Semiconductor Technology from A to Z

The exposed wafers are processed using either dip or spray development. While in the dip etch step an entire batch can be developed in a caustic solution and then rinsed, in spray development each wafer is processed individually. As in resist coating, the wafer sits on a chuck and is continuously sprayed with developer solution while rotating slowly. Once the resist has been fully developed, the wafer is rinsed with sprayed water to stop the development process. Some advantages of spray development over the dip process are as follows:

• the smallest structures can be exposed
• the developer solution is continuously renewed: contamination is prevented
• the amount of developer solution used is considerably lower

During development, certain areas of the resist dissolve, depending on the resist type, so that a patterned wafer remains at the end. Exposure triggers a reaction in the resist in which the sensitizer is converted into an acid. During development, this carboxylic acid is converted into a water-soluble salt with sodium hydroxide (NaOH) according to the following equation:

Illustration of the exposed resist types after development

Since residues of the developer remain on the wafer with potassium or sodium hydroxide solutions, metal-ion-free developers such as TMAH (tetramethylammonium hydroxide) are also used. A further curing step (hard-bake) makes the resist resistant for subsequent processes, such as etching or ion implantation.

The resist structure is now inspected. Under oblique light incidence, a microscope can be used to detect the uniformity of the resist layer, as well as poor focusing and resist accumulation. If the resist lines are too thick or too thin, the resist has to be removed and the process repeated. Likewise, the resist structure must be precisely aligned to the underlying layer, otherwise the coating and exposure processes also have to be repeated. For this purpose, there are various alignment marks for checking the alignment accuracy and the line width:

Illustration of alignment marks

The width of the resist lines is checked with a microscope: light rays strike the wafer perpendicularly and are not scattered back to the objective at edges. This makes black lines visible as boundaries of the structures, and by measuring their spacing with the aid of the microscope's magnification, the width of the lines can be calculated.

After the pattern beneath the resist has been transferred by etching, or after the resist has served as a masking layer during an implantation process, the resist has to be removed. This is done using strong etch solutions (remover), in a dry etch step, or with solvents. Acetone is suitable as a solvent for removing the resist layer, since it does not attack the wafer or other layers. However, an ion implantation or a dry etch process can further harden the resist layer, so that solvents can no longer attack and remove the resist.

In this case, the resist can be removed with a remover at about 80 °C in a dip process. If the resist has been heated to above 200 °C during processing, it can no longer be removed even by the remover. In that case, the resist has to be removed by ashing or burning.

In the presence of oxygen, a gas discharge is ignited by high-frequency excitation, generating excited oxygen atoms. These burn off the resist without leaving residue. However, the charged particles are strongly accelerated by the electric field and can thereby cause slight removal of the wafer surface or minor damage to the wafer. Resist ashing (stripping) can be carried out either directly after etching in the same process chamber (in-situ) or in a separate chamber (ex-situ).