Iron(II) and Manganese(II) Complexes with N4Py as Dioxygen Activators for α-Pinene Oxidation in Acetonitrile

Abstract

Iron(II) and manganese(II) complexes with N4Py [N4Py—N,N-bis(2-pyridylmethyl)-N-(bis-2-pyridylmethyl)amine] have been found to activate O₂ for the oxidation of α-pinene in acetonitrile. For example, for 1 M α-pinene, 0.5 mM [(N4Py)Fe^II]²⁺, and dioxygen as an oxidant, 90 mM α-pinene epoxide, 48 mM verbenol, and 50 mM verbenone have been formed, which, taking into account the concentrations of the minor products (myrtenol and myrtenal), gives a turnover number approximately equal to 400. Based on the amounts of products formed, the conversion of α-pinene is approximately 20% and 18% for iron and manganese catalysts, respectively. Although the manganese catalyst is somewhat less effective than the iron catalyst, the selectivity of the products is similar for both catalysts. Replacement of dioxygen with air as the oxidant causes the reaction yield to be lower. The proposed mechanism assumes the formation of a metal(IV)-oxo complex [(N4Py)M^IV=O]²⁺, M–Fe or Mn, during the simultaneous combination of a catalyst, O₂, and substrate, and its subsequent reactions toward the observed products.

1. Introduction

α-Pinene is one of the cheapest and most abundant biorenewable terpenes available in nature. It is obtained mainly in the distillation process from paper pulp and turpentine oil acquired from coniferous trees [1,2]. More than 400 essential oils contain α-pinene, and it can be extracted directly from these sources [3]. α-Pinene is a monoterpene (C₁₀H₁₆), which contains two isoprene units assembled into a structure consisting of a four-membered ring bridgehead connecting C-1 and C-5 of the cyclohexyl ring. The structure shows two chiral centers at positions C-1 and C-5. Due to the chiral centers, α-pinene has two major enantiomers: (−)-α-pinene and (+)-α-pinene (Scheme 1) [4].

Scheme 1.

Enantiomers of α-pinene: (a) (−)α-pinene; (b) (+)α-pinene.

The availability and relatively low price of α-pinene make it an important raw material for many sustainable processes. α-Pinene (and other terpenes) can be converted into biofuels [5,6] and bio-based polymers [2,4,7,8,9,10,11]. It is also used as a precursor in the synthesis of chiral ligands [4,12,13,14]. Recent reviews present detailed information on biotransformations [3,15] and chemical transformations [16,17,18] of α-pinene like isomerization, dehydroisomerization, hydration, and oxidation. These processes can lead to many valuable fine chemicals that are used in the pharmaceutical as well as in the flavor and fragrance industries.

Epoxidation and oxidation of α-pinene are the most relevant from the perspective of this paper. Epoxide is obtained by the oxidation of C=C bonds, but the process is usually accompanied by allylic oxidation [1,4,16,17,18]. The main products of these processes are usually α-pinene epoxide, verbenol, and verbenone; however, a mixture of other products is often formed due to the occurrence of follow-up processes. As oxidants, hydrogen peroxide (HOOH) and tert-butyl hydroperoxide (t-BuOOH) were mostly used. The use of mesoporous metal silicates as catalysts shows that t-BuOOH is a more stable and efficient oxidant, because the catalytic decomposition of HOOH to O₂ and water can easily occur [19], and the presence of O₂ during the oxidation carried out with HOOH does not always increase the reaction yield [20]. However, for environmental and economic reasons, O₂ is the most desirable oxidant, having the highest active oxygen content, the lowest cost, and producing essentially no waste products. From the perspective of this article, the oxidation of α-pinene by O₂ catalyzed by metal complexes under homogeneous conditions is the most relevant.

In early research on the use of O₂ for the oxidation of α-pinene, Co(II)-pyridine complexes were used as catalysts [21]. A seven-day experiment with continuous flow of O₂ by mixture of α-pinene and catalyst [Co(4-Me-py)₂Br₂] at 72 °C produced 76% verbenone. In subsequent articles [22,23,24], in addition to Co(II)-pyridine complexes, simple Co(II) salts were used as catalysts, and the effect of small amounts of t-BuOOH was investigated. On a shorter time scale of the experiment, up to 24 h, the formation of α-pinene epoxide, verbenol, and verbenone was reported. Cobalt acetate Co(OAc)₂ with the addition of NaBr was also used as a catalyst [25]. The process was carried out in a water matrix at an elevated temperature of 50 °C with continuous flow of O₂ within 6 or 24 h. Verbenone, myrtanal, and trans-verbenyl acetate were the main products.

One of the first “true” homogeneous systems consisted of Co(II) Schiff’s base complexes and 2-methylpropanal, which were used to activate of O₂ in acetonitrile (MeCN) [26]. α-Pinene epoxide was formed as the main product (70%) when N,N′-bis[(pyridin-2-yl)methylidene]ethane-1,2-diamine (see SI for structure, Figure S1a) was used as the ligand, whereas a mixture of allylic alcohols, myrtenol, and pinocarveol was produced primarily when the ligand based on 2-methanimidoylphenol (Figure S1b) was applied. In the last case, about 40% of unidentified products were also present in the reaction mixture. Transition metal acetyloacetonates used for the oxidation of α-pinene with O₂ in the presence of isobutyraldehyde in 1,2-dichloroethane show completely different behavior [27]. Whereas the Fe(III), Mn(III), and Ni(II) complexes cause complete conversion of the substrate with approximately 90% selectivity toward α-pinene epoxide, in the presence of the Cu(II), Cr(III), and Rh(III) complexes, no reaction occurs under the same experimental conditions.

Glacial acetic acid has also been used as a solvent for oxidation [28,29]. Experiments have been carried out at elevated temperatures of 60 or 80 °C using as catalysts either the combination of PdCl₂, CuCl₂, and LiCl [28] or CoCl₂ [29]. However, the selectivity of the processes is poor; a relatively large number of acetylated and chlorinated products were formed. It is interesting that the first catalysts yielded a relatively large amount of limonene (between 14 and 19% depending on the experimental conditions); the other main products were α-terpinyl acetate (Figure S2a) (21–30%) and bornyl chloride (Figure S2b) (13–29%). The main products for the reaction catalyzed by CoCl₂ are campholene aldehyde (Figure S2c), verbenone, and verbenyl acetate. When MeCN was replaced by glacial acetic acid as a solvent, α-pinene epoxide, verbenone, and verbenol are formed as the main products in almost the same amounts [29].

