806 SAMPLE PREPARATION 29. G. M. Hearne and D. O. Hall, Amer. Lab., 25(1) (1993) 28H. 30. R. Majors, LC/GC, 23 (2005) 358. 31. R. E. Majors and A. Shukla, LC/GC, 23 (2005) 646. 32. A. A. Boyd Boland and J. B. Pawliszyn, Anal. Chem., 68 (1996) 1521. 33. R. Shirey, Advances and Applications of Solid Phase Microextractios (SPME),paper 23602 at Analytica, Munich, 1996, pp. 23–26. 34. E. Balrussen, P. Sandra, F. David, and C. A. Cramers, J. Microcol. Sep., 11 (1999), 737. 35. B. A. Bidlingmeyer, Liq. Chromatogr., 2 (1984) 578. 36. T. A. Dirksen, S. M. Price, and S. J. St. Mary, Amer. Lab., 25(18) (1993) 24. 37. G. E. Platoff and J. A. Gere, Forens. Sci. Rev., 3 (1991) 117. 38. X H. Chen, J P. Franke, K. Ensing, J. Wijsbeek, and R. A. De Zeeuw, J. Chromatogr. 613 (1993) 289. 39. V. Dixit and V. M. Dixit, J. Liq. Chromatogr., 14 (1991) 2779. 40. R C. Hsu, I. Biggs, and N. K. Saini, J. Agric. Food Chem., 39 (1991) 1658. 41. T. Tanaka, T. Hori, T. Asada, K. Oikawa, and K. Kawata, J. Chromatogr. A, 1175 (2007) 181. 42. N. J. K. Simpson, ed., Solid-Phase Extraction: Principles, Techniques, and Applications, Dekker, New York, 2000. 43. E. M. Thurman and M. S. Mills, Solid-Phase Extraction: Principles and Practice, Wiley, New York, 1998. 44. J. S. Fritz, Analytical Solid-Phase Extraction, Wiley-VCH, New York, 1999. 45. N. Simpson and K. C. van Horne, Sorbent Extraction Technology Handbook,Varian, Harbor City, CA, 1993. 46. M. J. Telepchak, Forensic and Clinical Applications of Solid Phase Extraction,Humana Press, Totowa, NJ, 2004. 47. R. Bland, Proceedings of the 3rd Annual International Symposium on Sample Prepara- tion and Isolation using Bonded Silicas, Analytichem International, Harbor City, CA, 1986, pp. 93–116. 48. E. J. Rook, M. J. X. Hillebrand, H. Rosing, J. M. van Ree, and J. H. Beijnen, J. Chromatogr. B, 824 (2005) 213. 49. M. Raisglid and M. F. Burke, abstract 653, 48 th Pittsburgh Conference on Analytical Chemistry and Applied Spectroscopy, Atlanta, GA, 1997. 50. K S. Boos and A. Rudolphi, LCGC, 15 (1997) 602. 51. K S. Boos and A. Rudolphi, LCGC, 15 (1997) 814. 52.N.M.Cassiano,V.V.Lima,R.V.Oliveira,A.C.dePietra,andQ.B.Cass,Anal. Bioanal.Chem., 384 (2006) 1462. 53. D. R. Doerge, M. L. Churchwell, and K. Berry Delclos, Rapid Commun. Mass Spec- trometry, 14 (2000) 673. 54. K. Ensing, C. Berggren, and R. E. Majors, LCGC, 19 (2001) 942. 55. S. G. Dmitrienko, V. V. Irkha, A. Yu. Kuznetsova, and Yu. A. Zolotov, J. Anal. Chem., 59 (2005) 808. 56. C. Baggiani, L. Anfossi, and C. Giovannoli, Current Pharma. Anal., 2 (2006) 219. 57. J. O. Mahony, K. Nolan, M. R. Smyth, and B. Mizaikoff, Anal.Chim.Acta, 534 (2005) 31. 58. P. A. G. Cormack and A. Z. Elorza, J. Chromatogr. B, 804 (2004) 173. 59. S. Piletsky and A. Turner, Molecular Imprinting of Polymers, Landes Bioscience, Austin, TX, 2006. REFERENCES 807 60. M C. Hennion and V. Pichon, J. Chromatogr. A, 1000 (2003) 29. 61. V. Pichon, N. Delaunary-Bertoncini, and M C. Hennion, in Comprehensive Analytical Chemistry, Vol. 37 J. Pawliszyn, ed., Elsevier, Amsterdam, 2002, p. 1081. 62. N. Delaunay, V. Pichon, and M C. Hennion, J. Chromatogr. B, 745 (2000) 15. 63. D. Stevenson, J. Chromatogr. B, 745 (2000) 39. 64. N. Zolotarjova, P. Mrozinski, and R. E. Majors, LCGC, 25 (2007) 118. 65. R. E. Majors, LCGC, 25 (2007) 16. 66. J. Stroka, E. Anklam, U. J ¨ orissen, and J. Gilbert, JAOAC Int., 83 (2000) 320. 67. M. Anastassiades, S. J. Lehotay, D. Stajnbaher, and F. J. Schenck, JAOAC Int.,86 (2003) 412. 68. M. Anastassiades, http://www.quechers.com/docs/quechers recov.pdf (2004). 69. S. J. Lehotay, A. de Kok, M. Hiemstra, and P. Van Bodegraven, JAOAC Int., 88 (2005) 595–614. 70. C. F. Fagerquist, A. R. Lightfield, and S. J. Lehotay, Anal.Chem., 77 (2005) 1473. 71. N. Jakubowska, Z. Polkowska, J. Namiesnik, and A. Przyjazny, Crit.Rev.Anal.Chem. 35 (2005) 217. 72. Method 3550C, Ultrasonic Extraction, US EPA, http://www.epa.gov/epawaste/hazard/ testmethods/sw846/online/3 series.htm. 73. E. L. Randall, JAOAC Int., 57 (1974) 1165. 