The current standard for organ-on-chip substrate materials is polydimethylsiloxane (PDMS). PDMS has many beneficial properties such as biocompatibility, transparency, elasticity, and easy prototyping. The main disadvantages of PDMS are its hydrophobicity (reducing cell attachment and cell proliferation), quick hydrophobic recovery (within a few hours) after surface modification, and unselective absorption of small molecules, altering drug concentration during testing or causing cross-contamination between channels. To overcome these shortcomings, an alternative material lacking the disadvantages and retaining the advantages of PDMS is needed in the field. Ostemer 324, Ostemer 220 and a variation of PDMS (PDMS+) were characterized to determine whether they would be a suitable alternative to PDMS as substrate material. The materials were compared to PDMS and characterized for spin-coating and curing on silicon wafers, moulding, etching, surface wettability before and after surface modification (plasma treatment), stiffness, optical transparency, biocompatibility and absorption of small molecules. The results of this work show that Ostemer 324 appears to be a promising alternative, with similar characteristics to PDMS. The main advantage is its hydrophilic surface, which becomes even more hydrophilic after plasma treatment, and slow hydrophobic recovery rate. However, its curing time is longer, the material is stiffer and not easy to mould. Ostemer 220 shows very poor adhesion on Teflon coated silicon wafers, is hard to mould, and shows inconsistent results during biocompatibility testing. PDMS+ is very similar to PDMS, and improves upon the latter in terms of surface wettability after surface modification, hydrophobic recovery and small molecule absorption.

The lack of reliable disease models hampers the development of safe and effective drugs for humansas they fail to precisely identify the cardiotoxicity profile of drug compounds. Recent technological advancements have led to the development of the"Organ-On-Chip"(OOC) technology. OOCs are miniaturized micro-engineered devices which aim to accurately recapitulate the physiological conditions of different organs. In order to develop a Heart-On-Chip (HOC) device as an in-vitro screening platform for cardiotoxicity, it is essential to integrate micro-electrodes (MEAs) in a flexible polymeric substrate that can be manufactured with standard micro-fabrication techniques. Besides electrical stimulation and recording, the device should be capable of accommodating the different cellular structures of the human heart along with its vasculature, which is in continual exposure to perfusion with fluid. In this manner, a platform that is cleanroom-compatible, scalable and physiologically representative of the human heart can be developed.The OOC device of Bi/ond provides an ideal platform in which MEAs can be successfully integrated. The device also offers microfluidic channels that supply the cardiac co-culture with the necessary perfusion to mimic the blood flow through the heart. In order to integrate TiN MEAs in the Bi/ond platform,several technological challenges need to be tackled. In previous work, polymers were often utilized as the insulation layer for the metal interconnects. Processing of polymers in a silicon-based cleanroom environment requires additional dedicated equipment. Therefore, it is necessary to investigate the feasibility of alternative dielectric, cleanroom-compatible materials, such as silicon nitride. Secondly, the metal lines experience considerable amounts of stress at the interface of the flexible membrane and the rigid silicon frame during the fabrication and post-fabrication process. By customizing the mechanical design of the PDMS membrane as well as the metal interconnects, the stress developed at the interface can be reduced, which can ultimately improve the yield of the devices due to the formation of fewer cracks in the metal lines.In this thesis, three goals were defined and achieved in order to develop a proof-of-concept for a HOC device with silicon nitride encapsulated TiN MEA in a thick PDMS layer, which can be further integrated in a microfluidic device. First, a study of the compatibility and characterization of SiNx insulation layers with PDMS membranes was performed. The adhesion between the nitride and PDMS membrane was improved by sandwiching a thin silicon oxide layer in-between the nitride and the PDMS. Additionally, the optimum deposition techniques for the nitride layers were demonstrated. Secondly, the yield of the devices was improved by optimizing the designs of the interconnects and the shape of the PDMS membranes. A novel turtle shaped design for the membrane was developed which helped to redistribute the stress at the interface. This novel design resulted in a yield higher than 75%.Thirdly, a silicon-based cleanroom compatible fabrication protocol was optimized and defined for processing a HOC device with a TiN MEA encapsulated by SiNx in a 200μm thick PDMS membrane.Moreover, based on the techniques developed by Bi/ond, this device module was successfully integrated in a microfluidic device. Finally, Electrochemical Impedance Spectroscopy (EIS) measurements were carried out to confirm the functionality of the electrodes. The average impedance value at 1 kHz of three working electrodes was measured to be in the order of 800 kΩ, proving that an electrical connection was successfully established between the embedded TiN electrodes and the bondpads on the silicon frame.