Introduction to Modern Liquid Chromatography, Third Edition part 25 ppt

196 DETECTION Figure 4.37 shows an example of the popular post-column derivatization of amino acids with ninhydrin followed by UV detection at 405 nm. In this case 45 amino acids and related substances are separated on an ion exchange column followed by reaction with ninhydrin at 130 ◦ C in a 500-μL reactor (residence time ≈45 sec). Sample derivatization and reaction detectors have gradually lost popularity as other HPLC detectors have been introduced and been made more effective. REFERENCES 1. G. Hesse and H. Weil, eds., Michael Tswett’s First Paper on Chromatography, M. Woelm, Eschwege, 1954. 2. V. G. Berezkin, ed., Chromatographic Adsorption Analysis: Selected Works of M. S. Tswett, Ellis Horwood, New York, 1990. 3. A. Tiselius and D. Claesson, Arkiv. Kemi Mineral. Geol., 15B (18) (1942) 1. 4. A. J. P. Martin and S. S. Randall, Biochem. J., 49 (1951) 293. 5. C. G. Horv ´ ath and S. R. Lipsky, Nature, 211 (1966) 748. 6. J. J. Kirkland, Anal. Chem., 40 (1968) 391. 7. E. Haahti and T. Nikkari, Acta Chem. Scand., 17 (1963) 2565. 8. R. P. W. Scott and J. G. Lawrence, J. Chromatogr. Sci., 11 (1970) 65. 9. B. G. Julin, H. W. Vandenborn, and J. J. Kirkland, J. Chromatogr., 112 (1975) 443. 10. J. W. Dolan and J. N. Seiber, Anal. Chem., 49 (1977) 326. 11. L. R. Snyder and J. J. Kirkland, Introduction to Modern Liquid Chromatography,1st ed., Wiley, New York, 1979, p. 136. 12. J. W. Dolan, LCGC, 14 (1996) 378. 13. L. R. Snyder and J. W. Dolan, High-Performance Gradient Elution, Wiley, New York, 2007. 14. J. J. Kirkland, W. W. Yau, H. J. Stoklosa, and C. H. Dilks, Jr., J. Chromatogr. Sci.,15 (1977) 303. 15. J. C. Sternberg, in Advances in Chromatography, Vol. 2, J. C. Giddings and R. A. Keller, eds., Dekker, New York, 1966, p. 205. 16. D. Parriott, A Practical Guide to HPLC Detection, Academic Press, San Diego, 1993. 17. R. P. W. Scott, in Handbook of HPLC (Chromatographic Science Series, Vol. 78), E. Katz, R. Eksteen, P. Schoenmakers, and N. Miller, eds., Dekker, New York, 1998, ch. 15. 18. D. T. Bach, US Patent 4,867,559 (Sept. 19, 1989). 19. M. H. Garrett, US Patent 6,542,231 (Apr. 1, 2003). 20. G. Cameron, P. E. Jackson, and M. V. Gorenstein, Chem. Aust., 60 (1993) 288. 21. A. G. Frenich, J. R. Torres-Laasio, K. De Braekeleer, D. L. Massart, J. L. Martinez-Vidal, and M. M. Galera, J. Chromatogr. A, 855 (1999) 487. 22. Waters 996 Photodiode Detector: Peak Purity III: Interpretation of Peak Purity Plots Appl. Note 970983, Waters, 1996. 23. F. K. Glowka and M. Karazniewicz-Lada, J. Chromatogr. B, 852 (2007) 669. 24. P. Vinas, N. Balsalobre, C. Lopez-Erroz, and M. Hernandez-Cordoba, J. Agr. Food Chem., 52 (2004) 1789. 25. S. R. Bakalyar, M. P. T. Bradley, and R. Honganen, J. Chromatogr., 158 (1978) 277. 26. Y T. Wu, T R. Tsai, L C. Lin, and T H. Tsai, J. Chromatogr. B, 853 (2007) 281. REFERENCES 197 27. I. N. Ackworth, Coulometric Electrode Array Detectors for HPLC (Progress in HPLC-HPCE, Vol. 6), Brill Academic, Boston, 1997. 28. R. E. Parks and R. L. Marietta, US Patent 4018562 (1975). 29. Model 8060 Nitrogen Specific HPLC Detector, brochure A-8060-99-25M, Antek Instru- ments, Houston, TX, 1999. 30. K. Petritis, C. Elfakir, and M. Dreux, LCGC Europe, 14 (2001) 389. 31. Jasco Incorporated, LCGC Applications Notebook (June 2003) 26. 32. L. Kott, W. B. Holzheuer, M. M. Wong, and G. K. Webster, J. Pharm. Biomed. Anal., 43 (2007) 57. 33. F. Glowka, M. K. Lada, G. Grund, and J. Wachowiak, J. Chromatogr. B, 850 (2007) 569. 34. L. B. Allen, J. A. Koropchak, and B. Szostek, Anal. Chem., 67 (1995) 659. 35. Quant Technologies, QT-500 datasheet (2/ 2007). 36. D. Asa, Amer. Lab., 38(7) (2006) 16. 37. T. Gorecki, F. Lynen, R. Szucs, and P. Sandra, Anal. Chem., 78 (2006) 3186. 