Protection analysis in refining

Independent pilot plant testing can serve as an independent protection layer to reduce the technical risks of new process technology

Robert Absil
Intertek PARC

Viewed : 4048

Article Summary

Factors contributing to 
technical risks when implementing new process technologies in the refining industry include the complex compositions of crude oil fractions and of heterogeneous catalysts used in the conversion of those fractions to marketable products. Further-more, reactor hydrodynamic performance and operational issues can pose significant technical risks. The approach proposed in this article is to use layer of protection analysis, which has its origin in the process plant safety field, to mitigate these technical risks. The motivation behind this approach is to semi-quantify the risks that are associated when implementing new process technologies so that they can be controlled, with decisions being made based upon objective evidence. To illustrate this approach, one operational issue — plugging of reactor beds — is analysed and a risk mitigation programme is proposed. Furthermore, independent pilot plant testing is proposed as an independent protection layer to reduce technical risks.

Refinery process implementation risks
Feedstock risks

When considering technical risk in the refining industry, the crude oil feedstock is one major contributor. Conventional crude oil is a complex mixture of hydrocarbons that boil over a wide range of temperatures. Reaching beyond the maximum cutpoints of vacuum-packed column distillation (~1040°F, 560°C) and high-vacuum, short-path distillation (~1300°F, 700°C), sequential elution fractionation indicates that the maximum atmospheric equivalent boiling points (AEBP) of crude oils can exceed 2500°F (1370°C). The hydrocarbons include paraffins, cycloparaffins, aromatics, resins and asphaltenes. They also contain heteroatoms, such as sulphur, nitrogen and oxygen, in their structures. Furthermore, crude oils contain metals, such as vanadium, nickel, iron and copper. In addition, crude oils contain salts, which are dissolved or suspended as crystals in very small water droplets emulsified in the crude oil.

To illustrate the complexity of a crude oil or its fractions, Altgelt and Boduszynski listed the number of isoparaffins that can exist as a function of carbon number (see Table 1).1 Similar structural variations exist for cycloparaffins, aromatics, resins and asphaltenes at a given carbon number.

Moreover, hydrocarbons may contain one or more heteroatoms in their structures. Sulphur exists, for example, as mercaptans, sulphides, disulphides, benzothiophenes, thiophenes and dibenzothiophenes; oxygen as phenolics and carboxylic acids; and nitrogen as indoles, pyridines, amines and quinolines.

The technical risk associated with the complexity of feedstock composition is that the potential outcome of a process technology implementation involving a new feedstock may not be completely known, since no two crude oils are the same. Altgelt and Boduszynski pointed out that a complete compositional analysis of heavy fractions — and thus also of the crude oil — is impossible and that simplifications must be made.  

Instead of focusing on individual compounds, groups or lumps of compounds that fall into similar chemical classifications have been used with success.2 However, the less is known about the feedstock, the greater the technical risk and the higher the probability of failure on demand of the programme.

Catalyst risks
Another important factor affecting risk assessment is the heterogeneous catalyst used in the conversion process. The composition of a solid catalyst can be complex; the catalyst is often dual-functional, in that the acidic function provided by the zeolite catalyses cracking and isomerisation reactions, and the metal function catalyses hydrogenation reactions, as is the case with hydrocracking catalysts. Moreover, a binder, such as alumina, is used as support or to give the solid catalyst particle mechanical strength. Although substantial advances have been made in understanding heterogeneous catalysts, factors such as diffusion limitations, poison sensitivity, limited understanding of reaction mechanisms 
and of the structure of the 
active sites contribute to technical risk.

Again, the technical risk associated with catalyst complexity is that the potential outcome of new process technology involving a new catalyst is not completely known.

Reactor hydrodynamic performance risks
In a trickle-bed reactor, the ideal flow pattern is that the liquid flows as a continuous film over the catalyst particles, with the hydrogen flowing through the remaining void space and diffusing through the liquid film to the catalyst surface. However, commercial trickle-bed operations have been shown to be non-ideal and are affected by, for example, liquid and gas maldistribution and mass transport limitations.3

Since no actual statistics were given in the McGovern report,3 a probability of failure on demand cannot be calculated. However, for the sake of argument, if 10 out of the 50 hydrotreaters started up industry-wide were not able to meet the performance predicted by properly packed pilot plant units, the probability of failure on demand of the commercial trickle-bed processes in this application would be 0.2.

Operational risks
Another risk often overlooked is associated with operational issues, which are due to problems pertaining to the operation of an existing refinery unit, such as a co-current, trickle-bed reactor, and can play a key role on whether the implementation of a new process technology is a success. These include issues related to plugging of reactors, excessive wear of control valve stems or unanticipated corrosion of equipment. Switching to processing feedstocks other than conventional crude oil may lead to unanticipated operational problems. Early reports indicated that hydroprocessing of residues was affected by reactor bed plugging due to fines, such as dirt, sand, salt and corrosion products.4 Hydrotreating of gas oil to produce ultra-low sulphur diesel was impacted by fines accumulation.3 Plugging of hydrotreating catalyst beds by colloidal and non-colloidal fines has been reported in the processing of oil sands-sourced gas oil.5

Add your rating:

Current Rating: 3

Your rate: