(S)-Metolachlor FromDreamto ProductionProcess · JOMC 90 (1975) 353 Rh / diop No reporton...

(S)-Metolachlor

From Dream to Production Process

Hans-Ulrich Blaser, SOLVIAS AG, Basel Switzerland

Uni Rostock, Gastvorlesung Asymmetrische Katalyse, 7.-8. Dez. 2007

Outline

� Background

The Molecule, the situation, possible approaches

� In Search of the Ideal Catalyst

�Moving in a Labyrinth (in the Fog)�

� The Technical Process

Ligand Scale-up, Imine Synthesis, Choice of Reactor,

MetolachlorThe Molecule

(R,S)

N

CH3

OCH3

CH3 CH3

O

Cl

� Herbicide for maize

� > 20�000 t/y

� Only (S)-enantiomers active

Ca. 35% less loadingfor enriched form

The Four Stereoisomers of Metolachlor

H. Moser, G. Rihs, H.P. Sauter 1982

N

CH3O

CH2Cl

O

N

CH3O

CH2Cl

O

R,1'S S,1'S

2 active stereoisomers

N

CH3O

CH2Cl

O

N

CH3O

CH2Cl

O

R,1'R S,1'R

2 inactive stereoisomers

Herbicidal Activity of Metolachlor Stereoisomers

H. Moser, G. Rihs, H.P. Sauter 1982

0%

20%

40%

60%

80%

100%

0 200 400 600 800 1000

application rate (g/ha)

R,1'S S,1'S rac-metolachlor S,1'R R,1'R

CH3

N

Cl

O

CH3

H3COCH3H

The History of rac-Metolachlor and (S)-Metolachlor

1970 Discovery of biological activity

1978 Full-scale plant >20�000 t/y

1982 Bioactivity of (S) enantiomers detected

1983 First attempts to make (S)-metolachlor

1985 Rh - cycphos (UBC Vancouver)

1987 Ir - diphosphine (F. Spindler; J.A. Osborn)

1993 Ir - ferrocenyl diphosphine catalysts

1993/4 Patents of rac. metolachlor expired

1995/6 Pilot results: e.e. 79%, ton 1�000�000, tof >200�000/h

16. Nov. 1996 First production batch

Metolachlor: The Problem

O

NH2

NOCH3H

ClCOCH2Cl

N

CH3O

CH2ClO

N

CH3O

CH2ClO

N

CH3O

CH2ClO

N

CH3O

CH2ClO

H2 - Pt/C

NAA

aR,1'S aS,1'S

aR,1'R aS,1'R

active stereoisomers

inactive stereoisomers

OCH3

Production process for racemate

?NH2

O

ClCOCH2Cl

+

+

OCH3

(S)-MetolachlorThe Challenge

Productionprocess

Enantioselectivity

Catalystproductivity (ton)

>1'000'000(70 recycles)

Space-time yield very high

-

Catalyst activity(tof)

>4000/hat 50°C, 5 bar

Minimumrequirements

ee >80%

>50'000(s/c ratio)

high

>10'000/h

100

80

60

40

20

0

Moving in the "ee � ton" Space

10 100 1000 10�000 100�000 1�000�000

ee >80%

ton > 50�0000

20

40

60

80

100

10 100 1'000 10'000 100'000 1'000'000

100

80

60

40

20

0

Moving in the "ee � ton" SpaceA Labyrinth!

10 100 1000 10�000 100�000 1�000�000

0

20

40

60

80

100

10 100 1'000 10'000 100'000 1'000'000

ton > 50�000

ee >80%

Development Phases for EPC Synthesis

Phase 1: Design and assessment of synthetic routes

Phase 2: Demonstrating chemical feasibility

Phase 4: Optimizing the over-all process

Phase 3: Optimizing the key (catalytic) reaction(s)

N

OCH3

COCH2Cl N

CH3O

COCH2Cl+ H2or

?

chiral cat

Hydrogenation / substitution

Hydrogenation of enamide (isomers) Hydrogenation of imine (isomers)

Routes to (S)-Metolachlor

OH

OMe

NH2

+chiral cat

?

