Skip to main content

SU-8 Information Page

SU-8 comes in many varieties with different densities and viscosities, allowing for the fabrication of structures with a wide range of thicknesses. The correct SU-8 variety and spin speed must be used to achieve a specific thickness. Other fabrication procedures also change due to thickness and the type of SU-8.

The table below shows thickness, spin speed, pre-bake times, post-exposure bake times, and development times based off of data available in MicroChem's datasheets. Some recipes developed at BYU are also in the table. Also below are notes on SU-8 processing and helpful links and datasheets.

 

 

SU-8 Thickness
(in µm)
Spin Speed
(in rpm)
Minutes to Pre-Bake
@65°
Minutes to Softbake
@95°
Minutes to PEB
@65°
Minutes to PEB
@95°
Development Time

             

*All values approximate      **Indicates a BYU exclusive recipe - see SU-8 processing notes for more detail

SU-8 is a negative, epoxy based, near-UV photoresist designed for MEMS and other microelectronic applications. It was originally developed and patented by IBM. SU-8 is processed using standard lithography techniques. When SU-8 is exposed to UV light its molecular chains cross-link, causing the SU-8 to solidify. SU-8 is highly transparent in the ultraviolet range. This allows for the fabrication of relatively thick (hundreds of micrometers) structures with nearly vertical side walls. Two companies have licenses from IBM to sell SU-8: MicroChem and Gersteltec.

Modifications to SU8 Recipes when using the South Aligner

The south aligner in the BYU Cleanroom now has new filters. The following are changes that should be made to compensate for this new filter.

Photoresist / ApplicationThickness (um)Original Exposure Time (s)New Optimal Exposure Time (s)Percent Difference (%)
SU8-10 / Core5152140
SU8-10 / Pedestal Core11162340
SU8-10 / Ridge10172440
SU8-10 / Pedestal Ridge18162340
SU8-3005 / Micropore5152140

SU-8 Processing

1. Substrate Pretreat: Clean wafer/substrate Dehydration Bake - 200° C for at least 5 minutes **BYU Dehydration Bake - 150° C for 15 minutes (Optional) Apply Omnicoat.

2. Spin Coat: Put wafer in spinner and set spin speed for desired thickness - see table for more information Microchem recommends: ramp to 500 rpm at 100 rpm/second then ramp to the final speed at 300 rpm/second and hold final speed for 30 seconds **BYU 5 µm SU-8 10 core: 500 rpm at 100 rpm/sec (6 sec) then 4400 rpm at 1200 rpm/sec (60 sec) and 6000 rpm at 6000 rpm/sec (2 sec) to remove edge beads Apply approximately 1ml of SU-8 per inch of substrate diameter to the center of the wafer to spread out the SU-8. After applying the SU-8, avoid grasping the wafer with tweezers, as this will push up ridges in the SU-8.

3. Softbake: Pre-bake the wafer at 65° C Ramp to 95° C and bake for more time. A hot plate is the recommended baking method **BYU tip: cool back down to 65° C on the hotplate then to room temperature on a level nonmetal surface. For bake times see the table.

4. Exposure: SU-8 is optimized for near UV (350-400nm) exposure. For best results energy below 350nm should be filtered out. Expose the wafer to UV light on an aligner. If the time is too short, the features will come off during development. Overexposure will increase the width of features. Thicker SU-8 requires more exposure energy. For more information on exposure energy check the MicroChem datasheets in the "Links and Datasheets" section.

5. Post Exposure Bake (PEB): Bake the wafer at 65° C Slowly ramp to 95° C and bake for more time. A hot plate is the recommended baking method. Do not rapidly cool the wafer after the PEB **BYU tip: cool back down to 65° C on the hotplate then to room temperature on a level nonmetal surface. For bake times see the table.

6. Development: Pour some SU-8 developer into a glass dish. Place the wafer in the developer and gently agitate the developer, the whole time strong agitation is recommended for structures with a high aspect ratio or large thickness. Development rates vary widely with agitation, temperature, and other processing parameters. Pour the SU-8 developer into the waste container and rinse the dish with IPA when finished. Typical development times are in the table.

7. Rinse and Dry: Rinse the wafer with IPA. If a white film is visible while rinsing then more development is needed. Dry with a nitrogen gun.

8. Hard Bake/Cure(optional): A hard bake is recommended if the SU-8 is to be left on as part of the final device or if there will be further thermal processing. A hard bake may also help to anneal any surface cracks after development. Typical hard bake temperatures are in the range of 150 °C to 250 °C. Typical times are between 5 and 30 minutes. Better results seem to be achieved by ramping from room temperature to the bake temperature and then ramping back down **BYU 5 µm core process: **Hard bake 1 - ramp on hotplate from RT to 200°C, bake at 200°C 10 min, ramp down to 95°C **Descum - PE2, 90 sec 50 W, 100 sccm Oxygen **Hard bake 2 - ramp on hotplate from RT to 250°C, bake at 250°C 5 min, ramp down to 95°C.

9. SU-8 Process Diagram from MicroChem.

su8_process2.jpg

SU-8 Spin Speed vs. Thickness - Graphs from experiments done at BYU
SU-8 Curing Images - Images of Cured SU8 from experiments done at BYU

SU-8 Links and Datasheets

Table of Properties for SU-8 2000 and 3000 from MicroChem
SU-8 2-25 Datasheet from MicroChem
SU-8 50-100 Datasheet from MicroChem
SU-8 2000-2015 Datasheet from MicroChem
SU-8 2025-2075 Datasheet from MicroChem
SU-8 2100 and 2150 Datasheet from MicroChem
SU-8 3000 Datasheet from MicroChem
MicroChem SU-8 page - An SU-8 supplier with useful SU-8 data sheets and FAQ
Gersteltec SU-8 page - Another SU-8 supplier with data sheets
MEMScyclopedia - Webpage dedicated to collecting all known data on SU-8
SU-8 Forum - A forum created by Sporian Microsystems for discussing SU-8