Comparison of Batch and Continuous

Process
Royal Society of Chemistry Symposium
Continuous Flow Technology
Chemspec Geneva 2011

Dr. Sergio Pissavini

Business Director
 Micro&Meso reactors:UnderstandInteractionsofTransports&
Reaction

AFRcanintensifyreactionwithvery
Reaction Fast shortresidencetime(secondstofew
with high
Mass/Heat transfer Reaction minutes)withsimilartemperatureas
Resistances Kinetics batch!Transportcontrolledprocess

Good mixing of multiphase is essential

AFR allows elevated temperature without leading to

high impurity due to short residence time/better heat transfer

AFRcanintensifyreactionwithshort
residencetime(fewminutes)butwith
Reaction Mediate
with high highertemperatureand/or
/Slow
Mass/Heat transfer
Reaction concentrationasbatch!(+30~100C)
resistances
Kinetics
Elevated T and/or C & good mixing are essential

Corning Incorporated 2010 2

 Micro&Meso reactors:Mixing&MassTransferFine Design

Gas-Liquid two phase mixing

Mi 2 D
llim De
ete sign
Immiscible liquid-liquid mixing

rs
ca
le
Met
e
3D r scale
De s
ign

Courtesy of Fluent Computational Fluid Dynamics (CFD)

Reactor Design & Flow Hydrodynamics Quantification

Corning Incorporated 2010 3

Micro&Meso Reactors:Efficient MixingforMultiphaseReaction
Systems

Mixing Zones Gas/Liquid

Superior mixing quality for L-L, G-L, G-L-S

Optimized designs for multiphase reactions
High resistance to fouling
Smooth surface for easy clean

Corning Incorporated 2010 4

Micro&Meso Reactors:HeatTransferFine Design&Control
Model Prediction Accuracy:
T < 3 C; HT Coefficient < 8%
Advanced thermal & mechanical

Fu
ll 3
designs became possible at fine

D
Te
dimensional scale

m
pe
ra
tu
(500 m few millimeters)

r
eD
ist
rib
ign

ut
s
lde cale

io
a

n
rm s

Pr
the m

e
ale erm

di
sc

ct
m tilay

ed
l
o na mul

by
ti d
ven grate

C
n
Co nte

FD
i
to 20 80 140
T, 0C

Corning Advanced-Flow
Courtesy of Fluent Glass Fluidic Modules

Corning Incorporated 2010 5

 Micro & Meso Reactor: The Impact of Dimensions
P Q Pressure drop Heat transfer Mixing Quality
= 128 Simple tube U x (S/V) Villermaux
L d4 (bar) (kW/m3.K) (%)
Laminar flow100 000 000

10 000 000
100

1 000 000
90
100 000

10 000
80
80 %
1 000

100

20 bar 10

100 m
0,01 0,1 1 10 100 1000
1 mm 1 cm 1m
450 m 6 mm Corning Incorporated 2010 6
Meso Reactor: Fluidic modules: concept and library

Heat exchange layer Heat transfer Heat exchange fluid

Reaction layer
Heat exchange layer 700 microns Reactants

Mixing
300 microns

700 microns
Pressure drop
Reactants
1 millimeter
4 mm
Corning Incorporated 2010 7
Numberingupandscalingout,InsteadofScalingupforProduction
Production Batch FUNCTIONAL Production Bank: - glass
FLUIDIC MODULES - ceramic
(SSiC)

Reactor: pilot
Pilot scale

Scale-Up Effect
Yes NO Fluidic Modules

Low Flow
feasibility
Lab scale

Corning Incorporated 2010 8

 Not to forget cost and scale up risk in process defintion
Manufacturing plant

Pilot Plant
Cost

Kilo Lab

Lab scale
xity
ple
Com

Advanced Flow
Reactor Technology
Time Production Banks
8 -15 months, X MM$

Laboratory Pilot Plant Scaleup Technology transfer large scale production

One tool engineered to fit all needs

Scale up failure, no scale up
Limited scale up, management of numerous batch
Key Development of second Risks at the no limitation
activities generation process manufacturing site
Complex transfer to regular production
Transfer to
manufacturing no issues
Regular production Final process parameter definition for large scale
production simple

Corning Incorporated 2010 9

Numberingupandscalingout,InsteadofScalingupforProduction

340
320
300 Internal volume
280
260
240
220
200
milliliter

180
160
140
120
100
80
60
40
20
0
0 LF
1 2G1 G2
3 G3
4 G4
5 6

Corning Incorporated 2010 10

 Industrial Plant capability: a flexible tool to adapt to
flexible needs Those estimated values are provided for indication

