Tech Briefs: What got you started on this project?
Professor Ron Miles: I have always been interested in acoustics in general, ever since I was really young. When I started here at Binghamton, I got connected with Professor Ron Hoy, a faculty member at Cornell, who worked on hearing in insects. We started working together on a really interesting story about a fly that happens to be very good at hearing, which is unusual. So, I got interested in insects and for the past 30 years I’ve been working on engineering applications of hearing, how animals hear, particularly insects. The way I got into this spider silk story is that most insects, which is actually most animals because there are so many insects, hear using hairs, not eardrums — hairs respond to the air flow in a sound field. When a sound is propagating, there's motion in the air along with the sound pressure. Although humans hear the pressure with their eardrums, insects hear the motion.
All microphones function a lot like human ears do because Alexander Graham Bell and Thomas Edison, the people who created the telephone and microphones in the first place, knew about the discoveries about hearing at that time. We have an eardrum that responds to sound and inside the ear we have stuff that turns that into a neural signal. So, they tried to make a microphone that works like ears. Microphones therefore have pressure sensing membranes that convert the pressure into an electrical signal. I'd been wondering, what about all the other ears in the world? Maybe it would be a good idea to take a look at those to mimic them. After all, microphones are biomimetic since they're inspired by ears. What about all the other ears? Why don't we sense flow instead of pressure? And since I know something about acoustics, I know that pressure and flow are related — the equations that govern pressure are pretty much the same ones that govern flow in the sound field, so why not sense flow?
So, I was looking around for ways to do that and I thought that since insects use fine hairs, maybe we should make something like a fine hair. So, I messed around with the equations and was learning how the viscosity of the air, at a small scale, makes small things move — you see it all the time — dust floats around and small things move with the air. I had a new grad student at that time, Jian Zhou, and I was talking to him about this, trying to come up with fibers that would respond to sound. We then started working with a guy in chemistry who was electrospinning fibers, but they were very fragile and hard to work with.
One day, Jian was taking a walk in our nature preserve and noticed a spider web that was blowing in the breeze, and he thought: maybe spider silk, you know, it's really strong and moves with the air. So, he borrowed a spider from the nature preserve, brought it back to the lab and figured out how to get it to spin silk. Then he harvested the silk from it, and we tested its acoustic response — we have a really good facility for measuring sound. We placed the spider silk in our anechoic chamber, played sound, and measured its motion using a laser. We found that its motion was very close to the motion of the air in the sound field, so it acted like a really good microphone.
We published a paper about this, which is a cool story because this spider silk microphone was amazing — it had a flat frequency response from 1 Hz to 50 kHz, and it was very simple.
I have another student, Junpeng Lai, who was looking at this, and a question came up from that work: If the spider silk responds to sound and a spider is sitting on the web, can the spider hear the sound because of the motion of the web? Spiders are known to be really good at detecting the motion of their webs; that's how they live. So, we worked out how to find out if the spider could hear based on the sound-induced motion of the web, and showed in fact that they can.
Tech Briefs: How did you figure out that it can?
Miles: There were two ways to do that. The first was in collaboration with our biology friends at Cornell. The normal way to find out if an animal can hear sound is to insert electrodes in their sensory cells and if you get spikes, you know it’s responding. So, we tried to do this with the spiders, but it's really difficult, obviously.
The other way to do it is to find out if they respond in some way through their behavior. If they orient themselves toward the sound, then they're hearing it — that's the approach we took. So, we placed the spider on a web in the anechoic chamber, turned out the lights, and had sound coming out of a loudspeaker three meters away, and the spider responded. We found several different behaviors that the spider had in response to that sound, which showed that the spider can in fact hear.
It was actually pretty difficult, there are lots of questions that can come up. Is this spider responding to the sound of the web? Is the web being driven by the sound? However, we did basically show that the airborne sound from the loudspeaker drives the web, and the motion of the web then is picked up by the spider. This suggests that this spider could actually hear because of its web.
Although this didn’t prove that’s the way spiders actually hear in nature, it suggested that there might be other ways to make a microphone. Thinking about it as engineers, pressure is one way to sense sound, but maybe we should investigate whether it's feasible, or even a good idea, to try to make devices that respond to flow instead of pressure. There's been relatively little research on the sensing of acoustic flow. There are lots of microphones in this world, and they all sense pressure. And I think it's a good idea to be open to the possibility of sensing acoustic flow. Maybe that would be very helpful in some situations.
Tech Briefs: Do you have any sense of why the frequency response is so much better than with a pressure-sensitive microphone?
Miles: I do know why the frequency response of the spider silk is so good. It's because the spider silk is so lightweight. It basically rests in a fluid that is moving, and if the spider silk is lightweight enough, the motion of the fluid, the air, will exert forces on it. We’ve shown that if the spider silk is heavy, it tends to not move very well at high frequencies. But if it's light enough that its mass is on the order of the mass of the air that moves with it, then its frequency response is really good. So, it's a very effective way to detect sound.
There are lots of practical issues in sensing sound. Getting it to respond well is one thing but making an actual microphone that way is kind of hard, it’s not real practical. So, having the idea and then turning it into a practical engineered device is really hard.
Tech Briefs: Along those lines, are you going to use actual spider silk or synthesize something? And then how do you support it?
