Dear Michele,
it is very helpful and I FINALLY got some nice results.
good day!

2015-06-01 21:33 GMT+02:00 Michele Dolfi <dolfim@phys.ethz.ch>:
Dear Cathy,

I’m not sure I fully get what if your goal. Here is anyway what I would do.

Usually for inhomogeneous models we define lattice sites with different type and set different values of the model parameters for each type.

Quick example:
In mylattice.xml you define the graph
<LATTICES>
 <GRAPH name="my great lattice" dimension="1" vertices=“3" edges="2">
  <VERTEX id="1" type="0"><COORDINATE>0</COORDINATE></VERTEX>
  <VERTEX id="2" type="1"><COORDINATE>1</COORDINATE></VERTEX>
  <VERTEX id="3" type=“2"><COORDINATE>2</COORDINATE></VERTEX>
  <EDGE source="1" target="2" id="1" type="0" vector="1"/>
  <EDGE source="2" target="3" id="2" type="0" vector="1"/>
 </GRAPH>
</LATTICES>


In the parameter file:
LATTICE_LIBRARY="mylattice.xml"
LATTICE="my great lattice"
MODEL=“fermion Hubbard” 
t=1
U=8
mu0=0
mu1=10
mu2=5


This will have a different chemical potential for each of the three sites.


Best regards,
Michele

--
ETH Zurich
Michele Dolfi
Institute for Theoretical Physics
HIT G 32.4
Wolfgang-Pauli-Str. 27
8093 Zurich
Switzerland



On 01 Jun 2015, at 18:31, Cathy Wu <cathy.em2013@gmail.com> wrote:

Dear everyone,
Does anyone know how to use the "coordinate " parameter in the models.xml file.
I am trying to defining a quasi-periodic potential. But the "sparsediag" function does not recognize site "i".
best regards!