OR WAIT 15 SECS
John V. Hinshaw is senior staff engineer at Serveron Corp., Hillsboro, Oregon, and a member of LCGC's editorial advisory board. Direct correspondence about this column to "GC Connections," LCGC, Woodbridge Corporate Plaza, 485 Route 1 South, Building F, First Floor, Iselin, NJ 08830, e-mail LCGCedit@ubm.com. For an ongoing discussion of GC issues with John Hinshaw and other chromatographers, visit the Chromatography Forum discussion group at http://www.chromforum.com.
This month, John Hinshaw examines the environment, setup, and operating conditions necessary to ensure high gas chromatography detector performance and reliability.
After analytes transit the inlet system and separate in the column, a detector transduces the contents of the exiting mobile phase from a variable chemical signal into an electrical signal that can be recorded and processed to yield qualitative and quantitative information about the analyzed mixture. As chemical to electrical transducers, detectors are susceptible to interference from chemical and electrical sources that originate outside as well as inside the chromatographic system. Such interferences can reduce minimum detectability, compromise peak area accuracy and repeatability, and detract from qualitative peak identification confidence. Gas chromatography (GC) detectors must be set up and maintained correctly. They require a suitable operating environment in the form of appropriate gases and electrical supplies, and temperatures and flows must fall within proscribed limits. Although many analyses do not require the highest available sensitivities, analysts should establish and maintain a healthy performance safety margin so that nominal performance degradation during regular maintenance intervals is accommodated.
John V. Hinshaw
Requirements for the Best Detector Operation
Obtaining consistent high detector performance over extended periods requires the analyst to adhere to specific requirements for setup and operation. Some, such as hydrogen and air flow, are determined largely by the manufacturer and design and cannot be changed significantly without loss of detector performance; straying too far can cause a failure to operate at all. Others, such as amplification range, are dictated by sample size, analyte-specific response, and their relationship to minimum detection requirements. By encoding these items in a well-written method and adhering to them, chromatographers can achieve the detector noise, sensitivity, drift, and reliability that their analytical methodology requires.
Each detector has unique gas supply requirements as well as other external requirements that chromatographers must establish and maintain. Instrument manufacturers specify basic requirements for the electrical supply and environmental conditions for their instruments, and they will communicate these conditions upon request. These prerequisites affect all aspects of an instrument and should be adhered to as closely as possible.
Electrical: As electrical transducers that operate in very sensitive ranges, GC detectors are particularly sensitive to electrical supply quality and to external interferences. In general, each gas chromatograph requires a dedicated grounded alternating current (AC) electrical supply with its own circuit breaker, usually 120 V at 20 A or 240 V at 10 A. The appropriate voltage depends upon the original country or area for which the instrument was manufactured. GC users should not attempt to change the voltage supply level for which the instrument was manufactured. The internal electronics usually are capable of running with multiple line supply voltages, but in most units the inlet, detector, auxiliary, and oven heaters are voltage-dependent, so it is not possible to switch from 120 to 240 V or vice versa without replacing these heaters with those designed for the target line voltage. The line voltage also must stay within a certain window above and below the nominal voltage level: one manufacturer specifies ±10 %, for example. The majority of modern GC systems are not affected by the AC supply frequency — they use internal crystal clocks for a time base — so either 50- or 60-Hz supplies are appropriate.
Most importantly for detectors, the electrical supply must be free of significant electrical spikes, high-frequency signals, and rapid voltage fluctuations. Some manufacturers specify line quality and, in cases in which such interferences are suspected, a line voltage monitor can reveal the nature of any unusual fluctuations. In addition to the quality of directly connected supply lines, the vicinity must be free from sources of significant radio-frequency (RF) interference. Various country and international standards specify both the degree to which instruments can be influenced by incoming RF signals, as well as the frequency distribution and signal strength of any potentially interfering RF signals that originate from the instrument. Instruments will meet these specifications only if all covers and attached grounding wires are in place. Even so, some cell phones and two-way radios can produce significant detector interference if operated very close to an instrument. It is good practice to require that cell phones be turned off or simply not present in the proximity of sensitive laboratory instruments.
Laboratory temperature: In general, most laboratory heating and cooling systems provide the laboratory temperature constancy necessary for stable detector operation — about ±5 °C. Constant laboratory temperatures also are necessary for the best retention time reproducibility, due to temperature coupling with the oven temperature as well as with pneumatic controls and tank regulators. The gradual fluctuations associated with normal air conditioning and heating system thermostat cycling will not affect detector output stability if sensitive detector components are isolated thermally in separately heated zones, under instrument covers, or in external boxes. However, it is a good idea not to permit heating and cooling room air to flow directly over a GC system. Temperature fluctuations are more abrupt inside the airflow, and some detectors will exhibit increased noise when operated in a turbulent air stream. Avoid exposing an instrument to direct sunlight, which can heat portions of the instrument far above the recommended 5 °C temperature extremes. GC instruments also have maximum and minimum allowable operating temperature and humidity levels that should not be exceeded. One company specifies a temperature range of 10 °C (50 °F) to 32 °C (90 °F) at noncondensing relative humidity levels of 5–80%.
Sometimes installing a GC system in an otherwise adequately controlled room will increase the room temperature significantly. Bench-top gas chromatographs give off in excess of 3000 British thermal units (Btu) when operating under average temperature-programmed conditions. In the absence of sufficient cooling power delivered into the room, the ambient temperature can exceed the maximum instrument operating temperature. In cold environments, on the other hand, the GC system can be a welcome additional heat source for chilled analysts.
In the absence of well-controlled laboratory temperature and humidity, GC users might experience instability in retention times and detector responses that do not meet expected performance criteria. Besides upgrading the heating, ventilation, and air conditioning systems, GC users can circumvent temperature-related problems by choosing to make critical separations during periods of relative temperature stability. A simple room fan or open window sometimes serve to enhance the removal of excess heat from a GC system and control the laboratory temperature sufficiently, especially when the instrument is located in a small room or trailer.
Gas supplies: All gases flowing into a detector must meet minimum purity and flow-control specifications. In general, detector requirements — not column or inlet system considerations alone — determine appropriate gas purities. Even when only the carrier gas enters the detector—as in a mass spectrometry (MS) detector—the carrier must meet the possibly more stringent requirements for the detector and not just those for the column alone. Table I summarizes carrier, combustion, and makeup gas purity, purification, and regulation requirements for some common detector systems.
Table I: Recommended gases for carrier, combustion, and makeup gas.
In addition to procuring sufficiently pure gases, chromatographers must ensure that appropriate gas pressure regulators and purifiers are in place. An inexpensive regulator can add impurities to the gas passing through and might fail to regulate the gas pressure with sufficient accuracy. Without proper purification, minute leaks between the tank and the instrument can compromise gas purity and eliminate the advantage of more expensive, high-purity gases. On the other hand, do not rely solely on filter purification to remove impurities from less expensive, low-grade gases. The objective of gas purification is to ensure that gases reach the instrument without any added impurities from regulators, tubing, and microleaks, not to act as an inexpensive high-purity gas preparation system.
With appropriate gases, regulators, and filters on hand, chromatographers must establish leak-free connections to clean tubing. GC consumables suppliers offer suitable fittings and appropriately cleaned copper or stainless steel tubing. Always try to use new ferrules and nuts when connecting gas supplies. If older ferrules already on a length of tubing are distorted in shape, or if the fitting nut does not turn smoothly and seat firmly, then installers should cut off the tubing cleanly and use new ferrules and nuts. The bulkhead fittings on the back of instruments should be in good condition as well. Follow fitting manufacturers' guidelines for making up new fittings. In particular, smaller swaged gas fittings generally require less than one turn past finger-tight for leak-free performance. Over tightening a fitting can cause it to leak, and once leaking from being overtightened, the only option is to cut the tube, replace the fitting components, and start over.
Once the gas supplies are connected, pressurize the tubing and check for leaks with an electronic leak detector. Never use a leak-checking solution that contains surfactants. A leak is like a two-way street, and solutions, as well as atmospheric oxygen and contaminants, will enter the gas stream through even a small leak. Use a volatile solvent such as isopropanol or water if necessary, and fix all leaks before operating the GC oven above 100 °C.