Typical tonnage needs Number of reactors

(Isolated product) Feeds concentration in wt%
3 5 20 50 100
10 kg / day (24h) 2 1 1 1 1
100 kg / week (150h) 2 2 1 1 1
5 mT / month (600h) 10-3-1 17-6-2 5-2 2 1
100 mT / year (7200h) 16-6-2 10-3-1 7-3-1 3-1 2
50 mT / 3 months (1800h) 11-4 19-7-2 14-5-2 6-2 3-1
400 mT / year 21-7 13-5 10-3-1 12-4-2 6-2
1,000 mT /year 18 11 24-8-3 10-3-1 14-5-2
10,000 mT /year 26 11 16-6

# Gen1 at 10 kg/h processed per reactor # Gen3 at 90 kg/h processed per reactor

# Gen2 at 30 kg/h processed per reactor # Gen4 at 270 kg/h processed per reactor

Corning Incorporated 2010 11

 Fromprocesstargettoimplementation:anintegratedChemistry
ChemicalEngineeringview
Pressures exceeding the reactor capability?
Temperatures exceeding the reactor capability?
Zone 1
Increasing
Preliminary economic analysis: look to the full plant
Likely no applicability
Chemistry not compatible with reactor materials? benefits and/or Energy, water, raw material, solvent (running & cleaning)
Equilibrium reactions? not doable
Solid precipitates formed? Labor cost
Very slow kinetics? Zone 2
Solid reactants or catalysts? Possible Capital depreciation cost
benefits, but
Gas reactants?
technical
Homogeneous catalysts? challenges
product yield and quality (including off-spec reduction)
Gas evolution?
Reaction benefits from pressures > 120 Pa
product throughput
Zone 3
Unstable intermediates?
Plant safety
Fast kinetics? Advantages likely
Highly toxic reactants or byproducts? speed of scaling up
Reaction requires or benefits from T < -10 C?
Rapid mixing required to optimize the reaction? flexibility to turn throughput up and down
Highly exothermic?

Primary Choice:
Engineering watch out:
hazardous reaction path
intermediate unstability feeding system (pumps selection, pressurized vessels, .)
intermediate severe, extreme condition distribution system for parallelized reactors line
new domains
process control system
minimum footprint must
kinetics complexity: diluition, dosing, criogenic temeperature
on line analytics
continuous equipment for work up

Consider flow chemistry as early in design as possible

Involve a multidisciplinary team: chemist, chemical
engineering, process control

Corning Incorporated 2010 12

Scale out, numbering up: manage distribution

Internal distribution

External and active

bifurcation

External & passive

Typical internal fluid
distribution in a plate heat
exchanger consecutive
Doc Schmidt
(www.apiheattransfer.com/)

Corning Incorporated 2010 13

each solution having its pros and cons.

INTERNAL EXTERNAL
MANIFOLD TYPE
PASSIVE ACTIVE

APPEARS SIMPLE APPEARS COMPLEX

COMPLEXITY
COMPLEXITY HIDDEN
ACTUALLY SIMPLER THAN IT APPEARS
AND REAL

DIRECT MEASUREMENT NOT POSSIBLE EASY

FLOW ADJUSTMENT NOT POSSIBLE POSSIBLE YES

COST LOW LOW HIGH

Corning Incorporated 2010 14

 together
Cornings with an adapted
approach relies on external fluid
an internal distribution.
numbering-up..
A combination of passive and active distribution system.
(EP 2 193 839 A1). Conventional internal
numbering-up: no way to
correct a wrong distribution

Taking
Corning intowith
approach account actual
connecting process
paths betweenneeds and
channels chemical
(EP 2 172 260 A1)
engineering know-how
Corning Incorporated 2010 15
On line analytics, PAT an enabler of continuous flow

Courtesy of Dr.
Brian Marquardt
CPAC Seattle

Corning Incorporated 2010 16

Economicscomparison:case1,Nitration
Reducedsolventusage,higheryieldofsaferoperation
Shortening of Development Cycle
Value generated from: reduced solvent usage,
higher yield & significant improvement in safety

HO HO O2NO
OH + HNO3 ONO2 X ONO2
R R R
Product By-Product
Caused Safety Issue

Feed
Excellent Mixing of
preparation immiscible liquids
Substrate

Nitration Quench and neutralization

Solvent

Flush
H2O

HNO3

H2O NaOH NaOH NaOH

Ref: Chemistry Today, 26 (5), 1-4, Sept~Oct (2008) Commercial scale demonstration Product

Corning Incorporated 2010 17

Economicscomparison:case1,Nitration
Reducedsolventusage,higheryieldofsaferoperation

