Multimaterial microrobots for hyperlocalized sensing

Chemical sensing at the microscale is important for the study, monitoring, and control of e.g. biological and microfluidic systems. Hydrogen ions (H+) are among the most important analytes detectable in all aqueous solutions. The concentration of H+ is typically expressed as a value on the pH scale. Despite its importance, pH sensing at the microscale is not straightforward as on the macroscale. Currently, there are several techniques for performing microscale pH sensing. However, they all have their own limitations. Therefore, there is a need to develop new approaches towards microscale pH sensing. For this Ph.D. project, we selected microrobots as promising candidates for microscale pH sensing because of their small size and ability to maneuver in confined spaces.

The main objective of this Ph.D. project was to develop microrobots for localized pH sensing. To accomplish this, multimaterial microrobots were designed, fabricated, and characterized. The microrobots are composed of two materials: a hard polymer backbone and pH-responsive hydrogel. The pH-responsive hydrogel acts as the sensing element, while the hard polymer backbone enables transportation using optical trapping. Two-photon polymerization (2PP) printing was used to fabricate the multimaterial microrobots and optical trapping was used to manipulate the microrobots in microchannels.

Multimaterial microrobot designs were conceptualized and fabricated based on the requirements for optical trapping and pH sensing. Two promising designs were chosen after the design exploration phase. Since the method for determining local pH is by observing the shape change of the sensing elements, the shape change must be as large as possible. The print parameters used affect the extent of the shape change, and so an investigation of how certain print parameters affect the shape change was needed. For this purpose, a response surface methodology was performed and successfully used to optimize the print parameters of the pH-responsive hydrogel.

During the characterization phase of the remaining microrobot design, the sensing elements exhibited a 32 ± 11% size increase from pH 2.3 to pH 8.8, most of which occurs from ~ pH 5 to ~ pH 7. It was also observed that the swelling performance showed no signs of degradation after repeated cycling and long-term storage for up to 2 months. When manipulating the microrobots using optical trapping, speeds of ~14 μm s-1 were reached in a phosphate buffered saline solution. These results show that the microrobots are promising tools for qualitative pH sensing in microfluidic devices and can be stored and reused as needed. Their use as quantitative pH sensors is limited by the large variance of the shape change as well as the pH range that the shape change occurs.

To be able to visualize the in situ shape change of the sensing elements of optically trapped microrobots, a microfluidic setup was designed, constructed and integrated with our optical trapping setup. However, challenges related to leakages and bubbles in the microfluidic system could not be overcome, and so the setup could not be used for its intended purpose.

Before developing the microrobots, an understanding of how different print parameters affect the quality of 2PP-printed structures was needed. For this reason, several test structures were printed, with the aim of optimizing their print quality. Each test structure required a unique set of print parameters to obtain the best print quality. Parameter sweeps played a major role in understanding how different print parameters affected the print quality of the structures and for optimizing them. It was also seen that 2PP printing is an advantageous method to fabricate complex features and perform rapid prototyping to optimize such features.

Besides the print parameters, another key aspect to understand for the multimaterial microrobot fabrication were smart materials and their processing. A literature study was conducted about smart material systems for microscale 4D printing and was incorporated in a review paper. The focus of the review paper is on the different stimuli that smart materials respond to, how different shape responses can be obtained, and their applications.

Performing multimaterial printing is required to fabricate the microrobots. In practice, this process is often tedious and time-consuming. This is because every print step involves printing with a different photoresist, and the user must manually align the printer stage such that the new print job is precisely aligned with the previously printed structures. Hence, our group developed an automated alignment procedure based on computer vision to autonomously align the stage. Not only was this procedure helpful for fabricating the microrobots, but I then improved this procedure in various ways. For example, I designed new alignment markers and made the identification of the markers via computer vision more reliable. Furthermore, the procedure was expanded for use with any microscope objective and many different photoresists, which was demonstrated by printing multiscale and multimaterial structures. In addition, the alignment accuracy was characterized to be ~0.4 μm. This shows that the procedure not only reduces user workload but also provides high alignment accuracy.

Overall, this Ph.D. project has laid the groundwork for the development of advanced microrobots for chemical sensing. The microrobots presented in this project are mobile, respond to pH changes in their environment, and show good stability. However, their sensing performance is limited by issues such as reproducibility and the pH range in which they can be used. To mitigate these issues, suggestions for additional improvements are given in the thesis. These are based both on literature studies and on personal experience from the experimental work performed as part of the project. Successful implementation of these suggestions could strengthen the appeal of microrobots as solutions for microscale sensing.