Sulphur coverage

Figure 6.4. Examples for the four types of global classical promotion behaviour. Work function increases with the x-axis. (a) Steady-state (low conversion) rates of ethylene oxide (EtO) and C02 production from a mixture of 20 torr of ethylene and 150 torr of 02 for various Cs predosed coverages on Ag(lll) at 563 K19 (b) Rate of water-gas shift reaction over Cu(l 11) as a function of sulphur coverage at 612 K, 26 Torr CO and 10 Torr H202° (c) Effect of sodium loading on NO reduction to N2 by C3H6 on Pd supported on YSZ21 at T=380°C (d) Effect of sodium loading on the rate of NO reduction by CO on Na-promoted 0.5 wt% Rh supported on Ti02(4% W03).22...

At equilibrium, the sulphur coverage of the nickel surface, 0, can be calculated by a Temkin-like adsorption isotherm dependent on the temperature and the H2S/H2-ratio 1,9) with the parameters found by Alstrup et al (P) equation 5. 

Rostrup-Nielsen found that the intrinsic reaction rate, rj, for methane steam reforming is correlated with the sulphur coverage by equation 6 (2). In the adiabatic prereformer, the sulphur acts as a pore mouth poison and as the reactions are restricted by pore diffusion 2,8), the effective activity of the sulphur poisoned catalyst pellet can be described by an empirical relation, equation 7, between the effective pellet reaction rate, rp, and the average sulphur coverage, 0av (7/... 

In summary, the results of TDS [13], photoemission [13,45] and scanning tunnelling microscopy [24,45] indicate that at low sulphur coverages the interactions between S and Ag on Ru(OOOl) can be classified as repulsive, in the sense that there is weakening of the Ru-Ag bond and no mixing of the adsorbates. Once the ruthenium substrate becomes saturated with sulphur, then attractive interactions between silver and sulphur are possible and AgS is formed [13,45]. Very similar trends are observed for the coadsorption of sulphur and copper on Ru(OOOl) [13,23]. 

If it is assumed that the adsorption heat is independent of the sulphur coverage 0s, a Temkin isotherm results ... 

This result may appear surprising as the heat of ehemisorption of oxygen is high (approximately 430 kJ/mol O2). However, the value is less than the heat of ehemisorption for sulphur, whieh is about 480 kJ/mol S2. This means that Equation (5.15) can be applied to estimate sulphur coverages at equilibrium in the presence of steam as illustrated in Figure 5. 42 for at3q)ical ammonia plant reformer [387] [389]. 

Figure 5.42 Sulphur coverage of nickel surface of catalyst in an anunonia plant tubular reformer, H20/CH4=3.3, Pexit=34 bar abs, Tii/Texit=505°C/800 C, no prereformer [387]. Reproduced with the permission of Brill.

It was shown [390] that the Equation (5.15) for sulphur coverage in the range of interest can be simplified to ... 

Therefore the retarding effect of sulphur is a d5mamic phenomenon. This means that carbon may be formed under certain conditions in spite of sulphur passivation — although at markedly reduced rates. However, the thermodynamic potential (approximately 30 kJ/mol) required to initiate carbon [389] formation was foimd to depend on the sulphur coverage as illustrated in Figure 5.47. This energy is much higher than that (approximately 2 kJ/mol) (refer to Table 5.4) required to initiate carbon formation on sulphur-free catalysts. 

Figure 5.47 Sulphur coverage and supersaturation (-AGc) for start of carbon formation [389]. TGA studies. -AG calculated from equihbrated gas at (H20/CH4)exit [389]. Reproduced with the permission of Springer.

The feed gas in the heating up zone is far from equilibrium because of nearly complete sulphur coverage. When the reaction starts, the interior of the pellets is filled with equilibrated gas, but the gas has no longer potential for carbon. 

Sulphur also inhibits the dissociation of carbon monoxide (the Boudouard reaction). TGA studies [385] on nickel catalysts showed that methanation as well as carbon from the Boudouard reaction were strongly inhibited with increasing sulphur coverage. Other studies [186] found that sulphur eliminates the dissociation of carbon monoxide at a H2S/Ni coverage above 0.33. 

This is reflected by the supersaturation required for on-set of the whisker growth depending on sulphur coverage [390] as illustrated in Figure 5.47. 

The step sites are the most active for the reforming reaction and the carbon formation as well. However, the nucleation of carbon requires a critically large group of surface carbon atoms on the step to form a stable carbon island above approximately 2.5 nm as outlined above. This may explain the different dependency of the sulphur coverage of the two reactions shown in Figure 5.46. 

Figure 9. Sulphur Coverage of the Nickel Surface of a Reforming Catalyst at various Sulphur Levels. Ammonia Plant Reformer (Rostrup-Nielsen, 1984a).

The progress in surface science has provided a basis for a detailed description of the phenomena (Rostrup-Nielsen, 1991a). The adsorbed sulphur atom results in a deactivation of the neighboring nickel atoms. The rate of carbon formation decreases more with sulphur coverage than does the reforming rate reflecting that the ensemble for the reforming reaction is smaller than that required for the nucleation of the carbon whisker. 

---