James P. Grinias
As electronic devices and computers have continued to shrink in size over the past few decades, many analytical methods have as well, including liquid chromatography (LC). Although 4.6-mm and 2.1-mm i.d. columns still dominate the field of LC, smaller diameter columns have found utility in a number of applications, especially for the analysis of small volumes of complex biological samples by LC–multiple-stage mass spectrometry (MSn) (1). When using these columns with internal diameters of 5–500 µm, considerations must be made in terms of the column format and the type of stationary-phase support (2). Capillaries (usually made of fused silica) and microfabricated devices containing microfluidic channels are the two main formats that can be implemented. By far, capillaries are the more common of the two, with many commercial vendors offering a wide selection of capillary-scale (or “nano”) LC columns. Microfluidic LC columns are not as widely available because their fabrication and preparation process can be far more labor intensive (3,4). Despite difficult production, the integration of connections between the pump, column, and detector into a single device can simplify column installation in commercial microfluidic LC systems.
In both formats, the stationary phase is typically bonded either directly to the column wall, to particles, or to a monolithic structure. Open-tubular columns promise extremely high efficiencies, especially at very small diameters, but limited sample capacity can sometimes make detection sensitivity of low-concentration analytes an issue (5). For particles, many that have been used in analytical-scale columns can also be packed into smaller diameter channels. Recent research has correlated packing techniques to both bed structure and chromatographic performance having led to the preparation of very high efficiency columns (1,6).
Submicrometer particles in this format have shown phenomenal performance, especially when interactions between the bonded and mobile phases promote slip-flow conditions (7). Particle-packed microfluidic LC columns usually have lower efficiencies than cylindrical capillaries because of differences in their cross-sectional area, difficulties associated with fabricating circular channels, and lower achievable packing pressures that affect bed porosity (8). Monoliths provide an integrated structure that can better fill these different cross sections and show promise for use in microfluidic LC, especially since the support can be localized to specific areas of the device where separation is desired (9). Microfabrication of the actual column structure has also been demonstrated through the use of pillar-array columns, which can be designed to exact specifications that optimize chromatographic performance (10). This technology only recently became available commercially, and while it is still more expensive compared to other column formats, the use of these microfabricated columns likely will only grow as the price decreases. Potential ways to reduce fabrication costs include using an embossing process from a microfabricated master (11) or three-dimensional (3D)-printing (12), although a reduction in defects and improvements to feature resolution will be needed to achieve the same efficiency observed in silicon-based devices.
The small-volume columns described thus far still require large instrumentation for both flow generation and detection (13). However, new technologies have started to enable the miniaturization of the entire LC system and this trend will likely continue. Smaller pumps that are capable of achieving ultrahigh pressure flow in the nanoliter- to microliter-per-minute range show promise in not only reducing the pump footprint, but also provide opportunities for developing new, fully portable liquid chromatographs (14). Here, either traditional piston or electroosmotic pumping can be utilized to generate the flow needed for capillary and microfluidic LC columns. Decreases in the size of optical detectors have also been reported, mainly through improvements in the design and operation of light-emitting diode (LED) sources, with good sensitivity for on-column detection despite the reduced pathlength compared to standard flow cells (15). Contactless conductivity (16) and electrochemical (17) detectors can also be implemented in smaller systems or even integrated into a microfluidic device that also contains a separation column, although the number of analyte classes that respond to these detectors is smaller than with optical detection. Mass spectrometry, the most important detection method for capillary-scale separations, has a number of challenges related to miniaturization, including the size of the vacuum pumping system and mass analyzer. Advances toward reducing the size of these components has led to both small benchtop and portable MS detectors (18). Improving the mass resolution of these systems will be the biggest challenge to matching the performance of the more standard MS systems used in complex biological analyses. No matter which of the detectors is used, data acquisition and instrument control interfaces are also needed. As both smartphones and single-board miniaturized computers become more common and continue to grow in computing power, replacing large desktop systems with much smaller setups for data collection and processing will also help reduce system size (19).
With column and instrument miniaturization, several aspects of the method must be improved compared to standard techniques to achieve wide use. Although there are several recent reports of exceptional chromatographic efficiency with capillary columns, commercial systems must be designed to ensure that any performance gains are not lost to instrumental band broadening. Smaller detectors must still provide adequate detection limits for the samples being analyzed, which can be difficult because injected volumes are typically well below 1 µL. Perhaps most importantly, new miniaturized liquid chromatographs must be as robust (if not more so) and user-friendly as standard size instruments to encourage their adoption for routine analysis. As further advances in column and instrument technology are made to improve chromatographic efficiency and reduce system size, enhanced separation performance at the nanoliter scale will be both easier and more affordable to attain.
James P. Grinias is an assistant professor in the Department of Chemistry and Biochemistry at Rowan University in Glassboro, New Jersey. Glenn A. Kresge is a is a fourth-year student majoring in biochemistry at Rowan University.