0
$\begingroup$

Background

I am studying the paper Lee et al (2012), located at arXiv:1207.7174. In this paper, we study the spin-boson model under a polaron transformation. The Hamiltonian is $$H = \frac{\epsilon}{2}\sigma_z + \frac{\Delta}{2}\sigma_x + \sum_k\omega_k b_k^{\dagger}b_k + \sigma_z\sum_k \omega_k (b_k^{\dagger}-b_k)$$ and the polaron operator is defined by $$F = \sigma_z\sum_k\frac{g_k}{\omega_k}(b_k^{\dagger}-b_k).$$

Problem

I've derived the transformed Hamiltonian $H' = e^FHe^{-F}$ without difficulty. However, I am having trouble calculating Eq. 6 in the paper, which defined the expectation of the bath operators $$D_{\pm} = \exp\left[\pm\sum_k \frac{g_k}{\omega_k}(b_k^{\dagger}-b_k)\right]$$ as $R = \langle D_{\pm}^2\rangle_{H_B},$ where $H_B = \sum_k \omega_kb_k^{\dagger}b_k$. If we introduce the spectral density $$J(\omega) = \pi\sum_kg_k^2\delta(\omega-\omega_k),$$ the paper claims we can write $R$ as $$R = \exp\left[-2\int_0^{\infty}\frac{d\omega}{\pi}\frac{J(\omega)}{\omega^2}\coth(\beta\omega/2)\right].$$

Attempted Solution

For simplicity, I temporarily write $D_+ = e^C$, where $C = \sum_k\frac{g_k}{\omega_k}(b_k^{\dagger}-b_k)$. Then, I can write $R$ as $$R = \langle D_+^2\rangle_{H_B} = \sum_{n=0}^{\infty}\frac{1}{n!}\langle C^n\rangle_{H_B},$$ which reduced the problem to calculating the moments of C. However, at this point I am unsure how to proceed. Any help given would be immensely appreciated.

$\endgroup$

1 Answer 1

1
$\begingroup$

if you have $A$ an operator which is linear in bosonic operators, and you want to calculate its expectation value when the bosons are non-interacting, then $$ \langle e^A \rangle = \exp\left[\langle A \rangle + \frac{1}{2}\left(\langle A^2 \rangle-\langle A \rangle^2\right) \right]$$ which is a consequence of Wick's theorem.

$\endgroup$
2
  • $\begingroup$ Thank you! Do you have a reference where I can see the full calculation? I’m still uncomfortable using Wick’s theorem. $\endgroup$ Commented Jan 8, 2020 at 15:15
  • $\begingroup$ googling, I found this: imperial.ac.uk/media/imperial-college/… (section 4). However, thinking about it you don't even need Wick's theorem: each mode $b_k$ is independent, and what you have is a coherent state (using the displacement operator en.wikipedia.org/wiki/Displacement_operator) and the matrix elements in this specific case can be calculated. $\endgroup$
    – user245141
    Commented Jan 8, 2020 at 15:28

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.