The role of sulfur in commercial iron-based Fischer–Tropsch catalysis with focus on C2-product selectivity and yield

This paper presents some insight into the role of sulfur, a well-known Fischer–Tropsch (FT) catalyst poison, regarding the selectivity of ethylene:ethane and C₂-product yield in a commercial, Fe-based, high-temperature, 2-phase FT process.

It is well known that in these kind of processes, the level of sulfur allowed in the feed gas is normally very low (ppb range) and aggressively kept low due to the poisoning effects. The practice generally is to get sulfur levels down as close as possible to zero.

The poisoning effects are well studied in research work and significant information exists in that regard. It is, e.g. known that the poisoning action of sulfur would typically increase selectivity towards more hydrogenated ethane relative to ethylene (the more attractive product). At high enough levels it would kill catalyst activity completely.

This paper highlights the role of sulfur in a typical commercial operation. Data from a period of almost 2 years was used. It was found that under the prevailing actual commercial scale operating conditions, sulfur did not behave as a poison to the catalyst. However, its effect did not become completely negligible either. There were strong indications of sulfur promoting selectivity shifts towards more olefinic C₂-product. These results were also independently confirmed on a different set of data and a different kind of commercial reactor.     

Comparing the results from various performance models of IEEE 802.11g DCF

IEEE 802.11 DCF is the most widely-used CSMA/CA access control mechanism. Recent analytic performance models for DCF have received acclaim for both their simplicity and reported accuracy. Most of these models share the assumptions of full single-hop connectivity among all stations, that DCF back-off may be modeled as a Markov process and that the network is saturated with traffic. In order to verify the accuracy of existing analytic models we developed a discrete-event simulator to record the performance of the DCF protocol and ensure that every detail of the standard is represented. Simultaneously we set up a hardware test bed to measure the same performance metrics in an environment that makes none of the simplifying assumptions of either the analytic models or the simulation. In the test bed, as in the simulator, we used the same physical parameter settings prescribed by the standard. As is the case for the analytic models we used, we subjected the simulator and the test bed to the same saturated workload for both basic and RTS/CTS access modes. Finally, we also implemented a non-saturating Markov Modulated Arrival Process (MMAP) workload model for our simulator to test the performance of DCF subject to more realistic internet traffic conditions. We describe both the simulator and the test bed in some detail in order to testify to the accuracy and detail of our results. The results show that the analytic models are mostly pessimistic for small numbers of nodes and optimistic for larger numbers of nodes. The performance measurements from the test bed, in turn, indicate that the simulation results are similarly optimistic when large numbers of nodes are concerned. Since the test bed uses an error-prone wireless channel, this latter result is, in principle, not surprising. The rate of deterioration in the actual performance is however something that is not widely known and is much more rapid than analytic models would suggest.     

A probabilistic dynamic technique for the distributed generation of very large state spaces

Conventional methods for state space exploration are limited to the analysis of small systems because they suffer from excessive memory and computational requirements. We have developed a new dynamic probabilistic state exploration algorithm which addresses this problem for general, structurally unrestricted state spaces.

Our method has a low state omission probability and low memory usage that is independent of the length of the state vector. In addition, the algorithm can be easily parallelised. This combination of probability and parallelism enables us to rapidly explore state spaces that are an order of magnitude larger than those obtainable using conventional exhaustive techniques.

We derive a performance model of this new algorithm in order to quantify its benefits in terms of distributed run-time, speedup and efficiency. We implement our technique on a distributed-memory parallel computer and demonstrate results which compare favourably with the performance model. Finally, we discuss suitable choices for the three hash functions upon which our algorithm is based.     

SNAPL/1: A language to describe and evaluate queuing network models

Recent theoretical developments in queuing theory have made multiclass queuing network models a viable alternative to established simulation methods for the analysis of dynamic systems. Computer software is required to describe and solve such network models. This paper describes a language which provides the human interface to a particular multiclass queuing network modelling package. The discussion is illustrated with an example from the language and concludes with suggestions for improvements.     

Two-Shockley unfaulting of two- and three-layer faulted defects

The principles of butterfiy-unfaulting of Frank dislocation loops, i.e. unfaulting by the simultaneous action of two Shockley partial dislocations (Swart & Kritzinger, 1974a), are applied to two-layer faulted defects in thin foils of quenched aluminium to explain the structure and contrast behaviour of some dislocation configurations observed by transmission electron microscopy. Direct electron microscope evidence is presented for the butterfiy-unfaulting of three-layer faulted defects in the same specimens. Positive evidence is furnished that Shockley partial dislocations can be nucleated at sites in the stacking fault other than the hexagon corners of the defects.     

Facility for rapid preparation of silicon and silicide TEM specimens

A simple chemical jet polishing arrangement, for the thinning of semiconductors or metals for transmission electron microscopy (TEM), is described for the specific case of silicon and silicides. The effect of variation in three mechanical parameters on the profile and quality of the specimen is described, and the optimum conditions are determined. A proposed polishing solution is one part 48% HF mixed with one part fuming HNO₃. Reproducibility is only achieved in the presence of a relatively large concentration of nitrous acid in the polishing solution. The solution must also be relatively concentrated, as the reaction rate falls off rapidly with decreasing concentration.     

A generalisation of Norton's theorem for multiclass queueing networks

Consider an arbitrary subset σ of service centres in a locally balanced multiclass queueing network for which a parametric analysis is to be undertaken. It is shown that the complete network only has to be solved once using the convolution algorithm, after which statistical measures for σ can be calculated repeatedly without re-evaluating the rest of the network. It has also been proved that the concept of replacing a subnetwork in a closed multiclass queueng network with a single composite service centre with a state dependent service rate (also called Norton's Theorem for queuing networks) is a very special case of the more general result mentioned above. An example is given in which the theoretical results are applied when σ is a subnetwork of a closed multiclass queueing network satisfying local balance.     

Local beam heating in metallic electron microscope specimens

A very sensitive method is described to measure the local rise in the temperature of a metal specimen during investigation in the electron microscope. The method is based on the known temperature dependence of the climb rate of Frank dislocation loops in materials with relatively high stacking fault energy. The climb rate of loops subject to the influence of the condensed electron beam is compared to the corresponding rate for loops in an area of the same specimen which is remote from the observation spot. From the difference in the logarithms of these rates the amount of local beam heating is determined. For a typical Al-1.5 wt% Mg specimen under typical normal working conditions an average local rise in temperature of approximately 6°C was determined.