Adsorbed 2-methylpentane

Around a value of the gas-phase fraction of 2-methylpentane of about 0.83, the influence of the acid sites on the n-hexane diffusivity is not dominant anymore in comparison to the pore occupation of slow-diffusing 2-methyl-pentane. Figure 14 shows the dependence of the diffusivities of both components versus the concentration of adsorbed 2-methylpentane in terms of molecules per unit cell. The diffusivities of n-hexane in silicalite-1 and H-ZSM-5 become nearly equal when the concentration of 2-methylpentane reaches approximately 2.75 molecules per unit cell. For 2-methylpentane we And that the self-diffusivity in silicalite-1 becomes very close to the value in H-ZSM-5 at the same loading. 

When the reactions of alkane molecules larger than the butanes or neopentane are studied, and in particular when the molecule is large enough to form a Cs or a Ce ring, the complexity of the reaction pathway is considerably increased and an important feature is the occurrence, in addition to isomerization product, of important amounts of cyclic reaction products, particularly methylcyclopentane, formed by dehydrocycliza-tion this suggests the existence of adsorbed cyclic species. The question is whether the reaction paths for dehydrocyclization and isomerization are related. There is convincing evidence that they are. Skeletal interconversions involving n-hexane, 2- and 3-methylpentane may be represented. 

Referring first of all to the reactions over 0.2% platinum/alumina (Table V) the major features of the product distributions may be explained by a simple reaction via an adsorbed C5 cyclic intermediate. For instance, if reaction had proceeded entirely by this path, 2-methylpentane-2-13C would have yielded 3-methylpentane labeled 100% in the 3-position (instead of 73.4%) and would have yielded n-hexane labeled 100% in the 2-position (instead of 90.2%). Similarly, 3-methylpentane-2-I3C would have yielded a 2-methylpentane labeled 50% in the methyl substituent (instead of 42.6%), and would have yielded n-hexane labeled 50% in the 1- and 3-positions (instead of 43.8 and 49% respectively). The other expectations are very easily assessed in a similar manner. On the whole, the data of Table V lead to the conclusion that some 80% or so of the reacting hydrocarbon reacts via a simple one step process via an adsorbed C5 cyclic intermediate. The departures from the distribution expected for this simple process are accounted for by the occurrence of bond shift processes. It is necessary to propose that more than one process (adsorbed C6 cyclic intermediate or bond shift) may occur within a single overall residence period on the catalyst Gault s analysis leads to the need for a maximum of three. The number of possible combinations is large, but limitations are imposed by the nature of the observed product distributions. If we designate a bond shift process by B, and passage via an adsorbed Cs cyclic intermediate by C, the required reaction paths are... 

Unlike the behavior over 0.2% platinum/alumina, the main features of the labeled product distributions obtained over 10% platinum/alumina and over platinum film catalysts (Tables VI and VII respectively) cannot be explained in terms of a single dominant reaction pathway via an adsorbed C6 cyclic intermediate. Again, parallel, multiple-step reaction pathways are involved. The results from 2-methylpentane-2-13C have been qualitatively accounted for (84) by the pathways... 

Only nonpolar solvents, e.g., cyclohexane (most commonly employed with silica gel and silicic acid), methylcyclohexane, methylpentane, and carbon tetrachloride, can be utilized in these slurries since solvents of greater polarity will compete with the intended adsorbate for available binding sites and will result in incomplete sustrate adsorption. 

The idea of the evidence is rather simple and can be elucidated by means of the following experiment. Let us consider, for example, a molecule of 2-methylpentane labeled in a branched position by 13C 2-methyl- 13C(2)-pentane. If the consecutive reactions in the adsorbed state are with a given metal of low extent, and this is certainly true for Pt or Pd, then the appearance, among the product, of 3-methyl-l3C(3)-pentane is very strong evidence of the operation of the 5C (cyclic) intermediates. Only via a ring closure at one place and an opening at another place of the molecule can a label move simultaneously with the branch. On the other hand, when the branch and labeled atom become separated by isomerization, this is evidence of the operation of the 3Cay complexes (see Fig. 5). 

The product distributions in the ring opening of substituted cyclopentanes show striking similarities to those in isomerization of the corresponding alkanes (n-hexane, 2- and 3-methylpentane),301 which led to the formulation of the cyclic mechanism involving adsorbed cyclopentane intermediates to interpret alkane isomerization15,251,252 (see Section 4.3.1). 

For some reason unknown to us, a large amount of 2-methylpentane became adsorbed on the molecular sieves at its critical point. 

Another study examined the NH3 and CO2 adsorption heats on several zirconia catalysts, differing in their preparation procedure and/or in the addition of dopants [46]. The differential heats of NH3 and CO2 adsorption show a wide range of variability, displaying either a plateau of constant heat or a continuous decrease indicative of adsorption heterogeneity [12]. The ratio between the number of the basic and acidic sites, Ub/ua, was calculated for each catalyst from the microcalorimetry results, by dividing the amount of adsorbed CO2 by the amount of adsorbed NH3. These catalysts were used to produce alk-l-ene from 4-methylpentan-2-ol. Alk-1-ene selectivity was found to first increase with the b/Wa ratio, reach a maximum and then decrease, whereas ketone formation continuously increased, being negligible for low Ub/ua values. 

