The Current Status and Future of LC Column Technology: What the Experts Are Saying

The Significance of This Article-Then and Now

During the course of my writing for LC and LCGC magazine, and coming from a technical marketing background, I always tried to keep the readers in mind when it came to understanding trends in the marketplace. In my 33 years as a columnist, with the help of the publishers who provided the resources, I conducted 22 surveys of readers and experts in liquid chromatography (LC) columns (12 total), sample preparation and automation (six total), gas chromatography (GC) columns (three total), and thin-layer chromatography (one total). Out of the 22 surveys, six of them addressed experts in the field but the remainder were sent to practitioners. The survey cited here was the last expert survey that we conducted and involved a good cross section of current LC column experts: old and young alike; academic and industrial; European and North and South American. Since the “hot” new columns for high performance liquid chromatography (HPLC) and ultrahigh-pressure liquid chromatography (UHPLC) had been on the market for some time, the sub-2-µm totally porous particle (TPP, 2003) and the new breed of superficially porous  particle (SPP, 2008) experts had a chance to reflect on the contributions made by them on chromatographer needs. The experts were asked about the biggest contributions in the last decade, whether these columns are meeting the needs of the users, where this technology is heading, the role of nano and capillary columns, whether instruments are ahead or behind column developments, and the biggest problem facing column users today. The survey also gave the experts a chance to make a parting statement. Besides the SPP versus TPP controversy, a number of excellent suggestions and concerns were brought up by the experts and the article serves as interesting reading two years later.

ABSTRACT

High performance liquid chromatography (HPLC) and ultrahigh-pressure liquid chromatography (UHPLC) stationary-phase technology have arrived at a crossroads, with several competing approaches vying to become the standard column type for use in future routine separations. To understand the path forward of sub-2-µm porous particles, superficially porous (core-shell) particles, silica monoliths, and polymeric monoliths, a survey of 14 leading HPLC–UHPLC column experts was conducted in June 2013. A series of questions was directed to the experts, and this report serves as a summary of their inputs on recent column advances, deficiencies and problems, instrument capabilities, and future directions of the technologies.

More than 25 years ago (1), when high performance liquid chromatography (HPLC) was in its teenage years and many chromatographers were unsure of which way column technology was going to go, I conducted a survey of leading experts at the time in column technology. The survey was entitled “The Future of HPLC Column Technology: A Survey of Experts” and predictions were made. Then, 20 years later in 2007 (2), I revisited these predictions, discussing where the experts were “right on” and where they missed the boat. Rather than repeating those earlier observations and comparisons, the reader is referred to those original publications. 

Today, I believe that liquid chromatography (LC) column technology, particularly in stationary-phase formats, is at another crossroads. Since the original survey (1), porous particles have become much smaller; now sub-2-µm porous particles are available from more than 30 companies. As a result of these smaller particles, column pressure drop has become a bigger issue. Ultrahigh-pressure liquid chromatography (UHPLC) has now become a household word with instruments capable of pressures of nearly 20,000 psi. Superficially porous (core-shell) particles (SPPs), previously referred to as pellicular packings, mostly larger in diameter than the sub-2-µm particles, have come into their own as lower-pressure-drop replacements for the smaller sub-2-µm columns but with similar efficiencies. Silica monoliths are now into their second generation and show better efficiency, but at the expense of greater pressure than the first-generation monoliths. Polymeric monoliths are starting to show some promise for both large and small molecule separations and offer greater chemical stability than silica-based packings. 

At the HPLC 2013 conference in June, I connected with a selection of today’s experts, and most agreed to respond to a questionnaire on the subject of HPLC column technology. This time I decided not to include experts from companies specializing in column development because their answers were somewhat biased in the first survey and some were reluctant to express their predictions on future column developments as that might divulge proprietary product development information. This installment of “Column Watch” is a summary of the responses to those questions and presents not only a current status report on LC stationary phases, but also a look into the future. In this survey, the “column” refers to the stationary phase inside the hardware and not necessarily to the hardware itself. 