74. Method 3541, Automated Soxhlet Extraction, US EPA, http://www.epa.gov/ epawaste/hazard/testmethods/sw846/online/3 series.htm. 75. J. L. Luque-Garcia and L. de Castro. J. Chromatogr A., 998 (2003) 21. 76. J. L. Luque-Garc ´ ıa,M.J.Ramos,M.J.Mart ´ ınez-Bueno, and M. D. Luque de Castro, Chromatographia, 62 (2005) 69. 77. J. L. Luque-Garcia and M. D. Luque de Castro., Anal. Chem., 73 (2001) 5903. 78. J. M Levy, LCGC, 17(6S) (1999) S14. 79. R. Smith, J. Chromatogr. A, 1000 (2003) 3. 80. L. T. Taylor, Techniques in Analytical Chemistry: Supercritical Fluid Extraction, Wiley, New York, 1996. 81. M. D. Luque De Castro, M. Valcarcel, and M. T. Tena, Analytical Supercritical Fluid Extraction, Springer, New York, 1994. 82. E. D. Ramsey, ed., Analytical Supercritical Fluid Extraction Techniques,KluwerAca- demic, Dordrecht, 1998. 83. Method 3545A, Pressurized Fluid Extraction (PFE), US EPA, http://www.epa.gov/ epawaste/hazard/testmethods/sw846/online/3 series.htm. 84. Accelerated Solvent Extraction (ASE) of Active Ingredients from Natural Products, Application note 335, Dionex Corp., Sunnyvale, CA. 85. G. LeBlanc, LCGC, 17(6S) (1999) S30. 86. Method 3546, Microwave Extraction, US EPA, http://www.epa.gov/epawaste/hazard/ testmethods/sw846/online/3 series.htm. 87. Method 3540C, Soxhlet Extraction, US EPA, http://www.epa.gov/epawaste/hazard/ testmethods/sw846/online/3 series.htm. 88. N. Tang and C. Miller, An Integrated Approach to Improve Sequence Coverage by Combining LC-MALDI MS/MS and NanoLC-LC-MS/MS, Agilent Technologies, Santa Clara, CA, Publication Number 5989-3500EN (2005). 89. S. R. Souverain and J L. Veuthey, J. Pharm. Biomed. Anal., 35 (2004) 913. 808 SAMPLE PREPARATION 90. R. Herr ´ aez-Hern ´ andez, P. Camp ´ ıns-Falc ´ o, and A. Sevillano-Cabeza, Chromatographia, 33 (1992) 177. 91. L. Ynddal and S. H. Hansen, J. Chromatogr. A, 1020 (2003) 59. 92. J.Ayrton,G.J.Dear,W.J.Leavens,D.N.Mallett,andR.S.Plumb,Rapid Comm. Mass Spec., 11 (1997) 1953. 93. K. N. Frayn and P. F. Maycock, Clin. Chem. 29 (1983) 1426. 94. Y J. Xue, J. B. Akinsanya, J. Liu, S. E. Unger, Rapid Comm. Mass Spec., 20 (2006) 2660. 95. J. Porath, J. Protein Chem., 16 (1997) 463. 96. F. Laborda, M. V. Vicente, J. M. Mir, and J. R. Castillo, Anal. Bioanal. Chem., 357 (1997) 1618. 97. C. Shaw, Meth.Mol.Biol., 64 (1996) 101. 98. Y. Wang and J. Seneviratne, Curr. Proteomics, 5 (2008) 104. 99. J. Martosella, N. Zolotarjova, H. Liu, G. Nicol, and B. E. Boyes, Proteomics, 13 (2005) 3304. 100. N. Zolotarjova, J. Martosella, G. Nicol, J. Bailey, B. E. Boyes, and W. C. Barrett, Electrophoresis, 25 (2004) 2402. 101. Y. Ikeguchi and H. Nakamura, Anal. Sci., 16 (2000) 541. 102. A. Apffel, S. Fishcher, G. Goldberg, P. C. Goodley, and F. E. Kuhlman, J. Chromatogr. A., 712 (1995) 177. 103. S. Ahuja, Selectivity and Detectability Optimization in HPLC, Wiley, New York, 1989. 104. B. King and G. S. Graham, Handbook of Derivatives for Chromatography, Heyden, Philadelphia, 1979. 105. H. Lingeman, Detection-Oriented Derivatization T echniques in Liquid Chromatogra- phy, Dekker, New York, 1990. 106. J. F. Lawrence and R. W. Frei, Chemical Derivatization in Liquid Chromatography, 3rd ed., Elsevier, Amsterdam, 1985. 107. T. Toyo’oka, Modern Derivatization Methods for Separation Science, Wiley, West Sussex, UK, 1999. 108. K. Blau and J. M. Halket, eds., Handbook of Derivatives for Chromatography, 2nd ed, Wiley, West Sussex, UK, 1993. CHAPTER SEVENTEEN TROUBLESHOOTING 17.1 INTRODUCTION, 810 Quick Fix, 809 17.2 PREVENTION OF PROBLEMS, 811 17.2.1 System Performance Tests, 811 17.2.2 Periodic Maintenance, 812 17.2.3 System-Suitability Testing, 813 17.2.4 Historical Records, 813 17.2.5 Tips and Techniques, 814 17.3 PROBLEM-ISOLATION STRATEGIES, 819 17.3.1 Divide and Conquer, 819 17.3.2 Easy versus Powerful, 820 17.3.3 Change One Thing at a Time, 820 17.3.4 Address Reproducible Problems, 820 17.3.5 Module Substitution, 820 17.3.6 Put It Back, 821 17.4 COMMON SYMPTOMS OF HPLC PROBLEMS, 821 17.4.1 Leaks, 822 17.4.2 Abnormal Pressure, 830 17.4.3 Variation in Retention Time, 833 17.4.4 Peak Area, 838 17.4.5 Other Problems Associated