Phase Contrast Illumination
Gregor T. Overney writes about his experiences with phase contrast microscopes in this article. He supplements his text with informative pictures and images that help the reader understand and make comparisons. He touches a little on the history of phase contrast microscopes. He pays tribute to the man who first described the method in 1934, a Dutch physicist called Frits Zernike. This invention was so highly regarded and paved the way for so many discoveries that Zernike was awarded a Nobel Prize for it. Phase contrast microscopes allow for the clear and detailed visualization of otherwise invisible microorganisms in their own habitat. When compared to bright field microscopy, which requires the specimens to be stained, often killing and immobilizing them. With phase contrast microscopy, living cells can be observed directly without any preparation at all. The author explains that phase contrast microscopy is made possible by a phase ring and a corresponding annular ring. He subsequently provides an illustration of the illuminator, condenser and objective. The rest of the mechanism is also aptly described, explaining the direction of the light rays and its effect on the image. For further understanding of the mechanics of the phase contrast microscope, a figure showing the condenser annulus and the phase plate from a different angle is shown. He also describes in depth how this works with illustrations and images of the wave fronts. The author admits to comparing bright field illumination with a green interference filter and a dark field illumination with a blur filter and phase contrast illumination without green interference filter with images from Stauroneis phoenicenteron. He highlights the results in images called photomicrographs he took himself. He then concludes that using phase contrast microscopy is useless if the specimens require contrast using bright field illumination. Phase contrast allows for the limitation because of the condenser annulus. He then continues to compare epithelial cell images from dark field microscopes, bright field microscopes and phase contrast microscopes. He highly recommends the use of phase contrast illumination for studies and researches involving protests. He suggests reading up on interesting articles that have previously been published in Micscape Magazine about phase contrast methods. In summary, the phase plate decreases the S-wave amplitude by a lot. The more this is decreased, the more contrast is possible for phase contrast illumination. But as the reduction is enhanced, so is the softening of the image in bright field. Manufacturers have been alerted to this limitation and thus they have created phase contrast lenses that are more suitable for bright field. Although this may decreased the contrast of phase contrast, the use is highly optional. The author further supplements his point by comparing resolution qualities between phase contrast at 40x objective and a bright field illumination at 40x objective. A stained histology in bright field is also compared to a “high contrast” phase contrast objective with accompanying images. If there are no phase contrast microscopes available in the laboratory, a bright field microscope can be modified to do the tasks.
ORIGINAL TEXT: The topic of phase contrast microscopy is well explained in good textbooks and on many websites (such as [1]). The purpose of this paper is not to write another introduction to phase contrast illumination. This article shall merely be about some of my experiences I collected using this interesting type of illumination.
Phase contrast microscopy was invented in 1934 by Dutch physicist Frits (Frederik) Zernike (1888 - 1966). The main aspect of this invention is to convert phase differences into amplitude variations that can easily be detected. For instance, phase contrast (PC) microscopy can be used to produce high-contrast images of transparent specimens, such as living epithelial cells. It is an interference technique that requires at least partially coherent light to illuminate the specimen (more precisely, partial, longitudinal coherence is required to make PC work).
The numerical aperture (NA) of the condenser is reduced by the geometrical arrangement of the annular ring. However, this reduction in the condenser’s NA does in no way diminish the benefit of phase contrast illumination for transparent specimens that mainly change the phase of the diffracted light beam rather than its amplitude.
Some Details about Phase Contrast Illumination A practical implementation of PC illumination consists of a phase ring (located in a conjugated aperture plane somewhere behind the front lens element of the objective) and a matching annular ring, which is located in the primary aperture plane (location of the condenser’s aperture).
The figure to the left shows a cross section of the illuminator, condenser and objective. Two selected light rays (indicated by blue lines), which are emitted from one point inside the lamp’s filament, get focused by the field lens exactly inside the opening of the condenser annular ring. Since this location is precisely in the front focal plane of the condenser, the two light rays are then refracted in such way that they exit the condenser as parallel rays. Assuming that the two rays in question are neither refracted nor diffracted in the specimen plane (location of microscope slide), they enter the objective as parallel rays as illustrated in this figure. Since all parallel rays are focused in the back focal plane of the objective, the back focal plane is a conjugated aperture plane to the condenser’s front focal plane (also location of the condenser annulus). To complete the phase setup, a phase plate is positioned inside the back focal plane in such a way that it lines up nicely with the condenser annulus. Since this figure shows the condenser annulus and the phase plate from the side, the circular shape of the phase ring and condenser annulus is not easily recognized. The figure below shows the condenser annulus (left) and the phase plate (right) from a more appropriate angle.