<strong>What do you feel has been the biggest contribution to HPLC–UHPLC column technology in the past 10 years?</strong>


With agreement among most of the experts, the development of new particle technologies was, by far, the biggest contributor to column technology. Surprisingly, out of these technologies, the sub-2-µm porous particles were cited as the most significant development. Besides offering an order of magnitude in column efficiency over the popular 5-µm columns of the 1980s and 1990s, sub-2-µm particles kicked off a whole new type of LC instrument that was not only capable of reliable flow rates at high-pressure operation but had decreased extracolumn (band dispersion) contributions than previous 400-bar instruments, which were originally developed for 5-µm microparticulate columns. In fact, three of the experts thought that the instrumentation developments were key to the acceptance of the higher efficiency sub-2-µm and SPP columns and that both went hand-in-hand to bring HPLC–UHPLC to where it is today. In particular, Gerard Rozing, retired from Agilent Technologies and now a consultant in the field, felt that practitioners “. . . could hardly imagine the engineering task to provide pumps that deliver mobile phase at 1000+ bar reliably.” He also cited breakthrough designs in sealing materials, embedded software for instrument control, flow cells, capillary connections, and experimental aspects like frictional heating and the pressure effects on physical and chemical properties of solutes that were irrelevant in the 400-bar realm that needed to be dealt with. 

Nevertheless, the general opinion among our panel is that SPP columns may, in the long run, replace the totally porous particle columns for routine use. Already, we have seen both smaller (d_p≈ 1.3 µm) and larger (d_p≈ 5 µm) SPP columns come into the market, supplementing the more popular 2.6–2.7 µm products. Despite this, Pat Sandra of the Research Institute for Chromatography in Belgium is not entirely convinced of the robustness of SPPs with diameters less than 1.5 µm. The reduced frictional heating observed for SPPs relative to sub-2-µm particles by Fabrice Gritti and Georges Guiochon of the University of Tennessee may also prove useful for smaller SPPs. Frictional heating arises when small particle columns are operated at very high pressure and flow rates and can cause undesirable retention and reproducibility problems. Gritti also pointed out that to get effective performance out of 1.3-µm particles, higher flow rates and pressure requirements are needed to achieve the optimum linear velocity, further emphasizing the importance of making sure that these small-particle columns are rugged and reliable.

Monoliths were hardly mentioned by our panelists, although Gritti did mention that the “. . . second generation of silica monolithic columns should not be left to oblivion.” Recent advances in silica monoliths demonstrated at HPLC 2013 showed structures that are more uniform from the column center to the wall and can intrinsically generate up to 600,000 plates/m. But to reach these efficiencies, Gritti feels that new hardware must be designed to minimize band dispersion and the adhesion between the silica structure and the column hardware wall must be improved to withstand higher pressure (up to 500 bar). He expects another breakthrough in the future for silica monolithic columns. Davy Guillarme of the University of Geneva, University of Lausanne in Switzerland felt that there was a breakthrough 10 years ago, but there is “still not enough choice in terms of column dimensions (length and diameter) and, above all, chemistries, for the monoliths to succeed.”


Do you feel that the present generation of HPLC–UHPLC analytical columns for small-molecule separations is meeting the needs of chromatographers today? If not, where do you feel that there are deficiencies?

In general, the experts agreed that for the most part the columns of today are meeting the needs of most users. John Dolan, a seasoned LCGC columnist and founder of LC Resources, believes that for a pharmaceutical audience “. . . the current offering of columns fills almost all the needs.” But Michael Dong of Genentech finds that highly basic molecules still continue to present peak shape challenges “. . . particularly under high loading conditions.” Sandra noted that his laboratory no longer has problems with basic compounds, but acidic compounds are creating problems now.

Several of the experts mentioned that, in the absence of predicting phase selectivity, chromatographers will always need more theoretical plates to improve their resolution. Gert Desmet of the Vrije Universitat Brussel in Belgium pointed out that “. . . method development skills are declining in the everyday lab . . .” and this is another reason that increased efficiency will still be a driving force in column development. In the laboratory of Mary Ellen McNally of Dupont’s Crop Protection Division, they are using higher temperatures and pressures and, therefore, want columns that are rugged under these conditions.

Both Sandra and Guillarme said there is a need for better quality 1.0-mm i.d. columns. Sandra brought up the 1987 survey in which the future use of microbore columns was a topic of interest and that they still are today, with not many improvements to talk about in 25 years. He feels that 1-mm columns should become more important not only for “exotic” solvents, but for solvent savings in general. Guillarme’s point was more related to performance. He would like to see 1-mm (and 2-mm) columns perform equally as well as 4.6-mm i.d. columns. It is generally well known that these smaller-diameter columns do not show the expected efficiency when packing the same stationary phases, whether they be SPPs or sub-2-µm particles. Dong was a bit more conservative in his thinking and sees smaller-diameter columns in the future, but feels that 3 mm may become the standard internal diameter taking over from today’s 4.6-mm i.d. columns. Paul Ferguson of Astra Zeneca brought up the need for SPP columns that have both high pH stability and high mechanical strength.

