Photolithography
Created | Updated Feb 28, 2002
Photolithography as is commonly used in semiconductor engineering is a process (similar to photography) that allows complex layers of material (e.g. metal,silicon,insulator etc) to be patterned into a variety of micro/nano electronic devices and/or circuitry.
The basic steps involved in photolithography are:
1. Substrate choice
- choice of base material which forms the backplane of any fabricated device - commonly silicon, but could be glass, metal or another material.
2. Substrate preparation
- involves polishing, cutting, slicing, cleaning substrate before subsequent processing steps
3. Deposition of 1st material (metal,insulator etc) "Layer 1" - commonly vacuum coating techniques are used (e.g. sputtering, thermal evaporation, PECVD..etc)
4. Photoresist coating/spinning - A layer of photoresist is deposited onto the substrate. This is normally done by holding the rear side of substrate onto a vacuum chuck and dropping liquid photoresist onto the top side of the substrate. The vacuum chuck spins at high velocity (typically 4000rpm) and leaves a thin layer (1 - 2 microns) of photoresist covering the surface of the substrate.
5. Baking - After coating with photoresist, the substrate is gently baked to remove any residual solvent and help to cross-link chains in the photoresist.
6. Exposure - Photoresist changes chemically upon exposure to Ultra-violet (UV)light. Hence by exposing some parts of the substrate to Ultraviolet light and blocking other parts it is possible to write patterns into the photoresist. Generally no physical change in appearance is observed, simply some parts have been exposed (changed in chemical structure) and other parts remain unchanged. Complex patterns can be transferred from a "shadowmask" or "photomask" to the photoresist. Shadowmasks or (also known as photomasks) are used as a method of transmitting/blocking UV light to the photoresist layer and are usually fabricated of Black/Chrome areas on glass/quartz plates. Transparent areas will transmit UV and darkened areas will block UV.
There are two types of photoresist - positive and negative. Basically positive will allow you to make a copy of the pattern on the shadowmask (ie WYSIWIG - What You See is What You Get) whereas negative will result in an inverse pattern.
7. Developing.
- After exposure to UV, the photoresist comprises parts which have been chemically altered and parts which remain unchanged. Using developer it is possible to remove either the parts which have or have not been exposed (depending on choice of positive/negative resist). The result being a layer of patterned photoresist on top of "Layer 1".
8. Postbake - allows for the photoresist to be hardened before exposing to harsh etching environments.
9. Etching - After developing resist and postbaking the next step is to transfer the pattern to Layer 1 by removal of the unwanted material. This is done by using selective etching techniques. The etch will remove the parts of Layer 1 which are not protected by photoresist.
10. Photoresist removal - after Layer 1 has been etched, it is no longer required to protect the material by photoresist, hence any remaining photoresist should be removed. The result being a substrate which now is now patterned in the material chosen for Layer 1 (see step 3 above)
Thus one Layer has now been processed. Many electronic devices/circuitry rely on many different layers, and hence the whole process has to be repeated again (starting from step 3). Often it is critical that subsequent patterned layers are aligned to already existing layers. This process is known as "mask aligning" and essentially involves an extra step between
The basic steps involved in photolithography are:
1. Substrate choice
- choice of base material which forms the backplane of any fabricated device - commonly silicon, but could be glass, metal or another material.
2. Substrate preparation
- involves polishing, cutting, slicing, cleaning substrate before subsequent processing steps
3. Deposition of 1st material (metal,insulator etc) "Layer 1" - commonly vacuum coating techniques are used (e.g. sputtering, thermal evaporation, PECVD..etc)
4. Photoresist coating/spinning - A layer of photoresist is deposited onto the substrate. This is normally done by holding the rear side of substrate onto a vacuum chuck and dropping liquid photoresist onto the top side of the substrate. The vacuum chuck spins at high velocity (typically 4000rpm) and leaves a thin layer (1 - 2 microns) of photoresist covering the surface of the substrate.
5. Baking - After coating with photoresist, the substrate is gently baked to remove any residual solvent and help to cross-link chains in the photoresist.
6. Exposure - Photoresist changes chemically upon exposure to Ultra-violet (UV)light. Hence by exposing some parts of the substrate to Ultraviolet light and blocking other parts it is possible to write patterns into the photoresist. Generally no physical change in appearance is observed, simply some parts have been exposed (changed in chemical structure) and other parts remain unchanged. Complex patterns can be transferred from a "shadowmask" or "photomask" to the photoresist. Shadowmasks or (also known as photomasks) are used as a method of transmitting/blocking UV light to the photoresist layer and are usually fabricated of Black/Chrome areas on glass/quartz plates. Transparent areas will transmit UV and darkened areas will block UV.
There are two types of photoresist - positive and negative. Basically positive will allow you to make a copy of the pattern on the shadowmask (ie WYSIWIG - What You See is What You Get) whereas negative will result in an inverse pattern.
7. Developing.
- After exposure to UV, the photoresist comprises parts which have been chemically altered and parts which remain unchanged. Using developer it is possible to remove either the parts which have or have not been exposed (depending on choice of positive/negative resist). The result being a layer of patterned photoresist on top of "Layer 1".
8. Postbake - allows for the photoresist to be hardened before exposing to harsh etching environments.
9. Etching - After developing resist and postbaking the next step is to transfer the pattern to Layer 1 by removal of the unwanted material. This is done by using selective etching techniques. The etch will remove the parts of Layer 1 which are not protected by photoresist.
10. Photoresist removal - after Layer 1 has been etched, it is no longer required to protect the material by photoresist, hence any remaining photoresist should be removed. The result being a substrate which now is now patterned in the material chosen for Layer 1 (see step 3 above)
Thus one Layer has now been processed. Many electronic devices/circuitry rely on many different layers, and hence the whole process has to be repeated again (starting from step 3). Often it is critical that subsequent patterned layers are aligned to already existing layers. This process is known as "mask aligning" and essentially involves an extra step between
