Fe3O4?l-Proline/Pd nanocomposite for one-pot tandem catalytic synthesis of (±)-warfarin from benzyl alcohol: Synergistic action of organocatalyst and transition metal catalyst

Article abstract of DOI:10.1039/c9cy01497g

One-pot synthesis of (±)-warfarin, an anticoagulant, has been achieved from benzyl alcohol in a 'green way' by using a multicomponent catalyst. For the purpose, l-proline capped Fe₃O₄ nanoparticles (Fe₃O₄?l-proline NPs) were synthesized and metallic palladium was loaded on its surface (Fe₃O₄?l-proline/Pd NCs). The morphology, particle size and shape were studied by using FESEM and TEM analysis. The Pd present on the surface was responsible for oxidation of benzyl alcohol and its derivatives to the corresponding aldehyde in situ. This in turn, condensed with acetone to form the aldol condensation product, benzylideneacetone, at 70 °C due to the presence of the l-proline organocatalyst on the surface of Fe₃O₄ NPs. Later, 4-hydroxycoumarin was introduced to condense with in situ generated benzylideneacetone by a Michael addition to form the target product (±)-warfarin. It was established that benzyl alcohol can be converted into the final product, (±)-warfarin, with an overall 35% yield within 5 days in a single-pot process. This process requires a rise in temperature in stages to a maximum of 100 °C and 1 atm pressure of dioxygen gas. An important aspect of the developed process is the avoidance of loss of costly Pd by leaching and catalyst recovery by the use of a magnetic field. The use of a solvent like PEG-400 makes the process green in a true sense. The interaction of l-proline with Fe₃O₄ NPs and the presence of Pd on the surface were confirmed by the FTIR and XRD patterns, respectively. The present study hereby suggests a combined 3-step mechanism for the production of the target product warfarin. Pilot-scale one-pot production of (±)-warfarin was carried out and a flow diagram with various unit processes is presented.

Full text of DOI:10.1039/c9cy01497g

Catalysis
Science &
Technology
View Article Online
View Journal | View Issue
PAPER
Fe3O4@L-Proline/Pd nanocomposite for one-pot
tandem catalytic synthesis of ( )-warfarin from
benzyl alcohol: synergistic action of
Cite this: Catal. Sci. Technol., 2019,
9, 6517
organocatalyst and transition metal catalyst†‡
*
Sanjiv O. Tomer
and Hemant P. Soni
One-pot synthesis of ( )-warfarin, an anticoagulant, has been achieved from benzyl alcohol in a ‘green
way’ by using a multicomponent catalyst. For the purpose, L-proline capped Fe3O4 nanoparticles
(Fe3O4@L-proline NPs) were synthesized and metallic palladium was loaded on its surface (Fe3O4@L-
proline/Pd NCs). The morphology, particle size and shape were studied by using FESEM and TEM analysis.
The Pd present on the surface was responsible for oxidation of benzyl alcohol and its derivatives to the
corresponding aldehyde in situ. This in turn, condensed with acetone to form the aldol condensation
product, benzylideneacetone, at 70 °C due to the presence of the L-proline organocatalyst on the surface
of Fe3O4 NPs. Later, 4-hydroxycoumarin was introduced to condense with in situ generated
benzylideneacetone by a Michael addition to form the target product ( )-warfarin. It was established that
benzyl alcohol can be converted into the final product, ( )-warfarin, with an overall 35% yield within 5 days
in a single-pot process. This process requires a rise in temperature in stages to a maximum of 100 °C and
1 atm pressure of dioxygen gas. An important aspect of the developed process is the avoidance of loss of
costly Pd by leaching and catalyst recovery by the use of a magnetic field. The use of a solvent like PEG-
400 makes the process green in a true sense. The interaction of L-proline with Fe3O4 NPs and the
presence of Pd on the surface were confirmed by the FTIR and XRD patterns, respectively. The present
study hereby suggests a combined 3-step mechanism for the production of the target product warfarin.
Pilot-scale one-pot production of ( )-warfarin was carried out and a flow diagram with various unit
processes is presented.
Received 27th July 2019,
Accepted 12th October 2019
DOI: 10.1039/c9cy01497g
rsc.li/catalysis
of one of the enantiomers.1–4 The S-enantiomer is shown to
be 2–5 times more active than the R-enantiomer. Moreover, it
1. Introduction
Warfarin (Coumadin® or Marevan®) is an anticoagulant drug
most commonly prescribed by physicians for the prevention
of thrombosis and embolism. It functions by inhibiting the
vitamin K-mediated biosynthetic pathway leading to the
formation of blood clotting proteins. Most commonly, it is
synthesized by Michael addition of 4-hydroxycoumarin to
has a different metabolic pathway in humans and also has a
shorter half-life than the R-enantiomer (21–40 h for S- and
37–89
h
for R-warfarin).5–7 However, in general, it is
benzylideneacetone in the presence of
a base. Various
organocatalysts having primary amines, amino acid
functionalities and natural cinchona-derived chiral amines
and diamines are reported to favor the asymmetric synthesis
Department of Chemistry, Faculty of Science, The Maharaja Sayajirao University
of Baroda, Vadodara-390 002, Gujarat, India. E-mail: drhpsoni@yahoo.co.in;
Tel: +91 0265 2795552
† This work is dedicated to Dr. D. P. Bharambe, a humble teacher and mentor,
on the occasion of his 62nd birthday and retirement.
‡ Electronic supplementary information (ESI) available: XRD patterns, TEM
images with SAED pattern, FTIR spectra of all synthesized materials, synthesis
scheme, combined reaction mechanism, 1H NMR and 13C NMR are provided in
the ESI. See DOI: 10.1039/c9cy01497g
Scheme 1 Planning for one-pot synthesis of ( )-warfarin from benzyl
alcohol through 3-way catalysis by using a single nanocomposite
catalyst having multiple components (I, II and III).
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol., 2019, 9, 6517–6531 | 6517

