Halide-promoted dioxygenolysis of a carbon-carbon bond by a copper(II) diketonate complex

Article abstract of DOI:10.1021/ja502577b

A mononuclear Cu(II) chlorodiketonate complex was prepared, characterized, and found to undergo oxidative aliphatic carbon-carbon bond cleavage within the diketonate unit upon exposure to O₂ at ambient temperature. Mechanistic studies provide evidence for a dioxygenase-type C-C bond cleavage reaction pathway involving trione and hypochlorite intermediates. Significantly, the presence of a catalytic amount of chloride ion accelerates the oxygen activation step via the formation of a Cu-Cl species, which facilitates monodentate diketonate formation and lowers the barrier for O₂ activation. The observed reactivity and chloride catalysis is relevant to Cu(II) halide-catalyzed reactions in which diketonates are oxidatively cleaved using O₂ as the terminal oxidant. The results of this study suggest that anion coordination can play a significant role in influencing copper-mediated oxygen activation in such systems.

Full text of DOI:10.1021/ja502577b

Communication
pubs.acs.org/JACS
Halide-Promoted Dioxygenolysis of a Carbon−Carbon Bond by a
Copper(II) Diketonate Complex
Caleb J. Allpress,† Anna Miłaczewska,‡ Tomasz Borowski,‡ Jami R. Bennett,§ David L. Tierney,§
Atta M. Arif,|| and Lisa M. Berreau*,†
†Department of Chemistry & Biochemistry, Utah State University, Logan, Utah 84322-0300, United States
‡Jerzy Haber Institute of Catalysis and Surface Chemistry, Polish Academy of Science, Krakow 30-239, Poland
§Department of Chemistry and Biochemistry, Miami University, Oxford, Ohio 45056, United States
||Department of Chemistry, University of Utah, Salt Lake City, Utah 84112-0850, United States
S
* Supporting Information
Scheme 1. Examples of Oxidative Aliphatic Carbon−Carbon
Bond Cleavage Catalyzed by Cu(II) Salts and O2
ABSTRACT: A mononuclear Cu(II) chlorodiketonate
complex was prepared, characterized, and found to
undergo oxidative aliphatic carbon−carbon bond cleavage
within the diketonate unit upon exposure to O2 at ambient
temperature. Mechanistic studies provide evidence for a
dioxygenase-type C−C bond cleavage reaction pathway
involving trione and hypochlorite intermediates. Signifi-
cantly, the presence of a catalytic amount of chloride ion
accelerates the oxygen activation step via the formation of
a Cu−Cl species, which facilitates monodentate diketonate
formation and lowers the barrier for O2 activation. The
observed reactivity and chloride catalysis is relevant to
Cu(II) halide-catalyzed reactions in which diketonates are
oxidatively cleaved using O2 as the terminal oxidant. The
results of this study suggest that anion coordination can
play a significant role in influencing copper-mediated
oxygen activation in such systems.
Almost no mechanistic details are currently known for the
reactions shown in Scheme 1. Of particular note is the
complete lack of understanding of the role of counterion or
coligands in allowing the reaction to proceed in a facile manner.
In fact, mechanistic studies of oxidative C−C bond cleavage
reactions using copper and O2 have been limited to model
complexes for copper-containing quercetin dioxygenase
(QDO).7 Herein we report our discovery of dioxygenase-type
aliphatic C−C bond cleavage reactivity involving a Cu(II)
chlorodiketonate complex upon reaction with O2 at ambient
temperature. Importantly, initial mechanistic and computa-
tional studies of this reaction indicate that chloride ion can
serve as a catalyst to lower the barrier of the rate-determining
oxygen activation step in the reaction pathway. This is an
intriguing result that suggests that simple anions can play a key
role in Cu(II)/O2 systems that promote oxidative C−C bond
cleavage reactivity.
ne of the most important challenges for chemists in the
O21st century is to develop new ways for converting
chemical feed stocks (including biomass) into useful products,
including pharmaceuticals, polymers, and fuels, in an environ-
mentally benign and cost-efficient manner.1 Still under-
developed and of relevance to this goal are methods for the
selective oxidative cleavage of carbon−carbon bonds.2 Of
particular current interest in this area are reactions catalyzed by
earth-abundant metals where O2 is used as the terminal
oxidant.3 In this regard, a few reports have appeared
demonstrating aliphatic C−C bond cleavage mediated by
copper and O2 at room temperature and above.4 For example,
oxidative cleavage reactions involving cyclic ketones (Scheme
1a) and cyclic 1,3-diones (Scheme 1b) catalyzed by simple
Cu(II) salts and O2 have been reported.5 The reaction pathway
proposed in these systems involves the generation of a Cu(I)−
ketonyl radical pair which intercepts dioxygen. A very recent
report outlines the copper-catalyzed oxidative cleavage of
acyclic β-diketones (Scheme 1c) in a dioxygenase-type
pathway.6 This reaction proceeds in moderate to high yield
by combining a catalytic amount of CuBr or CuBr2 salt,
pyridine, the diketone substrate, and oxygen in toluene/
methanol at 90 °C.
