Enzyme-inspired functional surfactant for aerobic oxidation of activated alcohols to aldehydes in water

Article abstract of DOI:10.1021/cs5020018

We describe an enzyme-inspired catalytic system based on a rationally designed multifunctional amphiphile. The resulting micelles feature metal-binding sites and stable free radical moieties as well as fluorous pockets that attract and preconcentrate molecular oxygen. In the presence of copper ions, the micelles effect chemoselective aerobic alcohol oxidation under ambient conditions in water, a transformation that is challenging to achieve nonenzymatically.

Full text of DOI:10.1021/cs5020018

Letter
pubs.acs.org/acscatalysis
Enzyme-Inspired Functional Surfactant for Aerobic Oxidation
of Activated Alcohols to Aldehydes in Water
Ba-Tian Chen,† Konstantin V. Bukhryakov,† Rachid Sougrat,‡ and Valentin O. Rodionov*,†
‡
†KAUST Catalysis Center and Division of Physical Sciences and Engineering, Imaging and Characterization Lab, King Abdullah
University of Science and Technology, Thuwal, 23955-6900, Kingdom of Saudi Arabia
S
* Supporting Information
ABSTRACT: We describe an enzyme-inspired catalytic system based on a rationally
designed multifunctional amphiphile. The resulting micelles feature metal-binding sites
and stable free radical moieties as well as fluorous pockets that attract and preconcentrate
molecular oxygen. In the presence of copper ions, the micelles effect chemoselective
aerobic alcohol oxidation under ambient conditions in water, a transformation that is
challenging to achieve nonenzymatically.
KEYWORDS: micellar catalysis, oxidation, copper, TEMPO, fluorous surfactant
Cu/TEMPO-catalyzed alcohol oxidations rely on organic
ature’s enzymes are extremely efficient catalysts. Their
Nremarkable properties result from precise folding of solvents, especially favoring acetonitrile.13 The transfer of this
catalytic system to water could be advantageous from a green
chemistry standpoint (especially if the aqueous media could be
reused).14 The use of water for performing oxidations will also
circumvent the usual safety concerns associated with combining
oxygen or air with flammable organic solvents. Several groups
have reported attempts to introduce Cu/TEMPO and related
systems into water and water−organic mixtures. Typically,
heating, elevated O2 pressure, special strategies to increase
dissolved O2 concentration, and the presence of water-soluble
ligands were required.15,16 A recent report highlighted the use
of a nonionic surfactant to perform the oxidations of activated
alcohols under mild conditions.17
To improve upon the state of the art, we have aimed to
design a functional surfactant architecture incorporating both
TEMPO moieties and Cu-binding sites. We envisioned that a
more efficient catalyst could result if a high local concentration of
O2 were to be created in the vicinity of the metal and free radical
sites. Because perfluorocarbons and their aqueous dispersions are
capable of dissolving significant amounts of molecular oxygen,18
we chose a fluorous hydrophobic “tail” as another desirable
structural feature. As an added benefit, an “everything-phobic”
perfluorocarbon moiety is expected to prevent the accumulation
of the starting materials/products within the micelles, thus
improving the accessibility of the active sites and preventing
undesirable overoxidation. Previously, the fluorous tagging
strategy19 has been used in Cu/TEMPO alcohol oxidations in
organic solvents20 as well as in biphasic ClO−/TEMPO alcohol
oxidations.21 The structure of 1, the surfactant we chose to
explore, is shown in Figure 1. It was readily prepared in three
constituent polypeptide chains; preorganization of the local
environment and functional groups around the catalytic sites;
and participation of metal ions, prosthetic groups, and
cofactors.1 The active sites and cofactor binding pockets are
often isolated from the bulk solvent and buried in the hydro-
phobic interior of the enzyme.
Enzyme-inspired nanoscale catalysts,2 including dendrimers,3
functional polymers,4 and supramolecular assemblies,5 have
been the subject of intensive investigation. Like enzymes, many
of these chemist-designed catalysts feature spatially preorgan-
ized functional groups, hydrophobic interiors/hydrophilic
exteriors, and catalytic sites isolated from the bulk solvent.
