Oxidative cleavage of functionalized cyclic olefins with H 2O2 using a peroxotungstate complex

Article abstract of DOI:10.1021/op200130d

The oxidative cleavage of functionalized cyclic olefins with aqueous hydrogen peroxide was studied. A mild protocol was developed using a readily available diperoxotungstate catalyst that allowed otherwise unstable functionalities to be present during the

Full text of DOI:10.1021/op200130d

COMMUNICATION
pubs.acs.org/OPRD
Oxidative Cleavage of Functionalized Cyclic Olefins with H2O2 Using a
Peroxotungstate Complex
Mark J. Ford,* Helmut Kohlhepp, Stephen Miller, Sven Riess, Jan Peter Schmidt, and Lothar Wisplinghoff
Bayer CropScience AG, Industriepark H€ochst G836, 65926 Frankfurt am Main, Germany
S Supporting Information
b
was prepared in near quantitative yield on a 2.5 mol scale
ABSTRACT: The oxidative cleavage of functionalized
cyclic olefins with aqueous hydrogen peroxide was studied.
A mild protocol was developed using a readily available
diperoxotungstate catalyst that allowed otherwise unstable
functionalities to be present during the oxidation. During
the optimization of the procedure (3R*,4R*)-3-[(benzyloxy)-
carbonyl]-4-(ethoxycarbonyl)hexanedioic acid was synthesized
on a 400 g scale furnishing the product in 79% yield.
(Scheme 1). Here successful alkylative trapping of the inter-
mediate monoester carboxylate salt is crucial to achieving
this high yield since otherwise the carboxylate-ester is only
present as part of an equilibrating mixture with the starting
anhydride.
We then went about investigating the oxidative cleavage of the
cyclohexene moiety. Heterogeneous oxidation using potassium
permanganate (3.15 equiv) gave us access to the desired bis-acid
1 in a moderate 64% yield (on a 2 mol scale). However, it was
very difficult to isolate the product from the aqueous solvent. In
addition, filtration of the manganese dioxide waste product was
challenging and the amounts of solid waste generated would have
disfavored further scale-up. Changing to catalytic ruthenium(IV)
oxide and stoichiometric sodium periodate, we were able to synthe-
size the target product more efficiently. Bis-acid 1 was obtained in
a pleasing 90% yield (on a 0.1 mol scale). However, again the use
of the stoichiometric oxidant, sodium periodate, was unattractive
from an environmental and financial point of view. Additionally,
a problematic kinetic behavior of the reaction posed a threat for
process safety, and for this reason alone this approach would have
been abandoned. This kinetic behavior manifested itself as an
unpredictable 3ꢀ5 h delayed strongly exothermic oxidation of
the dialdehyde which only starts after all of the olefin appears to
have been consumed. Encouraged by the ability of ruthenium to
catalyze the oxidative cleavage of cyclohexene 2, we became
aware of a report by Che and co-workers in which the catalytic
ruthenium(III) species [(N,N0,N00-trimethyl-1,4,7-triazacyclononane)-
(CF3CO2)2Ru(III)(H2O)]CF3CO2 was stoichiometrically reoxi-
dized by aqueous hydrogen peroxide and is reported to oxidatively
cleave olefins.5 Disappointingly, this catalyst is not easily pre-
pared and in our eyes the reaction protocol would not lend itself
to large scale synthesis.
In our search for a robust system, we came across pioneering
work by Ishii et al. in which the oxidative cleavage of olefins into
carboxylic acids with hydrogen peroxide by tungstic acid is
reported.6 This work has been carried on by numerous groups
in both industry and academia.7aꢀf Tungstic acid itself cannot
oxidize organic substrates in water due to the fact that it is only soluble
in water whereas the oxidation substrates are usually immiscible
with water. It was found that certain organic acids may form complex
catalysts with tungstic acid and enable good catalytic perfor-
mance in nonoleophilic systems. Jiang and co-workers demon-
strated that a large range of organic acids may be employed for
this oxidation. Yields were high for cases in which the pKa of the
’ INTRODUCTION
Polycarboxylic acids are important building blocks in the
chemical industry. This is particularly true for adipic acid
which is prepared by the oxidative cleavage of cyclohexene and
finds its use as a raw material of polyesters and 6,6-nylon.
Adipic acid is manufactured on a million tons per year scale.
Despite increasing environmental concerns, many polyacidic
fine chemicals are still manufactured by classical stoichio-
metric methods that are unacceptable for practical syntheses.
