The Rate of Chemisorption

The term activated adsorption is, as originally designated by H. S. Taylor, a type of adsorption that takes place at a measurably slow rate, associated with a certain temperature coefficient. For the same type of adsorption, the term chemisorption has been used more frequently, in later years. We shall use here the term chemisorption when the heat of adsorption is comparable with the heat evolved in ordinary chemical reactions. 

In studying the chemisorption of hydrogen on carefully reduced nickel the author has actually observed that a minute quantity of the vapor of stop-cock grease or of mercury vapor from a pressure gage appreciably affect the rate of chemisorption in so far as these contaminants reduce considerably the rate of adsorption and produce the effects typical for the so-called activated adsorption. Incomplete reduction of nickel oxide to the metal leads to a similar result. This can be avoided by repeated reduction and subsequent evacuations of the metal sample at 400°C. for a week. A typical result obtained with an exhaustively reduced nickel specimen is shown in Fig. 1. In view of these findings, the activated adsorption of hydrogen on other reduced metal catalysts frequently reported in the earlier literature might have been caused by contamination effects. 

Results pointing in the same direction were obtained recently by Schuit and De Boer (17), who found that activated adsorption of hydrogen occurs only on a partially oxidized surface of nickel supported on silica (3 1) but not on a thoroughly reduced surface. According to Schuit and De Boer, very prolonged evacuation or heating of a reduced nickel catalyst in an inert atmosphere leads to a slow activated hydrogen adsorption. This effect, however, disappears on renewed careful reduction 

We now turn to the numerous observations made on the activated adsorption of hydrogen on reduced copper. In this case a definite, large temperature coefficient or activation energy Ae has been found. It was calculated by the conventional formula 

The possibility of a contamination of the copper specimens used in these investigations is also indicated by the experimental fact that the adsorbed quantity of hydrogen per unit weight of copper was extremely small compared with that for other metals, in spite of the large heat of adsorption whose value exceeded 30 kcal./mole according to Ward (21). These findings have to be correlated to those of Beeck et al. (14), who observed that no perceptible amount of hydrogen is chemisorbed over an evaporated copper film which consisted probably of the pure metal. 

---

The presence of alkali promoters on the substrate surface can affect both the rate of chemisorption, (e.g. on K/Rh(100))55 and the adsorptive capacity... 

C/min to 140°C, (3) hold for 2 hours, and (4) heat to 500°C at 3°/min. Oxygen was introduced at the time the temperature reached 140°C. The increase in temperature after the isothermal (140°C) region led to an increase in the rate of chemisorption, up to the temperature at which combustion (burn-off) becomes the dominant process resulting in rapid weight loss (ca. 270°C). 

Here, again, the exponential term is of primary importance in determining the rate of chemisorption and the amount of the surface concentration of the chemisorbed atoms. Considering the other terms to be constant, therefore, integration gives... 

According to the experimental result that the rate of reaction is proportional to Pn,o, the electron consuming step (37a) is the slowest, and is therefore rate determining for the chemisorption. From (37a) we obtain for the rate of chemisorption, in the case of a p-type catalyst... 

Weisz (22) and Morrison (31) have pointed out that if there is a barrier at the surface of a semiconductor, and if electrons from the bulk of the material must cross this barrier for chemisorption to occur, then the rate of chemisorption may be limited by the rate at which electrons can cross the barrier. The rate at which electrons can cross a barrier is proportional to the number of electrons with energy greater than the barrier height, or... 

If the activation energy for chemisorption is appreciable, the rate of chemisorption at low temperature may be so slow that, in practice, only physical adsorption is observed. 

Figure 5.3 shows how the extent of gas adsorption on to a solid surface might vary with temperature at a given pressure. Curve (a) represents physical adsorption equilibrium and curve (b) represents chemisorption equilibrium. The extent of adsorption at temperatures at which the rate of chemisorption is slow, but not negligible, is represented by a non-equilibrium curve, such as (c), the location of which depends on the time allowed for equilibrium. 

Molecular hydrogen does not dissolve easily into iron with a smooth surface at temperatures below 200°C. Atomic hydrogen, however, enters the iron easily even at room temperature (218). In the case of molecular hydrogen it is the activation energy at the surface which governs the process. On a smooth or contaminated iron surface, it is the rate of chemisorption which governs the total rate. We shall return to this special case in Sec. X,4. 

The rate of chemisorption at a pressure of 20 cm. mercury could be represented by... 

The rate of chemisorption may be estimated from the kinetic theory formula for wall collisions [Eq. (VII.6.6)] ... 