Certain classes of compounds are still presenting challenges to separation chemists. In particular, Dong and Guillarme both mentioned chiral compounds for which method development is still a “hit or miss” encounter. Guillarme added that zwitterionic compounds and very hydrophilic compounds cannot be easily handled by hydrophilic interaction liquid chromatography (HILIC), even though this technique has made great strides in handling compounds unretained or poorly retained by reversed-phase LC.

Looking into the future, what direction do you see column technology going in the next five years? 

Rozing expects the “quest for more separation power” to continue and likewise most of the other experts predicted that smaller-diameter particles for both sub-2-µm particles and especially SPPs will be the most likely development, particularly particles with d_p smaller than 1.5 µm. The expert panel felt that more stationary phases, such as higher temperature and pressure-resistant phases suggested by McNally, are needed for SPP to become the favored column type. Furthermore, Dong, Guillarme, and Gritti all expressed a need for hybrid phases with a pH range of 1–11, something currently lacking in the marketplace.

Sandra agreed that more unique stationary phases will become available, but what he would really like to see is a reduction in further proliferation of these phases to just a few that provide the efficiency, selectivity, lifetime, and reproducibility that the majority of chromatographers use. Ferguson suggested we go beyond reversed-phase chromatography and mentioned that improved mixed-mode and HILIC columns are needed along with more polar phases specifically designed for supercritical fluid chromatography (SFC) “. . . where separation mechanisms are probably less well understood than in other separation areas.” Joseph Glajch, one of the original interviewees in the 1987 survey (1) and now at Momenta Pharmaceuticals, agrees with Sandra that with all the advances made in the past few years, a consolidation of these column developments is in order and sees this happening over the next five years or so.

Several predicted that further instrument development with respect to reductions in band dispersion will play a key role in the further reduction in particle diameter of both SPPs and sub-2-µm particles. Gritti suggested improvements in frit and endfitting designs is needed. Dwight Stoll of Gustavus Adolphus College went one step further and suggested that to achieve these expected reductions, the column should be better integrated with the instrument hardware as has occurred with microfluidic formats. Such an integration may take a while and may not be so popular  “. . . because moving to an integrated platform dramatically reduces the flexibility of operation that chromatographers really like to have,” Stoll said. Ferguson and Gritti suggested that improvements in column hardware design for both particulate and monolithic columns could also provide better performance. In particular, they suggested that the design presented at HPLC 2013 by Andrew Shalliker of the University of Western Sydney in Australia (3) termed “active flow technology” might be worthy of consideration. In this paper, describing a virtual wall-less column, Shalliker suggested that if one samples only the central core of the parabolic flow profile and splits the flow at the column outlet (using specially designed endfittings), the gap between the performance of real columns and the infinite diameter column suggested years ago by John Knox would be narrowed. Shalliker showed improvements in efficiency for both SPPs and sub-2-µm particles compared to normal endfittings used on most column configurations. Rozing suggested that, unlike gas chromatography (GC), HPLC has still not reached its theoretical performance. There is still enormous room for improvement, and one way to achieve it has been demonstrated by Karger and coworkers (4) with their high efficiency porous layer open tubular (PLOT) columns. These columns could prove attractive for biological analyses in which the amount of sample is very limited.

Matthias Pursch of Dow Germany believes that silica monoliths may still command attention, especially if plate counts of the next generation continue to increase and more phases become available. The trade-off between available plates and generated pressure will continue to confound monolith developments. No mention was made of polymeric monoliths by any of the experts.

If you had to choose one of these technologies, where would you put your efforts: smaller porous particles (d_p< 1.5 µm), superficially porous particles (d_p< 1.5 µm), silica monoliths, or polymeric monoliths?

This question was less open-ended to force the respondents choose among the four distinct research areas that seem to be in the current domain of possibility with some further research and potential commercial development. There was a mixed bag of responses, which were somewhat influenced by the degree of familiarity of the respondents with respect to the technology. Of the four technologies, silica monoliths generated the most positive response, most likely because commercial development is further behind the SPP and sub-2-µm phases because silica monoliths were caught up in intellectual property (patent) concerns and thus very little competitive development took place for a long time. So in a way, a deeper understanding and further development of monoliths and thus increased efforts would result in greater strides than investing more time in studying SPP and sub-2-µm technology. Polymer monoliths, which are also less developed, weren’t mentioned much in the responses. 

