The overall goal of this procedure is to demonstrate a simple application of droplet microfluidics fluorescence detection using optical fibers. This method can help answer key questions in high throughput biological screening, such as the identification of rare phenotypes and the directed evolution of enzymes. The main advantage of this technique is that it allows for a simple and inexpensive application of fluorescence detection in droplet microfluidics.
To begin, place a precleaned three inch diameter silicon wafer onto the vacuum chuck of a spin coater. Then, apply one milliliter of SU-8 3050 in the center of the wafer, and spin the wafer to provide a layer of photoresist that is 80 micrometers thick. Remove the coated wafer from the spin coater and place it on a hotplate preheated to 135 degrees Celsius.
After 30 minutes remove the wafer and allow it cool back down to room temperature. Next, place the wafer onto a flat surface and align the first layer mask onto the wafer. Expose the wafer under a collimated 190 milliwatt, 365 nanometer LED for three minutes.
After exposure, place the wafer on a 135 degree Celsius hotplate for one minute and then allow it to cool back down to room temperature. Once cool, place the wafer back onto the spin coater and affix it to the chuck. Apply an additional milliliter of SU-8 3050 in the center of the wafer and spin for 20 seconds at 500 r.p.m.
followed by 30 seconds at 5, 000 r.p.m. Next, bake the wafer on a hotplate. Allow it to cool back down to room temperature and then place it back onto the mask alignment surface.
Use the dissecting scope to manually align the second layer mask with the patterned alignment marks on the wafer. Once properly aligned, expose the coated wafer for 4-1/2 minutes using the same conditions as before. After exposure, place the wafer onto a hotplate at 135 degrees Celsius for one minute, then cool the wafer and affix it back onto the spin coater.
Apply one milliliter of SU-8 3050 to the wafer for a third time and spin to add an additional 100 micrometers of photoresist. Remove the wafer and bake on a 135 degree Celsius hotplate for 30 minutes. Then, cool it to room temperature and place it onto the alignment surface.
Align the third layer mask with the previous geometry and expose the coated wafer using the previous conditions for nine minutes. After exposure, place the wafer onto a 135 degree Celsius hotplate for one minute and then cool it to room temperature. Once cool, develop the masks by immersing the wafer in a stirred bath of developer for 30 minutes.
Finally, wash the wafer in isopropanol and then bake it one last time for one minute on a hotplate at 135 degrees Celsius. Store the developed master in a 100 millimeter Petri dish until molding. Prepare a 10:1 mixture of PDMS by combining 50 grams of silicon base with five grams of curing agent in a plastic cup.
Mix the contents with a rotary tool, and then degas the mixture inside desiccator for 30 minutes, or until all of the air bubbles are removed. Pour the degassed PDMS over the prepared master to give a thickness of three millimeters. Then, place the Petri dish back into the desiccator for further degassing.
Once all bubbles are removed, bake the device at 80 degrees Celsius for 80 minutes. Once cooled, cut the device from the mold using a scalpel so that both the 120 micrometer and 220 micrometer telegeometries are accessible from the side of the device. Then punch the fluidic inlets and outlets with a 0.75 mm biopsy punch.
Plasma treat the device with feature side up along with a precleaned glass slide and 1 millibar oxygen plasma for 20 seconds in a 300 watt plasma cleaner. Bond the device by firmly placing the patterned side of the PDMS device onto the plasma treated side of the glass slide. Then, place the device in an 80 degree Celsius device oven and back the assembled device for 40 minutes.
Once baked, use a syringe to flush the device with a fluorinated surface treatment fluid to render the channels hydrophobic. Then, immediately bake the device at 80 degrees Celsius for 10 minutes to evaporate the solvent. Prepare two laser excitation fibers by removing the insulation from the last five millimeters of the optical fibers.
Also, prepare the fiber for collecting the fluorescent signal by removing the insulation from the last five millimeters of a larger optical fiber. Inspect the tips of all the fibers under a microscope. If the tips do not end in a flat surface, recleat the ends with a fiber scribe.
Next, attach a laser fiber coupler to a 50 milliwatt, 405 nanometer laser and attach one of the 105 micrometer core fibers to the laser. Direct the stripped end of the sensor of a laser power meter and use the fine adjustments of the laser coupler to maximize the laser power. Perform the same process to tune the other fiber using a 50 milliwatt, 473 nanometer laser.
Then, mount a quad bandpass filter on the photo multiplier tube using lens tubes to block the laser light and transmit emitted fluorescence. Attach a fiber coupler so that light travels through the filters before hitting the photo multiplier tube. Next, attach the collecting fiber to the fiber coupler.
Place the fabricated microfluidic chip onto the stage of an inverted microscope coupled with a digital camera capable of at least 100 microsecond shutter speeds. Working carefully from the sides, insert the fiber coupled to the 473 nanometer laser into the farthest upstream 120 micrometer channel. Take care not to puncture through to the main flow channel.
Then, insert the fiber coupled to the 405 nanometer laser into the farthest downstream 120 micrometer high side channel, providing fiber spacing of 300 micrometers. Lastly, insert the larger photo multiplier tube coupled fiber into the 220 micrometer tall channel normal to the two laser excitation fibers. Fill a five milliliter syringe filled with HFE 7500 containing 2%ionic fluorosurfactant and mount it to the spacer oil inlet of the detection device using PE2 tubing.
Next, fill a syringe with a mixed FITC and blue dye emulsion and mount it on a vertically oriented syringe pump coupled to the device as droplet reinjection inlet using PE2 tubing. Connect a length of PE tubing from the device exit to a waste container. Then, prime the device by running each of the pumps at 1, 000 microliters per hour until both oil and droplets are seen to be regularly combining in the device and flowing downstream.
Adjust the flow rates such that the spacer oil runs at 6, 000 microliters per hour, and the droplets at 100 microliters per hour, providing significant spacing between droplets traveling through the detection region. Then, turn on the lasers and start the data acquistion program. Next, adjust the photo multiplier tube gain to provide signals that are more than 100 times the baseline noise floor.
Finally, adjust the laser power so that all of the doublet peaks are clearly visible on a single, linearly scaled time trace. A droplet flowing through the detection channel encounters discrete excitation regions in front of each of the laser coupled optical fibers. The detection fiber records emitted fluorescence with a temporal shift with a peak separation that correlates with the time it takes a droplet to move from one excitation region to the next.
This removes the need for further light filtering and separate photo multiplier tubes. The efficacy of the detection scheme was tested by detecting a mixed emulsion of droplets containing dyes that excite at 405 nanometers and at 473 nanometers. This graph shows clear separation between the four reinjected droplet types when being flowed at roughly 50 hertz.
The dynamic range and absolute sensitivity of the detector was investigated by measuring the fluorescein only droplets. The data shows the ability of the system to detect dye concentrations that range from 0.1 to 100 nanomolar. After watching this video you should have a good understanding of how to implement a simple and compact fluorescence detection system by utilizing an array of inserted optical fibers and a single photo detector.