with the Chromatogram, 841 17.4.6 Interpretation of System Performance Tests, 856 17.5 TROUBLESHOOTING TABLES, 865 QUICK FIX If your goal is to solve a problem quickly and not to read about troubleshooting, one of two approaches is recommended. Look through the detailed table of contents at Introduction to Modern Liquid Chromatography, Third Edition, by Lloyd R. Snyder, Joseph J. Kirkland, and John W. Dolan Copyright © 2010 John Wiley & Sons, Inc. 809 810 TROUBLESHOOTING the front of the book and find the section of this chapter that relates to your problem. Alternatively, go to Section 17.5 at the end of this chapter and use Tables 17.2 through 17.11 to guide you through the troubleshooting process and give you a cross-reference to the text for more information. Each chapter in this book contains troubleshooting information relative to the topic of discussion; these are easily located by consulting the appropriate topic in the index at the end of the book. 17.1 INTRODUCTION When the second edition of this book was written, troubleshooting was a major part of the job of the HPLC operator. Problems related to excessive pump-pressure pulsation, pressure shocks from manual injectors, short lifetimes for pump seals and detector lamps, and general instrument problems abounded. Columns were subject to failure as a result of shipping damage or column-bed collapse, and column inlet-frit replacement was so common that most manufacturers shipped several extra frits with each column. A detailed understanding of gradient elution was in its infancy, and this lack of understanding resulted in unexpected changes in the chromatogram with changes in flow rate or other gradient adjustments, and the transfer of gradient methods was quite difficult. Fast-forward 30 years, and the situation has changed dramatically. HPLC hardware is much more reliable, with routine maintenance intervals of 6 to 12 months or longer, instead of on a weekly or monthly basis. With the advent of higher purity, type-B column packings, improved particles, and better column-packing procedures, column problems are a small fraction of those of their ancestors. Gradient elution is well understood, and with a little care, it can perform as well as isocratic techniques, even in the hands of relatively inexperienced users. This chapter focuses on HPLC problems and how to correct them. At the core of troubleshooting is the prevention of problems (Section 17.2)—a major strategy for reliable HPLC system operation. Problems are most easily isolated by use of a disciplined approach (Section 17.3) to identify specific symptoms (Section 17.4). Troubleshooting and problem prevention are greatly aided by a good under- standing of how the HPLC system operates; readers are encouraged to get this information by reading Chapter 3 (equipment) and the appropriate part(s) of Chapter 4 (detectors). At a minimum, Section 3.10 (maintenance) should be reviewed; main- tenance practices tie in closely to troubleshooting and will be cross-referenced regularly in this chapter. Of course, do not forget to check the troubleshooting and preventive maintenance sections of the instrument manuals for specific instructions that apply to your HPLC system. Another aspect of troubleshooting and preventive maintenance is the question of who will make the necessary repairs. A discussion of this topic is covered in Section 3.10.3.1; Table 3.7 lists recommended repair activities for different personnel. The final determination of who is qualified to make repairs is made by a combination of laboratory policy, regulatory requirements, training, and personal mechanical aptitude. In any event, a good understanding of instrument operating principles and the troubleshooting process will benefit workers at all levels of competence and responsibility. 