38. W. M. A. Niessen, Liquid Chromatography-Mass Spectrometry (Chromatographic Science Series, Vol. 97), 3rd. ed., Taylor and Francis, London, 2006. 39. M. C. McMaster, LC-MS: A Practical User’s Guide, Wiley, Hoboken, NJ, 2005. 40. Current Practice in Liquid Chromatography-Mass Spectrometry,W.M.A.Niessenand R. D. Voyksner, eds., Elsevier, Amsterdam, 1998. 41. I. D. Wilson and U. A. Th. Brinkman, J. Chromatogr. A., 1000 (2003) 325. 42. S. W. Huffman, K. Lukasiewicz, S. Geldart, S. Elliott, J. F. Sperry, and C. W. Brown, Anal. Chem., 75 (2003) 4606. 43. G. W. Somsen, G. Gooijer, and U. A. Th. Brinkman, J. Chromatogr. A., 856 (1999) 213. 44. S. Down, Spectro. Europe, 16 (2004) 8. 45. K. Albert, J. Chromatogr. A , 856 (1999) 199. 46. A. J. Simpson, L H. Tseng, M. J. Simpson, M. Spraul, U. Braumann, W. L. Kingery, B. P. Kelleher, and M. H. B. Hayes, Analyst, 129 (2004) 1216. 47. K. Blau and J. M. Halket, eds., Handbook of Derivatives for Chromatography, 2nd ed., Wiley, New York, 1993. 48. Application Manual: Amino Acids, Pickering Laboratories, Mountain View, CA, 2002. 49. Z. Xie, W. Shi, L. Liu, and Q. Deng, J. Chromatogr. B, 857 (2007) 53. CHAPTER FIVE THE COLUMN 5.1 INTRODUCTION, 200 5.2 COLUMN SUPPORTS, 200 5.2.1 Particle Characterization, 201 5.2.2 Silica Supports, 203 5.2.3 Porous Polymers, 212 5.2.4 Monoliths, 212 5.2.5 Other Inorganic Particles, 214 5.3 STATIONARY PHASES, 217 5.3.1 ‘‘Bonded’’ Stationary Phases, 218 5.3.2 Other Organic-Based Stationary Phases, 223 5.3.3 Column Functionality (Ligand Type), 225 5.4 COLUMN SELECTIVITY, 227 5.4.1 Basis of RPC Column Selectivity, 227 5.4.2 Column Reproducibility and ‘‘Equivalent’’ Columns, 235 5.4.3 Orthogonal Separation, 236 5.4.4 Other Applications of Column Selectivity, 237 5.5 COLUMN HARDWARE, 238 5.5.1 Column Fittings, 238 5.5.2 Column Configurations, 239 5.6 COLUMN-PACKING METHODS, 240 5.6.1 Dry-Packing, 240 5.6.2 Slurry-Packing of Rigid Particles, 240 5.6.3 Soft Gels, 244 5.7 COLUMN SPECIFICATIONS, 244 5.7.1 Manufacturing Standards, 244 5.7.2 Column Plate Number, 245 5.8 COLUMN HANDLING, 246 Introduction to Modern Liquid Chromatography, Third Edition, by Lloyd R. Snyder, Joseph J. Kirkland, and John W. Dolan Copyright © 2010 John Wiley & Sons, Inc. 199 200 THE COLUMN 5.1 INTRODUCTION The column—the ‘‘heart’’ of the HPLC system—has changed greatly from the beginning of HPLC in the mid-1960s. Columns have been improved for greater separation speed and efficiency, as well as increased stability and reproducibility. New stationary phases have been introduced for the extension of HPLC to a wider range of sample types, or for better separations of compounds that have in the past proved problematic. Today it is rare that a column cannot be found for solving a particular HPLC separation problem. Early columns were made of glass, but the need for higher pressure operation quickly led to the exclusive use of metal columns. With the passage of time, columns became shorter and particles smaller (Fig. 1.5g, h). Column lengths of 30 to 250 mm are commonly used today, with particles that are 1.5 to 5 μm in diameter (larger particles are used for preparative separations; see Chapter 15). In this chapter we will examine the HPLC column, including the bare particle or support, the added (‘‘bonded’’) stationary phase, and the column hardware. Porous-silica particles in packed beds (as in Fig. 2.2) are most commonly used. Particles can also be made from solids other than silica; the main value of non- silica particles is their greater stability for use with extremes of temperature or mobile-phase pH. Of special current interest are columns packed with very small particles (<3 μm) for carrying out fast separations (Section 2.5.3.2). Although very few workers will produce their own bonded-phase particles (‘‘column packings’’), we will describe briefly how these materials are made; this information can prove helpful in choosing a column or for troubleshooting. Column characteristics that affect reversed-phase retention and selectivity (Section 5.4) can also influence our selection of a column for a particular application. Column-packing methods (Section 5.6) are described as further background for how ‘‘good’’ columns are prepared. Column testing is covered in Section 5.7. Finally, column specifications and column handling techniques are covered in Sections 5.7 and 5.8, as an aid for good laboratory practice. 