Carbonꢀcarbon double bonds cleavage may be achieved through
the use of heavy metal oxidants. However, these oxidants typically
lead to the formation of large amounts of toxic waste, while known
organic stoichiometric oxidants are usually expensive.1 Nitric acid,
the most conventional industrial oxidant, is cheap but unavoidably
forms the greenhouse gas nitrous oxide. As a consequence, great
attention has been paid to the development of alternative green
methods for the preparation of polyacids. Earlywork byVenturello
and Ricci suggested that hydrogen peroxide could be used in
the synthesis of polycarboxylic acids from lipophilic olefins.2
Their work was greatly expanded by Noyori and co-workers3a,b
and has been the focus of green researchinoxidationchemistryand
industrial technology ever since.4 However, the methodologies
developed so far showed one great weakness. They did not
allow acid sensitive substrates to be used in the oxidative
cleavage protocols. Therefore, we had great interest in making
this oxidation technology more applicable to a wider range
of substrates. For this, we concentrated on a substrate that
was previously difficult to cleave oxidatively under the known
conditions.
’ RESULTS AND DISCUSSION
At the onset of our work, there was great interest in the efficient
synthesis of dicarboxylic acid 1 (Scheme 3), where the two ester
groups are functionally distinct. Starting from readily available
1,2,3,6-cis-tetrahydrophthalic anhydride (3), cyclohexene-diester 2
Received: May 17, 2011
Published: June 22, 2011
r
2011 American Chemical Society
883
dx.doi.org/10.1021/op200130d Org. Process Res. Dev. 2011, 15, 883–885
|

Organic Process Research & Development
COMMUNICATION
Scheme 3. Oxidative cleavage to furnish bis-acid 1a
Scheme 1. Preparation of cyclohexene 2
Scheme 2. Reaction mechanism ꢀ in analogy to the pro-
posed reaction pathway for the synthesis of adipic acid by
oxidation of cyclohexene with 30% hydrogen peroxide
(Noyori et al.)3a
a See ref 11 for details of L1 and A1.
organic solvents and a phase transfer catalyst.7d The preparation
of their peroxotungsten catalyst was in close analogy to the one
described previously by Bhattacharyya et al. for the oxidation of
simple alcohols, sulfides, and amines.10 In the search for the
optimum system for our substrate 2, we screened a number of
ligands L1 and organic acids A1 as depicted in Scheme 3.11 Iterating
twenty nitrogen heterocyclic ligands (L1) and eight carboxylic
acids (A1), we found a set of conditions in which L1 = L2 =
pyridine in the presence of acetic acid (A1) that allowed smooth
oxidative cleavage of our functionalized cyclohexene 2 with
minimal saponification side products being formed. It should
be noted that the concentration of acetic acid in water in this
system was limited to a maximum of 40% since this mixture has
no flash point.15 There is no certainty of the exact nature of the
catalytic species because it remained in solution.
However, based on literature reports, it is assumed that whilst
L1 and L2 in the original freshly prepared catalyst are pyridine it
could well be that at least one of these ligands is exchanged for
acetic acid or water under the reaction conditions. After optimi-
zation, the protocol was finally carried out on a 400 g scale, giving
reliable results on upscale.16
acid ligand lay in the range of 1ꢀ3.7b However, it was found that
neutral or basic ligands inhibited the reaction.
Pleasingly, the procedure described by Jiang et al. led to the
formation of bis-acid 1 albeit in modest yield. Noyori proposed
that the reaction proceeds via six steps involving three kinds of
oxidation reactions (epoxidation, dehydrogenation, and a Bayerꢀ
Villiger oxidation) and two hydrolytic reactions (Scheme 2).3a
Clearly, many steps in this scheme are facilitated by (strongly)
acidic conditions. Interestingly, in contrast to Noyori’s observa-
tions, we found that for our substrate the oxidation of cyclohe-
nanediols 5 (four diastereoisomers) was slow (and thus appears
for at least one diastereomeric pair to be the rate-limiting step)
whereas Noyori reported a possible slow hydrolytic cleavage of
the intermediate epoxide: Only at temperatures of 40ꢀ50 °C did
we see significant accumulation of the epoxide intermediates
(two isomers). As under the Noyori conditions, during our rate-
limiting step, large amounts of reactant were saponified to
ultimately yield 1,2,3,4-butanetetracarboxylic acid and benzoic
acid (which is derived from the benzyl alcohol). This is not
unexpected at a pH of ∼0 in refluxing water. For our substrate,
we were in need then of a milder system that would allow us to
balance a reaction which progresses through many acid catalysed
reaction steps with an acid sensitive substrate. Catalytic systems
of tungstate/cocatalyst for the oxidative cleavage of cyclohexene
to adipic acid have been reported as indicated above, but direct
catalysis by peroxotungsten complexes has received less attention.
Yet, reports by Topich and Shechter indicated that oxidation
essentially occurred with peroxotungstate species.8,9 Li and co-workers
’ CONCLUSION
An efficient diperoxotungstate catalyzed oxidative cleavage
protocol was developed that allowed acid sensitive cyclohexene
derivatives to be converted into their ω-dicarboxylic acid analogues.
