Adsorption difference spectra

Figure A3.10.24 UPS data for CO adsorption on Pd(l 10). (a) Clean surface, (b) CO-dosed surface, (c) Difference spectrum (b-a). This spectrum is representative of molecular CO adsorption on platinum metals [M].

A change in potential can cause any of several effects, including migration of ions into or out of the thin layer, adsorption, desorption, and faradaic reactions consuming or producing species adsorbed on the surface or in solution. For these reasons a difference spectrum (see Eq. (1.3) can exhibit both negative bands due to species formed and positive bands due to species consumed at the sample potential. 

A different spectrum having gzz = 2.002, gxx = 2.009, and gyv = 2.020 was observed following the adsorption of oxygen on Ti02. Mikheikin and co-workers (46) suggested that this spectrum was consistent with that of O- however, to achieve the small value for gxx it wTas necessary to assume a rather large value for AEi. As will be discussed in Section IY.A.3, the spectrum was later proven to be that of 02 . 

The first electron spectroscopic study of adsorbed hydrocarbons was that reported by Eastman and Demuth (78) who used He radiation to probe the valence electrons of benzene, acetylene, and ethylene. Figure 17 shows the difference spectrum of C2H4 adsorbed on Ni(lll) at 100 and 230 K compared with the results of Clarke et al. (79) for ethylene adsorption on Pt(lOO) at 290 K, propylene adsorption on Pt(lOO), and ethylene adsorption on Pt(lll). 

The adsorptions of H, O, and S04 on Pt/C electrocatalyst electrodes have been further investigated by O Grady and Ramaker by comparing the XANES data at the Pt L2 and L3 absorption edges. In their analysis, the difference spectrum, which they term AS for antibonding state, is obtained as follows ... 

Fig. 11. UP-spectra from the CO/W(110) adsorption system at hv = 40.8 eV (a) Clean surface (b) difference spectrum after exposure of 1L CO (c) difference spectrum after 10L CO (d) difference spectrum with surface in equilibrium with 10-6 Torr CO (e) difference spectrum after pumping CO out and heating to approximately 600 K.

Fig. 33. Changes in the carboxylate and amide IR bands with hydration. Shown are the IR absorption of the carboxylate band at 1580 cm" in the difference spectrum (O) and the sum of the absolute value of the absorption of the amide I difference band at 1645 and 1690 cm" (X). Data are shown for 38°C (top) and 27°C (bottom). Curves A, D O sorption onto strong sorption sites curves B, D2O sorption onto weak sorption sites curve C, DjO multimolecular adsorption. Curves A-C were derived from the sorption isotherm by htting data to a three-site model. Ordinate units represent percentages of the values at 0.33 h. From Careri etal. (1979b).

EMIRS studies of ethanol on platinum electrodes have demonstrated the presence of linearly bonded carbon monoxide on the surface [106]. An important problem in the use of EMIRS to study alcohol adsorption is the choice of a potential window where the modulation is appropriate without producing faradaic reactions involving soluble products. Ethanol is reduced to ethane and methane at potentials below 0.2 V [98, 107] and it is oxidized to acetaldehyde at c 0.35 V. Accordingly, a potential modulation would be possible only within these two limits. Outside these potential region, soluble products and their own adsorbed species complicate the interpretation of the spectra. The problem is more serious when the adsorbate band frequencies are almost independent of potential. In this case, the potential window (0.2-0.35 V) is too narrow to obtain an appropriate band shift and spectral features can be lost in the difference spectrum. 

Figure 2. Hydroxyl stretching vibration spectra of sample 1 (a), 2 (b), 4 (c) and 6 (d) before adsorption of pyridine (1) and after evacuation of pyridine at 4 23K (2) (3) is the difference spectrum between (1)...

At pH 9 the DRIFT difference spectrum for As(V) on am-Fe(OH)3 appears to have two peaks, at 872 and 820 cm", as shown in Figure 11b. Similar results were obtained with the FTS-7 and the Csl beamsplitter in a completely replicated experiment, except that additional peaks were identified around 700, 605, and 480 cm. The two peaks at 605 and 480 cm are tentatively identified as the and V4 modes, which are below the detection of our FTS-175 system. Interpretation of the spectra obtained at pH 9 is not clear as the two peaks at 872 and 820 cm appear split much more than expected for the V3 split assigned to H2ASO4". It also does not appear reasonable to consider HAs04 " adsorption at pH 5 and H2ASO4 adsorption at pH 9. The following reaction... 

Figure 3.17 contains the difference spectrum between ATH and ATH which had been coated, from solution, with the same grade of MPBD. In this spectrum relatively weak adsorptions at 1860 cm and 1781 cm are present indicating some unreacted MPBD, but in addition a strong peak at 1700 cm exists which corresponds to the absorption frequency of the carbonyl stretch in the carboxylic acid. Clear evidence that the anhydride has hydrolysed, but the lack of a strong peak at 1580 cm, the carbonyl stretching frequency of the salt, shows that the acid had not reacted with the surface of the ATH to form salt bridging between the polybutadiene backbone and the filler surface. 

During adsorption at low temperature, benzene molecules interact only weakly (van der Waals adsorption) with Ni(lll) and as a result the UPS difference spectrum (c) at 150 K looks very similar to the gas-phase spectrum. The species on the surface is basically unaltered condensed benzene. Comparison of difference spectrum (c) with that in (b), for room-temperature adsorption, reveals both that the benzene n orbitals have shifted with respect to the other orbitals, and that most of the nickel rf-electron density has been lost. Clearly there has been a dissociative and strong interaction of the benzene with the Ni(ll 1) surface since the electronic structure of the benzene molecule has been disrupted and the Ni cf-electrons have been involved in the bonding. This type of change in electronic structure is indicative of chemisorption. 

Loader 38) studied the Raman spectra of styrene adsorbed on different silicas—chromatographic grade silica gel, Cab-O-Sil, and Aerosil 380. The author utilized the fact that chemisorption will bring about marked changes in the spectrum whereas physical adsorption will cause only a broadening of the Raman lines accompanied in some cases by a frequency... 

Angell (1) has investigated the Raman spectra of acetonitrile, propylene, and acrolein on a number of zeolites and found that physical adsorption occurred. There are sufficient differences between the spectrum of the liquid and of the adsorbed species (e.g. the carbon-carbon double bond stretching in the case of propylene and the carbon-nitrogen triple bond stretching in the case of acetonitrile) to make it quite clear that it was not merely a case of condensation in the pores of the solid adsorbent. 

---