[gmx-users] g_rdf and structure factor

[=?gb2312?b?wO61wrL9?=]

Dear gromacs users,
   I am using Gromacs 3.3 to simulate a coarse-grained molecule which con=
tains atom B and L. The atom B and L are defined myself. When the simulat=
ion finished, I tried to use g_rdf to calculate the static structure fact=
or. But it gave the error message:
  =20
-------------
Program g_rdf, VERSION 3.3
Source code file: gmx_rdf.c, line: 817

Fatal error:

Error: atom type (B) not in list (18 types checked)!

-------------------------------------------------------

   I've read the source code of g_rdf, found that the variable CM_t[] onl=
y contains 18 atom types. I think it is the root of the above error, righ=
t? I want to add my user-defined atoms into the code, but I found that th=
e program uses the Cromer Mann formula to reproduce the experimental scat=
tering factors, can anyone tell me how I choose the nine Cromer Mann form=
ula coefficients of my atoms?
   Why it don't use the following formula to calculate the structure fact=
or?
   S(k)=3D1+4*pi*density*integral{r^2*sin(kr)/k/r*g(r)*dr}

   Thank you for your answers.  ^_^