Efficient "on water" organocatalytic protocol for the synthesis of optically pure warfarin anticoagulant

Article abstract of DOI:10.1039/c1gc15118e

The present development leads to an efficient protocol for the "on water" formation of one of the most widely used anticoagulants warfarin: the presented scalable and environmentally friendly organocatalytic approach affords the target drug in optically pure form.

Full text of DOI:10.1039/c1gc15118e

View Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Green Chemistry
Cite this: Green Chem., 2011, 13, 1155
www.rsc.org/greenchem
COMMUNICATION
Efficient “on water” organocatalytic protocol for the synthesis of optically
pure warfarin anticoagulant
Maria Rogozin´ska,a Anna Adamkiewiczb and Jacek Mlynarski*a,b
Received 30th January 2011, Accepted 21st March 2011
DOI: 10.1039/c1gc15118e
The present development leads to an efficient protocol
for the “on water” formation of one of the most widely
used anticoagulants warfarin: the presented scalable and
environmentally friendly organocatalytic approach affords
the target drug in optically pure form.
Scheme 1 Ideal green reaction “on water”.
involving suspensions of solids on water: perfect reaction of
two practically insoluble reactants at ambient temperature. This
system is close to the requirements of ideal green chemistry, and
it is noteworthy that such reactions are possible, although rare.5
As one of the most effective anticoagulants, warfarin
The design of straightforward and practical chemical syntheses
of drugs and fine chemicals that satisfy economic and ecological
criteria is a major challenge in industry nowadays. Ideally, a
sustainable synthesis should be safe, environmentally benign,
reasonable in terms of cost, and careful in the use of resources
and energy.1 An important goal for asymmetric catalysis is
still to develop new reactions that afford optically active
compounds from simple and easily available starting materials
by using environmentally friendly and possibly cheap catalysts
and solvents.
Among available methodologies asymmetric organic synthe-
sis using metal-free organic molecules as catalyst is broadly
recognised as promising tool which easily meet green chemistry
criteria. This approach, known as organocatalysis, has blos-
somed rapidly since the turn of the century.2 Organocatalysis
is now presented to be an enabling technology for the efficient
synthesis of drug and several biologically relevant molecules.3
Recent achievements in chemical efficiency have focused on
economic and ecological aspect of organocatalysis. From a
green chemistry perspective, the use of water instead of even
a small additive of organic solvent is preferred to decrease
environmental contamination. Thus since 2003, efforts have
been also made to develop organocatalysts for asymmetric
reactions in water.4
R
R
(Coumadinꢀ, Marevanꢀ) has been introduced for clinical use as
a racemate for more than half a century. Activity and metabolism
are markedly dissimilar for the two enantiomers, and the (S)-
form is proved to be more active than its mirror image.6
In 2003, Jørgensen reported the first one-step synthesis
of enantiomerically enriched warfarin from simple materials
catalysed by secondary amine catalyst.7 Further, primary8 and
secondary9 amino-acid based amines have been showed to be
good efficient in this asymmetric procedure. However, 20 mol%
catalyst loading was needed and the reaction time was long. In
2006, Chin demonstrated that application of 10 mol% of (R,R)-
diphenylethylenediamine in THF resulted in the formation of
(R)-warfarin in 94% yield after 48 h (ca. 80% ee).10 Recently,
efficiency of this catalyst was further improved by metal
salt additive.11 Again, reaction in THF delivered warfarin in
64% yield after 24 h with 5 mol% of amine and metal salt
combination. In all cases harmful organic solvents were used
as reaction medium, making any attempt at green chemistry
unsuccessful.12
In this communication, we present an example of the straight-
forward synthesis of chiral warfarin via Michael addition of
4-hydroxycumarin to benzylideneacetone on water using com-
mercially available amines (Scheme 2). In elaborated scalable
procedure purification using silica gel is also excluded from the
protocol.
First, by using 4-hydroxycoumarin (1) and benzylideneace-
tone (2), a series of amine catalysts were screened at room
temperature in homogeneous solutions including also aqueous
environment (Table 1).
