An understanding of the processes involved in adsorption are vital to gaining an insight of the mechanisms involved. The principles of adsorption onto porous adsorbents, in particular activated carbons, are considered with reference to gas and vapour adsorption processes.

Principles of Adsorption
Meaning of Porosity
Classification of Pores
Origin of Porosity
Pore Size Distribution
Adsorption Isotherms

Principles of Adsorption

The term adsorption is said to have been first used by Kayser [1] in 1881 in order to explain the condensation of gases on surfaces, in contrast to gas absorption in which gas molecules penetrate the bulk phase of the absorbing solid. The term ‘sorption’ was proposed by McBain [2] as a complete description of mass transport into a solid, encompassing surface adsorption, absorption by penetration into the solid and condensation within pores.

Adsorption is described as the enrichment of one or more components in the interfacial layer,[3] i.e. an excess of molecules exists at the adsorbate/adsorbent interface, upon exposure of an adsorbing solid to a gas or vapour. It is the selective collection and concentration onto solid surfaces of certain molecules contained in a vapour or gas stream. Hence, vapours or gases referred to as adsorbates when adsorbed, even of mixed systems and at low concentrations, may be captured, often selectively, and removed from the effluent stream using a material of the category of adsorbents.

Adsorption is divided into the two sub-categories of physical adsorption (physisorption) or van der Waals adsorption and chemical adsorption (chemisorption) and the adsorption process can be determined whether chemical bonds are formed during the process. Physisorption is applicable to all adsorbate-adsorbent systems provided the conditions of pressure and temperature are suitable whereas chemisorption may only occur if the system is capable of making a chemical bond.


Physical Adsorption (Physisorption)

The process is a dynamic one where an equilibrium state exists with molecules and the interaction between the adsorbate and adsorbent.

No chemical bonds are formed during physical adsorption; attraction between the adsorbate and adsorbent exists by the formation of intermolecular electrostatic, such as London dispersion forces, or van der Waals forces from induced dipole-dipole interactions, or may be dependent on the physical configuration of the adsorbent such as the porosity of activated carbons. Dispersion forces are the result of rapid fluctuations in the electronic density of one adsorbent molecule inducing an electrical moment in a second atom.[4] If the adsorbate possesses a permanent dipole, or even a multipole, then additional interactions may occur, as charge distributions are induced in the adsorbent and interactions of these moments with any permanent field of the solid.

The process is a very general one and is analogous with that of condensation. Physisorption occurs to varying extents for all adsorbates, gases and vapours, with all adsorbing solids and the effect increases with decreasing temperature or increasing pressure. Physical adsorption is based on certain basic considerations and adsorption on a heterogeneous surface, that is a surface on which the sites are different, occurs at the sites of highest adsorption potential. The process of physical adsorption into the microporous structure of activated carbon follows the theory of Dubinin.[5-7]

The mechanism of adsorption is dependent upon the size of the admolecule in comparison with the pore width due to the energetic interactions between the chosen adsorbate and the pores. Admolecules initially adsorb into the pores with the highest energy, ignoring activated diffusion effects, then adsorption proceeds via filling of progressively larger, or decreasing energy, porosity. Some pores are capable of accommodating two or three admolecules and, therefore, may undergo co-operative adsorption effects by reducing the volume element thus increasing the energy and adsorptive potential of the pore.

The process of adsorption is always exothermic due to the increased ordering of the adsorbate on the adsorbent surface, reducing the entropy, as:


Thus the amount adsorbed should decrease with increasing temperature as a reduction in the thermal energy supplied to the process, by Le Chatelier’s principle, favours the exothermic process of adsorption increasing the equilibrium uptake, except in the case of activated diffusion.

It has been proposed by Lamond and Marsh,[8] by the interpretation of data for physical adsorption of nitrogen on both polar and non-polar surfaces that physical adsorption is independent of the surface chemistry of the adsorbent.


Chemical Adsorption (Chemisorption)

Chemisorption involves the transfer of electrons between the adsorbent and the adsorbate with the formation of chemical bonds, by chemical reaction, between the two species causing adhesion of the adsorbate molecules. Chemical adsorption is far less common than physical adsorption and due to the chemical bonds formed regeneration of the adsorbent for subsequent re-use is often difficult or impossible.[9]

Due to the fact that chemical bonds are formed during the adsorption process, desorption of the adsorbed phase may yield products which are chemically different to the original adsorbate. For example oxygen may chemically bond to the surface of a carbon, which upon desorption may evolve CO and CO2 as products.