Fe(III)-catalyzed aerobic oxidation of α-pinene in methanol (MeOH) gives myrtenol methyl ether as the main product [30]. This product was accompanied by mixtures of α- and γ-terpinenes and α- and γ-terpinyl methyl ethers (Figure S2d–g), both in almost equimolar amounts. In a recent report [31], Cu(II) complexes with pyridine dicarboxylates have been used as catalysts for the oxidation of pure α-pinene (without solvent) by O₂ with small amounts of t-BuOOH. Under optimized conditions, the conversion of α-pinene reached approximately 90%, producing α-pinene epoxide, verbenyl hydroperoxide, verbenol, and verbenone as the main products.

An interesting problem is the comparison of the results obtained by homogeneous and heterogeneous catalysis. The results obtained for several transition metal complexes with salen-type ligand based on cyclohexanediamine (Figure S1c) have shown that there are no substantial differences in the yields of α-pinene epoxide formation for homogeneous and immobilized complexes [32]. The other report [33] has indicated that for Co(II) and Ru(III) complexes with N,N′-bis(salicylidene)-1,2-phenylenediamine (Saloph) (Figure S1d) and its derivatives, the yields obtained for the zeolite-encapsulated complexes are almost twice as much as for these complexes used in homogeneous conditions. However, in both reports, the experimental conditions are not described in detail, which makes the interpretation difficult. In analogous research, the conversions and yields of α-pinene oxidation process with air at 130 °C have been compared for Mn(III) acetate and modified with this salt metal–organic framework (MOF) based on MIL-53NH₂(Al) [34]. Both the Mn-containing MOF catalyst and Mn(III) acetate have shown similar activity in a mixture of diethyl carbonate and dimethylformamide as the solvent. In both cases, a conversion of 31% and the formation of 17% α-pinene epoxide (55% selectivity) were observed after 6 h. The heterogeneous catalyst had not lost its catalytic activity while being reused for at least five cycles [34]. Finally, based on O₂ uptake, the initial reaction rate of α-pinene oxidation has been measured primarily for Co(II) complexes with N,N′-bis(salicylidene)ethylenediamine (Salen), acetyloacetonate (acac) and 2-methylpyridine as catalysts in the homogeneous system and for polymer-supported ammonium heptamolybdate [(NH₄)₆Mo₇O₂₄] and CoCl₂ in the heterogeneous system [35]. The products of α-pinene oxidation have not been determined. Generally, the measured initial reaction rates for the homogeneous catalysts were an order of magnitude higher than those for the heterogeneous catalysts.

Heterogeneous catalysts have been used much more frequently for the oxidation of α-pinene by O₂. Since the present research refers to homogeneous catalysis, the experimental conditions and major products for the known transformations performed in heterogeneous systems only have been summarized in Table 1.

Table 1.

Conditions of α-pinene oxidations by dioxygen catalyzed by heterogeneous catalysts.