View Article Online
Paper
Catalysis Science & Technology
prescribed in the racemate form1,5 and its dosage is adjusted
on an ‘individual basis’ considering the patient's disease
history. Herein, we have attempted to synthesize ( )-warfarin
from scratch in a one-pot synthesis by 3-way catalysis, that is,
by (1) oxidizing benzyl alcohol to aldehyde and then (ii)
condensing the latter with acetone directly in an aldol fashion
leading to benzylideneacetone; (iii) it is then condensed with
4-hydroxycoumarin resulting in the target racemic product
(Scheme 1) by developing a catalytic nanocomposite. A
properly designed nanocomposite material can do its
and solvent immobilization techniques.26 These catalytic NCs
were evaluated for solvent-less oxidation of benzyl alcohol to
benzaldehyde in the liquid phase by using 35% H2O2 as an
oxidizing agent. They were able to achieve 88% selectivity for
the target oxidation in the presence of Au–Pd/C NCs prepared
by the impregnation method. Enache et al. developed a
Au@Pd core–shell supported on a TiO2 matrix for oxidation
of a 1° alcohol to aldehyde with O2 as the oxidant at 373 K.
They obtained a very high TOF (270 000 per h) from the
process.27 However, it is too costly and commercially
unviable. It is established that the recovery of Pd becomes
efficient when it is anchored/doped/supported on a solid
matrix which is magnetic or superparamagnetic in nature.
Zamani et al. reported a Pd/Fe3O4/amino acid nanocomposite.
They obtained more than 99% selectivity for benzyl alcohol
to benzaldehyde under solvent-free conditions.28 Wang et al.
function
effectively
and
efficiently
ranging
from
hydrodesulfurization of liquid fuels, photocatalysis, and
catalysis in organic transformations to magnetic resonance
imaging and photothermal ablation therapy.8–12
Controlled oxidation of benzyl alcohol to benzaldehyde is
very well reported in the literature.13,14 Many oxidizing agents
are explored for the purpose. For example, a stoichiometric
quantity of CrIJVI) (John's reagent, Collins reagent, PCC, PDC
etc.), MnIJVII) (aqueous, acidic or basic solution of KMnO4),
OsIJVIII) (OsO4), and Ru(VI, VIII) (RuO4) metal-based reagents
are all used for the purpose.15 Attempts have also been made
to use H2O2 or O2 as an oxidant for a neat, clean and atom-
efficient oxidation process.16–20 These oxidants work in the
presence of early transition metals with d0 configuration like
MoIJVI), WIJVI), TiIJV) or PdIJII) complexes. However, these
methods involve chlorinated solvents and generate metallic
salts as solid waste that are toxic and pollute the
environment. Oxidation methods mediated by TEMPO
(2,2,6,6-tetramethylpiperidine-1-yloxyl) and related stable
nitroxide radicals (Anelli oxidation)21 have also been
developed. However, these methods contain bleach (NaOCl)
and HOCl which are generated in situ that harm the
environment.22 Hence, the development of neat, clean and
atom efficient oxidation processes are the need of the
day.23,24 From the mechanistic points of view, the above
oxidizing agents donate an O atom, an intrinsic part of their
molecular structure, to the substrate and oxidize it. However,
the same function can be carried out by using pure O2, air or
peroxides (H2O2 is desirable) as an oxidant for ‘green’
oxidation processes. However, molecular oxygen in the
ground state (triplet state) is ‘inactive’ and needs to be
activated to act as an oxidizing agent.25 This requires early
transition metals with d0 configuration (as mentioned above)
and their complexes with sterically hindered ligands like
PPh3 mediated catalysis. For example, R. A. Sheldon's group
prepared Pd supported on Al2O3 and TiO2 by
a wet
impregnation method and used it as a catalyst for solvent-
free selective oxidation of benzyl alcohol by molecular
oxygen. The developed catalyst showed high activity and
selectivity for benzaldehyde on pretreatment like
calcination.29
Drawing inspiration from these studies and considering
the need for the recovery of the precious metal catalyst
without any loss, it was planned to load the metallic Pd on
the surface of magnetic Fe3O4 nanoparticles (NPs) in the
present study.
Generally, aldol condensation requires
a base (dil.
NaOH) catalyst and activation of substrates to react in a
desired way to form the C–C bond.30,31 It takes place in
two steps, namely, aldolization and dehydration. Practically,
with a stronger base, high temperature and longer reaction
time, the aldol product can become dehydrated to form an
α,β-unsaturated carbonyl compound. The development of a
direct catalytic asymmetric aldol reaction (limited to the
first step) under mild conditions in the presence of
L-proline like amino acids as organocatalyst,32 aldolase
enzyme33 or catalytic antibodies34 in a ‘green solvent’ like
water is a subject of research and reported very well in the
literature. For the present, required strategy for the one-pot
synthesis of warfarin, an aqueous solution of base cannot
be used due to its chance of attacking the Fe3O4 host
matrix. List, Lerner and Barbas reported for the first time
an enamine-based organocatalyst for asymmetric aldol
reaction.35 Encouraged by this, a strategy was devised such
that L-proline can be loaded onto the surface of Fe3O4
either by using a carboxylic acid group as a ‘handle’ or by
developed
a
water-soluble PdIJII) bathophenanthroline
complex for selective aerobic oxidation of a wide range of
alcohols to aldehydes, ketones and carboxylic acids in a
biphasic water–alcohol system. The use of water as a solvent
and air as the oxidant made the process economical and
environment friendly.20 However, as the catalyst is water
soluble, there is a chance of loss of precious Pd metal during
catalyst recovery. A number of efforts have been made on Au-
based materials for alcohol oxidation. For example, T. Saleh's
group developed a series of NCs made of Pd and Au NPs
supported on carbon (Au–Pd/C), prepared by impregnation
H-bonding with
a surface oxygen atom, leaving the
secondary imine group, the catalytic site, free. The loading
of L-proline onto the Fe3O4 surface would serve two
purposes: (i) it would provide a catalytic site for the aldol
reaction which can further proceed to α,β-unsaturated
ketone by adjusting the time and temperature of the
reaction; (ii) it can also serve as a base required in a
catalytic
amount
for
final
condensation
with
4-hydroxycoumarin in step III (Scheme 1).
6518 | Catal. Sci. Technol., 2019, 9, 6517–6531
This journal is © The Royal Society of Chemistry 2019

View Article Online
Catalysis Science & Technology
Paper
It can be concluded from the above discussion that for the
development of a green one-pot tandem catalytic process for
the production of ( )-warfarin from benzyl alcohol, the
following parameters should be taken care of: (i) sufficient
Pd sites should be available on the surface of the magnetic
matrix; (ii) there are chances of formation of by-products like
carboxylic acid, esters, or ethers in the first step,
dibenzylideneacetone (dimer) in the second step and other
condensation products in the third step, which should not
interfere during the reaction or in the preceding steps; (iii)
there should be an immediate desorption of products from
the surface and should not block the catalytic sites; (iv) the
product of the first step should easily be available as a
reactant for the second step and so on for the continuation
of the tandem process; (v) the selection of the solvent system
for this tandem process is crucial and it should be
compatible with all the reaction components and should be
green in nature as well as recyclable; (vi) corrosion of the
matrix and leaching of the catalytic components should be
monitored throughout the process. All these aspects can be
set by process optimization. Overall, economic and ecological
aspects should be considered for the development of the one-
pot multi-stage organic transformations on the industrial
scale.
The purpose of this work is to develop a tandem one-pot
catalytic process involving a cascade of chemical reactions
carried out by using ‘green solvents and reagents’ leading to
racemic warfarin. For this, we have developed Fe3O4@L-
proline/Pd nanocomposite (NCs) and used it as single catalyst
for the proposed tandem process beginning from benzyl
alcohol. The process parameters like temperature, reaction
time, solvent and the stoichiometry of the reagents required
at each stage are optimized considering the environmental
aspects.
On introduction of Pd in Fe3O4@L-proline NPs, new peaks
at 2θ values of 40.43°, 40.65° and 68.12° emerged
corresponding to the (111), (200) and (220) planes of Pd,
respectively (JCPDS 46-1043). The major peak at 40.43°
due to the (111) planes has high intensity while the other
two peaks are of low intensity due to a very small amount
of Pd in the host matrix (Fig. 1A). This confirms that the
Pd metal is in zero oxidation state (also, due to
hydrazine, a reducing agent used to load Pd on the host
matrix). The emergence of new peaks in the XRD pattern
on loading Pd indicates the presence of Pd as a separate
phase in the form of small islands on the host surface
which
transformations.
can
act
as
The
catalytic
nitrogen
sites
for
organic
adsorption–desorption
isotherms for the synthesized Fe3O4@L-proline NPs and
Fe3O4@L-proline/Pd NCs, their calculated BET surface area
and pore volume are presented in Fig. 1(B and C) and
Table 1. It can be observed from Fig. 1 that the isotherms
of Fe3O4@L-proline NPs are almost similar to that of type
IV (BDDT classification) with the H4 type of hysteresis
loop in which the two branches are nearly horizontal and
parallel over a wide range of relative pressures (P/Po). The
quantity of gas adsorbed, in this case, increases with
increase in P/Po. The type H4 loop is often associated
with narrow slit-like pores.37,38 The adsorption average
pore width (4 V Å−1 by BET) values indicate a mesoporous
material. It can be observed from Table 1 that on loading
Pd, the BET surface area increases. The BET surface area
and pore size studies reveal that the material is suitable
for catalytic activities. N2 adsorption–desorption analyses
could
not
provide
structural
and
morphological
information about the material. Therefore, the structure
and morphology of the samples were directly observed by
field effect scanning electron microscopy (FESEM) and
transmission electron microscopy (TEM). Fig. 1(D and E)
show FESEM images of Fe3O4@L-proline NPs and Fe3O4@L-
proline/Pd NCs. They indicate a spherical morphology, and
on loading Pd, the surface area and porosity of the
material increase while maintaining the morphology,
supporting the BET results. The TEM images for pristine
Fe3O4, Fe3O4@L-proline NPs and Fe3O4@L-proline/Pd NCs
also show spherical particles and are in the size range of
10–15 nm (Fig. 1F and G and S2‡). The spots having
equal intensity in selected area electron diffraction (SAED)
indicate the highly crystalline nature of the material. The
lattice fringes of the exposed (200) planes of magnetite
present in Fe3O4@L-proline/Pd are at 0.15 nm distances
each and that for Pd at 0.25 nm, corresponding to (111)
planes. From the TEM images and SAED patterns it can
be concluded that the L-proline organic layer and Pd
loading did not alter the lattice parameters of the host
Fe3O4, retaining the crystallinity of the material which is
necessary for the catalytic activity and recycling ability due
to the magnetic nature of the material. TEM in
combination with EDS elemental mapping analysis was
employed to determine the distribution of elements in the
2. Results and discussion
The synthesized L-proline capped Fe3O4 NPs were
characterized by XRD to confirm the phase, purity, particle
size and crystallinity of the material (Fig. S1‡). The XRD
pattern manifests predominant diffraction peaks at 2θ
values of 30.51°, 35.78°, 43.55°, 47.13°, 53.50°, 57.40°,
62.92°, 65.74°, 70.92°, 73.96°, 74.96° and 78.92°
corresponding to the (220), (311), (400), (331), (422), (333),
(440), (531), (620), (533), (622) and (444) planes,
respectively (JCPDS 82-1533), confirming the magnetite
phase of Fe3O4 (Fig. 1A). The magnetite phase is known
for its inverted spinel structure having a face centered
cubic (FCC) arrangement and Fe cations occupying the
interstitial Td and/or Oh sites. The particle size was
calculated from the Debye–Scherrer formula (L = 0.9λ/β cos
θ)36 and the FWHM (full width at half maximum) value
of the major peak for (311) planes in the range of 10
nm. There is no change in the XRD pattern due to the
coating of an L-proline layer over Fe3O4, suggesting only
the surface interaction/adsorption of organic molecules.
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol., 2019, 9, 6517–6531 | 6519

