Transforming Water Quality with Sensor Electrochemistry: Potential Benefits for Pollution Reduction in the Potomac River
Advisor: Dr. Grove
Internship Location: The MITRE Corporation, McLean Campus
Onsite Mentor: Dr. Joe Roberts
The Potomac River, which supplies drinking water to the 5.1 million people living on its watershed, has become increasingly polluted. According to Military Poisons and WILPF US Earth Democracy, a Potomac water sample in October 2021 revealed the contaminants include 7.8 ppt PFOS and 6.1 ppt PFOA. These concentrations are over 100 times each compounds’ maximum exposure level and over the maximum contaminant level of 4.0 ppt allowed in drinking water the EPA established for both this past April. PFOS and PFOA are types of PFAS, a family of man-made toxins with negative health effects such as higher risk of kidney cancer and liver damage. MITRE is working to develop a deployable sensor that will detect and remediate PFAS from natural waterways. Analytical electrochemistry allows for such sensors to be portable, so by manipulating factors like cell potential, measuring the current of the sensor, and working with different samples of PFAS, we can analyze the characteristics of an optimal sensor. I plan to conduct a literature review of sensor technology’s use to improve water quality in other rivers and compare electrochemical sensors with other types to conclude on the best potential practice of applying sensors to the Potomac. In the future, sensor technology could be applied to waterways all over America – experts estimate more than 200 million Americans are served by water systems with PFOS or PFOA in their drinking water at a concentration of 1 ppt or higher according to the Environmental Working Group.
Project Blog
Post 1: Introductory Post: Welcome to My Senior Project Blog
Hi, my name is Fiona Xu and welcome to my first Senior Project blog post!
For my project, I plan to conduct a literature review of the use of sensor technology to clean up PFAS contamination and improve water quality in rivers and the potential applicability of electrochemical sensors like the ones developed by my internship (MITRE’s McLean campus) to the Potomac River.
The first day of my internship is March 4, which is orientation. I'll receive all of the equipment I’ll be using such as a laptop (soon I’ll have two laptops!), RSA token, and a badge; I also expect to sit through some dense but informative presentations. At my internship, I’ll work closely with Dr. Joe Roberts, my on-site advisor, two senior chemists, and a lead materials scientist. After orientation, I'll undergo training for the first few days because I’ll eventually be conducting my own experiments and collecting, analyzing, and presenting my findings in weekly meetings.
I’m especially looking forward to working with the equipment they have. Most of my internship is heavy wet chemistry in the lab, directly working with PFAS (which may get nasty at times) and electrodes (the sensors!). My understanding is I will be manipulating the electrode’s cell potential and measuring things like its current. Some of my tasks will include cleaning electrodes, sampling electrochemical compounds like ferrocene, running experiments, and making stock solutions.
I’m really excited to start working at MITRE, even if I have to arrive there at 8am. It’s definitely going to be an adjustment from my usual start-of-the-day at 10:15am I was privileged to have in the first two trimesters. Nevertheless, I’m sure I will have an eventful and rewarding third trimester.
I hope you will take the time to read more of my blog posts in the future,
Fiona
Post 2: Week One: All About the Ducks
I have officially finished my first week at MITRE! My week started with lots of employee orientation activities. The bulk of those activities were introductory presentations and training videos. After receiving my badge and laptop (my arm is sore from carrying it around), I received a tour of the entire campus. There are four buildings to navigate (and get lost in), but I think I’ve done a pretty good job of finding my way around so far. Then, I met with my on-site mentor, Dr. Joe Roberts (Joe), and met my other coworkers who will work with and guide me throughout the next three months. They are wonderful and have been so incredibly kind and welcoming to me.
Additionally, I toured the Bio-Nanotechnology lab I’ll work in (my "home base"). Joe and I got to eat free from the cafeteria on Monday; I personally think the dessert is especially great. While I’m still getting used to all the websites and procedures, there’s already a lot of other things I’m excited about. Waking up early is becoming easier, and I take advantage of the free tea and coffee provided. MITRE also has these fancy elevators that I’ve never seen before – instead of pressing a button for up and down, you select your desired floor on a touch screen, and it redirects you to one of six carts. The last time I was in an office building was with my mom during “Bring Your Child to Work Day,” so it’s surreal that I’m back in one for this internship.
Other than orientation activities, I spent time familiarizing myself with the project I’m interning on and the main technique used in the lab: Cyclic Voltammetry (CV). I read an introductory guide to CV my coworkers shared with me and researched the basics of CV. CV is the technique I mentioned in my abstract and blog post last week. The current of an electrochemical cell is measured as an external power source – potentiostat – manipulates the cell potential (voltage) applied. Thus, we can observe the electrochemical behavior, or reduction and oxidation (the gain and loss of electrons, respectively), of a molecular species.
In this case, the molecule ortho-phenylenediamine (o-PD) is electropolymerized (joined together by electricity) on the electrode surface. PFAS molecules interact with the o-PD, causing a decrease in current that can be associated with the presence/absence of PFAS. The sensors are very small, rectangular gold-plated electrodes called screen printed electrodes (SPE); the chemistry only happens on the bottom half of it. I’ve included a simplified diagram of an SPE below for visual reference.
