Unveiling Non-Covalent Interactions in Novel Cooperative Photoredox Systems for Efficient Alkene Oxidation in Water

A new cooperative photoredox catalytic system, [RuII(trpy)(bpy)(H2O)][3,3′-Co(8,9,12-Cl3-1,2-C2B9H8)2]2, 5, has been synthesized and fully characterized for the first time. In this system, the photoredox catalyst [3,3′-Co(8,9,12-Cl3-1,2-C2B9H8)2]− [Cl6-1]−, a metallacarborane, and the oxidation catalyst [RuII(trpy)(bpy)(H2O)]2+, 2 are linked by non-covalent interactions. This compound, along with the one previously synthesized by us, [RuII(trpy)(bpy)(H2O)][(3,3′-Co(1,2-C2B9H11)2]2, 4, are the only examples of cooperative molecular photocatalysts in which the catalyst and photosensitizer are not linked by covalent bonds. Both cooperative systems have proven to be efficient photocatalysts for the oxidation of alkenes in water through Proton Coupled Electron Transfer processes (PCETs). Using 0.05 mol% of catalyst 4, total conversion values were achieved after 15 min with moderate selectivity for the corresponding epoxides, which decreases with reaction time, along with the TON values. However, with 0.005 mol% of catalyst, the conversion values are lower, but the selectivity and TON values are higher. This occurs simultaneously with an increase in the amount of the corresponding diol for most of the substrates studied. Photocatalyst 4 acts as a photocatalyst in both the epoxidation of alkenes and their hydroxylation in aqueous medium. The hybrid system 5 shows generally higher conversion values at low loads compared to those obtained with 4 for most of the substrates studied. However, the selectivity values for the corresponding epoxides are lower even after 15 min of reaction. This is likely due to the enhanced oxidizing capacity of CoIV in catalyst 5, resulting from the presence of more electron-withdrawing substituents on the metallacarborane platform.


Introduction
Epoxides are an important and versatile part of intermediate and basic components that are used to obtain more elaborate chemical products in both organic synthesis and in the industrial production of fine and bulk chemical products [1,2].For example, epoxides can be transformed into a variety of functionalized products as diols, aminoalcohols, allylic alcohols, ketones, etc. [3][4][5][6].The chemical oxidation of alkenes to obtain the corresponding epoxides has been carried out using different metal catalysts such as Mn-salen catalyst and NaClO or variations of these [7,8], Ti(iPrO) 4 [9,10], and Ru-aqua complexes [11,12], among others.Most of these processes have been studied in organic media.
The development of photocatalytic methods and systems for organic transformations is challenging given the significant environmental and economic impact that this entails [13].Cooperative photoredox catalysis represents a significant advancement in this field.It involves two catalysts: one that is photochemically active and another that is redox active, even in the absence of light [14].The redox-active metal complex serves as a catalyst to to activate either a water molecule or an organic substrate through a proton-coupled electron transfer (PCET) mechanism.Currently, photochemical systems studied for the oxidation of substrates involve a photocatalyst [15][16][17] or a photocatalyst combined with a transition metal catalyst based on polypyridyl compound [18][19][20][21].However, a few examples of photochemical epoxidation of alkenes have been carried out with Earth-abundant metals [22][23][24].Manganese and cobalt compounds all containing the carboranylcarboxylate ligand, [1-CH3-2-CO2-1,2-closo-C2B10H10] − , have been tested in the epoxidation of aliphatic and aromatic alkenes using peracetic acid as the oxidant [25].The catalytic results highlight the role of the carboranylcarboxylate ligand in the selectivity of the processes, and it was found that coordination of the carboranylcarboxylate ligand to the metal ions is key to their catalytic performance.
On the other hand, it is well known that some boron clusters interact with light [26][27][28] and that they have been studied as catalysts in different processes.The most wellknown metallacarborane is the anionic cobaltabis(dicarbollide) [29] that can be synthesized in high yield by a fast and environmentally friendly solid-state reaction [30].The sandwich compound Na [3,3′-Co(1,2-C2B9H11)2], Na [1], found by Fuentes et al. [31], has many possibilities to form hydrogen bonds (e.g., Cc-H•••O or Cc-H•••X (X = halogen) as well as dihydrogen bonding Cc-H•••H-B and B-H•••H-N (Cc stands for the cluster carbon atoms)) [32][33][34].Na [1] is highly stable in water, but even at low concentrations it forms aggregates (vesicles and micelles) [35][36][37] and can form ion-pair complexes through hydrogen and dihydrogen interactions.These supramolecular interactions appear to be significant in electron transfer processes and therefore in the performance and efficiency of photocatalytic systems.
Recently, we have shown that Na [1] and its dichloro (Na[Cl2-1]) and hexachloro (Na[Cl6-1]), acting both as catalyst and photosensitizer, have been highly efficient in the photooxidation of alcohol [38] and Na [1] in the oxidation of alkenes [24] in water, through single electron transfer processes (SET).We have also supported the metallabis(dicarbollide) catalyst on silica-coated magnetite nanoparticles [39].This system has proven to be a green and sustainable heterogeneous catalytic system, highly efficient, and easily reusable for the photooxidation of alcohols in water.Finally, we have synthesized a cooperative system where the cobaltabis(dicarbollide) was linked by non-covalent interactions to a redox active oxidation catalyst that represented an efficient photocatalyst for the photooxidation of alcohols in water through PCET [40].This specific cooperative photocatalytic system has not yet been explored for the photooxidation of alkenes in an aqueous environment.