The Cu(II) chlorodiketonate complex [(6-Ph2TPA)Cu-
(PhC(O)CClC(O)Ph)]ClO4 (1)8 was assembled as shown in
Figure 1 (top). X-ray crystallographic studies revealed a five-
coordinate copper center in which one of the phenyl-appended
pyridyl arms of the chelate ligand is dissociated from the metal
Received: March 13, 2014
Published: May 14, 2014
© 2014 American Chemical Society
7821
dx.doi.org/10.1021/ja502577b | J. Am. Chem. Soc. 2014, 136, 7821−7824

Journal of the American Chemical Society
Communication
Figure 2. Reaction of 1 with O2 in CH3CN at 25 °C. Benzil
(PhC(O)C(O)Ph) is generated via Lewis acid-promoted migration
chemistry of diphenylpropanetrione.15,16 The formation of this
byproduct provides evidence for the involvement of the trione in
the reaction. Previous studies have shown that divalent metal
complexes of the 6-Ph2TPA ligand promote this type of migration
chemistry.17
Figure 1. (Top) Synthetic route for the preparation of 1. (Bottom) A
thermal ellipsoid representation of the cationic portion of 1. Ellipsoids
are plotted at the 50% probability level. Hydrogen atoms are omitted
for clarity.
center (Figure 1, bottom), and the second phenyl-appended
pyridyl arm is elongated relative to the other nitrogen donors.
The geometry of the center is a distorted square-based pyramid
(τ = 0.40).9 Bond distances within the six-membered
diketonate chelate ring are consistent with a fully delocalized
monoanion, and the C−Cl bond distance (1.747(2) Å) is in the
range typically found for other chlorodiketonates (1.739−1.755
Å).10
Complex 1 has a prominent absorption feature at 363 nm (ε
= 10 200 M−1 cm−1) that is likely due to a π−π* transition
within the diketonate, but also may include some LMCT
character. The X-band EPR spectrum of 1 at 20 K (Figures S1
and S2) is consistent with retention of the distorted square
pyramidal geometry in solution.11
Exposure of a wet CH3CN solution of 1 to O2 leads to the
decay of the 363 nm absorption band (Figure S3) within 1 h at
ambient temperature, which is consistent with destruction of
the diketonate unit.12 Vapor diffusion of Et2O into a
concentrated CH2Cl2 solution of the product mixture
generated crystals of [(6-Ph2TPA)CuCl]ClO4 (2). Complex
2 contains a five-coordinate Cu(II) center in a distorted
trigonal bipyramidal geometry (τ = 0.61, Figure S4).9 The Cu−
Cl bond distance in 2 (2.1989(7) Å) is the shortest reported for
a mononuclear Cu(II)−Cl complex ligated by a tris(2-
methylpyridyl)amine-based ligand.13 A comparison of the
EPR spectra of the crude product mixture and independently
synthesized 2 (Figure S5) indicated that complex 2 was the
only Cu(II)-containing product generated in the reaction.
The fate of the diketonate ligand was determined by GC-MS
analysis of the products.14 In the presence of water, analysis
revealed the formation of benzoic acid, benzil and diphenyl-
propanetrione (Figure 2). Under dry conditions, an additional
product, benzoic anhydride, was identified. GC analysis of the
headspace of the reaction mixture revealed the formation of
CO2 (0.80 equiv) and CO (0.05 equiv). 18O2-labeling studies
revealed the incorporation of at least one labeled oxygen atom
into 60% of the benzoic acid, while an 16O2/H218O experiment
shows incorporation of labeled oxygen into 30% of the benzoic
acid (Figure S6). Combined, these results strongly suggest a
reaction pathway in which an α-substituted diketonate reacts
with dioxygen to produce a trione intermediate, as has been
observed in model systems for Ni(II)- and Fe(II)-containing
acireductone dioxygenases.17 Importantly, under dry condi-
tions, quantitative 18O incorporation was found in the
diphenylpropanetrione intermediate. Additionally, we have
found that admixture of equimolar amounts 6-Ph2TPA,
Cu(ClO4)2, diphenylpropanetrione, and sodium hypochlorite
gave the same product mixture and CO2:CO ratio as the
reaction starting from 1 and O2.
