Dr. Gary Hunter, who joined NASA in 1990, is an expert in the design, fabrication, and testing of sensors, especially chemical species gas sensors. In 1995 and 2005, he led the development of sensor systems that won R&D 100 awards, which recognizes the 100 most significant inventions or products of those years. Dr. Hunter currently serves as Intelligent Systems Hardware Lead, and Technical Lead of Chemical Sensors, for the Sensors and Electronics Branch at the NASA Glenn Research Center.
NASA Tech Briefs: Dr. Hunter, one of your areas of expertise is chemical species gas sensors. What are chemical species gas sensors and how do they work?
Dr. Gary Hunter: Simply speaking, chemical species gas sensors measure the chemical species in the atmosphere, rather than in the ground or water. There are a number of species in the atmosphere – nitrogen, oxygen (O2), carbon dioxide (CO2), carbon monoxide (CO), and others. So, some of the chemical sensors we work on are sensors that measure hydrogen, hydrocarbons, oxygen, carbon monoxide, carbon dioxide, nitrogen oxides (NOx), and other species, which have relevance to aerospace applications.
How do they work? There are different ways of measuring chemical species, and each technique has its advantages and disadvantages. The technique that we use has to do with small microsensors, and they work by coming in contact with the gas in the environment and having a chemical reaction with it. That chemical reaction produces an electrical signal that we measure, so one can begin to understand what’s going on in the environment.
What we try to measure depends on the application. For example, for fuel leak applications, we measure fuels like hydrogen and hydrocarbons, as well as oxygen. For fire detection, two major species are CO and CO2. For jet engine emissions, a mixture of species are of interest: NOx, CO, CO2, hydrocarbons, and oxygen. Finally, for monitoring someone's breath to determine their health, we’d be interested first in NOx, and then other species that might make up the breath. Each one of these sensors have different materials and work by different chemical reactions.
Overall, we try to tailor the sensor for the application. We develop core technology that we can then adapt to measuring conditions related to aircraft or the spacecraft, or inside the vehicle habitat, or outside on the engine. We also try to make the sensor selective – or orthogonal – for the species they’re trying to measure. So we try to get a CO sensor as best we can to measure only CO. There will be some cross-sensitivity, but the approach is to have a CO sensor be sensitive only to CO and likewise oxygen sensors to be sensitive to oxygen, and so on. In this way, by combining a number of sensors together, the approach is to be able to measure well the species that are important to your application, the species that would help you understand the environment and be able to make better decisions about what the conditions are. Does that answer your question?
NTB: I think it does, thank you. Last year you were part of a team that was awarded a U.S. patent for making miniaturized Schottky diode hydrogen and hydrocarbon sensors. What is unique about this technology and what do you envision these sensors being used for?
Dr. Hunter: Silicon carbide is one of the tools we have in our toolbox. That is, it’s one of several different sensor platforms that we work on to measure the range of chemical species just described. And I should note that we work with collaborators in both university and industry in order to bring these various sensors together. So, one platform might be an electrochemical cell that we would modify to measure carbon dioxide in one configuration or oxygen in another. A second platform might be metal oxide resistors to measure hydrogen or hydrocarbons. Another platform that we might use is the silicon carbide-based Schottky diode platform. What’s special about silicon carbide as a gas sensor is the idea that it can operate as a semiconductor at higher temperatures. Thus, operating at higher temperatures, it will be able to detect species like hydrocarbons or NOx depending on the materials you use with the silicon carbide.
One of the issues of operating a sensor at high temperatures is that chemical reactions can occur within the sensor structure itself. That is, if one heats some metals up on silicon carbide, you may get a chemical reaction that can degrade the sensor performance. So, if one puts, for example, palladium, which is a metal we use for hydrocarbon detection, right on top of silicon carbide and heat it up, you may get chemical reactions that change the structure of the sensor.
One of the things one would like to do with a sensor is have it be stable, have it be reliable, be able to give you the same results every time, and be able to – if at all possible – eliminate these types of drifts that are associated with higher temperature chemical reactions. That’s work that we’ve done for a period of time with respect to silicon carbide gas sensors.
One of the approaches to decreasing the chemical reactions that happen at high temperatures is to put barrier or interface layers in between, for example, the silicon carbide and the metal alloy. The barrier layer associated with this particular patent that you mentioned is palladium oxide. The patent describes the idea that one of the ways to solve this problem of higher temperature reactions to give a sensor with better stability and good sensitivity all at the same time, for silicon carbide, is to use this palladium oxide barrier layer. We found what we believe to be good stability with respect to the sensor at higher temperatures for extended periods of time using palladium oxide.
Now, what would you use it for? Well, there are two sets of applications that one might take a look at, two sets of gasses that I mentioned a few moments ago with respect to silicon carbide. One is the measurement of hydrocarbons, which can be used for a range of different purposes. For example, with respect to fire detection applications, hydrocarbons are emitted. With respect to leak detection, a fuel is a hydrocarbon. These sensors can also be used to measure hydrogen or toxic species such as hydrazine. So, the ability to make a stable silicon carbide sensor leads to a range of possible applications from fire to leak detection to measuring toxic gases. That’s some of the type of work that we’ve done with respect to the hydrocarbon detector.