5.2 COLUMN SUPPORTS Column packings consist of a rigid support plus an attached stationary phase (as in Fig. 2.2b, which shows a silica particle with attached C 18 groups). In some cases the support and stationary phase are the same; for example, an unbonded silica particle is often used for normal-phase chromatography (Chapter 8). Silica monoliths (Section 5.2.4) represent an alternative column support of more recent vintage. Monoliths refer to columns composed of an interconnected, porous bed, as opposed to columns packed with distinct particles (a monolith may be thought of as one big particle that fills the entire column). In this section we will describe column supports, their characterization, and how their properties affect their final use. Later, we will discuss (1) how these supports can be modified to create column packings for different purposes (Section 5.3) and (2) how the final particles are packed into the column (Section 5.6). 5.2 COLUMN SUPPORTS 201 5.2.1 Particle Characterization Particles can be characterized by their configuration or type, physical dimensions (particle diameter), the nature and size of pores within the particle, and surface area. The importance of particle diameter in influencing separation has been discussed (Fig. 1.5 and Section 2.4). Pore size and surface area are usually related reciprocally; as pore diameter increases, the surface area decreases in roughly the same proportion. The phase ratio Ψ (Section 2.3.1) is roughly proportional to surface area, and retention is proportional to Ψ (Eq. 2.3); therefore retention increases as pore size decreases and column surface area increases. The maximum weight of sample that can be injected is also proportional to surface area (Section 2.6.2), so a greater surface area is usually desirable—which suggests the use of the smallest possible pore diameter (and smallest pores). However, solute molecules must be able to enter the pores without hindrance, and this requires pores that are larger than the solute molecule. For compounds with molecular weights <500 Da, the average pore diameter should preferably be about 9 nm or larger. Larger molecules require larger pores; for example, proteins are usually separated with 30-nm-pore particles (Section 13.3.1.1). The interstitial volume of the column is the space between particles; it is usually about 40% of the total column volume. The physical characteristics of particles are important, and manufacturers monitor these particle properties in various ways. The size of the particles is especially important, as this largely determines the efficiency of the packed column. While optical microscopy and air or liquid elutriation (or classification) methods can provide this information, instruments for the measurement of particle size (e.g., the Coulter counter) are more convenient and quantitative; they can also provide a particle-size distribution. A narrow particle-size distribution is preferred, as this favors larger values of N (Section 5.2.2.1), and provides lower column pressure for comparable efficiency. Particle surface-area, average pore-diameter, and pore-diameter distribu- tion typically are measured by the adsorption of nitrogen or argon, using the Brunauer–Emmett–Teller (BET) procedure. Particles with average pore-diameters > 30 nm (which are used less commonly) are preferably characterized by mercury intrusion, which also can be used to measure pores as small as 3 nm. Mercury intrusion does not work well for fragile particles (with a large pore volume) or for soft polymer particles. Reference [1] should be consulted for additional information on the measurement of the physical characteristics of HPLC particles. 