Skip to main content
NCBI home page
ACS Omega logoLink to ACS Omega
. 2020 Mar 30;5(13):7613–7626. doi: 10.1021/acsomega.0c00447

Sustainable Oxidation of Cyclohexane and Toluene in the Presence of Affordable Catalysts: Impact of the Tandem of Promoter/Oxidant on Process Efficiency

Alexander Pokutsa †,*, Sergiy Tkach ∥, Andriy Zaborovsky †, Pawel Bloniarz ‡, Tomasz Paczeŝniak ‡, Jacques Muzart §
PMCID: PMC7144162  PMID: 32280905

Abstract

graphic file with name ao0c00447_0004.jpg

The oxygenation of cyclohexane and toluene by O2 and H2O2 catalyzed by VO(acac)2 and Co(acac)2 was studied at 40–100 °C and 1–10 atm. Upon such conditions, the process can be remarkably (30× times) enhanced by the minute (6–15 mM) additives of oxalic acid (OxalH) or N-hydroxyphthalimide (NHPI). The revealed effect of OxalH on H2O2-piloted oxidation is closely associated with the nature of the catalyst cation and boosted by VO(acac)2. Whereas the effectiveness of Co(acac)2-based systems was curbed by the addition of OxalH and remained much below the one displayed with the previous system. The observed conspicuous difference in activity was attributed to the substantially higher solubility of in situ formed VO(IV)oxalate compared to that of Co(II)oxalate. The exploration of H2O2 for the NHPI-promoted process leads to the decisively lower (5–7 times) yield in comparison to the O2-driven reaction. Similarly, for the O2-operated protocol, the yield cannot be improved by addition of OxalH either to VO(acac)2 + NHPI or to Co(acac)2 + NHPI mixture. By contrast, the combination of NHPI with VO(acac)2 or Co(acac)2 and particularly with the above two mixtures in O2-piloted oxidation enhances the yield of the aimed products 3–6 times regardless of the substrate used. The revealed significant synergetic effect of the cobalt + vanadyl bicomponent catalyst was due to the participation of each of its moiety in the different stages of the process mechanism. Only benzyl alcohol and benzaldehyde were identified in VO(acac)2- or Co(acac)2-catalyzed toluene oxidation, while cyclohexane oxidation yields cyclohexylhydroperoxide in line with cyclohexanol and cyclohexanone. The putative mechanism of investigated processes is highlighted and discussed.

1. Introduction

Selective oxidation of hydrocarbons with strong C–H bonds remains in the focus of interests of both academia and industry. This curiosity is stipulated by the assortment of manufactured products and their widespread use, as well as the relative abundance of feedstock (natural and technological gaseous and liquid fuels), moderate cost, and toxicity. On the other hand, the inertness of such hydrocarbons makes them appropriate for the final trial of designed oxidation systems. For example, the bond dissociation energy (BDE) value of aliphatic C–H bonds in PhCH3 (93 kcal mol–1) and C6H12 (98 kcal mol–1)3 makes these substrates a proper candidate for such a role. Then, pursuing the enhancement of cyclohexane (C6H12) and toluene (PhCH3) oxidation would provide a deeper insight into the issues challenging both fundamental and applied chemistry.

The products of C6H12 and PhCH3 oxidation—cyclohexanol (C6H11OH), cyclohexanone (C6H10(O)), cyclohexylhydroperoxide (C6H11OOH), benzyl alcohol (PhCH2OH), and benzaldehyde (PhC(O)H)—are utilized as commodities for the manufacture of artificial fibers, dyes, solvents, perfumeries, plasticizers, preservatives, and flame retardants.1,2 To date, high temperatures and expensive oxidants and additives are needed to maintain process productivity. Such factors are also responsible for side reactions leading to moderate selectivity, low yield, and unacceptable environmental and economic costs.4 The significance of this issue is reflected by the number of papers published worldwide encompassing theoretical study,2,3 synthesis of sophisticated ligands, and catalyst preparation.5−9

The main disadvantage of the artificial catalytic systems concerns the low ability to afford the direct oxygen transportation from oxidant (O2 or H2O2) to oxidizing substrate that is kinetically prohibited by the spin conservation rule.10 To overcome this obstacle, organic initiators (generally peroxides, peracids, diazo-compounds, etc.)12,13 and/or promoters (e.g., aliphatic aldehydes) are usually explored.14,15 According to the mechanism, the enlisted compounds speed up the oxidation using the free radicals generated in the course of their homolysis. In a similar, although not identical, manner, N-hydroxyphthalimide (NHPI)17,18 enables to transfer electrons by the way of mimicking biological oxidation.10,16

The key steps of substrate/oxidant activation by initiators, promoters, and mediators are charted in Scheme 1. The main benefit of the application of initiators is their relatively low price (particularly when H2O2 or t-BuOOH are explored)19 and the shortest way (one step) it impacts the process (route I, Scheme 1). Promoters act rather in a similar way except for the number of steps (four) necessary to generate active oxy-radicals (route II, Scheme 1).20 It is most beneficial in terms of process rate and yield to use NHPI as a promoter/mediator (route III, Scheme 1) because this compound possesses a weaker (compared to oxidizing substrates) O–H bond owing to the presence of electron-withdrawing (carbonyl) groups. The dissociation energy (BDE) of the O―H bond in NHPI (88 kcal mol–1) is lower than that of CH2―H in PhCH3 (90 kcal mol–1) or CH―H in C6H12 (100 kcal mol–1).3,4 Compared to the BDE of HO―OH or t-BuO―OH (about 50 kcal mol–1), the corresponding value of 88 kcal mol–1 is certainly higher than and closer to that of the O–H one (90 kcal mol–1) of the referred hydroperoxides developed.21 Nevertheless, the abstraction of the hydrogen atom from the hydroxyl group of NHPI is facilitated in comparison to HOO―H17,22 by the electrophilic character of NHPI, which, in turn, rendered with the polar effect of attached carbonyl groups. On the other hand, NHPI can act as an initiator, promoter, and catalyst concomitantly, whereas H2O2 affords the initiation step only. Alongside, the generated phthalimide-N-oxi radicals (PINO) trigger the cascade of radical-steered oxidative transformations, assist the catalytic cycle, and ultimately enhance the yield (route III, Scheme 1).23 (Oxalic acid primarily grants the hidden-radical mechanism of the process, catalyst reduction, and renders proper electrochemical characteristics to the reaction medium.)24,25 Then, the use of NHPI allows us to reduce the toxicity of the operation and expensive organic initiators and promoters.26

Scheme 1. Peculiarities of Initiator (I), Promoter (II), and Promoter/Mediator (III) Action in Transition-Metal Cation-Catalyzed Oxidation.

Scheme 1

Open in a new tab

Elucidating the mechanism of oxidation in the presence of NHPI, OxalH, its mixture and different oxidants may be beneficial for the process efficiency as well as the impact of natures of oxidants on process productivity looks having a sense. For example, the competitive stages a21–a23 of the III cycle (Scheme 1) may be responsible for the irreversible catalyst oxidation, which ultimately leads to retardation of the yield.

Such issues motivated us to study the relationships among OxalH and/or NHPI additives, oxidant, temperature/pressure, and process effectiveness. This study aims also to elaborate the oxidation protocol that allows us to avoid the high load (10 mol %, according to ref (17)) of NHPI as well as metallocomplexes. The oxidation tests investigated the oxidation of C6H12 and PhCH3 by O2 or H2O2 in the presence of small-molecule VO(acac)2 and Co(acac)2 catalysts within 40–100 °C and 1–10 atm intervals. The results of the efforts made are reported therein.

2. Results and Discussion

2.1. Cyclohexane Oxidation

Both VO(acac)2 and Co(acac)2 taken together with either oxalic acid or NHPI to catalyze the oxidation of cyclohexane to cyclohexanol, cyclohexanone, and cyclohexylhydroperoxide although with different efficiencies and product proportions. (In a cation-free process, no products were identified despite the presence of promoters). Oxalic acid enhances the productivity of the VO(acac)2-based process, but it is detrimental to that catalyzed by Co(acac)2 (compare entry 3 with 6 of Table 1; curves 1, 2, 3 with curve 4 of Figure 1). The revealed feature could be a consequence of different solubilities of the VO2+ and Co2+ oxalates formed in situ by OxalH and metal–salt interaction. For example, the addition of Co(acac)2 to the oxalic acid solution led to a quick precipitation of fine salmon powder (insoluble in water too). (According to the literature it may be Co(II)oxalate dihydrate.27) While the light blue mixture VO(acac)2 + H2C2O4 remains transparent for a week, the insolubility of Co(II)oxalate causes a shortage of Co2+ ions in the solution, a lower rate of cation-assisted steps (Scheme 2, path A), and ultimately a reduced yield (curve 4, Figure 1; entry 6, Table 1). (For the moment, we can only hypothesize what may happen if the preprepared VO oxalate or Co oxalate are explored as oxidation catalysts. Studies to answer this query are being planned in the nearest future.)

Table 1. Results of C6H12 Oxidation Depending on Initial System Composition, Oxidant, Temperature, and Pressure.

