The Incredible Shrinking Mass SpectrometersMiniaturization is on track to take MS into space and the doctor's officeA quadrupole mass spectrometer the size of a shoe box? An ion trap on a chip? A time-of-flight mass analyzer with a 1 -in. flight path? As
little as five years ago, these things would
have been the stuff of science fiction. However, today they are within the realm of
possibility.One of the major forces behind the move toward miniaturization has been the space program. For the people developing mass spectrometers for space applications, size is a primary consideration. Smaller instruments are being driven by the National Aeronautics and Space Administration's mantra of "better, faster, cheaper". NASA scientists are working under the rule of thumb that an instrument should be smaller than 1 kg, occupy no more than 1 or 2 L, and consume less than 5 W--and that's for the complete system. Other areas of interest for miniaturized mass spectrometers are chemical and biological weapons detection, environmental monitoring, and medical diagnostics. "If you can make the overall device truly small, then all the power of mass spectrometry goes with you when you go out into the field," says R. Graham Cooks of Purdue University. "Basically, you put the instrument in your overcoat pocket and go from there. Obviously we're a long way from that with any commercial mass spectrometer." J. Michael Ramsey's group at Oak Ridge National Laboratory has been interested in the miniaturization of analytical devices for a number of years--mostly in the microfluidics area. "Microfluidics devices appear to offer the rare win/win/win combination of smaller, better, and cheaper at the same time," says Ramsey. "My feeling is that this will not be the case for microscale gas-phase chemical measurement devices. You can make one of these devices small and less expensive, but you will probably lose some performance." Up to this point, the miniaturization efforts have been limited to a small segment of the mass spectrometry community. "The [mass spectrometry] culture says that an instrument should do all the measurements and do them at absolutely top performance," says Robert J. Cotter of Johns Hopkins University. "Yet there is indeed a need to select particular performance characteristics related to specific tasks. I don't think the culture has pushed us in that direction yet, but it's moving there." Shrinking mass analyzers"In theory, probably just about any mass analyzer can be miniaturized, but some are better [suited for miniaturization] than others," says Cotter. Quadrupole, magnetic-sector, time-of-flight (TOF), quadrupole ion trap, and ion cyclotron resonance (ICR) mass analyzers are among those that have been miniaturized. "My theory about making small mass spectrometers is that one must determine what the objective is," Cotter adds. "The difference between a small mass spectrometer and a large mass spectrometer is that the large mass spectrometer can be designed to do many different measurements. With the smaller mass spectrometer, you must choose what you want to do." For each type of mass analyzer, miniaturization sacrifices some aspect of performance. "What do you give up?" asks Cotter. "That's why there's an issue. Otherwise, mass spectrometers would have been small all along." Other things being equal, mass analyzers lose mass resolution and usable mass range (except for the ion trap and the quadrupole, which actually have a higher maximum m/z than the larger version) as they shrink in size. For TOF analyzers, mass resolution is tied to the time resolution, which in turn depends on the length of the flight path. "You cannot simply address it by slowing the ions down," says Cotter. "If you slow the ions down, the differences in their energies become more appreciable and more important, and the resolution drops again." Nevertheless, Cotter notes, all other things being equal, miniaturized TOF analyzers should be more sensitive because the chances are greater that a given ion will reach the detector. Cotter's group, in collaboration with
researchers at the Johns Hopkins University Applied Physics Laboratory and the
University of Maryland, has designed a
miniature TOF instrument with a flight
path of approximately 1
Mahadeva Sinha of NASA's Jet Propulsion Laboratory (JPL) is working on miniaturizing a magnetic-sector mass spectrometer for both space and terrestrial applications. Ions of different m/z values are focused simultaneously along a focal plane and measured with an array detector. Sinha says that this instrument is useful for applications where conditions in the ion source or sample are changing or when the sample is fugitive. Both cases prohibit the use of a scanning instrument. The instrument is also suited to isotope ratio measurements, where sequential measurements of isotopes can permit bias if conditions change. "[With simultaneous isotope measurements], any fluctuations will make all the isotope intensities go up and down in sync," says Sinha. Sinha's current miniaturized mass
