-D contacts may be related to disordered or  
 misplaced H-atoms. Experience has shown that any intermolecular H...H 
 separation of less than 1.8 Angstroms between unit-occupancy H atoms is a
 clear indicator that one or both of these H atoms may be wrongly placed.

 Potential hydrogen bond donors are checked for the presence of suitable
 acceptors using commonly used (Jeffrey) H-bond criteria. As a general rule 
 there should be an acceptor for each donor. Exceptions are very rare for O-H   
 and more common for -NH and -NH2. A common error is an -OH with Hydrogen atom
 on a calculated position with O-H pointing in the wrong direction.  

 This test alerts for possibly missed Hydrogen bonds as indicated by short  
 (i.e. shorter than sum of the van der Waals radii - 0.2 Angstrom)  
 Donor - Acceptor atom distances.
 Note: Short C=O .. O=C are observed sometimes when part of three-centre
 O-H, N-H or C-H O..O bridging. 

 This test reports on short intermolecular Halogen .. Donor/Acceptor
 atom-type distances.   

 This test raised an ALERT for short intermolecular contacts.   
 In general, intermolecular contact distances should be not much smaller than   
 the sum of the associated van der Waals Radii. More often than not, such   
 short contacts can be a warning sign for errors. All short contacts should 
 therefore be examined in some detail. Interesting exceptions are carbonyl- 
 carbonyl interactions that often feature short O...C contacts (see Allen   
 et al. (1998) B54, 320-329, short NO2 O...O interactions and BF4(-) to 
 (aromatic) carbon contacts.

 This test raised an ALERT for short intermolecular contacts between minor  
 disorder components. In general, intermolecular contact distances should be
 not much smaller than the sum of the associated van der Waals Radii. More  
 often than not, such short contacts can be a warning sign for errors. All  
 short contacts should therefore be examined in some detail. Interesting   
 exceptions are carbonyl-carbonyl interactions that often feature short O...C   
 contacts (see Allen et al. (1998) B54, 320-329).

 This test reports on short intermolecular Halogen .. Halogen type distances.   
 Note: Bridging I- ions in a X-I...I-...I-Y chain may invoke level B ALERTs. 

 Check this (unrealistically) long reported H..A contact.   
 Jeffrey criterium: Contact  <   vdWR(H) + vdWR(A) - 0.12 Angstrom. 

 Check this (unrealistically) long reported D..A contact.   
 Jeffrey criterium: Contact  <   vdWR(D) + vdWR(A) + 0.50 Angstrom. 

 Check this unrealistically small reported D-H..A Angle.
 Jeffrey criterium: D-H..A Angle > 100 degrees. 

 Crystal structures in general do not contain large solvent accessible voids
 in the lattice. Most structures lose their long-range ordering when solvent
 molecules leave the crystal. Only when the remaining network is strongly   
 bonded (e.g. zeolites and some hydrogen bonded networks) the crystal   
 structure may survive. Residual voids in a structure may indicate the
 omission of (disordered) density from the model. Disordered density may go
 undetected when smeared since peak search programs are not designed to locate
 maxima on density ridges. The presence or absence of residual density in the
 void may be verified on a printed/plotted difference Fourier map or with
 PLATON/SQUEEZE. Voids of 40 Ang**3 may accommodate H2O. Small molecules such
 as Tetrahydrofuran have typical volumes in the 100 to 200 Ang**3 range. This
 test reports the volume of the largest solvent accessible void in the 
 structure. A paper reporting a crystal structure with a significant solvent 
 accessible void should at the least discuss the issue.

 This test reports on a solvent accessible void in the structure, too large 
 or too time consuming for the current PLATON version for a more detailed   
 analysis as part of the validation run. Use the SOLV option for more details.  
 Such a warning might also indicate that the symmetry is incomplete 
 e.g. should have been specified as P-1 and not P1, leaving out half of 
 the unit cell content. 

 No search for solvent accessible VOIDS done as part of VALIDATION in   
 view of large unit-cell.   

 Too many solvent accessible VOIDS. 

 Crystal structures in general do not contain large solvent accessible voids
 in the lattice. Most structures lose their long-range ordering when solvent
 molecules leave the crystal. Only when the remaining network is strongly   
 bonded (e.g. zeolites and some hydrogen bonded networks) the crystal   
 structure may survive. 
 Residual voids in a structure may indicate the omission of (disordered)
 density from the model. Disordered density may go undetected when smeared  
 since peak search programs are not designed to locate maxima on density
 ridges. The presence or absence of residual density in the void may be 
 verified on a printed/plotted difference Fourier map or with PLATON/SQUEEZE.   
 Voids of 40 Ang**3 may accommodate H2O. Small molecules such as
 Tetrahydrofuran have typical volumes in the 100 to 200 Ang**3 range.   
 This test reports the volume of the largest solvent accessible void in the 
 structure. 
 A paper reporting a crystal structure with a significant solvent accessible
 void should at the least discuss the issue.
 Note: The use of PLATON/SQUEEZE was reported in the CIF.

 This test reports on a solvent accessible void in the structure, too large 
 or too time consuming for the current PLATON version for a more detailed   
 analysis as part of the validation run. Use the SOLV option for more details.  
 Such a warning might also indicate that the symmetry is incomplete 
 e.g. should have been specified as P-1 and not P1, leaving out half of 
 the unit cell content. 

 Void test Skipped. 

 The ADDSYM test is skipped as part of the checkCIF run when the symmetry   
 generated cell content gets too large and the calculation too time consiming.  
 The calculation can still be done via the ADDSYM tool in PLATON.   

 The presence of an ABIN instruction in the SHELXL .ins file implies the use 
 of the PLATON/SQUEEZE or OLEX2/MASK tool to take into account the disordered
 solvent contribution to the calculated structure factors. The corresponding
 procedure details, generated by those programs, appear to be absent from or
 incorrectly modified as part of the CIF embedded .fab file. 

 This test reports the use of the SHELXL/SWAT instruction for the modelling 
 of the disordered solvent contribution to the calculated structure factors.

 Examples of expected values are 'MoKa', 'synchrotron', 'neutron', 'electron'   

 A number of blank records interspersed in the CIF embedded hkl file have been
 removed.

 The supplied CIF contains a '_shelx_include_file' record but not a valid   
 associated '_shelx_include_file_checksum' record or valid value..  
 This include file is called from .ins and is needed for the refinement  
 and re-creation of the associated .fcf file (used for a detailed analysis of   
 the refinement result). The calculated and reported checksums should be
 identical. Only characters with an ASCII value higher than 32 contribute to
 the checksum.  An embedded include file might be broken, either due to 
 transfer errors or to deliberate or accidental post-refinement editing of its   
 content.   

