Student Paper Competition: Environmentally Friendly Acoustic Design, Spatial Impulse Response Measurements, and Acoustic Spectrometers

This episode showcases the latest winners of the POMA Student Paper Competition: First, Jonathan Michael Broyles (University of Colorado, Boulder) discusses his database to help acoustical consultants design more environmentally friendly spaces. Next, John Latta (University of Nebraska - Lincoln) shares his work regarding spatial impulse response measurements. Finally, Michelle Ruth Crouse (California State University, Dominguez Hills) talks about the acoustic spectrometer she created using off-the-shelf parts. 

Associated papers:

Jonathan Michael Broyles and Wil Srubar, III. "A comprehensive dataset of environmental emissions, health, and manufacturing information of building acoustic products in North America." Proc. Mtgs. Acoust. 55, 015002 (2024) https://doi.org/10.1121/2.0001997.

John S. Latta and Lauren M. Ronsse. "An analysis of spatial impulse response measurements and their ability to validate spatial features within acoustic models." Proc. Mtgs. Acoust. 55, 015001 (2024) https://doi.org/10.1121/2.0002004.

Michelle R. Crouse, Małgorzata Musial, Jason A. Widegren, Jacob Pawlik, Bryan Bosworth, Nathan Orloff, Aaron Hagerstrom, Angela C. Stelson, and Robert Lirette. "A low-cost ultrasonic absorption spectrometer mainly using off-the-shelf parts."Proc. Mtgs. Acoust. 55, 030002 (2024) https://doi.org/10.1121/2.0002003.

Learn more about entering the POMA Student Paper Competition for the Spring 2025 meeting in New Orleans.

Read more from Proceedings of Meetings on Acoustics (POMA).

Learn more about Acoustical Society of America Publications.

Transcript

ASA Publications (00:26)

It's time for the next round of POMA student paper competition winners, this time from the online meeting we held in November. First up, I'm talking to someone we've heard from on this show before, Jonathan Michael Broyles. We'll discuss his article, “A Comprehensive Dataset of Environmental Emissions, Health, and Manufacturing Information of Building Acoustic Products in North America.” Congrats on another award and thanks for taking the time to speak with me today. How are you?

Jonathan Broyles (00:49)

I'm well, Kat. Thanks again for having me.

ASA Publications (00:52)

Yeah, of course. So first tell us a bit about your research background. What have you been up to since we last talked to you?

Jonathan Broyles (00:58)

Yeah, kind of a loaded question to answer about the research background. So do a lot of different research projects. My research is highly interdisciplinary. Those who don't know me, I have a background in structural engineering, architectural acoustics, computational design, and sustainability. I do a lot of projects at the intersection of all those different topics, but more recently, I've been kind of picking and choosing certain ones or combinations of specific topics. 

So since we last talked, there's been a number of things that have happened me professionally. So I was still a graduate student at Penn State, completed my PhD in architectural engineering. I have since successfully I passed and graduated. So that's a bit, yeah—

ASA Publications (01:38)

Congrats! That's exciting.

Jonathan Broyles (01:40)

Thank you. Not insignificant, now I'm currently a postdoctoral research associate at the University of Colorado in Boulder. I'm focusing on embodied carbon emissions in buildings, which is kind of a broad topic, but I'm doing a handful of projects, so it kind of summarizes that well. But more recently, maybe a new update, and I know this is dropping in the summer, is I will be a post-doctoral impact fellow at MIT's Climate and Sustainability Consortium later this summer. So that'll be really fun. It will continue the sustainability thread and interests of mine.

It'll be more on buildings in the building sector, but it's not limited to the building sector.

ASA Publications (02:22)

Awesome, that sounds very exciting and sounds like good, important work. So let's get into what you were talking about. Your research has to do a lot with the decarbonization of building materials. Can you give us a bit of background on what this entails?

Jonathan Broyles (02:35)

Sure, yeah. So specifically for this project and paper, I'm looking at ways to reduce embodied carbon emissions, which are the carbon emissions associated with like the material extraction, manufacturing, and transportation of different materials or products, specifically for the built environment. So I was looking at acoustic products in my paper. But embodied carbon emissions account for 7 % to upwards of like 11 or 12 % of global carbon emissions, which is huge. So because we're trying to reduce our carbon footprints in the built environment, we also have to consider these embodied carbon emissions in addition to operational carbon emissions.

