continuous crystallisation is right here and right no crystallisation is right here and right now...

Continuous crystallisation is right here and right now

Professor Xiong-Wei Ni, BSc, PhD, CEng, CSci, FIChemEProfessor Xiong-Wei Ni, BSc, PhD, CEng, CSci, FIChemENiTech Solutions Ltd

10 Cammo Place, Edinburgh, Scotland

www.nitechsolutions.co.uk

Paris, 3rd October 2010

Structure

1. Crystallisation science

© NiTech Solutions Ltd, 2010

2. Challenges in industrial crystallisation

3. Continuous oscillatory baffled crystalliser

4. Case studies

5. Forward remarks

Detail from an 1850 drawing of the London sugar refineries of Messrs. Fairrie and Co showing a copper crystallising pan for sugar.

The worker to the right of


The worker to the right of the pan is holding a mallet that was used to bang on the pan to induce nucleation of the crystallisation process –Book of Joel

• Crystallisation is a relatively simple process

• But the governing science is not well understood

• In industrial scales, operators’ experience still plays a major part

Under-saturated

Solubility

Metastable Zone (MSZW)

Concentration

Supersolubility

Labile

Crystallisation “pathway”

1. Crystallisation Science1. Crystallisation Science


part

Supersolubility is thermodynamically not found and kinetically not well defined, depending on:

Temperature

• Temperature profile• Rate of generating supersaturation• Solution history• impurities• Fluid dynamics

Scale dependent parameter

Ulrich & Strege, J. of Crystal Growth, Vol.237, p2130-2135, 2002

Science

Supersaturation

Concentration

Temperature

Cooling profile

Mixing

Solvent/additives

Seeding

Crystal size and shape

Morphology

PurityProcess

ProductProduct


Nucleation mechanism

Nucleation rate

Growth mechanism

Growth rate

Seeding

Science

Functionality (materials)

Bio-availability (drugs)

Surface activity (catalysts)

Texture (foods)

• Secondary nucleation

– attrition, seeding

Nucleation•What initiates primary nucleation?

• not well understood

• 3-D assembly of molecular clusters on nm scale

• very fast kinetics On-set nucleation

• very fast kinetics

• not easy to detect Tsat

On-set dissolution

nucleation

Tcry

crysat TTMSZW −=

Detection of post-nucleation

concentration = 30 g/L, cooling rate = 0.5 oC/min

MetastableMetastable zone width (MSZW) zone width (MSZW) MSZW depends on reactor, scale and operating conditions


concentration = 45 g/L, f = 2 Hz, xo = 10 mm

• Dissolution – a thermodynamic process

• Crystallisation –a kinetic process

•Mass transfer control (mixing dependent ) or surface integration control (less mixing dependent)

• Crystal growth phase must take place within the MSZW

Crystal growth

Metastable

Crystallisation “pathway”


• Nucleation dominates �� small crystals

• Growth dominates �� large crystals

Under-saturated

Solubility

Metastable Zone (MSZW)

Temperature

Concentration

Supersolubility

Labile

• Nucleation via collision breeding:

• Walls of vessel

• Free surface of liquid

• Internals, e.g. stirrer, baffle

• Impurities, e.g. particulates,

Mixing affects

0

5

10

15

20

25

30

35

0 5000 10000 15000

Reynolds Number, Re o [-]

MS

ZW [

oC

]

30g/L 0.5°C/min

35g/L 0.5°C/min

30g/L 2.0°C/min

35g/L 2.0°C/min

200


• Impurities, e.g. particulates, seeds

• Crystal/crystal & crystal/vessel collision

• Inhomogeneities due to mixing

0

20

40

60

80

100

120

140

160

180

200

0 5000 10000 15000 20000 25000 30000

Reynolds Number, Re o [-]

Cry

stal

Siz

e [

µµ µµm]

30g/L 2.0°C/min

35g/L 2.0°C/min•MSZW

• Crystal size distribution

Crystal Size

Supersaturation

MSZW

PolymorphicForm

ProcessConditions

Shape

Crystal Size

Polymorphic

Supersaturation

Shape

MSZW

Growth Kinetics Nucleation Kinetics

PolymorphicForm

ProcessConditions

Shape

Hydrodynamics Heat Transfer

Chemicals Behaving Badly (CBB)Chemicals Behaving Badly (CBB)