No.	Catalysts	Oxidants	Solvents	Main Oxidations Products	Ref.
1.	Titanium-silicate catalysts	O2 (85 °C)	w/o solvent	α-pinene epoxide, verbenol, verbenone	[36]
2.	Titanium-silicalite TS-1 catalysts	O2 (80 °C)	w/o solvent	α-pinene epoxide, verbenol, verbenone	[37]
3.	Titanium-silicate catalysts: Ti-SBA-15- and Ti-SBA-15-modified orange peels	O2 (80, 90, 100, 110, 120, 130 °C)	w/o solvent	α-pinene epoxide, verbenol, verbenone	[38]
4.	Titanium silicate catalysts with tailored pore architectures (MTS-1, HTS-1)	O2 (110, 115, 120, 125, 130, 135, 140 °C)	w/o solvent	α-pinene epoxide, verbenol, verbenone	[39]
5.	Polymer-supported Co(II) (polymers: polyaniline, poly-o-toluidine, poly-o-anisidine)	O2 (65 °C)	MeCN	α-pinene epoxide, verbenone	[40]
6.	Cr-MCM-41 [mesoporous molecular sieves containing Cr(III)]	O2 (60 °C)	w/o solvent	α-pinene epoxide, verbenol, verbenone	[41]
7.	Dioxo-Mo(VI) complex with an aminothiazole ligand on TiO2 NT	O2 (RT)/UV–Vis	MeCN, ethanol, chloroform, dichloromethane	α-pinene epoxide (verbenol, verbenone)	[42]
8.	Mo(VI)Cl2O2Bipy/TiO2 NT	O2 (RT)/UV–Vis	MeCN	α-pinene epoxide (verbenol, verbenone)	[43]
9.	Mo(VI)Cl2O2Bipy/TiO2 NT	O2 (19 °C)/UV–Vis	MeCN	α-pinene epoxide (verbenone, verbenol)	[44]
10.	Metal-free catalysts like activated carbon nanotubes (CNTs) and nitrogen-doped NCNTs	O2 (80 °C)	MeCN	α-pinene epoxide, verbenol, verbenone, verbenyl hydroperoxide	[45]
11.	Co-POM/MIL-101 (Co- monosubstituted Keggin heteropolyanions bound to the chromium terephthalate polymer matrix)	O2 (50 °C)	MeCN	verbenol, verbenone	[46]
12.	Fe-MIL-101, Cr-MIL-101	O2 (40–60 °C)/initiator (TBHP)	w/o solvent	verbenol, verbenone (α-pinene epoxide, verbenyl hydroperoxide, camphelene aldehyde)	[47]
13.	Fe- and Cr-based MOFs, MIL-100 and MIL-101	O2 (40–60 °C)/initiator (TBHP)	w/o solvent	verbenol, verbenone (α-pinene epoxide, verbenyl hydroperoxide)	[48]
14.	Cu-MOF-2	O2 (40 °C)	MeCN	α-pinene epoxide	[49]
15.	Co-MOF	air (90 °C)/initiator (TBHP, CHP)	DMF	α-pinene epoxide	[50]
16.	Co(III)-HMS (HMS—hexagonal mesoporous silica)	air (60, 80, 100 °C)	dioxane	α-pinene epoxide, verbenol, verbenone	[51]
17.	Mesoporous Co-based nanocatalyst (Cox/N−C)	O2 (85 °C)	w/o solvent	verbenone (α-pinene epoxide, verbenol)	[52]
18.	Co(II) cation-exchanged zeolite Y-based	O2 (80, 100 °C)	DMF	α-pinene epoxide (verbenol, verbenone)	[53]
19.	PrAlPO-5 (bi-functional molecular sieves)	air (30, 50, 70 °C)	chloroform, dichloroethane, MeCN, acetone, dimethylacetamide, n-hexane, cyclohexane	α-pinene epoxide (campholenic aldehyde, verbenone)	[54]
20.	Mn(III)/Co(III)-containing molecular sieves: CoAlPO-36, CoAlPO-5, MnAlPO-36, MnAlPO-5	dry air (50 °C)	α-pinene: benzaldehyde mole ratio 1:3	α-pinene epoxide (verbenol, verbenone)	[55]
21.	Co–Fe3O4, Mn–Fe3O4 (Co- and Mn-substituted ferrites)	O2 (60 °C)	w/o solvent	verbenone, verbenol, α-pinene epoxide	[56]
22.	Co/SiO2	O2 (60 °C)	w/o solvent	verbenol, verbenone (α-pinene epoxide)	[57]
23.	Fe(III)/SiO2	O2 (58, 78, 84 °C)	MeCN	verbenol, verbenone, α-pinene epoxide	[58]
24.	Mn(III)–salen-sulphonato complex intercalated into a Zn(II)–Al(III) layered double hydroxide (LDH)	O2 (25 °C)/initiator (pivalaldehyde)	toluene	α-pinene epoxide	[59]
25.	Metal (Fe/Cr/V/Co/Mn)–salen-like zeolite	O2 (RT)/initiator (pivalaldehyde)	fluorobenzene	α-pinene epoxide	[60]
26.	[Co(NH3)6]3+-zeolites, [Co(NH3)6]3+-mordenite	air (90 °C)/initiator (CHP)	DMF, tert-butanol, DMA, THF, toluene, dioxane	α-pinene epoxide	[61]
27.	Co(II)-exchanged ZSM-5 and beta zeolites	air (90 °C)/initiator (TBHP)	DMF	α-pinene epoxide (verbenone, verbenol)	[62]
28.	Co-MOR with bi-,tridentate Schiff base ligands	dry air (90 °C)/initiator (TBHP)	DMF, DMA, toluene, cyclohexanone, dioxane	α-pinene epoxide (verbenone, verbenol)	[63]
29.	Schiff base (Saloph) complexes of Ru(III) and Co(II) immobilized inside zeolite Y	air (100 °C)/initiator (AIBN)	MeCN	α-pinene epoxide, verbenone	[33]
30.	CoOx/mordenite	air (90 °C)/initiator (CHP, TBHP)	DMF, toluene, acetyl acetone, dioxane, cyclohexanone, butyl-acetate	α-pinene epoxide	[64]
31.	Nanosized Co3O4	air (90 °C)/initiator (TBHP)/ultrasonic wave	DMF, toluene, dioxane, MeCN	α-pinene epoxide (verbenone, verbenol)	[65]
32.	CeO2 with Cu, Co, Ni, Mn, and Fe metal oxides	photocatalytic oxidation with O2 (25 °C)	MeCN	α-pinene epoxide, verbenol, verbenone	[66]
33.	CeO2 nanostructures (CeO2 with nanoparticles, nanorods, and nanocubes)	photocatalytic oxidation with O2 (RT)	MeCN, MeOH, water, DMF, dioxane, dichloromethane	α-pinene epoxide, verbenol, verbenone	[67]
34.	MoO2Cl2@ COMOC-4 (gallium-based MOF, modified with MoO2Cl2)	photocatalytic oxidation with O2 (RT)	MeCN, DCM, chloroform, ethanol,	α-pinene epoxide	[68]
35.	SalenMnIIICl-aminodeoxycellulose	O2 (25 °C)	MeCN	α-pinene epoxide	[69]
36.	Vanadium-containing nickel phosphate molecular sieves (V-VSB-5)	O2 (60 °C)	MeCN	α-pinene epoxide, verbenone, verbenol	[70]
37.	Carbon catalysts from pine cones	O2 (80, 90, 100, 110, 120, 130, 140 °C)	w/o solvent	α-pinene epoxide, verbenone, verbenol	[71]
38.	Ni-modified carbonaceous materials (carrier—coffee grounds)	O2 (90, 100, 110 °C)	w/o solvent	α-pinene epoxide, verbenone, verbenol	[72]
39.	Clinoptilolite—zeolite	O2 (80–110 °C)	w/o solvent	α-pinene epoxide, verbenone, verbenol	[73]
40.	FeCl3-modified carbonaceous catalysts from waste orange peel	O2 (80, 90, 100, 110, 120 °C)	w/o solvent	α-pinene epoxide, verbenone, verbenol	[74]
41.	Metallodeuteroporphyrin	O2 (60, 90, 105, 120 °C)	w/o solvent	verbenol, verbenone, myrtenol, α-pinene epoxide	[75]
42.	MIXMIL-53-NH2(50)-Mal-Mn (mixed-linker MOF)	air (130 °C)	DMF/DEC	α-pinene epoxide (verbenone, verbenol)	[34]
43.	Pd/C catalyst	O2 (70–120 °C)	w/o solvent	α-pinene epoxide, verbenyl-hydro peroxide, verbenol	[76]

Abbreviations used in Table 1: TBHP—tert-butyl hydroperoxide, CHP—cumene hydroperoxide, AIBN—azobisisobutyronitrile, RT—room temperature ~25 °C, NT—nanotube, Bipy—4,4′-dicaboxylato-2,2′-bipyridine, MOF—metal–organic framework, THF—tetrahydrofuran, DMF—N,N-dimethylformamide, DMA—N,N-dimethylacetamide, DCM—dichloromethane, and DEC—dimethylformamide.

