A Dicopper Dioxygenase Model Immobilized in Mesoporous Silica

Apr 15, 2019 - A Dicopper Dioxygenase Model Immobilized in Mesoporous Silica Nanoparticles for Toluene Oxidation: A Mechanism to Harness Both O ...
0 downloads 0 Views 4MB Size
Subscriber access provided by OCCIDENTAL COLL

Article

A Dicopper Dioxygenase Model Immobilized in Mesoporous Silica Nanoparticles for Toluene Oxidation: A Mechanism to Harness Both O Atoms of O for Catalysis 2

Chih-Cheng Liu, Yi-Fang Tsai, Chung-Yuan Mou, Steve S.-F. Yu, and Sunney I. Chan J. Phys. Chem. C, Just Accepted Manuscript • Publication Date (Web): 15 Apr 2019 Downloaded from http://pubs.acs.org on April 15, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 58 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

A Dicopper Dioxygenase Model Immobilized in Mesoporous Silica Nanoparticles for Toluene Oxidation: A Mechanism to Harness Both O Atoms of O2 for Catalysis Chih-Cheng Liu,† Yi-Fang Tsai,† Chung-Yuan Mou,*,‡ Steve S.-F. Yu,† and Sunney I. Chan*,†

†Institute

of Chemistry, Academia Sinica, No. 128, Sec. 2, Academia Road, Nankang,

Taipei 11529, Taiwan ‡Department

of Chemistry, National Taiwan University, No. 1, Sec. 4, Roosevelt Road,

Daan District, Taipei 10617, Taiwan

Corresponding authors *Sunney I. Chan (email: [email protected]; telephone: +886-2-5572-8654); *Chung-Yuan Mou (email: [email protected]; telephone: +886-2-3366-8205).

1 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 58

Abstract Recently, we have reported on a dicopper system a (CuIII(μ-O)2CuIII complex immobilized in mesoporous silica nanoparticles) that can mediate the catalytic conversion of toluene into benzaldehyde by O2, in which the oxidizing power of both O atoms is harnessed for catalytic turnover. This is the first example of a CuIII(μ-O)2CuIII complex capable of functioning like a “dioxygenase” in hydrocarbon oxidation. We have undertaken a mechanistic study to clarify how this catalytic conversion is accomplished without the input of sacrificial reductants. While the first O atom in the CuIII(μ-O)2CuIII complex can actively insert into a C–H bond, the second O atom left in the CuII(μ-O)CuII complex is inert. We show that a second molecule of O2 is involved in activating the dicopper catalyst, forming an O2 complex with the CuII(μ-O)CuII intermediate to give a species with the [Cu2O3]2+ core, which then mediates the transfer of the remaining O atom of the original O2 molecule to the organic substrate to complete the catalytic turnover. The study offers a mechanistically characterized analogue of the heterogeneous metal oxide catalyst that oxidizes organic substrates with the lattice oxygens by the Marsvan-Krevelen (MvK) mechanism at significantly higher temperatures.

2 ACS Paragon Plus Environment

Page 3 of 58 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Introduction The rich chemistry of copper complexes with dioxygen (O2) is often exploited to develop oxidation catalysts for controlled oxidation of hydrocarbons and other organics.1-3 Mononuclear,4,5 dicopper,6-8 as well as tricopper complexes,9-12 have all been considered. These complexes are intended to serve as biomimics of the active sites of copper monoxygenase enzymes13-16 that mediate the efficient oxidation of aliphatic C−H bonds under ambient conditions. The best studied dicopper biomimics are the CuIII(μ-O)2CuIII or CuII(μ-η2:η2-peroxo)CuII species formed by self-assembly of two tripodal mononuclear copper complexes using O2, or by O2 activation of dicopper complexes prepared from binucleating ligands capable of sequestering two CuI ions in close juxtaposition.1,2 Recently, a series of tricopper cluster complexes have also been prepared to mimic the active site of the particulate methane monooxygenase, which contains a tricopper cluster that catalyzes the conversion of methane into methanol efficiently under ambient conditions.9-11 As further evidence of the richness of copper/O2 chemistry, the CuII(μO)CuII sites in Cu-ZSM-5 and Cu exchange mordenites can also support selective methane oxidation at temperatures near 200 °C.8 The reactivity of the CuIIIO (or [Cu−O]1+) species toward methane oxidation in the gas phase has been known for some years.17,18 A tricopper-oxo cluster stabilized in a copper exchange mordenite has also recently demonstrated the efficacy of the [Cu3(μ-O)3]2+ trinuclear copper system toward methane oxidation at 200 °C.19 As with the active sites of copper monooxygenase enzymes, all of the model copper systems are CuI complexes or species with CuI sites that can be used to activate O2. Upon activation, one of the O atoms is typically transferred to a C−H bond of the hydrocarbon

3 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 58

while the O atom remaining in the CuII(μ-O)CuII species is reduced by two additional reducing equivalents to form a molecule of H2O with the assistance of two protons. The CuII(μ-O)CuII species as written is expected to be relatively stable and inert, at least at room temperature. However, it has been proposed that the structure can rearrange to give the more reactive CuII oxyl or CuIII-oxo, namely, CuI···O-CuII or CuI···O=CuIII.2 If so, either the oxyl or oxene is poised for abstraction of a hydrogen atom from an aliphatic C−H bond of the organic substrate followed by geminal recombination of the OH and the alkyl radicals to form the alcohol product. This is the chemistry that has been proposed to take place at the CuII(μ-O)CuII sites in Cu-ZSM-5 and Cu-mordenite toward methane oxidation at 200 °C.8 In these zeolites, it appears that the first of the two O atoms is reduced to form water and the oxidation chemistry is mediated by the remaining CuII(μ-O)CuII species as just described.8 It would be an advance in the field of copper/O2 chemistry if we could harness the oxidizing power of both O atoms of the O2 molecule and exploit it for the hydrocarbon oxidation process. The copper complex would then function as a dioxygenase, which uses both O atoms of the O2 to mediate the oxidation of a substrate. Only one copper dioxygenase has been firmly established; however, the chemistry appears to be quite different and more complex than what we are describing here.20, 21 We have recently reported a CuIII(μ-O)2CuIII complex in mesoporous silica nanoparticles (MSN), henceforth referred to as CuIII(μ-O)2CuIII@MSN, that is capable of catalyzing aerobic oxidation of the aliphatic C−H bonds in toluene to yield benzyl alcohol (BnOH) and further to benzaldehyde (Bz) at ambient temperatures under mild conditions (Scheme 1).22 The CuIII(μ-O)2CuIII@MSN is formed by immobilizing the

4 ACS Paragon Plus Environment

Page 5 of 58 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Cu(II) complex of a tripodal tridentate histidine-like ligand, CuIIImph (Imph = bis(4imidazolyl methyl)benzylamine) (Chart 1), in the nanochannels of the MSN, followed by reduction of the Cu(II) complexes and activation by O2. The CuIII(μ-O)2CuIII dicopper complex, with only O2 as the bridging ligand, is apparently stabilized by confinement within the nanopores of the MSN. An interesting and significant finding related to the catalytic system is that no additional reducing equivalents are required to sustain multiple turnovers of the catalyst for the conversion of toluene into BnOH and Bz other than the initial minute amounts of sodium ascorbate required to reduce the CuIIImph to CuIImph before activation of the complex by O2 to generate the CuIII(μ-O)2CuIII intermediate 1, namely, the [{CuIIIImph}2(μ-O2−)2]2+ species. Thus, both O atoms of the O2 molecule used to produce the activated dicopper species are participating in the substrate oxidation. In other words, the [{CuIIIImph}2(μ-O2−)2]2+ complex is functioning like a “dioxygenase”. Here, we describe a kinetic and mechanistic study on the reaction illustrated in Scheme 1 in order to develop an understanding of how this dicopper system carries out this interesting and unique chemistry. Since the catalytic cycle for the oxidation under consideration is analogous to the Mars-van-Krevelen mechanism for O2 activation and replacement of the lattice oxygen atoms in heterogeneous metal-oxide catalyzed oxidations at high temperatures,23-25 the outcome of the present study on a well defined heterogenized system can bridge our understanding between heterogeneous and homogeneous oxidation of C-H bonds mediated by copper-oxo species.

Scheme 1. Conversion of toluene into BnOH and Bz mediated by a CuIII(μ-O)2CuIII complex formed by immobilization of the positively charged tripodal tridentate Cu(II) 5 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 58

complex CuIIImph in negatively charged MSN (MSN-TP-3) followed by reduction using sodium ascorbate and activation by O2.