17.2 PREVENTION OF PROBLEMS 811 Finally, it is important to recognize that HPLC troubleshooting is difficult to condense into a single chapter, as in the present case. Other resources exist, and you are encouraged to explore these for additional help. There are numerous books on the subject. References [1, 2] are a good place to start, but new material is constantly being written—a Google of ‘‘HPLC troubleshooting books’’ at the time of this writing yielded 13,600 hits! (but not 13,600 books). Another excellent source of regular advice on HPLC troubleshooting is the ‘‘LC Troubleshooting’’ column, found each month in LCGC magazine and written by one of the authors [3]. On-line discussion groups also provide troubleshooting support. One of the best of these is Chromatography Forum [4], a discussion of troubleshooting and related problems that is monitored and contributed to by experts in all aspects of HPLC. There also are on-line expert-system tools, such as the HPLC Wizard [5] that can help you isolate and identify HPLC problems. Each laboratory should establish its own preventive maintenance program. A list of possible items to include can be found under the ‘‘Best practices’’ entry in this book’s index. 17.2 PREVENTION OF PROBLEMS The best kind of HPLC problem is the one that does not occur. For this to be the general pattern for an HPLC system, a structured preventive-maintenance program, such as that described in Section 3.10.2, should be established. Four elements are important in this process. First, a system performance test (Section 17.2.1) is used to establish that the HPLC system can operate properly under ideal conditions. Second, periodic maintenance (Section 17.2.2) needs to be performed to repair or replace parts that have a limited lifetime due to normal wear. Third, a system suitability test (Section 17.2.3) should be run prior to each batch of samples to ensure that the method and equipment are working at a level that will produce acceptable data. Finally, a repair and maintenance record (Section 17.2.4) should be kept as proof of maintenance and to establish failure patterns for each HPLC system. 17.2.1 System Performance Tests An HPLC system performance test is described in Section 3.10.1; a brief summary is included here—consult Section 3.10.1 for details. There are three types of tests that can be used to test system performance: Installation Qualification, Operational Qualification, and Performance Qualification (Section 17.2.1.1). All three groups of tests should be carried out when the HPLC system is new. The gradient tests (Section 17.2.1.2) and additional tests (Section 17.2.1.3) should be performed every 6 to 12 months to ensure continued reliable HPLC operation. Throughout this book it is assumed that the HPLC system is operating in reversed-phase mode with a UV detector. Many of the tests that are described work best with UV detection, including the tests listed and cross-referenced in this section. It may be possible to devise ways to accomplish the same testing goals with non-UV detectors, but some tests—such as the various gradient performance tests—will be very difficult to perform with certain detectors (e.g., LC-MS or LC-MS/MS). For this reason we recommend that every laboratory have at least one UV detector available to facilitate performance testing and troubleshooting. 812 TROUBLESHOOTING 17.2.1.1 Installation Qualification (IQ), Operational Qualification (OQ), and Performance Qualification (PQ) Tests A combination of tests designed by the instrument manufacturer (IQ, OQ) and the user (PQ) show that the HPLC system works as it was designed to and performs according to the published instrument specifications. These tests are performed when the instrument is new, and the PQ test is repeated periodically afterward. The results of these tests play an important role in the divide-and-conquer problem-isolation strategy (Section 17.3.1), by allowing the user to check the current performance of the instrument against its original performance when it was known to be working properly. This helps to answer the ‘‘system or method?’’ question that often arises when an HPLC problem occurs. 