As can be seen, most of the research on α-pinene oxidation by O₂ has been carried out using heterogeneous catalysts, and α-pinene epoxide, verbenol, and verbenone are the dominant products. However, the application of homogeneous conditions in these investigations allows for an easy assessment of the performance of the catalyst investigated.

We have previously used different Fe(II), Fe(III), and Mn(II) complexes for the activation of O₂ in alkene oxidation [77,78,79,80,81,82,83,84]. The Fe(II) complex with the pentadentate ligand N4Py [N4Py—N, N-bis(2-pyridylmethyl)-N-(bis-2-pyridylmethyl)amine] (Scheme 2), with a structure inspired by nonheme enzymes, shows promising properties in terms of effectiveness in the oxidation of cyclohexene and limonene by O₂ [83]. For these reasons, Fe(II) and Mn(II) complexes with N4Py have been used as catalysts for the dioxygen oxidation of α-pinene, one of the most abundant terpenes.

Scheme 2.

Structures of (a) N4Py ligand and (b) [(N4Py)M^II]²⁺_MeCN complex (M = Fe or Mn), visualized in GaussView 5.0.

The complexes of [(N4Py)Fe^II]²⁺, [(N4Py)Mn^II]²⁺, and their adduct with oxygen have been the subject of many examinations. The available literature provides detailed information on their structures derived from UV–Vis [85,86,87,88,89,90,91], EPR [87,92], NMR [86,89,93], and EXAFS [94] spectra, as well as X-ray crystallography [86,93,95], XRD [92], electrospray ionization mass spectra [88], and Mössbauer parameters [96].

DFT (Density Functional Theory) methods allow us to calculate the energy and geometry of the structures analyzed, the study of reaction mechanisms, and the identification of intermediate and final reaction products. These methods are widely used in the study of reactions with α-pinene as a substrate. Functional and basis sets such as M06-2X and various Pople basis sets— including 6-31G(d), 6-31⁺G(d,p), and 6-311⁺⁺G(d,p) [97]—have been used to study the reactions of α-pinene with ^∙OH ([97,98,99,100]) and ^∙NO₃ [99] radicals as well as with O₃ [101], and the calculations showed good agreement with the experimental results. DFT methods (ωB97X-D/6-31⁺⁺G, ωB97X-D/aug-cc-pVTZ, and M062x/aug-cc-pVTZ [100]) were also used in studies of the autoxidation reaction of α-pinene induced by ^∙OH radicals. The methods B3LYP/6-31⁺G(d), and ωB97X-D/aug-cc-pVTZ were used to study the ozonolysis reaction of α-pinene [102] and also in the case of low-volatility organic compounds formed as products of ozonolysis [103].

2. Results and Discussion

2.1. [(N4Py)Fe^II]²⁺-Catalyzed Oxidation of α-Pinene with Dioxygen

In MeCN, [(N4Py)Fe^II]²⁺ activates molecular oxygen for the oxidation of α-pinene. The main products are α-pinene epoxide, verbenol, and verbenone; myrtenol and myrtenal are formed in much smaller amounts (Scheme 3).

Scheme 3.

The products of α-pinene oxidation.

Table 2 presents the product concentrations after 24 h of reaction time for different catalyst and substrate concentrations using dioxygen and air as oxidants. In a deaerated solution of α-pinene containing the catalyst, no oxidation products were found. Similarly, in the solution without the catalyst, O₂ causes the formation of only trace amounts of the main products.

Table 2.

Oxidation of α-pinene with dioxygen (p_O2 = 1 atm) and air (p_O2 = 0.2 atm) catalyzed by [(N4Py)Fe^II]²⁺. Reaction time 24 h, with MeCN as the solvent.

Catalyst, mM	Substrate, M	O2, atm	α-Pinene Epoxide, mM	Verbenol, mM	Verbenone, mM	Myrtenol, mM	Myrtenal, mM	TON
0.5	1	1	90	48	50	4	6	396
0.5	1	0.2	41	30	23	3	4	202
1	1	1	93	48	43	5	5	194
1	1	0.2	61	38	37	6	2	144
2.5	1	1	67	37	37	4	5	60
2.5	1	0.2	56	37	34	4	4	54
5	1	1	54	39	35	4	5	27
5	1	0.2	36	30	23	3	4	19
7.5	1	1	58	43	37	4	4	19
7.5	1	0.2	21	22	14	3	3	8
10	1	1	33	30	29	4	4	10
10	1	0.2	18	20	12	3	2	6
1	0.1	1	1	1	n.d.	1	0	4
1	0.1	0.2	1	1	n.d.	1	0	4
1	0.25	1	13	6	5	2	1	28
1	0.25	0.2	12	5	6	2	1	25
1	0.5	1	41	19	18	3	2	83
1	0.5	0.2	30	15	17	3	2	68
1 a	1	1	93	48	43	5	5	194
1 a	1	0.2	61	38	37	6	2	144
1	1.5	1	93	55	53	5	4	211
1	1.5	0.2	77	43	45	7	3	176

^a—data repeated from rows 3 and 4 of this table, TON—product molecules per catalyst molecule, and n.d.—not detected.

The best yields have been obtained for catalyst concentrations equal to 0.5 and 1 mM using dioxygen as an oxidant. For example, for 1 M α-pinene, 0.5 mM [(N4Py)Fe^II]²⁺, and dioxygen as an oxidant, 90 mM α-pinene epoxide, 48 mM verbenol, and 50 mM verbenone have been formed, which, taking into account the concentrations of the minor products, gives a turnover number approximately equal to 400. The replacement of dioxygen with air as an oxidant causes the reaction yield to be lower. The reaction yield is also lowered by increasing the catalyst concentration. The same effect, caused by the decomposition of catalysts, has been observed in our previous research on metal-induced O₂ activation for the oxidation of organic substrates [83]. As expected, a decrease in the concentration of α-pinene causes a decrease in the reaction yield. However, it should be noted that for 0.1 M α-pinene, practically no products are formed (see also Figure S3).

The concentration profiles of the main products of α-pinene oxidation with dioxygen and air, catalyzed by 1 mM [(N4Py)Fe^II]²⁺ in MeCN, are presented in Figure 1. The amounts of products increase almost linearly up to approximately 8 h, and then a plateau begins to form. It is characteristic that the amounts of verbenol and verbenone formed are almost the same, especially when dioxygen is used as an oxidant. The results presented indicate that the products form independently during the oxidation process. Excess substrate was used to avoid subsequent reactions.

