The University of Western Ontario

Department of Chemical and Biochemical Engineering

Wencor LTD (LTEE).

Simulation of Process for

Manufacturing of Clopidogrel
Bisulfate with SuperPro
Designer Software Package
Advisor:
Dr. Shazhad Barghi

Project Engineers:

Milana Trifkovic
306-916 Wonderland Rd.S.
London, ON N6K 2V8
519.471.2675
milanat@gmail.com
250174208

Sandra Cardoso
19 Cherish Court
London ON N6K 4H2
519.472.3061
sandra.cardoso@gmail.com
250114116
Table of Contents

1 Abstract..........................................................................................................6

2 Acknowledgments.........................................................................................7

3 Introduction....................................................................................................8

3.1 History and Background...........................................................................8

3.2 Current Trends and Future Directions......................................................9

4 Need Analysis..............................................................................................10

4.1 Market Study..........................................................................................10

4.2 Economic and Cost Factors...................................................................13

4.3 Plant Capacity and Time Basis..............................................................13

5 Review of Process Options........................................................................14

6 Selected Process.........................................................................................15

6.1 Active Pharmaceutical Ingredient (API) Production...............................15

7 Process Description and Model Development with SuperPro Software

17

7.1 Coupling Reaction (Strecker Synthesis)................................................19

7.2 Efficient Hydration of Nitriles to Amides.................................................22

7.3 Resolution..............................................................................................25

7.4 Freebasing to the Amide........................................................................27

7.5 Methylation.............................................................................................29

7.6 Acid Addition...........................................................................................31

8 Simulation Results ...................................................................................35

8.1 Material and Stream Balances...............................................................35

8.2 Scheduling..............................................................................................36

2
9 Process Economics....................................................................................40

9.1 Fixed Capital Cost..................................................................................40

9.2 Annual Operating Costs.........................................................................43

9.3 Profitability Analysis................................................................................45

10 Conclusion...................................................................................................47

11 References.................................................................................................49

12 Appendix........................................................................................................50

3
List of Figures

Figure 1. Chemical Structure of Clopidogrel Molecule...........................................8

Figure 2. Results of CAPRIE Study (Harker, Boissel, Pilgrim & Gent, 1996).......11

Figure 3. CURE Trial Results (Gerschutz & Bhatt, 2002)....................................12

Figure 4. Simplified Diagram of Process..............................................................15

Figure 5. Summary of Reaction Scheme..............................................................16

Figure 6. Gantt chart for Manufacturing Process of Clopidogrel Bisulfate...........37

Figure 7. Overlap between Step 2 and 3 due to CIP............................................39

4
List of Tables

Table 1. Unit Opearations for Step I.....................................................................21

Table 2. Unit Operations in Step 2........................................................................23

Table 3. Unit Operations for Step 3......................................................................26

Table 4. Unit Operations for Step 4......................................................................28

Table 5. Unit Operations in Step 5........................................................................30

Table 6. Unit Operations for Step 6......................................................................33

Table 7. Overall Process Data..............................................................................36

Table 8. Bulk Raw Materials Requirements (Entire Process)...............................36

Table 9. Purchase Costs (PC) of Major Units.......................................................41

Table 10. Fixed Capital Estimate Summery.........................................................43

Table 11. Annual Operation Cost..........................................................................44

Table 12. Profitability Analysis for Manufacturing CPG-BS Plant.........................45

Table 13. Executive summery of economic evaluation.........................................46

5
1 Abstract

A complex pharmaceutical process for the manufacturing of Clopidogrel

Bisulfate (an anti-platelet drug) was successfully designed and simulated using

SuperPro Designer (SPD) software package. The plant was designed based on

lab-scale data provided by Sanofi-Synthelabo (US patent: 6858734).

Multifunctional equipment operating at moderate temperatures and atmospheric

pressure were designed to reduce the production and maintenance costs. The

performance of each unit as well as the overall process were evaluated by the

software which facilitated diagnosis and solution of the problems. The software

was extremely useful not only in design and simulation but also in economical

analysis. The simulation was the authors enhancement of the fourth year design

project at undergraduate level for Chemical Engineers in the department of

Chemical and Biochemical Engineering at the University of Western Ontario.

6
2 Acknowledgments

The project engineers would like to thank their advising professors, Dr. Shazhad

Barghi, Dr. Sohrab Rohani as well the invaluable help of Mehdi Sheikhzadeh,

from The University of Western Ontario. We would also like to thank the

Chemical Engineering Department for purchasing the SuperPro Designer

(Academic Site Edition) v. 6.0. for this project after the authors request. The

help provided by Dr. David Stradiotto of Apotex Pharmachem has also been

invaluable and the project engineers would also like to express their thanks to

him as well.

7
3 Introduction

A brief history and background of the process as well as current trends in this

market are shown in the section below.

3.1 History and Background

Clopidogrel (CPG) is a potent anti-platelet drug, which was launched in 1997 by

Sanofi-Synthelab (US patent: 6858734) with the brand name Plavix. It is

indicated for the prevention of vascular thrombotic events in patients at risk.