Based on the observed CO2:CO ratio, the quantitative
incorporation of a labeled oxygen into the trione intermediate,
the observation of benzoic anhydride under dry conditions, and
the congruence of products from the reaction of 1 with O2 and
the independent 6-Ph2TPA/Cu(II)/diphenylpropanetrione/
hypochlorite reaction, we propose that aliphatic C−C bond
cleavage in this system results primarily from the reaction of
hypochlorite ion with diphenylpropantrione, with these
reagents being formed via O−O cleavage and O−Cl bond
formation, respectively, within a Cu(II)−peroxo intermediate
(A, Scheme 2). Notably, this hypochlorite pathway differs from
the hydroperoxide pathway previously identified for aliphatic
C−C bond cleavage involving a coordinated diketonate having
a hydroxyl group in place of the chloride substituent.17
UV−vis kinetic studies of the reaction of 1 with O2 under
pseudo-first-order conditions with respect to oxygen revealed a
slow induction phase (kobs ≈ 3.0 × 10−5 s−1, 21 °C) followed by
a first-order decay process (Figure 3, left; kobs ≈ 3.5 × 10−3 s−1,
21 °C). This behavior suggested that a product formed in the
reaction could serve as a catalyst for the reaction. With that in
mind, we found that the induction period is not observed in the
presence of a catalytic amount of chloride ion (Figure 3, right).
A linear dependence of the observed rate constant on chloride
concentration was found (Figure S7). Overall, these initial
kinetic studies suggest that chloride ion lowers the activation
barrier for reaction of the Cu(II) diketonate complex with O2.
This is significant in terms of Cu(II) halide-mediated C−C
bond cleavage reactions (Scheme 1) involving O2 in that, while
7822
dx.doi.org/10.1021/ja502577b | J. Am. Chem. Soc. 2014, 136, 7821−7824

Journal of the American Chemical Society
Communication
to be most feasible for the initial steps of the reaction involves
the copper center exclusively in the Cu(II) oxidation state and
parallels the mechanism proposed for Cu(II)-containing
quercetin dioxygenase.20 Thus, first the enolate ligand changes
its coordination mode from bidentate to monodentate, which
opens up a coordination site on the copper center for
interaction with O2. As shown in Scheme 3, opening the
Scheme 2. Hypochlorite versus Hydroperoxide Pathways for
Aliphatic Carbon−Carbon Bond Cleavage
Scheme 3. Comparison of the Lowest Energy O2 Activation
Pathways for 1 and 1-Cl
coordination site and binding of O2 in the second coordination
sphere is less unfavorable for the chloride-binding complex 1-
Cl. This can be attributed, at least in part, to the different
orientation of the Jahn−Teller axes in 1 and 1-Cl. From the
analysis of spin density plots (Figures S8 and S12), it follows
2
that in 1 the doubly occupied dz orbital is perpendicular to the
O1−Cu−O2 plane, whereas in 1-Cl it points along the Cu−O2
bond vector, and hence cleavage of the Cu−O2 bond is
promoted by the Jahn−Teller effect. Once the coordination site
is open, O2 attacks the copper center and C2 in an
asynchronous manner, first forming a bond to copper and
partially oxidizing the enolate to enolate radical, and then
developing a bond between C(2) and O(4). The computed
Gibbs free energy of activation amounts to 25.8 and 22.8 kcal/
mol for TS and TS-Cl, respectively, which through transition
state theory corresponds to rate constants of 7.8 × 10−7 and 1.3
× 10−4 s−1 (25 °C), suggesting an expected >150-fold rate
enhancement in the presence of Cl−. The experimentally
observed ∼120-fold rate enhancement observed for the
reaction of 1 with O2 following the slow induction phase is
consistent with this proposal.
Overall, the studies reported herein provide evidence for an
intriguing role for chloride ion in Cu(II)/O2-promoted
oxidative carbon−carbon bond cleavage reactions involving a
diketonate substrate. The well-defined nature of 1 bodes well
for further systematic kinetic and mechanistic investigations of
analogues that will provide additional insight into the reaction
pathways involved in Cu(II)/O2-promoted oxidative cleavage
chemistry.