Chart 1. Chemical structures of CuIIImph and the copper-oxo species 1−3 derived from the reaction of this complex with O2 in the CuImph@MSN nanoparticles. 6 ACS Paragon Plus Environment

Page 7 of 58 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Materials and Methods Materials All solvents and chemicals used in the studies were of commercially available analytical grade, if not mentioned otherwise. Solvents for air sensitive reactions were distilled under argon. All preparations of the materials (including the ligand Imph, the CuIIImph, MSN-TP-3, and the procedures used to immobilize the CuIIImph in the MSN-TP-3) have been previously reported.22 The MSN-TP-3 was synthesized using tetraethylorthosilicate as a precursor under basic conditions (ammonia solution). In order to facilitate anchoring of the positively charged copper complexes, the MSN was functionalized by 5 % anionic 3-(trihydroxysilyl)-propylmethyl-phosphonate (TP) to generate a negatively charged surface within the MSN channels. The CuIIImph complex was immobilized in the nanochannels of silica by the ion-exchange method via electrostatic coulomb attraction. The loading of the copper complex in the nanoparticles was determined by measuring the copper content by ICP-MS and C, N elemental analysis. The amount of CuIIImph encapsulated by the nanosized MSN-TP-3 sample is 3.23 × 10−4 mol g−1. This corresponds to a loading efficiency of 27.6 %. The excessive Cu species are removed by washing with enthanol. The BET surface area of the CuIIImph@MSN-TP-3 is 630 m2 g−1 and the BET pore volume is 0.32 cm3 g−1. The zeta-potential is −26 mV. Toluene oxidation by CuImph@MSN-TP-3 during single-turnover 30 mg of the CuIIImph@MSN-TP-3 sample was well suspended in O2-free THF (5 ml) in a 10 ml Schlenk flask, and sodium ascorbate was added as a reducing agent (4

7 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 58

equiv. based on the amount of CuIIImph (9.7 μmol) in the MSN-TP-3 that was freshly prepared as a 1 M solution in de-ionized water) under a N2 atmosphere. The heterogeneous mixture was stirred vigorously at room temperature to generate the CuIImph@MSN-TP-3 sample in situ. The mixture was then degassed, and O2 gas (99.999 %) was bubbled into the flask and stirred for 5 min. The sample was then frozen (ca. −120 oC), and the “excess” O2 was pumped away. Then the reacting substrate toluene (0.03 ml, technical grade) (29 equiv.) was added and the products analyzed periodically by using GC-MS. In a control experiment, after 60 minutes of incubation, when a turnover number of 0.93 was obtained for the initial product BnOH without producing any of the final product Bz, the system was re-purged with pure O2 and the products analyzed once again at various intervals. Five additional CuImph@MSN-TP-3 samples were then studied. In one experiment, the conversion of toluene into BnOH and Bz was followed at room temperature. In two other experiments, the samples were incubated for 40 min at either 10 °C or 50 °C, following an initial 70-min reaction with toluene at room temperature, and the products of the toluene oxidation were analyzed at various intervals up to 5−7 h. In two final experiments, the effects of introducing a stoichiometric pulse of N2 or O2 during incubation of the system at room temperature were also studied. 0.1 ml (NTP) of pure N2 gas or pure O2 gas was injected into the sample at 90 min into the single-turnover experiments. Preparation of the dicopper species 1, 2, and 3 for spectroscopic characterization Experimental details on the preparation of the dicopper speices 1, 2, and 3 in the CuImph@MSN-TP-3 samples for UV-Visible spectra and EPR measurements are given

8 ACS Paragon Plus Environment

Page 9 of 58 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

in Supporting Information. For Cu K-edge X-ray absorption measurements, samples of the dicopper species were prepared as below. (i) Species 1. The formation of the bis(μ-oxo)dicopper(III) intermediate, species 1, was carried out as follows. 50 mg of the CuImph@MSN-TP-3 sample was suspended in O2free THF (5 ml) in a 10 ml Schlenk flask, and sodium ascorbate was added as a reducing agent (4 equivalents based on the amount of CuIIImph (16.2 μmol) in the MSN that was freshly prepared as a 1 M solution in de-ionized water) to generate the CuIImph@MSNTP-3 sample in situ under a N2 atmosphere. The heterogeneous mixture was then directly vacuum-dried and the solid was saturated with O2 by using a gas syringe for 5 min. (ii) Species 2. To obtain the dicopper species 2, a solid sample of the CuIII(μ-O)2CuIII species 1 was first prepared. 5 ml O2-free THF was then added and when a high-quality suspension was obtained, the mixture was frozen (ca. −120 °C) and the “excess” O2 in the system was pumped away by vacuum drying and re-purging with N2(g) for 5 times. Species 2 was produced upon performing the single-turnover experiment under O2limiting conditions at room temperature, the reaction time suggested by the results of a controlled single-turnover experiment as described earlier, and the UV-Visible spectrum recorded. Then, the mixture was vacuum-dried overnight at room temperature to remove the solvent, the toluene substrate, the product BnOH, and all the O2 in the sample, including any O2 that might be trapped in the micropores of the Cu-complex-loaded MSN. (iii) Species 3. After performing the single-turnover experiment under O2-limiting conditions and forming the CuII(μ-O)CuII species 2 (as confirmed by UV-Visible spectrum of the suspension, vide infra), the mixture was directly vacuum-dried at room temperature overnight to remove the solvent, the toluene substrate, the product BnOH,

9 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 58

and all the O2 in the sample, including any O2 that might be trapped in the micropores of the Cu-complex-loaded MSN. At this point, O2 was introduced to directly activate species 2 to species 3 for X-ray absorption measurements. X-ray absorption measurements on the CuImph@MSN-TP-3 samples The CuImph@MSN-TP-3 samples were subjected to Cu K-edge X-ray absorption measurements during different stages of the single-turnover experiments. The X-ray absorption near edge structure (XANES) and extended X-ray absorption fine structure (EXAFS) were recorded on the wiggler beamline BL-17C1 at the National Synchrotron Radiation Research Center (NSRRC) in Hsinchu, Taiwan. Fluorescence data were collected using an Ar-filled ionization chamber detector equipped with a Ni filter and Soller slits (Lytle detector in the fluorescence detection mode) in the energy range from 8.779 to 9.779 keV. All spectra were recorded at room temperature on solid samples of the dicopper species as prepared above after transferring to polyethylene (PE) zipper bags. The EXAFS was extracted from each spectrum in the standard way (see Supporting Information). Typically, 3 or 5 scans were sufficient to obain EXAFS data suitable for analysis. The X-ray data collected were merged and subjected to data analysis. Other experimental procedures and methods The experimental procedures used for characterization of the CuImph@MSNTP-3 samples, GC-MS analysis of the products during catalytic turnover, recording of the UV-Visible, electron paramagnetic resonance (EPR), and Raman spectra of the MSN samples; and details pertaining to the analyses of the EPR and EXAFS data are given in the “Materials and Methods” section of Supporting Information. 10 ACS Paragon Plus Environment

Page 11 of 58 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Results Nano-confinement enables immobilization of the CuIII(μ-O)2CuIII species and control its stability and reactivity Many CuIII(μ-O)2CuIII or CuII(μ-η2:η2-peroxo)CuII systems have now been structurally characterized and their reactivity toward organic substrates examined under varying conditions. Generally speaking, these oxygenated dicopper species are relatively inert. Not only do these complexes exhibit low reactivity toward exogenous substrates, but also they do not tend to support multiple catalytic turnovers. Under ambient conditions, the oxygenated dicopper species are unstable and dissociate at various stages of the chemistry. To circumvent these difficulties, we have immobilized the Cu(II) complex of a tripodal tridentate histidine-like ligand (Chart 1) in the nanochannels of the MSN-TP-3 and formed the CuIII(μ-O)2CuIII species upon reduction of the Cu(II) complexes followed by activation by O2. The immobilized CuIII(μ-O)2CuIII complex in MSN-TP-3 thus formed is capable of catalyzing aerobic oxidation of the aliphatic C−H bonds in toluene to yield BnOH and further to Bz at ambient temperatures under mild conditions (Scheme 1).22 In the presence of O2, we observe good catalytic activity with multiple turnovers, and the O2 complex exhibits high stability. Evidently, the confinement of the monocopper model complexes in the nanochannels affords the proper distance and geometry constraints to form the stable CuIII(μ-O)2CuIII species, creating the reactive intermediate 1 (Chart 1) at room temperature for O-atom transfer to the C−H bond of an appropriate

11 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 58

substrate. Confinement is also important to avert dissociation of the dicopper complex at various stages during the catalytic cycle.26-28 Interestingly, the system is capable of multiple catalytic turnovers without input of additional reducing equivalents. In other words, the catalytic oxidation of toluene to Bz by O2 mediated by our Cu(I) catalyst can proceed without the consumption of sacrificial reductants. Thus, it appears that the CuII(μ-O)CuII species 2, namely, the ([{CuIIImph}2(μ-O2−)]2+) intermediate (Chart 1) formed after the O-atom transfer from the CuIII(μ-O)2CuIII species 1 to toluene, is mediating the further oxidation of the BnOH to Bz. Single turnover experiments offer mechanistic insights To elucidate the mechanism, we have undertaken single-turnover experiments to follow the individual kinetic steps in the catalytic cycle. Based on the time course of formation of the products and the product yields, the CuIII(μ-O)2CuIII species 1 mediates the oxidation of toluene to Bz according to the consecutive reaction shown in Scheme 1, with the “O” equivalents provided by the CuIII(μ-O)2CuIII species 1. However, Scheme 1 does not provide mechanistic insights, as the two steps merely describe the conversion of toluene into BnOH and Bz without consideration of changes in the dicopper system mediating the toluene oxidation reaction. (i) The initial step in the toluene oxidation. As before, we have initiated the reduction of CuIIImph to CuIImph by sodium ascorbate and the formation of the CuIII(μ-O)2CuIII species 1 by O2 in the nanochannels of the MSN-TP-3. To verify the conversion of toluene into BnOH, we have now performed a single-turnover experiment. Upon formation of intermediate 1, we quickly remove any excess O2 available to the activated 12 ACS Paragon Plus Environment

