Dear ALPS friends,
I'm trying to generate Hamiltonian matrix elements for lattice models using the kitchen sink of symmetries in ALPS. Unfortunately, I can't figure out where I can get them by reading the source code.
Specifically, I'm looking for the quantity M(i,j,k,l) which defines the Hamiltonian H = sum(i,j,k,l) M(i,j,k,l) c(i)+ c(j)+ c(l) c(k). I doubt it is ever stored complete in memory, but I just want to read it out element-by-element in sparse format into an external file. If no part of this quantity ever exists but can be obtained by some modifying the code, that works, too.
If someone can point me to the right place and make a suggestion for what to do, that would be fantastic. If this is not a simple request at all, but someone still wants to help me, I can give a much more detailed explanation of what I'm trying to do.
Jeff Hammond The University of Chicago
Hi Jeff,
On Dec 3, 2006, at 7:07 PM, Jeff Hammond wrote:
Dear ALPS friends,
I'm trying to generate Hamiltonian matrix elements for lattice models using the kitchen sink of symmetries in ALPS. Unfortunately, I can't figure out where I can get them by reading the source code.
We can help but I do not understand the Hamiltonian:
H = sum(i,j,k,l) M(i,j,k,l) c(i)+ c(j)+ c(l) c(k).
This seems to split trivially into single-site terms c(i) and 2-site terms c(l) c(k)
Also, what is c?
I doubt it is ever stored complete in memory, but I just want to read it out element-by-element in sparse format into an external file. If no part of this quantity ever exists but can be obtained by some modifying the code, that works, too.
If someone can point me to the right place and make a suggestion for what to do, that would be fantastic. If this is not a simple request at all, but someone still wants to help me, I can give a much more detailed explanation of what I'm trying to do.
We'll first need to understand what yur question above is. I do not fully understand your H yet?
Matthias
Sorry for my failure to communicate. My notation was wholly insufficient.
In Latex format, it would be
\begin{equation} \hat{H}=\sum_{i,j,k,l}M^{i,j}_{k,l}\hat{a}^{\dag}_{i}\hat{a}^{\dag}_{j}\hat{a}_{l}\hat{a}_{k} \end{equation}
which I attached as a GIF just in case.
In the original equation, the "plus sign" was supposed to represent the dagger of a creation operator, while c(k) was an annihilation operator for site/function/orbital k.
The matrix elements could be broken up into 1- and 2-body terms for my purpose as well.
Jeff
On 12/3/06, Matthias Troyer troyer@phys.ethz.ch wrote:
Hi Jeff,
On Dec 3, 2006, at 7:07 PM, Jeff Hammond wrote:
Dear ALPS friends,
I'm trying to generate Hamiltonian matrix elements for lattice models using the kitchen sink of symmetries in ALPS. Unfortunately, I can't figure out where I can get them by reading the source code.
We can help but I do not understand the Hamiltonian:
H = sum(i,j,k,l) M(i,j,k,l) c(i)+ c(j)+ c(l) c(k).
This seems to split trivially into single-site terms c(i) and 2-site terms c(l) c(k)
Also, what is c?
I doubt it is ever stored complete in memory, but I just want to read it out element-by-element in sparse format into an external file. If no part of this quantity ever exists but can be obtained by some modifying the code, that works, too.
If someone can point me to the right place and make a suggestion for what to do, that would be fantastic. If this is not a simple request at all, but someone still wants to help me, I can give a much more detailed explanation of what I'm trying to do.
We'll first need to understand what yur question above is. I do not fully understand your H yet?
Matthias
comp-phys-alps-users@lists.phys.ethz.ch