Clopidogrel Bisulfate is inactive in vitro, and a hepatic biotransformation is

necessary to express the full anti-aggregating activity of the drug. Six different

polymorphs have been identified so far, but only Form I and Form II are used in

pharmaceutical formulations as they are the only polymorphs that exhibit

bioavailability (Koradia, et al. 2004). The chemical structure of CPG (methyl(+)-

(S)--(o-chlorophenyl)-6,7-dihydrothiento[3,2-c]pyridine-5(4H)-acetate), in 2D

and 3D, is shown below in Figure 1.

Figure 1. Chemical Structure of Clopidogrel Molecule

8
3.2 Current Trends and Future Directions

Cardiovascular disease accounted for the deaths of almost 75,000 Canadians in

2002, more than any other disease (W1, 2005). Heart disease is one of the

leading causes of death for men and women in Canada today. The exact number

of Canadians who have cardiovascular disease is unknown, but it is estimated

that one in four Canadians have some form of heart disease, disease of the

blood vessels or is at risk for stroke (W1, 2005). If this estimate is accurate,

approximately 8 million Canadians are affected.

The economic burden worldwide in treating patients who have suffered and

survived a stroke has been extensively studied and found to average 3% to 5%

of all national healthcare costs (Evers et al., 2004). It is no doubt because of

these costs that the move to prevent thrombotic events in patients at risk through

means such as drug therapy has become so common.

In a British study, the efficacy of 39 different anti-platelet drugs, including

Aspirin and Plavix were studied in patients at risk of stroke. Overall, it was

found that those treated with any anti-platelet drug had an 8.9 % reduction in

stroke versus those treated with no drugs. It was also reported that patients

treated with Plavix, in combination with other anti-platelet drugs, experience

the greatest reduction in occurrence of stroke (Robless et al., 2001)

This report summarizes the design simulation and economical analysis of the

pharmaceutical process using SuperPro Design software.

9
4 Need Analysis

The section below summarizes the various aspects of a full needs analysis

including a market study, possible environmental issues and finally economic and

cost factors to be considered.

4.1 Market Study

The patent for Plavix expires in 2007, and many generic pharmaceutical

companies are currently in the process of filing patents on the production of

clopidogrel bisulfate. Interest in the production of this drug comes from the fact

that sales of anti-thrombotic drugs have more than doubled in the past few years

(W2, 2005).

As of March 2004, the anti-thrombotic market reached annual sales of $13 billion

(USD) and has shown constant growth since 1999 (W2, 2005). Plavix is

currently the market leader with sales reaching nearly $4 billion in 2004 (W2,

2005).

Various studies comparing the effectiveness of different drugs, including

clopidogrel bisulfate, at preventing the occurrence of stroke in patients who had

recently suffered a thrombotic event have been conducted

The CAPRIE (Clopidogrel vs. Aspirin in Patients at Risk of Ischemic Events)

study was a double blind trial which compared the effect of Plavix (75 mg daily)

to Aspirin (325 mg daily) in preventing stroke. The study followed

10
approximately 19, 000 patients with recent histories of myocardial infarction 1

(within 35 days) and recent histories of ischemic stroke 2(within 6 months) for an

average of 1.6 years. Figure 2 below summarizes the results of this study.

Fatal or Non-Fatal vascular events

Months of follow up

Figure 2. Results of CAPRIE Study (Harker, Boissel, Pilgrim & Gent, 1996)

It can be seen from this graph that patients using Plavix experienced less fatal

and non-fatal vascular events compared to the patients treated with Aspirin.

Compared with aspirin, clopidogrel reduced the combined risk of ischemic stroke,

myocardial infarction or vascular death by 8.7 % (Harker, et al. 1996).

The CURE (Clopidogrel in Unstable Angina to Prevent Recurrent Ischemic

Events) trial studied the efficacy of a combination of Plavix and Aspirin on

patients exhibiting unstable angina3. Approximately 12, 500 patients were

1
Heart attack. Damage to the heart muscle due to insufficient coronary artery blood supply
(www3.uta.edu/sswtech/sapvc/information/teens13_15/Teens_(ages13-15)_Glossary.htm)
2
The most common kind of stroke; caused by an interruption in the flow of blood to the brain
(wordnet.princeton.edu/perl/webwn)
3
Unstable angina occurs at rest and usually means that a coronary artery has narrowed to such a
critical degree that the heart is not receiving enough oxygen even at rest
(www.med.miami.edu/glossary/art.asp).

11
randomized to receive Plavix (75 mg daily) and Aspirin (75 325 mg daily)

or placebo and Aspirin (75 325 mg daily) and were treated for up to one year

(Gerschutz & Bhatt, 2002).

Figure 3 depicts the results from the CURE trials and is shown below.

Figure 3. CURE Trial Results (Gerschutz & Bhatt, 2002).

From this graph, it is clear that Plavix combined with Aspirin is more effective

at reducing various detrimental vascular events compared to Aspirin alone.

Overall, there was a 20 % relative risk reduction (95% CI of 10%-28%) for the

Plavix treated group (Gershutz & Bhatt, 2002).

From both the CAPRIE and CURE studies it is clear that Plavix is effective at

reducing stroke, myocardial infarction or cardiovascular death in patients at risk.

12
4.2 Economic and Cost Factors

Currently the cost of one Plavix tablet is $2.50 (CAD) which is very costly for a

patient that requires to take one tablet/day. Recent sales forecasts made by IMS

Health, a reputable health-care consulting company, estimates that worldwide

sales of generic drugs will grow 14 % to 17 % annually, compared to the 8 %

growth expected in the overall drug market (IMS Health, 2006).

