CHAPTER 5

DESIGN OF CHEMICAL DEVICES,

FUNCTIONAL PRODUCTS AND
FORMULATED PRODUCTS

Nor Alafiza Yunus, Universiti Teknologi Malaysia

1
 PART I OVERVIEW
• Introduction
• Product specification
• Mathematical modelling
• Elements of the Product Design Framework

PART II
• Design of Chemical Devices
• Design of Functional Product

PART III
• Design of Formulated Products
• Design a Sunscreen Cream
• Design of insect repellent spray
2 • Design a Cream Manufacturing Process
 OUTCOMES OF THIS TOPIC
After studying this chapter, student should:
1. Be cognizant of the characteristics of devices, functional products, and
formulated products
2. Be familiar with the procedures, methods, and tools for the conceptual
design of these products.
3. Be able to design chemical devices and functional products with specified
product performance by modeling the dominant physicochemical
phenomena and by accounting for the material properties and the
product structure, form, shape, or configuration.
4. Be able to design formulated products with specified product attributes
by experimental iteration guided by heuristics, correlations, and
computer-aided design tools.
3
 CLASSES OF CHEMICAL PRODUCTS

Molecules

Devices/Functional
Products

Formulated Products

4
 CHEMICAL DEVICES
• Chemical devices are those chemical products that
perform a particular purpose, especially those
with mechanical and electrical parts. Often, a
feed stream to a chemical device is transformed
into an outlet stream with characteristics specified
in the product attributes by performing reactions,
fluid flow, heating/cooling, and/or separations.
• For chemical engineers, we can view a chemical
device as a single miniaturized chemical
processing equipment item, a group of
interconnected equipment items, or even a
5 chemical plant.
 EXAMPLES OF CHEMICAL DEVICES

A reverse osmosis module that produces A water filter that removes organics
filtered water free of contaminants and metal ions from water.

6
 EXAMPLES OF CHEMICAL DEVICES

A mosquito repellent dispenser that lifts the

solvent containing an active ingredient to
A wine aerator that sucks in air the electric heater using a wick, followed by
into the passing wine to soften the vaporizing the active ingredient to repel
tannins in a red wine mosquito
7
 FUNCTIONAL PRODUCTS
• Functional products are those chemical products
made up of materials that perform a desired
function. Generally, these products do not have feed
and outlet streams, and do not involve mechanical
and electrical parts.
• Functional products serve as barriers (food
packaging), delivery vehicles (controlled release
granule), absorbers (silica gel), etc.
• The performance of functional products depends on
the material properties and the structure, form,
shape or configuration of the product.
8
 EXAMPLES OF FUNCTIONAL PRODUCTS

Metallized polyethylene is used in

Silica gel is used in an enclosed
food packaging to prevent oxygen
closet to reduce humidity
penetration
9
 EXAMPLES OF FUNCTIONAL PRODUCTS

Active ingredient
Polymer matrix

A controlled release granule that gradually releases

10 an herbicide over a period of time
 FORMULATED PRODUCTS
• Formulated products are obtained by mixing selected components
together to get the desired product attributes.
• Many products, such as antifreeze, do not possess a microstructure, but
some, such as a skincare lotion, do. The lotion can be an oil-in-water or
water-in-oil emulsion with different emulsion droplet sizes.
• The manufacture and use of formulated products generally do not
involve reaction or separation.

11
 EXAMPLES OF FORMULATED PRODUCTS

Skin Care Creams Cosmetics Detergents

Paints Inkjet Printer Inks Adhesives

12
 PRODUCT SPECIFICATIONS
Chemical engineers are responsible for designing a product that meets the
product specifications.

DEVICE FUNCTIONAL FORMULATED

The product specifications should The product specifications required

include the type and amount of are the same as those for a device The product specifications are the
the key ingredients and the except for the absence of composition of the mixture and the
structure with which these mechanical and electrical parts, microstructure.
ingredients are configured. and feed/outlet streams.

Modelling is a key technique in product conceptualization and specification

13
 PRODUCT DESIGN VS. PRODUCT ANALYSIS
Product Analysis
Input • In product analysis, the type and
(Known) Output?
Fixed Product amount of the ingredients and the way
in which the ingredients are configured
are fixed. The results of modelling are
Conceptual Product Design
the output information for given input
Input Material Type, Output information.
(Known) Amount and (Known)
Configuration
• In conceptual product design, the input
? and output information is fixed. The
Kinetics, Transport Processes
outcome of modelling is the material
type, amount and configuration.
Thermodynamics, Material Sciences

14
 WHAT IS A “MATHEMATICAL” MODEL IN SCIENCE AND
ENGINEERING?
The real world The conceptual world
Mathematical model : a representation in
Observation mathematical terms of the behavior of real
devices and objects

Phenomena Models (analyses)

Engineers need to model products and processes to

Predictions
design and fabricate these products and processes
that have not yet come to life

15 http://www.sfu.ca/~vdabbagh/Chap1-modeling.pdf
 WHAT DOES MATHEMATICAL MODELLING MODEL?
Conservation Principles

Q represents some property such as

mass, momentum, and energy.
qin = inlet flow rate of Q
qout = outler flow rate of Q
g = the Q generation rate
c = the Q consumption rate
t = time.
16
 WHAT DOES MATHEMATICAL MODELLING MODEL?
Constitutive equations
Fick’s law – Law of viscosity
Fourier’s law –
Describe the molar The law of heat
flux due to conduction. The
concentration rate of heat
gradient transfer

17
 WHAT DOES MATHEMATICAL MODELLING MODEL?
Experimental correlations

Langmuir Hinshelwood Power law fluid Newtonian

n=1; Shear thinning, n <1;
Shear thickening, n>1.
τ = η (γ )γ
η = mγ n −1

Langmuir isotherm

18
 MODELS IN PRODUCT DESIGN
• Models have different degrees of complexity and accuracy .
• The simplest is the black-box model.
• Simplified models may not be very accurate. However, by accounting for
the dominant mechanisms, a simplified model can be expected to
capture the correct dependence of product performance on the
influencing factors, thus reducing tremendously the required amount of
experimental work.
• Rigorous models can account for most, if not all, of the product features
and physicochemical phenomena involved when the underlying physics
and chemistry are perfectly understood. Models obtained this way are
predictive, and experimental data are only needed for verification
19 purposes.
 COMPARISON OF VARIOUS APPROACHES IN PRODUCT DESIGN
Approach Black-box model Simplified model Rigorous model

