Towards a greener epoxidation method: Use of water-surfactant media and catalyst recycling in the platinum-catalyzed asymmetric epoxidation of terminal alkenes with hydrogen peroxide

Article abstract of DOI:10.1002/adsc.200600518

Remarkable improvements in enantioselectivity as well as recycle were observed in the catalytic asymmetric epoxidation of terminal alkenes with a chiral, electron-poor platinum(II) catalyst with hydrogen peroxide as terminal oxidant in water-surfactant media.

Full text of DOI:10.1002/adsc.200600518

COMMUNICATIONS
DOI: 10.1002/adsc.200600518
Towards a Greener Epoxidation Method: Use ofWater-Surfactant
Media and Catalyst Recycling in the Platinum-Catalyzed Asym-
metric Epoxidation ofTerminal Alkenes with Hydrogen Peroxide
Marco Colladon,a Alessandro Scarso,a and Giorgio Strukula,*
a
Department of Chemistry, Università Caꢀ Foscari di Venezia, Dorsoduro 2137, 30123 Venice, Italy
Fax: (+39)-041-234-8517; e-mail: strukul@unive.it
Received: October 10, 2006
Dedicated to Professor Mauro Graziani in occasion of his 70th birthday.
Supporting information for this article is available on the WWW under http://asc.wiley-vch.de/home/.
Abstract: Remarkable improvements in enantiose-
lectivity as well as recycle were observed in the cat-
alytic asymmetric epoxidation of terminal alkenes
with a chiral, electron-poor platinum(II) catalyst
with hydrogen peroxide as terminal oxidant in
water-surfactant media.
zation of reagents thus improving local concentration
and reactivity, thirdly in some cases they impart
unique chemo- regio- and stereoselectivities.[8] While
ꢀ
hydrogenation, hydroformylation and C C coupling
reactions have been already investigated in water and
in micellar media,[7a,9] oxidation reactions still need to
be thoroughly explored both in water[10] and in micel-
lar media,[11] in particular stereoselective oxidation re-
actions in water or in micellar media are reported in
only a few contributions to date.[12,13,14]
Keywords: aqueous phase catalysis; asymmetric cat-
alysis; epoxidation; hydrogen peroxide; platinum
Oxidation processes are also experiencing a smooth
but profound transformation from the employment of
hazardous oxidants to greener ones, such as molecular
Water as an alternative reaction medium[1] for organic oxygen and hydrogen peroxide.[15] Soft Lewis acids
reactions has been a well studied topic for several are intrinsically compatible with reactions in water as,
[2]
ꢀ
years, in particular for C C coupling reactions ex- at variance with early transition metal Lewis acid cen-
ploiting the peculiar properties of such solvent, in ters, they suffer neither from deactivation by water
primis polarity and hydrophobic effect.[3] More re- nor competition for substrate coordination.
cently, the employment of water as solvent is experi-
Recently, we reported some interesting results ob-
encing a sort of renaissance in the scientific communi- served in the asymmetric sulfoxidation of prochiral
ty because of the growing stringent requirements con- thioethers with a dimeric chiral Pt(II) complex in
cerning chemical production,[4] associated with possi- water-SDS medium (SDS=sodium dodecyl sulfate)
ble catalyst recycling that is a common advantage of with yields up to 99% and ees up to 88%.[14] Exploit-
liquid water/organic biphasic reactions. To ensure the ing the same fruitful approach, herein we present the
solubilization of metal catalysts in water, common employment of water-surfactant media in the catalytic
transition metal-catalyzed organic transformations asymmetric epoxidation of terminal alkenes[16] with
like hydrogenation[5] require ligand modification[6] the chiral Pt(II) complex {[(S,S)Chiraphos]Pt
with polar moieties that often turns out to involve te- (H2O)}
dious and time-consuming synthetic steps. Much more terminal oxidant (Scheme 1).
