Onboarding New LC Users: A Top Ten List

Over the last year, there has been a lot of turnover of people in my laboratory, so we’ve been spending a lot of our time training and onboarding new people. I’ve found myself frequently saying things like “I find that this is one of the top three things that is most difficult to teach new users.” Upon realizing that these topics are spread out over multiple “LC Troubleshooting” articles, I decided it was time to put together a “Top Ten” list. My hope is that this list can be a good resource for anyone onboarding or training a new liquid chromatography (LC) user, whether in an academic laboratory or an industrial setting. If you are a trainer or teacher yourself, what topics are on your Top Ten list? I’d love to hear about them (email to dstoll@gustavus.edu).

In the following sections, I provide an overview of each item, as well as references to full “LC Troubleshooting” articles for readers interested in a deeper dive on specific topics. Finally, before closing, I provide a short list of resources and materials that we use regularly in my group when onboarding new folks.

The Top Ten

1. Making Good Connections, Part I: Too Loose is Better Than Too Tight

In the May, 2018 installment of “LC Troubleshooting,” I wrote about fittings and connections in detail (1). Readers interested in the full picture should take a look at it, but here I’d like to emphasize two points. First, especially when making “swaged” connections involving a nut and ferrule type of connection, in my laboratory we emphasize that “too loose is better than too tight.” The worst possible outcome is that it will leak if the swaged connection is too loose. This will produce a little puddle of mobile phase, but the remedy is simple—just tighten until the leak stops. New LC users typically do not have a sense for how tight is tight enough. For brand new connections, we teach to assemble the connection to finger tightness, and then tighten another 3/8 turn (135°) with wrenches. For reconnection of a previously swaged ferrule, we teach to assemble the connection to finger tightness, and then another 1/8 of a turn (45°) with wrenches. These practices should prevent connections that are too tight. The major problem we try to avoid is ferrules that become stuck in fittings because of over-tightening. This is especially tragic in cases where the connection being made is to an expensive part such as a valve stator, column, or pump (all >> $1k in replacement cost).

2. Making Good Connections, Part II: A Necessary Condition for Good Peak Shapes (and Widths)

The second aspect of good connections, which I think is one of the most difficult things for new users to fully appreciate, is the importance of avoiding unintended gaps or “dead volumes” in connections. Figure 1 illustrates where these gaps are most likely to occur. The effects of such gaps are usually excessive peak tailing, and larger than expected peak widths (see reference (1) for examples). These can be particularly dramatic when working with low volume (that is, less than 100 mm in length, and less than 4.6 mm i.d.) columns or small particles that are expected to produce very low volume peaks. Especially when making brand new connections where the ferrule moves freely on the connecting tubing, it can be quite difficult to ensure that the end of the tubing is fully seated at the bottom of the receiving port. Advanced designs of fittings introduced by several manufacturers in recent years have made it easier to make good connections. In my laboratory, we do not use these very often, partly because of their cost, but also because we think it is important for students to learn how to properly make good connections. We teach that the right thing to do is to have another person assist with the connection by holding the connecting tubing in place, especially the first time.

3. Diagnosing Pressure Problems

Aside from the importance of making good connections to avoid undesirable dead volumes (#2 above), I find that diagnosing problems with high pressure is the most difficult topic to teach to new users. In my view, this is mostly because users don’t have access to the information that would really make this easy, which would be a measure of the pressure at each connecting point along the flow path from the pump to the detector outlet. The only thing the instrument gives us is the pressure measured at the pump outlet, which really reflects the pressure drop from the pump to atmospheric pressure. If we could see all the pressure drops along the way, it would be straightforward to identify which piece of the flow path (for example, a filter, a guard column, or a flow cell) is producing the higher-than-expected pressure. However, we don’t have this, so new users need to learn how to approach the diagnosis process systematically. To this end, Jim Grinias and I wrote a two-part series of articles dedicated to this systematic approach, entitled: “Treat It Like a Circuit” (3,4). The strategy leverages concepts more commonly studied in electronics, where the “voltage drop” in an electrical circuit is analogous to the “pressure drop” we have in a fluidic circuit such as an LC system. I encourage both new LC users and trainers alike to consider adopting this approach to diagnosing problems with high pressure; it will make your troubleshooting experience more efficient and cost effective.