  VO(acac)2 Co(acac)2 OxalH NHPI   selectivity (mol %)
 TON
    
entry × 103 (M) ΔC6H12% C6H11OH C6H10(O) C6H11OOH by cation by NHPI ΔH2O2b (%) EffH2O2c (%)
H2O2-Piloted Oxidationa
1 0.6       1.2 33 62 5 23   48 4.2
2 0.6   1.0   1.7 1 8 91 51   52 6.5
3 0.6   15   5.3 48 7 45 169   96 8.4
4 0.6   50   7.6 43 13 44 253   90 13.2
5 0.6   150   5.2 47 10 43 173   80 9.9
6   0.6 15   <1 trace trace trace <1   20 <1
7d 0.3     6 0 0 0 0 0 0 0  
8d   0.3   6 0 0 0 0 0 0 0  
O2-Piloted Oxidatione
9 0.3       <0.1 trace trace trace <1      
10 0.3   15   0.9 22 75 3 40      
11 0.3   15 6 1.9 13 84 3 87 4    
12 0.3     6 3.3 5 80 15 153 8    
13   0.3 15   0.3 14 51 35 14      
14   0.3 15 6 1.2 24 64 12 57 3    
15   0.3   6 4.9 5 70 25 229 11    
16 0.3     6 4.2 12 83 5 200 10    
Open in a new tab
a

[C6H12]0 = [H2O2]0 = 1.4 M; 40 °C, 1 atm, MeCN, 5 h.

b

Amount of H2O2 consumed.

c

EffH2O2 is the ratio of the stoichiometric (due to the product yield) amount of H2O2 divided by the H2O2 consumed.

d

[C6H12]0 (1.4 M), [H2O2]0 (0.14 M).

e

[C6H12]0 (1.4 M), O2 bubbling (100 mL min–1), 100 °C, 10 atm, 5 h, MeCN. For Co(acac)2-catalyzed oxidation by H2O2, only traces of products were detected regardless of the presence of oxalic acid additives. ΔC6H12—substrate conversion. Due to almost 100% process selectivity, the total product yield was calculated as: ΔC6H12/100% × 1.4 M.

Figure 1.

Figure 1

Open in a new tab

Kinetics of C6H11OH (1), C6H10(O) (2), and C6H11OOH (3) accumulation upon VO(acac)2-catalyzed C6H12 oxidation by H2O2 in the presence of 5 mM oxalic acid (4—the kinetics of total yield alteration when 0.6 mM Co(acac)2 was used instead of VO(acac)2). Inset: time traces of C6H11OH/total yield (1), C6H10(O)/total yield (2), and C6H11OOH/total yield (3) ratios. Oxidation conditions: C6H12 = H2O2 (1.4 M), VO(acac)2 (0.6 mM), OxalH (5 mM), MeCN, 40 °C, 1 atm.

Scheme 2. Key Putative Steps of H2O2 (A)- and O2 (B, C)-driven C6H12 and PhCH3 oxidation.

Scheme 2

Open in a new tab

The numbers next to the formulas of C6H11OOH, C6H11OH, C6H10(O), PhCH2OH, and PhC(O)H denote the selectivity of the respective product obtained when H2O2 and O2 (values in brackets) were used as oxidant. (For the sake of clarity, the presentation of Russel’s-type dibenzylenetetraoxide formation was omitted from consideration.)

A similar of effect oxalic acid is demonstrated in the case of C6H12 oxidation by O2 at increased temperature and pressure (compare curves 1 and 2, Figure 2). However, in the absence and presence of OxalH, the yield differed not so conspicuous as in the process piloted by H2O2 (Figure 1). For example, the substrate conversion (ad hoc the yield because of more than 90% process selectivity) catalyzed by VO(acac)2 and Co(acac)2 O2-piloted oxidation differed roughly by 3 times (0.9/0.3% = 3, entry 10 and 13, Table 1; curves 2 at Figures 2 and 3). At the same time, only traces of products were detected if the reaction mixture comprised Co(acac)2 + OxalH + H2O2 (entry 3 and 6, Table 1). Increasing the reaction temperature for the Co(acac)2/OxalH/O2 oxidation system leads to a slightly higher yield compared to the one obtained at 40 °C (confer curve 4, Figures 1–3).

Figure 2.

Figure 2

Open in a new tab

Kinetics of the sum C6H11OH + C6H10(O) + C6H11OOH accumulation upon VO(acac)2-catalyzed C6H12 oxidation with O2: 1—without mediator; 2—in the presence of 15 mM OxalH; 3—15 mM OxalH + 6 mM NHPI; 4—6 mM NHPI. Insets: time traces of C6H11OOH/total yield, C6H11OH/total yield, and C6H10(O)/total yield ratios alteration: 1—in the presence 15 mM OxalH; 2—15 mM OxalH + 6 mM NHPI; 3—6 mM NHPI. Oxidation conditions: C6H12 (1.4 M), VO(acac)2 (0.3 mM), O2 bubbling (100 mL·min–1), total reactants volume = 50 mL, MeCN, 100 °C, 10 atm, stainless steel reactor.

Figure 3.

Figure 3

Open in a new tab

Kinetics of C6H11OH + C6H10(O) + C6H11OOH accumulation upon Co(acac)2-catalyzed C6H12 oxidation with O2: 1—without activators; 2—in the presence of 15 mM OxalH; 3—15 mM OxalH + 6 mM of NHPI; 4—6 mM NHPI. Insets: time traces of C6H11OOH/total yield, C6H11OH/total yield, and C6H10(O)/total yield ratios alteration: 1—in the presence 15 mM OxalH; 2—15 mM OxalH + 6 mM of NHPI; 3—6 mM NHPI. Oxidation conditions: C6H12 (1.4 M), Co(acac)2 (0.3 mM), O2 bubbling (100 mL·min), total reaction volume = 50 mL, MeCN, 100 °C, 10 atm, stainless steel reactor.

On the other hand, the different nature and thereof the activity of species originated in O2- and H2O2-driven oxidation can impel the whole process mechanism. For example, the oxidation by O2 may involve the initial Mn+–OO• species generation (Scheme 3). By contrast, in the H2O2-piloted process, the initially formed putative Mn+1–OOH adduct afterward decomposed to the active metal-oxo (Mn+1–O•) and HO• species, which are evidently far more active than the metal-peroxo ones.28 For example, the H-atom-abstracting ability of hydroxyl radicals is approximately 102–103 times higher compared to that of Mn+–OO•.29

Scheme 3. Key Putative Steps of NHPI-Promoted Oxidation by O2 (Routes in Black), NHPI Decomposition Upon Acidic pH (Routes in Blue), and β-Scission of PINO Radical (IIa) and the Putative Intermediate IIb (Routes in Red).

Scheme 3

Open in a new tab

Using NHPI as a process promoter speeds up the O2-driven C6H12 oxidation and notably increases the yield of both VO- and Co-catalyzed processes (compare curve 2 with curve 4 in Figures 2 and 3, as well as entry 10 with entries 11 and 12 of Table 1). The primary intermediates responsible for this are PINO radicals (Schemes 2 and 3).17 In opposite, oxalic acid, once mixed with NHPI, diminishes the product yield regardless of the type of catalyst and oxidant used (compare curve 3 with curve 4 in Figures 2 and 3, as well as entry 11 with entry 12 of Table 1). In the case of a H2O2-piloted process catalyzed by Co2+, the negative impact of oxalic acid on the reaction becomes more evident (compare pairs of curves 3 and 4 in Figures 2 and 3, as well as runs 11 and 13 of Table 1). Such a trait could be a consequence of the much lower solubility of Co(II)Oxalate in comparison to VO(Oxal)2 (above). Moreover, the observed peculiarity may have resulted from the proton-driven mechanism of the NHPI hydrolysis (Scheme 3 and the related discussion below).

The systems constituted of C6H12 and NHPI show clearly worse yields when piloted by H2O2 regardless of temperature, pressure, and catalyst used (Table 1, pairs of entries 12, 15 and 8, 16). It can occur due to: (i) quick NHPI destruction by, e.g., hydrolysis caused by both hydroperoxide itself and water, presented in H2O2 as well as originated by the catalytic cleavage of the later. The last reaction (extremely fast at pH < 7) may trigger the cascade of NHPI decompositions (may involve free-radical-dependent ones), resulting in phthalic acid and hydroxyl amine (Scheme 3, routes in blue).30 (ii) β-scission of PINO (IIa) radicals17 as well as the intermediate IIb (Scheme 3, upper route in red). This pathway could be an alternative to the conventional classic mechanism (routes in black).31 (According to the literature, the energy required to break the single bond in β-position in either hydrocarbon •CH2–CH2―H radicals or oxy-carbon •CH2–O―H, •O–CH2―H ones is 2.5–4 times less than the one that has to be applied in α-position.)32 The last scenario implies the formation of compound III with o-quinonoid structure.33 Although the calculated rate of β-scission of > N―C bond is 7 orders of magnitude lower compared to the value intrinsic of C―C,34 the former one may become higher upon the actual experimental conditions (d-elements compounds, H2O2, polar medium, pH < 7). All of that leads to the prevalence of free-valent species generation. The last product (III) could have also undergone further β-scission (Scheme 3, lower route in red). According to this scheme, free radicals, particularly hydroxyl ones (the average rate constant of H-atom abstraction by HO• is 107–108 M–1 s–1),29 are able to attack any C–H bonds. In this case, the initial stage of NHPI destruction may involve hydrogen atom abstraction from the phenyl ring (Scheme 3, lower route in red), has 4 times more places (four H atoms) for •OH attacks compared to that of the >N–O―H group. Consequently, the intermediate formed may come into play for β-scission conjugated with the rearrangement of formed radical species leading to the same ultimate product III. In addition, HO• radicals35,36 are extremely powerful oxidizing and acidic agents with the reduction potential E•OH0 of 2.7 V (H+/H2O, pH < 7).11 (This value is only slightly lower than that of intrinsic fluorine EF–/F20 = 2.87 V.)37 Therefore, such species can smoothly destroy NHPI by itself or/and afford the circumstances beneficial for such process.38 Experimental data of the mechanism of Scheme 3 are presented in Section 2.3.