Characteristics Associated with Physical/Chemical Adsorption [10]

Physical Adsorption
Chemical Adsorption
Heat of Adsorption / kJmol-1

20 - 40
c.f. heats of liquefaction

> 80 c.f. bulk-phase chemical reactions

Rate of Adsorption
(at 273K)

Temperature Dependence of Uptake (with Increasing T)



Easy- by reduced pressure or increased temperature
Difficult - high temperature required to break bonds
Desorbed Species
Adsorbate unchanged
May be different to original adsorptive
Very Specific
Monolayer Coverage
Mono or multilayer condition dependent



The word pore comes from the Greek word ‘poros’ which means passage. This indicates the role of a pore acting as a passage between the external and the internal surfaces of a solid, allowing material, such as gases and vapours, to pass into, through or out of the solid. Almost all adsorbents that are used in catalysis or for purification/separation purposes possess porosity, and this is the only practical method of introducing greatly enhanced surface areas into a solid.

Activated carbons were the source of several observations made by Bussy.[11] He reported that porosity, although he could not measure it directly, was extremely important and that the decolourising power of a carbon was dependent on the parent precursor and the processes used in its manufacture.

Porosity has been studied extensively by the process of adsorption [4] where exposure of a porous solid to an adsorbate creates a concentration of the adsorbate at the surface, within molecular or atomic distance, greater than the adsorptive concentration in the gas phase. An excess of molecules exists dynamically at the interface in the process called adsorption.

There are four terms generally used to describe the accessibility of porosity:

‘open pore’ - a pore which is connected to the external surface of a solid and allows the passage of an adsorbate through the solid

‘closed pore’ - a void within the solid which is not connected to the external surface and hence is isolated

‘transport pores’ - these connect different parts of the external surface of the solid to the inner microporosity

‘blind pores’ - are connected to transport pores but do not lead to any other pore or surface

It may be useful to explain the terms of internal and external surfaces:

‘internal surface’ - composed of the area which surrounds the closed pores as well as all fissures and cracks which penetrate deeply into the interior of the adsorbent which are deeper than they are wide, and

‘external surface’ - consists of the protrusions and superficial cracks which are wider than they are deep.

The total surface of a carbon material is predominantly composed of internal surface and activated carbons owe their adsorptive properties to the internal surface area and pore size distribution with the external surface area and functional groups playing a comparatively minor role.[12]

The figure below shows the different components of a typical mass transport system of a solid. The total pore volume is essentially the sum of two fractions:

Vt = Vo + Vc

where Vt is the total pore volume, Vo is the pore volume due to the open porosity in the solid, including transport and blind pores, and Vc is the closed pore volume.

Different Types of Porosity

The total pore volume of a carbon is independent of the adsorbate, provided the amount adsorbed is expressed as a liquid volume and excluding selective effects, according to the Gurvitsch rule.[13] An exception to this rule is water as it is assumed that water molecules can not hydrogen bond effectively in the narrow micropores effectively reducing the density of the adsorbed phase hence reducing the pore volume. This phenomenon has been frequently observed in adsorption studies.[14-16] Unfortunately this rule does not take size exclusion effects into account as is often the case with CMS but it can be assumed that this rule is applicable ignoring all other effects.


Meaning of Porosity

Pores are minute openings in solids that are accessible to vapours and gases. Porosity is a collective term for these pores and their distribution in the structure of a solid. Volume element is another term used to describe pores which arises from the theoretical model of carbon structure where pores are created by the irregular packing of the lattice. It is probable that pores are irregular in shape and they may also be interconnected, completely closed from the particle exterior and other porosity or terminated at some point within the particle depending on the arrangement of the surrounding lamellae.

There is a common misconception that porosity in carbon is a series of interconnecting tubes or slits [17] and the tree-branch model may also be misleading. A more realistic approach is the view that the carbon atoms of a porous carbon form a covalently bonded three-dimensional network within which imperfect lamellar type arrangements may be recognised, where interstices between carbon sheets give rise to the porous structure.[18-22] Interconnections do join parts of the porosity but in doing so must be classed as part of the porosity themselves forming part of both the adsorption system and the transportation system.

The internal surface area of a porous solid is composed of several classes of pores and may often be in excess of the external surface area. The volume is accessible to gases and vapours with which reactions may occur and it is not the porosity itself, which interacts directly with the adsorbate but the internal surface of the adsorbent.


