DESIGN OF CHEMICAL

REACTORS

Required Reading: Chapter 22 (reactors) of

Turton et al. (2018)
Basic Reactor Models in Hysys
Non-kinetic models: Provide no information about reactor size
• Conversion reactor: The reactions and the percent conversion of a key
component must be given. The conversion can be specified as a function of
temperature.
• Equilibrium reactor: The reactions must be specified and their equilibrium
constants either given or calculated by Hysys.
• Gibbs reactor: The products are given, without specifying the reactions.
Hysys then calculates the equilibrium composition of the product stream by
minimizing its Gibbs free energy, constrained by conservation of atomic
species using the composition of the feed stream.
Reactor Models in Hysys
Kinetic models: Provide reactor size information
• Continuous stirred-tank reactor (CSTR): Assumes perfect mixing.
• Plug flow tubular reactor (PTFR): Assumes no backmixing (dispersion). A
packed bed catalyst will be inside the tube(s), with a heat-transfer medium
flowing co-current outside the tube(s).
Reaction Stoichiometry

where Aj : reactant or product

nj : stoichiometric coefficient (>0 for products, <0 for
reactants)
C : number of components participating in the
reaction

• Example: manufacture of methanol

2H2 + CO = CH3OH
- 2H2 - CO + CH3OH = 0
nj = [-2, -1, 1]
Conversion, Yield, and Selectivity
• Conversion: Amount of reactant reacted / Amount of reactant fed
• Limiting reagent: Reagent whose initial amount limits the amount of
product that would be formed if the reaction went to completion.
• Theoretical yield: Amount of product that would be formed if the reaction
went to completion (i.e., if all the limiting reagent was consumed).
• Percent yield = 100 x (Actual yield / Theoretical yield)
• Selectivity (1) = Amount of reagent reacting to give the desired product /
Total amount of reagent reacting
• Selectivity (2) = Amount of reagent reacting to give the desired product /
Amount of reagent reacting to give undesired products
Example
The stoichiometric reaction of 1-propanol with phosphorus trichloride to produce 1-
chloropropane is as follows:
3 CH3CH2CH2OH + PCl3  3 CH3CH2CH2Cl + H3PO3
1-propanol 1-chloropropane phosphoric acid

We start with 2 moles of 1-propanol and 1 mole of PCl3 . If 1.5 moles of 1-

chloropropane are produced, calculate a) the conversion of PCl 3 , b) the limiting
reagent, c) the theoretical yield of 1-chloropropane, and d) the percent yield.
Example
The stoichiometric reaction of 1-propanol with phosphorus trichloride to produce 1-
chloropropane is as follows:
3 CH3CH2CH2OH + PCl3  3 CH3CH2CH2Cl + H3PO3
1-propanol 1-chloropropane phosphoric acid

We start with 2 moles of 1-propanol and 1 mole of PCl3 . If 1.5 moles of 1-

chloropropane are produced, calculate a) the conversion of PCl 3 , b) the limiting
reagent, c) the theoretical yield of 1-chloropropane, and d) the percent yield.

SOLUTION
a. One mole of PCl3 is consumed per 3 moles of 1-chloropropane produced. Hence,
1.5/3 = 0.5 moles of PCl3 reacted. The conversion of PCl3 is 0.5/1 = 0.5 or 50%
b. 1-propanol is the limiting reagent because its amount is less than would be
required to react with all the PCl3.
c. If the reaction went to completion, all the 1-propanol would react. The theoretical
yield is therefore 2 moles of 1-chloropropane.
d. The actual yield is 1.5 moles of 1-chloropropane. Therefore, the percent yield is
100 x (1.5/2) = 75%.
Example
Consider a reactor where the following two reactions take place:

N2O5  2 NO2 + ½ O2 (Reaction A)

2 NO2  N2O4 (Reaction B)

We start with 2 moles of dinitrogen pentoxide (N2O5). The reactions

produce 2.5 moles of nitrogen dioxide (NO2) and 0.5 mole of dinitrogen
tetroxide (N2O4).