Scheme 1. Synthetic strategy for preparation of 4 and 5. (a) Synthesis of 2 obtained following method described in [42].(b) Synthesis of Ag [1] following method described in [40] and Ag[Cl6-1].(c) Synthesis of compounds 4 and 5 by reaction between 2 and Ag [1] and Ag[Cl6-1], respectively.The IR of 4 and 5 show vibrations around 2900 and 3030 cm −1 that can be assigned to υC-H stretching modes for the aromatic rings of the cationic moieties and to the υC-H stretching of the Cc-H bonds in the different rotamers of the dicarbollide anions, respectively.A band can be seen over 3500 cm −1 which corresponds to the υO-H stretching of the water ligands coordinated to ruthenium atoms in both compounds.We have also observed significant vibrations around 2530 cm −1 that correspond to υB-H stretching mode for the B-H bonds in the compounds.Unlike the spectrum of compound 4, the spectra of compound 5 and Ag[Cl6-1] presents bands at 900 cm −1 corresponding to υB-Cl stretching of the B-Cl bonds, present in the hexachloro cobaltabis(dicarbollide) anions.The 1D 1 H NMR spectra of the complex Ag[Cl6-1] and compound 5, along with the COSY spectrum of 5, were recorded in acetone-d6.These are displayed in Figures 2, S3 and S4, respectively.The 1 H NMR spectrum of 5 shows two set of signals, similar to complex 4 [40], (i) one in the aliphatic region that can be assigned to the Cc-H of the cobaltacarborane anions, which resonances appear around δ = 4.34 ppm, showing a shift of 0.36 ppm to low field with respect to complex 4; and (ii) the second one, in the aromatic region corresponding to the protons of the bypyridine and terpyridine ligands of the ruthenium cation.
The 1 H{ 11 B} NMR spectrum exhibits the H-B resonances over a wide range of chemical-shift in the region from δ = 1.70 to 3.5 ppm (Figure S4b).The UV-vis spectrum of Ag[Cl 6 -1] in CH 2 Cl 2 shows one strong absorption band at 327 nm and three others with less intensity at 280, 390, and 469 nm, (Figure S5) in accordance with [43][44][45].Figure 3 shows the UV-visible spectra of 5 and 2. The former exhibits ligand-based π-π* bands of the cationic part below 350 nm that are partially eclipsed by strong absorptions corresponding to the hexachloro cobaltabis(dicarbollide) anionic moiety.Above 350 nm, the spectra show less-intense bands that correspond to dπ(Ru)-π*(L) MLCT transitions [46].It is worth mentioning the shift to higher energy absorptions observed for the MLCT bands of 5 regarding those of complex 2. This seems to indicate that the substitution of the chloride counterions by the metallacarboranes provokes the stabilization of the dπ (Ru) donor orbitals.The UV-vis spectrum of Ag[Cl6-1] in CH2Cl2 shows one strong absorption band nm and three others with less intensity at 280, 390, and 469 nm, (Figure S5) in accord with [43][44][45].Figure 3 shows the UV-visible spectra of 5 and 2. The former exhibits lig based π-π* bands of the cationic part below 350 nm that are partially eclipsed by s absorptions corresponding to the hexachloro cobaltabis(dicarbollide) anionic m Above 350 nm, the spectra show less-intense bands that correspond to dπ(Ru)-MLCT transitions [46].It is worth mentioning the shift to higher energy absorption served for the MLCT bands of 5 regarding those of complex 2. This seems to indicat the substitution of the chloride counterions by the metallacarboranes provokes the lization of the dπ (Ru) donor orbitals.The UV-vis spectrum of Ag[Cl6-1] in CH2Cl2 shows one strong absorption band at 327 nm and three others with less intensity at 280, 390, and 469 nm, (Figure S5) in accordance with [43][44][45].Figure 3 shows the UV-visible spectra of 5 and 2. The former exhibits ligandbased π-π* bands of the cationic part below 350 nm that are partially eclipsed by strong absorptions corresponding to the hexachloro cobaltabis(dicarbollide) anionic moiety.Above 350 nm, the spectra show less-intense bands that correspond to dπ(Ru)-π*(L) MLCT transitions [46].It is worth mentioning the shift to higher energy absorptions observed for the MLCT bands of 5 regarding those of complex 2. This seems to indicate that the substitution of the chloride counterions by the metallacarboranes provokes the stabilization of the dπ (Ru) donor orbitals.  TBAH) shows one reversible one-electron-redox process at E 1/2 = −0.89V versus Ag/AgCl as reference electrode, which can be assigned to Co III /Co II (see Figure S6).Unlike the derivative Ag [1], the wave corresponding to Co IV /Co III is not observed, probably because it is outside the working range of the solvent.
The CV of complex 5 in CH 3 CN + 0.1 M TBAH exhibits different redox processes due to the ruthenium and cobalt ions.Two quasi-reversible monoelectronic Co III Ru III /Co III Ru II and Co III Ru IV /Co III Ru III redox waves at E pa1 = 0.82 V; E pc1 = 0.74 V and E pa2 = 1.17 V; E pc2 = 0.93 V vs. Ag/AgCl (E pa1 = 0.38 V; E pc1 = 0.31 V and E pa2 = 0.73 V; E pc2 = 0.49 vs. Fc + /Fc) respectively and one quasireversible monoelectronic Co III Ru II /Co II Ru II redox wave at E pa3 = −0.76V; E pc3 = −0.97V vs. Ag/AgCl, (E pa3 = −1.19V; E pc3 = −1.40V vs. Fc + /Fc) (Figure S6).The redox wave corresponding to the Co IV Ru IV /Co III Ru IV is not observed.The two successive one-electron oxidation waves that correspond to Co III Ru III /Co III Ru II and Co III Ru IV /Co III Ru III redox couples can be assigned to two PCETs as it was observed in complex 4 [40].These values have been assigned to the Ru III /Ru II and Ru IV /Ru III couples of the catalytic unit based in the values presented by the mononuclear [Ru II (terpy)(bpy)(H 2 O)] 2+ compound, [42,47] and in accordance with the values observed in complex 4.This last compound, unlike 5, shows one more oxidation wave attributed to Co IV Ru IV /Co III Ru IV redox couple a E 1/2 = 1.38 vs. Ag/AgCl (0.95 V vs. Fc + /Fc).In 5, this potential would likely be outside the working range due to the greater oxidizing character of the anion [Cl 6 -1] − compared to [1] − .As has been observed with 4 in water [42], it is expected that in both compounds 4 and 5 the photogenerated Co IV could oxidize [Ru III -OH] 2+ to [Ru IV =O] 2+ , since the potential for Co IV /Co III is high enough compared to the Ru IV /Ru III couple.Then, it is expected that for photocatalytic oxidation of alkenes, both [1] − and [Cl 6 -1] − could act as good photosensitizers.

