While a polished slab of dielectric material is one type of dielectric mirror, we assume your question concerns mirrors fabricated by layering dielectric thin films on a substrate. These mirrors’ effects on the reflected polarization state can be relatively minor or severe, depending on factors including the dielectric stack design as well as the light’s angle of incidence, wavelength spectrum, and beam diameter. In some applications, reflection from a dielectric mirror can significantly depolarize the light, and a metallic mirror would be preferred. The source of the reflected light's polarization change is the thin-film dielectric layer structure. Some light is reflected and some is transmitted at each interface in the stack, following Fresnel’s equations. The total reflected beam is the sum, determined using thin-film interference analysis, of all of the different components of light that have reflected from the different layers. Each component in the total reflected beam has a phase shift that depends on both its phase change at reflection and the phase accumulated while propagating through the stack. Multi-layer stacks are often constructed so that different wavelengths reflect more strongly at different depths in the stack. When light is incident at an oblique angle, single-wavelength S- and P-polarized light can also be reflected at different depths. A consequence is that S- and P-polarized reflected beams of the same wavelength can be laterally shifted from one another when they exit the stack, in addition to having accumulated different phases while propagating through the stack. For many mirror designs, the lateral shift and accumulated phase differences will increase with the angle of incidence and the depth of the reflection in the stack. Both effects can influence the phase difference between S- and P-components in the reflected beam.
Assuming you are thinking of the type of dielectric mirrors that consist of thin-film multi-layers deposited on a substrate, the answer is complicated. One reason for this is that the mirror’s effect depends strongly on its precise design, which specifies parameters like thin-film compositions and layer widths. Many different designs, typically optimized for specific applications, are used. Due to this, different mirrors can have significantly different effects on the reflected light. Another reason the answer is complicated is that the mirror’s effect is also highly dependent on the illumination conditions, including angle of incidence, the bandwidth of the light, and the beam diameter. Generally speaking, especially for larger angles of incidence, beams reflected by these mirrors can exhibit wavelength-dependent and polarization-dependent lateral shifts and related effects that include beam distortion and depolarization. (See our response to the question posted by YoseobYoon for additional information.) Let us know if you would recommend this topic for a future Video Insight, and what application would be interesting to you :)
When light is normally incident, the incident and reflected rays overlap. As a consequence, it is not possible to define a plane of incidence, so S-polarized light cannot be distinguished from P-polarized light. In other words, the S- and P-polarizations are degenerate. Under this condition, mirrors are not polarizing, meaning all of the reflected light will accumulate the same amount of phase delay. Often light that is normally incident on a mirror is described as having a pi phase shift after reflection. The pi phase shift is not physical. Instead, it is a mathematical convention that is followed when using local right-hand coordinate systems to describe the incident and reflected light. Because reflection reverses the direction of the propagation vector, the math needs to add a 180° (pi) shift to the polarization orientation of the reflected light to ensure mathematical consistency.
Very nice demonstration of the phase shifts caused by metal mirrors. Thank you, Thorlabs!
We're glad you enjoyed this Insight!
What about dielectric mirrors? Is the phase shift between s and p polarizations worse than metallic mirrors?
While a polished slab of dielectric material is one type of dielectric mirror, we assume your question concerns mirrors fabricated by layering dielectric thin films on a substrate. These mirrors’ effects on the reflected polarization state can be relatively minor or severe, depending on factors including the dielectric stack design as well as the light’s angle of incidence, wavelength spectrum, and beam diameter. In some applications, reflection from a dielectric mirror can significantly depolarize the light, and a metallic mirror would be preferred.
The source of the reflected light's polarization change is the thin-film dielectric layer structure. Some light is reflected and some is transmitted at each interface in the stack, following Fresnel’s equations. The total reflected beam is the sum, determined using thin-film interference analysis, of all of the different components of light that have reflected from the different layers. Each component in the total reflected beam has a phase shift that depends on both its phase change at reflection and the phase accumulated while propagating through the stack. Multi-layer stacks are often constructed so that different wavelengths reflect more strongly at different depths in the stack.
When light is incident at an oblique angle, single-wavelength S- and P-polarized light can also be reflected at different depths. A consequence is that S- and P-polarized reflected beams of the same wavelength can be laterally shifted from one another when they exit the stack, in addition to having accumulated different phases while propagating through the stack. For many mirror designs, the lateral shift and accumulated phase differences will increase with the angle of incidence and the depth of the reflection in the stack. Both effects can influence the phase difference between S- and P-components in the reflected beam.
@@thorlabs Thank you for the detailed response. As always, your videos and replies are extremely helpful!
What happens if we use dielectric mirrors?
Assuming you are thinking of the type of dielectric mirrors that consist of thin-film multi-layers deposited on a substrate, the answer is complicated. One reason for this is that the mirror’s effect depends strongly on its precise design, which specifies parameters like thin-film compositions and layer widths. Many different designs, typically optimized for specific applications, are used. Due to this, different mirrors can have significantly different effects on the reflected light. Another reason the answer is complicated is that the mirror’s effect is also highly dependent on the illumination conditions, including angle of incidence, the bandwidth of the light, and the beam diameter. Generally speaking, especially for larger angles of incidence, beams reflected by these mirrors can exhibit wavelength-dependent and polarization-dependent lateral shifts and related effects that include beam distortion and depolarization. (See our response to the question posted by YoseobYoon for additional information.)
Let us know if you would recommend this topic for a future Video Insight, and what application would be interesting to you :)
What if the incident light is normal?
When light is normally incident, the incident and reflected rays overlap. As a consequence, it is not possible to define a plane of incidence, so S-polarized light cannot be distinguished from P-polarized light. In other words, the S- and P-polarizations are degenerate. Under this condition, mirrors are not polarizing, meaning all of the reflected light will accumulate the same amount of phase delay.
Often light that is normally incident on a mirror is described as having a pi phase shift after reflection. The pi phase shift is not physical. Instead, it is a mathematical convention that is followed when using local right-hand coordinate systems to describe the incident and reflected light. Because reflection reverses the direction of the propagation vector, the math needs to add a 180° (pi) shift to the polarization orientation of the reflected light to ensure mathematical consistency.