Bulk Formates

Although less known than diatomite, these products have been in wide use for many years so that there exists a sound body of applications knowledge upon which to base grade selection, dosage, and procedures. Perlite and Solka-floc have the same availability in bagged, semi-bulk, or bulk formats as diatomite. 

It is interesting to compare these thermodynamic magnitudes with the bulk formation of the hydroxide, given by ... 

In this case, the values of the thermodynamic magnitudes can be also determined from (7.15)-(7.17), but these results should be compared with the bulk formation of the oxide ... 

Unfortunately, the above thermodynamic approach has only been followed for two systems, namely As-Pt(lll) and Bi-Pt(lll) [Blais et al., 2001, 2002]. Table 7.3 summarizes the main results. Thermodynamic data about the bulk formation of As(OH)s and Bi(OH)2 are not available for comparison. The only data available is the standard enthalpy for the bulk formation of Bi(OH)3 (A/7 = 711.3kJ/mol). 

With the aim of overcoming the major issues of the bulk format, new polymeric materials, such as beads, particles and membranes, were developed, prompting the application of these new formats to well known imprinting systems. 

Wulff and collaborators, for instance, reported the preparation of TSA imprinted beads for the hydrolysis of carbonate and carbamate [61, 62], exploiting the amidine (33) functional monomer previously developed by the same group and successfully applied to the bulk format [63]. The polymers were prepared using a suspension polymerisation that produced beads with sizes in the range 8-375 pm, depending on the polymerisation conditions. The pseudo-first order reaction rate of the imprinted beads (Tyrrp/ soin) was enhanced by a factor of 293 for the carbonate hydrolysis and 160 for the carbamate, when compared with the background. 

However, about the exact structure of the intermediate only guesses could be made Schwab, from his work with alloys (see Section III, B), and also Dowden and Reynolds (51), concluded that it was a positively loaded particle, whereas Rienacker (3) came to suppose that it was a (negative) formate ion, on account of his film resistance work published earlier [Rienacker and Hansen (3a)]. In this respect it is interesting that from their study of the decomposition of bulk formates Hofmann and Schibsted (52, 53) already in 1918 arrived at the view that also in the catalytic decomposition of HCOOH, the formation of formates may play a role. 

This result points to a decomposition of bulk formate with both CO and C02 as primary products, and to a further equilibration of these products via the water gas reaction, if the contact time is long enough. 

It follows from the first four results tabulated that, unlike in the experiment with bulk formate, the product gases did not react to establish the water gas equilibrium in a measurably short time. This may have been due to the fact that the decomposition temperatures were lower than in the bulk experiment. In two respects the results differ from those obtained by Walton and Verhoek with Ni-films ... 

We think that in the experiments with the Ni-on-silica catalyst (and perhaps also with the bulk-formate) the situation is somewhat obscured by the fact that the CO C02 ratio may be influenced by adsorption of H20 on the silica carrier, and of H2 and CO on the nickel. A reinvestigation with a pure nickel surface is desirable. The expression [CO] [H20]/[C02] [H2] will be a better guide in this last case, than the CO C02 ratio as the former is insensitive to adsorption of the reaction products. 

In both dehydrogenation and dehydration formate ions on the surface thus seem to play a central role. It is therefore of interest to study the decomposition of bulk formates, although it is obviously to be expected that there will always be differences between two-dimensional surface compounds and bulk compounds. The decomposition reactions of various formates will be discussed in the following section. Knowledge of the properties of these compounds will enable us to deal more extensively with the reaction mechanisms of the dehydration and the dehydrogenation of formic acid. 

The decomposition of bulk formates has drawn special attention, in connection with the function of promoters in iron and cobalt catalysts for the Fischer-Tropsch synthesis. In the investigations made on this point the interest was, naturally, focused on the organic products formed during the decomposition of the formate [Hofmann and Schibsted (53), Marec and Hahn (126)]. 

The affinity for oxygen of the metal involved determines whether reactions of type I or type II occur. Nickel formate produces nickel, magnesium formate produces magnesium oxide. Of special interest to us, however, is the extent to which reactions of type a or b occur, and whether the Ia/Ib or the Ila/IIb ratio is in any way related to the selectivity of the catalytic decomposition of formic acid on the metals and oxides in question. Furthermore it is worth while to investigate whether the stability of the bulk formates (e.g., the decomposition temperature) bears a relation to the catalytic activity of the corresponding metals or oxides. 

