askesis

Advanced chemical engineering: 50 concepts for design and practice

Upper-level chemical engineering: reactor design, transport analogies, phase and reaction equilibrium, shortcut distillation, control, safety and economics.

EngineeringAdvanced50 cardsFree

Practice now

Study without an account

0 of 50 seen0 graduated
Levenspiel plot

Scheduled with the default Askesis timings. Progress is saved in this browser.

All 50 cards

  1. 1
    Levenspiel plot
    A plot of FA0−rA\dfrac{F_{A0}}{-r_A} against conversion XX
    A PFR’s volume is the area under the curve; a CSTR’s is the rectangle whose height is the curve’s value at the exit conversion.
  2. 2
    Residence time distribution function E(t)
    E(t) dtE(t)\,dt is the fraction of the fluid leaving a reactor that spent between tt and t+dtt + dt inside it
    ∫0∞E(t) dt=1\displaystyle\int_0^\infty E(t)\,dt = 1. It is measured with a pulse or step tracer test.
  3. 3
    E(t) for an ideal CSTR and an ideal PFR
    CSTR: E(t)=1τe−t/τE(t) = \dfrac{1}{\tau}e^{-t/\tau}. PFR: E(t)=δ(t−τ)E(t) = \delta(t - \tau)
    Every fluid element spends exactly τ\tau in a PFR, while a CSTR has a broad exponential spread.
  4. 4
    Damköhler number
    The ratio of reaction rate to convective transport rate; for a first-order reaction, Da=kτDa = k\tau
    For order nn, Da=k CA0 n−1τDa = k\,C_{A0}^{\,n-1}\tau. In a first-order CSTR, X=Da1+DaX = \dfrac{Da}{1 + Da}.
  5. 5
    Parallel reactions: choosing the reactant concentration
    If the desired reaction has the higher order in A, keep CAC_A high; if it has the lower order, keep CAC_A low
    Instantaneous selectivity SD/U=kDkU CAαD−αUS_{D/U} = \dfrac{k_D}{k_U}\,C_A^{\alpha_D - \alpha_U}. High CAC_A: batch reactor or PFR. Low CAC_A: CSTR or diluted feed.
  6. 6
    Series reactions A→B→CA \rightarrow B \rightarrow C: time for maximum BB
    τopt=ln⁡(k2/k1)k2−k1\tau_{\text{opt}} = \dfrac{\ln(k_2/k_1)}{k_2 - k_1}
    For first-order steps in a batch reactor or PFR. After this time, B is consumed faster than it forms.
  7. 7
    Adiabatic reactor: conversion from the energy balance
    X=∑iΘiCp,i (T−T0)−ΔHRxX = \dfrac{\sum_i \Theta_i C_{p,i}\,(T - T_0)}{-\Delta H_{\text{Rx}}}
    Θi=Fi0/FA0\Theta_i = F_{i0}/F_{A0}, assuming ΔCp≈0\Delta C_p \approx 0. For an exothermic reaction, temperature rises linearly with conversion.
  8. 8
    Multiple steady states in a CSTR
    The S-shaped heat-generation curve can cross the straight heat-removal line up to three times; the middle steady state is unstable
    The upper and lower states are stable, so the reactor can ignite or extinguish as the feed temperature changes.
  9. 9
    Thiele modulus
    ϕ=LkDe\phi = L\sqrt{\dfrac{k}{D_e}} for a first-order reaction in a catalyst pellet
    Its square compares the reaction rate to the diffusion rate. LL is a characteristic length (the half-thickness of a slab), kk a volume-based rate constant, DeD_e the effective diffusivity.
  10. 10
    Internal effectiveness factor
    The actual reaction rate in a catalyst pellet divided by the rate if the whole interior were at the surface concentration and temperature
    For a first-order reaction in a slab, η=tanh⁡ϕϕ\eta = \dfrac{\tanh\phi}{\phi}: near 1 for small ϕ\phi, about 1/ϕ1/\phi for large ϕ\phi.
  11. 11
    Fick’s first law
    JA=−DAB dCAdzJ_A = -D_{AB}\,\dfrac{dC_A}{dz}
    JAJ_A is the molar diffusion flux and DABD_{AB} the binary diffusivity, in m2/s\mathrm{m^2/s}.
  12. 12
    Prandtl number
    Pr=cpμk=ναPr = \dfrac{c_p\mu}{k} = \dfrac{\nu}{\alpha}
    Momentum diffusivity over thermal diffusivity. About 0.70.7 for air and 77 for water at room temperature.
  13. 13
    Schmidt number
    Sc=μρDAB=νDABSc = \dfrac{\mu}{\rho D_{AB}} = \dfrac{\nu}{D_{AB}}
    Momentum diffusivity over mass diffusivity. Of order 1 for gases and around 10310^3 for liquids.
  14. 14
    Nusselt number
    Nu=hLkNu = \dfrac{hL}{k}
    Convective over conductive heat transfer; kk is the fluid’s thermal conductivity, not the wall’s.
  15. 15
    Sherwood number
    Sh=kcLDABSh = \dfrac{k_c L}{D_{AB}}
    The mass transfer counterpart of the Nusselt number; kck_c is the mass transfer coefficient.
  16. 16
    Dittus–Boelter equation
    Nu=0.023 Re0.8 PrnNu = 0.023\,Re^{0.8}\,Pr^{n}
    n=0.4n = 0.4 when the fluid is heated and 0.30.3 when it is cooled. For fully developed turbulent pipe flow with Re>104Re > 10^4 and 0.6<Pr<1600.6 < Pr < 160.