Photocatalytic Alkene Oxidations
The photocatalysts 4 and 5 are [Ru][Co] 2 ion pairs based on Co, an abundant transition metal, whereas the redox parts are Ru-OH 2 complexes.The first photocatalytic alkene oxidation experiments were all performed using 4 as the photocatalyst by exposing the reaction quartz vials to UV irradiation (2.2 W, λ ~300 nm) at room temperature and atmospheric pressure using styrene as substrate and very relevantly using water as a solvent.The samples were made of 5 mL of water pH = 7 (K 2 CO 3 solution is used to adjust the pH) with a mixture of 4 (0.01 mM), styrene (20 mM), and Na 2 S 2 O 8 (26 mM) as oxidizing agent, (catalyst load of 0.05 mol%).Then, different reaction times were tested to assess the conversion as a function of time.The reaction products were extracted with dichloromethane six times, dried with Na 2 SO 4 and quantified by means of GC-MS analysis.As can be observed in Figure S7, after 30 min of reaction, even after 15 min, all the styrene was converted.Blank experiments after 30 min of reaction, in the absence of catalyst, light, or oxidizing agent, showed that no significant conversion of styrene occurred.Figure 4 displays (a) the yields of different oxidation products obtained at different times and (b) the molar fractions (X M ) obtained in function of the reaction time.As we can see, after 5 min the amount of epoxide obtained is high (67%) and after 15 min the amount of diol increases, although the amount of epoxide is still greater (35% diol vs. 57% epoxide).A slight decrease in epoxide is observed after 30 min, (34% diol vs. 51% epoxide), but in any case, it continues to be the majority product above diol and other byproducts such as benzaldehyde and benzoic acid.We observed a slight decrease in yield and selectivity for epoxide formation with longer reaction times.Therefore, we selected a 15-min reaction time for the photoredox oxidation of various alkenes.However, we also examined the catalyst's performance after a 30-min reaction in some instances.
Table 1 displays the performance of the photocatalyst 4 after 5, 15, and 30 min of reaction using 0.05 mol% and 0.005 mol% of catalyst.In general, when 0.05 mol% of catalyst is used, it can be observed that total conversion values were achieved after 15 min of reaction with, in many cases, total conversion after 30 min.However, moderate selectivity in the corresponding epoxide is obtained, which decreases when the reaction time increase, along with the TON values.However, with 0.005 mol% of catalyst the conversion values after 15 and 30 min are lower but the selectivity and TON values are higher.This happens simultaneously with an increase in the amount of the corresponding diol for most of the substrates studied (see Table S1).At this point we can assert that the selectivity towards the epoxide increases with lower loads of catalyst 4.
in the corresponding epoxide is obtained, which decreases when the reaction time increase, along with the TON values.However, with 0.005 mol% of catalyst the conversion values after 15 and 30 min are lower but the selectivity and TON values are higher.This happens simultaneously with an increase in the amount of the corresponding diol for most of the substrates studied (see Table S1).At this point we can assert that the selectivity towards the epoxide increases with lower loads of catalyst 4. The photoepoxidation of trans-β-methylstyrene (entry 2), trans-stilbene (entry 3) led to the formation of the corresponding trans-isomer without isomerization after 15 and 30 min; however, in the case of cis-β-methylstyrene, the stereoselectivity towards the formation of the cis-epoxide isomer decreases after 30 min of reaction.The oxidation of aliphatic alkenes as the cyclic cyclooctene (entry 5) together with linear 1-octene (entry 6) has also been tested.In both cases, a good performance is shown.The photoepoxidation of trans-β-methylstyrene (entry 2), trans-stilbene (entry 3) led to the formation of the corresponding trans-isomer without isomerization after 15 and 30 min; however, in the case of cis-β-methylstyrene, the stereoselectivity towards the formation of the cis-epoxide isomer decreases after 30 min of reaction.The oxidation of aliphatic alkenes as the cyclic cyclooctene (entry 5) together with linear 1-octene (entry 6) has also been tested.In both cases, a good performance is shown.
As we have commented previously, the selectivity towards the epoxide in water decreased with the reaction time whereas the amount of diol increases as well as, in some cases, other overoxidation products.In an aqueous medium, the epoxide ring undergoes ring opening.However, we aimed to determine whether this process is driven by water or by our catalyst.Different experiments were then performed by exposing the sealed reaction to UV irradiation for 15 and 30 min, operating with 0.05 mol% of catalyst, under the same conditions used before and using three different epoxides: 1,2-epoxyoctane, styrene oxide, and trans-stilbene oxide.The results shown in Table S2 indicate the formation of the corresponding diols in water after 15 and 30 min if the photocatalyst is present.When the catalyst is absent, lower or nonexistent yields of diols have been observed.In the case of styrene, only benzaldehyde of benzoic acid has been detected after 30 min.Thus, we can conclude that 4 acts as a photocatalyst in both processes, epoxidation, and hydroxylation in aqueous media.It is worth mentioning that ring opening of epoxides is a promising process to produce 1,2-diols, an important functional group to produce pharmaceuticals, surfactants, or their intermediates [48].
We have also studied the behavior of 5 as photocatalyst in the alkene oxidation in water using low catalyst load of 0.005 mol%.The results are presented in Tables 2 and S3, including those previously obtained for compound 4, to facilitate a comparative analysis.In general, we can observe moderate-to-high conversion values after 30 min of reaction in most cases.It is worth noting that while the conversion values of catalyst 5 are generally higher than those obtained with 4, in most of the substrates studied, the selectivity values in the corresponding epoxides are lower even after 15 min of reaction.This is probably due to the enhanced oxidizing capacity of Co IV in catalyst 5, as a consequence of the presence of more electron-withdrawing substituents on the cosane platform.At this point, we can assert that in most cases and under the studied conditions, photocatalyst 4 shows higher selectivity values towards the epoxidation of alkenes that photocatalyst 5.
