Desorption


Catalytic gas-phase reactions play an important role in many bulk chemical processes, such as in the production of methanol, ammonia, sulfuric acid, and nitric acid. In most processes, the effective area of the catalyst is critically important. Since these reactions take place at surfaces through processes of adsorption and desorption, any alteration of surface area naturally causes a change in the rate of reaction. Industrial catalysts are usually supported on porous materials, since this results in a much larger active area per unit of reactor volume.  [c.47]

The physical chemist is very interested in kinetics—in the mechanisms of chemical reactions, the rates of adsorption, dissolution or evaporation, and generally, in time as a variable. As may be imagined, there is a wide spectrum of rate phenomena and in the sophistication achieved in dealing wifli them. In some cases changes in area or in amounts of phases are involved, as in rates of evaporation, condensation, dissolution, precipitation, flocculation, and adsorption and desorption. In other cases surface composition is changing as with reaction in monolayers. The field of catalysis is focused largely on the study of surface reaction mechanisms. Thus, throughout this book, the kinetic aspects of interfacial phenomena are discussed in concert with the associated thermodynamic properties.  [c.2]

An interesting development is that of electron-stimulated desorption ion angular distribution (ESDIAD). The equipment is essentially that for LEED—a focused electron beam (20-1000 eV) impinging on a surface results in the desorption of atomic and molecular ions, which are accelerated linearly from the surface by means of suitably charged grids. The direction cones are largely determined by the orientation of the bonds that are ruptured, and the result is a somewhat diffuse but rather LEED-like pattern that may be seen on a fluorescent screen and photographed. If necessary, the ions may be mass selected, but often this is not necessary as some single kind of ion predominates. Figure Vin-13 shows a ESDIAD pattern of NH3 on an oxygen pretreated Ni(lll) surface [93]. The technique has been used to study hindered rotation of PF3 on the same surface [94].  [c.310]

ESD Electron-stimulated (impact) desorption [148, 149] An electron beam (100-200) eV) ejects ions from a surface Surface sites and adsorbed species  [c.315]

ESDIAD Electron-stimulated desorption ion angular distribution [150-152] A LEED-like pattern of ejected ions is observed Orientation of adsorbed species  [c.315]

PSD Photon-stimulated desorption [149, 162-165] Incident photons eject adsorbed molecules Desorption mechanisms and dynamics  [c.316]

TPD Temperature-programmed desorption [171, 172] The surface is heated and chemisorbed species desorb at characteristic temperatures Characterization of surface sites and desorption kinetics  [c.316]

TDS, FDS Thermal desorption spectroscopy. Flash desorption spectroscopy [173] Similar to TPD Similar to TPD  [c.316]

A somewhat subtle point of difficulty is the following. Adsorption isotherms are quite often entirely reversible in that adsorption and desorption curves are identical. On the other hand, the solid will not generally be an equilibrium crystal and, in fact, will often have quite a heterogeneous surface. The quantities ys and ysv are therefore not very well defined as separate quantities. It seems preferable to regard t, which is well defined in the case of reversible adsorption, as simply the change in interfacial free energy and to leave its further identification to treatments accepted as modelistic.  [c.352]

In the adsorption of long-chain surfactant molecules to charged surfaces, both chemical and electrical interactions may be important. These mechanisms are nicely reviewed by Berg [170], and we offer only a brief summary of important features here. The amphiphilic nature of surfactants causes them to follow Traube s rule (Section XI-1 A), and numerous studies have focused on the effect of chain length on adsorption [20-23,171]. Rusling and co-workers have used flow voltammetry to study the adsorption and desorption dynamics of a series of amphiphilic ferrocenes [172]. They find adsorption free energies for their surfactants to increase with length by about 750 J/mol CH2, in agreement with the findings for SAMs discussed in Section XI-IB. Ralston and co-workers  [c.414]

Derive Eq. XI-IS, assuming a Langmuir adsorption process described in Eq. XI-2, where ka and kd are the adsorption and desorption rate constants. Treat the solution  [c.420]

Irreversible adsorption discussed in Section XI-3 poses a paradox. Consider, for example, curve 1 of Fig. XI-8, and for a particular system let the equilibrium concentration be 0.025 g/lOO cm, corresponding to a coverage, 6 of about 0.5. If the adsorption is irreversible, no desorption would occur on a small dilution on the other hand, more adsorption would occur if the concentration were increased. If adsorption is possible but not desorption, why does the adsorption stop at 6 = 0.5 instead of continuing up to 0 = 1 Comment on this paradox and on possible explanations.  [c.421]