  17. 17
    Chilton–Colburn analogy
    jH=jD=fF2j_H = j_D = \dfrac{f_F}{2}, with jH=NuRe Pr1/3j_H = \dfrac{Nu}{Re\,Pr^{1/3}} and jD=ShRe Sc1/3j_D = \dfrac{Sh}{Re\,Sc^{1/3}}
    So a heat transfer correlation gives a mass transfer one by swapping NuNu for ShSh and PrPr for ScSc. The link to the Fanning friction factor fFf_F holds only without form drag.
  18. 18
    Two-film theory: overall gas-phase coefficient
    1Ky=1ky+mkx\dfrac{1}{K_y} = \dfrac{1}{k_y} + \dfrac{m}{k_x}
    Gas- and liquid-film resistances add in series; mm is the slope of the equilibrium line. For a very soluble gas (small mm), the gas film controls.
  19. 19
    Fugacity coefficient
    ϕ=fP\phi = \dfrac{f}{P}
    Fugacity ff is an effective pressure for a real fluid; ϕ=1\phi = 1 for an ideal gas and ϕ→1\phi \to 1 as P→0P \to 0.
  20. 20
    Criterion for phase equilibrium
    At the same TT and PP, each component’s fugacity is equal in every phase: f^i α=f^i β\hat{f}_i^{\,\alpha} = \hat{f}_i^{\,\beta}
    Equivalent to equal chemical potentials.
  21. 21
    Activity coefficient
    γi=f^ixifi\gamma_i = \dfrac{\hat{f}_i}{x_i f_i}
    It measures the departure from an ideal solution, where γi=1\gamma_i = 1. fif_i is the pure-component fugacity at the same TT and PP.
  22. 22
    Modified Raoult’s law
    yiP=xiγiPisaty_i P = x_i \gamma_i P_i^{\text{sat}}
    For nonideal liquids with an ideal vapor phase, at low to moderate pressure.
  23. 23
    Equilibrium constant from the standard Gibbs energy
    ΔG∘=−RTln⁡K\Delta G^\circ = -RT\ln K
    ΔG∘\Delta G^\circ is the standard Gibbs energy change of reaction at temperature TT.
  24. 24
    Van ’t Hoff equation
    dln⁡KdT=ΔH∘RT2\dfrac{d\ln K}{dT} = \dfrac{\Delta H^\circ}{RT^2}
    So KK falls with temperature for an exothermic reaction (ΔH∘<0\Delta H^\circ < 0) and rises for an endothermic one.
  25. 25
    Gibbs energy minimization
    At fixed temperature and pressure, the equilibrium composition minimizes the total Gibbs energy, subject to element balances
    Solved with Lagrange multipliers; it needs the list of species present, not a set of independent reactions.
  26. 26
    Fenske equation
    Nmin⁡=ln⁡[xD1−xD⋅1−xBxB]ln⁡αN_{\min} = \frac{\ln\left[\dfrac{x_D}{1 - x_D}\cdot\dfrac{1 - x_B}{x_B}\right]}{\ln\alpha}
    The minimum number of equilibrium stages, at total reflux, for a binary with constant relative volatility α\alpha. For multicomponent mixtures, apply it to the key components.
  27. 27
    Minimum reflux ratio
    The lowest reflux ratio that can achieve the separation: an operating line touches the equilibrium curve (a pinch), so infinitely many stages are needed
    Columns typically run at about 1.1 to 1.5 times Rmin⁡R_{\min}.
  28. 28
    Underwood equations
    ∑iαiziαi−θ=1−q,Rmin⁡+1=∑iαixi,Dαi−θ\sum_i \frac{\alpha_i z_i}{\alpha_i - \theta} = 1 - q, \qquad R_{\min} + 1 = \sum_i \frac{\alpha_i x_{i,D}}{\alpha_i - \theta}
    They give the minimum reflux ratio for multicomponent distillation. θ\theta is the root that lies between the relative volatilities of the two keys.
  29. 29
    Gilliland correlation
    An empirical link between the actual number of stages NN and the reflux ratio RR, given Nmin⁡N_{\min} and Rmin⁡R_{\min}
    It plots N−Nmin⁡N+1\dfrac{N - N_{\min}}{N + 1} against R−Rmin⁡R+1\dfrac{R - R_{\min}}{R + 1}. It is the last step of the Fenske–Underwood–Gilliland shortcut method.
  30. 30
    Light key and heavy key
    In multicomponent distillation, the two components whose split is specified: the light key goes mostly to the distillate, the heavy key mostly to the bottoms
    Components lighter than the light key usually end up almost entirely in the distillate, and heavier ones in the bottoms.
  31. 31
    HETP
    Height equivalent to a theoretical plate: the height of packing that gives the same separation as one equilibrium stage
    The packed height is Z=N×HETPZ = N \times \mathrm{HETP}.
  32. 32
    Height of a packed absorber from transfer units
    Z=HOG NOGZ = H_{OG}\,N_{OG}
    HOGH_{OG} is the height of a transfer unit (how effective the packing is) and NOGN_{OG} the number of transfer units (how hard the separation is), both on an overall gas basis.
  33. 33
    Ideal selectivity of a gas separation membrane
    αAB=PAPB\alpha_{AB} = \dfrac{P_A}{P_B}, the ratio of the two gases’ permeabilities
