Mixed-ligand aminoalcohol-dicarboxylate copper(II) coordination polymers as catalysts for the oxidative functionalization of cyclic alkanes and alkenes

Introduction

The development of new protocols and metal-complex catalytic systems for the oxidative C–H functionalization of saturated and unsaturated hydrocarbons into various types of value added products continues to be a topic of intense research in the fields of coordination chemistry, organic synthesis, and oxidation catalysis [1], [2], [3], [4], [5], [6], [7], [8], [9]. Among a diversity of coordination compounds used as catalysts or catalyst precursors in such hydrocarbon functionalization reactions, copper derivatives are of particular significance due to a relatively low cost of copper, its natural abundance, rich bioinorganic and coordination chemistry, as well as a recognized application in oxidation catalysis [10], [11]. Besides, copper is present in the active sites of different oxidation enzymes [12], [13], including a particulate methane monooxygenase, which is a unique multicopper enzyme capable of hydroxylating alkanes that are very inert and least reactive hydrocarbons [14], [15].

Despite a considerable progress made on the design and application of bioinspired and/or biomimetic discrete copper complexes in oxidation catalysis [16], [17], [18], [19], [20], the use of soluble Cu-containing coordination polymers that incorporate multicopper secondary building units still remains less explored [21], [22], [23]. However, such compounds can potentially exhibit better stability and higher catalytic activity in oxidative functionalization of hydrocarbons than the discrete mononuclear copper complexes. Bearing these points in mind and following our general interest in the synthesis and catalytic application of copper(II) coordination polymers, the main objective of the present work consisted in the evaluation of the catalytic potential of several mixed-ligand aminoalcohol-dicarboxylate Cu(II) coordination polymers in different types of oxidative transformations of hydrocarbons.

Hence, the present study reports the catalytic application of three 1D coordination polymers, namely [Cu₂(μ-dmea)₂(μ-nda)(H₂O)₂]_n·2nH₂O (1), [Cu₂(μ-Hmdea)₂(μ-nda)]_n·2nH₂O (2), and [Cu₂(μ-Hbdea)₂ (μ-nda)]_n·2nH₂O (3) that bear slightly different dicopper(II) aminoalcoholate cores, as well as a structurally distinct dicopper(II) discrete [Cu₂(H₄etda)₂(μ-nda)]·nda·4H₂O (4) derivative (Scheme 1). These coordination compounds were self-assembled from 2,6-naphthalenedicarboxylic acid (H₂nda) acting as a μ-nda linker and four aqua-soluble aminoalcohols [N,N′-dimethylethanolamine (Hdmea), N-methyldiethanolamine (H₂mdea), N-butyldiethanolamine (H₂bdea), and N,N,N′,N′-tetrakis(2-hydroxyethyl)ethylenediamine (H₄etda)] that possess one (Hdmea), two (H₂mdea, H₂bdea), or four (H₄etda) hydroxyethyl arms. These products were tested as homogeneous catalysts (or pre-catalysts) in three types of model catalytic reactions that proceed in H₂O/MeCN medium under mild conditions (50–60°C): (A) the oxidation of cyclohexane by H₂O₂ to cyclohexyl hydroperoxide, cyclohexanol, and cyclohexanone, (B) the oxidation of cyclic alkenes (cyclohexene, cyclooctene) by H₂O₂ to a mixture of different oxidation products, and (C) the single-pot hydrocarboxylation of cycloalkanes (cyclopentane, cyclohexane, cycloheptane, cyclooctane) by CO, H₂O, and K₂S₂O₈ into the corresponding cycloalkanecarboxylic acids. The obtained results on the catalytic activity of 1–4 in these reactions are discussed in detail.

Scheme 1:

Structural formulae of 1–4.