N

CH3Ochiral cat

+ H2

?

OTs

OMe

O

OMe

NH R

+ 2. T sX

R = H or COCH2Cl

?1. H2, chiral cat

?

Direct alkylation

(S)-MetolachlorRoute Assessment

route catalytic step

enamide close analogy ee >90%

substitution weak analogy ee >80%

imine weak analogy ee <30%

direct alkylation

no precedent

(S)-Metolachlor:Route Assessment

route catalytic step other steps

enamide close analogy ee >90%

enamide synthesis difficult

substitution weak analogy ee >80%

substitution difficult

imine weak analogy ee <30%

as in current process

direct alkylation

no precedent as in current process

(S)-Metolachlor:Route Assessment

route catalytic step other steps cost (ecology) priority

enamide close analogy ee >90%

enamide synthesis difficult

high (medium)

1

substitution weak analogy ee >80%

substitution difficult

high (bad)

2

imine weak analogy ee <30%

as in current process

medium (good)

3

direct alkylation

no precedent as in current process

low (very good)

4

Development Phases for EPC Synthesis

Phase 1: Design and assessment of synthetic routes

Phase 2: Demonstrating chemical feasibility

Phase 4: Optimizing the over-all process

Phase 3: Optimizing the key (catalytic) reaction(s)

The Enamide Route

H.U. Blaser, F. Spindler 1983

The Analogy: L-DopaRh-dipamp

(Monsanto)

The Result: (All) available Rh/P^P(up to 20 bar / 50°C)

ee 96%ton 10�000

NO activity at all!!

N COCH2Cl

MeO

N COCH2Cl

MeO

N COCH2Cl

OMe

NCOCH2Cl

MeO

oror

COOH

N

MeOMeO

H

O

From Dream to ProcessImportant Milestones (1983)

100

80

60

40

20

010 100 1000 10�000 100�000 1�000�000

ee >80%

ton > 50�000enamide

40

60

80

100

100 1'000 10'000 100'000 1'000'000

Substitution RouteHeterogeneous Hydrogenation

H.U. Blaser, H.P. Jalett 1983

The Analogy: Pt/Al2O3-Cinchonidine

(Orito, 1979) ee >85%

O

O

OCH3

activity okee 12%!!

The Result: Pt/Al2O3-Cinchonidine

O

OCH3

ee 0%!!N

OCH3

From Dream to ProcessImportant Milestones (1983)

100

80

60

40

20

010 100 1000 10�000 100�000 1�000�000

ee >80%

ton > 50�000Pt/cinch

0

20

40

60

80

100

10 100 1'000 10'000 100'000 1'000'000

enamid

Imine Hydrogenation

N

CH3O

NH

CH3O

???????

Enantioselective C=N ReductionState of the Art Around 1982

0

20

40

60

80

100

1940 1950 1960 1970 1980

ee (%)

N

HydrogenationLevi et al.CC (1975) 6

HydrosilylationKagan et al.JOMC 90 (1975) 353

Rh / diop

No report on N-aryl imine hydrogenation

mostly heterogeneous catalysts

Industry�s Approaches for Solving Difficult Problems

� UBC Vancouver (Cullen, Fryzuk, James, Kutney et al.)Search for a catalyst� ULP Strasbourg (J.A. Osborn)Investigate Ir-diphosphine catalysts (active species, deactivation behavior)

Do-it-yourself Outsource to a specialized department Outsource to a specialized company (Solvias!) Collaboration with universities

UBC: Rh-CycphosA First Success

Cullen, Fryzuk, James, Kutney, et al. UBC Vancouver 1984-89

D M A-im ine

NO

NOH

H

H2

R h-PP

s/c Temperature t Conv. tof ee[°C] [hrs.] [%] [h-1] [%]