    In the solution–diffusion model, each permeability is the gas’s solubility in the membrane times its diffusivity through it.
  34. 34
    Transfer function
    G(s)=Y(s)U(s)G(s) = \dfrac{Y(s)}{U(s)}: the Laplace transform of the output over that of the input, in deviation variables with zero initial conditions
  35. 35
    First-order transfer function
    G(s)=Kτs+1G(s) = \dfrac{K}{\tau s + 1}
    KK is the steady-state gain and τ\tau the time constant. After a step input, the output reaches 63.2% of its final change at t=τt = \tau.
  36. 36
    First-order plus dead time (FOPDT) model
    G(s)=Ke−θsτs+1G(s) = \dfrac{K e^{-\theta s}}{\tau s + 1}
    Dead time θ\theta is a pure delay before the output starts to respond. It adds phase lag and limits how tightly a loop can be tuned.
  37. 37
    Stability of a closed-loop system
    Stable if every root of the characteristic equation 1+GOL(s)=01 + G_{OL}(s) = 0 has a negative real part
    A root in the right half of the ss-plane gives a response that grows without bound. The Routh array tests this without finding the roots.
  38. 38
    Cascade control
    A primary (master) controller sets the set point of a secondary (slave) controller on a faster inner loop
    The inner loop corrects disturbances, such as swings in steam pressure, before they reach the primary variable.
  39. 39
    Feedforward control
    Measures a disturbance and adjusts the manipulated variable before the disturbance affects the controlled variable
    Ideally Gf=−GdGpG_f = -\dfrac{G_d}{G_p}, ignoring sensor and valve dynamics. It is usually combined with feedback to correct for model errors.
  40. 40
    HAZOP
    Hazard and operability study: a team works through the P&IDs node by node, applying guide words such as no, more, less and reverse to parameters like flow and pressure to find hazardous deviations
    For each deviation the team records causes, consequences, safeguards and actions.
  41. 41
    LOPA
    Layer of protection analysis: multiplies the frequency of an initiating event by the failure probability of each independent protection layer, then compares the result with a tolerable risk
    fmitigated=finitiating×∏jPFDjf_{\text{mitigated}} = f_{\text{initiating}} \times \prod_j \mathrm{PFD}_j, where PFD is the probability of failure on demand.
  42. 42
    Relief valve vs rupture disc
    A spring-loaded relief valve opens at its set pressure and recloses when the pressure falls; a rupture disc bursts once and stays open
    A rupture disc is sometimes fitted upstream of a relief valve to protect it from a corrosive process fluid.
  43. 43
    Maximum accumulation for a single relief valve (ASME Section VIII)
    10% above the maximum allowable working pressure (MAWP)
    16% when several relief devices protect the vessel, and 21% for fire exposure.
  44. 44
    Flammability limits (LFL and UFL)
    The lowest and highest concentrations of a fuel in air that can propagate a flame; outside them the mixture is too lean or too rich to burn
    Methane: about 5 to 15 vol% in air.
  45. 45
    Limiting oxygen concentration (LOC)
    The oxygen concentration below which combustion is impossible, whatever the fuel concentration
    Inerting with nitrogen or carbon dioxide keeps the oxygen below it, with a safety margin.
  46. 46
    Inherently safer design
    Removing or reducing hazards instead of controlling them: minimize (intensify), substitute, moderate (attenuate) and simplify
    Example: storing less of a toxic intermediate, or making it only as it is needed.
  47. 47
    CAPEX vs OPEX
    CAPEX: the one-time capital spent to build or buy a plant. OPEX: the recurring cost of running it, such as raw materials, utilities, labor and maintenance
  48. 48
    Net present value (NPV)
    NPV=∑t=0nCFt(1+i)t\mathrm{NPV} = \displaystyle\sum_{t=0}^{n} \frac{CF_t}{(1 + i)^t}
    CFtCF_t is the cash flow in year tt (negative for investment) and ii the discount rate. A project adds value if NPV>0\mathrm{NPV} > 0.
  49. 49
    Internal rate of return (IRR)
    The discount rate at which a project’s NPV is zero
    A project is attractive if its IRR exceeds the company’s minimum acceptable rate of return.
  50. 50
    Six-tenths rule
    C2=C1(S2S1)0.6C_2 = C_1\left(\dfrac{S_2}{S_1}\right)^{0.6}
    Scales the cost CC of equipment or a plant to a different capacity SS. The exponent varies by equipment type; 0.6 is a typical average.