Figure 3. Absorbance (363 nm) versus time plots for reaction of 1
with O2 in the absence (left) and presence (right) of a catalytic
amount of chloride ion. Insets: ln (Absorbance) vs time plots. Data
were collected using O2-purged solutions.
bulky carboxylate anions have been shown to accelerate model
reactions for Fe(III)-containing QDOs, a similar anion effect
has not been reported involving halides or in reactions
involving a non-flavonol substrate.18
We have performed preliminary DFT studies to probe the
reaction pathway leading to formation of the proposed bridging
peroxo species A (Scheme 2) and to evaluate the effect of
chloride ion on the oxygen activation chemistry. Using a
truncated model wherein the phenyl appendages of the 6-
Ph2TPA ligand were replaced by hydrogen atoms,19 several
different scenarios for O2 activation were probed both in the
absence and in the presence of chloride ion, including (1)
electronic excitation to a Cu(I)−enolate radical charge-transfer
state followed by reaction with O2 and (2) direct attack of O2
on the central carbon of the enolate ligand. Additionally, we
investigated whether dissociation of one of the pyridyl donors
would lower the energy. The reaction pathway that was found
ASSOCIATED CONTENT
* Supporting Information
■
S
Details of experimental procedures; crystallographic files in CIF
format for 1 and 2; results of EPR experiments/simulations and
isotope labeling studies; absorption spectra for the reaction of 1
with O2; thermal ellipsoid representation of 2; chloride
dependence studies of reaction of 1 with O2; spin density
7823
dx.doi.org/10.1021/ja502577b | J. Am. Chem. Soc. 2014, 136, 7821−7824

Journal of the American Chemical Society
Communication
(14) Diketonate-derived products were separated via a short silica
column with >80% recovery; see Supporting Information for further
details.
plots. This material is available free of charge via the Internet at
http://pubs.acs.org.
(15) Roberts, J. D.; Smith, D. R.; Lee, C. C. J. Am. Chem. Soc. 1951,
73, 618.
AUTHOR INFORMATION
■
Corresponding Author
lisa.berreau@usu.edu
(16) Hydrated triketone may also be present in the reaction mixture.
Lewis acid-promoted migration chemistry involving this intermediate
would result in the formation of CO2 and benzoin, the latter of which
would be oxidized to benzil in the presence of OCl−.15
(17) (a) Berreau, L. M.; Borowski, T.; Grubel, K.; Allpress, C. J.;
Wikstrom, J. P.; Germain, M. E.; Rybak-Akimova, E. V.; Tierney, D. L.
Inorg. Chem. 2011, 50, 1047. (b) Allpress, C. J.; Grubel, K.; Szajna-
Fuller, E.; Arif, A. M.; Berreau, L. M. J. Am. Chem. Soc. 2013, 135, 659.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the National Science Foundation for support of this
research (CHE-1301092 to LMB; CHE-1152755 to DLT). AM
and TB acknowledge support from the National Science
Centre, Poland (2011/03/B/NZ1/04999).
(18) Barat
Vertes, A. Chem. Commun. 2009, 3630.
́
́ ́ ́
h, G.; Kaizer, J.; Speier, G.; Parkanyi, L.; Kuzmann, E.;
(19) Preliminary calculations on the full cation (containing the 6-
Ph2TPA ligand) provide evidence that the use of the smaller TPA
model is valid. Specifically, geometry optimized structures of both 1
and its smaller TPA analog both exhibit two axial Cu(II)−N(pyridyl)
(or phenyl−pyridyl) bonds. However, in both cases there are two
distinct Cu−N distances, with the longer bond in the 6-Ph2TPA-
ligated model structure corresponding to the pyridyl appendage that is
dissociated in the X-ray structure of 1. This being the case, the X-ray
structure was used to guide the computations: e.g., the pyridyl donor
that is dissociated in the crystal structure was rotated away to make
space for O2 binding in the next step. Overall, our computational
studies thus far suggest that while the inaccuracy of the geometry
optimized TPA structures may affect the energies to some extent, the
qualitative conclusions drawn likely have substantial validity.
REFERENCES
■
(1) (a) Gettys, N. S. J. Chem. Educ. 1998, 75, 665. (b) National
Research Council. Bioinspired Chemistry for Energy: A Workshop
Summary to the Chemical Sciences Roundtable; The National Academies
Press: Washington, DC. 2008. (c) National Research Council. Health
and Medicine: Challenges for the Chemical Sciences in the 21st Century;
The National Academies Press: Washington, DC, 2004.
(2) (a) Allpress, C. J.; Berreau, L. M. Coord. Chem. Rev. 2013, 257,
3005. (b) Crabtree, R. H. Nature 2000, 408, 415. (c) Jun, C.-H. Chem.