Another interesting feature of C6H12 oxidation (which is in fact an intrinsic PhCH3-based process too) is the clear relationship among the type of oxidant, yield, and assortment of products. In the case of H2O2, the free radicals are produced instantly and become the principal oxidizing agents. HO• is a much more powerful oxidant (E0 = 2.7 V) than parental H2O2 (1.77 V) or O2 (1.21 V).39,40 Then, the outstanding oxidizing power of H2O2 is stipulated mostly by the ability to generate hydroxyl radicals. The last ones can easily abstract H atoms working as a trigger for the subsequent free-radical-dependent steps (route A, Scheme 2).

Contrarily, O2-piloted oxidation of nonactivated C–H bonds cannot be triggered by conventional d-metals catalysts solely. The inertness of such a system to oxidation (curve 1 at Figures 2 and 3) is explained by the much lower value of E0 (1.21 V, above), higher BDE (118.8 kcal mol–1), and shorter O–O distance (121 pm) of O―O in comparison to those of HO―OH (1.77 V, 49.5 kcal mol–1, and 147.5 pm, respectively).41 On the other hand, direct interaction of O2 (ground spin quantum number triplet) with organic molecules (singlet state) is restricted by the spin selection rule. Therefore, paramagnetic molecular oxygen readily reacts only with molecules in a doublet state (e.g., radicals). By contrast, diamagnetic organic molecules (particularly the one possessing strong C–H bonds, e.g., toluene and cyclohexane with BDE equal to 90 and 100 kcal mol–1, respectively)3,42 are practically inert to O2 attack also due to the lack of unpaired electrons. Then, the combination of metallocomplexes with promoters/mediators could be beneficial with respect to process efficiency.

The coordination of O2 on the catalyst metal center ends up in the metal–dioxygen Mn+1–OO• adduct that consequently activates the O–O bond. Next, these metal-peroxo species hypothetically can react directly with substrates to withdraw H atoms and produce free radicals. However, the rate of such interaction is several orders of magnitude smaller compared to that of radicals.43 Then, NHPI application helps to overcome the thermodynamic and quantum-chemical incompatibilities, retarding the direct Mn+1–OO•/substrate interaction. Such action consists of splitting the total reaction activation energy Eact into two consecutive steps with smaller Eact on each one (route III in Scheme 1 and B in Scheme 2). As a consequence, the overall rate of process and yield may get increased.

From Figures 1– 3, one can deduce that the kinetics of total products accumulation in the VO(acac)2-catalyzed process remains rather identical to that inherent of Co(acac)2-based one for both tested promoters and oxidants. On the other hand, the existing dissimilarities in the profiles of products ratio vs time dependencies (Figures 1–3 insets) point out the influence of mediator and the oxidant nature on the process mechanism. For example, using NHPI upon increased pressure of oxygen prompts C6H10(O) formation regardless of the nature of the catalyst (curves 2 and 3 in Figures 2 and 3). Such a feature is imprinted by the C6H10(O)/C6H11OH ratios: when OxalH in the VO(acac)2-catalyzed process was replaced by NHPI, it shifts such proportion from 1.1–2.0 to 2.5–8.0 (insets b and c, Figure 2). Generally, the C6H11OH yield dropped after a period of oxidation catalyzed by VO(acac)2 (compare insets b and inset c, Figures 2 and 3), while the content of C6H10(O) increases. It may be a consequence of self-termination of two secondary peroxy radicals resulting in cyclic tetraoxide intermediate (path B3),44 intermolecular rearrangements of cyclohexylperoxy radicals (path B2) and C6H11OOH (path B4) as well as a consecutive C6H11OH oxidation (path B5). By contrast, C6H11OH can be accumulated by paths B1 and B3 (Scheme 2) only. In turn, the change of C6H11OOH concentration following a parabolic manner45 (insets a, Figures 1 and 2) depicts the latter as a transient product gradually decomposed to C6H10(O) and C6H11OH (Scheme 2). It is worth noting that the concentration of C6H11OOH was much lower than C6H10(O) irrespective of the catalyst, mediator, and oxidant used (insets a and c in Figures 1–3). Examining the influence of different promoters, one can deduce that the presence of oxalic acid enhances the C6H11OOH yield in the VO(acac)2-catalyzed process. However, the last value still did not exceed 70–80% of the one obtained in the presence of Co(acac)2 (insets a, Figures 2 and 3).

The detected prevalence of C6H11OOH formation in H2O2-driven C6H12 oxidation catalyzed by VO(acac)2 and mediated by small additives of oxalic acid may be the consequence of the next matters. First of all, the production of HO• (primary active species) by H2O2 decomposition is stimulated by oxalic acid additives (path A, Scheme 2). The abstraction of H-atom from the substrate by these radicals triggers free-radical steps, leading consequently to C6H11OOH and improving the yields of both C6H11OH and C6H10(O) (Table 1). It may occur due to the acceleration of both forward and reverse reactions of the Haber–Weiss cycle in view of the noteworthy redox and acidic properties of oxalic acid.46 At the same time, the yield of C6H11OOH was lower in the O2-piloted process promoted by either NHPI or OxalH + NHPI. Again, the prevalence of peroxy radical disproportionation (path B2) and square termination by Russel’s mechanism (path B3) may be responsible for this.44 The last transformations that may proceed via the formation of C–H···O– intermolecular bonds resulted in ketone, alcohol, and O2 (Scheme 2).47

The kinetics of C6H11OOH accumulation catalyzed by Co(acac)2/NHPI O2-piloted oxidation was slightly different from that for the Co(acac)2/OxalH system (inset a, Figure 3). However, the most conspicuous differences were found for the time traces of C6H10(O) (inset c, Figure 3). This product shows a much higher yield detected for the VO(acac)2/OxalH system and subjected to certainly better solubility of generated in situ VO(IV)oxalates than Co(II)oxalates. Nevertheless, the kinetic profile of C6H10(O) accumulation remains generally the same (insets c, Figures 2 and 3). Contrary to C6H10(O), the kinetics of C6H11OOH and particularly C6H11OH production in the Co-catalyzed process contrasts with the ones inherent to the VO(acac)2-based process (compare insets a and b in Figures 2 and 3).

Figure 4 summarizes the acquired kinetic results, whose evaluation reveals that the relative contents of C6H11OOH, C6H11OH, and C6H10(O) varied when the process evolves. The sequence of limitation steps switching on and off during the process is responsible for these products formation order (that affects its relative content). The quantified products assortment alteration correlates well with the reactiveness of pointed components. For example, the majority of profiles (except curves 4 and 5) of C6H11OOH (the most unstable product) followed the classic parabolic (rainbow-shaped) trajectory imprints of this product’s sequential transformation (Figure 4a). (In fact, the rest of the curves, i.e., 4 and 5, exhibit the parabolic character, too, although with a negative half-loops shape.) The order of C6H11OH and C6H10(O) accumulation/consumption demonstrates a trend different from that of C6H11OOH (compare Figure 4a–c). For instance, most curves representing the kinetics of alcohol and ketone content alteration that develop the positive half-parabola character point out the origin of the probability (among others) of these products by C6H11OOH decomposition. The factor that decides the prevalent C6H11OH, and particularly C6H10(O), accumulation in the course of the process is, most probably, the higher resistance of the last two compounds to further transformation. Since that, their relative contents at tested oxidation conditions increased when the reaction proceeds. Quantitatively, such peculiarities are charted by the positive half-parabola loops in Figure 4b,c and also depict the origin of the order of these two products (Scheme 2).

Figure 4.

Figure 4

Open in a new tab

Time traces of C6H11OOH/total yield (a), C6H11OH/total yield (b), and C6H10(O)/total yield (c) ratios alteration depending on the nature of promoter, catalyst cation, and oxidant used. Conditions: 1—OxalH, VO(acac)2, H2O2; 2—OxalH, VO(acac)2, O2; 3—OxalH, Co(acac)2, O2; 4—NHPI, VO(acac)2, O2; 5—NHPI, Co(acac)2, O2.

The acquired data support the foregoing statements concerning curbing C6H11OOH homolytic cleavage (path B1), peroxyl radicals (B2), and C6H11OOH (B4) disproportionation as well as square-termination (path B3) reactions (Scheme 2). Since a slight increase of the C6H11OOH content was inherent to the Co-catalyzed process, it may mirror less radical-induced peroxide decomposition (insets a in Figures 2 and 3). However, some vague features of such mechanism still await more meticulous envisaging.

