Ferrocene and [3]ferrocenophane-based β-diketonato copper(II) and zinc(II) complexes: synthesis, crystal structure, electrochemistry and catalytic effect on thermal decomposition of ammonium perchlorate

Abstract

Ferrocene and [3]ferrocenophane-based β-diketones and their Cu(II) and Zn(II) complexes have been synthesized and characterized. Single-crystal X-ray diffraction analysis confirmed the molecular structures of the Cu(II) complexes. UV-vis spectroscopy and electrochemical measurements were obtained. All the new compounds showed quasireversible and diffusion-controlled redox processes. The complexes displayed high thermal stability according to the thermogravimetry (TG) measurements. The β-diketone ligands and complexes exhibited high catalytic effects on the thermal degradation of ammonium perchlorate (AP) which was evaluated by TG and differential scanning calorimetry techniques, whereas the addition of Cu(II) complexes lowered the thermal decomposition temperature of AP more dramatically and released more heat compared with the addition of β-diketone ligands or Zn(II) complexes. We expect that this kind of transition metal complexes containing ferrocenyl would has great value in preparing a high burning rate catalyst for composite solid propellants.

---

1 Introduction

As is well known, ammonium perchlorate (AP) is a common oxidizer in composite solid rocket propellants due to its excellent burning characteristics, good processability and storability,¹ generally accounting for 60–90% of the total mass of the propellant.² The thermal decomposition characteristics of AP directly influence the burning velocity and energy features of propellants.³ The thermal decomposition of AP includes low- and high-temperature modes of decomposition. The low-temperature decomposition (LTD) occurs below approximately 300 °C and results only in ∼30% decomposition. The high-temperature decomposition (HTD) is observed above 300 °C and leads to complete gasification of AP.^4,5 Lower HTD temperature of AP results in a shorter ignition delay time, a higher burning rate (BR) of these propellants and an enough driving force for the rocket.⁶ Considering its limitation in loading in composite solid propellants, it is important to improve decomposition efficiency of AP to satisfy the requirements of high energy generation at low burning temperatures. Therefore, a variety of catalysts have been investigated including nano-metal particles,^7–9 metal chelates,¹⁰ transition metal oxides^11,12 and ferrocene derivatives^13–15 to anticipate their catalytic properties for the thermal decomposition process of AP.

As high BR catalysts for composite solid propellants, ferrocene derivatives are superior to other transition-metal compounds.¹⁶ However, commercially available ferrocene-based BR catalysts such as n-butylferrocene, tert-butylferrocene, and 2,2-bis(ethylferrocenyl)propane (catocene) tend to migrate out of the binder matrix of the propellants into the insulation material, which leads to uneven combustion and reduces the resistance to aging with the formation of highly sensitive boundary layers.^17,18 Furthermore, the migration affects the pourability of the propellant slurry to a significant degree and also changes the mechanical and the ballistic properties of the finished propellant.¹⁹ To preclude these problems and also to enhance the efficiency of ferrocene-based BR catalysts, specially designed ferrocene based BR catalysts, such as ferrocene-containing polymers,²⁰ polar groups functionalized ferrocene compounds²¹ and ionic ferrocene compounds^22–24 were developed. Recently, transition-metal complexes containing ferrocenyl attracted significant interest, due to their excellent thermal stability and catalytic properties.^25–29 Among these, the transition metal complexes of ferrocene-containing β-diketones as polynuclear ferrocenyl derivatives, widely used in conjugate chemistry,³⁰ electrochemical studies³¹ and medicinal chemistry³² etc. are supposed to be potentially high BR catalysts due to containing large polar carbonyl and metal atoms.³³ On the other hand, bridged ferrocenophanes with a minimal structural variation from ferrocenes showed enhanced redox properties. Hence, we expect that incorporation of a ferrocenophane unit into the transition metal complexes can bring certain benefit to improving catalytic efficiency in composite solid propellants. However, to the best of our knowledge, there are no reports about the synthesis and properties of ferrocenophane-based transition metal complexes so far. In the present paper, we report the synthesis of ferrocene and [3]ferrocenophane-based β-diketonato and their Cu(II) and Zn(II) complexes (Fig. 1), and the catalytic effects of the complexes on the thermal decomposition of AP was researched in order to confirm their potential BR catalytic activity.

Fig. 1 The structures of ligands and complexes.

