CHEMSUSCHEM FULL PAPERS DOI: 10.1002/cssc.201402169

Nanostructured Manganese Oxides as Highly Active Water Oxidation Catalysts: A Boost from Manganese Precursor Chemistry Prashanth W. Menezes, Arindam Indra, Patrick Littlewood, Michael Schwarze, Caren Gçbel, Reinhard Schomcker, and Matthias Driess*[a] We present a facile synthesis of bioinspired manganese oxides for chemical and photocatalytic water oxidation, starting from a reliable and versatile manganese(II) oxalate single-source precursor (SSP) accessible through an inverse micellar molecular approach. Strikingly, thermal decomposition of the latter precursor in various environments (air, nitrogen, and vacuum) led to the three different mineral phases of bixbyite (Mn2O3), hausmannite (Mn3O4), and manganosite (MnO). Initial chemical water oxidation experiments using ceric ammonium nitrate (CAN) gave the maximum catalytic activity for Mn2O3 and MnO whereas Mn3O4 had a limited activity. The substantial increase in the catalytic activity of MnO in chemical water oxidation was demonstrated by the fact that a phase transformation occurs at the surface from nanocrystalline MnO into an amor-

phous MnOx (1 < x < 2) upon treatment with CAN, which acted as an oxidizing agent. Photocatalytic water oxidation in the presence of [Ru(bpy)3]2 + (bpy = 2,2’-bipyridine) as a sensitizer and peroxodisulfate as an electron acceptor was carried out for all three manganese oxides including the newly formed amorphous MnOx. Both Mn2O3 and the amorphous MnOx exhibit tremendous enhancement in oxygen evolution during photocatalysis and are much higher in comparison to so far known bioinspired manganese oxides and calcium–manganese oxides. Also, for the first time, a new approach for the representation of activities of water oxidation catalysts has been proposed by determining the amount of accessible manganese centers.

Introduction Efficient solar water splitting by abundant, environmentally friendly, stable, and low-cost materials is of extreme interest due to the rapidly increasing demand for renewable energy technologies.[1–5] Although numerous materials have already been investigated in this direction, the low overall conversion efficiencies and photoinstability are still matters for concern.[6–10] In nature, conversion of solar energy to chemical energy occurs at the Mn4CaO5 cluster of photosystem II (PS II).[11–13] In photosynthetic water oxidation, the oxygenevolving center (OEC) catalyzes the extraction of electrons and protons from water initiated by the absorption of solar photons.[14] The overall process of water splitting is described by Equation (1): 2 H 2 O ! O 2 þ 4 H þ þ 4 e

ð1Þ

Enormous efforts have been made for decades to understand the structure and functionality of OEC as well as the mechanism of water oxidation in PS II.[15–21] Recently, the crystal [a] Dr. P. W. Menezes, Dr. A. Indra, P. Littlewood, Dr. M. Schwarze, Dr. C. Gçbel, Prof. Dr. R. Schomcker, Prof. Dr. M. Driess Department of Chemistry Technische Universitt Berlin Strasse des 17 Juni 135, Sekr. C2, 10623 Berlin (Germany) E-mail: [email protected] Supporting Information for this article is available on the WWW under http://dx.doi.org/10.1002/cssc.201402169.

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

structure of PS II was determined at an unprecedented 1.9  precision, which fits well with the results obtained from spectroscopic, magnetic, and theoretical studies.[18, 22–28] Apart from extensive structural studies, a constant effort has also been to synthesize molecular species that could act as the functional mimic of PS II.[29–34] Although some of the synthetic homogenous molecular systems can mimic the structural motifs of the natural system, they are inefficient in functional mimicking and for catalytic water oxidation.[33, 35, 36] In heterogeneous media, materials based on iridium, platinum, ruthenium, and cobalt have been proclaimed as potentially suitable candidates for efficient water oxidation, but high costs, toxicity, and low natural abundance result in very limited use.[37–42] On the other hand, special attention has been given to manganese-based materials for photocatalytic water oxidation due to their potential redox ability and the combination of its economic and environmental factors. Because the OEC of PS II also contains a stable Mn4CaO5 subcluster, the level of interest to mimic the structural and functional motifs containing manganese at the core has substantially grown. Inspired by the principles of nature, over the years biomimetic manganese-based heterogeneous systems for water oxidation have been thoroughly studied and well established. Recently, Kurz et al. reported amorphous calcium–manganese oxides with a high surface area as effective water oxidation catalysts and demonstrated that the role of calcium ions in the lattice is vital for increasing the catalytic activity in photochemical water oxidaChemSusChem 2014, 7, 2202 – 2211

2202

CHEMSUSCHEM FULL PAPERS tion.[43] In due course, Najafpour et al. synthesized several nanocrystalline and amorphous manganese and calcium manganese oxides active for chemical water oxidation.[34, 44–50] In addition, Jiao and Frei used silica-supported manganese oxide[51, 52] whereas manganese oxide molecular sieves were successfully employed by Suib et al.[53] Nepal and Das discovered that the immobilization of an active manganese-terpyridine complex in the pores of a metal–organic framework (MOF) resulted in enhanced photostability and subsequently in activity for the photochemical water oxidation.[54] Electrochemical water oxidation using electrodeposited manganese oxides has been the main theme of interest in the group of Dau for a long time.[16, 55–58] Lately, electrochemical water splitting using layered and three dimensional crosslinked manganese oxides were studied by the same group; they correlated the catalytic activity with the responsible structural motifs.[59] In a similar manner, Nakamura et al. have shown that electrochemical water oxidation could be efficiently carried out under neutral pH by stabilizing the surface-associated intermediate MnIII species.[60] Recently, Spiccia et al. modeled water-oxidizing manganese(III/IV) oxide and prepared a screen-printed material containing nanostructured manganese by means of redox precipitation that was highly active towards electrochemical water oxidation whereas manganese oxide films electrodeposited from ionic liquids were also investigated for the same purpose.[61–63] In the meantime, Jaramillo et al. described excellent electrocatalytic activities of nanostructured manganese oxides towards oxygen evolution that could even act as a bifunctional catalyst for water oxidation with activities comparable to other highly active materials.[64–66] An attempt to correlate the structures of different crystalline polymorphs of manganese oxides with the catalytic activities for water oxidation was made by Dismukes et al.[67] Thorough investigation of the bonding sites indicated that the asymmetrically occupied antibonding electron eg led to Jahn–Teller distortion in the system, which induced flexibility in the structure that promotes effective water oxidation.[68] This finding is also closely related to the extended X-Ray absorption fine structure (EXAFS) of other known manganese oxide water oxidation catalysts.[58] Recently, we reported the photochemical and the electrochemical water oxidation using mixed-valent amorphous manganese oxide (MnOx ; 1 < x < 2) synthesized by partial oxidation (’corrosion’) of inactive crystalline manganese(II) oxide (MnO) nanoparticles using cerium(IV) as an oxidizing agent. Extensive investigation of the EXAFS structure of the catalyst revealed m-oxido-bridged MnOMn bonds and the flexibility of layered structure effecting the enhancement of water oxidation activity.[69] Our recent activities in the area of biomimetic water oxidation and continuous search for highly active water-oxidizing catalysts on larger scale prompted us to explore a singlesource precursor (SSP) approach. The advantages of SSP routes are immense in comparison to the arduous classical approach that comprises a low temperature synthesis and the precise control of the composition with a maximum of dispersion of the elements on the atomic level and of the oxidation states of the metals of the heterogeneous catalysts. Moreover, the  2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemsuschem.org SSP approach facilitates rational access and shape selectivity of potential heterogeneous catalysts on the nanoscale; this itself is a significant challenge in the field of advanced functional inorganic materials.[70–76] In addition, the SSP route is particularly promising for the synthesis of materials that do not exist on the microscale and of which the activity and electronic properties can be easily tuned through shaping. This versatile approach has already been demonstrated for the preparation of amorphous transparent conducting oxides, which have shown remarkable long-term stability as well as high optoelectronic performance.[77–79] Interestingly, SSPs for the synthesis of conducting electrode surfaces for metal oxide water oxidation has already been utilized a while ago.[80, 81] Herein, we present the facile synthesis of a versatile nanocrystalline manganese oxalate precursor by means of an inverse micellar approach. Thermal degradation of the oxalate precursor at different environments led to the desired mineral phases of nanostructured bixbyite (Mn2O3), hausmannite (Mn3O4), and manganosite (MnO) oxides (Scheme 1). Relatively

