Gases & Instrumentation - January/February 2009

It was actually the Institute’s environmental section that eventually formed the Environmental Protection Agency (EPA). There were a couple of scientists there, Andy O’ Keefe and Gordon Ortman working on methods to calibrate ambient air analyzers that made these little gadgets— permeation tubes. They sealed liquid SO and NO inside a piece of a

22

Teflon tube. The vapor leaking
out through the Teflon® was
a stable, very low continuous
flow of the SO or NO . They
22
measured the flow rate by peri-
odically weighing the tubes to
measure the rate of weight loss.
Using weight loss also gave a
handle for obtaining traceabil-
ity to NIST.

PN: You were involved with permeation tubes before Kin-Tek, correct?

JJM: I started doing sampling by permeation, back when I was with Monsanto. We had some streams that were just horrible to deal with. And one of the compounds they were interested in measuring was hydrogen. Then, I came across a paper talking about the permeability of Teflon® and other plastics, work that was done by the British UKAEA (The United Kingdom Atomic Energy Authority).

PN: So that got you started?

JJM: Yes. I started to play around with that concept and developed what turned out to be a patented hydrogen analyzer based on permeation. Then I saw this article by O’Keefe and Ortman about their work. I didn’t think anything about it for a couple of years or so until I formed Kin-Tek. Originally, those two NIH scientists were working primarily with SO and NO . I had been deal-

22 ing with analyzing industrial compounds and I knew that

there were a number of them that were needed at rather low concentrations. In particular, we had problems calibrating for trace contaminants in ethylene. Some of those ethylene contaminants could not be contained as a liquid sealed in a permeation tube, so right away we began expanding the horizons of the method by developing a gas-only tube and a method to measure the emission rate. And since then, I have been working with gases and permeation tubes.

PN: So gases and permeation tubes grabbed your interest?

JJM: I had started Kin-Tek with the intention of doing a number of different things, but because of the increasing applications, the permeation business was just about all I could handle. It turns out that permeation tubes are extremely adaptable to a really wide range of compounds. And we can do the development on a new permeation tube at a one hundredth or so of the cost to develop a cylinder standard. If you can transport the mixture at all, that is if the compound doesn’t immediately decompose or polymerize, then you have a good chance of making it work as a permeation tube. You don’t have to deal with interactions with the cylinder walls and you don’t even need to worry about slow reactions in the mixture. You make it and use it immediately before it deteriorates.

PN: How have the demands changed?

JJM: We started working at low ppm and some fractions of ppm; now a ppm is a moderate, easy concentration and low ppb is common. Now we are frequently at ppt, and we actually have made a few parts per quadrillion mixtures. The trouble with dealing with con-

centrations this low is there is no way of verifying it. There are tricks you can play to verify your basic numbers, but beyond a certain level, you really can’t measure it so you are totally at the mercy of those quirky non-linearities we stumble over from time to time.

PN: Are there new applications that will want that level of detection, even if they can’t have the verification?

JJM: That is happening. And in fact, all of the homeland security problems have pushed things in that direction. This is because they really wanted to detect extremely low levels so they can react quickly, when, for example, they want to find where explosives devices are being made, or detecting when toxic gases are released into the atmosphere. And incidentally, one of the people working with us has patented a computer program that can combine the detected signal levels from a target substance, with data from meteorological instrumentation and can pin-point the source. In a dynamic situation that is quite a trick.

PN: Detection is one thing, but locating the source is another.

JJM: That’s true. One test using that technology was at a superfund site where they were getting anomalous levels of benzene. They took this program and ran it using data from the existing in-situ monitoring system, and found that the source was an open gas can left by a landscaper about a quarter mile away from the superfund site, completely off the property.