(Fiona Xu) (ACS Publications – American Chemical Society)
One of my coworkers took me inside the lab to show me how to use CV to perform something called Molecularly Imprinted Polymer (MIP) acid fabrication. In simple terms, it’s basically preparing the area on the sensor (the gold circle) PFAS will interact with. I then saw the characteristic CV duck curves (shown above) on the lab computers. As the cell potential is manipulated or the o-PD is reduced and oxidized, the current peaks on the graph are explained by the Nernst Equation, which I learned in my AP Chemistry class (yay for real-life applications). Depending on the conditions of the CV experiment, some curves look more duck-shaped than others. I completed lab training on Thursday, so hopefully I’ll start conducting experiments in the lab next week.
As for my independent research, I started skimming through and compiling sources about the Potomac River’s general properties. I’m also in the process of researching the history and current situation of PFAS contamination in the Potomac River. Before I research the applicability of these sensors to Potomac River water samples and per Joe’s suggestion, I need to understand how severe PFAS contamination in the Potomac is first and whether it’s a persisting issue. This was definitely a memorable first week.
Thanks for reading all the science-speak,
Fiona
Citations
Elgrishi, N., Rountree, K., McCarthy, B., Rountree, E., Eisenhart, T., & Dempsey, J. (2017). A Practical Beginner’s Guide to Cyclic Voltammetry. Journal of Chemical Education, 95(2), 197–206. https://doi.org/10.1021/acs.jchemed.7b00361
Post 3: Week Two: Conducting My First Sensor Experiment!
Hello, and welcome back to my blog!
Personally, I’m not a huge fan of beginnings. I recognize they are important, but I usually don’t like the first week of anything new because you don’t know what to expect, which I guess is both a good and bad thing. For example, my least favorite week of the school year is the first week. During those starting days you don’t know your overarching schedule or a general idea of how your weeks are going to be yet because you’re still in a transition period to get used to that schedule. Thus, I always go into them with some degree of anxiety. However, now that I’ve finished my second week out of traditional high school classes and navigating my internship life off campus, I feel more “at peace.” Navigating the unknown is also a huge part of science, so I feel like it all ties together.
Despite not being in school every day, I still hang out with my BIM friends. I try to maintain a good balance of working at my internship and socializing with friends every week. I think this [skill? practice? you get the idea] will help me in college because for the next four years of my life, I will have to actively carve out time in my schedule to hang out with friends.
Speaking of transitioning into new routines, I think I’m slowly getting the hang of things at my internship – meetings and the lab work are starting to make more sense :). In addition, I finally have lab access, so I can scan myself in and out of the lab and conduct experiments. On Thursday, I got my hands wet and did my first Acid MIP Fabrication. It definitely wasn’t perfect, but it was a great learning experience. I essentially prepared the o-PD electrode surface the PFAS molecules will absorb (adhere to a solid surface) onto. The analogy that was described to me was creating Swiss cheese, and the tiny holes are where the PFAS would fit into. My coworker also helped me start a new section in OneNote to be my lab notebook, and I wrote down the entire procedure of my Acid MIP Fabrication, helping me reinforce the steps in my brain as well. I will finish the sensing process next Monday. A picture of my lab setup is below.
(Fiona Xu)
I talked about this briefly in my abstract, but I realized I should probably give another more in-depth introduction to PFAS and the exigence of creating methods of detecting and remediating them (i.e. sensors) in nature here on my blog. PFAS stands for per- or polyfluoroalkyl substances and are synthetic or man-made compounds found in items such as firefighting foams, waterproof cosmetics, paint, and food-packaging. PFAS are nicknamed “forever chemicals,” so when PFAS enters our waterways, they will take a long time or virtually never breakdown; this is due to their strong carbon-fluorine bonds. The longer the chain, the harder to breakdown, but even shorter chains pose a problem.
As for the project I’m interning on, other than PFAS being a threat to human health and contaminating our waterways and drinking water, traditional methods for detecting PFAS like Liquid Chromatography and Mass Spectrometry (LC-MS) are highly efficient, but they’re also expensive, cumbersome, and not user-friendly, meaning it requires help from experts. On the other hand, cheap, portable, and user-friendly sensors often have low sensitivity, so without getting too jargony, MITRE is working to find the middle ground. We want the sensors to detect as low of a concentration of PFAS as possible (ppt or parts-per-trillion concentrations) to meet EPA regulations.
If you want to learn more about PFAS, here is a link to a video about the Minnesota PFAS incident publicized last year: https://www.youtube.com/watch?v=t8qGtEVh7oQ (Bloomberg Originals). It notably included the trial to pass Amara’s Law: a PFAS ban named after outspoken PFAS activist Amara Strande. Strande was diagnosed with a rare form of liver cancer after exposure to PFAS and unfortunately passed away before the passage of the law.
This week for my independent research, I continued compiling and started reading those sources regarding the Potomac River’s general characteristics and current PFAS contamination. Most of my sources so far have been annual waterway reports and journal articles. Finding sources proves difficult sometimes because I’ll click on a title thinking it’s about one thing, but then when I read through it it’s actually about something completely irrelevant to my topic. Nevertheless, I’ve managed to find a good number of sources by referring to the citations of papers from the EPA. While I read, I take note of the important bits of information related to my research topic. Next week I plan to look more into how sensors or similar technologies have already been used to clean waterways.
See you next week,
Fiona
Citations
Ajasa, A. (2023, July 14). She died fighting ‘forever chemicals.’ they still linger in her town. The Washington Post. Retrieved from https://www.washingtonpost.com/climate-environment/2023/07/14/pfas-forever-chemicals-cancer/.