Corning Incorporated 2010 18

Case 1: Plant design Comparison Batch vs. Continuous
Selective Nitration
Comparison between Continuous and Discontinuous Plants -
INSTRUMENTATION AND CONTROL - TOTAL
Comparison between Continuous and Discontinuous Plants -
UTILITIES CONSUMPTION
550
800
500 3
m /d
450
400
350
5.000 500
Amount

500
300 3 3 3
m /h Nm /h m /d
250
200 3.000
300.000
150 kfrig/h kg/h 2.500

100 kg/h

50 100.000 1.000 1.000 10

kfrig/h kg/h 500 3
m /h 50 tons/h 4
0 3
kg/h Nm /h
CONTROLLERS LOCAL INDICATOR DCS INDICATORS tons/h

CONTINUOUS BATCH BRINE PROCESS WATER SCRUBBER NITROGEN STEAM ACQUEOUS WASTE WATER
WASTE TREATMENT
CONCENTRATOR

Comparison between Continuous and Discontinuous Plants -

BATCH CONTINUOUS
ESTIMATED TIME FOR REALIZATION

BASIC DESIGN

DETAIL DESIGN
Comparison between Continuous and Discontinuous Plants -
REGULATION LOOP
BUILDING CONSTRUCTION
220
CARPENTRY
200
MECHANICAL ASSEMBLING 180

WIRING 160
140
AUTOMATION
Amount
120
COMMISSIONING 100
80
VALIDATION
60
TOTAL WITH OVERLAP 40
20
TOTAL WITHOUT OVERLAP
0
FLOW RATE PRESSURE pH TOTAL
0 5 10 15 20 25 30 35

CONTINUOUS BATCH
BATCH CONTINUOUS

Corning Incorporated 2010 19

 Case 1: Plant design Comparison Batch vs. Continuous
Selective Nitration
Comparison between Continuous and Discontinuous Plants -
PRODUCT DIRECT COST
40,00

35,00

30,00

25,00

/kg 20,00

15,00

10,00

5,00 Comparison between Continuous and Discontinuous Plants -

INVESTMENT COST
0,00
RAW JOB WASTES PLANT COST BUILDINGS TOTAL 25 000 000
MATERIAL (5Y) (10Y)

CONTINUOUS BATCH 20 000 000

15 000 000

10 000 000

5 000 000

0
PROCESS STORAGE UTILITIES GENERAL TOTAL

CONTINUOUS BATCH

Corning Incorporated 2010 20

Case 1: Plant Footprint Comparison Batch vs. Continuous
Selective Nitration
NITRATION PLANT BATCH 400 ton/yr
42.80 meters 42.80 meters 42.80 meters
22.80 meters

22.80 mt

22.80 mt
30.80 meters
14.80 meters

NITRATION PLANT CONTINUOUS

400 ton/yr

Corning Incorporated 2010 21

Economics comparison: Case 2, Pantoprazole intermediate
A reference case: a nitration of an aromatic compound.

This step would lead typically to an amino group (NH2), and ultimately to the
formation of azotized cycles
This step may require an upstream reaction to protect the existing groups already
present on the aromatic group. (and of course, ultimately, a deprotection step)

Specifically we will evaluate the nitration of p-alkoxyaniline, which is an

intermediate for the manufacturing of pantoprazole.
Annual throughput: 50 t/y
Batch capacity; 350kg in 24h (in full campaign condition. Start-up and shut-down
for week-end have to be taken into account)

Acetylation, nitration an deprotonation steps are considered

Conversion: more than 99%

Yield: nearly 99% . By-products negligible and removed during washing/ extraction
operation.
Safety issue in batch: no safety issue.
Corning Incorporated 2010 22
Pantoprazole intermediate: Process flow diagram
ALKOXYANILINE
DCM
ACETYLATION
CH3COCl

NaOH 30%

H2O, NaOH 5% WASHING / SEPARATION SOLVENT CHANGE Et-OH

AZEO DISTILLATION HYDROLYSIS NaOH

HNO3 98%
NITRATION NEUTRALIZATION HCl

H2O, NaOH 5% DILUTION / DECANTATION CRISTALLIZATION H2O

NaOH 30% WASHING / SEPARATION FILTRATION

Corning Incorporated 2010 23

Washing flow diagram (batch)
PROCESS SOLUTION
DCM

H2O Water washing / decantation

Extraction / decantation

NaOH 5% Soda washing / decantation Extraction / decantation

DCM

Batch extraction, separation process with limited

H2O Water washing / decantation
efficiency due to stirring efficiency
Droplet size distribution is not uniform, different
decantion time
Vessel geometry is not favoring the decantation
process
AZEO DISTILLATION Interface management and phases contamination
batch process is inducing a phases back-mixing

