logo


Oct-2002

Improving hydrogen plant performance: Part II

Optimisation of operating conditions requires an in-depth study of the hydrogen plant, say the authors. They consider the optimisation of catalyst and absorbent to maximise hydrogen output beyond the capabilities of operations strategies

P V Broadhurst and P E J Abbott
Johnson Matthey Catalysts

Viewed : 6787


Article Summary

In Part I of this article, published in the Summer 2002 issue of PTQ, several strategies for optimisation of operating conditions to increase hydrogen output were discussed. Another approach to increase plant hydrogen output is to optimise the choice of absorbents and catalysts used in the hydrogen plant. The ability of absorbents and catalysts to deliver increased output can be viewed in three ways: improved conversion to hydrogen (through increased catalyst activity), increased plant rate (through increased catalyst activity and/or lower pressure drop) and increased online time (through increased robustness).

Changes to catalyst and absorbents will often impact on more than one of these factors. To examine how output may be improved, each section of the hydrogen plant flowsheet (Figure 1) has been examined. In order to illustrate the effect of the changes, examples are taken from Synetix’s experience with its own products, but the principles underlying these points are generally applicable to the selection process for catalysts and absorbents.

Hydrocarbon purification section
This section removes trace levels of mainly sulphur and chlorine compounds from the hydrocarbon feed to avoid poisoning of downstream catalysts. The section comprises up to three stages.
— Hydrodesulphurisation (HDS) or hydrogenation
— Chloride removal
— Hydrogen sulphide (H2S) removal

The typical layout for the section varies and some examples are shown in Figure 2. The HDS stage removes trace levels of organic sulphur compounds in the hydrocarbon feed by reaction with hydrogen (either in the feed or provided as a recycle stream from the hydrogen plant product). The sulphur is converted to H2S. Any traces of organo-chlorides are similarly converted. The usual catalyst is a sulphided cobalt oxide/molybdenum oxide/alumina (CoMo) operated at 350–400°C, although under special circumstances a nickel/molybdenum (NiMo) catalyst is preferred.

Changes to this catalyst cannot add directly to the hydrogen production. However, other catalyst changes may lead to an increase in throughput. Alternatively, additional hydrogen may be produced by changing the hydrocarbon feed and this could have a higher organo-sulphur content. In these cases, the capability of the installed catalyst type and volume must be checked against the amended duty.

An increase in rate of up to 10% can usually be accommodated with some adjustment to inlet temperature and expected catalyst life, as there is some margin in the design method. If the installed HDS catalyst cannot cope, it may be possible to install a more active material with a higher level of the active metals, either CoMo or NiMo. In regard to time online, a typical HDS catalyst has a long life and lasts a number of operational cycles. Lives of 10 years are not uncommon. Thus, early failure, which takes the hydrogen plant offline, is very uncommon.

 Many feeds do not contain chloride and in these cases there is no need for a chloride removal stage. Where the feed does contain chloride, it passes from the HDS stage as HCl. The material used in the chloride removal stage is, therefore, an absorbent for HCl. As the hot gas exits the HDS at >350°C, a physical adsorbent is inappropriate and the preferred absorbents react irreversibly to capture the chloride as a metal chloride.

As with the HDS catalyst, the HCl absorbent cannot influence the hydrogen output directly, but an increase in hydrogen plant rate (or possibly a change in hydrocarbon feed type) would lead to an increased chloride burden. The ability of the HCl absorbent bed to cope must therefore be evaluated.

If the plant features a single HCl absorbent bed, then the original design will have usually sized the bed to run between turnaround intervals. If throughput is increased, the bed life will be shorter and it may not be possible to run for the required time. In these circumstances two approaches are possible. One is to seek an absorbent with a higher chloride capacity so that the same installed volume will have a longer run length. A second approach is to review the layout of the purification section.

It may be possible to rearrange the location and/or volume of materials between the available reactors, such as adding HCl absorbent to the top of the H2S removal beds. It is also possible that the increased throughput may exceed design space velocity. If this is the case, then the installed volume of HCl absorbent should be increased. If the HCl absorbent is installed on top of the H2S removal absorbent in a lead/lag pair of beds, absorbent volumes can be adjusted if needed and changed out on line.

After the HDS stage the sulphur is present as H2S. This is removed using an absorbent based on zinc oxide (ZnO). This operates typically at >350°C and reacts with the H2S to form zinc sulphide. Again, the H2S absorbent cannot influence hydrogen output. An increase in hydrogen plant rate (or maybe a change in hydrocarbon feed type) would lead to an increased sulphur burden. The ability of the H2S absorbent bed to cope must be evaluated.

Most hydrogen plants have a pair of H2S absorbent beds in an interchangeable lead/lag arrangement. This allows renewal of the absorbent without taking the plant off-line and so an increase in sulphur level is not an issue unless there is a very significant increase.

Some hydrogen plants, however, feature only a single H2S absorbent bed. In this case, the approach required is as described for the HCl absorbent. Care is required in selecting the H2S absorbent when seeking higher sulphur capacity in a single bed situation. Generally, a ZnO based absorbent with higher saturation capacity for H2S (kg S/m3) has a higher density. As density is increased the absorbent porosity tends to decrease and this increases the mass transfer zone for H2S removal. Thus, at the point of sulphur breakthrough the average capacity over the whole bed may be less for the denser ZnO-based absorbent due to the increased depth of the mass transfer zone (the bed depth required to remove H2S to the required level).


Add your rating:

Current Rating: 3


Your rate: