Hi, my name is Marcos. I'm working with Professor Roman Stalker at MIT Civil Environmental Engineering and I'm gonna be showing you how to make a micro channel using self photography technique. And to begin, we have to design a mask.
So this is the mask, which is basically the final design of the microchannel that we intend to do. This is a design of an injector, so we can make a thin layer of nutrient where we put microorganisms swimming around it and we see how the chemo tax, and this is a channel of a side cavity to generate a micro vortex inside the cavity, which are, there'll be video showing it later on after getting a mask, we take wafers to the clean room and put it in the, on the petri dish. And we're gonna to the, we're going to the clean room now.
Okay, so after having our mask printed and ready, so now we are in the clean room to do the actual micro fabrications. And what I'm going to do now, I'm taking a piece of the clean wafer that I, that I just bought and I just want to make sure that it is extra clean by using the following tree solvents. The first step is to clean using an acetone and then immediately methanol.
And then the final step is isopropanol. And we're working on the shiny side of the wafer, which is prepared for the coating. And now I'm gonna blow dry it using nitrogen.
After drying the wafer, now we put the wafer inside an oven off about 130 Celsius and bake it for or five to 10 minutes. In the meantime, while we're waiting for the baking of the wafer, we can prepare the two hot plates, 1 7 65 Celsius and the other 1 95 Celsius. So after baking for five minutes, we take a wafer out from the oven and let it sit in the room temperature for about five minutes.
And then we're ready for spin coating. Before pouring the photo resist, we need to just to make sure there is no dust particle on top of it, we use nitrogen to squirt it on top of the wafer. The squirting motion is from the center out.
I'm now going to pour the photo resist on top of this wafer. We pour the amount of about three milliliters. Try to put the bottle as low as possible so that it doesn't generate bubble as you pour the photo resist on top of the wafer.
So now let the photo resist sit on the wafer for about a minute before we start spinning it. And as I mentioned, I'm gonna make a hundred micron thickness of coating. So we set the total spinning time to 55 seconds and we can start spinning.
First of all, we try to ramp the spinning speed from zero to 500 RPM in five seconds and let it spin for about 10 seconds such that you can see that the photo resist is coating the entire wafer and then ramp it up to 3000 RPM for about seven seconds and let it spin for 30 seconds. So after coating the wafer with the photo resist, I now place the wafer on top of the hot plate and I'll soft bake it at 65 degrees Celsius for five minutes and 95 degrees Celsius for 20 minutes. After the soft baking at 95 degrees is completed, we cool it down to the room temperature, but before that, we cool it back at 65 degrees Celsius for about two to three minutes so it doesn't have a rapid decrease in temperature.
And then we put it at the room temperature to cool down for about five minutes. And now I'm going to expose it with UV light so that the part that is exposed will be hardened and the part that is unexposed will be washed away later. So this is the part where we use the mask and as you can see, the mask has two parts, the part where it is transparent and the other part that is completely opaque or black in this, in this setting.
So I'm gonna put this mask on top of the coated wafer. So the mask that you get from the printing company have two sides. One is the one where they print with with the ink.
And this is a site that you want to put in contact with the wafer. Now I'm going to expose the wafer through the mask with UV light. And for safety purpose, I'm going to wear this safety goggles.
The amount of the exposure time depends on the SUA that you're using and also the thickness that you're trying to achieve. Now after exposing, I'm putting the wafer onto the 65 degrees hot plate and bake it again for five minutes and then 95 degrees Celsius for 20 minutes. And just remind you this baking time varies from the thickness desire and also different as way that you're using.
Following this point, we are going to develop the wafer. Develop means we remove the part of the photo resist that is undesired so that only the exposed region leaves on the wafer. In order to develop the wafer and to wash away the undesired part of the photo resist, we're using the developer.
In this case, I'm using pomonal acetate as the developer and I'm gonna pour this solvent into this beaker and placed the wafer and immerse it into the beaker for about 10 minutes. Again, the time varies depending on your thickness and the solvent that you're using. Now, 10 minutes later you can see that what is left on this wafer is the design of the channel that is desired and, and all the unwanted portion has been washed away just to give it a rinse.
We use the same solvent, a fresh one, the same polyon acetate, and just squirt on top of the wafer just to rinse it off. And finally, to wash the solvent away, we use iso propanol and just rinse it. And now we blow dry it using nitrogen.