Depending on the reactants’ nature, reactions can be per-
formed “in water” and “on water”. The latter protocol is far
more practical making separation of substrates/products and
water more efficient. Organic liquids that remained separated
from water in a clear organic phase were ideal reactants, but
solids could also be used, although being conceptually more
demanding. The synthetic reactions in Scheme 1 are cases
aInstitute of Organic Chemistry, Polish Academy of Sciences, Kasprzaka
44/52, 01-224, Warsaw, Poland
Initial examination indicated that structure of primary amine
was of great importance, and commercially available (S,S)-
diphenylethylenediamine (11) exhibited the highest yield and
bFaculty of Chemistry, Jagiellonian University, Ingardena 3, 30-060,
Krakow, Poland. E-mail: jacek.mlynarski@gmail.com; Fax: +48 12 663
2035
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 1155–1157 | 1155
©

View Online
Table 1 Catalyst screeninga
Table 2 Reaction on water accelerated by ultrasounda
Entry
Catalyst 11
Additive
% Yieldb
% eed
1
2
3
4
5
6
7
10 mol (%)
10 mol (%)
10 mol (%)
10 mol (%)
10 mol (%)
5 mol (%)
2 mol (%)
AcOH (10 mol%)
AcOH (20 mol%)
PhCOOH (10 mol%)
PhCOOH (20 mol%)
PhCOOH (20 mol%)
PhCOOH (10 mol%)
PhCOOH (4 mol%)
97
98
90
93
66
87
78
58
68
66
72
75c
68
70
Entry Cat. Additive (2 equiv.) Solvent
% Yieldb % eec
1
2
3
4
5
6
7
4
—
—
—
—
—
—
—
THF–H2O (9/1) 55
Rac.
5
THF–H2O (9/1) 30
THF–H2O (9/1) 34
THF–H2O (9/1) 56
THF–H2O (9/1) 22
THF–H2O (9/1) 22
THF–H2O (9/1) 40
24 (S)
20 (R)
44 (S)
18 (S)
20 (R)
30 (S)
62 (S)
55 (S)
62 (S)
86 (S)
84 (S)
70 (S)
6
7
a The reaction was performed by employing
1 (0.50 mmol), 2
8
(0.55 mmol), catalyst 11, and water (1 mL) at rt for 10 h in ultrasound
bath. b Isolated yield after silica gel chromatography. c Reaction time –
2 h. d Ee’s were determined by HPLC analysis on a chiral phase (Daicel
AD-H column).
9
10
11
11
11
11
11
11
8
9
—
—
THF
89
THF–H2O (9/1) 98
10
11
12
13
—
Water
THF
THF–H2O (9/1) 85
Water 30
8
92
AcOH
AcOH
AcOH
a The reaction was performed by employing
1 (0.50 mmol), 2
(0.55 mmol), catalysts (10 mol%), and solvent (1 mL) at rt for 20 h.
b Isolated yield after silica gel chromatography. c Ee’s were determined
by HPLC analysis on a chiral phase (Daicel AD-H column).
Scheme 2 Asymmetric warfarin synthesis.
enantioselectivity in both dry and aqueous THF solutions. This
is important to stress that the most promising catalyst 11 can
be obtained in both enantiomeric forms using reliable protocol
in multigram scale.13 Remarkably, with the use of acetic acid
additive, the reaction proceeded efficiently and afforded high
ee in both dry and aqueous solvents (entries 11 vs. 12). This
observation being contrary to previously published findings12
encouraged us to exclude organic solvents from the reaction
mixture.
Interestingly, by using pure water as a solvent the reaction
achieved good level of enantioselectivity (70% ee: 85%-S and
15%-R), but careful foraging of reactant was essential for
reasonable yield (entry 13). Reaction “on water” efficiency
depends on appropriate contact between reactants, and for
this reason intense mechanical stirring was applied. Further
extensive screening showed that the optimal reaction conditions
were only 2–5 mol% of the catalyst supported by acetic or better
benzoic acid when ultrasound bath was applied for the reaction
(Table 2).†
Scheme 3 Warfarin after reaction in ultrasound bath and HPLC of the
product purified using one crystallisation of crude product.
to 2 g with 2 mol% catalyst at rt in an open vessel, excellent
results were still obtained in 10 h (Table 3). To take advantage
of elaborated green protocol, solid product was separated from
water phase by filtration, and after one crystallisation from hex-
ane delivered optically pure warfarin.† This procedure excluded
Table 3 Reaction on water in 2 g scale accelerated by ultrasounda
Entry
Catalyst 11
Additive
% Yieldb
% eec
1
2
3
4
10 mol (%)
2 mol (%)
2 mol (%)
5 mol (%)
AcOH (20 mol%)
AcOH (4 mol%)
PhCOOH (4 mol%)
PhCOOH (10 mol%)
61
52
50
55
>99
>99
>99
>99
Sonication enables the rapid dispersion of solids on water
surface allowing better contact between water and reactants.