The energetics and kinetics of film formation appear to be especially important when two or more solutes are present, since now the matter of monolayer penetration or complex formation enters the picture (see Section IV-7). Schul-man and co-workers [77, 78], in particular, noted that especially stable emulsions result when the adsorbed film of surfactant material forms strong penetration complexes with a species present in the oil phase. The stabilizing effect of such mixed films may lie in their slow desorption or elevated viscosity. The dynamic effects of surfactant transport have been investigated by Shah and coworkers [22] who show the correlation between micellar lifetime and droplet size. More stable micelles are unable to rapidly transport surfactant from the bulk to the surface, and hence they support emulsions containing larger droplets.  [c.505]

There appear to be two stages in the collapse of emulsions flocculation, in which some clustering of emulsion droplets takes place, and coalescence, in which the number of distinct droplets decreases (see Refs. 31-33). Coalescence rates very likely depend primarily on the film-film surface chemical repulsion and on the degree of irreversibility of film desorption, as discussed. However, if emulsions are centrifuged, a compressed polyhedral structure similar to that of foams results [32-34]—see Section XIV-8—and coalescence may now take on mechanisms more related to those operative in the thinning of foams.  [c.506]

Davies [114] found that the rates of desorption of sodium laurate and of lauric acid films were in the ratio 6.70 1 at 21.5°C at molecular areas of 90 and 60 per molecule, respectively. Calculate o. the potential at the plane CD in Fig. XV-12.  [c.563]

Many solids have foreign atoms or molecular groupings on their surfaces that are so tightly held that they do not really enter into adsorption-desorption equilibrium and so can be regarded as part of the surface structure. The partial surface oxidation of carbon blacks has been mentioned as having an important influence on their adsorptive behavior (Section X-3A) depending on conditions, the oxidized surface may be acidic or basic (see Ref. 61), and the surface pattern of the carbon rings may be affected [62]. As one other example, the chemical nature of the acidic sites of silica-alumina catalysts has been a subject of much discussion. The main question has been whether the sites represented Brpnsted (proton donor) or Lewis (electron-acceptor) acids. Hall  [c.581]

Chemisorption may be rapid or slow and may occur above or below the critical temperature of the adsorbate. It is distinguishable, qualitatively, from physical adsorption in that chemical specihcity is higher and that the energy of adsorption is large enough to suggest that full chemical bonding has occurred. Gas that is chemisorbed may be difficult to remove, and desorption may be  [c.599]

As is made evident in the next section, there is no sharp dividing line between these two types of adsorption, although the extremes are easily distinguishable. It is true that most of the experimental work has tended to cluster at these extremes, but this is more a reflection of practical interests and of human nature than of anything else. At any rate, although this chapter is ostensibly devoted to physical adsorption, much of the material can be applied to chemisorption as well. For the moment, we do assume that the adsorption process is reversible in the sense that equilibrium is reached and that on desorption the adsorbate is recovered unchanged.  [c.601]

Fig. XVII-5. Schematic detector response in a determination of nitrogen adsorption and desorption. A flow of He and N2 is passed through the sample until the detector reading is constant the sample is then cooled in a liquid nitrogen bath. For desorption, the bath is removed. (From Ref. 28. Reprinted with permission from John Wiley Sons, copyright 1995.) Fig. XVII-5. Schematic detector response in a determination of nitrogen adsorption and desorption. A flow of He and N2 is passed through the sample until the detector reading is constant the sample is then cooled in a liquid nitrogen bath. For desorption, the bath is removed. (From Ref. 28. Reprinted with permission from John Wiley Sons, copyright 1995.)
Fig. XVll-19. Adsorption of CH4 on MgO(lOO) at 77.35 K. The vertical line locates each vertical step corresponds to the condensation of a monolayer. There was no hysteresis. Desorption points are shown as . (From Ref. 110.) Fig. XVll-19. Adsorption of CH4 on MgO(lOO) at 77.35 K. The vertical line locates each vertical step corresponds to the condensation of a monolayer. There was no hysteresis. Desorption points are shown as . (From Ref. 110.)
Turning to the multilayer region, the actual assumptions of the BET model are not realistic. As illustrated in Fig. XVII-21h, it does not give the correct variation of qd with 6 (but partly because of the surface heterogeneity) or the correct value of AI2. Basically, the available evidence suggests that the adsorbed film approaches bulk liquid in properties as P approaches and while the BET assumption as to the adsorption energy correctly reflects this behavior, the assumption of localized multilayers is not consistent with it and gives an erroneous configurational entropy. Related to this is a catastrophe that Cas-sel [134] has pointed out, namely that the integral I of Eq. X-40 is infinity in the BET model. Some further evidence of the liquidlike state of multilayer films was provided by Arnold [135], who found that his data on the desorption of oxygen-nitrogen mixtures on titanium dioxide could be accounted for by assuming the second and third layers to possess the molar entropy of normal liquid and that Raoult s law applied, whereas the BET treatment gave much poorer agreement.  [c.653]

