Magnetic superbasic proton sponges are readily removed and permit direct product isolation

Article abstract of DOI:10.1021/jo501913z

Workup in organic synthesis can be very time-consuming, particularly when using reagents with both a solubility similar to that of the desired products and a tendency not to crystallize. In this respect, reactions involving organic bases would strongly be

Full text of DOI:10.1021/jo501913z

Article
pubs.acs.org/joc
Magnetic Superbasic Proton Sponges Are Readily Removed and
Permit Direct Product Isolation
Elia M. Schneider, Renzo A. Raso, Corinne J. Hofer, Martin Zeltner, Robert D. Stettler, Samuel C. Hess,
Robert N. Grass, and Wendelin J. Stark*
Institute for Chemical and Bioengineering, Department of Chemistry and Applied Biosciences, ETH Zu
̈
rich, Wolfgang-Pauli Strasse
10, CH-8093 Zurich, Switzerland
*
S Supporting Information
ABSTRACT: Workup in organic synthesis can be very time-
consuming, particularly when using reagents with both a solubility
similar to that of the desired products and a tendency not to crystallize.
In this respect, reactions involving organic bases would strongly benefit
from a tremendously simplified separation process. Therefore, we
synthesized a derivative of the superbasic proton sponge 1,8-
bis(dimethylamino)naphthalene (DMAN) and covalently linked it to
the strongest currently available nanomagnets based on carbon-coated
cobalt metal nanoparticles. The immobilized magnetic superbase
reagent was tested in Knoevenagel- and Claisen−Schmidt-type
condensations and showed conversions of up to 99%. High yields of
up to 97% isolated product could be obtained by simple recrystallization without using column chromatography. Recycling the
catalyst was simple and fast with an insignificant decrease in catalytic activity.
INTRODUCTION
combine a large surface area with the potential for a simple
separation process. However, many magnetic reagents and
surface linkers for reagent attachment are unstable in acidic-,
basic-, or organic-solvent-containing reaction media. The
exceptional chemical stability of carbon-coated metal nano-
particles is based on a crystalline, graphene-like carbon
surface that effectively prevents core oxidation. The all-carbon
■
Organosuperbases, such as the classic proton sponge 1,8-
bis(dimethylamino)naphthalene (DMAN), Verkade’s base
proazaphosphatrane), and 1,8-bis(tetramethylguanidino)-
naphthalene (TMGN) have become very important reagents
4
(
19
7
in organic chemistry in past years. Their exceptional basicity is
associated with high levels of kinetic activity in proton-exchange
reactions. These properties are often manifested by low
nucleophilicity, making this kind of compound interesting for
surface further allows covalent functionalization of the particle
20
surface by using commercially available aryl diazonium salts.
8
−11
At present, a number of promising recyclable, stable, metal-
a wide scope of reactions,
and even heterogeneous cases are
21−31
based magnetic catalysts have been proposed.
Most
(so-
1
,2
known. However, the workup of such reactions on the
laboratory level can be difficult, time-consuming, and expensive.
Similarly, in industry, a broad variety of condensation reactions
are applied. At present, free bases, such as sodium hydroxide,
potassium hydroxide, and various organic bases, are the most
commonly used, and they often lead to corrosion, significant
waste production, and difficult workup. Thus, a chemically
stable magnetic base with the attributes necessary for rapid
quantitative separation would be very useful for industrial
applications, resulting in significant solvent savings, a reduced
amount of time spent, a decreased need for expensive
equipment, and reusability of the separation reagent. Moreover,
a ready-to-separate reagent is interesting for its potential use in
the synthesis of high-quality products, where impurity carryover
defines product performance. Magnetic nanoparticles have
fascinated scientists for several decades and have been used in a
32,33
recently, alkene hydrogenation palladium catalysts
34
called “catch-and-release” systems ) have been developed. In
this work, we present an organic superbase, coupled to
magnetic nanoparticles, with stability amenable for use under
such challenging reaction conditions. We also show that such
an easy-to-separate reagent is useful in a number of
condensation reactions and simplifies workup and product
isolation.
