Inclusion of companies in this article does not indicate endorsement by either AAAS orScience, nor is it meant to imply that their products or services are superior to those of other companies.
Peter Girguis is neither a mass spectrometrist nor a chemist. He's a microbial physiologist, and his interest is the biogeochemistry of the deep ocean.
"Our entire biosphere is run by microbes," Girguis, the John Loeb Professor of Natural Sciences at Harvard University, explains. "That's pretty much the bottom line."
But the vast majority of microbes cannot be cultured in the lab, making them refractory to standard analyses. Girguis tries to understand what these microbes do by studying their impact on the chemical composition of the ocean floor and correlating those data with gene expression analyses to figure out which microbes are doing what.
"That's where mass spectrometers are, I would argue, one of the single most advantageous tools, because with a single analyzer you can detect a wide array of compounds," says Girguis.
There's no denying the incredible power of mass spectrometry. Using these instruments researchers can tease apart proteins and peptides that differ by just a single chemical modification; they can scan complex biofluids and home in on the few molecules that make them different; and they can interrogate samples for hundreds of compounds at once.
But to perform that kind of research takes considerable expertise. And the instruments on which it is done, says R. Graham Cooks, the Henry Bohn Hass Professor of Chemistry at Purdue University and a leading light in the drive to miniaturize mass spectrometry, are almost always for laboratory use only.
Benchtop instruments "weigh several hundred pounds," Cooks says. They're expensive and power-hungry, coupled to gas lines and powerful vacuums, and often require front-end separation systems. On the analytical side, they produce incredibly detailed spectra that take specialized software to decipher. All of which makes it hard to get the technology into the hands of people who might benefit from it—physicians at the bedside, firefighters in a burning factory, and even food safety inspectors in a warehouse.
"The shrinking of mass spectrometers is really about doing in situ, on-site measurements," Cooks says. "And that calls for an instrument that is fully portable and ... that can be moved around at will."
Go small or stay home
In making mass specs smaller and friendlier, researchers empower a far wider circle of users to employ them. David Rafferty, president and chief technology officer at 1st Detect Corporation, in Webster, Texas, likens the resulting democratization to the personal computer revolution. "Previously, only large institutions and large universities and companies had computers, but now with the advent of the personal computer it was made available to the masses, so to speak," Rafferty says. "We want to do the same thing with the mass spec."
Staffed heavily by expats from aerospace engineering, 1st Detect intends its MMS-1000 for industrial applications such as quality control and food safety testing, and, ultimately, homeland security. In contrast, 908 Devices focuses its 3.75 lb, "high-pressure mass spectrometer" on first responders in the safety and security markets, says Chris Petty, the company's vice president of business development and marketing, while Microsaic Systems, based in Surrey, United Kingdom, targets its single-quad 4000 MiD at organic chemists in drug discovery.
Girguis' need was more esoteric. His research calls for quantifying dissolved gases such as methane, hydrogen, and oxygen on and beneath the sea floor. It is, of course, possible to do that by installing a benchtop mass spec on a boat, collecting samples at depth, and analyzing them on deck. But a sample of water 1 km below the ocean can hold considerably more gas than it can on the sea surface, a function of the differences in pressure and temperature. "The solubility of methane at one atmosphere, 5°C, is about 2 mmol. The solubility of methane on the sea floor is much higher."
He realized he would need a mass spec he could use on site, and being "a bit of a gearhead," decided to build it himself.
Girguis got his first experience with high-pressure mass spectrometry as a graduate student at the University of California, Santa Barbara, when he was interested in the animals that colonize hydrothermal vents and their symbionts. He studied those in pressure vessels. Later, as a postdoc, he wanted to investigate the influence of microbes on the methane and hydrogen content of the ocean, but realized he needed a special instrument. How, though, to make a mass spec small enough and robust enough to operate underwater?
"The real serendipitous moment came when a couple of companies built small turbopumps," he says. (One of those companies, Alcatel Vacuum, was subsequently acquired by the other, Pfeiffer Vacuum, based in Germany.)
