Diffusion of n-paraffins

Fig. 11. Diffusion of n-paraffins in a cis-polyisoprene as function of concentration at 51 °C. Lines are two-parameter fits of Eq. (17) to data. Paraffin carbon numbers are indicated. (Ref.70), with permission).

The rate of n-paraffin desorption generally controls the overall production rate (18, 19). The diffusion of n-paraffins in commercial 5A molecular sieves is reported to be controlled by either micropore diffusion or macropore diffusion, or both, depending on the molecular sieve crytal size and macropore size distribution of the adsorbent (20). A 5A molecular sieve adsorbent with smaller crystal size and optimum macropore size distribution would have a faster adsorption-desorption rate and, therefore, a higher effective capacity. 

Table 8 Transport diffusivities of n-paraffins in single crystals of H-ZSM-5...

Fig. 26 Arrhenius plot of transport diffusivities of n-paraffins (n-hexane, n-heptane, n-octane, n-nonane) in H-ZSM-5 single crystals...

SELF-DIFFUSION OF N-PARAFFINS IN A WIDE TEMPERATURE RANGE. 

Figure 14 Diffusivities of n-paraffins in FT wax (, experimental data at T=504 K —, Wilke-Chang correlation at 7=540 K ---correlation bylglesia etal. n, carbon number D , diffusivities of n-paraffins) (Erkey et al., 1990).

Examples of rate-selective adsorption are demonstrated using silicalite adsorbent for separation of Ciq-Cm n-paraffins from non- -paraffins [40, 41] and Ciq-Ch mono-methyl-paraffins from non-n-paraffins [42-45]. Silicalite is a ten-ringed zeolite with a pore opening of 5.4A x 5.7 A [22]. In the case of -paraffins/non-n-paraffins separation [40, 41], n-paraffins enter the pores of silicalite freely, but non-n-paraffins such as aromatics, naphthenes and iso-paraffins diffuse into the pores more slowly. However, the diffusion rates of both normal -paraffins and non-n-paraffins increase with temperature. So, one would expect to see minimal separation of n-paraffins from non-n-paraffins at high temperatures but high separation at lower temperature. 

The diffusion coefficients of n-paraffins with 12 to 22 carbon atoms in high density (HDPE) and low density polyethylene (LDPE) have been measured by a permeation method (Koszinowski, 1986). Methanol (MeOH) and ethanol (EtOH) were used as contacting liquid phases which minimized interaction between these polar solvents and the nonpolar polymers. No interaction was observed over the investigated temperature range of 6 to 40 °C for both solvents. 

Kinetic research on adsorption of n-decane from a solution of n-dec-ane in toluene to partially ion-exchanged type A Mg and Ca zeolites in batch processes showed that the rate of adsorption of the n-paraffin molecule on zeolite 5A depends on the degree of cation exchange. n-Dec-ane was not adsorbed below a cation exchange of 33%. The rate of diffusion of n-decane molecules increases with increasing cation exchange for the Mg zeolite, it reaches a maximum at about 49% cation exchange and then drops off somewhat, whereas it rises rapidly for the... 

Erionite and the related zeolite T exclude branched chain paraffins, so that n-paraffins are selectively cracked or hydrocracked over these catalysts. However, even n-paraffins have diffusities several orders of magnitude lower than in large pore zeolites. Earlier work on hydrocracking over erionite catalysts gave unexpected product distributions, with maxima at C4 and C12 and essentially no Cg product. Gorring measured diffusion coefficients for migration of n-paraffins in KT zeolite. An unexpected phenomenon, termed the window... 

Diffusion of linear paraffins was studied in a wide range of 5A zeolite crystals including both the small commercial (Linde) crystals and larger crystals synthesized in our laboratory by Charnell s method ( 31.). Whereas in the case of NaX we found no significant difference between the diffusivities for large laboratory synthesized and small commercial crystals (15), for 5A we see large differences. In particular, for n-butane, the large... 