Micelles, vesicles, and emulsion droplets are some of the simplest
and most versatile systems of this kind.6,7 They spontaneously
self-assemble from amphiphilic small molecules, which can be
tailored using the formidable arsenal of organic chemistry.
Hydrophobic catalysts can be easily solubilized inside micelle
interiors.8 Alternatively, catalytic moieties, either metal-based9 or
organic,10 can be incorporated into the amphiphile structures.6
Here, we describe an enzyme-inspired catalytic system based on
a rationally designed multifunctional amphiphile. The resulting
micelles feature metal-binding sites and stable free radical
moieties as well as fluorous pockets that attract and precon-
centrate molecular oxygen. In the presence of Cu ions, the
micelles effect chemoselective aerobic alcohol oxidation under
ambient conditions in water, a transformation that is challenging
to achieve nonenzymatically. A related catalytic system based on
oxygen preconcentrating block copolymer micelles has been
reported recently by us.11
The combination of Cu and TEMPO (2,2,6,6-tetramethyl-1-
piperidine-N-oxyl) has emerged as one of the most attractive
catalytic systems for selective aerobic oxidation of primary
alcohols to aldehydes.12,13 The established protocols for
Received: October 15, 2014
Revised: January 15, 2015
Published: January 20, 2015
© 2015 American Chemical Society
1313
DOI: 10.1021/cs5020018
ACS Catal. 2015, 5, 1313−1317

ACS Catalysis
Letter
To establish the baseline for Cu/TEMPO reactivity in
aqueous and micellar media, we initiated our study by
performing a range of model aerobic oxidation reactions in
the presence of common surfactants. The conditions were
selected to be similar to well-established organic solvent
protocols.13 Data for selected reactions is summarized in
Table 1 (see the Supporting Information (SI), Sections 3−4 for
the full set of experiments). The nature of the base and
polydentate ligand added (if any) has a profound effect on the
outcome of the reactions in organic solvents.22,16,23 We found
that the same holds true for water and micellar media. The
nature and amount of surfactant present had an equally
significant impact on reactivity. Generally, we found the
catalytic system is less well-behaved in water than in organic
solvents.
Although TEMPO is moderately soluble in water/benzyl
alcohol mixtures, the aerobic oxidation of the alcohol is
relatively slow in the absence of surfactants (Table 1, entry 1,
and Figure S5, SI) and does not proceed to completion. In the
presence of sodium dodecyl sulfate (SDS), the rate improves
markedly at the cost of a significant amount of overoxidation
(Table 1, entry 2). The overoxidation can be prevented by intro-
ducing 2,2′-bipyridine (bpy) (Table 1, entry 4; see Figure S4, SI,
for additional ligands). However, this necessitates the addition of
50 mol % of DMAP base (Table 1, entries 3 and 4, and Figure S1,
SI). DMAP was found to be an optimal base/cocatalyst, with
other bases we tried, resulting in low conversion, poor selectivity,
or both (Table 1, entries 8−12, and Figure S3, SI).
It is important to note that a high concentration of SDS
(100 mM, 12 times higher than the critical micellization
concentration, CMC) is necessary for the CuSO4/TEMPO/
bpy system to function effectively (Table 1, entries 4 and 5;
Table S2, entries 19−22, SI; Figure S6, SI). Anionic SDS and
the neutral Triton X-100 surfactants outperform the cationic
cetyltrimethylammonium bromide (CTAB), even though the
latter features a much lower CMC than SDS. This does suggest
Figure 1. Design of a trifunctional surfactant for aerobic oxidation of
alcohols in water. (A) The proposed design for self-assembled fluorous
“nanoreactors”. (B) Synthesis of TEMPO derivatives 1, 5, and 6.
steps, starting with commercially available amino-TEMPO. We
also synthesized compounds 5 and 6, each featuring only two of
1’s three functional groups.