straightforward would be the employment of organo- Initially a surfactant screening was performed
ACHTREUNG
A
ACHTREUNG
metallic catalysts developed expressly for organic sol- (Table 1) to assess the degree of compatibility of the
vents and their solubilization in water with the aid of surfactant additives with the asymmetric epoxidation
a surfactant, thus avoiding several synthetic steps and of 4-methyl-pentene as model substrate with one
allowing the use of ordinary and/or commercially equivalent of hydrogen peroxide catalyzed by com-
available ligands.
plex 1 (1% mol). The same reaction performed in
Surfactants play many roles at the same time,[7] 1,2-dichloroethane (DCE) showed 56% yield with
firstly they improve the solubilization of organic re- 58% ee.[16] Conversely, in pure water without surfac-
agents in water, secondly they favor compartmentali- tant catalyst solubilization did not occur and no reac-
Adv. Synth. Catal. 2007, 349, 797 – 801
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
797

Marco Colladon et al.
COMMUNICATIONS
sulfonate) provided appreciable catalyst solubilization
in water and the enantioenriched epoxide was pro-
duced in lower yield but with surprisingly enhanced
ee (entries 1 and 5). Among neutral surfactants large
differences were observed: while the sorbitane deriva-
tive SPAN60 showed no activity, alkyl polyoxyethy-
lene derivative showed interesting results (entries 6
and 7).
A wide range of performance was observed with
surfactants related to the Triton family, with large dif-
ferences in activity related to the chemical structure
and length of the surfactant itself (entries 8–10). The
above observations suggest that the right balance of
polarity, presence and size of functional groups in the
micellar aggregate are all critical parameters to
ensure good activity and selectivity. Moreover, the
effect of surfactant concentration was investigated
Scheme 1. {[(S,S)Chiraphos]Pt
asymmetric epoxidation of terminal alkenes with hydrogen
peroxide in water with surfactants.
A
G
A
tion was observed (entry 2). Anionic as well as cation- with Triton X100 (entries 10–13) observing a more
ic micelles were not suitable for such reaction (en- pronounced effect on the activity, while enantioselec-
tries 3 and 4), at variance with what is observed in sul- tivity was little influenced but always above the
foxidation in the presence of a bis-cationic dimeric values observed in the organic phase under identical
complex.[14] In the present case, a zwitterionic surfac- experimental conditions.
tant (N-dodecyl-N,N-dimethyl-3-ammonio-1-propane-
Table 1. Catalytic enantioselective epoxidation of terminal 4-methylpentene with H2O2 mediated by 1 in micellar media.
General conditions: 4-methylpentene:H2O2:1=100:100:1; [4-methylpentene]=0.83 mmol, [H2O2]=0.83 mmol, [1]=
0.0083 mmol 1% mol, solvent water 0.5 mL, room temperature in air.
Entry
Time [h]
Yield [%][a]
ee[b] [%]
Abs. Conf.[c]
Solvent/Additive
1
2
3
4
5
6
7
8
4
56
0
0
0
28
0
58
-
-
-
84
-
R-(+)
-
-
DCE[d]
24
24
24
6
24
6
24
6
6
H2O
H2O/SDS[e]
-
H2O/CTABr[f]
H2O/Zwitterionic[g]
H2O/SPAN60[h]
H2O/POA[i]
R-(+)
-
R-(+)
-
R-(+)
R-(+)
R-(+)
R-(+)
R-(+)
41
0
84
-
H2O/Triton-X405[j]
H2O/Triton-X114[k]
H2O/Triton-X100[l]
H2O/Triton-X100[m]
H2O/Triton-X100[n]
H2O/Triton-X100[o]
9
61
34
45
51
33
82
78
79
82
76
10
11
12
13
6
6
6
[a]
Yield in epoxide determined by extraction with hexane and GC analyses (column b-cyclodex).
Enantiomeric excess determined by CSP-GC (column b-cyclodex).
Absolute configuration determined by optical rotations and comparison of the retention orders with known literature
data.
Dichloroethane as solvent, [1]=2% mol.
Sodium dodecyl sulfate (75 mM, 1 mM in micelles).
Cetyltrimethylammonium bromide (168 mM, 1 mM in micelles).
N-Dodecyl-N,N-dimethyl-3-ammonio-1-propanesulfonate (150 mM, 1 mM in micelles).
