Page 5

Figure I.5 Stages of DOE fabrication

In this book, we deal only with the first stage of this technological pattern: calculation of phase function. The essence of the subsequent stages can be briefly described as follows.

After the inverse problem has been solved and the DOE phase function has been derived on the basis of a set of desired characteristics of the image, one needs to transfer it onto a physical carrier as a microrelief. There are quite a lot of techniques for translating the phase function onto a DOE microrelief. Note, however, that a method of microlithography making it possible to obtain a stepwise microrelief [4] has found the widest use.

The starting-point for microlithography is a photomask that can be produced using a photo- or e-beam image generator and represents a binary amplitude mask consisting of transparent and opaque areas of photofilm. Next, the surface of a substrate is coated with photoresist and exposed to UV light through the photomask. Under the action of radiation, the photoresist changes its structure and, upon the succeeding chemical treatment, either illuminated or non-illuminated areas are removed, depending on the type of photoresist. As a result, windows arise in the photoresist layer through which the etching of the substrate material takes place. After etching and removal of the remaining photoresist, a microrelief corresponding to the phase pattern on the photomask will be found on the substrate. The microrelief depth depends on the time of etching. To produce a multistep microrelief, the above procedure should be repeated. Note that subsequent binary photomasks should be aligned precisely with the preceding ones. In such a manner one can fabricate transmitting phase DOEs. If a metal microlayer is deposited on the microrelief, a reflecting DOE will result.

It is also noteworthy that the process of DOE fabrication is, as a rule, iterative in nature since, if necessary, each stage is repeated or one returns to a preceding stage.

I.2 Methods for Calculating DOEs

Obtaining the phase transmission function of a DOE is always connected with solving an ill-posed inverse task. This task is inverse because it relates to