Safely Delivering the Best Possible Carrier and Detector Gases to Your GC System

The quality of a gas chromatogram depends heavily on the quality of the separation and detection gases, among many other factors. In this month's installment, "GC Connections" discusses ways in which chromatographers can ensure a safe working environment while delivering gases that are up to the requirements of the separations at hand, in the context of moving a laboratory to a new location.

Earlier this year, the laboratory I use was moved to a new location several miles away. In the course of setting up the new laboratory, the gas chromatography (GC) carrier and detector gas supplies had to be torn down and rebuilt. This laboratory is similar to many industrial GC laboratories-it contains a number of GC systems, plus a variety of specialized test equipment. The laboratory has two double-wide gas cylinder corrals that hold helium, air, nitrogen, hydrogen, and an assortment of gas standards. Cylinders not in use are stored in gas safety storage cages in another room. Occasionally a cylinder is secured next to an instrument, but lengths of manifolded tubing anchored to the walls deliver the gases for permanent use from the corrals to the instruments.

The new laboratory is somewhat larger and requires longer tubing runs between the cylinder corrals and the instruments. The challenge was to reuse as much of the existing hardware-regulators, tubing, fittings and valves-as possible, to save costs, while maintaining the integrity of the connections and purity of the delivered gases. The leak-tight aspect is absolutely crucial for laboratory safety, because some of the gas standards contain high levels of toxic substances, and, of course, hydrogen is quite flammable.

How did it go, and what lessons were learned? Let's take a look.

Safety First

Gas cylinder safety has been addressed many times in this column as well as in multiple safety guides and government regulations. The topic was covered recently in two "GC Connections" installments from 2016 (1,2). Good safety practice centers around proper training and equipment. Gas safety training should include both general procedures and practices as well as topics specific to the gases in use, emergency procedures, and appropriate training on how to make and break the various gas-tight fittings found in the work environment. Beyond cylinder restraints and carts, safety equipment also includes goggles, gloves, and safety shoes, plus correctly sized and rated regulators, tubing, valves, gas filters, and fittings. Please see the two installments (1,2), as well as the references inside, for additional gas safety information.

In this laboratory, tanks of air, hydrogen, nitrogen, and helium are secured alongside gas standard cylinders. Liquid nitrogen tanks are used occasionally. Beyond dangers directly attributable to the gas cylinders, such as high internal pressures and risks from handling heavy objects improperly, the cylinder contents create hazards of toxicity, flammability, oxygen displacement, and cryohazards. See Table 1 in reference 1 for more details about the various commonly used gases and associated hazards.

The primary lines of defense against these gas hazards are proper cylinder handling and regular verification of the leak-free state of all gas feeds. These hazards are mitigated further by providing the laboratory with high-volume heating, ventilation, and air conditioning (HVAC) air flow and suitable ventilation of gas streams, along with toxic, flammable, and oxygen-depletion gas sensors wired to the building alarm system.

When the new laboratory was configured, the gas sensors were brought over intact, and a larger HVAC system was installed to create an improved gas-safety work environment. During the interim period between moving the gas sensors to the new building and leaving the old laboratory, sets of similar portable gas sensors were leased and placed in key spots in the old location, where some work continued right up to the move date.

Immediately after the move, some of the built-in sensors were found to be near their rated service period and were replaced. It is difficult to perform regular checks for this type of safety failure. Fire extinguishers, for example, are checked periodically for expiration as specified in fire safety regulations. Equipment like the gas sensors sits quietly for extended periods of time without alarming and so can fade into the background and not receive sufficient attention. Another example like this that I have encountered is eye wash equipment in a very expired condition, the type that has a fluid reservoir instead of a plumbed water connection. Thus gas safety sensors, as well as all of the other safety equipment related to gases or for other purposes, should be checked regularly for function and expiration date where appropriate.

Making the Move

The move was a multistep process. One of the first steps was to disconnect the GC gas filters. The incoming gas pressures on the carrier and detector lines were reduced, the gas lines to the instrument were disconnected, and the filter fittings were quickly capped under gas flow. After disconnecting and capping the bulkhead fittings at the back of the instruments, the gas line pressures were shut off at the regulator outlet valves, the tubing was disconnected from the filters one by one, and the filter inlet fittings were capped. This approach allowed as much pure gas as possible to be retained inside the filters. The intent was to reuse each filter at the new location.

The tubing, fittings, and valves were disconnected from the regulators, and the longer tubing runs with intermediate unions were disconnected as well. The first tubing sections, starting at the tank regulators, are six-foot lengths of flexible hose, which makes connecting the cylinders much easier and adaptable. The same is true for the gas connections at the outlet ends of the carrier gas lines. Where possible, fittings were left intact, because they were likely to remain leak-tight through the move. The 0.125-in. diameter tubing was coiled as smoothly as possible, while the 0.25-in. pieces had to be moved with straight sections and bends left intact. It was not practical to cover all of the exposed fitting ends, so they were taped over with low-residue blue painter's tape. Of course, the new laboratory configuration is not the same, and so a significant number of pieces of bent tubing would not fit anywhere at the destination.

The tank regulators were vented and then packed a few to a box with bubble wrap or foam surrounding them. Although cylinder regulators don't look very fragile, their gauges and valve stems are prone to impact damage during transport. It's a good idea to cover the regulator gas inlets with low-residue blue painter's tape as well, to prevent ingress of particles. Regulator outlets are best sealed with matching caps.

