Abstract
The human gut microbiota (GM) plays a vital role in the breakdown and digestion of indigestible carbohydrates, with the subsequent production of short-chain fatty acids (SCFA) and different gas species including carbon dioxide (CO2), hydrogen sulfide (H2S), hydrogen (H2), methane (CH4) nitric oxide (NO), and ammonia (NH3). These byproducts of microbial fermentation are key indicators of the impact that a certain diet can have on the GM. For instance, significant changes in NH3 are an indicator of the alterations in dietary protein components, that has been linked to several gastrointestinal diseases and toxicity. Thus, the rate and location of these compounds within the gastrointestinal (GI) tract are pivotal to investigating the relationship between gut health and diet. An easy, ethical, and efficient method for understanding and monitoring the complex transformation processes that take place during the microbial community fermentation process and their subsequent gaseous biomarkers are in-vitro fermentation models. These models are closed anaerobic environments used to investigate the capability of GM to synthesize several dietary fibers by rating the different byproducts. These analyses are usually measured offline by using expensive analytical techniques, such as high-performance liquid chromatography (HPLC) or gas chromatography coupled with mass spectrometry (GCMS). For this, the development of sensors able to monitor fermentation byproducts, such as gases, in in-vitro gut models represents a cost-effective, easy, and less invasive alternative to detect and monitor continuously and in-real time these important by-products of the gut fermentation process. This Ph.D. thesis focused on the development and evaluation of gas sensors for the continuous and real-time monitoring of gaseous biomarkers produced during microbial fermentation in the in-invitro fermentation model known as the Simulator of Human Intestinal Microbial Ecosystem (SHIME®). The latter is widely regarded by the scientific community as one of the most accurate models of the human microbial ecosystem. To effectively sense gaseous biomarkers in the GI tract, which is characterized by a highly humid, acidic, and anaerobic environment full of microbes, a systematic sensor development approach is needed. This thesis presents a holistic sensor development approach, starting with the evaluation of the sensor layout, substrate, and material optimization, to the final operation of the device in the SHIME®. For the development of the sensor, initially, a systematic study on the effect of the selected fabrication techniques of the custom-designed electrodes was performed. Specifically, three different additive manufacturing technologies (dispense printing, screen printing, and inkjet printing), have been extensively exploited, optimized, and compared. Conductive networks of single-walled carbon nanotubes (SWCNTs) were selected as suitable materials for the realization of the gas-sensing thin layer and an optimization of the deposition technique (i.e., spray coating) was performed. Optical, topographic, and electrical analysis were used for both, electrodes and sensing material characterization. From 300 µm nominal spacing between fingers, we obtained a decrease of 25%, 13%, and 5% on the printed spacings with dispense, screen, and inkjet printing, respectively. At 100 ppm of NH3 a response of 33%, 31%, and 27% with the dispense-, inkjet-, and screen-printed sensors, respectively were obtained. Screen printing was chosen as the best technique for carbon nanotubes (CNTs) -based NH3 gas sensor fabrication after taking into account the versatility of the fabrication process, the sensor response, as well as the target application (SHIME), and market perspective. In the second part, polydimethylsiloxane (PDMS) based membranes, as a gas selective and protective barrier, were fabricated for the improvement of the selectivity and sensitivity of the chemiresitive CNTs NH3 gas sensors previously developed. PDMS is a low-cost, flexible, bio-compatible, and easy-to-process polymer with outstanding gas permeation properties, however, a systematic fabrication process is needed to find the optimal membrane thickness (permeable to NH3 and impermeable to CO2, and CH4). The response to NH3 of the chemiresistive CNTs-based sensor coated with 12 µm thick membrane was enhanced by 1.13 % and 11.25 % for 3 ppm and 100 ppm with respect to the sensor without the PDMS layer. The results showed that the membrane inhibited the surrounding humidity, allowing a pure interaction between the CNTs network and the gas. Furthermore, the membrane improved the sensor selectivity as it restricted the permeation of CO2 and CH4. Finally, in collaboration with the Micro4Food Lab of the Free University of Bozen Bolzano, the PDMS-coated SWCNTs-based chemiresistive sensors were used to monitor in-line the gases produced in the gastrointestinal tract of the in-vitro model SHIME®. The developed sensors continuously monitored the different stages of gas production induced by the undergoing microbial fermentation activities during two weeks of SHIME® experiment. This demonstrates that the proposed sensor development approach paves the way to a deeper understanding of the difficulties and potential solutions associated with operating gas sensors in a GI tract environment. It also highlights the need of incorporating inline sensors to monitor the effects of various diets on the GM through the identification of other interesting fermentation byproducts.