Scheme 1. Synthesis of nanocrystalline manganese(II) oxalate precursor through the inverse micellar method and its facile transformation into three distinct manganese oxides under different conditions.

low degradation temperature of manganese oxalate to manganese oxides, low cost, and ease of production on a large scale makes this route very efficient and appealing for material synthesis. First of all, the catalytic activity of the as-synthesized pure phases of mineral manganese oxides for the chemical water oxidation was tested by using ceric ammonium nitrate (CAN) as an oxidizing agent. Then, photochemical water oxidation in the presence of [Ru(bpy)3]2 + (bpy = 2,2’-bipyridine) as a photosensitizer and peroxodisulfate (Na2S2O8) as a sacrificial electron acceptor was extensively studied using various buffer media. Excellent catalytic activities were achieved for all phases; these were much larger than that of any other literature-reported bioinspired manganese oxides and calcium manganese oxides. Interestingly, a highly active amorphous MnOx water-oxidizing catalyst was also produced by the corrosion of nanocrystalline MnO phase upon treatment with CAN (chemical oxidation, Scheme 2) and was then examined for photochemical water oxidation. Furthermore, for the first time, we introduce a new concept for the representation of the water oxidation activity by determining the actual amount of active catalytic centers responsible within the catalyst using temperature-programmed reduction (TPR). This method could potenChemSusChem 2014, 7, 2202 – 2211

2203

CHEMSUSCHEM FULL PAPERS

Scheme 2. Synthesis of the active amorphous MnOx water oxidation catalyst using MnO nanoparticles through the corrosion approach.

tially be useful for the future of water oxidation, for which to date only a bulk or specific-surface-normalized representation have been used without considering the amount of active centers involved. Finally, the correlations of catalytic activity versus the structural and bonding aspects have been described.

Results and Discussion The manganese oxalate dehydrate precursor was synthesized by means of an inverse micelle approach by mixing two microemulsions containing MnII and oxalate ions. All reflections obtained for the precursor by powder X-ray diffraction (PXRD; Figure S1 in the Supporting Information) could be indexed on the basis of a monoclinic cell (I2/a) reported for MnC2O4·2 H2O (JCPDS 25-544), with lattice parameters a = 12.01 , b = 5.63 , c = 9.96 , and Z = 4. No other reflections were detected, thus confirming the phase purity of the crystalline product. The crystal structure of MnC2O4·2 H2O consists of one-dimensional chains with each manganese atom being coordinated to two bidentate oxalate ligands and two water molecules as shown in Figure 1 a.[82] The morphology, particle sizes, and crystallinity

Figure 1. (a) Structure of MnC2O4·2 H2O consisting of one-dimensional chains where each manganese atom is coordinated to two bidentate oxalate and two water molecules. Atom codes for the spheres are given as follows: blue—Mn, red—O, grey—C, and dark green—H. A nanorod of the as-prepared oxalate precursor is shown as SEM (b) and TEM (c) images.

of the precursor were initially determined by SEM and TEM. Figure 1 b and c depict typical SEM and TEM micrographs of the as-prepared oxalate precursor; all showed nanorod structures with a diameter of about 100 nm and a length of several hundred nanometers. The reason for the formation of a rod 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemsuschem.org like morphology of the manganese oxalate precursor has already been discussed previously.[83, 84] It has been observed that manganese oxalate nanorods have negative surface charge and that the construction of such nanorods is aided by the templating effect of the cationic surfactant molecules [cetyltrimethylammonium bromide (CTAB), in this case], which organize themselves in a linear fashion around the MnII ions and the oxalate ligands. The chemical composition and the presence of manganese were determined by energy dispersive X-ray spectroscopy (EDX) and inductively coupled plasma atomic emission spectroscopy (ICP-AES) analyses (Figure S2 and Table S1). Elemental analysis was carried out to quantify the elements with lower atomic numbers (Table S2). The FTIR spectrum of the as-prepared oxalate precursor recorded between 2000 and 400 cm1 showed a band at 1714 cm1 attributed to the asymmetric OCO vibrations whereas a strong band at 1625 cm1 was ascribed to the HOH bending vibration. The presence of bridging oxalate ligands, proving that the four oxygen atoms are coordinated to the manganese centers, is confirmed by bands at 1359 and 1317 cm1 corresponding to CO asymmetric and symmetric vibrations. The assignment of the other bands at 791, 514, 492, and 410 cm1 are shown in Figure S3 (and in Table S3). Also, a comparatively large surface area of 23 m2 g1 was determined using Brunauer–Emmett–Teller (BET) experiments. A thermogravimetric analysis (TGA) was performed to study the degradation of MnC2O4·2 H2O in a nitrogen atmosphere. The thermal degradation of the behavior of the manganese(II) oxalate precursor is shown in Figure S4. Two distinct mass loss steps were observed between 80 and 600 8C. The first mass loss occurred between 80 and 150 8C, which corresponds to the release of structural water, thereby converting the hydrous into an anhydrous phase. The calculated weight loss (20.10 %) of the two water molecules is consistent with the experimental TGA values (20.03 %), confirming the chemical formula MnC2O4·2 H2O. The second mass loss was observed between 280 and 450 8C, which is cause by the transformation of the anhydrous oxalate phase into a manganese oxide. This weight loss (39.9 %) perfectly matches with a calculated loss of one carbon monoxide and one carbon dioxide molecule (39.7 %). After the TGA investigations, the product was analyzed using PXRD by which the phase could be identified as Mn3O4 (hausmannite, tetragonal I41/amd with unit cell parameters a = 5.76 , c = 9.46 , and Z = 4; JCPDS, 24-731) as shown in Figure S5. Hence, all results acquired by using PXRD, EDX, chemical analysis, FTIR, and TGA indicate that MnC2O4·2 H2O is a suitable and convenient precursor to produce Mn3O4. Remarkably, the TGA-monitored degradation of MnC2O4·2 H2O in other environments led to the formation of different manganese oxides. Indeed, Mn2O3 (bixbyite) was formed when MnC2O4·2 H2O was treated in the presence of synthetic air at 400 8C. The reflections in the PXRD pattern could be readily indexed to a pure isometric system crystallizing in space group Ia-3 with lattice parameters a = 9.43  and Z = 16 (JCPDS 41-1442). In contrast, a cubic MnO phase results from thermolysis of the precursor in vacuum at 400 8C (manganosite, Fm-3m with lattice constants ChemSusChem 2014, 7, 2202 – 2211