Soc. Rev. 2004, 33, 610.
(3) Constable, D. J. C.; Dunn, P. J.; Hayler, J. D.; Humphrey, G. R.;
Leazer, J. L., Jr.; Linderman, R. J.; Lorenz, K.; Manley, J.; Pearlman, B.
A.; Wells, A.; Zaks, A.; Zhang, T. Y. Green Chem. 2007, 9, 411.
(4) Zhang, L.; Bi, X.; Guan, X.; Li, X.; Liu, Q.; Barry, B.-D.; Liao, P.
Angew. Chem., Int. Ed. 2013, 52, 11303.
(20) Siegbahn, P. E. M. Inorg. Chem. 2004, 43, 5944.
(5) (a) Atlamsani, A.; Breg
J.; Belotti, D.; Bellosta, V.; Brocca, D. Tetrahedron Lett. 1994, 35, 6089.
(c) Bregeault, J.-M.; Launay, F.; Atlamsani, A. C. R. Acad. Sci., Ser. IIC:
́
eault, J.-M. Synthesis 1993, 79. (b) Cossy,
́
Chim. 2001, 4, 11. (d) Allen, S. E.; Walvoord, R. R.; Padilla-Salinas, R.;
Kozlowski, M. C. Chem. Rev. 2013, 113, 6234.
(6) Zhang, C.; Feng, P.; Jiao, N. J. Am. Chem. Soc. 2013, 135, 15257.
(7) (a) Kaizer, J.; Pap, J. S.; Speier, G. In Copper Oxygen Chemistry;
Karlin, K. D., Itoh, S., Eds.; John Wiley & Sons, Inc.: New York, 2011;
pp 23−52. (b) Pap, J. S.; Kaizer, J.; Speier, G. Coord. Chem. Rev. 2010,
254, 781.
(8) 6-Ph2TPA = N,N-bis((6-phenyl-2-pyridyl)methyl)-N-((2-pyrid-
yl)methyl)amine.
(9) Addison, A. W.; Rao, T. N.; Reedijk, J.; van Rijn, J.; Verschoor, G.
C. J. Chem. Soc., Dalton Trans. 1984, 1349.
(10) (a) Taylor, J. C.; McLaren, A. B. J. Chem. Soc., Dalton Trans.
1979, 3, 460. (b) Isakova, V. G.; Baidina, I. A.; Morozova, N. B.;
Igumenov, I. K. Polyhedron 2000, 19, 1097. (c) Ware, D. C.; Wilson,
W. R.; Denny, W. A.; Rickard, C. E. F. J. Chem. Soc., Chem. Commun.
1991, 1171. (d) Vock, C. A.; Renfrew, A. K.; Scopelliti, R.; Juillerat-
Jeanneret, L.; Dyson, P. J. Eur. J. Inorg. Chem. 2008, 10, 1661.
(e) Estes, E. D.; Scaringe, R. P.; Hatfield, W. E.; Hodgson, D. J. Inorg.
Chem. 1976, 15, 1179. (f) Kavounis, C. A.; Tzavellas, L. C.; Cardin, C.
J.; Zubavichus, Y. Struct. Chem. 1999, 10, 411. (g) Sans-Lenain, S.;
Gleizes, A. Inorg. Chim. Acta 1993, 211, 67. (h) Sharutin, V. V.;
Sharutina, O. K.; Zadachina, O. P.; Zakharova, A. N.; Reutov, V. A.;
Shapkin, N. P.; Velsky, V. K. Russ. J. Gen. Chem. 2000, 70, 1672.
(i) Allpress, C. J.; Arif, A. M.; Houghton, D. T.; Berreau, L. M.
Chem.Eur. J. 2011, 17, 14962.
(11) (a) Peisach, J.; Blumberg, W. E. Arch. Biochem. Biophys. 1974,
165, 691. (b) Bubacco, L.; van Gastel, M.; Groenen, E. J. J.;
Vijgenboom, E.; Canters, G. W. J. Biol. Chem. 2003, 278, 7381.
(12) The reaction proceeds in the presence or absence of light.
(13) (a) Harata, M.; Jitsukawa, K.; Masuda, H.; Einaga, H. Chem.
Lett. 1995, 24, 61. (b) Lucchese, B.; Humphreys, K. J.; Lee, D.-H.;
Incarvito, C. D.; Sommer, R. D.; Rheingold, A. L.; Karlin, K. D. Inorg.
Chem. 2004, 43, 5987.
7824
dx.doi.org/10.1021/ja502577b | J. Am. Chem. Soc. 2014, 136, 7821−7824