Figure 1.

Dependence of the concentrations of the main products on time for the oxidation of 1 M α-pinene with (a) dioxygen (p_O2 = 1 atm) and (b) air (p_O2 = 0.2 atm), catalyzed by 1 mM [(N4Py)Fe^II]²⁺ in MeCN.

Unfortunately, the catalyst has very limited solubility in pure α-pinene. The use of 1 mM [(N4Py)Fe^II]²⁺ shows the turbidity of the reaction mixture. After 24 h of oxidation with dioxygen, 35 mM pinene epoxide, 32 mM verbenol, and 18 mM verbenone were formed. Using air as an oxidant, the amounts of the products were equal to 28, 28, and 12 mM, respectively. The yield of the main products formed is reduced by approximately 60% compared to the systems containing solvent (1 M α-pinene in MeCN, the same amount of catalyst). It should also be noted that in the reaction environment without the solvent, the amounts of α-pinene epoxide and verbenol are almost the same, whereas the ratio of verbenol to verbenone is roughly equal to 2.

Table 3 presents the results of the oxidation of 1 M α-pinene with hydrogen peroxide catalyzed by 1 mM [(N4Py)Fe^II]²⁺ in deaerated MeCN. In this case, verbenol is the main product, and the increase in HOOH concentration has not increased the reaction yield. This suggests that the Fenton mechanism prevails in the oxidation process.

Table 3.

Oxidation of 1 M α-pinene with hydrogen peroxide catalyzed by 1 mM [(N4Py)Fe^II]²⁺ in MeCN. Reaction time 24 h.

HOOH, mM	α-Pinene Epoxide, mM	Verbenol, mM	Verbenone, mM	Myrtenol, mM	Myrtenal, mM	TON
25	6	6	1	n.d.	2	13
50	4	12	4	n.d.	2	19
100	4	12	4	n.d.	2	20

TON—product molecules per catalyst molecule; n.d.—not detected.

2.2. [(N4Py)Mn^II]²⁺ as a Catalyst for the Oxidation of α-Pinene with Dioxygen

The [(N4Py)Mn^II]²⁺ complex, similarly to its iron analog, catalyzes the oxidation of α-pinene by O₂ in MeCN. Regardless of the slightly reduced yield, the observed product concentration profiles are close to those obtained for the iron complex. As follows from Table 4, α-pinene epoxide is formed in the largest amounts, and the molar ratio of verbenol to verbenone is approximately equal to 1.

Table 4.

Oxidation of α-pinene with dioxygen (p_O2 = 1 atm) and air (p_O2 = 0.2 atm) catalyzed by [(N4Py)Mn^II]²⁺. Reaction time 24 h, with MeCN as a solvent.

Catalyst, mM	Substrate, M	O2, atm	α-Pinene Epoxide, mM	Verbenol, mM	Verbenone, mM	Myrtenol, mM	Myrtenal, mM	TON
0.5	1	1	72	43	53	4	5	355
0.5	1	0.2	38	24	23	5	3	185
1	1	1	67	35	45	4	5	157
1	1	0.2	46	32	30	3	6	116
2.5	1	1	27	14	20	2	2	26
2.5	1	0.2	11	13	9	2	2	15
1	0.1	1	n.d.	n.d.	n.d.	n.d.	n.d.	0
1	0.1	0.2	n.d.	n.d.	n.d.	n.d.	n.d.	0
1	0.25	1	3	2	1	1	1	9
1	0.25	0.2	4	3	3	2	1	12
1	0.5	1	24	16	15	2	4	60
1	0.5	0.2	30	21	20	4	3	77
1 a	1	1	67	35	45	4	5	157
1 a	1	0.2	46	32	30	3	6	116
1	1.5	1	67	40	53	7	5	172
1	1.5	0.2	47	35	32	3	7	123

^a—data repeated from rows 3 and 4 of this Table, TON—product molecules per catalyst molecule, and n.d.—not detected.

The reaction yield decreases with the increase in catalyst concentration; for the concentrations of the catalyst equal to 5 and 10 mM, no oxidation products were found. The same effect, attributed to the catalyst’s decomposition, was observed during O₂ activation by the [Mn^II(bpy)₂²⁺]_MeCN complex for the oxygenation of limonene in MeCN acetonitrile [80]. It is also worth noting that for the substrate concentration equal to 0.1 M, any oxidation product has been detected (see also Figure S4).

Figure 2 presents the concentration profiles of the main products of α-pinene oxidation with dioxygen and air catalyzed by 1 mM [(N4Py)Mn^II]²⁺ in MeCN. Similarly to the results obtained for the Fe catalyst, the concentration profiles indicate that the products form independently during the oxidation process. However, the observed plateau on the concentration profiles starts earlier than in the case of the iron complex used as a catalyst.

Figure 2.

Dependence of the concentrations of the main products on time for the oxidation of 1 M α-pinene with (a) dioxygen (p_O2 = 1 atm) and (b) air (p_O2 = 0.2 atm), catalyzed by 1 mM [(N4Py)Mn^II]²⁺ in MeCN.

Similarly to the iron catalyst, [(N4Py)Mn^II]²⁺ exhibits very limited solubility in α-pinene. An attempt to use 1 mM of this catalyst indicates that it is not completely soluble. In this condition, the oxidation of α-pinene with dioxygen catalyzed by [(N4Py)Mn^II]²⁺ gives a slightly lower yield compared to the iron analog; thus, 18 mM α-pinene epoxide, 28 mM verbenol, and 16 mM verbenone are formed after 24 h. It is characteristic that verbenol is the main product, and the use of air instead of dioxygen gives almost the same reaction yield.

In contrast to the iron catalyst, the use of HOOH for the oxidation of α-pinene in MeCN in the presence [(N4Py)Mn^II]²⁺ as a catalyst produces verbenol as the sole product, while α-pinene epoxide and verbenone are present in trace amounts.