a
Table 1. Aerobic Oxidations of Benzyl Alcohol Catalyzed by a Variety of Cu/TEMPO/Amphiphile Systems
d
bd
,
c
c
entry
surfactant (mM)
TEMPO derivative (mol %)
base (mol %)
ligand
CuSO4 (mol %)
O2 source
3a (%)
4a (%)
1
2
TEMPO/5%
TEMPO/5%
TEMPO/5%
TEMPO/5%
TEMPO/5%
TEMPO/5%
TEMPO/5%
TEMPO/5%
TEMPO/5%
TEMPO/5%
TEMPO/5%
TEMPO/5%
5·K/5%
DMAP/20%
DMAP/20%
DMAP/10%
DMAP/50%
DMAP/50%
DMAP/50%
DMAP/50%
NMI/50%
2
2
5
5
5
5
5
5
5
5
5
5
2
2
2
O2
O2
air
air
air
air
air
air
air
air
air
air
O2
O2
air
O2
48
83
4
4
SDS/100 mM
SDS/100 mM
SDS/100 mM
SDS/20 mM
10
3
3
bpy
bpy
bpy
bpy
bpy
bpy
bpy
bpy
bpy
bpy
3
0.5
4
100
5
57 1.5
6
TX-100/100 mM
CTAB/100 mM
SDS/100 mM
SDS/100 mM
SDS/100 mM
SDS/100 mM
SDS/100 mM
92
40
5.5
5.5
31
35
62
26
39
1
1
7
8
1
9
DABCO/50%
KOH/50%
1
10
11
12
13
14
15
16
7
12
9
3
DIPEA/50%
DBU/50%
2
1
1
4
19
DMAP/20%
DMAP/20%
DMAP/20%
DMAP/20%
1
6/5%
4
1/5 mM
1/5 mM
1/5%
93
2
1/5%
2
98
a
b
Reaction conditions: 0.5 mmol of BnOH in 5 mL of water, 25 °C, magnetic stirring, 1 h. Ligand:Cu ratio has been set to 1 for all ligands.
c
d
Determined by 1H NMR. Each result is an average from three independent experiments. DMAP, 4-dimethylaminopyridine; NMI, N-methylimidazole;
DABCO, 1,4-diazabicyclo[2.2.2]octane; DBU, 1,8-diazabicyclo[5.4.0]undec-7-ene; DIPEA, N,N-diisopropylethylamine; bpy, 2,2′-bipyridine.
1314
DOI: 10.1021/cs5020018
ACS Catal. 2015, 5, 1313−1317

ACS Catalysis
Letter
that surfactants may play a role in catalysis beyond simple
solubilization of hydrophobic TEMPO or bpy/Cu.
Table 2. Scope of Alcohol Oxidations Catalyzed by 1 in
Water
a
Next, we examined the catalytic activity of compounds 5
and 6. The hydrophobic, poorly soluble 6 was less active than
parent TEMPO (Table 1, entry 14). The readily soluble
potassium salt of 5 (5·K) was even less active: only 26% yield
was realized after an hour, and the reaction did not proceed to
completion. We attribute this loss of activity to the poor
accessibility of the hydrophobic benzyl alcohol substrate by
the highly hydrophilic catalyst. Although the reaction mixture
appears homogeneous, its phase behavior is known to be
complex, with up to three distinct phases coexisting.24 With 5·K
and Cu2+ confined to different phases, we expect the reactivity
will be impacted.
To our delight, we found that functional surfactant 1 was an
efficient and selective catalyst for alcohol oxidations. The
oxidation of benzyl alcohol proceeded to completion in the
presence of 5 mol % of 1, with significantly lower loadings of
DMAP (20 mol %) and CuSO4 (2 mol %) than were necessary
for reactions in SDS micellar media (Table 1, entry 16). No
additional ligand was necessary to avoid overoxidation. The
oxidations proceeded well with both air and pure O2 (Table 1,
entries 15, 16). Even lower amounts of CuSO4, down to 0.2 mol %,
could be used to effect the oxidation reactions, at the cost of
slower reaction rates (Table S3, SI). We found the parent
TEMPO/CuSO4 system in water loses its catalytic competency
completely with low Cu loadings.