With this predicted growth, it is likely that a generic version of Clopidogrel

Bisulfate would create a favorable profit for the manufacturing company.

Using the SuperPro Design software package, a full economic and cost analysis

of the process in terms of cost of equipment, raw material, utilities is shown later

on in the report.

4.3 Plant Capacity and Time Basis

The proposed plant capacity for the manufacturing of CPG is 91.25 million tablets

per year, although there will also be production of other active pharmaceutical

ingredient at the same site.

5 Review of Process Options

13
The process is related to a patented procedure therefore alternative processes, if

they exist at all, are not available. However, after reviewing the procedure

described in the patent and developing the process described in this report, the

main alternative considered was the number of reactors to be used.

The entire drug synthesis could be performed using one reactor, but the choice of

two reactors in parallel was made according to the length of the batch process.

Although the initial investment will be greater, the main advantage of two reactors

is the possibility of running two reaction steps or operations, such as cleaning,

simultaneously.

One other alternative for the reactor set up that was considered was the

possibility of an adjustable size reactor. That is, being able to attach different

sized bottoms to the reactor as dictated by the capacity required for a given

reaction. This option is advantageous since it would allow the plant to be adapted

for different processes with a smaller initial investment. However, the chances of

contamination would be high and could pose a problem in the quality of the final

product.

One main advantage with both of these alternatives is the shortening of the batch

length and consequently, the increase in the return on investment (ROI).

6 Selected Process

The selected process is a modified version of the method found in the Sanofi-

Synthelab patent (US patent: 6858734). Figure 4 below depicts a very simplified

14
representation of the main units required for manufacturing of Clopidogrel

Bisulfate.

Figure 4. Simplified Diagram of Process

6.1 Active Pharmaceutical Ingredient (API) Production

There are six chemical reactions involved in producing Clopidogrel Bisulfate from

the raw materials. Figure 5 summarizes these steps, including the operating

temperatures, and the estimated time required for the reactions.

15
 Figure 5. Summary of Reaction Scheme

The sample Process Flow Diagram for the Step 1 with is shown in Appendix.

16
7 Process Description and Model Development with SuperPro
Software

The following section describes the process and simulation with SuperPro

software. SuperPro is an advanced simulation and data management tool for the

modeling of complex, recipe-based batch processes such as those found in the

pharmaceutical industries. The tool supports a wide spectrum of engineering

workflows over the full lifecycle of a process. The software contains a

comprehensive library of unit operations and models that supports the

incremental development of a recipe at different scales and permits the

construction of a single composite model of the entire process (Petrides et al.,

2002).

Although not covered in this report, the software is very useful for secondary

processes in pharmaceutical industry; such are packaging and solids processes.

This report focuses on simulation of the primary process (the production of the

active pharmaceutical ingredient, Clopidogrel Bisulfate). The simulation of the

entire process for manufacturing Clopidogrel Bisulfate is shown in Figure 6.

17
 Manufacturing Process of Clopidogrel Bisulfate

Coupling Reaction Hydration Reaction

S-1091 S-202 Gaseous Stream
S-1071
S-102
S-103
S-104 S-201
S-204
S-213
S-101
S-105
S-106 P-4 / RV-102
Storage
P-1 / R-101 S-1061 S-108
P-2 / RV-101
Vessel Procedure Mixer-Settler Extraction
P-3 / R-101

Resolution Reaction Vessel Procedure

S-208 S-206
Gaseous Stream2 S-302
S-205
S-304 S-311

S-301
P-7 / RV-103
Storage
S-303
S-207 P-5 / RV-101
P-6 / R-102 Mixer-Settler Extraction
S-306 P-8 / F-101
S-305 Vessel Procedure
Nutsche Filtration

Freebasing Reaction S-501 S-505

S-405
Methylation Reaction
S-133
S-401 S-502 S-134
S-403 P-10 / RV-104
S-503
Storage S-504

S-402
S-404 S-137
P-12 / F-101
P-11 / R-101 S-506
P-9 / R-101
Nutsche Filtration
Vessel Procedure S-601
Vessel Procedure

S-608 Crystallization and Drying

S-607
S-604 S-602
S-603

CPG-BS
P-15 / DCDR-101 P-13 / R-102
S-606
Double Cone Drying Vessel Procedure
P-14 / F-101
Nutsche Filtration

Figure 6. Clopidogrel Bisulafate Process

18
7.1 Coupling Reaction (Strecker Synthesis)

O CN

T=40-50C
1.H2O N
NH H
+ NaCN +
S S Cl
Cl
6,7dihydro-4H-thieno[3,2-c]pyridine 2-chlorobenzaldehyde 2-(2-chlorophenyl)-2-(6,7-dihydrothieno[3,2-c]
pyridin-5(4H)-yl)acetonitrile
Scheme 1. Reaction Mechanism for Strecker Synthesis (Step 1)