Page 13 of 58 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

dicopper complex and follow the time course of the product formation upon the addition of the toluene substrate. (The overall reaction is sufficiently slow that the involvement of O2 in the subsequent steps of the catalytic turnover could be addressed in this manner.) We find that only one molecule of BnOH is produced per activated dicopper cluster within 1 h according to GC-MS analysis (Figure 1a). However, upon re-purging of the sample with O2 at the end of this 1-h period, additional BnOH is formed and Bz is detected. Interestingly, without the re-introduction of the O2, no further oxidation of the BnOH to Bz is observed for the next 2.5 h (reaction time points 1 to 3.5 h, Figure 1b). At the end of the 2.5-h period (reaction time points 3.5 to 6 h), however, the BnOH is finally converted into Bz with a turnover number of 1, as confirmed by GC-MS analysis (Figure 1b). This kinetic behavior of the single-turnover chemistry is unexpected. In any case, the pause following the formation of the BnOH before its eventual conversion into Bz suggests that the CuII(μ-O)CuII species 2 is relatively inert under conditions of limiting O2. That the dicopper species is still intact with a bridging μ-oxo at this point of the turnover is evident from Cu K-edge EXAFS experiments on the MSN-TP-3, where we observe strong Cu···Cu back scattering with a short Cu···Cu distance (vide infra). The UV-Visible spectrum of species 2 is also consistent with a CuII(μ-O)CuII species (vide infra). (ii) The second step in the turnover: Conversion of BnOH into Bz during incubation of the sample. To explore the origin of the hiatus mentioned earlier, we have incubated the sample, after the formation of the BnOH, at 10 °C and 50 °C (some twenty degrees centigrade below and above the temperature at which the original experiment is

13 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 58

conducted) for 40 min (reaction time points 70 to 110 min) before continuing the toluene oxidation reaction at room temperature. When the sample is incubated at 10 °C, there is no noticeable change in the hiatus period (Figure 2a). However, incubating the sample for 40 min at 50 °C shortens the hiatus period from 2.5 h to 2 h. In fact, if the sample is incubated at 50 °C for 2 h, the hiatus is abolished (Figure 2b). These observations have led us to surmise that the termination of the hiatus is linked to some subtle but cooperative structural change within the functionalized silica framework of the MSN-TP3, one that allows residual O2 occupying the micropores within the framework to be released into the nanochannels, where the catalysis mediated by the dicopper cluster is taking place. If so, the conversion of BnOH into Bz in the reaction scheme highlighted earlier is promoted by O2, possibly a molecule of O2.

Figure 1. (a) The oxidation of toluene to BnOH (black line) and BnOH to Bz (red line) were performed under O2 limiting conditions and the products detected by GC-MS from 0 to 60 min (Region I: pink shaded region). After that, the oxidation reactions were continued under excess O2 conditions and the products BnOH and Bz monitored by GC-

14 ACS Paragon Plus Environment

Page 15 of 58 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

MS up to 150 min (Region II: yellow shaded region). (b) Time course of the singleturnover experiment studied at room temperature at 30-min intervals up to 6 h.

Figure 2. Upper panels: Effects of a 40-min incubation of the sample at (a) 10 °C and (b) 50 °C, on the conversion of BnOH into Bz following the initial 1-h reaction with toluene at room temperature. Lower panels: Effects of introducing a stoichiometric aliquot of (c) dry N2 or (d) pure O2, 90 min into the hiatus period during incubation of the sample at room temperature.

15 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 58

(iii) Interrupting the “annealing” process in the single-turnover experiment. To verify that an additional O2 molecule is involved in the second step of the single-turnover reaction, we compare the effects of interrupting the incubation or “annealing” of the sample by introducing a pulse of gas consisting of either a stoichiometric aliquot of (a) dry N2, or (b) pure O2, into the sample 90 min into the hiatis period. As shown in Figure 2c-d, the continuation of the catalytic cycle and formation of Bz is immediately apparent only in the case of (b), emphasizing the role of O2 in the second kinetic step of the singleturnover reaction. These results provide definitive evidence for participation of a second O2 molecule in the catalytic cycle. Stoichiometry of the individual kinetic steps If a second O2 molecule is involved in the second step of the single-turnover, then the “annealing” process would culminate in a well-defined product stoichiometry of one Bz molecule and one H2O molecule as the BnOH formed in the first step is converted into Bz, concomitant with return of the dicopper catalyst to the CuIII(μ-O)2CuIII species 1. This is indeed the case. To demonstrate this, we have repeated the single-turnover experiment using d8-toluene. As shown in Figure 3, one molecule of d8-benzyl alcohol (d8-BnOD) is produced initially, and one D2O molecule is released in parallel with the one d6-benzaldehyde (d6-Bz) molecule formed during the progression of the “annealing”. Even after re-purging O2 into the catalytic system at the end of the single turnover experiment, the quantity of D2O produced corresponds with the stoichiometric amount of d6-Bz (d6-Bz: 155.92 μmol, D2O: 156.05 μmol) expected in the multiple turnover reaction, as shown in Figure S1 in Supporting Information. Thus, in either the singleturnover or multiple-turnover experiment, the eight deuterium atoms of the original d8-

16 ACS Paragon Plus Environment

Page 17 of 58 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

toluene can be seen to be fully accounted for (i) by the d8-BnOD produced initially; and (ii) in the two final products: six deuterium atoms in the d6-Bz plus two deuterium atoms in the D2O. However, only two O atoms have managed to find their ways into the two final products, one atom in each of the two substrate products. Clearly, one O atom has originated from the first O2 molecule, and the second is coming either from the first O2 as well, or from the second O2 molecule. Since there are no other substrate products, we surmise that the remaining two O atoms must have become part of the dicopper cluster; that is, the dicopper catalyst has reverted back to the CuIII(μ-O)2CuIII species 1 upon completion of the turnover cycle, as confirmed by UV-Visible spectrum of the dicopper catalyst (vide infra).

Figure 3. The time course of the single-turnover experiment in which d8-toluene is oxidized to d8-BnOD (black data points), and further converted to the final products, d6Bz (red data points) and D2O (blue data points), at room temperature over the 6-h incubation period.

17 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 58

Evidence for micropores within the CuIIImph@MSN-TP-3 Evidence for the existence of micropores within the CuIIImph@MSN-TP-3 has come from measurements of the nitrogen (N2) adsorption-desorption isotherms (BET) (Figure S2). A t-plot analysis (Figure S3) of the N2 adsorption data indicates that immobilization of the CuIIImph in the MSN-TP-3 has converted a large portion of the mesopores into micropores (pore size less than 0.5 nm). Whereas there are practically no micropores in MSN-TP-3, the micropore surface area and pore volume are ~50 % in the CuIIImph@MSN-TP-3. The total pore volume of the MSN-TP-3 has been reduced by almost 50 % (Table 1) after loading with CuIIImph to give a substantial fraction of micropores. The stored O2 molecules in the micropores are apparently resistant to the freeze-pumping (ca. −120 °C) in the single-turnover experiment and they participate in the activation of the CuII(μ-O)CuII species 2 upon “annealing“ of the sample under ambient conditions. It is possible to rid the sample of all the O2 associated with the Cu-complex-loaded MSN, including any O2 that might be trapped in the micropores, by direct overnight vacuum drying (vide infra). If this procedure is applied to the sample after the first step of the single-turnover experiment, the solvent, the initial toluene substrate, and the product BnOH, are also removed (vide infra). Upon re-suspension of the MSN in THF, an aliquot of BnOH can be added to attempt resumption of the catalytic turnover. However, no production of Bz is detected in this experiment, underscoring the inertness of species 2 and the requirement for additional O2 to sustain the catalysis (see Figure S4).

Table 1. Textural properties of the MSN-TP-3 and CuIIImph@MSN-TP-3 samples.a 18 ACS Paragon Plus Environment

Page 19 of 58 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

bA

Sample

BET

(m2g−1)

cMiA

BET

(m2g−1)

dV

p

eMiV

p

MiVp



(cm3g−1)

(cm3g−1)

(%)

(mV)

MSN-TP-3

860

49

0.58

0.013

2.2

–40

CuIIImph@MSN-TP-3

630

330

0.32

0.15

47

–26

a

Errors are estimated to be within ±2% for surface area and pore volume, and ±0.5 mV

for zeta potential; bABET: surface area; c MiABET: surface area of the silica micropores;

d

Vp: pore volume; e MiVp :pore volume of the silica micropores; f ζ: zeta potential. The surface area and pore volume of the micropores are calculated by the t-plot method (0.3~0.5 of N2 thickness).