Classification of pores

Porosity has a classification system as defined by IUPAC,[23] which gives a guideline of pore widths applicable to all forms of porosity. Distinctions in porosity class are not rigorous and they may often overlap in size and definition. The widely accepted I.U.P.A.C. classification is as follows:

Width less than 2 nm
Width between 2 and 50 nm
Width greater than 50 nm

Microporosity may then be subdivided into three subsequent categories:


Width less than 0.5 nm
Width between 0.5 - 1.4 nm
Width between 1.4 - 2.0 nm


Origin of Porosity

The total classification of porosity is tri-modal, as outlined previously, although not all adsorbents will contain all classes of pores, and the three types of pores are formed in different ways. The porosity in any one carbon is unique being created by the constituent basic structural units composed of carbon lamellae, which by definition are dependent on the precursor hence the porosity is parent material dependent. Poor alignment of the lamellae in three dimensions produces extensive internal porosity.[24] The pores are not uniformly shaped but are irregular shapes and sizes and these factors as well as the accessibility of the pores is dependent on the proximity of the constituent structural units.



The micropores are formed as the result of imperfect stacking of constituent molecules and packing arrangements of the bulk material, producing a lack of crystallite alignment, and small pseudo-graphitic crystallites.[25] The easiest way to imagine the microporosity, in keeping with the model outlined for general porosity, is that it exists as a series of interconnecting volume elements, rather like zeolites, but with each volume element of a different size and shape, connected in a completely random manner, with exception to precursor retention, three-dimensional structure. No two micropores will ever be identical and up to, although often much less than, 90% of adsorption in activated carbons takes place in the micropores. Their shape has been shown, by TEM studies [26] to be either slit-like or convoluted in shape.

The class of micropores may be subdivided into three separate groups:

i) ultramicroporosity (diameter < 0.5 nm)[27] - usually responsible for activated diffusion the pore diameter is comparable to that of the admolecule.

ii) microporosity (diameter ~ 0.5 –1.4 nm)[27] - fill quickly, within the first few minutes of adsorption. Overlap of the pore wall potentials results in a pore filling mechanism for adsorption in micropores.[6,28-30]

iii) supermicroporosity (size - 1.4 - 2.0 nm)[27] - co-operative pore filling is usually associated with a ratio of approximately 4:1 adsorbate molecule to pore diameter, as is typically found in the supermicropores. Monolayer formation occurs and the pore diameter is effectively reduced enhancing the adsorption potential of the pore hence increasing adsorption and completing the pore filling at low relative pressure. This class of porosity was proposed by Dubinin [30] who originally proposed the classification system.

The micropores provide sites of maximum adsorption potential for an admolecule/atom and within the pore an admolecule is usually influenced by approximately twelve adsorbent atoms. Due to the close proximity of the walls of micropores an interaction of the Polyani potentials, the result of overlapping of the dispersion fields, may occur resulting in a relatively deep potential energy well and enhanced adsorption at a given pressure. Hence, diffusion into the ultramicropores has a significant activation energy associated with it.

During the adsorption process an admolecule will wait until the vibration sequence of the pore wall opens up the porosity and then it will move ‘into’ the pore. Once inside the pore the admolecule has the ability to move ‘out’ but concentration gradients established within the porosity during the adsorption process should prevent such a situation.

The pore filling process may be divided into three steps:

i) monolayer formation

ii) pore filling by co-operative effects

iii) completion of the pore filling process.

Adsorption into micropores is completely reversible and no hysteresis loop is observed because of the inability for the adsorbate to condense in such narrow volume elements.



Mesopores are the result of major defects in the structure of a solid and serve as passages, providing a transport system, to the micropores. These are the pores which give rise to the phenomenon of capillary condensation with the adsorption/desorption isotherm, which is observed by the existence of an inherent hysteresis loop. The mesopores fill in the final stages of the isotherm by multilayer formation and a cylindrical meniscus. The pore diameters, greater than 2 nm but less than 50 nm according to IUPAC definition, are so large that at low relative pressures monolayer coverage occurs followed by further layers and the adsorbed film acts as a nucleus upon which capillary condensation may take place. This mechanism gives rise to hysteresis, as there are differences in the pore filling and emptying processes.

Determination of the volume and distribution of mesopores is via vapour adsorption, the subsequent evaluation performed is based on capillary condensation.



The micropores are considered important in the process of adsorption whereas the meso- and macropores primarily act as transport pores.[22,31,32] Major lattice structure defects, such as racks, fissures and etching channels, within a solid lead to the formation of macropores which may be treated as an open surface. It is possible to observe macroporosity by optical microscope and scanning electron microscopy as they are of the order 50 nm and greater. There is no actual upper limit to the diameter of the pores but it is usually 1 – 2 mm.