2.2. Toluene Oxidation

The total yield of PhCH2OH and PhC(O)H in VO(acac)2-catalyzed oxidation was substantially (about 40–45 times) higher compared to that of the promoter-free process (Table 2, Figures 5, and 6). Similarly to C6H12 oxidation, no products were identified in the absence of cations regardless of the promoter used. On the other hand, supplementing the reaction solution with either OxalH or H2O2 does not provide any advantages for NHPI-promoted process and were rather detrimental with respect to the yield (compare run 1 with run 2; run 6 with run 7; and run 6 with run 15, Table 2). The yield of aldehyde was significantly (10-fold in average) higher than that of alcohol (Figures 5, 6, and Table 2). Altogether, such features correlate with the dependencies intrinsic to the C6H11OH + C6H10(O)-targeted process (Table 1). Thus, one can infer the involvement of the same type of elemental reactions to the origin of these products (compare route C with route B, Scheme 2). However, in contrast to C6H12, no peroxo derivatives of toluene were detected irrespective of the oxidation conditions (Table 2). Furthermore, the ratio of benzaldehyde/benzyl alcohol (∼10) disclosed by the current study (Table 2, Figures 5, and 6) correlates with the one referred in ref (48) and contradicts with another one49 declared the mutually inverse PhC(O)H/PhCH2OH proportions.48,49 Such dual behavior may indicate the notable impact of experimental conditions on the mechanisms. In the case of C6H12, the free-radical-piloted transformations dominate,50 albeit the nonradical ways of substrate oxidation may not be fully excluded from consideration.25 By contrast, due to the lack of peroxides formation during the PhCH3 oxidation (Table 2), one may conclude that free radicals are not involved in the process mechanism on a large scale (Scheme 2). Nevertheless, at least, partial involvement of hidden-radical steps in the overall mechanism cannot be ignored.51 The beginning stages of PhCH3 oxidation encompass the formation of benzyl radicals followed by diffusion-controlled benzylperoxy radical generation (Scheme 2). Consequently, the latter species can either abstract hydrogen atom from toluene leading to benzyl hydroperoxide or couple easily with other peroxy radicals, e.g., by Russel’s mechanism (or, alternatively, by the putative C3 one), or undergo intramolecular rearrangement to produce alcohol and aldehyde. The detected increasing yield of PhC(O)H may be a consequence of the latter two opportunities denoted by paths C2, C3, and C4. For the moment, we can admit that the PhC(O)H yield substantially overcomes that of PhCH2OH due to the +I effect of the methyl group as well as the electron deficiency of initial PhCH3 (pKa = 43)52 (the highly symmetric and nonpolar C6H12 with pKa = 59 cannot be polarized by any means.)53 Furthermore, in contrast to the single H atom pendant at the secondary cyclohexylperoxo radical, the primary benzylperoxo radical when formed has two hydrogens, which doubles with much possibilities to create an intramolecular hydrogen bond with the neighboring oxygen. Overall, these determine the type of cyclic tetroxide intermediate formation followed by its subsequent decay to benzaldehyde and H2O2. Such a scenario would become an alternative way to the one proposed by Russel (paths C3 and B3).44,47 The experimental PhC(O)H/PhCH2OH ratio spans from 6:9 to 9:15 depending on whether H2O2 or O2 was used (Table 2) and corresponds well with these theoretical assumptions. Therefore, one may admit that the higher the value of the PhC(O)H/PhCH2OH proportion, the less important role the free-radical steps play in the whole mechanism of substrate transformation (Scheme 2). The last conclusion also holds true for the PhC(O)H/total product ratios (Figures 5 and 6, insets). These ratios slightly decrease in the course of the process regardless of the catalyst used, which apparently indicates the priority of free-radical steps in the mechanism. Summarizing, these results led us to a conclusion of quick homolysis of benzoilperoxide at oxidation conditions. The decrease of the benzaldehyde content can evidently occur due to its minor oxidation to benzylic acid. Again, based on experimental data, the process mechanism may be defined as mostly mixed-radical/hidden-radical dependent and may be charted by the set of elemental steps C2, C3, and C4 (Scheme 2).

Table 2. PhCH3 Oxidation Depending on Initial Reaction Mixture Composition and Nature of Oxidant.

  VO(acac)2 Co(acac)2 OxalH NHPI   selectivity (mol %)
 TON
    
entry ×103 (M) ΔPhCH3% PhC(O)H PhCH2OH peroxidesd by cation by NHPI ΔH2O2b (%) EffH2O2c (%)
H2O2-Piloted Oxidationa
1 0.06   15   1.6 86 14 u/d 350 - 37 4
2 0.06   15 6 1.3 89 11 u/d 300 3 30 4
3 0.3   15   2.0 86 14 u/d 93 - 50 4
4 0.3     6 <0.1 trace 0 0 <1 <1 10 0
5   0.3   6 <0.1 trace 0 0 <1 <1 8 0
O2-Piloted Oxidatione
6 0.3     6 5.0 94 6 0 233 12    
7 0.3   15 6 2.3 94 6 0 110 5.5    
8   0.3   6 10.7 90 10 0 500 25    
9 0.15 0.15   6 9.0 89 11 0 420 21    
10 0.075 0.075   6 10.4 92 8 0 973 24    
11 0.037 0.037   6 10.0 90 10 0 1867 23    
12 0.020 0.020   6 6.7 93 7 0 1247 15    
13   0.037   6 1.1 99 1 0 432 3    
14   0.075   6 11.0 92 8 0 2053 26    
15f 0.3     6 2.4 91 9 u/d 11 5 100  
Open in a new tab
a

[PhCH3]0 = [H2O2]0 = 1.4 M; 40 °C, 1 atm, MeCN, 5 h.

b

Amount of H2O2 consumed.

c

EffH2O2 is the ratio of stoichiometric (due to the product yield) amount of H2O2 divided by the H2O2 consumed.

d

Peroxides content was tracked by treating the samples with Ph3P (see the Experiments section). u/d—content undetected.

e

[PhCH3]0 (1.4 M), O2 bubbling (100 mL min–1), total reaction volume = 50 mL, 100 °C, 10 atm, 5 h, MeCN, stainless steel reactor. ΔPhCH3 (substrate conversion) corresponds to the product yield due to almost 100% process selectivity.

f

[PhCH3]0 = 1.4 M, [H2O2]0 = 0.14 M. Only traces of products were detected when OxalH was presented in Co(acac)2-catalyzed oxidation by O2 (100 °C, 10 atm). No products formed in a detectable amount in the absence of neither OxalH nor NHPI regardless of the catalyst and oxidant used. The combination of either VO(acac)2 or Co(acac)2 with NHPI in oxidation by H2O2 eliminates the advantages of the latter mediator developed in O2-driven processes (run 14).

Figure 5.

Figure 5

Open in a new tab

Kinetics of PhC(O)H (2, 4) and PhCH2OH (3, 5) accumulation upon VO(acac)2-catalyzed toluene oxidation with H2O2 (1—the sum of PhC(O)H + PhCH2OH yield obtained without promoter; 2, 3—in the presence of oxalic acid; 4, 5—oxalic acid + NHPI). Oxidation conditions: PhCH3 = H2O2 (1.4 M), VO(acac)2 (0.06 mM), oxalic acid (15 mM), NHPI (6 mM), MeCN, 40 °C, 1 atm. Inset: the kinetics of PhC(O)H/total yield ratios changes in the presence of OxalH (1) and OxalH + NHPI (2).

Figure 6.

Figure 6

Open in a new tab

Kinetics of PhC(O)H (2, 4) and PhCH2OH (3, 5) accumulation upon VO(acac)2 (2, 3)- and Co(acac)2 (4, 5)-catalyzed toluene oxidation by O2 promoted by NHPI (1—the sum of PhC(O)H + PhCH2OH obtained in the absence of promoter). Oxidation conditions: PhCH3 (1.41 M), VO(acac)2 = Co(acac)2 = 0.3 mM, O2 bubbling (100 mL min–1), total volume 50 mL, MeCN, 100 °C, 10 atm, stainless steel reactor. Inset: the kinetics of PhC(O)H/total yield ratio obtained in the presence of VO(acac)2 (1) and Co(acac)2 (2).

Switching to Co(acac)2 catalyst leads to a further substantial increase (by ∼2.5 times) in the yield of the O2-piloted process promoted with NHPI (Figure 6 and inset). A similar, although not so prominent, character demonstrates this catalytic system when applied to C6H12-targeted oxidation (the previous section). Such behavior may be a consequence of much higher redox potential E0 of Co(II)/Co(III) (1.72 V) compared to the one intrinsic of V(IV)/V(V) (1.00 V).54 On the other hand, the metal center of Co(acac)2 can donate electron density of five d-electrons through π backdonation to the empty p orbitals of O2, which strengthens the Mn+―O2 σ bond. For example, in Co(acacen)―O2 ethylenediamine/acetylacetone metallocomplex (analogue of Co(acac)2),55 the oxygen molecule is attached in a bent way with the cobalt atom. The Co―O and O–O distances have been taken as 1.86 and 1.26 Å, respectively, while the Co―O–O angle has been taken as 126°.56 Assuming a bent bonding of dioxygen, the authors suppose only a very weak interaction of x type, corresponding to a weak spin coupling and a negligible charge transfer in the bonding region. A total spin density of 1.36 electrons is localized on the oxygen moiety, while a spin density of −0.397 is localized on the dz2 orbital of cobalt. Despite such a situation leading to rather very weak bonding interaction between the cobalt cation and attached dioxygen, one can assume that the Co+2―O2 bond is still much stronger compared to the hypothetic VO+2―O2 bond. The matter is that the electron at d1 (VO+2) metal-oxo fragment of VO(acac)2 (metal d orbitals in an octahedral field) fills a nonbonding orbital, which destabilizes the complex. As a consequence, the probability of VO+2―O2 bond formation is much lower than that of the Co+2―O2 bond. Altogether, the higher E0 of Mn+/Mn+1 pair and spin density localized on the oxygen moiety and the longer (compared to O2) O–O distance intrinsic of the Co+2–O―O adduct weaken the O–O BDE. As a result, Co+2–O• species become more capable of abstracting hydrogen from NHPI than putative VO+2–O• ones.17 Ultimately, more PINO radicals generated lead to the additional enhancement of the yield (Scheme 1). The revealed facts also correlate with the literature data concerning the substrate-steered process mechanism.57

Another peculiarity of toluene oxidation (which in fact may be extrapolated on the different hydrocarbons) was the significant increase of the activities of tested catalysts when taken in combination with each other (runs 9–11, Table 2). In the presented data (preliminary for the moment), the notably lower load of VO(acac)2 + Co(acac)2 mixtures is able to afford the same product yields at 4 times higher concentration of the same catalysts taken solely (compare runs 8 and 11, Table 2). The revealed synergetic effect of Co(acac)2 + VO(acac)2 mixtures in comparison to that the individual components may be the result of mutual enhancing of the catalytic activity of each component. To date, we admit that the originated Co+2–O2 moiety of such a binary catalyst is responsible for the initiation step by means of H-atom abstraction from NHPI, whereas VO+2 cations may participate in the forthcoming stages.

