Self-Separating Biochip

  During the Summer of 2019, I worked in the Advanced Energy Systems and Microdevices Laboratory on this project. The main goal of this project is to develop a bio-chip that can, using only a few microliters of blood, can detect whether a person has a certain disease based on their antigens. For the chip to work, the blood must be placed into the hydrophilic silicone-based supply channel, seen to the left,  which will pull the blood across the channel in a self-sustaining manner. Upon reaching the middle of the supply channel, the plasma of the blood, containing the antigens, will be filtered into the filter region, also known as etched region, while leaving red blood cells in the supply channel. The filtered plasma enters the sensing channel which has the receptors to test for antigen presence. There are two filtration methods, one using a 3-d printed filter and one using a glass etching cream both of which created very small incisions in the glass slide to only allow plasma through.  

 My specific role focused on the glass etching cream method. I was tasked with characterizing the differences in flow-rates caused by altering parameters, such as the distance between channels or the amount of time the etching cream was on the glass slide, quantitatively. There were two components for my role, running the tests and developing a method to obtain quantitative flow-rate. 

The image on the right is an example of how the blood, in this case the blood mimicking fluid in the black and white dots, is flowing through the supply channel before being filtered through the etched region, in dark gray, before entering the supply channel (u shape figure. The filtered blood is seen as the white liquid within the supply channel.​

 While running the tests, I often ran into the same problem of the channels not sealing to the slide properly which enabled the testing fluid to leak out and ruin the test. I worked with my graduate mentor on possible solutions to this issue such as a laser-cut etching mask, stage clamp, channel placement device, and a slide holder. However, problems still persisted. After investigating the mold the channels were produced from, some residue was on some channels which may have caused the sealing issue. From here it was determined that either a new mold needs to be made or just the unaffected channels need to be used. 

 The other part of my role was developing a method to quantitatively determine the flow rate for successful tests. We were able to develop a MATLAB code that calculated the flow rate by inspecting, per frame, how much more fluid entered the channels. To see how accurate this method was, we compared the total volume given by MATLAB at a given time versus hand measuring the volume in ImageJ at the same time. The small sample size showed a 21.1% difference between the two methods showing the refinement to this code is needed.​Future work for this project includes running the tests again and then refining the code further.