The use of heavy metal oxidants and expensive organic oxidants was
avoided in favor of readily available hydrogen peroxide. The process
was readily scaled up with the target molecule being isolated in good
yield and high chemical purity. No attempts have been made to scale
the protocol up further. We acknowledge that different substrates to
the one used during this optimization might need further optimiza-
tion of the procedure and might actually favor another ligand/acid
combination.17
’ EXPERIMENTAL SECTION
General. All reagents and solvents were used directly as purchased
from commercial suppliers. Reaction progress was followed by
high performance liquid chromatography (HPLC) at 210 nm
using an Agilent 1100 series HPLC and a Phenomenex Luna C18(2)
(30 mm ꢁ 4.6 mm) column. NMR spectra were obtained on a
Bruker DRX 400 MHz or Bruker AvanceIII 600 MHz spectrometer.
Catalyst Stock Solution. Tungstic oxide dihydrate (WO3 2H2O)
3
was prepared as described by Freedman.12 Next, the diperoxotung-
state species [W(O)(O2)2(H2O)2] was prepared by dissolving the
freshly prepared tungstic oxide dihydrate (22.4 g, 83.6 mmol) in
aqueous hydrogen peroxide (35%, 224 mL, 2.56 mol) and stirring the
showed that the peroxotungsten complex [W(O)(O2)2 Phen]-
3
H2O was able to catalyze the oxidative cleavage of cyclohexene
with hydrogen peroxide in the absence of a cocatalyst and without
884
dx.doi.org/10.1021/op200130d |Org. Process Res. Dev. 2011, 15, 883–885

Organic Process Research & Development
COMMUNICATION
initial yellow suspension vigorously for 15 min until a colorless
solution was obtained.13 Now pyridine (13.5 mL, 167 mmol)
was introduced until the solution turned a pale yellow. The freshly
prepared [W(O)(O2)2(Py)2] was used directly.14
(2) (a) Venturello, C.; Ricci, M. J. Org. Chem. 1986, 51, 1599.
(b) Venturello, C.; Ricci, M. EP0122804, 1984.
(3) (a) Sato, K; Aoki, M; Noyori, R. Science 1998, 281, 1646.
(b) Noyori, R; Aoki, M.; Sato, K. Chem. Commun. 2003, 1977.
(4) Searching SciFinder for ‘adipic acid and hydrogen peroxide’
provides 299 hits between 1999 and 2009 (ca. 60% patents).
(5) Che, C.-M.; Yip, W.-P.; Yu, W.-Y. Chem.—Asian J. 2006, 1, 453.
(6) Oguchi, T.; Ura, T.; Ishii, Y.; Ogawa, M. Chem. Lett. 1989, 857.
(7) (a) Deng, Y.; Ma, Z.; Wang, K.; Chen, J. Green Chem. 1999, 275.
(b) Jiang, H.; Gong, H.; Yang, Z.; Zhang, X.; Sun, Z. React. Kinet. Catal.
Lett. 2002, 75, 315. (c) Cao, F.-B.; Jiang, H.; Gong, H. Chin. J. Org. Chem.
2005, 1, 96. (d) Zhu, W.; Li, H.; He, X.; Shu, H.; Yan, Y. J. Chem. Res.
2006, 774. (e) Fujitani, K.; Mizutani, T.; Oida, T.; Kawase, T. J. Oleo Sci.
2009, 58, 37. (f) Fujitani, K.; Mizutani, T.; Oida, T.; Kawase, T. J. Oleo
Sci. 2009, 58, 323.
(8) Topich, J.; Lyon, J. T., III. Inorg. Chem. 1984, 23, 3202.
(9) Li, J.; Elberg, G.; Gefel, D.; Shechter, Y. Biochemistry 1995,
34, 6218.
(10) Maiti, S. K.; Banerjee, S.; Mukherjee, A. K.; Malik, K. M. A.;
Bhattacharyya, R. New J. Chem. 2005, 29, 554.
(11) Screening: L1 = 2,6-lutidine, 2-methylbenzoxazole, 2-phenyl-
benzimidazole, 2-picoline, 3-chloropyridine, 3-picoline, 4-picoline, 4-tert-
butylpyridine, benzimidazole, benzoxazole, indazole, iso-quinoline, 2-N-
dimethylbenzimidazole, 2-N-dimethylimidazole, N-butylimidazole,
N-methylbenzimidazole, N-methylimidazole, N-methylpyrazole, pyridine,
quinoline, whereby the best results were obtained with pyridine,
N-methylimidazole, and N-methylbenzimidazole. Screening: A1 = Acetic
acid ≈ propionic acid g succinic acid > 1,4-phthalic acid ≈ benzoic acid ≈
iso-butyric acid > 1,2-phthalic acid > oxalic acid.
Benzyl Ethyl (1S*,2R*)-Cyclohex-4-ene-1,2-dicarboxylate (2).