Crystal Size

Supersaturation

MSZW

Growth Kinetics Nucleation Kinetics

FTIR

USSUV Vis

PolymorphicForm

ProcessConditions

Shape

Hydrodynamics Heat Transfer

VideoMicroscopy

PIV/LDAReactionCalorimetry

Flow thruXRD

CFD

Courtesy of Dr White, Heriot-Watt University

• Simultaneous multi-technique measurements during a crystallisation process

• Have promoted better understanding of crystallisation in lab scale

Lab scale batch crystallisation

• Promoted better operation of crystallisation in small scale

• Linear cooling identified as one of the key operational parameters

• Data rich

• Lack of correlations between data

• Disturbing flow conditions

• Our understanding in scaling up STR is limited

• There is no agreed rule or parameter on scaling

2. Challenges in industrial crystallisation

− Tip velocity of impeller− Mean velocity− Averaged shear− Ratio of diameter to height


− Ratio of diameter to height− Geometric direct− RPM− STR Reynolds number− Power density− CFD

• Velocity gradient or non-homogeneity increases with scale

•Mixing gradient leads to temperature and mass gradient

ScaleScale--up of stirred tanksup of stirred tanks

Laboratory

Industrial scale


Mixing cannot linearly be scaled in STR

Kinetic control

Mass/heat transfer control

Heat Transfer on ScaleHeat Transfer on Scale--UpUp

Volume (m3) Area (m2) Area / unit Volume (m2/m3)

Typical 90 litre STR

0.09 0.9 10


Specific heat transfer area decreases with scale


6.82 15.8 2.3

Industrial batch crystallisation

Inconsistent morphology

Wide size distribution

Mixing

gradient

Concentration

gradient

supersaturtion

gradientpolymorph

MSZWConcentration

∆∆∆∆C


Temperature

gradient

MSZW and

Crystallisation path

nucleation

& growth

Polymorph

& size

Wide size distribution

Difficult to filter

Temperature

∆∆∆∆

∆∆∆∆T

Practical issues

• Linear cooling (oC/min) is difficult to achieve in large STR


• Implementation of lab measuring techniques is practically impossible in industrial scale

i) consistent product quality can only be achieved in plug flows;

ii) it is very rare to obtain plug flow conditions in

From the viewpoint of fluid mechanics

3. Continuous crystallisation


ii) it is very rare to obtain plug flow conditions in batch STR!

iii) Plug flow conditions can only be attained in continuous operation.

a) Use a series of CSTRs

…

b) Employ a tubular reactor

The conventional ways to achieve plug flow include


…

..

Plug flow is achieved when the number of CSTRs goes to infinite

Significant increase in inventory, running & capital costs, yet far from plug flow

Near to plug flow is obtained operating at turbulent flows

Significant high flow rates lead to very long reactor length, very short reaction time and high costs

Oscillatory baffled reactor (OBR)

Demonstration of radial mixingDemonstration of radial mixingwith good dispersionwith good dispersion

Reo = 1250 (xo = 4 mm, f = 1 Hz)

Real system 3-D CFD simulation

NiTech’s Continuous OBR (COBRTM)

Tracer

Conductivity probes


Oscillation

Out

liquid

Probe 13.7 meters away from injectionProbe 27.9 meters away from injectionProbe 310.1 meters away from injection

D = 0.00007~0.003 m2/s

• Mixing is not controlled by net flow

• Plug flow achieved at laminar flows

Key differences from other tubular devices on the market


laminar flows

• Excellent heat & mass transfers

• No concentration gradients

• Good with solids

Equipment Selection - Reactor TechnologyE

xoth

erm

Many reactions fall into the band of exothermagainst reaction timeMicro

reactorsSpinning

disc reactors

Compact reactors

Reaction time

Exo

ther

m

ms secs mins hrs

Static mixer reactors Oscillating flow reactors

CSTR

Continuous oscillatory baffled crystalliser


baffled crystalliser (COBCTM)

Heat Transfer on Scale-Up

Volume (m 3) Area (m 2) Area / unit Volume (m 2/m3)


0.09 0.9 10


6.82 15.8 2.3


COBC™ has larger specific surface area for HT

Typical 72m long 100mm

diameter COBCTM

0.56 22.6 40

COBC offers

� Consistent fluid mechanical conditions for nucleation and growth

� Different mixing mechanism in comparison to traditional stirring


traditional stirring

� Excellent heat transfer characteristics

� Easy operation, e.g. cooling/heating at different sections

Control over cooling profile

90-10 oC Flow Rate 0.3 L/min Cooling Rate 2.0 oC/mi n

y = -1.7118x + 81.031

R2 = 0.993460

70

80

90

80-10oC Flow Rate 0.945 L/min Cooling Rate 6.5oC/mi n

y = -6.5517x + 83.727

R2 = 0.9918

y = -6.5524x + 83.696

R2 = 0.9916

y = -6.5657x + 83.8

R2 = 0.9919y = -6.5541x + 83.668

R2 = 0.9917y = -6.5629x + 83.699

R2 = 0.9919

0

10

20

30

40

50

60

70

80

90

0 2 4 6 8 10 12

Time (min)