ASA Publications (03:13)

Okay, okay, got it. So how can acoustic consultants help reduce a building's carbon footprint?

Jonathan Broyles (03:20)

Great question. So, acousticians can participate in decarbonizing buildings through their material selection. So, typically, acousticians have not been at this round table of being able to find levers to reduce carbon emissions in buildings. However, because acousticians can inform material selection, like they select specific acoustic products to be in buildings that have obviously acoustic benefits, we also need to consider the carbon emissions of those acoustic products or treatments. So being mindful of the carbon emissions of those acoustic products can help, again, further decarbonize our buildings.

ASA Publications (04:00)

Okay, okay, that makes a lot of sense. So what are the limitations of existing acoustic product databases?

Jonathan Broyles (03:20)

It's a good question. One of the biggest challenges with finding out the carbon emissions related to acoustic products is a lot of it is disjointed. You might have an environmental product declaration or an EPD that's available on a specific manufacturer's website for one specific type of acoustic product. However, that information might be on one website, but a different manufacturer's product could be on a different site, and it might be hard to just decipher and collect all of this information and put it in one specific space. So that was the point of this work was to help compile some of that.

ASA Publications (04:43)

Okay, that makes sense. Save the work for other people by doing it yourself. 

Jonathan Broyles (04:50)

Exactly.

ASA Publications (04:51)

So can you explain the environmental and health impact data that acousticians would want to consider when choosing building materials? And is there anything else that they would want to consider?

Jonathan Broyles (05:02)

Yeah, good question. So regarding environmental information, there's many different environmental metrics or midpoints that are used when conducting an LCA, which an LCA essentially is just a tool or a method to help quantify the environmental impacts of a specific product. So global warming potential is the metric that is most often used or referenced when talking about global climate change. A higher global warming potential generally means that there's a higher potential to increase global temperature. So minimizing our global warming potential, again, helps to decarbonize buildings and minimize potential to increase climate change. Now, there are many other things we have to consider in addition to GWP, or global warming potential, but that's kind of the big one to start with. 

I didn't mention this on the last question, but we have to consider more than just environmental metrics. So you mentioned health. So our acoustic products, we need to make sure that they are, I'm saying this with air quotes, but are healthy. We don't want to have acoustic products in our buildings that have potentially harmful contaminants or chemicals that could potentially be, again, harmful to the building users or occupants. Things like VOCs, which are volatile organic compounds, could be potentially harmful, again, to people in our buildings. So we want to make sure that the products that we implement and design with are safe, are healthy. So, now, that information is not typically in environmental product declarations. That's obviously more environmentally focused, but products, or more documents specifically, like health product declarations would, like, communicate that information. Like if there are any VOCs or any other harmful contaminants or if the product is safe. So making sure that that information is available for the product is good because if a product doesn't have a health product declaration associated with it, then typically there might be something the manufacturer doesn't want say or communicate. But if a product does have it, then it typically is safe. So that information is definitely useful when making wise and sustainable decisions with reference to acoustic products. 

And then lastly, manufacturing information is also helpful to know. And specifically, like where is the manufacturing plants of the acoustic product being made? Like, is the acoustic product being made in a neighboring city? If so, that's excellent because the transportation-related carbon emissions would be minimal, just because it's so close and locally based hopefully. Or are you shipping the product from, you know, another country or across, you know, the ocean. If so, then the carbon emissions associated with that transportation probably are much larger. So, manufacturing information, specifically where the manufacturer is located, is also important information to help make more sustainable decisions.

ASA Publications (08:00)

Right, right. That also makes quite a lot of sense. So how did you compile your acoustic product data set?

Jonathan Broyles (08:06)

Yeah, so I did the heavy lifting for us. I went on several different manufacturers’ websites, looked at all of these different environmental product declarations, health product declarations, and just tried to compile and synthesize this information into one spreadsheet, which is open source so people can access it for free.  It has information for over 100 products.

ASA Publications (08:29)

Awesome, yeah, you did the dirty work then for everybody else to benefit. So in this you presented a case study of how the acoustic product data set could be used. Can you walk us through the scenario?