Although the figure to the left clearly shows that the phase plate is located in the back focal plane of the objective, the location of the phase plate is usually at a suitable location inside the objective. What is important is that the phase plate and the condenser annular are located in conjugated aperture planes. (Conjugated planes are perpendicular to the optical axis.)
Only through correctly centering the two elements, phase contrast illumination can be established. A phase centering telescope that temporarily replaces one of the oculars is used to center the annular ring with the ring of the phase plate.
To get a better understanding of how phase contrast illumination works, we study two wave fronts (see the figure to the right). This figure simplifies a few things. First, the condenser annulus is just a small aperture located in the center (see the plane labeled ‘1') and the phase plate is also just covering a small aperture (located in the plane labeled ‘3'). Second, the optical system is greatly simplified by showing only two single lenses to represent all optical elements.
The plane labeled ‘1' is the front focal plane of the condenser. The light emanating from the small aperture ‘S’ is captured by the condenser and emerges as light with only parallel wavefronts from the condenser. When these plane waves (parallel wave fronts) hit the phase object ‘O’ (located in the object plane labeled ‘2'), some of this light is diffracted (and/or refracted) while moving through the specimen. Assuming that the specimen does not significantly alter the amplitudes of the incoming wavefronts but mainly changes phase relations with respect to the “unperturbed” wavefronts, newly generated spherical wave fronts that are retarded by 90° (?/4) emanate from ‘O’ (see the purple area that contains now “unperturbed” plane waves and spherical wave fronts). It is important to note that there are now two types of waves, the surround wave or S-wave and the diffracted wave or D-wave, which have a relative phase-shift of 90° (?/4). - The objective focuses the D-wave inside the primary image plane (labeled ‘4'), while it focuses the S-wave inside the back focal plane (labeled ‘3'). The location of the phase plate ‘P’ has now a profound impact on the S-wave while leaving most of the D-wave “unharmed”. In what is known as positive phase contrast optics, the phase plate ‘P’ reduces the amplitude of all light rays traveling through the phase annulus (mainly S-waves) by 70 to 90% and advances the phase by yet another 90° (?/4). However, the phase plate leaves most of the D-waves “untouched”. Hence the recombination of these two waves (D + S) in the primary image plane (labeled ‘4') results in a significant amplitude change at all locations where there is a now destructive interference due to a 180° (?/2) phase shifted D-wave. The net phase shift of 180° (?/2) results directly from the 90° (?/4) retardation of the D-wave due to the phase object and the 90° (?/4) phase advancement of the S-wave due to the phase plate. Without the phase plate, there would be no significant destructive interference that greatly enhances contrast. With phase contrast illumination “invisible” phase variations are hence translated into visible amplitude variations. The destructive interference is illustrated in the figure to the right. Blue and orange indicate D-wave and S-wave, respectively. The resulting wave (D + S), indicated by yellow, has a reduced amplitude.
Comparison of Various Illumination Techniques I compared brightfield (BF), BF with a green interference filter, circular oblique lighting (COL) ([2]), darkfield (DF) ([3]), DF with a blue filter, and phase contrast without green interference filter using images taken from Stauroneis phoenicenteron. I used a Plan Achromat 40x objective with NA 0.65 for all DF work. I used a Plan Fluor 40x objective with NA 0.75 for BF and COL and a DL Plan Achromat 40x objective with NA 0.65 for phase contrast. The results are depicted in the figure below. When taking these photomicrographs, I carefully focused on just one row of dots located towards the middle of this diatom.
Diatom Stauroneis phoenicenteron (click on image for larger version)
From the above data, it is perfectly obvious that the use of phase contrast illumination is rather pointless for looking at specimens that offer sufficient contrast in brightfield when trying to maximize contrast as well as resolution. While DF and COL offer at least the same resolution, phase contrast limits the resolution due to the condenser annulus.