As we have commented previously, the selectivity towards the epoxide in water decreased with the reaction time whereas the amount of diol increases as well as, in some cases, other overoxidation products.In an aqueous medium, the epoxide ring undergoes ring opening.However, we aimed to determine whether this process is driven by water or by our catalyst.Different experiments were then performed by exposing the sealed reaction to UV irradiation for 15 and 30 min, operating with 0.05 mol% of catalyst, under the same conditions used before and using three different epoxides: 1,2-epoxyoctane, styrene oxide, and trans-stilbene oxide.The results shown in Table S2 indicate the formation of the corresponding diols in water after 15 and 30 min if the photocatalyst is present.When the catalyst is absent, lower or nonexistent yields of diols have been observed.In the case of styrene, only benzaldehyde of benzoic acid has been detected after 30 min.Thus, we can conclude that 4 acts as a photocatalyst in both processes, epoxidation, and hydroxylation in aqueous media.It is worth mentioning that ring opening of epoxides is a promising process to produce 1,2-diols, an important functional group to produce pharmaceuticals, surfactants, or their intermediates [48].
We have also studied the behavior of 5 as photocatalyst in the alkene oxidation in water using low catalyst load of 0.005 mol%.The results are presented in Tables 2 and S3, including those previously obtained for compound 4, to facilitate a comparative analysis.In general, we can observe moderate-to-high conversion values after 30 min of reaction in most cases.It is worth noting that while the conversion values of catalyst 5 are generally higher than those obtained with 4, in most of the substrates studied, the selectivity values in the corresponding epoxides are lower even after 15 min of reaction.This is probably due to the enhanced oxidizing capacity of Co IV in catalyst 5, as a consequence of the presence of more electron-withdrawing substituents on the cosane platform.At this point, we can assert that in most cases and under the studied conditions, photocatalyst 4 shows higher selectivity values towards the epoxidation of alkenes that photocatalyst 5.
As we have commented previously, the selectivity towards the epoxide in water decreased with the reaction time whereas the amount of diol increases as well as, in some cases, other overoxidation products.In an aqueous medium, the epoxide ring undergoes ring opening.However, we aimed to determine whether this process is driven by water or by our catalyst.Different experiments were then performed by exposing the sealed reaction to UV irradiation for 15 and 30 min, operating with 0.05 mol% of catalyst, under the same conditions used before and using three different epoxides: 1,2-epoxyoctane, styrene oxide, and trans-stilbene oxide.The results shown in Table S2 indicate the formation of the corresponding diols in water after 15 and 30 min if the photocatalyst is present.When the catalyst is absent, lower or nonexistent yields of diols have been observed.In the case of styrene, only benzaldehyde of benzoic acid has been detected after 30 min.Thus, we can conclude that 4 acts as a photocatalyst in both processes, epoxidation, and hydroxylation in aqueous media.It is worth mentioning that ring opening of epoxides is a promising process to produce 1,2-diols, an important functional group to produce pharmaceuticals, surfactants, or their intermediates [48].
We have also studied the behavior of 5 as photocatalyst in the alkene oxidation in water using low catalyst load of 0.005 mol%.The results are presented in Tables 2 and S3, including those previously obtained for compound 4, to facilitate a comparative analysis.In general, we can observe moderate-to-high conversion values after 30 min of reaction in most cases.It is worth noting that while the conversion values of catalyst 5 are generally higher than those obtained with 4, in most of the substrates studied, the selectivity values in the corresponding epoxides are lower even after 15 min of reaction.This is probably due to the enhanced oxidizing capacity of Co IV in catalyst 5, as a consequence of the presence of more electron-withdrawing substituents on the cosane platform.At this point, we can assert that in most cases and under the studied conditions, photocatalyst 4 shows higher selectivity values towards the epoxidation of alkenes that photocatalyst 5.
As we have commented previously, the selectivity towards the epoxide in water decreased with the reaction time whereas the amount of diol increases as well as, in some cases, other overoxidation products.In an aqueous medium, the epoxide ring undergoes ring opening.However, we aimed to determine whether this process is driven by water or by our catalyst.Different experiments were then performed by exposing the sealed reaction to UV irradiation for 15 and 30 min, operating with 0.05 mol% of catalyst, under the same conditions used before and using three different epoxides: 1,2-epoxyoctane, styrene oxide, and trans-stilbene oxide.The results shown in Table S2 indicate the formation of the corresponding diols in water after 15 and 30 min if the photocatalyst is present.When the catalyst is absent, lower or nonexistent yields of diols have been observed.In the case of styrene, only benzaldehyde of benzoic acid has been detected after 30 min.Thus, we can conclude that 4 acts as a photocatalyst in both processes, epoxidation, and hydroxylation in aqueous media.It is worth mentioning that ring opening of epoxides is a promising process to produce 1,2-diols, an important functional group to produce pharmaceuticals, surfactants, or their intermediates [48].
We have also studied the behavior of 5 as photocatalyst in the alkene oxidation in water using low catalyst load of 0.005 mol%.The results are presented in Tables 2 and S3, including those previously obtained for compound 4, to facilitate a comparative analysis.In general, we can observe moderate-to-high conversion values after 30 min of reaction in most cases.It is worth noting that while the conversion values of catalyst 5 are generally higher than those obtained with 4, in most of the substrates studied, the selectivity values in the corresponding epoxides are lower even after 15 min of reaction.This is probably due to the enhanced oxidizing capacity of Co IV in catalyst 5, as a consequence of the presence of more electron-withdrawing substituents on the cosane platform.At this point, we can assert that in most cases and under the studied conditions, photocatalyst 4 shows higher selectivity values towards the epoxidation of alkenes that photocatalyst 5.