View Article Online
Paper
Catalysis Science & Technology
Fig. 1 (A) XRD pattern of Fe3O4@L-proline/Pd2 NCs (* and • indicate Fe3O4 and Pd respectively) (B and C) nitrogen adsorption–desorption
isotherms for Fe3O4@L-proline and Pd loaded Fe3O4@L-proline (D) FESEM image (before Pd loading) (E) FESEM image (after Pd loading) (F) TEM
image (inset image shows SAED pattern) and (G–I) lattice fringes at higher magnification of as-synthesized Fe3O4@L-proline/Pd2 NCs.
6520 | Catal. Sci. Technol., 2019, 9, 6517–6531
This journal is © The Royal Society of Chemistry 2019

View Article Online
Catalysis Science & Technology
Paper
Table 1 BET surface characterization of as-synthesized NPs and NCs
Sr. no.
Parameters
Fe3O4@L-proline NPs
Fe3O4@L-proline/Pd2 NCs
1
2
3
4
5
6
SBET (m2 g−1
)
59.39
114.99
127.94
0.19
0.17 (by DFT)
36.75 (by DFT)
65.23
111.67
111.20
0.18
0.16 (by DFT)
34.68 (by DFT)
Average pore diameter (Å)
Adsorption average pore width (4 V Å−1 by BET) (Å)
Pore volume (cm3 g−1
)
Total volume in pores (cm3 g−1
Total area in pores (m2 g−1
)
)
Fe3O4@L-proline/Pd nanocomposite. The results are given
in Fig. 2. The images clearly show the uniform
distribution of Pd metal on the surface of the Fe3O4 host.
FTIR spectroscopy is one of the best tools to study the
interaction of ligand molecules with the host surface
atoms.39–41 Fig. S3‡ shows the FTIR spectrum of L-proline.
The vibrations at 1622 and 1428 cm−1 correspond to νas and
νs carboxylate stretching, respectively, while the hump at
1335 cm−1 corresponds to –OH bending in carboxylic acid
when it is in the free form. When L-proline interacts with
Fe3O4, this hump disappears (Fig. S3‡), keeping other
carboxylate stretching vibrations intact (Fig. S3‡) and
suggesting that L-proline molecules orient on the surface of
Fe3O4 by H-bonding with surface oxygen atoms (Fig. 3). The
vibration at 636 cm−1 is due to Fe–O stretching. There is no
subtle change in FTIR spectra when Pd is loaded on
Fe3O4@L-proline.
As discussed earlier, 3-way catalysis is required for the
one-pot synthesis of ( )-warfarin from benzyl alcohol. The
benzylideneacetone formed by α-alkylation of ketone with a
1° alcohol in two steps can be directly condensed with
4-hydroxycoumarin under mild basic condition in a single
pot. For this purpose, the amount of L-proline and Pd loaded
onto the surface of the host Fe3O4 must be optimized. A
stoichiometric amount of L-proline was used as a capping
agent during the synthesis of Fe3O4 NPs. Hence, the amount
of Pd loaded on the surface of Fe3O4@L-proline was varied.
Accordingly, three sets of Fe3O4@L-proline/Pd NCs were
prepared (Scheme S1‡) and used for optimization of the
alcohol/ketone
condensation
reaction
leading
to
Fig. 2 Elemental distribution analysis of Fe3O4@L-proline/Pd2 NCs, elemental mapping of (A) Fe (red), (B) O (blue), (C) Pd (green) and (D) their
overlay.
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol., 2019, 9, 6517–6531 | 6521

