Separations Science

LONG AFTER technical innovations have been turned into successful commercial products, clever people continue to dream up ways to improve and refine the inventions. Automobiles, computers, and telephones, for example, have been redesigned and enhanced for decades, yet every new model boasts advantages relative to its predecessors.

The situation is much the same with laboratory tools for chemical separations. For years, analytical chemists have benefited from instrument makers' efforts to extend the capabilities of gas and liquid chromatography systems. From modest beginnings as simple analyzers that probe a relatively limited number of compounds, chromatography instruments have evolved to today's high level of sophistication, sensitivity, and automation. Nonetheless, the drive to advance the equipment's usefulness continues.

In gas chromatography, for example, researchers are developing molecular-beam methods that enhance separation and analysis of mixtures. And in the area of liquid chromatography, novel types of carbon electrodes are being developed to improve sensitivity and device durability.

For more than 50 years, researchers have used gas chromatographs coupled to mass spectrometers to detect and analyze the chemical components of mixtures. Commercial GC-MS systems are practically standard laboratory equipment in industry and academia as well as in many government and small-business labs. Despite broad popularity and generations of improvements, the powerful lab combo is still fraught with shortcomings.

"GC-MS suffers from a major Achilles heel in terms of the relatively small range of volatile and thermally stable compounds that are amenable to analysis." That assessment comes from Aviv Amirav, a chemistry professor at Tel Aviv University. Amirav, a specialist in analytical instrumentation development, has commercialized novel sample introduction devices and detectors for use in gas chromatography (C&EN, March 28, 2005, page 53).

Another limitation common to GC-MS analysis, according to Amirav, arises from the electron bombardment process typically used to ionize analyte molecules. In that widely used procedure, electrons emitted from a hot filament collide energetically with sample molecules and ionize those species by dislodging electrons. Mass spectrometry measurements based on that ionization method are straightforward and have been used to generate enormous spectral reference libraries. The trouble, however, is that the electron-bombardment technique often yields mass spectra in which the molecular ion—the unfragmented parent molecule—is indiscernible.

"Without knowledge of the molecular ion, we can never be completely confident that an unknown compound has been identified correctly," even if extensive spectral libraries are available for searching, Amirav argues. The reason for the uncertainty, he explains, is that the analyte molecule and its homologs, derivatives, and degradation products may all lead to the same or nearly the same mass spectrum.

More than 10 years ago, Amirav and his research group began addressing these issues by drawing upon the benefits of molecular-beam methodology. This is an area of experimental science that originated in chemical physics and spectroscopy and eventually came to be applied to analytical chemistry.

Following a number of investigations based on the group's earlier instrument designs, Amirav and Tel Aviv senior scientist Alexander B. Fialkov teamed up with researchers Urs Steiner and Larry Jones of Varian, an instrument manufacturer in Walnut Creek, Calif. Recently, the team developed and tested an advanced-generation GC-MS interface device that transports sample compounds eluting from a GC column to a mass spectrometer by way of a supersonic molecular beam.

THROUGH VARIOUS comparative studies, the team showed that due to the properties of molecular beams and the nature of the apparatus that generates them, their beam-based technique broadens the range of compounds suitable for GC-MS analysis relative to conventional GC-MS methods. In addition, it enhances molecular-ion signals and sensitivity and offers a number of other advantages compared with standard GC-MS methodology.

Basically, a supersonic molecular beam can be generated by seeding a carrier gas such as helium with a low concentration of sample molecules and then directing the gas mixture through a tiny orifice into a vacuum chamber. As the gas stream passes through the small opening (a nozzle) to the evacuated chamber, the lightweight helium atoms collide with the sample molecules and strip away some of the molecules' vibrational energy. The helium atoms, in turn, acquire additional translational energy.

Eventually, that process "cools" the analyte molecules, leaving them with very little vibrational energy. The upshot is that if the molecules are ionized in a state of reduced vibrational energy, they have a lower probability of fragmenting upon ionization. That increases the likelihood that molecular ions will survive the trip through the mass analyzer intact.

The molecular-beam advantage is further bolstered by ionizing sample molecules using a so-called fly-through ionizer. As its name implies, this type of electronimpact device is designed to ionize compounds in molecular beams as they fly through the ionizer while preventing the molecules from coming in contact with the device's hot surfaces. Such contact, according to Amirav, can occur as many as 50 times as a molecule bounces around or scatters from the surface of a conventional ionizer. This contact would warm the molecule and restore its vibrational energy.