Jonathan Broyles (08:40)

Yes. So in this case study, we were looking at an office lobby and we're trying to find the most sustainable acoustic product. So we wanted to find an acoustic baffle that had a high NRC. So we filtered the acoustic data sets to find an acoustic baffle with a high NRC. Then we looked at what baffles were also reported as safe, so didn't have harmful contaminants like VOCs, again. And then also just, again, had a low carbon footprint, so we essentially to conduct a life cycle assessment using the GWP information in the data set to find what was the lowest carbon solution.

ASA Publications (09:18)

Okay, got it. So what are the implications of your work for acoustic consultants?

Jonathan Broyles (09:24)

Yeah, the implication, the biggest thing is, you know, an acoustician can come to this data set and use it to make smart, informed, sustainable decisions. That's the biggest thing is that we can start to, again, have more information to make wise and sustainably conscious decisions.

ASA Publications (09:42)

Okay, okay, great. So what were the limitations of this project?

Jonathan Broyles (09:47)

It's a great question. And there's a handful. The biggest is that it's one data set with just a little over 100 products. I know there are many, many acoustic products out there, many hundreds, maybe even over 1,000, that we use in our building. So this is just a small sample size, which is unfortunate. But it's a start. So that's exactly what I wanted to do with this project, and hopefully build off of it. 

The other big limitation with this is the data set doesn't do the life cycle assessment for you. So an acoustician has to be very mindful and educated on how to do that, then to use this information that's in the data set correctly to conduct a life cycle assessment. So that's part of it. One of the future things, which I can elaborate a little bit more on in a minute, could be to actually help streamline that process. But, currently, an acoustician has to come in and know what information to extract to make a sustainable decision.

ASA Publications (10:43)

Okay, got it. So let's jump off of that. What are the next steps?

Jonathan Broyles (10:49)

One of the big things that I would like to do with this project moving forward is to help streamline or make it into a tool to streamline sustainable design decisions. So, building off of the tool idea, we could use the information in this data set to do the LCA for an acoustician, even if they aren't super well educated on what global warming potential is. So that's one idea, is just to have a tool that can help a designer do, or conduct and gather that information rather quickly. A second thing, obviously, would just be to add more acoustic products into the data set itself. So that is something I plan to do and hope to do. Again, currently it's a little over 100 products, but to get it to several hundred would be excellent.

And lastly, which is kind of related to this, I would say sustainability acoustic effort, is to continue to educate the profession. That wasn't the sole goal of this work, but of another work and some other things that I'm doing, like, it'd be really awesome to continue to educate the acoustic profession on embodied carbon emissions and sustainability, just because it is pretty complex and there's not a whole lot of education, like classes, especially in existing curriculum that teaches this information. A lot of people, to learn about this, have to do this, you know, after they've graduated from, you know, their university, and potentially have been in the profession for several years. Maybe they’ll take like a, you know, continued learning education class on this information. But, one of the pieces I'm becoming increasingly more passionate about is just to help educate acousticians on this stuff.

ASA Publications (12:26)

Right, so that everybody knows going into it without having to be necessarily a sustainability expert, right?

Jonathan Broyles (12:34)

Exactly.

ASA Publications (12:36)

So what was the most exciting or surprising aspect of this project?

Jonathan Broyles (12:40)

I would say the most exciting and fascinating part of this work to me has been just the people's response to it. There's been a lot of interest, which is great to see, that there's interest in sustainability. But I think a number of people have reached out to me about this work or other related sustainability, architectural acoustic-type projects and are really, really wanting to be a part of this, be a part of the effort to decarbonize our buildings from an acoustician standpoint. Which is excellent because again, we haven't, the last like 10 years or so, haven't been part of the conversation, but we need to be, especially if we want to have fully net zero type of buildings, all of the building stakeholders have to be a part of it. So seeing that response has been excellent, and I think there's already been, this is new since the last time we talked, there has been a number of people who have downloaded the dataset. So seeing that is also exciting.