Process Description of Step I

The raw materials are received and put under quarantine until they pass testing

criteria for purity. Sodium bisulfate (S-101) is mixed with purified water (S- 105) in

the reactor (R-101). The reactor is not pressurized (although is still equipped with

a pressure relief valve (PRV-101) as a safety precaution) and the addition of all

reactants, and solvents is done under atmospheric pressure (approximately

101.3 kPa) and room temperature (approximately 25 C). Once the sodium

bisulfate is dissolved, o-chlorobenzaldehyde (S-102) is pumped via (P-101) into

the same reactor, resulting in a white precipitate. 6, 7, dihydro-4-H thieno [3, 2, c]

pyridine (S- 103) is pumped into the reactor via (P-102) and sodium cyanide (S-

104) is also added to the reaction mixture. The reactor jacket temperature is

maintained at the desired temperature range of 40 -50 C by a combination of

steam (S-112) and cold water (S-110) and maintained for 6 hours until the

reaction is quenched with water (S-105) via pump (P-103). The product of this

reaction is transferred by gravity into a receiving vessel (RV-101). Ethyl acetate

(S-107) is pumped (P-107) into the same receiving vessel and the mixture

undergoes liquid-liquid extraction. The organic layer (S-108) is pumped (P-106)

19
back to the reactor while the aqueous layer is left in the receiving vessel (RV-

101). Here the aqueous layer is extracted a second time to maximize the

isolation of the product, which is then sent back to the reactor via a pump (P-

106). The aqueous layer (S-109) is sent to the waste treatment facility.

Process Simulation of Step 1

The process described above is simulated using SuperPro Designer (SPD)

application. The first step in the simulation is registering all the ingredients

involved in the process. This can be done by retrieving the specific component

from SPD pure component library, or by entering the physical data manually in

user databank.

The user enters the process flowsheet by putting together unit procedures

selected from the SPD library. The equipment included for this step is:

Stirred reactor (R-101) in which the Intermediate I is produced

Mixer-Settler extractor (RV-101) in which the extraction is carried out.

Each unit procedure mentioned above contains a series of a unit-operations as

shown below Table 1.The operation sequence was set according to the process

described for Step I.

Table 1. Unit Opearations for Step I

Equipment Tag Unit Operation

R-101 Charge (Water)

20
 Charge (Sodium Bisulfate)
Heat
Agitate
Charge (Chlorobenzaldehyde )
Charge (6,7 dihydro-4H-pyridine)
Charge (NaCN)
Batch Stoich. Reaction
Charge (Water-Qenching)
Transfer Out
RV-101 Mixer Settler Extraction (Extract Intermediate I)

Each operation was initialized with the appropriate engineering data such are

charge quantities, reaction stoichiometry, and scheduling relations. The operation

duration was fixed (US patent: 6858734). Furthermore, the operations start was

specified either based on the beginning of the batch (charging the raw material),

or related to the end of another operation (i.e. reaction was started at the end of

charging of the reactants). Reaction operation required entering the reaction

stoichiometry, and the extraction operation required a separation yield for each

stream component. The reaction conversion was known from the patent data

(96%), and the separation was assumed to be perfect for the purpose of

simplicity of the model.

The quenching operation was not available in the software, and thus water-

quenching charge operation was used to compensate for this operation.

21
7.2 Efficient Hydration of Nitriles to Amides

CN CONH2

T=80-82C
N 1.KOH N

2.t-BuOH
S S
Cl
(2-cholrphenyl)-(6,7-dihydro-4H-thieno
[3,2-c]pyrid-5-yl) acetonitrile
Cl
2-chloro-phenyl-(6,7-dihydro-4H-thieno
[3,2-c] pyrid-5-yl) acetamide

Scheme 2. Reaction Mechanism for Efficient Hydration of Nitriles to Amides (Step 2)

Process Description for Step 2

The excess solvent is first vaporized using vacuum conditions, and it is collected

into the receiving vessel (RV-102). The product of the first step remains in the

reactor where butyl alcohol (S-201) is pumped (P-201) and KOH (S-202) is

added. This mixture is stirred vigorously after which it is refluxed using the

condenser (C-101) for 3 hours at 80-82 C. The reactor jacket is then cooled (S-

212) to 30 C and purified water (S-204) is pumped by (P-204) to stop the

reaction. The product is not soluble in water, thus, in this case, water behaves as

an antisolvent. More purified water (S-204) is pumped by (P-204) and cooled by

(RV-203) to 5-10 C, and added slowly to the reactor for 15 minutes. The reaction

mixture is transferred by gravity into a receiving vessel (RV-101) and ethyl

acetate (S-206) is pumped (P-205) into the same receiving vessel where the

mixture undergoes liquid-liquid extraction. The organic layer (S-207) is pumped

(P-206) back to the reactor while the aqueous layer (S-208) is sent for waste

treatment. The residual solvents vapors are removed from the reactor (R-102)

22
under vacuum by the condenser (C-101), which is sent to solvent recovery in the

same fashion as in Step 1.

Process Simulation of Step 2

The flowsheet was expanded by adding three additional unit procedures selected

from the SPD library.

Stirred reactor (R-101) in which the Intermediate II is produced

Mixer-Settler extractor (RV-101) in which the extraction is carried out.

Storage Tank (RV-102) in which the excess solvent it collected

The unit-operations in the sequence needed to simulate the Step 2 are shown in

Table 2.