The scattering process also normally broadens chromatography peaks, causing peak tailing, and promotes sample decomposition on the hot surfaces—problems that are avoided in the fly-through ionizer design.

At present, the Tel Aviv group uses a commercially available triple-quadrupole GC-MS system (Varian 1200 series) that has been modified to accommodate the molecular-beam instrumentation. To evaluate the performance of the new system, Amirav and coworkers began by focusing on aliphatic hydrocarbons. In this class of compounds, the fragmentation pattern of one member often resembles those of other members, making identification difficult.

In a proof-of-concept demonstration using hexadecane, the Tel Aviv group showed that with the molecular beam method, the analyte's molecular ion, at 226 amu, is far and away the most prominent feature of the mass spectrum. In classic electron-impact mass spectra, the molecular ion shows up as a minuscule, hard-to-discern signal. But other than the large boost in the abundance of the molecular ion, the group's mass spectrum for that C₁₆ compound closely matches reference-library spectra, such as the one published by the National Institute of Standards & Technology.

The researchers also showed that even better matches with reference spectra can be obtained (if needed for certain applications) by altering experiment conditions on the fly—for example, by briefly reducing the carrier-gas flow rate to temporarily reduce the vibrational cooling effect.

After the C₁₆ demonstration, Amirav's group moved on to larger hydrocarbons. Compounds of that type are tough to analyze via standard GC-MS methods not only because of their tendency to yield nondistinct fragmentation patterns but also due to their low volatilities. Underscoring those points, Amirav notes that for compounds with carbon chains in the range of C₂₈ and larger, molecular-ion signals are absent or nearly absent under typical GC-MS conditions (carrier-gas flow rate of 1 mL per minute). The largest hydrocarbons that can be eluted from a chromatography column under those conditions are around C₄₀.

Now, the Tel Aviv's group's method has extended that upper limit significantly. Using a commercial mixture of large linear-chain hydrocarbons in test studies, Amirav and coworkers find that with their molecular-beam-based GC-MS technique, they can separate the components of the mixture readily, and they see dominant molecular-ion signals for compounds as large as C₈₄H₁₇₀ (1,179.3 amu) (Int. J. Mass Spectrom. 2006, 251, 47).

As Amirav explains, the principal feature of the new GC-MS method that overcomes the low-volatility challenge is the high carrier-gas flow rate, which can be up to 90 times greater than flow rates used in conventional GC-MS methodology. Essentially, the fast gas flow pushes sluggish, nonvolatile molecules through the GC column. Ordinary GC-MS systems cannot tolerate the onslaught of gas, which, in addition to other problems, would bog down the mass spectrometer's vacuum system. But being that high gas flow rates are essential for producing molecular beams, the GC-MS interface in the Tel Aviv lab is designed to accommodate (and continuously pump) all that gas.

THE HIGH FLOW RATE offers another key advantage: It drives thermally labile compounds through the GC column at relatively low temperatures. That ability extends the range of compounds amenable to GC-MS analysis by including analytes that would ordinarily decompose at temperatures encountered in typical GC experiments.

As a case in point, the Tel Aviv group showed that mixtures of aldicarb, oxamyl, and other carbamate pesticides can be separated easily and quickly via their method. Those compounds are typically separated and analyzed by customized liquid chromatography procedures. The team reports similar success with steroids, explosives, and other compounds that are not ordinarily analyzed via GC-MS methods due to their sensitivity to heat.

Amirav points out that high carrier-gas flow rates, low column temperatures, and short run times can reduce the separating power of the GC portion of the experiment for some types of analytes. "There is always a trade-off between speed, resolution, sample capacity, and other properties," he says. "It's just a matter of choice because the science of separations is the art of compromise." Reducing run time, for example, may cause some analytes to co-elute from a GC column. But if the co-eluting species can be detected adequately and confidently with a mass spectrometer—for example, by identifying their molecular ions—then the time benefit can be exploited without losing separating power.

The separating power of an analytical technique can be boosted by including multiple separation phases—for example, by coupling two GC separations (GC-GC) to resolve analytes that elute together from a single column. Alternatively, two mass spectrometers can be coupled to work in tandem (MS-MS) to select certain ions for focused analysis.

Recently, the Tel Aviv team members combined their molecular-beam method with the MS-MS capabilities of their commercial triple-quadrupole instrument to explore fast separation techniques. In the MS-MS mode of operation, molecules eluting from a GC column are ionized in the molecular-beam interface and directed to the first quadrupole mass analyzer, which is set to transmit ions of just a single mass (often the molecular ions of an analyte). Those ions are fragmented in a second quadrupole by collision with a gas such as argon. The daughter ions are then steered to a third quadrupole for mass analysis.