Calculate a) the conversion of N2O5, b) the theoretical yield of NO2, c) the

percent yield of NO2, and d) the selectivity (1) toward NO2.
Example
SOLUTION
a. The total amount of N2O5 reacted to produce 2.5 moles NO2 by reaction A is 2.5/2
= 1.25 moles.
The total amount of N2O5 reacted to produce 0.5 moles of N2O4 by reactions A and B
in series is 0.5 mole.
Hence, the total amount of N2O5 reacted is 1.25 + 0.5 = 1.75 moles.
Therefore, the conversion of N2O5 is 1.75/2 = 0.875 (87.5%).

b. The theoretical yield of NO2 is calculated by assuming that all of the initial N2O5
reacts by Reaction A and Reaction B does not occur because this would produce the
maximum amount of NO2. Hence, the theoretical yield is 4 moles of NO 2.

c. Since only 2.5 moles of NO2 were produced, the percent yield of NO2 is
100x(2.5/4) = 62.5%.

d. The amount of N2O5 reacted to form 2.5 moles NO2 by Reaction A is 2.5/2 = 1.25
mole. From a) we know that the total amount of N2O5 reacted is 1.75 moles. Hence
the selectivity (1) toward NO2 is 1.25/1.75 = 0.714 (71.4%).
 Tutorial

Conversion Reactor with Multiple

Reactions

SKIPPED – STUDY THIS

Chemical Equilibrium
• The reaction equilibrium equation is given by:

where âi = chemical activity of species i :

K = reaction equilibrium constant

• The equilibrium constant is related to the Gibbs energy change of reaction,

DG0, by:
Effect of Temperature on Equilibrium
• The van’t Hoff equation relates K to temperature:

where DH0 = heat (enthalpy change) of reaction

If DH0 < 0 (exothermic reaction), K decreases with T => The reaction

moves to the left when T increases.

If DH0 > 0 (endothermic reaction), K increases with T => The reaction

moves to the right when T increases.

• If DH0 is almost independent of T in the temperature range of interest,

integration gives:
Effect of Temperature on Equilibrium (cont’d)

Endothermic

Exothermic
ln K

1/T
Calculation of the Equilibrium Constant in
Hysys
• Fixed Keq: The equilibrium constant Keq is considered to be fixed, and is
thus independent of temperature.
• Gibbs Free Energy: The equilibrium constant is determined from the Ideal
Gas Gibbs Free Energy Coefficients in the Hysys library.
• Ln(Keq) Equation:

• Keq vs T table: The user provides values for temperature and equilibrium
constants in the K Table. Hysys estimates the equilibrium constant from
the pairs of data provided by the user and will interpolate when necessary.
Hysys will calculate the constants in the Ln(Keq) equation with the table
data.
Equilibrium of Gas-Phase Reactions
(Review)
• Consider a single reaction with all reactants and products in the
gas phase.
• The mole fractions of all species (, ) at reaction equilibrium are
given by:

: reaction equilibrium constant (depends on T, not on P)

: pressure at which the reaction takes place
: standard-state pressure (1 bar, in the same units as )
: sum of stoichiometric coefficients
: fugacity coefficient of species i (depends on T, P and composition)
Equilibrium of Gas-Phase Reactions
(Review)
• If the equilibrium mixture is an ideal solution:

: fugacity coefficient of pure species i (depends on T, P and not on

composition)

• If the equilibrium mixture behaves as an ideal gas (low pressure or

temperature sufficiently high):
 Equilibrium of Liquid-Phase
Reactions (Review)
• Consider a single reaction with all reactants and products in the
liquid phase
• The mole fractions of all species (xi, ) at reaction equilibrium are
given by:

: molar volume of pure species i

: activity coefficient of species i (depends on T, P and composition)
 Equilibrium of Liquid-Phase
Reactions (Review)
• At moderate pressures, the exponential term is close to unity and
can be omitted:

