Synthesis, Crystal Structure, and Electrochemistry of Mono- and Bis-Homoannular Ferrocene Derivatives

1. Introduction

Ferrocene and its derivatives play important roles in modern-day materials chemistry owing to their air stability, aromatic reactivity, reversible electrochemistry, and low toxicity [1]. After the discovery of ferrocene in 1951 [2], research was focused on the conjugation of carbocyclic rings to one or both cyclopentadienyl rings. The most straightforward method of such conjugation was the cyclization of ω-ferrocenylaliphatic acids. The ferrocenylaliphatic acids containing four or five carbon atoms on their side chain undergo electrophilic ring-closing to give homoannular ferrocene derivatives under electrophilic conditions [3]. Such ferrocene derivatives with unsymmetrically substituted cyclopentadienyl rings represented prototypes of planar chirality in metallocene chemistry due to the “sandwich” structure of the ferrocenyl moiety [4].

With our continuous interest in extending the π-conjugation of metallocenes [5], we synthesized 1’,2’,3’,4’,5’-pentamethylruthenocene-fused quinones from the double Friedel-Crafts acylation between metallocene-1,2-diacylchloride and organic aromatics [6]. Later, we realized that switching the functionality of two reaction partners can give such complexes in a much simpler method. Since ferrocene is an excellent nucleophile for Friedel-Crafts acylation, its reaction with succinic anhydride can eventually lead us to synthesize ferrocene-fused ring(s). These compounds can serve as the simplest possible model for extending the π-conjugation of metallocene. Further, ferrocene and its derivatives undergo reversible electrochemistry. Numerous studies have been concerned with the effects of substituents on the redox potential of the ferrocenyl moiety [7,8,9,10]. In general, electron-releasing substituents lower the oxidation potentials and electron-withdrawing substituents increase the potential in comparison to unsubstituted ferrocene [11]. In this contribution, we report the synthesis of 1,2-(tetramethylene)-ferrocene and 1,2,1’,2’-bis(tetramethylene)-ferrocene starting from ferrocene, their crystal structures, and their half-wave oxidation potentials.

2. Materials and Methods

2.1. General Procedures

Reactions were carried out using standard Schlenk line techniques under nitrogen unless otherwise mentioned. Solvents were used as received without further drying or purification. Ferrocene, succinic anhydride, aluminum chloride, zinc powder, mercuric chloride (Alfa Aesar), and trifluoroacetic anhydride (Oakwood Chemicals) were used as purchased. The organic phases were dried using anhydrous magnesium sulfate (Mallinckrodt). Flash chromatography was performed using 60-Å pore size, 230 x 400 mesh silica gel (Sorbent Technologies).

¹H and ¹³C NMR spectra were recorded on a JEOL – 400 ESZ spectrometer at ca. 22 °C and were referenced to residual solvent peaks. Infrared spectra were recorded on Bruker Alpha-E FT-IR spectrometer with a diamond crystal ATR accessory. Melting points were taken on a Thomas-Hoover capillary melting point apparatus and were uncorrected. The oxidation potentials of the compounds were measured with BASi Epsilon – Electrochemical Workstation.

2.2. X-ray Crystallography

X-ray diffraction data were measured at T = 90 K on a Bruker Kappa Apex-II diffractometer equipped with a sealed-tube MoKα source (λ=0.71073 Å), a Triumph focusing monochromator and a CCD detector. Structures were solved using SHELXT [12] and refined using SHELXL [13]. Hydrogen atoms were visible in difference maps and were placed in idealized positions during refinement and treated as riding. The structure of 3b’ had a disorder involving partially occupied carbonyl groups at CH₂ sites. The crystal structures and refinement data are presented in Table 1.