As we have commented previously, the selectivity towards the epoxide in water decreased with the reaction time whereas the amount of diol increases as well as, in some cases, other overoxidation products.In an aqueous medium, the epoxide ring undergoes ring opening.However, we aimed to determine whether this process is driven by water or by our catalyst.Different experiments were then performed by exposing the sealed reaction to UV irradiation for 15 and 30 min, operating with 0.05 mol% of catalyst, under the same conditions used before and using three different epoxides: 1,2-epoxyoctane, styrene oxide, and trans-stilbene oxide.The results shown in Table S2 indicate the formation of the corresponding diols in water after 15 and 30 min if the photocatalyst is present.When the catalyst is absent, lower or nonexistent yields of diols have been observed.In the case of styrene, only benzaldehyde of benzoic acid has been detected after 30 min.Thus, we can conclude that 4 acts as a photocatalyst in both processes, epoxidation, and hydroxylation in aqueous media.It is worth mentioning that ring opening of epoxides is a promising process to produce 1,2-diols, an important functional group to produce pharmaceuticals, surfactants, or their intermediates [48].
As we have commented previously, the selectivity towards the epoxide in water decreased with the reaction time whereas the amount of diol increases as well as, in some cases, other overoxidation products.In an aqueous medium, the epoxide ring undergoes ring opening.However, we aimed to determine whether this process is driven by water or by our catalyst.Different experiments were then performed by exposing the sealed reaction to UV irradiation for 15 and 30 min, operating with 0.05 mol% of catalyst, under the same conditions used before and using three different epoxides: 1,2-epoxyoctane, styrene oxide, and trans-stilbene oxide.The results shown in Table S2 indicate the formation of the corresponding diols in water after 15 and 30 min if the photocatalyst is present.When the catalyst is absent, lower or nonexistent yields of diols have been observed.In the case of styrene, only benzaldehyde of benzoic acid has been detected after 30 min.Thus, we can conclude that 4 acts as a photocatalyst in both processes, epoxidation, and hydroxylation in aqueous media.It is worth mentioning that ring opening of epoxides is a promising process to produce 1,2-diols, an important functional group to produce pharmaceuticals, surfactants, or their intermediates [48].
We have also studied the behavior of 5 as photocatalyst in the alkene oxidation in water using low catalyst load of 0.005 mol%.The results are presented in Tables 2 and S3, including those previously obtained for compound 4, to facilitate a comparative analysis.In general, we can observe moderate-to-high conversion values after 30 min of reaction in most cases.It is worth noting that while the conversion values of catalyst 5 are generally higher than those obtained with 4, in most of the substrates studied, the selectivity values in the corresponding epoxides are lower even after 15 min of reaction.This is probably due to the enhanced oxidizing capacity of Co IV in catalyst 5, as a consequence of the presence of more electron-withdrawing substituents on the cosane platform.At this point, we can assert that in most cases and under the studied conditions, photocatalyst 4 shows higher selectivity values towards the epoxidation of alkenes that photocatalyst 5.
As we have commented previously, the selectivity towards the epoxide in water decreased with the reaction time whereas the amount of diol increases as well as, in some cases, other overoxidation products.In an aqueous medium, the epoxide ring undergoes ring opening.However, we aimed to determine whether this process is driven by water or by our catalyst.Different experiments were then performed by exposing the sealed reaction to UV irradiation for 15 and 30 min, operating with 0.05 mol% of catalyst, under the same conditions used before and using three different epoxides: 1,2-epoxyoctane, styrene oxide, and trans-stilbene oxide.The results shown in Table S2 indicate the formation of the corresponding diols in water after 15 and 30 min if the photocatalyst is present.When the catalyst is absent, lower or nonexistent yields of diols have been observed.In the case of styrene, only benzaldehyde of benzoic acid has been detected after 30 min.Thus, we can conclude that 4 acts as a photocatalyst in both processes, epoxidation, and hydroxylation in aqueous media.It is worth mentioning that ring opening of epoxides is a promising process to produce 1,2-diols, an important functional group to produce pharmaceuticals, surfactants, or their intermediates [48].
As we have commented previously, the selectivity towards the epoxide in water decreased with the reaction time whereas the amount of diol increases as well as, in some cases, other overoxidation products.In an aqueous medium, the epoxide ring undergoes ring opening.However, we aimed to determine whether this process is driven by water or by our catalyst.Different experiments were then performed by exposing the sealed reaction to UV irradiation for 15 and 30 min, operating with 0.05 mol% of catalyst, under the same conditions used before and using three different epoxides: 1,2-epoxyoctane, styrene oxide, and trans-stilbene oxide.The results shown in Table S2 indicate the formation of the corresponding diols in water after 15 and 30 min if the photocatalyst is present.When the catalyst is absent, lower or nonexistent yields of diols have been observed.In the case of styrene, only benzaldehyde of benzoic acid has been detected after 30 min.Thus, we can conclude that 4 acts as a photocatalyst in both processes, epoxidation, and hydroxylation in aqueous media.It is worth mentioning that ring opening of epoxides is a promising process to produce 1,2-diols, an important functional group to produce pharmaceuticals, surfactants, or their intermediates [48].
As we have commented previously, the selectivity towards the epoxide in water decreased with the reaction time whereas the amount of diol increases as well as, in some cases, other overoxidation products.In an aqueous medium, the epoxide ring undergoes ring opening.However, we aimed to determine whether this process is driven by water or by our catalyst.Different experiments were then performed by exposing the sealed reaction to UV irradiation for 15 and 30 min, operating with 0.05 mol% of catalyst, under the same conditions used before and using three different epoxides: 1,2-epoxyoctane, styrene oxide, and trans-stilbene oxide.The results shown in Table S2 indicate the formation of the corresponding diols in water after 15 and 30 min if the photocatalyst is present.When the catalyst is absent, lower or nonexistent yields of diols have been observed.In the case of styrene, only benzaldehyde of benzoic acid has been detected after 30 min.Thus, we can conclude that 4 acts as a photocatalyst in both processes, epoxidation, and hydroxylation in aqueous media.It is worth mentioning that ring opening of epoxides is a promising process to produce 1,2-diols, an important functional group to produce pharmaceuticals, surfactants, or their intermediates [48].
