Fritless Chromatographic Microfluidic-Based Columns for Chemical Separations

The development of microfluidic chips and devices offers a number of advantages over conventional analytical techniques, including small sample volume requirements, portability, fast sampling times, ability to multiplex, and compatibility with other techniques. 1 Microfluidics has been the motivation for various biochemical application advancements in point-of-care diagnostics, bioterrorism detection, and drug discovery. Potential applications include biotechnology, pharmaceuticals, life sciences, defense, public health, and agriculture.2

There is currently high demand for miniaturized separations techniques that are easy to use, versatile, and inexpensive to fabricate. Recent studies have documented the development of both liquid chromatography and capillary electrochromatography (CEC)-based platforms-on-a-chip for separating proteins, peptides, DNA, and viral and bacterial cells.3–10 One area of interest is the use of microchips for separations chemistry. Packed-bed chromatography is one area of separations that is amenable to microfluidics- based techniques. Reversed-phase silica particles (e.g., C18), for example, are widely used as the stationary phase in HPLC and solid-phase extraction (SPE) for preconcentration and separation of analytes or to remove unwanted components from samples. Protein purification commonly employs the use of affinity columns for purification and separation.

This paper describes some of the authors’ work in chip-based separations by detailing their fabrication and use in small-molecule analysis.11 Specifically, they demonstrate a microfluidic-based LC technique to separate dyes using C18 particles as the stationary phase. A prototype protein purification microchip is also described.

Experimental

Materials

The reversed-phase chromatographic packing material consisted of porous, C18-modified, 10-μm particles (Western Analytical Products, Inc., Wildomar, CA). A suspension (20 μL) of C18 beads was prepared in methanol by sonication (10 min). Stock solutions of food dyes (FD&C blue #1 and FD&C yellow #5) (McCormick & Co., Inc., Sparks, MD) were prepared in water. All solutions (methanol, water) were degassed and filtered through a 0.45-μm syringe filter. For the protein microchip, a suspension (20 μL) of Ni-sepharose beads was prepared in methanol.

Microchip fabrication

The poly(dimethylsiloxane) (PDMS) chips were prepared by using a mold created by soft photolithography.12 For the small-molecule separations, the pattern consisted of 100-μm-wide channels designed by use of AutoCAD® software (AutoCAD, San Rafael, CA) and printed as a high-resolution (20,000-dpi) photomask (CAD/Art Services, Inc., Bandon, OR). The protein purification microchip consisted of a pattern of 100-μm-wide channels. Negative-type photoresist (SU-8 2025, Microchem, Newton, MA) was spin-coated onto a 3-in. silicon wafer at 3000 rpm for 30 sec to a thickness of 30 μm. The protein microchip was fabricated similarly. The photoresist-coated wafer was baked for 15 min at 95 °C. The pattern on the mask was transferred to the wafer through UV exposure for 2 min. The exposed wafer was baked at 95 °C for 5 min and unexposed areas were removed by rinsing with SU-8 developer (Microchem). The PDMS chip was fabricated by cast molding of a 10:1 mixture of PDMS oligomer and cross-linking agent (Sylgard 184, Dow Corning, Midland, MI). The PDMS mixture was degassed and baked for 30 min in an oven at 80 °C. The PDMS replicas were peeled off the mold.

The chips were prepared by simple cast molding. Holes (300-μm-diam) for the liquid connections were punched through the PDMS chip. The chip was irreversibly sealed onto a clean coverglass of 2 mm thickness (VWR micro cover glass, VWR, West Chester, PA) or onto a quartz slide of 0.5 mm thickness (SPI Supplies, West Chester, PA) using Ar plasma.

Packing of beads

Figure 1 - a, b) Schematic illustration of the packing of a microchannel in PDMS chip through pressing the top of the flexible PDMS chip to trap the chromatographic beads (not to scale). I: sample inlet, O: separation outlet. OR1 and OR2 are outlet reservoirs; the suspension of particles is pumped from O. c) Serpentine microfluidic chip packed with C18-modified silica particles prior to loading of dye mixture.

The packing of C18 particles in the channel is shown in Figure 1a and b. First, the chip was rinsed with degassed methanol for 3 min to clean the wall and to remove any residual water from the channel. The top of the flexible chip was then pushed downward using either a metal rod or magnetic valve13 around the point of the separation channel where the packing begins. A suspension (0.01– 0.1 μL) of methanolic C18 particles was manipulated through a small-bore tubing (0.3-mm-i.d.) using a peristaltic pump, and connected to the O-port (outlet of the separation channel of the chip) and washed with methanol (10 μL/min) for 2 min. A pressure of approximately 2 bar was applied intermittently (4–5 sec). The tapering was removed, and methanol and water were pumped through the channel from the reverse direction (I-port) to obtain a smooth front edge of the packing. The packed channel was then rinsed with water and heated at 115 °C for 2 hr. Figure 1c is a completed serpentine chip packed with beads.