RESULTS AND DISCUSSION
■
Preparation and Characterization of the Magnetic
Base. Magnetic, nanosized DMAN was prepared from
chemically modified magnetic nanoparticles, a linker, and a
derivative of DMAN. For covalent attachment of the base to
the magnetic particle (1), the bifunctional linker hexamethylene
diisocyanate (4) was first reacted with the primary amine of 4′-
1
4
15
plethora of applications, such as for drug delivery, in cancer
1
6
treatments, and as contrast agents for magnetic resonance
1
7
imaging. In chemical synthesis, magnetic nanoparticles have
Received: August 18, 2014
18
recently gained attention in the field of catalysis because they
©
XXXX American Chemical Society
A
dx.doi.org/10.1021/jo501913z | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
a
benzylamine-derivatized, graphene-coated metal nanoparticles
Scheme 1. Synthesis of Immobilized Magnetic Superbase 7
(
(
C/Co@amine) (3) to yield C/Co@hexamethyleneisocyanate
5). This intermediate contains a stable N-alkylurea bond and
an active isocyanate moiety. Subsequently, C/Co@hexamethy-
leneisocyanate (5) was reacted in situ with 4-amino-1,8-
bis(dimethylamino)naphthalene (DMAN-NH , 6), which was
2
35
prepared according to the literature. This afforded a magnetic,
covalently bound DMAN (C/Co@DMAN (7)) in proper yield
−1
with a base capacity of 0.11 ± 0.01 mmol g . For an
illustration of the synthesis, see Scheme 1. The chemical
identity of the herein synthesized reagent C/Co@DMAN (7)
was proven by diffuse reflectance infrared Fourier transform
spectroscopy (DRIFTS) and elemental microanalysis (C, H, N;
Figure S1 and Table S1). The fairly strong urea-stretching
−1
vibration at 1690 cm clearly confirms urea formation in the
case of C/Co@hexamethyleneisocyanate (5). The successful
attachment of DMAN to the isocyanate linker of C/Co@
hexamethyleneisocyanate (5) was investigated with IR spec-
troscopy and elemental microanalysis (Table S1). A substantial
increase in nitrogen content (ΔN = 0.82%) as well as an
increase in carbon content (ΔC = 3.43%) is in line with
successful functionalization. Vibrating sample magnetometry
(
VSM) measurements (at room temperature) showed an
−1
overall magnetization of 139.6 emu g , which is in good
agreement with the expected values, namely, a little lower than
the saturation magnetization of nonfunctionalized C/Co (lower
−1
36
content of organics or carbon; 158 emu g ). It is notable that
the saturation magnetization of this material is still much higher
than even the strongest conventional magnetite−silica nano-
−1
37
particles (30−50 emu g ).
The high saturation magnetization is of relevance because it
38
permits rapid and easy separation. To confirm the chemical
stability and robustness of C/Co@DMAN (7), we recorded
transmission electron microscopy (TEM) images of nanocarrier
system 7 before and after reaction runs. Particles remained
spherical and uniformly sized, in the diameter range of 20 to 60
nm (Figure S2). The reactivity of the herein proposed magnetic
base C/Co@DMAN (7) was compared to reactions using
similar catalysts from the literature. Another reference material,
polystyrene-based resin-supported DMAN (S2), was synthe-
sized (Scheme S2), fully characterized (Figure S3 and Table
S2), and used for comparison.
Reliable recycling and separation of a reagent requires a
robust confirmation of covalent attachment. Therefore, we used
several test reactions (Scheme S1) to prove the covalent nature
of DMAN’s attachment to the nanomagnets. As a first control
experiment, C/Co@hexamethyleneisocyanate (5), which bears
one free, active isocyanate moiety, was reacted with pure (not
aminated) DMAN. The lack of a free amine group in DMAN
does not allow covalent attachment. Using the same synthesis
and purification/washing procedure as for the herein proposed
reagents, product analysis (Table S1) revealed the absence of
both nitrogen incorporation and physisorption effects. As a
separate control experiment, C/Co@hexamethylene-isocyanate
a
C/Co (1) was coupled to 4-aminobenzylamine (2) via diazonium
chemistry to yield C/Co@amine (3), which was further reacted with
hexamethylendiisocyanate (4), resulting in C/Co@hexamethyleneiso-
cyanate (5), bearing an active isocyanate moiety. C/Co@hexamethy-
leneisocyanate (5) was coupled with 4-amino-1,8-bis(dimethylamino)
naphthalene (6) to yield C/Co@DMAN (7).
(5) was first quenched with water, leading to a primary amine
(i.e., the linker was deliberately rendered nonfunctional).
Nonfunctional 5 was then exposed to functionalized, amine-
bearing DMAN derivative 6; product analysis after applying the
same purification/washing procedure as in the preparation of
the functional magnetic base again confirmed the absence of
reactivity (Table S1). These two experiments show that both
the linker and the amination of the DMAN are necessary for
the covalent binding of a DMAN moiety to the nanomagnet.
The experiments further show the absence of base
physisorption, a potential experimental error in such reagent-
anchoring studies.
Knoevenagel and Claisen−Schmidt Condensation.
The literature states that DMAN indeed has catalytic activity
39
in Knoevenagel condensation. After identity confirmation, the
catalytic activity of C/Co@DMAN (7) for the Knoevenagel
B
dx.doi.org/10.1021/jo501913z | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Table 1. Knoevenagel Condensation of Benzaldehyde and Malononitrile under Different Reactions Conditions
a
b
entry
catalyst
catalytic amount
solvent
toluene
time (h)
conversion (%)
63
isolated yield (%)
purity (%)
1
2
3
4
5
6
7
8
9
7
2 mmol %
2 mmol %
2 mmol %
2 mmol %
0.2 mmol %
2 mmol %
2 mmol %
7.5
7.5
7
61
97
89
91
91
99
94
95
96
92
96
98
c
7
H O
2
99 (2)
7
MeOH
93 (85)
98 (95)
92 (90)
99
d
7
MeOH
MeOH
MeOH
4.5
6
d
7
e
f
DMAN
6
63
S2
toluene
MeOH
MeOH
MeOH
6
99
97
g
5.5
7.5
6
30
C/Co
CoCl2
100 mg
5 mg
90
77
g
86
10
33
a
b
Conversion was determined by HPLC with the reference product as the standard (benzaldehyde/malononitrile = 1:1). Isolated yield after
c
d
recrystallization. Numbers in parentheses represent the percent conversion without washing the particles once with 3 mL of toluene. Reaction
carried out using an ultrasound bath. Experiment was carried out using 2 mmol % of the free base. Isolated yield after column chromatography
e
f
g
(
DCM/MeOH = 20:1). No direct product isolation possible without column chromatography.