2.3. Experimental Procedures

Synthesis of C₁₄H₁₄FeO₃ (1a). In a 500 mL flask, a mixture of succinic anhydride (1.83 g, 18.1 mmol) and aluminum chloride (4.20 g, 32.1 mmol) suspended in dichloroethane (80 mL) was added to the solution of ferrocene (3.02 g, 16.3 mmol) in dichloroethane (40 mL) over 45 minutes. The reaction mixture was allowed to stir for an additional 2 hours at room temperature and then poured into water (50 mL). The organic phase was collected. The product was extracted in 1.0 M NaOH (3 x 20 mL) and acidified with concentrated HCl until the pH of the solution dropped below 7. The precipitate was collected by filtration, washed with water, and dried to give 1a (1.78 g, 38%) as an orange powder. Melting point: 165–166 °C (Lit. [14] 164 – 165). ¹H NMR (400 mHz, CDCl₃ ): δ 2.72 – 2.76 (t, 2 H, ³J = 6.8 Hz), 3.06 – 3.09 (t, 2 H, ³J = 6.4 Hz), 4.23 (s, 5 H, Cp), 4.51 – 4.52 (t, 2 H, ³J = 2.4 Hz), 4.80 – 4.81 (t, 2 H, ³J = 2.0 Hz). IR (ATR, cm ^{– 1}): 1705 (C=O), 1657 (C=O).

Synthesis of C₁₈H₁₈FeO₆ (1b). In a 500 mL flask, succinic anhydride (10.8 g, 108 mmol) and anhydrous aluminum chloride (28.8 g, 216 mmol) were mixed in dichloroethane (80 mL). Ferrocene (10.0 g, 53.8 mmol) dissolved in dichloroethane (80 mL) was added dropwise into the flask using a dropping funnel. The reaction was stirred for 48 hours at room temperature and poured into ice water (100 mL). The solutions were layered, and the organic phase was collected. The aqueous phase was extracted with dichloromethane (100 mL). The product was extracted in 2 M sodium hydroxide (3 x 100 mL). The aqueous phase was acidified with conc. HCl until precipitation was completed. The precipitate was collected by filtration to give a crude product. The crude product was suspended in boiling water (100 mL) and filtered to give 1b (14.9 g, 72%) as a dark red powder. Melting Point: 178 – 179 °C (Lit. [15] 179 °C). ¹H NMR (400 mHz, acetone-d₆, ppm): δ 2.60 – 2.64 (t, 4 H, ³J = 6.4 Hz), 3.02 – 3.05 (t, 4H, ³J = 6 Hz), 4.62 – 4.63 (t, 4H, ³J = 3.6 Hz), 4.89 – 4.90 (t, 4 H, ³J = 2.4 Hz).

Synthesis of C₁₄H₁₆FeO₂ (2a). In a 250 mL flask, zinc (6.0 g, 91.8 mmol), mercuric chloride (0.60 g, 2.21mmol), DI water (20 mL), and concentrated HCl (0.30 mL) were mixed for five minutes by hand, then five minutes with a stir bar. The suspension was allowed to settle, the supernatant was decanted, washed with deionized water, then transferred into a 500 mL flask along with concentrated HCl (8.8 mL), toluene (20 mL), and compound 1a (2.0 g, 7.00 mmol). The reaction was refluxed for 6 h. The reaction was cooled to room temperature and diluted with H₂O (20 mL). The product was extracted in ethyl ether (3 × 10 mL), then the organic layer was washed with H₂O (2 × 10 mL) and dried with anhydrous magnesium sulfate. The volatiles were removed in vacuo. The crude product was triturated with petroleum ether to give 2a (1.03 g, 54%). Melting Point: 115 – 116 °C (Lit. [14] 115 – 116 °C). ¹H NMR (400 MHz, CDCl_3, ppm): δ 1.80 – 1.87 (m, 4 H), 2.35 – 2.40 (m, 8 H), 3.98 – 4.09 (m, 8 H).