Rehman, A. U., Crimi, M., & Andreescu, S. (2023). Current and emerging analytical techniques for the determination of pfas in environmental samples. Trends in Environmental Analytical Chemistry, 37. https://doi.org/10.1016/j.teac.2023.e00198
YouTube. (2023, November 9). The Forever Chemical scandal | bloomberg investigates. YouTube. https://www.youtube.com/watch?v=t8qGtEVh7oQ
Post 4: Week Three: Patience Please
Hi there, and welcome to week three!
Spring is officially here, and we got some really nice weather this past week. The cherry blossom trees outside of the MITRE buildings are in full bloom, and the flowers look beautiful. I love catching a glance at them every time I go through the walkways between each building. My family and I are planning to see more in DC soon.
I’m going to start off this post by discussing my independent research instead of ending with it. I outlined my final paper and most likely the content in my final presentation. I found some articles on sensor technology’s use to clean water, but I want to keep searching for more. Ideally, I’m hoping to find a case study involving sensor technology cleanup of a specific waterway. Lastly, I began sorting my information on the Potomac River into more specific, water quality-assessing categories like pH and temperature levels; the Potomac River is split into several sites.
Continuing this theme of unconventionality, this week was different from usual because most of my coworkers were taking their extensive LC-MS training. Thus, there were no meetings, and I spent most of my time in the lab running experiments. On Monday, I finished the sensing process with the electrode I prepped with Acid MIP Fabrication last Thursday. I recorded my procedure in my lab notebook for future reference. I repeated those steps by doing the Acid MIP Fabrication on Tuesday and sensing on Wednesday; this was the first time I conducted the experiment independently or without guidance from a coworker at each step, but I still asked questions when necessary.
Each experiment is two steps: Acid MIP Fabrication and Current Sensing. To recap, the first step is preparing the electrode surface on which the PFAS will absorb onto. It creates the mold that the PFAS molecules will interact with – this is done by imprinting or creating little holes in the polymer (o-PD) sitting on the surface. The acid used is hydrochloric acid (HCl), which helps create an ideal pH for sensing PFAS. This process takes about two and a half hours, and it involves micro pipetting PFAS into water and making solutions. I must be precise and careful with each step because I’m working with very small samples in units such as milligrams and microliters.
The second step is conducting the actual sensing of PFAS in the [deionized] water sample created, which is what I did on Monday. The sensing has a much more straightforward procedure than the Acid MIP Fabrication, but it takes about five hours. This is because when we measure the current of or conduct CV on the redox reaction on the surface of the electrode (characterization), we run three iterations, so the same thing is repeated multiple times. Additionally, each iteration scans for and creates three graphs: CV, Differential Pulse Voltammetry (DPV), and the Open Circuit Potential (OCP). Thus, a lot of the sensing is waiting for the graphs to generate.
A lot of time is also spent waiting for the electrode to soak in water vials and then water spiked with PFAS. The electrode is then characterized after 20-minute intervals, and the methodology we’re doing calls for the electrode to spend 60 minutes total in each soak solution. The three graphs I mentioned above are then consolidated into one aggregate (mean) current for each soak, which I will expand on soon. Since I’m still new to this experiment, both the Acid MIP Fabrication and Sensing took a bit longer than expected, so you can imagine why I was busy in the lab much of this week.
(All images: Fiona Xu)
Above are images from my experiments. The yellow-ish solution the electrode sits in while being characterized is ferrocene methanol in sodium chloride, which I prepare a new solution of each time I conduct the Current Sensing step. The purpose of the ferrocene methanol is to act as the mediator molecule that gives us a signal (the current measured) indicating the presence of PFAS, as a decrease in current indicates detection of some PFAS concentration.
After I collected all of my experiment data on the lab computers, I exported them into RStudio, the programming software we’re using to create graphs and do statistical calculations as I mentioned above. This is useful for when I’ll create presentations to report my data to everyone. Next week, I get to do this all over again.
Thanks for reading,
Fiona
Post 5: Week Four: First Setback
Happy Friday and welcome back to my blog!
I can’t believe the first month of my Senior Project/internship has gone by! It’s crazy how fast time flies. This week I mainly focused on my internship.
I continued to conduct PFAS Acid MIP Fabrication and Sensing experiments, but this time I did it with a different type of PFAS – PFBS (PFAS with four carbon-fluorine chains) and PFOS in water (or PFOS Water MIP). That just means the PFOS stock solution is water-based instead of acetone or methanol-based; the PFOS I used last week was dissolved in acetone, and the PFBS I prepared this week is dissolved in methanol. I made a 500 ppb (parts-per-billion) stock solution of PFBS and then later diluted that concentration of PFBS in water to 1000 ppt (parts-per-trillion) for sensing/detection.
Because both steps take hours, in addition to the electrode having to soak in a solution overnight or over weekend, the whole experiment is a two-day process. My original agenda was to complete the PFBS MIP Fabrication on Monday, Sense on Tuesday, complete PFOS Water MIP Fabrication on Wednesday, and Sense on Thursday. However, as I was in the middle of sensing on Tuesday, my MIP “burned”; essentially, the way the system measures voltage stopped working resulting in significant current going through the MIP, ruining it. Thus, the past day-and-a-half’s worth of work was all in vain. It was not a fun experience to say the least, but apparently, it’s a “milestone” that everyone runs into in the lab at least once.