Spectroscopic signatures of the catalytic dicopper species 1, 2 and 3 in the MSN-TP3 The Cu K-edge XANES provides information on the coordination charge of the copper ions in the various dicopper species. For copper-oxo species, the UV-Visible spectrum offers ligand-to-metal charge transfer (LMCT) transitions that are diagnostic of the nature of the coordinating oxo species. In principle, additional structural information on the coordinating oxo-species can be derived from the resonance Raman (rR) spectrum. Here we compare the Cu K-edge XANES and the UV-Visible spectra of the dicopper species implicated in the catalytic cycle. We briefly describe our attempts to obtain rR spectra of these intermediates in the Supporting Information. (i) Cu Kα X-ray absorption near-edge spectra (XANES). Cu K-edge XANES in the preedge region of the CuIIImph immobilized in the MSN-TP-3, as well as the CuIII(μO)2CuIII species 1 and species 2 formed during the initial step of the turnover in the MSN-TP-3, are given in Figure S5. The positions of the pre-edges indicate that the Cu 19 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 58

oxidation states are Cu(II), Cu(III), and Cu(II) for these three species, respectively. The positions of the pre-edge and the near-edge features in the X-ray absorption spectrum of species 3 suggest the Cu(III) oxidation state for the copper ions in this O2 adduct. (Figure S5). (ii) UV-Visible spectra. The UV-Visible spectra of the dicopper species 1, 2, and 3 are compared in Figure 4. In our earlier work,22 we obtain the spectrum of CuIII(μ-O)2CuIII species 1 shown in Figure 4a after exposing a solid sample of the CuIImph@MSN-TP-3 to O2 at room temperature. An intense UV–Visible absorption band at 390 nm is observed.22 Both the position and intensity of this absorption band is characteristic of a LMCT transition for a species with the [CuIII(μ-O)2CuIII]2+ core, namely, the O2− (σu*) → Cu (III) (dxy) LMCT of an bis(μ-oxo)dicopper(III) species 1 (see Figure 4a, inset).22, 29, 30 The UV–Visible spectrum of species 2 (Figure 4b) exhibits an intense μ-O2− to Cu(II) charge transfer band near 300 nm (ε = 10,421 M−1 cm−1) expected of a species with the [CuII(μ-O)CuII]2+ core and a weak absorption at 642 nm (ε = 1,730 M−1 cm−1) originating from Cu(II) d–d transitions. Following earlier work (see Figure 4b, inset),22, 31 we assign species 2 to the CuII(μ-O)CuII intermediate, or the (μ-oxo)dicopper(II) species. The UV-Visible spectrum of species 3, formed upon the introduction of O2 to activate species 2 in the absence of substrate is shown in Figure 4c. In addition to the moderate μO2− to Cu(III) charge-transfer band at 390 nm (ε = 2,218 M−1 cm−1), an even more intense band is observed at 516 nm (ε = 3,207 M−1 cm−1), together with a well-defined weaker transition at 626 nm (ε = 1,017 M−1 cm−1). These spectroscopic results provide evidence for an end-on peroxide species with the [CuIII(μ-O)((μ-1,2)-peroxo)CuIII]2+ core, namely,32

20 ACS Paragon Plus Environment

Page 21 of 58 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

for the copper cluster in species 3. We have assigned the 390 nm band to the μ-O2− to Cu(III) LMCT transition, and the absorptions at 516 nm and 626 nm to the μ-O22− (π*σ)  Cu(III) (dx2-y2 + dx2-y2) and μ-O22− (π*ν) Cu(III) (dx2-y2 + dx2-y2) LMCT transitions, respectively, of the proposed peroxo-bridged CuIII(μ-O)CuIII complex in the putative O2adduct of the [CuII(μ-O)CuII]2+ species 2.22,31 Species 3 is light sensitive and reverts back to species 2. Upon recording the UVVisible spectrum of species 3 in successive scans, both the 390 nm and 516 nm bands diminish in intensity and the intensity of the 300 nm band of species 2 is restored. The time course of decay of species 3 upon exposure to UV light during three successive scans of the spectrum is depicted in Figure 4d. This sensitivity of species 3 to UV light precludes the possibility of obtaining the conventional rR spectrum for this species.

21 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 58

Figure 4. (a) Solid state UV-Visible absorption spectra of CuIIImph@MSN-TP-3 (black line) and [{CuIIIImph}2(μ-O2−)2]2+@MSN-TP-3 (species 1, red line) at room temperature, taken form our earlier work.22 For comparison, the UV-Visible spectrum of the bis(μ-oxo)dicopper(III) species [(TMED)2Cu(III)2(μ-O)2]2+ (black solid line) is shown in the inset.30 (b) The UV-Visible spectrum of [{CuIIImph}2(μ-O2−)]2+@MSNTP-3 (species 2) well suspended in THF at room temperature. The absorption at 642 nm is assigned to the (μ-oxo)dicopper(II) species by comparison to the reference [LCuII(μO)CuIIL] complex, L = Me2C6H3Xanthium in the inset.31 (c) The UV-Visible spectrum of the putative [{CuIIIImph}2(μ-O2−)(O22−)]2+@MSN-TP-3 (species 3). (d) The time course of decomposition of the light-sensitive species 3 with repeated scans of the UV-Visible spectrum: first scan at t = 0 (black); second scan 5 s later (red); and a third scan 15 s later (blue). The blue spectrum obtained after photobleaching corresponds to that of species 2,

22 ACS Paragon Plus Environment

Page 23 of 58 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

which is stable. Panels (a) and (b) are adapted with permission from ref. 22. (Copyright 2015 Elsevier B.V.)

Electron paramagnetic resonance (EPR) redox titration experiments support assignment of copper oxo cores The MSN sample containing species 1 is EPR silent. This result indicates all the CuIIImph complexes immobilized in the MSN have been converted into CuIImph or converted into the CuIII(μ-O)2CuIII or CuII(μ-η2:η2-peroxo)CuII species. CuIImph is EPR silent; however, CuIIImph is paramagnetic. If there are residual Cu(II) complex within the nanoparticles, namely CuIIImph, we would have detected them by EPR. Subsequent EPR redox titrations of this sample (vide infra) show that the intensity of the CuIIImphperoxo species (CuII-OO2−) detected is totally consistent with the level of species 1 expected based on the copper content of the MSN and the total conversion of all the CuIImph complexes into species 1. The two copper atoms in CuIII(μ-O)2CuIII are Cu(III), which is not paramagnetic. The two copper atoms in CuII(μ-η2:η2-peroxo)CuII are Cu(II), but they are antiferromagnetically coupled. These results indicate complete formation of species 1. Although the dicopper species 1, 2, and 3 are EPR silent, it is possible to elicit an EPR spectrum for each of these species by redox titrations and the outcome of these experiments can be used to offer support of our assignments of their oxo cores. To verify that species 1 is a dicopper intermediate with the [Cu(O)2Cu]2+ core, either the CuIII(μ-O)2CuIII or the CuII(μ-η2:η2-peroxo)CuII species, immobilized in the nanochannels of the MSN-TP-3, we have reduced species 1 (0.67 μmol) by one reducing

23 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 58

equivalent from decamethylferrocene (FeCp*2) to produce a Cu(II) species with a rhombic EPR feature (g// = 2.217; g, g = 2.029, 2.022; A// = 171 G; A = A ≅ 0), as shown in Figure 5, upper panel. (Note that the EPR of the corresponding decamethylferrocenium generated by the redox reaction is not detected at 77 K).33 Comparison of this EPR signature with the corresponding spectrum obtained by mixing the CuIIImph complex in O2-free THF with 1−5 equiv. of H2O2 (35 % in H2O) in the presence of Et3N (middle panel: g// = 2.218; g, g = 2.052, 2.022; A// = 171 G; A = A ≅ 0), we conclude that the redox titration of species 1 has led to the formation of a CuIIImph-peroxo species (CuII-OO2−) together with the EPR-silent valence-trapped CuIImph complex in the nanochannels of the MSN-TP-3. This is the expected scenario, if species 1 is a dicopper species with two Cu(III) ions bridged by a bis(μ-oxo) or two Cu(II) ions bridged by a peroxo. To emphasize that the EPR spectra of these species are distinct from that of the CuIIImph complex immobilized in the MSN-TP-3, we show the spectrum of the latter in the bottom panel of Figure 5. This complex exhibits the typical type 2 Cu(II) EPR spectrum with g// = 2.247, g = 2.055, A// = 162 G, and A ≅ 0, when it is immobilized in MSN-TP-3.

24 ACS Paragon Plus Environment

Page 25 of 58 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Figure 5. X-band EPR spectra at 77 K. Upper panel: The [{CuIIIImph}2(μ-O2−)2]2+ intermediate (species 1) (black line) and the spectrum recorded after reducing species 1 with 1 equiv. of decamethylferrocene (FeCp*2) (red line). Middle panel: Spectra obtained after reaction of the CuIIImph complex in O2-free THF with 1 equiv. of H2O2 (blue line), and 5 equiv. of H2O2 (green line) in the presence of Et3N. Lower panel: CuIIImph@MSN-TP-3 sample (pink line). g and A values of the various EPR spectra are obtained by computer simulations (see Figure S6).

