Chapter 1: What is Chemical Engineering?

This chapter explains why industrial-scale chemistry is a different discipline from bench chemistry, introduces the two ideas that organize the whole field — unit operations and balances — and teaches you to read a process flowsheet.

Unit Operations, Balances, and the Language of Processes

Learning Objectives

By completing this chapter, you will be able to:

• ✅ Explain why a reaction that works in a flask does not simply scale to an industrial reactor
• ✅ Define a unit operation and name the major ones by purpose
• ✅ Write and solve a steady-state material balance
• ✅ Read a block flow diagram and explain the role of a recycle stream
• ✅ Describe the shared structure of momentum, heat, and mass transfer

Reading Time: 20-25 minutes Code Examples: 1 Exercises: 3

1.1 From Chemistry to Chemical Engineering

A chemist gets a reaction to work in a 100 mL flask. The yield is good, the product is pure. Now make ten thousand tonnes of it a year.

This is not the same problem, and the reason is geometry. Run the reaction in a 100 m³ reactor — a million times the volume — and the vessel is about 100 times larger in each direction. Heat is generated throughout the volume, but removed through the wall:

• Heat generation ∝ volume ∝ L³ (grows 10⁶×)
• Heat removal ∝ surface area ∝ L² (grows 10⁴×)
• Surface-to-volume ratio ∝ 1/L — it falls by a factor of 100

An exotherm the flask shed harmlessly now has one-hundredth the relative cooling surface. Left unaddressed, the temperature runs away. The same argument applies to mixing (a second in a flask, minutes in 100 m³), to mass transfer between phases, and to the plain fact that a lab spill is a paper towel while an industrial one is an emergency.

Chemical engineering is the discipline of designing and operating processes that transform matter and energy at industrial scale — safely, reliably, and economically.

Chemistry answers what reaction occurs. Chemical engineering answers at what rate, in what equipment, at what cost, and what happens when something goes wrong.

1.2 Unit Operations: The Founding Abstraction

In 1915, Arthur D. Little — writing to MIT about how chemical engineering should be taught — argued that any process, however complicated, decomposes into a modest set of recurring physical steps. He called them unit operations. The idea reorganized the field: instead of teaching "sulfuric acid manufacture" and "soap manufacture" as separate recipes, you teach distillation, heat exchange, and filtration once, as engineering subjects in their own right.

The power of the abstraction is transferability. Learn distillation properly — vapor–liquid equilibrium, trays, reflux (returning part of the condensed overhead to the column), energy cost — and you can work on crude-oil fractionation in a refinery, an ethanol–water column in a bioethanol plant, or a cryogenic column separating nitrogen from oxygen in air. The fluids differ; the engineering is the same. Separations dominate the table for a reason: in most plants the reactor is a small fraction of the equipment, and most of the capital and energy goes into purifying what leaves it.

1.3 Mass and Energy Balances: The Grammar

If unit operations are the vocabulary, balances are the grammar. Every quantitative claim a chemical engineer makes rests on conservation, written for a defined region of space called the control volume:

accumulation = in − out + generation − consumption

Total mass is conserved, so generation and consumption appear only when you track a chemical species that reacts. At steady state — the normal condition of a continuous plant, where nothing changes with time — accumulation is zero and the balance becomes algebra you can solve by hand.

Worked Example: Splitting Ethanol and Water

A distillation column is fed F = 100 kg/h of a mixture containing 10 wt% ethanol. It produces a distillate D at 90 wt% ethanol and a bottoms stream B at 1 wt% ethanol. Nothing reacts. How much distillate does the column make?

Write two balances over the whole column — overall mass and ethanol:

F = D + B                →   100 = D + B
0.10 × 100 = 0.90 D + 0.01 B   →   10 = 0.90 D + 0.01 B

Substitute B = 100 − D into the ethanol balance:

10 = 0.90 D + 0.01 (100 − D) = 0.89 D + 1
0.89 D = 9
D = 10.11 kg/h      B = 89.89 kg/h

Always check: ethanol out = 0.90 × 10.11 + 0.01 × 89.89 = 9.10 + 0.90 = 10.0 kg/h, equal to the ethanol in. Of the 10 kg/h fed, 9.10 kg/h leaves in the distillate — a 91% recovery. Two equations, two unknowns, and nothing about the column's internals was needed. That is the characteristic move of the discipline: bound the problem with conservation first, then worry about mechanism. In code:

import numpy as np

# Unknowns: D (distillate, kg/h), B (bottoms, kg/h)
# Row 1: overall mass balance      D +      B = 100
# Row 2: ethanol balance      0.90 D + 0.01 B = 0.10 * 100
A = np.array([[1.00, 1.00],
              [0.90, 0.01]])
b = np.array([100.0, 10.0])

D, B = np.linalg.solve(A, b)
print(f"D = {D:.2f} kg/h,  B = {B:.2f} kg/h")       # D = 10.11 kg/h,  B = 89.89 kg/h
print(f"ethanol out = {0.90*D + 0.01*B:.2f} kg/h")  # 10.00 kg/h

An energy balance works identically, with enthalpy — the energy content of a stream — in place of mass, and it turns this tidy result into a cost: separating ethanol from water means boiling the mixture — and vaporizing 1 kg of water at 100 °C and atmospheric pressure takes about 2,257 kJ. Distillation is among the largest energy consumers in the chemical industry, so the energy balance, not the mass balance, usually decides whether a process is worth building. (A caution for later: ethanol and water form an azeotrope near 95.6 wt% ethanol at 1 atm, so ordinary distillation cannot exceed that purity.)

