Samenvatting
Liquid chromatography has significantly evolved throughout the last century, nowadays receiving the status of being a mature technology. One of the few ways to acquire a quantum leap in efficiency is to create perfect order in the stationary phase. The potential of additive manufacturing to create structures with an unlimited freedom of design has finally granted us the ability to create the desired perfect three-dimensional order. The only additive manufacturing technology capable of creating structures on a micrometer scale, needed to be competitive with commercially available analytical columns, is two-photon polymerization or 2PP.
The advantage of perfect order was first investigated in this work by performing high-accuracy computations of band broadening inside a face-centered cubic packed bed of spheres using computational fluid dynamics. The influence of the zone retention factor was investigated over a series of reduced velocities. A ready-to-use analytical expression describing the mobile phase mass transfer contribution (hCm) in packed bed columns was constructed. It was furthermore shown that the existing Sherwood-correlation of Wilson-Geankoplis is defective for chromatographic purposes and that the mass transfer from the mobile zone to the particles is much more complicated as expected.
In the present thesis, 2PP was used for the creation of micrometer-scaled stationary phase supports. The research was performed from the ground up, starting with the search for the ideal substrate. Fused silica was selected as the substrate of choice, showing good solvent resistance, an ideal refractive index for automatically finding the interface, and a broad process window of printing parameters. Using fused silica as the substrate in combination with the commercially available high-resolution printing photoresist IP-Dip, the influence of printing parameters on the creation of 2.5D pillar array columns and 3D tetrahedral skeleton models was examined based on the structure’s uniformity, size, and shape. The printing parameters were optimized to minimize the printed geometry, resulting in pillars with a minimal printed diameter of 900 nm and a tetrahedral skeleton model with a minimal skeleton diameter of 400 nm for an 80% porous monolith and 650 nm for a 60% porous monolith. These results clearly indicate that two-photon polymerization is capable of creating sub-micrometer scaled stationary phase supports, making these structures competitive with commercial columns.
After proving that 2PP is a suitable manufacturing technique, a real TSM column was fabricated within an etched microfluidic channel. The whole manufacturing process, from etching the channels to printing the column is described. Several challenges which were encountered during the column manufacturing are described and attempted to be solved or minimized, including the detachment of prints from the fused silica surface, the influence of the proximity effect, shrinkage, printing problems at the etched sidewalls, and loss of quality as a function of time. Finally, several chip sealing strategies were investigated and discussed in this work. It was found that the use of dry-film photoresist showed the greatest potential for sealing off the chromatographic column. Testing the printed columns by performing flow experiments, indicated that there are still many problems complicating the chip bonding. Most of these problems are related to the printing process and are unavoidable, such as shrinkage and the shadowing effect. Nevertheless, all results and challenges of initial flow experiments are discussed.
The advantage of perfect order was first investigated in this work by performing high-accuracy computations of band broadening inside a face-centered cubic packed bed of spheres using computational fluid dynamics. The influence of the zone retention factor was investigated over a series of reduced velocities. A ready-to-use analytical expression describing the mobile phase mass transfer contribution (hCm) in packed bed columns was constructed. It was furthermore shown that the existing Sherwood-correlation of Wilson-Geankoplis is defective for chromatographic purposes and that the mass transfer from the mobile zone to the particles is much more complicated as expected.
In the present thesis, 2PP was used for the creation of micrometer-scaled stationary phase supports. The research was performed from the ground up, starting with the search for the ideal substrate. Fused silica was selected as the substrate of choice, showing good solvent resistance, an ideal refractive index for automatically finding the interface, and a broad process window of printing parameters. Using fused silica as the substrate in combination with the commercially available high-resolution printing photoresist IP-Dip, the influence of printing parameters on the creation of 2.5D pillar array columns and 3D tetrahedral skeleton models was examined based on the structure’s uniformity, size, and shape. The printing parameters were optimized to minimize the printed geometry, resulting in pillars with a minimal printed diameter of 900 nm and a tetrahedral skeleton model with a minimal skeleton diameter of 400 nm for an 80% porous monolith and 650 nm for a 60% porous monolith. These results clearly indicate that two-photon polymerization is capable of creating sub-micrometer scaled stationary phase supports, making these structures competitive with commercial columns.
After proving that 2PP is a suitable manufacturing technique, a real TSM column was fabricated within an etched microfluidic channel. The whole manufacturing process, from etching the channels to printing the column is described. Several challenges which were encountered during the column manufacturing are described and attempted to be solved or minimized, including the detachment of prints from the fused silica surface, the influence of the proximity effect, shrinkage, printing problems at the etched sidewalls, and loss of quality as a function of time. Finally, several chip sealing strategies were investigated and discussed in this work. It was found that the use of dry-film photoresist showed the greatest potential for sealing off the chromatographic column. Testing the printed columns by performing flow experiments, indicated that there are still many problems complicating the chip bonding. Most of these problems are related to the printing process and are unavoidable, such as shrinkage and the shadowing effect. Nevertheless, all results and challenges of initial flow experiments are discussed.
| Originele taal-2 | English |
|---|---|
| Toekennende instantie |
|
| Begeleider(s)/adviseur |
|
| Datum van toekenning | 7 sep 2022 |
| Status | Published - 2022 |