We have also studied the behavior of 5 as photocatalyst in the alkene oxidation in water using low catalyst load of 0.005 mol%.The results are presented in Tables 2 and S3, including those previously obtained for compound 4, to facilitate a comparative analysis.In general, we can observe moderate-to-high conversion values after 30 min of reaction in most cases.It is worth noting that while the conversion values of catalyst 5 are generally higher than those obtained with 4, in most of the substrates studied, the selectivity values in the corresponding epoxides are lower even after 15 min of reaction.This is probably due to the enhanced oxidizing capacity of Co IV in catalyst 5, as a consequence of the presence of more electron-withdrawing substituents on the cosane platform.At this point, we can assert that in most cases and under the studied conditions, photocatalyst 4 shows higher selectivity values towards the epoxidation of alkenes that photocatalyst 5. ≥99(67) [b]  68(66) [b]   Molecules 2024, 29, x FOR PEER REVIEW 8 of 16 1344 [a] ≥99(57) [b]  73(56) [b]  1140 [b]  8200 [b]  ≥99(51) [c]  ≥99(61) [c]  1020 [c]  12,200 [c]   2 96(79) [b]  1520 [b]  6 ≥99(69) [b]  58(≥99) [b]  1380 [b]  11,600 [b]   ≥99(35) [c]  95(≥99) [c]  700 [c]  19,000 [c]   Conditions: 4 (0.01 or 0.001 mM), substrate (20 mM), Na2S2O8 (26 mM), 5 mL aqueous solution at pH = 7.
As we have commented previously, the selectivity towards the epoxide in water decreased with the reaction time whereas the amount of diol increases as well as, in some cases, other overoxidation products.In an aqueous medium, the epoxide ring undergoes ring opening.However, we aimed to determine whether this process is driven by water or by our catalyst.Different experiments were then performed by exposing the sealed reaction to UV irradiation for 15 and 30 min, operating with 0.05 mol% of catalyst, under the same conditions used before and using three different epoxides: 1,2-epoxyoctane, styrene oxide, and trans-stilbene oxide.The results shown in Table S2 indicate the formation of the corresponding diols in water after 15 and 30 min if the photocatalyst is present.When the catalyst is absent, lower or nonexistent yields of diols have been observed.In the case of styrene, only benzaldehyde of benzoic acid has been detected after 30 min.Thus, we can conclude that 4 acts as a photocatalyst in both processes, epoxidation, and hydroxylation in aqueous media.It is worth mentioning that ring opening of epoxides is a promising process to produce 1,2-diols, an important functional group to produce pharmaceuticals, surfactants, or their intermediates [48].
We have also studied the behavior of 5 as photocatalyst in the alkene oxidation in water using low catalyst load of 0.005 mol%.The results are presented in Tables 2 and S3, including those previously obtained for compound 4, to facilitate a comparative analysis.In general, we can observe moderate-to-high conversion values after 30 min of reaction in most cases.It is worth noting that while the conversion values of catalyst 5 are generally higher than those obtained with 4, in most of the substrates studied, the selectivity values in the corresponding epoxides are lower even after 15 min of reaction.This is probably due to the enhanced oxidizing capacity of Co IV in catalyst 5, as a consequence of the presence of more electron-withdrawing substituents on the cosane platform.At this point, we can assert that in most cases and under the studied conditions, photocatalyst 4 shows higher selectivity values towards the epoxidation of alkenes that photocatalyst 5.
As we have commented previously, the selectivity towards the epoxide in water decreased with the reaction time whereas the amount of diol increases as well as, in some cases, other overoxidation products.In an aqueous medium, the epoxide ring undergoes ring opening.However, we aimed to determine whether this process is driven by water or by our catalyst.Different experiments were then performed by exposing the sealed reaction to UV irradiation for 15 and 30 min, operating with 0.05 mol% of catalyst, under the same conditions used before and using three different epoxides: 1,2-epoxyoctane, styrene oxide, and trans-stilbene oxide.The results shown in Table S2 indicate the formation of the corresponding diols in water after 15 and 30 min if the photocatalyst is present.When the catalyst is absent, lower or nonexistent yields of diols have been observed.In the case of styrene, only benzaldehyde of benzoic acid has been detected after 30 min.Thus, we can conclude that 4 acts as a photocatalyst in both processes, epoxidation, and hydroxylation in aqueous media.It is worth mentioning that ring opening of epoxides is a promising process to produce 1,2-diols, an important functional group to produce pharmaceuticals, surfactants, or their intermediates [48].
As we have commented previously, the selectivity towards the epoxide in water decreased with the reaction time whereas the amount of diol increases as well as, in some cases, other overoxidation products.In an aqueous medium, the epoxide ring undergoes ring opening.However, we aimed to determine whether this process is driven by water or by our catalyst.Different experiments were then performed by exposing the sealed reaction to UV irradiation for 15 and 30 min, operating with 0.05 mol% of catalyst, under the same conditions used before and using three different epoxides: 1,2-epoxyoctane, styrene oxide, and trans-stilbene oxide.The results shown in Table S2 indicate the formation of the corresponding diols in water after 15 and 30 min if the photocatalyst is present.When the catalyst is absent, lower or nonexistent yields of diols have been observed.In the case of styrene, only benzaldehyde of benzoic acid has been detected after 30 min.Thus, we can conclude that 4 acts as a photocatalyst in both processes, epoxidation, and hydroxylation in aqueous media.It is worth mentioning that ring opening of epoxides is a promising process to produce 1,2-diols, an important functional group to produce pharmaceuticals, surfactants, or their intermediates [48].
We have also studied the behavior of 5 as photocatalyst in the alkene oxidation in water using low catalyst load of 0.005 mol%.The results are presented in Tables 2 and S3, including those previously obtained for compound 4, to facilitate a comparative analysis.In general, we can observe moderate-to-high conversion values after 30 min of reaction in most cases.It is worth noting that while the conversion values of catalyst 5 are generally higher than those obtained with 4, in most of the substrates studied, the selectivity values in the corresponding epoxides are lower even after 15 min of reaction.This is probably due to the enhanced oxidizing capacity of Co IV in catalyst 5, as a consequence of the presence of more electron-withdrawing substituents on the cosane platform.At this point, we can assert that in most cases and under the studied conditions, photocatalyst 4 shows higher selectivity values towards the epoxidation of alkenes that photocatalyst 5.