1.4 Reading a Process: Flowsheets

Processes are communicated as flowsheets, at three levels of detail:

Nearly every continuous process has the same skeleton:

The arrow returning from separation to the front is a recycle stream, one of the most consequential features in process design.

Why recycle is essential: reactors rarely convert all of the feed in one pass — equilibrium may forbid it, or the conditions needed to force high conversion may wreck selectivity. Discarding unconverted reactant would waste raw material, usually the largest operating cost. Recycling lets a reactor with modest single-pass conversion reach high overall conversion.

Why recycle makes design harder: it couples everything. Change the reactor temperature and the separation duty changes; change the separation and the reactor feed changes, which changes the reactor again. The flowsheet can no longer be solved unit by unit — it must be converged iteratively, which is what process simulators do. Recycle loops also accumulate whatever enters but cannot leave — an inert in the feed, or a by-product the separation misses — which is what the small purge stream is for: discarding a slice of the recycle to hold impurities steady. Recycle also carries disturbances back to the front of the plant, a control problem taken up in Chapter 3.

1.5 Transport Phenomena: The Physics Underneath

Unit operations look like a list of unrelated devices. They are not. In 1960, Bird, Stewart, and Lightfoot published Transport Phenomena, which unified them by showing that momentum, heat, and mass transfer obey laws of the same shape:

flux = coefficient × driving force

A flux is an amount passing through unit area per unit time; the driving force is a gradient — how steeply a property changes with position.

All three transport in the direction that flattens the gradient: momentum from fast fluid to slow, heat from hot to cold, a species from concentrated to dilute. The payoff is practical — a result derived for one transport problem often carries over to the others, and the same equipment features (turbulence, thin films, large interfacial area) improve all three at once.

1.6 Where Chemical Engineers Work

The toolbox is industry-agnostic, which is why the degree travels so well:

• Petroleum and petrochemicals — the classical home of large-scale continuous processing
• Pharmaceuticals — mostly batch: crystallization, drying, strict regulation
• Food and beverage — evaporation, drying, sterilization, fermentation
• Semiconductors — ultrapure gases and water, thin-film deposition, contamination control
• Batteries and energy — slurry mixing, coating, drying, electrolyte handling, recycling
• Environmental and water — pollutant absorption, membrane treatment, carbon capture

A distillation column in a refinery and an evaporator in a dairy are the same unit operation obeying the same balances. The chemistry changes; the engineering transfers.

1.7 Chapter Summary

• Scale changes the physics: heat generation grows with volume (L³) while heat removal grows with area (L²), so surface-to-volume ratio falls as equipment grows — bench results do not transfer automatically
• Unit operations (Arthur D. Little, 1915) decompose any process into reusable steps: separation, heat transfer, fluid handling, and reaction
• Balances (accumulation = in − out + generation − consumption) reduce at steady state to algebra, as in the 100 kg/h column giving D = 10.11 kg/h and B = 89.89 kg/h
• Energy balances, not mass balances, usually decide whether a process is economic
• Flowsheets come as BFD, PFD, and P&ID; recycle makes processes economical while coupling every unit to every other, requiring purges and iterative solution
• Transport phenomena unify the field: momentum, heat, and mass transfer all follow flux = coefficient × driving force

Next chapter: the reactor at the center of the flowsheet — how reaction rates, reactor types, and residence time set conversion and selectivity.

Exercises

1. Conceptual — scale-up: A reaction runs in a 1 L stirred flask and is scaled to a geometrically similar 1,000 L vessel. By what factor does the linear dimension grow, the wall heat-transfer area grow, and the surface-to-volume ratio change? Name two design responses that restore adequate cooling. Hint: take the cube root of the volume ratio first; then think about heat-transfer area that is not the vessel wall. Answer: Volume ×1,000, so the linear dimension grows by the cube root: ×10. Wall area scales as L², so ×100. Surface-to-volume scales as 1/L, so it falls to one tenth — each liter of contents has only 10% of the cooling surface it had in the flask. Design responses: add heat-transfer area that is not the wall (internal cooling coils, or an external heat-exchange loop pumping contents through an exchanger), or cut the heat-release rate by semi-batch dosing — feeding the limiting reactant slowly so generation never exceeds what the cooling can remove.

2. Quantitative — material balance: A dryer receives 500 kg/h of wet solid at 20 wt% water and produces product at 2 wt% water; water leaves only as vapor. Compute the dried-product flow and the water evaporated per hour, then check that total mass in equals total mass out. Hint: the dry solid passes through unchanged — balance it first and only one unknown remains. Answer: Dry solid in = 0.80 × 500 = 400 kg/h, and it all leaves in the product, where it is 98 wt%: product = 400 / 0.98 = 408.16 kg/h. Water evaporated = 500 − 408.16 = 91.84 kg/h. Check: 408.16 + 91.84 = 500 kg/h in = out. ✓

3. Discussion — recycle: A reactor converts only 40% of reactant A per pass, but the separation section recovers essentially all unconverted A and returns it. Explain why overall conversion can still approach 100%, why a small inert in the feed must be purged, and what goes wrong if the purge rate is too high or too low. Hint: distinguish single-pass from overall conversion; an inert has no exit but the purge. Answer: Overall conversion → ~100% because unconverted A is not lost: it is returned to the reactor and gets further chances, so at steady state essentially all fresh A fed eventually becomes product even though each pass converts only 40%. An inert entering with the feed does not react and is not taken off with the product, so the recycle loop is its only home — without a purge it accumulates without limit. The purge rate is a trade-off: too high wastes recycled reactant along with the inert (raw-material cost, the largest operating cost); too low lets inerts build up, diluting the reactor feed, lowering the reaction rate, and enlarging every recycle-loop unit.