A light source that is suitable for confocal illumination needs to fulfill a range of requirements: brightness, stability, focusability and appropriate wavelength – besides moderate cost, low energy consumption, and tolerable heat and noise generation.
Focusability is the most important parameter, as the target is to focus the light to a diffraction-limited spot. Extended light sources cannot be focused to a spot, and consequently a pinhole is needed to generate a virtual spot-shaped light source (the "spotness" is determined by the optical elements). The pinhole emits light in all (forward) directions, causing losses when finite apertures are used to collect the light. A (good) laser emits light with very low emittance, which means: the light rays are parallel – the beam is not divergent. Such a beam is very easy to focus, as all rays of a parallel beam will cross in the focal point. To obtain diffraction-limited illumination through a circular aperture (the objective lens is circular), the laser must emit only "transversal electromagnetic mode" TEM 00 (see Figure 1), that best corresponds to circular diffraction. The engineers designing confocal microscopes have to select and test new laser types for this performance before integrating them into a system.
Brightness is also an important factor. Not the total amount of light emitted by the source is the critical parameter, but the irradiance, which is the power of electromagnetic radiation per unit area. The laser beam diameter is usually in the range of less than a millimeter. An appropriate laser emits about 10 mW, corresponding to 10 lm blue light, which is emitted from 10–6 m2. That corresponds to 107 lux – 100 times brighter than a bright sunny day.
The intensity of illumination must be constant, both during acquisition of an image, in order to ensure homogenous illumination, and during longer periods of operation, in order to compare measurements from different lab days. Also, the beam has to be stable: the pointing direction must not change, otherwise the adjustment is lost, and the beam profile has to be constant, including the polarization properties.
Classical lasers are single-wavelength emission lasers. This restricts the applications in fluorescence and requires a series of lasers to be installed. The strong coherence of the laser emission is not needed for confocal imaging. It is instead a challenge for the engineers, as it causes interference artifacts that have to be eliminated.
Laser technology is still a very vivid area of innovations and new concepts. When confocal microscopy started to be a product that could be bought ready-made from industrial providers, which was in the mid eighties of the last century, the light sources were HeNe lasers. These lasers are comparably small and easy to handle. The most prominent red line at 633 nm was a little too far in the red for most fluorochromes used at that time. A second HeNe laser type emits at 543 nm, which is good for a series of mid-range fluorochromes. Unfortunately, these lasers are quite dim, typically emitting only 0.5 mW. Most successful was the Ar-Ion gas laser with two prominent emissions at 488 nm and 514 nm. Especially 488nm was the mostly used excitation line, it perfectly fits for fluorescein (FITC) and derivatives that were tailor made to be used with 488 nm excitation. Ar lasers are still used today, as they offer 5 different excitation lines in the blue/blue-green range of the visible spectrum.
Although gas lasers are still in use, solid state lasers and diode lasers currently have the best chances to replace them. They are more stable, produce less heat, need no cooling and emit more or less any wavelength. Most of them are pulsed lasers and therefore suitable for fluorescence lifetime imaging (FLIM). They are limited by emitting only one single wavelength and a range of combinable lasers is needed to fulfill the needs for multiparameter fluorescence experiments. Typically 5 different lines are needed (roughly) to cover most of the dyes that are excited in the visible range.
Short wavelength excitation was initially used for UV dyes, like DAPI nuclear stains. This is now possible with 405nm diode lasers. These lasers also fit deep-blue excited fluorescent proteins. Transiently, Ca2+ measurements were conducted with ratio-excitation by Ar-UV lasers emitting 351 nm and 364 nm. Modern biosensors based on FRET-FP pairs made this complex and complicated approach obsolete.
Biologists have always wanted to examine samples as alive as possible. One restriction in looking deep into living material is scattering, caused by the complex refractive index pattern that is resembled by crowded cellular organization in the specimen. Longer wavelengths have the advantagethat they evoke significantly less scattering and thus clearer images. Consequently, the current trend is to find red and infrared light sources and fluorescence labels that permit deeper penetration. A second booster for far-red dyes is the request for parallel fitting of as many dyes as possible in the used spectral range, to observe many parameters simultaneously. One solution is to use optical parametric oscillators (OPOs) to meet that target.
Multiphoton microscopy uses the nonlinear process of exciting a molecule by two quanta half the energy required to step the electronic system from ground state to the excited state. As the probability of this 2-particle process depends on the photon concentration in a square mode, the excitation inherently occurs only in a layer around the focus which inherently leads to optical sectioning without pinholes. The required high photon density is created by pulsing the laser energy. Mid-range fluorochromes are excited in the near