And now you have a clean wafer with the desired pattern on it. And we call this as the master. Now we are going back to our lab to pour PDMS on it and to continue the rest of of the procedure.
So now we're back from the clean room and we have, we get the master, which is basically the mold for what we're gonna do this mold. What we're gonna do next is to pour PDMS, which is a polymer and it's gonna cure. So as you can see that if we pour a polymer on top of this mold, the polymer will have some groove on it and eventually if we bond to a glass light, that polymer itself will be the channel.
So now we're gonna prepare the PDMS. So this is the PDMS, which is the polymer, the poly dile Sloane PDMS. And this is a curing agent.
What I'm gonna do is I'm gonna pour this curing agent into this cup and I'm gonna pour this PDMS into this cup and mix it with a ratio of one to 10. So we generally want to use a for to stir it homogeneously. As we stir, we generate lots of bubble, and the bubble is kind of like an indication to tell us whether we have stir it or not.
Now, after mixing the PDMS with the curing agent, we are ready to pour this mixture in on top of the master, which is a wafer. And just to make sure that the wafer doesn't go anywhere during the pouring, we taped the wafer with, with a, with a tape. Before pouring the mixture PDMS on top of the wafer, we want to try to make the wafer as clean as possible from the dust that might have fall on top of it.
So we turn on the compressor to dust the wafer. Okay, you can see that the PDMS has a lot of bubble. So we're gonna remove the bubble by putting this inside a vacuum chamber.
So the vacuuming takes about roughly 30 minutes to remove all the bubbles. And once we're done with removing the bubbles, we take this out and put it into the oven at 65 degrees Celsius and bake it for at least 12 hours. And once we're 12 hours later, we'll have the wafer ready for cutting.
I'm gonna use this knife to cut the PDMS and take the PDMS off the wafer and then use this whole puncher to punch the hole for the inlets and outlets of the tubing. So when you cut, you wanna cut, put your knife until it hit the wafer, but don't apply too much pressure so you won't break the wafer and cut around your channel. Now peel it off.
Do you have, so now you can see that this is the channel, which is the pattern from the wafer. So you can see they're the same. And now this pattern has been and grave on this polymer, this PDMS.
And once they bond this into onto a glass light, this will be the channel for our experiment. The next step, which is important, is to remember to punch holes so that we can insert inlets and outlets into the micro channels after punching the holes into the micro channels into the PDMS. So we used the scotch tape to tape on the channel side, so where the groove is and tape it.
And the reason we're taping this channel with the scotch tape is to, to remove all the dust particles. So the next step is the plasma bonding. And before the plasma bonding, which we're gonna do in the other lab, we're gonna remove the tape and expose the surface that we want to bond, which is this surface and the glass slide together and then expose it onto the oxygen plasma and afterwards, put them together and they will bond.
Hello, my name is Justin Seymour and I'm a postdoctoral researcher in Roman stockers lab at MIT. And my research is focused on microbial ecology, more specifically looking at how marine microbes are able to find patches of food in the ocean. And we're using microfluidics techniques to look at these kind of patterns.
So we're going to use the channel which Marcus demonstrated how to produce earlier and after the plasma bonding. We are then free to use this in a microfluidic experiment. Prior to this experiment, we need to get the substrates we're going to use in the experiment prepared.
And in today's experiment we're gonna be looking at how the marine phytoplankton donella tur electro is able to find microscale patches of food. In this case it's ammonium and we're going to provide it with a a hundred micromolar concentration of ammonium. So in these experiments, we are able to visualize the patches of food by adding a fluorescent stain to the cons to the nutrients which we provide them with.
So we use fluorescein dye. So here I'm pipetting fluorescein at a final concentration of 100 micromolar to the ammonium, which we use as the nutrient. And we can then visualize that nutrient under fluorescence microscopy and which allows us to look at how a patch of nutrients diffuses out, and then look at the response of organisms to that patch of nutrients.
So in the microfluidics experiments, we use glass syringes. So the first thing we want to do is add our substrates and organisms to the syringes we're going to use. The next step is to insert the tubing into the microfluidic setup.
And we do this over at the microscope stage. So we place our microfluidics channel onto the microscope stage and then insert tubing into the inlets and outlets. We then insert a waste reservoir and we ensure that the reservoir is filled halfway with fluid so that the tubing in the waste reservoir is completely submerged.