Synthesis of C₁₈H₂₂FeO₄​ (2b). In a 250 mL flask, zinc (12.0 g, 184 mmol), mercuric chloride (1.20 g, 4.42mmol), DI water (20 mL), and concentrated HCl (0.60 mL) were mixed for five minutes by hand, then five minutes with a stir bar. The suspension was allowed to settle, the supernatant was decanted, washed with DI water, then transferred into a 500 mL flask along with H₂O (8 mL), concentrated HCl (17.6 mL), toluene (20 mL), and compound 1b (4.0 g, 10.4 mmol). The reaction was refluxed for 19 hours. The reaction was cooled to room temperature and diluted with H₂O (20 mL). Unreacted zinc was separated by decantation and the residue was washed with diethyl ether (30 mL). The organic layer was collected, and the aqueous phase was extracted with diethyl ether (2 x 50 mL). The combined organic phase was dried with anhydrous magnesium sulfate. Volatiles were removed in vacuo. The final product was purified by trituration with petroleum ether to give 2b (3.36 g, 90%) as a yellow powder. Melting Point: 72 – 73 °C (Lit. [15] 73 °C). ¹H NMR (400 mHz, CDCl₃, ppm): δ 1.80 – 1.87 (m, 4H), 2.35 – 2.40 (m, 8H), 43.98 – 4.09 (m, 8H).¹³C{¹H} NMR (100 MHz, CDCl₃, ppm): δ 25.95, 28.58, 33.64, 67.89, 68.88, 77.32, 88.08, 179.97. IR (ATR, cm^–1): 1703 (C=O), 3000 – 3200 (OH).

Synthesis of C₁₄H₁₄FeO (3a). In a 100 mL flask, trifluoroacetic anhydride (0.620 mL) was mixed with dichloromethane (33 mL). A mixture of compound 2a (1.00 g, 3.67 mmol) and dichloromethane (33 mL) was added dropwise to the flask via a dropping funnel and the reaction was allowed to stir for 8 hours at room temperature. A saturated solution of NaHCO₃ (10 mL) was added and the mixture was layered in a separatory funnel. The organic layer was collected, dried with anhydrous MgSO₄ and volatiles were removed in vacuo to give 3a (0.758 g, 81%). Melting Point: 82 – 83 °C (Lit. [16] 83 – 85 °C) ¹H NMR (400 MHz, CDCl₃, ppm): δ 2.02 – 2.22 (m, 2H), 2.25 – 2.45 (m, 2H), 2.59 – 2.67 (m, 2H), 4.17 (s, 5H, Cp), 4.45 – 4.47 (m, 2H, Cp), 4.81 – 4.82 (m, 1H, Cp). ¹³C{ NMR (100 MHz, acetone-d₆, ppm): δ 23.4, 23.9, 38.8, 64.6, 70.0, 75.8, 92.5, 202.4. IR (ATR, cm⁻¹): 1664 (C=O). The product was analyzed by single crystal X-ray analysis.

Synthesis of C₁₈H₁₈FeO₂ (3b and 3b’). In a 100 mL flask, 8 (0.501 g, 1.40 mmol) was dissolved in dichloromethane (15 mL). Trifluoroacetic anhydride (0.778 mL, 5.59 mmol) in dichloromethane (15 mL) was added dropwise. The reaction was stirred for 20 hours at room temperature. Sodium bicarbonate solution (10%, 100 mL) was added to the mixture and layered with dichloromethane. The organic layer was collected, washed with water (2 × 20 mL), dried with magnesium sulfate, and filtered. The filtrate was evaporated to dryness to give 3b and 3b’ (0.190 g, 42%) as a bright red powder. The two diastereomers were separated by column chromatography in silica using a mixture of ethyl acetate and dichloromethane (2:1 by volume). The racemic mixture (dark red), 3b’ was followed by the meso (dark orange), 3b in the column. 3b (Meso): Melting Pont: 168 -189 °C (Lit. [17] 161 -167 °C. ¹H NMR (400 mHz, CDCl₃): δ 2.00 -2.11 (m, 4H, CH2), 2.15 – 2.40 (m, 6H, CH2), 2.56 – 2.66 (m 4H CH2), 4.42 – 4.45 (m, 2H, Cp), 4.73 (b, 2H, Cp). ¹³C{ NMR (100 MHz, CDCl₃, ppm): δ 22.92, 23.66, 39.27, 67.38, 72.37, 73.19, 76.58, 93.42, 204.33. IR (ATR, cm⁻¹): 1661 (C=O). The product was characterized by single crystal X-ray analysis. 3b’(Racemic): Melting Point: 156 - 157 °C (Lit. [17] 153 – 161 °C). ¹H NMR (400 mHz, CDCl₃): δ 2.01 – 2.15 (m, 2 H), 2.21 – 2.47 (m, 6H), 2.61 (dt, ²J = 15.6 Hz, ³J = 4.4 Hz, 4 H), 4.39 (d, ³J = 1.2 Hz, 2H), 4.42 – 4.43 (m, 4 H), 4.65 (t, 3J = 1.2 Hz, 2 H). ¹³C{ NMR (100 MHz, CDCl₃, ppm): δ 23.32, 23.52, 39.20, 68.03, 70.84, 71.85, 77.32, 94.00, 203.40 IR (ATR, cm ^{– 1}): 1663 (C=O). The product was analyzed by single crystal X-ray analysis.