The results collected in Table 3 finalize the undertaken studies by the influence of tested catalysts and promoter on the kinetics of C6H12 and PhCH3 oxidation. Given these data, one can conclude that the assistance of a promoter is essential and its impact on the oxidation of substrates (e.g., process rate and catalyst turnover frequency, TOF) depends on the nature of each constituent of the reaction mixture—catalyst cation, promoter, and oxidant. The most beneficial for the process rate and TOF was carrying out the oxidation in the presence of Co(acac)2+NHPI upon O2-bubbling. For example, at such conditions, observed that the rate (kobs) and TOF were almost 5 times higher compared to those obtained when VO(acac)2 was used (compare pairs of columns 5, 8 and particularly 11, 12). The calculated parameters depended also on the substrate taken for oxidation. Juxtaposing column 3 with 11 and 8 with 12 discloses about 5–10 times of multiplication for kobs and TOF when PhCH3 was taken instead of C6H12. The revealed feature may be referred to the higher BDE of C–H in methylene groups of C6H12 (100 kcal mol–1) compared to C–H of the PhCH3 methyl group (95 kcal mol–1). In most cases, TOF values correlate well with the change of kobs. Some discrepancy in the behavior of these two parameters was noted when using C6H12 (columns 3, 4) and PhCH3 (columns 9, 10). The matter is that due to the assumption, the growth of TOF should be proportional to kobs. For example, increasing kobs by n times should result in roughly the same proportion of TOF increasing (compare e.g., column 8 with 12). Whereas for the spotted pairs, the deviation is far from proportionality and can thus be referred to the different mechanism by which the products originate in the course of C6H12 and PhCH3 oxidation (Schemes 2 and 3). Both substrates were oxidized with the second-order rate constants (k2), which, according to Figure S4, were 0.0026 and 0.0055 M–1 s–1 for C6H12 and PhCH3, respectively. All products (C6H11OH, C6H10(O), C6H11OOH, PhCH2OH, and PhC(O)H) were detected simultaneously at the very beginning period of oxidation. Then, to date, we can admit that all of that may originate parallelly. Nevertheless, C6H11OOH seems to be the precursor for C6H11OH and C6H10(O). (A more thorough investigation concerning the manner of oxidant, promoter, and catalysts possibly affecting the kinetics and process mechanism is being planning.)

Table 3. Impact of Oxidation Conditions on the Observed Process Rate (kobs) and Catalyst Turnover Frequency (TOF) of C6H12 and PHCH3 Oxidationa.

  C6H12
 PhCH3
  H2O2 O2
 H2O2
 O2
  VO(acac)2 VO(acac)2
 Co(acac)2
 VO(acac)2
 VO(acac)2 Co(acac)2
parameter OxalH OxalH OxalH + NHPI NHPI OxalH OxalH + NHPI NHPI OxalH OxalH + NHPI NHPI NHPI
1 2 3 4 5 6 7 8 9 10 11 12
kobs × 103 (s–1) 10 1 1.5 2.5 0.1 0.3 3 2.2 1.3 7 37
TOF (h–1) 58 7 17 30 2 12 37 133 75 83 440
Open in a new tab
a

In the absence of either OxalH or NHPI, only traces of products have been detected regardless of the oxidant, catalyst, and substrate used. Similarly, no products were detected as promoted by NHPI oxidation of C6H12 and PhCH3 by H2O2 regardless of the nature of catalyst cations. TOF was calculated as the yield of products/catalyst per hour, mM/(mM·h).

2.3. Study of the Impact of NHPI on the Current vs Voltage and UV–Vis Dependencies

The involvement of the cyclic voltammetry (CV) technique for studying the redox properties of the reaction mixture can afford valuable information about the process mechanism, particularly of redox moiety.58 For that reason, the CV curves of solutions consisting of VO(acac)2, NHPI, and VO(acac)2 + NHPI have been drawn before and after supplementing it with H2O2. (The influence of H2O2 on the redox properties of the systems composed of VO(acac)2 and VO(acac)2 + OxalH has been reported previously25). Based on the acquired results, one can rationale that a combination of VO(acac)2 with NHPI notably modifies the voltammogram of initial VO(acac)2 (Figure 7). First of all, the forward anodic scan produces an oxidation peak at 0.98 V (assigned to the oxidation of VO2+),25 which is being diminished consecutively with the one elevated at 1.590 V (compare curves 1 and 3 at Figure 7). This strong oxidation signal becomes quasi-reversible to the reduction one (1.235 V), which appeared at the reverse CV of the VO(acac)2 + NHPI mixture. The presence of similarly shaped oxidation (1.540 V) and reduction (1.275 V) (curve 2) peaks also observed at the CV curve of solely NHPI (curve 3) points out that the last compound is responsible for these peaks (1.590 V, oxidation and 1.235 V, reduction). The continuation the reverse scans led to irreversible minute reduction signals at −0.820, −1.270, and −1.320 V, which are visible in the CV curves of both NHPI and NHPI + VO(acac)2 solutions (curves 2 and 3, respectively). (The irreversible reduction signals were found in the CV curve of VO(acac)2 too, but at −1.00 and −1.755 V, respectively). Further proceeding the reverse scans resulted in reduction peaks appearing at −1.795 V (curve 2) and −1.785 V (curve 3) quasi-reversible to the oxidation ones arising at −1.690 V and −1.685 V, respectively. The peak at −1.760 V emerges during the reverse scan of VO(acac)2 and possibly corresponds to the reduction of acac ligands (curve 1).51 The last one plausibly is buried in the much strong one arising at −1.795 V and attributed to NHPI (curve 2). The peaks corresponding to the oxidation of V+4 (0.98 V) and reduction of acac (−1.760 V) still remained detectable as in the case of VO(acac)2 + NHPI (compare curves 1 and 3 of Figure 6). Moreover, the reduction signals of V+5 (0.910 V) and acac (−1.760 V) were reversible to the oxidation ones developed at 0.980 V and −1.700 V, respectively (curve 2). The last observations allows us to assert the potent reduction properties of NHPI when it coupled with the transition-metal cation. In turn, it enhances the oxidation activity of such a system.

Figure 7.

Figure 7

Open in a new tab

CV of VO(acac)2 (1), NHPI (2), and NHPI + VO(acac)2 (3), anodic scan. Initial conditions: NHPI (6 mM), VO(acac)2 (0.6 mM), H2O2 (1.47 M), MeCN, 20 °C. Insets: CVs of VO(acac)2 were taken before (0) and respectively at 1, 3, 6, and 9 min after the addition of H2O2 (a); magnified inset of a (b).

Complementing the VO(acac)2 solution with H2O2 entirely diminished the oxidation peak at 0.98 V (forward anodic scan). The last fact points out the irreversible oxidation of the catalyst cation (compare curves 0 and 1 in Figure 7 insets). The reverse scan of this mixture demonstrates two reduction peaks: a very weak one at 0.520 V evidently originated due to the O2 + H+ + e– → HO2• reaction proceeding59 and the prominent second one disclosed at −1.00 V (Figure 7, inset a, curves 1–9). The last signal was almost 40 times higher with respect to the one situated at the same place and detected in the absence of H2O2 (Figure 7, inset b, curve 1). Both of them were subjected to the reduction of O2 released in the course of hydrogen peroxide catalytic cleavage. Instead, the 0.340 V oxidation peak, which in fact was absent in the CVs of initial VO(acac)2 (inset b, curve 1), appearing at the end of this reverse scan steadily disappeared within 6 min altogether with a sharp oxidation signal at −0.770 V and a prominent reduction one at −1.00 V (compare curves 0 with curves 3–9). Such a trait allowed us to conclude that the decreasing O2 concentration was a consequence of H2O2 exhausting. It may also depict that some electrode reactions would occur with released O2 (more speculation by this matter is brought below and in ref (59)).

Once the VO(acac)2 + NHPI solution was mixed with H2O2, the shapes of the resulting composition CVs were altered (compare curve 1 with curves 2–4, Figure 8) in a way fairly resembling that observed for solely NHPI-composed solution (Figure S1). Some of the signals depicting the CV of the starting VO(acac)2 + NHPI mixture (e.g., 1.235, 0.910, −1.325, −1.583, and −1.785 V reduction peaks, 0.98 V and 1.598 V oxidation peaks) vanished as H2O2 added (Figure 8), while others, e.g., the reduction (−0.128 V, −0.224 V, and 0.274 V) and oxidation (0.308, 0.548 V) signals emerge (Figure 8). The last series of peaks was absent on the CV of VO(acac)2 + H2O2. The character of curves representing the CVs of VO(acac)2 + NHPI + H2O2 mixtures and taken within first 30 min (Figure 8) also differs from the ones intrinsic to VO(acac)2 + H2O2 (inset b, Figure 7). The revealed divergence, according to the Scheme 3 prediction, may be a consequence of NHPI hydrolysis increased by the presence of H2O2 and led to o-phthalic acid.60

Figure 8.