Data fitting Simplified analysis of Detailed analysis of the

Technique
(Empirical correlations) the phenomena phenomena

Understanding of Partially or completely

Poorly understood Completely understood
physical phenomena understood

Limited to the range of Limited to the assumed

Range of validity Relatively wide
collected data conditions

Case dependent; Tend

Predictability of
Not attempted Case dependent to have more unknown
model parameters
parameters

Minimal, directed by None or for verification

Experimental effort Extensive
theory only
20
 ELEMENTS OF THE PRODUCT DESIGN FRAMEWORK FOR TASK
EXECUTION
Methods
Tools:
Rule-based (e.g., Quality
Excel
function deployment)
MATLAB
Model-based (e.g.,
ICAS software
Mechanistic model)
Hybrid method

Database
Chemical database Experiments
Equipment database
Thermodynamic database

21
 SUMMARY
• Identify the class of the product
• Understand the phenomena involve. How the product
functioning?
• Identify models to represent the phenomena
• Apply them to design the desired product

22
 PART I OVERVIEW
• Introduction
• Product specification
• Mathematical modelling
• Elements of the Product Design Framework

PART II
• Design of Chemical Devices
• Design of Functional Product

PART III
• Design of Formulated Products
• Design a Sunscreen Cream
• Design of insect repellent spray
23 • Design a Cream Manufacturing Process
DESIGN OF
CHEMICAL
DEVICES

24
 DESIGN PROCEDURE FOR CHEMICAL DEVICE AND FUNCTIONAL
PRODUCT
1) Specify the product performance.
2) Identify the key ingredients as well as the configurations that can accomplish the
targeted changes to the feed stream or deliver the desired function. Keep in mind
that more than one set of key ingredients and/or configuration may provide the
product performance, leading to different product alternatives.
3) Identify the physicochemical phenomena ranging from reactions to separations
involved in the product.
4) Use models, experiments and available data/knowledge to identify the product
specifications for meeting the desired product performance and compare product
alternatives.

25
 EXAMPLE: DESIGN OF AN AIR PURIFIER
Volatile organic compounds (VOCs) are emitted as gases VOCs
from certain solids or liquids especially from household Benzene
products. Paints, varnishes and wax all contain organic toluene
solvents, as do many cleaning, cooking, disinfecting, xylene
cosmetic, and hobby products. All of these products can ethylene glycol
release organic compounds while you are using them, formaldehyde
and, to some degree, when they are stored. methylene chloride
tetrachloroethylene
Health effects may include: 1,3-butadiene
• Eye, nose and throat irritation
• Headaches, loss of coordination and nausea
• Damage to liver, kidney and central nervous system
• Some organics can cause cancer in animals, some are
26 suspected or known to cause cancer in humans.
 EXAMPLE: DESIGN OF AN AIR PURIFIER
Design scenario:
A home appliance company plans to produce
an air purifier to remove indoor volatile
organic compounds (VOCs). As a measure of
product performance, it is expected to
reduce the concentration of VOCs from 150
mg/m3 to 30 mg/m3 in a room 200 m3 in size
within 2 hours.

Co = 150 mg/m3
Cf = 30 mg/m3
Vr = 200 m3
t=2h

27 HEPA = High Efficiency Particulate Absorption

 STEP 1 – SPECIFY PRODUCT PERFORMANCE
VOCs are made up of different compounds such
as acetone, benzene, toluene and methylene
chloride.

As a starting point, toluene is used as the

reference compound for quantifying the product
performance as stated above.

Co = 150 mg/m3  Cf = 30 mg/m3

t=2h

Toluene
28 HEPA = High Efficiency Particulate Absorption
 STEP 2 – IDENTIFY THE KEY INGREDIENTS
• How to remove VOCs?
o Adsorption – capturing
o Absorption – capturing
o Photocatalytic Oxidation (PCO) –
decomposing VOCs

29
 STEP 2 – IDENTIFY THE KEY INGREDIENTS
• What is the mechanism involves in PCO?
o The decomposition done by catalysts with present of UV light - TiO2 is the most widely
used PCO catalysts.
• How to place the catalyst within the air purifier system?
o It can be deposited on any catalyst support such as carbon, alumina and silica to obtain
a high surface area.
• How to design the air purifier?
• The support and the structure, form, shape or configuration of the product must
o provide intimate contact between the catalyst and the air flow with the VOCs,
o allow sufficient and relatively uniform UV light on the catalyst, and
o be compact, taking up as little space as possible.
• It is decided that a commercially available Pt-doped TiO2 catalyst be impregnated on a
HEPA filter, which is a highly porous material made up of randomly oriented synthetic
30
fibers.
 STEP 2 – IDENTIFY THE CONFIGURATION
• Three additional possible product configurations are shown below.

31
 STEP 3 – IDENTIFY THE PHYSICOCHEMICAL PHENOMENA
• The decomposition of VOCs on the surface of TiO2 proceeds
via several steps.
• Electron–hole pairs, generated by irradiating the TiO2 with
light, are separated by forming a hydroxyl radical (OH) with
adsorbed H2O. Then, redox reactions take place between
OH and adsorbed VOC. The last step is desorption of
reaction products such as CO2, water and hydrogen halides
(HX), and reconstruction of the surface.

Langmuir-Hinshelwood kinetics can be used to model the PCO reaction.

32
 STEP 4 – USE MODELS AND DATA TO DETERMINE PRODUCT
SPECIFICATIONS
• Product Specifications: The total amount of TiO2 coating for achieving the design
target.
• Assuming the air in the room is perfectly mixed, the mass balance of toluene can be
written as: 𝑑𝑑[𝐶𝐶]
𝑉𝑉𝑟𝑟 = 𝑟𝑟𝑇𝑇 𝑉𝑉𝑐𝑐𝑐𝑐𝑐𝑐 (5.1)
𝑑𝑑𝑑𝑑
where Vr and Vcat are the volume of the room and catalyst, respectively, [C] is the
toluene concentration in air, and rT is the rate of decomposition of toluene.
• The decomposition rate of toluene is given by
−𝑘𝑘𝑘𝑘𝑇𝑇 [𝐶𝐶]
𝑟𝑟𝑇𝑇 = (5.4)
1 + 𝐾𝐾𝑇𝑇 [𝐶𝐶]

where k is the surface reaction rate constant, and KT is the adsorption equilibrium
33 constant for toluene.
 STEP 4 – USE MODELS AND DATA TO DETERMINE PRODUCT
SPECIFICATIONS
• Substituting Eq 5.4 into Eq 5.1 and integrating t from t=0 to tf
𝐶𝐶𝑓𝑓
ln + 𝐾𝐾𝑇𝑇 𝐶𝐶𝑓𝑓 − 𝐶𝐶0
𝐶𝐶0
𝑡𝑡𝑓𝑓 = − (5.5)
𝑉𝑉𝑐𝑐𝑐𝑐𝑐𝑐
𝑘𝑘𝐾𝐾𝑇𝑇
𝑉𝑉𝑟𝑟

where tf is the total reaction time, [C0] and [Cf] are the initial and final concentration of
toluene, respectively.

