Do slits absorb photons at the edges?

[Swamp Thing]

Xiao et al say the following here: https://arxiv.org/pdf/1703.03818.pdf

When a spatial mask such as an iris is applied to the Gaussian beam, the quantum states of different modes
will interact. The interaction can be described as a transformation of annihilation or creation operators of input
modes into operators of output modes. This transforma-tion should be unitary, which preserves the commutation relations of the annihilation or creation operators. However, one problem needs to be solved. Generally, spatial masks (or other optical devices) are lossy. For example, an iris will absorb part of the input signal at the rim.

Is this correct? I'm finding it hard to believe that there is a probability that some irreversible events will occur that would ultimately result in heating up the mask (screen with slits). I mean, we could always make the screen out of highly reflective material that would either reflect most photons back towards the source, or let some pass with a small amplitude of detection on the other side.

But a reflective screen actually absorbing photons? With the absorption events clustured near the edges?

 

[DrClaude]

There is no such thing as a perfectly reflecting material. Some of the photons will ultimately be absorbed. In any case, I'm not sure why this would matter here: reflected photons will also affect u_output.