2204

CHEMSUSCHEM FULL PAPERS

www.chemsuschem.org

of a = 4.446  and Z = 4; JCPDS 71-1177). The PXRD patterns and the corresponding Miller indices for strong reflections of all three mineral phases are shown in Figure 2. Mixtures of Mn3O4 and MnO were formed when attempts were made to produce pure Mn3O4 at optimized temperatures (below 500 8C) under nitrogen (Figure S6).

Figure 2. PXRD patterns and Miller indices (hkl) of the intense diffraction peaks showing the crystal family of planes of monophasic Mn2O3 (top, JCPDS 41-1442), Mn3O4 (middle, JCPDS 24-731), and MnO (bottom, JCPDS 71-1177).

The crystal structure of Mn2O3 comprises of Mn3 + ions that are octahedrally coordinated whereas the oxygen atoms have four manganese neighbors (Figure 3 a).[85] This structure can be seen as close-packed arrangement of manganese and oxygen atoms filling 3=4 of the tetrahedral interstitials in a symmetrical fashion. Mn3O4 crystallizes in the normal spinel AB2O4 (MnIIMn2IIIO4) structure in which the manganese atoms are distributed among the tetrahedral (A) and octahedral (B) sites (Figure 3 b).[86] The octahedral sites are close to each other sharing edges, but tetrahedral sites share only corners with the octahedral sites. MnO has the rock salt structure with face-centered cubic arrangement where both MnII and O2 ions are in octahedral coordination (Figure 3 c).[87] All three mineral phases of manganese oxides described above are closely packed by lattice oxygen and differ only in the arrangement of octahedral and tetrahedral sites for the various oxidation states of Mn. One of the most common features of these oxides is the organization of edge-sharing manganese octahedra. The average MnO and MnMn distances within the octahedra and to the adjacent manganese centers are 2.033 and 3.111  for Mn2O3, 2.026 and 3.042  for Mn3O4, and 2.222 and 3.142  for MnO. The SEM and TEM micrographs of the mineral manganese oxide phases are shown in Figure 4. Highly porous (77 ) nanorods of 100 nm diameter particles and with large surface area (49 m2 g1) were obtained for the Mn2O3 phase whereas self-assembled particles with diameters 150–200 nm forming nanorods and with much lower surface areas (14 m2 g1) were acquired for Mn3O4. Most importantly, cubic-type nanoparticles (surface area 11 m2 g1) with 200 nm particle size were formed  2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Figure 3. Crystal structures of mineral phases of (a) Mn2O3, (b) Mn3O4, and (c) MnO. All three manganese oxides are closely packed by lattice oxygen and differ only in the arrangement of octahedral and tetrahedral sites with various oxidation states of manganese (see text). The green, blue, and red atoms represent MnIII, MnII, and O2, respectively.

ChemSusChem 2014, 7, 2202 – 2211

2205

CHEMSUSCHEM FULL PAPERS

www.chemsuschem.org

Figure 5. Dissolved-oxygen profiles measured by a Clark electrode in deoxygenated aqueous solutions containing manganese oxides catalysts and 0.5 m CAN as an oxidizing agent.

Figure 4. SEM (left column) and TEM (right column) images of manganese oxides prepared by the thermal degradation of the respective oxalate precursor: (a, b) Mn2O3, (c, d) Mn3O4, and (e, f) MnO.

for the MnO phase. The presence of the manganese in each case was confirmed by EDX analysis (Figure S7). The MnO bonds are more sensitive to infrared irradiation; therefore, FTIR spectra in the range from 400 to 1200 cm1 were measured Figure S8). In the far-IR range, Mn2O3 exhibited three shoulders at 664, 570, and 475 cm1 for Mn2O3. Two broad bands at 604 and 470 cm1 were seen for Mn3O4 whereas two shoulders at 669 and 470 cm1 were observed for MnO. Chemical water oxidation experiments were performed in deoxygenated aqueous solutions containing the different manganese oxides and CAN. The standard chemical potential of Ce4 + [E0 = 1.7 V vs. normal hydrogen electrode (NHE)] is adequate to oxidize water to oxygen in acidic conditions. Figure 5 shows the dissolved oxygen content in the reactant solution monitored for 15 min at room temperature; the rate of the oxygen evolution was determined from the slope of the linear fitting for the first 60 s. Maximum oxygen evolution was observed when Mn2O3 and MnO were used as catalysts (0.41 and 0.16 mmolO2 molMn1 s1). With Mn3O4 only a maximum rate of 0.02 mmolO2 molMn1 s1 was achieved. The rate of oxygen evolution of Mn2O3 was twice that of MnO, 20 times higher than that presented by Mn3O4, 12 times higher than that of any other Mn2O3 known to date and comparable with the active biomimetic CaMn2O4 systems (see Table S4).[43, 46, 47] The corresponding specific-surface-area-normalized plot is shown in Figure S9. From literature, it is evident that effective water oxidation likely requires the presence of manganese sites with a mean  2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

manganese oxidation state between 3.5 and 3.8. To our surprise, the MnO phase, which does not fulfill this requirement, showed very high catalytic activity towards chemical water oxidation. This curiosity led us to investigate the reason for its activity. Careful analysis of the concentration profiles of oxygen formation with respect to time disclosed that the reaction was slower at the initial stages (appearance of a time lag for the first 5 s) before oxygen evolution. At this stage, it was assumed that the apparent induction time was due to the transformation of the catalytically inactive MnO to higher-oxidized active MnOx catalyst. To authenticate this prediction, we started investigating the active MnOx catalyst extensively to gather the structural information and to correlate the catalytic activity with inactive MnO. As a first clue, the PXRD pattern revealed that the active MnOx phase was largely amorphous after 15 min of CAN treatment by losing most of its crystallinity as shown in Figure 6 a. This was further supported by microscopic studies during which the cubic nanoparticles of the MnO were converted into amorphous particles by treatment with CAN (Figure 6 b and c). In addition, BET analysis showed an unusual increase in the surface area from 11 m2 g1 for the inactive MnO to more than twice that value (25 m2 g1) for the active MnOx catalyst. Simultaneously, ICP-AES experiments were carried out to confirm the composition and the average oxidation state of manganese in the active amorphous MnOx catalyst. At this stage, the chemical formula appeared to be MnO1.25 with + 2.5 as the mean oxidation state of manganese (Table S5). Therefore, it is not the MnO that is catalytically active but the transformed amorphous and partially oxidized MnOx (1 < x < 2) responsible for chemical water oxidation. We subjected the asprepared active MnOx catalyst to further chemical water oxidation, and an increased oxygen evolution was observed, indicating that the transformed MnOx is indeed an active catalyst (Figure S10). Similarly, time-dependent studies were also performed to check whether the active MnOx was formed instantly when it comes into contact with CAN solution or whether a step-wise corrosion is favored. This study confirmed that the ChemSusChem 2014, 7, 2202 – 2211

2206

CHEMSUSCHEM FULL PAPERS

www.chemsuschem.org at higher concentration (1.8 m) and after prolonged treatments (ranging from a few hours to 4 weeks).[50, 88] The photocatalytic cycles of water oxidation for all mineral phases of the Mn2O3, Mn3O4, and MnO catalysts, including the newly formed amorphous MnOx, were investigated in the presence of a two-electron acceptor (Na2S2O8) and a photosensitizer ([Ru(bpy)3]2 + ) in aqueous buffer solution (Scheme 3). In the

Scheme 3. Photocatalytic cycle of water oxidation using the Na2S2O8 and [Ru(bpy)3]2 + system.

Figure 6. (a) PXRD patterns of MnO (top) and the transformed active amorphous MnOx phase (bottom) after treatment with 0.5 m CAN. (b) and (c) are the respective SEM and TEM images showing the amorphous nature of the active MnOx phase.

inactive MnO needs at least 15 min to transform itself from a crystalline to an amorphous phase, which could be seen in PXRD (Figure S11) analysis. Furthermore, an enhancement in the oxygen evolution profiles could also be seen after treating samples with CAN for different time intervals (Figure S12). As mentioned earlier in the introduction, this finding is similar to that of a study that we published recently in which an inactive MnO precursor synthesized solvothermally could be transformed into an active amorphous MnOx catalyst for effective water oxidation by using CAN as an oxidizing agent (corrosion agent).[69] X-ray absorption spectroscopy (XAS) of this MnOx catalyst showed an increase in the mean oxidation state of manganese accompanied by a considerable change in the structural rearrangement, which induces a series of defects and disorders at the di-m-oxido-bridges; the latter makes the catalyst analogous to layered calcium–manganese-oxide systems but for calcium sites being occupied by MnII or MnIII ions. Therefore, we firmly believe that a similar structural reorganization must also have occurred for MnO presented in this work and led to the formation of active layered amorphous MnOx. This could be further supported by the identical oxygen evolution profiles obtained in both cases. No such amorphous layered structure formation was observed after treatment of crystalline Mn2O3 with CAN as shown by PXRD, SEM, and TEM studies (Figure S13). The higher stability of Mn2O3 against CAN could be explained by the lower concentration of CAN (0.5 m) used and the shorter treatment time (15 min). This corroborates with the results illustrated by Najafpour et al. that it is indeed possible to convert some manganese oxides and calcium–manganese oxides structurally into their amorphous layered form by treatment with CAN but only  2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Ru(bpy)3]2 + –S2O82system, [Ru(bpy)3]2 + absorbs visible light and generates electron–hole pairs on the surface of the catalyst. The electrons produced were expelled by the sacrificial electron acceptor S2O82 by further oxidizing [Ru(bpy)3]2 + to [Ru(bpy)3]3 + and reducing S2O82 to SO42 and a sulfate radical (SO4C). Thus, the formed radical can subsequently further oxidize [Ru(bpy)3]2 + to give [Ru(bpy)3]3 + . Then, the [Ru(bpy)3]3 + molecule donates its holes to the catalyst and is reverted back to [Ru(bpy)3]2 + ; at this site, water molecules are oxidized to form oxygen molecules at the surface. The dissolved-oxygen content can then by analyzed by using a Clark oxygen electrode system. A series of experiments were conducted to analyze the catalytic activity of the manganese oxides in various buffer solutions. Additionally, optimization was achieved by changing the concentration of the sensitizer and the sacrificial electron acceptor. Buffer solutions including carbonate (pH 4.7), acetate (pH 5.8), phosphate (pH 7), and borate (pH 9) were probed to ensure the stability of the catalyst. Oxygen evolution was only observed when all constituents (i.e., the catalyst, the light source, the sacrificial electron acceptor, and the catalysts) were involved in the reaction in buffered solutions; the only exception was the borate buffer, which showed significant oxygen formation even without the presence of the catalyst. Maximum catalytic activity and best stability were found when using the phosphate buffer in neutral conditions; therefore, this buffer was used for the present work. Also, in all cases, the catalytic activities leveled after about 2–3 min of photolysis due to the use of low concentrations and the consumption of the sacrificial electron acceptor. Addition of sacrificial acceptor to the used solution and readjusting the pH value to 7 resulted in continued water oxidation at the similar rate. This confirms that the stability and prolonged activity of the studied catalysts are similar to previously reported silica-supported manganese oxides.[51, 52] All photochemical reactions were performed at least thrice to ensure good reproducibility and reliability. A comparison of catalytic activities of all manganese oxides for photochemical water oxidation is shown in Figure 7. In acChemSusChem 2014, 7, 2202 – 2211

2207

CHEMSUSCHEM FULL PAPERS

Figure 7. Dissolved-oxygen profiles measured by a Clark electrode in phosphate buffer (pH 7) using Na2S2O8 as two-electron acceptor and [Ru(bpy)3]2 + as photosensitizer.

