ASK A DOZEN CHEMISTS about the forces that drive innovation in high-performance liquid chromatography, and nearly all of them will agree: When it comes to chemical separations, the name of the game is speed.
“Nowadays, it comes down to being able to run as many samples as possible in a given amount of time,” says Dennis D. Blevins, a senior manufacturing chemist at Agilent Technologies, an analytical instrument maker based in Santa Clara, Calif. Pointing to the pharmaceutical industry’s ever-growing and impatient hunger for data, Blevins says today’s chemists are called upon by their employers to analyze a larger number of samples???often mixtures of active drug compounds, impurities, and synthesis intermediates???more quickly than ever before.
“The most critical success factor is speed to market,” says Jeremy B. Desai, executive vice president for R&D at Toronto-based Apotex, Canada’s largest generic drug manufacturer. Because chromatographic analysis plays a central role in every phase of drug development—from exploratory research to clinical testing to releasing commercial batches—reducing analysis times can mean the difference between a product’s success or failure, Desai says.
Industry’s pressing need for faster separations coupled with other researchers’ calls for greater separating power have prompted instrument manufacturers and other analysis experts to push the limits of high-performance liquid chromatography, or HPLC, where the P often can also stand for pressure.
Every area of chemistry could benefit from more powerful analysis methods. But the enormous popularity of HPLC has kept the technique in the cross hairs of instrument developers for years. The tool is so ubiquitous in today’s laboratories, according to chromatography expert John G. Dorsey, that “liquid chromatography instrumentation now ranks third behind analytical balances and pH meters in number.” Dorsey, a chemistry and biochemistry professor at Florida State University, adds that “every major pharma company has an LC instrument on virtually every bench.” No surprise there, he says, because most analytical tools—mass spectrometers, for example—don’t handle chemical mixtures very well.
Efforts to devise a new generation of HPLC tools have led to new types of column-packing materials, some of which have been commercialized recently; a new generation of instruments with advanced capabilities; and new HPLC applications. Overall, the demands for faster and more efficient chemical separations have guided innovators toward the extremes of particle size, pressure, temperature, and other chromatography parameters.
IT’S NO SURPRISE then that last month at Pittcon in New Orleans, “extreme” and “ultra” appeared in the titles of several symposia, workshops, and presentations on HPLC. The sessions gave researchers an opportunity to compare notes on the performance of new commercial and noncommercial instruments that operate under unconventional conditions. The sessions also provided chromatographers with a chance to take stock of the big changes that have taken place in HPLC in the past few years.
Some of the most talked-about changes are related to the sizes of the tiny particles that fill the inside of a chromatography column and the pressures needed to push liquid through such a column. Until recently, “in terms of capabilities and performance, liquid chromatography remained largely unchanged since the early 1970s,” says Douglas R. McCabe, a marketing manager at Waters Corp. In the ‘70s, HPLC columns were packed with irregularly shaped particles of roughly 10 µm in diameter (the stationary phase). A solvent (the mobile phase) containing the various components to be separated was forced through the packed column with a pump operating at up to 6,000 psi.
Instrument designers made incremental improvements over the years, McCabe says, but with the introduction of Waters’ Acquity UltraPerformance Liquid Chromatography (UPLC) system in 2004, the field was hit with what he terms “disruptive technology.”
The UPLC instruments, which were designed to provide faster and higher quality separations than conventional HPLC units, feature columns packed with highly spherical, uniform-sized silica particles measuring less than 2 µm in diameter. In addition, because smaller particles lead to more tightly packed columns with greater resistance to liquid flow, the mobile phase is driven through the columns at pressures of up to 15,000 psi.
In the few years since Waters introduced its UPLC instruments, a number of other manufacturers, including Agilent and Shimadzu, have also developed next-generation LC systems.
In one demonstration of Acquity’s enhanced performance, Waters chemists showed that a mixture of acetaminophen, caffeine, and a few other compounds could be separated into its components by a conventional HPLC system with 5-µm column-packing particles in just under seven minutes. But with the UPLC system’s smaller particles and greater operating pressure, they showed that the same mixture could be separated with greater resolution (sharper peaks) and increased sensitivity in just over one minute.