The adsorption branch of isotherms for porous solids has been variously modeled. Again, the DR equation (Eq. XVII-75) and related forms have been used [186,194]. With respect to desorption, the variety of shapes of loops of the closed variety that may be observed in practice is illustrated in Fig. XVII-29 (see also Refs. 195 and 197).  [c.665]

At c all such capillaries would be filled and on desorption would empty by retreat of a meniscus of curvature 2/r, so that at each stage of the desorption branch dea the radius of the capillaries emptying would be  [c.665]

The effect of the bundle of capillaries picture is to stress the use of the desorption branch to obtain a pore-size distribution. The basic procedure stems from those of Barrett et al. [201] and Pierce [196] see also Ref. 58 wherein the effective meniscus curvature is regarded as given by the capillary radius minus the thickness of ordinary multilayer adsorption expected at that P/P. This last can be estimated from adsorption data on similar but nonporous material, and for this de Boer s t-curve (see Table XVII-4) is widely used. The general calculation is as follows. After each stepwise decrease in (pressure on desorption), an effective capillary radius is calculated from Eq. XVII-135, and the true radius is obtained by adding the estimated multilayer thickness. The exposed pore volume and pore area can then be calculated from the volume desorbed in that step. For all steps after the first, the desorbed volume is first corrected for that from multilayer thinning on the sum of the areas of previously exposed pores. In this way a tabulation of cumulative pore volume of pores of radius greater than a given r is obtained and, from the slopes of the corresponding plot, a pore-size distribution results. Such a distribution was compared in Fig. XVI-2  [c.666]

More detailed information about the pore system can be obtained from scanning curves, illustrated in Fig. XVII-28c. Thus if adsorption is carried only up to point a and then desorption is started, the lower curve ab will be traced if at absorption is resumed, the upper curve ab is followed, and so on. Any complete model should account in detail for such scanning curves and, conversely, through their complete mapping much more information can be obtained about the nature of the pores. Rao [214] and Emmett [215] have summarized a great deal of such behavior.  [c.668]

The projection of a domain plot onto its base makes a convenient two-dimensional graphical representation for describing adsorption-desorption operations. Here, the domain region that is filled can be indicated by shading the appropriate portion of the 45° base triangle. Indicate the appropriate shading for (a) adsorption up to Xa - 0.8 (b) such adsorption followed by desorption to Xd - 0.5 and (c) followed by readsorption from Xd = 0.5 to Xa = 0.7.  [c.675]

Different types of chemisorption sites may be observed, each with a characteristic A value. Several adsorbed states appear to exist for CO chemisorbed on tungsten, as noted. These states of chemisorption probably have to do with different types of chemisorption bonding, maybe involving different types of surface sites. Much of the evidence has come initially from desorption studies, discussed immediately following.  [c.694]

A powerful technique in studying both adsorption and desorption rates is that of programmed desorption. The general procedure (see Refs. 36, 84) is to expose a clean metal filament or a surface to a known, low pressure of gas that flows steadily over it. The pressure may be quite low, for example, 10 mm Hg or less, so that even nonactivated adsorption can take some minutes for  [c.694]

Temperature-programmed desorption (TPD) is amenable to simple kinetic analysis. The rate of desorption of a molecular species from a uniform surface is given by Eq. XVII-4, which may be put in the form  [c.696]

