Dear users,

I am facing a strange problem with Artemis and i don't know how to get rid of it. I hope any expert in the mailing lists would help me solving this problem.

I have EXAFS data on Fe and Mn K-edge for Arrojadite mineral. Since the crystal structure of this mineral is not  known we did neutron diffraction to refine the structure. To support the neutron diffraction results, we planned to do EXAFS on this mineral (Fe and Mn k-edge). We have around 8 different atomic species (i.e. Na,K,Fe,Mn,Al,Cr,O,P) in the mineral and some of the sites are fractionally occupied.

Since ATOMS can not handle more than 7 potentials and site occupancy other than unity i have made following changes :
1) replaced Cr with Al, mainly because Cr is present in a few ppm only (relative importance).
2) set all site occupancy to 1

Even after doing the above changes, i got following error message when i run FEFF

**************
 Feff 6L.02
  WARNING:  TWO ATOMS VERY CLOSE TOGETHER.  CHECK INPUT.
   atoms    11   20
    11   -3.01973   -0.20361    1.44233
    20   -3.54653   -0.05083    0.83053
  Run continues...
  WARNING:  TWO ATOMS VERY CLOSE TOGETHER.  CHECK INPUT.
   atoms    46   62
    46   -0.40014   -0.20361    5.03633
    62    0.12666   -0.05083    5.64813
  Run continues...
 Sodium Potassium Aluminium Chromium Iron Manganese Phosphate Hydrate
 Al Cr Fe7.53 H2 K1.21 Mn6 Na3 O50 P12
 Kallfa\&s, Christoph
 Crystal structure, chemical composition and
 thermal behaviour of arrojadite
 Calculating potentials and phases...
     free atom potential and density for atom type    0
     free atom potential and density for atom type    1
     free atom potential and density for atom type    2
     free atom potential and density for atom type    3
     free atom potential and density for atom type    4
     free atom potential and density for atom type    5
     free atom potential and density for atom type    6
     free atom potential and density for atom type    7
     overlapped potential and density for unique potential    0
     overlapped potential and density for unique potential    1
     overlapped potential and density for unique potential    2
     overlapped potential and density for unique potential    3
     overlapped potential and density for unique potential    4
     overlapped potential and density for unique potential    5
     overlapped potential and density for unique potential    6
     overlapped potential and density for unique potential    7
     muffin tin radii and interstitial parameters
     phase shifts for unique potential    0
   Hard tests failed in fovrg.
  Muffin-tin radius may be too large; coordination number too small.
     phase shifts for unique potential    1
   Hard tests failed in fovrg.
  Muffin-tin radius may be too large; coordination number too small.
     phase shifts for unique potential    2
   Hard tests failed in fovrg.
  Muffin-tin radius may be too large; coordination number too small.
     phase shifts for unique potential    3
   Hard tests failed in fovrg.
  Muffin-tin radius may be too large; coordination number too small.
     phase shifts for unique potential    4
   Hard tests failed in fovrg.
  Muffin-tin radius may be too large; coordination number too small.
     phase shifts for unique potential    5
   Hard tests failed in fovrg.
  Muffin-tin radius may be too large; coordination number too small.
     phase shifts for unique potential    6
   Hard tests failed in fovrg.
  Muffin-tin radius may be too large; coordination number too small.
     phase shifts for unique potential    7
   Hard tests failed in fovrg.
  Muffin-tin radius may be too large; coordination number too small.
 Preparing plane wave scattering amplitudes...
   nncrit in prcrit       9
 Searching for paths...
     Rmax  7.0000  keep and heap limits   0.0000000   0.0000000
    Preparing neighbor table
     nfound  nheap  nheapx  nsc    r
       1000    469     469   3    5.2890
       2000    730     730   3    5.7147
       3000    907     907   3    5.9455
       4000   1040    1040   3    6.1157
       5000   1177    1177   3    6.2616
       6000   1319    1319   3    6.3835
       7000   1412    1413   3    6.4825
       8000   1443    1444   3    6.5749
       9000   1469    1473   3    6.6585
      10000   1432    1474   3    6.7377
      11000   1367    1474   3    6.8062
      12000   1202    1474   3    6.8680
      13000    870    1474   3    6.9289
      14000    251    1474   3    6.9856
.
.
.
.
.
..
.

    This warning indicates a problem in the Atoms list.  For some
    reason you have atoms that are separated by than 0.93 Angstroms
    (or 1.75 Rydberg).

    This may be due to the presence of hydrogen atoms in your feff.inp
    file, in which case this is an innocuous warning and may be ignored.

    However, this may indicate a problem constructing the Atoms
    list. The most common cause of this problem is a mistake in the
    crystallographic data used on the Atoms page.  You may have
    incorrect values for lattice constants or angles or incorrect
    values for site coordinates.  You may need a shift vector to move
    the lattice into its standard setting.

    Please be aware that Atoms works with 5 digits of precision.
    Thus, if you have a site with a coordinate of 1/3, you should use
    either "1/3" or "0.33333" on the Atoms page.  Using insufficient
    precision, say "0.333" is a common cause of this error message.

    Artemis has continued on the possibility that the warning is
    caused by hydrogen atoms, but be warned that the feff.inp may
    require your attention.

************************************
I have attached both original.cif file (with all 8 atomic species and fractional occupancy) and modified.cif file (Cr replaced by Al and all site occupancy set to 1) with this mail for your perusal.

I hope some experts in the mailing list would suggest me some solutions.

Many more thanks in advance for your valuable time and suggestions....

Best regards,
Rana