Experimental

Materials and methods. Chemicals and solvents were obtained from commercial sources and used as received. Compounds 1–4 were prepared according to a reported procedure [24]. Gas chromatography (GC) analyses were run on an Agilent Technologies 7820A series gas chromatograph (He as carrier gas) equipped with the FID detector and BP20/SGE (30 m×0.22 mm×0.25 μm) capillary column or on a GC 2010 Shimadzu (N₂ as carrier gas) equipped with the FID detector and BP20/SGE (10 m×0.10 mm×0.10 μm) capillary column. GC-MS analyses were run on a GC-MS QP 2010 Plus Shimadzu (He as carrier gas) instrument having BP20 (30 m×0.25 mm×1.0 μm) column; mode TIC (Total Ion Chromatogram).

Oxidation of cyclic alkanes and alkenes. Cyclohexane, cyclohexene, and cyclooctene oxidation reactions were performed in air atmosphere in thermostated glass reactors equipped with a condenser under vigorous stirring at 50°C and using MeCN as solvent (up to 5 mL total volume). In a typical experiment, catalyst 1–4 (10 μmol) and gas chromatography (GC) internal standard (MeNO₂, 50 μL) were introduced into the MeCN solution, followed by the addition of an acid promoter (TFA, 0.1 mmol) used as a stock solution in MeCN. An alkane or alkene substrate (2 mmol) was then introduced, and the reaction started upon the addition of hydrogen peroxide (50% in H₂O, 10 mmol) in one portion. The oxidation reactions were monitored by withdrawing small aliquots after different periods of time, which were then analyzed by GC (nitromethane as an internal standard) twice, before and after the treatment with PPh₃, following the Shul’pin’s method [25], [26], [27] that allows to estimate a concentration of organic hydroperoxides in the reaction medium. In fact, the introduction of PPh₃ leads to a complete reduction of remaining H₂O₂ and organic hydroperoxides that are typically formed as primary products in hydrocarbon oxidations. Attribution of peaks was made by comparison with chromatograms of authentic samples and by running GC-MS analyses for selected reactions. Blank tests confirmed that the hydrocarbon oxidations do not proceed in the absence of copper catalyst (only traces of products, <0.5% total yield can be detected).

Hydrocarboxylation of cycloalkanes. In a typical experiment, the reaction mixtures were prepared as follows: catalyst (10 μmol, 1–4) was placed into a 20.0 mL stainless steel autoclave, equipped with a Teflon-coated magnetic stirring bar, followed by the addition of K₂S₂O₈ (1.50 mmol), 2.0 mL of H₂O, 4.0 mL of MeCN (total solvent volume was 6.0 mL), and cycloalkane (1.00 mmol) [28], [29], [30], [31]. Then, the autoclave was closed and flushed with CO three times to remove the air, and finally pressurized with 20 atm of CO. CAUTION: Due to the toxicity of CO, all operations should be carried out in a well-ventilated hood! The reaction mixture was stirred for 4 h at 60°C using a magnetic stirrer and an oil bath, whereupon it was cooled in an ice bath, degassed, opened, and transferred to a flask. Diethyl ether (9.0 mL) and 45 μL of cycloheptanone (typical GC internal standard) were added. In the case of the cycloheptane hydrocarboxylation, cyclohexanone (45 μL) was used as a GC standard. The obtained mixture was vigorously stirred for 10 min, and the organic layer was analyzed by gas chromatography (internal standard method), revealing the formation of the corresponding monocarboxylic acids as major hydrocarboxylation products, along with cyclic alcohols and ketones as minor oxidation products. Attribution of peaks was made by comparison with chromatograms of authentic samples.

Results and discussion

Mild oxidation of cyclohexane

Cyclohexane was selected as a model substrate in view of the industrial importance of its oxidation products (cyclohexanone and cyclohexanol) in nylon production [2], [32], [33]. Hence, compounds 1–4 were applied as homogeneous catalysts (or pre-catalysts) for the mild oxidation of cyclohexane by hydrogen peroxide (50% in H₂O), in aqueous acetonitrile medium at 50°C in air, and in the presence of a catalytic amount of an acid promoter [10], [34], [35], [36], [37]. Trifluoroacetic acid (TFA) was used in view of its recognized promoting effect in alkane oxidations, which consists of activating a catalyst by partial protonation of aminoalcohol and/or carboxylate ligands (thus improving catalyst’s solubility in the reaction medium), facilitating proton transfer steps, and enhancing the oxidation properties of hydrogen peroxide [19], [21], [22], [34], [37].

