Early adoption of high-pressure systems and sub-two-micron columns was what one would expect from a “disruptive” analysis platform. Finding doubters was not too difficult, particularly among Waters’ competitors. But eventually even the most vociferous critics joined the bandwagon with their own versions of high-pressure, low-particle-size stationary-phase liquid chromatography systems.
“We’d like to think that UPLC was one of the more innovative developments in laboratory analytics,” says Bill Foley, senior director of separations product management for Waters. “The label disruptive was fairly accurate. UPLC changed how customers process samples; it improved productivity and provided more and better information about samples than conventional HPLC.”
After more than ten years of continuous investment and improvement, UPLC and UHPLC have branched out to include supercritical fluid chromatography (e.g., Waters’ UPC2® or convergence chromatography platform), polymer analysis and nanoscale and microscale LC analysis. Column chemistries suitable for ultrahigh- pressure LC have broadened to include size exclusion and gel permeation. Detector options, including Waters’ innovative ACQUITY QDa detector, have similarly multiplied.
Waters took significant risks in rolling out UPLC. “We placed a big bet on the technology,” Foley says. “Its adoption in key industries strongly suggests that this is the current state-of-the-art, and the future, of liquid chromatography.”
For Waters UPLC has grown into a family of products and a multiplicity of applications and settings. For example, the UPLC-based PATrol™ process analyzer duplicates in real time the capabilities of hours-long sampling and analysis of bioprocesses, to allow real-time decision-making during cell cultures.
Hardware and process choices
Many choices in LC, for example balancing column life against the time and costs of sample preparation, come down to economics. Estimating the costs of sample prep at 50 cents per sample, and of columns at $500, the break-even is at approximately 1,000 injections (preparation time not included). Trap columns shift the economics slightly in favor of trap and elute, as a trap column costs about $200. Clearly each lab must calculate relative economic benefits and costs for its own workflows, according to Phillip DeLand, global LC business manager at Bruker Daltonics (Freemont, CA).
For example, in environmental and food testing, which consist of “difficult” matrices, operators load the matrix onto a trap column, retain analytes of interest, wash away the background, and elute target analytes onto a separation column to reduce background and with it limits of MS detection.
One Bruker customer has been able to shoot human bodily fluids directly into the LC without sample prep or even a trap column. “You wouldn’t even think of doing that several years ago,” DeLand says. “But today, with the robustness of columns, you can perform that kind of analysis with no loss of resolution, even after one thousand injections.”
Despite years of experience from the vendor and user perspectives, no go-to column yet exists for SFC that approaches the utility of C18 columns in reverse-phase HPLC and UHPLC. “Even for chiral separations, there still isn’t one SFC column users can rely on as a first choice. A lot of research has gone into finding one or two columns with comparable selectivity to C18,” says D.J. Tognarelli, chromatography product specialist at JASCO (Easton, MD).
This drawback becomes even more acute as pharmaceutical industry chromatographers look to SFC for non-chiral separations to augment the technique’s superiority for separating chiral compounds. “Achiral separations are where SFC is really trying to catch up with conventional LC,” Tognarelli adds.
Since SFC most closely resembles normal phase separations, analysts usually begin their search for an appropriate separation medium with a silica column, but as Tognarelli notes, silica is nowhere nearly as selective, or as capable of modification to suit particular situations, as C18 is. It is therefore not unusual for analysts to have as many as six columns each for chiral and achiral SFC separations and to have to run through several to optimize their separations.