spectrometer has a 1-in. focal plane, and
the entire instrument, including the ion
source, mass analyzer, and detector,
weighs about 300 g. This instrument covers a mass range of 2-250 amu and has a
resolution >300 m/ Cooks of Purdue University and J. Michael Ramsey and William B. Whitten of Oak Ridge National Laboratory are working independently to miniaturize cylindrical ion traps. Cooks has made an ion trap with an inner radius of 2.5 mm, and the ion trap constructed by Ramsey's group has a radius of 0.5 mm. According to Cooks, his group's ion trap is 1/4 the radius and 1/64 the volume of commercial hyperbolic ion traps. "We're really quite pleased that the performance, in terms of signal strength, is still quite good," says Cooks. "I wouldn't want to go any smaller than that without using different types of rf supplies." Ramsey and Whitten's ion trap is more than an order of magnitude smaller than commercial traps, with one-thousandth the volume. According to Ramsey, if the trap depth can be maintained, the charge storage capacity reduces linearly with the trap radius. The voltage requirements drop off as the reciprocal of the radius squared. "The smaller voltages suggest the possibility of using solid-state, direct-drive rf sources, so the electronics surrounding the ion trap can be potentially reduced in size," says Ramsey.
Cooks envisions using the miniaturized ion traps in an array in which each trap is a slightly different size. "We call it our pipe organ trap," says Cooks. "It's a set of what look like miniature pipe organs with different sizes--different radii, different lengths--and they're scaled appropriately." Rather than scanning the rf amplitude as is done in the conventional mass-selective instability mode, a single-rf frequency and single-dc voltage are applied to the entire array of traps. Each element, because of its size, traps ions of a certain m/z range. The elements are then emptied with a dc pulse onto an imaging detector, and the spatial registration corresponds to the m/z information. According to Cooks, everyone assumes that, as ion traps are miniaturized, fewer ions will be stored, and the signal will be reduced. However, that has not been true in his experience. He says that the major tradeoff is reduced mass resolution. Therefore, he anticipates using such an instrument for environmental or process control applications that do not require the same mass resolution as biological applications. Because the trap dimensions and the maximum m/z are inversely related, the smaller ion traps actually have a greater mass range than the larger traps. However, the low-mass cutoff increases concomitantly. Cooks doesn't believe the extended mass range is necessary to all users, which he cites as one reason that commercial ion trap mass spectrometers have an m/z range of 5000. "Would people like 50,000? I think they would, and this is a straightforward way of getting there," says Cooks. "But, would people like to trade off from having 0.1 mass unit accuracy? Would they trade up on that number to get to 50,000? Some would, some wouldn't." Ease of construction is a distinct advantage of the cylindrical ion traps relative to the conventional hyperbolic ion traps used in commercial instruments. "The devices we're using are simple cylinders. One of the students machines them by running a drill and making a hole," says Cooks. The end caps are flat, and by adjusting the spacing between the cylinder and the end caps, the field can be made nearly identical to that in commercial ion traps. Dan Dietrich and co-workers at Lawrence Livermore National Laboratory have miniaturized an FT-ICR mass spectrometer. According to Maria C. Prieto, from Dietrich's group, the major advantage is the ability to maintain instrument resolution and sensitivity as the size is reduced. Another advantage of this type of instrument is that the entire process--from ionization to detection--occurs in a single chamber, thus reducing the number of components that need to be miniaturized. Other advantages are the possibility for tandem mass spectrometry and the ability to select and eject ions of certain masses. The disadvantages are the need for a permanent magnet and the low operating pressure, in the range of 10-8 torr, which is necessary to maintain the resolution. The Lawrence Livermore group has
constructed an instrument that fits into
one side of an 18 x 12 x 3.5-in. briefcase.
