Methanizer

Methanizer is an appliance used in gas chromatography, which allows to detect very low concentrations of carbon monoxide and carbon dioxide. It consists of a flame ionization detector, preceded by a hydrogenating reactor, which converts CO₂ and CO into methane CH₄.

Chemical Reaction

On-line catalytic reduction of carbon monoxide to methane for detection by FID was described by Porter & Volman,^[1] who suggested that both carbon dioxide and carbon monoxide could also be converted to methane with the same nickel catalyst. This was confirmed by Johns & Thompson,^[2] who determined optimum operating parameters for each of the gases.

CO₂ + 2H₂ ↔ CH₄ + O₂

2CO + 4H₂ ↔ 2CH₄ + O₂

Typical Design

The catalyst usually consists of a 2% coating of Ni in the form of nickel nitrate deposited on a chromatographic packing material (e.g. Chromosorb G). A 1½" long bed is packed around the bend of an 8"×1/8" SS U-tube. The tube is clamped in a block so that the ends protrude down into the column oven for easy connection between column or TCD outlet and FID base. Heat is provided by a pair of cartridge heaters and controlled by a temperature controller.

Hydrogen for the reduction can be provided either by adding it via a tee at the inlet to the catalyst (preferred), or by using hydrogen as carrier gas.

An alternative design has been suggested recently,^[3] which catalytically combusts all organic species to CO₂ before reduction to methane. This has several benefits including the detection of many more organic molecules and resistance to poisoning. A commercial version of the device called the Polyarc reactor is available from Activated Research Company.^[4]

Start-up

Since the raw catalyst is supplied in the form of nickel oxide, it is necessary to reduce it to metallic nickel before it will operate properly.

The following procedure is recommended:

• Condition all columns in the normal way. Never condition a column while connected to the catalyst.
• Connect columns to detectors(s) and catalyst as required by the application.
• Set normal carrier gas flow (25-30 mL/min for 1/8" columns). Either He or N₂ is satisfactory.
• Set H₂ flow to the catalyst at about 20 mL/min and H₂ to the FID at 10 mL/min. If H₂ is used as carrier, 20 mL/min N₂ or He make-up should be provided through the normal FID H₂ make-up line.
• When the detectors are up to operating temperature, set the column oven temperature as required, turn on the catalyst heater, and set to 400 °C.
• By the time the injector temperature reaches 400 °C, the catalyst will be reduced and ready for use.

Inject a sample containing known amounts of CH₄, CO and CO₂ to check conversion efficiency and peak shape. The retention times of these compounds should be known. If not, and light hydrocarbons are present in the sample, there might be some confusion in identification. The user should be aware, also, that the FID does respond slightly to O₂ so at high sensitivities an air peak might also be evident. As a very rough indication, 1% O₂ gives a signal similar to that of 1 ppm CO or CO₂.

If there is any doubt about retention times, the following pointers might be useful:

• On a Mol. Sieve 5A, CO retention time is about three times that of CH₄.
• On a Mol. Sieve 13X, CO has about 25% longer retention time than CH₄.
• On a porous polymers and silica gel, CO elutes with air just before CH₄, and CO₂ elutes between CH₄ and C₂H₆ except on Chromosorb 104, from which CO₂ elutes just after C₂H₆. It also elutes just after ethane from silica gel and the retention time is considerably longer than on porous polymers.
• For confirmation of CO₂, atmospheric air contains about 300 ppm and a breath sample 5-15%.

If necessary, adjust catalyst temperature to optimize conversion efficiency and peak symmetry. Also adjust the H₂ flow to optimize sensitivity. The H₂ flow through the catalyst and the ratio of H₂ to catalyst and H₂ to FID are not critical.

Operating Characteristics

Temperature

Conversion of both CO and CO₂ to CH₄ starts at a catalyst temperature below 300 °C, but the conversion is incomplete and peak tailing is evident. At around 340 °C, conversion is complete, as indicated by area measurements, but some tailing limits the peak height. At 360-380 °C, tailing is eliminated and there is little change in peak height up to 400 °C.

Although carbonization of CO has been reported at temperatures above 350°,^[5] it is rather a rare phenomenon.

Range

The conversion efficiency is essentially 100% from minimum detectable levels up to a flow of CO or CO₂ at the detector of about 5×10⁻⁵ g/s. These represent a detection limit of about 200 ppb and a maximum concentration of about 10% in a 0.5 mL sample. Both values are dependent upon peak width.