Other researchers report similar improvements in separations. At Eli Lilly & Co., for example, senior research chemist Todd D. Maloney says he completes a typical HPLC separation of an active drug compound and a total of six known impurities and synthesis intermediates via conventional HPLC methods in about 45 minutes. Now, using today’s smaller particles and newer instruments, he can do the equivalent separation and analysis in just under 10 minutes.
The improved chromatographic performance doesn’t only save time when running standard assays or other types of routine tests, Maloney says. It also cuts down on the time needed to optimize the separations and develop those assays and other tests.
Maloney finds that the souped-up LC instruments he’s worked with, which include units made by five manufacturers, are “well designed and very easy to use.” The new LC systems, however, are also more expensive than conventional ones. “These days, every instrument purchase is scrutinized,” Maloney remarks. “So you really need to be able to show the value in terms of return on investment.” And a surrogate measure of that value is the speed of the analyses.
At Apotex, Desai is quick to point out that value. In comparisons of HPLC with UPLC, he finds that UPLC reduces the average run time for standard analyses of impurity levels in drug compounds to one-eighth the conventional HPLC run time. And assays of the active components of drugs are completed two-and-a-half times faster by using UPLC than by HPLC.
Apotex has strong incentive to invest in the fast analytical technology. As Desai explains, in the U.S. the first company to receive approval from the Food & Drug Administration to market a generic pharmaceutical compound is often granted a 180-day exclusive marketing license. “Having that six-month advantage over the competition is huge in terms of its revenue impact on a generic drugmaker,” Desai stresses. To help secure that advantage, Apotex chemists are already using 30 Acquity systems, Desai notes. The company plans to buy at least 20 more systems this year, he adds.
ALTHOUGH the pharmaceutical industry is the heaviest user of the new technology, it’s not the only user. The new LC systems are being put to use in a wide variety of applications. For example, scientists now use them in environmental studies to measure pesticides and other contaminants in water; in food-safety tests to quantify drug residues and other impurities in meat, poultry, and seafood; and in biomedical applications to study amino acids, peptides, and other bioanalytes.
Explanations for why reducing the size of column-packing particles leads to enhanced separations come in a few forms. Classic chromatography theory, for example, shows mathematically that reducing the diameter or thickness of packing particles improves separations.
Another explanation relates the size of the porous particles to the time it takes analyte molecules to diffuse in and out of those particles. In general, small differences in the extent to which the components of a mixture interact with or are retained on the surfaces of the packing materials result in chemical separation. As a mixture flows through a column, those differences ultimately lead to a measurable separation because the analyte molecules repeatedly shuttle back and forth between the solvent and the particles, thereby interacting over and over again with the particles. Because molecules can diffuse in and out of small porous particles faster than large ones, smaller particles facilitate faster shuttling and hence more efficient separations.
Far and away, silica is the most common column-packing material for conventional- and ultra-HPLC instruments. The material, which is generally prepared from tetraethoxysilane (TEOS) or related silanes, is polymerized to form high-purity, monodisperse (same sized) silica spheres that are free from metals that would otherwise interfere with chromatographic separations. The particles are mechanically strong, have high surface area, and feature pore sizes that can be easily tailored. In addition, manufacturers customize the properties of the material for a particular application by modifying the surfaces with octyl, octadecyl, phenyl, and other types of functional groups that elicit separation-enhancing interactions between the solution’s components and the particles in the column.
Commercial examples of column-packing materials used in the new systems include Waters’ High Strength Silica (HSS) and Ethylene Bridged Hybrid (BEH) materials. Particles of both materials, which were designed for the UPLC systems, have diameters in the 1.7–1.8-µm range and feature pores with diameters of either 10 (HSS) or 13 (BEH) nm.
The materials’ compositions differ in a pivotal way. Unlike the all-silica HSS product, BEH particles contain C–C bridges between pairs of silicon atoms. The hybrid organic-inorganic particles, which are prepared via copolymerization of TEOS and bis(triethoxysilyl)ethane, were designed to retain silica’s mechanical strength while overcoming pure silica’s tendency to undergo column-damaging hydrolysis in alkaline environments (greater than pH 8). BEH’s covalently bonded Si–C–C–Si units render the hybrid material chemically stable up to a pH of 12, which is an ideal condition for analyzing some pharmaceutical agents and other types of compounds.