 Examples of expected values are 'block', 'needle', 'plate', 'sphere',  
 'cylinder'. In case of 'cylinder' or 'sphere' also provide a value for 
 '_exptl_crystal_size_rad'. 

 Bond distances given in the CIF are cross-checked with corresponding   
 values calculated from the coordinates.  Alerts are set at 1,2 and 3 sigma 
 deviation levels.  
 Note: Default s.u.'s are used where no s.u. given (e.g. for C-H)   
 In general, all differences should be within the associated s.u.   
 Small differences may arise from rounding. 
 Very large deviation (or zero distance) normally indicate incorrectly  
 specified symmetry operations on the associated atoms, or 'cut-and-pasting'
 of incompatible CIF's. 

 Bond Angles given in the CIF are cross-checked with corresponding values   
 calculated from the coordinates.  Alerts are set at 1,2 and 3 sigma
 deviation levels.  
 In general, all differences should be within the associated s.u.   
 Small differences may arise from rounding. Very large deviations normally  
 indicate incorrectly specified symmetry operations on the associated atoms,
 or 'cut-and-pasting' of incompatible CIF's.

 Torsion angles given in the CIF are cross-checked with corresponding values
 calculated from the coordinates.  Alerts are set at 1,2 and 3 sigma
 deviation levels.  
 In general, all differences should be within the associated s.u.   
 Small differences may arise from rounding. Very large deviations normally  
 indicate incorrectly specified symmetry operations on the associated atoms,
 or 'cut-and-pasting' of incompatible CIF's.

 Intermolecular contacts listed in the CIF are checked against the  
 coordinates in the CIF.
 Alerts are set at 1,2 and 3 sigma deviation levels.

 Hydrogen-Bond D-H listed in the CIF is checked.  Alerts are set at 
 1,2 and 3 sigma deviation levels.  

 Hydrogen-Bond H..A listed in the CIF is checked. Alerts are set at 1,2 and 
 3 sigma deviation levels.  
 This ALERT is generally related to incorrect symmetry codes. The symmetry  
 number s in the symmetry code s_pqr should correspond to the expression
 for s in the CIF. Those expressions can be different for different 
 software packages. E.g. pasting H-bond table data generated with PLATON
 into a CIF generated with SHELXL may raise this ALERT. Manual correction   
 of the symmetry code should be trivial.

 Hydrogen-Bond D..A listed in the CIF is checked. Alerts are set at 1,2 and 
 3 sigma deviation levels.  
 This ALERT is generally related to incorrect symmetry codes. The symmetry  
 number s in the symmetry code s_pqr should correspond to the expression
 for s in the CIF. Those expressions can be different for different 
 software packages. E.g. pasting H-bond table data generated with PLATON
 into a CIF generated with SHELXL may raise this ALERT. Manual correction   
 of the symmetry code should be trivial.

 Hydrogen-Bond Angle D-H..A listed in the CIF is checked. Alerts are set at 
 1,2 and 3 sigma deviation levels.  
 This ALERT is generally related to incorrect symmetry codes. The symmetry  
 number s in the symmetry code s_pqr should correspond to the expression
 for s in the CIF. Those expressions can be different for different 
 software packages. E.g. pasting H-bond table data generated with PLATON
 into a CIF generated with SHELXL may raise this ALERT. Manual correction   
 of the symmetry code should be trivial.

 Torsion angles specified in the CIF are checked for the
 'linear variety' where one or both of the 1-2-3 and 2-3-4 bond 
 angles are close to 180 Deg. SHELXL97 will generate those 'torsions' for   
 molecules containing linear moieties (E.g. Metal-C=O). 

 When labels are found on geometry items (bonds, angles etc.) that are not in   
 the coordinate list, and alert _71n is issued, related to alert _70n.  

 When labels are found on geometry items (bonds, angles etc.) that are not in   
 the coordinate list, and alert _71n is issued, related to alert _70n.  

 When labels are found on geometry items (bonds, angles etc.) that are not in   
 the coordinate list, and alert _71n is issued, related to alert _70n.  

 When labels are found on geometry items (bonds, angles etc.) that are not in   
 the coordinate list, and alert _71n is issued, related to alert _70n.  

 When labels are found on geometry items (bonds, angles etc.) that are not in   
 the coordinate list, and alert _71n is issued, related to alert _70n.  

 When labels are found on geometry items (bonds, angles etc.) that are not in   
 the coordinate list, and alert _71n is issued, related to alert _70n.  

 When labels are found on geometry items (bonds, angles etc.) that are not in   
 the coordinate list, and alert _71n is issued, related to alert _70n.  

 When labels are found on geometry items (bonds, angles etc.) that are not in   
 the coordinate list, and alert _71n is issued, related to alert _70n.  

 Up to 4 Character Labels of the type C11, H101, N10A, i.e. chemical symbol +   
 number + optional letter are to be preferred.  
 Note: SHELXL based refinements may use labels up to 7 characters (Including
 residue and or disorder information).

 Same as 701 but for distance without s.u. (esd). Difference is tested in
 terms of Angstroms.  

 Same as 702 but for angle without s.u. (esd). Difference is tested in terms
 of Degrees.   

 Same as 703 but for torsion without s.u. (esd). Difference is tested in terms  
 of Degrees.

 Same as 704, but for distance without s.u. (esd). Difference is tested in  
 terms of Angstroms.

 Same as 705 but for distance without s.u. (esd). Differences are tested in 
 terms of Angstrom. 

 Same as 706 but for distance without s.u. (esd). Differences are tested in 
 terms of Angstrom. 

 Same as 707 but for distance without s.u. (esd). Differences are tested in 
 terms of Angstrom. 

 Same as ALERT 708 but for angle without s.u. (esd). Differences are tested 
 in terms of Degrees.   

 A large ratio of the reported and calculated bond s.u.'s is found. 
 The use of a DFIX instruction might cause such a warning since 
 calculated s.u.'s are based on reported variances only.
 Note_1: s.u.'s on the unit-cell dimensions are taken into account in the   
 calculation of expected s.u.'s. This may result in large differences   
 between expected and reported s.u.'s when this contribution is not included
 in the reported s.u.'s, in particular for inaccurate unit-cells.   
 Note_2: Another source for the discrepancy between calculated and  
 reported s.u.'s can be that the validation software has access only to 
 the variances of the refined parameters as opposed to the full co-variance 
 matrix used by e.g. SHELXL for the calculation of derived parameters   
 with associated s.u.'s. Constrained/restrained refinement may cause large  
 co-variances.  

