Configuration of either a transmitted or reflected optical microscope for operation in differential interference contrast (DIC) using a de Sénarmont compensator offers far more latitude and accuracy for the introduction of bias retardation than is possible with systems that rely on translation of the objective Nomarski (or Wollaston) prism across the optical pathway. Virtually any microscope that contains polarizing elements and the necessary condenser and objective beamsplitting compound prisms can be easily converted for operation in de Sénarmont mode, regardless of whether the microscope was originally designed for this purpose.
Several of the major microscope manufacturers are now producing DIC accessory kits for their research-level microscopes that contain the necessary components for using a de Sénarmont compensator instead of a translatable objective prism to introduce bias retardation into the wavefront field. The basic de Sénarmont DIC microscope optical train is presented in Figure 1 for a transmitted light microscope. Semi-coherent bundles of non-polarized white light emitted from localized neighborhoods on the lamp filament (the source is usually a 100-watt tungsten-halogen bulb) first pass through a linear polarizer and quarter-wavelength retardation plate combined together in a de Sénarmont compensator housing, which is attached to the illumination port at the base of the microscope. Linear, elliptical, or circularly polarized light exiting the de Sénarmont compensator next passes through the condenser Nomarski (or Wollaston) beamsplitter prism, where it is first sheared into orthogonal components, and then rendered parallel by the condenser lens system. After passing through the specimen and its surrounding medium, light is gathered by the objective and focused at the interference plane of a second Nomarski prism located above the objective thread seat in the microscope nosepiece. Orthogonal wavefronts are recombined by the objective Nomarski prism, and travel up the optical train to the analyzer, which passes only those components parallel to the transmission azimuth. The parallel wavefronts admitted through the analyzer are able to undergo interference to produce an image that can be observed or recorded by a detector.
In the configuration illustrated in Figure 1, the de Sénarmont compensator is designed to fit and secure tightly onto a mounting bracket positioned inside the large circular knurled wheel that surrounds the field lens and controls the field diaphragm iris size. Mounted above the polarizer in the de Sénarmont compensator housing is a quarter-wavelength retardation plate positioned with the fast optical axis oriented East-West in relation to the microscope frame. In addition, the retardation plate is tilted off-axis by a few degrees from the perpendicular to the optical path in order to reduce reflections, and has a multiple coating of antireflection thin films applied to the upper surface.
The polarizer is contained in a rotatable cylinder sandwiched between the light port housing skirt and the fixed retardation plate (see Figure 2). When the polarizer transmission axis is aligned parallel to the fast axis of the retardation plate, no optical path difference is added to the sheared wavefronts traveling between the condenser and objective prisms (in effect, there is no bias retardation because linearly polarized light exits the compensator). However, the cylinder enclosing the polarizer is designed to be rotated around the microscope optical axis by approximately 180 degrees (plus or minus 90 degrees from the parallel alignment of the polarizer and the retardation plate fast axis) in order to introduce elliptically or circularly polarized light into the optical system. The exact degree of rotation is indicated on some de Sénarmont compensator housings by a graduated scale that contains a zero position (polarizer aligned parallel to the retardation plate) in the center, with linear graduations extending about 45 degrees in both the left and right orientations (as illustrated in Figure 2). Thus, positive and negative bias retardation can be introduced into a de Sénarmont DIC microscope having this type of compensator simply by rotating the polarizer back and forth within its range.
Upon initial installation, the de Sénarmont compensator illustrated in Figure 2 is first aligned with the polarizer transmission azimuth and retardation fast axis oriented in the proper position (East-West) and crossed with respect to the transmission axis of the analyzer (which is oriented North-South). After alignment, the compensator is secured to the light port knob with a locking setscrew. The polarizer axis orientation is marked on the front of the de Sénarmont compensator unit, as described above, and the graduated ruling enables the operator to qualitatively determine the approximate amount of bias retardation introduced into the system when the polarizer is rotated. A locking knob (see Figure 2) can be utilized to hold the polarizer immobile with respect to the quarter-wavelength plate. For brightfield or enhanced contrast techniques without differential interference contrast, the entire polarizer and retardation plate assembly can be removed from the optical path by swinging out the hinged upper section.
