by Dr J. J. Destafano, Dr T. K. Langlois & Dr J. J. Kirkland
This article describes the re-introduction of superficially porous particles to provide an alternative to sub-2 micron packings for achieving high-speed or high-resolution separations without the disadvantages of high pressure operation.
The introduction of superficially porous and pellicular particles of silica in the 1960s [1,2] was instrumental in enabling liquid chromatography to develop into the powerhouse analytical technique that it is today. These spherical particles with shallow pores allowed the speed of liquid-phase separations to be improved to levels that became competitive with gas-phase separations and led to increased popularity and rapid developments. Such packing materials were eventually displaced as the particles of choice when much smaller, totally porous particles became available in the early 1970s. In recent years, to satisfy the need for high productivity, there has been a strong movement toward the use of HPLC columns with even smaller, totally porous particles. It has been known for some time that high-speed separations can be achieved without loss in separating power by reducing the size of particles in columns [3]. However, while a twofold reduction in particle size can lead to a 1.4-times (square root of 2) increase in column efficiency, the pressure drops within columns of equal dimensions increases fourfold. This situation then forces a compromise between particle size, column length and pressure; this has been the subject of many studies [4,5].
Many users of HPLC have turned from workhorse columns containing 5-micron particles to columns containing 3.5- or 3-micron particles to produce faster separations and to increase sample throughput. The need for even higher sample throughput has led to a trend towards columns packed with particles of 2-microns or even smaller [6-8]. Of course, there are always tradeoffs that must be made to achieve this increased capability. Smaller particles mean that higher pressures must be used to deliver the desired flow rates of mobile phase liquids, resulting in increased stress on pumping systems and other instrument components. Columns packed with sub-2-micron particles are also more subject to pluggage because of the small porosity frits that are required to retain the packings within the columns. Extra care must be taken to filter the samples and mobile phase to prevent fouling of the 0.5-micron and smaller porosity frits on these columns. In many cases, new higher-pressure instruments capable of delivering up to 1000 bar (15,000 psi) of liquid pressure must be purchased to allow these columns to be used with optimum results. Columns of 2.7µm superficially porous silica particles have been shown to allow very fast separations of small molecules at the modest pressures available in most present HPLC instruments (400 bar). These highly-purified particles with 1.7µm solid silica cores and 0.5µm-thick shells of 9 nm pores exhibit efficiencies that rival those of totally porous sub-2-µm particles but at one-half to one-third of the column back pressure. The electron microscope photo of the cross-sectional structure of the particles is shown in Figure 1.
Unexpectedly high efficiency
The unusually high efficiency of the new superficially porous particle columns is illustrated in Figure 2. As expected from theory, the plate heights for 50 x 4.6 mm i. d. columns are increasingly smaller (separation power is increased) as particle size decreases from 5- to 1.8-µm. Unexpectedly however, the plate height for the 2.7-µm fused-core column is less than that of a column of smaller 1.8-µm totally porous particles. This increase in separation power for fused-core particles is significant. Figure 3 shows plots of reduced plate height (plate height H divided by particle diameter, dp) versus mobile phase linear velocity for columns of smaller totally porous particles and a column of fused-core particles. The 3- and 3.5-µm particles produce reduced plate height minima of about 2 for small molecules. This value of 2 has been traditionally considered as the practical “limit” for excellent columns of totally porous particles. On the other hand, the column of fused-core particles has a plate height minimum of about 1.5, which is consistently seen for columns of these particles. In fact, longer columns (e.g., 150 mm) of these particles have shown reduced plate heights as small as 1.2 because of reduced extra-column band broadening effects of the larger-volume column (not shown here). Figure 3 also shows the pressure required for the highest mobile phase velocity attempted for each column in this study. As expected, the larger particles show lower pressure drops, and the smaller particles higher pressure drops, with the column of 1.8-µm particles showing the highest pressure requirements at much smaller mobile phase velocities.
The larger reduced plate heights of 1.8-µm particles in Figure 3 suggest that there is difficulty in obtaining homogeneous packed beds of these very small particles. On the other hand, the unusual efficiency of the column of fused-core particles may be explained by the very narrow particle size distribution of this material, as shown in Figure 4. The standard deviation of the fused-core particle size distribution typically is 5 - 6%, while the standard deviation of high-quality 3-µm totally porous particles in Figure 5 is typically about 20%. The very narrow size particle distribution of the fused-core particles, coupled with the higher density of this material (30 - 70% more dense compared to totally porous particles, depending on porosity), apparently allows the formation of very homogeneous packed beds. The resulting excellent packed-bed homogeneity probably leads to the unusual efficiency of the fused-core particle columns.