Under the optimized conditions, 1 effectively catalyzes
oxidations of a wide range of primary alcohols to their cor-
responding aldehydes (Table 2). The aldehyde products could
be readily isolated in all cases by extracting the aqueous mixture
with ethyl ether, followed by a filtration through a pad of silica
gel. Although the oxidations proceed nearly as well with air, we
performed the experiments using pure O2 because of the easier
handling/better reproducibility achievable with our experimen-
tal setup.
Benzylic and heteroaromatic alcohols were somewhat more
reactive (Table 2, entries 1−9) than allylic substrates (Table 2,
entries 11−14). Most benzylic alcohols could be completely
oxidized within 1−3 h, whereas 20−24 h was required for allylic
substrates. The transformations of 2e, 2g, and 2h were more
sluggish, possibly because of the competing coordination of the
phenol, amine, and furyl groups with copper (Table 2, entries 5,
7, and 8). Consequently, only middling conversions were
achieved in 24 h. It is worth mentioning that the sulfur-
containing 2i could be readily oxidized (Table 2, entry 9).
Cinnamyl alcohol was found to be extraordinarily reactive, with
the reaction complete in less than an hour. For geraniol and
nerol, Z → E isomerization was observed (Table 2, entries 12
and 13). While this type of isomerization is facile for geranial
and neral, it has also been observed for Z-enals in Cu/TEMPO-
catalyzed oxidations in the presence of DMAP.25
Substrates lacking unsaturation a bond away from the OH
group could not be oxidized (Table 2, entries 15−17).
Although benzylic and allylic alcohols are known to be the
more active substrates in the established organic solvent
protocols, they can be oxidized in acetonitrile. We attribute the
difference in reactivity to a change in Cu speciation between
acetonitrile and water/micellar media.
a
Reaction conditions: 0.5 mmol of alcohol, 5 mol % 1, 2 mol %
b
CuSO4, 20 mol % DMAP, 5 mL of H2O, 25 °C, 1 atm O2. Isolated
yield. Determined by H NMR.
c
1
To test our oxygen preconcentration hypothesis, we
examined the solubility of oxygen in the aqueous solutions of
1 and SDS as well as the kinetics of O2 release from oxygen-
oversaturated solutions. Deionized water, as well as solutions of
SDS and 1, were stirred vigorously in vials filled with pure O2.
The vials were opened to air, and the evolution of dissolved
oxygen (DO) concentration was followed using an Inlab
1315
DOI: 10.1021/cs5020018
ACS Catal. 2015, 5, 1313−1317

ACS Catalysis
Letter
aqueous micellar media. Although this chemistry could be
coerced into water with the help of SDS, the optimal reaction
conditions were found to be impractical. To address this
problem, we synthesized an enzyme-inspired, trifunctional
surfactant 1 that is catalytically competent at practical,
millimolar concentrations. Unlike the TEMPO/SDS combina-
tion, 1 does not require additional Cu ligands or a high loading
of base. There is evidence for O2 preconcentration within the
fluorous cores of the micelles formed by 1. Investigations of
functional soft materials capable of sequestering O2 and other
gases for reactions in liquid phase are under way in our
laboratories.
ASSOCIATED CONTENT
■
S
* Supporting Information
The following file is available free of charge on the ACS
Figure 2. Kinetics of O2 release from O2-oversaturated solutions.
Deionized water, black markers; 14 mM SDS, green markers; 5 mM 1,
purple markers.
Publications website at DOI: 10.1021/cs5020018
Experimental details for organic synthesis, character-
ization (1H, 13C, and 19F NMR) and catalytic studies
(PDF)
605 immersion probe (Mettler Toledo) (Figure 2). We found
that O2-oversaturated solutions of SDS and 1 retain similar
amounts of O2 (∼28 mg·L−1), measurably more than pure
water. However, for both pure water and SDS solution, the DO
concentration dropped to its air-saturated value of ∼9 mg·L−1
within 20 min. As SDS solutions foam, the slightly slower O2
release (and larger error of DO determination) can be
attributed to gas retention within foam bubbles. In contrast,
DO concentration in a dilute 5 mM solution of 1 is ∼25 mg·L−1
20 min after venting the headspace oxygen. That oxygen must
reside in the aggregates of 1 in the vicinity of TEMPO and Cu
catalytic sites.