The cyclohexane oxidation leads to the formation of a mixture of cyclohexyl hydroperoxide (major primary product) and cyclohexanol and cyclohexanone (final products). Despite the catalytic activity of 1–4 is almost similar, compounds 1 and 4 are slightly more active and result in up to 22% total yields of the oxidation products based on substrate (Fig. 1); these yields are typically achieved within the first 2 h of the reaction. It should be mentioned that such a level of yields is rather good in the field of alkane functionalization, especially considering an intrinsic inertness of these saturated hydrocarbons and mild reaction conditions [1], [2], [3], [4], [10], [32]. The obtained herein values of the product total yields are also comparable or even superior than those achieved when using other related copper catalysts bearing aminoalcoholate and/or carboxylate type ligands [9], [10], [20], [28].

Fig. 1:

Accumulation of products (total yield) with time in the cyclohexane oxidation by H₂O₂ catalyzed by 1–4. Reaction conditions: 1–4 (10 μmol), TFA (0.1 mmol), C₆H₁₂ (2 mmol), H₂O₂ (10 mmol), CH₃CN (up to 5 mL of the total volume), 50°C.

Typical kinetic curves of products accumulation in the cyclohexane oxidation catalyzed by compounds 1–4 are presented in Fig. 2. By analyzing the same reaction mixture twice, before and after an addition of PPh₃ to reduce cyclohexyl hydroperoxide (Shul’pin’s method [25], [26], [27]), we estimated an amount of cyclohexyl hydroperoxide as a primary intermediate product in the cyclohexane oxidation. In fact, it is a main product at the beginning of the reaction but after 2 h the amount of cyclohexyl hydroperoxide falls down due to its decomposition, mainly to cyclohexanol and cyclohexanone, as well as to some unidentified products. An increase in the accumulation of cyclohexanol and cyclohexanone is observed up to 3–4 h of the reaction time (Fig. 2) and then remains almost constant.

Fig. 2:

Kinetic curves of product accumulation in the cyclohexane oxidation by H₂O₂ catalyzed by 1 (a), 2 (b), 3 (c), and 4 (d) in the presence of TFA promoter. Reaction conditions: 1–4 (10 μmol), TFA (0.1 mmol), C₆H₁₂ (2 mmol), H₂O₂ (10 mmol), CH₃CN (up to 5 mL of the total volume), 50°C.

Despite showing resembling levels of activity in the cyclohexane oxidation, catalysts 1–4 exhibit slightly different trends of the total yield along the time. In fact, in the reaction catalyzed by 1 (Fig. 2a), the total yield of products (cyclohexyl hydroperoxide, cyclohexanol, cyclohexanone) reaches a plateau at ~1 h and remains almost unchanged up to 6 h of the reaction, suggesting that an overoxidation of the formed products does not occur. However, a slight decline of the maximum total yield after 2 h is observed in the reaction catalyzed by 3 (Fig. 2c). This process becomes more pronounced when using catalysts 2 and 4 (Fig. 2b, d), suggesting an overoxidation of the formed products at prolonged reaction times. This observation suggests that at a certain stage the products already present in the reaction medium (cyclohexyl hydroperoxide, cyclohexanol, and cyclohexanone) compete as substrates with more inert hydrocarbon molecules. Besides, this observation indicates that the H₂O₂ oxidant (used in excess) is still present in the reaction medium at prolonged times, despite its partial decomposition due to the catalase activity that is commonly observed in related Cu-based catalytic systems [10], [20], [21].

