Adsorption is the process by which molecules of a substance from a gas mixture or liquid solution became attached to a solid or liquid surface. The substance being absorbed is an adsorbate and the absorbing substance, an adsorbent.
Adsorption is originated by the surface forces acting on the solid-gas, solid-liquid, gas-liquid or (in the case of immiscible liquids in contact) the liquid-liquid interface.
In adsorption of gas by a solid surface, one never distinguishes between physical adsorption and chemisorption. In physical adsorption, the main factors are the nature of interacting substances, the forces of interaction on the interface, and the interaction between adsorbate molecules. Of paramount importance is the degree of surface inhomogeneity and the capability of adsorbed molecules to be in translational, rotary, and oscillatory motion. The surface structure and, hence, adsorption properties may change during adsorption. This is most clearly pronounced when the so-called adsorption energy is of the same order as the surface energy of the adsorbent, e.g., ice, paraffin, and polymers.
Equilibrium in physical adsorption is rapidly established and is reversible. It should be born in mind, however, that the actual adsorption rate can be restricted by the mass transfer rate in the gaseous phase or inside the pores of a porous adsorbent, i.e., by the adsorbate supply to the surface. It is believed that physical adsorption is brought about by the same intermolecular forces as vapor condensation. Hence, a conclusion is drawn that the heat of physical adsorption is close to that of condensation and the amount of physical adsorption is particularly high when the gas has a temperature below critical, i.e., in a vapor.
Gas adsorption by a solid surface is expressed quantitatively as a gas volume adsorbed by the adsorbent per unit mass or unit surface of the adsorbent.
Chemisorption, in contrast to physical adsorption, involves chemical reactions on the surface, which may result in its rearrangement. Cases are also known where physical adsorption proceeds over a chemisorbed layer.
The parameter describing the physical adsorption dynamics is adsorption time, which is the mean lifetime of the molecule on the adsorbent surface. It is t = t0 exp Q/RT, where t0 is the time of molecular vibration (t0 = 10−13 - 10−12 s) and Q is the energy of the molecule-surface interaction, i.e., the heat of adsorption. If thermal equilibrium is established between the molecules and the surface, then the molecules being desorbed leave the surface in the direction independent of the incidence direction, i.e., the accommodation coefficient is assumed to be unity).
An important characteristic of adsorption is also a molecule concentration on an adsorbent surface. It is measured in mol/cm2 and calculated as Г = zt, where z is the number moles of gas colliding with unit surface in a second.
At a given temperature T, adsorption can be classified according to the parameters shown in Table 1.
Table 1. Regions of adsorption
|No adsorption, specular reflection
|of molecules, the accommodation
|coefficient is zero
|The region of physical adsorption, the
|accommodation coefficient is one
|The region of chemisorption
An isotherm, i.e., the dependence graphed as a function V = f(p), where V is the volume adsorbed and p the pressure, is the most common representation of adsorption. In 1918 Langmuir derived theoretically an adsorption isotherm equation that became known as the Langmuir equation. He assumed that the adsorbent surface possesses a certain number of adsorption sites s , part of which are occupied by adsorbed molecules s1 and the other part are vacant s2. He considered adsorption-desorption as a process of condensation and evaporation. The rate of condensation was taken to be proportional to the number of free surface sites s2, i.e., k2s2, while the rate of evaporation proportional to the number of occupied surface sites, i.e., k1s1. When the equilibrium was attained k1s1 = k2ps2 = k2p(s-s1 ) or θ = bp/(1+b ), where p is the partial pressure of an adsorbate, θ = s1/s is the fraction of the surface occupied by adsorbed molecules and b is the Langmuir constant. Taking into account the fact that the evaporation rate is proportional to the saturation vapor pressure pG , the Langmuir adsorption isotherm equation can be reduced to θ = (cx)/(1 + cx ), where c = exp Q/RT and x = (p/pG ) is the ratio of the vapor pressures of a given substance in the mixture to its saturation vapor pressure at the same temperature T. The heat of adsorption Q can be assumed to be equal to the extent of heat of vaporization per mole. R is the universal gas constant related to a mole of vapor.
The next step in the description of adsorption isotherm is the allowance for formation of multimolecular layer, which was realized in the BET (Brunauer-Emmet-Teller) equation.
where V and Vm are respectively the actually adsorbed gas volume and its limit in a monolayer. This equation has gained recognition as a most reliable adsorption isotherm equation and has been widely employed for determining the surface per unit volume from the experimental data. It can be transformed to the form
Plotting x/V(x − 1) against x gives a straight line the intercept and slope of which is used to find the constants Vm and c.
The diversity of experimental adsorption isotherms is depicted in Figure 1. Type 1 isotherms correspond to the Langmuir equation and are characterized by a monotonic approach to a limit value conforming to a complete monolayer surface coverage by a monolayer of adsorbed molecules. Type 2 corresponds to the case when first, a monolayer is formed, and then (from the point B) a multimolecular layer predominantly develops. Isotherms of Type 3 are relatively uncommon and correspond to the formation of multimolecular layer when heat of adsorption is less than or equal to adsorbate heat of condensation. Types 4 and 5, as a rule, are characteristic of capillary adsorbate condensation in porous solids.
Adsorption in porous solids with pores substantially exceeding the free path length of molecules is of the same nature as it is on an open surface of a nonporous solid. In particular, at temperatures higher than the critical temperature, no multimolecular layers are formed. An adsorbent with molecular size pores behaves like a Molecular Sieve, and its apparent surface per unit volume depends on the size of adsorbed molecules: small molecules pass into pores and adsorb on their surfaces, while large ones are trapped and no adsorption occurs in pores.