Synthesis of C₁₄H₁₆Fe (4a). In a 250 mL flask, zinc (5.0 g, 76.4 mmol), mercuric chloride (0.5 g, 1.84 mmol), and DI water (20 mL) were mixed for 10 min. The suspension was allowed to settle, the supernatant was decanted, and the residue was washed with DI water. To the amalgamated zinc, concentrated HCl (2.0 mL), toluene (20 mL), and compound 3 (200 mg, 0.787 mmol). The reaction was refluxed for 2 h and an additional amount of conc. HCl (2.0 mL) was added and continued reflux for another 3 h. The reaction was cooled to room temperature, the clear solution was decanted, and the unreacted zinc was washed with ethyl acetate (40 mL). The combined organic layer was washed with water (3 x 100 mL), dried with anhydrous MgSO₄, and filtered. The volatiles were removed in vacuo to give 4a (166 mg, 87%) as a yellow solid. Melting point: 39 °C (Lit. [18] 39 – 41 °C) ¹H NMR (400 MHz, Acetone-d₆, ppm): δ 1.53 – 1.64 (m, 2H), 1.85 – 1.92 (m, 2H), 2.17 – 2.26 (m, 2H), 2.60 – 2.67 (m, 2H), 3.85 (t, ³J = 2.4 Hz, 1H, CHCHCH), 3.94 (d, ³J = 2.4 Hz, 2H, CHCHCH), 3.95 (s, 5H, Cp). ¹³C{¹H}, acetone-d₆, ppm): δ 23.5, 24.6, 64.8, 65.1, 69.4, 84.9.

Synthesis of C₁₈H₂₂Fe​ (4b). In a 100 mL flask, a mixture of zinc (5.0 g, 76.47 mmol), mercuric chloride (0.50 g, 1.84 mmol), and DI water (20 mL) was stirred for 10 min. The clear solution was decanted and the residue was washed with DI water (20 mL). To the activated Zn, the diastereomeric mixture of 3b and 3b’ (200 mg, 0.62 mmol), toluene (20 mL), and conc. HCl (2.0 mL) was added. The reaction was refluxed for 3 h and an additional amount of conc. HCl (4.0 mL) was added. The reaction was again refluxed for 2 h. The mixture was cooled to RT and the soluble product was decanted. The residue was washed with ethyl acetate (2 x 10 mL). The extract was washed with water (2 x 80 mL); the organic layer was collected, dried over anhydrous MgSO₄, and filtered. Volatiles were removed in vacuo. The crude product was purified by silica gel column chromatography using petroleum ether as eluent to give 4b (107 mg, 59%) as yellow gum. ¹H NMR (400 mHz, acetone-d₆): δ 1.53 – 1.57 (m, 4H), 1.85 – 1.88 (m, 4H), 2.16 – 2.23 (m, 4H), 2.57 – 2.64 (m, 4H), 3.65 (t, ³J = 2.4 Hz, 2H, CHCHCH), 3.76 (d, 3J = 2.4 Hz, 4H, CHCHCH).¹³C{ NMR (100 MHz, acetone-d₆, ppm): δ 23.6, 23.9, 67.5, 68.3, 84.2.