2.3. Investigations on the Influence of α-Pinene and O₂ on the Electrochemical Properties of the Catalysts

In our previous work [83], it has been shown that the first oxidation peak of the [(N4Py)Fe^II]²⁺ complex is present at +1.05 V (vs. SCE), and its height remains unchanged in the presence of dioxygen. The addition of the substrate (cyclohexene or limonene) caused its height to decrease. Similarly, the addition of α-pinene to the oxygenated solution of [(N4Py)Fe^II]²⁺ leads to a decrease in its height (Figure 3a in the presence of air and Figure S5a in the presence of dioxygen).

Figure 3.

Cyclic voltammograms of 1 mM [(N4Py)Fe^II]²⁺ complex in MeCN with (a) anodic and (b) cathodic scans performed first. (I) Ar atmosphere, (II) air atmosphere (p_O2 = 0.2 atm), (III) as (II) after the addition of 1 M α-pinene, and (IV) as (III) after 3 h. Supporting electrolyte, 0.1 M (t-Bu)₄NClO₄; scan rate, 0.1 V/s; GCE (0.008 cm²); SCE vs. NHE +0.242 V.

The more important is the fact that after the addition of 1 M α-pinene to the oxygenated solution of 1 mM [(N4Py)Fe^II]²⁺ (this mimics the conditions of the oxidation process performed) in the first cathodic scan, a small reduction peak is present at approximately 0.0 V (Figure 3b in the presence of air and Figure S5b in the presence of dioxygen). As demonstrated in [83], the peak is attributed to the reduction of [(N4Py)Fe^IV=O]. This indicates that the Fe(IV)-oxo species are formed when the catalyst, substrate, and O₂ are present simultaneously in the reaction mixture. The broad cathodic peak visible in the voltammograms at about −0.5 V is probably caused by the reduction of the other Fe-oxygen species, which can be formed under experimental conditions.

Analogously, the [(N4Py)Mn^II]²⁺ complex is also oxidized at approximately 1.1 V, and the height of the peak is not affected by the presence of dioxygen. However, in contrast to the iron complex, the addition of α-pinene causes only a slight decrease in the height of this peak (Figure 4a in the presence of air and Figure S6a in the presence of dioxygen). On the other hand, in the first cathodic scan performed in the solution containing the Mn complex, dioxygen, and α-pinene, a small reduction peak is present at approximately 0.0 V, suggesting that a Mn-oxo species, [(N4Py)Mn^IV=O], is formed (Figure 4b in the presence of air and Figure S6b in the presence of dioxygen). However, the peak disappears after approximately 3 h.

Figure 4.

Cyclic voltammograms of 1 mM [(N4Py)Mn^II]²⁺ complex in MeCN with (a) anodic and (b) cathodic scans performed first. (I) In the inert gas (Ar) atmosphere, (II) in the air atmosphere (p_O2 = 0.2 atm), (III) as (II) after the addition of 1 M α-pinene, and (IV) as (III) after 3 h. Supporting electrolyte, 0.1 M (t-Bu)₄NClO₄; scan rate, 0.1 V/s; GCE (0.008 cm²); SCE vs. NHE +0.242.

2.4. The DFT Analysis of α-Pinene and Its Main Oxidation Products

We have used the B3LYP/6-311⁺⁺G(d,p) and ωB97XD/6-311⁺⁺G(d,p)/def2sv methods to evaluate the relative energies of the (−) or (+)enantiomers of the oxygenation products of α-pinene oxidation, α-pinene epoxide, verbenol, verbenone, myrtenal, and myrtenol (Schemes S1 and S2), with the results presented in Table S1. The results obtained indicate that the relative electronic energies and the relative Gibbs free energies are practically equal to zero for all substrates and products investigated except verbenol. The relative energy has been defined as the difference between the energies of the enantiomers of the same molecule, whereby the lower energy is arbitrarily set as zero. For verbenol, the parameters discussed have values up to 1.2 kcal/mol, which are within the error limits of the calculation method. For optimized molecules, their HOMO and LUMO surfaces have been visualized (Figure 5), and the obtained energies values were used to calculate chemical reactivity descriptors (CRD) (Table S2). CRDs (ionization potential—I, electron affinity—A, electronegativity—X, hardness—η, softness—S, electrophilicity index—ω, and chemical potential—μ) are defined by Parr, Pearson, and Yang [104].

Figure 5.

Energies of the HOMOs and LUMOs for the analyzed molecules of (−)-enantiomers with their visualized surfaces for the corresponding orbitals, B3LYP/6-311⁺⁺G(d,p).

The high energy difference between HOMO and LUMO levels ΔE_gap indicates that a molecule has a lower chemical reactivity and, consequently, a higher stability [105]. The data presented in Figure 5 and Table S2 indicate that the α-pinene oxidation products are generally not more reactive than the substrate, and therefore, taking into account the concentration difference, their formation does not alter the process of α-pinene oxidation.

2.5. Putative Reaction Mechanism

The possible reaction paths in the investigated system are outlined in Scheme 4. Based on the argumentation presented in our previous articles [83,84], the presence of the reduction peak at approximately 0.0 V in the first cathodic scan after the addition of the substrate (α-pinene) to the mixture of the catalyst [(N4Py)Fe^II]²⁺ or [(N4Py)Mn^II]²⁺ and the oxidant (O₂) suggests that the corresponding metal-oxo adduct [(N4Py)Fe^IV=O]²⁺ or [(N4Py)Mn^IV=O]²⁺ is formed (path A). The possibility of the occurrence of the process, in which an alcohol or epoxide is also formed, was indicated in [106,107,108].

Scheme 4.

Proposed route for the reactivity of Fe^IV- and Mn^IV-oxo adducts (M = Fe or Mn; L = N4Py).

However, the further reactivity of Fe-oxo and Mn-oxo adducts with organic substrates differs. For the Fe-oxo adduct, the reaction takes place mainly through an oxygen rebound mechanism (path B), which leads to the formation of alcohol and/or epoxide, and the original catalyst [(N4Py)Fe^II]²⁺ is released. In the case of Mn-oxo adducts, the non-rebound mechanism prevails (path C), which, except for the formation of alcohol and/or epoxide, also generates the dimeric μ-oxo complex [(N4Py)Mn^III-O-Mn^III(N4Py)]⁴⁺ [109,110,111].