Table 2. Knoevenagel Condensation with Various Substrates Catalyzed by C/Co@DMAN (7)
a
b
Reaction conditions: catalyst loading, 2 mmol %; basic substrate, 0.4 mmol; aldehyde or ketone, 0.4 mmol; MeOH, 5 mL; 16 h. GC conversion.
c
d
e
After three consecutive cycles, conversion was still 95%. Purity detected by NMR is given in brackets []. No direct product isolation was possible
f
without column chromatography. Isolated yield after recrystallization (DCM/hexane).
reaction of benzaldehyde (8) with malononitrile (9) was
evaluated in different solvents at room temperature (Table 1,
entries 1−3). If toluene was used as a solvent, then only 61%
yield could be obtained. However, in water, quantitative
conversion was achieved after 7.5 h. Still, the catalyst had to
be washed once with toluene to release the product from the
carbene surface. Therefore, methanol was tested as a solvent
with promising results (Table 1, entries 3−5). The use of
ultrasound during catalysis is known to ensure good reagent
dispersion and consistently leads to excellent yields at shorter
contact times. Indeed, when the reaction was conducted in an
ultrasonication bath, a higher yield (91%) was obtained in a
C
dx.doi.org/10.1021/jo501913z | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
shorter time span (Table 1, entry 4). If the catalyst loading was
decreased 10-fold, then similar yields were detected but a
longer reaction time was necessary (Table 1, entry 5).
Alternatively, running the same reaction using free DMAN
where product workup and purification were not considered,
quantitative conversion was obtained after 6 h (Table 1, entry
6). However, column chromatography had to be used to
separate the catalyst residues from the product, which resulted
in a significant drop in isolated yield. Magnetic base C/Co@
DMAN (7) was separated within seconds, and recrystallization
of the crude product (leading to the pure product) was
achieved within minutes. Using polystyrene-immobilized
proton sponge S2 gave good yields as well, but filtration of
the polymer was tedious and not very solvent-efficient. It is
remarkable that even in the absence of any catalysts for the
chosen reaction a low conversion (30%) was detected after 5.5
h (Table 1, entry 8). More surprisingly, the chosen reaction
showed enhanced conversion (90%) in the presence of
nonmodified (i.e., naked carbon surface) nanoparticles (Table
Figure 1. Kinetic plots for the Knoevenagel condensation of
benzaldehyde (1.5 equiv) with malononitrile (1 equiv). All reactions
used 2.5 mmol % 7 unless otherwise specified. Solvents and conditions
for each reaction include ■, methanol and ultrasonication; ●, toluene
1, entry 9). To study further this unexpected finding, we
included additional experiments. One probable hypothesis is
based on leached cobalt ions functioning as a source of activity
and stirring; ▲
, methanol and stirring; and ▼, C/Co instead of 7,
methanol and stirring.
(hypothesis 1). Also plausible is that the carbon surface itself
acts as an active catalyst (hypothesis 2). To exclude the first
hypothesis, the reaction was carried out with a deliberate
addition of cobalt(II) chloride (CoCl ; Table 1, entry 10), the
2
40
species that is supposedly leaching from the particles. The
addition of cobalt salts did not significantly affect the
conversion (Table 1, entry 10; conversion = 33%); therefore,
we do not assign significant activity of ionic cobalt in this kind
of reaction. With regard to the second hypothesis, however, it is
known that carbon-based surfaces (e.g., graphene, carbon
nanotubes (CNT), carbon nanorods (CNR), etc.) indeed have
been reported to support catalytic reactions (so-called
4
1,42
carbocatalysis).
Thus, our finding is in line with similar
carbon-surface activity and is subject to further extended studies
by our group.
In the next part of the investigation, we tested different
substrates (Table 2, entries 1−7). Ethyl cyanoacetate (Table 2,
entry 1) reacted smoothly with benzaldehyde, and the product
(
(
ethyl α-cyanocinnamate) was detected in high yield and purity
96 and 99%, respectively) after 16 h at 40 °C; even after three
Figure 2. Kinetic plots for reactions with different catalysts (2.5 mmol
%) under optimal conditions for each:
■
, 7 in methanol, using
consecutive cycles, the conversion was similarly high (95%).
The less acidic and thus more demanding substrate ethyl
acetoacetate (Table 2, entry 2) was not very active, and only
low yields could be obtained. Also, different aldehydes and
ketones were reacted with malononitrile. 4-Methoxy benzalde-
hyde performed very well (Table 2, entry 2), whereas the
electron-poor nitro substrate was slightly less active (Table 2,
entry 3). Surprisingly, 4-methoxy benzaldehyde reacted quickly,
whereas 4-hydroxy benzaldehyde showed a slow conversion
rate (Table 2, entry 6). An explanation may be the formation of
hydrogen bonds with the solvent; one can imagine a shell of
solvent molecules shielding the substrate from the catalyst.
