Fluorescence-Correlation Spectroscopy Study of Molecular Transport within Reversed-Phase Chromatographic Particles Compared to Planar Model Surfaces

Abstract

Reversed-phase liquid chromatography (RPLC) is a widely used technique for molecular separations. Stationary-phase materials for RPLC generally consist of porous silica-gel particles functionalized with n-alkane ligands. Understanding motions of molecules within the interior of these particles is important for developing efficient chromatographic materials and separations. To characterize these dynamics, time-resolved spectroscopic methods (photobleach recovery, fluorescence correlation, single-molecule imaging) have been adapted to measure molecular diffusion rates, typically at n-alkane-modified planar silica surfaces, which serve as models of chromatographic interfaces. A question arising from these studies is how dynamics of molecules on a planar surface relate to motions of molecules within the interior of a porous chromatographic particle. In this paper, imaging-fluorescence-correlation spectroscopy is used to measure diffusion rates of a fluorescent probe molecule 1,1'-dioctadecyl-3,3,3'3'-tetramethylindocarbocyanine perchlorate (DiI) within authentic RPLC porous silica particles and compared with its diffusion at a planar C18-modified surface. The results show that surface diffusion on the planar C18 substrate is much faster than the diffusion rate of the probe molecule through a chromatographic particle. Surface diffusion within porous particles, however, is governed by molecular trajectories along the tortuous contours of the interior surface of the particles. By accounting for the greater surface area that a molecule must explore to diffuse macroscopic distances through the particle, the molecular-scale diffusion rates on the two surfaces can be compared, and they are virtually identical. These results provide support for the relevance of surface-diffusion measurements made on planar model surfaces to the dynamic behavior of molecules on the internal surfaces of porous chromatographic particles.

Spatially Multiplexed Imaging: Fluorescence Correlation Spectroscopy for Efficient Measurement of Molecular Diffusion at Solid–Liquid Interfaces

“…The C 18 surfaces were illuminated by total-internalreflection, which limits excitation of fluorescent molecules to within a small distance (approximately 100 nm) from the glass surface. In previous experiments, 17 the region sampled by the detector was restricted to a small square, 8 Â 8 pixels (2.1 Â 2.1 mm) area, and fluorescence intensity time traces from this region were generated with 1.09 ms (exposure þ read time) resolution by summing the total intensity within the entire region. The average of 10 autocorrelations was determined for each experimental condition by co-adding their power spectra and inverse Fourier-transforming the result.…”

“…In multiplexed imaging-FCS, cross-talk can occur between contiguous FCS-acquisition regions, 11,13 when the fluorescence intensity from a molecule at the border of one acquisition region is also detected in an adjacent region. This could be a problem when using FCS as a method for quantifying molecular populations 7,17 where an individual molecule could simultaneously be counted in adjacent regions leading to over counting. Cross-talk was avoided in this work by setting the distance between adjacent interrogated regions to be six times the point spread function, where ten 8 Â 8 pixel regions were spaced evenly over the 53 mm length of the image.…”

“…The desorption rate of DiI from the C 18 surface may be estimated from the intercept of the line in the plot in Figure 3b, where k desorb ¼ 7.1 AE 2.6 s -1 ; both values are comparable to previously reported results determined by averaging 10 acquisitions at a single spot. 17…”

“…By limiting the acquisition region on the CCD to a small subset of pixels, the readout can be much more rapid than reading the entire CCD chip, allowing kHz-framing rates capable of following fast diffusion, comparable to small-molecule diffusion rates in free solution. 17,18 This approach is versatile since the location of the acquisition region can be varied to interrogate different regions of the sample surface and the area of that region can be varied to unambiguously determine the diffusion coefficient from the dependence of the relaxation time on probed area. 17 In this work, the multiplexing capability of imaging-FCS is used to simultaneously collect autocorrelation data from multiple regions for averaging, thus reducing the measurement time for achieving a useful S/N ratio.…”

“…17,18 This approach is versatile since the location of the acquisition region can be varied to interrogate different regions of the sample surface and the area of that region can be varied to unambiguously determine the diffusion coefficient from the dependence of the relaxation time on probed area. 17 In this work, the multiplexing capability of imaging-FCS is used to simultaneously collect autocorrelation data from multiple regions for averaging, thus reducing the measurement time for achieving a useful S/N ratio. Kannan et al first suggested using the multiplexing capability of imaging-FCS, 11 where a beam splitter would be used to generate an array of excitation spots that are simultaneously sampled by the EM-CCD and followed by autocorrelation of the signal from each region.…”