Frequently asked questions

Who is this deck for?

Senior undergraduates, graduate students and practicing engineers who want the key results of reactor design, transport phenomena, thermodynamics, separations, control, safety and economics at their fingertips.

What is the Fenske–Underwood–Gilliland method?

A shortcut for sizing a distillation column: the Fenske equation gives the minimum number of stages, the Underwood equations the minimum reflux ratio, and the Gilliland correlation the actual number of stages at a chosen reflux ratio.

Are the safety cards enough for process safety work?

No. They cover the core vocabulary and numbers, but real hazard studies and relief sizing follow codes and standards such as ASME and API, and need trained, experienced teams.

Beginner

Mechanical engineering basics

First-year mechanical engineering: statics, stress and strain, material properties, simple machines and gears, thermodynamics terms and manufacturing.

50 cards
Intermediate

Intermediate mechanical engineering

Core undergraduate mechanical engineering: bending and torsion, thermodynamic cycles, fluid flow, heat transfer, machine design and vibrations.

50 cards
Advanced

Advanced mechanical engineering

Upper-level mechanical engineering: fracture, fatigue, FEA, turbulence, turbomachinery, compressible flow, vibration, control, creep, composites and GD&T.

50 cards
Beginner

Electrical engineering basics

The basics of electrical engineering: Ohm’s and Kirchhoff’s laws, series and parallel circuits, capacitors, inductors, AC, components and safety.

50 cards
Intermediate

Intermediate electrical engineering

Core undergraduate circuits and electronics: Thevenin and Norton, transients, phasors, AC and three-phase power, op-amps, transistors, logic and filters.

50 cards
Advanced

Advanced electrical engineering

Upper-level electrical engineering: Laplace and Z transforms, Bode and Nyquist stability, DSP, transmission lines, Maxwell’s equations, power systems and noise.

50 cards
Beginner

Civil engineering basics

Learn the core ideas of civil engineering: loads, tension and compression, beams and trusses, concrete and steel, soils, foundations, surveying and roads.

50 cards
Intermediate

Intermediate civil engineering

Undergraduate civil engineering in 50 cards: beam formulas, buckling, concrete and steel design, soils, open channels, hydrology, traffic and water treatment.

50 cards
Advanced

Advanced civil engineering

Advanced civil engineering: indeterminate analysis, prestressing, seismic design, consolidation, slope stability, pile capacity, pavements and limit states.

50 cards
Beginner

Chemical engineering basics

The first ideas in chemical engineering: mass balances, stoichiometry, conversion and yield, gas laws, pressure and units, and the main process equipment.

50 cards
Intermediate

Intermediate chemical engineering

The core chemical engineering courses: energy balances, phase equilibrium, distillation, fluid flow, heat transfer, kinetics, reactor design and control.

50 cards