ASA Publications (13:33)

Yeah, very exciting. Yeah, it's funny because I know I've said this to you before, like talking to you on a previous interview was what made me understand like how much, carbon emissions impact, or related to, building materials, particularly like concrete, which you've talked about in the past. So it's great that you've come up with this data set for acousticians and hopefully it can be expanded upon further, like you said, and hopefully your education efforts for the acoustical community will also continue. Congratulations again, and I wish you the best of luck on your future endeavors.

Jonathan Broyles (14:08)

Thank you so much, Kat.

ASA Publications (14:12)

Our next POMA Student Paper Challenge winner also comes from the field of architectural acoustics. John Latta won the award for his paper, “An Analysis of Spatial Impulse Response Measurements and their Ability to Validate Spatial Features within Acoustic Models.” Thanks for taking the time to speak with me today, John. How are you?

John Latta (14:28)

I am good, how are you?

ASA Publications (14:31)

Good. So, first, tell us a bit about your research background.

John Latta (14:36)

Yeah, so I was a student University of Nebraska where I also worked as an undergrad assistant, where we looked at how room conditions affected speech intelligibility. By room conditions, I mean, you know, background noise levels, reverberation time, things like that. And then throughout my schooling and my internships and now in my career, I've done all different kinds of impulse response measurements, with a variety of different equipment and all different kinds of methodologies.

ASA Publications (15:08)

Okay, okay. So let's start off with that idea of impulse response, which I know comes up a lot in acoustics research. What is impulse response? And how is it used in understanding the acoustics of a space?

John Latta (15:20)

Yeah, so the simplest way to kind of describe it is in two parts and it's already in the name, right? It's an impulse, which is the sound generation in a particular space. And there's the response, which is capturing kind of what the room sounds like. So for impulse, when I say “sound generation,” this be done in a variety of ways. Popping a balloon that's loud enough to kind of fill the room with sound energy. You can also use pink noise, which is kind of a staticky sound. It almost sounds like white noise. The only difference with pink noise is that it has equal energy across octave bands. So that's important when we look at like a logarithmic analysis, which is what impulse response measurements use. And when we look at that through a logarithmic kind of lens, pink noise has a flat frequency response. So when we use it to kind of fill the room, it has an equal energy across all the octave bands that we're looking at. And so in terms of noise generation, seems like pink noise is almost built for impulse response measurements. You can also use a sine sweep, which is basically a single tone that starts at 20 Hertz and sweeps through the audible frequency range all the way up to 20 kilohertz. And again, these are all kind of filling the space with energy, exciting the room, and getting loud enough so that we can really see how the room responds to all the frequency range of the sound that we're generating. 

So that's kind of the impulse generation side, and part two is the response, and it's capturing how the room, I guess, responds to space. And what I mean by that is, when we're no longer generating sound in the room. So when we turn the speaker off, after the balloon has popped, we still kind of hear reflections and sound still exists in the space and it's decaying. And that is what we call an impulse response. And we can use this, after we've captured it with a microphone, we can kind of use this understand different acoustic metrics after some long complicated processing. Some the metrics we look at are reverberation time, similarly, early decay time, which tells us how long sound kind of takes to decay in a room. We also look at clarity and definition, which are metrics we use to analyze speech or music intelligibility. And so I think the important thing to know is it seems very simple, right? We're playing a sound in a space, it's filling the room, and then we capture it with a microphone. It sounds very simple, but it can actually unlock a lot of important information about the rooms that we're analyzing.

ASA Publications (18:05)

Yeah, that's so cool. So how is impulse response conventionally measured and what are the limitations of this process?

John Latta (18:13)

Yeah, so the typical setup is usually, like I mentioned, a balloon or kind of a speaker, some kind of impulse generator that again is filling the room with sound, exciting the sound energy in the room. And then for capturing, we typically use omnidirectional sound level meter, which means it captures sound coming in from all directions. One of the limitations these standard impulse response measurements is that it only captures really two aspects of the room's energy response. And that's the temporal, or the time domain, and the spectral ,or the frequency domain. And we're kind of missing this third component, which is direction. And so again we know where or how sound is kind of reacting over time. We know when sound is arriving. We know how it's reacting at different frequencies. But we don't know where it's happening. And so I think it's also worth mentioning that depending on what kind of impulse you're using, repeatability can also be an issue. For example, when you're generating pink noise or popping a balloon, these obviously aren't repeatable, right? Every time we pop a balloon, it's going to be slightly different. Pink noise is a random noise, so every time you use it, it's going to be slightly different. And you can get around this either by doing a ton of measurements and kind of averaging across, or like we did on this project, you just use a sine sweep signal. It's a repeatable signal. You can play it as many times as you want, and you're always going to have the same signal, the same response.