The use of ultrasound as a means of accelerating reactions is
an important technique and one which is rapidly developing for
green processes.14 In addition, when the reaction was scaled up
a The reaction was performed by employing
1 (5.00 mmol), 2
(5.50 mmol), catalyst 11, and water (10 mL) at rt for 10 h in ultrasound
bath. b Isolated yield after one crystalisation from hexane. c Ee’s were
determined by HPLC analysis on a chiral phase (Daicel AD-H column).
1156 | Green Chem., 2011, 13, 1155–1157
This journal is
The Royal Society of Chemistry 2011
©

View Online
use of expensive silica gel chromatography, and application of
small amount of hexane instead of chromatography eluents
seems to be promising from green chemistry perspective.
123.9, 123.8, 123.5, 123.0, 122.6, 116.6, 116.4, 116.1, 115.8, 115.5, 104.1,
101.1, 100.4, 98.9, 45.1, 42.5, 39.9, 35.3, 34.8, 34.1, 30.0, 28.1, 27.7,
26.9. Warfarin exists in a rapid equilibrium with a hemiketal forms and
therefore gives highly complicated and concentration-dependent NMR
spectra. Recorded spectra is in accordance with those recorded in ref.
7–11.
Rescaled procedure for the preparation of (S)-warfarin on water (Table 3):
A suspension of 4-hydroxycoumarin (1) (5 mmol), benzylideneacetone
(2) (5.5 mmol), (S,S)-diphenylethylenediamine (11) as a catalyst (2–
5 mol%), appropriate carboxylic acid (4–10 mol %) and water (10 mL)
were sonicated at rt for 10 h in ultrasound bath. The crude mixture was
filtered off from water and crystallised from hexane to give optically pure
warfarin (3) as a white solid (Table 3). Alternatively, the crude product
can be dissolved in warm EtOAc and precipitated with hexane.
Conclusions
In summary, we have described water compatible and scalable
protocol for the synthesis of (S)-warfarin by using primary
diamine as organocatalyst. By modifying the reaction condi-
tions, we were able to obtain the warfarin with an excellent
ee under mild conditions without application of silica gel
chromatography. Finally, the presented example showed that
further enhancements in the organocatalytic Michael reaction
may be discovered with the novel strategy based on the “solids
on water” green concept.
Project operated within the Foundation for Polish Science
TEAM Programme co-financed by the EU European Regional
Development Fund. Financial support from the Polish State
Committee for Scientific Research (KBN Grant N N204 093
135) is also gratefully acknowledged.
1 R. A. Sheldon, Green Chem., 2005, 7, 267.
2 A. Dondoni and A. Massi, Angew. Chem., Int. Ed., 2008, 47, 4638.
3 M. Christmann, Eur. J. Org. Chem., 2007, 2575.
4 (a) J. Paradowska, M. Stodulski and J. Mlynarski, Angew. Chem.,
Int. Ed., 2009, 48, 4288; (b) J. Mlynarski and J. Paradowska, Chem.
Soc. Rev., 2008, 37, 1502; (c) M. Gruttadauria, F. Giacalone and R.
Noto, Adv. Synth. Catal., 2009, 351, 33; (d) M. Raj and V. K. Singh,
Chem. Commun., 2009, 6687.
5 For the review on synthesis “on water” and “in water” see: (a) R. N.
Buttler and A. G. Coynote, Chem. Rev., 2010, 110, 6302, for
examples of reactions of solids “on water” see: Huisgen cycloaddition
reaction of 1,3-dipole with diphenylacetylene: (b) R. N. Butler,
A. G. Coyne and E. M. Moloney, Tetrahedron Lett., 2007, 48,
3501; (c) nucleophilic substitution of 1,4-quinone derivative: V. K.
Tandon and H. K. Maurya, Tetrahedron Lett., 2009; V. K. Tandon
and H. K. Maurya, Tetrahedron Lett., 2009, 50, 5896; (d) Wittig
reaction of aldehydes with stabilized phosphorus ylide: S. Tiwari and
A. Kumar, Chem. Commun., 2008, 4445.
6 (a) R. A. O’Reilly and N. Engl, J. Med., 1976, 295, 354; (b) L. B.
Wingard, R. A. O’Reilly and G. Levy, Clin. Pharmacol. Ther., 1978,
23, 212; (c) H. P. Rang, M. M. Dale, J. M. Ritter, P. Gardner,
Pharmacology, 4th ed., Churchill Livingstone, Philadelphia, 2001.