To build the mass spec itself, he worked with a mechanical engineer to package a commercial quadrupole mass analyzer from Stanford Research Systems, a Pfeiffer HiPace80 turbopump, and a custom gas extractor into a 25 cm x 90 cm cylinder. The result is the "in situ mass spectrometer" (ISMS), a 25 kg assembly that resembles a titanium-encased scuba tank, he says.
The extractor is a key element, Girguis says. Essentially a 10 &um;m thick Teflon membrane backed with a metal frit to provide structural support at high vacuum, this component degases the water being sampled by the mass spec, at up to 450 atmospheres of pressure. The resulting vapors are pulled into the instrument, ionized by electron ionization and mass analyzed, like a gas chromatography (GC)-coupled MS without the GC.
The ISMS has visited some enviable locales. Attached to either remotely operated vehicles or manned submersibles (like the Woods Hole Oceanographic Institute's Alvin), it has visited the Gulf of Mexico, hydrothermal vents off Washington state and the Azores (mid-Atlantic ridge in the North Atlantic), and the South Pacific. "I'm sure we've cleared over 100 dives at this point," Girguis says.
Using it, he has produced what he calls "geochemical maps" of dissolved gases at hydrothermal vents, collecting hundreds of data points both at different depths in the ocean sediment and across the floor. In one study, he discovered to his surprise that the charismatic deep-sea hydrothermal vents, sometimes called black smokers, often actually pump out less gas than do nearby "diffuse flows" on the ocean floor. "It just goes to show you that your eyes can deceive you," he says.
Girguis has published detailed plans and parts lists for the ISMS on his website, and anyone can build one. Total cost is about $15,000. But the housing is another matter. A simple polyvinyl chloride shell for relatively shallow dives (up to 50 m or so) might cost $1,000, but, "If you want titanium, to dive 4,000 m, you're going to have to shell out $20,000 for the housing alone."
Honey, I shrunk the spec!
Miniature mass specs have potential in other exotic locales, too. Rafferty says his company was approached by a museum looking to detect leakage of the preservative solution protecting an embalmed giant squid (though to date, no deal has been struck). Guido Verbeck, an associate professor at the University of North Texas who miniaturizes mass specs in his lab, envisions applications for his designs in homeland security and the military, such as being able to "toss" a mass spec into a burning industrial fire to have it report back what is burning, he says. "But you're going to destroy the device, so you have to make something that's cheap, small, [and] portable, with no moving parts."
As for Cooks, he targets the surgical suite. With colleague Nathalie Agar at Boston's Brigham and Women's Hospital, he already has demonstrated the feasibility of grading brain tumors using the lipid profiles they produce in a mass spectrometer (see "Mass Spec Imaging: From Bench to Bedside," scim.ag/1dCjmPx). But that experiment involved relatively simple benchtop instruments, Cooks says,Bruker and Thermo Fisher Scientific ion traps equipped with a Prosolia desorption electrospray ionization (DESI) source.
The logical next step, he says, is to shrink that ion trap down to a size (and cost) that would make it a practical addition in operating rooms everywhere.
As it turns out, one of the biggest challenges to shrinking a mass spectrometer is the vacuum, Cooks says. Mass specs function in a vacuum to eliminate background signal and avoid intermolecular collision events. But vacuum systems are large and heavy, and those parameters scale with the pressure differential needed. A Thermo Fisher Orbitrap requires three turbo pumps pulling some 900 L/sec in LC-MS modes to achieve a vacuum below 10-10 torr, according to a company representative.
Time-of-flight (TOF) mass analyzers also require high vacuum. As a result, most mini-mass specs are built from more forgiving mass analyzers, namely ion traps and quadrupoles—though at least two researchers have succeeded in miniaturizing a TOF, including Verbeck. Verbeck made a reflectron-based mini-TOF using a microelectromechanical system, or MEMS, technology, fashioning components out of boron-doped silicon wafers that he then assembled like old-fashioned tab-and-slot paper models. The analyzer measures just 2 cm x 5 cm, extending the ions' effective path length by moving ions back and forth for extended periods of time.
Cooks (with his associate at Purdue, Zheng Ouyang) built his miniature mass spectrometers using a linear (or quadrupole) ion trap, which operates at about 10-3 torr.