 A large ratio of the reported and calculated bond angle s.u.'s is found.   
 This check should warn for erroneous rounding: E.g. 105.5(19) to 105.5(2) or   
 105.0(5) to 105(5) etc. Note: Large differences are possible when certain  
 constraints/restraints were applied in the refinement (e.g. the FLAT option
 in SHELXL97).  
 Note_1: s.u.'s on the unit-cell dimensions are taken into account in the   
 calculation of expected s.u.'s. This may result in large differences   
 between expected and reported s.u.'s when this contribution is not included
 in the reported s.u.'s, in particular for inaccurate unit-cells.   
 Note_2: Another source for the discrepancy between calculated and  
 reported s.u.'s can be that the validation software has access only to 
 the variances of the refined parameters as opposed to the full co-variance 
 matrix used by e.g. SHELXL for the calculation of derived parameters   
 with associated s.u.'s. Constrained/restrained refinement may cause large  
 co-variances.  

 A large ratio of the reported and calculated torsion angle s.u.'s is found.
 This check should warn for erroneous rounding: E.g. 105.5(19) to 105.5(2)  
 or 105.0(5) to 105(5) etc. 
 Note_1: s.u.'s on the unit-cell dimensions are taken into account in the   
 calculation of expected s.u.'s. This may result in large differences   
 between expected and reported s.u.'s when this contribution is not included
 in the reported s.u.'s, in particular for inaccurate unit-cells.   
 Note_2: Another source for the discrepancy between calculated and  
 reported s.u.'s can be that the validation software has access only to 
 the variances of the refined parameters as opposed to the full co-variance 
 matrix used by e.g. SHELXL for the calculation of derived parameters   
 with associated s.u.'s. Constrained/restrained refinement may cause large  
 co-variances.  

 A large ratio of the reported and calculated contact distance s.u.'s is
 found. 
 Note: s.u.'s on the unit-cell dimensions are taken into account in the 
 calculation of expected s.u.'s. This may result in large differences   
 between expected and reported s.u.'s when this contribution is not included
 in the reported s.u.'s, in particular for inaccurate unit-cells.   

 A large ratio of the reported and calculated H-bond D-H distance s.u.'s is 
 found. The use of a DFIX instruction might cause such a warning since  
 calculated s.u.'s are based on reported variances only.
 Note: s.u.'s on the unit-cell dimensions are taken into account in the 
 calculation of expected s.u.'s. This may result in large differences   
 between expected and reported s.u.'s when this contribution is not included
 in the reported s.u.'s, in particular for inaccurate unit-cells.   

 A large ratio of the reported and calculated H-bond H..A distance s.u.'s   
 is found.  
 Note: s.u.'s on the unit-cell dimensions are taken into account in the 
 calculation of expected s.u.'s. This may result in large differences   
 between expected and reported s.u.'s when this contribution is not included
 in the reported s.u.'s, in particular for inaccurate unit-cells.   

 A large ratio of the reported and calculated H-Bond D...A distance s.u.'s is   
 found. 

 A large ratio of the reported and calculated H-Bond D-H..A angle s.u.'s
 is found.  
 Note: s.u.'s on the unit-cell dimensions are taken into account in the 
 calculation of expected s.u.'s. This may result in large differences   
 between expected and reported s.u.'s when this contribution is not included
 in the reported s.u.'s, in particular for inaccurate unit-cells.   

 Likely Missing s.u. on Bond in CIF.

 Likely Missing s.u. on Bond angle in CIF.  

 Likely Missing s.u. on Torsion angle in CIF.   

 Likely missing s.u. on contact Distance in CIF.

 Likely missing s.u. on H-Bond D-H distance in CIF. 

 Likely missing s.u. on H-Bond H...A distance in CIF.   

 Likely missing s.u. on H-Bond D...A distance in CIF.   

 Likely missing s.u. on H-Bond D-H..A angle in CIF. 

 An s.u. should not be given in the CIF for constrained distances.  
 Please check for proper refinement status flags (e.g. R)   

 An s.u. should not be given in the CIF for constrained angles. 
 Please check for proper refinement status flags (e.g. R)   

 An s.u. should not be given in the CIF for constrained torsion angles. 
 Please check for proper refinement status flags (e.g. R)   

 An s.u. should not be given in the CIF for constrained contact distances.  
 Please check for proper refinement status flags (e.g. R)   

 An s.u. should not be given in the CIF for constrained distances.  
 Please check for proper refinement status flags (e.g. R).  

 An s.u. should not be given in the CIF for constrained distances.  
 Please check for proper refinement status flags (e.g. R).  

 An s.u. should not be given in the CIF for constrained distances.  
 Please check for proper refinement status flags (e.g. R).  

 An s.u. should not be given in the CIF for constrained angles. 
 Please check for proper refinement status flags (e.g. R)   

 The CIF contains no X-H Bonds. This might be caused by not using the SHELXL  
 instruction BOND $H. Inclusion is required by Acta Cryst. but not necessarily
 so by other journals.  

 The CIF contains no X-Y-H or H-Y-H bond angles. This might be caused by not
 using the SHELXL instruction BOND $H. Those data should also be supplied   
 when H-atoms are introduced on calculated positions and/or refined riding  
 on their carrier atom. Inclusion is required by Acta Cryst. but not
 necessarily so by other journals. 

 Bond list in CIF likely incomplete.

 The CIF contains more bonds than the unique set, indicating redundancy.
 An example is redundancy due to the inclusion of symmetry related bonds.   

 
 The final .fcf associated with the .cif should have been created
 with the LIST 4 instruction (or just left out because already implicit with
 the ACTA instruction). The LIST 8 instruction will create an FCF file where 
 the observed data are detwinned making the generated .fcf unsuitable 
 for checking for missed (additional) twinning and validation.

 
 The '.fcf' associated with the '.cif' should have been created with a
 'LIST 4' instruction (or left out because already implicit with the ACTA
 instruction). LIST 6 will create an FCF file where the observed data are
 corrected for anomalous/resonant scattering, making the generated '.fcf'
 unsuitable for post refinement absolute structure analysis and validation.  

 The supplied CIF includes explicit atomic scattering factor date in the 
 embedded .res file. Those values may or may not be identical to the standard
 values as included in current versions of SHELXL. It should be noted that
 those values will not be present as CIF data items in CIF's created with
 SHELXL. PLATON/checkCIF will use its own build-in scattering factors, that
 are identical to those used by default by SHELXL for common wavelengths,
 for its validation calculations. 

 The supplied CIF contains explicit scattering factor data as values of their
 corresponding CIF datanames. Those scattering factor data are used by 
 PLATON/checkCIF for its validation calculations as opposed to using the
 neutral scattering factor data that are stored in PLATON (which are identical 
 to those in stored in SHELXL). Those CIF reported values should be identical
 to those provided in the embedded *.ins file. 
 Current SHELXL versions do not copy scattering factor data that are
 explicitly included in the *.ins instruction file into corresponding CIF data
 items. A reason for the use of non-standard scattering factors can be the
 modelling of charged atoms in the refinement.

 Report on unusual C-H bonds not caught in other tests. 

 Report on unusual N-H bonds not caught in other tests. 

 Report on unusual O-H bonds not caught in other tests. 
 Note: Exceptions can be H-atoms in acid O..H..O bridges or in H5O2+
       (Hydronium) species. 

 Report on unusual C-C bonds, possibly  not caught in other tests.  
 Exceptions include C-C distances of around 1.75 Ang. in e.g.   
 1,2-dicarba-closo-dodecaborane.

 Likely Erroneous Bond Entry.   

 Likely Erroneous Contact Entry.

 Likely Erroneous D-H Entry.

 Report N-H distance in special N..H..X+ bonds. The N .. X distance is  
 expected to be shorter than in a standard hydrogen bond.   

 Report O-H distance in special O..H..X+ bonds. The O .. X distance is  
 expected to be shorter than in a standard hydrogen bond.   

 Possibly erroneous (Bond)angle less than 45 degree. The angle might be 
 considered for elimination from the CIF when irrelevant. This ALERT can
 also be triggered when the assigned occupancy factors are incorrect.   

 Atoms given in a CIF should form a 'connected set', i.e. no symmetry   
 operations are needed to get atoms in a bonding position. A connected set of   
 atoms is not needed for the least squares refinement (unless hydrogen atoms
 are to be added at calculated positions). Geometry listings (bonds, angles,
 torsions & H-bonds) become unwieldy for non-connected atom sets.   

 A Flack parameter value is erroneously given for a structure reported  
 in a centrosymmetric space-group.  

 Warning for a possible misassignment of C-NO2 and C-CO2 moieties.  
 The geometry of the reported moiety appears to be unusual/inconsistent.
 The C-O bond distances in C-CO2 are expected to add up to about 2.5
 Test Criteria: ALERT for C-O bond sum < 2.48 and C-C < 1.48 Angstrom   
 The N-O bond distances in C-NO2 are expected to add up to about 2.4
 Test Criteria: ALERT for N-O bond sum > 2.48 and C-C > 1.48 Angstrom   

 A negative value of the _atom_site_disorder_group number of an atom indicates  
 whether that atom is part of a group disordered over a symmetry element.   
 Check whether this negative sign assignment applies.   

 Unless for a good reason, molecular species should be transformed  
 (by symmetry and/or translation) so that their centres of gravity  
 are close to or within the unit-cell bounds. This is a strict rule for the 
 main species. Deviations from this general rule are for smaller additional 
 species when relevant for intermolecular interactions with the main species.   

 A tentative (R/S) assignment is done on the basis of the geometrical and   
 atom type data in the CIF. A correct assignment depends on the algorithm   
 used to automatically detect double and triple bonds. For that reason it is
 important that the molecule is complete (i.e. all hydrogen atoms should be 
 included). In case of a chiral molecule in a centro-symmetric spacegroup, it
 is essential that the coordinate set used and the associated molecular
 (ORTEP) illustration are consistent. This can be verified by checking the  
 identical signs of relevant torsion angles. The absolute structure assignment
 should also be consistent with the lowest value of the Flack parameter and/or
 known absolute configuration.   

 This test addresses the consistency of the absolute structure assignment   
 (i.e. polarity etc.) in non-centrosymmetric structures in space groups 
 that include improper symmetry operations (e.g. mirror planes).
 Check the (R/S) absolute configuration assignment of this atom and 
 the consistency of the absolute configuration implicit in the CIF-data 
 with that in the 'ORTEP' illustration. 

 This test addresses the consistency of the absolute configuration assignment   
 of molecules in the reported asymmetric unit among coordinates,
 molecular presentations and chemical diagrams. 
 Check the (R/S) absolute configuration assignment of this atom and 
 the consistency of the absolute configuration implicit in the CIF-data 
 with that in the 'ORTEP' illustration. 

 This test reports the valency of an atom as predicted by the Valence   
 Bond Model. See:   
 N.E. Brese & M. O'Keeffe (1991) Acta Cryst. B47, 192-197.  
 I.D. Brown (2002). The Chemical Bond in Inorganic Chemistry:   
      The Bond Valence Model. Oxford University Press.  
 More explicit info on the calculations can be obtained by running the  
 calculations explicitly with the PLATON option BondValence.
 Note: The underlying theory is empirical and might not apply to the
       case at hand (e.g. charged species). 

 Atom labels are generally not a number (i.e. starting with one or two  
 characters indicating the atom type). Labels can be erroneously numeric
 due to typing errors (e.g. 'Oxygen' typed as 'zero').  

 Atom labels are generally not a number (i.e. starting with one or two  
 characters indicating the atom type). Labels can be erroneously numeric
 due to typing errors (e.g. 'Oxygen' typed as 'zero').  

 PLATON/CheckCIF has a problem with the Cell data. A possible reason can be 
 that the cell data are missing, incomplete or out-of-sequence. 
 PLATON/CheckCIF wishes to encounter the cell and symmetry data before any  
 coordinates are given. PLATON expects the values of all six cell parameters.   

 The CIF contains records longer than 80 characters. Not all software will  
 read beyond column 80. The CIF-1.1 definition specifies a maximum of 2048
 character per record.   

 Fatal Problem: Check loop data names and data for errors. There are likely
 too many or to few data in the loop.

 Problem: ARU representations turn out to be needed outside the ORTEP   
 style -5:5 unit-cell translation range. The Analysis might be incomplete.  
 The problem often occurs for structures with aliphatic chains stretching   
 over many unitcells or network structures. Transformation of the unitcell  
 content to a symmetry related position might solve the problem.

 Check Coordinate Data Loop.

 Check UIJ Data Loop.   

 PLATON can handle up to 'NP1' in the (expanded) ATOM list. This might happen
 with disordered or network structures in high symmetry space groups. Deletion
 of the symmetry information might solve part of the problem and provide a
 partial validation. Alternatively, clicking on 'NOSYMM' on the PLATON menu
 before invoking validation might address the problem.  

 The validation software did not succeed in finding/analyzing a parsable
 weighting scheme.  SHELXL style weight parameters are expected to be given in
 the format: 
 _refine_ls_weighting_details   
 'calc w=1/[\s^2^(Fo^2^)+(0.1000P)^2^+0.0000P] where P=(Fo^2^+2Fc^2^)/3'
 JANA style weight is expected to be given in the format:   
 _refine_ls_weighting_details 'w=1/(\s^2^(I)+0.0016I^2^)'   
  Do not edit this string or make it into a text block between ';'. 

 The software did not succeed in Parsing the SHELXL style weighting scheme.
 The string might have been edited or of the more than 2 parameter variety
 (see SHELXL manual).   

 Analysis for missing reflections may be incomplete due to an out-of-memory 
 problem.   

 The ADDSYM test for missed symmetry is not executed for structures with too
 many disordered atoms. 

 Non-standard labels are aliased into acceptable labels. The maximum number 
 of aliases is reached. 

 Check the HKLF record in the SHELXL style .res is embedded in the CIF. 
 This record should show either 'HKLF 4' or 'HKLF 5'. The information in this
 record is used to establish that a twinning model was refined.   

 PLATON/CheckCIF can not validate the CIFs associated with (in)commensurate 
 structure reports. 

 PLATON/checkCIF is currently dimensioned for up to 255 population parameters.
 Please contact the author at a.l.spek@uu.nl

 The CIF includes a loop with twin matrices. The embedded .res file however
 points to a HKLF 4 file with likely detwinned intensities. 

 Internal PLATON Problem. Please refer problem to author at a.l.spek@uu.nl  

 Please check the validity of the negative PART numbers in the CIF-embedded 
 SHELXL .res file. Their use may obscure disorder model errors.   

 This G_ALERT can be ignored in the case that the so-called 'on-the-cheap'  
 Flack parameter is reported as determined with SHELXL. 
 Exactly zero values are possible but may also be a software artefact.  
 The following should be checked:   
 Problem #1: Some SHELXL97 versions do not allow negative values of the 
 Flack parameter when determined using the BASF/TWIN instructions.  
 Negative values are set to 0.00001. Refinement may not converge completely.
 Problem #2: Some SHELXL97 versions put meaningless values in the CIF for the   
 Flack parameter when 'TWIN -1 0 0 0 -1 0 0 0 -1 2 / BASF' instructions (i.e.   
 an explicit matrix is specified on the TWIN instruction) are used. 
 Please check the value of BASF (in the list output)  against the Flack 
 parameter in the CIF. 	

 The use of restraints used in the refinement should be explained in
 the write-up of a structure analysis.  
 It is also recommended to include the refinement instructions in the   
 CIF (e.g. the final .res of a SHELXL refinement) as a comment: 
 _iucr_refine_instructions_details  
 ;  
 TITL   
 .. 
 (etc.) 
 ;  
 Note: An exception are restraints for floating origins (e.g. in P21).  

 ALERTS related to the use of Olex2/_smtbx_masks that can not (yet) be  
 accounted for as part of the VALIDATION algorithms have been suppressed.   

 ALERTS related to the use of PLATON/SQUEEZE that can not (yet) be accounted
 for as part of the VALIDATION algorithms have been suppressed. 

 ALERTS related to twinning effects that can not (yet) be accounted 
 for as part of the VALIDATION algorithms have been suppressed. 

 ALERTS related to the use of the Laue technique with a wavelength  
 range have been suppressed.

 ALERTS related to anharmonic refinement have been suppressed. This generally   
 concerns XD and some OLEX2 refinement based CIF's. Internally calculated   
 structure factors are based on the reported harmonic displacement parameters   
 only.  

 The total number of measured reflections is not reported.  

 The merging R-factor for equivalent reflections is missing in the CIF. 
 A value of 0.0 usually indicates that merging has been done outside the
 program that created the CIF. A proper value should be included as reported
 by the external merging program. Note: modern hardware will generally produce  
 redundant reflection data. 

 No value reported for either '_diffrn_reflns_av_unetI/netI' or the   
 older/superseded _diffrn_reflns_av_sigmaI/netI CIF item.   

 Please supply the proper info for '_atom_sites_solution_primary'. Values must 
 be one of the following: difmap, vecmap, heavy, direct, geom, disper, isomor,  
 notdet, dual, iterative, other.

 A second specification of the H-M symbol is found in the CIF. Only the 
 first one encountered is retained. 

 A more recent SHELXL version is now available. Final refinement with the   
 latest version is recommended since (IUCr) checkCIF is run against the latest 
 SHELXL version for the re-creation of the FCF and its validation. 

 Likely cause: Dataset names in the CIF and FCF differ. 
 Note: FCF Validation is Skipped for this Entry.

 Possible causes: wrong dataset, CIF or FCF parameters edited inconsistently
 or cell parameters for transformed cell (e.g. P21/n <-> P21/c).
 Note: FCF Validation is Skipped for this Entry.

 Either no reflections are found in the FCF or the FCF is uninterpretable   
 due to unknown format or editing.  
 Note: FCF Validation is Skipped for this Entry.

 Check the FCF file for F(obs) equal F(calc) [or F(obs)**2 equal F(calc)**2].   
 The FCF is obviously not created as the result of a refinement with regular 
 observed data.

 The number of reflections found in the reflection file is less than the
 number of parameters reported in the CIF.  

 Scale Factors (i.e. K = Mean[Fo**2] / Mean [Fc**2]) for selected groups of 
 reflections, as listed in the Analysis of Variance Section #7 of the FCF   
 validation report, are expected to have a value of about 1.000. Strong 
 deviations should be investigated, acted upon or explained.
 (see also SHELXL manual). A reason can be meaningless (weak) high order
 data where a cut-back of the resolution might be indicated. Incorrect  
 background treatment can lead to numerous negative observed intensities
 for weak reflections (i.e. F(calc) close to zero.  
 For details see the Analysis-of-Variance Section in the '.ckf' file.   

 Scale Factors (i.e. K = Mean[Fo**2] / Mean [Fc**2]) for selected groups of 
 reflections, as listed in the Analysis of Variance Section #7 of the FCF   
 validation report, are expected to have a value of about 1.000. Strong 
 deviations should be investigated, acted upon or explained. (see also SHELXL
 manual). For details see the Analysis-of-Variance Section in the '.ckf' file.   

 A Flack x value generally indicates that the structure needs to be inverted.   
 Check whether the Flack x value is unreliable due to a weak anomalous
 dispersion signal, systematic errors or weak data and against know absolute
 structure. 