Tem

p (o

C)


R2 = 0.9934

y = -1.7103x + 80.986

R2 = 0.9935

y = -1.7106x + 80.97

R2 = 0.9933

y = -1.7112x + 80.972

R2 = 0.9933

y = -1.7121x + 81.032

R2 = 0.9936

0

10

20

30

40

50

60

0 10 20 30 40 50

Time (min)

Tem

p (o

C)

Any cooling profile, e.g. linear, parabolic, non-continuous, step-wise and etc, can be obtained along COBCTM

Any lab monitoring techniques can directly be implemented along COBCTM without modification

a) Pharma API

4. Case studies

The residence time in the COBCtranslates to 12 minutes, compared to a batch cycle time of 8 hours, demonstrating a significant potential improvement in throughput relative to a batch manufacturing facility.

Product performance

�All continuously generated material filters very well–Material specific resistance ~106 m/kg

––Typically 107 – 109 m/kg for APIs in manufacture

�Agglomerated / Aggregated material appears compressible

�Rods appear relatively uncompressible following cake consolidation

Conceptual Flowsheet

�We could incorporate continuous dissolution and crystallisation

into an overall continuous pures process

An Alternative Process Design (2)Particle Engineering to remove milling step

Capital Costs (new builds) would be approx 50% lower

�Operating Team could be reduced by about a further 1/3rd

�Solvent consumption reduced�Solvent consumption reduced

�Overall impact on Cost of Goods–Reduced Depreciation costs

–People £622k/year

–10% solvent consumption reduction ~ £400k/year

–No losses in milling operation

Organic Process Research & Development, 2009, 13, 1357-1363

b) Filtration index for a pharma product

COBCTM deliver� Product characteristics/performance� Consistent filtration index of 25


STR COBCTM

c) Crystal moisture contents


Moisture content:50% 73%

Crystallisation achieved in 1/8th of the time

OBC STR

d) Filtration rate vs crystal size


Filtration rate (L/m 2.hr)

648.9 2125 24610

Purity by HPLC (%)

99.26 99.44 99.37

Yield by HPLC (%)

98.31 99.07 98.80

e) Fractionation of edible oils


� Two stage cooling (17 and 0.3 oC/min)

� Consistent IV achieved0

5

10

15

20

25

0 50 100 150 200

Process Time (min)

Filt

ratio

n ti

me

(m

in)

Benchmark

T05C

T08C

T10C

T11C

T12C

T13C

T14C

T16C

• Significant reduction of crystallisation time

• Improved filtration time

f) Crystallisation for purification

� Purity increased from 91.2% to 97.8%

© NiTech Solutions, 2010

to 97.8%� Customer’s

benchmark: 95.5%

g) Extended Operation 24 hrs a day for 7 days


Vanisal sodium data

0

20

40

60

80

100

120

140

160

0 50 100 150 200

Time (hour)

Van

isal

Sod

ium

(m

g/m

l)

Valve a

Valve f

As pirin tes t

0

50

100

150

200

250

300

0 50 100 150 200

Time (hour)

Ace

tyls

alic

ylic

Aci

d (m

g/m

l)

Valve a

Valve b

Valve c

Valve f

Aspirin run


SEM of crystals

Cycle 16Cycle 12 Cycle 24

Cleaning

Washing data for Aspirin

Cleaned during 3rd wash

1st order kinetics

a) How did nucleation take place in OBC without seeding, while seeding was essential in benchmarking cases?

b) Why did OBC have higher nucleation temperature &

Some interesting findings


b) Why did OBC have higher nucleation temperature & narrower MSZW than traditional crystallisers?

c) Why/how did OBC allow consistent crystal morphology?

d) Why/how did OBC achieve the same/better purities at faster cooling rates, while impurities increase with cooling rates in benchmark cases?

a) GSK led initiative with industrial companies, e.g. Pfizer, AZ

b) Supported by universities (Bath, Edinburgh, Glasgow, Heriot-Watt, Loughborough and Strathclyde)

Continuous Manufacturing And Crystallisation (CMAC)


Heriot-Watt, Loughborough and Strathclyde)

c) Supported by Scottish Government and UK funding bodies

d) A global “one-stop” shop for continuous crystallisation and manufacturing

• Plug flow tubular crystallisation technology, such as COBCTM, can be used for both pilot and industrial scale crystallisation operations

• Any lab scale monitoring techniques can directly be fitted along the COBCTM

4. Forward remarks


fitted along the COBCTM

• COBCTM can be incorporated into existing plants

Continuous crystallisation using COBCTM is not the future, but it is right here and right now.

Dr Gayle De Maria, Chemistry Today

Acknowledgement


Colleagues from NiTech

continuous crystallisation is right here and right no crystallisation is right here and right now...

Documents