View Article Online
Paper
Catalysis Science & Technology
Fig. 3 Interaction of L-proline molecules with the Fe3O4 NPs' surface (a) through H-bonding, revealed from FTIR spectroscopy and (b) cartoon
showing the overall molecular framework of the multicomponent catalyst deduced from FTIR spectroscopy and XRD analysis. Small dark-brown
semi-spheres show Pd islands on the surface of host Fe3O4 NPs (dark yellow).
benzylideneacetone in one pot. The amount of Pd loaded was
also determined from elemental EDAX analysis (Fig. S4‡).
It can be observed from Table 2 that Fe3O4@L-proline/Pd1
having the maximum amount of Pd loaded (A, 500 mg g−1
NPs) can convert 1 to 2 in 32 h in the presence of O2 gas at 1
atm and 60 °C in PEG-400 medium, but it failed to transform
2 into 3 in a separate reaction. This may be due to the
blocking of the catalytic imine site of L-proline with excess
Pd. When Fe3O4@L-proline/Pd3 NPs having a lesser amount
of Pd loaded (C, 50 mg g−1 NPs) was used as a catalyst, the
first conversion, 1 to 2, could not take place in a measurable
amount while it could convert 2 to 3 within 20 h when the
reaction was carried out separately. Optimum performance
could be achieved in the presence of Fe3O4@L-proline/Pd2 (B,
200 mg Pd per g NPs); it was able to transform 1 to 2 in 60 h
and 2 to 3 in 24 h in separate reactions. Hence, Fe3O4@L-
proline/Pd2 was selected to proceed further.
although it continued up to 2 days. In a literature survey, we
found that PEG-400 is a low cost, non-toxic and eco-friendly
liquid polymer and in addition, it can be recovered by
extraction with a suitable solvent or by direct distillation of
water or solvent.42,43 It can act as a recyclable reaction
medium for the Baylis–Hillman reaction,44 Heck reaction45
and asymmetric dehydroxylation.46 It also supports the aldol
condensation reaction by acting as a catalyst with anhydrous
K2CO3.47 S. Chandrasekhar et al. reported an L-proline
catalyzed asymmetric aldol reaction in PEG-400 as
a
recyclable medium.48 They were able to recycle this solvent
10 times without losing the activity of the L-proline catalyst.
In the present case, when PEG-400 was used as a solvent for
reaction I, it was unable to support the oxidation of 1 to 2 at
RT although the reaction was continued for 4 days; however,
it could complete the reaction at 60 °C to almost 90%
conversion within 3 days (Table 3). Similar results were
observed for reaction II for aldol condensation of 2 to 3 with
acetone (Table 4). PEG-400 was not able to sustain this
reaction at RT up to 40 °C when the reaction was continued
for 2 days, but could drive the reaction to 53% yield at 70 °C
within 24 h (Table 4). Hence, it is established that PEG-400 is
suitable as a solvent for one-pot α-alkylation of acetone with
benzyl alcohol. It can be assumed that reaction II may
proceed via β-hydroxyketone, but it directly converted into 3
due to dehydration at high temperature.
2.1 Screening of solvent
Both the processes, namely, oxidation of benzyl alcohol to
benzaldehyde (1 to 2, reaction I) and condensation of
benzaldehyde with acetone to benzylideneacetone (2 to 3,
reaction II), were carried out separately in the presence of as-
synthesized NCs as catalyst for solvent screening to achieve
optimum conversion by considering the ‘green aspects’. The
best choice for the purpose is to carry out the
transformations in water as a medium. It can be observed
from Table 3 that reaction I was successfully carried out in
water, and about 70% yield of 2 was achieved in 96 h at RT;
however, no reaction took place for reaction II in water at RT
up to 70 °C. Similar results were obtained when both the
reactions were carried out in toluene as a solvent. Also, both
the reactions were unable to proceed separately at RT in
ethanol as the reaction medium. We had also carried out the
reactions in PBS buffer at neutral pH 7.4. Interestingly, 76%
conversion was achieved at RT within 60 h for reaction I;
however, it was unable to support reaction II separately
2.2 Screening of substrate
Different derivatives of benzyl alcohol were evaluated for both
reactions I and II separately to understand the electronic
effect of substituent groups on the reaction (Tables 5 and 6).
p-Nitrobenzyl alcohol could not undergo oxidation under
optimized reaction parameters (O2, 1 atm/60 °C, PEG-400)
and resulted in poor yield (15%) in 3 days while 53% yield
was obtained for reaction II under optimized parameters
(acetone, 70 °C, PEG-400) in 2 days. When the electron
releasing –OH group was attached to benzyl alcohol, reaction
6522 | Catal. Sci. Technol., 2019, 9, 6517–6531
This journal is © The Royal Society of Chemistry 2019

View Article Online
Catalysis Science & Technology
Paper
I was completed within 60 h with almost 90% yield. However,
it was unable to carry out aldol condensation even though
the reaction was continued for 2 days. It is interesting to note
that p-OCH3, p-CH3 and p-Cl attached to benzyl alcohol can
undergo both the reactions (separately) with substantial
percentage yields (entries 4–6, Tables 5 and 6) in 60 and 24 h
for reactions I and II, respectively. Unsubstituted benzyl
alcohol undergoes oxidation with the highest (94%) yield in
60 h, while its corresponding aldehyde can condense with
acetone at 70 °C in PEG-400 medium to form 3 within 24 h
and with 53% conversion in a separate reaction (entry 2,
Tables 5 and 6).
From the above observations, benzyl alcohol was selected
to begin with, to carry out both reactions I and II in a single
pot in the presence of Fe3O4@L-proline/Pd2 NCs as the
catalyst and PEG-400 as the solvent. It can be seen from
Table
7 that benzyl alcohol yields the corresponding
benzylideneacetone at the end of the 4th day with 47% yield,
while p-CH3, p-Cl and p-OCH3 derivatives of 1 yield 52%, 52%
and 58% overall conversion to 3, respectively, at the end of
the 4th day in one pot.
In next set of experiments, the above two steps were
extended to the third sequential reaction, that is, the Michael
addition of 4-hydroxycoumarin (4) to 3. The pot reaction was
extended for a further 12 h at 100 °C maintaining the other
pot conditions the same (Scheme 2). At the end of the 5th
day (overall 120 h), ( )-warfarin could be separated by
fractional crystallization from the reaction mass with 35%
yield. The peculiarity of this one-pot one catalyst process is
that the solvent PEG-400 can be recycled (3 times in this
study) to carry out fresh pot reactions without the loss of
catalytic activity.
2.3 Discussion
Park et al.49 reported palladium (0.2 mol%) nanoparticles
entrapped in aluminium hydroxide, Pd/AlOIJOH). This catalyst
shows activity for alkene hydrogenation and aerobic alcohol
oxidation. It produced 1,3-diphenylpropan-1-one (6) in 97%
yield in α-alkylation of acetophenone 5 with 1.2 equiv. of 1 in
the presence of a base like K3PO4 (3 equiv.) for 8 h at 80 °C
under argon (Scheme 3). It produced chalcone 8 with a yield
of 95% in 20 h when the reaction was carried out under 1
atm O2 in a toluene medium. They compared the activity of
the catalyst with different commercially available
combinations like 5% Pd/C, 5% Pd/Al2O3, 5% Pd/BaCO3,
[RuCl2IJPPh3)3], [{IrIJcod)Cl}2] etc. However, these catalysts favor
6 as the major product with very little formation of 7. Product
8 was obtained in trace amounts or even none at all. In
contrast, the herein developed catalyst favors 8 as the major
product due to step-wise reactions.
Further, these catalytic processes require a base like K3PO4
(3 equiv.) or KOH (1 to 0.1 equiv.) for each cycle to achieve the
optimum reaction rate again. Such strong bases may attack the
host surface responsible for leaching of the precious metal
catalyst. The Ru grafted hydrotalcite was reported to bring about
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol., 2019, 9, 6517–6531 | 6523