Figure 8

Open in a new tab

CVs of VO(acac)2 + NHPI (black) and VO(acac)2 + NHPI + H2O2 taken at 1 min (red), 10 min (violet), and 30 min (pink) after addition of H2O2, anodic scan. Initial conditions: NHPI (6 mM), VO(acac)2 (0.6 mM), H2O2 (1.47 M, obtained right after its addition to the initial solution), MeCN, 20 °C.

The last suggestion obtained experimental support. The matter is that the profiles of CVs of o-phthalic acid + VO(acac)2 + H2O2 were similar to those of VO(acac)2 + NHPI + H2O2 (Figures 8 and 9). For example, the shape of the reduction (−0.128, −0.224, −0.274 V) and oxidation (0.308, 0.548 V) peaks (Figure 8) matches well with the reduction (−0.141, −0.226, −0.280 V) and oxidation (0.448, 0.770 V) peaks appearing in the CVs of the o-phthalic acid + VO(acac)2 solutions treated by H2O2 (Figure 9), despite the kinetics of the CV of NHPI + H2O2.

Figure 9.

Figure 9

Open in a new tab

CVs of VO(acac)2 + phthalic acid (black) and VO(acac)2 + phthalic acid + H2O2 taken at 0 min (black), 1 min (red), 10 min (violet), and 30 min (pink) after H2O2 addition, MeCN, anodic scan. Inset: CVs of phthalic acid (black) and VO(acac)2 + phthalic acid (red), MeCN, anodic scan. Initial conditions: phthalic acid (2 mM), VO(acac)2 (0.6 mM), H2O2 (1.47 M obtained right after its addition to the initial solution), MeCN, 20 °C.

Although the shape of CV of Figure S1 differs a bit from the phthalic acid + VO(acac)2 one, its trend remains similar.The last fact points out that in the course of NHPI oxidative hydrolysis catalyzed by VO(acac)2, o-phthalic acid is generated as an interim product and may undergo further oxidative transformations (Figure 9).

Contrary to current–voltage dependencies, the electronic spectra of the VO(acac)2 solution taken before and after the addition of NHPI exhibit rather minor changes (compare insets a and b of Figure 10). In addition, the solution of NHPI, similarly to NHPI + H2O2, was almost clear within the whole interval of visible wavelength (Figure 10, insets a and c). At the same time, supplementing the VO(acac)2 + NHPI mixture with H2O2 resulted in an almost instant evolution of a new protruding absorbance peak (ascribed to VO2+ species) with maximum at 440 nm (Figure 10). A similar character of response on the addition of H2O2 demonstrates the UV–vis spectra of solely VO(acac)2 (Figure S2), whereas the same absorbance band resulting on the addition of H2O2 to the mixture of VO(acac)2 + OxalH was 10 times slower (compare kobs of 440 nm bond evolving at Figures S2 and S3). As it was stated previously, such a feature evidently identifies decreasing the income from the free-radical counterpart in the mechanism of H2O2 decay.25

Figure 10.

Figure 10

Open in a new tab

Electronic spectra of the solution of VO(acac)2 (0.6 mM) + NHPI (6 mM) taken before (black) and after (red) supplementing it with H2O2 (15 mM), MeCN, 20 °C. The red line exhibits the spectra drawn respectively at 10, 20, 40, 60, and 120 s after H2O2 addition. Insets: spectra of VO(acac)2 solutions drawn at 10, 20, 40, 80, and 180 s after mixing it with NHPI (a); NHPI solely (b); and NHPI + H2O2 taken respectively at 10, 30, 60, 120, and 240 s after mixing (c). Different colors of lines represent the time of spectra taken.

Given these results, one can deduce that opposite to the electric current–voltage dependencies, there is rather very tiny impact of NHPI on the electronic spectra of VO(acac)2 or VO(acac)2 + H2O2. Nevertheless, it does not mean at all that such a correlation between CV and UV–vis spectra cannot be disclosed if more pedant investigations will be undertaken.

3. Conclusions

The efficiency of homogeneous cyclohexane and toluene oxidation piloted by O2 or H2O2 and catalyzed by small-molecule catalysts VO(IV)- and Co(II) acetylacetonates can be notably (30×) enhanced by the presence of relatively cheap and low-toxic oxalic acid or N-hydroxyphthalimide. The key novelty of this research is that such significant improvement (particularly in the case of the O2-piloted process) can be attained by a combination of increased pressure (10 atm) and moderate temperature (40–100 °C). Upon such conditions, the load of metal catalysts and promoters can be notably (by 10–15 times) decreased without the deterioration of the aimed product yield. The protocol achievement enables us to avoid the usage of radical initiator with no lack in the process productivity. Furthermore, the moderate reaction conditions minimize the side reactions maintaining the selectivity at a high level. As appeared, the nature of substrates as well as catalyst and promoter affects the process mechanism and product assortment. For example, no peroxide-type products were detected during toluene oxidation by O2 evincing benzyl alcohol and benzyl aldehyde as the primary ones. By contrast, in the case of cyclohexane, the process follows the traditional scheme and cyclohexylhydroperoxide was formed alongside with cyclohexanol and cyclohexanone. Oxalic acid boosts the yield of oxidation catalyzed by VO(acac)2, whereas the Co(acac)2-based process is dramatically hindered by the presence of this promoter. The last effect was subjected to the formation of insoluble (therefore eventually much less active) Co oxalate. Oxalic acid additives, similarly to H2O2, demonstrate the detrimental impact on the O2-driven process promoted by NHPI. Such an effect was rationalized by the fast decay of the latter upon acidic conditions either by radicals (originated by hydrogen peroxide catalytic cleavage) or oxalic acid dissociation. Significantly increasing the activity of the mixtures constituting cobalt and vanadyl catalysts was referred to in its participation in the different stages of the process mechanism, e.g., initiation, propagation and degenerative branching. The cyclic voltammetry study confirms that the decrease of the NHPI activity ensued from its fast hydrolysis to o-phthalic acid. On the other hand, the redox properties of the reaction medium as well as the process mechanism can be modulated by NHPI increasing the yield. Although UV–vis measurements did not exhibit a strict impact of NHPI on the electronic spectra of VO(acac)2, it does not mean that a more thorough scrutinizing of such a system cannot be surprising.

4. Experiments

4.1. Experimental Section

The commercial aqueous solution of hydrogen peroxide (35 wt %, Fluka) has been concentrated to 70 wt % by vacuum distillation at 45 °C/10 mm Hg (CAUTION: risk of explosion!) All glassware prior usage has to be thoroughly washed and rinsed with distilled or deionized water to remove traces of any heavy metals. As the purchased commercial VO(acac)2 and Co(acac)2 (Aldrich) was ca. 95%, it was additionally purified. For this aim, either VO(acac)2 or Co(acac)2 was dissolved in MeCN and both solutions heated to about 60 °C under stirring. (The weight of solutes taken should be a slightly lower than that of the saturated concentration at such temperature). The hot solutions were filtered through a micro-porous paper filter. Then, both the solutions were cooled in ice water and the blue crystals were filtered (in the case of VO(acac)2) via a No.3 sintered glass crucible or a micro-porous paper filter (in the case of Co(acac)2). The residues were washed with 2 × 5 cm3 portions of ice-cold MeCN, dried at ambient temperature and pressure overnight, and vacuumed at 20 °C for 48 h. The dark-green tiny (0.5–1.0 mm) crystals of VO(acac)2 or pale powder Co(acac)2 were used for further experiments. Cyclohexane, toluene, acetonitrile, triphenylphosphine, KI, Na2S2O3, and tetra-n-butyl ammonium perchlorate (Aldrich) were of analytical grade and used as purchased.

4.2. Characterization

Cyclic voltammetry (CV) experiments were carried at room temperature using a METROHN AUTOLAB PGSTAT model 302N. The detailed conditions for CV, UV–vis, GLC, and iodometry measurements were provided in refs (24, 25).

4.3. Activity Test

4.3.1. Oxidation of Cyclohexane and Toluene at 40 °C and 1 atm

These experiments were carried out in 25 mL round-bottom glass flasks equipped with a reflux condenser and a magnetic stirrer. In a typical experiment, cyclohexane or toluene (3.36 g, 40 mmol) was dissolved in MeCN (15 mL) containing VO(acac)2 or Co(acac)2 (0.0032 g, 0.012 mmol) and oxalic acid (0.038 g, 0.3 mmol) or NHPI (6 mM). The mixture was heated at 40 °C for 5 min before adding H2O2 (1.34 g, 40 mmol). This moment corresponds to the beginning of the oxidation reaction. Usually, the reaction was carried out for 5 h upon vigorous stirring. In the course of the process, the probes were withdrawn and analyzed by GLC and iodometry.

4.3.2. Cyclohexane and Toluene Oxidation at 100 °C and 10 atm

To determine the influence of increased temperature and pressure, the experiments were carried out at 100 °C and 10 atm of O2 in a stainless steel bubbling reactor (Scheme S1).