Although a detailed investigation of the catalytic intermediates lied out of the scope of this work, we analysed the ESI-MS(+) data for catalyst 1 (as a representative example) before and after its treatment with TFA and H₂O₂. Both spectra show resembling features and confirm the dissociation of 1 in MeCN/H₂O medium to give a main [Cu₂(dmea)₂(H₂O)₄]⁺ (m/z=374) fragment and related solvent containing adducts. Given resembling structures and catalytic performances of 1–3, such dicopper-aminoalcoholate cores may potentially constitute the catalytically active species. Taking into consideration that the fragmentation of such copper compounds can be ESI-induced, we also cannot exclude an involvement of higher molecular weight oligomeric species into catalytic cycles.

The formation of cyclohexyl hydroperoxide as a primary oxidation product (typical for a free-radical type process [25], [26], [27]) in the cyclohexane oxidation and the comparison of catalytic behavior of 1–4 with other related multicopper(II) systems that operate with an involvement of hydroxyl radicals [10], [19], [20], [21], [22], suggest that these can be principal oxidizing species in the present systems. On the basis of related literature background [10], [19], [20], [21], [22], [25], [26], [27], [34], [35], [36], we can assume the following main steps of the cyclohexane oxidation. Hence, a copper(II) catalyst participates in the generation of HO^• radicals from H₂O₂ which then abstract an H atom from cyclohexane (CyH) forming a cyclohexyl radical (Cy^•). This reacts with O₂ (e.g. from air) leading to a CyOO^• radical that is transformed to cyclohexyl hydroperoxide (CyOOH) as a primary intermediate product. This then rapidly decomposes to give cyclohexanol and cyclohexanone as final products [10], [21], [22], [34], [35], [36].

Mild oxidation of cyclohexene and cyclooctene

To explore the substrate scope of the catalytic systems comprising copper(II) compounds 1–4, we carried out the oxidation by H₂O₂ of unsaturated cyclic hydrocarbons, namely cyclohexene and cyclooctene. As in the case of cyclohexane, the oxidation of cyclohexene by H₂O₂ and in the presence of TFA promoter results in resembling total product yields (~15–18% based on substrate) when using compounds 1–4 as catalysts (Fig. 3). However, an overoxidation of products is more pronounced in the case of cyclohexene and its trend is different for 1–4.

Fig. 3:

Accumulation of products (total yield) with time in the cyclohexene oxidation by H₂O₂ catalyzed by 1–4. Reaction conditions: 1–4 (10 μmol), TFA (0.1 mmol), C₆H₁₀ (2 mmol), H₂O₂ (10 mmol), CH₃CN (up to 5 mL of the total volume), 50°C.

In contrast to cyclohexane, the oxidation of cyclohexene leads to the formation of a very complex mixture of different products, as shown in Scheme 2. The obtained kinetic curves for the accumulation of all the identified products in the oxidation of cyclohexene catalyzed by 1–4 are represented in Fig. 4.

Scheme 2:

Different paths for the formation of observed products in the copper-catalyzed cyclohexene oxidation by H₂O₂.

Fig. 4:

Accumulation of products with time in the cyclohexene oxidation by H₂O₂ catalyzed by 1–4: 1 (a/a′), 2 (b/b′), 3 (c/c′), 4 (d/d′); products IV, V, VI, IX, and X are given on plots (a–d); products XIV, XVII, and XVIII are given on plots (a′–d′). Product codes are those of Scheme 2: IV – 2-cyclohexenyl hydroperoxide, V – 2-ciclohexenol, VI– 2-cyclohexenone, IX – 3-cyclohexenyl hydroperoxide, X – 3-cyclohexenol, XIV – cis-2-hydroperoxycyclohexanol, XVIII – trans-1,2-cyclohexanediol, XIX – trans-2-hydroperoxycyclohexanol. Reaction conditions: 1–4 (10 μmol), TFA (0.1 mmol), C₆H₁₀ (2 mmol), H₂O₂ (10 mmol), CH₃CN (up to 5 mL of the total volume), 50°C. Concentrations of organic hydroperoxides were estimated by injecting the samples into GC before and after their treatment with PPh₃ (Shul’pin’s method, see refs. [25], [26], [27]), to allow the reduction of organic hydroperoxides to the corresponding alcohols.