Sorbitane monostearate (150 mM, 1 mM in micelles).
[b]
[c]
[d]
[e]
[f]
[g]
[h]
[i]
Polyoxyethylene alcohol (C12H25-C18H37)-(OCH2CH2)5 (150 mM, 1 mM in micelles).
Polyoxyethylene(40) isooctyl phenyl ether (150 mM, 1 mM in micelles).
Polyoxyethylene(8) isooctyl cyclohexyl ether (150 mM, 1 mM in micelles).
Polyoxyethylene(10) isooctyl phenyl ether (33 mM, 0.22 mM in micelles).
Polyoxyethylene(10) isooctyl phenyl ether (75 mM, 0.5 mM in micelles).
Polyoxyethylene(10) isooctyl phenyl ether (150 mM, 1 mM in micelles).
Polyoxyethylene(10) isooctyl phenyl ether (230 mM, 1.5 mM in micelles).
[j]
[k]
[l]
[m]
[n]
[o]
798
www.asc.wiley-vch.de
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 797 – 801

Platinum-Catalyzed Asymmetric Epoxidation of Terminal Alkenes
COMMUNICATIONS
The increment of enantiomeric excess observed be- the aqueous phase with a solvent in which 1 and the
tween chlorinated solvent (58%) and water-non-ionic surfactant are both insoluble. To this end, we explored
surfactant (84%) can be better expressed in terms of various experimental conditions characterized by dif-
enantiomeric ratio[18] (organic 79:21; water-surfactant ferent combinations of organic solvents for extraction
¼
92:8) and subsequently translated in terms of DDG6
and surfactants. We found that operating with Triton-
between the two transition states for the production X114 in a 2:1 substrate/oxidant excess, extracting the
¼
of the enantiomeric epoxides. DDG6 almost doubles water phase with hexane and simply adding fresh sub-
from ꢀ3.23 kJmolꢀ1 to ꢀ5.95 kJmolꢀ1 by simply strate and 35% H2O2 for the next runs, makes it pos-
switching the reaction from DCE to water-non-ionic sible to perform up to three enantioselective epoxida-
surfactant media, as a further evidence of the unique tion cycles with both constant yield and enantioselec-
environmental conditions provided by micelles.
tivity (Table 3). The different experimental conditions
The synthetic scope of the enantioselective epoxi- employed were required to favor easier phase separa-
dation reaction was investigated employing other ter- tion. This led to a slight decrease in ee and a more
minal alkene substrates, as reported in Table 2.
pronounced decrease in yield per individual cycle, but
From Table 2 it is clear that linear and g-branched the total turnover number achieved was definitely
terminal alkenes are all suitable substrates with good higher. The major limitation to further recycling was
conversions, comparable to those of reactions per-
formed in the organic phase,[16] and with moderate to
Table 3. Recycling experiments in the catalytic enantioselec-
good ees of the corresponding epoxides, even though
tive epoxidation of 4-methylpentene with H2O2 in water/
in these cases no marked enhancement of enantiose-
Triton-X114 solution mediated by 1. General conditions:
lectivity was observed between organic and water-
non-ionic surfactant media, probably because of the
lower steric hindrance of those alkenes with respect
to 4-methylpentene. It seems conceivable that a fine
tuning of the surfactant properties with respect to the
individual substrateꢀs steric properties is probably
necessary to improve enantioselectivity. On the con-
trary, allylbenzene derivatives that are suitable sub-
strates in organic solvent,[16] are not generally oxi-
dized in the micellar medium because of insufficient
solubilization. Only the more polar 3,4-dimethoxyal-
lylbenzene reacted providing the corresponding enan-
tioenriched epoxide with poor yield but interesting ee.
The use of biphasic catalysis opens the way to the
possible recycling of the chiral catalyst by extracting
substrate:H2O2:1=200:100:1;
[H2O2]=3.3 mmol, [1]=0.066 mmol, solvent water 4 mL,
surfactant Triton-X114 (150 mM, 1 mM in micelles).