View Article Online
Paper
Catalysis Science & Technology
Table 3 Optimization of solvent system and temperature for reaction I in the presence of as-synthesized Fe3O4@L-proline/Pd2 NCs
Solvent
Timec (h)
Temperature (°C)
Yielda,b (%)
Toluene
Ethanol
PBS buffer pH 7.4
Water
PEG-400
48
48
60
96
96
60
70
56
Trace
76
70
Trace
94
RT
RT
RT
RT
60
PEG-400
a
b
Isolated yields after column chromatography. The reactions were carried out in triplicate. Benzyl alcohol (1.0 mmol), solvent (5 mL) and
catalyst (100 mg) were stirred at the above-mentioned temperature under 1 atmosphere pressure of dioxygen in a sealed tube till the
disappearance of the starting material (from the TLC plate). The product was extracted in diethyl ether (10 mL × 3), concentrated and purified
c
by column chromatography eluting with cyclohexane/ethyl acetate. The progress of the reaction was monitored on the TLC plate after every 4
hours.
Table 4 Optimization of solvent system and temperature for reaction II in the presence of as-synthesized Fe3O4@L-proline/Pd2 NCs
Solvent
Time (h)
Temperature (°C)
Yielda,b (%)
Toluene
Ethanol
PBS buffer pH 7.4
PBS buffer pH 7.4 + DMSO
Water
Water
PEG 400
PEG 400
24
24
24
24
24
24
24
24
70
RT
RT
RT
RT
39
40
70
No reaction
No reaction
No reaction
No reaction
No reaction
16
Trace
53
a
Isolated yields after column chromatography. The reaction was carried out in triplicate for 24 h and the progress of the reaction was
b
monitored on the TLC plate after every 4 hours. Benzaldehyde (1.0 mmol), acetone (5 mL), and catalyst (100 mg) in 5 mL of the above solvent
were stirred at ambient temperature in a sealed tube till the disappearance of the starting material (from the TLC plate). The product was
extracted in diethyl ether (10 mL × 3), concentrated and purified by column chromatography eluting with cyclohexane/ethyl acetate.
α-alkylation of ketone to produce 6 without any requirement of
a base. However, it demands a very high temperature of 180
°C.50 Yanan et al. have reported α-alkylation of acetone with 1
in the presence of CrO3 at 56 °C in 10 h.51 Apart from benzyl
alcohol, they selected p-OCH3 and p-Cl derivatives as substrates
and achieved a high percentage yield of benzylideneacetone, in
agreement with our present results. They also obtained
dibenzylideneacetone as a by-product in various proportions. In
this case, CrO3 worked as the oxidant as well as the catalyst and
was consumed at the end of the reaction. Hence, from the
environmental point of view, this catalyst cannot be used in an
industrial scale. The importance of the presently developed
catalyst is that it can do 3-way catalysis leading to ( )-warfarin
in a single-pot process in the presence of all the products, by-
products and intermediates generated during various reaction
steps. The mechanism combining transition metal catalysis and
organocatalysis is shown in Fig. S5.‡
It is quite natural to think of chiral induction during step
3 of the tandem process due to the presence of L-proline on
the surface as a 2° amine organocatalyst to form enantiopure
(R)- or (S)-warfarin. Kristensen et al.52 had tested most
common
imidazolidinone
organocatalysts
(9) and
like
the
MacMillan-type
Jørgensen/Hayashi
the
diarylprolinols (10, 11 Fig. 4) for the synthesis of enantiopure
warfarin. They achieved an excellent percentage yield of
racemic warfarin but poor conversion to a single specific
stereoisomer even after 9 days. The Jørgensen group1 also
carried out this reaction in the presence of S-proline but they
obtained only racemic warfarin as the product. The possible
reason, as explained by Kristensen, may be the long reaction
time due to which the racemic pathway dominates over the
asymmetric one.52
To extend the application of the synthesized catalyst NCs
and the developed one-pot tandem reaction, we had carried
6524 | Catal. Sci. Technol., 2019, 9, 6517–6531
This journal is © The Royal Society of Chemistry 2019

View Article Online
Catalysis Science & Technology
Paper
Table 5 Oxidation of benzyl alcohol and its derivatives to the corresponding aldehyde (reaction I) under optimized reaction parameters in the presence
of Fe3O4@L-proline/Pd2 NCsa
Sr. no.
1
–R
Time (h)
60
Product
Yield (%)
15
–NO2
2
3
–H
60
60
94
90
–OH
4
5
–CH3
–Cl
60
60
92
80
6
–OCH3
60
78
a
The detailed protocol is discussed in the Experimental section.
out the reaction in a pilot plant. The flow diagram of the unit
processes used for the production of ( )-warfarin starting
from 10 g of benzyl alcohol is shown in Fig. 5.
The procedure used for the production of the target
product in a tandem process in a pilot plant is discussed in
detail in the Experimental section.
at any stage; (iv) the as-synthesized catalyst NCs maintained
their efficiency for three cycles (under this study) for 3-way
catalysis; (v) the reaction is ‘green’ as no noxious solvents are
used and the whole one-pot process can be carried out in
PEG-400 and water as the reaction medium. This study opens
new avenues for the development of multicomponent
catalysts to produce materials on a large scale in a ‘green
way’. The next study would be the development of a tandem
process for the large-scale production of enantiopure
warfarin by loading chiral 1° diaryldiamines on the magnetic
host surface with palladium.
3. Conclusion
We have demonstrated that multi-step synthesis of
pharmaceutically important molecules can be carried out in
a single pot by inserting multiple catalytic components into a
robust host and optimizing the reaction parameters. Various 4. Experimental
catalytic components present in a single nanocomposite work
All the chemicals were purchased from Aldrich Chemicals
synergistically such that organic transformations, like
oxidation of benzyl alcohol to benzaldehyde (Pd metal as
catalyst), which in turn later condensed with acetone in an
aldol fashion (L-proline as organocatalyst) resulting in
unless specified. FeCl3·6H2O was purchased from Loba
Chemie, India. FeSO4·4H2O was purchased from S. D. Fine
Chemicals Limited, India. Aqueous ammonia was purchased
from S. R. Enterprise, India. L-Proline was purchased from
TCI Chemicals and palladium acetate from Hindustan
Platinum.
benzylideneacetone
which
then
condensed
with
4-hydroxycoumarin by Michel addition (L-proline, as
secondary amine, provides the basic condition) leading to the
target product warfarin, are carried out in a synchronized
way. By implementing all the optimized process parameters
for the pilot-scale production of ( )-warfarin, it can be
concluded that (i) the developed process becomes cost-
effective compared to others based on bare noble metal
catalysts; (ii) the catalyst recovery becomes easy by applying a
magnetic field; (iii) the possibility of noble metal leaching
from NCs is nullified in the process as no strong base is used
4.1 Synthesis of L-proline coated magnetic nanoparticles
(Fe3O4@L-proline NPs)
Amino acid-coated magnetite nanoparticles were synthesized
by the one-pot synthesis method. In brief, ferric chloride
hexahydrate (5.0 g, 18.5 mmol) and ferrous sulphate
tetrahydrate (2.4 g, 10.7 mmol), molar ratio 1 : 0.58, were
added to 250 mL deionized water under an inert atmosphere.
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol., 2019, 9, 6517–6531 | 6525