This enables us to have a constant pressure within the channel. We then attach syringes into the tubing that feeds the microfluidics channel containing the organisms, which in this case is donella, lector, and the nutrient substrate, which in this case is ammonium. We then set up the microscope and look at the microfluidics channel in the appropriate position before we start the experiments.
Once the channel is in position, we then switch over to view the channel with the camera and we can visualize it on the computer. We can now perform finer adjustments with the focus and we can now visualize the injector from which we inject nutrients. And this consists of a 100 micrometer wired needle tip in the built into the microfluidics channel.
And we can produce a gradient of nutrients by injecting them. From this point prior to starting experiments, we need to remove any bubbles which may have got into the microfluidics system. And we do this by using a syringe filled with artificial seawater in this case, which we can attach to the microfluidics channel via a valve.
And we then push this syringe to remove bubbles from the channel. We can remove bubbles in one of two ways, either by forcing them out via the pressure produced by the syringe out the outlet, or we can also remove bubbles by forcing them through the walls of the PDMS material, which is permeable to gases. Once bubbles are removed and we are ready to start the experiment, we then set the appropriate flow rates for injection and materials into the channel using a syringe pump.
In this case, we use a flow rate of two microliters per minute, which is equivalent to a flow rate in the channel of 266 micrometers per second. Once we've set the appropriate flow rate, we can then start the experiment by beginning, beginning the flow, which creates a thin band of nutrients. So we can visualize the nutrients as they're injected into the channel from the injection point by adding fluorescein dye to the nutrients and visualizing it under APF fluorescence microscopy, we can then look at patch dynamics by stopping the flow in the microfluidics channel and visualizing the band of nutrients diffused out laterally from their original position.
We then switch to phase contrast microscopy to visualize the organisms swimming within the channel and looking at their positions in relation to the nutrient patch. We are able to do this by taking a sequence of frames at a frame rate of 10 per second, which produces a movie in which we can visualize these organisms. We can then visualize swimming organisms by taking time differences between two individual frames in which one frame is subtracted from the previous, so the only moving objects are visualized.
So we can remove the background noise from the images. We then record movies at regular intervals for 10 to 20 minutes after the initial release of the nutrient patch. And by using the image analysis software to superimpose the positions of organisms from one frame in the sequence to another, we can look at the changes in position of these organisms to gain information on their swimming trajectories from which we can then gain more detailed information on their swimming statistics, including speeds and changes in direction.
So these experiments have allowed us to, for the first time directly examine how marine microbes are able to find and take advantage of microscale nutrient patches. This could have important implications for trophic dynamics and nutrient cycling in the ocean. So following up to what Justin mentioned, how microorganisms, especially bacteria interact with their food in the ocean can have many effects on biochemical cycles.
So however, the ocean is, is very at steel, they're not addressed and there are many flow conditions such as turbulence, which is generated by the wind or the surface. And in the scale of microorganisms, what they're ober, what they're experiencing is not really the big turbulence, but it is the sheer and the smallest remnants of turbulence is known as the coog graph Eddie. And what we're trying to do here is that how to, what we want to study, what are the effect of this Eddie to the assuming behavior of microorganisms.
And to do that, we are making another channel using this mass that I showed you earlier on. And there is a, there is a cavity at the center. And let me just draw it to enlarge the diagram.
So this is how channel look like we, we have a, this is the view that we see from the top, and we inject the flow from the left to the right and due to the shear, it generates a re recirculating region inside this cavity. And this cavity is a representation of, of a graph Addis. We can just use the same setup that Justin has already explained earlier on when the syringe pump is off, and I didn't turn on any flow, you can see that the bacteria are swimming pretty much randomly and happily inside this cav cavity.
And there is no preferred orientation. And the, the movie that you're seeing is a time difference movie, which is exactly the same technique that Justin is using. So when the flow is very fast, as you can see, the bacteria are being vectored away by, by the vortex.
And the trajectory of the bacteria themself follows the streamline of the vortex. And what is actually more interesting is, is the following when we turn on the syringe pump. But the velocity is not so fast.
It's not too fast that bacteria, the bacteria are not able to fight it. But in a way, you can still pretty much see a vortex resemblance in this movie. So from this experiment, we've learned that in a, in the presence of a very strong, turbulent bacteria are in the mercy of the flow, they're not able to fight the flow.
However, in a mild turbulence region, they're still able to swim in their own accord.