 A low maximum percentage of reflections with I > 2*s(I) may indicate:  
 1 - Missed translation symmetry. E.g. all reflections hkl weak 
     for l = 2n +1  
 2 - Pseudo-merohedral twinning, index > 1. (e.g. non-spacegroup
     extinctions.   
 3 - Very weak observed data.   

 This ALERT Reports on whether there is still a significant level of
 observed data beyond the Theta cutoff of the Dataset. There should 
 be a good reason for a cutoff below sin(theta)/lambda = 0.6.   
 Reflection data beyond that value should not be removed when significantly 
 above the noise level. E.g. they may be very relevant in case of pseudo-   
 symmetry and (non)centrosymmetry refinement.   

 Possible causes: Experimental beamstop Theta(min) limit set too high; A large
 unit-cell causing reflections effected by the beamstop. A possible technical
 solution on CCD based equipment involves the collection of additional images
 with the detector at a larger distance from the crystal with the beamstop 
 setting changed accordingly. Alternatively, a large number of low-order
 'outlier' reflections might have been 'Omitted' deliberately from the (final)
 least-squares refinement. The test is against the actual Theta(Min) of the
 reflections in the '.fcf' file. See the *.ckf file for details of the missing
 reflections. Good quality low order reflections might be relevant for SQUEEZE
 and similar solvent modelling refinement techniques.

 Possible causes: Missing cusp of data (due to rotation about one axis),
 deleted (overflow) reflections or improper strategy (orthorhombic for  
 monoclinic crystal etc.) See the '.ckf' file for details.

 Possible causes: Missing cusp of data (due to rotation about one axis),
 deleted (overflow) reflections or improper strategy (orthorhombic for  
 monoclinic crystal etc.) See the .ckf file for details.

 This ALERT reports the number of missing reflections with Fc**2 values 
 greater than the largest Fc**2 value in the FCF. Possible causes: Missing  
 cusp of data (due to rotation about one axis), deleted (overflow)  
 reflections or improper strategy (orthorhombic for monoclinic crystal etc.)
 or behind the beamstop. See the .ckf file for details. 

 This ALERT reflects the notion that a dataset should contain a sufficient  
 number of Bijvoet (Friedel) pairs for the reliable determination of the
 absolute structure of a non-centrosymmetric crystal structure. 
 This test is invoked when a Flack parameter value is specified.
 Note: SHELXL97 will calculate/report a Flack parameter value even for  
 refinement against Friedel merged data. Remove the Flack entry from the CIF.

 This ALERT reflects the notion that a dataset should contain a sufficient  
 number of Bijvoet (Friedel) pairs for the reliable determination of the
 absolute structure of a non-centrosymmetric crystal structure. 
 A Friedel coverage that deviates significantly from 100 percent may
 bias/invalidate the value of the Flack parameter.  

 The Hooft y Parameter is calculated independently from the Bijvoet 
 differences and should have a value similar (observing the s.u.'s) to that of  
 the Flack x Parameter. See: Hooft, R.W.W, Straver, L.H. & Spek,A.L. (2008).
 J. Appl, Cryst. 41, 96-103. Thompson,A.L. & Watkin, D.J. (2009). Tetrahedron:
 Asymmetry, doi:10.1016/j.tetasy.2009.02.025 
 Large differences may arise in cases where the Flack parameter was not 
 determined with a BASF/TWIN refinement or with essentially centrosymmetric
 data. See: Flack, H.D., Bernardinelli, G, Clemente, D.A., Linden, A. Spek,
 A.L. (2006) Acta Cryst. B62, 695-701.  

 The contribution of F(-h,-k,-l) to F(h,k,l) is likely not included in the  
 FCF file. This usually indicates that the Flack parameter was NOT determined   
 with a BASF/TWIN type of refinement.   

 This ALERT reports on the number of reflections with   
 (Fo**2 - Fc**2) / Sigma(Fo**2) < - 100.0. In case of strong reflections this   
 might be due to extinction (to be addressed with the refinement of an  
 extinction parameter. Otherwise Those reflections are better removed from the  
 final refinement since they are in systematic error. Of course, a valid
 reason for this problem should be found.

 This ALERT reports the number of reflections with intensities seriously
 effected by the beamstop. Reflections are counted for which theta < 3  
 Degrees and (Fo**2 - Fc**2) / sqrt(weight) < - 10.0. Those reflections 
 are better removed from the final refinement since they are in systematic  
 error. 

 Check reflection statistics of the data in the FCF for consistency with
 the data reported in the CIF. A difference usually indicates an edited CIF or
 an FCF file that was not created in the same SHELXL run where the CIF was
 created. In rare cases this ALERT may point to using a SHELXL TWIN 
 instruction for handling a non-merohedral twin. 

 Please check whether the supplied FCF corresponds with the CIF produced in 
 the same least squares refinement job. See at the end of Section #10 in the
 .ckf report file for details. The test is based on the observed and
 calculated F**2 in the FCF and de weight parameters taken from the CIF.  

 Please check whether the supplied FCF corresponds with the CIF produced in 
 the same least squares refinement job. See at the end of Section #10 in the
 .ckf report file for details. The test is based on the observed and
 calculated F**2 in the FCF and de weight parameters taken from the CIF.  

 Please check whether the supplied FCF corresponds with the CIF produced in 
 the same least squares refinement job. See at the end of Section #10 in the
 .ckf report file for details. The test is based on the observed and
 calculated F**2 in the FCF and de weight parameters taken from the CIF.  

 Check & Explain why the Reported Rho(min) differs significantly from the  
 value calculated on the basis of the reported structure (using the AIM model)  
 Note: The Reported and Calculated values may differ slightly due to a
 differing peak interpolation algorithm, or significantly when the refinement
 is based on a non-spherical atom model.  

 Check & Explain why the Reported Rho(max) differs significantly from the  
 value calculated on the basis of the reported structure (using the AIM model)  
 Note: The Reported and Calculated values may differ slightly due to a
 differing peak interpolation algorithm, or significantly when the refinement
 is based on a non-spherical atom model.  

 Please check whether the R1 value that is reported in the CIF corresponds  
 with the R1 value calculated from the parameters supplied in the CIF. See at
 the end of Section #10 in the .ckf report file for details.  This test is
 based on the observed reflection data in the FCF and reflection data that are
 calculated with the parameters (i.e. coordinates, displacement and weight
 parameters) taken from the CIF.

 Please check whether the wR2 value that is reported in the CIF corresponds 
 with the wR2 value calculated from the parameters supplied in the CIF. See at 
 the end of Section #10 in the .ckf report file for details.  This test is
 based on the observed reflection data in the FCF and reflection data that are
 calculated with the parameters (i.e. coordinates, displacement and weight
 parameters) taken from the CIF.