The technically desirable conditions of anode potentials smaller than 1(X) mV vs. RHE, imply very small rates of process (5d) at either platinum or platinum-alloy PEFC anode catalysts, as can be seen, for example, from the RDE results reported in [18d,e]. The PEFC anode catalyst is thus required to electro-oxidize hydrogen in the presence of significant coverage by CO. The rate of sequence (5b) -I- (5c) can be enhanced by anodic overpotential as long as process (5c) significantly limits the rate of this sequence. Since reaction (5c) is a fast and potential-driven process, at relatively low anodic overpotentials the rate of sequence (5b) -I- (5c) could become fully controlled by the rate of chemisorption of H atoms (Eq. (5b)) on a catalyst surface with few CO-free sites. 

The chemisorption data determined gravimetrically were analyzed according to Elovich kinetics (J5). These kinetics are based on the assumption that the rate of chemisorption (dg/dt) declines as the more reactive sites are quenched according to the equation ... 

With a high number of different active sites the rate of chemisorption on heterogeneous surface can be described using a simpler Crickmore-Wojciechowski equation... 

Show that the Elovich equation for the rate of chemisorption can be integrated with respect to surface coverage (at constant pressure of the adsorbing gas) to give... 

The rate of chemisorption is governed by the frequency of collisions with the surface and the probability of sticking with chemical bond formation. The former is a physical phenomenon, dependent on temperature and pressure. For example, at one atmosphere pressure and 25 C, 3 x 10 molecules strike each square centimeter of surface each second. I fall stick, the surface is covered in 3 x 10" sec. The probability of chemical bonding is exponentially proportional to the enthalpy change, and activation... 

In this complex sequence of events the most important rate is the intrinsic rate of reaction which is composed of three rates the rate of surface reaction, the rate of chemisorption and the rate of desorption. The three processes are called the intrinsic kinetics i.e. the rate of reaction in the absence of any mass or heat transfer resistances. Intrinsic kinetics is almost always predicted experimentally in differential or integral laboratory reactors using fine powders to avoid diffusional resistances. This is not an easy task especially with regard to temperature gradients for highly exothermic reactions. Experimental difficulties associated with the development of models for intrinsic kinetics are not discussed in this book. However, the kinetic modelling of the intrinsic rates is described in chapter 3. 

In principle, therefore, the study of catalyzed reactions is intimately bound up with studies of chemisorption. The rate of the reaction may be controlled by the rate of chemisorption of the reactants, or the rate of reaction between chemisorbed molecules or by the rate of desorption of the product. 

It is clear that the material given in this chapter is quite classical and has been known in the literature since the 1930s and 1940s in the field of surface chemistry and catalysis. In fact this is the extent of knowledge used to date in the derivation of rate equations for gas solid catalytic reactions. To be more specific most of the studies on the development of gas-solid catalytic reactions do not even use the information and knowledge related to the rates of chemisorption (activated or non-activated) and desorption. Even the most detailed kinetic studies, usually rely on the assumption of equilibrium adsorption-desorption and use one of the well known equilibrium isotherms (usually the Langmuir isotherm) in order to relate the surface concentration to the concentration of the gas just above the surface of the catalyst. 

Compare the rate computed in problem 1 with the rate of chemisorption of nitrogen molecules, at the same conditions of temperature and pressure, on an hypothetical surface with site density, n of lO cm at halfmonolayer surface coverage. The activation energy for chemisorption is 5kcal/mol, and cr = 0.1. 

The principle of microscopic reversibility applies at equilibrium, which states that the rate of chemisorption equals the rate of desorption. It is important to emphasize that individual rates of the forward and backward steps are not zero at equilibrium but that these rates are nonzero and equal to each other. Adsorption isotherms predict surface coverage at equilibrium where iRadsoipaon = 0. Hence,... 

It might have been logical to start a disquisition on the interaction of hydrogen with supported metals with a paragraph or two on the rates of chemisorption, as was the case with unsupported metals (Section 3.22), but the quantitative measurement of rates of adsorption (and of desorption) on highly porous materials presents formidable difficulties, and the general opinion seems to be that the reward does not justify the effort needed. It is true that some years ago it was discovered that the... 

Similar calculations have been made for hydrogen chemisorption by nickel surfaces (49). In both cases the theoretical value of the activation energy is far higher than the experimental, and the method is of little quantitative use in predicting the rate of chemisorption processes. It does, however, show that rates of reactions in monolayers, and of... 

Preliminary to oxidation, we can also consider the situation of oxygen adsorbing rapidly as physically adsorbed gas, followed by conversion at a slower rate to chemisorbed oxygen atoms. The chemisorbed oxygen, in turn, reacts with underlying metal to form metal oxide, a reaction that is activated mechanically by asperities moving over the metal surface. Chemisorption limits the amount of oxide that is formed in such a process, the rate of chemisorption following an equation identical in form to that of (29.57) [6]. Hence, whichever process apphes, the form of the final equation is essentially the same. ... 

---