Separation of Hydrogen and Carbon Monoxide

Another important application is the removal of carbon monoxide from the valuable hydrogen produced in steam reformed natural gas. 

Alumina membranes made by anodic oxidation were tested for separating hydrogen and carbon monoxide at a temperature up to 9T C [Itaya, 19S4], The ideal separation factor only achieved 3.5. See Table 7.4. Wu et al. [1993] also examined the use of alumina membranes for this application at an even higher temperature of 300 C and obtain a maximum separation factor of only 2.2. The alumina membranes used in the above studies all have pore diameters larger than 4 nm. Their separation factors for the H2/CO gas pair are no higher than the Knudsen diffusion values of 3.7 most likely as a result of the limitation imposed by the relatively large pores involved. 

Separation of hydrogen and carbon monoxide by inorganic membranes 

Wu et al. (1993] have developed a mathematical model based on Knudsen diffusion and intermolecular momentum transfer. Their model applies the permeability values of single components (i.e., pure gases) to determine two parameters related to the morphology of the microporous membranes and the reflection behavior of the gas molecules. The parameters are then used in the model to predict the separation performance. The model predicts that the permeability of carbon monoxide deviates substantially from that based on Knudsen diffusion alone. Their model calculations are able to explain the low gas separation efficiency. Under the transport regimes considered in their study, the feed side pressure and pressure ratio (permeate to feed pressures) are found to exert stronger influences on the separation factor than other factors. A low feed side pressure and a tow pressure ratio provide a maximum separation efficiency. 

On the contrary, silica membranes with pore diameters in the 0.3-0.8 nm on porous glass supports have been claimed to exhibit an H2/CO separation factor as high as 30. This is significantly higher than what Knudsen diffusion alone predicts. It is possible that some molecular sieving takes place in the reportedly small pores in the silica membranes. This postulation appears to be further reinforced by another study using hollow fiber silica 

---

Gu C., Gao G. H., Yu Y. X. and Nitta T., Simulation for separation of hydrogen and carbon monoxide by adsorption on single-walled carbon nanotubes. Fluid Phase EquUibria 9A-W1 (2002) pp. 297-303. 

CA blend membranes for gas separation are commercially available from Envlrogenlcs Systems Co. (El Monte, CA), Separex Corp. (Anaheim, CA) and Grace Membrane Systems (Houston, TX), and are applied In spiral-wound modules for the separations of acidic gaseous components from natural gas, for the recovery of carbon dioxide In enhanced oil recovery processes for gas dehydration or the separation of hydrogen from carbon monoxide (21-23). 

High pressure technology transfer and diversification took many avenues, though most new innovations continued to appear from BASF. First, in 1923, was methanol production at the Leuna ammonia factory, and based on the work of Matthias Pier. BASF had patented a high pressure methanol process in 1914, but no further studies were carried out until after G. Patart in France applied for a similar patent (1921). In this case the same equipment could be used to manufacture ammonia or methanol, according to demand. Synthesis gas, the mixture of hydrogen and carbon monoxide, was used directly, without separation, to prepare methanol. In a similar way, isopropanol was manufactured under high pressures. 

As is visible in Figure 4.14, a usage of 20 h is enough to produce an evident porosity of the dense layer also, an effect of superficial corrosion, due to the filamentous carbon, can be seen from the scanning electron microscopy (SEM) analysis (Galuszka et al., 1998). All the experimental evidence showed that these kinds of membranes are useful for the separation of methane and carbon monoxide-free streams, because while the swelling, due to hydrogen, can be controlled in a certain way, the same cannot be said for filamentous carbon formation. 

Another technique to separate a mixture of hydrogen and carbon monoxide is to, under pressure, selectively adsorb the carbon monoxide on zeolites or some other adsorbent. Purified hydrogen passes through the adsorbent. Release of the pressure then desorbs the carbon monoxide. This is called Pressure Swing Adsorbtion or PSA. 

The anodic oxidation of fuels in low temperature cells, mainly on platinum metals, platinum metal alloys and alloys of platinum metals with other metals, is the subject of this chapter. Most oxidation studies were made on these metals because the efficiency of other electrocatalysts is too low. The mechanism for the oxidation of carbon monoxide, nlixtures of hydrogen and carbon monoxide, formic acid, methanol, higher alcohols, hydrocarbons, and hydrazine is discussed in separate sections. 

Separation of Methane from Hydrogen and Carbon Monoxide by an Absorption/Stripping Process... 

Figure 1 shows the vapor pressure of some of the relevant compounds as a function of temperature. At -230.8°F and -184°F, the vapor pressure of propane is about 0.1 mm Hg and 1 mm Hg respectively. Absorption/stripping process is a conventionally practical process so it is decided to evaluate the separation of methane from hydrogen and carbon monoxide by this process. Since the operating pressures of the absorber and stripper are about 500 psia and 487 psia respectively, the mole fractions of propane in the outlet gas streams of the absorber and stripper are about 6.75 x 10 6 and 1 x 10 4 at -230.8°F and -184°F respectively. 

---