Table 2. Unit Operations in Step 2

Equipment Tag Unit Operation

Transfer in
Batch Vaporization
Charge (KOH)
Charge (Butanol)
Agitation
R-101 Batch Heating
Batch Stoich. Reaction
Batch Cooling (Chilled Water)
Batch Stoich. Reaction
Transfer Out
In Place Cleaning (CIP-1)
Transfer In
RV-102
Store
RV-101 Mixer Settler Extraction (Extract Intermediate
II)

23
The excess solvent from the first step had to be transferred to the receiving

vessel (RV-102) before the next reaction was initiated, for the purpose of

optimizing the reactor volume. This was simulated by using Batch Vaporization

operation. The components needed to be evaporated were specified, as well as

the percentage of their removal. The reactor (R-101) was shared between the

procedures for Step 1 and Step 2, and this was specified in the reactor data, and

also taken into account when scheduling the various operations. The reaction

stoichiometry was specified in a similar manner as in the Step 1, and the

Intermediate II was produced (conversion 94%). The In Place Cleaning operation

was scheduled after the reactor content was transferred to the receiving vessel

(RV-102). Methanol was chosen as a cleaning agent, and the duration of

cleaning was also specified. As mentioned in the process description the reaction

mixture is refluxed by using condenser for three hours, and this is a quite

frequent way of achieving the desired conversion in pharmaceutical processes.

However, this operation was not available in the SPD library. The extraction was

simulated in the same way as in Step 1; the partition coefficients for several

components were selected. For the purpose of the simplicity of the calculation,

the total separation of the desired component was assumed, and the partition

coefficients assigned accordingly.

24
7.3 Resolution

CONH2 CONH2
O

N N S
OH
O
S S Cl
Cl O
2-chloro-phenyl-(6,7-dihydro-4H- (+)-(S-(2-chlorophenyl)-(6,7-dihydro-4H-thieno-[3,2-c]pyrid-5-yl)
thieno[3,2-c] pyrid-5-yl) acetamide acetamide (1S-(+)-camphor-10-sulfonic acid ) salt
Scheme 3. Reaction Mechanism for Resolution (Step 3)

Process Description for Step 3

The excess of ethyl acetate that comes in the reactor with the Intermediate II is

first removed by vacuum into the receiving vessel (RV-103). Acetone (S-301) and

CSA-acetone solution (S-302) are pumped, by (P-301) and (P-302) respectively

into the reactor (R-102) containing the product from the previous reaction. This

addition is done at 15-20 C over the course of 4 hours. After the reaction is

completed, the solvents are removed from the reaction mixture under reduced

pressure by condenser (C-101) and the reactor jacket is cooled to below 8 C to

encourage precipitation of the product. The slurry (S-303) is transferred via

gravity to a filter (F-101) where it is filtered and washed twice with acetone (S-

304). The filtrate (S-306) is sent to the waste treatment and the filter cake (S-

305) is transferred to the second reactor (R-101).

Process Simulation of Step 3

The Step 3 is simulated using the following unit procedures from the SPD library:

Stirred reactor (R-102) in which the Intermediate III is produced

25
 Nutsche Filter (F-101) in which the separation of the Intermediate III

occurs

Storage Tank (RV-103) in which the excess solvent it collected

The operations in the sequence needed to simulate the Step 3 are shown in

Table 3.

Table 3. Unit Operations for Step 3

Equipment Tag Unit Operation

Transfer in
Agitation
Batch Vaporization
Charge (CSA)
Charge (Acetone)
R-102
Batch Stoich. Reaction
Batch Vaporization
Batch Cooling (NaCl Brine)
Transfer Out
In Place Cleaning (CIP-1)
Transfer In (Ethyl Acetate + Residuals from Step
RV-103 2)(Acetone)
Transfer In
Store
Transfer In
Cloth Filtration
F-101 Cake Wash (Acetone)
Transfer Out
In Place Cleaning (CIP-1)

The Batch Vaporization was used twice in this step to simulate the removal of

solvents under reduced pressure. The reduced pressure of 50 KPa was

specified, and the components to be removed were specified as well. The first

26
time, the excess of the solvent from the extractor was vaporized, and the second

time acetone was removed from the reaction mixture. The stoichiometric reaction

was simulated using the conversion of 60% of the Intermediate II, which was a

reference component. The product of Step 3 was filtered using Nutshce filter unit

procedure. The cake thickness as well as the cake dryness had to be specified in

order to simulate the procedure. The cake was also washed with acetone.

7.4 Freebasing to the Amide

CONH2
CONH2
O
N
N S
OH
O S
S Cl
Cl O

(+)-(S-(2-chlorophenyl)-(6,7-dihydro-4H-thieno-[3,2-c]pyrid-5-yl) (+)-(S-(2-chloro-phenyl)-(6,7-dihydro-4H-
acetamide (1S-(+)-camphor-10-sulfonic acid ) salt thieno[3,2-c] pyrid-5-yl) acetamide

Scheme 4. Reaction Mechanism for Freebasing to the Amide (Step 4)

Process Description for Step 4

Once the solid product from Step 3 is transferred to the second reactor (R-102),

purified water (S-403) and sodium carbonate solution (S-401) are added using

pumps (P-401) and (P-402) respectively. This mixture is stirred for 2 hours and

after the reaction is completed, ethyl acetate (S-402) is pumped (P-103) into the

reactor for extraction of the product. The reaction occurs at 40 C. The aqueous

layer contains the undesired (R) - enantiomer and is removed from the bottom of

the reactor (S-404) where it may undergo further purification. The solvents from

the organic layer are removed by condenser (C-101) and sent to the receiving

27
vessel (RV-104) via (S-405). This process results in obtaining the solid product

(S-enantiomer).