17.2.1.2 Gradient Performance Test If the HPLC system has gradient capabilities, the gradient performance test (Section 3.10.1.2) should be run, even if the system is used only for isocratic applications with on-line mixing of the mobile phase. The gradient performance test checks the linearity and accuracy of mobile-phase preparation as well as measures a value of the gradient dwell volume. If the system is used only for isocratic methods with on-line mixing, an alternative test is to inject a standard solution under isocratic conditions. Then exchange the A- and B-solvent reservoirs and adjust the program to deliver the same mobile phase (e.g., 30% A = water + 70% B = MeOH becomes 30% B = water and 70% A = MeOH) and inject again. The retention times in both cases should be the same. These tests ensure that when a mobile-phase mixture is programmed into the system controller, the desired mixture is delivered to the column. See Section 17.4.6.1 for specific examples of gradient performance test failures. 17.2.1.3 Additional System Tests A few more tests will round out the overall testing of the performance of the HPLC system (Section 3.10.1.3). A flow-rate check will determine if the pump is operating properly, and the pressure bleed-down test checks the outlet check-valves and/or pump-seal integrity. A retention-reproducibility test double-checks the flow rate and verifies that on-line mobile-phase preparation is consistent. The peak-area reproducibility test ensures that the autosampler is working as it should. See Section 17.4.6.2 for specific examples of failure of the additional system tests. 17.2.2 Periodic Maintenance A regular maintenance program will allow the HPLC system to work more reliably and encounter fewer breakdowns. A list of recommended preventive maintenance items is contained in Table 3.6 and discussed in Section 3.10.2.1. As a rule, time spent in preventive maintenance, especially to ensure that the system is cleaned out regularly and normal-wear parts (e.g., pump seals) are replaced before failure, will be time well spent and will reduce overall operating costs. Laboratories that always wait for a failure to occur before performing HPLC maintenance usually spend more time and money on troubleshooting and repair than those who regularly maintain their HPLC systems. 17.2 PREVENTION OF PROBLEMS 813 17.2.3 System-Suitability Testing System suitability refers to a series of injections, pre-defined for a specific method, that are made prior to running samples to ensure that the HPLC system is working properly and that the system is able to generate data that will meet the acceptance criteria of the method. To accomplish this task, the system-suitability sample must be able to adequately assess method performance—usually a synthesized sample made from standards mixed in a real or formulated sample matrix. These tests may include a subset of the performance qualification test (Sections 17.2.1, 3.10.1.1) as well as other tests, such as retention-time and/or peak-area reproducibility, that demonstrate that the HPLC is working suitably as a system. System-suitability testing should assess the most meaningful parameters of the separation (e.g., resolution, accuracy, reproducibility) to ensure the quality of the chromatographic data. In other words, system-suitability testing should answer the question, ‘‘Does the HPLC system currently work acceptably for this method?’’ Whether or not system-suitability tests are required by a regulatory agency or local operating procedures, it is wise to make some kind of a test prior to running samples to minimize the risk of collecting meaningless data and wasting precious samples—as well as time and resources. A historical record (Section 17.2.4) of system-suitability tests can also be useful to track down the source of a problem. See Section 12.3 for further information on system-suitability testing. 