ASA Publications (19:54)

Okay, okay. So what are spatial impulse response measurements?

John Latta (19:59)

It's in the name, right? It's basically an impulse response with a spatial element. So again, we're getting all this information about time and about the frequency spectrum. This is now providing a directional component on top of all the information we're already capturing. So the easiest way that I've found of thinking about it is we're now on top of, you know, finding what frequencies and what time sound is arriving at the microphone. We're now also capturing where that sound is coming from when it eventually reaches the microphone.

ASA Publications (20:35)

Okay, got it. So what is ODEON and how does it tie into your interest in impulse measurements?

John Latta (20:43)

Yeah, so ODEON is a 3D modeling tool that we use for room acoustic analysis, and it's typically used for spatial impulse response analysis. So again, in ODEON, you can kind of rebuild your measurements to find reverberation time, early decay time, clarity, definition, and it uses a bunch of methods, different methods of ray tracing, but it's essentially giving you the same results that you would find if you took and impulse response measurement in person. What’s also nice about ODEON, and why it's important for this project, is that it has some spatial features. And these features can be used similar to standard impulse response measurements. They can be used to analyze and, I guess, rebuild your spatial impulse response measurements. And so I guess kind of the goal of this project was to kind of compare what we were finding in our spatial impulse response measurements to what we were finding when we eventually rebuilt them in an ODEON model. And really what we were looking for was what's different between what's happening in person and what's happening in a computer and why that was happening.

ASA Publications (22:00)

Okay, okay, got it. How did you test the acoustic model created by Odeon compared to spatial impulse response measurements?

John Latta (22:08)

Yeah, so because it's, as you can imagine, really hard to convey spatial information in a textual way or in an Excel spreadsheet, we had to do, actually, a graphical comparison. So conveniently, both the measurements that we took and the model that we made, they produced very similar graphics. I believe in ODEON they're called a “decay rose.” But essentially, it's a circular plot that shows you how sound is captured by the microphone, and specifically it tells you kind of what direction sound is arriving at your microphone, in a kind 360 degree circle. It tells you how long it takes that sound to arrive at the microphone. So for example when we play a sine sweep through a speaker, it's going to bounce off a bunch of different walls and eventually arrive at our microphone. So it tells us how long it took from when it was generated to when it was captured. And then it also tells us what the direction of that sound it was coming from. And the other thing it tells us is how loud that sound is. So it might be arriving really fast. It might only bounce off of one surface and come right back to the microphone. It might be bouncing off a really absorptive surface. So it also tells us how loud that sound is when it eventually reaches the microphone as well. So that was kind of the basis our graphical analysis. And so using these graphics, we could kind of look for visual differences between what we were finding in the measurements and what we were finding in the model. 

So we were particularly looking for patterns. And by patterns, I mean any kind of repeat discrepancies that popped up across different frequencies, different arrival times. So sound was arriving really late after bouncing off a couple surfaces, is that different from when it was arriving after bouncing off of one surface? And then also across locations. Was the certain discrepancies we were finding location specific, or was it happening everywhere in the space? But those were kind of the things we were looking for when we did these kind of graphical analyses.

ASA Publications (24:24)

Okay, okay. So how did the model end up performing compared to both the conventional and spatial measurements?

John Latta (24:31)

Yeah, so for the standard, or non-directional, impulse response measurements, they actually lined up pretty well. And by lined up, I mean we found the reverberation time, the early decay time, clarity, definition. We found those metrics in person. We found them in the model. And then we kind of compared them. And they seemed to line up pretty well. We did do a little bit of I'll call calibration. What I mean by that is we slightly adjusted scattering and absorption coefficients in the ODEON model until we got near perfect matches between what we were finding in person, what we were finding in the ODEON model. This is kind an industry standard practice. It's just to make sure that your model is accurate to what you're measuring. 