Cyclohexene possesses three different functionalities that can be oxidized, namely the C_sp3−H, C_sp2−H, and C=C bonds (Scheme 2), all of them are susceptible to an attack of the hydroxyl radical that is believed to be a main oxidizing species in the present type of systems [38], [39], [40], [41], [42], [43], [44]. Additionally, the C=C double bond can undergo an epoxidation via a non-radical mechanism [45], [46], leading to epoxide and diol (mainly in acidic media). To explain the obtained results (Fig. 4), we proposed the following paths for the cyclohexene oxidation (Scheme 2).

The 2- and 3-cyclohexenyl hydroperoxides (IV and IX, hereinafter product codes are those of Scheme 2) are formed via a radical hydroperoxidation of C_sp2−H and C_sp3−H bonds with the participation of HO^• radical, via paths (a) and (b). The concentration of these products reaches a maximum in 1−2 h of the reaction (Fig. 4) and then decreases on prolonging the reaction time; a decomposition of these cyclohexenyl hydroperoxides occurs, leading to other products that were not detected by GC. Various carboxylic acids can be formed from alkyl hydroperoxides via the C–C bond cleavage at a high temperature of the GC injector, and we believe that alkenyl hydroperoxides could also show a similar type of behavior [25], [26], [27]. Diacids, oligomers and CO₂ could be generated. The peroxide (IV) is obtained in a higher amount than its isomer (IX), since it is formed via an allyl radical (II) that is more stable than the corresponding radical (VII), an intermediate during the formation of (IX). 2-Cyclohexenone (VI) can be obtained directly as a decomposition product of 2-cyclohexenyl hydroperoxide (IV) in the reaction medium, or even as a result of the overoxidation of 2-cyclohexenol (V). 3-Cyclohexenone (XI) was not identified, most likely due to its formation in a very small amount, since the oxidation at this position is much less favorable in comparison with an allylic oxidation.

Other products such as trans-2-hydroperoxycyclohexanol (XVII) and cis-2-hydroperoxycyclohexanol (XIV) were also found in the reaction medium, being probably obtained via a radical mechanism as a result of the HO^• attack at the C=C bond of cyclohexene. This reaction forms an intermediate radical (XII) (path C) that could be attacked by molecular oxygen under or over the ring, finally giving the two isomeruc organic peroxides, trans- (XVII) and cis-2-hydroperoxycyclohexanol (XIV). However, there is almost no information in the literature about the stability or the analysis of these peroxides [47], in contrast to alkyl peroxides that are known to decompose to alcohols and ketones in the reaction medium, or even in the GC column and/or injector [25], [26], [27]. Since these organic peroxide isomers, trans (XVII) and cis (XIV), start to disappear during the reaction after nearly 60–90 min (Fig 4, a′–d′), we initially believed that they were forming the corresponding trans-1,2-cyclohexanediol (XVIII) and cis-1,2-cyclohexanediol (XV) in the reaction medium. However, cis-1,2-cyclohexanediol (XV) was not observed. Despite its lower thermodynamic stability compared to the trans-isomer (XVIII) [44], we would expect to observe the formation of, at least, a minor amount of this cis-diol product. Thus, we believe that the trans- and cis-2-hydroperoxycyclohexanol do not selectively decompose in the reaction medium to their respective diols, but instead suffer stronger decomposition to less volatile compounds not detectable by GC (e.g. diacids, oligomers, or even CO₂). Therefore, the main possibility to explain the selective formation of the trans-diol would be via an epoxide hydrolysis, as shown in the path (d) (Scheme 2). Nevertheless, the epoxide was not observed in the reaction, but it could be formed and then immediately hydrolyzed to give trans-1,2-cyclohexanediol (XVIII).