3. Results and discussion

3.1. Synthesis and Structural Elucidation of Compounds 1a, 1b – 4a, 4b

Synthesis of the target compounds (4a and 4b) was performed by following the reaction sequence as shown in Scheme 1. Our attempts to reproduce the procedures reported in the literature [19,20] resulted in the formation of a mixture of both mono-and bis(3-carboxypropenoyl)ferrocene in a ca. 1:1 ratio. To avoid the tedious separation of these two keto-carboxylic acids, we synthesized (3-carboxypropenoyl)ferrocene, 1a (38%) by slowly adding a suspension of succinic anhydride and aluminum chloride in dichloroethane to the solution of ferrocene in the same solvent. The procedure left some unreacted ferrocene, but the product was purified by solvent extraction using NaOH solution. Synthesis of bis(3-carboxypropenoyl)ferrocene, 1b (72%) was performed by adding ferrocene into the suspension of succinic anhydride and aluminum chloride and stirring the solution at room temperature for 48 hours. After the base extraction, the product was found to contain 1a as an impurity which was removed by washing it with hot water. The Clemmensen reduction of both keto carboxylic acids was performed by following the literature procedures [14,15] with slight modifications to give mono- and bis(3-carboxypropyl)ferrocene 2a (54%) and 2b (90%), respectively.

The ring closing of 2a and 2b was performed in the presence of trifluoroacetic anhydride [17,21] to give homoannularly cyclized products: 1,2(α-ketotetramethylene)ferrocene, 3a (81%) and 1,2,1’,2’-bis(α-ketotetramethylene)ferrocene, 3b (42%) respectively. Although Nesmeyanov[22] and Rinehart [3,17] independently reported the synthesis of 3a and 3b, their full characterization data were not available. The diketone 3b must exist as two geometric isomers (racemic and meso) but the authors were unable to assign the configurations of the diastereomers [17]. Based on ¹H NMR analysis of the crude product, we observed the formation of these isomers in ca. 1:1 ratio. Assignment of the stereochemistry of these geometrical isomers (3b and 3b’) was performed by single crystal X-ray analysis (vide infra). The ¹H NMR of 3a, 3b and 3b’ exhibited a distinctive ABC pattern of substituted cyclopentadienyl rings. Similarly, the diastereotopic nature of exo- and endo-protons of the methylene groups exhibited a complex coupling pattern giving a set of multiplets.

The keto groups of 3a, 3b, and 3b’ were reduced to methylene under Clemmensen conditions to give 1,2-(tetramethylene)-ferrocene 4a (87%) and 1,2,1’,2’-bis(tetramethylene)-ferrocene 4b (59%), respectively. The loss of carbonyl stretching in IR and the appearance of triplets (1H) and doublets (2H) on the substituted cyclopentadienyl rings in ¹H NMR indicated the formation of desired products. Like ketones (3a, 3b, and 3b’), the diastereotopic nature of the exo- and endo-hydrogens of the methylene group gave a complicated coupling pattern. Compound 4b exhibited four sets of CH₂ protons and two sets of cyclopentadienyl protons indicating the free rotation of two Cp rings with respect to the Fe(II) center. Moreover, both signals of cyclopentadienyl rings are shifted upfield for both 4a and 4b but the shift is more prominent for 4b (triplet at 3.65 and doublet at 3.75 ppm). Compounds 4a and 4b were previously synthesized by the catalytic hydration of (cyclopentadienyl)(indenyl)iron [18] or bis(indenyl)iron [23]. Synthesis of 4b was also reported from the condensation of spiro[4·4]nona-1,3-diene with iron[24] as well as the treatment of the lithium salt of 4,5,6,7-tetrahydroindene with FeCl₂ [25].

