In situ measurement of the transversal dispersion in ordered and disordered two-dimensional pillar beds for liquid chromatography.

Using a fully transparent micropillar array chip and an optical "injection" setup capable of writing pulsed and continuous patterns into the flow by uncaging a fluorescent dye, highly detailed measurements of the transversal dispersion process in two-dimensional (2D) chromatographic beds could be made. With the use of water-glycerol mobile phase spanning a wide range of viscosities, the obtained data cover a broad range of reduced velocities (0 < ν < 3000) and show a clear leveling-off of the transversal dispersion coefficient at large reduced velocities. With dependence on the packing density, this leveling-off occurs at a value of about Dtrans/Dmol = 10 (ε = 0.4), Dtrans/Dmol = 4 (ε = 0.6) and Dtrans/Dmol = 2.5 (ε = 0.8). Another interesting observation that could be made is that the effect of the bed order on the observed transversal dispersion process is relatively small. The observed leveling-off in the relation between the measured Dtrans values and the reduced liquid velocity furthermore clearly invalidates the classical Galton-board model, predicting a linear increase of Dtrans with the reduced velocity. On the other hand, it corroborates a recently proposed series-connection transport model for Dtrans in 2D porous media.

Exploring the speed limits of liquid chromatography using shear-driven flows through 45 and 85 nm deep nano-channels.

We explored the possibility to perform high speed and high efficiency liquid chromatographic separations in channels with a sub-100 nm depth. The mobile phase flow through these nano-channels was generated using the shear-driven flow principle to generate high speed flows which were the equivalent of a 12,000 bar pressure-driven flow. It was found that the ultra-fast mass transfer kinetics prevailing in this range of small channel depths allow to drastically reduce the C-term contribution to band broadening, at least up to the upper speed limit of our current set-up (7 mm s(-1) mobile phase velocity), leaving the inescapable molecular diffusion (i.e., B-term band broadening) as the sole detectable source of band broadening. Due to the greatly reduced mass transfer limitations, 50,000 to 100,000 theoretical plates could be generated in the span of 1 to 1.5 seconds. This is nearly two orders of magnitude faster than the best performing commercial pressure-driven UHPLC-systems. With the employed channel depths, we appear to have struck a practical lower limit for the channel miniaturization of shear-driven flows. Despite the use of channel substrates with the highest grades of optical flatness, the overall substrate waviness (on the order of some 5 to 10 nm) can no longer be neglected compared to the etched channel depth, which in turn significantly influenced the local retention factor and band broadening.

Separations using a porous-shell pillar array column on a capillary LC instrument.

We investigated the achievable separation performance of a 9-cm-long and 1-mm-wide pillar array channel (volume = 0.6 µL) containing 5 µm diameter Si pillars (spacing 2.5 µm) cladded with a mesoporous silica layer with a thickness of 300 nm, when this channel is directly interfaced to a capillary LC instrument. The chip has a small footprint of only 4 cm × 4 mm and the channel consists of three lanes that are each 3 cm long and that are interconnected using low dispersion turns consisting of a narrow U-turn (10 µm), proceded and preceded by a diverging flow distributor. Measuring the band broadening within a single lane and comparing it to the total channel band broadening, the additional band broadening of the turns can be estimated to be of the order of 0.5 µm around the minimum of the van Deemter curve, and around some 1 µm (nonretained species) and 2 µm (retained species) in the C-term dominated regime. The overall performance (chip + instrument) was evaluated by conducting gradient elution separations of digests of cytochrome c and bovine serum albumin. Peak capacities up to 150 could be demonstrated, nearly completely independent of the flow rate.

Capillary liquid chromatography separations using non-porous pillar array columns.

We report on a series of explorative experiments wherein a non-porous pillar array column (NP-PAC) is coupled to a commercial capillary LC instrument. The performance of the system was evaluated by both non-retained and retained experiments using several types of samples. In order to minimize interfacing related dispersion, relatively large pillars (d(p)=11 µm) were defined so that a considerable depth could be achieved (50 µm), resulting in an equivalent cylindrical internal diameter of 252 µm. Connecting 20 channel tracks of 1mm wide and 7 cm long by previously developed distributor-based turns, a large channel length (1.4m) and volume (28µl) could be achieved without compromising the separation performance excessively. Establishing a van Deemter curve under non-retained conditions with off-chip injection and detection, a minimal plate height of 13 µm was established, resulting in some 100,000 plates obtained in 30 min. To demonstrate the practical applicability of the NP-PAC, high pressure operation was applied to perform a number of example separations (parabens, phenones, sulfonamides, steroids and BSA digest) during a continuous operation period of 3 months wherein some 500 injections were performed. In the gradient mode, the NP-PAC approach allowed to achieve good to reasonable peak capacities (n(p)=100-140 in 50-70 min) and symmetries for both large and small solutes and for both gradient and isocratic separation modes, with figures of merit for the quantification of the peaks in the ppm range, opening more perspectives for microfluidics-based small molecule analysis.