We also work with silicon carbide sensors to make them sensitive to NOx. Again, start with the same platform but modify some the materials on that platform to target NOx instead of hydrocarbons. Reducing engine emissions are a good application for NOx detection. We are also doing work in monitoring a patient’s breath in order to determine their health – the application of a NOx sensor based on silicon carbide or other approaches might allow one to, for asthma conditions, note the amount of NOx in their breath and begin to understand issues and conditions related to asthma.
NTB: Since we’re talking about silicon carbide, what is atomically flat silicon carbide and how does it improve the performance of a Schottky diode sensor?
Dr. Hunter: The advantages of silicon carbide have to do with its ability to operate at high temperatures. Silicon carbide is a high-temperature semiconductor, so it can operate at temperatures in excess of 600 degrees C. We typically operate between 500 – 600 degrees C with our sensors. One can then have high temperature electronics, as it develops, that can be coupled with a silicon carbide gas sensor, or a pressure sensor, or other types of systems. Then one can combine electronics with these high-temperature sensors and start to make smart systems.
One of the challenges with silicon carbide is that it is not as mature an electronic material as silicon. The present silicon carbide wafers can have defects associated with them. Those defects can affect device performance. In particular, they can affect gas sensor device performance. So, one of the things we’ve shown is that if you go to an atomically flat surface, if you go to a surface that doesn’t have defects, then you are able to get better response out of the sensor than you would with a rougher surface with defects. An analogy I’ve used before is that it’s like trying to put a sensor on the surface of the moon versus a putting green — pretty rough with potholes vs smooth and flat. So, the idea is that if you deal with materials that have less defects, that are smoother, that are flatter, that have more uniform surface properties, it allows you to better control the sensor response, from what we’ve seen so far, than if you were to try to deal with a less controlled, defect-filled surface.
NTB: There is a lot of talk these days about "Lick and Stick" sensor systems. What are Lick and Stick sensors?
Dr. Hunter: Well, other folks may have different versions and different interpretations of this approach, and even different names other than ”Lick and Stick”, but what we mean is as follows.
A direction that we’d like to see for sensor technology is to make them small and make them smart — be able to integrate within a sensor system the components one needs to be able to make the sensor operate and make that self-contained and independent. So, for example, within the sensor you have a sensing element, say a hydrogen sensor, but in order to be able to operate that sensor you may also want to have a microprocessor, power, signal conditioning, and communications. You may have multiple sensors. You may want to read several parameters at the same time. So, the bottom line is, what we mean by Lick and Stick sensors are systems that are small, self-contained, the size of a postage stamp, and which we are able to literally place where you need them when you need them without having to rewire the vehicle or without having to provide new power lines for the sensor.
So what we mean by Lick and Stick is the concept of being able to, like a postage stamp, put sensor systems where you need them and when you need them as a self-contained, smart, individual unit. To be clear, we’re not talking about literally gluing things onto surfaces for all aerospace applications. For some applications, the idea that the “stamp” might fall off could cause real issues. Often, they might be screw-mounted or use some other formal mounting system. But we do want to significantly decrease all of the baggage that often comes into play in implementing sensors.
That’s the direction we’re trying to head in. What we'd like to do overall with a vehicle system, or whatever it is we're dealing with, is make it smart, and drive that intelligence from the lower component level, from individual smart systems, leading up to a smart vehicle overall.
NTB: What are microfabricated sensors and, their size notwithstanding, what advantages do they have over conventional sensors?
Dr. Hunter: Microfabrication, as we use the term, is associated with the idea of using the silicon processing techniques used in the electronics industry and applying those to sensor technology. The idea behind a microfabricated sensor system, then, is to apply the silicon processing technology where one can batch fabricate things, that is, make many of the same thing at the same time, make them repeatable again and again, and be able to put multiple components into a very small space. This type of work was pioneered by one of our collaborators, C. C. Liu, at Case Western Reserve University. So, for example, say for our hydrogen sensor, one may have two sensors on that unit – one for low concentrations and one for higher concentrations – and a temperature detector and a heater. All of that is microfabricated, that is, using silicon processing techniques put onto a very small surface.
The advantages, then, are that first, they're smaller. They may be able to have more features associated with them. In silicon processing, you would hope for reproducibility each time. And issues such as power and weight are minimized because, for example, if you’re heating a microsensor, heating something small takes less power than a comparable element that’s larger.
In summary then, silicon processing applied to miniaturized sensor technology, which we refer to as microfabricated sensors, have advantages related to the capabilities you can build into the sensor structure, as well as the minimal impact that it has — at least as an ideal — associated with the application that you’re trying to meet. If you are going to start putting more sensors into a vehicle, they are going to have be small with minimal impact to the system.
NTB: Do sensors used in space have to be radiation hardened like semiconductors?
Dr. Hunter: This question I'm going to partly punt on. I am not the best person to talk about that, but I will give you an answer that is not quite the question you asked about, and do with it as you wish.