7 N. Halland, T. Hansen and K. A. Jørgensen, Angew. Chem., Int. Ed.,
2003, 42, 4955.
8 T. E. Kristensen, K. Vestli, F. K. Hansen and T. Hansen, Eur. J. Org.
Chem., 2009, 5185.
9 Z. Dong, L. Wang, X. Chen, X. Liu, L. Lin and X. Feng, Eur. J. Org.
Chem., 2009, 5192.
10 H. Kim, C. Yen, P. Preston and J. Chin, Org. Lett., 2006, 8, 5239.
11 H.-M. Yang, L. Li, K.-Z. Jiang, J.-X. Jiang, G.-Q. Lai and L.-W. Xu,
Tetrahedron, 2010, 66, 9708.
12 K. A. Jørgensen tested activity of imidazolidine catalyst in water and
observed formation of warfarin in 22% yield and 49% ee – ref. 7.
X. Feng, showed that dosage of water additive increased reaction
yield and enantioselectivity – ref. 9.
Notes and references
† Typical procedure for the preparation of (S)-warfarin on water (Table 2):
Suspension of 4-hydroxycoumarin (1) (81.0 mg, 0.5 mmol), benzyli-
deneacetone (2) (87.6 mg, 0.6 mmol), (S,S)-diphenylethylenediamine
(11) as a catalyst (0.05 mmol, 10 mol%), appropriate carboxylic acid
(0.1 mmol, 20 mol %) and water (1 mL) was sonicated at rt for
10 h in ultrasound bath. After extraction between EtOAc and water
crude product was purified by column chromatography on silica gel
(hexane–ethyl acetate 4 : 1) to give pure warfarin (3) as a white solid.
The enantiomeric excess was determined by HPLC analysis of the
purified product with an Daicel AD-H column (hexane/i-PrOH (4 : 1),
1.0 mL min-1, l = 254 nm): tR = 6.3 min (minor) and tS = 15.2 min
(major).
Data for sample with ee > 99% (S): [a]D = -10.7 (c 1.0, acetonitrile);
1H NMR (600 MHz, CDCl3): d (ppm) = 9.47 (0.3 H, s, OH, keto), 7.93
(0.3 H, d, J = 7.6, ArH), 7.88 (0.9 H, d, J = 7.8, ArH), 7.80 (1.1 H,
d, J = 7.6, ArH), 7.55 (1.0 H, t, J = 7.6, ArH), 7.47 (1.5 H, m, ArH),
7.34–7.18 (18.0 H, m, ArH), 4.69 (0.3 H, d, J = 10.0, CH, keto), 4.27
(1.0 H, dd, J = 6.2, 2.4, CH, ketal), 4.15 (1.1 H, dd, J = 11.2, 6.7, CH,
ketal), 3.85 (0.3 H, dd, J = 10.3, 19.5, CH2, keto), 3.38 (1.0 H, bs, OH,
ketal), 3.30 (0.3 H, d, J = 19.1, CH2, keto), 3.21 (1.1 H, bs, OH, ketal),
2.53 (1.0 H, dd, J = 14.3, 2.9, CH2, ketal), 2.46 (1.1 H, dd, J = 13.8,
6.8, CH2, ketal), 2.40 (1.0 H, dd, J = 14.0, 6.9, CH2, ketal), 2.28 (0.9
H, s, CH3, keto), 2.00 (1.1 H, dd, J = 13.5, 11.8, CH2, ketal), 1.70 (3.3
H, s, CH3, ketal), 1.66 (3.0 H, s, CH3, ketal); 13C (150 MHz, CDCl3)
d (ppm) = 162.1, 161.2, 159.6, 158.7, 152.9, 152.8, 143.1, 141.4, 131.9,
131.7, 131.5, 129.2, 128.6, 128.1, 127.9, 127.2, 127.0, 126.9, 126.6, 126.4,
13 (1R,2R)-(+)- and (1S,2S)-(-)-1,2-diphenyl-1,2-ethylenediamine were
prepared according to: S. Pikul and E. J. Corey, Org. Synth., 1992,
71, 22.
14 (a) P. Cintas and J.-L. Luche, Green Chem., 1999, 1, 115; (b) T. J.
Mason, J. P. Lorimer, Applied Sonochemistry, Wiley-VCH,
Weinheim, 2002.
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 1155–1157 | 1157
©