Of great interest are zeolites composed of tetrahedral groups (Al, Si) O4 . They have large voids into which the molecules of adsorbed gas have to pass through much smaller holes. Thus, in chabazite Ca Al2Si4O12, there are six inlet holes about 4 Å in diameter in the void 10 Å in diameter. These holes accommodate only monatomic or diatomic gases, H2O, and H-alkanes, but trap larger molecules. Changing the hole size makes it possible to separate gases.
At temperatures below critical, for multimolecular adsorption in small size pores not only is the number of layers restricted but also so-called capillary condensation may be observed. For capillary condensation to occur, the effective curvature of the meniscus in the micropores must be sufficient to entirely fill the pores with condensing liquid; however, since the curvature is variable even with a narrow range of pore size, the desorption and adsorption curves do not coincide leading to a hysteresis in adsorption (Figure 2).
Chemisorption gives rise to bonds between adsorbate and adsorbent approaching chemical bonds in strength. In this case, the chemical properties of adsorbate and adsorbent may differ from their initial ones. Chemisorption conspicuously shows up in formation of the first molecular layer because in this case, an enhanced heat of adsorption characteristic of chemisorption comes into play. Considerable variation may occur because of inhomogeneities in the surface or in the energy distribution of adsorption centers. The inhomogeneity may be attributed to surface defects and impurity inclusions, i.e., it depends on the material treatment and quality.
As the second and subsequent molecular layers are formed, chemisorption is reduced to physical adsorption The distinguishing feature of chemisorption, in contrast to physical adsorption, is the fact that it can proceed with gases, whose temperature is above critical.
The rate of chemisorption is described by the expression
where f(θ) is a function depending on the nature of adsorption; k2, the constant of adsorption rate; k2 = k exp( /RT); and is the activation energy which, taking into account its dependence on the surface coverage θ, is described by the binomial = . Hence, at a given temperature, the rate of adsorption varies with surface coverage by molecules in proportion to f(θ) exp(−αθ).
The activation energies of desorption and adsorption differ by the heat of adsorption Q: = . The rate of desorption is described as Rd = dθ/dτ = kf'(θ) exp( + Q/RT), where f'(θ) is analogous to f(θ).
Chemisorption is an important underlying phenomenon in heterogeneous catalysis.
In adsorption from solutions, the behavior depends greatly on whether the solutions are of nonelectrolytes or electrolytes. Adsorption from nonelectrolyte solutions depends on adsorbate concentration and, in the case of dilute solutions, is similar to gas adsorption. The solvent properties manifest themselves at high concentrations.
Two basic models of nonelectrolyte adsorption exist. Adsorption in terms of the first model is restricted to a monolayer directly on the surface, whereas the consecutive layers are merely an ordinary solution. This resembles gas chemisorption, but the heat of adsorption from a solution is low in relation to the energy of chemical reaction. In the second model, adsorption occurs in a fairly thick (up to 100 Å) multimolecular interphase layer in a decreasing potential field of the solid surface.
Both models are consistent with experiment and yield the same mathematical expressions, e.g., an adsorption isotherm: θ/(1 − θ) = ka2/a1, where a1 and a2 are activities of the solvent and the solute in the solution. This is obviously an expression similar to the Langmuir isotherm equation.
Adsorption in electrolyte solutions is considered using several approaches. If an electrolyte is assumed to be adsorbed as an entity bound by opposite charged ions, then this resembles adsorption in nonelectrolyte solutions. However, more frequently, the ions of one sign are more strongly attracted by the surface than the ions of the other sign. The latter produce near the surface a diffuse layer through which the adsorbed ions are forced to diffuse.
A wide range of adsorbents are used for adsorption from gases and liquids. Special adsorbents are often developed for specific purposes. Among the most well-known adsorbents are activated carbon used for adsorbing gases, including hydrocarbons, and silica gel for adsorbing water vapor.
A technological apparatus in which adsorption is implemented is called an adsorber. Adsorbers can have an immovable adsorbent bed and operate either with periodic regeneration under an adsorption-desorption regime or without regeneration but with periodic change of the adsorbent.
In fixed-bed adsorbers, the adsorption process is represented by the breakthrough curve (Figure 3). The concentration of pollutant C in the exit gas stream is plotted versus time. Initially, the adsorbent adsorbs the pollutant gas from the stream readily and efficiently, so that the pollutant concentration of the outlet stream is close to zero (period 0-a on the abscissa in Figure 3). In this period, the bulk of the adsorption is taking place near the inlet with the rest of the bed removing the traces. Eventually, the inlet region becomes saturated and the main region of adsorption moves towards the outlet. Traces of the adsorbate begin to appear at the outlet. When the outlet concentration begins to rise rapidly (B), the so-called “break through point” has been reached. Though, at this time, the exit pollutant concentration may be below the required emission standard Cstd after the point of breakthrough, the exit concentration rises rapidly towards the inlet concentration C0. Thus, at time B, regeneration is required. A steep breakthrough curve is more desirable than a flat one. For a steep curve, the bed saturation may reach 80 per cent, but with a flat curve only 15-20 per cent may be typical before breakthrough. The calculation methods for adsorption, particularly in porous adsorbents, mist allow for heat and mass transfer to the adsorbed substance in pores.
Adamson, A. (1976) Physical Chemistry of Surfaces, 3rd ed., Wiley, New York.
Perry, R. H., Green, D. W., and Maloney, J. O. (eds.) Perry’s Chemical Engineers Handbook (1984), McGraw-Hill, New York.
- Adamson, A. (1976) Physical Chemistry of Surfaces, 3rd ed., Wiley, New York.
- Perry, R. H., Green, D. W., and Maloney, J. O. (eds.) Perryâ€™s Chemical Engineers Handbook (1984), McGraw-Hill, New York.