Co = 150 mg/m3
Cf = 30 mg/m3
Vr = 200 m3
t=2h
34
 STEP 4 – USE MODELS AND DATA TO IDENTIFY PRODUCT
SPECIFICATIONS
• The surface reaction rate constant for toluene k can be expressed by the following
empirical relation:
𝑘𝑘 = 𝐾𝐾𝑚𝑚 (𝐾𝐾3 + 𝐾𝐾4 𝐶𝐶0 + 𝐾𝐾6 𝐶𝐶0 2)(𝐾𝐾5 +𝐾𝐾2 𝑅𝑅𝑅𝑅) (5.6)
where Km, K2, K3, K4, K5 and K6 are fitting parameters, and RH is the relative humidity
of the air.
• Parameter fitting for toluene degradation experiments conducted at a RH between 14-
82%, a light intensity of 2.2 mW/cm2, and an initial concentration of 150 mg/m3, we
get
Km KT K2 K3 K4 (m3/mg) K5 K6 (m3/mg)2
(mg/m3.min) (m3/mg)
5.78 0.0189 -14.599 183873.42 -3033.00 1614.37 13.1871

35 • Assume RH = 50%, k = 238,103,199.5 mg/(m .min)

3
 STEP 4 – USE MODELS AND DATA TO IDENTIFY PRODUCT
SPECIFICATIONS Co = 150 mg/m 3

Cf = 30 mg/m3
• Since 200 m3 air needs to be cleaned within 2 hours, tf ≤ 2h Vr = 200 m3
and Vr = 200 m3. Substituting k and KT into Eq 5.5, we get t=2h
k = 238,103,199.5 mg/(m3.min)
𝐶𝐶𝑓𝑓 KT =0.0189 m3/kg
ln + 𝐾𝐾𝑇𝑇 𝐶𝐶𝑓𝑓 − 𝐶𝐶0
𝐶𝐶0
𝑡𝑡𝑓𝑓 = − (5.5)
𝑉𝑉𝑐𝑐𝑐𝑐𝑐𝑐
𝑘𝑘𝐾𝐾𝑇𝑇
𝑉𝑉𝑟𝑟

30 m3 mg 3
ln + 0.0189 × 30 − 150 3 200 m
150 mg m ×
𝑡𝑡𝑓𝑓 = − mg m3 VCat ≤ 2 h (5.7)
238103199.5 3 × 0.0189
m � min mg

or Vcat ≥ 1.436 x 10-6 m3

36
 STEP 4 – USE MODELS AND DATA TO IDENTIFY PRODUCT
SPECIFICATIONS
• If a HEPA filter area of 0.1 m2 is assumed, the thickness of the 0.1 m2
catalyst coating, s, is given by HEPA filter area

𝑉𝑉𝑐𝑐𝑐𝑐𝑐𝑐 1.436 × 10−6 m3 D=10 cm

𝑠𝑠 ≥ 2
= 2
= 14.4 μm (5.8)
0.1 m 0.1 m

• If assumed product configuration selected is Fig. 5.8b,

L=32 cm
the inner diameter of cylindrical HEPA filter is 10 cm and length
would be 32 cm.

37
 STEP 4 – USE MODELS AND DATA TO IDENTIFY PRODUCT
SPECIFICATIONS
• The calculation assumes that the coating is so thin that the light HEPA Filter TiO cat.
intensity across the coating is uniform.
• Assumption validation: According to Beer’s Law, the light intensity,
I, at a certain depth, z, of the catalyst coating is UV

𝐼𝐼𝑠𝑠 𝐼𝐼𝑜𝑜
𝐼𝐼 = 𝐼𝐼0 𝑒𝑒 −𝜀𝜀𝜀𝜀𝜌𝜌𝑏𝑏
s

where I0 is the light intensity at the surface (2.2 mW/cm2) , ε is

the UV light absorption coefficient of the catalyst (ε = 0.0386
cm2/g), and ρb is the bulk density of the catalyst (3.56 mg/cm3).
38
 STEP 4 – USE MODELS AND DATA TO IDENTIFY PRODUCT
SPECIFICATIONS
HEPA Filter TiO cat.
• Light intensity across the catalyst bed with a thickness of 14.4 μm

cm2 mg UV
−0.0386 g ×0.0015 cm× 3.56 3
𝐼𝐼14.4𝜇𝜇𝜇𝜇 = 𝐼𝐼0 e cm ≈ 𝐼𝐼0
𝐼𝐼𝑠𝑠 𝐼𝐼𝑜𝑜
s
• This suggests negligible light attenuation for a 14.4 μm thick
PtTiO2 catalyst coating and the assumption of a uniform rate of
toluene decomposition rT in Eq 5.1 is valid.

39
 ACTIVITY 5.1: DESIGN OF WATER FILTER
A home appliances company decides to produce an all-in-one water treatment system for
treating tap water. It contains a filter to remove suspended solids, a reverse osmosis (RO)
membrane unit to remove dissolved ions, an ultraviolet disinfection system, and a granular
activated carbon filter for final polishing of taste and odor. We have been assigned to
design the RO module (Figure 5.1). The water treatment system is attached to a faucet with
a water flow rate of 7.5 L/min. The water flow rate across the membrane (i.e., permeate
flow rate) is directly proportional to the pressure difference (Pfc − Pp) minus the osmotic
pressure difference (𝜋𝜋fc − 𝜋𝜋p),