Well, it was not entirely all in vain because I used the Sensing data I collected before my MIP burned to create a graph. The connection (white wire circled in the picture below) on the reference electrode broke, so the reference electrode became disconnected, and the computer could no longer recognize it. The reference electrode is the long, glass structure with a wire of silver inside behind the gold electrode, and it serves as the reference voltage for the experiment. There is an extremely thin and delicate wire inside the harness, and the continual usage sometimes causes them to break.
Usually, when characterizing I get the beautiful duck curves seen during CV as well as the DPV and OCP graphs, but the malfunction caused me to only see a single dot, and the resulting graph was just blank. Since the system can’t measure voltage, it overcompensates which usually damages the electrode. After my coworker helped me replace the wires, I was able to obtain measurements of the current again.
Electrode in the PFBS Solution (Fiona Xu)
Determined to make up for the time already lost, I quickly (but efficiently) redid the PFBS MIP Fabrication during the remainder of the day. I was proud of myself for finishing the Fabrication in record time; before, I would stop between each step to triple-check I was doing each task correctly, but now I’m becoming more confident in the procedure. Then, I sensed on Wednesday, and Thursday I completed the PFOS Water MIP Fabrication. When I return from the weekend I will sense with it. Even though everything in my agenda was pushed back one day, I’m glad I didn’t let the setback faze me.
I also presented my own experimental data in our weekly meetings for the first time: I presented my PFOS Acid MIP and Sensing Run data from last week on Monday and the PFBS Acid MIP and Sensing data on Friday. After I finish my data analysis in RStudio, I compile the graphs and statistics into a PowerPoint to show my results. My results agreed with previous experiments from the lab! There is a steady drop in current during the experiment which sometimes makes it difficult to determine if PFAS is detected. There is still a lot to investigate regarding the reasoning behind such behaviors (i.e. why does the current drop X amount from this step to the next?).
That concludes my work for this month and I’m looking forward to April!
Fiona
Post 6: Week Five: New Developments
Hi, and welcome to week five!
It was another busy week in the lab – the experiments I did this week were Dip Method Sensing with last Thursday’s PFOS Water MIP on Monday, PFOS Water MIP Fabrication and Time Trial Method Sensing on Tuesday and Wednesday, and another PFOS Water MIP Fabrication on Thursday. One of my coworkers will sense my Thursday PFOS Water MIP on Friday.
The Dip Method Sensing I performed on Monday was slightly different than what I’ve previously done as well. I still let the electrode soak in an aqueous sodium chloride solution over the weekend, but this time, instead of switching and letting the electrode soak in water before characterizing, I let it sit in the same sodium chloride solution for 20-minute intervals before characterizing or scanning for the current voltammograms.
I characterized the electrode after 20, 40, 60, 80, 100, and 120 minutes, for a total of six data points. Then, I inputted this data into RStudio to obtain the numerical peak current values. Additionally, instead of creating a box-and-whisker plot, I took the six current values from RStudio and plotted them against the time intervals on a scatterplot in Excel. The general trend I saw was stability in the first hour and a slight current decay in the second.
However, the 120-minute data point is most likely erroneous because I accidentally bumped into the fume hood cover after rinsing it and dropped it. At first, I panicked a little because I thought I’d lost it but turns out it was just behind the plastic glove box container right underneath (phew!). The MIP on the electrode surface is extremely sensitive, though, and because I dropped the electrode, the MIP must’ve dried and/or accumulated foreign particles like dust, affecting the current measured. It definitely wasn’t the greatest way to start the week, but I was still able to provide useful data.
Letting it sit in the same over weekend sodium chloride solution was an important factor because we were attempting to determine if switching between vials was having an adverse effect on the stability of the background (the soak solution). My experiment supports the fact that it does, as keeping the MIP in the same solution seemed to provide more stability. This is important because we want the PFAS to be the only cause of current decay as much as possible so that a decrease in current is indicative of PFAS detection.
For the PFOS Water MIP Fabrications, I increased the amount of 500 ppm PFOS in water (seen in the image below) from 5 mL to 15 mL this week. Basically, that’s a huge amount of PFOS in the water sample. What was more interesting is that I kept acquiring CVs that showed currents that were way lower than what was expected. The current was expected to be low with such a high concentration of PFOS, but the CVs I obtained were somewhat wonky, too. I don't think I'm completing the procedure incorrectly, but I'm going to keep investigating why I'm getting such results.
(Fiona Xu)
As for the Time Trial Sensing on Wednesday, the procedure was the simplest I’ve done by far. After preparing the ferrocene methanol and characterizing the working reference electrode to make sure everything was working smoothly, all I needed to do was take my electrode out of the sodium chloride overnight soak and run 13 iterations of the characterization. Essentially, most of the time was spent waiting for the characterization to finish. After it was finished, I repeated the same data analysis process as the Dip Sensing and created a 13-data point scatterplot on Excel.
Other than lab work, I organized my literature review sources to the appropriate section of my paper and started identifying five specific sites of the Potomac River that will be the focus of my project. The Potomac surrounds four states - Maryland, Pennsylvania, Virginia, and West Virginia – in addition to Washington D.C., so I chose one site in each state and D.C. to research PFAS contamination and water quality characteristics on. Ideally, after I finalize the five sites, I’ll collect water samples from each to bring them into the lab, but I’ll see what time, resources, and conditions allow. I plan to collect the Virginia Potomac water sample very soon. I’ll have to be careful because I don’t unintentionally contaminate my water sample when I collect it, so that means not wearing or bringing any products that have PFAS in them such as waterproof clothing.
