Editing Diode-pumped solid-state laser (section)

== Coupling ==
{{uncited section|date=January 2023}}
The wavelength of laser diodes is tuned by means of temperature to produce an optimal compromise between the absorption coefficient in the crystal and [[Energy efficiency (physics)|energy efficiency]] (lowest possible pump photon energy). As waste energy is limited by the [[thermal lensing|thermal lens]] this means higher power densities compared to [[high-intensity discharge lamp]]s.

High power lasers use a single crystal, but many laser diodes are arranged in strips (multiple diodes next to each other in one substrate) or stacks (stacks of substrates). This diode grid can be imaged onto the crystal by means of a [[lens (optics)|lens]]. Higher brightness (leading to better beam profile and longer diode lifetimes) is achieved by optically removing the dark areas between the diodes, which are needed for cooling and delivering the current. This is done in two steps:
# The "fast axis" is collimated with an aligned grating of [[cylindrical]] micro-lenses.
# The partially collimated beams are then [[image]]d at reduced size into the crystal. The crystal can be pumped [[wiktionary:longitudinal|longitudinally]] from both end faces or [[wiktionary:transverse|transversely]] from three or more sides.
The beams from multiple diodes can also be combined by coupling each diode into an [[optical fiber]], which is placed precisely over the diode (but behind the micro-lens). At the other end of the fiber bundle, the fibers are fused together to form a uniform, gap-less, round profile on the crystal. This also permits the use of a remote power supply.

=== Some numbers ===

High power laser diodes are fabricated as bars with multiple single strip laser diodes next to each other.

Each single strip diode typically has an active volume of:
{| class="wikitable"
|-
|1&nbsp;μm || 2&nbsp;mm || 100&nbsp;μm
|-
|Height || Depth || Width
|-
|fast axis || optical axis || slow axis
|}

and depending on the cooling technique for the whole bar (100 to 200) μm distance to the next laser diode.

The end face of the diode along the fast axis can be imaged onto strip of 1&nbsp;μm height. But the end face along the slow axis can be imaged onto a smaller area than 100&nbsp;μm. This is due to the small divergence (hence the name: 'slow axis') which is given by the ratio of depth to width. Using the above numbers the fast axis could be imaged onto a 5&nbsp;μm wide spot.

So to get a beam which is equal divergence in both axis, the end faces of a bar composed of 5 laser diodes, can be imaged by means of 4 (acylindrical) cylinder lenses onto an image plane with 5 spots each with a size of 5&nbsp;mm x 1&nbsp;mm. This large size is needed for low divergence beams. Low divergence allows paraxial optics, which is cheaper, and which is used to not only generate a spot, but a long beam waist inside the laser crystal (length = 50&nbsp;mm), which is to be pumped through its end faces.

Also in the paraxial case it is much easier to use gold or copper mirrors or glass prisms to stack the spots on top of each other, and get a 5 x 5&nbsp;mm beam profile. A second pair of (spherical) lenses image this square beam profile inside the laser crystal.

A volume of 0.001&nbsp;mm<sup>3</sup> active volume in the laser diode is able to saturate 1250&nbsp;mm<sup>3</sup> in a Nd:YVO<sub>4</sub> crystal.