Development of Thermal Sensors and Sensing Principles for Monitoring Physical or Chemical Processes and the Rapid Quantification of Biological Samples

Abstract

The thermal properties of a material describe how a material responds to changes in temperature and the application of heat. For many applications, thermal properties are a determining factor in the selection of materials. Examples range from selecting insulation materials for buildings to selecting textiles with specific thermal properties for optimal wearing comfort. Within the field of materials engineering, the composition of materials is often altered to obtain new materials with more desirable thermal properties. To compare the thermal performance of most materials, researchers can rely on commercially available sensors. These sensors can accurately determine absolute thermal properties using a straightforward, direct current (DC) pulse­based measurement approach. However, a measurement of thermal properties can reveal more about a sample than only its ability to conduct or dissipate heat. Thermal properties can vary based on the composition of a material. Consequently, thermal measurements can also be used to obtain information about the composition of unknown samples. This alternative purpose of thermal sensing, for example, is used routinely in gas chromatography to monitor the composition of gas samples. Recently, several studies have suggested that measuring thermal properties can also reveal valuable information about biological samples. Findings of such studies showed that a measurement of thermal properties could be used to determine the size of a cell population or even distinguish between live and dead cells. These results highlight the potential of measuring thermal properties in the field of biology. However, in these works, a more complex alternating current (AC) based measurement approach is used for determining the thermal properties. The complexity of this measurement technique forms a barrier for biologists to fully exploit thermal measurements for research purposes. This doctoral research aimed to develop new thermal sensors and sensing principles that are better suited for biological applications. For this purpose, we first looked into the context in which current biological research is performed and noticed two main trends. Firstly, there is the ongoing trend of large­scale parallel testing using microplates. Secondly, there is a more recent trend towards lab­on­chip applications that focuses on obtaining information about ever smaller liquid sample sizes. This thesis presents DC­based sensing solutions for both formats. The chapters of this work are therefore categorized into non­microfluidic and microfluidic applications. Initially, a study was done on the usability of a DC­based thermal sensing principle for monitoring variations in the thickness of a generic thin film at the sensor interface. The intention of this study (Chapter 3) was to serve as a proof­ofprinciple for later possible studies on biofilm formation or cell proliferation. From Finite Element Modeling (FEM) simulations, a novel method was suggested for deriving variations in thin film thickness from pulse­based thermal measurements. This so­called Transient Thermal Offset (TTO) method was later demonstrated in practical experiments. The results suggested that pulse­based measurements might indeed be used for real­time monitoring of biofilm formation and cell proliferation. Moreover, the experimental results were obtained using thermal sensors that were designed in­house and fabricated externally at a fraction of the cost of commercially available pulse­based sensors using standard flexible Printed Circuit Board (PCB) manufacturing technology. This cost reduction opens possibilities for high­throughput parallel thermal sensing applications that were previously not possible for economic reasons. In the following study (Chapter 4), it was first demonstrated that the in­house developed sensors are sensitive towards varying numbers of cells at the interface. This was done by performing cell sedimentation experiments using thermal sensors that were incorporated into the bottom of milliliter­sized reservoirs. Later, cell proliferation experiments were performed. The results confirmed that the thermal sensors were able to monitor the proliferation process by exploiting the thermal insulating effect of a growing cell culture at the sensor interface. In a final non­microfluidic study (Chapter 5), a novel sensor design was presented. In this work, thermal sensors were designed specifically to fit within the well diameter of standard 96­well microplates. However, this work was not only limited to reducing the sensor size. Also, new sensor functionality was presented. In addition to measuring thermal properties, this work demonstrated the possibility to simultaneously measure temperature. This double functionality was first demonstrated in proof­of­principle experiments and was later used during cell proliferation experiments. The results of the proliferation experiments confirmed earlier findings that changes in cell number at the interface resulted in measurable changes in thermal insulation. Additionally, the sensors were able to pick up temperature changes caused by the metabolic activity of the cells. In short, the thermal sensors were able to simultaneously monitor variations in cell number and cell activity inside a microplate format. The increasing interest in lab­on­chip applications generates a growing demand for reliable and robust sensing principles that can be implemented into a microfluidic channel. However, previously, pulse­based measurements were not well suited for microfluidic samples due to issues related to the thermal probing depth. Based on our earlier findings that led to the introduction of the TTO measurement principle, a solution was suggested for eliminating these issues. A novel idea for a microfluidic device was proposed for performing the TTO method on small liquid samples (Chapter 6). The functionality of this device was first evaluated successfully using FEM. Next, a prototype microfluidic device was fabricated. The experimental results were in accordance with the FEM simulation results. With the prototype device, it was possible to determine the absolute thermal conductivity of fluid samples smaller than 3 µl with an accuracy of 0.5% and a standard deviation of below 0.009 W/mK. In a following study (Chapter 7), it was demonstrated that the TTO method can also be used for measuring thermal conductivity in flow, as well as for measuring the flow rate of liquids inside a microchannel. These additional functionalities might later be used for continuousflow microfluidic applications. In summary, this dissertation pushed the boundaries of thermal sensing. The most significant contributions are the demonstrated reduction in sensor cost, the demonstration of parallel thermal sensing in a microplate format, and the introduction of the TTO method, which knows applications on thin films as well as in microfluidics. The presented sensors are ready to be used by biological researchers for exploring thermal­based microplate assays or thermal­based sensing on lowvolume biological samples. All sensors can be used for long­duration monitoring as well as for short­duration measurements for rapid quantification. Even though this work focused on biological applications, the results suggest that the presented sensors and sensing principles can also be used for monitoring various other types of physical or chemical processes.