• If the equilibrium mixture is an ideal solution, then . Therefore, the

law of mass action applies:
 Tutorial

Equilibrium Reactor
Homogeneous Reaction Kinetics
Consider the stoichiometric reaction:

where Ai = reactant or product

ni = stoichiometric coefficient

Power-law kinetic equation:

where -rj = rate of disappearance of component j (mol/vol-time)

Cj = concentration of component j (mol/vol)
t = time
k = reaction rate coefficient
ai = order of the reaction with respect to component i
C = number of components

VERY IMPORTANT: In general, ai  ni

Example: Toluene Hydrodealkylation
Stoichiometric reaction:
H2 + C6H5CH3  CH4 + C6H6
toluene methane benzene

Elementary steps:
H2 <-> 2H• (fast => reaches equilibrium very quickly)

H• + C6H5CH3  C6H6 + CH3• (slow); CH3• + H2  CH4 + H• (fast)

toluene benzene
Arrhenius Equation
• The reaction rate coefficient is a function of temperature:

where A = pre-exponential factor

E = activation energy
T = temperature in Kelvin

Example:
For toluene hydrodealkylation

with concentrations in mol/L, t in s, T in K, and R = 1.987 cal/mol-K

• Chemical reaction rates approximately double every 10ºC.

Rate Models in Hysys –
Kinetic Reaction

forward backward
 Tutorial

Setup of Homogeneous Reaction

Kinetics
Heterogeneous Reaction Kinetics
For solid catalyzed reactions:

where -rj = rate of disappearance of component j (moles/mass of catalyst - time)

Nj = moles of component j reacted
W = mass of catalyst

Important note: In Hysys, heterogeneous reaction rates are based on moles/void

volume of reactor-time.

where V = void volume of reactor

rb = bulk catalyst density (catalyst mass / reactor volume)

e = void fraction (reactor void volume / reactor volume)
Heterogeneous Reaction Kinetics (cont’d)

where rb = bulk catalyst density (catalyst mass / reactor volume)

rcat = catalyst specific density (catalyst mass / total volume of all
catalyst particles)
= non-void fraction (total volume of all catalyst particles
/ reactor volume)
Heterogeneous Reaction Kinetics (cont’d)
The series of resistances that can govern the rate of catalytic chemical
reactions are:
• Mass film diffusion of reactant from bulk fluid to external surface of
catalyst
• Mass diffusion of reactant from pore mouth to internal surface of
catalyst
• Adsorption of reactant on catalyst surface
• Chemical reaction on catalyst surface
• Desorption of product from catalyst surface
• Mass diffusion of product from internal surface of catalyst to pore mouth
• Mass diffusion of product from pore mouth to bulk fluid
Heterogeneous Reaction Kinetics (cont’d)
For solid-catalyzed reactions, the power-law is often replaced by the
Langmuir-Hinshelwood equation:

where
k : reaction rate constant
Ki : adsorption constant
k and Ki are catalyst specific and must be obtained by fitting reaction data.
The denominator includes the concentrations of reactants that are strongly
adsorbed on the catalyst.
Rate Models in Hysys –
Heterogeneous Kinetic Reaction
𝑘 ∏ 𝐶𝑖 − 𝑘 ′ ∏ 𝐶𝑖
′
𝛼𝑖 𝛼𝑖

𝑖 𝑖
− 𝑟 𝑗=
[ ]
𝛼 𝑛
1+ ∑ 𝐾 ∏ 𝐶𝑖
𝑗, 𝑖

𝑗
𝑗 𝑖
 Tutorial

Heterogeneous Reaction Kinetics Set-

up
Relationship Between Forward and Reverse Reaction
Kinetics Limited by Equilibrium
Consider a reversible reaction mechanism: A + B   C
The forward reaction rate is given by: rf = kfCACB
The reverse reaction rate is given by: rr = krCC

At equilibrium: rf = rr and CC/(CACB) = K where K is the thermodynamic

equilibrium constant. Therefore:
kfCACB = krCC => kr = kfCACB/CC = kf/K

Hence, the overall reaction rate is given by:

r = rf – rr = kf(CACB – CC/K)

In general:

This equation is called the “simple rate model” in Hysys.

