Carbon dioxide absorption columns

Eckert et al. report test data for carbon dioxide absorption from air in a packed column. Conditions for a particular test are as follows ... 

First, the overall mass transfer coefRcient k a of the microreactor was estimated to be 3-8 s [43]. For intensified gas liquid contactors, kj a can reach 3 s while bubble columns and agitated tanks do not exceed 0.2 s Reducing the flow rate and, accordingly, the liquid film thickness is a means of further increasing kj a, which is limited, however, by liquid dry-out at very thin films. Despite such large mass transfer coefficients, gas-liquid microreactors such as the falling film device may still operate between mass transfer and kinetic control regimes, as fundamental simulation studies on the carbon dioxide absorption have demonstrated [44]. Distinct concentration profiles in the liquid, and even gas, phase are predicted. 

Example 143.a-1 Design of a Packed Column for Carbon Dioxide Absorption after Danckwerts and Sharma 15 )... 

Design of a packed column for carbon dioxide absorption. After Danckwerts and Sharma [1966]. 

Tepe, J. B. and B. F. Dodge, Absorption of Carbon Dioxide Sodium Hydroxide Solutions in a Packed Column, Trans. Amer. Inst. Chem. Engrs. 39, 255 (1943). 

Massimilla et al. (M5) measured the rate of absorption of carbon dioxide in water from a mixture of carbon dioxide and nitrogen. Used as solid phase were silica sand particles of average equivalent diameter 0.22 mm, or glass ballotini of average equivalent diameter 0.50 and 0.80 mm. Columns of 30-and 90-mm i.d. were used, and the column height was varied from 100 to 1200 mm. 

Example 11.7 Carbon dioxide is sometimes removed from natural gas by reactive absorption in a tray column. The absorbent, typically an amine, is fed to the top of the column and gas is fed at the bottom. Liquid and gas flow patterns are similar to those in a distillation column with gas rising, liquid falling, and gas-liquid contacting occurring on the trays. Develop a model for a multitray CO2 scrubber assuming that individual trays behave as two-phase, stirred tank reactors. 

From the quench scrubber (B) the gas passes to an absorption column (C) in which the acrylonitrile is absorbed in water to produce a 3 wt per cent solution. The carbon dioxide, unreacted propylene, oxygen, nitrogen and unreacted hydrocarbons are not absorbed and are vented to atmosphere from the top of column (C). 

Absorption experiments in columns packed with spheres, 37.8 mm diameter, were also carried out by Davidson el al.<6> who absorbed pure carbon dioxide into water. When a small amount of surface active agent was present in the water no appreciable mixing was... 

In an attempt to test the surface renewal theory of gas absorption, Danckwerts and Kennedy measured the transient rate of absorption of carbon dioxide into various solutions by means of a rotating drum which carried a film of liquid through the gas. Results so obtained were compared with those for absorption in a packed column and it was shown that exposure times of at least one second were required to give a strict comparison this was longer than could be obtained with the rotating drum. Roberts and Danckwerts therefore used a wetted-wall column to extend the times of contact up to 1.3 s. The column was carefully designed to eliminate entry and exit effects and the formation of ripples. The experimental results and conclusions are reported by Danckwerts, Kennedy, and Roberts110 who showed that they could be used, on the basis of the penetration theory model, to predict the performance of a packed column to within about 10 per cent. 

It is difficult to compare the performance of various spray towers since the type of spray distributor used influences the results. Data from Hixson and Scott 33 and others show that KGa varies as G70-8, and is also affected by the liquid rate. More reliable data with spray columns might be expected if the liquid were introduced in the form of individual drops through a single jet into a tube full of gas. Unfortunately the drops tend to alter in size and shape and it is not possible to get the true interfacial area very accurately. This has been investigated by Whitman et a/. 34 , who found that kG for the absorption of ammonia in water was about 0.035 kmol/s m2 (N/m2), compared with 0.00025 for the absorption of carbon dioxide in water. 

Tepe, J. B. and Dodge, B. F. Trans. Am. Inst. Chem. Eng. 39 (1943) 255. Absorption of carbon dioxide by sodium hydroxide solutions in a packed column. 

Ammonium bicarbonate is made by passing carbon dioxide through an aqueous solution of ammonia in an absorption column or a packed tower ... 

In many of these experiments, interfacial turbulence was the obvious visible cause of the unusual features of the rate of mass transfer. There are, however, experimental results in which no interfacial activity was observed. Brian et al. [108] have drawn attention to the severe disagreement existing between the penetration theory and data for the absorption of carbon dioxide in monoethanolamine. They have performed experiments on the absorption of C02 with simultaneous desorption of propylene in a short, wetted wall column. The desorption of propylene without absorption of C02 agrees closely with the predictions of the penetration theory. If, however, both processes take place simultaneously, the rate of desorption is greatly increased. This enhancement must be linked to a hydrodynamic effect induced by the absorption of C02 and the only one which can occur appears to be the interfacial turbulence caused by the Marangoni effect. No interfacial activity was observed because of the small scale and small intensity of the induced turbulence. 

Carbon dioxide at an inlet concentration of 0.1 mole per cent is to be removed from an air stream at a total pressure of 1 bar by bubbling through a 0.0SM (0.05 kmol/m3) solution of NaOH at 20°C in a bubble column. The caustic soda solution passes through the column at such a rate that its composition is not significantly affected by the absorption. The superficial velocity of the gas will be 0.06 m/s. The ratio of outlet concentration of C02 in the air to inlet concentration is to be calculated for the following cases the height of the gas-liquid dispersion will be 1.5 m in each case ... 

Reactive absorption is very old as a processing technique and has been used for production purposes in a number of classical bulk-chemical technologies, such as nitric or sulfuric acid. The Raschig process for the production of hydroxylamine, an important intermediate in classical caprolactam technologies (Stamicarbon, Inventa), is also an example of a multistep reactive absorption process. Here, water, ammonia, and carbon dioxide react together in an absorption column to give a solution of ammonium carbonate, which subsequently forms an alkaline... 

Coke oven gas consists mainly of a mixture of carbon monoxide, hydrogen, methane, and carbon dioxide. It is contaminated with a variety of organic and inorganic compounds that have to be separated in absorption columns before its further use as a synthesis gas. The selective absorption of coke plant gas contamination results from a complex system of parallel liquid-phase reactions. Instantaneous reversible reactions ... 

---