5.2.1.1 Particle Type As illustrated in Figure 5.1, several particle configurations are currently available for HPLC. Totally porous silica particles (Fig. 5.1a) are the most common because of their greater column capacity (allowing the injection of a larger sample weight) and availability in a wider variety of options (stationary phase, particle and pore size, column dimensions, etc.). The most popular particles have diameters in the 1.5- to 5-μm range. Today these particles are often prepared by the aggregation of much smaller spheres (Section 5.2.2.3). Pellicular particles (Fig. 5.1b) consist of solid, spheres that are covered with a very thin surface layer of stationary phase. These silica- or polymer-based particles (presently 1.5 to 2.5 μm in diameter, but much larger in the past) display larger 202 THE COLUMN Porous-shell particle Porous shell solid core throughpores Perfusion particle diffusive pore Micro-pellicular particle solid core (b) (d) (c) Porous particle (a) Figure 5.1 Different particle configurations. values of N for macromolecules, because of better stationary-phase mass-transfer (smaller C-term, Section 2.4.1.1; [2, 3]). Because of their very low surface area, these particles exhibit greatly reduced retention and are limited to much smaller weights of injected sample (with a proportional decrease in detection sensitivity). Pellicular particles are best suited for the analysis of major components (rather than trace impurities), and have been used mainly for the separation of large biomolecules (Chapter 13). Superficially porous particles (also called fused-core™ particles, shell particles, or controlled-surface-porosity particles) have a solid core with a porous outer shell (Fig. 5.1c). These particles typically have diameters of 2 to 5 μm, with porous shells of 0.25 to 0.5 μm in thickness. Superficially porous particles have much greater surface areas than pellicular particles (∼3/4 as large as fully porous particles), providing longer retention and allowing a larger weight of injected sample. The thin, outer porous shell permits faster separations—especially for macromolecules 5.2 COLUMN SUPPORTS 203 [4, 5]. These particles also provide somewhat higher values of N than totally porous particles (Section 5.2.2.1). Perfusion particles (Fig. 5.1d) contain very large (e.g., 400- to 800-nm) pores (through-pores), connected to a network of smaller (e.g., 30- to 100-nm) pores. By comparison, most other particles have pore diameters of 8 to 30 nm (diffusive pores). At high flow rates, solutes are carried into and out of perfusion particles by a combination of diffusion and flow of mobile phase through the particle [6, 7]. This additional contribution to stationary-phase mass transfer is claimed to reduce band broadening, especially for large molecules at higher flow rates. Perfusion particles are usually larger (e.g., 10 μm), permitting their use at lower pressures. These particles are best suited for preparative-scale separations of macromolecules such as proteins; they are little used for the analysis of small molecules. 