We have also studied the behavior of 5 as photocatalyst in the alkene oxidation in water using low catalyst load of 0.005 mol%.The results are presented in Tables 2 and S3, including those previously obtained for compound 4, to facilitate a comparative analysis.In general, we can observe moderate-to-high conversion values after 30 min of reaction in most cases.It is worth noting that while the conversion values of catalyst 5 are generally higher than those obtained with 4, in most of the substrates studied, the selectivity values in the corresponding epoxides are lower even after 15 min of reaction.This is probably due to the enhanced oxidizing capacity of Co IV in catalyst 5, as a consequence of the presence of more electron-withdrawing substituents on the cosane platform.At this point, we can assert that in most cases and under the studied conditions, photocatalyst 4 shows higher selectivity values towards the epoxidation of alkenes that photocatalyst 5.
Based on the photocatalytic results exposed above, we have postulated a mechanism, displayed in Figure 5, for the photoepoxidation of alkenes carried out by complexes 4 and 5.
Based on the photocatalytic results exposed above, we have postulated a mechanism, displayed in Figure 5, for the photoepoxidation of alkenes carried out by complexes 4 and 5.In the mechanism that is consistent with the products generated, the absorption of light by the photosensitizer Cop III produces the excitation to form Cop III *, which undergoes oxidative quenching by S2O8 2− (oxidizing agent) generating Cop IV .This photogenerated strong oxidizing Co IV can oxidize Ruc II -OH2 to Ruc III -OH.Then, the SO4 − radical oxidizes a new Cop III to Cop IV which oxidizes Ruc III -OH to Ruc IV =O species.The Ruc IV =O species reacts 9000 [a]  18,400 [a]   85(41) [b]  96(55) [b]  6970 [b]  10,600 [b]   Conditions: 4 or 5 (0.001 mM), substrate (20 mM), Na 2 S 2 O 8 (26 mM), 5 mL aqueous solution at pH = 7. Ratio 1:2000:2600 or 1:20,000:26,000, [a] 15 min of reaction, [b] 30 min of reaction.
Based on the photocatalytic results exposed above, we have postulated a mechanism, displayed in Figure 5, for the photoepoxidation of alkenes carried out by complexes 4 and 5.In the mechanism that is consistent with the products generated, the absorption of light by the photosensitizer Cop III produces the excitation to form Cop III *, which undergoes oxidative quenching by S2O8 2− (oxidizing agent) generating Cop IV .This photogenerated strong oxidizing Co IV can oxidize Ruc II -OH2 to Ruc III -OH.Then, the SO4 − radical oxidizes a new Cop III to Cop IV which oxidizes Ruc III -OH to Ruc IV =O species.The Ruc IV =O species reacts   [40] were also prepared according to literature procedures.All synthetic manipulations were performed under nitrogen atmosphere using vacuum line techniques.
All reagents used in the present work were obtained from Aldrich Chemical Co. and were used without further purification.Reagent grade organic solvents were obtained from SDS, and high purity deionized water was obtained by passing distilled water through a nano-pure Mili-Q water purification system.
UV-vis spectroscopy was performed on a Cary 50 Scan Varian (Santa Clara, CA, USA) UV-vis spectrophotometer with 1 cm quartz cells or with an immersion probe of 5 mm path length.NMR spectra have been recorded with a Bruker ARX 300 instrument (Bruker Biospin, Rheinstetten, Germany) and 1 H NMR spectra were recorded in acetone-d 6 .Chemical shift values were referenced to SiMe 4 .Elemental analyses were performed using a CHNS-O Elemental Analyser EA-1108 from Fisons (Waltham, USA).ESI-MS experiments were performed on a Navigator LC/MS chromatograph from Thermo Quest Finnigan (Toronto, Canada) using acetonitrile as mobile phase.Cyclic voltammetric (CV) or differential pulse voltammetry (DPV) was performed in an IJ-Cambria 660C potentiostat using a three-electrode cell.Glassy carbon electrode (3 mm diameter) from BAS was used as working electrode, platinum wire as auxiliary, and Ag as pseudo-reference electrode or SCE as the reference electrode.All cyclic voltammograms presented in this work were recorded under nitrogen atmosphere.The complexes were dissolved in deoxygenated solvents containing the necessary amount of TBAH as supporting electrolyte to yield a 0.1 M ionic strength solution.All E 1/2 values reported in this work were estimated from cyclic voltammetry experiments as the average of the oxidative and reductive peak potentials (E pa + E pc )/2.Unless explicitly mentioned, the concentration of the complexes was approximately 1mM.

Photocatalytic Studies
A quartz tube containing an aqueous solution (5 mL) at pH 7 (K 2 CO 3 solution is used to adjust the pH) with 4 or 5, as catalysts, alkene as substrate, and Na 2 S 2 O 8 as sacrificial acceptor was exposed to UV light (2.2 W, λ = 300 nm or 352 nm) for different times.The complex: substrate: sacrificial oxidant ratios used (1:2000:2600 and 1:20,000:26,000 corresponding to concentrations of 0.01:20:26 mM and 0.001:20:26 mM, respectively).The concentrations were varied according to the study.For each experiment, a light reactor supplied light illumination with 12 lamps that produce UVA light at room temperature.The resulting solutions were extracted with CH 2 Cl 2 six times.The solution was dried with anhydrous sodium sulfate and the solvent was reduced to a minimum volume under reduced pressure, then 100 µL of biphenyl 100 mM as internal standard was added to the resulting solution, 2 mM in the resulting 5 mL solution.To check the reproducibility of the reactions, all the experiments were performed in triplicate and analyzed by Gas Chromatography.

Gas Chromatography Studies
Studies were performed with a GC-2010 Gas Chromatograph from Shimadzu, equipped with an Astec CHIRALDEX G-TA column, 30 m × 0.25 mm (i.d) (Fukuoka, Japan); FID detector, 250 • C; injection: 250 • C; carrier gas: helium; rate: 1.57 mL/min; area normalization.For alkenes, substrates and products of catalysis were detected under the following conditions: styrene and derivatives: column temperature, 80

Figures 1 ,
Figures 1, S1 and S2 display the IR spectra of complexes 4, 5, and Ag[Cl 6 -1].The IR of 4 and 5 show vibrations around 2900 and 3030 cm −1 that can be assigned to υ C-H stretching modes for the aromatic rings of the cationic moieties and to the υ C-H stretching
The 11 B{ 1 H} NMR spectra displays signals corresponding to the non-equivalent boron atoms avoiding the B-H signals coupling.On the other hand, the 11 B NMR spectrum displays B-H signals as doublets and B-Cl signals as singlets.Both spectra show the typical pattern of the hexachloro cobaltabis(dicarbollide) cluster in the range from δ = 12.62 to −24.38 ppm, as shown in Figure S4 [34].