To check this possibility, we performed the epoxidation of cyclooctene by H₂O₂ and using compound 3 as a catalyst (Fig. 5). Cyclooctene epoxide is known to be one of the most stable epoxides, because every reaction at the epoxide group is retarded by the steric hindrance of the cyclooctene ring. On the other hand, cyclohexene epoxide is a very reactive epoxide, since the steric influence of the cyclohexene ring promotes the ring opening of the epoxide [48]. For simplicity, we did not quantify the organic peroxides in the present reaction and the GC analysis was made only after the addition of PPh₃ (to reduce all organic peroxides). The obtained results (Fig. 5) show that cyclooctanediol is the major product, followed by 2-cycloocten-1-ol, and cyclooctene epoxide formed in a lower amount. The observation of the epoxide in the case of cyclooctene oxidation indirectly proves that the path (D) of Scheme 2 is viable, wherein the cyclohexene epoxide is formed as an intermediate leading then to a selective generation of trans-cyclohexanediol (XVIII).

Fig. 5:

Accumulation of products with time in the cyclooctene oxidation by H₂O₂ catalyzed by 3. Product codes: (a) 1,2-cyclooctanediol, (b) 2-cyclooctenol, (c) cyclooctene epoxide. Reaction conditions: 3 (10 μmol), TFA (0.1 mmol), C₈H₁₄ (2 mmol), H₂O₂ (10 mmol), CH₃CN (up to 5 mL of the total volume), 50°C.

Single-pot hydrocarboxylation of cycloalkanes

Apart from the oxidation reactions, compounds 1–4 were also tested in the hydrocarboxylation of cycloalkanes. This process consists of reacting a cycloalkane C_n with CO as a carbonyl source and H₂O as hydroxyl source, in the presence of a copper(II) catalyst and a peroxodisulfate anion S₂O₈²⁻ as an oxidant, to form in a single-pot a C_n+1 carboxylic acid [28], [29], [30], [31]. The reaction typically requires a CO pressure of 20 atm and proceeds at 60°C in water-acetonitrile medium (1:2, v/v).

In the presence of 1–4, the hydrocarboxylation of cyclopentane (Table 1), cyclohexane (Table 2), cycloheptane (Table 3), and cyclooctane (Table 4) results in the formation of one monocarboxylic acid as a major hydrocarboxylation product due to the presence of a single type of carbon atom in the cycloalkane substrate molecules. No formation of a dicarboxylic acid product neither the cycloalkane ring opening were detected. For all the tested substrates, the formation of the corresponding cyclic ketones and alcohols in minor amounts was also observed as a result of a partial alkane oxidation. The total yield of oxygenate products (ketone is formed predominantly to alcohol) increases with the hydrocarbon size.

Table 1:

Cu-catalyzed single-pot hydrocarboxylation of cyclopentane^a.

Cu catalyst	Product yield, %b
	Carboxylic acid	Ketone	Alcohol	Totalc
1	20.9	0.5	0.2	21.6
2	22.7	0.7	0.2	23.6
3	30.0	0.7	0.2	30.9
4	17.6	0.9	0.2	18.7

^aReaction conditions: cyclopentane (1.00 mmol), Cu catalyst (10 μmol), p(CO)=20 atm, K₂S₂O₈ (1.50 mmol), H₂O(2.0 mL)/MeCN(4.0 mL), 60°C, 4 h in an autoclave (20.0 mL capacity). ^b(Moles of products/mol of cyclopentane)×100%. ^cYield of all products.

Table 2:

Cu-catalyzed single-pot hydrocarboxylation of cyclohexane^a.

Cu catalyst	Product yield, %b
	Carboxylic acid	Ketone	Alcohol	Totalc
1	38.9	2.4	0.5	41.8
2	39.7	2.9	0.7	43.3
3	47.4	2.0	0.6	50.0
4	33.2	1.4	0.6	33.2

^aReaction conditions: cyclohexane (1.00 mmol), Cu catalyst (10 μmol), p(CO)=20 atm, K₂S₂O₈ (1.50 mmol), H₂O(2.0 mL)/MeCN(4.0 mL), 60°C, 4 h in an autoclave (20.0 mL capacity). ^b(Moles of product/mol of cyclohexane)×100%. ^cYield of all products.

