LCGC North America
A number of frequently asked questions about gases and their delivery to a gas chromatography instrument are addressed.
Gases for gas chromatography (GC) have become a hot topic in recent months, primarily because of concerns over short supplies of helium. Already one of the top discussion items, hydrogen as a carrier gas has garnered much of the attention as chromatographers' awareness of these issues continue to expand. Just last month (October 2012), on-line web seminars from Agilent and CHROMacademy were dedicated to conversion from helium to hydrogen carrier gas. A web search on "helium hydrogen carrier" yields guidelines and instructions from every major GC manufacturer and supplier as well as myriad topical threads on all of the GC blogs, boards, and discussion groups. Go to Pittcon, Analytica, the Eastern Analytical Symposium (EAS), or any other conference where GC is on the agenda and the helium issue will be featured prominently.
Among all the discourse, I have noticed that the essential related topic of good practices for deployment of carrier gases and, for that matter, all gases used in GC, is largely missing. Although much of this good advice is easy enough to find in instrument installation guides and supplier catalogs, the connection between obtaining the information and putting it into practice is often missed in many laboratories. Most laboratories will install gas filters in-line, but many will fail to obtain the right type of regulator, make the gas connections correctly, or maintain the filters and check the regulators on a regular schedule.
Questions That Should Be Asked Frequently
Here, then are some guidelines and recommendations about GC gases, in a question-and-answer format. This is not an exhaustive list, but rather it covers some of the more frequently asked questions as well as some that are not asked as often as they should be. The list starts at the gas source and moves onward to the instrument. Questions about the instrument internals are not addressed because of space limitations.
Table I: Recommended carrier gas purity*
What Are the Recommended Gas Purities for Carrier and Detector Gases?
The exact requirements for gas purity should follow the instrument manufacturer's guidelines as found in their site preparation and installation manuals. If that information is not available, then Tables I and II will serve as a general guideline. The gas purities are stated as percent levels rather than referring to supplier-specific names, which can be ambiguous or inconsistent.
Table II: Recommended detector support gas purity*
How Pure Are My Gases, Really?
The purity of a gas when it reaches the back of the instrument depends on the supply quality, regulators, filters, fittings, and connecting tubing. Filters will clean up minor contamination, but they are not intended to take gas to a higher purity level. Most of the time the purity of cylinder gas is as labeled on the bottle, but occasionally a contaminated cylinder may make it to delivery. Although it is bad practice on the part of cylinder users, a cylinder might be left open to the atmosphere for hours when empty and removed from service. If not cleaned up by evacuation and baking before filling and then filled with gas to 2450 psig (166 mPa), such a cylinder will contain approximately 6000 ppm of air, which degrades the gas purity to 99.4%. Although it is extremely unlikely to arrive in a cylinder at the receiving dock, this level of contamination represents a conceivable upper limit. When placed in service, the resulting onslaught of oxygen, water, and possibly hydrocarbons will completely exhaust a high-capacity gas filter before the contaminated tank is empty.
This potential for contamination is an excellent reason to use indicating filters on all gas supply lines. The indicator will change color as the filter reaches capacity. As long as the color change is noticed, a new filter is installed, and a pure gas supply is restored, the GC instrument will be spared the indignity of gas contamination and resultant high detector background, irregular baselines, and accompanying loss of signal-to-noise and repeatability.
It is possible, but expensive, to order purity analyses of individual cylinders. This step is only significant when it is difficult to observe the filters or replace the cylinder, such as at remote unattended locations.
How Often Should the Gas Delivery System Be Checked?
Given the slight chance of a contaminated cylinder, check indicating filters every time a tank is replaced or at least once every two weeks if using a gas generator. That's also a good time to check the gas lines for unusual bends or kinks. Check for leaks any time a fitting is disconnected and reconnected, and of course on all new connections.