AUTHOR INFORMATION
■
Corresponding Author
*Phone: +966 (12) 8084592. E-mail: valentin.rodionov@kaust.
edu.sa.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We gained further insight into the structure of our catalytic
system using cryogenic transmission electron microscopy
(cryo-TEM) images of the aggregates of 1 (Figure 3). In
The authors are grateful to Dr. Virginia Unkefer and Ms. Sarah
Almahdali for their help in preparing the manuscript, and to
Prof. Mathias Christmann for a helpful discussion. This
research was supported by King Abdullah University of Science
and Technology.
REFERENCES
■
(1) Zhang, X.; Houk, K. Acc. Chem. Res. 2005, 38, 379−385.
(2) (a) Raynal, M.; Ballester, P.; Vidal-Ferran, A.; van Leeuwen, P.
Chem. Soc. Rev. 2014, 43, 1734−1787. (b) Vriezema, D.; Aragones, M.;
Elemans, J.; Cornelissen, J.; Rowan, A.; Nolte, R. Chem. Rev. 2005,
105, 1445−1489.
(3) (a) Astruc, D.; Chardac, F. Chem. Rev. 2001, 101, 2991−3024.
́
(b) Helms, B.; Frechet, J. Adv. Synth. Catal. 2006, 348, 1125−1148.
Figure 3. Cryo-TEM images of the aqueous solutions of 1: (A) 1
spiked with KCN; (B) 1 with added DMAP and CuSO4; (C) 1 with
added DMAP, CuSO4, and benzyl alcohol.
(4) (a) Lu, A.; O’Reilly, R. Curr. Opin. Biotechnol. 2013, 24, 639−645.
(b) Wulff, G.; Liu, J. Acc. Chem. Res. 2011, 45, 239−247.
(c) Terashima, T.; Sawamoto, M. In Progress in Controlled Radical
Polymerization: Materials and Applications; Matyjaszewski, K.,
Sumerlin, B., Tsarevsky, N., Eds.; ACS Symposium Series 1101;
American Chemical Society: Washington, DC, 2012: pp 65−80.
(5) (a) Breslow, R. Acc. Chem. Res. 1995, 28, 146−153. (b) Wiester,
M.; Ulmann, P.; Mirkin, C. Angew. Chem., Int. Ed. 2011, 50, 114−137.
(6) Dwars, T.; Paetzold, E.; Oehme, G. Angew. Chem., Int. Ed. 2005,
44, 7174−7199.
5 mM aqueous solutions of 1 (spiked with KCN to scavenge
any residual Cu ions), vesicular structures with an average
diameter of 42.4 9.1 nm were observed. After the addition of
20 mM DMAP and 2 mM CuSO4, the vesicles changed to more
complex, electron-dense aggregates (d = 118.8
33.2 nm).
Interestingly, the reaction mixture with BnOH (100 mM)
exhibited a completely different morphology: high-aspect-ratio
cylindrical micelles (d = 8.2
0.5 nm) were observed. In
(7) Holmberg, K. Eur. J. Org. Chem. 2007, 731−742.
addition to the impact of BnOH and the aldehyde product, this
structural rearrangement might be ascribed to the change in the
coordination preference of Cu, which is expected to exist
mostly in its reduced CuI state upon addition of the alcohol.
Furthermore, the reduced TEMPO-H form of 1 could be
detected in the reactions by LC/MS. This compound is more
polar than 1 and likely has different self-assembly priorities.
In conclusion, we explored the reactivity of the Cu/TEMPO
catalytic system for oxidation of alcohols to aldehydes in
(8) Lipshutz, B.; Aguinaldo, G.; Ghorai, S.; Voigtritter, K. Org. Lett.