25 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 58

Similarly, an EPR signal is obtained upon reduction of species 2 with FeCp*2 (Figure 6). This EPR spectrum corresponds to that of the [(CuIIImph)(μO2−)(CuIImph)]1+ complex, or the CuII(μ-O)CuI species, with one unpaired spin delocalized equally over the two copper ions as manifested by the

63/65Cu

nuclear

hyperfine splitting pattern of 7 lines with intensity ratio 1:2:3:4:3:2:1 in the parallel region, corresponding to coupling to two 63/65Cu (I = 3/2) ions with A// ≈ 80 Gauss. This observed A// is ½ of the corresponding value in the CuIIImph complex by itself, as expected. The intensity of this EPR spectrum reaches a maximum with one equiv. of the reductant (EPR intensity corresponds to the amount of FeCp*2 added); with additional equivalents of FeCp*2, both CuIIImph centers of the CuII(μ-O)CuI species are reduced to CuIImph and the EPR intensity vanishes. A simulated spectrum with g// = 2.296, g = 2.047; and A// = 80 G, A ≅ 20 G is shown in Figure 6a for comparison. The same EPR signal is obtained upon reduction of species 3 with 1 equiv. of FeCp*2 (Figure S7), indicating that species 3 is merely an O2-adduct of species 2. Upon reduction of species 3 by an electron, the bound O2 is released and the [{CuIIImph}(μO2−){CuIImph}]1+ species is generated within the nanochannels of the MSN-TP-3. The outcomes of the above EPR experiments, namely [{CuIIIImph}2(μ-O2−)2]2+ + e−1  CuIIImph-OO2− + CuIImph [{CuIIImph}2(μ-O2−)]2+ + e−1

 [{CuIIImph}(μ-O2−){CuIImph}]1+

{[{CuIIIImph}2(μ-O2−)(O22−)]2+ + e−1  [{CuIIImph}(μ-O2−){CuIImph}]1+ + O2. are entirely consistent with the proposed [Cu(O)2Cu]2+ core for species 1; the loss of one O-atom to produce the [Cu(O)Cu]2+core in species 2 after the first step of the catalytic

26 ACS Paragon Plus Environment

Page 27 of 58 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

turnover; and species 3 is just a simple O2-adduct of species 2, with the capability to transfer a second O-atom to the substrate in the second step of the catalytic turnover.

Figure 6. X-band EPR spectra at 77 K: (a) The simulated spectrum of the [{CuIIImph}(μ-O2−){CuIImph}]1+ species (g// = 2.296, g = 2.047; and A// = 80 G, A ≅ 27 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 58

20 G). (b) The [{CuIIImph}2(μ-O2−)]2+ intermediate (species 2) (black line) and the spectrum recorded after reducing species 2 with 1 equiv. of FeCp*2 (red line); 1.5 equiv (blue line); and 2 equiv. (olive line). The spectrum of the CuIIImph complex is also shown for comparison (purple line).

Structural characterization of the dicopper species in the MSN-TP-3 by Cu K-edge EXAFS Since the dicopper species 1, 2, and 3 are embedded within the nanochannels of mesoporous silica nanoparticles, Cu K-edge EXAFS is the only structural tool that can offer a glimpse of the three-dimensional structures of these catalytic species. In Figure 7 and Table 2, we compare the Cu K-edge EXAFS (both the k3χ(k) and its Fourier transform) of (a) the CuIIImph complex; (b) the CuIII(μ-O)2CuIII species 1; (c) the CuII(μO)CuII species 2; and (d) the putative CuIII(μ-O)((μ-1,2)-peroxo)CuIII species 3. The results of the best fits of various structural models to the observed data are also presented here. The very low values of fitting error Rfit (as defined in SI) in Table 2 indicate excellent fit to the structure models. With the CuIIImph complex, we use its known Xray structure 22 as the point of departure and fine tune the distances of the first-shell and second-shell elements to best fit the observed EXAFS. With the dicopper species 1, 2, and 3, we begin the EXAFS analyses with the ligand structure of CuIIImph and systematically improve the quality of the fitting (Rfit) by adjusting the ligand distances, adding additional light-elements (O, C) to the first and second coordination shells, and including Cu…Cu backscattering. When the Fourier transforms of the experimental k3χ(k) of the various copper species are compared in R space, it is evident that the structures of these species differ primarily in the second coordination shells. In all three cases, better

28 ACS Paragon Plus Environment

Page 29 of 58 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

fits are obtained with the inclusion of Cu…Cu backscattering, providing some degree of assurance that 1, 2, and 3 are dicopper cluster species. For comparision, details of the various analyses of the EXAFS data with alternative structures are given in Figure S8S10. We find that the alternative structures always give worse fittings of the EXAFS data with significantly higher values of Rfit. In any case, we use the structural models summarized in Table 2 and Figure 7 to augment the structural conclusions derived earlier from the UV-Visible spectroscopy and the EPR redox titration experiments.

Table 2. Best-fit parameters used in fitting the k3-weighted EXAFS data obtained for the various copper species a in the MSN-TP-3: (a) CuIIImph; (b) species 1; (c) species 2, and (d) species 3. (a) CuIIImph

(b) Species 1

Bond type

N

R(Å)

σ2(Å2)

Bond type

N

R(Å)

Cu−N

3

1.94(7)

0.001(0)

Cu−O

2

1.90(6)

0.003(1)

Cu−Namine

1

2.08(0)

0.001(1)

Cu−Nb

3

1.96(2)

0.005(1)

Cu−C

7

3.04(5)

0.008(2)

Cu−C

7

3.14(8)

0.003(3)

Cu−Cu

1

2.64(7)

0.002(3)

Δk (Å−1)

[1.913, 12.480]

[1.970, 12.480]

ΔR(Å)

[1.00, 3.00]

[1.00, 3.25]

cR fit

0.025 %

0.035 %

(c) Species 2

σ2(Å2)

(d) Species 3

Bond type

N

R(Å)

σ2(Å2)

Bond type

N

R(Å)

σ2(Å2)

Cu−O

1

1.89(2)

0.004(7)

Cu−O(1)

2

1.93(0)

0.004(6)

Cu−Nb

3

1.97(0)

0.003(9)

Cu−Nb

3

1.97(3)

0.002(8) 29

ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 58

Cu−C

7

3.14(2)

0.03(3)

Cu−O(2)d

1

2.92(7)

0.002(0)

Cu−Cu

1

2.84(0)

0.009(6)

Cu−C

7

3.15(0)

0.008(4)

Cu−Cu

1

3.64(0)

0.003(3)

Δk (Å−1)

[1.996, 12.480]

[1.927, 12.480]

ΔR(Å)

[1.00, 3.25]

[1.00, 3.60]

cR fit

0.040 %

0.051 %

a Errors

are estimated to be within ±0.01 Å for distances (R), and ±0.0005 Å2 for Debye–

Waller factor (σ2) bThe

number of coordinating N atoms includes the Nimidazole and Namine atoms in the

ligand. c

The goodness-of-fit factor (Rfit) defined in reference 34 and in supporting information.

dThe

Cu−O(2) corresponds to the backscattering from the O in the second coordination-

sphere associated with the (μ-1,2)-peroxo unit in species 3.

30 ACS Paragon Plus Environment

Page 31 of 58 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Figure 7. Cu K-edge EXAFS data obtained on the various copper species in the MSNTP-3. Fourier transforms of k3χ(k): (a) CuIIImph; (b) species 1; (c) species 2; and (d) species 3. Inset: the k3-weighted EXAFS data. Black circle: raw data; red line: fitting results.

The best fits to the EXAFS data for CuIIImph and the three dicopper species implicated in the catalytic cycle are detailed in Supporting Information. For sake of the present discussion, it suffices to note that the EXAFS structures for the three dicopper species are consistent with the chemistry that we are proposing here. In particular, for species 3, the EXAFS data (Figure 7d and Table 2d) show an additional O(1) in the first coordination sphere of each Cu ion, namely, 2 Cu−O with average bond distance of 1.93 Å, and a third O(2) in the second coordination sphere with a bond distance of 2.93 Å, 31 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 58

together with lengthening of the Cu···Cu distance to 3.64 Å. These results lend credence to the structure of the O2 adduct proposed above for the dicopper cluster in the putative CuIII(μ-O)((μ-1,2)-peroxo)CuIII species 3. As noted earlier, the positions of the pre-edge and the near-edge features in the X-ray absorption spectrum of this species suggest the higher oxidation state of Cu(III) for the copper ions in this O2 adduct. The cis- end-on addition of an O2 molecule across the two Cu(II) ions of species 2, namely, CuII(μ-O)CuII, is expected to withdraw sufficient electron density from the Cu(II) atoms and formally oxidize them to Cu(III) concomitant with formation of the bridging peroxide.

