Confocal Microscopy

The big idea of a confocal microscope is to illuminate and image a sample just one spot at a time. A complete two-dimensional image is then built up by scanning, in other words illuminating and imaging row upon row of individual spots in a plane. The resulting image is then a "slice" through the sample, typically at a distance of a few microns into the sample. It's possible to get some really clear images this way, and to use sets (or stacks) of slices at different depths to build a three-dimensional picture of the sample.

Below is a schematic of how a confocal microscope works:

The light source that illuminates the sample is the laser at the top of the picture. The fine laser beam (blue line) travels to the sample via a beamsplitter, then two scanning mirrors, and finally the actual microscope lenses. Those scanning mirrors are a key ingredient. Each mirror twists about its axis very rapidly, deflecting the laser beam. One mirror is responsible for scanning the light one way across the sample. The other mirror scans the light beam at right angles to the first, so that a complete region of the sample is illuminated.

Light reflected by the illuminated portion of the sample (green line) passes back through the lenses and mirror, through the beamsplitter and into the detector. Each sweep of the sample by the illuminating beam gives a corresponding image of the sample in the detector. Crucially, the illuminating light and the image pass through the same microscope optics - so where the laser is focused on the sample, the imaging optics (being one and the same) are focused too.

Rendered three-dimensional image of an aggregated emulsion

There are a couple more magic ingredients. Imagine the illuminating laser beam is focused some way into the sample, as is usually the case in confocal microscopy. Then although the brightest illumination is at the focus, regions of the sample immediately before and after the focus will also be lit up and will also reflect light. But these regions will not be in focus, and so would make the final image fuzzy. Light from these regions needs to be removed, and this task is performed by the pinhole, shown in the diagram close by the detector.

The pin hole is located such that light from the spot of interest is focused exactly on the pin hole, so passes through and on to the detector beyond. However light from a point deeper into the sample, say, will be focused in front of the pinhole, and will have begun to spread out again by the time it reaches the pin hole. Being too spread out by the time it reaches the pin hole, most will not make it through. A similar fate befalls light emanating from points not far enough into the sample. So the pinhole filters out light coming from points above and below the plane of interest.

The final trick is to use a special dye added to the sample which, when illuminated by one colour of light, will fluoresce and emit light of a different colour. For example the dye Nile Red will give off red light when illuminated with blue laser light. The illuminating blue laser light enters the system via a "dichroic beamsplitter", which acts as a mirror for high frequency light (eg blue) but like a window for low frequency light (eg red). So the beamsplitter diverts the laser light into the system, but the red light returning to it merely passes through and on to the detector.

Confocal microscopes are really useful if you want to look into a sample instead of at its surface, for creating clear images, and for building three-dimensional images. They are widely used in biology, for example. Typically, a confocal microscope is mounted vertically above the sample, looking down into it. In our work we are interested in the role of gravity, so we have arranged our microscope to view the image sideways on, which is pretty unusual.