It's also a good idea to check regulator function at least once a year. The easiest way to do this is to establish a flow of 500 mL/min or greater on each gas cylinder. Note the pressure output, then turn the tank valve off. Wait until the outlet pressure starts to drop off, and then turn the tank valve back on. The same pressure should be reestablished. Next, turn the regulator pressure knob down and back up while observing how smoothly the knob moves, how the output pressure gauge reacts, and that the knob is not screwed in all the way when the original pressure is restored. Don't forget to set the flow back to normal at the gas chromatograph.
Check that the correct cylinders are connected through to the correct fittings on the instrument. Some cylinders, such as helium, argon, and nitrogen, use the same gas fitting (at least in the United States); hydrogen and high-level combustible gas mixtures also may share a fitting type. Thus, it is possible to connect the wrong cylinder. Chromatographers should not rely on the uniqueness of cylinder fittings to ensure the correct gas type.
What Kinds of Regulators Should Be Used for GC?
Generally, dual-stage high-purity regulators with stainless steel diaphragms are the correct choice in all cases. For economy, less expensive dual-stage regulators may be used for detector air, hydrogen, and make-up gas if separate from the carrier supply. The added cost of a dual-stage compared to a single-stage regulator does not justify the potentially poor pressure regulation of a single-stage regulator as the cylinder pressure decays.
A regulator outlet valve is a handy accessory that I like to order on all regulators. However, in laboratories where the downstream connections remain in place permanently, the outlet valve can be omitted.
Always dedicate a regulator to its intended use. Never change the cylinder fitting on a regulator to switch it from inert-gas to detector-gas service or the other way around. Making the cylinder fitting connection correctly so that it will maintain high pressures is best left to the regulator manufacturer.
What Types of Fittings and Tubing Are Required?
The fittings and ferrules used in any GC installation should all be new, as should the tubing. The fittings should be of a suitable type for high-purity gas lines, such as those that are available from instrument company and supplier catalogs. I always like to order some spare fittings to have on hand for the usual problem connection that refuses to seal, as well as for later on when the gas setup needs some modifications.
Only metal tubing should be used for GC gases, with one exception. Copper tubing is the easiest to install, while stainless steel tubing is more robust and generally can be more organized visually. Either type of tubing must be precleaned before installation, and can be ordered that way as "GC" or "Chromatography" cleaned tubing. "Refrigerator" designated tubing is not suitable. For installations with tanks or gas generators dedicated to one or two instruments, ⅛-in. or 3-mm o.d. tubing is appropriate. If multiple instruments share a tank or generator then consider using ¼-in. or 6-mm o.d. tubing up to the point where the flow path splits to the individual instruments.
Polymer tubing is appropriate only in one case. If the GC system uses air-actuated sampling valves then that air supply can be connected with polymer tubing. But if the air tank is shared with a flame ionization detection (FID) system — which is not such a good idea, but it is done — then metal tubing is required throughout. Polymer tubing may allow air and airborne contaminants as well as monomers from the plastic into the gas stream. This is unacceptable for carrier and detector gases, even for FID air.
A nice touch when installing tubing is to mark either end of each tube with differently colored electrical tape. This makes it clear which tube goes to which gas inlet and regulator and avoids some of the more hazardous mix-ups such as swapping FID hydrogen and air. (Yes, I have seen that situation and the result of igniting the flame was, well, exciting!)
What Is the Right Way to Make Gas Connections?
Three gas connection types are encountered in a GC system. Occasionally chromatographers encounter other fitting types, but these three are the most common. First is the high-pressure tank fitting, identified by a letter and number designation such as CGA-580, DIN 477-6, or BS-341-3 for inert gases. With few exceptions no additional sealing is required — just assemble the fitting to the cylinder and tighten to seal. Most regulator fittings used in GC have a torque specification of 40–60 ft-lb (54–81 Nm). Some types of high pressure fittings, notably for liquid CO2 in the United States, require a plastic washer that is usually suppled with each tank. If a washer is present then assume it must be used, and the tightening torque must be reduced by half. If not in place, the absence of a washer will immediately be evident upon opening the cylinder valve! Never use polyfluorocarbon tape or liquid sealant anywhere on a high-pressure tank fitting; this will just make it leak. If either side of the fitting is scratched or deformed then replace the fitting.