Polarized light having linear, elliptical, or circular characteristics, depending upon the polarizer orientation with respect to the retardation plate, exits the de Sénarmont compensator and next encounters a Nomarski or Wollaston prism positioned within the condenser turret (Figures 1 and 3). DIC condenser prisms, which act as beamsplitters to produce an angular shear to incoming polarized wavefronts, have a similar design characteristics regardless of whether they are intended for use with de Sénarmont compensators or translatable objective prisms. These prisms are usually mounted in a revolving turret designed to house at least three or four individual prism units, similar to the model illustrated in Figure 3. Turret specifications and configurations vary according to the manufacturer, but they generally contain slots for four to eight auxiliary components, including Wollaston or Nomarski prisms, phase contrast annular rings, Hoffman modulation contrast slits, or darkfield light stops. The condenser turret illustrated in Figure 3 contains seven openings, three of which are filled with phase contrast annuli and three with DIC Nomarski prisms. The open slot is employed for brightfield observations.
Each condenser DIC prism (which are also termed compensators or auxiliary prisms) must be specifically matched to a narrow range of objective numerical apertures, so a particular prism may only work for one or two objectives (for example, the 20x and 40x). As a result, between three and five condenser prisms must be utilized to match the entire objective magnification range between 10x and 100x in a typical compound microscope. Some manufacturers design each condenser prism specifically for a particular objective, thus requiring up to seven condenser prisms to span the entire spectrum of dry and oil immersion objectives having varied numerical apertures. The Nikon condenser (Nomarski) prism specifications for de Sénarmont DIC microscopy are catalogued in Table 1, including the prism and objective identification letters, color codes, shear distances, and numerical aperture ranges.
Condenser DIC prism inserts are constructed with anodized circular aluminum or brass plates having the combination prism (usually oval in shape) secured by optical cement with the shear axis positioned in a fixed, precise orientation. DIC prism wedges are very thin and cut with close tolerances to ensure that angular shear values match those required by the objective numerical aperture. The polished plates must be handled carefully to avoid contamination from fingerprints, oil, dirt, and debris. Each prism frame contains a positioning slot or pin that mates to a corresponding partner in the condenser turret in order to define and secure alignment of the condenser prism with respect to the objective prism and the axes of the polarizer, analyzer, and the de Sénarmont compensator retardation plate. After the prisms are inserted into the condenser turret, they are held firmly in place with a spring or a locking screw. Magnetic or polymer labels with adhesive backings are provided with most universal condensers to allow identification of components in the turret openings after the unit has been assembled.
Universal turret condensers are usually provided with a top lens swing-out mechanism that enables the condenser to be employed with both high (10x through 100x) and low (2x through 5x) magnification objectives. The top lens is placed into the optical path by pulling or pushing the swing lever for the higher magnifications, and removed from the path for use with the lower magnification objectives. A majority of the turret condenser top lens swing arms are threaded to accept lens assemblies designed for both dry and oil immersion objectives. Typically, a dry condenser top lens element will have a numerical aperture value ranging between 0.75 and 0.90, while the corresponding assembly designed for use with oil has a much higher numerical aperture (1.3 to 1.4). Often, there is not enough clearance between the condenser and the microscope stage to allow the swing mechanism to operate properly when the oil top lens element is inserted.
Table 1 - Nikon Condenser Nomarski Prism Specifications
|Dry||DIC L||Less Than 0.5||0.6||Green|
|Dry||DIC M||0.5 to 1.0||0.3||Green|
|Dry||DIC H||1.0 or More||0.15||Green|
|Oil||DIC L||Less Than 0.5||0.6||Yellow|
|Oil||DIC M||0.5 to 1.0||0.3||Yellow|
|Oil||DIC H||1.0 or More||0.15||Yellow|
Sheared wavefronts exiting the condenser prism are rendered parallel by the condenser optical system and pass through the specimen before being intercepted by the objective front lens element. The optical path difference between orthogonal wavefronts passing through the system (bias retardation) is pre-determined by the position of the polarizer transmission axis with respect to the retardation plate fast axis in the de Sénarmont compensator before light enters the condenser. After experiencing wavefront field distortions by the specimen, light is gathered by the objective optical system and focused onto the interference plane (conjugate to the objective rear focal plane) of a second Nomarski or Wollaston prism positioned above the objective.
In a properly aligned DIC microscope (irrespective of the mechanism utilized to introduce bias retardation), the condenser prism is imaged onto the objective prism by the combined action of the condenser and objective lens systems. As a result, the wavefront shear produced by the condenser prism is exactly matched at every point along the surface of both prisms, which are inverted with respect to one another. Rotating the polarizer in a de Sénarmont compensator produces a wavefront mismatch (bias retardation), which, in turn, introduces an optical path difference that is uniform across the microscope aperture.