Until next week,
Fiona
Post 7: Week Six: PFAS Makes the EPA Go Round
Hello!
Welcome to week six of my blog. I’m not sure why, but this week felt longer and slower than usual. I guess from the solar eclipse on Monday to me taking Wednesday off and coming in on Friday instead, my normal rhythm was “thrown off.” On a completely unrelated note, I’ve gotten a lot of my coworkers into Hi-Chew candies, and I even traded Hi-Chews with one. I just think it’s funny how that’s become a frequent discussion topic.
I conducted three experiments: a PFOS Water MIP Fabrication and Time Trial Sensing on Monday and Tuesday, a Control Sensing Experiment on Thursday, and another Control on Friday. The PFOS Water MIP and Time Trial Sensing was a repeat of what I did last week. I still combined o-PD with 15 mL of 500 ppm PFOS in water for the MIP; however, the CV I obtained for the Acid MIP Fabrication was the expected current! Last week, I consistently saw “wonky” CVs with currents lower than expected and/or had unexpected spikes, so I was happy to finally obtain the “normal” one (see image below).
(all images: Fiona Xu)
The Time Trial I did this week was also the same procedure as last week’s, but I changed the iteration and wait time to four iterations and 38 minutes, respectively. Thus, I was measuring the current every 40 minutes because the scan takes two minutes. Afterwards, I created a four-data point scatterplot on Excel. There seems to be a slight decay in current over time (~0.5 microamps), but overall, it remained pretty stable.
It was my first time doing the Control Experiment. For this experiment, I was sensing with a bare gold electrode, so there was no molecularly imprinted polymer or electropolymerized o-PD with PFAS molecule imprints on the electrode surface. After I ran the first three characterizations, I let the electrode soak in a water vial for 20 minutes and then characterized again after placing the electrode back in the ferrocene methanol.
I repeated these steps until it soaked for a cumulative time of 120 minutes. The purpose was to have a standard of comparison we can compare the data from the MIP experiments to. We want to confirm that the electron mediator (the ferrocene methanol) is stable so that it doesn’t interfere with the sensor’s PFAS detection. In addition, I was also attempting to replicate one of my coworker’s results of the same experiment. We suspect that one of the potentiostats is not as stable as the others, which is causing there to be extra decay in the current.
My Friday Control Experiment was the same as my Thursday one, but I measured the current after every 40 minutes of letting it soak in water. I don’t have all the data points yet, so I’m not exactly sure what the trend of the graph is; I plan on analyzing the data on Monday.
I think it’s also worth mentioning that this past Wednesday, the EPA set the first ever national, enforceable legal limit of PFAS in drinking water. Most notably, the Maximum Contaminant Level (MCL), which the EPA defines as “the highest level of a contaminant that is allowed in drinking water. MCLs” for PFOS and PFOA is 4.0 ppt. This means that 4.0 ppt is the highest tolerated level, and public water systems are now legally required to act and reduce PFAS levels in water if its current levels exceed the MCL. This new policy is a testament to the growing need for PFAS monitoring and detection methods.
The values or “compounds’ safe level in drinking water” (0.004 ppt for PFOA and 0.02 ppt for PFOS) I referred to in my abstract were not enforceable. I’d also like to make a correction to my abstract because those values are the maximum exposure values to PFOS and PFOA from any source (not just water). Any level of PFOS and PFOA in water is technically “bad.” The MCL goal of PFOS and PFOA in drinking water the EPA set is 0 ppt.
That’s it for now, and next week I will discuss points I found interesting in my source readings.
Fiona
Citations
Environmental Protection Agency. (2024). PFAS National Primary Drinking Water Regulation [Fact sheet]. FACT SHEET (epa.gov)
Environmental Protection Agency. (n.d.). Questions and Answers: Drinking Water Health Advisories for PFOA, PFOS, GenX Chemicals and PFBS. EPA. https://www.epa.gov/sdwa/questions-and-answers-drinking-water-health-advisories-pfoa-pfos-genx-chemicals-and-pfbs#q4
Post 8: Week Seven: A Rough Week
Hi blog, and welcome to week seven!
This week has been quite eventful. I came into the week with the plan to run two experiments: a PFOS Water MIP but with a slower scan rate and soak time of 40 minutes on Monday and Tuesday and the same thing but with HCl acid soaks on Thursday and Friday, but that plan quickly went awry.
Everything went as planned on Monday. I did the Acid MIP Fabrication (“Fab”) with the same measurements I had last week – o-PD, HCl acid, and 15 mL of PFOS in water. I created the MIP, characterized the electrode at each necessary step, saved all my files, and uploaded the information to PowerPoint; nothing went wrong. However, Tuesday afternoon is when things started going downhill, and it unfortunately remained that way for the rest of the day.
As I was in the middle of sensing, the white wire connected to the reference electrode broke. I had the exact same thing happen to me not long ago, which I wrote about in a previous blog post. The CV and DPVs I was getting were the same jagged ones I had seen a little less than a month ago. Once the wire was switched, and I made a couple tiny adjustments (i.e. flipping the direction the gold electrode was facing), things seemed to be working again. This is my second MIP that got burned, and we’re going to look more in-depth into the problem with the wires so that hopefully no one will have to face this issue anymore.