View Article Online
Paper
Catalysis Science & Technology
Table 6 Aldol condensation of benzaldehyde and its derivatives to the corresponding benzylideneacetone (reaction II) under optimized reaction
parameters in the presence of Fe3O4@L-proline/Pd2 NCsa
Sr. no.
1
–R
Time (h)
24
Product
Yield (%)
–NO2
No desired product
2
3
–H
24
24
53
–OH
No reaction
4
5
6
–CH3
–Cl
24
24
24
58
68
65
–OCH3
a
The detailed protocol is discussed in the Experimental section.
The mixture was heated to 70 °C to obtain a clear yellow
in an oven at 60 °C overnight. Dry weight of nanoparticles:
solution. The mixture was stirred vigorously with
a
1.01 g.
mechanical stirrer to avoid interparticle magnetic interaction.
After 30 min, L-proline (5.3 g, 46.0 mmol) dissolved in 50 mL
deionized water was rapidly added. The clear reaction
mixture was further stirred for another 30 min. Then 27 mL
ammonium hydroxide (25%) was rapidly added which in turn
changed the reaction color to black. The stirring was
continued for 1.5 h. Finally, the black nanoparticles were
separated magnetically, washed with distilled water three
times, and dried in an oven at 100 °C overnight. Dry weight
of nanoparticles: 3.0 g.
4.3 Oxidation of benzyl alcohol (1) to benzaldehyde (2)
Benzyl alcohol (100 mg, 0.92 mmol) and Fe3O4@L-proline/Pd
(100 mg) catalyst were added to PEG-400 (5.0 mL) with
magnetic stirring. The reaction mixture was heated to 60 °C
under 1 atm dioxygen pressure in a sealed tube. The progress
of the reaction was monitored by TLC. After the
disappearance of the starting material, the product was
extracted in diethyl ether (10 mL × 3), concentrated under
vacuum and purified by silica gel column chromatography,
and eluted with 20% ethyl acetate in cyclohexane to yield the
product as a colourless liquid (92 mg, 94%). 1H NMR (300
MHz, CDCl3): δ 7.48–7.53 (m, 2H), 7.58–7.64 (m, 1H), 7.85–
7.88 (m, 2H) 10.00 (s, 1H). 13C NMR (75 MHz, CDCl3): δ
128.39, 128.99, 129.70, 130.03, 134.45, 136.38, 192.38.
4-Chlorobenzaldehyde. Prepared as described for 2.
Starting from 4-chlorobenzylalcohol (100 mg, 0.70 mmol),
after purification with silica gel column chromatography
eluted with 20% ethyl acetate in cyclohexane, a white solid
was produced (79 mg, 80%), mp 46–47 °C. 1H NMR (300
MHz, CDCl3): δ 7.50–7.53 (m, 2H), 7.80–7.85 (m, 2H), 9.8 (s,
1H). 13C NMR (75 MHz, CDCl3) δ 129.45 (2C), 130.90 (2C),
134.72, 140.92, 190.84.
4.2 Synthesis of palladium loaded amino acid functionalized
magnetic nanocomposites (Fe3O4@L-proline/Pd NCs)
Palladium acetate (0.2 g, 0.89 mmol) was dissolved in 200 mL
of ethanol at room temperature with vigorous mechanical
stirring to obtain an orange colored transparent solution. To
this solution was added 1.0 g L-proline coated magnetic
nanoparticles (Fe3O4@L-proline). This suspension was stirred
vigorously and refluxed for 2 h. As the reaction proceeded,
the initial brown suspension turned dark. This reaction
mixture was cooled to room temperature and 2.5 mL 80%
hydrazine hydrate was added. The reaction mixture was
further refluxed for 1.5 h. The generated nanocomposites
were separated magnetically, washed with ethanol three
times followed by distilled water three times, and then dried
p-Hydroxybenzaldehyde. Prepared as described for 2.
Starting from 4-hydroxybenzylalcohol (100 mg, 0.80 mmol),
6526 | Catal. Sci. Technol., 2019, 9, 6517–6531
This journal is © The Royal Society of Chemistry 2019

View Article Online
Catalysis Science & Technology
Paper
Table 7 One-pot conversion of benzyl alcohol and its derivatives to benzylideneacetone in the presence of Fe3O4@L-proline/Pd2 NCs under optimized
reaction parametersa
Sr. no.
1
–R
Time (h)
60 + 24
Expected product
Yield (%)
–NO2
No product
2
–H
60 + 24
47
3
4
5
–OH
–CH3
–Cl
60 + 24
60 + 24
60 + 24
No product
52
52
6
–OCH3
60 + 24
58
a
The detailed protocol is discussed in the Experimental section.
Scheme 2 Synthesis of ( )-warfarin through 3-way catalysis in the presence of Fe3O4@L-proline/Pd2.
after purification with silica gel column chromatography
eluted with 20% ethyl acetate in cyclohexane a light brown
solid was obtained (88.5 mg, 90%), mp 110–113 °C. H NMR
δ 3.82 (s, 3H), 6.94 (d, 2H), 7.76 (d, 2H), 9.82 (s, 1H). 13C
NMR (75 MHz, CDCl3) δ 55.58, 114.32, 129.94, 131.98, 164.62,
190.84.
1
(300 MHz, CDCl3): δ 6.98–7.01 (m, 2H), 7.79–7.82 (m, 2H),
9.83 (s, 1H), 10.67 (brs, 1H). 13C NMR (75 MHz, CDCl3) δ
116.66, 128.89, 132.54, 164.20, 191.31.
p-Nitrobenzaldehyde. Prepared as described for 2. Starting
from 4-nitrobenzyl alcohol (100 mg, 65 mmol), after
purification with silica gel column chromatography eluted
with 20% ethyl acetate in cyclohexane, a yellow solid was
p-Methylbenzaldeyde. Prepared as described for 2. Starting
from 4-methylbenzylalcohol (100 mg, 0.82 mmol) after
purification with silica gel column chromatography eluted
with 20% ethyl acetate in cyclohexane, a colourless liquid
1
obtained (15 mg, 15%), mp 102–104 °C. H NMR (300 MHz,
CDCl3) δ 8.10 (d, 2H), 8.41 (d, 2H), 10.18 (s, 1H). 13C NMR
(75 MHz, CDCl3) δ 124.32, 130.51, 140.06, 151.12, 190.37.
1
was obtained (90 mg, 92%). H NMR (300 MHz, CDCl3) δ 2.41
(s, 3H), 7.30 (d, 2H), 7.75 (d, 2H), 9.94 (s, 1H). 13C NMR (75
MHz, CDCl3) δ 21.81, 129.08, 134.20, 145.51, 191.93.
4.4 One-pot synthesis of benzylideneacetone (3) from benzyl
alcohol (1)
p-Methoxybenzaldehyde. Prepared as described for 2.
Starting from 4-methoxybenzyl alcohol (100 mg, 0.72 mmol),
after purification with silica gel column chromatography
eluted with 20% ethyl acetate in cyclohexane, a colourless
liquid was obtained (76 mg, 78%). 1H NMR (300 MHz, CDCl3)
Benzyl alcohol (100 mg, 0.92 mmol) was added to PEG-400 (5
mL) and stirred for 10 min. Then, the catalyst Fe3O4@L-
proline/Pd2 (100 mg) was added and stirred for 10 min. The
reaction temperature was raised to 60 °C and the reaction
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol., 2019, 9, 6517–6531 | 6527