5.2.1.2 Particle Size and Pore Diameter From 1980 to 2000, particles with diameters of ∼5 μm were generally preferred for routine separation. Particles of this size represent a good compromise in terms of efficiency, pressure drop, convenience, equipment requirements, and column lifetime. However, columns with particle diameters of 3 μm and smaller are becoming more popular, mainly as a means for decreasing run time and increasing sample throughput (Sections 2.5.3.2, 9.3.9.2). Because of the very narrow peaks generated by small-particle columns, any extra-column peak broadening must be minimized (Sections 2.4.1.1, 3.9), especially when smaller diameter columns are used. Columns with particle diameters <2 μm are used mainly with equipment that is capable of higher pressure operation (e.g., up to 15,000 psi; Section 3.5.4.3). Table 5.1 summarizes some physical characteristics of HPLC particles and their effect on separation from favorable (‘‘4’’) to unfavorable (‘‘1’’). Totally porous and superficially porous particles with pore diameters of 8 to 12 nm are used most often for the separation of molecules smaller than 10,000 Da. These column-packings have surface areas of about 125 to 400 m 2 /g, which allow injected sample weights of ≈50 μg (for column inner-diameters of 4–5 mm). Because peaks as small as a few nanograms can be measured, both major and trace components can be assayed in a single separation (or ‘‘run’’) with these columns. Larger molecules (usually biomolecules such as proteins) require larger pores, to avoid restricted diffusion within the pore, with a resulting decrease in column efficiency (Section 13.3.1.1). Column packings with ≈30-nm pores typically are used for separating molecules larger than 10,000 Da. If the pore diameter is at least four times larger than the hydrodynamic diameter of the solute molecule, hindered diffusion and lower values of N will be avoided [8]. Wider pores also mean smaller surface areas and decreased sample weights. C 18 columns are used most often for reversed-phase chromatography (RPC); the physical properties of some popular columns of this type are shown in Table 5.2. 5.2.2 Silica Supports Silica in the form of either particles or monoliths is the most commonly used support for the production of HPLC packings. High mechanical strength is a strong advantage for silica particles, allowing the formation of packed beds that are stable for long periods and high operating pressures. Silica-based columns also provide 204 THE COLUMN Table 5.1 Separation Characteristics of Particles for HPLC Columns Particle Separation Pressure Ruggedness Operator Column Type Speed Convenience Stability a 5-μm totally porous 1 4 4 4 4 3.5 totally porous 2 3 4 4 4 2- to 3-μm totally porous 3 2 2 3 3 <2-μm totally porous 4 1 2 1 3 5-μm superficially porous 2 4 3 3 3 2- to 3-μm superficially porous 4 2 4 3 4 1.5-μm pellicular (nonporous) 4 1 2 1 1 Pore diameter 7- to 12-nm pores Small-molecule separations (<10, 000 Da) (150–400 m 2 /g) 15- to 100-nm pores Large-molecule separations ( > 10, 000 Da) (5–150 m 2 /g) Note: Ratings in terms of advantage from moderate (1) to high (4). a Ability to tolerate high pressures or a rapid change in pressure. Table 5.2 Properties of Some Commercial C 18 Particles Packing Material Pore Diameter Surface Area Carbon Bonded-Phase (nm) (m 2 /g) Load (%) Coverage (μmol/m 2 ) Ace C 18 10 300 15.5 na c Ascentis C 18 10 450 25 3.7 Halo C 18 a 9 150 8 3.5 Hypersil Gold C 18 17.5 220 10 na Luna C 18 (2) 10 400 17.5 3.0 Sunfire C 18 10 340 16 na TSK-GEL ODS-100V 10 450 15 na XBridge C 18 b 13.5 185 na 3.1 Zorbax XDB-C 18 Plus 9.5 160 8 na a Superficially porous particles; other particles are totally porous. b Hybrid stationary phase (Section 5.3.2.2). c Data not available. 