Optical microscopes are the most common class of microscopes. Any piece of optical equipment used to look at small regions of space can be called an optical microscope. In fact, primitive microscopes were nothing else but magnifying lenses mounted on a holder. This design is still widely used by goldsmiths and philatelists today, but this is not the common gizmo most people would call a microscope. What most people associate with the term 'microscope' is the common biology class prop, which in a strict sense should be called a 'compound microscope,' because it is made out of numerous lenses, mirrors and apertures, all assembled together in a casing of a very complicated shape and with a number knobs on the side. This entry will explain the parts a microscope is made of, along with some nomenclature stuff, and it will roughly explain its working mechanism. At the end of this entry there is a short guide on how to get the best out of a microscope, as well as a chapter on modern optical microscopes. Historical aspects can be found in The History of Optical Microscopy.
The Parts of a Microscope
A little abstract imagination is required to see what's on the following picture. The gizmo to the left is supposed to represent a generic optical microscope.## ———————————— Ocular (eye or detector) __// / | —————————————— Filters (polarisation or colour) ===== | | ——————————————— Tubus | |__ | |-( ) ——————————— Focus adjust knob |__| \\ //||\\ —++——————————— Objective lenses (here, three on a drum) `´ || ========|| —————————— Object-holder table ___LL____||__ |_____\_____| \______________ Light source (bulb, laser, mirror)
Ocular - A lens for comfortable viewing1, sometimes it is also called the 'eye-piece'. High-tech microscopes have the ocular replaced by a detection device such as a digital camera or a photomultiplier. Some designs have a double ocular, so the viewer can use both eyes to inspect an object. This is even more comfortable and gives some 3D-impression but is also expensive. Cheaper versions have a single eye-piece, which is a lot easier to manufacture. Normally the eyepiece is movable, often slightly bent towards the observer to add comfort, but this also requires some extra mirrors.
Filters - Sometimes it is useful to filter the light stemming from the studied object. For that reason, filters can be incorporated into the light-path just before the ocular lens. This is a common variation seen in many microscopes. Filters include colour filters (to cut off undesired colours) polarization filters (which enhance contrast in certain probes) or generic filters (to dim brightness).
Focus Adjust Knob - To obtain a sharp image of the probe it must be placed at the right distance from the objective lens (in the focal spot). Images become blurry if the object is some microns away from the focus. For that reason almost all modern microscopes have a 'fine' distance adjustment knob. This mechanically lifts or lowers the object holder (or sometimes the whole microscope) towards or away from the objective lens. Many microscopes have two knobs - a big chunky one and a small one. The big one is for coarse adjustment (movement in the centimetre-range) and the small one for fine tuning (with the range of a few microns).
Objective Lenses - These are the lenses that will collect the light from the tiny probe. This is the really expensive, vital, high-precision part of a microscope - the heart of a microscope. There are many variations of objective lenses and for that reason a microscope is commonly equipped with a revolving drum on which various objectives are mounted. The user can then simply change the objective by turning the drum, instead of unscrewing and re-installing objectives.
Object-holder - This is the table where the probe to be examined is positioned. Since most of the studied probes are observed by shining light through them (like an overhead projector) the table has a hole in its middle where the light will shine through. Right above this hole, on the top of the table, there is the probe-holding clamp which holds the immobilises the probe. Since most of the studied probes are placed on (more or less standarised) glass plates, there is a groove according to the plate size. Another common feature of that table is that it can be mechanically moved, laterally on the focal plane, with two adjustment knobs. This can be used to scan wider regions of the probe in the search for cells or other interesting features.
Light Source - From here the light is channelled through the probe. Some microscopes have a lamp adaptor at this position, which is merely a light bulb and a translucent surface, to generate diffuse light. Sometimes one can find a diversity of filters between the light source and the hole in the object-holder. Many microscopes do not have an internal lamp, but a mirror which will direct external light (eg, laser, neon-light, sunlight, etc.) to the opening. This has the advantage that the properties of the light can be adjusted by the user according to the user's needs (eg, modulation of intensity, plane of polarisation, etc...)
General Working Mechanism
A beam of light is sent to the object that is to be observed. It can be shone through it if it is translucent (as with an overhead projector) or scattered by its surface if it is illuminated from the top. In both cases this light is collected by the objective. When the object lies in the focus of the objective lens then the light coming from the object is turned into parallel light. A third lens (the ocular) is then used to focus this parallel light onto another plane. If the distance between this plane and the ocular is greater than the focal length of the objective lens, then the overall impression of magnification is achieved. Normally objectives have a number followed by an 'x' engraved on them. That's the magnification factor. Just from this principle one might think that any imaginable magnification could be achieved. However, there is a certain loss of information during the collection of light by the objective lens, which results in the resolution limit. Details follow:
The Resolution Limit
Only a cone of light coming from a certain point of the object can reach the objective lens. The remaining light goes somewhere else outside of the lens boundaries and is lost forever. This implies a loss of information. The light excluded from the collection cone will be missing for the reconstruction of the image after the third lens. This is not an intensity problem as it might appear at a first glance, but a problem of interference (this is also where the wavelength of light starts to play a role). Due to this missing interference, information from the edges of the image point will become blurry. For this reason, the bigger the collection cone, the less blurry the image will become. However, no objective lens can come infinitely close to the object (thus increasing the cone), and even so, it will never go around the object. There is a number that characterises the size of this cone - it is called the 'numeric aperture.' This is the second number (usually close to 1) engraved on the objective lens.