Nonaromatic substrates such as isobutyraldehyde and cyclo-
hexanone were tested as well (Table 2, entries 6 and 7), and
isobutyraldehyde performed better because of steric reasons
ultrasonication (taken from Figure 1); ●, free DMAN in methanol; ▲,
polymer-bound DMAN (S2) in toluene; ▼, without catalyst in
methanol.
of mixing). To put the performance of C/Co@DMAN (7) in
an appropriate context and to provide a fair comparison, we
measured reaction kinetics for free DMAN (no mass-transfer
limitation expected) and polystyrene-bound superbase S2
(under optimal conditions in toluene) (Figure 2). C/Co@
DMAN (7) initially catalyzes the chosen reaction at least as
well as free DMAN, which leads to comparable or even slightly
higher efficiency after 1 h of reaction time. The conversions
converge with time, leading to high final conversion for all
catalysts. The immobilized magnetic base C/Co@DMAN (7)
can therefore be considered tp be directly competitive with
existing or alternative catalytic systems. The minor changes in
activity (i.e., rapid first part) may be a result of the altered
electronic structure (note that DMAN was derivatized at
position 4 to anchor the nanoparticle).
(i.e., hindrance of nucleophilic attack on ketones).
Kinetic plots (Figures 1 and 2) confirmed the different
reaction rates of C/Co@DMAN (7) when different solvents
and mixing conditions (stirring vs sonication) were used. As
expected, the reaction rate increased when the superbase was
mixed more thoroughly with the reactants (i.e., a higher degree
D
dx.doi.org/10.1021/jo501913z | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
To determine the relevance of and to illustrate a broad
applicability of magnetic organic bases such as C/Co@DMAN
(
7), another type of condensation reaction (Claisen−Schmidt)
between acetophenone and benzaldehyde was tested (Table 3).
Table 3. Claisen−Schmidt Condensation of Acetophenone
and Benzaldehyde
conversion
isolated yield
time
(h)
Co
c
a
b
entry catalyst run
(%)
(%)
(ppm)
1
2
3
4
5
6
7
8
9
7
1
2
3
4
5
1
1
1
1
66
78
71
68
76
<1
58
<1
<1
62 [97.6]
16
20
18
18
20
20
21
21
20
19
16
5
7
d
7
d
7
d
44
64
7
75 [98.6]
DMAN
Co/C
e
e
e
e
42
S2
a
Conversion was determined by HPLC using the reference product as
b
the standard. Isolated yield after solvent evaporation; purity as
detected by NMR is given in brackets []. ICP-OES detection limit for
Co = 0.2 ppb. No product isolation has been carried out. No direct
product isolation was possible without column chromatography.
c
d
e
This well-known reaction yields trans-chalcone, an important
precursor for flavonoids that are widely used in medicine as
4
3
44
45
anti-inflammatory, antiviral, and anticancer agents. Im-
mobilized magnetic base C/Co@DMAN (7) catalyzed the
reaction with a yield of up to 75% and high levels of purity.
Additionally, the same catalyst was reused in five iterative runs
with similar yields. The harsh temperature (130 °C) leads to
the conclusion that C/Co@DMAN (7) indeed remains stable
using covalent-linker systems (Table 3, entries 1−5).
Surprisingly, free DMAN does not catalyze the above
condensation reaction (Table 3, entry 6; yield = 1%). This
behavior is explainable considering the pK values. Because
a
46
acetophenone has a higher pK value (pK = 25) than does
a
a
2
DMAN (pK = 12.1), it can be concluded that the basicity of
a
DMAN is not the key factor governing this catalysis.
Furthermore, this result is in line with earlier work in the
literature. Corma et al. experimentally investigated a similar
case and reported that heterogeneous base catalysts work
47
considerably better than free bases.
Comparison to Known Solid-Supported Catalytic
Base Reagents. In the recent literature, a number of excellent
1
,2,6,13,47,48
examples describe immobilized base catalysts.
Mag-
netic supports have been realized using modified magnetite and
cobalt ferrite with good yields (Table 4, entries 6−7). In
comparison to metal-based C/Co@DMAN (7), the oxidic
supports exhibit a much lower saturation magnetization (133
49
emu/g for C/Co@DMAN (7), ca. 70 emu/g for magnetite ).
This is a direct result of the inherently lower level of
magnetization of most oxides. This value is important because
it correlates directly with separation efficiency/speed and has
recently enabled magnetic metal nanoparticles to purify water
50
on a ton per hour scale. In a typical laboratory setting, the
improved magnetization facilitates workup, as demonstrated
E
dx.doi.org/10.1021/jo501913z | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
using commercially available silica−iron oxide particles and the
herein described metal reagent (Figure 3). The high level of
Figure 4. Immobilization linker stability tests under harsh reaction
conditions overnight. The amount of cleaved linker was determined by
analyzing the nitrogen content of the materials before and after
exposure using quantitative elemental microanalysis. pH 4 buffer, 0.1
M potassium hydrogen phthalate with HCl; pH 9.5 buffer, 0.1 M
carbonate-bicarbonate.