In DIC microscopes designed to introduce bias retardation using a de Sénarmont compensator, the objective Nomarski (or in some cases, Wollaston) prisms are secured with precise orientation in fixed mounts that slide into the nosepiece above the objectives, as illustrated in Figures 1 and 4. The Nomarski prisms presented in Figure 4 were imaged between crossed polarizers to illustrate the relative sizes of interference patterns as a function of shear distance (see Table 1). Microscopes using de Sénarmont compensators require a separate objective prism for each objective, but can often utilize the same condenser Nomarski prism for two or more objectives, as described above. The fixed objective prism frames (Figure 4) are easily removed from the optical path for observation in other contrast enhancing modes by sliding the frame out of the slot and away from the microscope nosepiece.
After exiting the objective Nomarski prism, recombined wavefronts next encounter the analyzer (a second polarizer), which is usually placed in a convenient location between the microscope nosepiece and the observation tubes. The analyzer serves to pass wavefront vector components that are parallel to the transmission azimuth and capable of constructive and destructive interference to form a DIC image at the fixed eyepiece diaphragm or camera projection lens. Several microscope nosepiece designs incorporate a slot for insertion of a simple fixed-position linear analyzer oriented North-South with respect to the microscope frame and the polarizer (illustrated in Figure 5(b)). In this configuration, the analyzer is positioned with the transmission axis perpendicular to that of the polarizer (to yield crossed polarizers). Other microscopes utilize fixed analyzers that are inserted into the optical train throughout a number of locations, including intermediate tubes and vertical illuminators (for reflected light configurations). Fixed analyzers are mounted in rectangular frames that can be easily installed or removed from the microscope optical pathway to allow rapid changeover of imaging modes.
Several analyzer designs feature the same frame style (see Figures 5(a) and 5(b)), but enable rotation of the analyzer element by means of a thumbwheel that is often graduated in 5, 10, 45, or 90-degree increments (Figure 5(a) presents an example). Polarized light microscopes equipped for DIC observation often house the analyzer in an intermediate tube (Figure 5(c) and Figure 6) located between the objective nosepiece and the observation tubes. These units are often designed for precision measurements in polarized light and feature 360-degree graduated vernier scales wrapped around the circumference of the tube (Figure 5(c)), or adjacent to a graduated thumbwheel (Figure 6). In addition, a locking mechanism is included to tightly secure the analyzer in the desired transmission azimuth. In addition, the analyzer is usually mounted on a slider so that it can be conveniently removed from the light path for linearly polarized or brightfield observation. Intermediate tubes for polarized light and DIC microscopy also usually contain a 20 x 6 millimeter DIN standard slot for a quarter-wavelength, full-wave, or de Sénarmont compensator (Figures 5(c) and 6), although several manufacturers use proprietary slot sizes.
An exact determination of the amount of bias retardation introduced to the wavefront field is very difficult to determine in a differential interference contrast microscope that employs a translating Nomarski prism in the objective rear focal plane (traditional Nomarski DIC). For quantitative estimates, Nomarski prism frames having highly accurate micrometer-controlled translation devices have been developed, and these components can be utilized to accurately gauge the amount of bias introduced by moving the prism in small, recorded increments. A more precise determination (approaching one-hundredth of a wavelength) can be made by utilizing a graduated analyzer (or polarizer) coupled to a vernier scale (Figure 5(c) and 6) and a quarter-wavelength retardation plate in a de Sénarmont DIC configuration.
For quantitative determination of bias retardation values, a configuration similar to the one illustrated in Figure 6 is ideal. The intermediate tube presented in Figure 6 is designed to be installed between the objective Nomarski prism in the nosepiece and the microscope observation tubes. Inserted into the lower slot of the tube is a standard 550 nanometer de Sénarmont compensator (originally designed for polarized light measurements) housed in a fixed frame with the fast axis parallel to the polarizer (East-West) and the slow axis parallel to the analyzer (North-South). Above the de Sénarmont compensator is a rectangular frame containing a 360-degree adjustable linear polarizing element (acting as the analyzer), and graduated in single degrees of rotation. The circular control knob is stationed adjacent to a vernier scale that enables accurate determination of the polarizer transmission azimuth orientation, and provides the microscopist with the ability to control the level of bias retardation in the optical system to a few fractions of a wavelength.