 Please check whether the S value that is reported in the CIF corresponds   
 with the S value calculated from the parameters supplied in the CIF.   
 See at the end of Section #10 in the .ckf report file for details. 
 This test is based on the observed reflection data in the FCF and  
 reflection data that are calculated with the parameters (i.e. coordinates, 
 displacement and weight parameters) taken from the CIF.

 SHELXL weight parameters are expected to be given in the format below: 
 _refine_ls_weighting_details   
 'calc w=1/[\s^2^(Fo^2^)+(0.1000P)^2^+0.0000P] where P=(Fo^2^+2Fc^2^)/3'
 JANA style weight is expected to be given in the format:   
 _refine_ls_weighting_details 'w=1/(\s^2^(I)+0.0016I^2^)'   
  Do not edit this string or make it into a text block between ';'. 

 Check the proposed Twin Law. The entry in () represents the proposed rotation
 axis in reciprocal space and the one in [] the corresponding rotation is
 direct space. The relevant Twin Matrix can be found in the file '.ckf'.  
 Note: This analysis is based on Fobs**2/Fcalc**2 differences with Fcalc**2 
 data taken from the .fcf file (i.e. Fobs, Fcalc listing).   
 ALERT-930 is expected to generate an related ALERT-931 as well.

 Check the proposed Twin Law. The entry in () represents the proposed rotation
 axis in reciprocal space and the one in [] the corresponding rotation is
 direct space. The relevant Twin Matrix can be found in the file '.ckf'.  
 Note: This test is based on Fcalc**2 values calculated with the data in the
 CIF. This ALERT can be ignored when twinning has been addressed in the
 refinement (As indicated by the Absence of ALERT 930). Please make sure that
 twinning is mentioned in the write-up of the structure report and paper. 

 This ALERT reports on the number of 'OMIT h k l' records (Reflections
 omitted from the least-squares refinement) in the CIF embedded .res file.
 Generally, there should be a good reason for excluding those observed
 experimental data point such as (partial) obscuration by the beam stop. 

 This ALERT reports on the number of reflections for which I(obs) and I(calc)   
 differ more that 10 times SigmaW. (The latter being the square root of 
 1.0/weight for that reflection in the L.S. refinement). The reason for those   
 deviations should be investigated. When shown to be systematic errors, those
 reflections are best removed from the refinement and their omission from the
 refinement reported in the experimental section of an associated paper. 

 Both significantly positive and significantly negative values should   
 invoke a search for a likely cause and a corrective action.

 The reason for the use of a DAMP instruction in the final refinement job   
 should be discussed/reported. 'DAMP 0.0' should not be used for small  
 molecule refinement since it masks non-convergence.

 The use of an exponential term in the SHELXL weight expression is not  
 recommended in the final refinement stage (i.e. the third parameter in the 
 SHELXL WGHT record).   

 SHELXL suggests/optimizes two weight parameters aiming at bringing the S
 value (Goodness-of-Fit) close to 1.0. This 939_ALERT reports the S value when
 based on the supplied sigma(I) values only. A large value of this not weight
 optimized S generally indicates the presence of large outliers in the data
 set. Examples are reflections (partially) 'measured' behind the beam stop.
 (See also ALERT_919). The latter low order reflections are best left out from 
 the final refinement with an OMIT hkl instruction.  

 Apparently, observed data with I > n * sigma(I) were used in the F**2  
 least squares refinement, rather than all observed data. Please ignore
 when the dataset is essentially complete (and all reflections used in the
 refinement 'observed'.    

 Multiple hkl measurements are advised, either of symmetry related reflections  
 and/or at different psi angle values around the diffraction vector normal to
 the diffracting plane. A good multiplicity is needed for a meaningful  
 multi-scan based correction for absorption. Details can be found in the
 '.ckf' file.  

 Check the validity of the observed reflection data.

 Reported (in the CIF) and Calculated (from Theta-max in CIF) Max(Hmax,-Hmin)   
 values differ by at least one unit. Check the consistency of wavelength and
 reported resolution data items.

 Reported (in the CIF) and Calculated (from Theta-max in CIF) Max(Kmax,-Kmin)   
 values differ by at least one unit. Check the consistency of wavelength and
 reported resolution data items.

 Reported (in the CIF) and Calculated (from Theta-max in CIF)  Max(Lmax,-Lmin)  
 values differ by at least one unit. Check the consistency of wavelength and
 reported resolution data items.

 Reported (in the CIF) and Actual (in the FCF)  Max(Hmax,-Hmin) values differ  
 by more than one unit. Check for data set truncation.   

 Reported (in the CIF) and Actual (in the FCF) Max(Kmax,-Kmin) values   
 differ by more than one unit. Check for data set truncation.   

 Reported (in the CIF) and Actual (in the FCF) Max(Lmax,-Lmin) values   
 differ by more than one unit. Check for data set truncation.   

 Calculated (From Theta-Max in the CIF) and Actual (in the FCF) Max(Hmax,-Hmin) 
 values differ by more than one unit. Check for data set truncation.

 Calculated (From Theta-Max in the CIF) and Actual (in the FCF) Max(Kmax,-Kmin) 
 values differ by more than one unit. Check for data set truncation.

 Calculated (From Theta-max in the CIF) and Actual (in the FCF)
 Max(Lmax,-Lmin) values differ by more than one unit. Check for data set
 truncation.

 Multiple strongly negative intensities may be indicative for poor integration
 of the  diffraction images. Too many negative intensities may result in
 higher than usual wR2 values.  

 Generally, both positive and slightly negative intensities are expected in a
 data set. Resetting negative intensities to zero may bias the refinement
 results and the 'analysis-of-variance' as reported e.g. in the  SHELXL output
 listing. 

 Reflections with Sig(I) = 0.0 are suspect and best left out of the  
 refinement. This type of reflections will result in multiple R & S-value
 difference ALERTS. Note: Sometimes such a 0.00 value may be related to the
 limited number of reported decimals in the FCF file as opposed to those in 
 the unscaled unmerged HKL file.

 Check unusual reported/refined zero SHELXL weight parameter values.
 Note: SHELXL will not refine to negative values.   

 Two weight parameter values reported in the CIF weight expression string
 '_refine_ls_weighting_details' are found to differ from those archived in the
 WGHT parameter value list in the embedded ',res' file. The latter values are
 assumed to be the correct ones as part of the final refinement. 
 Recreation of the FCF from the embedded '.ins' & '.hkl' files will result in
 inconsistent wR2 and S values and associated ALERTS in the checkCIF report
 when those weight parameter values differ. Please also check the proper
 format of the SHELXL weight expression string.