Fast mass transfer
Compared with totally porous particles, the thin porous shell on fused-core particles results in improved mass transfer (kinetic) effects where solutes diffuse more quickly in and out of the porous structure for interaction with the stationary phase. For smaller molecules (e.g., MW <600) this effect is not so dramatic, as shown in Figure 5. Here, columns of fused-core particles with C18 and C8 stationary phases show improved reduced plate heights over a 3-µm totally porous particle column for virginiamycin (MW = 574) at the reduced plate height minimum. This effect is mainly due to the homogeneous packed bed for the fused-core particles. Note also that the slope of the plot with the fused-core column is somewhat less steep than for the totally porous particle column. This lower slope of the plot suggests that improved mass transfer effects are beginning to take place because of shorter diffusion paths in the porous shell.
For the present fused-core particles, mass transfer improvement begins to be obvious at about 600 molecular weight and is significant for molecules larger than about 1000 molecular weight. For still larger solutes, the improved mass transfer of the fused-core particles becomes impressive at higher mobile phase velocities. At higher reduced velocities, the fused-core particle column exhibits reduced plate heights almost half that of totally porous particles of about the same size. This improved mass transfer characteristic of the fused-core particles is especially attractive for carrying out very fast separations of larger compounds that can access the porous structure.
Improved column stability
Rigorous tests have confirmed the unusual stability of fused-core particle packed beds. Figure 6 shows the overlaid first and 250th chromatograms of a pharmaceutical sample separated on a 100 x 2.1 mm C18 fused-core particle column at 30 oC, 1.6 mL/min and a column pressure of 12,000 psi. Also shown on the figure are the standard deviations found for variation in retention times, plates, peak tailing and resolution for peaks 1 and 2 during the study.
Separations in minutes
The capability of a fused-core column to carry out very fast separations of mixtures is illustrated in Figure 8. This mixture of eight pesticides is separated in less than one minute using conditions that are readily available in most instruments. An interesting, often unrecognised additional advantage of fused-core particle columns is that, because of the solid core and the smaller porous volume of the thin outer shell, unretained solutes (t0) are more quickly eluted than totally porous particles of the same size. This further increases separation speed by reducing the “dead” time of the column, about one-half that for comparable columns of totally porous particles.
Conclusions
Columns of new silica 2.7-µm fused-core particles with solid cores and a 0.5-µm-thick shell of 9 nm pores exhibit unusual chromatographic efficiency, with reduced theoretical plates of 1.5 or lower for small molecules. Presumably, this is due to the ability to form very homogeneous packed beds as a result of an extremely narrow particle size distribution and higher particle density. These columns with homogeneous packed beds are also highly stable at operating pressures of at least 600 bar. Sample loading characteristics of fused-core columns for practical separations closely resemble those of the totally porous particles with higher surface areas. Fused-core particles exhibit highly improved mass transfer (kinetic) effects because of the thin porous shell surrounding a solid core, allowing solutes to rapidly diffuse in and out of the porous structure containing the stationary phase for interaction. Compared to totally porous particles, the fused-core configuration significantly improves column efficiency at high mobile phase velocities for compounds above about 600 daltons MW. For solutes above approximately 1000 Daltons, column efficiency can about double at higher mobile phase velocities, compared to columns of totally porous particles. Columns of the fused-core particles exhibit theoretical plates nearly comparable to those of sub-2-µm particles, but with much reduced pressure requirements. Stated otherwise, the fused-core particles generally develop about twice the theoretical plates/bar pressure when measured at the plate height minimum, compared to sub-2-µm particles.
References
1. C G Horvath et al. Anal Chem 1967; 39: 1422.
2. J J Kirkland. US Patent No 3 505 785.
3. J C Giddings. Dynamics of Chromatography, Dekker,
New York, NY, 1965, Ch. 2
4. G J Kennedy and J H Knox. J Chromatog Sci 1972;10: 549
5. JH Knox. J Chromatog Sci 1972;15:35215 (1977) 352.
6. RE Majors. LC-GC. 2006; 24: 248.
7. J W Thompson et al. LC-GC; 2006; 24: 16
8. J J Kirkland. J Chromatogr Sci.2000; 38: 535.
9. J J Kirkland et al. Am Lab. 2007; 39: 18.
The authors
Drs Destefano, Langlois and Kirkland are at
Advanced Materials Technology
3521 Silverside Road, Suite 1-K, Quillen Building Wilmington DE 19810, USA.
e-mail: joedestefano@advanced-materials-tech.com
Acknowledgement: All figure in this article are reproduced from the Journal of Chromatographic Science by permission of Preston Publications, a Division of Preston Industries, Inc.