3.2. X-ray Crystal Structures

The structures of ferrocene derivatives 3a, 3b, 3b’, and 4a were determined by X-ray crystallographic methods. All crystals were grown by slow evaporation of concentrated ethyl ether solution in air at room temperature. Thermal ellipsoid plots with numbering schemes are shown in Figure 1, Figure 2, Figure 3 and Figure 4. The crystal structure and refinement data for these compounds are given in Table 1. The crystal structure of 3a has been reported by Fleischer, et al. at room temperature [26]. We have redetermined its crystal structure at 90 K with a much higher precision level. The compound 3a exhibits planar chirality since two different substituents are connected to the cyclopentadienyl ring [27]. As the compound crystallizes in centrosymmetric space group (P−1), the two enantiomers Rp/Sp are present in equal numbers within the crystals. The iron atom in compound 3b lies on a crystallographic inversion center. Compound 3b’ crystallizes with Z’ = 2 in which one of the molecules has a disorder of O atom in two positions. The 3b’ crystallizes as a racemic mixture in a centrosymmetric space group P2₁/n. The two molecules have a roughly perpendicular orientation. Each molecule possesses an approximate crystallographic two-fold rotational axis (Figure 3). Compound 4a crystallizes in the monoclinic space group P2₁/c with one molecule per asymmetric unit.

In the molecules of 3a, 3b, and 3b’ the ferrocenyl moiety is fused with α-keto tetramethylene rings with one of both cyclopentadienyl rings while in the molecules of 4a, the ferrocenyl moiety is fused with tetramethylene ring. In all molecules, the angle between Fe and two Cp ring centroids is nearly 180° with a maximum deviation of 4.85° from linear geometry in 3b’. The Cp rings in the ferrocene system are almost parallel; as the dihedral angle between the planes of two Cp rings in 3a, 3b’, and 4a are 1.84°, 3.46° (average of two), and 1.92°, respectively. The Cp rings display nearly eclipsed conformation on 3a, 3b’and 4a, as demonstrated by C – Cg1 – Cg2 – C average torsion angles of 9.12°, 0.26°, and 6.12° respectively. The Cp rings in 3b are in perfectly staggered conformation (torsional angle 36.00°) as required by the inversion center located at Fe.

Within the ferrocenyl moieties, the mean Fe – C bond distances for unsubstituted Cp are unremarkable. However, the Fe–C distance for the ipso carbon-bearing keto group next to it is slightly shorter than the rest of the distances. For instance, Fe1 – C1 bond in compound 3a is 2.0397(9) Å while the average distance of Fe1 with C2 to C5 is 2.0505(7) Å. The slippage of the Fe center towards C1 has been observed in other α-keto ferrocenyl derivatives [28,29]. The slipping of Fe1 towards C1 presumably occurs to maximize the interaction of the Fe center with the exocyclic double bond in the resonance of such ketones [30]. The attached carbonyl group lies almost co-planar with the plane of the substituted cyclopentadienyl ring in 3a 3b and 3b’ as given by the torsional angle [C5-C1-C6-O1 = 1.52(11)°] in 3a. The C=O bond length in 3a 1.2265(9) Å, 3b 1.2230(9) Å, and 3b’ 1.225 (17) Å are in normal range of similar α-ferrocenyl ketones.[31,32] In 4a, the average of Fe1-C1 and Fe1-C2 distances is higher by 0.023(3) Å than the rest of Fe-C bonds of the substituted cyclopentadienyl ring. The distance of the Fe center from the centroid of cyclopentadienyl rings ranges from 1.647 Å to 1.660 Å.

The six-membered rings in all structures adopt half-chair conformations. The Cremer & People puckering parameters [33] of the six-membered rings, namely, 3a, 3b, 3b’ and 4a are Q_T = 0.4340(8) Å, θ = 131.43(11)° and φ = 2.47(14)°; Q_T =0.4227(8) Å, θ = 129.68(10)° and φ = 358.42(13)°; Q_T = 0.4356(12) Å, θ = 53.54(14)° and φ = 178.20(18)°; and Q_T = 0.521(3) Å, θ = 51.7(3)° and φ = 219.0(4)°, respectively. In 3a, atoms C6/C7/C9 lie on the same plane of the substituted cyclopentadienyl ring with a dihedral angle of 1.80°. The C8 projects inward from the main plain of Cp by 0.582 Å. Similar folding of six-membered rings can be seen in 3b and 3b’ as well. The two α-ketotetramethylene groups are oriented with dihedral angles of 180° and 72° in 3b and 3b’, respectively. The six-membered ring of 4a is more twisted than in 3a or 3b. In this molecule, C6 and C9 are almost coplanar with the substituted Cp [deviation: C6 = 0.018 Å, C9 = 0.033 Å]. The C7 is projected down by 0.498 Å and C8 is projected up by 0.305 Å from the Cp plane.