2008, 10, 1325−1328.
(9) (a) Lipshutz, B.; Ghorai, S. Org. Lett. 2008, 11, 705−708.
(b) Lipshutz, B.; Ghorai, S. Tetrahedron 2010, 66, 1057−1063.
(c) Lipshutz, B.; Isley, N.; Moser, R.; Ghorai, S.; Leuser, H.; Taft, B.
Adv. Synth. Catal. 2012, 354, 3175−3179.
(10) Lipshutz, B.; Ghorai, S. Org. Lett. 2011, 14, 422−425.
(11) Mugemana, C.; Chen, B.; Bukhryakov, K.; Rodionov, V. Chem.
Commun. 2014, 50, 7862−7865.
1316
DOI: 10.1021/cs5020018
ACS Catal. 2015, 5, 1313−1317

ACS Catalysis
Letter
(12) (a) Semmelhack, M.; Schmid, C.; Cortes, D.; Chou, C. J. Am.
Chem. Soc. 1984, 106, 3374−3376. (b) Gamez, P.; Arends, I.; Sheldon,
R.; Reedijk, J. Adv. Synth. Catal. 2004, 346, 805−811. (c) Parmeggiani,
C.; Cardona, F. Green Chem. 2012, 14, 547−564.
(13) Hoover, J.; Stahl, S. J. Am. Chem. Soc. 2011, 133, 16901−16910.
(14) Blackmond, D.; Armstrong, A.; Coombe, V.; Wells, A. Angew.
Chem., Int. Ed. 2007, 46, 3798−3800.
(15) (a) Figiel, P.; Sibaouih, A.; Ahmad, J.; Nieger, M.; Raisanen, M.;
Leskela, M.; Repo, T. Adv. Synth. Catal. 2009, 351, 2625−2632.
(b) Zhang, G.; Han, X.; Luan, Y.; Wang, Y.; Wen, X.; Xu, L.; Ding, C.;
Gao, J. RSC Adv. 2013, 3, 19255−19258. (c) Mase, N.; Mizumori, T.;
Tatemoto, Y. Chem. Commun. 2011, 47, 2086−2088.
(16) Hoover, J.; Ryland, B.; Stahl, S. ACS Catal. 2013, 3, 2599−2605.
(17) Lipshutz, B.; Hageman, M.; Fennewald, J.; Linstadt, R.; Slack,
E.; Voigtritter, K. Chem. Commun. 2014, 50, 11378−11381.
(18) (a) Dias, A.; Freire, M.; Coutinho, J.; Marrucho, I. Fluid Phase
Equilib. 2004, 222−223, 325−330. (b) Riess, J. Chem. Rev. 2001, 101,
2797−2920. (c) Lowe, K. Comp. Biochem. Physiol., Part A: Mol. Integr.
Physiol. 1987, 87A, 825−838.
(19) Horvath, I. Acc. Chem. Res. 1998, 31, 641−650.
(20) Ragagnin, G.; Betzemeier, B.; Quici, S.; Knochel, P. Tetrahedron
2002, 58, 3985−3991.
(21) Gheorghe, A.; Cuevas-Yanez, E.; Horn, J.; Bannwarth, W.;
Narsaiah, B.; Reiser, O. Synlett 2006, 2767−2770.
(22) Allen, S.; Walvoord, R.; Padilla-Salinas, R.; Kozlowski, M. Chem.
Rev. 2013, 113, 6234−6458.
(23) Kumpulainen, E.; Koskinen, A. Chem.Eur. J. 2009, 15,
10901−10911.
(24) Sjoblom, J.; Blokhus, A.; Sun, W.; Friberg, S. J. Colloid Interface
̈
Sci. 1990, 140, 481−491.
(25) Konning, D.; Hiller, W.; Christmann, M. Org. Lett. 2012, 14,
̈
5258−5261.
1317
DOI: 10.1021/cs5020018
ACS Catal. 2015, 5, 1313−1317