Discussion We have immobilized the Cu(II) complex of the tripodal tridentate histidine-like Imph ligand in the nanochannels of MSN and show that we can form the CuIII(μ-O)2CuIII complex upon reduction of the Cu(II) complexes followed by activation by O2.22 We show that the immobilized CuIII(μ-O)2CuIII complex in MSN thus formed is capable of catalyzing aerobic oxidation of the aliphatic C−H bonds in toluene to yield BnOH and further to Bz at ambient temperatures under mild conditions with multiple turnovers but without the injection of sacrifical reductants.22 In other words, the catalytic system is capable of harnessing both O atoms of the O2 molecule for hydrocarbon oxidation and is functioning like a “dioxygenase”. This result is unprecedented. The mechanism of catalytic turnover A plausible mechanism of the catalytic turnover may be written according to the following reaction scheme: (1) Activation of the mononuclear CuIImph by O2:

32 ACS Paragon Plus Environment

Page 33 of 58 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry



2 [CuIImph]1+ + O2

1

(1)

(2) Oxo-transfer to the methyl group of toluene: C6H5−CH3

+ 1



C6H5−CH2OH + 2

(2)

(3) Oxidation of the BnOH to Bz to complete the catalytic cycle: 

C6H5−CH2OH + 2

C6H5−CHO + 2 [CuIImph]1+ + H2O .

(3)

Step 3 is predicated on the assumption that the CuII(μ-O)CuII species 2 has sufficient oxidizing power to transfer the second O atom of the activating O2 molecule to one of the remaining aliphatic C−H bonds of the BnOH. However, we show in this study that this species is in fact relatively inert at room temperature so that the conversion of BnOH into Bz could not be described according to Step 3 of the reaction scheme as written. Instead, a second O2 molecule is involved in the second step of the single-turnover. Accordingly, Step 3 in the above reaction scheme needs to be amended and replaced by the following two reactions: 2 + O2



3

C6H5−CH2OH + 3

(3’a) 

C6H5−CHO +

1 + H2O

(3’b) ,

where species 3 is an O2 adduct of the CuII(μ-O)CuII species 2, namely, the putative CuIII(μ-O)((μ-1,2)-peroxo)CuIII species 3 mentioned earlier. Species 3 could be a transient kinetic species, but we show that it is a stable intermediate, and in fact, have successfully trapped the intermediate for structural analysis (vide supra). The structure of species 3 is not unprecedented: a dicopper-dioxygen species with the structure akin to species 3 has recently been reported.35 Since the dicopper catalyst reverts back to the CuIII(μ-O)2CuIII species 1 upon completion of the turnover cycle (Scheme 2), the catalytic cycle does not involve Step 1, but only Step 2 together with Step 3’a and Step 3’b above. 33 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 34 of 58

In the presence of O2, the CuIImph, once formed and converted into the CuIII(μ-O)2CuIII species 1, no longer becomes part of the catalytic cycle.

Scheme 2. Reversion of the proposed CuIII(μ-O)((μ-1,2)-peroxo)CuIII species 3 to CuIII(μO)2CuIII species 1 after transfer of an O atom to the organic substrate.

As expected, the dicopper cluster in the putative CuIII(μ-O)((μ-1,2)-peroxo)CuIII species 3 is also capable of oxidizing toluene to BnOH. This is evident from the burst in the production of BnOH (Figure 1a) after we attempt to simulate multiple turnover conditions again by purging the sample with O2 in the single-turnover experiment. Implications of the findings The results of the present study illustrate how it is possible to harness the oxidizing power of both O atoms in the CuIII(μ-O)2CuIII species 1 for hydrocarbon oxidation, or to transform the catalytic system from a “monooxygenase” to a “dioxygenase”. The activation of the CuII(μ-O)CuII species 2 by an O2 molecule in the second step of the catalytic cycle of the dicopper catalyst would not only provide additional driving force for the transfer of the remaining oxidizing equivalents from the original activating O2 to an aliphatic C−H bond of a substrate molecule, but also would lower the activation barrier for the O-atom transfer process. Without the formation of the O2 adduct, the

34 ACS Paragon Plus Environment

Page 35 of 58 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

second O-atom transfer reaction is uphill thermodynamically. The cis- end-on peroxide bridging the two CuIII ions would also facilitate the transfer of a “singlet” oxene from the copper cluster to the C−H bond of the substrate in the transition state. During this O-atom transfer, we surmise that there would be structural rearrangement of the dicopper cluster from end-on to the side-on peroxo to convert the copper cluster back to the CuIII(μO)2CuIII species 1, completing the catalytic cycle. This structural rearrangement of the peroxide could occur in concert with the O-atom transfer, but it is possible that the two steps are distinct. It is also possible that there is some scrambling of these O atoms during this process. In any case, the energy derived from O2 activation of the two CuIImph complexes can be exploited to drive the significantly less favorable transfer of the O atom from the CuII(μ-O)CuII species 2, and the catalyst remains a dicopper species without returning to two individual [CuIImph]1+complexes. These findings illustrate some principles that might be exploited for the rational design of copper catalysts toward the controlled oxidation of aliphatics based on a new strategy that no longer requires sacrificial reductants. Such a strategy would represent a departure from the mechanism used by the traditional copper monooxygenases as noted earlier. It would be more efficient in terms of atom economy, simpler and cheaper, and catalytically more efficient. However, for a dicopper system, the turnover rates are relatively slow, as we are demonstrating here for our [{CuIIIImph}2(μ-O2−)2]2+@MSNTP-3 (species 1). Moreover, our catalyst shows no activity toward the aliphatic C−H bonds in ethyl benzene, not to mention the stronger C−H bonds in methane, at least at room temperature. It is in principle possible to redesign the tricopper cluster catalyst that we have recently developed for facile conversion of methane into methanol to harness the

35 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 36 of 58

total oxidizing power of the O2 molecule for methane oxidation, either to oxidize two molecules of methane per O2 or convert methane into formaldehyde. The reaction scheme for the catalytic turnover Armed with the insights derived from the single-turnover experiments, we can now formulate the reaction scheme shown in Scheme 3 for the catalytic aerobic oxidation of the aliphatic C−H bonds in toluene to yield BnOH and further oxidation to Bz by the O2activated CuIImph@MSN-TP-3 in Scheme 1. At the outset, two CuIImph complexes sharing the same nanochannel in the MSN-TP-3 are activated by O2 to form the CuIII(μO)2CuIII species 1, which oxidizes the toluene substrate to BnOH. With the loss of two of the four oxidizing equivalents, the dicopper catalyst is converted into the relatively inert CuII(μ-O)CuII species 2. In the presence of excess O2, however, the latter is rapidly activated to give the significantly more reactive CuIII(μ-O)((μ-1,2)-peroxo)CuIII species 3 which can oxidize the BnOH to Bz, or an additional toluene substrate to BnOH. Eventually, with the buildup of BnOH, the latter is also oxidized by the CuIII(μ-O)2CuIII species 1 to yield Bz. Since BnOH is also a substrate of our catalytic system with ~100 % conversion,22 it is evident that the CuIII(μ-O)2CuIII species 1 can convert BnOH into Bz as well. Thus, the conversion of toluene into Bz mediated by the CuImph@MSN-TP-3 nanoparticles is a complex parallel reaction of the two oxidation steps (toluene to BnOH and BnOH to Bz) with the dicopper catalyst running in sequence between two states, the CuIII(μ-O)2CuIII species 1 and the CuIII(μ-O)((μ-1,2)-peroxo)CuIII species 3, with the product yields of BnOH and Bz at a given time determined by the rate constants of the two oxidation steps mediated by each of the two states of the catalyst and the concentrations of the reactants and products prevailing at that instant. With the dicopper

36 ACS Paragon Plus Environment

Page 37 of 58 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

catalyst cycling between these two “activated” states, it never reverts back to the mononuclear CuIImph complex so long as O2 is available for the catalytic turnover.

Scheme 3. The reaction scheme for the toluene oxidation reaction mediated by the CuImph@MSN-TP-3 catalytic system.

The activation of the Cu(II)-O-Cu(II) complex by O2 and the subsequent transfer of the oxo species to the C–H bond of the organic substrate in Scheme 3 is an analog of the Mars-van-Krevelen(MvK) mechanism in the O2 activation of lattice oxygens of metal 37 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 38 of 58

oxides in heterogeneous oxidation catalysis.23-25 Typically, the MvK mechanism in metaloxide heterogeneous catalysis is observed at high temperatures due to the difficulty in activating the lattice oxygens in metal oxides and the process is not well understood. The present work offers a possible analog of this mechanism by way of a molecularly confined bis-μ-oxo dicopper species. Our species 3 could also offer an example of the inter ediate that might be involved in the oxygen re-entry step in the MvK mechanism, which is poorly understood at the moment in heterogeneous catalysis. Our catalytic system gives excellent selectivity in the partial oxidation of toluene since it operates at room temperature. Furthermore, we have a complete understanding of the catalytic mechanism, which might inspire future developments of supported metal-oxo clusters that can operate under mild conditions for the controlled oxidation of hydrocarbons and other organic substrates.