100 20 44 99 2.3 53

100 -10 20 100 5 69

1000 -10 168 67 4 69

100 -25 70 100 1.4 73

PPh2

Ph2P

(R)-cycphos

From Dream to ProcessImportant Milestones (1985)

Rh/cycphos

100

80

60

40

20

010 100 1000 10�000 100�000 1�000�000

ee >80%

ton > 50�000enamid Pt/cinch

0

20

40

60

80

100

10 100 1'000 10'000 100'000 1'000'000

New Idea: Ir - P^P Complexes

F. Spindler 1986

Ph2P PPh2 N

PPh2

CH2PPh2

COO-t-Bu

P POCH3CH3O

Ph Ph

CH3

OHH

PPh2

PPh2

Fe

PP

O

OH

H

R

R

R'

R'

R

R

bdpp (skewphos)dipamp bppmbppfohsubstituted diop

2

2

Ligand tof (4h) tof (24h) eediop 165 32 61subst-diop 89-146 28-32 56-61bppm ( tBuOMe) 6 79bppm (MeOH) 30 6dipamp 19 7bppfoh 21 28bdpp 114 26 78pN(Me)2-bdpp 31 rac

10-20 °C, 30-80 bar, conversion: 17-96%

Why Iridium?

[Ir(cod)(py)(Pcy3)]PF6: Extremely activecatalysts for olefin hydrogenation

Olefin maximum tof (1/h)

6400

8300

4000

Drawback: Very fast deactivation via dimerization

R. Crabtree et al. J. Organomet. Chem. 141 (1977) 205

1997: Chiral Iridium � PN Catalysts: Effective for C=C Hydrogenation

Review: A. Pfaltz, W.J. Drury, Proc. Nat. Acad. Sci. USA 101 (2004) 5723

R2

R1

R'R3

P N

O

R3R2

R1N

OO

R1 R1

R3

R2

PPh2

R2

R1

R'R3

Ir+ / PN / X-

A. Pfaltz

ee up to >99%ton >5000

tof 1250 h-1

Iridium: Best Results afterOptimization

N

CH3O

NH

CH3O

Ir - P^P, iodidePPh2

PPh2

O

O

CH2PPh2

CH2PPh2

bdpp diop

bdpp ee 84% ton 100 (at 0°C)

diop ee 52% ton 10'000 (at 30°C)

F. Spindler, B. Pugin 1986

PROBLEM

CATALYSTDEACTIVATION

(Dimerization??)

From Dream to ProcessImportant Milestones

Rh/cycphos

100

80

60

40

20

010 100 1000 10�000 100�000 1�000�000

ee >80%

ton > 50�000enamid Pt/cinch

Ir/diop

Ir/bdpp

0

20

40

60

80

100

10 100 1'000 10'000 100'000 1'000'000

Fighting Catalyst Deactivation

Understanding

Nature of active species: Collaboration with JAO

Stabilizing Additives

Avoid dimerization by complexation

Immobilization

Avoid dimerization by �site-isolation�

Fighting DeactivationImmobilization of Ir - bppm

B. Pugin 1988

0

20

40

60

80

100

120

0 3 6 9 12 15 18raction time [h]

imin

e [%

]

soluble catalyst; ee = 38%

NP

P

O

OIr

HN

P

P

O

OIr

H

inactive dimer???

