When you’re a white guy in a J. Crew shirt, headed toward the outer boroughs from Hell’s Kitchen, it’s tempting to see police officers as set dressing — and tempting to believe that they see you the same way. Like the tag lines for ill-fated sitcoms written on the subway walls, transit police are just another native species of the MTA. That’s privilege speaking, of course, but there’s only so much we can ask of ourselves when we commute. The shortest distance between point A and point B is, after all, a pre-coffee haze.

So I was probably more shocked than I should have been when I got stopped. The initial punch in the gut, Me, really?, was followed by the one-two of, These dudes have jobs to do and Don’t be an entitled piece of shit. I put my bag on the white plastic table and went to open it because that’s what I presumed the point was, to see what was inside (red curry in Tupperware). But that wasn’t the point, or what the cops want. They swabbed the bag with a piece of fabric and placed that fabric in a small black box. I asked what the device was designed to find.

“Bombs,” said the cop, before making that great law enforcement gesture that means “move on.”

In 2006, the New York City Police Department purchased several Sabre 4000 explosive trace detectors from Smiths Detection Inc. It’s certainly possible, even likely, that my bag was analyzed by one of these systems — though the NYPD declined to confirm their continued usage. According to its technical specifications, the Sabre 4000 can detect particles of Semtex, TNT, sarin, methamphetamine, THC, cocaine, and other substances within 15 seconds. So, yes, they were looking for bombs, but they might have found something else.

The first step in the Sabre’s ion mobility spectrometry (IMS) process is the fabric swab. This is where most false negatives pop up — if the swab doesn’t pick up an explosive molecule, there’s nothing to detect. Once inside the IMS, a bit of nickel-63 or other source incinerates the sample to ions.

An electrical gradient bulldozes the ionized molecules through the IMS, where they collide and bounce off one another at a rate of billions of collisions per second. The molecules with higher charge affinities gobble up the charges from the other ions. You can think of it as 20 kids on a playground fighting for a ball. “In the end,” says David Atkinson, Ph.D., an analytic chemist and trace detection expert at the Pacific Northwest National Laboratory in Richland, Washington, “the biggest kid always ends up with the ball.” Based on its charge, an ion will go from the beginning of the tube to a detector at the end in a very specific amount of time. This is the ion mobility. “All of the TNT ions [in a sample] have the same mobility,” according to Atkinson, “they’ll show as a peak in the spectrum.”

As chemistry would have it, nitro-type, military explosives — nitroglycerine, TNT, RDX, PETN, HMX — have charge affinities that stand out from more mundane materials.

“Let’s say we have 999 gasoline molecules. Gasoline is a very complex mixture of molecules: It has thousands of chemicals in it—benzene, toluene, xylene, ethyl benzene, pentane, propane, octane — they’re all in there,” Atkinson says. “We have 999 gasoline molecules and one explosive molecule. You will only see the explosive molecule — every single time — because of that chemical selectivity.”

The ionization process is very selective, with scientists and industry specialists boasting about 99 percent accuracy for IMS in the field. But it’s not perfect. “If you look around the internet — I’m not going into details about which things may cause false alarms — you will see occasional blog postings and news stories,” says Atkinson, of toiletries or other chemicals that might register as explosives. The American Civil Liberties Union speculated in 2013 that bed-bug killing spray could have caused a Hindu man to set off an IMS detector in an airport, leading to an ugly misunderstanding with JetBlue.

“We’d like to make false alarms zero. Because every time there’s an alarm you have to deal with it,” Atkinson says. “You have to have an alarm resolution procedure, which often involves a search, a pat down, everything else.” His goal is a combination of more sensitivity (“femtograms would be wonderful”) and more selectivity. The big question in Atkinson’s mind is if mass spectrometry — similar to IMS, but the tube is a vacuum — could work.

Mass spectrometry, as Atkinson demonstrated in 2013, has plenty of promise: A device in his lab detected vapor — not a swab — of the explosive RDX at 25 parts per quadrillion. It is the closest a man-made device has come, in this sense, to the sniffing abilities of a bomb-detection canine. But the device is large and expensive, and even if it made it out of a laboratory setting, would require some 45,000 officers to be retrained.

Smiths stopped manufacturing the Sabre 4000 in 2011, replacing it with the Sabre 5000. The 5000, as Smiths’ Head of Product Ken Fredeen told Inverse, can suss out one nanogram of an explosive and has a false-positive rate of less than one percent. But what about the poor 4000? If those devices were still on the streets — and, if Smiths knew the NYPD had upgraded, he wasn’t admitting it — should they be?

“Customers with Sabre 4000 units can generally expect them to remain in service for 10 or more years from their date of manufacture,” Fredeen said, “depending on usage, care, and maintenance.”

So, yeah, given a date of manufacture prior to the date of purchase (a reasonable given, I think) it’s possible my bag was being analyzed by a machine nearing its expiration date. After years of use, a trace explosive detection system might need maintenance on the mundane components — air purifiers, say — or slightly more exotic repair. Inside every Sabre 4000, like most other ion mobility spectrometers, sits a radioactive lump of nickel-63. The radioactive nickel will ionize particles, illicit or otherwise, for analysis. Being radioactive, that source decays; in nickel-63’s case, there’s a half-life of about 100 years.

“I have IMSs that are 30 years old and still work fine,” says Atkinson.

What does this mean for the temporarily inconvenienced? Not much. I had my bag screened for trace elements of bad particles, but I’ll never know if the machine was fully operational. (The lifespan of a device sitting in an expert’s lab is, presumably, a bit longer than one banged around the subway.) Did the swab miss the few specks of THC that I’m fairly certain sticks to most objects belonging to a cross-section of the New York population? Or do you get to pass on cannabinoids from bomb cops? Some questions can never be answered. The only certainty is I’m not touching bed bug spray until Atkinson perfects his mass spectrometer.

Photos via Giphy.com