It didn’t end there, though. Because my MIP was burned, I started over with the fabrication process. Everything was ran smoothly and the Acid MIP Fabrication CV looked great (like the one pictured in my blog post last week), but right as I was about to finish the experiment and head out for the day, I observed an abnormal CV (image below) and DPV for the last step - the extraction (removing the PFAS molecules to make the “swiss cheese mold”/imprinted polymer). What I should have seen was a normal duck curve, so the CV I got suggests that my MIP was not fully extracted. Thus, I had to scrap my second attempt at the experiment, too. I feel like I’ve set a record with my two failed MIPs in one day.
(Fiona Xu)
Long story short, I was stuck trying to conduct the same experiment for this entire week. It makes me feel better to know that the experiment going south was due to something out of my control and not because of a procedural error I made, but it was still irritating that I basically got nothing done on Tuesday :(.
I also had plans to collect a water sample from the Great Falls Park (Patowmack Canal) this week, but things didn’t fall into place. I will most likely collect it next week, as I have the supplies prepared for field sampling: gloves, plastic container, cooler, and a permanent marker. Additionally, I started drafting my final Senior Project paper.
Despite the bad luck I ran into this week, I want to also focus on the positive things that happened. On the bright side, I had no issue completing the experiment on Wednesday and Thursday. It’s also taking me less and less time to prepare the solutions needed for experiments. I feel like I’m developing the “muscle memory” of measuring out the o-PD and ferrocene methanol and pipetting the aqueous stock solutions and acid.
My coworker also shared with me a tip on how to make the sonicating step quicker: wiggle the vial to accelerate the dissolving of the solid pieces. For example, the o-PD, PFOS, and acid solution would take up to 15 minutes to sonicate at times, but now it’s consistently shortened to five minutes. The sonicator uses high frequency sound waves to break down large particles (in our case solid o-PD or ferrocene methanol) and homogenize solutions.
Vial sitting in the sonicator bath (Fiona Xu)
Hopefully I’ll be able to follow through the plans I set up next week. See you then!
Fiona
Post 9: Week Eight: Sensing, Sensing, and More Sensing
Hi blog,
Wow, it’s the last April post and week eight already! This second month has truly flown by in a blink-of-an-eye.
In contrast to last week, this week I was able to get a lot done and completed three experiments. First, I repeated the same PFOS Water MIP experiment I did last week, with Fabrication on Monday and Sensing on Tuesday, but with an acid soak instead. Basically, instead of letting the electrode soak overnight in pure water, I added 200 uL of hydrochloric acid or HCl to the water vial (see image below).
Then, for the sensing, I characterized the electrode after taking it out of the overnight soak for the first time, and after that was done, I let the electrode sit in the same overnight water-acid solution for 40-minute intervals before characterizing. It spent a cumulative time of 160 minutes soaking. The scan after 80 minutes was sensing for 50 ppt PFOS, a very small amount as when I was micro-pipetting the PFOS into the vial, only a tiny drop was dispensed, and the scan after 120 and 160 minutes was sensing for 250 ppb PFOS. I micro-pipetted more PFOS into the vial after the 80-minute scan finished.
On Wednesday, I conducted another control sensing experiment with a bare electrode surface, but this time it was with an acid soak. For this experiment, I also prepared a vial of water with 200 uL HCl. Thus, during sensing, I repeatedly dipped the electrode back into the same vial for 40-minute intervals for a total of 120 minutes in the soak solution and four data points. This experiment only took one day because since this is the control, there was no overnight soak, so the first characterization (t0) was a “fake” post overnight soak scan.
Later that afternoon after I finished the control experiment, I fabricated a PFBS Water MIP. It was also a completely different procedure because I was doing it with a different ratio of o-PD:PFBS in solution and with an extraction solution of acetone and water. Before, all of the PFOS Water MIP extractions had been with methanol and water. PFBS is more soluble (dissolves to a greater degree) in acetone than methanol, so using more soluble solvent allows for more complete removal of template (PFBS).
I finished the sensing on Thursday; the electrode soaked in solution for a cumulative time of 160 minutes, and the t120 and t160 characterizations were sensing 500 ppb PFBS. I talked about my results for the PFBS Sensing with my coworker, and she and I have seen similar trends in our data. Lastly, I consolidated the data for all these experiments into a PowerPoint, which I am improving at organizing into slides every week.
I'm still having a little bit of trouble with the thin white wire connecting the reference electrode to the potentiostat in the fume hood (the other station's harness is working fine for me). My coworker helped me troubleshoot and look for anything that could be disrupting the scan such as not having water at the top of the electrode (where it connects to the port on the lid connected to the potentiostat), so I think as long as I watch out for those details and apply the tips she gave me, I shouldn't have any issues characterizing.
As for my research, I finally found a case study regarding the use of sensors to detect water contamination. Ten sensors were placed in multiple nodes of a contaminated water system in a town in Madrid, Spain to investigate how successful the detection ability of the sensors based on two factors: the volume of contaminated water and time detection. I'm also in the midst of writing my final paper.
To end this post, I just wanted to show a picture of the scrumptious, red-velvet waffle-on-a-stick that was being sold this week :).