Fighting DeactivationImmobilization of Ir - bppm

0

50

100

150

200

250

300

350

0 0.05 0.1 0.15 0.2Loading (mmol Ir/g)

tof

0

10

20

30

40

50

60

ee

B. Pugin 1988

NP

P

O

OIr

HN

P

P

O

OIr

H

inactive dimer

N

P P

O N

SiO

O O

Ir

N

P P

O N

SiO

O O

Ir

dilution

Fighting DeactivationHomogeneous Ir � bppm Catalyst

B. Pugin 1988

0

20

40

60

80

100

120

0 3 6 9 12 15 18raction time [h]

imin

e [%

]

soluble catalyst; ee = 38%

Immob. catalyst; ee >50%

From Dream to ProcessImportant Milestones

Rh/cycphos

100

80

60

40

20

010 100 1000 10�000 100�000 1�000�000

ee >80%

ton > 50�000enamid Pt/cinch

Ir/diop

Ir/bdpp

0

20

40

60

80

100

10 100 1'000 10'000 100'000 1'000'000

Ir/bppmimmob

Intermezzo 1988-1992

1988: Project stopped by Agro Division

A. Togni: Ligands for Au-Aldol reaction

Enlargement of ligand library

Systematic immobilization studies

Technical ligands syntheses

Various process developments

A. TogniLigands Studies: Josiphos

A. Togni, F. Spindler, B. Pugin

N(CH3)2

Fe

N(CH3)2

PR 2Fe

PR'2

PR 2Fea) BuLi

b) ClPR2

HPR'2

AcOH

OAc

PPh2

X

SAc

PPh2

PPh2

NRR'

PPh2

PPh2

PCy2

PPh2

H

Fe

X = H, PPh 2

Fe

Fe

Fefirst Josiphos

ligands for Au-Aldol

Modular, tunable ligands

The Final BreakthroughIr � Xyliphos / AcOH / Iodide

Best results (laboratory)

R´ = p-tBu-phenylee 87% low tof at -15°C

R´ = 3,5-xylylee 76-80% ton 1�000�000; tof ca. 30'000 h-1

F. Spindler, H.P. Jalett, H.P. Buser 1992

BUT:only on presenceof acid AND iodide!!

PR'2

PPh2Fe

Ir - Xyliphos: Effect of Solvent

Solvent t (100% )(h)

initialrate

ee

thf 7 0.3 74CH 2Cl2 4 0.3 74(CH 3)3CO CH 3 12 0.2 75acetone 6 0.4 73to luene 12 0.4 73i-PrO H 0.75 1.1 79t-BuO H 1 1.0 77CH 3CO O Et 8 0.7 72none 10 0.3 73CH 3CO O H 0.5 1.5 79R eaction conditions: MEA-Im ine; s/c : 800; 150 mg TBAI; solvent: 2 m l; 25 bar H 2; 30°C .

H.P. Jalett, F. SpindlerHH69

Why Acetic Acid

solvent ee EtOH 82%

toluene 87% acetic acid 92-94%

R

H

COOEt

OH

R COOEt

OPt/Al2O3 - cinchona

+ H2

R = CH3 and PhCH2CH2 R-hydroxyester

H.P. Jalett, H.U. Blaser, Wiehl 1991

Ir - XyliphosEffect of AcOH and I-

General acid effect:- CF3COOH- H2SO4

CH3

PH

P(C6H5)2 Fe 2

Xyliphos

H.P. Jalett, F. Spindler 1993

------

---AcOH

Iodide---

IodideAcOH

0

200

400

600

800

1000

1200

1400

1600

1800

0

20

40

60

80TOF ee

100

80

60

40

20

0

enamid

From Dream to ProcessImportant Milestones

Ir/xyliphosAcid / iodide

Ir/PPF-P

Ir/bppmimmob

Ir/bdpp

Ir/diopRh/cycphos

10 100 1000 10�000 100�000 1�000�000

ee (%)

tonPt/cinch

ee 80%ton >50�000tof >10�000/h

0

20

40

60

80

100

10 100 1'000 10'000 100'000 1'000'000

Reductive Alkylation of MEAIr-xyliphos, AcOH: Solvent Effect

NH2

NO

H

H

OO

H2

Ir-PP+

MEA MOA S-NAA

Solvent Time Rate Conv. ee[h] [mmol/min] [%] [%]

none 22 0.5 87 76 2 phases

Cyclohex (10 ml) 21 0.9 92 77 2 phases

EtOH (10 ml) 18 0.3 24 15 1 phaseReaction conditions: 0.1 mol MEA; 0.12 mol MOA (dry); AcOH: 2.5 ml; 20 mg TBAI; 80 bar H2;