cordance with chemical water oxidation using CAN, Mn2O3 showed the highest oxygen generation, which was almost twice as active as the amorphous MnOx and six times better than the Mn3O4 phase. As expected, the MnO phase showed negligible oxygen evolution because of its unsuitably low oxidation state for water oxidation. The rate of oxygen formation for the respective catalysts increased in the series Mn2O3 (1.90 mmolO2 molMn1 s1) > active MnOx (0.80 mmolO2 molMn1 s1) > Mn3O4 (0.30 mmolO2 molMn1 s1) > MnO 1 1 (0.02 mmolO2 molMn s ). The values obtained here are remarkably larger than those reported for all other manganese oxides[50, 89, 90] as well as for the active biomimetic CaMn2O4·H2O mineral phase (see Table S4 for comparison).[43] In fact, our results underline again that biomimicking heterogeneous manganese oxides can be highly efficient even in the absence of calcium or any other alkali and alkaline earth metals. Because Mn2O3 and the active MnOx were confirmed to act as effective water oxidation catalysts, the evolution of oxygen in the photochemical system collected in the head space of the reaction mixture was quantitatively measured by means of GC analyses. A maximum rate of 1.23  103 mmolO2 molMn1 s1 was achieved for Mn2O3, whereas 0.99  103 mmolO2 molMn1 s1 could be detected for the active MnOx (Table S6). The trend observed here also confirms the reliability of the measurement of dissolved-oxygen concentration using a Clark electrode. H218O labeling studies on Mn2O3 and active MnOx were also performed using GC equipped with a mass spectrometer, which confirmed that only water is the source of the evolved oxygen (Figure S14). Due to the differences in the BET surface area of the materials, the catalytic activities were normalized to the geometric surface area (Figure S15) to allow for direct comparison with the other highly active crystalline polymorphs of manganese oxides. The catalytic activity of Mn2O3 (0.041 mmolO2 molMn1 m2 s1) was higher than that of amorphous MnOx (0.031 mmolO2 molMn1 m2 s1) and twice that of  2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemsuschem.org Mn3O4 (0.020 mmolO2 molMn1 m2 s1); the latter values were found to be higher than those reported so far for other manganese oxide phases and calcium–manganese oxides. A direct representation of the catalytic activity has been a point of discussion for a long time because of varying conditions (such as concentration of the photosensitizer, nature of the sacrificial agent, catalyst amount, use of various buffers, reaction volumes, intensity of the light source, surface areas, etc.) involved in the chemical or photochemical water oxidation. However, many groups have been applying the concept of representing catalytic activities without considering the geometrical surface assuming that the complete bulk is involved in the catalysis whereas some have successfully normalized the surface against the bulk materials assuming that only the metal centers at the surface are active. The disadvantage in both cases is that it is very difficult to calculate the actual amount of bulk or surface responsible for the catalytic water oxidation. This encouraged us to provide a unique concept for which we attempted to practically measure the amount of active material responsible for water oxidation by reducing the active catalysts without considering the bulk or the surface activity. The percentage of reduction of the active material to its highly reducible form gives us the amount of actual catalytically active metal centers accessible for catalysis. To determine the amount of active material within the catalyst, temperature-programmed reduction (TPR) using hydrogen was carried out from room temperature to 700 8C as shown in Figure S16. A high heating rate was chosen to keep the experiment time low to minimize reaction with any internal lattice oxygen, which would be inaccessible at lower temperature; this causes an overall peak shift to higher temperatures. The TPR profile of the Mn2O3 phase clearly shows two structural reduction peaks at 324 and 420 8C.[91, 92] The former peak can be assigned to the reduction of Mn2O3 to Mn3O4 and the latter is correlated to the transformation of Mn3O4 to MnO (Figure S16 a). The TPR of the Mn3O4 phase has a relatively narrow single reduction peak at 560 8C due to its high crystallinity, which corresponds to the direct structural change of Mn3O4 to MnO (Figure S16 b).[93] The active amorphous MnOx phase shows a strong peak at 482 8C due to the conversion of the MnOx to the MnO phase (Figure S16 c) with a shoulder at lower temperature, which is probably due to the higher surface area of MnOx ; such an observation has already been reported earlier.[91] In each case, the presence of MnO after complete reduction at 700 8C was confirmed by PXRD (Figure S17). As only the manganese sites within the catalyst were reducible, the stoichiometrically adjusted total hydrogen consumption was divided by the amount of manganese atoms in the sample to give the proportion of material accessible for catalytic reaction. It is not possible to predict whether all accessible manganese sites are catalytically active towards water oxidation, but this approach provides a fast, indiscriminate alternative to adsorption techniques and was used to normalize the catalytic activity per gram bulk to a “per mass of active catalyst” basis (Table S7). The proportions of Mn2O3, the active amorphous MnOx, and Mn3O4 accessible for reactions were 80.4 %, 38.4 %, and 82.2 %, respectively. ChemSusChem 2014, 7, 2202 – 2211

2208

CHEMSUSCHEM FULL PAPERS Normalized activities of all synthesized materials are depicted in Figure 8. Considering the amount of active material involved in catalysis, the values were further increased 1.25 times for Mn2O3 (2.37 mmolO2 molMn1 s1), 2.6 times for the

Figure 8. Normalized plot of dissolved-oxygen profiles measured by a Clark electrode on the basis of TPR studies. The percentage of reduction of the active materials to its highly reducible form gives the amount of actual catalytically active metal centers accessible for water oxidation.

active MnOx (2.08 mmolO2 molMn1 s1), and 1.2 times for Mn3O4 (0.36 mmolO2 molMn1 s1). A similar trend was observed when the corresponding specific surface was normalized against the activity except for the active amorphous MnOx, which showed an enhancement in the catalytic activity due to the lower reduction capacity resulting in a lower amount of active material available for water oxidation (Figure S18). With reference to a previous reports[67] we also tried to correlate the catalytic activity with crystallographic aspects to understand the structural and bonding aspects. It has already been stated that materials such as Mn2O3 (bixbyite) and Mn3O4 (hausmannite) are more active than other manganese polymorphs and have longer MnO bonds at edge-sharing octahedra containing manganese(III) species (Jahn–Teller distorted) with electronically degenerated eg1 configuration. This leads to a structural flexibility on the surface of manganese oxides and increases the activity for water oxidation. This proposed hypothesis matches quite well with the presented manganese oxides herein. Moreover, the structural investigations (from XAS) of amorphous MnOx have also shown longer MnO distances that could be attributed to Jahn–Teller-elongated MnIII O bonds. On the contrary, due to the lack of structural flexibility, MnO does not exhibit any activity for catalytic water oxidation..

Conclusions We reported the facile preparation of a manganese(II) oxalate single-source precursor (SSP) and its low temperature degradation to various manganese oxides on a large scale, representing an inexpensive and reliable route for the synthesis of  2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemsuschem.org highly active manganese oxide catalysts for water oxidation. Three different nanostructured mineral phases of manganese oxides (Mn2O3, Mn3O4, and MnO) were produced through degradation of the SSP in various environments. The as-synthesized materials were characterized using state-of-the-art techniques and tested for chemical and photochemical water oxidation. Chemical water oxidation measurements were performed using ceric ammonium nitrate (CAN). Interestingly, the conversion of catalytically inactive nanostructured MnO into active amorphous MnOx was attained by using CAN as an oxidizing agent (“corrosion agent”). Photocatalytic water oxidation activities of as-prepared catalysts were investigated using the [Ru(bpy)3]2 + –S2O82 system. The results obtained from the photocatalytic experiments showed that both Mn2O3 and amorphous MnOx represent exceptionally active catalysts. The latter catalytic activities were substantially larger than those hitherto reported for manganese oxides and calcium–manganese-oxide mineral phases. This further confirms that through biomimicking using earth-abundant heterogeneous manganese oxides highly efficient water oxidation catalysts can be obtained even in the absence of calcium or any other alkali and alkaline earth metals. Similarly, a new approach was proposed for the representation of water oxidation activity on the basis of temperature-programmed reduction (TPR) with hydrogen to determine the amount of accessible manganese sites within the catalyst. The latter method could potentially be useful in describing the different catalytic activities of related systems in the future. Further studies to test the electrocatalytic activities of the presented manganese oxides for water oxidation are currently in progress. Also, we are aiming to study the conversion of inactive MnO to highly active layered amorphous MnOx using various corrosion agents (dichromate, peroxides, etc.) and to investigate the influence of the structural changes on chemical, photochemical, and electrochemical activities.

