Spin waves and magnetic nanoparticles for gas sensing applications
Conductometric semiconducting metal oxide gas sensors are currently the most commonly studied group of gas sensors. However, such conductometric sensors have poor sensitivity at room temperature and a high sensitivity at high temperatures, and thus require high levels of energy consumption. Different types of magnetic materials (e.g., ferrites) have also been used in gas sensing applications.1, 2 Only a few experimental studies, however, have so far reported the perturbation of the magnetic properties of these materials that are caused by interactions with gases.3
In most of the previous research, only the electrical properties of the magnetic materials in gas sensors have been considered. This relatively slow advance in the development of magnetic chemical sensors seems to be caused by the absence of inexpensive solutions for accurate measurement of the weak induced variations in the magnetic characteristics of the sensitive material. However, tunable magnetostatic surface spin wave (MSSW) oscillators, which are based on yttrium iron garnet (YIG),4, 5 are known to possess sensitivity that is more than sufficient to register such weak variations.
In this work we have developed a new type of chemical sensor that is based on an MSSW oscillator (as a magnetic field detector) combined with a layer of magnetic nanoparticles that acts as a sensitive gas layer (i.e., its magnetic properties change), as shown in Figure 1. We call these devices ‘magnonic gas sensors.’6 We use magnetic nanoparticles because of their very large surface-to-volume ratio, which provides a significant improvement in sensor sensitivity. This increased sensitivity is due to the surface character of the gas–nanoparticle interactions.
Some important properties of MSSWs include their very low level of propagation loss at microwave frequencies, high-loaded ‘Q value’ (indication of under-damping), small wavelength, and high tunability (from 0.2–20GHz). We also note that the frequency of a spin-wave oscillator can be tuned by changing the magnitude of a bias magnetic field (HB), while the MSSW wavelength remains constant. In our case, we applied a bias magnetic field (about 200Oe) perpendicularly to the wave propagation direction and parallel to the YIG film plane. We are therefore able to approximate the oscillation frequency of our sensor (f) as: 
where f0 is the unperturbed oscillation frequency, δfSL is the frequency shift caused by the interaction between the sensitive layer and a toxic gas, γ is the gyromagnetic constant (about 2.8MHz/Oe), HSL is the static magnetic field induced by the sensitive layer, and δHSL is the variation of the magnetic field induced by the interaction between the sensitive layer and the gas.
We have tested our sensor with different concentrations of volatile organic compounds (VOCs) in dry air. The VOCs tested were dimethylformamide, isopropanol, ethanol, benzene, toluene, and xylene. Our testing cycles were conducted with a constant exposure and a purge of 1min. During the exposure of each concentration to the toxic gases, the frequency of the sensor shifts significantly. It is therefore clear that during the exposure period, the interaction with the VOCs produces a change that causes a frequency shift in the oscillating device. This results in a change to the total magnetic field applied in the YIG film.
In the next step, we purged our sensor with synthetic dry air. This causes the frequency of the sensor to shift back to its initial value. The gas-induced magnetic changes in the nanoparticle layer are thus reversible, which is an essential property for practical sensors. In addition, we measured each gas concentration twice and we obtained repeatable results from the sensor each time. As an example, the results from the case of 100ppm dimethylformamide are illustrated in Figure 2. We find that the maximum frequency shift of the sensor—which occurred after 1min of exposure—was 53kHz. This value is much greater than the system's noise factor of 0.1kHz. The results of our tests also indicate that the magnonic sensors have higher sensitivity at room temperature than more common conductometric sensors, which is an important advantage for practical applications.
The magnonic sensor we have presented in this work exhibits fast detection and recovery responses, with high sensitivity and reproducibility for detecting gases at room temperature. From our experimental results we observe several issues that will be addressed in our future work. First—to discriminate and classify different gases—it will be interesting to investigate the selectivity of different ferrite nanoparticles for each gas. Second, we can improve the sensitivity of our current device to detect lower gas concentrations. This work may involve using higher operating frequencies, different sensor temperatures, smaller nanoparticles, or new settings for magnetostatic spin wave propagation. We believe that our demonstration of this new sensor type opens up a promising field of research for so-called e-nose applications. It should be possible to produce ‘magnonic e-noses’ that include a variety of nanostructured materials with appropriate magnetic properties, e.g., ferrites with different morphologies (nanoparticles, nanorods, core–shell microspheres, and thin films).
This work was supported through Programa de Apoyo a Proyectos de Investigación e Innovación Tecnológica (PAPIIT) UNAM projects IA103016 and IG100314.
Daniel Matatagui is currently a research professor in the Sensor Group. He works on the development of sensors for volatile compounds and biosensors.
Oleg Kolokoltsev is a research professor and is currently head of the Microwave Photonics Group. He received his PhD in physics in 1991 from the Taras Shevchenko National University, Ukraine, where he was also faculty and head of the Microwave Devices Laboratory. He has previously been a Royal Society Research Fellow at the University of Salford, UK.
Naser Qureshi is a research professor and part of the Microwave Photonics Group. He received his BA from Princeton University and his PhD from the University of California, Santa Barbara, in 2002. He has also worked as a postdoctoral fellow at the University of California, Santa Cruz. His primary research interest is the development of novel methods for high-resolution microscopy and their application in nanoscience.
Elsi Mejía-Uriarte obtained her MS in physics from the National University of Trujillo, Peru, in 1998 and her PhD in materials science and engineering from UNAM in 2003. She is currently a full professor in the Department of Optics and Microwaves. Her research interests include luminescent ions in solid photonic crystals, high-pressure sensors, metal and luminescent nanostructures, optical characterization under high pressure and high temperature, as well as instrumentation.
José Saniger is a research professor. He received his PhD in chemical sciences from the Complutense University of Madrid, Spain, in 1998. He is also the founder of the Materials and Nanotechnology Group of CCADET. His research areas of interest are molecule/substrate interactions, nanostructured and catalytic materials, gas-opto electrical sensors, and molecular spectroscopy.