𝑄𝑄𝑝𝑝 = 𝑘𝑘𝑤𝑤 𝐴𝐴 𝑃𝑃𝑓𝑓𝑓𝑓 − 𝑃𝑃𝑝𝑝 − (𝜋𝜋𝑓𝑓𝑓𝑓 − 𝜋𝜋𝑝𝑝 )

where kw is the mass-transfer coefficient of water across the membrane, A is the membrane
surface area, and Pfc and Pp (𝜋𝜋fc and 𝜋𝜋p) are the pressure (osmotic pressure) at the feed-
concentrate side and the permeate side of the membrane, respectively.
40
 The osmotic pressure 𝜋𝜋 is related to the solute concentration C as follows.
𝜋𝜋 = CRT
where R is the universal gas constant and T is the temperature. Similarly, the solute flow
rate across the membrane is proportional to the concentration difference,
𝑄𝑄𝑝𝑝 𝐶𝐶𝑝𝑝 = 𝑘𝑘𝑠𝑠 𝐴𝐴 𝐶𝐶𝑓𝑓𝑓𝑓 − 𝐶𝐶𝑝𝑝
where ks is the mass-transfer coefficient of solute across the membrane and Cfc and Cp
are the solute concentration in the feed-concentrate side and the permeate side of the
membrane, respectively.
As a measure of product performance, a solute rejection ratio (Rej =1 − Cp∕Cf ) of
99% is required for a feed stream with 100 ppm of Ca2+. Determine the product
specifications, particularly the membrane area and the permeate flowrate.

The following parameters are available:

Pfc = 30 atm Pp = 1 atm T = 298 K
kw = 2.87 × 10−3 m∕h atm ks = 6.14 × 10−4 m∕h
41
42
43
 DESIGN OF
FUNCTIONAL
PRODUCTS

44
 EXAMPLE: DESIGN OF A CONTROLLED-RELEASE HERBICIDE
Active ingredient
GRANULE
Design scenario: Polymeric matrix
Herbicides such as 2,4-dichlorophenoxyacetic acid (2,4-D) are
commonly used for killing unwanted broadleaf weeds while leaving
the desired crop (usually grass) to grow unmolested. However, 2R
herbicides spread on soil are easily lost due to leaching,
volatilization, or biodegradation. Only a portion of the herbicide is
used in killing weeds. An agrochemical company wants to develop a controlled-
release granule with the active ingredient uniformly dispersed within a polymeric
matrix (Figure 5.10). The controlled-release granule should release the herbicide at
a fast rate at the beginning to kill weeds quickly and at a lower rate thereafter to
retard the regrowth of the weeds for a long period of time.
An agrochemical company wants to develop a controlled-release granule with the
45 active ingredient uniformly dispersed within a polymeric matrix.
 STEP 1 – SPECIFY PRODUCT PERFORMANCE
Based on experimental tests, it is decided that the controlled-release granules should
deliver 0.5-2 mg/m2 day of the herbicide from day 1-10, followed by a lower dose
of 0.1-0.5 mg/m2 day from day 11-90. Thus, a total maximum amount of around 60
mg of herbicide should be applied to 1 m2 of soil.

These product attributes can be obtained by varying the granule size, the granule
material, and the concentration of herbicide within the granule.

STEP 2 – IDENTIFY THE KEY INGREDIENTS

Cellulose and lignin are commonly used in the formulation of controlled-release
herbicides. The polymer matrix controls the release rate of herbicide from the
granules.
46
 STEP 3 – IDENTIFY THE PHYSICOCHEMICAL PHENOMENA
• The release of herbicide from the granules is controlled by two different mechanisms
–dissolution/degradation of the granule and the diffusion of herbicide within the
granule.
1. For cellulose granules, the release profile is controlled by the dissolution rate of
cellulose in soil because the dissolution rate of cellulose in water is fast compared
with the diffusion rate of herbicide.
2. Although lignin is biodegradable, its degradation rate is slow and the release
profile is controlled by the diffusion rate of herbicide within the granule.

47
 STEP 4 – USE MODELS AND DATA TO IDENTIFY PRODUCT
SPECIFICATIONS
1. Dissolution of cellulose matrix
The rate of herbicide release from the granule depends on the dissolution rate of
cellulose, which is linearly dependent on the surface area of the granule,
𝑑𝑑𝑀𝑀𝑐𝑐
= 4𝜋𝜋𝑘𝑘0 𝑟𝑟 2 (5.11)
𝑑𝑑𝑑𝑑
where Mc is the mass of cellulose, k0 is the dissolution rate constant, and r is the
radius of the granule at time t.
The amount of dissolved cellulose, Mcd, can be obtained by mass balance:
4𝜋𝜋𝜋𝜋𝑥𝑥𝑐𝑐
𝑀𝑀𝑐𝑐𝑐𝑐 = 𝑅𝑅3 − 𝑟𝑟 3 (5.12)
3
where ρ is particle density, xc is the weight fraction of cellulose, and R is the initial
radius of the granule.
48
 STEP 4 – USE MODELS AND DATA TO IDENTIFY PRODUCT
SPECIFICATIONS

Integrating Eq. (5.11) over time after substituting Mc into it gives

𝑘𝑘0
𝑟𝑟 = 𝑅𝑅 − 𝑡𝑡 (5.13)
𝜌𝜌𝑥𝑥𝑐𝑐

Because the herbicide is dispersed uniformly in the matrix, the cumulative amount of
herbicide released from the granule, Mh, can be obtained by replacing xc by xh, the
weight fraction of herbicide in the granule, and substituting r from Eq. (5.13) into Eq.
(5.12),

3
4𝜋𝜋𝜋𝜋𝑥𝑥ℎ 3
𝑘𝑘0
𝑀𝑀ℎ = 𝑅𝑅 − 𝑅𝑅 − 𝑡𝑡 (5.14)
3 𝜌𝜌𝑥𝑥𝑐𝑐

49
 STEP 4 – USE MODELS AND DATA TO IDENTIFY PRODUCT
SPECIFICATIONS

3
4𝜋𝜋𝜋𝜋𝑥𝑥ℎ 𝑘𝑘0
𝑀𝑀ℎ = 𝑅𝑅3 − 𝑅𝑅 − 𝑡𝑡 (5.14)
3 𝜌𝜌𝑥𝑥𝑐𝑐

𝑘𝑘0 = 1 × 10−6 𝑘𝑘𝑘𝑘/𝑚𝑚2 𝑠𝑠

𝜌𝜌 = 2000 𝑘𝑘𝑘𝑘/𝑚𝑚3
𝑥𝑥ℎ = 0.1

Figure 5.11 shows that all herbicide release

profiles do not meet the desired product
performance. The amount of herbicide released is
too much during the first few days of application
and the release can only be sustained for less than
25 days.
50
 STEP 4 – USE MODELS AND DATA TO IDENTIFY PRODUCT
SPECIFICATIONS
2. Herbicide diffusion in lignin matrix
Because a lignin granule degrades very slowly, the release of herbicide occurs by diffusion
from the interior of a granule into the soil, followed by diffusion to the surroundings.
Diffusion within a spherical granule, the rate-determining step, can be modelled using Fick’s
second law,