Rate Models in Hysys –
Simple Rate Reaction
Catalyst Selection
• Catalysts increase the reaction rate.
• Catalysts are not affected in quantity or composition by the reaction.
• In a system of multiple reactions, the catalyst may affect the rate of
different reactions differently => Catalysts can have a major influence on
selectivity.
• For reversible reactions, catalysts do not alter the equilibrium composition.
ONLY INCREASES REACTION RATE, NOT EQUILIBRIUM.
CATALYST ONLY AFFECTS KINETICS, NOT THERMODYNAMICS.
• The selection of catalysts is still mostly done by trial and error.
• Catalytic processes can be homogeneous or heterogeneous.
Homogeneous Catalysts
• The reaction occurs entirely in the vapour or liquid phase.
• Example: production of ketene from acetic acid in the gas phase at 700°C
using triethyl phosphate as a homogeneous catalyst:
CH3COOH  CH2=C=O + H2O
Acetic acid ketene water
• The separation and recycling of homogeneous catalysts is often very
difficult.
• Loss of homogeneous catalyst increases catalyst cost and creates an
environmental problem.
• When possible, heterogeneous catalysts are preferred to homogeneous
catalysts.
Heterogeneous Catalysts
• The catalyst is in a different phase from the reacting species.
• When the catalyst is a solid and the reacting phase is a liquid or a gas, the
following steps occur:
– Reactants diffuse to the catalyst surface
– Reactants adsorb on the catalyst surface
– The reaction takes place
– The products desorb from the catalyst surface
– The products diffuse back to the bulk gas or liquid phase

• The reaction tends to be rate limiting at low temperatures.

• Diffusion tends to be rate limiting at high temperatures.
• Solid catalyst types:
– Bulk catalyst materials: the gross composition does not change significantly through the
material (e.g., platinum wire mesh).
– Supported catalysts: the active catalytic material is deposited over the surface of a porous
solid (e.g, alumina, silica or aluminosilicates).
Important Parameters for Solid Catalysts
• Specific surface area of solid (usually > 100 m2/g)
• Catalyst particle size
• Pore structure of the particles
• Stability to impurities and degradation over time
• Ability to regenerate the catalyst activity
• Physical strength to sustain handling and abrasion
• Resistance to abnormally high temperatures that can reduce physical
strength and catalytic activity
Catalytic Degradation
• Physical loss
– Homogeneous catalysts (discussed previously)
– Heterogeneous catalyst attrition in fluidized-bed reactors causes catalyst
particles to be broken down in size. Fine catalyst particles cannot be effectively
separated from the reactor effluent and recycled to the bed.
• Deposits on catalyst surface
– e.g., coke [carbon] deposits on catalysts in hydrocarbon reactions.
– Catalyst can usually be regenerated by air oxidation of the carbon deposit at
high temperatures.
• Sintering
– Molecular rearrangement occurs below the catalyst melting point (usually starts
at 50% of the catalyst melting point)
– Causes a reduction in the effective surface area of the catalyst.
– Accelerated by poor heat transfer or poor mixing leading to hot spots.
Catalytic Degradation (cont’d)
• Poisoning
– Some impurities in the raw materials or products of reaction can form strong
chemical bonds with the catalyst.
– These reactions degrade the catalyst and reduce its activity.

• Chemical change
– Slow chemical change of the catalyst causes reduction in its activity.
Solid-Fluid Reactor Types (1)
Solid-Fluid Reactor Types (2)
• Fixed-bed reactors
– Used for heterogenous catalysis
– Approach plug flow
– Ease of catalyst access for maintenance and regeneration
– High pressure drops
– Not suitable if high rates of heat transfer are required because of difficulty controlling
the temperature (heat loads vary through the bed) => hot spots

• Fluidized-bed reactors
– Gas flows at a rate sufficient to fluidize suspended solid particles.
– Performance is intermediate between well-mixed and plug-flow models due to ease of
backflow in the gas.
– Suitable for high rates of heat transfer and temperature uniformity
– Operating difficulties due to flow and suspension issues

