Recycle Loops and Purge Streams

Introduction

A reversible reaction may leave much of its feed unreacted in one trip through a reactor. Industrial processes can still use fresh feed efficiently by separating product and returning unreacted reactants. That recycle changes the inlet to the next pass. It also creates a new problem: inert substances that do not react can build up unless some loop gas leaves through a purge. A clear process boundary makes both effects easy to reason about.

Core explanation

In an ammonia synthesis loop, nitrogen and hydrogen enter a catalytic reactor and an effluent leaves containing ammonia plus unreacted gases. A product-separation stage removes much of the ammonia, and a compressor returns most of the remaining nitrogen and hydrogen. A small purge may remove argon or other inerts that entered with the feeds, along with some valuable reactants. The purge composition, recycle flow and fresh-feed rate are linked by steady-state material balances.

Single-pass conversion is defined using the reactor inlet , which contains both fresh and recycled reactants. If 100 mol h⁻¹ of a reactant enters a pass and 20 mol h⁻¹ reacts, single-pass conversion is 20%. Overall conversion uses the fresh feed entering the whole plant and counts all product and material leaving its boundary. Recovered unreacted feed can traverse the reactor repeatedly, so overall fresh-feed utilisation can be much higher than 20%. Reporting only one number without its boundary invites confusion.

Recycle does not change the equilibrium constant at a fixed temperature. Product removal changes the composition of material returned to the reactor; the next pass therefore starts from a different state and again moves toward the equilibrium relation. The total plant can convert more fresh feed than any single reactor pass because it is a sequence of reaction and separation operations, not one closed equilibrium vessel. The energy cost of separating and compressing recycled gas must be included in judging whether the loop is worthwhile.

Inerts need special attention. If an inert enters a closed recycle loop at 1.0 mol h⁻¹ and no inert can leave with product, its inventory increases continuously unless a purge removes it. At steady state, inert input must equal inert output. If a purge removes 10% of the well-mixed loop gas after the separator, the inert flow in the gas before the purge must be 1.0/0.10 = 10 mol h⁻¹, so that 1.0 mol h⁻¹ leaves in the purge. This simple result shows why a small purge fraction can coexist with a substantial circulating inert inventory. It also shows why a purge loses some reactants: the purged gas is a mixture, not a pure inert stream.

The same mass-balance principle applies to impurities that react very slowly or to by-products that cannot be fully separated. They may accumulate until their purge or other removal matches their addition or formation. If inerts become too concentrated, they dilute reactive-gas partial pressures and can lower reactor performance at a given total pressure. Choosing the purge fraction is therefore a tradeoff between controlling inert build-up and losing valuable feed.

For a quantitative ammonia loop, use separate balances for N₂, H₂, NH₃ and each inert. The reactor reaction extent obeys N₂ + 3H₂ → 2NH₃, while separator and purge split each species according to the equipment's performance. A single “recycle percentage” is not enough to compute everything unless separation efficiencies, purge location and compositions are specified. A process-flow diagram provides the structure before the algebra.

Step-by-step reasoning

1. Draw the reactor, product separator, recycle line, fresh-feed connection and purge branch. 2. Choose a boundary around the reactor to define single-pass conversion and a boundary around the entire loop to define overall conversion. 3. Write component balances, including N₂, H₂, NH₃ and each inert rather than only a total-mass balance. 4. Set inert input equal to inert purge output at steady state. 5. Apply the purge fraction to the correct pre-purge mixed stream. 6. Account for reactant loss in purge and the energy required for separation and recompression.

Visual explanation

Sketch a reactor feeding a condenser or product separator. Draw an ammonia product arrow leaving the separator and a gas arrow continuing to a T-junction. At that junction, a small arrow leaves as purge and the larger arrow loops back to the reactor, joining fresh N₂ and H₂. Mark the loop with an inert dot that repeatedly travels around it; without the purge arrow, the dots multiply each cycle.