Catalyst Poisoning

Some elements and compounds can deactivate the catalyst:

• H₂S. Very small amounts of H₂S, SF₆, and probably any other sulfur containing gases, cause immediate and complete deactivation of the catalyst. It is not possible to regenerate a poisoned catalyst that has been deactivated by sulfur, by treating with either oxygen or hydrogen. If sulfur containing gases are present in the sample, a switching valve should be used either to bypass the catalyst, or to back-flush the column to vent after elution of CO₂.
• Air or O₂. Reports of oxygen poisoning seem to be rather rumors than real facts. Small amounts of air through a catalyst will not kill it but anything over about 5 cc/min will cause an immediate and continual degradation of the catalyst. This has been seen first hand on several systems over 30 years of personal experience with a catalytic FID designed for analysis of U.S. EPA Method 25 and 25-C samples.
• Unsaturated hydrocarbons. Samples of pure ethylene cause immediate, but partial, degradation of the catalyst, evidenced by slight tailing of CO and CO₂ peaks. The effect of 2 or 3 samples might be tolerable, but since it is cumulative, such gases should be back-flushed or by-passed. Low concentrations do not cause any degradation. Samples of pure acetylene affect the catalyst much more severely than does ethylene. Low concentrations have no effect. Probably some carbonization with high concentrations of unsaturates occurs, resulting in the deposit of soot on the catalyst surface. It is likely that aromatics would have the same effect.
• Other compounds. Water has no effect on the catalyst, as well as various Freons and NH₃. Here again, with NH₃, there is conflicting evidence from some users, who have seen a degradation after several injections, but other researchers were not able to confirm it. As with sulfur containing gases, NH₃ can be back-flushed to vent or by-passed if desired.

Troubleshooting

In general, the catalyst works perfectly unless it is degraded by sample components, possible minute amounts of sulfur gases at otherwise undetectable levels. The effect is always the same — the CO and CO₂ peaks start to tail. If only CO tails, it might well be a column effect, e.g., a Mol. Sieve 13X always causes slight tailing of CO. If the tailing is minimal, raising the catalyst temperature might provide enough improvement to permit further use.

With a newly packed catalyst, tailing usually indicates that part of the catalyst bed is not hot enough. This can happen if the bed extends too far up the arms of the U-tube. Possibly a longer bed will improve the upper conversion limit, but if this is the aim, the packing must not extend beyond the confines of the heater block.

Catalyst Preparation

Dissolve 1 g of nickel nitrate Ni(NO₃)₂•6H₂O in 4-5 mL of methanol. Add 10 g of Chromosorb G. A/W, 80-100 mesh. There should be just enough methanol to completely wet the support without excess. Mix the slurry, pour into a flat Pyrex pan and dry on a hot plate at about 80-90 °C with occasional gentle shaking or mixing. When dry, heat in air at about 400 °C to decompose the salt to NiO. Note that NO₂ is emitted during baking — provide adequate ventilation. About an hour at 400 °C, longer at lower temperatures, will be needed to complete the process. After baking, the material is dark gray, with no trace of the original green.

Pour the raw catalyst into both arms of an 8"×1/8" nickel U-tube, checking the depth in both with a wire. The final bed should extend 3/8" to 1/2" above the bottom of the U in both arms. Plug with glass wool and install in the injector block.

Disadvantages

Methanizers are limited by their ability to react only CO and CO₂ to methane and their deactivation by compounds commonly found in chemical samples. These include olefins and sulfur containing compounds. Thus, the use of methanizers typically require complex valve systems that may include backflush and heartcuts. These systems can work well, but can add cost and complexity, and the potential for leaks and adsorption in the chromatographic flow path.

Alternative Solution

An alternative methanizer variant that overcomes these limitations and allows for the direct injection of all compounds without backflush or heartcuts is a two-step oxidation and subsequent reduction reactor to convert nearly all organic compounds to methane.^[6] This technique enables the accurate quantification of any number of compounds that contain carbon beyond just CO and CO₂, including those with low sensitivity in the FID such as carbon disulfide (CS₂), carbonyl sulfide (COS), hydrogen cyanide (HCN), formamide (CH₃NO), formaldehyde (CH₂O) and formic acid (CH₂O₂). In addition to increasing the sensitivity of the FID to particular compounds, the response factors of all species become equivalent to that of methane, thereby minimizing or eliminating the need for calibration curves and the standards they rely on. The reactor is available exclusively from the Activated Research Company^[4] and is known as the Polyarc reactor.

References

1. Porter, K. and Volman, D.H., Anal. Chem 34 748-9 (1962).
2. Johns, T. and Thompson, B., 16th Pittsburgh Conference on Analytical Chemistry and Applied Spectroscopy, Mar. 1965.
3. Maduskar, S., Teixeira, AR., Paulsen, A.D., Krumm, C., Mountziaris, T.J., Fan, W., and Dauenhauer, P.J., Lab Chip, 15 (2015) 440-7.
4. "Activated Research Company". ARC.
5. Hightower F.W. and White, A. H., Ind. Eng. Chem. 20 10 (1928)
6. Dauenhauer, Paul (January 21, 2015). "Quantitative carbon detector (QCD) for calibration-free, high-resolution characterization of complex mixtures". Lab Chip. 15 (2): 440–7. doi:10.1039/c4lc01180e.