Experimental Section Synthesis of manganese oxalate dihydrate Microemulsion containing CTAB (2.0 g) as surfactant, 1-hexanol (20 mL) as co-surfactant and hexane (35 mL) as lipophilic phase was prepared and mixed separately with an aqueous solution of 0.1 m manganese acetate. Similarly, the second microemulsion with the same constituents but with 0.1 m ammonium oxalate was prepared. Both microemulsions were mixed slowly and stirred overnight. The resulting white precipitate was then centrifuged and washed with a 1:1 mixture of chloroform and methanol (200 mL) and subsequently dried at 60 8C for 12 h.

Syntheses of manganese oxides The manganese oxalate precursor was subjected to thermal treatment at 400–550 8C in various environments to obtain the desired mineral phases of manganese oxides. Monophasic Mn2O3 was obtained when the oxalate precursor was heated in dry synthetic air (20 % O2, 80 % N2) at 400 8C for 8 h (2 8C min1) whereas heating in nitrogen under the same conditions led to the formation of a mixture of Mn3O4 and MnO. Prolonged annealing of the precursor under nitrogen at 550 8C for 24 h produced Mn3O4 in pure form. ChemSusChem 2014, 7, 2202 – 2211

2209

CHEMSUSCHEM FULL PAPERS For the synthesis of MnO, the precursor was filled in a quartz ampoule and sealed under vacuum. The quartz tube was then placed in a furnace and heated slowly (2 8C min1) for 8 h to 400 8C and cooled down to ambient temperature.

Chemical oxygen evolution from water Chemical oxygen evolution experiments were carried out using aqueous solutions containing CAN as sacrificial one-electron acceptor (oxidizing agent). The oxygen evolution was measured using a Clark-type oxygen electrode system (Strathkelvin, 1302 oxygen electrode and 782 oxygen meters). Prior to the experiments, the electrode was calibrated in an air-saturated water solution and in zero-oxygen (sodium sulfite in water) solution. In a typical reaction, the catalyst (1 mg) was placed in the reactor, which was then degassed by purging with nitrogen for 30 min. An anaerobic solution (2 mL) of 0.5 m CAN was then injected into the reactor to initiate the chemical water oxidation, and the liberation of oxygen was monitored using the Clark electrode under stirring. In each case, the maximum rate of oxygen evolution was calculated using the total amount of oxygen yield after the first 60 s of reaction.

www.chemsuschem.org Acknowledgements Financial support by the BMBF (L2H project) and the DFG (Cluster of Excellence UniCat) is gratefully acknowledged. Keywords: catalysis · manganese oxide · photochemical · photosystem II · water oxidation

[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]

Photocatalytic oxygen evolution from water Photochemical water oxidation experiments were performed in a 10 mL quartz reactor fitted with a water jacket, which maintained the temperature at 20  0.5 8C. Na2S2O8 was chosen as sacrificial electron acceptor and [Ru(bpy)3]2 + as a photosensitizer. Various buffer solutions including carbonate (Na2SiF6–NaHCO3 ; pH 4.7), acetate (CH3COONa–CH3COOH; pH 5.8), phosphate (Na2HPO4–KH2PO4 ; pH 7), and borate (H3BO3–Na2B4O7; pH 9) were tested; the phosphate buffer in neutral conditions was found to be most suitable, not only due to its stability but also because of its higher rates of oxygen evolution. A typical run was carried out using [Ru(bpy)3]Cl2·6 H2O (1.5 mg), Na2S2O8 (3.5 mg), and phosphate buffer (1 mL) along with the catalyst (0.5 mg). The reactants were purged with nitrogen for almost 1 h to remove all dissolved oxygen from the aqueous solution. The quartz reactor was then illuminated using a continuous output xenon lamp with the power of 300 W. Visible light was prepared by placing a long pass filter of 395 nm between the quartz reactor and the light source. Dissolved oxygen formed during the reaction was measured by using the Clark electrode, and the maximum rate of oxygen evolution was calculated in the same way to that of chemical water oxidation. A separate set of control experiments was carried out in similar conditions as above to quantify the oxygen gas obtained in photochemical water oxidation by placing the catalyst (30 mg), [Ru(bpy)3]Cl2·6 H2O (25 mg), Na2S2O8 (100 mg) and phosphate buffer solution (6 mL) in a quartz reactor with a headspace of 15 and 12 mL. The reactor was then irradiated for 2 h using a xenon lamp (300 W); the resulting oxygen molecules in the head space were injected thrice into GC and quantitatively analyzed. A labelled-oxygen study was also performed using similar conditions as for oxygen detection by GC. The catalyst (30 mg), [Ru(bpy)3]Cl2·6 H2O (25 mg), Na2S2O8 (100 mg), phosphate buffer (5.6 mL) and H2O18 (0.4 mL) solution was placed in a quartz reactor with a headspace of 15 mL. After an irradiation period of 2 h, the resulting values of m/z = 36 (O18O18), 34 (O16O18), and /z 32 (O16O16) were systematically monitored using a GC equipped with mass spectrometer.  2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

[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]