DIET Desorption induced by electronic transitions [147a] General class of desorption and reaction phenomena induced by electron or photon bombardment Same as ESD and PSD  [c.315]

LISC Laser-induced surface Similar to LIF chemical Pbotochemistry, desorption  [c.317]

Protein adsorption has been studied with a variety of techniques such as ellipsome-try [107,108], ESCA [109], surface forces measurements [102], total internal reflection fluorescence (TIRE) [103,110], electron microscopy [111], and electrokinetic measurement of latex particles [112,113] and capillaries [114], The TIRE technique has recently been adapted to observe surface diffusion [106] and orientation [IIS] in adsorbed layers. These experiments point toward the significant influence of the protein-surface interaction on the adsorption characteristics [105,108,110]. A very important interaction is due to the hydrophobic interaction between parts of the protein and polymeric surfaces [18], although often electrostatic interactions are also influential [ 116]. Protein desorption can be affected by altering the pH [117] or by the introduction of a complexing agent [118].  [c.404]

There are numerous references in the literature to irreversible adsorption from solution. Irreversible adsorption is defined as the lack of desotption from an adsoibed layer equilibrated with pure solvent. Often there is no evidence of strong surface-adsorbate bond formation, either in terms of the chemistry of the system or from direct calorimetric measurements of the heat of adsorption. It is also typical that if a better solvent is used, or a strongly competitive adsorbate, then desorption is rapid and complete. Adsorption irreversibility occurs quite frequently in polymers [4] and proteins [121-123] but has also been observed in small molecules and surfactants [124-128]. Each of these cases has a different explanation and discussion.  [c.404]

Fig. XI-8. Adsorption of BaDNNS on TiOi at 23°C from n-heptane solution. , x, A, D, O, adsorption points for indicated equilibration times. , desorption points following 12-hr and 20-min equilibrations, respectively. (From Ref. 124.) Fig. XI-8. Adsorption of BaDNNS on TiOi at 23°C from n-heptane solution. , x, A, D, O, adsorption points for indicated equilibration times. , desorption points following 12-hr and 20-min equilibrations, respectively. (From Ref. 124.)
Fig. XVI-7. Dielectric isotherms of water vapor at 15°C adsorbed on a-FeiOa (solid points indicate desorption). A complete monolayer was present at P/P = 0.1, and by P/P = 0.8 several layers of adsorbed water were present. (From Ref. 110.) Fig. XVI-7. Dielectric isotherms of water vapor at 15°C adsorbed on a-FeiOa (solid points indicate desorption). A complete monolayer was present at P/P = 0.1, and by P/P = 0.8 several layers of adsorbed water were present. (From Ref. 110.)
First, it is entirely possible that surface heterogeneities and imperfections undergo some reversible redistribution during adsorption. As noted in Section VII-4B, Dunning [116] has considered that since the presence of an adsorbed molecule should alter the energy of special sites (such as illustrated in Fig. VII-5), above some critical temperature for surface mobility, their distribution should depend on the extent of adsorption. Also, the first few layers of the crystalline surface of a solid are distorted (see Section VII-3B), and this distortion should certainly be altered if an adsorbed layer is present here no more than motions perpendicular to the surface may be involved. There is a scattering of early literature indication that changes in surface structure, specifically, surface reconstruction, can occur on adsorption. In the 1960 s Lander and Morrison [117] concluded from a LEED study that considerable surface rearrangement of germanium surfaces took place on adsorption of iodine. Similarly, field emission work with tungston led Ehrlich and Hudda [118] to conclude that reconstruction could accompany the adsorption or desorption of nitrogen. Hydrogen atom adsorption on metals can lead to complex surface reconstructions, as for H on Ni(llO) [56] and on W(IOO) [119]. Somoijai [120] gives a somewhat more detailed account of such effects.  [c.590]

The third region is one for which the Q values are of the order of chemical bond energies the r values become quite large, indicating that desorption may be slow, and F as computed by Eq. XVII-3 becomes preposterously large. Such values are evidently meaningless, and the difficulty lies in the assumption embodied in Eq. XVII-3 that the collision frequency gives the number of molecules hitting and sticking to the surface. As monolayer coverage is approached, it is to be expected that more and more impinging molecules will hit occupied areas and rebound without experiencing the full Q value. One way of correcting for this effect is taken up in the next section, which deals with the Langmuir adsorption equation.  [c.603]