This suggests that in the case of the manganese catalyst, its decomposition is more probable. The results obtained are in agreement with the above statement. In the case of the manganese catalyst, the reaction is inhibited at a lower concentration of the catalyst than for the iron catalyst (approximately 2.5 mM compared to 10 mM), and the higher concentration of the substrate is necessary for the oxidation reaction to occur (compare Table 2 and Table 4). This also indicates that the rate of the path H (regeneration of the catalyst) is relatively slow.

To explain the formation of a ketone during the oxidation process, it is necessary to assume that in the reaction mixture, the combination of M(IV)-oxo species (M = Fe or Mn), O₂, and substrate produces a hypothetical adduct (1) (path D). The assumption is based on our previous research [77,78,79,80], in which this type of structure was postulated as an intermediate step in the oxidation of cyclohexene and limonene with O₂ catalyzed by coordinately unsaturated Fe or Mn catalysts, e.g., [M^II(bpy)₂²⁺]_MeCN (bpy—2,2′-bipyridine). In spite of N4Py being a pentadentate ligand, it has been shown [86] that in a water solution of the [(N4Py)Fe^II]²⁺ complex, a nitrogen atom of the N4Py ligand can free the iron ion coordination site for a water molecule or a hydroxyl ion. By analogy, it can be assumed that such release of the iron coordination place can also be possible in MeCN. It is necessary to emphasize that the proposition of species (1) formation is hypothetical but gives a reasonable explanation of the observed experimental data.

Possible transformations of the hypothetical adduct (1) can involve its decomposition to a ketone and M(IV)-oxo species (path E), reaction with the substrate molecule to produce a ketone, alcohol, and catalyst (path F), and interaction with the catalyst that gives a ketone and dimeric M^III μ-oxo complex, which in many cases is catalytically inactive (path G) [112,113]. The increased amount of verbenol obtained in the case of the use of HOOH as an oxidant suggests that the Fenton mechanism plays a substantial role in product formation.

3. Materials and Methods

3.1. Equipment

A gas chromatograph equipped with an HP-1 capillary column (cross-linked methyl silicone rubber phase, 30 m × 0.53 mm internal diameter), Hewlett-Packard 4890A series (Palo Alto, CA, USA), was used to identify and separate the reaction products. ¹H and ¹³C NMR analyses were performed in CDCl₃ or DMSO-d₆ using a Bruker Avance 500 MHz spectrometer (Bruker, Karlsruhe, Germany) at 25 °C. The Princeton Applied Research Model 273A potentiostat (PAR, Oak Ridge, TN, USA) and Metrohm Autolab (Utrecht, The Netherlands) model PGSTAT 302 N potentiostat were applied to perform electrochemical experiments using the cyclic voltammetric method.

3.2. Chemicals and Regents

The reagents used for the experiments were of the highest available purity and were applied without further purification. The solvent in all experiments was acetonitrile (≥99.9%, HPLC grade) from Sigma-Aldrich (now Merck, Poznań, Poland). High-purity argon (5.0) supplied by Linde (Kraków, Poland) was used to remove oxygen from the solutions. Dioxygen (5.0) was provided by SIAD (Ruda Śląska, Poland). Iron(II) perchlorate, Fe(ClO₄)₂·6H₂O, manganese(II) perchlorate, Mn(ClO₄)₂·6H₂O (99%), biphenyl (PhPh, ≥99%), sodium hydroxide (98%), α-pinene (98%), α-pinene epoxide (97%), hydrogen peroxide (50 wt% in H₂O, stabilized), (S)-(−)-verbenone (94%), (S)-cis-verbenol (95%), (1R)-(−)-myrtenol (95%), (1R)-(−)-myrtenal (98%), di-2-pyridyl ketone oxime (98%), 2-picolyl chloride hydrochloride (98%), and chloroform (98%) were obtained from Sigma-Aldrich. Magnesium sulfate anhydrous (99%), ammonium acetate (97%), sodium sulfate anhydrous (99%), and ammonia solution (25%) were provided by POCH (Gliwice, Poland).

The N4Py ligand and its iron complex [(N4Py)Fe^II]²⁺ were prepared according to the procedure given in the literature [95,114,115], applying the Schlenk line system and glovebox. ¹H NMR analysis was performed with the use of DMSO-d₆ or CDCl₃ for the synthesized ligand; N4Py and the [(N4Py)Fe^II]²⁺ complex confirmed that the characteristic bands [83,95] were obtained. ¹H NMR N4Py (500 MHz, DMSO-d₆, 25 °C): δ (ppm) = 8.50 (d, 2H, Py), 8.48 (d, 2H, Py), 7.42 (m, 8 H, Py), 7.22 (m, 4H, Py), 5.25 (s, 1 H, CH), 3.86 (s, 4H, CH). There were four doublets in the spectrum with chemical shifts in the range of 7.42–7.67 ppm originating from the hydrogen of the pyridine group connected to the methine group of the ligand. ¹H NMR N4Py (500 MHz, CDCl₃, 25 °C): δ (ppm) = 8.63 (d, 2H, Py), 8.48 (d, 2H, Py), 7.63 (m, 8 H, Py), 7.11 (m, 4H, Py), 5.23 (s, 1 H, CH), 3.87 (s, 4H, CH). ¹³C NMR analysis C (N4Py): δ (ppm) = 159.85–162.36 (C, 2-pyridine), 122.9–149.0 ppm (CH, 2-pyridine), 71.60 (CH, aliphatic), 68.73 (CH₂, aliphatic), 64.71 (CH₂, aliphatic). The synthesis of [(N4Py)Mn^II]²⁺ was carried out in a glovebox, according to the procedure in [116], in which the ligand and Mn(II) salt were dissolved in MeCN and stirred at room temperature overnight in an atmosphere of inert gas, and the solution was used for further studies.

3.3. Methods

Preparative experiments: The appropriate amounts of catalyst were dissolved in deaerated MeCN (p_O2 = 0 atm). The solution was aerated (p_O2 = 0.2 atm) or oxygenated (p_O2 = 1 atm), and the standard and substrate were successively added. The total volume of the sample was 2.5 mL, using a 25 mL reaction vial with a cutout cap and a Teflon septum. The reactions were carried out up to 24 h with constant stirring at room temperature (23 ± 1 °C). Samples (0.2 µL) were taken periodically to monitor the progress of the reaction, using gas chromatography with an FID detector. Biphenyl (10 mM) was used as an internal standard. The concentrations of the obtained products were determined using standard curve equations presented in Table S3. Product concentration values are the average values of three independent experiments.