Real-world analogy

Imagine recycling wash water in a factory while a small amount of dissolved salt enters with every new batch. Recycling saves water, but salt concentration would keep rising unless some water is discarded or treated. The ammonia loop likewise conserves useful reactant gas, while a purge controls substances that cannot be consumed by the desired reaction.

Real-world example

An ammonia producer may purify nitrogen and hydrogen but still allow a little inert gas into the synthesis loop. Product ammonia is removed, and the remaining reactive gases are recycled. A deliberate purge prevents inert build-up, even though it sends some nitrogen and hydrogen out of the loop too. The best purge amount depends on purification quality, separation performance and the value of lost feed.

Why?

Why is a purge necessary if the loop is well sealed? Sealing prevents leaks but does not remove inert material continually introduced with fresh feed. At steady state, every species needs an exit or a consuming reaction to balance its entry. An inert has no consuming reaction, so a purge or dedicated separation is its required outlet.

Common misconception

“Twenty-percent single-pass conversion means 80% of fresh feed is wasted.” If unreacted feed is recovered, much of it returns for additional passes. Waste or loss depends on purge, leaks, separation inefficiency and other exits. Single-pass conversion and overall fresh-feed use have different denominators and process boundaries.

Worked example

An inert enters an idealised synthesis loop at 1.0 mol h⁻¹. It is not consumed or removed with condensed product. The gas leaving product separation is well mixed, and 10% of it is purged. At steady state the purge must carry 1.0 mol h⁻¹ inert, matching the inlet. Therefore the inert flow in the gas immediately before the purge is 1.0/0.10 = 10 mol h⁻¹. The remaining 9.0 mol h⁻¹ inert circulates in the recycle branch. This calculation does not determine the total purge flow until the inert mole fraction of that gas is known.

Quick check

1. An inert enters at 2.0 mol h⁻¹ and a purge removes 5% of a well-mixed loop stream. What pre-purge inert flow is needed at steady state? Answer: Inert output must equal 2.0 mol h⁻¹, so pre-purge inert flow is 2.0/0.05 = 40 mol h⁻¹.

Exam focus

Label every boundary and stream before setting up equations. Write “reactor inlet” in the denominator for single-pass conversion and “fresh feed” for overall conversion. In a purge calculation, balance inert at steady state and remember the purge removes a fraction of all gas species, not only the inert.

Advanced insight

Recycle loops can amplify disturbances. A small change in feed impurity or separator efficiency may propagate through repeated circulation before reaching a new steady state. Process control therefore monitors loop composition as well as product rate and reactor temperature. An economic optimisation may choose greater feed purification to reduce purge losses, or accept a larger purge to simplify purification. Both choices require a whole-loop balance and cost model.

Summary

Product removal and recycle let reactants make multiple passes, so whole-plant fresh-feed conversion can exceed single-pass reactor conversion without changing equilibrium thermodynamics. Inerts entering a loop accumulate unless removed. A steady-state inert balance sets purge removal equal to inert input, while the purge also wastes some reactive gas. Clear process boundaries and component balances are essential for correct calculations.

Practice questions

1. A reactor inlet contains 200 mol h⁻¹ N₂ and consumes 30 mol h⁻¹ in one pass. Find single-pass N₂ conversion. Answer: It is 30/200 × 100 = 15% for that reactor pass. 2. Why can overall fresh-feed conversion exceed the answer to question 1? Answer: Unreacted N₂ can be separated and recycled through additional reactor passes. 3. An inert enters at 0.50 mol h⁻¹ and 2% of the well-mixed loop is purged. Find pre-purge inert flow at steady state. Answer: It is 0.50/0.02 = 25 mol h⁻¹ so that purge removes 0.50 mol h⁻¹. 4. Why does increasing purge usually increase fresh reactant demand? Answer: The purge contains useful N₂ and H₂ as well as inert gas, so more of those reactants must be replaced.