Figure 3. Recovery of organic base bound to magnetic nanoparticles
using a low-cost external magnet. Model system in H O; same particle
2
−1
A number of control reactions were conducted to prove that
the reagent was indeed chemically linked to the nanoparticles,
and these confirmed the importance of linker type and
reliability under relevant, harsh reaction conditions. We further
demonstrated the experimental advantages of magnetic bases in
a number of Knoevenagel condensations, reporting quantitative
conversion and rapid product isolation in minutes. The
demanding Claisen−Schmidt condensation afforded conver-
sions of around 70% at 130 °C. The base catalyst was separated
from the reaction mixtures in less than 1 min and could be
recycled through at least five iterative runs. A quantitative
comparison between siloxane linkers used in many immobiliza-
tion studies and the herein described carbon shell-based
chemistry revealed the important role of linker stability when
designing reliable reagents for rough, everyday laboratory use.
concentration as used in all experiments (0.7 mg mL ). A, metal-
based reagents derived from Co/C; B, silica-coated Fe O . Note the
3
4
different times above and below.
magnetization of the carbon-coated cobalt allows clean,
quantitative separation of the organic base within seconds.
Established nonmagnetic immobilized catalyst supports are
frequently based on zeolites and silica beads. The active site/
5
1,52
moiety is predominantly attached by a siloxane-type linker.
These catalytic systems usually show excellent performance and
reusability (Table 4, entries 2−4, 9−10). In the case of harsh
reaction conditions (e.g., strong base), the weakest of these
otherwise well-performing materials are the siloxane-type
moieties. It is well known that extremely acidic or basic
environments (i.e., pH <4 or >9) hydrolyze silane linkers. In
contrast, the herein presented dialkyl urea moiety exhibits a
high stability over a broad range of pH.
EXPERIMENTAL SECTION
■
General Experimental Details. Carbon-coated cobalt nano-
To confirm this statement experimentally, we directly
compared siloxane linkers to the herein used C/Co@DMAN
2
0
particles were suspended by the use of an ultrasonic bath and
subjected to various reactions. After a reaction or a pretreatment step,
nanoparticles were recovered from the reaction mixture with the aid of
a conventional magnet (neodymium-based magnet, side length = 12
mm). Silica-coated magnetite nanoparticles (350 nm) were purchased
from a chemical supplier. Commercially available reagents were used
as received. All air-sensitive reactions were carried out under an argon
atmosphere. The nanoparticles were analyzed by FT-IR spectroscopy
(
7) in a number of representative acidic and basic environ-
ments. After 24 h at room temperature and pH 4, more than
0% of the siloxane linkers were cleaved, but less than 5% of the
2
urea-type linkers were cleaved (Figure 4). At pH 9.5, the dialkyl
urea linkers afforded lower stability (about 10% cleaved), but it
is notable that under the same conditions >75% of the siloxane
linkers were cleaved. Furthermore, the stability of these linkers
was tested at elevated temperatures in different solvents
(
5% in KBr) and elemental microanalysis. HPLC measurements were
2
performed (particle size, 5 μm; column size, 2.1 × 150 mm ) to
determine the conversion of the catalytic reactions. GC−MS
(
toluene and water). The dialkyl urea linkers performance
2
measurements were performed (capillary, 30 m × 250 × 0.5 μm )
under these conditions was superior to that of their silane-based
pendants (Figure 4). The intrinsic stability of these particles in
acidic or basic media has already been investigated and has
also been proven to be superior to that of silica-coated
magnetite.
with split injection (250 °C; ratio = 50:1; injection volume, 1 μL) and
−1
a temperature program (80 °C for 2 min; increase at 20 °C min until
19
250 °C). The cobalt concentration present in solution was measured
by inductively coupled plasma−atomic emission spectroscopy (ICP-
OES).
Synthesis of C/Co@amine (3). Carbon-coated cobalt nano-
particles (1) were purchased from a chemical supplier. The particles
were functionalized according to reported procedures (i.e., 3 g of 1 was
CONCLUSIONS
■
We have described the preparation of a covalently bound,
chemically stable, magnetic nanoparticle-immobilized organic
superbase and its extensive characterization using quantitative
elemental microanalysis, infrared spectroscopy, vibrating
sample magnetometry, and transmission electron microscopy.
suspended in 20 mL of H O, and 4-aminobenzylamine (2) (0.7 mL,
2
5
.3 mmol) were added). Then sodium nitrite (0.8 g, 11.5 mmol) was
added to the slurry. The vessel was put in the sonication bath, and 2
mL of concentrated HCl was slowly added dropwise. Upon
completion of the reaction (30 min), the nanoparticles were recovered
F
dx.doi.org/10.1021/jo501913z | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
from the reaction mixture with a magnet and washed with toluene (2
General Procedure for Claisen−Schmidt Condensations. A
solvent-free mixture of acetophenone (2 mL) and C/Co@DMAN (7,
0.1 g, 0.01 mmol) was sonicated in an ultrasonication bath for 5 min
and then stirred at 130 °C under a nitrogen atmosphere. Benzaldehyde
(0.1 mmol) was added, and the progress of the reaction was monitored
by HPLC (MS to ensure product identity, UV at 280 nm for
quantification) using a commercially available product as a reference.
×
10 mL), EtOH (1 × 10 mL), and acetone (2 × 10 mL) and dried for
−1
2
4 h at 50 °C in vacuo. FT-IR: 1603, 1503, 1015, 831 cm .