𝜕𝜕𝜕𝜕 𝐷𝐷 𝜕𝜕 2 𝜕𝜕𝜕𝜕 where C is the concentration of

= 𝑟𝑟 (5.15) herbicide in the granule,
𝜕𝜕𝜕𝜕 𝑟𝑟 2 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕
D is the diffusion coefficient, and
with the initial and boundary conditions: Cs is the concentration on the surface
0 ≤ 𝑟𝑟 ≤ 𝑅𝑅; C = C0 at t = 0 of the granule.
r = R; C = Cs for t > 0 (5.16)
51 r = 0; C = finite for t > 0
 STEP 4 – USE MODELS AND DATA TO IDENTIFY PRODUCT
SPECIFICATIONS
The cumulative amount of herbicide released at time t is given by

𝑡𝑡 ∞
3
𝜕𝜕𝐶𝐶 8𝑅𝑅 𝐶𝐶𝑠𝑠 − 𝐶𝐶0 1 𝑚𝑚2 𝜋𝜋 2 𝐷𝐷𝐷𝐷
𝑀𝑀ℎ = − � 𝐴𝐴𝐴𝐴 � 𝑑𝑑𝑑𝑑 = � 2
exp − 2
−1 (5.18)
0 𝜕𝜕𝑟𝑟 𝑟𝑟=𝑅𝑅
𝜋𝜋 𝑚𝑚 𝑅𝑅
𝑚𝑚=1

The diffusion coefficient in lignin is assumed to be 1 × 10−10 cm2∕s (Pereira et

al.,2002), and Cs is assumed to be the same as the herbicide concentration in
the soil, which is kept at around 2 ppm to be effective (Xin et al., 2012).

52
 STEP 4 – USE MODELS AND DATA TO IDENTIFY PRODUCT
SPECIFICATIONS
Figure 5.12 shows the herbicide
release profile for different sizes of
granules with an initial herbicide
concentration of 200 kg∕m3. It clearly
shows that the 1.4-mm granules with
lignin matrix provide a higher dosage
(0.5–2 mg∕m2 day) in the first 10 days
and a lower dosage (0.1–0.5 mg∕m2
day) in the next 80 days, meeting the
specified product performance.
Smaller granules release too much
herbicide in the first 10 days.
53
 ACTIVITY 5.2: DESIGN OF MOISTURE ABSORBER
A home supplies company decides to produce a reusable moisture absorber product for
removing the water content in air in an enclosed space. A reduction in humidity is
expected to help inhibit the growth of mold and mildew. As a measure of product
performance, the moisture absorber should reduce the relative humidity of air in a closed
coat cabinet with dimensions of 2m × 1m × 1m from 90 % to 50 % at 25 ∘C.

a) What is the minimum amount of silica gel to lower the humidity to 50 %?

b) What is the required amount of silica gel if the reduction is to be completed in two
hours?

Additional information from literature:

The maximum moisture content of silica gel, defined as weight of water per unit weight
of silica gel, is 0.26. The density of silica gel is 2.64 g∕cm3. The diffusion coefficient of
moisture in silica gel is 0.37 × 10−8 m2∕s. L value for Art Sorb is 0.26 at T is 25°C and the
54
relative humidity is 50%. Art Sorb® is a commercially available product.
 MODELS RELATED TO THE PROBLEM
The relative humidity (∅) is given by:
𝑃𝑃
∅= ∗
𝑃𝑃
where P is the partial vapor pressure and P* is the saturated vapor pressure of the
moisture in the air, which can be estimated by the Antoine equation:
𝐵𝐵
log10 𝑃𝑃∗ = 𝐴𝐴 −
𝐶𝐶 + 𝑇𝑇
For water at T = 25°C, A = 8.0713, B = 1730.63, and C = 233.426 (given T in °C
and P* in mmHg).
𝑚𝑚𝑚𝑚𝑚𝑚
𝑃𝑃𝑃𝑃 =
Given the ideal gas law: 𝑀𝑀
where m is the amount of water in the air, M is the molar mass of water (= 18
g/mol), V is the volume of the coat cabinet, R is the universal gas constant (= 8.314
J/mol K).
55
 MODELS RELATED TO THE PROBLEM
The equilibrium moisture content (L) of an absorbent quantifies the maximum amount of
moisture that can be absorbed at a specific temperature and relative humidity:
∗
𝑚𝑚𝑚𝑚
𝐿𝐿 =
𝑚𝑚𝐴𝐴
where 𝑚𝑚m∗ is the equilibrium amount of moisture absorbed by 𝑚𝑚𝐴𝐴 g of absorbent.

56
 SUMMARY

57
 PART I OVERVIEW
• Introduction
• Product specification
• Mathematical modelling
• Elements of the Product Design Framework

PART II
• Design of Chemical Devices
• Design of Functional Product

PART III
• Design of Formulated Products
• Design a Sunscreen Cream
• Design of insect repellent spray
58 • Design a Cream Manufacturing Process
 DESIGNDesign
OF FORMULATED
of
PRODUCTS
Formulated Products

59
 INTEGRATED EXPERIMENT-MODELLING PROCEDURE FOR FORMULATED
PRODUCT DESIGN
1) Collect the relevant knowledge and experiences in terms of basic scientific
understanding, typical ingredients and the selection criteria, heuristics, mechanistic
models, empirical correlations, computer-aided tools, and causal table.
2) Convert consumer preferences into quantitative and measurable technical
requirements.
3) Generate a base-case formula. This can be obtained from the open literature, the
database within the company, or calculations using computer-aided tools.
4) Verify the predicted physicochemical properties with experiments if necessary.
Measure those properties that cannot be predicted accurately by experiments.
5) Perform experimental iteration guided by a causal table until the desired product
attributes are achieved.
60
 EXAMPLE: DESIGN A SUNSCREEN CREAM
Design Scenario
A cosmetic company wants to produce a skin-care cream with the following product
attributes:
• Keep the skin in a moisturized condition
• Block UV light using nano ZnO
• Supplement the skin cream with an herbal anti-oxidant to ameliorate UV-induced skin
damage
̶ The product should be smooth, nonirritating to
the skin, and stable
̶ It should not feel oily or greasy
̶ The product should not run during and after
application.