In Table XIII a survey is given of data on the decomposition of bulk formates and of the direction of the formic acid decomposition on the corresponding metals or oxides. 

As regards selectivity it is clear that there exists a general correlation, as was already concluded by Komarov et al. (92). Exceptions were observed in the cases of the metals nickel and cobalt. For the catalytic reaction on both metals practically complete dehydrogenation is found, whereas in the decomposition of the bulk formates CO is also produced in appreciable amounts. In interpreting this result caution should be exercised, however. Dependent on the experimental conditions, the water-gas shift reaction can occur on the metal phase already formed. Careful experiments by Fahrenfort et al. (12) have shown that under conditions which exclude a secondary reaction the product distribution of the decomposition of Ni-formate and of the catalytic reaction on Ni are equal within the experimental error (see also Section IV, B). 

Fig. 28. Relative activity versus stability of the bulk formates. (For the definition of parameters see text).

As to the second point introduced above, that of a relation between the stability of the bulk formates (expressed in the value of the decomposition temperature) and the catalytic activity of the relevant metal or metal oxide, it is clear that no more than a qualitative agreement may be expected. In the decomposition of formates sometimes appreciable amounts of oxalates and carbonates are formed, which, at least on the dehydrogenating (basic) oxides, are more stable than their formates. The presence of these anions during the catalytic reaction has, however, never been clearly demonstrated. Furthermore, during the decomposition of formic acid the surface intermediates may... 

Nevertheless it is encouraging that in an appreciable number of cases a fair correlation can be demonstrated, as is shown in Fig. 28. Here the temperature at which the total reaction rate amounts to 1.6 x 1014 molecules/cm2 sec, (Tr), is plotted versus the temperature of decomposition of the bulk formates (Td). It is clear that the equal-rate temperatures of the various catalysts are related to the decomposition temperatures of the corresponding bulk formates. The decomposition temperatures of Au, Pt, and Pd formates are not known, these compounds being unstable at room temperature. In accordance with the general picture, the Tr values on these metals are far below those observed on Ag, Ni, and Cu. 

Summarizing we may say that in a great number of cases a strong analogy can be demonstrated between the chemical properties of bulk formates and the catalytic properties of the corresponding metals or oxides in the decomposition of formic acid. 

It is certain, however, that both for metals and metal oxides in the zero-order region there is a close correlation between the catalytic activity and the stability of the bulk formates. All substances investigated show a reaction rate of the order of 1014 molecules/cm2 sec, at a temperature just above the decomposition temperature of the corresponding bulk formate. 

It should not be inferred that the per cent mullite or the per cent gamma-alumina is a direct measure of the amount of active catalytic ingredient destroyed by sintering, since it is well known that only a small fraction (less than 1 %) of a cracking catalyst is responsible for its cataljrtic activity at any one time. Nevertheless, the bulk formation of mullite and gamma-alumina approximately parallel the destruction of the active portions of the catalyst when it is sintered in the 800-1000° temperature range. 

Another type of preform involves the bulk formation of chopped glass fiber against a screen that is shaped to the contour of the mold. Pressure against the screen can be applied by air or by water. Continuous glass fibers are also preshaped for mat molding. A preform binder is a resin apphed to the chopped strands of a preform, usually during its formation, and cured so that the preform will retain its shape and can be handled. 

While the bulk formation energy of metallic Rh is -555 kJ/at, the most stable Rhis cluster results in a corresponding value of only -299 kJ/at. The differences may reflect the much lower average coordination number of the cluster atoms compared with the bulk. One also notes that a few selected clusters such as Rha, D h] RJ14, Dha] Rh4, D h, Rhg, Oh] Rhi3, Ih have similar formation energies, whereas the atoms in these clusters have very different coordination numbers. The planar configurations appear to be the preferred clusters at least for the smaller sized clusters. 

Table 15.1 Vibration frequencies of isolated dimers of various nanosolids and their redshift upon bulk formation derived from simulating the size-dependent redshift of Raman optical modes, as shown in Fig. 15.1...

Fig. 16.4 a Evolution of the vth atomic energy level from E iO) to the vth band EJxt =12) with a shift of AEy(zb) = (Zv + ZbPv and a width of E,w = 2zbPv y k,R) upon bulk formation. The amounts of energy shift and band expansion depend on the cohesive energy per bond at equilibrium and the quantum number v of the band, b A typical XPS spectrum of the core band with addition of entrapment (7) and polarization (P) components to the bulk (B) shift. (Reprinted with permission from [3])... 

---