17.2.4 Historical Records The value of historical records in establishing instrument failure patterns and preven- tive maintenance programs cannot be underestimated. Section 3.10.3.2 described a suggested set of records, comprising at least three elements that should be maintained for each HPLC system. A record of the system configuration will include sufficient information to identify each system component (pump, autosampler, detector, etc.) that makes up a specific HPLC system. This, combined with sample batch-records (e.g., sample identification associated with injection number), should allow correla- tion of each sample run with a specific HPLC configuration. Maintenance records will provide a written record of all system maintenance activities. The results of system checks , such as performance qualification and gradient performance tests (Section 17.2.1) also should be included in the records for each HPLC system. Historical records serve two purposes. The first is to provide documentation to a regulatory agency that proper procedures for instrument qualification, use, and maintenance were in place and were followed. The second is to provide data to help design a preventive maintenance program. After a sufficient period of time, such as one to two years, depending on the intensity of HPLC system use, enough data should be available to determine failure patterns. If several HPLC systems in the laboratory are nominally identical (i.e., same brand and model), data from several systems may be pooled. For example, to determine a pump-seal replacement cycle, the pump-seal replacement interval records could be pooled for all instruments. Usually it is best to replace the pump seals prior to failure, so that particles generated from seal wear do not block tubing or frits downstream from the pump. If seal replacement was performed at 7, 6, 8, 12, and 11 months during 5 different repair incidents, the data suggest that it would be prudent to proactively replace the seals every 6 months so as to avoid the consequences of seal failure. A similar procedure 814 TROUBLESHOOTING can be used to identify other replacements or repairs that should be added to a preventive maintenance list. As a rule, if the lifetime of a part can be anticipated, replacement of the part at 70–80% of its anticipated lifetime is a good target for preventive service activities—this gets most of the useful service from the part while reducing risk of lost data when the part fails. 17.2.5 Tips and Techniques This section contains a collection of tips and techniques that can be used to isolate and correct HPLC problems, as well as prevent them from happening in the future. The topics highlight some of the most common problem areas in HPLC operation—by paying special attention to these, you should be able to reduce problem frequency, as well as the amount of time spent to correct problems when they occur. These are individually cross-referenced throughout this chapter. 17.2.5.1 Removing Air from the Pump The internal parts of the HPLC pump and associated hardware have many small, often angular, passages that can trap air bubbles. Sometimes a sharp tap on a pump head with a wooden or plastic object, such as a screwdriver handle, will dislodge bubbles. A system flush with a thoroughly degassed, low-viscosity, low-surface-tension solvent such as methanol will sometimes dissolve bubbles that resist displacement using other techniques. The use of degassed solvents on a routine basis will prevent the accumulation of bubbles in the system, since the solvent will have an additional capacity for dissolved gas that will solubilize tiny bubbles before they become a problem. Every HPLC system will work more reliably if the mobile phase is degassed. 