61
 STEP 1 – COLLECT RELEVANT KNOWLEDGE
Ingredients Ingredients Examples Suggested conc.
(wt%)
Selection criteria and considerations

contained in a Emollient
[HLB value]
Synthetic oils: Caprylic/Capric triglyceride [5],
C12-C15 alkyl benzoate [13]
10-40 • Natural oils are usually more acceptable by
the consumers.
cream, along with Natural oils: Sunflower oil [7], sweet almond oil • Silicone oils are non-sticky and highly
recommended [7], coconut oil [8]
Silicone oils: Cyclomethicone [5], dimethicone
spreadable. Its water repellent property is
good for waterproof sunblock creams.
concentrations and [5]

selection criteria Emulsifier Steareth-2 [4.9], oleth-20 [15], glyceryl 1-6 • Non-ionic emulsifiers are preferred in skin
[HLB value] stearate [3.8], PEG-100 stearate [16], care products.
Polysorbate 20 [16.7]
Stabilizer Sodium chloride, EDTA disodium salt dihydrate 0.01-0.2 --

Neutralizer Triethanolamine, citric acid 0.01-0.5 --

Humectant Glycerol, propylene glycol, butylene glycol, 1-5 • Materials with low freezing point are
sorbitol preferred.
Film former PVP/ dimethylaminoethyl-methacrylate 0.1-2.5 • Film formers can increase film thickness and
copolymer, PVP/ hexadecene copolymer, aloe water resistance, but also leaves a sticky feel.
vera
Thickener Carbomers, xanthan gum, carboxymethyl 0.1-0.5 • Carbomers show thickening function in the pH
cellulose range of 5-9.
Preservative Tocopheryl, diazolidinyl urea, iodopropynyl 0.01-0.5 --
butylcarbamate, methylparaben,
62 propylparaben
 STEP 1 – COLLECT RELEVANT KNOWLEDGE
Model for determining cream viscosity
Oldroyd model for dilute emulsions with a dispersed phase volume fraction (φ) less than
0.7. The viscosity of the emulsion (µe) is affected by φ, the viscosity of the continuous
phase (µ c), viscosity ratio (κ), that is, the viscosity of the dispersed phase to that of the
continuous phase, and the capillary number NCa.
2
5𝜅𝜅+2 5𝜅𝜅+2 2 2 1+𝜆𝜆1 𝜆𝜆2 𝑁𝑁𝐶𝐶𝐶𝐶
𝜇𝜇𝑒𝑒 = 𝜇𝜇𝑐𝑐 1 + 𝜙𝜙 + 𝜙𝜙 2 (5.21)
2 𝜅𝜅+1 10 𝜅𝜅+1 2 1+𝜆𝜆21 𝑁𝑁𝐶𝐶𝐶𝐶

where
19𝜅𝜅+16 2𝜅𝜅+3 19𝜅𝜅+16 19𝜅𝜅+16 2𝜅𝜅+3 3 19𝜅𝜅+6
𝜆𝜆1 = 1+ 𝜙𝜙 ; 𝜆𝜆2 = 1 − 𝜙𝜙
40 𝜅𝜅+1 5 𝜅𝜅+1 2𝜅𝜅+3 40 𝜅𝜅+1 10 𝜅𝜅+1 2𝜅𝜅+3
𝜇𝜇𝑐𝑐 𝑑𝑑𝑝𝑝 𝛾𝛾̇
𝑁𝑁𝐶𝐶𝐶𝐶 =
63 2𝜎𝜎
 STEP 1 – COLLECT RELEVANT KNOWLEDGE
Here, dp is emulsion droplet size, and s is interfacial tension.

The Herschel-Bulkley model shows the dependency of viscosity on shear rate and exhibits
shear-thinning behavior when n < 1,
𝜏𝜏𝑜𝑜
𝜇𝜇 = + 𝐾𝐾𝛾𝛾̇ 𝑛𝑛−1 (5.20)
𝛾𝛾̇
Here, to and K are model parameters and 𝛾𝛾̇ is the shear rate.

64
 STEP 1 – COLLECT RELEVANT KNOWLEDGE
Heuristics for designing lotion and cream
• They are derived from experience, but are often based on sound scientific rationales.
• They also help relate sensorial attributes described in qualitative terms to
quantifiable technical requirements
• Use O/W emulsion if the product should not feel greasy.
• Use W/O emulsion if the product should be resistant to washing and/or
perspiration.
• Prefer a product showing shear-thinning behavior if the product should be
thick at rest but spread easily upon shearing.
• Aim for a viscosity of 20 Pa.s at rest and 0.025 Pa.s at an application shear
rate of 1,000 s-1 to obtain the best consumer acceptance.
• A mixture of emulsifiers with high and low HLB values gives better stability
results than a single emulsifier with the exact required HLB value.
65
 STEP 1 – COLLECT RELEVANT KNOWLEDGE
Test Results Adjustments
Causal table Emulsion cannot • Adjust concentrations of emulsifiers
suggests the form • Consider using different emulsifiers
necessary changes instantaneously
in cream Too viscous • Reduce thickener concentration
Consider another thickener
formulation or
•

Not viscous enough • Increase thickener concentration

processing • Consider another thickener or a mixture of thickeners
conditions that Greasy • Reduce emollient concentration
Change emollients
would nudge the Not smooth enough
•

• Reduce droplet size by increasing the number of cycles and

product attributes pressure of the homogenizer
towards the target • Adjust the emulsion viscosity
Sticky Reduce emollient concentration
attributes. •

• Reduce film former concentration

• Change emollients
pH too high • Add acidic pH controlling compound such as citric acid
pH too low • Add alkaline pH controlling compound such as
triethanolamine
Emulsion not stable • Consider other emulsifiers
66
 STEP 2 – CONVERT CONSUMER PREFERENCES INTO TECHNICAL
REQUIREMENTS

Consumer Preferences Technical Requirements

Smooth cream Emulsion particle size < 5 µm
Non-irritating to skin pH is kept at or below around 7,
preferably around 5
Avoid an oily feel An oil-in-water emulsion is chosen

Feels good when applied Viscosity around 0.015 Pa s

on skin
Avoid appearing runny Viscosity around 20 Pa s

67
 STEP 3 – GENERATE BASE CASE HLB
Value
Ingredient Type
Base Case
wt%
Oil Phase
• Key ingredients: nano ZnO and herbal extract ZnO particles (Type 1) 5
Sunblock agent
4