Acknowledgments

The authors thank the Faculty of Chemistry Rzeszow University of Technology for analytical facilities and Prof. Andrzej Sobkowiak for stimulating discussions. The authors expressed their gratitude to Prof. Volodymyr Skorohoda for experimental materials support. Technical assistance of Dariya Maksym is indebted.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.0c00447.

  • CVs of NHPI and NHPI + H2O2 (Figure S1); UV–vis spectra of VO(acac)2 and VO(acac)2 + H2O2 (Figure S2); UV–vis spectra of VO(acac)2 + OxalH and VO(acac)2 + OxalH + H2O2 (Figure S3); kobs vs [substrate] plots (Figure S4); and chart of the lab unit for substrate oxidation at medium temperature and pressure (Scheme S1) (PDF)

The authors declare no competing financial interest.

Author Status

⊥ This study was done while Sergij Tkach was working at Institutea.

Supplementary Material

ao0c00447_si_001.pdf (532KB, pdf)

References

  1. Ma B. C.; Zhang Z. X.; Song W. F.; Xue X. L.; Yu U. Z.; Zhao Z. S.; Ding Y. Solvent-free selective oxidation of C–H bonds of toluene and substituted toluene to aldehydes by vanadium-substituted polyoxometalate catalyst. J. Mol. Catal. A: Chem. 2013, 369, 152–158. 10.1016/j.molcata.2012.11.022. [DOI] [Google Scholar]
  2. Metzger J. O.; Eissen M. Concepts on the Contribution of Chemistry to a Sustainable Development—Renewable Raw Materials. C. R. Chim. 2004, 7, 569–581. 10.1016/j.crci.2003.12.003. [DOI] [Google Scholar]
  3. Luo Y.-R.Comprehensive Handbook of Chemical Bond Energies; CRC Press: Boca Raton, FL, 2007. [Google Scholar]
  4. Musser M. T.UllmannQs: Encyclopedia of Industrial Chemistry; Wiley-VCH: Weinheim, 2000. [Google Scholar]
  5. Da Silva J. A. L.; Frausto Da Silva J. J. R.; Pombeiro A. J. L. Amavadin, a vanadium natural complex: its role and applications. Coord. Chem. Rev. 2013, 15–16, 2388–2400. 10.1016/j.ccr.2013.03.010. [DOI] [Google Scholar]
  6. Kesavan L.; Tiruvalam R.; Rahim M. H. A.; Saiman M. I.; Enache D. I.; Jenkins R. L.; Dimitratos N.; Lopez-Sanchez J. A.; Taylor S. H.; Knight D. W. Solvent-free oxidation of primary carbon-hydrogen bonds in toluene using Au-Pd alloy nanoparticles. Science 2011, 331, 185–199. 10.1126/science.1198458. [DOI] [PubMed] [Google Scholar]
  7. Lv J. G.; Shen Y.; Peng L. M.; Guo X. F.; Ding W. P. Exclusively selective oxidation of toluene to benzaldehyde on ceria nanocubes by molecular oxygen. Chem. Commun. 2010, 46, 5909–5911. 10.1039/c0cc00777c. [DOI] [PubMed] [Google Scholar]
  8. Arakawa H.; Aresta M.; Armor J. N.; Barteau M. A.; Beckman E. J.; Bell A. T.; Bercaw J. E.; Creutz C.; Dinjus E.; Dixon D. A.; Domen K.; DuBois D. L.; Eckert J.; Fujita E.; Gibson D. H.; Goddard W. A.; Goodman D. W.; Keller J.; Kubas G. J.; Kung H. H.; Lyons J. E.; Manzer L. E.; Marks T. J.; Morokuma K.; Nicholas K. M.; Periana R.; Que L.; Rostrup-Nielson J.; Sachtler W. M. H.; Schmidt L. D.; Sen A.; Somorjai G. A.; Stair P. C.; Stults B. P.; Tumas W. Catalysis research of relevance to carbon management: progress, challenges, and opportunities. Chem. Rev. 2001, 101, 953–996. 10.1021/cr000018s. [DOI] [PubMed] [Google Scholar]
  9. Lenze M.; Sedinkin S. L.; Bauer E. B. Polydentate pyridyl ligands and the catalytic activity of their iron(II) complexes in oxidation reactions utilizing peroxides as the oxidants. J. Mol. Catal. A: Chem. 2013, 373, 161–171. 10.1016/j.molcata.2013.03.006. [DOI] [Google Scholar]
  10. Rinaldo D.; Philipp D. M.; Lippard S. J.; Friesner R. A. Intermediates in Dioxygen Activation by Methane Monooxygenase: A QM/MM Study. J. Am. Chem. Soc. 2007, 129, 3135–3147. 10.1021/ja0654074. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Augusto O.; Miyamoto S.. Oxygen Radicals and Related Species. In Principles of Free Radical Biomedicine; Pantopoulos K.; Schipper H. M., Eds.; Nova Science Publishers, 2011; Vol. 1, pp 19–42. [Google Scholar]
  12. Piera J.; Bäckvall J. E. Catalytic oxidation of organic substrates by molecular oxygen and hydrogen peroxide by multistep electron transfer—a biomimetic approach. Angew. Chem., Int. Ed. 2008, 47, 3506–3523. 10.1002/anie.200700604. [DOI] [PubMed] [Google Scholar]
  13. Zhou B.; Chen W. A Mild Catalytic Oxidation System: FePcOTf/H2O2 Applied for Cyclohexene Dihydroxylation. Molecules 2015, 20, 8429–8439. 10.3390/molecules20058429. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Murahashi S.-I.; Zhou X.-G.; Komiya N. Chlorinated phthalocyanine iron(II) complex catalyzed oxidation of alkanes and alkenes with molecular oxygen in the presence of acetaldehyde. Synlett 2003, 321–324. 10.1055/s-2003-37128. [DOI] [Google Scholar]
  15. Piera J.; Bäckvall J.-E. Catalytic Oxidation of Organic Substrates by Molecular Oxygen and Hydrogen Peroxide by Multistep Electron Transfer—A Biomimetic Approach. Angew. Chem., Int. Ed. 2008, 47, 3506–3523. 10.1002/anie.200700604. [DOI] [PubMed] [Google Scholar]
  16. Berkessel A. Diversity-Based Approaches to Selective Biomimetic Oxidation Catalysis. Adv. Inorg. Chem. 2006, 58, 1–28. [Google Scholar]
  17. Ishii Y.; Sakaguchi S.; Iwahama T. Manganese Dioxide and N-Hydroxyphthalimide. An Effective Catalytic System for Oxidation of Nitrotoluenes with Molecular Oxygen. Adv. Synth. Catal. 2001, 343, 393–427. . [DOI] [Google Scholar]
  18. Coseri S. Phthalimide-N-oxyl (PINO) Radical, a Powerful Catalytic Agent: Its Generation and Versatility Towards Various Organic. Catal. Rev. 2009, 51, 218–292. 10.1080/01614940902743841. [DOI] [Google Scholar]
  19. Mac Leod T. C. O.; Kirillova M. V.; Pombeiro A. J. L.; Schiavon M. A. Mild oxidation of alkanes and toluene by tert-butylhydroperoxide catalyzed by an homogeneous and immobilized Mn (salen) complex. Appl. Catal., A 2010, 372, 191–198. 10.1016/j.apcata.2009.10.035. [DOI] [Google Scholar]
  20. Lisovska M.; Timokhin V.; Pokutsa A.; Kopylets V. Cyclohexane oxidation catalyzed by variable-valence metal compounds in the presence of propionic aldehyde. Kinet. Catal. 2001, 41, 223–232. 10.1007/BF02771421. [DOI] [Google Scholar]
  21. Ideas in Chemistry and Molecular Sciences: Advances in Synthetic Chemistry; Pignataro B., Ed.; Wiley-VCH Verlag GmbH & Co. KGaA: Weinheim, 2010; p 332. [Google Scholar]
  22. Brégeault J.-M. Transition-metal complexes for liquid-phase catalytic oxidation: some aspects of industrial reactions and of emerging technologies. Dedicated to Professor William P. Griffith for his innovative and imaginative science. Dalton Trans. 2003, 3289–3302. 10.1039/B303073N. [DOI] [Google Scholar]
  23. Punta C.; Gambarotti C.. N-Hydroxy Derivatives: Key Organocatalysts for the Selective Free Radical Aerobic Oxidation of Organic Compounds. In Ideas in Chemistry and Molecular Sciences: Advances in Synthetic Chemistry; Pignataro B., Ed.; Wiley-VCH Verlag GmbH & Co. KGaA: Weinheim, Germany, 2010; pp 3–24. [Google Scholar]
  24. Pokutsa A.; Fliunt O.; Kubaj Y.; Paczesniak T.; Blonarz P.; Prystanskiy R.; Muzart J.; Makitra R.; Zaborovskyi A.; Sobkowiak A. Relationships between the efficiency of cyclohexane oxidation and the electrochemical parameters of the reaction solution. J. Mol. Catal. A: Chem. 2011, 347, 15–21. 10.1016/j.molcata.2011.07.003. [DOI] [Google Scholar]