It follows from the above results that small-pore molecular sieves yield principally hydrocarbons and very small amounts of Cg compounds. This may be related to diffusion constraints and cavity dimensions. Gorring has determined that the diffusion coefficients of n-paraffins in T-zeolite at 340 C decrease as the number of chain C-atoms increase from 2 to 8. Also, for most of the cases the length of zeolite cavities is less or equal to the length of the n-heptane molecule. Thus, it may be assumed that the cavity length imposes a restriction on the formation of Cg linear compounds. The combination of cavity dimensions and pore opening permits attaining high selectivities for C2-C4 linear hydrocarbons. 

From the uptake curves of the four n-paraffins (as shown, e.g., in Figs. 24 and 25 for diffusion of n-heptane and n-hexane, respectively) into H-ZSM-5 as a function of adsorption temperature, the corresponding Arrhenius plots were derived and are shown in Fig. 26. 

Gdrring [146] has determined that the diffusion coefficients of n-paraffins in T-zeolite at 340°C decrease as the number of chain C-atoms increase from 2 to 8. Also, for most of the small-pores zeolites and zeotypes, the length of the cavities is less or equal to the length of... 

In addition to CN and ON, the smoke point (SP), which is the maximum smoke-free laminar diffusion flame height, has been employed widely to evaluate the tendency of different fuels to form soot. This tool was first applied to kerosenes, later diesel, and then jet engine fuels.19,20 Researchers have tried to relate smoke points of pure compounds to their molecular structure. It was found that the inverse of smoke point, which measures the potential of a fuel to form soot, increases from n-paraffins to iso-paraffins to alkylbenzenes to naphthalenes.21,22 Since smoke points vary with experimental conditions, the concept of a threshold soot index (TSI), which is calculated from the smoke point, molecular weight, and experimental constants, has been used to compare the soot-formation tendencies of different fuel molecules.23... 

For the rate-selective separation of Ciq-Ch mono-methyl-paraffins from non-n-paraffins [42-45], diffusion simulations were carried out using the Solids Diffusion module in the Accelrys Insight II molecular modeling package [44]. The modeling results from the diffusion simulations of four paraffins of varying carbon numbers in siUcalite are summarized in Table 6.9. 

Figure 6.2 illustrates the separation of n-Csis and non-n-Cs/is in CaA molecular sieves or 5A. The separation mechanism is obvious when the kinetic diameter of the molecules and molecular sieve pore size opening are compared. n-Csjc have kinetic diameters of less than 4.4 A which can diffuse freely into the 4.7 A pores of the CaA molecular sieve, while non-n-Cs/ have kinetic diameters of 6.2A. A commercial example of shape-selective adsorption is the UOP Molex process, which uses CaA molecular sieves to separate Cio-C n-paraffins from non- -parafHns (aromatics, branched, naphthenes). 

The MMP Sorbex process has many similarities but also some differences when compared to the detergent Molex process. As with all of Sorbex processes, the MMP Sorbex process operates in the Uquid phase, employing suitable conditions (pressure, temperature) to overcome any diffusion constraints to achieve target performance. Table 8.4 highlights and contrasts the different characteristics of the detergent Molex and MMP Sorbex processes. The process was successfully demonstrated in a continuous countercurrent moving bed separation pilot plant using commercial n-paraffin-depleted kerosene (Molex raffinate) feedstock. A typical gas... 

The C4 Olex process is designed with the full allotment of Sorbex beds in addition to the four basic Sorbex zones. The C4 Olex process employs sufficient operating temperature to overcome diffusion limitations with a corresponding operating pressure to maintain liquid-phase operation. The C4 Olex process utilizes a mixed paraffin/olefin heavy desorbent. In this case it is an olefin/paraffin mix consisting of n-hexene isomers and -hexane. A rerun column is needed to remove heavy feed components such as Cs/C because they would contaminate or dilute the hexene/hexane desorbent. Table 8.5 contains the typical feed and product distributions. 

