Crystallisation

Learning Outcomes
After this lecture you should be able to..
Describe crystal growth and nucleation
Define the constituents of a solution and degrees of saturation
Describe the solubility curve
List three methods of achieving supersaturation/crystallisation
Describe nucleation and crystal growth
Give an example of crystallisation
Analyse a crystallisation process in a pharmaceutical plant
Explain how to crystallise by cooling

Crystallisation - Introduction
Crystallization refers to the formation of solid crystals from
a homogeneous solution.
It is a solid-liquid separation technique
Used to produce
Sodium chloride
Sucrose from a beet solution
Desalination of sea water
Separating pharmaceutical product from solvents
Fruit juices by freeze concentration
Crystallisation requires much less energy than evaporation
e.g. water, enthalpy of crystallisation is 334 kJ/kg and
enthalpy of vaporisation is 2260 kJ/kg

What is a crystal?
A crystal is a solid form of substance (ice)
Some crystals are very regularly shaped and can be
classified into one of several shape categories such rhombic,
cubic, hexagonal, tetragonal, orthorhombic, etc.
With pharmaceuticals, crystals normally have very irregular
shapes due to dendritic growth which is a spiky type
appearance like a snowflake. It can be difficult to
characterise the size of such a crystal.
Crystals are grown to a particular size that is of optimum
use to the manufacturer. Typical sizes in pharmaceutical
industry are of the order of 50m.

Crystallisation
In general, crystallisation should be a straightforward
procedure. The objective is to grow crystals of a particular
size or crystal size distribution (CSD). If this is not
successful, problems that can occur are:
Inconsistency from batch to batch
Difficult to stir and filter
Crystals damaged in filtration/agitation
Creation of polymorphs
Difficult to dry

Crystal Size Distribution - CSD

A crystal
Paracetamol crystals
precipitated from acetone
solution with compressed
CO2 as antisolvent using
the GAS technique

Source
http://www.ipe.ethz.ch/laboratories/spl/research
/crystallization/project05
Accessed 131106

Solutions, Solubility and Solvent

A solid substance (solute) is termed soluble if it can dissolve
in a liquid (the solvent) to create a solution
The solution is a homogenous mixture of two or more
components
Solubility is normally (but not always) a function of
temperature
Solubility can change if the composition of the solvent is
changed (e.g. if another solvent is added)
Solubility is usually measured as how many grams of
solvent can be dissolved in 100 grams of solute

Solubility curve sucrose

Ref: http://www.nzifst.org.nz/unitoperations/conteqseparation10.htm

Solubility curve NaCl

Solubility Curve NaCl in H2O
100
90

Solubility g/100g H2O

80
70
60
50
40
30
20
10
0
0

10

20

30

40

50

60

Tem p Deg C

70

80

90 100

Saturation
An Unsaturated or Undersaturated solution can dissolve
more solute
A saturated solution is one which contains as much solute as
the solvent can hold
A Supersaturated solution contains more dissolved solute
than a saturated solution, i.e. more dissolved solute then can
ordinarily be accommodated at that temperature
Two forms of supersaturation
Metastable just beyond saturation
Labile very supersaturated
Crystallisation is normally operated in the metastable region

Concentration
Kg solute/100kg solvent

Solubility curve - Saturation diagram

Supersaturated
Or Labile

Metastable
Stable

Temperature
Stable zone crystallisation not possible
Metastable zone MSZ crystallisation possible but not spontaneous
Labile crystallisation possible and spontaneous
We need a supersaturated solution for crystallisation

Concentration
kg solute/100kg solvent

Supersaturated
Or Labile

Metastable

Stable
Undersaturated

Temperature

5
Metastable
Stable
Undersaturated
Temperature

Concentration
kg solute/100kg solvent

Concentration
kg solute/100kg solvent

Supersaturated
Or Labile

Supersaturated
Or Labile
3

Metastable
Stable
Undersaturated
Temperature

Concentration
kg solute/100kg solvent

Supersaturated
Or Labile

1
X

Metastable
Stable
Undersaturated

Temperature

Concentration
kg solute/100kg solvent

Determination of MSZW
Cool
Heat

1
2

Dilute

3
4

Temperature
MSZW is a function of kinetics (wider for faster cooling)

Achieving Supersaturation
Concentration

Labile

C
B
Metastable

A
Stable

Temperature
ABC - If A is cooled, spontaneous nucleation not possible
until C is reached. No loss of solvent
ADE If solvent is removed, nucleation occurs at E
Can combine cooling and evaporation

Crystallisation Techniques
In general crystallisation is achieved by
Cooling a solution
If supersaturation is a function of temperature
Removal of the solvent by evaporation
Where supersaturation is independent of temperature
(e.g. common salt)
Addition of another solvent to reduce solubility
When solubility is high and above methods are not
desirable, or in combination with above methods
The new solvent is called the anti solvent and is chosen
such that the solubility is less in this new solution than
it was before

Change in Concentration
Concentration
kg solute/100kg solvent

Supersat, C = C C*
Nucleation

Crystal Growth
C*

Temperature

Crystallisation

Ref: http://www.cheresources.com/cryst.shtml

Concentration
kg solute/100kg solvent

Activity
B

C
D
Temperature

Explain what is happening in each stage from A to D in the

above crystallisation

Supersaturation, C
Supersaturation is the driving force for
Nucleation
Crystal Growth
Creation and control of supersaturation is the key to
successful crystallisation
High C High Crystal Growth + High Nucleation
High nucleation means a lot of fines (filtration problems)
High crystal growth means inclusion of impurities
C is usually maintained at a low level in the
pharmaceutical industry so the right CSD is achieved

Activity How would you crystallise?