Corning Incorporated 2010 24

 Washing flow diagram (continuous)
Continuos process in flow
PROCESS reactor/
SOLUTION

an AFR mixer is highly efficient allowing to have small

DCM
decanter (25l
volume) with proper geometry to facilitate the phase separation
Water washingprocess,
being a continuos
H2O / decantation the phase are extracted from top and
Extraction / decantation
bottom, far away from the interface layer, avoiding phase contamination
consequently the water washing phase and related azeotropic
distillation
NaOH 5% is not
Soda required
washing anymore
/ decantation Extraction / decantation
DCM

more efficient technologies could also be envisioned, such as counter-

current continuous liquid-liquid extraction column, or rotating contactor-
REDUCED
centrifugalWater
H2O separator devices
washing / decantation

additional saving could be envisioned by less extraction solvent usage.

The same will apply to the separation steps/ washing after the nitration
step. AZEO DISTILLATION

Corning Incorporated 2010 25

 Continuous PLANT Batch PLANT
22.60 mt

28.00 mt
mt mt
16.7037.10

Corning Incorporated 2010 26

Pantoprazole intermediate: batch vs continuous
Comparison between Continuous and Discontinuous Plants -
UTILITIES CONSUMPTION
200.000
kfrig/h

100
Nmc/h

500
40.000 kg/h 4
kfrig/h 300 3.000 30 tons/h
kg/h mc/h Nmc/h 2
1.000 tons/h
mc/h
Comparison between Continuous and Discontinuous Plants -
BRINE PROCESS WATER SCRUBBER NITROGEN STEAM EQUIPMENT VOLUMES

BATCH CONTINUOUS
80 000
75 000
70 000
65 000
60 000
55 000
50 000
45 000
Liters

40 000
35 000
30 000
25 000
20 000
15 000
10 000
5 000
0
BATCH CONTINUOUS

REACTION VOLUME TANKS TOTAL

Corning Incorporated 2010 27

 Pantoprazole intermediate: batch vs continuous
Comparison between Continuous and Discontinuous Plants -
INSTRUMENTATION AND CONTROL - TOTAL

250

200

150
Amount

100

50

Comparison between Continuous and Discontinuous Plants -

0 REGULATION LOOP
CONTROLLERS LOCAL INDICATOR DCS INDICATORS

CONTINUOUS BATCH
110
100
90
80
70
Amount

60
50
40
30
20
10
0
FLOW RATE TEMPERATURE PRESSURE LEVEL WEIGHT TOTAL

CONTINUOUS BATCH

Corning Incorporated 2010 28

 O
PE
R
AT
IO
N
A
L

0
10
20
30
40
50
60
70
80
90
100
C
O
S
T
so
lv
en
ts
re
ac
t an
ts

w
at
er

w
as
te
m
an
po
w
er
PL Pl En
AN an e

0
20
40
60
80
100
120
T ta rg
y
C m
O Bu or
ST
ild tiz
eq in at
ui g io
pm am n
en or
t tiz
at
io
re n
ac
to
r
st
or
ag
e
Batch

ut
ilit
ie
s
Continuous

w br
in
as e
te
tre
at
m
en
ts
Operational cost and capital cost comparison:

pi
pi
Pantoprazole intermediate: batch vs continuous

ng bu
an ild
in
d g
as
w se
iri
n m
g bl
in
an g
d
au
to
m
at
io
n
Corning Incorporated 2010
Batch

29
Continuous
Pantoprazole intermediate: batch vs continuous
Overall, translated in the effective cost, the saving (50 tons per year plant)
are summarized as:

Operating cost saving: average 650000 /year

Capital investment savings for a grass root plant:

2300000

Based on optimized overall plant equipment and operation related to

the switch from batch to continuous

Corning Incorporated 2010 30

 Economic Index (E1/E0)-1-1
lo
w
la
bo
ur
co
le st
ss
ru
n
do
w
le n
ti
ss m
ca e
pi
t al
le i nv
ss
es
ca tm
pi
t en
al t
r is
k
in
sc
al
in
g
up
le
ss
lo

-2,5
-2,1
-1,7
-1,3
-0,9
-0,5
-0,1
0,3
0,7
1,1
1,5
1,9
2,3
2,7
3,1
3,5
3,9
4,3
4,7
5,1
5,5
5,9
6,3

w en
er
ra g
w y
m co
at st
er
ia
lc
on
su
m
pt
le io
ss n
w
as
te
tr
ea

Greener Index
tm
en
t
le
ss
re
w
le or
ss k
co
en st
er
av gy
oi co
d ns
do
w u m
n pt
st io
re n
am
op
er
a ti
ng
le un
ss it
w
Advanced Flow Reactors: Greener and More Economical

as
hi
ng
Corning Incorporated 2010

so
lv
en
t
31