To produce that vacuum, he and Ouyang obtained the smallest commercial turbopump they could find, capable of about 10 L/sec. "You have to have a way of working with small vacuum pumps," he says. "This is the hardest part, the part that most people have stumbled over."
Such a pump is too small to allow continuous sample introduction, so the team built a discontinuous sample inlet system called DAPI (discontinuous atmospheric pressure introduction), which takes ions from the system's ionization source—in this case, a DESI wand—and holds them on one side of a pinch-valve, which opens periodically to introduce them into the mass analyzer en masse.
The result, Cooks says, is a fully self-contained device, the Mini-11, which weighs just 8.5 kg and yet contains a vacuum, pumps (a turbo pump and backing pump), ionization system, battery, electronics, and communications, in a single portable device. A backpack-mounted, 25 kg Mini-12 also exists, and Cooks has hinted that an even smaller device, perhaps powered by an iPhone, is in the works for at-home diagnostics.
Yet despite their small sizes, these devices are surprisingly powerful. The Mini-11 and -12 offer unit resolution mass spectra up to m/z 600, a range that makes it useful for studying metabolites, lipids, and other small molecules.
Little mass specs, big problems
Besides the vacuum, miniaturizing a mass spec poses other difficulties, too. The central electrode of an ion trap, for instance, is traditionally curved—picture an aluminum can that is pinched in the middle. As it gets smaller, the shape becomes harder and harder to manufacture precisely, yielding imperfections that can negatively affect ion motion.
1st Detect circumvented that problem by swapping the traditional "hyperbolic" design for a more easily fabricated cylindrical device—basically a smooth hole bored into the electrode. "You can make it smaller more easily without having to follow that precise curve," Rafferty says.
Another problem, says Stephen Lammert, director of research and development at Torion Technologies, is that as traps get smaller, squeezing the same number of ions into them becomes harder and harder. "The grand challenge of miniaturizing mass spectrometers, and especially ion traps, is: How do you make the trap smaller without losing ion capacity?"
Torion's solution, embodied in its Tridion-9 mass spectrometer, is the toroidal ion trap, which transfers the trapping characteristics of a traditional trap into a doughnut-shaped volume that can hold up to 400 times more ions. "The toroidal ion trap we have in our instrument is one-fifth the radius of what would be considered a conventional laboratory ion trap, and yet it still has the ion capacity of the conventional trap because of the expanded geometric storage shape." It also uses 25 times lower voltage and 125 times less power overall.
For Verbeck, the primary challenge in making a smaller mass spec was electrical.
"We've gotten to the point where the devices are so small that [with] one wire next to another wire, there's cross-talk," he says. His team had to go back and redesign the electrical system, "making cleaner channels in between the conductive pads, and making them wider," among other things.
But perhaps the biggest issue when miniaturizing mass spectrometry devices is the tradeoff it requires in power and flexibility. Ion traps, for instance, are attractive candidates for miniaturization not only because they are so simple, but also because they have the built-in capacity for tandem mass spec analyses, enabling sophisticated structural analyses.
But a mass spectrometer intended for soldiers, firefighters, and physicians must be simple enough to be used by someone who knows nothing about such nuances and have the built-in intelligence to automatically switch into tandem mode as the data require.
Such a system must be efficient enough to run on a battery and yet accessible enough to mass-spec novices to hide the complexity of a mass spectrum behind a friendly interface. Be not a device that requires instructions from its user, but one that, as Rafferty puts it, can scan a sample and go "beep, pesticide."
That's not to say mini mass specs don't have a home in the lab. An inexpensive mass spectrometer that can fit inside a fume hood would be a welcome addition to any organic chemist's toolbox, and Microsaic Systems, at least, intends its 4000 MiD for exactly that purpose. But the most exciting applications for mini mass specs surely lie outside the lab.
"We don't even want the device to be called a mass spectrometer," Rafferty says of the MMS-1000. "We'd prefer to refer to it as a sensor or a chemical detector." And really, when you strip away all the bells and whistles, isn't that what a mass spec is?