Synthesis of C/Co@hexamethyleneisocyanate (5). C/Co@
amine (3, 2 g) was degassed three times in a Schlenk flask. Fifty
milliliters of dry DMF, hexamethylendiisocyanate (HDMI) (6 mL, 37
mmol, 97 equiv, 98%), and triethylamine (NEt , 0.01 mL, 99%) were
3
added under an argon atmosphere. Then the solution was dispersed in
a sonication bath for 5 min. The dispersion was heated to 70 °C and
stirred for 6 h. Upon completion of the reaction, the nanobeads were
recovered from the reaction mixture with a magnet and washed with
1,3-Diphenyl-2-propen-1-one (13). Isolated product was obtained
1
by evaporation in high vacuum at 65 °C. H NMR (CDCl , 200 MHz,
3
2
5 °C): δ 7.93−7.97 (m, 2H), 7.18−7.71 (m, 10H).
General Washing/Drying Procedure for Catalyst Recovery.
−1
Used catalyst was washed by sonication (5 min) with the reaction
solvent of the catalytic reaction and acetone (2×) and dried for 4 h at
anhydrous DMF (2×). FT-IR: 2927, 2857, 1689, 1553, 1018 cm .
Synthesis of C/Co@DMAN (7). Freshly synthesized C/Co@
hexamethyleneisocyanate (5, 2 g) was used directly in a 250 mL
Schlenk flask under an argon atmosphere. Forty milliliters of dry DMF
5
0 °C in vacuo.
General Procedure for Linker Stability Tests. A sample (30
mg) was placed in a 10 mL round-bottomed flask, and solvent (10
mL) was added. The solution was dispersed via ultrasonication bath
and then shaken overnight. The solid sample was washed by either
sonication (magnetic samples) or filtration (silica samples) with
and triethylamine (NEt , 0.01 mL, 99%) were added. The solution was
3
dispersed in a sonication bath for 5 min. 4-Amino-1,8-bis-
(
dimethylamino)naphthalene (DMAN-NH (6), 1 g, 4.3 mmol) was
2
degassed three times in a Schlenk flask, dissolved in 10 mL of
anhydrous DMF, and added dropwise to the C/Co-HMDI solution.
The dispersion was heated to 35 °C and stirred overnight (16 h).
Upon completion of the reaction, the nanoparticles were recovered
from the reaction mixture with a magnet, washed with DMF (2 × 10
mL), EtOH (1 × 10 mL), acetone (2 × 10 mL), and dried for 24 h at
EtOH, H O, and acetone and dried for 4 h at 50 °C in vacuo.
2
ASSOCIATED CONTENT
Supporting Information
■
*
S
Spectroscopic data, elemental analysis data, IR spectra, TEM
and SEM micrographs of the magnetic and polymeric
compounds, scheme depicting the test reactions carried out
to verify covalent linkage, and linker stability tests with a
duration of 1 week. This material is available free of charge via
the Internet at http://pubs.acs.org.
−1
5
0 °C in vacuo. FT-IR: 2931, 2857, 2783, 1676, 1532, 1249 cm .
Synthesis of Polystyrene-Supported DMAN (S2). Chloro-
methyl polystyrene (1 g, 2% DVB, 100−200 mesh, 0.9−1.5 mmol/g,)
was used in a 50 mL Schlenk flask. Twenty milliliters of anhydrous
DMF and triethylamine (NEt , 0.01 mL, 99%) were added under an
3
argon atmosphere. DMAN-NH (6) (0.6 g, 2.58 mmol) was degassed
2
three times in a Schlenk flask, dissolved in 5 mL of anhydrous DMF,
and added dropwise to the polystyrene slurry. The dispersion was
heated to 40 °C and stirred overnight (20 h). Upon completion of the
reaction, the functionalized polymer was filtered and intensively
AUTHOR INFORMATION
Corresponding Author
*E-mail: wstark@ethz.ch. Web: www.fml.ethz.ch. Fax: +41 44
633 10 83. Tel: +41 44 632 09 80.
Notes
The authors declare the following competing financial
interest(s): W.J.S. and R.N.G. declare financial interests,
because they are shareholders of TurboBeads LLC, a company
active in magnetic nanoparticles.
■
washed with acetone, 0.1 M NaOH, H O, EtOH, and acetone and
2
dried for 24 h at 50 °C in vacuo. FT-IR: 3030, 2927, 2775, 1944, 1874,
−
1
1
726, 1602 cm .
Synthesis of Silica-Amine (S4). Silica gel 230−400 mesh (1 g)
was placed in a 50 mL Schlenk flask and degassed three times. Twenty
milliliters of anhydrous toluene was added, and the slurry was heated
to 60 °C. N1-(2-Aminoethyl)-N2-(3-(trimethoxysilyl)propyl)ethane-
1
,2-diamine (0.3 mL, 11.6 mmol) was added dropwise to the solution,
which then was stirred for 24 h. The resulting solid was filtered with 50
mL of toluene, 100 mL of DCM, and 20 mL of Et O and dried for 24
ACKNOWLEDGMENTS
■
2
We thank Andreas Dutly and Nikita Kobert for GC
measurements, ICB/ETH Zurich, the EU-ITN network
Mag(net)icFun (PITN-GA-2012-290248), and the Swiss
National Science Foundation (no. 200021-150179) for financial
support.
−1
h at 50 °C in vacuo. FT-IR: 3289, 2940, 2828, 1863, 1474, 1109 cm .