View Article Online
Paper
Catalysis Science & Technology
Scheme 3 Palladium-catalyzed coupling of acetophenone and benzyl alcohol in the presence of Park's catalyst.49
was stirred for about 60 h under 1 atmosphere pressure of
dioxygen. After the disappearance of the starting material
(confirmed from the TLC plate), the dioxygen flow was
stopped and the reaction was flushed with N2 gas. At this
time, acetone (5 mL) was introduced and the reaction
mixture was stirred for 5 min under an inert atmosphere.
The reaction vessel was sealed with a rubber septum and the
mixture was stirred at 70 °C for about 24 h. The product was
extracted in diethyl ether (10 mL × 3), concentrated and
purified by column chromatography eluted with 25% ethyl
acetate in cyclohexane to give 3 as a light-yellow low melting
solid (64 mg, 47%). Rf 0.68 (1 : 4 ethyl acetate/cyclohexane).
1H-NMR (300 MHz, CDCl3) δ 2.36 (s, 3H), 6.70 (d, 1H), 7.38
(q, 3H), 7.49 (s, 1H), 7.52 (q, 2H). 13C NMR (75 MHz, CDCl3)
melting solid was obtained (68 mg, 52%). Rf 0.65 (1 : 4 ethyl
acetate/cyclohexane). 1H NMR (300 MHz, CDCl3) δ 2.33 (d,
6H), 6.65 (d, 1H), 7.16 (d, 2H), 7.42 (d, 2H), 7.43 (d, 1H). 13C
NMR (75 MHz, CDCl3) δ 21.46, 27.37, 126.19, 128.28, 129.71,
131.64, 140.99, 143.53, 198.44.
4.5 One-pot synthesis of ( )-warfarin (4)
Benzyl alcohol (100 mg, 0.92 mmol) was added to PEG-400 (5
mL) and stirred for 10 min. Then the catalyst Fe3O4@L-
proline/Pd 2 (100 mg) was added and stirred for 10 min. The
reaction temperature was raised to 60 °C and the reaction
was stirred for about 60 h under 1 atmosphere pressure of
dioxygen. After the disappearance of the starting material on
TLC, the flow of dioxygen was stopped and the reaction was
flushed with N2 gas. To this reaction mass, acetone (5 mL)
was introduced and stirred for 5 min under an inert
atmosphere. The reaction vessel was sealed with a rubber
septum and stirred at 70 °C for about 24 h. Acetone was
removed from the reaction mass under vacuum at 50 °C. At
this time, 4-hydroxycoumarin (75 mg, 0.46 mmol) and water
(5 mL) were added to the reaction mixture and stirred at 100
°C. After stirring for 12 h the reaction mass was cooled to
room temperature and the catalyst was removed
magnetically. The clear solution was dumped in 5 g ice. The
product was filtered and recrystallized from acetone/water to
give ( )-warfarin, (99 mg) as a white solid having 35% overall
yield from benzyl alcohol. Melting point: 159–161 °C. 1H
NMR (300 MHz, CDCl3) δ 2.04 (s, 3H), 3.29 (m, 2H), 4.89 (m,
1H), 6.99–7.40 (m, 8H), 7.82 (d, 1H). 13C NMR (75 MHz,
CDCl3) δ 30.19, 36.15, 46.85, 99.45, 115.62, 121.77, 124.04,
125.07, 125.22, 127.70, 128.12, 129.66, 147.40, 154.14, 164.25,
172.46, 209.50.
δ
27.49, 127.11, 128.28, 128.98, 130.55, 134.39, 143.51,
198.50.
4-Methoxybenzylideneacetone. Prepared as described for
3. Starting from 4-methoxybenzyl alcohol (100 mg, 0.72
mmol), after purification with silica gel column
chromatography eluted with 25% ethyl acetate in cyclohexane
a white low melting solid was obtained (84 mg, 58%). Rf 0.65
1
(1 : 4 ethyl acetate/cyclohexane). H NMR (300 MHz, CDCl3) δ
2.22 (s, 3H), 3.70 (s, 3H), 6.47 (d, 1H), 6.78 (d, 2H), 7.34 (d,
1H), 7.36 (d, 2H). 13C NMR (75 MHz, CDCl3) δ 27.33, 55.32,
114.41, 124.93, 129.95, 143.22, 161.58, 198.29.
4-Chlorobenzylideneacetone. Prepared as described for 3.
Starting from 4-chlorobenzyl alcohol (100 mg, 0.70 mmol),
after purification with silica gel column chromatography
eluted with 25% ethyl acetate in cyclohexane a tan low
melting solid was obtained (66 mg, 52%). Rf 0.55 (1 : 4 ethyl
acetate/cyclohexane). 1H NMR (300 MHz, CDCl3) δ 2.28 (s,
3H), 6.57 (d, 1H), 7.25 (d, 2H), 7.35 (d, 1H), 7.36 (d, 2H). 13C
NMR (75 MHz, CDCl3) δ 27.56, 127.42, 129.14, 129.38, 132.91,
136.21, 141.70, 197.88.
4-Methylbenzylideneacetone. Prepared as described for 3.
Starting from 4-methylbenzyl alcohol (100 mg, 0.82 mmol),
after purification with silica gel column chromatography
eluted with 25% ethyl acetate in cyclohexane a yellow low
4.6 Production of ( )-warfarin at pilot plant
To a jacketed glass reactor of 1 L capacity, benzyl alcohol
(10 g) and PEG-400 (500 mL) were added at ambient
temperature and the mass was stirred for 10 min. To this was
added the catalyst Fe3O4@L-proline/Pd2 (10 g) and the
mixture was stirred for 10 min. The reaction mass was heated
to 60 °C by applying hot water in the reactor jacket. This
suspension was stirred for about 60 h at 60 °C under 1 atm
pressure of dioxygen. After the disappearance of the starting
material, the reaction was cooled to ambient temperature by
applying chilled water in the jacket. Dioxygen was removed
and the reaction was flushed with nitrogen. To this reaction
mass, acetone (500 mL) was added and was stirred for 5 min
under a nitrogen atmosphere. The reactor was closed and
Fig.
4 MacMillan type imidazolidinone (9) and Jørgensen/Hayashi
diarylprolinols (10, 11) as organocatalysts.
6528 | Catal. Sci. Technol., 2019, 9, 6517–6531
This journal is © The Royal Society of Chemistry 2019

View Article Online
Catalysis Science & Technology
Paper
Fig. 5 Flow diagram for the pilot-scale production of ( )-warfarin from benzyl alcohol by the developed tandem process in the presence of
Fe3O4@L-proline/Pd2 NCs.
stirred at 70 °C for about 24 h. Acetone was removed from
the reaction mass by applying a vacuum of about 100 millibar
at 50 °C. The distilled acetone was collected in a receiver for
further use. 4-Hydroxycoumarin (7.5 g), and water (500 mL)
were added to the reaction mass and stirred at 100 °C. After
stirring for 12 h the reaction mass was cooled to room
temperature by applying chilled water circulation in the
reactor jacket, and the catalyst was removed by a specially
designed Neutch filter having magnetic bars. The clear
solution was dumped in 500 g ice. The product was filtered
and recrystallized from acetone/water to give ( )-warfarin
(12.3 g) as a white solid with 43% overall yield from benzyl
alcohol.
in the relative pressure (P/Po) range of 0.0–1.0. The
desorption isotherm was used to determine the pore-size
distribution using the Barret–Joyner–Halenda (BJH) method,
assuming a cylindrical pore model. The nitrogen adsorption
volume at the relative pressure (P/Po) range of 0.997 was used
to determine the pore volume and average pore size. The
FTIR (PerkinElmer, Spectrum Two) spectra of the synthesized
1
products were obtained in the range of 400 to 4000 cm−1. H
NMR and 13C NMR spectra of the products were recorded on
a Bruker Avance 300 MHz spectrometer using TMS as an
internal standard in CDCl3. All the spectra are reported in
the ESI.‡
Abbreviations
4.7 Characterization
Fe3O4@L-Proline NPs L-Proline capped Fe3O4 nanoparticles
Fe3O4@L-Proline/Pd NCs L-Proline capped and Pd loaded
Fe3O4 nanocomposites
Reaction I Oxidation of benzyl alcohol to benzaldehyde
Reaction II Aldol condensation of benzaldehyde with
acetone
An X-ray powder diffraction (XRD) pattern was obtained
using an X-ray powder diffractometer (Philips X'pert MPD
system) with Cu Kα radiation,
λ = 0.15418 nm. The
morphology with elemental analysis of the samples was
examined by FESEM (JEOL-JSM-7100F). The particle size
determination with elemental mapping was carried out by
TEM (JEOL-JEM-2100) at 200 kV. The samples were degassed
at 80 °C prior to Brunauer–Emmett–Teller (BET)
measurements. The BET specific surface area (SBET) was
determined by nitrogen adsorption (Micrometrics ASAP 2020,
USA) via a multipoint BET method using the adsorption data
Conflicts of interest
There are no conflicts to declare.
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol., 2019, 9, 6517–6531 | 6529