Conclusions In summary, we have used a catalytic system consisting of a mesopore-immobilized dicopper complex activated and bridged by O2 for room-temperature conversion of toluene into Bz to illustrate how it is possible to harness both O atoms for hydrocarbon oxidation. Through the pore-confinement effect, various stages of the oxygen-bridged dinuclear copper complex can be stabilized and identified. Exploiting the slow turnover of the catalytic system, it is possible to follow the individual kinetic steps of the toluene oxidation reaction. These single-turnover experiments have allowed us to examine the details of the catalyst during turnover and to clarify how the “O” equivalents are being transferred from the activated dicopper cluster to the toluene substrate in the two kinetic

38 ACS Paragon Plus Environment

Page 39 of 58 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

steps. We have performed UV-Visible spectral measurements and EXAFS analysis, as well as EPR redox titrations, on the intermediates in the catalytic cycle during the turnover to link the chemistry occurring at each step of the cycle to the structure of the dicopper cluster. A key discovery in this work is that the μ-oxo bridged dicopper intermediate (species 2) can be activated via the insertion of an O2 molecule to give a unique CuIII(μ-O)((μ-1,2)-peroxo)CuIII species 3 that is active in the transfer of the second O-atom in the catalytic cycle. By way of this kinetic/mechanistic study, we have shown how it is possible to convert the CuIII(μ-O)2CuIII catalytic system from a “monoxygenase” into a “dioxygenase”.

ASSOCIATED CONTENT Supporting Information. Detailed experimental procedures and data on the measurements of N2 adsorption-desorption isotherms (BET), t-plot analysis of the N2 adsorption data, the XAS Cu K-edge spectra, computer simulations of the EPR spectra, EXAFS analyses of the various copper species immobilized in the MSN participating in the catalysis, mass spectra of all the products observed during catalytic turnover. This material is available free of charge via the Internet at http://pubs.acs.org. AUTHOR INFORMATION Corresponding Author *E-mail: [email protected] (Sunney I. Chan). *E-mail: [email protected] (Chung-Yuan Mou). Notes The authors declare no competing financial interest.

39 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 40 of 58

ACKNOWLEDGMENTS This work is supported by Academia Sinica, the National Synchroton Radiation Research Center, funds from the Nanoscience and Nanotechnology Program of Academia Sinica, and grants from the Ministry of Science and Technology of the Republic of China (National Nanotechnology Project NSC 100-2120-M-002-001 to CYM, and MOST 1012113-M-001-007-MY3 to SSFY).

REFERENCES 1. Lewis, E. A.; Tolman, W. B. Reactivity of dioxygen−copper systems. Chem. Rev. 2004, 104, 1047-1076. 2. Haack, P.; Limberg, C. Molecular CuII-O-CuII complexes: Still waters run deep. Angew. Chem. Int. Ed. 2014, 53, 4282-4293. 3. Chen, P. P.-Y.; Nagababu, P.; Yu, S. S.-F.; Chan, S. I. Development of the tricopper cluster as a catalyst for the efficient conversion of methane into MeOH. ChemCatChem. 2014, 6, 429-437. 4. Donoghue, P. J.; Gupta, A. K.; Boyce, D. W.; Cramer, C. J.; Tolman, W. B. An anionic, tetragonal copper(II) superoxide complex. J. Am. Chem. Soc. 2010, 132, 1586915871. 5. Conde, A.; Vilella, L.; Balcells, D.; Diaz-Requejo, M.M.; Lledos, A.; Perez, P.J. Introducing copper as catalyst for oxidative alkane dehydrogenation. J. Am. Chem. Soc. 2013, 135, 3887–3896.

40 ACS Paragon Plus Environment

Page 41 of 58 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

6. Arii, H.; Saito, Y.; Nagatomo, S.; Kitagawa, T.; Funahashi, Y.; Jitsukawa, K.; Masuda, H. C–H activation by Cu(III)2O2 intermediate with secondary amino ligand. Chem. Lett. 2003, 32, 156-157. 7. Würtele, C.; Sander, O.; Lutz, V.; Waitz, T.; Tuczek, F.; Schindler, S. Aliphatic C−H bond oxidation of toluene using copper peroxo complexes that are stable at room temperature. J. Am. Chem. Soc. 2009, 131, 7544-7545. 8. Vanelderen, P.; Vancauwenbergh, J.; Sels, B. F.; Schoonheydt, R. A. Coordination chemistry and reactivity of copper in zeolites. Coord. Chem. Rev. 2013, 257, 483-494. 9. Chan, S. I.; Lu, Y.-J.; Nagababu, P.; Maji, S.; Hung, M.-C.; Lee, M.; Hsu, I.-J.; Minh, P. D.; Lai, J. C.-H.; Ng, K. Y. et al. Efficient oxidation of methane to methanol by dioxygen mediated by tricopper clusters. Angew. Chem. Int. Ed. 2013, 52, 3731-3735. 10. Liu, C.-C.; Mou, C.-Y.; Yu, S. S.-F.; Chan, S. I. Heterogeneous formulation of the tricopper complex for efficient catalytic conversion of methane into methanol at ambient temperature and pressure. Energy Environ. Sci. 2016, 9, 1361-1374. 11. Liu, C.-C.; Janmanchi, D.; Wen, D.-R.; Oung, J.-N.; Mou, C.-Y.; Yu, S. S.-F.; Chan, S. I. Catalytic oxidation of light alkanes mediated at room temperature by a tricopper cluster complex immobilized in mesoporous silica nanoparticles. ACS Sustainable Chem. Eng. 2018, 6, 5431–5440. 12. Liu, C.-C.; Ramu, R.; Chan, S. I.; Mou, C.-Y.; Yu, S. S.-F. Chemistry in confined space: A strategy for selective oxidation of hydrocarbons with high catalytic efficiencies and conversion yields under ambient conditions. Catal. Sci. Technol. 2016, 6, 7623-7630. 13. Decker, H.; Schweikardt, T.; Tuczek, F. The first crystal structure of tyrosinase: All questions answered? Angew. Chem. Int. Ed. 2006, 28, 4546-4550.

41 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 42 of 58

14. Blackburn, N. J. Chemical and spectroscopic studies on dopamine-β-hydroxylase and other copper monooxygenases. In bioinorganic chemistry of copper; Karlin, K. D., Tyeklár, Z., Eds.; Chapman & Hall, New York: 1993, p 164-183. 15. Chan, S. I.; Yu, S. S.-F. Controlled oxidation of hydrocarbons by the membranebound methane monooxygenase: The case for a tricopper cluster. Acc. Chem. Res. 2008, 41, 969-979. 16. Hemsworth, G. R.; Taylor, E. J.; Kim, R. Q.; Gregory, R. C.; Lewis, S. J.; Turkenburg, J. P.; Parkin, A.; Davies, G. J.; Walton, P. H. The copper active site of CBM33 polysaccharide oxygenases. J. Am. Chem. Soc. 2013, 135, 6069-6077. 17. Dietl, N.; van der Linde, C.; Schlangen, M.; Beyer, M. K.; Schwarz, H. Diatomic [CuO]+ and its role in the spin selective hydrogen and oxygen atom transfers in the thermal activation of methane. Angew. Chem. Int. Ed. 2011, 50, 4966-4969. 18. Schwarz, H. Thermal hydrogen-atom transfer from methane: A mechanistic exercise. Chem. Phys. Lett. 2015, 629, 91-101. 19. Grundner, S.; Markovits, M. A. C.; Li, G.; Tromp, M.; Pidko, E. A.; Hensen, E. J. M.; Jentys, A.; Sanchez-Sanchez, M.; Lercher, J. A. Single-site trinuclear copper oxygen clusters in mordenite for selective conversion of methane to methanol. Nat. Commun., 2015, 6, 7546, DOI: 10.1038/ncomms8546. 20. Steiner, R. A.; Kalk, K. H.; Dijkstra, B. W. Anaerobic enzyme⋅substrate structures provide insight into the reaction mechanism of the copper-dependent quercetin 2,3dioxygenase. Proc. Natl. Acad. Sci. USA 2002, 99, 16625-16630. 21. Kaizer, J.; Pap, J. S.; Speier, G. Copper dioxygenases. In copper-oxygen chemistry; Karlin, K. D., Itoh, S., Eds.; John Wiley & Sons, Inc., Hoboken, NJ: 2011, pp 23-52. 42 ACS Paragon Plus Environment

Page 43 of 58 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

22. Liu, C.-C.; Lin, T.-S.; Chan, S. I.; Mou, C.-Y. A room temperature catalyst for toluene aliphatic C–H bond oxidation: Tripodal tridentate copper complex immobilized in mesoporous silica. J. Catal. 2015, 322, 139-151. 23. Genuino, H. C.; Dharmarathna, S.; Njagi, E. C.; Mei, M. C.; Suib, S. L. Gas-phase total oxidation of benzene, toluene, ethyl- benzene, and xylenes using shape-selective manganese oxide and copper manganese oxide catalysts. J. Phys. Chem. C 2012, 116, 12066−12078. 24. Widmann, D.; Behm, R.J. Dynamic surface composition in a Mars-van Krevelen type reaction: CO oxidation on Au/TiO2. J. Catal. 2018, 357, 263-273 25. Wu, L.-N.; Tian, Z.-Y.; Qin, W. DFT study on CO catalytic oxidation mechanism on the defective Cu2O(111) surface. J. Phys. Chem. C 2018, 122, 16733−16740. 26. Lee, C.-H; Wong, S.-T.; Lin, T.-S.; Mou, C.-Y. Characterization and biomimetic study of a hydroxo-bridged dinuclear phenanthroline cupric complex encapsulated in mesoporous silica:  Models for catechol oxidase. J. Phys. Chem. B 2005, 109, 775-784. 27. Lee, C. H; Lin, H. C.; Cheng, S. H..; Lin, T. S.; Mou, C. Y. Hydroxo-bridged dinuclear cupric complexes encapsulated in various mesoporous silicas to mimic the catalytic activity of catechol oxidases: Reactivity and selectivity study. J. Phys. Chem. C 2009, 113, 16058−16069. 28. Fang, Y.-C.; Lin, H.-C.; Lin, T.-S.; Mou, C.-Y. Bioinspired design of a Cu–Zn– imidazolate mesoporous silica catalyst system for superoxide dismutation. J. Phys. Chem. C 2011, 115, 20639-20652. 29. Mirica, L. M.; Ottenwaelder, X.; Stack, T. D. P. Structure and spectroscopy of copper−dioxygen complexes. Chem. Rev. 2004, 104, 1013−1046.