Process Simulation for Step 4

The Step 4 is simulated using Stirred reactor (R-101) in which the Intermediate

IV is produced. The operations in the sequence needed to simulate the Step 3

are shown in Table 4.

Table 4. Unit Operations for Step 4

Equipment Tag Unit Operation

Transfer in
Agitation
Charge (Sodium Carbonate)
Charge (Water)
R-102
Batch Stoich. Reaction
Charge (Ethyl Acetate)
Batch Extraction (Phase Split)
Transfer Out
Transfer In
RV-104
Store

The reaction conversion was 60% based on Intermediate III as a reference

component. The excess solvent is removed in this case using Batch Extraction /

Phase Split for the separation of the Intermediate IV. The percentage of

component removal was specified in the model and the undesired components

were transferred into the receiving vessel (RV-104).

28
7.5 Methylation

CONH2 COOMe

N N

S S Cl
Cl

2-chlorphenyl-(6,7-dihydro-4h- (+) Methyl (2-chlorophenyl)-(6,7-dihydro-4H

thieno[3,2-c]pyrid-5yl) acetamide -thieno[3,2-c] pyrid-5-yl) acetate
Scheme 5. Reaction Mechanism for Methylation (Step 5)

Process Description for Step 5

Methanol (S-501) is pumped by (P-501) to the second reactor (R-101) containing

the solid product from the previous step and stirred. Once the product is

dissolved concentrated sulfuric acid (S-502) was added in portions via (P-502) to

the reaction mixture over a period of 1.5 hours. This reaction mixture is then

refluxed at 80 C for 26 hours using (C-101) to obtain desired conversion. Note,

this information was taken from the patent, and those experiments have not been

optimized. It is also applicable to the lab scale, but it is most likely different in the

pilot plant. After the reaction is complete the residual solvents are removed under

reduced pressure using (C-101). A solution of KOH (S-504) is pumped (P-504)

into the reactor and mixed for 0.5 hours. This solution is filtered (F-101) and the

filtrate is sent back to the reactor using a pump (P-506) and the solid waste from

the cake (S-506) is sent to waste treatment. The aqueous layer is separated in

the reactor and sent to waste treatment via (S-508). The remaining solvents are

removed under vacuum using (C-101), and the oily product is obtained and sent

back to (R-101) via (S-509).

29
Process Simulation for Step 5

The simulation of Step 5 requires two additional unit procedures.

Stirred reactor (R-101) in which the Intermediate II is produced

Nutsche Filter (F-101) in which separation of the Intermediate V occurs

The unit-operations in the sequence needed to simulate the Step 2 are shown in

Table 5.

Table 5. Unit Operations in Step 5

Equipment Tag Unit Operation

Transfer in
Agitation
Charge (MeOH)
Charge (Sulfuric Acid)
Charge (KOH)
R-101
Batch Stoich. Reaction
Batch Vaporization
Batch Cooling (Chilled Water)
Transfer Out
In Place Cleaning (CIP-1)
Transfer In
Charge (Acetone)
F-101
Cloth Filtration
In Place Cleaning (CIP-1)

The charge operations as well as the scheduling of the various operations were

performed in the similar manner as in the previous steps. The reaction extent

was 64% based on Intermediate IV as the reference component. The excess

solvent was removed using Batch Vaporization operation under reduced

30
pressure of 50 KPa. The reactor (R-101) is shared between several steps, and

this was also specified when scheduling the operations in various sections. The

In Place Cleaning operation was scheduled after the reactor content was

transferred to the filter (F-101). Methanol was chosen as a cleaning agent, and

the duration of cleaning was also specified.

7.6 Acid Addition

COOMe
COOMe

N
N
HSO4
S Cl
S Cl
(+) Methyl (2-chlorophenyl)-(6,7-dihydro-4H (+) Methyl (2-chlorophenyl)-(6,7-dihydro-4H
-thieno[3,2-c] pyrid-5-yl) acetate -thieno[3,2-c] pyrid-5-yl) acetatehydrogen sulfate salt
Scheme 6. Reaction Mechanism for Acid Addition (Step 6)

Process Description for Step 6

Ice cold acetone (S-601) is pumped by (P-602) and cooled by (RV-602) into the

reactor (R-101). Acetone is used to dissolve clopidogrel freebase made in the

last step. Concentrated sulfuric acid (S-602) is also pumped and cooled by (P-

603) and (RV-603). The reaction occurs at 40 C. After desired conversion is

obtained, the reaction mixture is cooled down to 5 C to enhance crystallization

of the CPG bisulfate. Water by itself is not a sufficient coolant to obtain this

temperature inside the reactor, and thus the mixture of methanol and water will

be used as a coolant for this reaction step. The reaction mixture (S-605) is sent

to a filter (F-101) to separate the solid product and the filter cake is washed twice

with acetone (S-604). The solid is sent to a vacuum rotary cone dryer (D-101) at

31
25 C where it is dried by contact with the hot surface of the dryer walls. The

acetone vapour is removed by the vacuum pump and condensed.