A pressure-drop leak test also is effective, particularly for gases such as air or nitrogen, to which electronic leak detectors are not very sensitive. Make sure that the gas controllers are turned off at the GC system, then shut off the high-pressure supply at the tank. Note the tank pressure, then wait 10 min. If the tank pressure has fallen significantly and increases noticeably upon reopening the tank valve, then there can be a significant leak in the regulator, line, or filter connections.
Installing the column: Follow the manufacturers' guidelines for column installation into a detector and try not to insert the column into a heated detector, especially if carrier-gas flow has not been established. Each detector has a nominal distance for positioning fused-silica capillary columns: see the instrument manual for exact specifications. Avoid introducing dirt into the detector by first threading the nut and then a new ferrule over the column tip. Cut the column off squarely, and measure the distance from the back of the nut to the freshly cut column exit. A small piece of masking tape serves to mark the correct position below the column nut as well as to support the nut during installation. Remove the tape before heating the detector or column. White correction fluid also works well, but take care that the dried solution does not become trapped between the ferrule and column as they slide over each other.
It is a good idea to wear clean gloves when handing and preparing the column ends, although gloves will make it more difficult to manage the small parts adroitly. Carefully cleaning the column end sections from the nuts towards the column tips with a lint-free methanol-moistened laboratory towelette helps remove any oils or dirt before they can enter the detector. If this is not done, then the column is likely to show more noise, offset, and bleed initially. Baking the system out will help reduce the contaminants.
Tighten the fittings no more than ¼–½ turn past finger tight, pressurize the column, and then, before heating, leak-check the column connections with an electronic leak checker and tighten a little more if necessary. While in the area, leak-check the bottom seals of the inlet and detector to ensure that tightening the column fitting did not disturb the next fitting up towards the inlet or detector. Be sure to reinstall the column and detector insulating cups if so equipped. Check the inlet septum as well.
The column can be damaged by performing bake-out runs with the detector gases turned off. GC operators should ensure that all detector supply and makeup gases (as well as the carrier gas) are flowing before heating and then cooling the column. The flow at a capillary column exit can reverse momentarily as the carrier gas contracts upon commencing oven cool-down. If the detector internal volume is not purged with oxygen-free hydrogen, carrier, or makeup gas, then a significant amount of oxygen can be sucked back into the column, which potentially could then damage the stationary phase close to the column's end.
Some ferrules will shrink slightly upon first heating, so leak check again and retighten the column fittings if necessary after the first temperature-programmed cycle. If more than ¼ turn is required to reseal the fitting, then it might be better to take it apart and try again.
Choosing optimum detector temperatures, flows, and electrical settings requires consideration of the detector's requirements as well as the column limits and the GC method.
Temperature: The detector operating temperature should be at least 20 °C above the highest operating column temperature, as long as this does not exceed the maximum allowable column-operating temperature or the maximum detector operating temperature, as specified by its manufacturer. Thus, a detector can sometimes impose a lower operating temperature limit than the column limit itself. Some manufacturers recommend always running an electron-capture detector at higher temperatures — up to 350 °C — to reduce the accumulation of high-boiling residue that might contaminate the detector. Electron-capture detector sensitivity drops off somewhat at lower temperatures, which provides an additional reason for higher temperature operation.
In the case of a thermal conductivity detector, the detector filaments have a maximum temperature. As filament current increases, the filament temperatures rise above the detector block temperature. Higher block temperatures mean that the filaments need less current to reach their maximum temperatures, so that thermal conductivity detector operation at higher block temperatures implies lower available filament currents and concomitantly lower available detector sensitivity.
Most detectors have a minimum operating temperature as well. A flame ionization detector should operate above about 150 °C to prevent condensation of the water of combustion inside the detector, so even if the column operates at only 50 °C, the detector should still operate at 150 °C or higher.
