+1 vote
asked by (680 points)
edited by

I'm studying the Hubbard Model and I want to calculate the following expected value:

$$
\left< c_{i,\sigma_{i}}^{\dagger}c_{j,\sigma_{j}}^{\dagger}c_{k,\sigma_{k}}c_{l,\sigma_{l}}\right>,
$$

which represents the two-particle density matrix by setting @@ \left(i,\sigma_{i}\ j,\sigma_{j}\right) \rightarrow \text{row; and }\left(k,\sigma_{k}\ l,\sigma_{l}\right) \rightarrow \text{column} @@. I'm doing in the most optimized way, following -Correlator- and transforming the spinfull fermion operators into bosonic ones -Jordan Wigner-. I rewrite the Jordan Wigner to compress the different spin cases:

$$
c_{i,\sigma_{i}} = F_{1}\cdots F_{i-1}\left(F_{i}^{\sigma_{i}}a_{i,\sigma_{i}}\right),
$$

with @@ \sigma_{i} = 0\text{ corresponds } \uparrow \text{ and } \sigma_{i} = 1\text{ corresponds } \downarrow @@. When @@ i\neq j\neq k\neq l @@ the algorithm works flawlessly, but when two or more indexes are equal It doesn't.

Here's an example when (i==k) < j < l. I have to sort all operators in an increasing way. Using anti-commutator of fermions:

$$
\left< c_{i,\sigma_{i}}^{\dagger}c_{j,\sigma_{j}}^{\dagger}c_{k,\sigma_{k}}c_{l,\sigma_{l}}\right> = -\left< c_{i,\sigma_{i}}^{\dagger}c_{k,\sigma_{k}}c_{j,\sigma_{j}}^{\dagger}c_{l,\sigma_{l}}\right>.
$$

Substituting the transformation of Jordan Wigner we have:

$$
\begin{align}
=-\left( a_{i,\sigma_{i}}^{\dagger}F_{i}^{\sigma_{i}} \right) \underbrace{\left( F_{i-1}\cdots F_{1} \right)\left( F_{1}\cdots F_{k-1} \right)}_{=1} \left( F_{k}^{\sigma_{k}}a_{k,\sigma_{k}} \right)
\left( a_{j,\sigma_{j}}^{\dagger}F_{j}^{\sigma_{j}} \right) \underbrace{\left( F_{j-1}\cdots F_{1} \right) \left( F_{1}\cdots F_{l-1} \right)}_{\left(F_{j}\cdots F_{l-1}\right)} \left( F_{l}^{\sigma_{l}}a_{l,\sigma_{l}} \right)
\end{align}
$$

$$
=-\left(a_{i,\sigma_{i}}^{\dagger}F_{i}^{\sigma_{i}+\sigma_{k}}a_{k,\sigma_{k}}\right)\left(a_{j,\sigma_{j}}F_{j}^{\sigma_{j}+1}\right)\left(F_{j+1}\cdots F_{l-1}\right)\left(F_{l}^{\sigma_{l}}a_{l,\sigma_{l}}\right).
$$

But I'm not getting the correct results. Here's a sample code:

IQIndex il = commonIndex(psi.A(i),psi.A(i+1),Link);
IQTensor C = psi.A(i);

C  *= (sigmai) ? sites.op("Adagdn",i) : sites.op("Adagup",i);
if ((sigmai+sigmak+1) % 2) {
    C.noprime(Site);
    C *= sites.op("F",i);
}
C.noprime(Site);
C *= (sigmak) ? sites.op("Adn",i) : sites.op("Aup",i);
C *= dag(prime(prime(psi.A(i),Site),il));

for(int m = i+1; m < j; ++m) {
    C *= psi.A(m);
    C *= dag(prime(psi.A(m),Link));
}

C *= psi.A(j);
C *= (sigmaj) ? sites.op("Adagdn",j) : sites.op("Adagup",j);
if (!sigmaj) {
    C.noprime(Site);
    C *= sites.op("F",j);
}
C *= dag(prime(prime(psi.A(j),Site),Link));

for(int m = j+1; m < l; ++m) {
    C *= psi.A(m);
    C *= sites.op("F",m);
    C *= dag(prime(prime(psi.A(m),Site),Link));
}

IQIndex kl = commonIndex(psi.A(l),psi.A(l-1),Link);
C *= psi.A(l);
if (sigmal) {
    C *= sites.op("F",l);
    C.noprime(Site);
}
C *= (sigmal) ? sites.op("Adn",l) : sites.op("Aup",l);
C *= dag(prime(prime(psi.A(l),Site),kl));

double result =  -C.real();

I still do not know where I'm going wrong. Whether it is in the Jordan Wigner transformation for spinfull fermions, or if in some part of the code. Maybe you can help me with some reference to this calculation. I thank you all!

1 Answer

+1 vote
answered by (680 points)
selected by
 
Best answer

Well, after posting this problem I ended up solving it. I decided to keep the post here if other people do the same silly thing as me. Another reason to leave the post here is that people who want to make a similar calculation can see and even exchange ideas with me.

The calculation is right, but the order of the operators' application is wrong in the code. Simple error, but since the correlator is calculated starting from the lowest to the highest site index (usually), I did not realize that in the same site the application of operators is obviously in the anti lexicographic order. The order of application is like this:

$$
\begin{align}
=-\underbrace{\left(a_{i,\sigma_{i}}^{\dagger}F_{i}^{\sigma_{i}+\sigma_{k}}a_{k,\sigma_{k}}\right)}_{\leftarrow}\underbrace{\left(a_{j,\sigma_{j}}F_{j}^{\sigma_{j}+1}\right)}_{\leftarrow}\underbrace{\left(F_{j+1}\cdots F_{l-1}\right)}_{\rightarrow}\underbrace{\left(F_{l}^{\sigma_{l}}a_{l,\sigma_{l}}\right)}_{\leftarrow}.
\end{align}
$$

commented by (70.1k points)
Glad to hear you solved it, and thanks for posting the solution. These things are tricky to get right. We are hoping to add some features to ITensor later this year to make working with fermions easier
Welcome to ITensor Support Q&A, where you can ask questions and receive answers from other members of the community.

Formatting Tips:
  • To format code, indent by four spaces
  • To format inline LaTeX, surround it by @@ on both sides
  • To format LaTeX on its own line, surround it by $$ above and below
  • For LaTeX, it may be necessary to backslash-escape underscore characters to obtain proper formatting. So for example writing \sum\_i to represent a sum over i.
If you cannot register due to firewall issues (e.g. you cannot see the capcha box) please email Miles Stoudenmire to ask for an account.

To report ITensor bugs, please use the issue tracker.

Categories

...