1,2,3,6-cis-Tetrahydrophthalic anhydride (3) (95%, 400.4 g,
2.50 mol) and ethanol (460.7 g, 10.0 mol) were dissolved in N,
N-dimethylacetamide (1.3 L). To this was added at room
temperature potassium carbonate (380.1 g, 2.75 mol) whereby the
temperature rose to 40 °C with evolution of gas. The suspension
was stirred for 15 min until the temperature had dropped to
35 °C when benzyl chloride (97%, 342.6 g, 2.63 mol) was added
over 5 min. The final reaction mixture was heated at 90 °C for 2 h
before it was allowed to cool to room temperature. Potassium
chloride was removed by filtration, and the filtrate was condensed
under reduced pressure. The oily residue was dissolved in toluene
(1.5 L) and washed with water (3 ꢁ 300 mL). The organic
solvent was removed in vacuo to give the target compound 2 as a
colorless liquid (724.9 g, 99%), HPLC purity >99% a/a.
(3R*,4R*)-3-[(Benzyloxy)carbonyl]-4-(ethoxycarbonyl)hexan-
edioic Acid (1). Benzyl ethyl (1S*,2R*)-cyclohex-4-ene-1,2-dicar-
boxylate (2, 400 g, 1.39 mol) and water (400 mL) were placed in a
2 L four-necked flask. The suspension was brought to a boil before
glacial acetic acid (320 mL) was added. Over the following 3 h,
reagents were added every 15 min: t0: 56.0 mL of catalyst solution
(CS), t15: 56.0 mL of 35% aqueous hydrogen peroxide (HP), t30:
56.0 mL (HP), t45: 56.0 mL (HP), t60: 56.0 mL (CS), t75: 56.0 mL
(HP), t90: 56.0 mL (HP), t105: 56.0 mL (HP), t120: 56.0 mL (CS),
(12) Freedman, M. L. J. Am. Chem. Soc. 1959, 81, 3834.
(13) Zhu, W.; Li, H.; He, X.; Shu, H.; Yan, Y. J. Chem. Res. 2006,
12, 774.
t135: 56.0 mL (HP), t150: 56.0 mL (HP), t165: 56.0 mL (HP), t180
:
56.0 mL (CS) ꢀ the total tungsten added equates to ∼4 mol %. The
reaction was further stirred at reflux for 1 h at which point the rate of
the oxidation had slowed to the point that saponification began to
compete with the oxidative cleavage of the diols. The reaction mixture
was cooled to 45 °C at which point water (400 mL) and a few
product seed crystals were added. Crystallization was complete after
stirring at 45 °C for 5 h and then at room temperature overnight. The
precipitate was removed by filtration, and the solids were washed
with water (2 ꢁ 800 mL) and n-heptane (640 mL) and dried in the
vacuum oven at 35 °C overnight providing the target compound 1 as
a colorless powder (388.0 g, 79%, HPLC purity 98.2% a/a, mp:
127ꢀ128 °C).
(14) Tarafder, M. T. H.; Fong, L. W. Orient. J. Chem. 2000, 16, 371.
(15) Garland, R. W.; Malcolm, M. O. Process Safety Progress 2002, 21
(3), 254.
(16) Ford, M. J.; Schmidt, J. P. EP2010162077.
(17) For example: Synthesis of 3-[(Benzyloxy]carbonyl]pentan-1,
5-dioic acid: A solution of benzyl cyclopent-3-ene carboxylate (1.026 g)
in water (1 mL) and acetic acid (0.8 mL) was heated to 100 °C and then
treated over 3 h with aqueous hydrogen peroxide (35%) (7 portions)
and catalyst solution (4 portions). After a further 1 h the mixture was
cooled and the product taken up in ethyl acetate (25 mL). The organic
phase was washed with dilute HCl and dilute sodium bisulfate (and
checked for the absence of peroxide!) before being dried over sodium
sulfate and concentrated in vacuo. Purification on silica (Biotage: heptane/
ethyl acetate gradient) afforded 1.08 g (80% yield) of the desired product.
’ ASSOCIATED CONTENT
Supporting Information. 1H NMR data and HPLC data
S
b
for cyclohexene 2 and bis-acid 1. This material is available free of
charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: mark.ford@bayer.com. Telephone: +49 (0)69 305 15797.
’ ACKNOWLEDGMENT
We wish to thank the analytical department at Bayer CropScience
AG for their assistance and particularly the NMR team of Matthias
Jank and the MS team of Peter Z€ollner.
’ REFERENCES
(1) Hudlicky, M. Oxidations in Organic Chemistry; ACS Monograph
Ser. 186; American Chemical Society: Washington, DC, 1990.
885
dx.doi.org/10.1021/op200130d |Org. Process Res. Dev. 2011, 15, 883–885