Adsorption is via the layer by layer mechanism typical of non-porous solids and they act as transport pores allowing access to the internal surfaces and microporosity. These pores do not contribute considerably to the surface area of the adsorbent, typically less than 2 m2g-1.


Pore Size Distribution

The distribution of pore sizes may be described in terms of micro-, meso- and macropores but are not necessarily tri-modal, exhibiting three peaks of varying ratios, this result formed the basis of Dubinin’s classification.[30]

A complete distribution is not always present in a carbon and some species may exhibit only some types of porosity. Distinct peaks are not always observed and some systems exhibit wide distributions over all pore diameters or a constant decrease/increase in pore frequency with variation in pore diameter.

Schematic of the tri-modal pore-size distribution found in many carbons


Classification of Adsorption Isotherms

Adsorption isotherms should conventionally be plotted on the basis of relative pressure, p/po (x-axis) versus amount adsorbed expressed as a molar quantity (y-axis) in mmolg-1, to allow comparisons to be made. The experimental procedure involves the use of partial pressure, where the actual pressure is expressed with respect to the saturation vapour pressure at a constant temperature of adsorption, hence the process is isothermal. Adsorption data may alternatively be expressed in terms of an isobar, the variation in uptake with temperature at constant pressure, or an isostere, the change in temperature with pressure at a constant surface coverage. Isotherms provide a significant amount of information about the adsorbent used and the interaction with the adsorbate in the system, including:

i) assessment of the surface chemistry and fundamentals involved in the adsorption process;

ii) estimates of the surface area, pore volume and pore size distribution;

iii) efficiency profiles for carbons used in industrial processes.

The interpretation of adsorption isotherms can yield a large amount of information about the processes involved, as outlined above, but this is only possible upon careful analysis of the data obtained and this can often lead to confusion in the interpretations made. An understanding of the adsorption mechanisms involved in the different classes of porosity is essential to explain the shapes observed.

The extent of adsorption on a surface, usually denoted n, is generally a function of the temperature, pressure and nature of the adsorbent and adsorbate:

n = f(P, T, adsorbate, adsorbent)

For isothermal adsorption of a particular system this simplifies to

n = f(P)T, adsorbent, adsorbate

By working within the limits of vacuum and the saturation vapour pressure, the pressure may be expressed in terms of relative vapour pressures, i.e. p/po:

n = f(p/po)T, adsorbent, adsorbate

This gives a universal basis upon which all isotherms may be displayed in order to make the comparison of results easier. The above relationship may be used to graphically represent uptake profiles in the form of an ‘adsorption isotherm’.

Diagrammatic Representation of Isotherm Classification

All adsorption isotherms should fit at least one, or at least a combination of two or more, of the six recognised types classified by Brunauer, Deming, Deming and Teller [33] (B.D.D.T. system). The figure above shows the possible shapes and information which may be drawn from them is outlined below:

Type I Isotherm - these are typical of adsorbents with a predominantly microporous structure, as the majority of micropore filling will occur at relative pressures below 0.1. The adsorption process is usually complete at a partial pressure of ~0.5. Examples include the adsorption of nitrogen on carbon at 77K and ammonia on charcoal at 273K.

Type II Isotherm - physical adsorption of gases by non-porous solids is typified by this class of isotherm. Monolayer coverage is followed by multilayering at high relative pressures. Carbons with mixed micro- and meso-porosity produce Type II isotherms.

Type III Isotherm - the plot obtained is convex to the relative pressure axis. This class of isotherm is characteristic of weak adsorbate-adsorbent interactions [34] and is most commonly associated with both non-porous and microporous adsorbents. The weak interactions between the adsorbate and the adsorbent lead to low uptakes at low relative pressures. However, once a molecule has become adsorbed at a primary adsorption site, the adsorbate-adsorbate interaction, which is much stronger, becomes the driving force of the adsorption process, resulting in accelerated uptakes at higher relative pressure. This co-operative type of adsorption at high partial pressures is known as cluster theory and examples include the adsorption of water molecules on carbon where the primary adsorption sites are oxygen based.

Type IV Isotherm - A hysteresis loop, which is commonly associated with the presence of mesoporosity, is a common feature of Type IV isotherms, the shape of which is unique to each adsorption system. Capillary condensation gives rise to a hysteresis loop[3] and these isotherms also exhibit a limited uptake at high relative pressures.

Type V Isotherm - these isotherms are convex to the relative pressure axis and are characteristic of weak adsorbate-adsorbent interactions.[34] These isotherms are indicative of microporous or mesoporous solids. The reasons behind the shape of this class of isotherm are the same as those for Type III and again water adsorption on carbon may exhibit a Type V isotherm.