Discussion. Fixed bed cracking reactors as well as commercial moving bed reactors operate under steady state or pseudo-steady state conditions ( ). Observed selectivity (eg., ratio of yield of branched to n-paraffin) in a steady state catalytic reactor is independent of space velocity (1, 17). The selectivity depends on intrinsic rate constants and diffusivities of the reacting species which depend on temperature. Thus, the selectivity observations reported here are applicable to commercial FCC units operating at space velocities different from that employed in this study. 

The first PGSE investigation of a rubber-based ternary solution was described by Ferguson and von Meerwall31), who measured diffusion of C6F6(19F NMR) and n-paraffin (n-dodecane or n-hexatriacontane 1H NMR) in a commercial polybutadiene as function of both concentrations. They showed that both concentration dependences in the ternary region can be derived from the measured diffusivity of each diluent i = 1, 2 in binary solution in the rubber. To do this it was necessary to extend the Fujita-Doolittle expression, as follows ... 

Hydrocarbon distributions in the Fischer-Tropsch (FT) synthesis on Ru, Co, and Fe catalysts often do not obey simple Flory kinetics. Flory plots are curved and the chain growth parameter a increases with increasing carbon number until it reaches an asymptotic value. a-Olefin/n-paraffin ratios on all three types of catalysts decrease asymptotically to zero as carbon number increases. These data are consistent with diffusion-enhanced readsorption of a-olefins within catalyst particles. Diffusion limitations within liquid-filled catalyst particles slow down the removal of a-olefins. This increases the residence time and the fugacity of a-olefins within catalyst pores, enhances their probability of readsorption and chain initiation, and leads to the formation of heavier and more paraffinic products. Structural catalyst properties, such as pellet size, porosity, and site density, and the kinetics of readsorption, chain termination and growth, determine the extent of a-olefin readsorption within catalyst particles and control FT selectivity. 

Reference Eq. (6-20) for an infinite chain of covalently bonded methylene groups can be considered to be an asymptotic limit for the homologous series of n-alkanes. By substitution of w into the exponent of Eq. (6-20) by the corresponding term, wI (., which represents a matrix composed of a paraffin with i carbon atoms, an equation for the diffusion coefficient Dski for trace amounts of a paraffin with k carbon atoms results ... 

It has been known for many years that molecular structure of a fuel has a direct bearing on the tendency of that fuel to smoke, i.e., to form carbon or soot in a flame. For example, in 1954 Schalla (41), reporting on a study of diffusion flames, indicated that the rate at which hydrocarbons could be burned smoke free varied in the order n-paraffins — mono-olefins — alkynes — aromatics. This same phenomena has been reconfirmed by many authors in a variety of systems and always in the same general order (j6, J3, J 5, 1 7, J 9, 26, 43, 45). Paraffins have the least tendency to smoke, whereas the naphthalene series have the greatest tendency to smoke. 

The concept of shape selectivity relies on the control of the adsorption and diffusion of the molecule in the pore of the catalytic material. Linear paraffins have the smallest critical diameter among hydrocarbons and can therefore penetrate smaller pore structures. In the figure, for example, the n-heptane molecule penetrates the tubular... 

Matthews-Akgerman The free-volume approach of H ildebrand was shown to be valid for binary, dilute liquid paraffin mixtures (as well as self-diffusion), consisting of solutes from Cj to Cie and solvents of Ce and C12. The term they referred to as the diffusion volume was simply correlated with the critical volume, as Vp = 0.308 V . We can infer from Table 5-15 that this is approximately related to the volume at the melting point as Vp = 0.945 V . Their correlation was valid for diffusion of linear alkanes at temperatures up to 300°C and pressures up to 3.45 MPa. Matthews et al. and Erkey and Akger-man completed similar studies of diffusion of alkanes, restricted to n-hexadecane and n-octane, respectively, as the solvents. 

Figure 6. Comparison of diffusivities of MFI type between calculated by lattice model and the experimental data. B benzene, T toluene, p-X para-xylene, m-x wt /a-xylene, o-x c r/to-xylene, n-Ci (i=6-8) /-paraffin of carbon number i, i-C8 /.SY -octane.

Fig. 6 Variation of diffusivity with carbon number for n-paraffins in...

---