Intermediate tubes containing a graduated analyzer and a de Sénarmont compensator can be retrofitted onto microscopes originally designed to introduce bias retardation by translation of the objective Nomarski prism (on a sliding frame). After initially configuring the optical system by matching the objective and condenser prisms for maximum overlap of the interference fringes (maximum extinction), the analyzer can be rotated to introduce bias retardation. In fact, microscopes already fitted with the original equipment de Sénarmont compensator (similar or identical to the model presented in Figure 1) can also be configured with the intermediate tube illustrated in Figure 6 (or Figure 5(c)) for quantitative determination of bias retardation. It is important to remove the fixed analyzer if an intermediate tube containing another analyzer is added to the optical system.
In summary, bias retardation in de Sénarmont DIC microscopy can be qualitatively controlled by a simple rotating polarizer unit, as presented in Figures 1 and 2, or with the more quantitative intermediate tubes illustrated in Figures 5(c) and 6. It is not necessary to alter the wavefront field relationship (linear, elliptical, or circularly polarized light) in differential interference contrast prior to entry into the condenser prism. The same effect can be achieved by changing the optical path difference using a quarter-wavelength retardation plate and rotating analyzer after the wavefronts have emerged from the objective prism. In fact, bias retardation can be introduced anywhere in the DIC optical system, provided the change is made somewhere between the polarizer and analyzer, and the proper components are employed with the correct orientation. Traditional DIC microscope designs rely on translation of the objective, or in rare cases, the condenser prism. Newer de Sénarmont compensator microscopes employ a rotating polarizer and a retardation plate near the field lens (primarily for ergonomic measures) to achieve the same effect.
Modern polarizing and DIC microscopes position the polarizer and analyzer in strategic locations with respect to the field lens, condenser, objective, and observation tubes. In older microscopes, these polarizing elements can be found installed in a wide variety of locations. It should be noted, however, that placing polarizing elements in or very near a conjugate image plane (the field diaphragm, specimen stage, or eyepiece fixed aperture) is not a good idea, because scratches, imperfections, dirt, and debris on the glass or polymer surface can be imaged along with the specimen.
de Sénarmont DIC for Inverted Microscopes
Inverted transmitted light (tissue culture) microscopes are often equipped with DIC optical components to aid in the visualization, photography, and digital imaging of a variety of transparent specimens, including living cells, embryos, and tissue slices. Both traditional and de Sénarmont DIC optical systems have been adapted for use with inverted microscopes. The major difference between DIC configurations in upright and inverted microscopes is the condenser system, which generally requires specialized long working distance optical components for the inverted instruments. In many cases, inverted microscope objectives must also be designed to operate with long or extra-long working distances to match condenser aperture illumination cones.
The most popular inverted microscope configuration places the objective revolving nosepiece beneath the mechanical stage, where it is attached to the main body through an intermediate tube or direct port to the main internal optical train. In some models, fixed Nomarski DIC prisms on slider frames (see Figure 4), which are identical to the prisms used in upright microscopes, can be inserted into the nosepiece. Other microscopes employ a single sliding Nomarski prism, mounted in a long rectangular frame, which is used for all objectives (10x through 100x; traditional Nomarski DIC). The Nomarski prism frame is mounted beneath the nosepiece in an intermediate tube that enables the beamsplitter to be easily removed from the optical pathway for brightfield, phase contrast, Hoffman modulation, or other imaging modes. Bias retardation is introduced by means of a control knob positioned at the end of the prism frame. In the latter configuration, a single Nomarski prism serves to recombine wavefronts for all of the objectives in the nosepiece, although many microscope designs require axial relocation of the prism at higher magnifications to accommodate variations in the location of the objective rear focal plane.
A typical inverted microscope designed for differential interference contrast using a de Sénarmont compensator is presented in Figure 7. Semi-coherent light waves emitted by the tungsten-halogen lamp on the illumination pillar first pass through a cascade of filters (neutral density, color balancing, and interference) before being deflected by a right-angle prism through the field lens and into the condenser system. Attached on top of the condenser is a de Sénarmont compensator (see Figure 8) consisting of a linear polarizer and a quarter-wavelength retardation plate. After light passes through the de Sénarmont compensator, it is focused at the condenser aperture iris diaphragm (site of the condenser focal plane), which is conjugate with the interference plane of a Nomarski prism located in the condenser turret (Figure 8). Sheared wavefronts are focused parallel to each other by the condenser lens system and illuminate the specimen before being collected by the objective front lens, which is positioned beneath the specimen stage.