43 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 44 of 58

30. Kang, P.; Bobyr, E.; Dustman, J.; Hodgson, K. O.; Hedman, B.; Solomon, E. I.; Stack, T. D. P. Bis(μ-oxo) dicopper(III) species of the simplest peralkylated diamine: Enhanced reactivity toward exogenous substrates. Inorg. Chem. 2010, 49, 11030-11038. 31. Haack, P.; Limberg, C.; Ray, K.; Braun, B.; Kuhlmann, U.; Hildebrandt, P.; Herwig, C. Dinuclear copper complexes based on parallel β-diiminato binding sites and their reactions with O2: Evidence for a Cu−O−Cu entity. Inorg. Chem. 2011, 50, 2133-2142. 32. Pavlova, S.V.; Chen, K. H.-C.; Chan, S. I. Spectroscopic characterization of the oxotransfer reaction from a bis(μ-oxo)dicopper(III) complex to triphenylphosphine. Dalton Trans. 2004, 20, 3261-3272. 33. Miller, J. S.; Glatzhofer, D. T.; O'Hare, D. M.; Reiff, W. M.; Chakraborty, A.; Epstein, A. J. Ferromagnetic behavior in linear charge-transfer complexes. Structural and magnetic characterization of octamethylferrocene salts: [Fe(C5Me4H)2]+ [A]- (A = TCNE, TCNQ). Inorg. Chem. 1989, 28, 2930-2939. 34. Newville, M. IFEFFIT : interactive XAFS analysis and FEFF fitting. J. Synchrotron Rad. 2001, 8, 322−324. 35. Cao, R.; Saracini, C.; Ginsbach, J.W.; Kieber-Emmons, M.T.; Siegler, M.A.; Solomon, E.T.; Fukuzumi, S.; Karlin, K.D. Peroxo and superoxo moieties bound to copper ion: Electron-transfer equilibrium with a small reorganization energy. J. Am. Chem. Soc. 2016, 138, 7055−7066

44 ACS Paragon Plus Environment

Page 45 of 58 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

TOC Graphic

45 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 1. (a) The oxidation of toluene to BnOH (black line) and BnOH to Bz (red line) were performed under O2 limiting conditions and the products detected by GC-MS from 0 to 60 min (Region I: pink shaded region). After that, the oxidation reactions were con-tinued under excess O2 conditions and the products BnOH and Bz monitored by GC-MS up to 150 min (Region II: yellow shaded region). (b) Time course of the singleturnover experiment studied at room temperature at 30-min intervals up to 6 h. 272x115mm (150 x 150 DPI)

ACS Paragon Plus Environment

Page 46 of 58

Page 47 of 58 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Figure 2. Upper panels: Effects of a 40-min incubation of the sample at (a) 10 °C and (b) 50 °C, on the conversion of BnOH into Bz following the initial 1-h reaction with toluene at room temperature. Lower panels: Effects of introducing a stoichiometric aliquot of (c) dry N2 or (d) pure O2, 90 min into the hiatus period during incubation of the sample at room temperature. 206x185mm (150 x 150 DPI)

ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 3. The time course of the single-turnover experiment in which d8-toluene is oxi-dized to d8-BnOD (black data points), and further converted to the final products, d6-Bz (red data points) and D2O (blue data points), at room temperature over the 6-h incubation period. 166x126mm (150 x 150 DPI)

ACS Paragon Plus Environment

Page 48 of 58

Page 49 of 58 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Figure 4. (a) Solid state UV-Visible absorption spectra of CuIIImph@MSN-TP-3 (black line) and [{CuIIIImph}2(μ-O2−)2]2+@MSN-TP-3 (species 1, red line) at room tempera-ture, taken form our earlier work.22 For comparison, the UV-Visible spectrum of the bis(μ-oxo)dicopper(III) species [(TMED)2Cu(III)2(μ-O)2]2+ (black solid line) is shown in the inset.30 (b) The UV-Visible spectrum of [{CuIIImph}2(μ-O2−)]2+@MSN-TP-3 (spe-cies 2) well suspended in THF at room temperature. The absorption at 642 nm is assigned to the (μ-oxo)dicopper(II) species by comparison to the reference [LCuII(μ-O)CuIIL] complex, L = Me2C6H3Xanthium in the inset.31 (c) The UV-Visible spectrum of the putative [{CuIIIImph}2(μ-O2−)(O22−)]2+@MSN-TP-3 (species 3). (d) The time course of de-composition of the light-sensitive species 3 with repeated scans of the UV-Visible spec-trum: first scan at t = 0 (black); second scan 5 s later (red); and a third scan 15 s later (blue). The blue spectrum obtained after photobleaching corresponds to that of species 2, which is stable. Panels (a) and (b) are adapted with permission from ref. 22. (Copyright 2015 Elsevier B.V.) 266x203mm (150 x 150 DPI)

ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 5. X-band EPR spectra at 77 K. Upper panel: The [{CuIIIImph}2(μ-O2−)2]2+ in-termediate (species 1) (black line) and the spectrum recorded after reducing species 1 with 1 equiv. of decamethylferrocene (FeCp*2) (red line). Middle panel: Spectra ob-tained after reaction of the CuIIImph complex in O2-free THF with 1 equiv. of H2O2 (blue line), and 5 equiv. of H2O2 (green line) in the presence of Et3N. Lower panel: CuIIImph@MSN-TP-3 sample (pink line). g and A values of the various EPR spectra are obtained by computer simulations (see Figure S6). 114x159mm (150 x 150 DPI)

ACS Paragon Plus Environment

Page 50 of 58

Page 51 of 58 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Figure 6. X-band EPR spectra at 77 K: (a) The simulated spectrum of the [{CuIIImph}(μO2−){CuIImph}]1+ species (g// = 2.296, g⊥ = 2.047; and A// = 80 G, A⊥ ≅ 20 G). (b) The [{CuIIImph}2(μ-O2−)]2+ intermediate (species 2) (black line) and the spectrum recorded after reducing species 2 with 1 equiv. of FeCp*2 (red line); 1.5 equiv (blue line); and 2 equiv. (olive line). The spectrum of the CuIIImph complex is also shown for comparison (purple line). 122x220mm (150 x 150 DPI)

ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 7. Cu K-edge EXAFS data obtained on the various copper species in the MSN-TP-3. Fourier transforms of k3χ(k): (a) CuIIImph; (b) species 1; (c) species 2; and (d) species 3. Inset: the k3-weighted EXAFS data. Black circle: raw data; red line: fitting re-sults. 324x249mm (150 x 150 DPI)

ACS Paragon Plus Environment

Page 52 of 58

Page 53 of 58 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Chart 1. Chemical structures of CuIIImph and the copper-oxo species 1−3 derived from the reaction of this complex with O2 in the CuImph@MSN nanoparticles. 195x176mm (150 x 150 DPI)

ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Scheme 1. Conversion of toluene into BnOH and Bz mediated by a CuIII(μ-O)2CuIII com-plex formed by immobilization of the positively charged tripodal tridentate Cu(II) com-plex CuIIImph in negatively charged MSN (MSN-TP-3) followed by reduction using sodium ascorbate and activation by O2. 234x257mm (150 x 150 DPI)

ACS Paragon Plus Environment

Page 54 of 58

Page 55 of 58 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Scheme 2. Reversion of the proposed CuIII(μ-O)((μ-1,2)-peroxo)CuIII species 3 to CuIII(μ-O)2CuIII species 1 after transfer of an O atom to the organic substrate. 190x46mm (150 x 150 DPI)

ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Scheme 3. The reaction scheme for the toluene oxidation reaction mediated by the CuImph@MSN-TP-3 catalytic system. 338x363mm (150 x 150 DPI)

ACS Paragon Plus Environment

Page 56 of 58

Page 57 of 58 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

47x47mm (150 x 150 DPI)

ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

TOC Graphic 85x47mm (150 x 150 DPI)

ACS Paragon Plus Environment

Page 58 of 58