[molpro-user] Bug in the CUBE program

[Andy May]

Many thanks to Zdenek who discovered the problem in the code relating to
symmetry 5 and higher. This has been fixed in the development version,
and will be backported to 2010.1 after testing.

Best wishes,

Andy

On 22/10/10 17:23, Z.Masin wrote:
> Dear Molpro users,
> 
> I am having problems with the CUBE program which is supposed to write a selected charge/spin density to a file in the CUBE format. I am attaching the Molpro output for one selected run. 
> 
> I need to process in another program charge densities of selected virtual orbitals, but the CUBE program does not allow me to write densities of orbitals of symmetries 5 and higher (my molecule is of D2h symmetry). If I want to write densities of an orbital of symmetry 5 and higher it fails with the error message (in this case I required the orbital 1.5):
> 
>  orbital=       1.5  (Input string=1.5)
>  ? Error
>  ? Illegal orbital specification. Orbital record must be given as RECORD=number.file
>  ? The problem occurs in cube
> 
> I would like to point out that there is no problem in the syntax for the command (unlike the error message suggests), because I am using the same syntax for writing orbitals of symmetries 4 and lower and it works fine.
> 
> On the other hand if I use the keywords 'HOMO', 'LUMO' or 'OCC' (section 33.7.3 of the Molpro manual) the program writes the correct orbitals, so the problem only occurs when I want to specify individual orbitals. Also I am unable to write densities of all available orbitals using the keyword 'ALL' because the program fails with the error message:
> 
>  ? Error
>  ? Too many files
>  ? The problem occurs in cube
> 
> suggesting that it cannot write to many (in this case 156) files at once.
> 
> I have been thinking about a way how to get around the problem with selecting the individual orbitals for output by producing a set of orbitals which would contain only a selected number  of orbitals (perhaps only one orbital) whose densities I would like to get and then use the keyword 'ALL' to write densities of these orbitals. Is there any way how to produce a record of orbitals which contains only a selected number of orbitals or even only one orbital?
> 
> I have a second question which is related to the format of the CUBE file produced by Molpro. If I write a cube file containing charge density of an electronic state then the format of the CUBE file is exactly the same as explained in  the manual and all values of charge density in that cube file are positive. 
> But if I look into a cube file containing charge density of a single orbital then there is one extra line in the CUBE file (following the data on geometry of the molecule) which contains two integers whose meaning is unknown to me and there is nothing on these numbers in the manual as well. Also this cube file contains negative values of the density which is something I do not understand as well. What is the meaning of these negative values of density and of the two extra integers in the CUBE file?
> 
> I would appreciate any help on these issues. Thank you very much.
> 
> Zdenek Masin
> PhD student
> Department of Physics and Astronomy
> The Open University
> Milton Keynes
> United Kingdom
> 
> 
> 
> 
> _______________________________________________
> Molpro-user mailing list
> Molpro-user at molpro.net
> http://www.molpro.net/mailman/listinfo/molpro-user