Gas sensors based on carbon nanofibers: a low power consumption approach

Abstract

Gas sensors can be found in many activities ranging from environment protection, risk prevention, agriculture and even in food, chemical, and petrochemical industries. There exist different technologies for gas sensors depending on the transduction mechanism: mass-sensitive, optical, calorimetric, magnetic, electrochemical and conductometric. In this work, conductometric (or resistive) gas sensors are studied. Conductometric devices base its operating principle on the variation of the electrical conductivity (resistivity) or conductance (resistance) of a reactive (active) material interacting with gas. A chemical reaction between the active material (surface or bulk) and the gas occurs. This reaction induces a variation on some electrical property of the material resulting in a change on the electrical signal (conductivity or resistivity of the active material) of the sensor. Therefore, the sensor material should be compatible with the mentioned properties above. A carbon based material was chosen to be the reactive compound for the conductometric sensors. This material, a specific type of carbon nanofibers (CNFs), shares some suitable properties with other trendy carbon based materials such as carbon nanotubes or graphene. Conductometric gas sensors usually are composed of two main parts: the already mentioned reactive material and the heater device. The heater is required in order to stabilize the temperature of operation and to activate a desired chemical reaction. Unfortunately, despite the efforts to improve the heater technology, this component is still the most power demanding part of the overall device. The here studied sensors have been characterized with a heater device, but also alternative energy sources and other sensing strategies have been tested in order to reduce the energy cost. Among these, the use of ultraviolet and visible light sources were tested in order to modulate the sensor properties. In addition, another non-common strategy was used to operate the sensor: the so called self-heating effect (or Joule effect). To obtain the electrical signal of a sensor, the reactive material have to be scanned, usually a current (or voltage) is applied to the sensor, then, the voltage (or current) is read. If the probing magnitude is increased, the power dissipation through the sensing material, and its temperature, also increases. Therefore, the sensor could be operated without a heater device with a considerable reduction of its power consumption. Moreover, the self-heating also allows reducing the fabrication complexity, as there is no need of the heater element.

In summary, the main objective of this work was to characterize the CNFs as a reactive material for conductometric sensors for low cost applications. First, the CNFs properties (electrical,

mechanical, response to light and gases) were screened with the aim to assess the applicability of the sensing material (O. Monereo et al., 2013, Flexible sensor based on carbon nanofibers with multifunctional sensing features). Then, the sensor was tested with the use of temperature modulation (S. Claramunt et al., 2013, Flexible gas sensor array with an embedded heater based on metal decorated carbon nanofibres). At this point, a more detailed characterization of the gas sensing properties with O2, H2O, NO2 and NH3 was conducted. Then, the use of continuous self-heating operation (O. Monereo et al., 2015, Self-heating effects in large arrangements of randomly oriented carbon nanofibers: Application to gas sensors) and pulsed self-heating application (O. Monereo et al., 2016, Self-heating in pulsed mode for signal quality improvement: application to carbon nanostructures-based sensors) were found to be efficient methodologies to modulate the sensing characteristics of sensor devices, based on large arrays of nanostructures. Among the benefits achieved, the sensor presented improvements on stability, specificity, the detection time modulation, all along the simplification of device fabrication and the reduction of the power consumption. Finally, the phenomenon of self-heating in carbon nanofibers and its origin was studied (O. Monereo et al., 2016, Localized self-heating in large arrays of 1D nanostructures). In addition, the use of ultraviolet and visible light as alternative energy sources was also assessed and compared with the self-heating operation. Finally, the applicability of self-heating was also tested in graphene based (reduced graphene oxide) and metal oxide based (ZnO) devices to test the applicability of self-heating in other relevant sensing materials.

El objetivo principal de esta tesis es la caracterización de las nanofibras de carbono (CNFs) como material reactivo para sensores resistivos de gas para aplicaciones de bajo consumo. Primero, las propiedades eléctricas, mecánicas y respuesta a luz y gases de las CNFs fueron evaluadas para comprobar la aplicabilidad del material sensor (O. Monereo et al., 2013, Flexible sensor based on carbon nanofibers with multifunctional sensing features). Posteriormente, la respuesta del sensor a gases fue estudiada con modulación de temperatura (S. Claramunt et al., 2013, Flexible gas sensor array with an embedded heater based on metal decorated carbon nanofibres). En este punto, una caracteritzación más detallada de la respuesta del sensor a gases modulados con temperatura se realizó con O2, H2O, NO2 y NH3. A continuación, el uso de la metodología de auto-calentamiento continuo (O. Monereo et al., 2015, Self-heating effects in large arrangements of randomly oriented carbon nanofibers: Application to gas sensors) y pulsado (O. Monereo et al., 2016, Self-heating in pulsed mode for signal quality improvement: application to carbon nanostructures-based sensors) han sido probados como formas energéticamente eficientes para modular la respuesta de sensores basados en grandes matrices de CNFs. Entre los beneficios encontrados, consta una mejora de la estabilidad, especificidad, la modulación del tiempo de detección; todo añadiendo la simplificación de la fabricación. Finalmente, el origen del fenómeno de auto-calentamiento en CNFs fue estudiado en detalle (O. Monereo et al., 2016, Localized self-heating in large arrays of 1D nanostructures). Además, la aplicabilidad de la metodología fue también probada en nanotubos de carbono, óxido de grafeno reducido y nanohilos de óxido de zinc. Finalmente, el uso de luz ultraviolada y visible ha sido estudiado como a energías alternativas para la modulación de los sensores de gases de CNFs.