Flame ionization detector optimization

Here is a challenge McMinn and co-workers investigated the effect of five factors for optimizing an H2-atmosphere flame ionization detector using a 2 factorial design. The factors and their levels were... 

Catalysts were tested for oxidations of carbon monoxide and toluene. The tests were carried out in a differential reactor shown in Fig. 12.7-1 and analyzed by an online gas chromatograph (HP 6890) equipped with thermal conductivity and flame ionization detectors. Gases including dry air and carbon monoxide were feed to the reactor by mass flow controllers, while the liquid reactant, toluene was delivered by a syringe pump. Thermocouple was used to monitor the catalyst temperature. Catalyst screening and optimization identified the best catalyst formulation with a conversion rate for carbon monoxide and toluene at room temperature of 1 and 0.25 mmolc g min1. Carbon monoxide and water were the only products of the reactions. 

Helium is the most common carrier gas and is compatible with most detectors. For a flame ionization detector, N2 gives a lower detection limit than He. Figure 24-11 shows that H2, He, and N2 give essentially the same optimal plate height (0.3 mm) at significantly different flow rates. Optimal flow rate increases in the order N2 < He < H2. Fastest separations can be achieved with H, as carrier gas, and H2 can be run much faster than its optimal velocity with little penalty in resolution.11 Figure 24-12 shows the effect of carrier gas on the separation of two compounds on the same column with the same temperature program. 

Soon after the flame ionization detector was announced in 1958, many different research groups began systematic studies to optimize the detector. In only a few years, the device was mature. By the mid-1960s all of the commercial instruments began to resemble each other in the inportant respects. 

Procedure (See Chromatography, Appendix IIA.) Use a gas chromatograph equipped with a flame-ionization detector and a 4-m x 2-mm (id) stainless-steel column, or equivalent, packed with 15%, by weight, methyl trifluoropropyl silicone (DCFS 1265, or QF-1, or OV-210, or SP-2401) stationary phase on 80- to 100-mesh Gas Chrom R, or the equivalent. Condition a newly packed column at 120° and with a 30-mL/ min helium flow for at least 2 h (preferably overnight) before it is attached to the detector. For analysis, maintain the column isothermally at 105° the injection port and detector at 250° the carrier gas flow rate at 11 mL/min with fuel gas flows optimized for the gas chromatograph and detector in use. Change the experimental conditions as necessary for optimal resolution and sensitivity. The signal-to-noise ratio should be at least 10 1. 

When first put into use—and every few months thereafter— the flow conditions for optimum response of the FID should be determined. This can be done by the time-honored method of repeated injections while varying the flow rate of air and especially of hydrogen or by a faster method recently publicized (18). This test takes but a few minutes to execute but can improve analytical results considerably. To even mention FID optimization may well be redundant. It has been my experience, however, that most gas chromatographs equipped with flame ionization detectors are run under less than ideal flow conditions. In trace analysis, this oversight may be crucial. 

The manufacturer s instructions regarding flow rates of carrier gas and the optimization of gas flows to, say, the flame ionization detector, are usually satisfactory, but the following points, while seemingly rather obvious, are nevertheless very important if the apparatus is to function properly. 

Among the variety of detectors, only the thermal conductivity detector (TCD) and the flame ionization detector (FID) are in broad use in PGC. The flame photometric detector, typically used for measuring trace sulfur containing species, and the photoionization detectoi predominately used in environmental monitoring, also see some usage. The variety of detector types available for PGC tends to be limited because of the requirements for robustness and sensitivity to a variety of stream components. In addition, many PGC detectors are not optimized for use with capillary columns. 

Milk fats provide a unique challenge with their high content of short-chain FAs (C4 to CIO) that show a lower than expected flame ionization detector (FID) response in GC. Therefore, the short-chain FAMEs require appropriate correction factors (64-67). In addition, the short-chain FAME are water soluble and can be easily removed by using an aqueous wash. Isopropyl and butyl esters have been used for the analysis of short-chain FA to eliminate the use of correction factors (24,25,64—71), but this requires merging the results of the butyl (or isopropyl) esters with FAMEs (24,25,69,70). In addition to the differences in the FID response of short-chain FAME, attention should also be focused at optimizing the accuracy and reliability of the hydrogen flame in the FID (72). 

Analysis of Fatty Acid Methyl Esters with High Accuracy and Reliability. I. Optimization of Flame-Ionization Detectors with Respect to Linearity, 247 47-61... 

The flame ionization detector (FID) is the most widely used GC detector. Its operation is simple, sensitivity is of the order of lOpg carbon per second with a linear range of 10, the response is fast, and detector stability is excellent. The carrier gas is mixed with hydrogen, and this mixture is combusted in air at the exit of a flame jet. Ions are formed that are collected at an electrode producing a current that is proportional to the amount of sample compound in the flame. Analytes such as permanent gases, nitrogen oxides, sulfur oxides, carbon oxides, carbon disulfide, water, formic acid, formaldehyde, etc., do not provide a significant FID response. The flows of carrier and combustion gas should be set properly for optimal FID operation. Typical flow rates are 1 1 12... 

PerkinElmer Corp. produces the Photovac MicroFID Handheld Flame Ionization Detector (Figure 11.7). This detector was selected for testing based on a survey of existing detection devices by Battelle Memorial Institute. The survey identifled the detectors most likely to be used by local responders in the event of a terrorist incident involving CWA(s). No attempt was made to optimize chemical agent detection capability. No pretest theoretical assessment was made on the detectors except to learn operating procedures from the manufacturer s user manual. 

These workers used a prototype spectraspan III dc plasma echelle spectrometer, 510-512, (Spectrametrics Inc., Andover, Mais.). They adapted a Varian 1200 gas chromatograph for on-column injection onto a 6 ft X g in. o.d. stainless steel column packed with 2% Dexsil 300 GC on Chromosorb 750, 100 120 mesh (Johns-Manvilie Corp., Denver, Col.). Column effluent was split by an approximately 1 1 ratio between the flame ionization detector of the gas chromatograph and a heated, thermal, and electrically insulated 1/16-in. o.d. stainless steel transfer line to the dc plasma. Preheated argon sheath gas was required in addition to the argon supplied to sustain the plasma, in order to optimize spectral sensitivity. The column and injection port temperature were set at 130 and 160 C, respectively, and the interface temperature was 170 0. Helium carrier gas flow rate was 25 ml/min. 

Nitrogen-phosphorous detector (NPD, thermionic detector, alkali flame ionization detector) 4 X 10- g to 1 X 10"gof nitrogen compounds 1 X 10" g to 1 X 10" gof phosphorous compounds 1 X 10 10 to 10 by mass selectivity of N or P over carbon Does not respond to inorganic nitrogen such as Nj or NH Jet gas flow rates are critical to optimization Response is temperature dependent Used for trace analysis only, and is very sensitive to contamination Avoid use of phosphate detergents or leak detectors Avoid tobacco use nearby Solvent-quenching is often a problem... 

Gas chromatography conditions. Gaseous pyrolyzate separated on a 10 foot x 1.8 inch o.d. stainless steel column packed with beta,beta -oxydipropionitrile Porasil C Durapak (Waters Associates), 80 to 100 mesh. The carrier gas is argon flowing at the rate of 15 cm 3 min" The temperature of the column oven is initially 90 C 3 min after pyrolysis, the temperature of the column oven is raised to 120 C and maintained there. The flame ionization detector is optimized for fluorocarbons. The complete elution of all fragments takes 30 min. 

A final point about factors. They need not be continuous random variables. A factor might be the detector used on a gas chromatograph, with values flame ionization or electron capture. The effect of changing the factor no longer has quite the same interpretation, but it can be optimized— in this case simply by choosing the best detector. 

---