• Moving-bed reactors
– Can process particles that are too large to be handled in fluidized-bed reactors.
– Favorable when catalyst requires frequent regeneration because of ease to
continuously remove and add catalyst particles (as in a refinery catalytic cracking unit
where coke deposits on catalyst particles).
Solid-Fluid Reactor Types (3)
Solid-Fluid Reactor Types (4)
• Slurry (stirred-tank) reactors
– Simple
– Good mass transport via mixing if fluid is not too viscous
– Good heat transfer via external jackets, internal coils, or external heat exchanger
in flow loop.
– Good access to catalyst
– Equipment degradation due to impingement of catalyst particles on impellers,
nozzles, and pipes
– Not good for high pressures
– Unfavorable when byproduct forms via series reactions (wide distribution of
residence times)
– Not desirable when reactants or products are very hazardous due to large
material inventory

• Multitubular reactors
– Catalyst filled tubes with heat conducting fluid flowing outside the tubes.
– Good thermal control and uniform residence time distribution
– Good for high pressures
– Unfavorable for multiphase reactions because of difficulty achieving good
mixing between phases
– Catalyst inaccessibility
Solid-Fluid Reactor Types (5)
• Shallow- or thin-bed reactors
– Suitable for fast reactions or where large heats of reaction are involved.
– Good access to catalyst for reactivation or maintenance

• Gas or liquid dispersion reactors

– Simple and inexpensive
– Require careful planning due to their sensitivity to flow behaviour.
Heuristics For Selection of Reactors
• CSTRs are usually used for slow liquid-phase or slurry reactions. A battery
of four or five in series is most economical.
• The optimum proportions of stirred tank reactors are with liquid level equal
to the tank diameter, but at high pressures slimmer proportions are
economical.
• Power input to a homogeneous reaction stirred tank is 0.1-0.3 kW/m 3, but 3
times this amount when heat is to be transferred.
• Ideal CSTR behaviour is approached when the mean residence time is 5 to
10 times the time needed to achieve homogeneity, which is accomplished
with 500-2000 revolutions of a properly designed stirrer.
• Batch reactors are best suited for small-scale production, very slow reactions
(long reaction times), fouling reactions, or reactions requiring intensive
monitoring or control.
Heuristics For Selection of Reactors
(cont’d)
• Tubular flow reactors are suited to high production rates at residence times of
seconds to minutes and when substantial heat transfer is needed. Embedded
tubes or shell-and-tube construction is then used.
• The rate of reaction in a heterogeneous system is often controlled by the rate
of heat transfer.
• The typical sizes of catalyst particles are approximately 2-5 mm for fixed bed
reactors, 1mm for slurry reactors, and 0.1 mm for fluidized bed reactors.
• The value of a catalyst may be to improve selectivity more than to improve
the overall reaction rate.
Plug-Flow Tubular Reactor (PFR) Model
• The composition changes down the length of the reactor.
• No temperature or composition gradients in the radial direction
• Axial dispersion of mass and heat are neglected.
• The mass balance equation is:

where V = reactor volume

FA0 = molar flow rate of limiting reactant A entering
the reactor
- rA = rate of disappearance of component A (rA < 0)
XA = fractional conversion of A
 Tutorial

Design of a PFR Reactor

Question a.
Pressure Drop in PFRs
Ergun equation:
Pressure Drop in PFRs (Cont’d)
• Try to keep the pressure drop below 10% of inlet pressure.

• If the pressure drop is too large:

– Decrease the reactor length

This will require increasing the number of tubes or the tube diameter to keep the
same reactor volume => increased area for flow => lower fluid velocity, which
reduces pressure gradient.