The researchers have seen the isotopes of
xenon (m/z 132), but Prieto says with a
0.4-T permanent magnet, there is the possibility of a 300- to 500-m/z range. At a
pressure of 7 x 10-7 torr, the mass resolution is 500-1000 m/ Although some mass spectrometrists have denigrated quadrupoles in recent years because they are scanning instruments, they remain the undeniable workhorses of MS. Thus, it is not surprising that they too are being miniaturized. Quadrupole mass spectrometers that serve as residual gas analyzers are already available from companies such as Ferran Scientific (San Diego, CA). Richard R. A. Syms of the department of
electrical and electronic engineering at
Imperial College of Science, Technology,
and Medicine (U.K.) is developing a micromachined quadrupole mass spectrometer
that is based on cylindrical electrodes
mounted in etched silicon substrates. The
electrode rods have a 500-
Syms's group has chosen to work with silicon because it allows them to use microfabrication techniques that maintain the necessary precision as the filter is miniaturized. In addition, Syms says, "Microfabrication is inherently a batch-fabrication method, allowing costs to be reduced." Ara Chutjian, Otto Orient, Vachik Garkanian, and Murray Darrach of NASA's JPL have developed a miniaturized 4 x 4 quadrupole array using rods that are 25 mm long and 2 mm in diameter. By tuning the nine quadrupolar regions to the same m/z value, the array recaptures some of the lost ion aperture. It also lends itself to multiplexing--that is, tuning different regions to transmit different mass ranges. However, Chutjian says, they don't see an application for multiplexing right now because "it's just as easy to jump from mass to mass". According to Chutjian, the quadrupole array is being packaged for a space shuttle mission later this year, which will work on the international space station. The mass of the sensor, the electronics boards, and the pumps is about 1100 g, and the entire package is 4 x 6 x 8 in.--approximately the size of a shoe box. The instrument's primary applications will be detecting nitrogen and oxygen leaks from the space station hull, detecting ammonia leaks from the space station refrigeration system, and looking for hydrazine (used as the rocket and thruster fuel) that might adsorb to astronauts' space suits. The instrument is small enough to clip onto the front of the astronauts' space packs. More than mass analyzersThe most difficult parts of any mass spectrometer to miniaturize will probably be the vacuum system and the electronics. The system pressure is an issue in both space and terrestrial applications. "There is less atmosphere on Mars--it's about 1% of Earth's pressure--but it's still there," says Stephen Fuerstenau of JPL. "We still have to pump against it and find strategies for handling it." Fuerstenau has been involved in miniaturizing TOF and quadrupole mass analyzers for space applications, but he has focused on miniaturizing the other portions of the system as well. The good news for builders of miniature mass spectrometers is that the pressure doesn't need to be as low as in conventional instruments. Mass spectrometers are run under vacuum to minimize the number of collisions an ion experiences along its trajectory. The ion trajectories must be significantly shorter than the mean free path. A shorter ion trajectory means that the mean free path and the operating pressure can be correspondingly increased. Thus, miniature mass spectrometers may not require the same pumping capabilities as conventional laboratory instruments. Fuerstenau is working with "getter pumps" as low-mass alternatives to conventional turbomolecular and mechanical pumps. Getters are typically powdered metal alloys with fairly active surfaces. These powders will chemisorb or physisorb gases. According to Fuerstenau, some getters can absorb as much as 10% of their weight in hydrogen. He admits that they are fairly massive, but he points out that 100 g of getter would probably pump several grams of hydrogen, which is a fairly substantial volume at 10 torr. Fuerstenau says that for Mars probes, time is to their advantage. "You wouldn't want the pump down for any given sample here [on Earth] to take more than a couple of minutes. But on Mars, if it takes several hours to pump down, it's not a problem because we have plenty of time." Sinha says, "What makes a GC/MS
[instrument] big is the pumping requirement. The gas that you load into the mass
spectrometer really makes it a big pump
that you need. We are using a microbore
capillary column that has a 50- Cotter believes that already available pump technology could make instruments about the size of a computer laser printer. "That's smaller than anybody has out now, and it's small enough that people could comfortably move it around." Such an instrument would include a small mechanical pump and a turbo pump. The challenge, Cotter says, will be in making instruments small enough to be conveniently carried in the field. "There are some experimental pumps around that are being developed by the pump manufacturers. I don't think they're moving very fast on this, but there are some around that are the size of a videotape box," says Cotter. "I think we need--and nobody's been motivated to do this--to combine a mechanical pump and a turbo pump into an integrated system."