And so once we had our model kind of quote unquote calibrated, then we could go into spatial, our or graphical analysis. And what we found is that these higher order reflections and these later arrival times, there was a noticeable difference in what we were finding between the two. In Odeon, we were finding that at these later arriving sounds, the model was finding much lower levels, we're talking almost half as loud, than we were finding in the in-person measurements. And so, for example, when we generate a sound at the speaker, it travels, you know, bounces off a couple walls and arrives at our microphone in a specific location. After, you know, three or four bounces, we were measuring maybe 60 dB in person. Well, when we rebuilt that in the ODEON model, we were only finding it was 30 dB. We also found that at these later arriving sounds, they were much more diffuse. What I mean by that is in the in-person measurements, on our graphics that we produced, there were noticeable spikes of energy at specific directions and at specific arrival times that weren't being shown in the ODEON model. Now, this could be any number of things. This could be specific surfaces in person, you know, harsh reflective surfaces that we weren't specifically modeling in ODEON. But the point is that the two discrepancies we found couldn't really be explained because there are just so many factors with something like this. You know, there's the model, that could be the reason. There's the methodology, the processing software we use. There could be something going on there. And then, of course, the equipment we use. There might be things going on with that as well. 

So one of the key takeaways I kind of found is that ODEON models are really limited by their quality in that, you know, we're not robots. Won't make perfectly accurate models. We're not detailing every single surface and putting in perfect absorption coefficients. And then the measurements…  So ODEON models are limited by quality, and measurements are kind limited by their quantity. Again, we're not robots. We're not taking tens of thousands of measurements, right? We're only doing a couple at specific receiver locations. And so I think it's important to remember that both the measurements we take and the models we make are just kind of tools in an acoustician's tool belt. And just because, you know, they don't agree and they're not perfectly accurate to each other, it doesn't mean they're not useful. Even when it comes to something like spatial impulse response analysis, I think, you know, they could still be used together kind of in conjunction to find, you know, meaningful results.

ASA Publications (28:30)

Right, right, I was just thinking that. I was like, well, if they're both limited in slightly different ways, then you could kind of come to a middle ground when you compare the data from both. 

John Latta (28:37)

Right, yeah.

ASA Publications (28:39)

Okay. So, speaking of kind of limitations and that kind of thing, what are the next steps for this project?

John Latta (28:47)

Yeah, so my answer is, in all caps, MORE TESTING, right? This is a small sample size. It was just me going out, taking measurements by myself, making a model by myself. You know, with something like this, there's always more spaces to be analyzed. I only did one recital hall. There's always more measurements. You know, I only did four receiver positions and two source positions. There’s different equipment, right? I only used one microphone, one spatial ambisonic microphone. I only used one speaker. There's obviously different processing methodologies. I used one processing software. And then of course the modeling software itself. You know, ODEON isn't perfect, as we all know, and there's other modeling softwares out there. But I think the general idea is that there's always more to be done when it comes to testing and this is no exception.

ASA Publications (29:47)

Yeah, lots of opportunity ahead of you, it sounds like. So what was the most exciting, surprising, or interesting aspect of this research for you?

John Latta (29:56)

Well, I think exciting was sitting at my computer looking at MATLAB for three hours and eventually having a picture pop up that actually made sense and was what I was trying to measure. That was both thrilling and relieving. No, I think the most, for me, the most interesting part was definitely that inaccuracies that we find can be dependent on the arrival time or the order of reflection of the sound that we're measuring, right? And so we kind of found this through diffusivity, so how diffuse the sound is versus how specular it is. There can be differences there and also in the levels that we're measuring, right? In certain cases, we were measuring really loud sounds at late arrival times. And then in certain cases, we were measuring, or finding in the model, that they weren't nearly as loud.

ASA Publications (30:54)

Yeah, yeah, that is really interesting. And it's funny just how having things that work out the way you hoped they would could be so exciting, you know? Well, thank you for giving us some insight into spatial impulse response measurements and modeling, which sounds like they have some potential for being a useful tool in architectural acoustics, as you mentioned. I wish you the best of luck in your future research. And I realized I did not say this before, Congratulations.

John Latta (31:18)

Hahaha. Thank you, thank you.