Results and discussion

Packing of C18-modified particles

PDMS is a rubber-like material by nature of its low Young’s modulus (E = 0.5–4 MPa depending on the curing conditions) and high Poisson ratio (v = 0.5, that, is essentially incompressible).14 Hence, it is relatively straightforward to pack C18 particles into the microfluidic channels due to its hydrophobicity and high elasticity. In the current work, an 80% taper traps the particles, allowing fluid to flow under the tapered region yet trapping the beads (10-μm-diam). Variable lengths of packing can be achieved by loading the desired amount of particles into the channel.

Figure 2 - a) Schematic representation of the (i) keystone, (ii) clamping, and (iii) anchor effects. b) Wall effect.

There are four reasons why packing of particles into the microfluidic channels was so facile, and they are denoted as the keystone, clamping, anchor, and wall effects. The density of particles increases at the taper, causing aggregation without the need for a frit or other physical barrier. The initial particles manipulated into the microchannel act as keystones blocking subsequent particles, thereby forming a region of packed particles (Figure 2a[i]). The keystone effect is temporary in the authors’ system and only occurs during the initial packing of beads.

To compress the packing, small pressures (2 bar) were applied, causing a deformation (enlarging) in the channel. Releasing the pressure shrinks the channel, causing a constant strain around the packed particles (Figure 2a[ii]). The particles press against each other due to the elastic strains acting perpendicularly from the wall and pointed inward toward the middle of the channel. Finally, these forces clamp the packing into the microfluidic channel. This clamping effect is presumably the main source of the stability of the packed particles in the microchannel.

The strong particle–wall interaction between the C18 silica and the hydrophobic surface of the PDMS chip is another source of stability of the packed particles. Particles adjacent to the wall of the channel penetrate the wall, acting as anchors for the packing (Figure 2a[iiI]). The extent of penetration is dependent on the intensity of the initial pumping pressures and composition and elasticity of the PDMS.

Previous work in liquid chromatography has shown that the wall effect is a result of the looser packing of the stationary phase adjacent to the walls of rigid columns.15,16 The mobile phase tends to flow faster near the wall due to the increased permeability. Solute molecules nearest the wall are more mobile than the average molecules that make up the solute band, resulting in band spreading. This effect is attenuated in the authors’ chips since the beads deform the PDMS wall (Figure 2b).

The results demonstrate that the packed beads remain static and do not degrade under electric fields or high-pressure pumping of organic media. Upon completion of the loading process, the chips were heated (115 °C) for 2 hr to maximize the packing of the beads.

Separation of dyes

Injection of sample plug and flow visualization

In order to reduce sample bias, injection was via hydrodynamic pressure using a single-channel peristaltic pump. A small volume (0.1–2 μL) of solution was manipulated into the pump tubing initially filled with water. The sample was then injected at the sample inlet (I) port and manipulated into the other three channels with different flow rates, depending on the hydraulic resistance of each channel. Due to the high hydraulic resistance of the packing, a reduced flow rate was observed in the separation channel, yielding an injected sample of less than 1 nL.

Figure 3 - Illustrations of the serpentine separation channel of the microchip packed with C18-modified silica particles. A) A mixture of blue and yellow food dyes was injected by pressure applied from a side inlet port (I). B–F) While the yellow dye flows through the packing, the blue dye migrates slowly through it, not yet reaching the middle of the packing (E) (carrier: water, yellow dye: FD&C yellow #5, blue dye: FD&C blue #1).

HPLC and SPE commonly employ reversed-phase silica particles (e.g., C18) as the stationary phase for preconcentration and separation of analytes or to remove unwanted materials from samples. In recent work, the authors used mixtures of two food dyes (FD&C blue #1 and FD&C yellow #5) to demonstrate the use of the chip-based reversed-phase chromatographic separation. Figure 1c shows a microfluidic serpentine channel packed with C18 particles and used to separate the mixture of food dyes. The dyes were injected by pressure from one of the sample inlet ports into the main microfluidic channel via a cross-T junction and manipulated to the chromatographic packing (Figure 3A). The blue dye was retained on the chromatographic packing due to hydrophobic effects as the yellow dye migrated through the packing (Figure 3B–F); hence, separation was achieved within the first few millimeters of packing. Upon elution of the yellow dye, methanol was pumped through the channel to elute the retained blue dye. The overall capacity was calculated to 7.5 × 10–11 mol for the blue dye.

Development of a prototype protein purification microchip

Figure 4 - Prototype microfluidic chip used for protein purifications. Before (a) and after (b) packing with Ni-sepharose beads.

A natural extension to the fritless chromatographic microchips is to employ them in other molecular separations and not limited to small molecules. For example, DNA, viral, bacteria, and protein separations should also be conducive to current technology. Protein synthesis, separation, purification, and isolation can be a laborious task. For those designing mutants of proteins, new analytical techniques that can yield material of high purity, and in sufficient quantities for a subsequent biochemical assay, are warranted. Figure 4 shows a prototype microfluidic chip packed with Ni-sepharose resin (30–100 μm diam) used in the purification of the enzyme adenosine diphosphate glucose pyrophosphorylase (ADPGlc Ppase). Due to the variance in diameter of the beads, a microfluidic channel of decreasing diameter was fabricated in order to pack the beads. Current work is focused on optimizing experimental conditions for purification.