Until next week,
Fiona
Citations
Shahra, E. Q., & Wu, W. (2020). Water contaminants detection using sensor placement approach in Smart Water Networks. Journal of Ambient Intelligence and Humanized Computing, 14(5), 4971–4986. https://doi.org/10.1007/s12652-020-02262-x
All images: Fiona Xu
Post 10: Week Nine: Things Are Heating Up
Happy May, blog!
It’s starting to feel more like summer because it was so hot this week - both outside and inside MITRE. This entire week we’ve consistently gotten temperatures in the high 80s to low 90s; while driving, I constantly had the AC on full blast. In addition, it was also extremely toasty inside the Bio-Nano lab. On Monday, it was super warm in the conference room, and starting Tuesday, the lab itself was hot as well because the AC unfortunately broke. The lab staff are working hard to fix it, and two huge fans were brought in as substitute so at least the temperature wasn’t unbearable, but what a week for the AC to break.
Even with the whole AC fiasco in the lab, I performed three experiments this week. On Monday and Tuesday, I did two more control experiment with the bare electrode in the acid soak, and then I finished the week with a PFBS Water MIP experiment by fabbing (performing the fabrication) on Wednesday and sensing on Thursday.
The Monday experiment was an exact copy of the one I did last Thursday. I prepped the vial with 20 mL water and 200 uL HCl, and after characterizing the bare electrode before it spent time soaking in the vial, I let the electrode soak in the same vial and characterized every 40-minutes. The bare electrode spent a cumulative time of 120 minutes in the soak solution, so there were four data points for each 40-minute increment (0, 40, 80, 120).
My original intention was to carry out this experiment to 200 minutes, but I had to leave before I could get to that time point. Thus, this experiment ended up serving the purpose of seeing if my results are consistent with last week’s and being another trial of the acid soak Bare Dip series experiment. Comparing the trials, they both showed good stability and there was less than a 0.5 microamp net change in current for both. However, the net change in current was negative last week while the net change in current for my Monday trial was positive (something my other coworker was seeing as well).
On Tuesday, I had more time to complete the experiment, so I performed the same one with the bare electrode soaking in the water-acid soak for a total of 200 minutes. The current only decayed ever so slightly for the first 160 minutes and increased by a tiny amount during the 200-minute scan for a small net change of about -0.1 microamps. Overall, there’s still a lot to investigate regarding the background and electron mediator (ferrocene methanol) stability.
I worked with PFBS again the second half of the week, but this time the PFBS Water MIP was with a 10:1 ratio of o-PD:PFBS; last week’s was a 5:1 ratio. Something interesting I noticed was that during the Acid MIP Fabrication, there was an orange hue around with working electrode (see in image below). It stained the clear solution after I took the electrode out to characterize the full MIP in the ferrocene methanol. My coworker said she has also seen this during her experiments, but right now she’s not entirely sure why this occurs. For sensing, the electrode soaked in solution for a cumulative time of 160 minutes, and the t120 and t160 characterizations were sensing 250 ppb PFBS.
Oh, and an update regarding the lingering issue with the white wires I mentioned in previous posts: we finally found a solution for them! The lab bought a chipboard with several alligator clips, which serves as a different way to connect the screen-printed electrode to the potentiostat without having to use the fragile harness. An alligator clip is used to connect the reference electrode to the potentiostat (see in image below). Temporarily, there is only one in-use right now, but three more are expected to arrive and will eventually replace all the white wires.
I also plan to collect water samples this weekend so I can bring them into the lab next week. I hope to use the screen-printed electrodes on the samples, as the data collected would help me finish my final Senior Project product.
See you next week,
Fiona
Citations
All images: Fiona Xu
Post 11: Week Ten: When It Rains, It Pours
Hi Blog!
It’s been a crazy (in a busy and eventful way) week. Why? Taylor Swift has a new setlist for the European leg of her tour!
I’m just kidding (but in all seriousness, the sentence above is big news). I finally collected water samples from the Potomac River this past weekend, and it was a pretty long process. I originally planned to go out and collect them on Saturday, but it was raining all day, so I couldn’t. I wanted to avoid collecting it on a rainy day because the rainwater forms a layer on top of the river before fully mixing in with its water, so if the origin of the sample doesn’t accurately reflect the rest of the water in the river, the sample won’t either. Ideally, the water sample should be representative of the entire river.
Then, after an entire weekend of gloomy skies and rain, my dad graciously drove me to three different sites on the Potomac River in the DMV on Monday.
-Georgetown Waterfront Park (Georgetown, Washington D.C.)
-Riverbend Park (Great Falls, VA)
-Algonkian Regional Park (Sterling, VA)
First, I collected the sample from Georgetown Waterfront Park, followed by the Riverbend Park, and ended with the Algonkian Regional Park because it's closest to my home.
One disclaimer I’d also like to point out is that the collection of water samples was not perfect due to resource limitations (i.e. specifics of where you should collect them – I had to collect them from the riverbank instead of from the middle of the river), but I tried my best to mimic the ideal conditions. For example, after I collected each sample, I stored them in a cooler before transferring them to the fridge in the MITRE Bio-Nano lab. They have to be kept in cold temperatures because the cold slows down natural processes that change the water sample's content, preserving the original composition as much as possible. I’ve included images of both below.
I’ve gotten the samples into the lab, but there’s still one more step. Because the river water has a lot of sediment and bacteria, amongst the other gunk in the water, the water needs filtering, so the sensor doesn’t detect for other things besides PFAS. Thus, my coworkers helped me use vacuum filtration to clean out the water samples. I stored the filtered water in new plastic bottles and relabeled them with the site name, my name, and date. As you can see in the image below, the filter paper was quite dirty.
