Question 5 of 6: Ten Hierarchical Elements Intrinsic to a Chemical Process

Question 5: Ten Hierarchical Elements Intrinsic to a Chemical Process (20 marks)

Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.

Chemical processes are not assembled all at once; they are built up in layers, and the layers have a fixed order. The chemistry sits at the centre; around it go the reactor, then the recycle structure, then the separations, then heat recovery, then the utilities, and finally the environmental, safety and control systems that wrap the whole plant. This is the “onion” or hierarchical view of process design (Douglas; Turton et al.). It matters because each layer fixes the degrees of freedom available to the next: the reactor’s conversion and selectivity decide what the separation train has to do, the separation train decides what heat has to be moved, and the whole flowsheet decides what has to be treated before it leaves the fence. Ten elements, in that hierarchical order, are described below.

1. Input information and the reaction chemistry (route selection). The innermost element: the reaction path itself — stoichiometry, feedstock slate, product and by-product specifications, physical-property and phase-equilibrium data, and any constraints on materials or conditions. Choosing a route is the single most consequential decision in a design, because raw materials typically dominate cost and the atoms that do not end up in the product are the plant’s waste by definition. A route is screened on its economic potential (product value minus raw-material cost) before anything is sized, exactly as in Question 1.
2. Batch versus continuous mode of operation. The first structural decision. Small tonnages, multi-product campaigns, short product life, long reaction times or solids handling push toward batch; large, stable tonnages push toward continuous. The choice determines whether the process is described by a steady-state flowsheet or by a schedule, and it changes the equipment, the control philosophy and the way inventory (and therefore hazard) accumulates.
3. Input–output structure of the flowsheet. The plant drawn as a single box: what enters (fresh feeds, air, water, inerts), what leaves (products, by-products, purges, effluents) and the overall material balance that connects them. At this level you decide whether to purify a feed, whether a gas recycle needs a purge for inerts, and whether a by-product is sold, recycled or destroyed. The economic potential is re-evaluated here with by-product credits and feed-losses included.
4. The reaction system (reactor). Reactor type (CSTR, tubular, fluidized bed, batch), catalyst, temperature, pressure, phase, conversion and selectivity. This element consumes the chemistry decided at level 1 and creates the separation problem at level 7. Selectivity is the key economic variable: unconverted feed can be recycled, but a molecule lost to a by-product is lost as raw material and reappears as a separation duty and a waste stream.
5. Recycle structure. How unconverted reactant is returned, how many recycle loops there are, where a purge is taken and how large it must be, and what the recycle costs in compression, pumping and equipment size. The classic trade-off lives here: lower per-pass conversion improves selectivity but enlarges the recycle and every unit it passes through, so the reactor and the recycle must be optimised together, not separately.
6. Vapour recovery and the gas-separation system. The first of the two separation layers: condensation, absorption, adsorption, membranes or cryogenic recovery applied to the reactor’s vapour effluent, deciding what returns to the recycle, what is purged and what is vented. Its position in the hierarchy matters — it is specified before the liquid train, because it sets the liquid train’s feed.
7. The liquid-separation system. The distillation, extraction, crystallisation or drying train that takes the liquid streams to product specification. The decisions are which separations are needed, in what sequence, and by what method — distillation unless the relative volatility approaches one, an azeotrope intervenes or the material is heat-sensitive. This layer usually owns the largest share of the plant’s energy consumption.
8. Energy integration — the heat-exchanger network. Only once the streams and their duties exist can hot and cold streams be matched against each other. Pinch analysis fixes the minimum hot and cold utility demand for a chosen approach temperature and identifies the pinch across which heat must not be transferred. Good integration cuts both operating cost and emissions, but it also couples the plant together and reduces its operability, which is why it sits outside the reactor and separation layers rather than inside them.
9. Utilities, storage and offsites. Steam and power generation, cooling water and refrigeration, compressed air and inert gas, fuel, tankage and feed/product storage, loading facilities and site services. These are chosen to serve the duties left over after integration, and their selection (steam levels, refrigerant, cooling-water versus air cooling) is itself an economic and environmental decision.
10. Environmental control, safety and process control — the outer layer. Effluent treatment, air-emission abatement, solid-waste handling and water reuse; hazard identification and the inherently safer decisions of minimising, substituting, moderating and simplifying, together with relief, containment and layers of protection; and the instrumentation and control system that keeps the plant inside its safe and specification envelope. This layer is the last to be detailed but must be considered from the first, because the cheapest emission and the cheapest hazard are the ones designed out at levels 1–5 rather than treated here.

The hierarchy is what makes a design tractable: at each level a small number of decisions is made on the information then available, the economics are re-tested, and unpromising alternatives are discarded before they are costed in detail. It is also where sustainability enters. Roughly speaking, the inner levels commit the money and the molecules — route, reactor and recycle decide raw-material efficiency, and therefore both the cost and the waste — while the outer levels can only manage what the inner ones have already created. Designing so that the process “retains the capacity of ecosystems to support both industrial activity and life into the future” therefore means pushing environmental and safety judgements inward, into the route and reactor decisions, instead of leaving them to the treatment plant at level 10.