ASA Publications (31:21)

The final winner of the POMA student paper competition from the virtual meeting is Michelle Crouse who won with the paper, “A low cost ultrasonic absorption spectrometer mainly using off-the-shelf parts.” Congrats on winning the award, Michelle, and thanks for taking the time to speak with me today. How are you?

Michelle Crouse (31:35)

Thank you for having me. I'm enjoying the spring season,

ASA Publications (31:39)

I know, same here. It's so nice out today. So first, tell us a bit about your research background.

Michelle Crouse (31:43)

Well, I studied biochemistry as an undergrad at California State University. I had done some research in the Department of Biochemistry and Chemistry where I developed standard operating procedures to prepare short nucleic acid structures for nuclear magnetic resonance spectroscopy. I also dabbled in drug discovery looking for natural products from fungal strains that can treat brain cancer.

In the Department of Mathematics, I studied the dynamics of cancer drugs in the simulation of the biological environment. I formulated the combination of two drug treatments into the differential mathematical model for chronic myeloid leukemia. 

This fall, I'll begin my graduate studies in the pharmaceutical sciences. I worked on these previously mentioned research projects as an undergrad over a couple of years while primarily conducting research in the Department of Physics for physics education. My focus was improving the retention rates of diverse undergraduate students in STEM based on their exposures and experiences. My background in physical chemistry brought me to the National Institute of Standards and Technology, or NIST, for the Summer Undergraduate Research Fellowship (SURF) program. There, I focused on ultrasonic absorption in liquids.

ASA Publications (32:57)

Awesome, awesome. It's not often that we have people sort of outside of the normal realm of our acoustical areas, like a chemist as you are. So what are acoustic spectrometers, and what are they typically used for?

Michelle Crouse (33:12)

Well, sound absorption provides information about a liquid's properties, such as sound speed, intermolecular interactions, and molecular relaxations. A relaxation is a finite amount of time to come to equilibrium. So if you use an oscillatory source, like a sound wave, to make that happen, there will be a peak in the response when the source takes exactly that amount of time to complete one cycle. One over that time is the relaxation frequency.

Studying relaxation phenomena makes it possible to determine a direct relationship between structure, the nature of the thermal motion of macromolecules, their segments and side groups, and structural mechanical properties. This acoustic spectroscopy method is used to study the dynamic mechanical properties of various materials. An acoustic absorption spectrometer measures sound absorption versus frequency in a liquid.

It can be used in the same way an optical absorption spectrometer is used to study interactions between solutes and solvents in solutions. Also, since absorption largely depends on viscosity, acoustic spectrometers can double as rheometers, and rheometers measure the viscosity of fluids. There are a few types of acoustic spectroscopy, including resonator and pulse echo. 

This acoustic spectrometer uses a method called pulse through-transmission The technique works by transmitting ultrasonic pulses from the transmitting transducer through the liquid to the receiving transducer. The liquid absorbs some of the energy in the pulse, and we can determine the absorption coefficient by analyzing the pulse's intensity after it has passed through.

ASA Publications (34:45)

Very cool. That's a great explanation. Interesting. So your project stems from the fact that spectrometers are so expensive. A lot of scientific research equipment is expensive, though, so why is this particular concern with acoustic spectrometers?

Michelle Crouse (34:59)

Well, there is currently only one commercially available acoustic spectrometer that I'm aware of. The cost of that system puts it out of reach for many university researchers and students wanting to get into acoustic spectroscopy. The commercial system also uses the variable path pulse through-transmission technique of spectroscopy. The sound wave travels through the liquid, and as it propagates and interacts with the liquid and some energy is absorbed. This absorption is related to molecular relaxation processes, viscosity, thermal conduction, and the internal friction within the material.

ASA Publications (35:33)

So what made you realize that you could make a spectrometer more affordably?

Michelle Crouse (35:38)

While the basic physics is simple, we measure the acoustic loss over a distance. The complex part is the analysis of the data. By analyzing the changes in amplitude, time of flight, and waveform shape, the spectrometer determines the sound speed and acoustic absorption coefficient, which tells us how much of the sound was absorbed by the liquid versus the frequency. With this, we can observe relaxation behavior.

ASA Publications (36:02)

Okay, okay. So how did you build the spectrometer?

Michelle Crouse (36:05)

We used transducers designed for non-destructive evaluation, or the NDE market, and mounted them using off-the-shelf optics bench components. We used a micrometer translation stage to move one transducer precisely relative to the other, and there's a flexible tube between them to hold our liquid samples. One transducer sends an ultrasonic pulse, and the other receives the pulse on the other side of the tube. The electronics include an NDE pulser/receiver connected to the transducers, which generates very short, high-voltage and precisely timed electrical pulses to the first transducer and amplifies the received pulse from the second transducer. 

We connected an oscilloscope to the pulser/receiver to visualize and record the received pulse. We then transfer the data from the oscilloscope to a computer for further analysis. For the data collection, we set the spacing between the transducers to a known value and record a pulse that has traveled through the liquid. Then we move the transmitting transducer to a known distance using the micrometer and record another pulse. We repeat this process a total of thirty-one times.

ASA Publications (37:10)

So what did you do to validate that the spectrometer actually works?

Michelle Crouse (37:14)

We compared the data we collected with data we obtained from the commercial spectrometer. The frequency range for the spectrometer we built overlaps with the range of the commercial spectrometer. Some of the liquids that we tested included solutions of salts, polyvinyl alcohols, and celluloses, which are of interest to the pharmaceutical manufacturing industry. Additionally, we compared our results to historical published data using some of the same samples.

ASA Publications (37:40)

That's so cool. Okay. So how did your spectrometer end up performing?

Michelle Crouse (37:45)

It performed exceptionally well. The uncertainties were well below the stated uncertainty of the commercial spectrometer.

ASA Publications (37:52)

That's exciting. So are there any limitations on your instrument?

Michelle Crouse (37:56)

Yes, there are. Frequencies below around one MHz have a higher uncertainty because our sample volume is relatively small, approximately twenty milliliters. This means that our transducers can't be too far apart to measure at low frequencies. Also, our relative uncertainty is increased for liquids with extremely low acoustic absorption since it largely depends on the signal-to-noise of our captured pulses.

ASA Publications (38:19)

So what are the next steps of your research?

Michelle Crouse (38:22)

We plan on automating the measurement process using a motorized translation stage. As I said previously, we were using a manual micrometer translation stage. We also want to figure out different transducer combinations to extend our frequency range.

ASA Publications (38:37)

Awesome. Okay, yeah, sounds like a couple good options there. What was this project's most exciting, surprising, or interesting aspect?

Michelle Crouse (38:46)

The most exciting part, or you know,  moment, was when we observed a relaxation in scandium sulfate that was nearly identical to one measured around fifty years ago using a much more sophisticated setup. This relaxation could only be captured at the time by our system because our spectrometer can measure absorption down to 0.6 megahertz.

This is great that we can use our spectrometer at lower frequencies in addition to providing absorption spectroscopy access to institutions.

ASA Publications (39:16)

Yeah, that is so cool. It’s also really cool that as an undergraduate researcher, you were able to come up with this inexpensive replacement for a costly piece of equipment. 

Michelle Crouse (39:27)

Yes

ASA Publications (39:28)

Good thing to put on your CV, right? 

Michelle Crouse (39:30)

Yes, yes. It's a little different than the pharmaceutical sciences, but you know, it makes for more and better conversation, I think.

ASA Publications (39:40)

Yeah, right, totally. Well, congrats again on winning the award and I wish you the best of luck in grad school and your future endeavors.

Michelle Crouse (39:46)

Thanks. Thank you, Kat.

ASA Publications (39:48)

Of course. You're welcome.

Before we wrap up this episode, I'd like to share a couple messages with our listeners. One, if you liked what you heard in this episode, please text it or email it to someone who may enjoy it as well. Second, for any students or mentors listening around the time this episode is airing, we're actually holding another student paper competition for the 188th ASA meeting in New Orleans. So students, if you gave a presentation or had a poster at the meeting, now's the time to submit your POMA. We're accepting papers from all the technical areas represented by the ASA. Not only will you get the respect of your peers, you'll win $300 and, perhaps the greatest reward of all, the opportunity to appear on this podcast. And if you don't win, this is a great opportunity to boost your CV or resume with an editor-reviewed proceedings paper. The deadline is June 22nd, 2025. We'll include a link to the submission information on the show notes for this episode.