– Increase catalyst particle size

• Be careful to keep the fluid in one phase throughout the reactor, or

Hysys will be unable to calculate the pressure drop (viscosity
determination error). As the pressure drops in the reactor, 2 phases may
appear if the temperatures and/or pressures are not selected
appropriately.
Modeling Heat Transfer in Plug-Flow
Tubular Reactors

Toutj Tbulkj
Modeling Heat Transfer in PFR (Cont’d)
• Heat-transfer fluid (cooling or heating utility) flows co-current outside
the tubes.
• The largest temperature driving force, and therefore heat transfer rate,
occurs near the reactor inlet where reaction rates are the largest.
• Only set the inlet T and P of the utility fluid. Hysys will calculate the
outlet T and P.
• Hysys calculates the temperature profile of the process stream and
utility fluid => improved estimation of reaction rates throughout the
reactor => more precise reactor sizing
• Knowing the temperature profiles allows:
– Avoiding hot spots
– precisely estimate temperature driving forces => precise estimate of utility fluid flow
rate
Heat Transfer Page in Hysys (Formula
Option)

Shell-side HT coefficient
Utility fluid flow, heat
capacity and inlet temp
information

Tube-side HT coefficient
Heat Transfer Parameters
Shell-Side
Heat Transfer Parameters
Tube-Side
 Estimating Heat Transfer Coefficients
• Shell-side heat transfer coefficients depend on the utility
fluid:
– For a boiling utility fluid, 1100 W/m2-K (200 Btu/h.ft2.oF) is
reasonable.
– For molten salts, literature values vary between 700 W/m2-K and
16,000 W/m2-K.

• For tubes packed with solid catalyst particles, reasonable

heat transfer coefficient values range from ~40 W/m2-K to
200 W/m2-K (1).
• For tubes containing only a fluid, use the default values for
the Standard Tube Side Heat Transfer (A=1.6, B=0.51,
C=1/3).

(1)
Peters, P.E. et al., Heat Transfer in Packed-Tube Reactors, Ind. Eng. Chem. Res. 1988, 27, 226-233.
 Estimating Tube-Side Heat Transfer
Coefficients
• For catalyst particles having shapes of spheres, cylinders and rings(1):

Where:
inner wall heat transfer coefficient (W/m2-K)
: particle diameter (m)
: fluid thermal conductivity (W/m-K)
: fluid density (kg/m3)
: fluid velocity (m/s)
: fluid viscosity (Pa.s)
: fluid heat capacity (J/kg-K)

(1)
Peters, P.E. et al., Heat Transfer in Packed-Tube Reactors, Ind. Eng. Chem. Res. 1988, 27, 226-233.
 Estimating Tube-Side Heat Transfer
Coefficients More Precisely
• For catalyst particles having shapes of spheres(1):

• For catalyst particles having shapes of cylinders and rings(1):

where:
: tube inner diameter (m)

(1)
Peters, P.E. et al., Heat Transfer in Packed-Tube Reactors, Ind. Eng. Chem. Res. 1988, 27, 226-233.
Simulating Cooling or Heating by Vaporization or Condensation
of Utility Fluid at Constant Temperature
• Only a non-boiling utility fluid is permitted in Hysys.
• Simulate boiling at constant temperature by using a fictional utility liquid
having an arbitrarily high heat capacity and flow rate so that its
temperature changes very little when flowing through the reactor.
• If the actual utility fluid will be vaporizing water for cooling or
condensing steam for heating, select an inlet temperature on the
Design/Heat Transfer page corresponding to one of the standard steam
pressure values, ~150oC for LPS, ~190 oC for MPS, or ~253 oC for HPS.
• Divide the reactor duty by the latent heat of vaporization of the actual
utility fluid (e.g., saturated boiler feed water) to get its actual flow rate.
 Tutorial

Design of a PFR Reactor

Questions b. and c.
Continuous-Stirred-Tank Reactor (CSTR)
Model
• The reactor contents are assumed to be perfectly mixed.
• Temperature and composition are homogeneous in the reactor.
• The composition and temperature of the effluent are the same as inside the
reactor.
• The mass balance equation is:

where V = reactor volume

FA0 = molar flow rate of limiting reactant A entering
the reactor
FA = molar flow rate of limiting reactant A exiting
the reactor
- rA = rate of disappearance of component A (rA < 0)
XA = fractional conversion of A
 Tutorial

CSTR Reactor