It is agreed with literary data of absence of steric hindrance for adsorption of normal hydrocarbons on zeolites of a pentasile type [5-8]. Adsorption isotherms of 3-methylpentane lay below ones of n-hexane and starting US-69 sample at identical p/ps, the decrease of adsorption volume capacity made about 10%. In the investigated p/ps range adsorption isotherm of benzene on US-69 sample lays below, than for 2,3-dimethylbutane, and in initial area - even is lower than for cyclohexane contrary to a ratio of there kinetic diameters. At the same time, benzene isotherm increases faster, so that level of saturation for benzene can lay above, than for 2,3-dimethylbutane. Apparently, such character of benzene isotherm is connected to a feature of packing of molecules in pentasile channels, and also with stronger interaction adsorbate-adsorbate in comparison with interaction adsorbate-adsorbent. 

Adsorption isotherms of 3-methylpentane and benzene on the modified samples are close, though for BP-US-69 (2) they lay a little above. As to hydrocarbons with the greatest kinetic diameter - cyclohexane on the modified samples did not keep. With a sufficient reliability it was possible to receive only on one point of isotherm, for which the adsorption value has made about 0.45-0.50 from value for US-69. The greatest distinctions between the modified samples are found under adsorption of 2,3-dimethylbutane, which was kept enough well on BP-US-69 (2) and was not adsorbed on BP-US-69 (1). 

In order to examine the possible participation of adsorbed cyclopropanes in the bond shift mechanism, the relative contributions of Paths A and B in chain lengthening were determined for a series of 2-methylalkanes (40, 43, 54). The contribution of Path B was found to decrease regularly from isopentane to 2-methylpentane and 2-methylhexane. The decreasing contribution of Path B from isopentane to 2-methylpentane is readily explained by the decreasing number of methyl substituents in the cyclopropane intermediate, but the difference between 2-methylpentane and 2-methylhexane cannot be accounted for by the cyclopropane mechanism (43). 

For 2-methylpentane cracking the assumed reaction mechanism was based on that proposed by Zhao et al [8], Reactions are initiated by adsorption of 2-methylpentane feed on to Bronsted acid sites. Subsequent protolysis produces a carbenium ion on the surface of the catalyst (an adsorbed olefin) and a smaller gas phase paraffin molecule. Propagation reactions can then occur by hydride transfer between carbenium ions on the surface of the catalyst and gas phase feed molecules. Zhao experimentally determined certain bimolecular reactions to be more significant than others, the most significant were implemented in our model. Reactions are terminated by desorption of carbenium ions to yield a gas phase olefin molecule... 

Many other adsorbates have been used to test pore accessibility and adsorption capacity changes upon coke deposition. Some examples are trimethylamine and ethyldiisopropylamine on ZSM-5 , m-xylene on USHY and HZSM-5, 3-methylpentane in H-Offretite xenon on ZSM-5 and USY zeolites. ... 

Theoretically calculated values of the heat of adsorption for n-hexane and 2-methylpentane are 70 kj mol and 65 kj mol, respectively [46,47], which is in agreement with the average values determined by Zhu et al. [48]. As the heats of adsorption of these alkanes are very close, the difference in adsorption is caused by an entropic effect. Indeed, the conformations of the bulkier branched alkanes are much more restricted in the narrow pores of the medium-pore MEI zeoUte. Eor the branched isomer in siUcaUte-1 there is a large difference in the adsorption entropy between the molecular locations in the intersections and in the channels as shown by Zhu et al. [48]. Therefore, the adsorption of 2-methylpentane from the gas phase leads to a higher reduction in entropy compared to adsorption of n-hexane. This makes it en-tropically less favorable to adsorb the branched isomer [44]. 

Table 2 shows the adsorbed concentrations of the pure components. At a partial pressure of 6.6 kPa the amount of n-hexane is just slightly higher than that of isohexane in silicalite-1, while the linear alkane is obviously adsorbed more strongly than 2-methylpentane in H-ZSM-5 due to the stronger interaction with the acid sites. The maximum loading of each component has been measured by a separate adsorption study. The sorption capacity of n-hexane (7 molecules per unit cell), in agreement with earlier studies [48,59-61] exceeds that of 2-methylpentane (4 molecules per unit cell). The latter value equals the number of channel intersections in the MFI pore system per unit cell. Indeed, the sorption of isohexane molecules at... 

Figure 15 displays the loadings of n-hexane and 2-methylpentane in both zeolites. Under similar conditions, the adsorbed concentration of n-hexane is higher than that of 2-methylpentane, especially at lower temperatures. The interaction with n-hexane results in higher loadings for H-ZSM-5 than for silicalite-1. From the temperature dependence of the diffusivity of n-hexane in both zeolites, the apparent activation energy has been deduced and the results are collected in Table 3. Corresponding Arrhenius plots are shown... 

A sample of results illustrating the diffusion/reaction kinetics approach applied to the cracking of alkanes on H-ZSM-5 is given in Table 2. The shape-selectivity of hexane versus 3-methylpentane is exclusively due to transition state shape-selectivity (S )=l S j transition-state complex for bimolecular hydride transfer between the reactant and an adsorbed sec. propyl cation [29]. For hexane, the cross-section of this transition state complex is estimated to be 0.49 x 0.6 nm, while for 3-methylpentane, it is ca. 0.6 X 0.7 nm. gem-Dimethylbranched alkanes have reduced diffusional mobility in... 

---