[substrate]=6.6 mmol,
Cycle
Time [h]
Yield [%][a]
ee [%][b]
Abs. Conf.[c]
1
2
3
8
8
8
26
26
30
75
72
77
R-(+)
R-(+)
R-(+)
[a]
Yield in epoxide determined by extraction with hexane
and GC analyses (column b-cyclodex).
Enantiomeric excess determined by CSP-GC (column b-
cyclodex).
Absolute configuration determined by optical rotations
and comparison of the retention orders with known liter-
ature data.
[b]
[c]
Table 2. Catalytic enantioselective epoxidation of terminal alkenes with H2O2 in water/Triton-X100 solution mediated by 1.
General conditions: substrate:H2O2:1=100:100:1; [substrate]=0.83 mmol, [H2O2]=0.83 mmol, [1]=0.0083 mmol, solvent
water 0.5 mL, surfactant Triton-X100 (150 mM, 1 mM in micelles), room temperature in air.
Entry
Substrate
Time [h]
Yield [%][a]
ee [%][b]
Abs. Conf.[c]
1
2
3
6
6
6
84
78
43
74
57
52
R
R
R
4
6
81
61
R
5
20
23
74[d]
R-(ꢀ)[e]
[a]
1
Yield in epoxides determined by H NMR.
[b]
1
Enantiomeric excess determined by integration of the H NMR spectrum in the presence of the chiral shift reagent euro-
pium tris-[3-(heptafluoropropylhydroxymethylene)-(+)-camphorate] Eu(hfc)3 at room temperature in CDCl3.
G
[c]
Absolute configuration assigned by comparison to the deshielding effect in the presence of Eu
A
observed for other terminal epoxides.[16]
[d]
[e]
Enantiomeric excess determined by chiral HPLC analysis on a Chiracel OD-H column.
Absolute configuration determined by optical rotations and comparison with literature data.
Adv. Synth. Catal. 2007, 349, 797 – 801
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
799

Marco Colladon et al.
COMMUNICATIONS
not due to catalyst decomposition, but rather to the Experimental Section
reaction medium losses that are unavoidably involved
in carrying out the different operations manually on a General Remarks
small lab scale. Anyhow, the results of Table 3 prove
Water was purified according to the milliQ technique, 35%
hydrogen peroxide as well as all the alkene substrates are
commercial products (Aldrich) and were used without fur-
ther purification. The chiral complex {[(S,S)Chiraphos]Pt-
the principle that catalyst separation and recycling
without loss of activity and selectivity is indeed possi-
ble using the micellar medium.
In order to prove the solubilization of catalyst 1 in
A
G
A
dure reported in the literature.[17]
the micellar medium, as well as investigating its posi-
1
1
tion in the micelles we carried out 31P, H NMR and
tra were recorded at 298 K on a Bruker AVANCE 300 spec-
trometer operating at 300.15 and 121.50 MHz, respectively.
2D-NOESY spectra of 1 in H2O-D2O/Triton-X100 so-
lution (see Supporting Information). The 31P NMR
spectrum shows the typical resonances of the Pt(II)
complex 1 with chemical shifts similar to those found
d values in ppm are relative to Si(CH3)4 and 85% H3PO4, re-
E
spectively. A 2D-NOESY experiment was acquired with a
spectrum width of 10 ppm, a relaxation delay d1 of 1 s, using
2 K data points in the t2 dimension and 512 data points in
the t1 dimension, with subsequent weighting with the sine-
bell function using 160 scans for each t1 increment. The
mixing time d8 employed was 0.4 s.
1
for 1 in chlorinated solvent but with a smaller JP,Pt for
P trans to the catalytic site (3776 Hz instead of
4272 Hz). This observation might be consistent with
the coordination of a hydroxy group instead of a neu-
tral water molecule,[17] however, the unusual solvent
(micellar) medium suggests these conclusions be
GLC measurements were taken on a Hewlett–Packard
5890 A gas chromatograph equipped with an FID detector
(carrier gas He). All reactions were monitored either by GC
1
taken very cautiously. In the H NMR spectrum the
1
or H NMR. The enantiomeric excess was determined as re-
signals of the aromatic rings of 1 close to the intense
resonances of the surfactant are clearly visible. Insight
into the position of catalyst 1 in the micelles is provid-
ed by a 2D-NOESY experiment where clear cross-
peaks between the aromatic moieties of the complex
and the terminal tert-butyl group of the surfactant are
observed, indicating that the catalyst is located deeply
in the apolar core of the micelles that behave as
nanoreactors.
ported in Tables 1, 2 and 3.