  25. Pokutsa A.; Kubaj Y.; Zaborovskyi A.; Maksym D.; Paczesniak T.; Mysliwiec B.; Bidzinska E.; Muzart J.; Sobkowiak A. V(IV)-catalyzed cyclohexane oxygenation promoted by oxalic acid: Mechanistic study. Mol. Catal. 2017, 434, 194–205. 10.1016/j.mcat.2017.02.013. [DOI] [Google Scholar]
  26. Petroselli M.; Franchi P.; Lucarini M.; Punta C.; Melone L. Aerobic Oxidation of Alkylaromatics using a Lipophilic N-Hydroxyphthalimide: Overcoming the Industrial Limit of Catalyst Solubility. ChemSusChem 2014, 7, 2695–2703. 10.1002/cssc.201402132. [DOI] [PubMed] [Google Scholar]
  27. Bacsa J.; Eve D.; Dunbar K. R. catena-Poly[[diaquacobalt(II)]-μ-oxalato]. Acta Crystallogr., Sect. C: Cryst. Struct. Commun.. 2005, 61, 58–60. 10.1107/S0108270104030409. [DOI] [PubMed] [Google Scholar]
  28. Sutradhar M.; Shvydkiy N. V.; Guedes da Silva M. C.; Kirillova M. V.; Kozlov Y. N.; Pombeiro A. J. L.; Shul’pin G. B. A new binuclear oxovanadium (V) complex as a catalyst in combination with pyrazinecarboxylic acid (PCA) for efficient alkane oxygenation by H2O2. Dalton Trans. 2013, 42, 11791–11803. 10.1039/c3dt50584g. [DOI] [PubMed] [Google Scholar]
  29. Kwan W. P.; Voelker B. M. Rates of hydroxyl radical generation and organic compound oxidation in mineral-catalyzed Fenton-like systems. Environ. Sci. Technol. 2002, 36, 1467–1476. 10.1021/es011109p. [DOI] [PubMed] [Google Scholar]
  30. Tesla B.; Mayer J. M.. Hydrolysis in Drug and Prodrug Metabolism; Wiley VCH: Zürich, 2003. [Google Scholar]
  31. Ishii Y.; Sakaguchi S. TCIMAIL 2003, 172, 2–17. [Google Scholar]
  32. Blanksby S. J.; Ellison G. B. Bond dissociation energies of organic molecules. Acc. Chem. Res. 2003, 36, 255–263. 10.1021/ar020230d. [DOI] [PubMed] [Google Scholar]
  33. Bauer L.; Miarka S. V. The Chemistry of Hydroxylamines, Oximes and Hydroxamic Acids. J. Am. Chem. Soc. 1957, 79, 1983. 10.1021/ja01565a061. [DOI] [Google Scholar]
  34. Wood G. P. F.; Easton C. J.; Rauk A.; Davies M. J.; Radom L. Effect of Side Chains on Competing Pathways for β-Scission Reactions of Peptide-Backbone Alkoxyl Radicals. J. Phys. Chem. A 2006, 110, 10316–10323. 10.1021/jp062916j. [DOI] [PubMed] [Google Scholar]
  35. Fenton H. J. H. On a new reaction of tartaric acid. Chem. News 1876, 33, 190. [Google Scholar]
  36. Haber F.; Weiss J. On the Catalysis of Hydroperoxide. Naturwissenschaften 1932, 51, 948–950. 10.1007/BF01504715. [DOI] [Google Scholar]
  37. Hill J. W.; Kolb D. K.. Chemistry for Changing Times, 9th ed.; Prentice Hall, 2001. [Google Scholar]
  38. Amorati R.; Lucarini M.; Mugnaini V.; Pedulli G. F.; Minisci F.; Recupero F.; Fontana F.; Astolfi P.; Greci L. Aerobic oxidation of N-alkylamides catalyzed by N-hydroxyphthalimide under mild conditions. Polar and enthalpic effects. J. Org. Chem. 2002, 67, 2671–2676. 10.1021/jo016398e. [DOI] [PubMed] [Google Scholar]
  39. Hess W. T.Hydrogen Peroxide, in Kirk-Othmer Encyclopedia of Chemical Technology, 4th ed.; Wiley: New York, 1995; Vol. 13, pp 961–995. [Google Scholar]
  40. Pegis M. L.; Roberts J. A. S.; Wasylenko D. J.; Mader E. A.; Appel A. M.; Mayer J. M. Standard Reduction Potentials for Oxygen and Carbon Dioxide Couples in Acetonitrile and N,N-Dimethylformamide. Inorg. Chem. 2015, 54, 11883–11888. 10.1021/acs.inorgchem.5b02136. [DOI] [PubMed] [Google Scholar]
  41. Housecroft C. E.; Sharpe A. G.. Inorganic Chemistry, 2nd ed.; Pearson Prentice-Hall, 2005. [Google Scholar]
  42. Tian Z.; Fattahi A.; Lis L.; Kass S. R. Cycloalkane and cycloalkene C–H bond dissociation energies. J. Am. Chem. Soc. 2006, 128, 17087–17092. 10.1021/ja065348u. [DOI] [PubMed] [Google Scholar]
  43. Denisov E. T.Liquid-Phase Reaction Rate Constants; Plenum: NY, 1974. [Google Scholar]
  44. Russell G. A. Deuterium-isotope effects in the autoxidation of aralkyl hydrocarbons. Mechanism of the interaction of peroxy radicals. J. Am. Chem. Soc. 1957, 79, 3871. 10.1021/ja01571a068. [DOI] [Google Scholar]
  45. Burko L. M.; Price R. H. Ballistic trajectory: Parabola, ellipse, or what?. Am. J. Phys. 2005, 73, 516. 10.1119/1.1866097. [DOI] [Google Scholar]
  46. Srinivasan V.; Rocek J. Formation of a longlived chromium(V) intermediate in the chromic acid oxidation of oxalic acid. J. Am. Chem. Soc. 1974, 96, 127–133. 10.1021/ja00808a021. [DOI] [Google Scholar]
  47. Lee R.; Gryn’ova G.; Ingold K. U.; Coote M. L. Why are sec-alkylperoxyl bimolecular self-reactions orders of magnitude faster than the analogous reactions of tert-alkylperoxyls? The unanticipated role of CH hydrogen bond donation. Phys. Chem. Chem. Phys. 2016, 18, 23673–23679. 10.1039/C6CP04670C. [DOI] [PubMed] [Google Scholar]
  48. Kesavan L.; Tiruvalam R.; Hasbi Ab Rahim M.; Izham bin Saiman M.; Enache D. I.; Jenkins R. L.; Dimitratos N.; Lopez-Sanchez J. A.; Taylor S. H.; Knight D. W.; Kiely C. J.; Hutchings G. J. Solvent-free oxidation of primary carbon-hydrogen bonds in toluene using Au-Pd alloy nanoparticles. Science 2011, 331, 195–199. 10.1126/science.1198458. [DOI] [PubMed] [Google Scholar]
  49. Du B.; Kim S.-I.; Lou L.-L.; Jia A.; Liu G.; Qi B.; Liu S. A simple and efficient zeolite catalyst for toluene oxidation in aqueous media. Appl. Catal., A 2012, 425–426, 191–198. 10.1016/j.apcata.2012.03.021. [DOI] [Google Scholar]
  50. Garcia-Bosch I.; Siegler M. A. Copper-Catalyzed Oxidation of Alkanes with H2O2 under a Fenton-like Regime. Angew. Chem., Int. Ed. 2016, 55, 12873–12876. 10.1002/anie.201607216. [DOI] [PubMed] [Google Scholar]
  51. Todre Z. V.Ion-Radical Organic Chemistry: Principles and Applications, 2nd ed.; CRC Press: Boca Raton, 2009. [Google Scholar]
  52. Bordwell F. G.; Algrim D.; Vanier N. R. Acidities of anilines and toluenes. J. Org. Chem. 1977, 42, 1817–1819. 10.1021/jo00430a039. [DOI] [Google Scholar]
  53. Bordwell F. G.; Drucker G. E.; Fried H. E. Acidities of carbon and nitrogen acids: the aromaticity of the cyclopentadienyl anion. J. Org. Chem. 1981, 46, 632–635. 10.1021/jo00316a032. [DOI] [Google Scholar]
  54. Brown W. H.; Iverson B. L.; Anslyn E.; Foote C. S.. Organic Chemistry; 7th ed.; Brooks/Cole Cengage Learning: U.K., 2014. [Google Scholar]
  55. Fantucci P.; Valenti V. Molecular Orbital Study of a Cobalt Reversible Oxygen Carrier. J. Am. Chem. Soc. 1976, 98, 3832–3838. 10.1021/ja00429a016. [DOI] [Google Scholar]
  56. Rodley G. A.; Robinson W. T. Structure of a monomeric oxygen-carrying complex. Nature 1972, 235, 438–439. 10.1038/235438a0. [DOI] [PubMed] [Google Scholar]
  57. Xue G.; Pokutsa A.; Que L. Jr. Substrate-Triggered Activation of a Synthetic [Fe2(μ-O)2] Diamond Core for C–H Bond Cleavage. J. Am. Chem. Soc. 2011, 133, 16657–16667. 10.1021/ja207131g. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Sawyer D. T.; Sobkowiak A.; Roberts J. R. Jr.. Electrochemistry for Chemists, Wiley: New York, 1995. [Google Scholar]
  59. Harris D. C.Quantitative Chemical Analysis, 7th ed.; W.H. Freeman & Co: New York, 2007. [Google Scholar]
  60. Selvakumar M. S.; Xavier N.; Jeyabal V. Electroanalytical studies on ethoxylation of phthalic acid. Der Chem. Sin. 2014, 5, 34–37. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ao0c00447_si_001.pdf (532KB, pdf)

Articles from ACS Omega are provided here courtesy of American Chemical Society

RESOURCES