I unfortunately don’t have a picture of it, but the Algonkian Regional Park sample was especially filthy, as the entire filter paper was a deep brown color with sediment dispersed on throughout it. Now that the water’s been filtered out, I started performing sensing experiments with them; I created PFOS MIPs, so I’m seeing if I can detect PFOS from the samples. This week I started with the Georgetown Waterfront sample (as seen below), and next week I’ll continue with the other two sites. I’ll talk more about the specifics of the experiment in my next post. The data from these experiments is what I’ll analyze and show in my final presentation.
I ended the week by performing a PFBS Water MIP with a 1:1 ratio of o-PD:PFBS. What’s different about this PFBS experiment compared to last week’s is I’m letting the electrode dry for 30 minutes post fabrication and using methanol in the extraction solution instead of acetone. I’m still in the process of completing it, but from the DPVs I’ve seen while characterizing, it seems like the methanol wasn’t strong enough to extract the MIP compared to the acetone.
That’s it for this week and since there’s t-minus two weeks since the Senior Project showcase, I’m going to spend a lot of time working on my paper and presentation between now and then! It’s insane to think that my Senior Project experience is coming to an end.
See you next week for my last blog post (I can’t believe I’m saying that),
Fiona
Citations
Daniels, M., B.J. Austin, and B.E. Haggard. 2017. Collecting Water Samples Is So Easy, Anyone Can Do It!. Arkansas Water Resources Center, Fayetteville, AR, FS-2018-01: 07 pp.
Metzger, M. R. (2014, October 9). Best practices for collecting water samples. Water Technology. https://www.watertechonline.com/home/article/15546332/best-practices-for-collecting-water-samples
All images: Fiona Xu
Post 12: Week Eleven: The End of the Beginning
Hello, blog!
Welcome to my last blog post. After 11 weeks of researching PFAS, experimenting in the lab, connecting electrodes, keyboard typing, and making and learning from mistakes, I can’t believe we’re finally here.
In this final post, I’ll expand on the experiments with the Potomac River environmental samples I started last week. I continued sensing with my two other water samples: Riverbend Park (Great Falls) and Algonkian Regional Park (Sterling). In effort to minimize the effects of any outside variables, I performed these experiments using the same computer and potentiostat setup in the fume hood and followed the exact procedure I had with the Georgetown Waterfront sample.
This means all of the electrodes are fabricated with a PFOS Water MIP before sensing PFAS in the Potomac River samples. The MIP solution included:
-500 ppm in PFOS in water
-o-PD
-small amount of HCl
Then, after the MIP on the electrode surface was fully prepared and characterized, I placed the electrode in a methanol and water extraction solution to remove the PFOS molecules from the MIP.
For sensing, I first characterize the MIP after it sits in water overnight (which in this case is only 30 minutes because long story short, the extraction solution wasn’t fully extracting the MIP). The electrode sits overnight in a water vial so that any residual methanol on the MIP gets rinsed off. While the characterization was running, I prepared the vial of the desired Potomac Water sample. Since the samples have been sitting in the fridge, they need to warm up to room temperature for the PFAS molecules to diffuse into the MIP imprints because warmer temperatures increase the rate of diffusion.
Then, I let the electrode soak in the same water vial for 40 minutes and characterized immediately after; I performed this step twice. Now, it’s time to place the electrode into the water sample. Like the previous step, I leave it in for 40 minutes and characterize. I’ve included images of this below.
However, because the concentration of PFOS in these samples are unknown, I performed an extra procedure called standard addition in order to trace back the concentration of PFOS present. The idea is to spike the water sample with known concentrations of PFOS, measure the currents after the electrode is exposed to them, and create a calibration curve with the data points to find the unknown concentration. Current is on the y-axis, and concentration on the x-axis; the unknown concentration is the x-intercept.
Thus, I micro-pipetted in 100 ppt of PFOS into the sample, characterized the electrode after it soaked in the PFOS-spiked water vial for 40 minutes, and repeated the same thing with 500 ppt and 1000 ppt PFOS in the sample. Once the data was exported to my laptop, I analyzed the data with RStudio and plotted the points on a scatter plot with Excel. From this first experiment, the sensor seems to be in the ballpark by a factor of ten for sensing the actual amount of PFOS present (referenced from online sources).
I ended the week by performing another trial of the Georgetown Waterfront sample experiment. I want to have at least two trials for all three samples, so I hope to complete the second trials for Riverbend and Algonkian next week. This data will enhance and help me finish my Senior Project paper and presentation.
It’s time to go, and I hope my blog was both interesting and informative. I’ve can confidently say I learned a lot, and I thoroughly enjoyed the entire Senior Project experience. I feel like I’ve also gained lots of valuable experience that I’ll bring with me to my future work.
Even though the official Senior Project is coming to an end, this doesn’t mean my journey with this project/topic is over. My internship will be going on for a couple of more weeks, and I’ll continue following any new updates and research regarding PFAS sensing after my time with this project and MITRE ends. Exactly what that looks like, whether it’s simply through reading online news or picking up similar projects in college, I’m not sure yet, but I’m excited for what the future holds. :)
Bye for now,
Fiona
Citations
All images: Fiona Xu