Features

• Optic Axis Alignment Not Required
• Ideal for Broadband Light Sources and Large Diameter (>6 mm) Monochromatic Beams
• Air-Gap Design Allows for use with High-Power Beams
• Available Uncoated (190 - 2500 nm) or with One of Three AR Coatings
• 350 - 700 nm (-A Coating)
• 650 - 1050 nm (-B Coating)
• 1050 - 1620 nm (-C Coating)

Thorlabs' DPU Series of Achromatic Depolarizers convert a polarized beam of light into a pseudo-random polarized beam of light. The term pseudo-random is used since the transmitted beam doesn't become unpolarized; instead the polarization of the beam is randomized. Linearly polarized light from a monochromatic source that is transmitted through a DPU depolarizer will have a polarization that varies spatially. Linearly polarized light from a broadband source that is transmitted through a DPU depolarizer will have a polarization that varies spatially as well as with wavelength. While the DPU-25 is designed for use with linearly polarized light (as shown in the Tutorial tab), it will also produce pseudo-random polarization if the input is elliptically or circularly polarized. DPU polarizers will work with light incident on either side of the optic.

Design
These achromatic depolarizers consist of two crystal quartz wedges, one of which is twice as thick as the other, that are separated by a thin metal ring. The assembly is held together by epoxy that has been applied only to the outside edge (i.e., the clear aperture is free from epoxy), which results in an optic with a high damage threshold. These depolarizers are available uncoated for use in the 190 - 2500 nm range or with one of three antireflection coatings deposited on all four surfaces (i.e., both sides of the two crystal quartz wedges). Choose from AR coatings for the 350 - 700 nm (-A coating), 650 - 1050 nm (-B coating), or 1050 - 1620 nm (-C coating) range.

The optic axis of each wedge is perpendicular to the flat for that wedge. The orientation angle between the optic axes of the two quartz crystal wedges is 45°. The unique design of the DPU series depolarizers eliminates the need to orient the optic axes of the depolarizer at any specific angle, which is especially useful if the depolarizer is used in an application where the initial polarization of the light is unknown or varies with time.

Edge on view of the DPU Depolarizer

Free-Space Applications
Although the DPU series of achromatic depolarizers can convert a polarized monochromatic or broadband source into a pseudo-random polarization, depolarization of narrowband (monochromatic) light sources can only be achieved if the incident beam is greater than 6 mm in diameter. This minimum beam diameter is needed for monochromatic sources because randomization of the output beam's polarization is achieved by producing a spatial variation in the beam's polarization (see the Tutorial tab).

Depolarization of a broadband source does not have the same beam diameter restrictions as found with the narrowband or monochromatic source. This is because the polarization of the output beam will be randomized as a result of the wavelength-dependent retardation of the light transmitted through the quartz crystal wedges, in addition to the spatial variation in the polarization of the output beam.

The pseudo-random polarization generated by these achromatic depolarizers may be more suitable than linearly polarized light for polarization-sensitive devices and experiments, such as Raman Amplification and reduction of polarization-dependent losses. Because of their effectiveness over a wide spectral range, with both narrow- and broadband sources, the DPU series of depolarizers have been used with favorable results in applications involving polarization-sensitive spectrometers and LCD test systems. In addition, the depolarizer's design eliminates the need to align the optic with respect to the orientation of the linear polarization axis of the incident beam, making them highly adaptable to varying input polarizations.

Fiber Applications
Thorlabs' achromatic depolarizers are not recommended for applications where the depolarized beam is going to be coupled into a single mode optical fiber. Consider the output beam as two superimposed beams. Since the optic axes of the two wedges are not aligned, there will be a non-zero divergence/offset of the propagation vectors of the two output beams due to the birifringence of the quartz crystal. If the diverging/offset beams are then focused onto the tip of a fiber, each beam will be imaged at a slightly different position. As a result, optimizing the coupling of light into the fiber can result in the preferential coupling of one beam over the other and the polarization of each individual beam is not completely randomized. The PL100S and DPC5500-T are fiber based polarization controllers that have a depolarization mode, which rapidly varies the output polarization in time.

The DPU-25 is also available from stock with three different AR coatings. As the plots below show, the AR coating significantly reduces surface reflections, and since the DPU series depolarizers are made from two quartz crystal wedges, the AR coating minimizes the reflections at four surfaces. For example, the loss due to surface reflections at 1400 nm for the uncoated DPU-25 is approximately 15%. However, the AR coating on the DPU-25-C reduces the loss due to reflections to approximately 2%.

Click to Enlarge
Theoretical transmission as a function of wavelength through an uncoated DPU-25 depolarizer.

Click to Enlarge
Theoretical percent reflectivity as a function of wavelength for the AR coated surfaces of the DPU-25-A depolarizer.

Click to Enlarge
Theoretical percent reflectivity as a function of wavelength for the AR coated surfaces of the DPU-25-B depolarizer.