4. Inline Filters

Simply stated, being disciplined about using inline filters after sample injectors and before LC columns will save money and time, and increase the up-time of your instrument. In my laboratory, we use quite a few filters, but the upside is that we rarely see an LC column die because of plugging, or experience plugged connecting capillaries. Modern LC systems incorporate filters of various kinds at several points in the flow path that runs from the mobile phase solvent bottle to the detector outlet. By the time the mobile phase arrives at the sample injector, it should be nominally free from particulate matter of the size that could harm the LC column. However, the sample injector and injection step is a point where the mobile phase can be contaminated with particulates. The two main problems are: 1) injection of the sample itself can introduce particulates (precipitates, or other insoluble material) into the mobile phase stream. Particles on the order of microns can be harmful to the system, and are difficult to detect by eye; and 2) most injection valves involve a rotor seal that is some kind of plastic or polymeric material. As this part wears, it occasionally sheds small pieces of the seal into the mobile phase stream. Installing an inline filter downstream from the injector and upstream from the LC column can catch these particulate materials before they cause trouble downstream. The first and most important benefit of this is that it protects the column, leading to better and more consistent peak shapes, as well as longer column lifetimes. Second, catching the particulates prevents them from getting stuck in connecting capillaries that can lead to pressure problems and taking the instrument offline while waiting for a repair. So, our practice is to emphasize to new users the importance of making sure an inline filter installed in their LC system. Readers interested in reading more about this topic are referred to the February, 2017 issue of “LC Troubleshooting” (5).

5. Checking Flow and Pressure Eyeballometrically

My good friend and Professor Emerita Sarah Rutan is a chemometrician who loves statistics and data analysis. But she is also fond of the term “eyeballometrically,” which to us means “analysis by eye.” In most LC systems, there is no actual measurement of the flow rate by a means other than what the pump reports. If a mobile phase solvent bottle is empty, the pump may tell you it is pumping at 1.0 mL/min when in fact there is no flow coming from the pump whatsoever. With modern computer networking there is a lot we can do with remote control of instruments, sometimes with an ocean between us and the instrument. But, there is no replacement for walking to the instrument and seeing with our eyes what is happening, physically. Generally speaking, I find that new users have a little too much faith in the instrument control software when it shows that it is “ready” to start an analysis. What constitutes “ready” for the software, and for us, are somewhat different things. We need to check, eyeballometrically, so see that: (1) the flow through the system is at least close to what it should be; and (2) the pressure reported by the pump is stable over time. An approximate measure of the flow rate for analytical scale work can be made by simply collecting the effluent at the detector outlet into a graduated cylinder and recording the volume collected over a couple of minutes. For the pressure, we should see that the short-term variation in the pressure measurement doesn’t exceed about 1% of the nominal value. If the variation is larger than this, we need to pause and address the problem before starting an analysis. Readers interested in learning more about diagnosing problems with flow and pressure are referred to prior installments of “LC Troubleshooting” (6).

6. Check the Solvents, Also Eyeballometrically

This one is simple—is there actually solvent in the solvent bottles feeding the pump? Although the situation is changing with some commercial offerings of LC systems that actually measure the amount of solvent in the bottles, most LC systems don’t do this. Again, the pump may say it is “ready,” even if one or more solvent bottles are actually empty. Before pressing “start” on the analysis, we need to check, eyeballometrically, to make sure there is solvent in the bottles, and that there is enough there to complete the analysis.