Have a look at the solubility curves provided
List the three techniques for achieving supersaturation
Which would you use and why?

Nucleation
Crystallisation starts with Nucleation
There are two types of nucleation Primary and Secondary
Primary relates to the birth of the crystal, where a few tens
of molecules come together to start some form of ordered
structure
Secondary nucleation can only happen if there are some
crystals present already. It can occur at a lower level of
supersaturation than primary nucleation.
Often, industrial crystallisers jump straight to secondary
nucleation by seeding the crystalliser with crystals
prepared earlier

Primary Nucleation
The birth of a new crystal is complex and involves the
clustering of a few tens of molecules held together by
intermolecular forces
Homogeneous small amounts of the new phase are formed
without any help from outside
Heterogeneous nucleation is assisted by suspended
particles of a foreign substance or by solid objects such as
the wall of the container or a rod immersed in the solution
these objects catalyse the process of nucleation so it occurs
at lower levels of supersaturation
Homogenous conditions are difficult to create so
heterogeneous nucleation is more normal in industrial
crystallisation (if it is not seeded)

Primary Nucleation

Nucleation rate

Heterogeneous

Homogeneous

Supersaturation

Secondary Nucleation
Secondary nucleation is an alternative path to primary nucleation and
occurs when seed crystals are added
Nucleation occurs at a lower supersaturation than primary when crystals
are already present
Secondary nucleation is due to:
Contact nucleation crystals are created by impact with agitator or
vessel wall. Nuclei are created by striking a crystal the number
created is related to the supersaturation and the energy of impact.
Can occur at low supersaturation.
Shear nucleation shear stresses in the boundary layer of fluid flow
create new crystals/nulcei. Embryos are created and swept away
that would have been incorporated into an existing crystal
Very important in industrial crystallisers as this is the main type of
crystal growth used
Difficult to predict or model nucleation rates

Supersaturation and Crystal Growth

For low supersaturation primary nucleation is not
widespread. Secondary nucleation on existing crystals is
more likely. Result is small numbers of large crystals
For high supersaturation primary nucleation is widespread.
This results in many crystals of small size.
Slow cooling with low supersaturation creates large crystals
Fast cooling from high supersaturation creates small crystals
Agitation reduces crystal size by creating more dispersed
nucleation
Rate of cooling can affect purity of product - see handout on
slow cooling v rapid cooling

Seeding
The type or quality of seed used can influence the
crystallisation process
Good seed results in a good crystallisation, i.e. a particle
size distribution that does not include fines
Bad seed can increase the amount of fines produced
Good and Bad can be defined by the seed crystal size
Source of seed can be
Material left from the last batch (no tight control on
particle size)
Specially prepared material or material from a good
batch (tight particle size distribution)

When to Seed?
Seed can be added dry to the crystalliser
Allow time for dispersion throughout the crystalliser this
can take several hours
Never seed to the right of the solubility curve the solution
is not yet ready
Never seed to the left of the solubility curve nucleation is
already happening
Seed half way between the two

No. of Particles

Seeded V Unseeded

Unseeded

Seeded

Time
With unseeded nucleation does not occur till later when
supersaturation is higher. High rate of nucleation follows.

Seeding advantages, disadvantages

Advantages include
Point of nucleation from batch to batch is repeatable
Reduces the number of fines
Improves predictability of scale up
Can prevent polymorphism
Disadvantages
Experience has shown that not any seed will do, good
quality seed is needed
Extra addition point on vessel or hand hole is usually
opened to manually add seed which could create health
and safety issues

Crystal Growth
Once nucleation has occurred crystal growth can happen
The objective of crystallisation is to produce the required
crystal size distribution (CSD)
The actual CSD required depends on the process
Crystal growth rate has proved difficult to model and
empirical relationships developed from laboratory tests are
generally used
Two steps to crystal growth
Diffusion of solute from bulk solution to the crystal
surface
Deposition of solute and integration into crystal lattice

CSD

Counts

The following CSD is very common. 50 m crystals are the desired outcome in this
crystallisation. However, some fines are created also.

Fines problem

50m

Crystal Size

Impurities and Crystal Growth

Impurities can prevent crystal growth
If concentration of impurities is high enough crystals will
not grow
Should not be an issue in the pharmaceutical industry
For example, the production of non crystalline sweets such
as lollipops (sugar crystals give an unwanted grainy texture)
Addition of acid breaks sucrose into fructose and glucose
This makes it difficult for sucrose crystals to form
because the impurities damage the structure
Addition of other sugars creates the same result