 Additional refinement cycles are advised using the (SHELXL) suggested new   
 WGHT parameter values. 

 
 The conventional OMIT threshold is I>2\s(I). This ALERT reports the use of
 a deviating value effecting e.g. the reported R1 value.

 
 In general refinement with all experimental reflection data is advised. A
 theta value cutoff may be considered to avoid refinement on noise for
 resolution shells with  < 2*sigma. 

  
 Report on predicted wR2, based on counting statistics, or SHELXL style
 weighting as compared to the actual refined wR2.
 See: J.Henn & A.Schonleber, Acta Cryst. (2013) A69, 549-558.

 A structure based on electron diffraction is reported. 
 Please check for the use of proper electron scattering factors in the  
 refinement. X-ray scattering factors are not appropriate for a SHELXL based
 refinement.

 Larger than expected residual density maximum outside metal atom locations.
 This might be caused by unaccounted for twinning, wrongly assigned atom
 types, unaccounted for solvent and other model errors.
 Note: This value is based on an Atom-in-Molecule (AIM) calculation and may
 differ significantly for a refinement using a non-spherical atom model.

 Larger than expected residual density minimum outside metal atom locations.
 This might be caused by unaccounted for twinning, wrongly assigned atom
 types and other model errors. 
 Note: This value is based on an Atom-in-Molecule (AIM) calculation and may
 differ significantly for a refinement using a non-spherical atom model.

 Larger than expected residual density maximum on metal atom location.  
 This might be caused by unaccounted for twinning, wrongly assigned 
 atom types and other model errors. Another cause may be a SHELXL 'DAMP 0 0'
 instruction for a non-converged refinement.

 Larger than expected residual density minimum on metal atom location. This  
 might be caused by unaccounted for twinning, wrongly assigned atom types 
 and other model errors. Another cause may be a SHELXL 'DAMP 0 0' instruction
 for a non-converged refinement.

 Positive density found in a difference density map at a position within
 bonding distance for a H-atom from a nitrogen or oxygen atom. A
 possible reason can be a missing hydrogen atom. Also check for tautomerism.

 Negative density found in a difference density map at a location within   
 bonding distance for a H-atom from a nitrogen or oxygen atom. A possible 
 reason can be a misassigned H-atom. Also check for tautomerism.

 Negative density found in a difference density map at a H-atom location.   
 A reason might be an incorrect AFIX instruction.   

 Difference density maps generally show residual densities on C-C bonds.
 A significant number of those may indicate good data. Cases where this number
 is zero should be investigated for a reason. The use of aspherical scattering
 factors in the refinement model may be one.  

 Non-spherical scattering factors were used in the refinement. Some ALERTS 
 that assume spherical scattering factors are suppressed. 

 Check for missing anomalous scattering factors.

 Check for non-zero f" anomalous scattering factor values in the CIF.   
 Note: Zero values are correct for SHELXL MERG 4 refinements.   

 Check the supplied anomalous scattering factor f' value against the    
 International Tables tabulated value.   

 Check the supplied anomalous scattering factor f'' value against the 
 International Tables tabulated value.   

 Check the supplied anomalous scattering factor f' value for the non
 CuKa, GaKa, MoKa, AgKa, InKa wavelength against those of S. Brennan & 
 P.L. Cowan (1992). Rev. Sci. Instr., 63, 850-853.

 Check the supplied anomalous scattering factor f'' value for the non   
 CuKa, GaKa, MoKa, AgKa, InKa wavelength against those of S. Brennan & P.L.
 Cowan (1992). Rev. Sci. Instr., 63, 850-853.

 Check for non-zero f' anomalous scattering factor values in the CIF.   

 A BASF/TWIN refinement of the Flack x parameter is indicated when its value
 deviates significantly from zero as measured against its associated s.u.
 value. This applies in particular for structures containing significant
 anomalous scatterers. With a default SHELXL refinement, the Flack x parameter
 value is not taken into account in the refinement model as part of the
 calculation of F(calc) and the creation of the FCF file. Inclusion of a Flack
 x in the refinement model that deviates significantly from zero will lead to
 lower R-values and a better model.  

 Check the supplied anomalous scattering factor f' value for the reported
 wavelength. (No internally calculated f' value available).   

 Check the supplied anomalous scattering factor f'' value for the reported
 wavelength. (No internally calculated f'' value available).   

 SQUEEZE jobs should as of the availability of SHELXL20xy be based on the   
 use of a CIF+FCF created in a SHELXL20xy refinement.   
 See A.L.Spek (2015) Acta Cryst. C71, 9-18. 

 Warning that the reflection data are not experimental but calculated. Those
 data are just meant to be used as test data. 

 The CIF Reported and actual (in FCF) '_reflns_number_gt' values are found to   
 differ. R1 values are calculated by checkCIF for cross-checking with the R1
 value as reported in the CIF. The number of 'observed' reflections that goes
 into that calculation is based on the criterium I > 2.0 * sigma(I). A 
 refinement program like SHELXL uses F(obs) > 4.0 * sigma(F(obs)) as
 'observed' criterium in its calculation of R1. Both numbers may differ
 slightly for numerical reasons. Large differences may point to
 inconsistencies between the CIF and FCF to be investigated. 

 The CIF-embedded .res file includes a call for a .bodd include file. Such a
 file is not found externally or CIF-embedded. Without this file, the   
 details of the refinement are not complete and no .fcf can be created. 

 A SHELXL job generally includes a MERG instruction. If not, the refinement 
 and 'LIST 4' FCF will include redundant reflections (i.e. refinement on more 
 than the unique set of reflections.)   

 A SHELXL 'LIST 4' Style FCF File should comply with the SHELXL Format.   
 This implies that an FCF should include only a unique set of reflections.  
 For non-centrosymmetric structures, Friedel pairs should be included.  

 The CIF file reports refinement with SHELXL whereas the supplied reflection
 file is not of the SHELXL 'LIST 4' or 'LIST 8' type. Proper FCF validation is
 not possible with the alternative LIST options.   

 IUCr CheckCIF validation requires SHELXL 'LIST 4', 'LIST 8' or Equivalent  
 Fo**2,Fc**2, sigma(Fo**2)  reflection files.   

 IUCr CheckCIF validation requires SHELXL 'LIST 4', 'LIST 8' or equivalent  
 Fo**2,Fc**2, sigma(Fo**2) reflection files. The 'LIST 6' style file has the
 dispersion contribution removed making this file unsuitable for external
 Analysis-of-Variance type of analysis and validation of the final refinement.