A. Singh, L. Spiccia, Coord. Chem. Rev. 2013, 257, 2607 – 2622. K. Maeda, K. Domen, J. Phys. Chem. Lett. 2010, 1, 2655 – 2661. M. W. Kanan, D. G. Nocera, Science 2008, 321, 1072 – 1075. K. Maeda, J. Photochem. Photobiol. C 2011, 12, 237 – 268. M. Grzelczak, J. S. Zhang, J. Pfrommer, J. Hartmann, M. Driess, M. Antonietti, X. C. Wang, ACS Catal. 2013, 3, 383 – 388. F. E. Osterloh, Chem. Soc. Rev. 2013, 42, 2294 – 2320. A. Kudo, Y. Miseki, Chem. Soc. Rev. 2009, 38, 253 – 278. A. Kudo, H. Kato, I. Tsuji, Chem. Lett. 2004, 33, 1534 – 1539. F. E. Osterloh, Chem. Mater. 2008, 20, 35 – 54. K. Takanabe, K. Domen, ChemCatChem 2012, 4, 1485 – 1497. H. Dau, I. Zaharieva, Acc. Chem. Res. 2009, 42, 1861 – 1870. B. Kok, B. Forbush, M. McGloin, Photochem. Photobiol. 1970, 11, 457 – 475. G. W. Brudvig, W. F. Beck, J. C. Depaula, Annu. Rev. Biophys. Biophys. Chem. 1989, 18, 25 – 46. J. F. Allen, W. Martin, Nature 2007, 445, 610 – 612. H. Dau, M. Haumann, Coord. Chem. Rev. 2008, 252, 273 – 295. H. Dau, I. Zaharieva, M. Haumann, Curr. Opin. Chem. Biol. 2012, 16, 3 – 10. J. P. McEvoy, G. W. Brudvig, Chem. Rev. 2006, 106, 4455 – 4483. E. M. Sproviero, J. A. Gascon, J. P. McEvoy, G. W. Brudvig, V. S. Batista, J. Am. Chem. Soc. 2008, 130, 6728 – 6730. A. Sartorel, M. Carraro, F. M. Toma, M. Prato, M. Bonchio, Energy Environ. Sci. 2012, 5, 5592 – 5603. W. Rttinger, G. C. Dismukes, Chem. Rev. 1997, 97, 1 – 24. J. Yano, J. Kern, K. Sauer, M. J. Latimer, Y. Pushkar, J. Biesiadka, B. Loll, W. Saenger, J. Messinger, A. Zouni, V. K. Yachandra, Science 2006, 314, 821 – 825. S. Schinzel, J. Schraut, A. V. Arbuznikov, P. E. M. Siegbahn, M. Kaupp, Chem. Eur. J. 2010, 16, 10424 – 10438. V. Krewald, F. Neese, D. A. Pantazis, J. Am. Chem. Soc. 2013, 135, 5726 – 5739. P. E. M. Siegbahn, Acc. Chem. Res. 2009, 42, 1871 – 1880. L. V. Kulik, B. Epel, W. Lubitz, J. Messinger, J. Am. Chem. Soc. 2007, 129, 13421 – 13435. H. Dau, A. Grundmeier, P. Loja, M. Haumann, Philos. Trans. R. Soc. London Ser. B 2008, 363, 1237 – 1243. N. Cox, D. A. Pantazis, F. Neese, W. Lubitz, Acc. Chem. Res. 2013, 46, 1588 – 1596. V. Krewald, F. Neese, D. A. Pantazis, J. Biol. Inorg. Chem. 2014, 19, S499 – S499. S. Mukhopadhyay, S. K. Mandal, S. Bhaduri, W. H. Armstrong, Chem. Rev. 2004, 104, 3981 – 4026. K. Wieghardt, Angew. Chem. Int. Ed. Engl. 1989, 28, 1153 – 1172; Angew. Chem. 1989, 101, 1179 – 1198. C. S. Mullins, V. L. Pecoraro, Coord. Chem. Rev. 2008, 252, 416 – 443. G. Christou, Acc. Chem. Res. 1989, 22, 328 – 335. G. C. Dismukes, R. Brimblecombe, G. A. N. Felton, R. S. Pryadun, J. E. Sheats, L. Spiccia, G. F. Swiegers, Acc. Chem. Res. 2009, 42, 1935 – 1943. M. M. Najafpour, F. Rahimi, E.-M. Aro, C.-H. Lee, S. I. Allakhverdiev, J. R. Soc. Interface 2012, 9, 2383 – 2395. R. Brimblecombe, A. Koo, G. C. Dismukes, G. F. Swiegers, L. Spiccia, J. Am. Chem. Soc. 2010, 132, 2892 – 2894. R. Brimblecombe, G. F. Swiegers, G. C. Dismukes, L. Spiccia, Angew. Chem. Int. Ed. 2008, 47, 7335 – 7338; Angew. Chem. 2008, 120, 7445 – 7448. J. Barber, P. D. Tran, J. R. Soc. Interface 2013, 10, 20120984 . M. Hara, T. E. Mallouk, Chem. Commun. 2000, 1903 – 1904.

ChemSusChem 2014, 7, 2202 – 2211

2210

CHEMSUSCHEM FULL PAPERS [39] M. Hara, C. C. Waraksa, J. T. Lean, B. A. Lewis, T. E. Mallouk, J. Phys. Chem. A 2000, 104, 5275 – 5280. [40] M. Murakami, D. Hong, T. Suenobu, S. Yamaguchi, T. Ogura, S. Fukuzumi, J. Am. Chem. Soc. 2011, 133, 11605 – 11613. [41] R. Nakamura, H. Frei, J. Am. Chem. Soc. 2006, 128, 10668 – 10669. [42] D. G. Nocera, Acc. Chem. Res. 2012, 45, 767 – 776. [43] M. M. Najafpour, T. Ehrenberg, M. Wiechen, P. Kurz, Angew. Chem. Int. Ed. 2010, 49, 2233 – 2237; Angew. Chem. 2010, 122, 2281 – 2285. [44] M. M. Najafpour, J. Photochem. Photobiol. C 2011, 104, 111 – 117. [45] M. M. Najafpour, Dalton Trans. 2011, 40, 3793 – 3795. [46] M. M. Najafpour, K. C. Leonard, F.-R. F. Fan, M. A. Tabrizi, A. J. Bard, C. K. King’ondu, S. L. Suib, B. Haghighi, S. I. Allakhverdiev, Dalton Trans. 2013, 42, 5085 – 5091. [47] M. M. Najafpour, A. N. Moghaddam, Y. Sakha, Dalton Trans. 2013, 42, 11012 – 11020. [48] M. M. Najafpour, B. Pashaei, S. Nayeri, Dalton Trans. 2012, 41, 4799 – 4805. [49] M. M. Najafpour, F. Rahimi, M. Amini, S. Nayeri, M. Bagherzadeh, Dalton Trans. 2012, 41, 11026 – 11031. [50] M. M. Najafpour, D. J. Sedigh, B. Pashaei, S. Nayeri, New J. Chem. 2013, 37, 2448 – 2459. [51] F. Jiao, H. Frei, Energy Environ. Sci. 2010, 3, 1018 – 1027. [52] F. Jiao, H. Frei, Chem. Commun. 2010, 46, 2920 – 2922. [53] A. Iyer, J. Del-Pilar, C. K. King’ondu, E. Kissel, H. F. Garces, H. Huang, A. M. El-Sawy, P. K. Dutta, S. L. Suib, J. Phys. Chem. C 2012, 116, 6474 – 6483. [54] B. Nepal, S. Das, Angew. Chem. Int. Ed. 2013, 52, 7224 – 7227; Angew. Chem. 2013, 125, 7365 – 7368. [55] H. Dau, C. Limberg, T. Reier, M. Risch, S. Roggan, P. Strasser, ChemCatChem 2010, 2, 724 – 761. [56] M. Wiechen, I. Zaharieva, H. Dau, P. Kurz, Chem. Sci. 2012, 3, 2330 – 2339. [57] I. Zaharieva, P. Chernev, M. Risch, K. Klingan, M. Kohlhoff, A. Fischer, H. Dau, Energy Environ. Sci. 2012, 5, 7081 – 7089. [58] I. Zaharieva, M. M. Najafpour, M. Wiechen, M. Haumann, P. Kurz, H. Dau, Energy Environ. Sci. 2011, 4, 2400 – 2408. [59] A. Bergmann, I. Zaharieva, H. Dau, P. Strasser, Energy Environ. Sci. 2013, 6, 2745 – 2755. [60] T. Takashima, K. Hashimoto, R. Nakamura, J. Am. Chem. Soc. 2012, 134, 18153 – 18156. [61] M. Fekete, R. K. Hocking, L. Y. Chang, C. Italiano, A. F. Patti, L. Spiccia, Energy Environ. Sci. 2013, 6, 2222 – 2232. [62] F. Zhou, A. Izgorodin, R. K. Hocking, L. Spiccia, D. R. MacFarlane, Adv. Energy Mater. 2012, 2, 1013 – 1021. [63] R. K. Hocking, R. Brimblecombe, L. Y. Chang, A. Singh, M. H. Cheah, C. Glover, W. H. Casey, L. Spiccia, Nat. Chem. 2011, 3, 461 – 466. [64] Y. Gorlin, T. F. Jaramillo, J. Am. Chem. Soc. 2010, 132, 13612 – 13614. [65] Y. Gorlin, T. F. Jaramillo, ECS Trans. 2011, 41, 1701 – 1707. [66] Y. Gorlin, B. Lassalle-Kaiser, J. D. Benck, S. Gul, S. M. Webb, V. K. Yachandra, J. Yano, T. F. Jaramillo, J. Am. Chem. Soc. 2013, 135, 8525 – 8534. [67] D. M. Robinson, Y. B. Go, M. Mui, G. Gardner, Z. Zhang, D. Mastrogiovanni, E. Garfunkel, J. Li, M. Greenblatt, G. C. Dismukes, J. Am. Chem. Soc. 2013, 135, 3494 – 3501.

