Godin:Protocols/Single layer PDMS - Nov 8th
This protocol details the fabrication of a 1 layer PDMS device: a microfluidic channel. It is assumed that masters for respective layers have already been fabricated (see Godin:Protocols/SU-8 Master, Godin:Protocols/AZ Master). The idea is to modify the silicone elastomer base to curing agent ratio, thereby modifying PDMS elasticity.
For most devices, 80g of PDMS is sufficient to produce reasonably-sized devices. Weigh 70g of silicone elastomer base on the scale using the weighing dish, then add 7g of curing agent. Though for these devices the ratio does not need to be perfect, try to keep it around 10:1. If too much curing agent is added, add more base as required. Use a pipette tip to mix the solution for 10 min. Blow dry the channel master with nitrogen, and place in a polystyrene dish. Use Scotch tape to fix the wafer at the center of the dish by taping all sides. We want to prevent the wafer from floating up, or PDMS from getting under the wafer once it's poured in. (These last steps are done only the first time a new wafer is used. We try to reuse them as much as possible.)
Fill the dish with PDMS. Place the dish in the vacuum chamber (without lid might work better), and allow it to degass for 30 minutes, or until it stops bubbling. To prevent overflow, you may have to periodically break vacuum and allow some bubbles to escape.
The dish is now placed in the oven for at least 3 hours at 80°C. The idea is to cure the PDMS so that it becomes solid. Let the PDMS cool for about an hour afterwards.
Cut out each device from the wafer. Be careful to avoid cutting near master features, as it will be reused for future devices. Place the cut devices in a clean petri dish, channels facing up, with the lid. Closing the petri dishes will prevent an excess of dust of going on the devices.
Valve holes are punched using the green 0.75 mm diameter tool. Punch from the channels' side, as it is easier to see the entryways. Use spare solid PDMS underneath to protect the punching tool. Be careful to punch each holes only once, as straight as possible, and being careful not to tear the PDMS on the way in or out of the punching.
Rinse devices thoroughly with distilled water, making sure the water goes through the valve hole. Next, sonicate the devices and the glass slides that will serve as the floor of the channels for 5 min in each of these solutions: soapy water, deionised water, ethanol and isopropanol. Dry with nitrogen, store in a closed clean dish.
Enter the clean room with both the clean glass slides and clean channels. Follow the instructions on the sheet (on the window) for the operation of the plasma binder. Once ready, place the devices in a glass petri dish and into the binder; do the same with the glass slides. Close the door of the binder and start procedure. 30 seconds at power 150 watts should be good, precisions are on the instruction sheet. Once the plasma ionizing is over, quickly take out the slides and devices, avoiding to touch the upper surface, and place them together. The two surfaces will quickly bond together once ionized, so each device has one chance of being placed on the slide: once they are bonded, it will be close to impossible to separate the two parts. Gently press evenly on the devices to secure the bonds, and place in petri dish with the lid. There is usually enough space for 2 devices on one glass slide. Once all devices are bonded, exit the clean room.
Place your wafer in the oven for at least 2 hrs at 80°C. If possible, leave it overnight.
An additional step that has helped prevent leaks between the glass and PDMS is the addition of a PDMS seal at the perimeter of the PDMS block. You prepare a 11g mixture of 10:1 PDMS as before, but without the degassing step. Pour the PDMS in a syringe with a large needle. Remove devices from the oven and carefully squeeze the PDMS out along the perimeter of each PDMS device. An alternative would be to use a pipette tip to "paint" this PDMS at the joints of the devices. Place the devices back in the oven.
NB: This cleaning part is not always necessary for this procedure of single layer.
After a few uses of the master, whether it has been cleaned many times or not, you should repeat the aminosilane surface treatment.