View Article Online
Paper
Catalysis Science & Technology
Acknowledgements
23 Z. Zhang, Y. Wang, M. Wang, J. Lu, C. Zhang, L. Li, J. Jiang
and F. Wang, Catal. Sci. Technol., 2016, 6, 1693–1700.
24 Y. Su, L. C. Wang, Y. M. Liu, Y. Cao, H. Y. He and K. N. Fan,
Catal. Commun., 2007, 8, 2181–2185.
25 M. Sono, M. P. Roach, E. D. Coulter and J. H. Dawson,
Chem. Rev., 1996, 96, 2841–2887.
26 S. Tareq, Y. H. T. Yap, T. A. Saleh, A. H. Abdullah, U. Rashid
and S. M. Izham, J. Mol. Liq., 2018, 271, 885–891.
27 D. I. Enache, J. K. Edwards, P. Landon, B. Solsona-
Espriu, A. F. Carley, A. A. Herzing, M. Watanabe, C. J.
Kiely, D. W. Knight and G. J. Hutchings, Science,
2006, 311, 362–365.
28 F. Zamani and S. M. Hosseini, Catal. Commun., 2014, 43,
164–168.
29 X. Wang, G. Wu, N. Guan and L. Li, Appl. Catal., B,
2012, 115–116, 7–15.
30 T. K. Hollis and B. Bosnich, J. Am. Chem. Soc., 1995, 117,
4570–4581.
31 T. Bach, Angew. Chem., Int. Ed. Engl., 1994, 33, 417–419.
32 A. Erkkilä, I. Majander and P. M. Pihko, Chem. Rev.,
2007, 107, 5416–5470.
HPS thanks the Science & Engineering Research Board
(SERB), New Delhi, India for financial support under a
sponsored scheme (No. EMR/2016/007638). The authors
thank DST-FIST for providing the BET surface area
measurement facility in the Department of Chemistry, Faculty
of Science, The M. S. University of Baroda. We also thank
Prof. A. V. Ramachandran for necessary suggestions and
corrections in the manuscript.
References
1 N. Halland, T. Hansen and K. A. Jørgensen, Angew. Chem.,
Int. Ed., 2003, 42, 4955–4957.
2 J.-W. Xie, L. Yue, W. Chen, W. Du, J. Zhu, J.-G. Deng and
Y.-C. Chen, Org. Lett., 2007, 9, 413–415.
3 H. Kim, C. Yen, P. Preston and J. Chin, Org. Lett., 2006, 8,
5239–5242.
4 R. M. de Figueiredo and M. Christmann, Eur. J. Org. Chem.,
2007, 2575–2600.
5 T. Meinertz, W. Kasper, C. Kahl and E. Jahnchen, Br. J. Clin.
Pharmacol., 1978, 5, 187–188.
6 R. A. O'Reilly, N. Engl. J. Med., 1976, 295, 354–357.
7 L. B. Wingard, R. A. O'Reilly and G. Levy, Clin. Pharmacol.
Ther., 1978, 23, 212–217.
8 T. A. Saleh, S. A. AL-Hammadi, I. M. Abdullahi and M.
Mustaqeem, J. Mol. Liq., 2018, 272, 715–721.
9 H. Chen, W. Liu and Z. Qin, Catal. Sci. Technol., 2017, 7,
2236–2244.
10 Y. Li, X. Feng and Z. Li, Catal. Sci. Technol., 2019, 9,
377–383.
33 P. Clapés and J. Joglar, in Modern Methods in
StereoselectiveAldol Reactions, ed. R. Mahrwald, Enzyme-
Catalyzed Aldol Additions, Wiley-VCH Verlag GmbH & Co.
KGaA, 2013, ch. 8., pp. 475–527.
34 J. Wagner, R. A. Lerner and C. F. Barbas III, Science,
1995, 270, 1797–1800.
35 B. List, R. A. Lerner and C. F. Barbas III, J. Am. Chem. Soc.,
2000, 122, 2395–2396.
36 R. Jenkins and R. L. Snyder, Introduction to X-ray
PowderDiffractometry, John Wiley & Sons, NY, 1996.
37 K. S. W. Sing, D. H. Everett, R. A. W. Haul, L. Moscou, R. A.
Pierotti, J. Rouquerol and T. Siemieniewska, Pure Appl.
Chem., 1985, 57, 603–619.
11 G. Gedda, Y.-Y. Yao, S.-H. Chen, A. V. Ghule, Y.-C. Ling and
J.-Y. Chang, J. Mater. Chem. B, 2017, 5, 6282–6291.
12 T. A. Saleh, Desalin. Water Treat., 2016, 57,
10730–10744.
38 J. Yu, L. Zhang, B. Cheng and Y. Su, J. Phys. Chem. C,
2007, 111, 10582–10589.
13 For a review of alcohol oxidation, see: A. M. F. Phillips,
A. J. L. Pombeiro and M. N. Kopylovich, ChemCatChem,
2017, 9, 217–246.
14 J. Muzart, Tetrahedron, 2003, 59, 5789–5816.
15 G. Tojo and M. Fernández, in Oxidation of alcohol to
aldehyde and ketones: A guide to current common practice,
Springer, USA, 2006.
16 K. Sato, M. Aoki and R. Noyori, Science, 1998, 281,
1646–1647.
17 B. M. Trost, Science, 1991, 254, 1471–1477 (Angew. Chem.,
Int. Ed. Engl., 1995, 34, 259–281).
18 J.-E. Backvall, R. L. Chowdhury and U. Karlsson, J. Chem.
Soc., Chem. Commun., 1991, 473.
19 K. P. Peterson and R. C. Larock, J. Org. Chem., 1998, 63,
3185–3189.
20 G.-J. ten Brink, I. W. Arendsand and R. A. Sheldon, Science,
2000, 287, 1636–1639.
21 J. A. Cella, J. A. Kelley and E. F. Kenehan, J. Chem. Soc.,
Chem. Commun., 1974, 943.
39 E. Shah and H. P. Soni, RSC Adv., 2013, 3, 17453–17461.
40 E. Shah, P. Upadhyay, M. Singh, M. S. Mansuri, R.
Begum, N. Sheth and H. P. Soni, New J. Chem., 2016, 40,
9507–9519.
41 M. Kakihana, T. Nagumo, O. Makoto and H. Kakihana,
J. Phys. Chem., 1987, 91, 6128–6136.
42 J. Chen, S. K. Spear, J. G. Huddleston and R. D. Rogers,
Green Chem., 2005, 7, 64–82.
43 E. Santaniello, A. Manzocchi and P. Sozzani, Tetrahedron
Lett., 1979, 20, 4581–4582.
44 S. Chandrasekhar, C. Narsihmulu, B. Saritha and S. S.
Sultana, Tetrahedron Lett., 2004, 45, 5865–5867.
45 S. Chandrasekhar, C. Narsihmulu, S. S. Sultana and N. R.
Reddy, Org. Lett., 2002, 4, 4399–4401.
46 S. Chandrasekhar, C. Narsihmulu, S. S. Sultana and N. R.
Reddy, Chem. Commun., 2003, 1716–1717.
47 Y.-Q. Cao, Z. Dai, R. Zhang and B.-H. Chen, Synth. Commun.,
2005, 35, 1045–1049.
48 S. Chandrasekhar, N. R. Reddy, S. S. Sultana, C. Narsihmulu
and K. V. Reddy, Tetrahedron, 2006, 62, 338–345.
22 J. Muzart, Tetrahedron, 2003, 59, 5789–5816.
6530 | Catal. Sci. Technol., 2019, 9, 6517–6531
This journal is © The Royal Society of Chemistry 2019

View Article Online
Catalysis Science & Technology
Paper
49 M. S. Kwon, N. Kim, S. H. Seo, I. S. Park, R. K. Cheedrala
and J. Park, Angew. Chem., Int. Ed., 2005, 44, 6913–6915.
50 K. Motokura, D. Nishimura, K. Mori, T. Mizugaki, K. Ebitani
and K. Kaneda, J. Am. Chem. Soc., 2004, 126, 5662–5663.
51 L. Yanan and C. Daoyong, Chin. J. Chem., 2011, 29,
2086–2090.
52 T. E. Kristensen, K. Vestli, F. K. Hansen and T. Hansen, Eur.
J. Org. Chem., 2009, 5185–5191.
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol., 2019, 9, 6517–6531 | 6531