17.2.5.2 Solvent Siphon Test All HPLC systems will perform more reliably if the reservoirs are elevated relative to the pump, so that a slight siphon head-pressure helps deliver mobile phase to the pump. To ensure a free flow of solvent to the mixer, it is important to check the solvent inlet-line frits occasionally. Because many pumps have an ‘‘asymmetric’’ duty cycle in which they spend more time delivering solvent than refilling, the flow rate during the intake stroke can be much higher than the overall average flow rate. Therefore one should expect the reservoir to be able to deliver several times more solvent by siphon action than will be required by the pump. To test this feature, disconnect the solvent inlet-line at the mixer (low-pressure mixing) or the pump inlet (high-pressure mixing) and allow the solvent to siphon through the tubing. A 10-fold excess of solvent is a good rule of thumb for adequate delivery. For example, if the typical operation of the system is 1 mL/min, at least 10 mL/min of solvent through the siphon should be expected. This will supply enough solvent so that starvation of the pump will never be an issue. If the flow is lower than expected, check for a blocked solvent inlet-line frit, a pinched inlet-line, or a poorly vented reservoir. A restricted solvent supply in low-pressure-mixing systems also can cause mobile-phase proportioning errors (Section 17.4.6.1). 17.2 PREVENTION OF PROBLEMS 815 Time % organic 85 95 1 2 Figure 17.1 Programmed shallow-gradient profile ( ), profile actually observed (—). Mobile phase is stronger at point 1 than 2, even though point 1 occurs earlier in the gradient. 17.2.5.3 Pre-mixing to Improve Retention Reproducibility in Shallow Gradients Complex mixtures often require high-resolution separations that utilize shal- low gradients. Also gradient slopes of <1%/min often are required for high- molecular-weight compounds, such as peptides and proteins so that reasonable k ∗ -values can be obtained for these samples with large S-values (Sections 9.4, 13.4.1.4). In some cases the HPLC equipment cannot generate these shallow gra- dients with sufficient accuracy to obtain an acceptable separation. Low-pressure mixers generate the gradient by mixing alternate pulses of the A- and B-solvents; because on-line mixing is never complete, a small residual variation in mobile phase composition (and strength) remains when the mobile phase reaches the column. This is visualized in Figure 17.1 as the solid trace overlaid on the programmed gradient (dashed line). Thus there is an oscillation of values of %B around the programmed gradient throughout the separation. For this very shallow gradient, the value of %B at point 1 is greater than at point 2, even though point 1 appears earlier in the gradient. To correct this problem, the mobile phases can be pre-mixed. For example, when pure solvents are used for the A- and B-solvents, a gradient of 40–50% B over 50 minutes requires that the system accurately deliver the gradient at a slope of 0.2%/min, which may challenge the precision of some equipment. When A and B are pre-mixed to 40% B (in the A-reservoir) and 50% B (in the B-reservoir), the system can be programmed to deliver a solvent gradient of 0–100% in 50 minutes, or 2%/min, a much easier job for most gradient equipment. Thus pre-mixing the mobile phase in this way can overcome the limitations of the equipment. The same pre-mixing technique can be used to reduce baseline noise in both isocratic and gradient separations when the detector is operated at maximum sensitivity. For example, isocratic methods with refractive index detection (Section 4.11) will have less baseline noise if the mobile phase is pre-mixed instead of using on-line mixing. Gradient baselines generally are improved if 5% of the B-solvent is mixed in the A-reservoir and 5%A in the B-reservoir. 17.2.5.4 Cleaning and Handling Check Valves As an alternative to the replacement of pump check valves with new parts, faulty check valves often can be rejuvenated by sonication in alcohol. Note that the inlet . in Liquid Chromatogra- phy, Dekker, New York, 1990. 106. J. F. Lawrence and R. W. Frei, Chemical Derivatization in Liquid Chromatography, 3rd ed., Elsevier, Amsterdam, 1985. 107. T. Toyo’oka, Modern. is recommended. Look through the detailed table of contents at Introduction to Modern Liquid Chromatography, Third Edition, by Lloyd R. Snyder, Joseph J. Kirkland, and John W. Dolan Copyright. end of the book. 17.1 INTRODUCTION When the second edition of this book was written, troubleshooting was a major part of the job of the HPLC operator. Problems related to excessive pump-pressure pulsation,