Electrochemical measurements: Cyclic voltammetry (CV) studies were performed in a 2 mL electrochemical cell with conical bottom and the ability to control the presence of O₂ using an argon flushing system. A three-electrode system was used: the working electrode was glassy carbon in PEEK (Cypress Systems, Division of ESA, Inc., Tokyo, Japan) with a diameter of 1 mm; the counter electrode was a platinum wire; and the reference electrode was Ag/AgCl (Ag/AgCl wire in an aqueous solution of tetramethylammonium chloride, which concentration was adjusted to obtain a potential of 0.00 V vs. SCE). The reference electrode was placed in a Pyrex tube with a Vycor tip, inside a Luggin capillary [117]. Before each electrochemical measurement, the working electrode was polished using a Buehler Micropolish Alumina Gamma 3B polisher (Buehler, Leinfelden-Echterdingen, Germany) and a Buehler Microcloth polishing cloth and then rinsed with deionized water and dried.

DFT calculations: Thermodynamic parameters were calculated using Gaussian 09 [118] with DFT methods and hybrid potentials, such as the Becke 3-parameter hybrid density functional and the Lee–Yang–Parr correlation, using methods from the Pople basis sets, including B3LYP/6-311⁺⁺G(d,p) and ωB97XD/6-311⁺⁺G(d,p)/def2sv. The structures were visualized in GaussView 05. This paper reports values that combine these electronic energies with the ZPE correction and the Gibbs free energy correction.

4. Conclusions

Both the [(N4Py)Fe^II]²⁺ and [(N4Py)Mn^II]²⁺ complexes catalyze the oxidation of α-pinene with O₂ in acetonitrile. However, the systems are not selective, and α-pinene epoxide is the main product that is produced, approximately twice as much as verbenol or verbenone, whose molecular ratio is approximately equal to 1. Myrtenol and myrtenal were also detected but in small amounts. The iron catalyst shows higher efficiency than the manganese catalyst. The use of air instead of dioxygen as the oxidant reduces the reaction yield. The results of the voltammetric measurements indicate that for both catalysts investigated, M(IV)-oxo adducts are formed when a catalyst, α-pinene, and O₂ are combined in MeCN. The adducts are probably reactive intermediates whose interaction with the ingredients of the system initiates a series of reaction paths leading to the formation of observed products. The manganese catalyst is less stable, and its decomposition to the less reactive dimeric μ-oxo Mn(III) complex is faster than in the case of the iron catalyst.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/molecules31050908/s1, Figure S1: Ligands used in the complexes that were discussed in the Introduction; Figure S2: Selected products of α-pinene oxidation, discussed in the Introduction; Figure S3: Dependence of the amounts of the main products on the α-pinene concentration, formed during its oxidation with (a) dioxygen (p_O2 = 1 atm) and (b) air (p_O2 = 0.2 atm), catalyzed by 1 mM [(N4Py)Fe^II]²⁺ in MeCN; Figure S4: Dependence of the amounts of the main products on the α-pinene concentration, formed during its oxidation with (a) dioxygen (p_O2 = 1 atm) and (b) air (p_O2 = 0.2 atm), catalyzed by 1 mM [(N4Py)Mn^II]²⁺ in MeCN; Figure S5: Cyclic voltammograms of 1 mM [(N4Py)Fe^II]²⁺ complex in MeCN with (a) anodic and (b) cathodic scans performed first. (I) In the inert gas (Ar) atmosphere, (II) in the O₂ atmosphere (p_O2 = 1 atm), (III) as (II) after addition 1 M α-pinene, and (IV) as (III) after 3 h. Supporting electrolyte = 0.1 M (t-Bu)₄NClO₄, scan rate = 0.1 V/s, GCE (0.008 cm²), and SCE vs. NHE +0.242 V; Figure S6: Cyclic voltammograms of 1 mM [(N4Py)Mn^II]²⁺ complex in MeCN with (a) anodic and (b) cathodic scans performed first. (I) In the inert gas (Ar) atmosphere, (II) in the O₂ atmosphere (p_O2 = 1 atm), (III) as (II) after addition 1 M α-pinene, and (IV) as (III) after 3 h. Supporting electrolyte = 0.1 M (t-Bu)₄NClO₄, scan rate = 0.1 V/s, GCE (0.008 cm²), and SCE vs. NHE +0.242 V, cathodic scan; Scheme S1: (−)-Enantiomers of α-pinene and its oxidation products; Scheme S2: (+)-Enantiomers of α-pinene and its oxidation products; Table S1: The energies (with and without zero point correction), enthalpies, free energies (G), and respective relative values for different molecules calculated with the B3LYP/6-311⁺⁺G(d,p) and ωB97XD/6-311⁺⁺G(d,p)/def2sv methods; Table S2: Calculated CRD for molecules of the presented molecules, with structures optimized using the methods B3LYP/6-311⁺⁺G(d,p) and ωB97XD/6-311⁺⁺G(d,p)/def2sv. Table S3: Equations of the calibration curves determined for the individual products of the α-pinene oxidation. Internal standard—biphenyl (10 mM), MeCN.

Author Contributions

Conceptualization, K.R.-C. and A.S.; methodology, K.R.-C. and A.S.; formal analysis, K.R.-C. and A.S.; investigation, K.R.-C.; resources, K.R.-C. and A.S.; data curation, K.R.-C.; writing—original draft preparation, K.R.-C. and A.S.; writing— review and editing, K.R.-C. and A.S.; visualization, K.R.-C.; supervision, A.S.; project administration, K.R.-C.; funding acquisition, K.R.-C. and A.S. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Data are provided in the article and Supplementary Materials.

Conflicts of Interest

The authors declare no conflicts of interest.

Funding Statement

This study was financed by the Minister of Science and Higher Education, Republic of Poland, within the program “Regional Excellence Initiative” under grant number RCD.RB.24.002.