The real question to ask, said Rozing, is “How may plates do I need and how long am I willing to wait for delivery of these plates?” Desmet has investigated this question very thoroughly by looking at kinetic plots where columns can be compared based on separation time, available pressure, and efficiency required (5). In fact, in his response, Desmet indicated that, based on the smaller impedance of monolithic columns, in the long term, these columns might turn out to be the best format provided that the future monoliths have much smaller and much more uniform domain sizes than is currently achievable. Gritti believes monoliths would also be useful if they could withstand higher pressures (500 bar) and with a domain size of 1.5 µm and the right column hardware, could generate 750,000 plates/m. Guillarme advocated the development of a very long monolith capillary column of several meters with a format similar to that used in capillary GC. Rozing noted that Tanaka and colleagues have been able to generate more than a million plates (albeit in a longer time period, ~530 min) using long silica monolith columns (6). Other experts are not keen on putting more efforts into monoliths based on the slow progress shown to date, poor implementation of the technology, and limited acceptance by most chromatographers.

Surprisingly, Desmet pointed out that a more straightforward outcome of the kinetic plot is that it doesn’t pay to go to even smaller porous or superficially porous particles sizes, but to settle on something in the 2–3 µm range but with very long lengths (coupled columns) with pressure ratings up to 2000 bar. Some form of cooling would be needed to abate the viscous heating problem. Desmet feels that “. . . efficiencies of a few hundred thousand plates could be achieved with a total analysis time on the order of 10 min.” Such a large plate number in a short time would make method development much easier, analogous to GC separations. Others feel that further work on getting more out of SPP and sub-2-µm columns, provided instrumentation can meet the needs, would be worth the effort. The preparation of submicrometer SPP and sub-2-µm columns for large molecules could bring some advantages at moderate pressures since the flow rates above the optimum linear velocity could be used advantageously. Stoll believes that submicrometer particle columns may approach the self-assembled submicrometer beds being studied by Mary Wirth and coworkers (7), whose columns are intended mainly for very high efficiency biomolecule separations. Guillarme thinks that the frictional heating problems with particles with diameters smaller than 1.5 µm may necessitate using smaller-diameter capillary columns.

Sandra felt that taking today’s SPP and sub-2-µm particles and packing them into 1-mm i.d. columns with resultant solvent savings would be more beneficial than reducing particle diameters further. He reminded us that there is only a gain in resolution proportional to the square root of the reduction in particle size. Besides the solvent savings, techniques like two-dimensional tandem liquid chromatography (2D LC×LC) can greatly benefit from small-diameter columns.

Do you feel that column technology is ahead of or behind instrument capability, and why?

An overwhelming majority of the experts feel that column technology is ahead of instrument technology. However, their supporting comments showed that they are mostly thinking of the contribution of instrumental band dispersion of currently available instrumentation. Certainly for low k value analytes in short- (<50-mm), small-internal-diameter (2 mm and below), small porous particle (<2 µm), and SPP (~2.7 µm and less) columns have difficulty in providing the expected performance when peak widths are a fraction of a second wide and peak volumes are 1 µL or lower. Dong stated that in his laboratory “UHPLC, in general, is barely able to keep up with 2-mm columns in terms of extracolumn band broadening under isocratic conditions.” Being the most aggressive, Gritti feels that system dispersion of modern UHPLC instruments must decrease by an order of magnitude to further advance column technology. Stoll, Gritti, and Desmet reiterated their belief that integrated columns should take the place of today’s endfittings (including frits), a source of band broadening. Dolan feels that in the real world, many laboratories still use older instruments with high system dispersion and aren’t ready to take advantage of these newer columns. “We need lower dispersion detectors for all LC detectors, not only just UV,” he added. Stoll feels that in his 2D work, where sub-1-min separations are needed in the second dimension, both columns and instrumentation come up short in performance with neither being optimized at the present time.