Range, attenuation, and autozero: Detector range and attenuation settings vary from one instrument to another. The variable attenuation settings in some instruments affect only the seldom-used analog chart recorder outputs, so it is a good idea to become familiar with how these settings function. In general, chromatographers should choose range and attenuation so that the highest anticipated analyte peak does not exceed the maximum output voltage level of the instrument. If that is not possible, then consider switching the detector range as necessary during analysis to bring the very large peaks back on scale. It usually is acceptable to allow a solvent peak to extend off scale. When monitoring the detector signal for off-scale conditions, remember to set the displayed signal range so it spans the entire output signal range from the detector — that is, zoom out all the way. Do not make the mistake of believing that a peak has been clipped at the detector when it only appears that way due to the on-screen signal presentation.
An instrument's autozero function serves to bring the quiescent detector output signal to a predetermined level just before the start of each run. This will prevent the initial signal from drifting below zero for detectors that drift significantly from run to run, which would render much of the subsequent data unusable. The autozero output signal level is configurable in most GC systems. Do not choose 0 V for this parameter, but instead set a positive value, such as 5 mV, that will provide sufficient room for a slight downward drift during analysis.
Flow rate: Detector gas flow rates strongly influence detector performance. A detector might exhibit lower sensitivity and increased noise levels if flow rates are set incorrectly. The flame ionization detector flame might not light or stay lit when the hydrogen flow is too far from the specified level. If detector flows are not well regulated, the baseline can wander or drift sufficiently to disturb peak detection and quantitation. Most detector flow controllers do a good enough job of regulation when operating correctly, but the question of measuring and setting the correct flows is more complex.
Accurate flow measurements depend upon the correct calibration and operation of the flow-measuring device. When using an electronic flowmeter, make sure it has been calibrated to a reference standard. Most flowmeter manufacturers supply this information with each meter, and they also can provide a recalibration service. Many flowmeters include a gas-selection setting, which must match the subject gas for accurate flow determination. This means that only one gas at a time should flow through a detector when determining flow rates. First, as mentioned previously, connect the column to the detector, purge the column with carrier gas, and check for leaks. Then, while keeping the oven door open, turn off the column flow and allow the pressure to dissipate. Next, turn on the hydrogen, air, or make-up gas, choose the correct gas type on the flowmeter, and record the flow, correcting as necessary. Then turn off the measured gas, turn on the next one, set the flowmeter gas type, and measure again. If a measured flow is significantly different than its setpoint, then either there is a leak or the flow controller or meter are not calibrated.
It is a very good idea to calibrate or verify the flows delivered by electronic or manual flow controllers both upon installation and periodically thereafter. Also, recalibrate if the instrument is moved or the gas supply pressure changes significantly.
The most accurate flow measurements are meaningless if the detector or the flow meter connecting tubing leaks. In such cases, the measured flow will be significantly less than the total gas flow that is delivered to the detector. Check for such leaks with an electronic leak detector while measuring hydrogen or carrier gas flows. The leak detector should give little or no response in the area around the detector base and the flowmeter connection after sufficient time is allowed for any gas to dissipate from the area. Correct leaks by resealing the detector base and side arms and by using a flowmeter tubing adapter that fits the detector tightly.
Taking steps to ensure that detectors are installed and operated correctly will go a long way towards obtaining the best possible detector sensitivity, noise level, and stability. The electrical supply, ambient temperature and humidity, gas supplies, pressure regulators, purifiers, tubing, and connections must meet the minimum criteria for the detector and column combination in use. Proper column installation is especially critical. Operating conditions — including range, attenuation, autozero, temperature, and flow settings — all must be selected appropriately for the detector and data system. Once an instrument system is up and running satisfactorily, record a benchmark chromatogram for future reference when something does go wrong: it will be an excellent aide for diagnosing a problem.
John V. Hinshaw "GC Connections" editor John V. Hinshaw is senior Research Scientist at Serveron Corp., Hillsboro, Oregon, and a member of LCGC's editorial advisory board. Direct correspondence about this column to "GC Connections," LCGC, Woodbridge Corporate Plaza, 485 Route 1 South, Building F, First Floor, Iselin, NJ 08830, e-mail firstname.lastname@example.org. For an ongoing discussion of GC issues with John Hinshaw and other chromatographers, visit the Chromatography Forum discussion group at http://www.chromforum.com.