Fig. XVII-27. Nitrogen adsorption at 77 K for a series of M41S materials. Average pore diameters squares, 25 A triangles, 40 A circles, 45 A. Adsorption solid symbols desorption open symbols. The isotherms are normalized to the volume adsorbed at Pj = 0.9. (From Ref. 187. Reprinted with kind permission from Elsevier Science-NL, Sara Burgerhartstraat 25, 1055 KV Amsterdam, The Netherlands.) Fig. XVII-27. Nitrogen adsorption at 77 K for a series of M41S materials. Average pore diameters squares, 25 A triangles, 40 A circles, 45 A. Adsorption solid symbols desorption open symbols. The isotherms are normalized to the volume adsorbed at Pj = 0.9. (From Ref. 187. Reprinted with kind permission from Elsevier Science-NL, Sara Burgerhartstraat 25, 1055 KV Amsterdam, The Netherlands.)
A potentially powerful approach is that of Everett [216], who treats the pore system as a set of domains, independently acting in a first approximation. Each domain consists of those elements of the adsorbent that fill at a particular Xa(j) and empty at a particular other relative pressure Xd(j), the associated volume being V). Each domain is thus characterized by these three variables, and a plot of the function V(Xa,xj) would produce a surface in three dimensions something like a relief map. On increasing the vapor pressure from Xa to Xa+dxa, all domains of filling pressure in this interval should fill, but these domains can and in general would have a range of emptying pressure Xd ranging from Xd = 0 to Xd = Xa- Since the Xd of any domain cannot exceed its Xa (it cannot empty at a higher pressure than it fills ), the base of the topological map must be a 45° triangle. The detailed map contains in principle full information about the adsorption and desorption branches, as well as about all possible scanning loops. The problem of deducing such a map from data is a massive one, however, and seems not yet to have been done. Other approaches to the problem treat a porous adsorbent as a network of various size capillaries [217, 218]. Finally, capillary condensation where the vapor consists of two or more components—a rather complex subject—is discussed in several papers in Ref. 219.  [c.668]

A concluding comment might be made on the temperature dependence of adsorption in such systems. One can show by setting up a piston and cylinder experiment that mechanical work must be lost (i.e., converted to heat) on carrying a hysteresis system through a cycle. An irreversible process is thus involved, and the entropy change in a small step will not in general be equal to q/T. As was pointed out by LaMer [220], this means that second-law equations such as Eq. XVII-107 no longer have a simple meaning. In hysteresis systems, of course, two sets of st values can be obtained, from the adsorption and from the desorption branches. These usually are not equal and neither  [c.668]

The nature of reaction products and also the orientation of adsorbed species can be studied by atomic beam methods such as electron-stimulated desorption (ESD) [49,30], photon-stimulated desoiption (PDS) [51], and ESD ion angular distribution ESDIAD [51-54]. (Note Fig. VIII-13). There are molecular beam scattering experiments such  [c.691]

If the heating is quickly to a high temperature, or a flashing, all adsorbed gas is removed indiscriminately. If, however, the heating is gradual, then separate, successive desorptions may be observed. Thus, as illustrated in Fig. XVIII-9, hydrogen leaves flat, stepped, and kinked Pt surfaces in stages, indicating the presence of different adsorption sites. The presences of successive desorption stages is fairly common. No less than four are found for H2 chemisorbed on Pd(llO), as shown in Fig. XVIII-10 notice how successive maxima appear with increasing exposure. Xu and Koel [87a] report three different states for NO desorption from Pt(III), possibly due to different bonding geometries (as in Fig. XVlII-5). Yates [88] has reviewed the subject.  [c.696]


See pages that mention the term Desorption : [c.324]    [c.395]    [c.602]    [c.638]    [c.665]    [c.668]    [c.694]    [c.695]    [c.696]   
Physical chemistry of surfaces (0) -- [ c.0 ]

Modeling of chemical kinetics and reactor design (2001) -- [ c.30 ]

Introduction to chemical engineering analysis using mathematica (2002) -- [ c.0 ]

Industrial ventilation design guidebook (2001) -- [ c.1429 ]

Computational methods in surface and colloid science (2000) -- [ c.389 , c.392 , c.404 , c.405 , c.417 , c.418 , c.421 , c.439 , c.477 , c.851 , c.869 , c.881 , c.908 ]