Typically with sensors in space, there are many parameters that might be measured and there may be a number of different types of sensors. There are sensors, for example, that measure temperature, strain, and the like. In some cases, and this is a little bit beyond my scope, in terms of radiation hardening, some materials are not necessarily easily damaged by radiation. Something like a platinum resistor that is used as a temperature detector is not something one usually worries about in terms of radiation hardening and radiation damage. However, a more complicated system with electronics can be affected because there are materials, for example, the silicon electronics, which can be affected by the radiation.
One thing that we've been working on is that we want sensors that are smart. We want, for example, the Lick and Stick approach with a microprocessor associated with it. And so as we have done our work towards making smart sensor systems ready for space, what we have encountered is that we have to take into account the fact that we have silicon electronics associated with our work. We then have to make sure that our smart sensor systems – if we want to try to move towards space implementation – are radiation hardened and able to handle a radiation environment. So, some of the work has involved radiation testing to make sure this basic Lick and Stick platform doesn't have anything that would be necessarily problematic associated with radiation. Now, the radiation environments vary and over time one will have to qualify each individual system.
NTB: Make it mission specific, I would imagine?
Dr. Hunter: Yes. Match it to the specific radiation environment.
NTB: In 1995 you were the co-recipient of an R&D 100 award for developing an automated hydrogen leak detection system that Ford installed on its automotive assembly lines, so you know some of the stuff you’re working on has potential commercial applications. What other types of sensor systems being developed by NASA do you envision finding their way into commercial applications?
Dr. Hunter: I'll note , for example, that we're working on the fire detection systems for which we received a 2005 R&D 100 award. It was meant to have applications associated with commercial aircraft cargo bay fire detection.
The high-temperature emissions work– we're working with the Navy to make what we call a high-temperature electronic nose to measure emissions coming from the back-end of a jet engine, and we’re interacting with, on that project, the Propulsion Instrumentation Working Group, which is a consortium of engine companies. So one would hope for a commercial product you might be able to use on jet engine test stands.
The work that we're doing in human health breath monitoring is part of a State of Ohio project that is meant to have a home monitoring system for asthma. We will move toward commercialization during the time of that project.
So, I've just noted three things right there associated with commercial applications. In general, given that we work on things like fuel leak detection, explosive combinations of gasses, environmental monitoring, measurement of toxic gasses, one can imagine other possible applications associated with our work. I would suggest that there are a significant number but, as with aerospace applications, one would have to tailor the sensor for the application.
Over the years, because of an STTR (Small Business Technology Transfer) program award, we have worked with a company by the name of Makel Engineering towards the commercialization of some of these products.
NTB: If you could look into the future, where do you see sensor technology going in the next five years?
Dr. Hunter: I think there are two parallel directions one has to think about in terms of sensor technology. One is the idea that it has to be reliable, and often folks are concerned that sensor systems can add to unreliability of the overall system. They worry that sensors may not give you the correct data — that sensors may not be something that you can trust.
So, one of the things that I think you have to look at in terms of sensor technology and its development in the next five years is to concentrate on reliability, to be able to make it so that sensor systems can be believed, and so that people are not trying to remove sensor systems from their vehicles, but instead realize that when you remove a sensor, you remove information about what's going on in the environment. You become less aware of the conditions under which you're operating and less aware of information that might help you make a better decision. So, reliability is one direction that certainly sensor suppliers have to have in their minds as things progress in the next five years.
The other aspect that needs to be developed for sensor technology, I think, is the idea that one is trying to, overall, make smart sensor systems that not only include the sensor itself but also elements like power, signal conditioning, communications, multiple sensors, and sensors that tell you multiple parameters about the environment. You want to have all of those come into play into one, as we called them earlier, Lick and Stick type system. And that direction is not only for near-room-temperature applications, for which some of the electronics and some of the technology is more readily available, but also one of the directions we're trying to move towards are high temperature technologies that, not in the immediate term but over time, allow high-temperature Lick and Stick type systems. Again, we're not talking glues or adhesives or anything; what we really mean here is the concept of being able to demonstrate signal conditioning, power, telemetry, sensing, all in the same package and all be able to communicate at higher temperatures.
So, within the next five years one of the things that we would hope to do is, in the near room temperature area, be able to have a maturation and, by demonstration of our capabilities, more of an acceptance of the integrated smart sensor system approach for room temperature applications. But also, in the high temperature area, begin to demonstrate the possibility of high-temperature systems with, for example, signal conditioning, power, telemetry, and the like at higher temperatures. In fact, that’s a milestone we have in 2011 for the IVHM (Integrated Vehicle Health Management) program in Aviation Safety. It’s to demonstrate the basic capabilities of telemetry, signal conditioning, sensors, and power, all in a higher temperature system at 500 degrees C. That’s the first step in the high temperature area towards what we hope to be maturation of being able to put sensors, not just where it’s convenient and where the technology already exists, but being able to put them where they’re needed, where previously, harsh environments have limited the capabilities of sensor systems. Being able to, over time, implement smart sensor systems in the high-temperature regime, as well as near room temperature regimes as well.
To download this interview as a podcast, click here