Miles: Using actual spider silk is certainly not practical. But you can make micro devices that have very small dimensions using silicon microfabrication. You can make beams and things that are very, very lightweight, and very thin. So, even though spider silk itself is not a real, practical kind of material to make an engineered device out of, there are man-made devices, structures that could be used that could be lightweight enough and respond to sound.
It just takes a bit of research to figure how to actually make this. What are its essential dimensions and properties? We've been looking at a few ideas of how you can make something that responds to flow.
There's been so little work done on sensing acoustic flow. In fact, I think that we’re the only ones doing it using thin, viscous-driven structures. If you compare us to all the work that's been done on pressure microphones, which has been going on for 150 years, they've got a bit of a head start, acoustic flow is a neglected field.
There's a lot that could be learned about how to practically make an engineered device this way. The structure needs to be made from a lightweight material. We know that it needs to have certain mechanical properties, but the hardest part is how to transduce its motion into an electrical signal. In our ears, the structures that convert the motion of the diaphragm into an electronic signal for our brains is insanely complex. There are only a few ways that we have for converting a microphone-sensed structure into an electrical signal, and I've always wanted a better one. You know, we have capacitance. This is the way it's normally done and there's strain sensing, there's optical sensing and not a whole lot of options for doing that.
It's actually hard to design a microphone. Modern microphones generally need to be pretty small, because they're made out of silicon, and silicon wafers cost money to process, so you want to stuff as many microphone devices on a wafer as you can, to make them as small as possible in order to make them cheap. That adds a lot of challenges to the design. So, I think that there's a lot of potential in looking at new ways to make microphones. The basic structure that we have in microphones that are used in cell phones and hearing aids and all kinds of electronic gadgets all really look the same in a way. They all have to somehow transduce the pressure-sensing diaphragm into an electrical signal. In principle, they're all really a lot like what Alexander Graham Bell worked on — they're different, but they use the same physics.
Tech Briefs: Could you give me an idea about the microphones inside a cell phone. Are they nanostructures?
Miles: They’re not nano, they're micro. The diaphragms are maybe half a millimeter in diameter but they're thin — maybe a micron thick. There’s usually a back plate to detect the change in capacitance between the two. It's a complicated structure with lots of engineering challenges.
Tech Briefs: Are they like MEMS accelerometers?
Miles: Yes, microphone structure is very similar to an accelerometer. The only real difference between a microphone and an accelerometer is that in a microphone you want the mass of the moving structure, the diaphragm, to be really small because it's got to be driven by small pressures in the sound field. In an accelerometer, it's desirable to have the moving mass be fairly heavy so that it stays still as the structure moves. You then detect the changing capacitance between the moving structure and something that's fixed. So, they're kind of similar things, but there are different challenges.
Tech Briefs: How far along are you in meeting any of those challenges at this point?
Miles: We've got structures, there are lots of tricks and complications, but we have a prototype microphone that senses flow. It's not quite as far along as a pressure microphone in terms of the overall performance, but it detects sound, and it works well. In addition, we patented the idea of sensing flow that we got from looking at spider silk. There’s a company in Canada that has licensed the patent, and they're currently developing microphones based on that idea. In fact, they're currently working on implementing it in a product for a customer. They've built a company around this called Soundskrit, which is located in Montreal. They’re working hard to turn it into a practical device.
In addition, we've also been funded through the National Institutes of Health to create sensors that detect acoustic flow in order to improve our ability to measure sound, with applications toward people who are hearing impaired — to try to better understand hearing in humans. It’s a scientific application, not necessarily a large-scale commercial product. We're working on ways to sense acoustic flow inside the human ear canal.
This particular study is based on the idea of detecting the sound that is emitted by the eardrum in order to understand the cochlea inside the ear. It's very strange but when you play sound to the ear, the eardrum moves. That ultimately causes vibration within the cochlea, which is where the transduction occurs. Then the signal from the cochlear goes off to the brain, the brain detects it and sends a signal back to the cochlear to kind of adjust things, which improves your hearing. The signal from the brain ultimately causes motion of the eardrum and the motion of the eardrum radiates sound and sends the sound back out through the ear canal. So, you play a sound, you detect the sound, then that's re-radiated by the ear, and it tells you a lot about how the ear works. It's actually used for newborns — they do this in order to find out if the newborn has a hearing problem. If the ear is not working, you don't get this re-radiated sound. So, measuring it is important, they do it on all newborns. It's called otoacoustic emission and is an important diagnostic tool. It also can tell you a lot about the physics of the cochlea and how it works. So, we received funding from the National Institutes of Health to work on ways to detect that sound that's coming out of the ear using flow sensing along with pressure.
Hearing is a big deal, it’s really important. A lot of technology has been created in order to help people who have hearing problems. It's the more important sense we have because it connects us with people. Cochlear implants have advanced tremendously in the past 15 years or so to enable people who have really serious hearing troubles to hear — it's amazing.
I keep thinking about a quote from Helen Keller. She was asked which is the more important sense, vision or hearing? She said that hearing is definitely the more important because if you lose your vision, it takes you away from things but if you lose your hearing, it takes you away from people.
I remember when I first took acoustics as an undergraduate, my professor said that there was an argument about which was the older science: optics or acoustics. His answer was that it's obviously acoustics, because in the beginning, God said…