7. Extra-Column Peak Broadening

In addition to diagnosing pressure problems and making good connections, this is the third topic in my “top three most difficult to teach.” The issue here is that peak broadening that happens outside the column (hence the name “extra-column” broadening) can be significant compared to the broadening that happens inside the column, and effectively rob our separations of the efficiency (and resolution) they should have. I think the major challenge with this is that we cannot see inside the parts of the flow path where this type of broadening occurs (for example, injectors, detectors, and connecting capillaries). If we could see the broadening that was happening inside, it would be quite obvious that we should take steps to mitigate the sources of peak broadening (for example, reducing and detector volumes, and reducing the length and diameter of connecting tubing). To help overcome this challenge, we have developed a web-based calculator that estimates the contribution of different parts of the flow path to peak broadening, based on the best available theory. Readers interested in learning more about the theory and how to use the web-based calculator are referred to a series of “LC Troubleshooting” articles where we address this topic in detail (7–10).

8. Protecting Columns

In many ways, the LC column itself is the most precious part of an LC system, and we should be digilent about protecting it. This includes avoiding conditions that are too acidic, too alkaline, and so on. Several “LC Troubleshooting” installments have been written about this over the years (11,12), and readers interested in learning more about these topics are referred to those articles. What I want to emphasize here is that we also need to be careful about how we treat the column between uses. As a matter of routine, mobile phases containing additives should be flushed out of the column after use, so that the column only contains organic solvent and water (assuming the column is compatible with the solvents; please check the user guide inside the column box for guidance about solvent compatibility). This flushing step avoids two things: 1) precipitation of salts if the solvent evaporates; and 2) growth of microbes inside the column. In the best case, an analysis can be programmed to automatically switch to a mixture of organic solvent and water after all samples have been analyzed so that the column is flushed before the pump turns off. In my laboratory, we try to flush columns if they are not going to be used again within about two days. Readers interested in learning more about column storage are referred to the July 2017 installment of “LC Troubleshooting” (13).

9. Expected Retention Behavior

An apparently simple, yet powerful, piece of knowledge when practicing any mode of LC is the expected retention behavior, particularly the dependence of retention on mobile phase composition. In the case of reversed-phase separations, we expect that the retention will decrease as we increase the fraction of organic solvent in the mobile phase. Conversely, in hydrophilic interaction liquid chromatography (HILIC) separations, we expect that increasing the water content of the mobile phase should decrease analyte retention, and so on. Of course, there are exceptions to these general trends, but they will be correct, and predictable, in the vast majority of cases we will encounter. When working with RP separations, if we increase the fraction of organic solvent in the mobile phase and retention increases, either we’ve got some unusual chemistry on our hands, or (more likely) something is wrong with the instrument or column. At this point, we have to pause and find out (troubleshoot) why we are seeing this behavior before moving on with method development or analysis of unknown samples.

10. The Purge Takes Longer Than You Think

In the field of LC, we talk about small volumes a lot—µL-level injection volumes, flow cells with single-digit µL volumes, and so on. When it comes to changing mobile phase solvents, though, we need to forget those small volumes. The 1/8” outer diameter (o.d.) tubing that is typically used to connect the solvent bottles to the pump inlet typically has an i.d. of about 1.5 mm. The volume contained per meter length of this tubing is about 2 mL, and it is common to see about 1 m of this tubing running from the solvent bottles to the pump inlet. To thoroughly flush this tubing when changing solvents, we need to pull several multiples of this volume through the tubing, in part because these tubes are usually straight enough that the flow in them is probably laminar, and the flushout is not very efficient. In my laboratory we teach that when changing solvents we need to purge each solvent line with 15 mL of the new solvent. This might seem like overkill, but I’d rather overdo it a bit here than risk having a situation where the previously-used solvent is not completely washed out of the solvent line; this can lead to drifting retention times for hours.

Training Resources and Materials

There are many excellent resources available for teaching and learning about all aspects of LC. Here I want to highlight just three that we use routinely in my group when onboarding new people.

Chapter 1 of Mark Vitha’s book entitled Chromatography: Principles and Instrumentation (14): An excellent, approachable introduction to theory of separations for those with no experience in chromatography.