The first number (followed by the 'x', the 'magnification factor') indicates how big a blob will look. The second one explains how blurry the blob will appear. There's a theoretical limit for the second effect. One can visualise this resolution effect by printing different zoomed-in jpeg images on an ordinary computer printer using different DPI numbers (dots per inch). At low DPIs no matter how much one zooms-in, all one will see is just bigger pixels instead of more details. Common magnification factors range between 20 and 100, and common numeric apertures range from 0.6 to 1.4 (latter requires special immersion objectives, cf: Objectives ). The resolution limit is roughly given by the wavelength divided by twice the numeric aperture, so in most cases it's roughly the wavelength divided by two. Note that this is a limit - the resolution can't be better, but it can be worse.
Getting the Best Results
As described in the previous chapter the ultimate limitation of a microscope is given by the resolution limit, which is given by the laws of optics. The resolution can be worse, which depends on the quality of the used objective lens. Here, no tricks can be applied. One has to have good objective lenses. However, this does not automatically mean that images will be superb, just because one is using expensive state-of-the-art objective lenses with big numeric apertures. There are more problems.
Contrast - Probes with poor contrast, such as cells in water, are constantly annoying microscopists. For that reason some tricks have been developed to circumvent low contrast. One of them is the use of polarised light and polarisation filters. Light passing a medium normally gets its polarisation plane changed, which can't be noticed unless one filters this difference out. By filtering, (with polarisation filters) these fine differences between different media become more pronounced.
Another trick is to selectively stain parts of the object, usually cells, which is a science in itself. Through trial and error, people figured out how to stain specific parts of a cell, thus making diverse membranes and organelles visible. Under normal conditions one will not be able to recognise most of a cell's organelles under a microscope.
Aberrations - The resolution limit mentioned above also depends on the used wavelength. In this case a spot will look more blurry for red light than it will look for blue light. If one is using white light to illuminate the sample, small features usually will have an annoying rainbow-halo around them. If the objective does not compensate well for this so-called 'chromatic aberration', then the other way around this is to use monochromatic, or at least colour-filtered light. A second type is the spherical aberration. This arises from the manufacture of the lens. Since lenses are derived from spheres, the edges of a lens will distort the image. One way to circumvent this problem is to use only the inner part of the lens at the cost of numeric aperture (which gives lower resolution). Another way is to use aspheric objective lenses, which are cleverly arranged to minimise the aberration. However, these lenses are very sensitive towards differences in refractivity on the sample. Using coverslips of different thicknesses will cause trouble. Fortunately, coverslips nowadays have a standard thickness (between 0.13 mm and 0.17 mm) and modern aspheric objective lens arrays have an adjustment ring.
Sample Preparation - At first it might seem of low importance, but the preparation of samples is at least as important as having a good microscope. Besides the staining techniques mentioned above, it is of great help if the samples can be cut into thin slices, because of the low depth resolution of conventional microscopes. The correct immobilising (or embedding) of samples is also of importance. Many cells, for instance, are sensitive towards the medium in which they are located (cf: osmosis). The concentration of stuff dissolved in a cell-growth medium can rise abruptly because of the heating effect of the illumination at the sample holder.
Modern Optical Microscopy: Confocal Microscopy
By 1954 a very smart man named Marvin Minsky came up with the idea of confocality, which sounds absurdly simple, but was quite difficult to put into practice. Confocal means that one only illuminates and observes things that lie in the focus of the microscope's objective lens. It tremendously improves the resolution of a microscope, but it needs a very sharp beam of light (preferably monochromatic and very parallel). For that reason it was not until the development and improvement of lasers (lasers are very parallel and quite monochromatic) that confocal microscopes were made available.
Confocal microscopes have an improved resolution of up to 200nm on the observation plane and are - as an optical method - not restricted to surfaces or dead stuff. Living cells and the inner part of crystals can easily be observed using a confocal microscope. One problem is that the confocal spot has to be scanned over the sample in order to obtain conventional images. This makes the inner mechanics of such a microscope a little more complicated than common microscopes, and increases the time needed to acquire one image. Furthermore, the image cannot be seen with the naked eye, since laser radiation is especially harmful to the eye. For that reason detectors must be controlled by computer and the images are only visible from a computer monitor. However, this method has proven very useful in the characterisation of new nano-structured materials and in single molecule spectroscopy, where individual detectable molecules (that are located in the confocal spot) are addressed and steered by the laser-light.