collapse operator or lindblad dissipator

[4psireal2]

I'm doing a simulation of a dissipative chain. Based on the documentation , c_ops is simply a list of C_n. I also came across the lindblad_dissipator function. I tested out the 2 approaches and the results are different.

dissipation_op_1 = []
for i in range(L):
	dissipation_op.append(self.gamma * lindblad_dissipator(a=destroy_list[i]))

dissipation_op_2 = []
for i in range(L):
	dissipation_op_2.append(self.gamma * destroy_list[i])

result_1 = mesolve(H, state0, tlist=self.time, c_ops=dissipation_op_1, options=options, progress_bar=True)
result_2 = mesolve(H, state0, tlist=self.time, c_ops=dissipation_op_2, options=options, progress_bar=True)

It seems to me that dissipation_op_1 is the way to define c_ops for wavefunction method while dissipation_op_2 is for density matrix. Any comments / feedbacks are very much appreciated.

[nwlambert]

The issue is probably because method 2 requires you to take the square root of the rate gamma, see just below equation 3 in the documentation and the example code there.

The logic is that the ''collapse operator'' in that approach is actually c = sqrt(gamma) * a, which will give the lindblad dissipator equivalent to method 1 when the construction of the dissipator is made internally in mesolve(), i.e., c \rho c^{\dagger}+... = gamma a \rho a^{\dagger}....