Process Simulation for Step 6

For this step the following unit procedures are needed to simulate the process

described above:

Stirred reactor (R-102) in which the final product is

Nutsche Filter (F-101) in which the extraction is carried out.

Double-Cone Dryer (D-101) in which the excess solvent it collected

The unit-operations in the sequence needed to simulate the Step 6 are shown in

Table 6.

32
 Table 6. Unit Operations for Step 6

Equipment Tag Unit Operation

Transfer in
Charge (Acetone)
Agitation
Batch Heating
Charge (Sulfuric Acid)
R-102
Batch Stoich. Reaction
Crystallization
Batch Cooling
Transfer Out
In Place Cleaning (CIP-1)
Transfer In
Cloth Filtration
F-101 Cake Wash (Acetone)
Transfer Out
In Place Cleaning (CIP-1)
Transfer In
D-101 Charge
Drying

The reactive crystallization occurs in Step 6, and this was still simulated by using

Batch Stoichiometric Reaction operation and Crystallization operation in series.

This was done in order to obtain the product first with the conversion of 100%,

and then to indicate the formation of solids. The crystal form component was

indicated, and the final temperature of 5 C. The cleaning of the reactor was

simulated using In Place Cleaning operation, so that the reactor is ready for the

next running the next batch. The excess of the solvent was vaporized under the

pressure of 50 KPa. The final product of was filtered using Nutshce filter unit

33
procedure. The cake thickness as well as the cake dryness had to be specified in

order to simulate this procedure. The cake was also washed with acetone.

Finally, the product was dried using Double Cone Dryer unit procedure. In the

real process the drying is performed under vacuum, but this unit procedure was

not available in SPD library.

34
8 Simulation Results

The material and energy balance, the scheduling information, as well as the

economic analysis for the entire process are presented in this section.

8.1 Material and Stream Balances

The software package track material balances across the entire process. The

solution of material and energy balances involves the calculation of the flowrate,

composition and thermodynamic state of all streams and equipment contents in

the flowsheet. The composition of the intermediates and outlet process streams

is calculated assuming that the information about the feed streams were entered

and initialized for all the unit operations. The advantage of SPD over other batch

software packages used for this project was that material balance results are

easily accessed by selecting the stream of interest in the process-flow diagram.

This opens a window with the detail description of the composition, flowrate and

density of the stream.

The generated report for material and stream balance contains the overall

material balance, the stream material balance, the material balance related to

separate sections, raw materials, and the equipment contents.

The overall process data and the raw material requirements are shown below in

Tables 7 and 8.

35
 Table 7. Overall Process Data

Table 8 depicts the raw material requirements for the entire process.

Table 8. Bulk Raw Materials Requirements (Entire Process)

8.2 Scheduling

The SPD perform process scheduling and generates the Gantt chart which is

shown below in Figure 7. Note that in this Gantt chart, only the main procedures

were shown, but there is an option of showing each operation within the specific

procedure.

36
 Figure 7. Gantt chart for Manufacturing Process of Clopidogrel Bisulfate

As seen from the Figure 6, the longest procedure is P-11 (Methylation Reaction)

in the reactor (R-101). This indicates that the main bottleneck of the entire

process is in this procedure, and that main optimization of the process should be

addressed to reduce the duration of this procedure.

It can be also seen that there is an overlap between some of the steps. This is

due to the cleaning in place operation of one reactor, while the reaction for the

following step is carried out in the other reactor. Figure 8 depicts this occurrence

between Step 2 and 3.

37
 R-101 is being cleaned
while R-102 is in
operation

Figure 8. Overlap between Step 2 and 3 due to CIP

39
9 Process Economics

As previously mentioned, the entire process was simulated using SuperPro

Designer which also allows for an extensive economic analysis. With the

successful compilation, the Total Capital Investment, Annual Operating Costs and

full Profitability Analysis were completed and shown below in more detail.

9.1 Fixed Capital Cost

The calculations for fixed capital cost were done using SPD. The purchase cost

is first factor for cost estimation. This can be evaluated either by using built-in

model in SPD which is working based on Power law correlation or entering

known value for all equipments. In this study, the purchase costs of all major

units are shown below in Table 9.

40
 Table 9. Purchase Costs (PC) of Major Units

The model was in the design mode, so the equipment was sized appropriately.

The purchase cost calculated by SuperPro Designer takes into account the

size of the vessel and material of construction. Each material of construction is

allocated a different factor which is multiplied by the base purchase cost. For

stainless steel (SS 316) the material factor is 1.0 and for a glass-lined vessel, like

the reactors, the material factor is 1.30.

Once the total purchase cost in calculated, the next step in the calculations

requires evaluating the Direct Costs (DC) associated with the purchase cost of

the unit. The Direct Costs (DC) are calculated as a percentage of the PC and

include things such as piping, instrumentation and so on.

Indirect Costs (IC) included construction and engineering costs, and were a

percentage of the total Direct Costs (DC). Other Costs (OC) were made up of the

contractors fee and contingency fees and were a percentage of the sum of the

41
Direct and Indirect Costs. Each of these costs and their associated percentage

are shown below in Table 3.