• Sunflower oil and sweet almond oil were chosen ZnO particles (Type 2) 13
Sunblock agent
12
• Film former was used to form uniform sunscreen Herbal extract -- Antioxidant 2
film on skin C12-15 alkyl benzoate
Sunflower oil
13
7
Emollient
Emollient
1
1
• Emulsifiers are selected based on HLB method. Sweet almond oil 7 Emollient 1
15.5 Co-emulsifier /
The required HLB was ~10 by accounting key Cetyl alcohol opacifier
1

ingredients and emollients Dimethicone

Steareth-2
4
4.9
Emollient
Emulsifier
2
2.5
Oleth-20 15 Emulsifier 2.5
𝐻𝐻𝐻𝐻𝐻𝐻𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟 = � 𝑤𝑤𝑖𝑖 𝐻𝐻𝐻𝐻𝐻𝐻𝑖𝑖 (5.19) Polyvinyl pyrrolidone / -- Film former 0.2
dimethylaminoethyl-methacrylate
copolymer
Here, wi is the weight fraction. Aqueous Phase
Carbomer -- Thickener 0.14
• Steareth-2 and Oleth-20 were mixed to Glycerol -- Humectant 3
Propylene glycol -- Humectant 2
provide an HLB of 9.95 to match the required -- Stabilizer 0.2
value of the oil mixture EDTA disodium salt dehydrate
Triethanolamine -- Neutralizer 0.1
Water -- -- Balance
68 Preservative Phase
Liquid Germall Plus -- Preservatives 0.7
 STEP 4 – VERIFY BASE CASE PERFORMANCE

Emulsion Characteristics • A panel of judges found that the emulsion has

pH 7.5-8.0 acceptable softness and a non-greasy feel
Average droplet size (µm) 1.4 • Average emulsion droplet size was measured to
Viscosity at 0.01 s-1 (Pa.s) 2.14 be 1.4 µm
Viscosity at 1,000 s-1 (Pa.s) 0.014 • The viscosity at a shear rate of 1000 s-1 was
Stickiness Sticky estimated to be 0.014 Pa s, which was close to
Stability Phase separation the desired value
• It also appeared runny on standing, as the
viscosity at low shear rate (0.01 s-1) was much
smaller than 20 Pa s.
• The cream was sticky after applying on skin.
• The cream emulsion was unstable.

69
 Base
HLB Prototype 2 Prototype 3
Ingredient Type Case
Value
wt%

STEP 5 – PERFORM ITERATIONS Oil Phase

ZnO particles (Type 1) 5 Sunblock agent 4 4 4

BY EXPERIMENTS ZnO particles (Type 2) 13 Sunblock agent

Antioxidant
12 12 12
Herbal extract -- 2 2 2
C12-15 alkyl benzoate 13 Emollient 1 2 2
Emulsion Base Case Prototype 2 Prototype 3 Sunflower oil 7 Emollient 1 0.5 0.5
Characteristics Sweet almond oil 7 Emollient 1 0.5 0.5
15.5 Co-emulsifier /
UV blocking -- -- OK 1 1.5 1.5
Cetyl alcohol opacifier
Transparency -- -- OK Dimethicone 4 Emollient 2 2 2
OK Cocoa butter 11.5 Emollient -- 1 2
Antioxidant activity Polysorbate 20 16.7 Emulsifier -- 0.16 0.82
pH 7.5-8.0 7.5-8.0 7.5-8.0 11 Emulsifier -- 4.84 4
GMS 165 (PEG-100 stearate
Viscosity at 0.01 s-1 2.14 5.09 15.2 and Glyceryl stearate)
(Pa.s) Steareth-2 4.9 Emulsifier 2.5 -- 0.18
Viscosity at 1,000 s-1 0.014 0.015 0.018 Oleth-20 15 Emulsifier 2.5 -- --
-- Film former 0.2 -- --
(Pa.s) PVP/dimethylaminoethyl-
1.4 1.4 1.4 methacrylate copolymer
Average droplet size
Aqueous Phase
(mm)
Carbomer -- Thickener 0.14 0.35 0.35
Sticky Non-sticky Non-sticky Xanthan gum -- Thickener -- -- 0.2
Stickiness
Glycerol -- Humectant 3 3 3
Phase Slight phase Stable Propylene glycol -- Humectant 2 2 2
Stability separation separation -- Stabilizer 0.2 0.1 0.1
EDTA disodium salt dehydrate
Triethanolamine -- Neutralizer 0.1 0.1 0.2
-- -- Balance
Water
70 Preservative Phase
Liquid Germall Plus -- Preservatives 0.7 0.7 0.7
 STEP 5 – PERFORM ITERATIONS BY EXPERIMENTS, CTD.
Prototype 2
• To remove the sticky feel, the amount of film former is reduced to zero
• The amount of carbomer (thickener in aqueous phase) was increased to 0.35wt% to
improve the cream viscosity.
• To improve stability, steareth-2 and oleth-20 were replaced by Polysorbate 20 and
GMS 165. A solid emollient, cocoa butter, was added to obtain a final HLB value of
11.7 for prototype 2.
• The viscosity at low shear rate (0.01 s-1) and the stability did not match the
requirements.

71
 STEP 5 – PERFORM ITERATIONS BY EXPERIMENTS, CTD.
Prototype 3
• To further increase the viscosity, xantham gum was used along with carbomer for a
total thickener concentration of 0.55wt%. This resulted in a viscosity of 15.2 Pa.s which
was within the acceptable range albeit still on the low side.
• The stability was also improved by adding steareth-2 back in. The pH was on the high
side but could be easily adjusted using a neutralizer.
• The cream texture was acceptable and it was shown to be stable for a period of more
than three years.

72
 Traditional Procedure for Designing Formulated Products
(an Insect Repellent Spray)
Desired product attributes

Generate a base-case Database within the

formula company

Formulation
Experimental Iterations
specialists
Panel for touch-and-
Performance Tests feel product
attributes; Various
characteristics
Formulated products

73 The search is inefficient and the search space is limited in size!

Design Procedure for an Insect Repellent Spray (to return)

74 Conte et al., AIChE J., 57, 2431 (2011)

How Modeling and Experiments Complement Each Other in the Product Design
Framework
System Optimized
under design?
study Experiments
REALITY
Provide Guidance Verify
VIRTUAL realistic model and insights theoretical
REALITY parameters for solutions
experiments
Model Simulations Optimized
system design?

Integrated modeling, experiments and

synthesis help find innovative yet
realistic solutions!
75
76
 EXAMPLE: A CREAM MANUFACTURING PROCESS
Design scenario:
One of the requirements for a cream product is that the emulsion droplets should
have a diameter smaller than 5 µm to obtain a smooth cream. Develop a conceptual
process design along with the key equipment characteristics and operating
conditions.