5.2 COLUMN SUPPORTS 205 higher values of N, compared to other support materials (Section 5.2.5). Spherical particles can be synthesized with a wide choice of pore sizes (e.g., 10, 30, 100 nm), particle sizes (e.g., 1.5, 2.7, 3.5, 5 μm), and in different particle configurations (as in Fig. 5.1). Silica also can be used for monolithic columns (Section 5.2.4). Another advantage of silica is that it can be bonded with different ligands (e.g., C 8, C 18 , phenyl, and cyano; Section 5.3.1) for use with different samples and to change separation selectivity. Silica-based columns are compatible with all organic solvents and water, and do not swell or shrink with a change of solvent (as can be the case for polymeric particles; Section 5.2.3); silica particles are especially suited for gradient elution, where the mobile-phase composition changes during the separation. Silica begins to dissolve in the mobile phase at pH > 8 [9], which can result in a short lifetime for columns packed with silica particles. However, special bonded-phases are available that stabilize silica particles at higher pH (Section 5.3.2.1). Alternatively, supports other than silica can be used for increased column stability at high pH (Section 5.2.5). In the case of basic solutes, another limitation of silica has been a tendency for tailing peaks. This problem has largely been resolved by the use of higher purity silica (Section 5.2.2.2). Most silica particles in use today are spherical in shape. Spherical particles are more easily and reproducibly packed to yield efficient columns, and spherical particles tend to be stronger. Large, irregular particles are used mainly for preparative separations because of the lower cost of these materials and less need for high values of N (Section 15.3.1.2). Silica-based particles are available that can withstand pressures of 15,000 psi (especially superficially porous and pellicular particles, because of their solid core). Spherical silica particles can be synthesized by several different methods (Section 5.2.2.3 [1, 4]). Figure 5.2a shows the visual appearance (surface topography, shape and size distribution) of some commercially available silica particles, while a closer look at some popular (≈3-μm-diameter) particles is shown in Figure 5.2b. Significant deviations from a true spherical shape can be seen in some of these examples. 5.2.2.1 Column Efficiency Particle size is a primary factor in determining column efficiency as measured by the plate number N. This is illustrated in Figure 5.3 for several columns of varied particle diameter; the plate height H (inversely proportional to N) is plotted versus mobile-phase velocity u (proportional to flow rate). As the diameter d p of the porous particles decreases from 5 to 1.8 μm, the plate height H decreases—corresponding to an increase in column efficiency per mm of column length. For well-packed columns of totally porous particles, the reduced plate height h = H/d p for a small molecule is ≈2, which until about 2006 was accepted as a lower limit for well-packed columns filled with totally porous particles (Section 2.4.1). The type of particle and its particle-size distribution can also affect column efficiency, presumably by influencing the homogeneity of the packed bed. An example is shown in Figure 5.3 for the 2.7-μm superficially porous column. This column (with h ≈ 1.5) is more efficient than the column packed with 1.8-μm fully porous particles. The superior efficiency of the 2.7-μm column may result from two separate factors. First, the particle-size distribution of these particles is extremely narrow (±5–6%, 1 SD), as shown in Figure 5.4. This can be compared with a . of Particles for HPLC Columns Particle Separation Pressure Ruggedness Operator Column Type Speed Convenience Stability a 5-μm totally porous 1 4 4 4 4 3.5 totally porous 2 3 4 4 4 2- to 3-μm totally. Rigid Particles, 240 5.6.3 Soft Gels, 244 5.7 COLUMN SPECIFICATIONS, 244 5.7.1 Manufacturing Standards, 244 5.7.2 Column Plate Number, 245 5.8 COLUMN HANDLING, 246 Introduction to Modern Liquid Chromatography,. led to the exclusive use of metal columns. With the passage of time, columns became shorter and particles smaller (Fig. 1.5g, h). Column lengths of 30 to 250 mm are commonly used today, with particles