However, in terms of other instrumental parameters needed for modern columns, such as pressure capability and detector speed, instruments are for the most part ahead of columns, a point made by several of the experts. Only in the smallest particle columns are pressure limits compromised. As Guillarme points out for the commercial 1.3-µm SPP columns “. . . current LC instruments are too restricted in terms of the upper pressure limit as a ΔP_max of 2000 bar will be required.” With the best low dispersion instruments on the market, Sandra’s laboratory can achieve reduced plate heights of 2.2, even for 1-mm i.d. columns. Unfortunately, these columns show poor reproducibility. 

A few of the experts made a very valid point that columns will always be ahead of instruments. Development of a new column can be achieved quickly relative to the time (2–3 years) it takes to come up with a new instrument or detector. Rozing summed it up nicely. “Instrument manufacturers will adapt to new column technology when they see a large enough market potential that will deliver a positive return on investment in a short period,” he said.

Nano (50 µm and below) and capillary columns (50–200 µm) have always been thought to have great potential, yet our surveys show a very low adoption rate for these smaller-volume columns. Do you think that such columns have a future in liquid chromatography?

Our experts agreed that nano and capillary columns will have an advantage when the sample size is limited and one is working with concentration-sensitive detectors. Thus, these small column types will mainly serve in niche applications, such as in proteomics and other “omics” and, perhaps, in toxicology studies. Rozing feels these columns are best interfaced with mass spectrometry (MS), where ultralow flow rates provide enhanced ionization and reduced ion suppression. Specialized instruments to date, particularly the chip-based instruments, appear to have the best potential in this market. However, so far only a few instrument companies have chosen to invest in this market and all have proprietary hardware impeding easy interfacing to other MS systems, much to the dismay of users. The nano and capillary instruments have never made it into the quality control (QC) environment, which also limits the market size for other instrument companies to enter into the nano or capillary space. Dolan stated that such instruments are “not for routine work in the average lab.” From talking to big pharma, Stoll doesn’t feel like they are ready to “. . . seriously invest in nano and capillary technology and thus, we will not see a big move in this direction.”

The instruments that can interface with nano and capillary columns must be fully optimized with extremely low band dispersion and pumps capable of accurate delivery at nanoliter-per-minute flow rates. The biggest problem noted by several of the experts is the lack of robustness and stability of these smaller-diameter columns. Ferguson noted that “much more care is required in aspects of column handling and installation.” McNally feels that to be acceptable, the columns “need to be in a more durable format” such as “encased in a cartridge that can be interchanged in a very easy way.” Sandra explained further. “Robustness issues limit their use to a research environment,” he said. “[There is] no future for small molecule analysis and for applications in which SOPs and validation are of utmost importance.” All in all, most of the experts felt that this market will continue to have a limited future.

What is the single biggest problem with HPLC–UHPLC columns today?

There were a lot of opinions on this topic, but the one that surfaced the most was the proliferation of HPLC–UHPLC columns on the market. “Selecting the right column must be a nightmare for neophytes in LC,” said Sandra. Others had similar feelings, but some actually personalized it to their own laboratory. Glajch mentioned his biggest problem is “the wide diversity of columns, many of which provide marginal improvement over other columns which should be similar.” McNally felt that a universal index of LC columns should be adopted to allow users to select the optimum column for the job at hand. She suggested that manufacturers could test a universal and standardized set of probes on their columns that help users select the best column for their application. The Snyder–Dolan hydrophobic subtraction model (8) attempted to do this, but her feeling was that it never became so user friendly that it could help with the initial purchase. Ferguson stated that because of the plethora of phases, diverse methods on a large variety of column types and dimensions are often developed for drug candidates on columns that may be difficult to obtain in certain parts of the world, adding to the costs and difficulty of method transfer.

The second biggest issue was the age-old problem of lot-to-lot variation in column performance. Pursch seemed to feel it was a universal problem whereas Dong said it was a problem of specific types of phases such as some polyfunctional C18 bonded phases and polar-embedded phases. In Dolan’s experience, the problems in the “reproducibility of specialty columns (for example, chiral, multimode, and so on)” was noted but felt that improvements will come with time. However, users generally feel that these columns should perform as well as alkylsilica columns and they don’t yet. “For complex samples (such as biosamples, metabolomics), we still observe batch-to-batch variability, creating problems for critical pairs,” said Sandra. “For example, UV cannot be applied and MS has to be used to differentiate these pairs.” 

For certain providers, column lifetime of UHPLC columns compared to HPLC columns is a problem according to Guillarme, and Desmet feels that the pressure stability of the UHPLC columns is not up to par with current HPLC columns. Column hardware design (frit and endfittings) is a hot button for Gritti who feels that there is room for improvement since “this hardware controls at least two-thirds of the plate height above the optimum velocity for small molecules.”