The design process includes three steps:

- Process synthesis
- Selection of equipment units
- Selection of equipment operating conditions

77
 STEP 1 – SELECTION OF FLOWSHEET STRUCTURE
Continuous-phase
ingredients
Insoluble
ingredients
Dispersed-phase
ingredients • Incompatible ingredients are mixed separately to
manufacturing creams and pastes

form the continuous and dispersed phases in the pre-

General Flowsheet Structure for

Pre- Pre- Pre-

mixing treatment mixing mixing units.
• Insoluble solid ingredients may need pre-treatment
Heating Mixing Heating
such as size reduction or surface modification before
Cooling being dispersed into the liquid.
Homogen- • The separate phases are then mixed in the mixing unit
ization
at an elevated temperature (e.g., 70°C) to form a
Cooling
pre-emulsion with relatively large droplet sizes.
Processing-
Mixing sensitive • A homogenization step is often necessary to reduce
ingredients
the droplet size.
Post
treatment • Heat-sensitive ingredients are added after
homogenization when the temperature is dropped to
Filling about 40-50°C.
78
Product
 FLOWSHEET FOR THE CREAM PRODUCT
~75 °C dp ~ 5mm
Mineral oil
Stearic acid
Pre- Since an O/W emulsion is to be
Cetyl alcohol Homogen-
I. palmitate mixing & Mixing formed, the oil phase should be
heating ization
Petrolatum added to the aqueous phase
Emulsifiers dp~100mm
Pre-
Water
Glycerol
mixing & Cooling
Carbomer heating

Filling

Product

79
Selection of Emulsification Equipment
Feed
p ~ 3-20 MPa
Feed Emulsion

Orifice Vibrating knife

f ~ 20-50 kHz Gap width
h ~ 0.1 mm
10 12

Emulsion

ultrasonic pressure Valve

109
ε (W/kg ) homogenizer homogenizer

Feed
10 6
toothed disc disperser

Gap width
colloid mill b ~ 0.05-1 mm

103 Rotor

agitated vessel Rotation speed

N ~ 103-104 rpm
Emulsion
100
10-3 10-2 10-1 100 101 102
80 Capacity (m3/h)
 PHYSICAL PHENOMENA FOR MIXING (PRE-EMULSIFICATION) AND
HOMOGENIZATION
• Imagine a drop under mixing. Two competing phenomena in emulsion formation

breakup coalescence
• Droplet breakup
− Occurs when NWe > NWe*

• Droplet coalescence
− Prevented when τ adsorption τ collision << 1
and when τ drainage τ collision > 1
81
 BREAKAGE AND COALESCENCE MECHANISTIC MODELS
Laminar flow Turbulent flow
Weber number γµ c d p C3 ρ cε 2 / 3d 5p / 3
ms = surfactant concentration
(Walstra, 1983) N We = N We = s = interfacial tension
2σ 4σ
Adsorption time 20 Γ 10 Γµ c1 / 2
ε = turbulent energy dissipation
τ adsorption = τ adsorption =
(Walstra, 1983) d p msγ d p ms ρ c1 / 2ε 1 / 2 rate
Collision time π d p2 / 3
τ collision = τ collision = Г = excess surface concentration
(Walstra, 1983) 8φγ 15φε 1/ 3

of surfactant
Drainage time - 40.84 µ cξ
τ drainage =
(Tsouris and ρ c ε 2 / 3 d p2 / 3
Tavlarides,
1994) ξ = 1.872 ln (0.2295κ + 1)
+ 0.127 ln (1.0136κ + 1)

82
Depiction of Breakup and Coalescence Criteria
• Turbulent flow in an agitated vessel for mixing/pre-emulsification
10
12 Mechanistic models for
breakage and
Re-coalescence
9
coalescence
10 below line
coalescence
criterionFeasible • Assume a pre-emulsion droplet size
10
6
(drainage)region of 100 µm
ε (W/kg )

Breakup  ε ~ 25 W/kg
3
10 above line • εav ~ 0.25 W/kg (about 1/100
times of ε)
0 coalescence criterion
10
(surfactant adsorption) breakup
criterion
-3
10
0.1 1 10 100 1000

dp (µ m)
83
 EQUIPMENT CORRELATIONS
• Empirical correlations to relate required shear rate or power density to equipment
design and operating conditions

Equipment Design equations References

Agitated ε av = C4 N 3 D 2 N P Zhou and Kresta
vessel N P = N P (N Re , geometry) (1998)
Colloid mill 2πNRr Wieringa et al.
γ =
b (1996)
Pressure C6 p 3 / 2 Dickinson (1994)
homogenizer ε av = hρ 3 / 2
e

D = impeller diameter
h = homogenizer valve gap width
Rr = colloid mill rotor radius
84
 Impeller Speed and Diameter of the Agitated Vessel
• Agitation speed of around 63 rpm gives εav ~ 0.25 W/kg and the corresponding power
requirement for a 5-tonne vessel
100 1Empirical correlation for
ε = C N 3D2 N agitated vessel
av 4 P

Power requirement (kW)

0.8
• Desirable to use high impeller-to-
Agitation speed (rpm)

80 vessel diameter ratio because

0.6
the emulsion is shear thinning. A
ratio of 0.8 was selected.
0.4
60
T = vessel diameter
0.2
ε av = 0.25 W/kg
(fixed by production
amount)
40 0
0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
D/T
85
 ROTATIONAL SPEED AND GAP WIDTH OF COLLOID MILL,

109

Feasible Region

Similar breakup – coalescaence breakup

Shear rate (s-1)

6
10 criterion
analysis using mechanistic models for
breakage and coalescence for colloid
mill (laminar flow) for a dp of 5 mm 103
led to a shear rate ~ 3⋅104 s-1 coalescence criterion
(surfactant adsorption)

100
0.1 1 10 100 1000
dp (µm)

86
 ROTATIONAL SPEED AND GAP WIDTH OF COLLOID MILL
6000

• Rotor rotational speed (N) and 5000 105 s-1

4 -1
gap width (b) which provide the 5.10 s
4000
required shear rate were chosen

N (rpm )
accordingly 3000 3.10 s
4 -1

2000

1000

0
0 200 400 600 800
b (µ m)

Empirical correlation for a

87 colloid mill
 Workflow diagram showing key tasks and supporting methods and tools in the design of
formulated products, devices and functional products

88
89