Oxidation Reactions
These were carried out in a 2-mL vial equipped with a side-
arm fitted with a screw-capped silicone septum to allow
sampling. Stirring was performed by a Teflon-coated bar
driven externally by a magnetic stirrer (700 rpm). A con-
stant temperature (208C) was maintained by water circula-
tion through an external jacket connected with a thermostat.
The concentration of the commercial 35% H2O2 solution
was checked iodometrically prior to use. Typically, the
proper amount of surfactant was dissolved in deionized
milliQ water (0.5 mL), followed by catalyst 1 (0.0083 mmol).
After 10 min the substrate (0.83 mmol) was added and the
mixture stirred for 10 min. To this 35% hydrogen peroxide
was added in one portion (0.83 mmol) and the mixture stir-
red at room temperature. After a certain time, 1 mL of or-
ganic solvent (pentane or hexane) was added to extract the
reagents and products and the organic phase was analyzed.
In conclusion, the methodology reported herein
represents a viable way for carrying out asymmetric
epoxidations in water with “ordinary” transition
metal catalysts, thereby avoiding the need to modify
the catalyst with hydrophilic functional groups. The
extremely mild and environmentally friendly experi-
mental conditions (room temperature, low catalyst
loading, use of water as solvent, use of 35% H2O2 as
oxidant) together with the easy isolation of the enan-
tioenriched product by simple hexane two-phases ex-
traction and the possibility to recycle the catalyst are
all key features of the present system that demon-
strate the viability of this “greener” synthetic method.
Yields and enantioselectivities were from moderate to
good, and in some cases the aqueous medium allowed
a significant improvement in the asymmetric induc-
tion compared to the use of organic solvents. This is
particularly interesting for a class of substrates that
have been only sparingly investigated.[19] Enantiose-
lectivity as well as yield and recyclability are all fea-
tures that result from a critical balance between cata-
lyst, substrate and surfactant properties and these re-
lationships deserve further investigations that are cur-
rently underway.
Recycling Procedures
These were carried as for normal oxidation reaction on a 4-
mL volume, extracting the solution with an identical volume
of hexane and leaving the system unstirred for 6 h to allow
phase separation. The organic supernatant phase was then
removed for analysis while the aqueous phase was recycled
for a new catalytic run by addition of fresh substrate and ox-
idant.
Supporting Information
1H and 31P NMR spectra and 2D-NOESY spectrum of cata-
lyst 1 in the micellar medium water/Triton-X100 solution.
800
www.asc.wiley-vch.de
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 797 – 801

Platinum-Catalyzed Asymmetric Epoxidation of Terminal Alkenes
COMMUNICATIONS
J. T. F. Keurentjes, Ind. Eng. Chem. Res. 2001, 40, 5240;
b) Z. Bourhani, A. I. Malkov, Chem. Commun. 2005,
4592; example of organocatalytic oxidations: R. Liu, C.
Dong, X. Liang, X. Wang, X. Hu, J. Org. Chem. 2005,
70, 729.
Acknowledgements
A.S. and G.S. wish to thank MIUR and Università Caꢀ Fo-
scari di Venezia for financial support (PRIN 2003), and
Johnson-Matthey for the loan of platinum.
[11] a) H. Yao, D. E. Richardson, J. Am. Chem. Soc. 2003,
125, 6211; b) A. Masuyama, K. Fukuoka, N. Katsuya-
ma, M. Nojima, Langmuir 2004, 20, 82; c) A. Masuya-
ma, T. Yamagichi, M. Abe, M. Nojima, Tetrahedron.
Lett. 2005, 46, 213; d) A. Blaskó, C. A. Bunton, S.
Wright, J. Phys. Chem. 1993, 97, 5435.
References
[1] D. Adams, P. Dyson, S. Tavener, Chemistry in Alterna-
tive Reaction Media, John Wiley & Sons, London, 2004.
[2] a) C. J. Li, Chem. Rev. 1993, 93, 2023; b) C. J. Li, Chem.
Rev. 2005, 105, 3095.
[12] M. Bonchio, T. Carofiglio, F. Di Furia; R. Fornasier,J.
Org. Chem.1995, 60, 5986.
[3] a) U. M. Lindstrçm Angew. Chem. Int. Ed. 2006, 45,
548; b) M. C. Pirrung Chem. Eur. J. 2006, 12, 1312.
[4] a) P. T. Anastas, J. C. Warner, Green Chemistry: Theory
and Practice, Oxford University Press, New York 1998;
b) P. T. Anastas, M. M. Kirchhoff, Acc. Chem. Res.
2002, 35, 686; c) C. Bolm, O. Beckmann, O. G. A.
Dabard, Angew. Chem. Int. Ed. 1999, 37, 1198; Angew.
Chem.1998, 37, 1198–1209.
[13] One example of stereoselective oxidation in reversed
micelles in toluene is known: H. Tohma, S. Takizawa,
H. Watanabe, Y. Fukuoka, T. Maegawa, Y. Kita, J. Org.
Chem. 1999, 64, 3519.
[14] A. Scarso, G. Strukul, Adv. Synth. Catal. 2005, 347,
1227.
[15] a) B. S. Lane, K. Burgess, Chem. Rev. 2003, 103, 2457;
b) G. Grigoropoulou, J. H. Clark, J. A. Elings, Green
Chem. 2003, 5, 1; c) R. Noyori, M. Aoki, K. Sato,
Chem. Commun. 2003, 1977.
[16] M. Colladon, A. Scarso, P. Sgarbossa, R. A. Michelin,
G. Strukul, J. Am. Chem. Soc. 2006, 128, 14006.
[17] E. Pizzo, P. Sgarbossa, A. Scarso, R. A. Michelin, G.
Strukul, Organometallics 2006, 25, 3056.
[18] In order to better compare the enantioselectivity under
different experimental conditions for asymmetric cata-
lytic reactions, enantiomeric excess ee=100·{([R]ꢀ[S])/
([R]+[S])} is replaced by enantiomeric ratio er=[R]/
[S] which more straightforwardly provides a direct
measurement of selectivity as recently outlined: R. E.
Gawley, J. Org. Chem. 2006, 71, 2411.
[5] T. Dwars, G. Oehme, Adv. Synth. Catal. 2002, 344, 239.
[6] N. Pinault, D. W. Bruce, Coord. Chem. Rev. 2003, 241,
1.
[7] a) T. Dwars, E. Paetzold, G. Oehme, Angew. Chem. Int.
˘
Ed. 2005, 44, 7174; b) S. Tas¸cioglu, Tetrahedron 1996,
52, 11113; c) J. B. F. N. Engberts, Pure Appl. Chem.
1992, 64, 1653; d) K. Holmberg, Eur. J. Org. Chem.
2007, 731.
[8] a) U. M. Lindstrom, Chem. Rev. 2002, 102, 2751; b) S.
Otto, J. B. F. N. Engberts, Org. Biomol. Chem. 2003, 1,
2809; c) S. Kobayashi, K. Manabe, Acc. Chem. Res.
2002, 35, 209; d) D. Sinou, C. Rabeyrin, C. Nguefack,
Adv. Synth. Cat. 2004, 345, 357.
[9] a) C.-J. Li, L. Chen, Chem. Soc. Rev. 2006, 35, 68; b) D.
[19] I. W. C. E. Arends, Angew. Chem. Int. Ed. 2006, 45,
6250.
Sinoi, Adv. Synth. Catal. 2002, 344, 221.
[10] Examples of metal catalyzed oxidations: a) L. J. P.
van den Broeke, V. G. de Bruijn, J. H. M. Heijnen,
Adv. Synth. Catal. 2007, 349, 797 – 801
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
801