Pillar Density Modulation in a Semi-packed MEMS Column

Gas chromatography is a technology that is constantly moving forward. The current trend for this field is moving towards miniaturizing the columns towards achieving high-speed separations. Groups are constantly researching different geometries, topographies, and stationary phases in order to make these separations more efficient. In order to achieve this goal of increased efficiency, we have taken two ideas from previous works, namely being width modulated channels as well as semi-packed columns, and tried to use the redeeming qualities that exist in both and combine them to achieve something that shares their positive points. Semi-packed columns have been shown to increase the plate height for a column. However, the pressure drop that occurs in these types is a significant drawback that makes high-speed separations very difficult to achieve with lower inlet pressures. Width modulation of the column channel can be used to change the overall velocity of the gas as it travels throughout the column resulting in a lower overall pressure drop. The method that was used to combine these two ideas was to change the density of the pillars in the semi-packed column along the length of the channel to achieve the same effect of the width modulation. This was achieved by changing the number of pillars across in the channel as well as the pitch distance between sets of pillars. Controls were also fabricated so the results could be compared to the two extreme designs. After extensive testing, the results indicated that between the minimum and maximum density, the maximum density allowed only ~50% of the total flow that the minimum density design showed at the same inlet pressure. In comparison, the design that changed pillar density throughout the channel allowed ~89% of the flow the minimum density design. This is a significant increase in flow and allows for much more to pass through the chip in a given time span increasing the efficiency of high-speed separations. The other factor that was tested for was the number of theoretical plates for each of the designs. It was shown through testing with a constant sample that the minimum density column behaves very similarly to an open column without pillars whereas the density modulated column has plate numbers between those of the minimum density and the maximum density. This is even further data that supports the use of a pillar density modulated chip for high-speed separations.