Adsorption process design pressure

The basic adsorption process design. Sub-tasks within that include the adsorbent selection, made in view of aU of the requirements imposed on the dehydration process. The adsorption step time, regeneration and cooHng step times all need to be settled and these in view of mechanical details. The overall vessel configuration, for example, the vessel ID and length, which quantities are typically sized based on pressure drop. Finally we need to make some estimate of the expected service Hfetime for the adsorbent product. 

Mechanical design of the adsorber then takes up the remainder of the engineering effort to produce a workable adsorption process design. Once a vessel is sized to provide the required inventory of adsorbenf we need to provide the mechanical details, which include flow distribution devices, bed supports and the required vessel wall thickness to withstand the working pressure and added stresses encountered during regeneration and repeated de-pressurization and re-pressurization. 

The range of applications for gas and liquid separation and purification by adsorption is large and growing. The strong research and development activity in this area is facilitated by the flexibility of practical adsorptive process designs such as pressure and thermal swing adsorption, and SMB adsorption, as well as by the availability of a large spectrum of new and old micro-and mesoporous adsorbents. 

FIG. 16-46 Pressurized adsorber vessel. (Reptinted with peirrussion of EPA. Reference EPA, Process Design Manual for Carbon Adsorption, U.S. Envir. Protect. Agency., Cincinnati, 1973.)... 

Data used for the design of adsorption processes are normally derived from experimental measurements. The capacity of an adsorbent to adsorb an adsorbate depends on the compound being adsorbed, the type and preparation of the adsorbate, inlet concentration, temperature and pressure. In addition, adsorption can be a competitive process in which different molecules can compete for the adsorption sites. For example, if a mixture of toluene and acetone vapor is being adsorbed from a gas stream onto activated carbon, then toluene will adsorbed preferentially, relative to acetone and will displace the acetone that has already been adsorbed. 

For optimum efficiency, humidity levels, temperature, and pressure should be monitored and controlled during the adsorption. The adsorption process of VOCs removal is exothermic in the most cases, which should be considered as a significant design parameter, since there is a risk of fire in the removal of high loads of organic compounds that exhibit high heats of adsorption. 

Another adsorption system evaluated was a high-volume, high-pressure, macroporous-resin-based concentrator system designed to provide a 10,000-fold concentrate. This system used four stainless steel columns in series. Columns one through four were filled with AG MP-1 (Bio-Rad), AG MP-50 (Bio-Rad), XAD-2 (Rohm and Haas), and XAD-7 (Rohm and Haas), respectively. Unlike the other adsorption processes, a pump system was employed for both the adsorption and desorption phases of the evaluation. The use of acetonitrile as the elution solvent permitted UV monitoring of the eluant. 

Adsorption equilibria is important fundamental property to design and develop the adsorption process. We have measured the adsorption equilibrium constant of limonene and linalool in SC-C02 by an impulse response technique [9]. Figure 1 shows the adsorption equilibrium constant at a temperature of 313 K - 333 K and a pressure of 11.8 MPa - 23.5 MPa. Linalool was adsorbed more selectively than limonene. Adsorption equilibrium constants were correlated linearly in log-log plot as a function of the density of SC-C02 independent of pressure and temperature. Adsorbed amounts decreased with the increase in the solvent density for both limonene and linalool. These results suggest the possibility of a process where oxygenated compounds are selectively adsorbed on the adsorbent at a lower pressure and then desorbed at a higher pressure. 

Selectivity is a key variable that affects the adsorption process and is essential for design. The variation of selectivity of ethane with pressure and composition is shown in a 3D graph in Fig. 2. Pure component data yields only the line AB at zero pressure, which is the ratio of Henry s constants. Using only this information it is not possible to accurately estimate the variation in selectivity. The two models differ substantially with respect to selectivity predictions. In the Langmuirian approach the selectivity is constant and is given by the ratio of Henry s constants (along a horizontal plane through AB). Selectivity by lAST approaches the same limit at zero pressure but rapidly decreases with pressure. 

Pressure swing adsorption processes are also designed to produce high-purity (99.95+ %) H2 products from refinery-off gases containing H2 (65-90%) and C1-C5 hydrocarbon impurities with high H2 recoveries ( 86+ %). Silica gel and activated carbons are used as adsorbents. 

Seider, W.D., Seader, J.D. and Lewin, D.R. (2004) Product Process Design Principles, Synthesis, Analysis and Evaluation, 2nd edn, John WUey Sons Inc. New York Knaebel, K.S. and Cussler, E.L. (1996) A novel pressure swing adsorption system for ammonia synthesis, in Fundamentals of Adsorption (ed. M.D. Le Van), Kluwer Academic Publishers, New York USA. 

Alternative separation operations can be inserted into Figure 3.5. When distillation is used, it is also possible to recover the least volatile species, dichloroethane, from the first column, and separate HCl from vinyl chloride in the second column. Yet another possibility is to use a single column with a side stream that is concentrated in the vinyl chloride product. Absorption with water, at atmospheric pressure, can be used to remove HCl. The resulting vapor stream, containing vinyl chloride and dichloroethane, could be dried by adsorption and separated using distillation. With so many alternatives possible, the process designer needs time or help to select the most promising separation operations. As mentioned previously, this topic is considered in detail in Chapter 7. 

Adsorption measurement for multicomponent systems is a function of the composition, temperature, pressure, and properties of adsorbate and adsorbent. As the number of components increases, the number of measurements needed to define the adsorption equilibrium increases rapidly and eventually becomes infeasible. Adsorption equilibrium models are therefore needed to correlate and predict the multicomponent adsorption equilibria. These models should be able to predict the mixture equilibria using the information available on pure component equilibria, as the latter are relatively easy to measure and furthermore there is an abundance of pure component isotherm data available in the literature. As a result, predictive models for gas mixture adsorption are necessary in the design and modeling of adsorption processes. 

---