Type VI Isotherm
- introduced primarily as a hypothetical isotherm, the shape is due to the complete formation of monomolecular layers before progression to a subsequent layer. It has been proposed, by Halsey,[35] that the isotherms arise from adsorption on extremely homogeneous, non-porous surfaces where the monolayer capacity corresponds to the step height. One example known to exist is the adsorption of krypton on carbon black (graphitised at 3000 K) at 90 K.[36]



1 Kayser, H. Wied. Ann. 1881, 14, 450.
2 McBain, J.W. Phil. Mag. 1909, 18, 916.
3 Sing, K.S.W.; Everett, D.H.; Haul, R.A.W.; Moscou, L.; Pierotti, R.A.; Rouquerol, J.; Siemieniewska, T. Pure Appl. Chem. 1985, 57, 603.
4 Gregg, S.J.; Sing, K.S.W. Adsorption, Surface Area and Porosity; 2nd ed.; Academic Press: London, 1982.
5 Dubinin, M.M. Zh. Fiz. Khimii 1965, 1305.
6 Dubinin, M.M. In Chemistry and Physics of Carbon; Walker, P.L. Jr., Ed.; Marcel Dekker: New York, 1966; Vol. 2.
7 Dubinin, M.M. In Progress in Surface and Membrane Science; Cadenhead, D.A., Ed.; Academic Press: New York, 1975; Vol. 9.
8 Lamond, T.G.; Marsh, H. Carbon 1964, 1, 281.
9 Cheremisinoff, N.P.; Cheremisinoff, P.N. Carbon Adsorption for Pollution Control; Prentice Hall: NJ, USA, 1993.
10 Atkins, P.W. 5th ed.; Oxford University Press: Oxford, 1994.
11 Bussy, A. Journal de Pharmacie et des Sciences Accessoires 1822, 8, 257.
12 Cheremisinoff, P.N.; Ellerbusch, F. Eds., Carbon Adsorption Handbook; Ann Arbor, Science Publishers Ins., 1978.
13 Gurvitsch, L. J. Phys. Chem. Soc. Russ. 1915, 47, 805.
14 Bradley, R.H.; Rand, B. Carbon 1991, 29, 1165.
15 O'koye, I.P.; Benham, M.; Thomas, K.M. Langmuir 1997, 13, 4054.
16 Rodriguez-Reinoso, F. Carbon 1998, 36, 159.
17 Byrne, J.F.; Marsh, H. In Porosity in Carbons: Characterisational Applications; Patrick, J. W., Ed.; Edward Arnold: London, 1995.
18 Bansal, R.C.; Donnet, J.; Stoeckli, H.F. In Active Carbon; Marcel Dekker: New York, 1988.
19 Doying, E.G. Patent 3 256 206 USA, 1965.
20 Peters, E.M. Patent 3 235 323 USA, 1966.
21 Bailey, A.; Maggs, F.A.P. 1 301 101 Britain, 1971.
22 McEnaney, B. Carbon 1988, 26, 267.
23 IUPAC J. Colloid Interface Chem.; Pure Appl. Chem. 1972, 31, 578.
24 Mahajan, O.P. Coal Structure Meyers, R.A., Ed.; Academic Press: New York, 1982.
25 Walker, P.L. Jr. Carbon 1986, 24, 379.
26 Thomas, J.M. and Thomas, W.J., Introduction to the Principles of Heterogeneous Catalysis, Academic Press: New York, 1967.
27 IUPAC, Carbon 1985, 23, 601.
28 Dubinin, M.M.; Eraehko, I.T.; Kadlec, O.; Ulin, V.I.; Voloshchuk, A.M.; Zolotarev, P.P. Carbon 1975, 13, 193.
29 Dubinin, M.M. Carbon 1981, 19, 321.
30 Characterisation of Porous Solids; Dubinin, M.M., Ed.; Society of Chemical Industries: London, 1979; Vol. 1.
31 Barton, S.S.; Evans, M.J.B.; Harrison, B.H. J. Colloid Interface Sci. 1974, 49, 462.
32 Stoeckli, H.F. Carbon 1990, 28, 1.
33 Brunauer, S.; Deming, L.S.; Deming, W.S.; Teller, E. JACS 1940, 62, 1723.
34 Kiselev, A.V. J. Colloid Interface Sci. 1968, 28, 430.
35 Halsey, G.D. J. Chem. Phys. 1948, 16, 931.
36 Kratschmer, W.; Rathousky, J.; Zukal, A. Carbon, 1999, 37, 301.



©2008 Ashleigh Fletcher