General Procedure for Knoevenagel Condensation Reac-
tions. A mixture of solvent (5 mL) and C/Co@DMAN (7, 0.1 g, 0.01
mmol) was sonicated in an ultrasonication bath for 5 min in a Schlenk
flask and then stirred at room temperature. Benzaldehyde (0.5 mmol)
and malononitrile (0.5 mmol) were added, and the progress of the
reaction was monitored by HPLC (MS to ensure product identity, UV
at 280 nm for quantification) or GC-FID using a commercially
available product as a reference.
REFERENCES
■
(
1) Corma, A.; Iborra, S.; Rodriguez, I.; Sanchez, F. J. Catal. 2002,
211, 208.
(2) Gianotti, E.; Diaz, U.; Velty, A.; Corma, A. Eur. J. Inorg. Chem.
2012, 5175.
(3) Phan, N. T. S.; Jones, C. W. J. Mol. Catal. A: Chem. 2006, 253,
123.
(4) Alder, R. W.; Bryce, M. R.; Goode, N. C.; Miller, N.; Owen, J. J.
Chem. Soc., Perkin Trans. 1 1981, 2840.
(5) Rondinone, A. J.; Samia, A. C. S.; Zhang, Z. J. J. Phys. Chem. B
1999, 103, 6876.
Benzylidenemalononitrile (10). Isolated product was obtained by
1
recrystallization in hot EtOH and hexane, followed by filtration. H
NMR (CDCl , 200 MHz, 25 °C): δ 7.86−7.82 (m, 2H), 7.71 (s, 1H),
7
3
.57−7.43 (m, 3H).
α-Cyanocinnamic Acid Ethyl Ester. Pure isolated product was
1
obtained by evaporation at 40 °C. H NMR (CDCl , 200 MHz, 25
3
°
2
C): δ 8.19 (s, 1H), 7.9 (m, 2H), 7.47−7.43 (m, 3H), 4.37−4.27 (q,
H), 1.37−1.30 (t, 3H).
-(4-Methoxybenzylidene)malononitrile. Pure isolated product
2
(6) Zhang, Y.; Zhao, Y.; Xia, C. J. Mol. Catal. A: Chem. 2009, 306,
107.
1
was obtained by evaporation at 40 °C. H NMR (CDCl , 200 MHz,
3
2
(
5 °C): δ 7.86−7.75 (m, 2H), 7.58 (s, 1H), 6.96−6.92 (m, 2H), 3.84
s, 3H).
-(4-Nitrobenzylidene)malononitrile. Isolated product was ob-
(7) Ishikawa, T. Superbases for Organic Synthesis; John Wiley & Sons:
Chichester, U.K., 2009; p 1.
(8) Gwaltney, S. L.; Sakata, S. T.; Shea, K. J. J. Org. Chem. 1996, 61,
2
tained by evaporation of the solvent, followed by recrystallization in
7438.
1
DCM/hexane. H NMR (CDCl , 200 MHz, 25 °C): δ 8.73 (s, 1H),
(9) Ireland, R. E.; Liu, L. B.; Roper, T. D.; Gleason, J. L. Tetrahedron
1997, 53, 13257.
3
8
.45−8.42 (m, 2H), 6.96−6.92 (m, 2H).
G
dx.doi.org/10.1021/jo501913z | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(
10) Nielsen, M.; Jacobsen, C. B.; Jørgensen, K. A. Angew. Chem., Int.
(48) Karaog
Mater. Res. Bull. 2012, 47, 2480.
(49) Goya, G. F.; Berquo, T. S.; Fonseca, F. C.; Morales, M. P. J.
Appl. Phys. 2003, 94, 3520.
(50) Rossier, M.; Schreier, M.; Krebs, U.; Aeschlimann, B.; Fuhrer,
R.; Zeltner, M.; Grass, R. N.; Gunther, D.; Stark, W. J. Sep. Purif.
Technol. 2012, 96, 68.
̆
̧
enel, M.; So
̈
zeri, H.; Toprak, M. S.
Ed. 2011, 50, 3211.
(
(
11) Tur, F.; Saa,
́
J. M. Org. Lett. 2007, 9, 5079.
12) Mori, K.; Kanai, S.; Hara, T.; Mizugaki, T.; Ebitani, K.;
Jitsukawa, K.; Kaneda, K. Chem. Mater. 2007, 19, 1249.
13) Luo, S.; Zheng, X.; Xu, H.; Mi, X.; Zhang, L.; Cheng, J.-P. Adv.
Synth. Catal. 2007, 349, 2431.
(
̈
(
14) Gao, J. H.; Gu, H. W.; Xu, B. Acc. Chem. Res. 2009, 42, 1097.
(51) Corma, A.; Garcia, H. Adv. Synth. Catal. 2006, 348, 1391.
(52) Han, Y.-J.; Stucky, G. D.; Butler, A. J. Am. Chem. Soc. 1999, 121,
9897.
(
15) Xu, Z. P.; Zeng, Q. H.; Lu, G. Q.; Yu, A. B. Chem. Eng. Sci. 2006,
6
(
1, 1027.
16) Pankhurst, Q. A.; Connolly, J.; Jones, S. K.; Dobson, J. J. Phys.
D: Appl. Phys. 2003, 36, R167.