In their crystal structures, compounds 3a, 3b, and 3b’ display similar intermolecular interactions. In these molecules, the most prominent interactions are weak intermolecular C-H···O hydrogen bonds and C-H···π interactions. For example, molecules of 3b exhibit intermolecular C4-H4···O1, C-H5···O1 interactions along the crystallographic b-axis. Similarly, atom H8A is positioned almost perpendicular above the cyclopentadienyl ring centroid of the adjacent molecule (Figure 5 (I). In the crystal structure of 4a, there are C-H···π interactions between methylene hydrogen and the cyclopentadienyl rings (Figure 5 (II)).

Figure 5. (I) Packing diagram of 3b along the b axis; (II) packing of 4a along a axis (II).

Figure 5. (I) Packing diagram of 3b along the b axis; (II) packing of 4a along a axis (II).

3.3. Electrochemical studies

To investigate the effects of tetramethylene substituents on the oxidation potential of the ferrocene moiety, we measured the half-wave redox potentials (E_1/2) of 4a and 4b by cyclic voltammetry using 0.1 M tetrabutylammonium hexafluorophosphate in dichloromethane as a supporting electrolyte at a scan rate of 50 mV s⁻¹ in 10⁻³ M concentration. All measurements were carried out at room temperature under a dry nitrogen atmosphere by the use of a three electrode system: glassy carbon as the working electrode, Ag/AgCl as the reference electrode, and platinum wire as a counter electrode.

Figure 5. Cyclic voltammogram of Ferrocene (red), 4a (blue), and 4b (purple).

Figure 5. Cyclic voltammogram of Ferrocene (red), 4a (blue), and 4b (purple).

Like ferrocene, the cyclic voltammetry of 4a and 4b shows a single-electron reversible redox process (Figure 5). As expected, the redox potentials of both complexes were lower than that of ferrocene due to the electron-releasing ability of tetramethylene rings. The higher electron density at the iron center due to the tetramethylene ring causes the iron center to lose an electron more easily in comparison to ferrocene [34]. Figure 6 shows a plot of E_1/2 vs. the number of tetramethylene groups. Under the experimental conditions, the E_1/2 of ferrocene, 4a and 4b are 499 mV, 303 mV, and 183 mV, respectively versus Ag/Ag⁺. Although there is a sharp decrease in the potential, the relationship between the number of substituents and oxidation potential is not linear. The difference in potential between ferrocene and 4a is 196 mV, while the difference between 4a and 4b is just 120 mV. The number of tetramethylene bridges in 4b is double that of 4a. However, the expected decrease in potential is less than half. A slightly higher potential of 4b than its predicted value might be due to the steric bulk of additional tetramethylene rings around the iron center that render the interaction of the iron atom with the electrode difficult [35].

Figure 6. Effects of substituents in the redox potential of ferrocene derivatives.

Figure 6. Effects of substituents in the redox potential of ferrocene derivatives.

4. Conclusions

We have synthesized and characterized mono- and bis-tetramethyleneferrocenes as viable precursors of π-extended ferrocene derivatives. Four compounds of the reaction sequence have been studied with single-crystal X-ray analysis. The effects of tetramethylene groups on the half-wave oxidation potential have been studied by cyclic voltammetry. Our attempts to dehydrogenate the final products of this reaction sequence with 2,3-Dichloro-5,6-dicyano-1,4-benzoquinone (DDQ) were unsuccessful. Currently, we are working on alternative methods to aromatize the ferrocene-bound aliphatic rings.