Once all of these costs were calculated the Direct Fixed Capital Costs (DFC) was

calculated using the following equation:

Direct Fixed Capital Costs = Direct Cost (DC) + Indirect Cost (IC) + Other Cost

(OC)

The total capital cost detail information for manufacturing CPG-BS project are

presented on Table 10. Once the software performed all above calculations, the

total capital investment for the proposed plant was found to be approximately $29

million based on 2006 prices.

42
 Table 10. Fixed Capital Estimate Summery

9.2 Annual Operating Costs

The economic estimation performed by SuperPro Designer also calculated the

associated operating costs of the proposed facility. Some of the main factors for

annual operating calculation are as follows:

1. Labor Costs: It is sum of the itemized estimate for operating labor on a

step-by-step process and additional labor defined on lumped time basis.

2. Laboratory/QC (Quality Control)/ QA (Quality Assurance): It is equal to

15% of total labor cost.

3. Raw materials

43
4. Utilities: It includes all the costs for electricity and heat transfer agents

(steam, cold/chilled water and NaCl brine).

5. Facility Dependant Cost: Main factors for this item are maintenance,

depreciation, insurance, local taxes and factory expenses.

6. Waste Treatment and disposal Cost: CPG-BS plant has organic and

aqueous wastes, solid disposals and emission and annual operating cost

for processing these material should be considered in this item.

Once this was done, the estimated operating costs for this plant was found to

be approximately $61 million (CAD) based on 2006 prices.

Table 11 presents the final calculation annual operation cost for

manufacturing CPG-BS project.

Table 11. Annual Operation Cost

44
9.3 Profitability Analysis

The final step in the economic analysis was to calculate the profitability of the

proposed plant. This was done using SuperPro Designer economic estimator

and simply took the total revenues generated by the product and subtracted the

costs associated with manufacturing the product. The Research and

Development (R&D) represents an important cost in any pharmaceutical industry

and this cost was fixed for all process steps. The gross margin of the project was

calculated 72.59%, and the payback period for this plant was found to be a little

over 1.38 years. Table 12 depicts profitability analysis for this project.

Table 12. Profitability Analysis for Manufacturing CPG-BS Plant

45
SPD has very flexible and comprehensive cost analysis tool and also it has
updated library for cost information. There is this possibility to choose SPD
suggested variable for cost analysis or change them to other value or correlation.
SPD is generating detail reports for total capital investment, annual operating
cost and profitability which are very helpful for tracking results and detail
analysis. Table 13 shows the executive summery of cost analysis for
manufacturing CPG-BS project.

Table 13. Executive summery of economic evaluation

46
10 Conclusion

The SuperPro Designer software package was successfully applied for design

and simulation of a batch process for production of Celopidegrol Bisulfate (an

anti-platelet drug). The reactor was designed with a high degree of flexibility to

accommodate for all reactions and crystallization of final product. The design

also included a rotary vacuum dryer to avoid contamination and loss of the drug

in the final stage of production. SuperPro Designer software package facilitated

the design and simulation of this process in different ways, some of which have

been summarized below.

- The performance of units can be evaluated individually and in the process,

which was extremely helpful in diagnosis of the problems

- Scheduling of various operations was a great help as a number of

reactions with different residence time must have been carried out in the

process. This was especially useful for evaluating different options with

respect to economics of the process (carrying out the process in one

reactor or two reactors in parallel)

- The errors during the design and simulation were explicitly explained by

the software in a very simple way. (i.e. the specific stream has to be

initialized first in order to perform the material balances)

- The Gantt chart provided by the software was also valuable in

determination of possible mistakes in the process scheduling

47
Economic analysis is one of the strongest points of the SPD software package.
The software accounts for operational, equipment, and overhead cost. It also has
an option of incorporating R&D cost, which is an important part of capital cost in
pharmaceutical industry. The economic analysis yielded the period of 1.38 years
for the return on the investment.

48
11 References

1. Silve, R.A., Preparation of (S)- Clopidogrel and related compounds , 2003,

(US patent: 6858734)

2. Koradia, V., Chawla, G., & Bansal, A.K. Qualitative and quantitative analysis

of clopidogrel bisulphate polymorphs. Acta Pharm, 54 (1), 2004, 193204

3. Evers, S.M., Struijs, J,N. & Ament, A,J. International comparison of stroke

cost studies Stroke 35 (6), 2004, 12091215.

4. Robless, P., Mikhailidis, D. P.& Stansby, G. Systematic review of antiplatelet

therapy for the prevention of myocardial infarction, stroke or vascular death in

patients with peripheral vascular disease. British Journal of Surgery 88 (6),

2001, 787-798

5. Harker, Boissel, Pilgrim, Gent. A randomized, blinded, trial of clopidogrel

versus aspirin in patients at risk of ischemic events (CAPRIE). The Lancet 348

(9038), 1996, 1329-1339.

6. Gerschutz, G.P. & Deepak, B.L.. The CURE trial: Using clopidogrel in acute

coronary syndromes without ST-segment elevation. Cleveland Clinic Journal of

Medicine 69 (5), 2003,377-385.

7. Petrides D. P., Koulouris A., Lagonikos P. T. The Role of Process Simulation

in Pharmaceutical Process Development and Product Commercialization

Pharmaceutical Engineering 22 (1), 2002

49
12 Appendix

50
51
52