Stoll mentioned other problems that remain, including lack of high efficiency columns with high pH stability and poor mechanical stability of intermediate-bore columns (such as 1 mm i.d.) in agreement with earlier concerns by Sandra. Cost was brought up as a problem by two experts, but Dolan disagreed. “Columns that last greater than 500 injections have a trivial (such as <1–2%) contribution to the cost of analysis,” he said.

Even though their biggest problems were noted, in general, the experts felt that today’s columns are a definite improvement over earlier generation columns. The columns are generally physically and chemically stable, reproducibly packed, and exhibit great performance. “Chromatographers are familiar with phases, underlying base materials and . . . the operating ranges in which columns can be employed and how to maximize chromatographic resolution,” said Ferguson.

Do you have any parting comments?

The experts were asked to provide parting comments on column topics or anything that they felt would add to the survey. Some had no comments, but others were quite prolific. For example, in comparing developments in HPLC–UHPLC to the field of DNA sequencing, Dong mentioned terrific advancements in DNA sequencing (100,000-fold) whereas HPLC struggles to obtain 3–5-fold performance enhancements. Ferguson also believes that significant chromatography developments are not keeping pace with other analytical techniques such as MS. “Potential advances in ion mobility MS could result in less need for complex separation capabilities like 2D chromatography,” said Ferguson. In his field (pharmaceuticals), he is seeing more instances “. . . where tests that were previously undertaken using chromatography are being replaced by spectrophotometric and spectrometric techniques (such as dissolution testing and raw material confirmation).” Does this mean that separation science will be less important in the longer term?

Gritti believes that system integration is where chromatography should be proceeding. “We should never separate column from system technologies,” he said. “Both should progress at the same rate.” Furthermore, with more high efficiency miniaturization approaching, he feels that the opportunity exists to revisit the entire HPLC–UHPLC system design and sample preparation–injection–separation–detection–data collection “should be integrated into the same unique block to minimize losses of resolution and sensitivity due to the current empty connecting tubing.” Along these lines, Rozing believes that instrument manufacturers should be working on autonomous HPLC–UHPLC systems. The system will know what LC method a certain sample requires, will select the column, mobile phase, run parameters and even “check results for plausibility and report errors, warnings, and inconsistencies.” With the recent trend in decreased chromatography skill level of operators, such a system would relieve users of becoming chromatography experts and allow them to work on other tasks in the laboratory that can better utilize their skillsets.

Although stationary-phase chemistry was not specifically queried in this survey, which dealt more with stationary-phase formats, many of the experts were impressed with the improvements noted in stationary phase chemistries in recent years. “We hardly ever experience problems for analysis of most compounds (bases and some acids) at high pressure and elevated temperature,” said Sandra. “More and more selectivities are introduced but they are hardly unique and do not help to solve problems. No way that we can evaluate 20 different chemistries for each new application.” 

Although Lloyd Snyder of LC Resources and Carol Collins of UNICAMP did not actively participate in the survey, both brought up the fact that the subject of new columns of differing selectivity was not queried. Following the caution of Sandra, both agreed that presently there are more selectivity options available than people need or are using. In reality, most people are using a limited number of phases. “GC saw a similar pattern in its history,” said Collins. “In the 1970s, one could get over 300 different phases for packed column GC. Then came along fused-silica capillary columns where initially there were three to six phase options and after a vast proliferation of phases, today one can separate most samples on just a handful of phases.” In her view, with a couple of decades delay, the same scenario is playing out in HPLC–UHPLC. Unlike GC, the mobile phase in LC provides a great deal of variation in influencing selectivity. “The problem generally is that searches for a better column selectivity are made either by trial and error or by choosing columns of different functionality,” said Snyder. “There is considerable variation of selectivity within a given class of columns that is generally unappreciated.” All of these observations “suggest that use could be made of available software for comparisons of column selectivity,” he added.

Earlier, a somewhat similar suggestion was made by McNally. Stoll agrees that a tremendous amount of selectivity information already exists but it is not localized and compiled in a way that is accessible to a large number of people. He relates it to his own area of specialization, 2D separations, in which it is difficult to suggest complementary phases that are likely to be effective for a given separations problem because selectivity data are too scattered and need to be compiled in a way that nonexperts can understand. Rozing also believes that a comprehensive map of column selectivity like the solvent selectivity triangle (9) is lacking. “Peter Carr’s work (10) should be adapted by manufacturers to allow users to make reliable choices of columns that are equivalent to the one that they are using in a particular analytical separation,” he added.