Realization of 1 × 10(6) theoretical plates in liquid chromatography using very long pillar array columns.

We report on the possibility to achieve ultra high efficiencies (order of 1 million theoretical plates) in liquid chromatography in a relatively short time of 20 min (elution time of unretained marker). This was achieved using a micropillar array column with optimized pillar diameter (5 µm) and interpillar distance (2.5 µm) to operate close to the Knox and Saleem limit of micropillar array columns in the region of the 1 million theoretical plate mark under the prevailing pressure restriction (350 bar in the present study). The obtained efficiency was slightly affected (some 15 to 20% around the optimal flow rate) by the turns that were inevitably needed to arrange a 3 m long column on a 4 in. silicon wafer.

Fabrication and chromatographic performance of porous-shell pillar-array columns.

We report on a new approach to obtain highly homogeneous silica-monolithic columns, applying a sol-gel fabrication process inside a rectangular pillar-array column (1 mm in width, 29 microm in height and 33.75 mm in length) having a cross-sectional area comparable to that of a 200 microm diameter circular capillary. Starting from a silicon-based pillar array and working under high phase-separation-tendency conditions (low poly(ethylene glycol) (PEG)-concentration), highly regular silica-based chromatographic systems with an external porosity in the order of 66-68% were obtained. The pillars, 2.4 microm in diameter, were typically clad with a 0.5 microm shell layer of silica, thus creating a 3.4 microm total outer pillar diameter and leaving a minimal through-pore size of 2.2 microm. After mesopore creation by hydrothermal treatment and column derivatization with octyldimethylchlorosilane, the monolithic column was used for chip-based liquid-chromatographic separations of coumarin dyes. Minimal plate heights ranging between 3.9 microm (nonretaining conditions) and 6 mum (for a retention factor of 6.5) were obtained, corresponding to domain-size-reduced plate heights ranging between 0.7 and 1.2. The column permeability was in the order of 1.3 x 10(13) m(2), lower than theoretically expected, but this is probably due to obstructions induced by the sol-gel process in the supply channels.

Effect of the presence of an ordered micro-pillar array on the formation of silica monoliths.

We report on the synthesis of siloxane-based monoliths in the presence of a two-dimensional, perfectly ordered array of micro-pillars. Both methyltrimethoxysilane- and tetramethoxysilane-based monoliths were considered. The obtained structures were analyzed using scanning-electron microscopy and can be explained from the general theory of surface-directed phase separation in confined spaces. The formed structures are to a large extent nearly exclusively determined by the ratio between the bulk domain size of the monolith on the one hand and the distance between the micro-pillars on the other hand. When this ratio is small, the presence of the pillars has nearly no effect on the morphology of the produced monoliths. However, when the ratio approaches unity and ascends above it, some new types of monolith morphologies are induced, two of which appear to have interesting properties for use as novel chromatographic supports. One of these structures (obtained when the domain size/inter-pillar distance ratio is around unity) is a 3D network of linear interconnections between the pillars, organized such that all skeleton branches are oriented perpendicular to the micro-pillar surface. A second interesting structure is obtained at even higher values of the domain size/inter-pillar distance ratio. In this case, each individual micro-pillar is uniformly coated with a mesoporous shell.

Estimation of surface desorption times in hydrophobically coated nanochannels and their effect on shear-driven and pressure-driven chromatography.

The present paper provides a detailed analysis of the analyte-wall adsorption effects in nanochannels, including a random walk study of the analyte-wall collision frequency, and uses these insights to estimate wall desorption times from chromatographic experiments in nanochannels. Using coumarin dye analytes and using a methanol/water mixture buffered at pH 3 in 120-nm deep channels, the surface desorption times on naked fused-silica glass were found to be maximally of the order of 60 to 150 mus, while they were found to be on the order of 100 to 500 mus on a hydrophobically coated wall. These nonzero adsorption and desorption times lead to an additional band broadening when conducting chromatographic separations. Shear-driven flows, requiring a noncoated moving wall and a stationary coated wall, intrinsically turn out to be more prone to this effect than pressure-driven or electro-driven flows for example. The present study also shows that, interestingly, the number of analyte-wall collisions increases with the inverse of the channel depth and not with its second power, as would be expected from the Einstein-Smoluchowski relationship for molecular diffusion.