17) Huh, Y.-M.; Jun, Y.-w.; Song, H.-T.; Kim, S.; Choi, J.-s.; Lee, J.-
(
H.; Yoon, S.; Kim, K.-S.; Shin, J.-S.; Suh, J.-S.; Cheon, J. J. Am. Chem.
Soc. 2005, 127, 12387.
(
8
(
18) Schat
950.
19) Schaetz, A.; Zeltner, M.; Michl, T. D.; Rossier, M.; Fuhrer, R.;
Stark, W. J. Chem.Eur. J. 2011, 17, 10565.
20) Grass, R. N.; Athanassiou, E. K.; Stark, W. J. Angew. Chem., Int.
Ed. 2007, 46, 4909.
̈
z, A.; Reiser, O.; Stark, W. J. Chem.Eur. J. 2010, 16,
(
(
(
21) Baruwati, B.; Guin, D.; Manorama, S. V. Org. Lett. 2007, 9, 5377.
22) Gleeson, O.; Tekoriute, R.; Gun’ko, Y. K.; Connon, S. J.
Chem.Eur. J. 2009, 15, 5669.
23) Keller, M.; Colliere, V.; Reiser, O.; Caminade, A.-M.; Majoral, J.-
P.; Ouali, A. Angew. Chem., Int. Ed. 2013, 52, 3626.
24) Polshettiwar, V.; Baruwati, B.; Varma, R. S. Chem. Commun.
009, 1837.
25) Riente, P.; Yadav, J.; Pericas, M. A. Org. Lett. 2012, 14, 3668.
26) Schatz, A.; Long, T. R.; Grass, R. N.; Stark, W. J.; Hanson, P. R.;
Reiser, O. Adv. Funct. Mater. 2010, 20, 4323.
27) Stevens, P. D.; Fan, J. D.; Gardimalla, H. M. R.; Yen, M.; Gao, Y.
Org. Lett. 2005, 7, 2085.
28) Stevens, P. D.; Li, G. F.; Fan, J. D.; Yen, M.; Gao, Y. Chem.
Commun. 2005, 4435.
29) Zeltner, M.; Schaetz, A.; Hefti, M. L.; Stark, W. J. J. Mater. Chem.
011, 21, 2991.
30) Cantillo, D.; Moghaddam, M. M.; Kappe, C. O. J. Org. Chem.
013, 78, 4530.
(
̀
(
2
(
(
̈
(
(
(
2
(
2
(
31) Zheng, Y.; Stevens, P. D.; Gao, Y. J. Org. Chem. 2005, 71, 537.
(
32) Kainz, Q. M.; Linhardt, R.; Grass, R. N.; Vile, G.; Perez-Ramirez,
J.; Stark, W. J.; Reiser, O. Adv. Funct. Mater. 2014, 24, 2020.
33) Linhardt, R.; Kainz, Q. M.; Grass, R. N.; Stark, W. J.; Reiser, O.
RSC Adv. 2014, 4, 8541.
34) Wittmann, S.; Schat
Angew. Chem., Int. Ed. 2010, 49, 1867.
35) Ozeryanskii, V. A.; Pozharskii, A. F. Russ. Chem. Bull. 1997, 46,
437.
36) Zeltner, M.; Grass, R. N.; Schaetz, A.; Bubenhofer, S. B.;
Luechinger, N. A.; Stark, W. J. J. Mater. Chem. 2012, 22, 12064.
37) Lu, A. H.; Salabas, E. L.; Schuth, F. Angew. Chem., Int. Ed. 2007,
6, 1222.
38) Schumacher, C. M.; Herrmann, I. K.; Bubenhofer, S. B.;
(
(
̈
z, A.; Grass, R. N.; Stark, W. J.; Reiser, O.
(
1
(
(
4
(
Gschwind, S.; Hirt, A.-M.; Beck-Schimmer, B.; Guenther, D.; Stark, W.
J. Adv. Funct. Mater. 2013, 23, 4888.
(
39) Rodriguez, I.; Sastre, G.; Corma, A.; Iborra, S. J. Catal. 1999,
83, 14.
40) Schumacher, C. M.; Grass, R. N.; Rossier, M.; Athanassiou, E.
K.; Stark, W. J. Langmuir 2012, 28, 4565.
1
(
(41) Machado, B. F.; Serp, P. Catal. Sci. Technol. 2012, 2, 54.
(42) Schaetz, A.; Zeltner, M.; Stark, W. J. ACS Catal. 2012, 2, 1267.
(43) Gomes, A.; Couto, D.; Alves, A.; Dias, I.; Freitas, M.; Porto, G.;
Duarte, J. A.; Fernandes, E. BioFactors 2012, 38, 378.
44) Cushnie, T. P. T.; Lamb, A. J. Int. J. Antimicrob. Agents 2005, 26,
43.
(
3
(
(
(
45) Woo, H. D.; Kim, J. PLoS One 2013, 8, e75604.
46) Novak, M.; Loudon, G. M. J. Org. Chem. 1977, 42, 2494.
47) Climent, M. J.; Corma, A.; Iborra, S.; Primo, J. J. Catal. 1995,
1
51, 60.
H
dx.doi.org/10.1021/jo501913z | J. Org. Chem. XXXX, XXX, XXX−XXX