Natalia, You are correct in saying that all codes do add a compensating background to charged systems. However, without the SimulateDoping option, the background added is homogeneous, while with the abovementioned option,
"This background charge is constant at points of the mesh near the atoms, and zero at points far from the atoms." Cheers, Marcos On Fri, Dec 2, 2011 at 1:39 PM, Natalia Martsinovich < [email protected]> wrote: > > Dear SIESTA users and developers, > > I have a question about the SimulateDoping option. > > The manual says: "This option instructs the program to add a background > charge density to simulate doping. > The new “doping” routine calculates the net charge of the system, and adds > a compensating > background charge that makes the system neutral." > > As I understand, SIESTA (or any code) always (with or without this option) > adds a compensating background charge density for charged cells - otherwise > charged systems would have an infinite positive or negative charge. > > Then, what is the effect of SimulateDoping? > > We have done calculations of charged cells (slab+adsorbate) with and > without the SimulateDoping option. The results (projected density of > states) are somewhat different, but neither of them look physically > meaningless. > > Is this option applicable to slab+adsorbate systems, where the charge is > believed to be on the adsorbate? > > I hope that someone can explain this or refer me to some papers. > > Best regards, > > Natalia >
