CCL:Re: How to calculate molecular volume

To: CCL <>
 Subject: CCL:Re: How to calculate molecular volume
 The ASV freeware computes analytically surfaces and volumes
 of any union of spheres. The precision is near the machine
 real*8 precision, and thus the volumes values may be used for
 numerical calculations of gradients. The reliability is
 good, since the method handles all topological situations
 for the set of spheres, including those which are never
 encountered in chemistry. If your input set contains
 some uncertainty, you can automatically generate perturbated
 inputs and see the effects on the analytical values.
 If you need real*16 precision, I can recompile for you
 the f77 source, provided that this is on a platform enabling
 it (e.g. the IBM autodblpad compiler option).
 The freeaware ASV includes also a Monte-Carlo calculation of
 surfaces and volumes (with precision calculus), which is useful
 if you need a short computing time for large sets, and provided
 that a limited precision suffice, such as for empirical correlations.
 See programmes, documentation, and ref. of the method on:
 The rectangular grid containing the molecule may be approximated
 > from the xmin,xmax, ymin,ymax, zmin,zmax via random rotations.
 Before doing that, since the parallelepiped box is convex, you
 can first extract the extremal points which the RADI freeware,
 which computes also the minimal sphere containing your set of
 points. See programmes, documentation and reference on:
 Michel Petitjean,                     Email:
 ITODYS (CNRS, UMR 7086)            
 1 rue Guy de la Brosse                Phone: +33 (0)1 44 27 48 57
 75005 Paris, France.                  FAX  : +33 (0)1 44 27 68 14
 =?gb2312?q?Jinsong=20Zhao?= <> wrote:
 >Dear all,
 >I hope to know if there is any program or routine
 >could calculate the molecular volume with high
 >precision and reliability, moreover, to figure out the
 >length, width and high of the grid that contain the
 >Any comments, suggestion are welcome!
 >Thanks for essential responses!
 >Best regards,