50°C.

s/c 10'000

H.U. Blaser, H.P. Jalett

Functionalization of Xyliphos

NMe2

Fe

NMe2

PPh2

Si

Cl

Fe

Pxyl2PPh2

Si

Cl

Fe

Immobilization

1) BuLiBuLi /TMEDA

2) mixture of

Si ClCl

andCl-PPh2

Pxyl2PPh2

Si

N

Fe

H-Pxyl2

(53%)

(60-70%)

(90%)

Gabriel

xyl=

B. Pugin, H. Landert 1995

homogeneous heterogeneous

Metolachlor: Recycling??

free extractable immobilized silicagel

immobilized polystyrene

P(xyl)2

PPh2Fe

P(xyl)2

PPh2

Si

COO-

-OOC

Fe

Silicagel

P(xyl)2

PPh2

Si NH

ONH

SiO OO

Fe

P(xyl)2

PPh2

NH O

NHNH

O

NH

Fe

Polystyrene

S/C 50'000 120'000 120'000 50'000 120'000 50'000 time 1h 2.1h 3h 8h 10h 30h ee 79% 80% 79% 78% 78% 74% separation distillation extraction

> 90% filtration

95% filtration

95%

B. Pugin, H. Landert, F. Spindler, H.U. Blaser, Adv. Synth. Catal. 344 (2002) 974

Hydrogenation of ImineBest Ir / Xyliphos Systems

F. Spindler, B. Pugin, H.U. Blaser, H.P. Jalett

ImmobilizedHomogeneous Reductivealkylation

No acid

0

2'000

4'000

6'000

8'000

10'000

12'000

14'000

tof

0

20

40

60

80

100

ee (%

)

s/c 100'000s/c 10'000

s/c 2'000'000

AcOH

Development Phases Metolachlor Imine Route

Phase 1: Assessing synthetic routes

Phase 2: Demonstrating chemical feasibility

Phase 4: Optimizing the over-all process

Phase 3: Optimizing the key (catalytic) reaction(s)

Process Optimization

Ligand fine tuning

Optimization of reaction conditions

Strategy for the development of the over-all process

The Production of the MEA Imine in the Required Quality

Technical Ligand Synthesis

Choice of Reactor Technology

Scale up

Work-up

Separation of the Catalyst from the Product

Fine Tuning(S)-Metolachlor Process

NH

CH3O

N

CH3OMEA imine (S) -NAA

Ir - PP

Catalyst: [Ir(COD)Cl]2, NaI, H2SO4, (R)-(S)-R2PF-PR�2; p(H2), 80 bar, 50°C

ee %tontof [h-1]

P

P

HCH3Fe

792�000�000>400�000

P

P

HCH3

F3C

F3C

Fe

82800400

P

P

HCH3Fe

875�000

80

P

P

HCH3

N

N

Fe

83100�00028�000

F. Spindler

Ligand Finetuning: N Substituents at Xylyl Group

F. Spindler, H.P. Buser 1994

CH3

CH3

e.e.: 76%rate: 26 mmol/min

CH3

CH3

N

e.e.: 83%rate: 1 mmol/min

CH3

CH3

N

e.e.: 69%rate: 10 mmol/min

N

CH3

CH3

e.e.: 76%rate: 2 mmol/min

CH 3

CH3

N

e.e.: 82%rate: 0.5 mmol/min

CH3

CH3

N

e.e.: 80%rate: 25 mmol/min

CH3

PR2

H

P(C6H5)2Fe

Reaction ConditionsEffect of H2 Pressure

0

2

4

6

8

10

12

14

0 20 40 60 80 100 120 140

Pressure (bars)

r max

(mm

ol H

2/m

in)

ee: 75 - 76%

F. Spindler, H.P. Jalett 1994

Development Phases for EPC Synthesis

Phase 1: Design and assessment of synthetic routes

Phase 2: Demonstrating chemical feasibility

Phase 4: Optimizing the over-all process

Phase 3: Optimizing the key (catalytic) reaction(s)