However, this changes dramatically if the specimen is flat and appears rather transparent in brightfield. While DF and COL often offer enough contrast, BF is not of much help. The picture below shows an epithelial cell in BF using a Plan Fluor 40x lens (NA 0.75) (left) and with phase contrast using a DL Plan Achromat 40x (NA 0.65) (right). A green interference filter is used for both images.
Unstained epithelial cell in BF (left) and in PC (right)
When using COL and DF, much more contrast can be obtained as can be seen in the picture below. The first image shows an epithelial cell in COL using a Plan Fluor 40x objective (NA 0.75); the second one shows the same cell with phase contrast illumination using a DL Plan Achromat 40x (NA 0.65); and the third image shows again the same cell in DF using a DL Plan Achromat 40x (NA 0.65). A green interference filter is used for all cases.
Unstained epithelial cell in COL (left), PC (middle) and DF (right) (click on image for larger version)
Phase contrast is most ideal for this application of thin, transparent samples such as epithelial cells. - When looking at the pictures above, some of the cellular features appear dark in front of a brighter background when observed in PC. This is usually the case when the S-wave is phase shifted by +90° (known as positive phase contrast). Essentially, objects with a higher refractive index than the surrounding medium appear dark (assuming, of course, the same thickness). - Lipid droplets and vacuoles in plant cells and protozoa usually appear brighter in the surrounding cytoplasm when observed with positive phase contrast.
When looking at the images in phase contrast depicted above, we can also easily find certain image artifacts due to the effects of halo and shade-off. Halos occur because the ring in the phase plate (”located” in the back focal plane of the objective) also receives some diffracted light from the specimen. To alleviate this problem, the phase plate annulus (or its image) is wider than the condenser annulus when both are observed with a phase centering telescope. Shade-off, another optical effect, is more obvious in extended phase objects.
Phase contrast illumination is recommended by many microscopists for the study of protists (for instance, see [4]). - Many interesting articles have been published in the Micscape Magazine about this contrast method (for example, see [5-10]).
Phase Contrast Objective Used for Brightfield In the last paragraph, we look at the performance of a phase contrast objective for use in brightfield illumination. Many amateur microscopists find themselves in a financial dilemma. Shall they purchase both, a dedicated phase contrast objective and a general purpose objective for brightfield (as well as COL and DF) or can a phase contrast lens reasonably well be used as general purpose objective lens?
Let us focus on positive phase contrast. As mentioned above, the phase plate reduces the amplitude of the S-wave (non-diffracted wave) by a significant amount. The stronger this reduction, the more contrast one obtains with PC illumination. However, the stronger this reduction, the more is also the image softened in brightfield. Interesting, the manufactures have recognized this limitation and developed phase contrast lenses with phase plates that do no longer reduce the amplitude this significantly. Unfortunately, this is just a compromise that reduces contrast with PC while making these lenses more suitable for brightfield. Most makers of Plan Fluor lenses have followed this example to lower the contrast of phase contrast while allowing the user to use his lens for better quality work in brightfield or fluorescence microscopy. If the goal is to maximize contrast with phase contrast illumination, a dedicated phase contrast lens should be purchased.
The following is a comparison between a “high contrast” Plan Achromat 40x phase contrast lens and a Plan Achromat 40x lens. As we can see from the image below, there is no difference in resolution.
Resolution comparison between a phase contrast 40x objective (right) vs. a BF 40x objective (left) in brightfield. Both objectives have a numerical aperture (NA) of 0.65.
This summarizes the results when looking at diatom frustules. Even a “high contrast” Plan Achromat phase contrast lens is perfectly suitable for this kind of work in brightfield. - But when looking at stained histology sections, the phase contrast lens does not perform as well as its brightfield sibling. The images below show a stained histology section observed with the same two lenses and photographed with a digital camera using a CMOS image sensor. This image sensor is not too suitable in discerning small changes in contrast. While the difference is significant when observed through the eyepieces, the digital camera’s limitation in recording contrast features shows only a small change.
Stained histology section (human kidney t.s.) in brightfield. The top image was recorded with a Plan Achromat 40x objective (NA 0.65), while the bottom image was obtained with a “high contrast” Plan Achromat phase contrast 40x objective (NA 0.65) (click on image for larger version).