In a demonstration study, Amirav's group found that the molecular-beam GC-MS-MS technique can be used to separate and detect the organophosphate insecticide diazinon in a fruit and vegetable extract at the nanogram-per-gram-of-produce level in less than 10 seconds. That separation is several minutes shorter than the time required by conventional techniques. For even more demanding applications, the team is exploring molecular-beam-based GC-GC-MS-MS separations.

Varian continues to collaborate with Amirav to explore and develop the technique, but for now, the instrument maker is not announcing commercialization plans. John D. Mills, Varian's scientific instruments marketing vice president, notes that the molecular-beam method appears to offer a number of advantages that may be especially beneficial for analyzing hydrocarbons by the petrochemical industry. "As with all projects at this stage of development, the ultimate commercialization of the technology will depend upon the results achieved by the research team," Mills says.

Meanwhile, on the solution-phase side of separations, chemists have used electrochemical cells as detectors for LC systems since the 1970s. The cells detect a variety of analyte molecules whose functional groups are electro-oxidized or electro-reduced at characteristic potentials. As with other analytical lab tools, LC detectors have undergone numerous modifications over the years to improve their sensitivity, reliability, and durability.

AT THE HEART of an electrochemical cell is the electrode upon which the reactions take place. Carbon electrodes are widely used to analyze various types of analytes including aromatic compounds with amine and hydroxyl groups, aliphatic thiols, and other species. Nonetheless, the electrodes suffer some shortcomings.

According to Jun Cheng, a staff chemist at Dionex, Sunnyvale, Calif., today's carbon electrodes exhibit poor electrode-to-electrode reproducibility. Furthermore, the sensitivity of individual electrodes drops markedly after just a small number of chromatography runs, forcing lab workers to try to restore electrode performance through time-consuming polishing and reconditioning of the device.

To tackle these problems, Cheng and coworker Petr Jandik have developed an inexpensive, disposable type of carbon electrode that's designed to be replaced after a few weeks of use. Unlike conventional glassy-type carbon electrodes, which are prepared commercially via high-temperature pyrolysis, the new disposable electrodes are prepared by a vacuum technique in which a thin film of carbon is deposited onto a thin polymeric material.

In an evaluation study, the Dionex team incorporated the new electrodes in flow-through electrochemical cells that were used to detect a mixture of neurotransmitter compounds, including dopamine, epinephrine, and norepinephrine, as they eluted from an LC column. The study showed that those analytes are readily detected by the new electrodes at concentrations below the nanomolar range—a two- to threefold improvement in detection sensitivity compared with standard carbon electrodes.

The team also tested the electrodes on mixtures of amino acids, metabolites, and other compounds. They report that the devices are stable and provide reproducible results from electrode to electrode.

Dionex Chief Science Officer and Vice President for R&D Christopher A. Pohl points out that, compared with conventional carbon electrodes, the disposable ones offer improved response linearity. "We believe that the improved performance is a result of the deposition process, which controls the purity of the electrode material," Pohl suggests. He adds that the new products should be generally available in two to three months.

Rather than making electrodes from thin films of carbon, researchers at Eksigent, Dublin, Calif., are developing methods for fabricating inexpensive high-surface-area electrodes based on packed beds of tiny carbon particles. Eksigent researcher Nicole E. Hebert explains that the motivation for the high-surface-area feature is the need to construct a robust device that withstands fouling from electrogenerated products that stick to electrode surfaces.

Made via microfabrication, the new electrodes, which feature picoliter-sized packed-bed volumes, can be connected in series—but controlled independently-to add resolving power to chromatographic separations. The idea is to set the potentials on the electrodes in a series to successively higher values-for example, to 100, 200, and 300 mV. In that way, if analytes co-elute from a chromatography column, then as they flow to the electrochemical cell, they can be detected selectively when they reach an electrode that's set to a suitable potential.

Hebert notes that tests based on mixtures of neurotransmitters and other compounds demonstrate the new electrodes' robustness and selectivity. She says the company is now working to reduce cell volume and improve sensitivity.

Inventions and innovations that survive the test of time eventually come to be referred to as "mature technologies." Some technology users prefer to abide by the old adage, "If it ain't broke, don't fix it." But not chemists. They know that even if technologies in separations science are mature, there's always room for improvement.

[Top of Page]