On a separate note, Sandra suggested that “the green character of analytical SFC should not be overemphasized (and even should be questioned) since on commercial SFC instrumentation 2-mm and smaller internal diameter columns with sub-2-µm particles (with their intrinsically lower solvent consumption) cannot be used.” Thus, chromatographers using this technique are forced to use large-diameter columns requiring higher flow rates and therefore greater solvent consumption of organic solvent-modified carbon dioxide.

Summary

Overall, the experts feel that column efficiencies have come a long way in the past 10 years, with sub-2-µm stationary phases paving the way followed by the superficially porous (core-shell) particles. Experts predicted that in the near future the core-shell particles may become the standard analytical column, but if current problems are solved for silica monoliths they could be optimal for complex samples requiring high plates but at long separation times. The debate about time, pressure, and efficiency compromises continues. Interestingly, intermediately sized particles (2–3 µm) operated at high pressures in long columns could be the best compromise in efficiency and speed if frictional heating problems can be abated. Overall, in terms of band dispersion, experts feel that columns are ahead of the present generation of instruments, which are mostly adequate for high pressure and fast peak operation. The biggest problem facing users is the proliferation of stationary phases, which makes choosing an optimal column difficult, especially for neophytes. A single source database or selectivity triangle that would help users decide on an optimal selectivity for a given separation was suggested. After all these years, lot-to-lot variation still plagues users, especially in the bio world and for certain specialty columns. Nano and capillary columns got little endorsement by the experts.

Acknowledgments

The author would like to thank the following team of experts for their thoughts on the present and future of HPLC–UHPLC column technology:

• Professor Carol Collins, UNICAMP, Campinas, Brazil

• Professor Davy Guillarme, University of Geneva, University of Lausanne, Switzerland

• Professor Gert Desmet, Vrije Universiteit Brussel, Brussels, Belgium

• Dr. John Dolan, LCGC columnist and LC Resources, Walnut Creek, California

• Dr. Michael Dong, Genentech, South San Francisco, California

• Dr. Paul Ferguson, Astra Zeneca, Macclesfield, United Kingdom

• Dr. Joseph Glajch, Momenta Pharmaceuticals, Cambridge, Massachusetts

• Dr. Fabrice Gritti, University of Tennessee, Knoxville, Tennessee

• Dr. Mary Ellen McNally, E.I. Dupont de Nemours, Newark, Delaware

• Dr. Matthias Pursch, Dow Deutschland, Rheinmuenster, Germany 

• Dr. Gerard Rozing, Rozing.com, Karlsruhe, Germany

• Professor Pat Sandra, Research Institute for Chromatography (RIC), Kortrijk, Belgium

• Dr. Lloyd R. Snyder, LC Resources, Walnut Creek, California

• Professor Dwight Stoll, Gustavus Adolphus College, St. Peter, Minnesota

References

(1)    R.E. Majors, LCGC5(6), 454–462 (1987).
(2)    R.E. Majors, LCGC North Am.25(8), 692–702 (2007).
(3)    R. Shalliker and H. Ritchie, Paper OR29, “Active Flow Technology Presenting the ‘Infinite Diameter Column’ in a Curtain Flow Mode of Operation,” presented at HPLC 2013, Amsterdam, The Netherlands, 2013.
(4)    B. Karger, Paper KN33, “The Integration of Liquid Chromatography and Mass Spectrometry for Ultratrace Analysis,” presented at HPLC 2013, Amsterdam, The Netherlands, 2013.
(5)    G. Desmet, D. Clicq, and P. Gzil, Anal. Chem.77, 4058–4070 (2005).
(6)    R.E. Majors, LCGC North Am. 25(9), 920–942 (2007).
(7)    B. Wei, B.J. Rogers, and M.J. Wirth, J. Am. Chem. Soc.134, 10780–10782 (2012). 
(8)    L.R. Snyder and J.W. Dolan, LCGC North Am.20(11), 1016–1026 (2002).
(9)    Y. Zhang and P.W. Carr, J. Chromatogr. A1216, 6685–6694 (2009)

How to cite this article

R.E. Majors, LCGC North Am.31(8), 596–603 (2013).