Table 3:

Cu-catalyzed single-pot hydrocarboxylation of cycloheptane^a.

Cu catalyst	Product yield, %b
	Carboxylic acid	Ketone	Alcohol	Totalc
1	17.9	6.8	2.3	27.0
2	17.1	7.1	2.0	26.2
3	20.0	8.5	2.0	30.5
4	17.9	7.3	3.6	29.1

^aReaction conditions: cycloheptane (1.00 mmol), Cu catalyst (10 μmol), p(CO)=20 atm, K₂S₂O₈ (1.50 mmol), H₂O(2.0 mL)/MeCN(4.0 mL), 60°C, 4 h in an autoclave (20.0 mL capacity). ^b(Moles of product/mol of cycloheptane)×100%. ^cYield of all products.

Table 4:

Cu-catalyzed single-pot hydrocarboxylation of cyclooctane^a.

Cu catalyst	Product yield, %b
	Carboxylic acid	Ketone	Alcohol	Totalc
1	6.5	12.3	6.2	25.0
2	8.7	10.7	5.3	24.7
3	12.4	11.0	5.4	28.8
4	7.2	9.4	4.6	21.2

^aReaction conditions: cyclooctane (1.00 mmol), Cu catalyst (10 μmol), p(CO)=20 atm, K₂S₂O₈ (1.50 mmol), H₂O(2.0 mL)/MeCN(4.0 mL), 60°C, 4 h in an autoclave (20.0 mL capacity). ^b(Moles of product/mol of cyclooctane)×100%. ^cYield of all products.

Among the tested catalysts, compound 3 shows the highest activity in the hydrocarboxylation of all the studied cycloalkanes, resulting in the maximum yield of a carboxylic acid product in the case of cyclohexane (47% yield of cyclohexanecarboxylic acid; Table 2), followed by cyclopentane (30% yield of cyclopentanecarboxylic acid; Table 1) and cycloheptane (20% yield of cycloheptanecarboxylic acid; Table 3) [all yields are based on substrate]. The reaction with cyclooctane catalyzed by 3 leads to the formation of only 12% of cyclooctanecarboxylic acid (Table 4), while also generating a rather high amount of the oxidation products (~16% total yield of cyclooctanone and cyclooctanol).

In comparison with 3, compounds 2 and 1 exhibit an inferior activity which is, however, higher than that of 4 (least active catalyst) in the case of the cyclopentane and cyclohexane hydrocarboxylation. For example, the hydrocarboxylation of cyclopentane results in the formation of cyclopentanecarboxylic acid in 23, 21, and 18% yields when using 2, 1, and 4, respectively (Table 1), while cyclopentanone and cyclopentanol are formed in very low amounts in all cases (less than 1% total yield). Similarly, the hydrocarboxylation of cyclohexane leads to the formation of cyclohexanecarboxylic acid in 40, 39, and 33% yields for 2, 1, and 4, respectively (Table 2). Low amounts of cyclohexanone and cyclohexanol are also generated (2–3% total yield). A better activity of coordination polymers 1–3 in comparison with a discrete copper(II) dimer 4 can be associated with a higher stability of 1–3 due to their polymeric nature. The obtained products yields are higher or comparable to those achieved when using other Cu-based hydrocarboxylation catalysts [10], [29], [49], [50].

In the case of the cycloheptane and cyclooctane hydrocarboxylation, compounds 1, 2, and 4 show a resembling level of activity, resulting in 17–18% yields of cycloheptanecarboxylic acid and in 7–9% yields of cyclooctanecarboxylic acid. As mentioned before, these substrates more easily undergo the competing oxidation reactions, leading to the formation of rather high amounts of cyclic ketones and alcohols (14–18% total yields).

Based on the obtained results and previous literature background on the copper-catalyzed hydrocarboxylation of alkanes [10], [28], [29], [30], [31], a simplified free-radical mechanism for the hydrocarboxylation of C₅–C₈ cycloalkanes can be suggested. Hence, thermolysis of the S₂O₈²⁻ oxidant results in a powerful sulfate radical SO₄^•− that abstracts an H atom from a cycloalkane. The generated cycloalkyl radical R^• then reacts with CO to give an acyl radical RCO^•. This is oxidized by Cu(II) species to furnish an acyl cation RCO⁺ which is finally hydrolyzed by water to form a cycloalkanecarboxylic acid RCOOH as a major product. Besides, it is known that O₂ can very quickly react with cycloalkyl radicals and thus can be used as a radical trap [19]. Indeed, the addition of O₂ into the hydrocarboxylation system under standard reaction conditions resulted in a strong suppression of the yield of cycloalkanecarboxylic acid (cyclohexane was selected as a model substrate), while the yields of cycloalkane oxidation products increased. This additional test also supports a radical nature of the reaction mechanism.

Conclusion

The current study has widened the use of copper(II) coordination polymers in oxidation catalysis], also contributing to the area of aqueous homogeneous catalysis [10], [19], [20], [21], [22], [23], [30], [49], [51]. In particular, the mixed-ligand aminoalcohol-dicarboxylate compounds 1−4 were applied as rather efficient homogeneous catalysts for the oxidative functionalization of saturated and unsaturated C₅−C₈ cyclic hydrocarbons.

In the oxidation of cyclohexane by hydrogen peroxide, compounds 1−4 show a comparable level of activity to give a mixture of cyclohexyl hydroperoxide, cyclohexanol, and cyclohexanone in up to 22% total yields based on substrate. A catalytic efficiency of 1−4 is superior in the mild hydrocarboxylation of cyclohexane by CO, H₂O, and K₂S₂O₈, resulting in up to 47% yields of cyclohexanecarboxylic acid (major carboxylation product) along with the formation of a minor amount (~3%) of cyclohexanol and cyclohexanone as oxidation products. In this system, the highest activity was shown by coordination polymer 3, followed by compounds 2, 1, and 4. Interestingly, catalyst 3 is also the most active in the hydrocarboxylation of cyclopentane, cycloheptane, and cycloalkane, leading to 30, 20, and 12% yields of the corresponding cycloalkanecarboxylic acids, respectively. Despite not being very high in some cases, these yields are good in the field of mild alkane functionalization [1], [2], [3], [4], [10], particularly in view of the alkane’s high inertness and mild reaction conditions applied. It should also be mentioned that even the industrial processes involving alkanes as substrates are carried out with low conversions (often not exceeding 10%) to minimize an overoxidation of the formed products that are typically more reactive than the parent alkane substrate [2], [52].

Another interesting feature of the tested catalytic systems concerns their broad substrate versatility that can also be extended from alkanes to alkenes. In fact, compounds 1−4 are also capable of catalyzing the oxidation of cyclohexene and cyclooctene, resulting in the generation of various oxidation products in up to ~18% yields based on substrate. A complex mixture of oxidation products is a result of the oxidation at the C_sp3−H, C_sp2−H, and C=C bonds, in addition to the epoxidation of the latter. Hence, for both the cycloalkane and cycloalkene oxidation reactions, the obtained product distribution profiles, along with a related literature background, suggest the free-radical reaction mechanisms that operate with the participation of hydroxyl radicals as a main oxidizing species.

Further work on extending the substrate scope in the present hydrocarbon C−H functionalization reactions along with the optimization of the reaction conditions will be pursued, namely by using a variety of mixed-ligand copper(II) coordination polymers and related derivatives as catalysts. Another research direction will also focus on the heterogenization of the selected copper(II) catalysts onto a solid support [53], [54], followed by an application of the obtained materials as heterogeneous catalysts.