The second type of fitting found in GC systems is the pipe-thread type. This fitting involves matching internally and externally threaded sides with no washers or ferrules. After making sure the threads are clear of old sealing tape, wrap two layers — no more — of polyfluorocarbon pipe sealing tape (available from instrument suppliers) onto the externally threaded side of the fitting. Holding the exit end of a right-hand threaded fitting toward you, smoothly wrap the tape in a clockwise direction while stretching it slightly. Then thread the taped piece into the internally threaded side and tighten.
The third type of GC fitting is the swaged tube fitting. This fitting consists of a threaded receiving piece with an internally beveled surface, a matching hexagonal internally threaded nut, and a one- or two-piece washer set. The swaged fitting directly connects the end of a tube to one side, and then includes one or two additional connections that may be pipe-thread, another swaged fitting, or a tube. These fittings can be of different sizes so that a swaged fitting can be used with different diameter tubes. Swaged fittings are available from several companies as well as in brass and stainless steel. Always use matched parts of the same metal from the same company.
Connecting a new metal swaged fitting is slightly more complex than the other two types. First, the tubing end must be cut squarely and free of burrs, scratches, or cutting debris. Check that the hexagonal nut threads onto the fitting smoothly, especially if reusing the nut or fitting. Using the tube as a guide, first slide the nut onto the tubing with the threads facing outward, then the circular washer (if needed) with the narrow side outward, and finally the conical washer with the narrow end facing outward. Fit the parts together and hand-tighten.
Now, comes the dexterous part: while holding the tubing all the way into the fitting with one hand, take the right-sized wrench in the other hand — such as a 7/16-in. wrench for a ⅛-inch swaged fitting. Hold the hexagonal part of the fitting with the wrench. Now, take another wrench in the other hand and . . . wait hold on, that's three hands! In the absence of a vise, chromatographers soon learn how to hold a wrench, the fitting, and the tubing in one hand while tightening the assembly with another wrench in the other hand. They didn't teach that in instrument class.
A metal swaged fitting must be tightened a certain number of turns, and not to a particular torque. For a common type of this fitting, the ⅛-inch size is tightened ¾ turn when new while the ¼ size is tightened 1¼ turns. The fitting manufacturer may make available a maximum-clearance tool that helps gauge when the fitting is sufficiently tightened. In any case, instructions are available from the manufacturers and should be followed closely. Overtightening these fittings will reduce the number of times they can be reconnected or even cause them to fail to seal altogether.
Common capillary column connections are also of the swaged type, but instead of hard steel or brass washers a soft metal or polymer ferrule is used. Sometimes the same polymer ferrule type is used to connect small diameter tubes inside of an instrument as well.
Regulator and pipe-thread fittings may be reused unless damaged. Swaged fittings, if treated correctly, also can be reused but must be carefully examined beforehand. Do not reuse if the nut does not thread smoothly onto the fitting or the existing tube end is distorted or bulging out of the ferrule. Instead, substitute all new parts and recut the tubing to start over.
Always leak-check fittings after making the connection. In the case of the regulator or pipe-thread fitting types, additional tightening up to the maximum specified level may help secure the seal, but never exceed that amount of torque; check the regulator on another cylinder and replace as needed.
Many More Questions
I've gotten only halfway down my list of questions that are or should be asked about gases for GC. This discussion will be continued in an upcoming installment next year. In the meantime, readers are encouraged to submit their questions about gases or anything else GC-related to firstname.lastname@example.org.
John V. Hinshaw "GC Connections" editor John V. Hinshaw is a Senior Scientist at BPL Global, Ltd., in Hillsboro, Oregon, and a member of LCGC's editorial advisory board. Direct correspondence about this column to the author via e-mail: email@example.com.
John V. Hinshaw