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemsuschem.org [68] U. Maitra, B. S. Naidu, A. Govindaraj, C. N. R. Rao, Proc. Natl. Acad. Sci. USA 2013, 110, 11704 – 11707. [69] A. Indra, P. W. Menezes, I. Zaharieva, E. Baktash, J. Pfrommer, M. Schwarze, H. Dau, M. Driess, Angew. Chem. Int. Ed. 2013, 52, 13206 – 13210; Angew. Chem. 2013, 125, 13447 – 13451. [70] S. Heitz, Y. Aksu, C. Merschjann, M. Driess, Chem. Mater. 2010, 22, 1376 – 1385. [71] S. Heitz, J.-D. Epping, Y. Aksu, M. Driess, Chem. Mater. 2010, 22, 4563 – 4571. [72] S. Jana, Y. Aksu, M. Driess, Dalton Trans. 2009, 1516 – 1521. [73] S. Polarz, A. V. Orlov, M. W. E. van den Berg, M. Driess, Angew. Chem. Int. Ed. 2005, 44, 7892 – 7896; Angew. Chem. 2005, 117, 8104 – 8109. [74] A. Roy, S. Polarz, S. Rabe, B. Rellinghaus, H. Zahres, F. E. Kruis, M. Driess, Chem. Eur. J. 2004, 10, 1565 – 1575. [75] K. Samedov, Y. Aksu, M. Driess, Chem. Mater. 2012, 24, 2078 – 2090. [76] K. Samedov, Y. Aksu, M. Driess, ChemPlusChem 2012, 77, 663 – 674. [77] Y. Aksu, M. Driess, Angew. Chem. Int. Ed. 2009, 48, 7778 – 7782; Angew. Chem. 2009, 121, 7918 – 7922. [78] Y. Aksu, S. Frasca, U. Wollenberger, M. Driess, A. Thomas, Chem. Mater. 2011, 23, 1798 – 1804. [79] M. Tsaroucha, Y. Aksu, J. D. Epping, M. Driess, ChemPlusChem 2013, 78, 62 – 69. [80] Y.-H. Lai, C.-Y. Lin, Y. Lv, T. C. King, A. Steiner, N. M. Muresan, L. Gan, D. S. Wright, E. Reisner, Chem. Commun. 2013, 49, 4331 – 4333. [81] J. D. Blakemore, M. W. Mara, M. N. Kushner-Lenhoff, N. D. Schley, S. J. Konezny, I. Rivalta, C. F. A. Negre, R. C. Snoeberger, O. Kokhan, J. Huang, A. Stickrath, T. Lan Anh, M. L. Parr, L. X. Chen, D. M. Tiede, V. S. Batista, R. H. Crabtree, G. W. Brudvig, Inorg. Chem. 2013, 52, 1860 – 1871. [82] J. Soleimannejad, H. Aghabozorg, S. Hooshmand, M. Ghadermazi, J. A. Gharamaleki, Acta Crystallogr. Sect. E 2007, 63, m2389 – m1294. [83] A. K. Ganguli, A. Ganguly, S. Vaidya, Chem. Soc. Rev. 2010, 39, 474 – 485. [84] T. Ahmad, K. V. Ramanujachary, S. E. Lofland, A. K. Ganguli, J. Mater. Chem. 2004, 14, 3406 – 3410. [85] W. Hase, Phys. Status Solidi 1963, 3, K446 – K449. [86] D. Jarosch, Miner. Petrol. 1987, 37, 15 – 23. [87] A. Bystrçm, A. M. Bystrom, Acta Crystallogr. 1950, 3, 146 – 154. [88] M. M. Najafpour, M. Zareh-Kompany, A. Zahraei, D. J. Sedigh, H. Jaccard, M. Khoshkam, D. R. Britt, W. H. Casey, Dalton Trans. 2013, 42, 14603 – 14611. [89] Y. Okuno, O. Yonemitsu, Y. Chiba, Chem. Lett. 1983, 815 – 818. [90] V. B. R. Boppana, F. Jiao, Chem. Commun. 2011, 47, 8973 – 8975. [91] E. R. Stobbe, B. A. de Boer, J. W. Geus, Catal. Today 1999, 47, 161 – 167. [92] K. Mette, A. Bergmann, J.-P. Tessonnier, M. Haevecker, L. Yao, T. Ressler, R. Schloegl, P. Strasser, M. Behrens, ChemCatChem 2012, 4, 851 – 862. [93] D. Dçbber, D. Kiessling, W. Schmitz, G. Wendt, Appl. Catal. B 2004, 52, 135 – 143.

Received: March 12, 2014 Published online on July 8, 2014

ChemSusChem 2014, 7, 2202 – 2211

2211