Process Development

Ligand fine tuning

Optimization of reaction conditions

Strategy for the development of the over-all process

The Production of the MEA Imine in the Required Quality

Technical Ligand Synthesis

Choice of Reactor Technology

Scale up

Work-up

Separation of the Catalyst from the Product

Imine Production

NH2 N

CH3O

O

OMe + H2O+

H+

Simple chemistry

BUT

Scale up difficult due to thermal instability of the imine Fast removal of water neccessary

Significant catalyst deactivation High purity required

Effect of MEA Concentration

Ir/xyliphos, Acetic Acid, TBAI

F. Spindler, H.P. Jalett

0

1

2

3

4

5

6

7

0 1 2 3 4 5MEA (%)

inita

l rat

e (m

mol

H2/

min

)

t(100%): 18 h

t(100%): 0.5 hNH2N

O

MEA imine

MEA

ee not affected!

Imine Production: Water Separators

Process Development

Ligand fine tuning

Optimization of reaction conditions

Strategy for the development of the over-all process

The Production of the MEA Imine in the Required Quality

Technical Ligand Synthesis

Choice of Reactor Technology

Scale up

Work-up

Separation of the Catalyst from the Product

Technical Ligand Synthesis

OO

Fe

OH

Fe

Fe PPh2

NMe2

Fe PPh2

PH

FeO

Fe

OH

Fe

NMe2

Fe

+

2

enzymatickinetic

resolution

Ugi amine

30 Kg scale unproblematic

Process Development

Ligand fine tuning

Optimization of reaction conditions

Strategy for the development of the over-all process

The Production of the MEA Imine in the Required Quality

Technical Ligand Synthesis

Choice of Reactor Technology

Scale up

Work-up

Separation of the Catalyst from the Product

Choice of Reactor Technology

Issues

Fast exothermic reaction -> heat removal important

High pressure (80 bar) -> high investment

Sensitive catalyst -> handling and loading

Alternatives

Loop reactor

Stirred tank reactor

Coice of Reactor

pump

stirred tank loop reactor

Coice of Reactor

pump

cooling

Process Development

Ligand fine tuning

Optimization of reaction conditions

Strategy for the development of the over-all process

The Production of the MEA Imine in the Required Quality

Technical Ligand Synthesis

Choice of Reactor Technology

Scale up

Work-up

Separation of the Catalyst from the Product

16. November 1996Happy End: First Production Batch

NO

NOH

H

amount mol

MEA-imine 10'000 kg 48'700[Ir(COD)Cl]2 34 g 0.05Ligand 68 g 0.11NaI 92.5 g 0.6H2SO4 250 g 0.5

N. Pericles, R. Hanreich

CH3

Pxyl2

H

PPh2Fe

reaction time 2h

conversion 99.6%

ee 79%

100

80

60

40

20

0

enamid

From Dream to ProcessImportant Milestones

Ir/xyliphosAcid / iodide

Ir/PPF-P

Ir/bppmimmob

Ir/bdpp

Ir/diopRh/cycphos

10 100 1000 10�000 100�000 1�000�000

ee (%)

tonPt/cinch

ee 80%ton >50�000tof >10�000/h

0

20

40

60

80

100

10 100 1'000 10'000 100'000 1'000'000

Some Lessons

Enantioselectivity is not always the major problem ton and tof can be just as critical!!

Screening capabilities are crucial Parallel screening equipment

NO SUCCESS WITHOUT TOP EXPERTISE

Success often depends on availability of ligands Modular ligands are an optimal solution Solvias Ligand Kit

The Key Players

B. PuginF. Spindler H.P. Jalett

Think catalytic!