Figure 1 .
Figure 1.IR spectra of complexes 4 (blue) and 5 (green).The 1D 1 H NMR spectra of the complex Ag[Cl 6 -1] and compound 5, along with the COSY spectrum of 5, were recorded in acetone-d 6 .These are displayed in Figures2, S3and S4, respectively.The 1 H NMR spectrum of 5 shows two set of signals, similar to complex 4[40], (i) one in the aliphatic region that can be assigned to the C c -H of the cobaltacarborane anions, which resonances appear around δ = 4.34 ppm, showing a shift of 0.36 ppm to low field with respect to complex 4; and (ii) the second one, in the aromatic region corresponding to the protons of the bypyridine and terpyridine ligands of the ruthenium cation.The 1 H{ 11 B} NMR spectrum exhibits the H-B resonances over a wide range of chemicalshift in the region from δ = 1.70 to 3.5 ppm (FigureS4b).The 11 B{ 1 H} NMR spectra displays signals corresponding to the non-equivalent boron atoms avoiding the B-H signals coupling.On the other hand, the 11 B NMR spectrum displays B-H signals as doublets and B-Cl signals as singlets.Both spectra show the typical pattern of the hexachloro cobaltabis(dicarbollide) cluster in the range from δ = 12.62 to −24.38 ppm, as shown in Figure S4 [34].The UV-vis spectrum of Ag[Cl 6 -1] in CH 2 Cl 2 shows one strong absorption band at 327 nm and three others with less intensity at 280, 390, and 469 nm, (FigureS5) in accordance with[43][44][45].Figure3shows the UV-visible spectra of 5 and 2. The former exhibits ligand-based π-π* bands of the cationic part below 350 nm that are partially eclipsed by strong absorptions corresponding to the hexachloro cobaltabis(dicarbollide) anionic moiety.Above 350 nm, the spectra show less-intense bands that correspond to dπ(Ru)-π*(L) MLCT transitions[46].It is worth mentioning the shift to higher energy absorptions observed for the MLCT bands of 5 regarding those of complex 2. This seems to indicate that the substitution of the chloride counterions by the metallacarboranes provokes the stabilization of the dπ (Ru) donor orbitals.

Figure 3 .Figure 3 .
Figure 3. UV-vis spectrum of 5 (green line) and 2 (brown line) in dichlorometane.The electrochemical behavior of complexes Ag[Cl6-1] and 5 have been studied by means of cyclic voltammetry (CV).The CV curve of Ag[Cl6-1] in CH3CN + 0.1 M [n-Figure 3. UV-vis spectrum of 5 (green line) and 2 (brown line) in dichlorometane.The electrochemical behavior of complexes Ag[Cl 6 -1] and 5 have been studied by means of cyclic voltammetry (CV).The CV curve of Ag[Cl 6 -1] in CH 3 CN + 0.1 M [n-Bu 4 N][PF] 6(TBAH) shows one reversible one-electron-redox process at E 1/2 = −0.89V versus Ag/AgCl as reference electrode, which can be assigned to Co III /Co II (see FigureS6).Unlike the derivative Ag [1], the wave corresponding to Co IV /Co III is not observed, probably because it is outside the working range of the solvent.The CV of complex 5 in CH 3 CN + 0.1 M TBAH exhibits different redox processes due to the ruthenium and cobalt ions.Two quasi-reversible monoelectronic Co III Ru III /Co III Ru II and Co III Ru IV /Co III Ru III redox waves at E pa1 = 0.82 V; E pc1 = 0.74 V and E pa2 = 1.17 V; E pc2 = 0.93 V vs. Ag/AgCl (E pa1 = 0.38 V; E pc1 = 0.31 V and E pa2 = 0.73 V; E pc2 = 0.49 vs. Fc + /Fc) respectively and one quasireversible monoelectronic Co III Ru II /Co II Ru II redox wave at E pa3 = −0.76V; E pc3 = −0.97V vs. Ag/AgCl, (E pa3 = −1.19V; E pc3 = −1.40V vs. Fc + /Fc) (FigureS6).The redox wave corresponding to the Co IV Ru IV /Co III Ru IV is
p III produces the excitation to form Co p III *, which undergoes oxidative quenching by S 2 O 8 2− (oxidizing agent) generating Co p IV .This photogenerated strong oxidizing Co IV can oxidize Ru c II -OH 2 to Ru c III -OH.Then, the SO 4 − radical oxidizes a new Co p III to Co p IV which oxidizes Ru c III -OH to Ru c IV =O species.The Ru c IV =O species reacts with the corresponding alkene to afford the oxidized products, with the regeneration of the corresponding catalyst Ru c II -OH 2 .With the pathway proposed in Figure 5, the exchange of two electrons and two protons takes place in the oxidation of alkenes.
• C for 5 min, raising to 170 • C in a rate of 10 • C/min, holding 170 • C for 6 min.Trans-β-methyl-styrene, cis-β-methylstyrene, cis-cyclooctene and derivatives: column temperature, 40 • C for 5 min, raising to 170 • C in a rate of 5 • C/min, holding 170 • C for 2 min.1-octene and derivatives: column temperature, 30 • C for 5 min, raising to 170 • C in a rate of 10 • C/min, holding 170 • C for 3 min.Trans-stilbene and derivatives: column temperature, 50 • C for 1 min, raising to 150 • C in a rate of 15 • C/min, holding 150 • C for 2 min, then raising to 170 • C in a rate of 4 • C/min, holding 170 • C for 12 min.The product analyses in the catalytic experiments were performed by GC with biphenyl as internal standard.