Before the move, the gas cylinders were inventoried, and any that were no longer needed were returned to the supplier. When the time came to break down and pack the laboratory gas systems, rather than attempting to put the cylinders on a truck and run afoul of state or federal Department of Transportation, Occupational Safety and Health Administration (OSHA), and who knows what other regulations while creating a true public safety hazard, the commercial gas suppliers were engaged instead to move their cylinders themselves. The cylinders were disconnected from their regulators, capped, and then packed by the gas suppliers onto suitable pallets for the short journey to the new location. Once on site they were unloaded from the pallets and placed back into the in-lab corrals or into one of the gas storage cages.

Reassembly

At the destination, the lengths of tubing were assembled as best they could be positioned to bring the carrier and other gases to the instrumentation. Nearly all the tubing is stainless steel, so there was little concern that the tubing would fail as a result of stress fracturing across multiple bends. A few of the lines are the heat-treated copper 0.125-in. outer diameter type provided by some instrument manufacturers. These lines were left intact and were disconnected only at the instrument bulkhead and the corresponding gas filter, then reassembled to exactly the same fittings with as little rebending as possible. No fractures occurred.

New tubing was used in some places where nothing fit the required lengths. When all was complete, about 75% of the tubing in the new laboratory was recovered from the old lab. Much of the new tubing used was needed to extend the tubing runs for the increased distance between the tanks and the instruments.

Special attention was paid to any reused swaged fittings. Each fitting was inspected first for over-tightening symptoms of a bulging tubing end, or distorted ferrules. Another problem can arise from mis-matched fittings from two different manufacturers. A recent installment of "GC Connections" (3) has some good photos of what to look for in this regard as well as an informative discussion on how to make the connections.

A trial attempt at making each connection was performed. If the fitting nut did not engage the threaded union or valve thread smoothly and without requiring the force of a wrench, then the nut and ferrule portion on the tubing were discarded. If the union or valve was not new, it was also discarded since it was likely that the damage extended to both sides of the connection.

After making a new clean cut on the tubing, and using a new union or valve, a new connection was made following the fitting manufacturer's procedures. Overall a good recovery of used fittings was achieved, around 80%. This good recovery was due to having paid attention to the quality of the original installation of the fittings in the old laboratory, which paid off handsomely for the move to the new lab.

After a tubing run was complete, the exit was sealed temporarily with a plug and the line was pressurized with helium. The first leak check consisted of turning off the tank valve after pressurization, leaving the regulator's outlet valve open, and observing the high-pressure tank gauge for up to 30 min. If any observable pressure drop was seen, a quick check of the fittings with a helium leak detector usually revealed one or more leaking fittings, which were duly repaired or replaced. Sometimes all that was needed was the audible hiss of an untightened fitting! If no pressure drop was observed, then the fittings were checked more carefully to be sure there were no microleaks. Liquid leak checking solutions were not used.

Finally the intended gas, if other than helium, was connected and another pressure drop check was made for air or nitrogen, or a leak tester check for hydrogen, after which the line was deemed ready for service. Note that flushing a hydrogen line with helium for the initial leak check is a good idea, as it avoids potentially venting a lot of hydrogen into the air uncontrollably in the event of a large unintentional leak.

Leak checking was performed before connecting tubing to the inlets of any gas-scrubbing filters, to avoid forcing any more air than necessary into them. After a line was leak tight, it was purged with the appropriate gas before connecting to the filter, and then the filter was purged before connecting to the instrument. Fortuitously, none of the water or oxygen indicating filters exhibited significant degradation from before to after the move after following the above procedures.

Caution is advised when venting hydrogen lines. Hydrogen diffuses away into the room air quite rapidly, because of its buoyancy and high diffusivity. Using a low pressure in the line during purging helps limit the amount of hydrogen that is released. I can say, from the experience of unintentionally testing a combustible gas detector, that it takes about two seconds for hydrogen to make its way up to a combustible gas detector near a 10-ft ceiling and halfway across the lab!

After the gas lines were set, one of the GC systems was powered on and a series of baseline runs were made. They followed a normal sequence with some ghost peaks and baseline instability in the first couple of runs, and then settled down nicely. The move was deemed successful, and we resumed our normal work.

Conclusions

One of the lessons learned in the move was that it is absolutely necessary to maintain accurate records of the age of consumable components in the laboratory. In this case, expired gas detectors were discovered and replaced. The GC gas filters, at least the indicating ones, appeared to survive the move well, but the non-indicating filters may or may not be in good shape today. It is difficult to track how well they perform, unless a small indicating filter is inserted downline. These will be replaced as necessary and feasible.

Another observation: Treat your fittings well and they will repay you with multiple make–break cycles. You will avoid having to replace them often to keep the gas system leak tight, and reduce correspondingly the expense of new fittings.

Although I would not choose to move a workplace very often, with proper planning, organization, and attention to the technical requirements for both safety and gas handling, a move can be made without major unplanned interruption or equipment losses.

References

(1) J.V. Hinshaw, LCGC North Am. 34(10), 804–811 (2016).

(2) J.V. Hinshaw, LCGC North Am. 34(11), 844–852 (2016).

(3) N.H. Snow, LCGC North Am. 36(1), 38–44 (2018).

ABOUT THE COLUMN EDITOR

John V. Hinshaw

John V. Hinshaw "GC Connections" editor John V. Hinshaw is a Senior Scientist at Serveron Corporation in Beaverton, Oregon, and a member of LCGC's editorial advisory board. Direct correspondence about this column to the author via e-mail: LCGCedit@ubm.com