A second Nomarski prism (illustrated in Figure 7) is housed beneath the objective in the microscope nosepiece, and serves to recombine sheared wavefronts after they are deformed by the specimen and focused by the objective. The recombined waves then pass through a second polarizer (the analyzer; see Figure 7) located in a sliding rectangular frame and inserted into an intermediate tube or slot between the nosepiece and the microscope base. After leaving the analyzer, the wavefronts (which are now capable of undergoing destructive and constructive interference) traverse through the microscope internal optical train to form an image at the fixed diaphragm of the eyepieces or the projection lens of a traditional or digital camera system.
As discussed above, microscopes equipped for DIC with de Sénarmont compensators have fixed objective Nomarski prisms mounted into frames similar to those illustrated in Figure 4. Bias retardation is introduced into the optical system by rotating the polarizer transmission azimuth on the de Sénarmont compensator (Figure 8) attached to the condenser. The Nomarski condenser prisms are mounted in a turret, similar to the design for upright microscopes (see Figure 3), but with a different frame design. As is the case for upright microscopes, the condenser Nomarski prisms are specifically matched to a narrow range of objective numerical apertures, so several prisms must be employed to cover the entire magnification range (Table 1).
Condenser prisms (usually circular in shape) are mounted with a precise orientation of the shear axis, using optical cement, into pie-shaped aluminum wedges that bolt into the condenser turret. Analogous to upright microscope condenser prisms, the inverted microscope Nomarski prisms are very thin and cut with close tolerances to ensure that angular shear values match those required by the objective numerical aperture. The condenser turret can accept between three and five wedges, which is enough to cover the entire microscope magnification range. In addition to Nomarski prisms, the condenser turret can accommodate phase contrast annulus plates and Hoffman modulation contrast slit plates. Any combination of phase annuli, slit plates, and Nomarski prisms can be utilized in the condenser.
First-Order Compensation Plates
Differential interference contrast microscopy with de Sénarmont compensators enables the introduction of bias retardation values between plus and minus one-half wavelength. In order to increase the path difference between sheared wavefronts, a full-wave retardation plate can be added to the optical system (yielding retardation values up to 1.5 times the wavelength of green light). Compensating plates bestow greater control for adjusting the contrast of specimen details in relation to the background intensity and color values, and also enable more precise tuning of the bias retardation between wavefronts. In addition, full-wavelength compensators are frequently employed for optical staining of thicker transparent specimens, which are normally imaged over a limited range of grayscale intensities.
On upright or inverted microscopes equipped with a de Sénarmont compensator for DIC observations, a full-wave retardation plate can be inserted into the optical pathway between either the objective prism and the analyzer or the de Sénarmont compensator and the condenser prism. Upright microscopes usually have a slot beneath the condenser or in an intermediate tube above the objectives, which is designed to receive a full-wave retardation plate. Inverted microscopes are more restricted (due to design constraints) in the placement of auxiliary compensators. The inverted microscope condenser illustrated in Figure 8 can accept a full-wavelength retardation plate in a slot beneath the de Sénarmont compensator housing, but does not have a provision for introducing auxiliary compensators between the objective and analyzer. A majority of the inverted microscope condensers offered by other manufacturers are similar in design.
In conclusion, de Sénarmont compensators enable more control of bias retardation in DIC microscopy than do the traditional Nomarski objective prism designs that rely on translation of the prism across the microscope optical axis. By coupling a graduated polarizer (or analyzer) to a fixed quarter-wavelength retardation plate, the amount of bias retardation introduced into the optical system can be determined with a high degree of precision. Finally, microscope designs that incorporate a de Sénarmont compensator near the field lens at the base of the instrument have ergonomic advantages over traditional configurations.
Douglas B. Murphy - Department of Cell Biology and Anatomy and Microscope Facility, Johns Hopkins University School of Medicine, 725 N. Wolfe Street, 107 WBSB, Baltimore, Maryland 21205.
Stanley Schwartz - Bioscience Department, Nikon Instruments, Inc., 1300 Walt Whitman Road, Melville, New York 11747.
Edward D. Salmon - Department of Cell Biology, The University of North Carolina, Chapel Hill, North Carolina 27599.
Kenneth R. Spring - Scientific Consultant, Lusby, Maryland, 20657.
Michael W. Davidson - National High Magnetic Field Laboratory, 1800 East Paul Dirac Dr., The Florida State University, Tallahassee, Florida, 32310.