Date of Award
Doctor of Philosophy in Chemistry
Jason R. Dwyer
A nanopore—typically defined as a through-hole with dimensions <100 nm in all directions that functions as the sole path between two electrolyte reservoirs—is a robust single molecule sensor element which has enjoyed a wealth of applications spanning genomics and proteomics, with fledgling contributions to glycomics over the past two decades. Two classes of nanopores exist—biological and solid state. Biological nanopores, for example, α-hemolysin, are highly reproducible and precise—with nanopore lengths and critical constriction sizes that are well known and reproducible. This is not the case with solid state nanopores. Assuming total nanopore length is equal to the nominal thickness of the membrane provided by the manufacturer is a standard practice in the nanopore field. However, given fabrication tolerances, there is some room for error, in certain instances close to 60% of the provided nominal thickness. Any error in nanopore length will couple to errors in the radius calculation. Another two key assumptions are: i) the nanopore has a cylindrical shape unless (and often even if) the shape is otherwise known and ii) a single nanopore through the membrane is formed when one is intended. These issues were addressed by developing a framework that shows errors in harboring such geometric assumptions and eventual consequences for nanopore-based sensing experiments.
The focus of nanopore-based sensing has been predominantly on DNA and protein profiling with only fledging contributions to glycomic profiling. Silicon nitride-based solid state nanopores were used to understand translocation conditions related to alginate and then to study source variability associated with alginates. The two alginates used (from two different sources) gave two distinct signal patterns. Heparin, a common anticoagulant was contaminated in 2008 with over sulfated chondroitin sulfate (OSCS)—a structurally similar adulterant—which lead to ~120 deaths in the United States. Nanopores, with sizes ranging from ~8.6 nm- ~13.6 nm were used to test the ability to flag the presence of OSCS in a contaminated heparin sample—all four unique nanopores used were able to flag the presence of the OSCS contaminant proving the diagnostic capability associated with nanopore sensing.
Surface modification techniques, for example, hydrosilylation, silane chemistry and electroless gold plating not only tune the size (minimum radius, �0, and total nanopore length, �) but also change the intrinsic surface chemistry. Hydrosilylation on planar silicon nitride—a less challenging and less volume-constricted environment compared to nanopore inner walls—has been shown to be possible photochemically and thermally. The photochemically driven hydrosilylation was scaled down to the nanopore level—decorating inner nanopore walls in a challenging zeptoliter volume—using a range of functional groups to potentially overcome unfavorable conditions such analyte “sticking” problem while tuning analyte residence time favorably. Choice of molecule plays a significant role—one with a reactive terminal group such as hydroxyl or amine allowed for subsequent reactions, through condensation and click reactions, respectively, which are fast and facile, thus allowing for further modification of the size and surface chemistry of the pore. We observed the residence time of λ-DNA to increase with positive charge of the pore surface at pH 7, with bare, hydroxyl terminated, and amine terminated functional pores having peak residence times of ~250 μS, ~450 μS and ~1000 μS respectively.
A carefully configured electroless plating procedure was used to deposit gold directly on silicon nitride. Since silicon nitride is an insulator, conventional electroplating would be futile—hence electroless plating. The plating was done at both 3° C and 10° C. The mean grain size of the gold grains plated at 3° C were found to be ~20 nm in radii. These nanostructured plated surfaces were also used to enhance the Raman signal of 4-nitrobenzothiol (test-molecule). The same plating method was extended to paper, nanocellulose, acrylate polymer grafted silicon nitride, nanoporous silicon nitride and Silmeco (a commercial substrate with a pillar like architecture) to create low cost surface enhance Raman active substrates. Enhancement values as high as ~106 for both acrylate polymer grafted silicon nitride and Silmeco was observed.
Patterned solution-phase gold depositions have great promise for electronics, photonics, and sensors such as nanopores as well—especially considering augmenting nanopore function with structures such as transverse electrodes. For nanopores and other fragile architectures, mechanical non-contact and cleaning ease (especially by simple rinsing) are key elements in designing modification and fabrication methods. Hydrosilylation meets these expectations as it can be guided and restricted to specific regions by manipulating the exposure of light (UV) to the surface. Hydrosylilated alkanes were used as a suppressing layer, for metal deposition in combination with electroless deposition to create spatial patterns of gold on silicon nitride. However, key modifications to the existing gold plating scheme had to be made. Key washing steps and replacement of Sn(II) chemistry with Pd(II) chemistry was done to increase the spatial selectivity of the plated patterns. Spatially selective patterns with lateral spacings as small as ~30 μm have been fabricated using this method.
Bandara, Nuwan Dhananjaya, "Profiling and Modification of Silicon Nitride Based Planar Substrates and Nanopores" (2018). Open Access Dissertations. Paper 718.