What's it good for?Cotter sees miniature mass spectrometers following two distinct paths--detectors and diagnostics. The first category of instruments includes those used in battlefield situations to detect chemical or biological warfare agents or on the space station to monitor air quality. "You could imagine small inexpensive MS devices being used for exhaust stack monitoring or even headspace monitoring of wells drilled into aquifers," says Ramsey. "Our efforts are directed toward environmental and field-monitoring activities associated with the production of weapons of mass destruction, in addition to potential forensic applications." Diagnostic instruments could be used to read DNA chips or perform antigen-based tests. "It's possible to precipitate these antigens in a process called immunoprecipitation. Most tests now that look at specific antigen-antibody [combinations] can only look at one thing at a time. However, as long as the antigens have different masses, a very small portable instrument in a doctor's office could look at a panel of antigens and find out which ones are being expressed. If somebody comes to us and says, 'I think I've got a virus,' we're not going to guess whether it's this year's Asian flu or last year's Asian flu. We're going to be able to say, 'Let's look at your antigens.'" According to Cotter, the real work needed to produce such a system is probably not instrumental. "If we want a diagnostic instrument that tells what kind of virus you have," he says, "we're going to have to know the antigens. People deal with these antigens all the time, but they haven't the foggiest notion of what their molecular weight is." For example, T-cells can be grown that respond to particular viruses without knowing what the antigens actually look like. However, such an approach results in a test that works only for that specific virus. With an MS approach, says Cotter, it is not necessary to bring in a whole host of different antibodies to test one at a time. Cotter also believes that a portable mass spectrometer could be used to diagnose diseases by monitoring RNA expression, which determines the proteins that are produced. He says, "There will be diagnostic instruments based on looking at selected parts of differential proteome expression. This peptide or protein is expressed when this disease state is present. It will be possible to look at a whole panel of those." Natural evolutionAlthough the push to miniaturize mass spectrometers is still only occurring in well-defined pockets of the MS community, Cotter sees it is as a natural progression. He compares the way mass spectrometers have shrunk since they were first introduced to the path followed by computers. "We've moved from the huge dinosaur era--the Univac era with huge mass spectrometer facilities; to GC/MS, which was kind of like mainframes; to instruments moving into biology laboratories, which I would consider like desktops; to the laptop era, with instruments that are much smaller and [that provide] dedicated processing," says Cotter. Cotter says that he has attended meetings where people predicted the demise of MS as a separate field of study. However, Cotter thinks that, following the computer analogy, the field will be healthy for the foreseeable future. "If you think about computers, it's not just programmers using them anymore. They're finding their way everywhere, including your oven and your automobile. The computer field will become healthier despite the fact that computers get taken more and more for granted," he says. "I think that will happen with mass spectrometers, too. There's still room for development, even though we're moving to a time when mass spectrometers will become pieces of other things." Celia Henry is an associate editor of Analytical Chemistry. |
m.
m diameter and
are 3 cm long. When driven at 6 MHz, the
mass filter provides 2.7-amu mass resolution at 10% peak height for the mass 40-amu ion of argon. Syms notes that the performance is "improving with experience".