Conclusion

A simple protocol for designing fritless chromatographic microchips for small-molecule and protein purifications was described. The chips were readily fabricated from PDMS and the packing of particles was easily accomplished. A complete separation of two dyes could be easily achieved in the first 25% length of the packing. A prototype protein purification chip was shown. Further investigation and characterization of the packing and use of other types of particles are under investigation.

References

  1. Tabeling, P. Introduction to Microfluidics; Oxford University Press: Oxford, U.K.; 2005.
  2. Ehlert, S.; Tallarek, U. High-pressure liquid chromatography in lab-on-achip devices. Anal. Bioanal. Chem.  2007, 388, 517–20.
  3. Ying, H.; Killeen, K.; Brennan, R.; Sobek, D.; Werlich, M.; Van de Goor, T. Microfluidic chip for peptide analysis with an integrated HPLC column sample enrichment column, and nanoelectrospray tip. Anal. Chem.  2005, 77, 527–37.
  4. Pamme, N. Continuous flow separations in microfluidic devices. Lab Chip  2007, 7, 1644–59.
  5. Ghitun, M.; Bonneil, E.; Fortier, M.-H.; Yi, H.; Killeen, K.; Thibault, P. Integrated microfluidic devices with enhanced separation performance: application to phosphoproteome analyses of differentiated cell model systems. J. Sep. Sci.  2006, 29, 1539–49.
  6. Foote, R.S.; Khandurina, J.; Jacobson, S.C.; Ramsey, J.M. Preconcentration of proteins on microfluidic devices using porous silica membranes. Anal. Chem.  2005, 77, 57–63.
  7. Koesdjojo, M.T.; Tennico, Y.H.; Remcho, V.T. Fabrication of a microfluidic system for capillary electrophoresis using a two-stage embossing technique and solvent wetting on poly(methyl methacrylate) with water as a sacrificial layer. Anal. Chem.  2008, 80, 2311–18.
  8. Li, S.F.Y.; Kricka, L.J. Clinical analysis by microchip capillary electrophoresis. Clin. Chem.  2006, 52, 37–45.
  9. Liu, D.; Shi, M.; Huang, H.; Long, Z.; Zhou, X.; Qin, J.; Lin, B. Isotachophoresis preconcentration integrated microfluidic chip for highly sensitive genotyping of the hepatitis B virus. J. Chromatogr. B  2006, 844, 32–8.
  10. Bandilla, D.; Skinner, C.D. Capillary electrochromatography of peptides and proteins. J. Chromatogr. B  2004, 1044, 113–29.
  11. Gaspar, A.; Piyasena, M.E.; Gomez, F.A. Simple fabrication of fritless chromatographic microchips packed with conventional reversed-phase silica particles. Anal. Chem.  2007, 79, 7906–9.
  12. Duffy, D.C.; McDonald, J.C.; Schueller, O.J.A.; Whitesides, G.M. Rapid prototyping of microfluidic systems in poly(dimethylsiloxane). Anal. Chem. 1998, 70, 4974–84.
  13. Gaspar, A.; Piyasena, M.; Daroczi, L.; Gomez, F.A. Magnetically controlled valve for flow manipulation in polymer microfluidic devices. Micro. Nano.  2008, 6, 525-31.
  14. Lotters, J.C.; Olthuis, W.; Veltink, P.H.; Bergveld, P. The mechanical properties of the rubber elastic polymer polydimethylsiloxane for sensor applications. J. Micromech. Microeng. 1997, 7, 145–7.
  15. He, B.; Tait, N.; Regnier, F. Fabrication of nanocolumns for liquid chromatography. Anal. Chem.  1998, 70, 3790–7.
  16. Knox, J.H.; Parcher, J.F. Effect of column to particle diameter ratio on the dispersion of unsorbed solutes in chromatography. Anal. Chem.  1969, 41, 1599–1606.

Dr. Gaspar is with the Department of Inorganic and Analytical Chemistry, University of Debrecen, Debrecen, Hungary. Ms. Baghdachi, Mr. Goldberg, Ms. Stevens, Mr. Torres, Ms. Salgado, and Prof. Gomez are with the Department of Chemistry and Biochemistry, California State University—Los Angeles, 5151 State University Dr., Los Angeles, CA 90032-8202, U.S.A.; tel.: 323-343-2368; fax: 323-343-6490; e-mail: [email protected]. The authors gratefully acknowledge financial support for this research by grants from the National Science Foundation (CHE-0515363, DMR-0351848, and CBET-0723271) and the European Community for the Marie Curie Fellowship (MOIFCT-2006-021447) of A. Gaspar.