Analytical Separation Science, by Pirok and Schoenmakers (15): One of the most recent entrants into the array of comprehensive texts on separation science, this book is organized by level (that is, B, M, and A levels) to help guide readers (and teachers) to the material that is most appropriate for their level and learning goals.

Introduction to Modern Liquid Chromatography, by Snyder, Kirkland, and Dolan (16): The truly comprehensive resource for both fundamental and practical aspects of LC, this book is probably the single most commonly found text on LC users’ shelves.

Acknowledgments

Thanks to Jim Grinias and Emanuela Gionfriddo for sharing items on their Top Ten lists.

Corrigendum

There were two errors associated with Figure 3 in the August, 2025 installment of “LC Troubleshooting.” In Panels A and B, the plate heights for the 3.5 µm column were calculated incorrectly. In Panel C, the axis labels appeared with units, however reduced velocity and reduced plate height are dimensionless parameters, and should not have any units. These errors do not affect the explanations given in the text, but I apologize for any confusion this may have caused. The errors have been corrected in the online version of the article.

References

1. Stoll, D. R. Fittings and Connections for Liquid Chromatography—So Many Choices! LCGC North Am. 2018, 36 (5), 304–311.
2. Li, T.; Zeko, D.; Fuehrer, M. Optimizing HPLC and UHPLC Performance with Agilent A-Line Quick Connect Fittings; in: Shanghai, China, 2014.
3. Stoll, D. R.; Grinias, J. P. Treat It Like A Circuit, Part I: Comparison of Concepts from Electronics to Flow in LC Systems. LCGC Int. 2024, 1 (4), 6–10. DOI: 10.56530/lcgc.int.ol2372p5
4. Stoll, D. R.; Grinias, J. P. Treat It Like a Circuit, Part II: Applications and Troubleshooting. LCGC Int. 2024, 1 (5), 6–12. DOI: 10.56530/lcgc.int.wy6073a3
5. Stoll, D. R. Filters and Filtration in Liquid Chromatography-What To Do. LCGC North Am. 2017, 35 (2), 98–103.
6. Stoll, D. R. Essentials of LC Troubleshooting, Part I: Pressure Problems. LCGC North. Am. 2021, 39 (12), 572–574.
7. Stoll, D. R.; Broeckhoven, K. Where Has My Efficiency Gone? Impacts of Extracolumn Peak Broadening on Performance, Part I: Basic Concepts. LCGC North Am. 2021, 39 (4) 159–166.
8. Stoll, D. R.; Broeckhoven, K. Where Has My Efficiency Gone? Impacts of Extracolumn Peak Broadening on Performance, Part II: Sample Injection. LCGC North Am. 2021, 39 (5), 208–213.
9. Stoll, D. R.; Broeckhoven, K. Where Has My Efficiency Gone? Impacts of Extracolumn Peak Broadening on Performance, Part III: Tubing and Detectors, LCGC North Am. 2021, 39 (6) 252–257.
10. Stoll, D. R.; Lauer, T. J.; Broeckhoven, K. Where Has My Efficiency Gone? Impacts of Extracolumn Peak Broadening on Performance, Part IV: Gradient Elution, Flow Splitting, and a Holistic View. LCGC North Am. 2021, 39 (7) 308–314.
11. Dolan, J. W. Column Care and Feeding. LCGC Magazine 1993, 11, 22–24.
12. Stoll, D. R. The Evolution of LC Troubleshooting: Strategies for Improving Peak Tailing. LCGC North Am. 2023, 41 (10), 404–408. DOI: 10.56530/lcgc.na.br4774h5
13. Stoll, D. R. Column Care for the Long Haul-Considerations for Column Storage. LCGC North Am. 2017, 35 (7), 434–439.
14. Vitha, M. F. Chromatography: Principles and Instrumentation; Wiley, 2017.
15. Pirok, B. W. J. Analytical Separation Science, 1st ed.; Royal Society of Chemistry, 2025.
16. Snyder, L. R.; Kirkland, J. J.; Dolan, W. Introduction to Modern Liquid Chromatography, 3rd ed.; Wiley, 2010.