Click to Enlarge
Theoretical percent reflectivity as a function of wavelength for the AR coated surfaces of the DPU-25-C depolarizer.

Click Here to download reflectivity data for all three AR coatings available.

Spatial Periodicity in the Polarization of the Output Beam

The spatial periodicity in the polarization of the output beam of the polarizer is easily observed when a linearly polarized monochromatic light source is used. The linearly polarized monochromatic beam must have a beam waist larger than 6 mm to generate this spatial periodicity. To observe this effect, place the DPU depolarizer in the beam so that propagation direction of the beam is normal to and incident upon the thicker wedge of the depolarizer.

To detect the beam's spatial periodicity, place a linear polarizer (analyzer) after the beam exits the DPU depolarizer. The intensity variation will be banded as shown in the false-color plots shown below. The banding occurs because the degree to which the state of polarization is rotated is identical along any line perpendicular to the thick wedge incline. Several examples of different incident and depolarizer angles are discussed below.

Example 1:

In the first example, a 635 nm, linearly polarized source with a 20 mm beam waist is used. The source light is incident on and propogating normal to the thick wedge of the depolarizer. The light is polarized in the same direction as the optic (fast) axis of the thick wedge. The optic axis of the thin wedge is at a 45^o angle with respect to both the incident light polarization and the optic axis of the thick wedge. In this orientation, the wedge incline is also oriented parallel to the incident light polarization.

The orientation of the input polarization and optic axis of the thick and thin wedges is shown in Figure 1. The input polarization is shown in red, the thick wedge optic axis in blue, and the thin wedge optic axis in green. The flat on the top of the depolarizer drawing (solid blue line) is cut perpendicular to the thick wedge optic axis. The flat on the side of the drawing (dashed blue line) is cut perpendicular to the thin wedge optic axis.

A theoretical plot showing the spatially-dependent polarization angle is shown in Figure 2. This false-color plot shows the polarization angle of the output beam from the DPU depolarizer based on the input beam characteristics described above.

Legend

The bands in the false-color plot represent different regions of identical polarization with polarizations varying from being completely along the horizontal axis (perpendicular to the input polarization) to completely along the vertical axis (parallel to the input beam polarization). The separation between the two polarizations in this case is approximately 2 mm. The orientation of the bands are always perpendicular with the thick wedge optic axis, and in this case, are also perpendicular to the input polarization.

Figure 1. Input polarization in a parallel arrangement with the optic axis of the thick wedge.

Click to Enlarge
Figure 2. False-color plot of a calculation showing a linearly polarized input beam parallel to the optic axis of the thick wedge after passing through the depolarizer.

Example 2:

In this example, the same linearly polarized source is used, except the polariation is oriented parallel to the optic axis of the thin wedge and at a 45 degree angle from the optic axis of the thick wedge. The orientation of the input polarization and optic axes of the thick and thin wedges are shown in Figure 3. The input polarization is shown in red, the thick wedge optic axis in blue, and the thin wedge optic axis in green. The flat on the top of the depolarizer drawing (solid blue line) is cut perpendicular to the thick wedge optic axis. The flat on the side of the drawing (dashed blue line) is cut perpendicular to the thin wedge optic axis.

The calculated result for the source and orientation passing through the DPU depolarizer is shown in Figure 4. As described in Figure 2, the bands in the false-color plot represent the the spatially-dependent regions of identical polarization. Note the orientation of the banding is perpendicular to the optic axis of the thick wedge (and wedge incline as well). The orientation of the input beam has no effect on the orientation of the banding.

Rotating the input polarization only affects the specific polarization of the band. This is easily observed by comparing the bands at 0 mm on the vertical axis in Figs. 2 and 4. In Figure 2, the band at 0 mm has a 45^o orientation while in Figure 4, the band at 0 mm has a 90^o orientation.

Figure 3. Input polarization in a parallel arrangement with the
optic axis of the thin wedge.

Click to Enlarge
Figure 4. False-color plot of a calculation showing a linearly polarized input beam parallel to the optical axis of the thin wedge after passing through the depolarizer.

Example 3:

In this last example, the input beam has the same characteristics as in Example 1. However, the DPU depolarizer is rotated 45^o so the input polarization is parallel to the optic axis of the thin wedge. This arrangement is shown in Fig. 5. The input polarization (red arrow) is still incident on the thick wedge; however, it is now perpendicular to the thin wedge flat.

The false-color plot in Fig. 6 shows the calculation result of the input beam described above passing through the DPU depolarizer. The orientation of the banding is clearly different from the banding in Figs. 2 and 4. The 45^o tilt in the pattern is because the banding occurs perpendicular to both the optic axis of the thick wedge and wedge incline.

Figure 5. Input polarization in a parallel arrangement with the
optic axis of the thin wedge and 45^o to the wedge plane.

Click to Enlarge
Figure 6. False-color plot of a calculation showing a linearly polarized input beam parallel to the optic axis of the thin wedge after passing through the depolarizer at a 45^o angle to the wedge plane.