MAKING NEW TYPES of separation media isn’t the only challenge in developing next-generation HPLC columns. Packing the particles into a column such that the end result works well chromatographically can also be difficult. That’s one of the areas that James W. Jorgenson studies with his research group at the University of North Carolina, Chapel Hill.
“Chromatography theory tells us that we could be doing faster and more efficient separations by using even smaller particles than the ones we use now,” Jorgenson says. So rather than drawing the line at 1.7 µm, the UNC Chapel Hill team works with silica particles measuring 1.0–1.5 µm in diameter and makes use of 100,000-psi pumps to do its work.
“As the particles get smaller, they become immensely harder to pack into columns that provide efficient separations,” Jorgenson observes. He notes, for example, that with 1.5-µm particles, his group can readily make columns that do their jobs. “But by the time we get down to 1.0 µm, the results aren’t nearly as good.”
The biggest problem, he says, is the tendency of tiny particles to agglomerate. The group uses various solvents and ultrasonication to prepare well-dispersed slurries and then packs capillary columns at high pressure. They test the columns chromatographically and examine the microscopic packing structure and other properties to discover the most effective column-packing methods.
In addition to commercially available packing materials with sub-2-µm particles and materials comprising even smaller particles studied by academic researchers such as Jorgenson, the LC community is developing other types of packing materials with properties and features that make the words “ultra” or “extreme” applicable. For example, Fused-Core particles (depicted on the cover illustration of this issue of C&EN), are unusual in that they are composed of a solid (nonporous) silica core (1.7 µm in diameter) surrounded by a 0.5-µm-thick porous silica shell, amounting to an overall diameter of 2.7 µm.
Recently developed by Wilmington, Del.-based Advanced Materials Technology (AMT) and marketed by Mac-Mod Analytical, these core-shell materials were designed to provide high-speed LC separations in a durable column. Joseph (Jack) Kirkland, AMT’s vice president for R&D, says the 2.7-µm Fused-Core materials are competitive with sub-2-µm particles in terms of separation efficiencies but don’t require special ultra-high-pressure LC equipment.
Head-to-head comparisons by users are consistent with Kirkland’s claim. Maloney reported recently that Fused-Core columns offer a much lower pressure alternative to sub-2-µm columns with only “a slight sacrifice” in separation efficiency (J. Sep. Sci. 2007, 30, 3104).
Still other types of materials fit the “extreme” LC bill. According to University of Minnesota chemistry professor Peter W. Carr, zirconia makes an incredibly stable column-packing material, much more so than silica. “You can cook zirconia in strong acid and strong base and it won’t dissolve,” Carr says. In fact, his desire to do just that as a means of cleaning HPLC columns led him to help found ZirChrom Separations, a zirconia-based HPLC supplies vendor, with which he is no longer affiliated.
ALSO EXTREME are the meters-long LC columns found in Nobuo Tanaka’s lab at the Kyoto Institute of Technology. Tanaka makes one-piece, porous, monolithic HPLC columns that reach several meters in length. Although these superlong columns are unlikely to find commercial use any time soon, they can be used to separate extremely similar compounds. Demonstrating just how similar the compounds can be, Tanaka showed that benzene and monodeuterated benzene, which are nearly identical, can be separated readily on an 8-meter silica column.
“Extreme” solvent temperatures can also significantly boost LC performance. Raising the solvent temperature lowers its viscosity and in turn decreases the pressure needed to force the liquid quickly though a packed column. By using that approach together with small column particles and short columns, Carr carries out two-dimensional separations on biological mixtures of hundreds of compounds in one-tenth the time required to separate the compounds via conventional 2-D LC methods.
Instrument makers first launched speed- and efficiency-enhancing LC products that significantly outperform conventional instruments only a few years ago. Yet LC aficionados are already wondering whether future commercial instruments might incorporate some of the more “extreme” forms of HPLC now being pioneered in academic labs. For example, Jorgenson wonders whether the next generation of commercial instruments should run at 40,000 or 50,000 psi. “The jury is still out” on that question, Lilly’s Maloney says, because the “ultra” instruments are still new to the market. Jorgenson concurs: “Right now,” he says, “it’s way too early to decide.”