17 Décembre – Thesis defense - Ioannis Nikolaou

10 h Conference room - laboratory IMS (Talence campus)

Properties of graphene by chemical recognition on Surface Acoustic Wave (SAW) sensors: Application to the detection of chemical compounds in the gaseous state.

The aim of this work is to introduce first the overall interest of Surface Acoustic Wave (SAW) devices based on carbon allotropes, such as CNTs and graphene, especially used for environmental or bio-sensing applications i.e. Volatile Organic Compounds (VOCs) as biomarkers in breath. Further, this work is illustrated with a versatile acoustic wave transducer, functionalized with Graphene Oxide (GO), synthesized for ethanol, toluene, ammonia, carbon dioxide, nitrogen dioxide and humidity detections. The devices were designed, simulated, fabricated and characterized according to the target selection and the aspects of the SAW devices. For example, particular ratios between the length and volume of the deposited graphene on SAW devices were selected in order to detect sub-ppm concentration levels of NH3 and NO2, respectively. Further, the novel properties of the graphene-based acoustic wave devices were studied and modified according to their optimal detection levels and validated over any further electrical and vapor characterization measurements.
Moreover, the devices were characterized by a Vector Network Analyzer (VNA), Raman spectroscopy, Scanning Electron Microscopy (SEM) and Atomic Force Microscopy (AFM), performed after each step of the fabrication, attesting that our method has a significant impact to the quality performance of the graphene-based surface acoustic wave devices. We have subsequently employed particular analytical modeling to investigate the electro-mechanical properties of the GO, as well as to extract the elastoviscosity parameters of several GO layers and their impact on the acoustic devices, theoretically and experimentally. While that, we observed a strong correlation of the results with the number of coatings of the GO solution on the supported SiO2, since the properties of the GO fabricated materials were highly dependent on the specifically designated thicknesses, themselves intrinsically influenced by the material viscosity and Young’s modulus. It has been previously reported in the literature that Young’s modulus generally decreases with increasing of the GO thickness. Instead, we conclude that the number of the GO layer samples display a compressive internal strain, which does not fully relax after the fabrication process. We attribute this behavior to the uncontrollable large number of the graphene-based oxide sites during the fabrication process, as it has been observed later on the Raman spectra, respectively.
Finally, we have studied the gas sensing characteristics of our devices at room temperature as well as at higher temperatures up to 60 °C. The main reason that the temperature was kept at low levels was to test our devices in a very competitive way based on the current industrial demands, minimizing energy consumption, and/or to overcome some of the latest literature detection levels. The measurements confirmed that some of the detections were efficient based on the graphene devices, and as a result, it is possible to open a new discussion regarding the geometry and the morphology of the very specific GO materials. Some of the device measurements were attributed to a better understanding of the detection mechanisms such as physical or chemical adsorption and further in many cases by distinguishing the adsorption and absorption phenomena. At very low concentration levels of the VOCs, we have observed signatures of few Hz variations for a 100 MHz resonant frequency, but high enough, which implies that further investigation is needed to identify selectivity or specificity levels of certain target analytes. Based on the different geometry and thickness levels, the dominant mechanisms may vary in our samples. At higher concentration levels, the sensitivity showed frequency/temperature-robust results according to the very stable oscillation levels which could be identified as the baseline or initial detection levels of each target analyte, subsequently.

Event localization