2 Experimental

2.1 Materials and measurements

Detailed procedures of synthesis and characterization of β-diketone ligands Fc8–Fc16 and Fp8–Fp16 were described in ESI.† IR spectra were recorded as KBr pellets on a Bruker-ALPHA spectrometer. NMR spectra were recorded on an Avance 500 Bruker (500 MHz) spectrometer using tetramethylsilane as internal standard. MS spectra were recorded on a Varian QFT-ESI mass spectrometer. Elemental analysis was performed on a Perkins-Elmer 2400 elemental analyser. Electronic absorption spectra were recorded on Shimadzu UV2600 spectrophotometer. Thermal stability and catalytic activity of synthetic compounds were measured by SDT Q600 V20.5 Build 15 simultaneous thermogravimetric and differential scanning calorimetry (TG-DSC) measurement under N₂. Cyclic voltammetry (CV) experiments were recorded on AUTOLAB PGSTAT302 voltammetric analyzer and performed at room temperature in dry CH₂Cl₂ solutions with the concentration of 10⁻³ M containing 0.1 M tetra-n-butylammoniumhexafluorophosphate (TBAPF₆) as a supporting electrolyte. A three-electrode configuration consisting of a glassy carbon working electrode, a Pt wire counter electrode, and an Ag/AgCl (with a saturated KCl solution) couple reference electrode was used. The ferrocene/ferrocenium (Fc/Fc⁺) couple was used as internal reference and showed a peak at +0.484 V vs. Ag/AgCl. Prior to experiments, the system was purged with purified nitrogen gas to exclude dissolved oxygen from the solution.

Yellow single crystal of Fc8Cu, suitable for X-ray diffraction analysis, was obtained by recrystallization from CH₂Cl₂. Data were collected on a Bruker Smart APEX II CCD area detector with a MoKα radiation (λ = 0.71073 Å) at 294 K in the φ–ω scan mode. The structure was determined by direct method (SHELXS-97) and successive Fourier synthesis and refined by using the SHELXL-97 program.³⁴ The nonhydrogen atoms were refined anisotropically, whereas the hydrogen atoms were placed in calculated positions and refined isotropically. Crystal data and structure refinement parameters are listed in Table S1.† Selected bond distances and angles are listed in Table S2.†

2.2 General procedure for the synthesis of complexes

A hot solution of M(OAc)₂·nH₂O (0.2 mmol) in ethanol (10 mL) was added dropwise to a stirring solution of β-diketone ligand (0.2 mmol) in ethanol (8 mL), then NH₃·H₂O was added in order to adjust the pH of solution (about 7–8) and stirred under reflux for 3 h. The mixture was cooled to room temperature and the precipitates were separated. The precipitates were isolated by centrifugation, washed several times with distilled water, anhydrous ethanol, and petroleum ether, filtered to give the complex.

Complex Fc8Cu. Yield 81.2%. IR (KBr, ν, cm⁻¹): 3129, 2924, 2855, 1605, 1583, 1526, 1503, 1449, 1119, 843. ESI-MS m/z: calcd 982.26 for [M + H]⁺; found: 982.23. Anal. calcd for C₅₄H₆₂CuFe₂O₆: C, 66.03; H, 6.36%, found: C, 65.49; H, 6.48%.

Complex Fc12Cu. Yield 82.6%. IR (KBr, ν, cm⁻¹): 3086, 2924, 2853, 1606, 1583, 1525, 1449, 1117, 842. ESI-MS m/z: calcd 1094.38 for [M + H]⁺; found: 1094.30. Anal. calcd for C₆₂H₇₈CuFe₂O₆: C, 68.04; H, 7.18%, found: C, 68.40; H, 7.12%.

Complex Fc14Cu. Yield 81.4%. IR (KBr, ν, cm⁻¹): 3083, 2923, 2851, 1605, 1583, 1539, 1503, 1449, 1118, 842. ESI-MS m/z: calcd 1150.45 for [M + H]⁺; found: 1150.22. Anal. calcd for C₆₆H₈₆CuFe₂O₆: C, 68.89; H, 7.53%, found: C, 68.62; H, 7.59%.

Complex Fc16Cu. Yield 80.7%. IR (KBr, ν, cm⁻¹): 3104, 2921, 2851, 1606, 1583, 1539, 1504, 1471, 1118, 842. ESI-MS m/z: calcd 1206.51 for [M + H]⁺; found: 1206.42. Anal. calcd for C₇₀H₉₄CuFe₂O₆: C, 69.67; H, 7.85%, found: C, 69.38; H, 7.97%.

Complex Fc14Zn(NH₃·H₂O). Yield 73.5%. IR (KBr, ν, cm⁻¹): 3120, 2923, 2852, 1605, 1586, 1524, 1449, 1109, 841. ¹H NMR (500 MHz, CDCl₃, δ ppm): 7.94 (d, J = 8.5 Hz, 4H, ArH), 6.90 (d, J = 8.5 Hz, 4H, ArH), 6.31 (s, 2H, CH), 4.85 (s, 4H, FcH), 4.39 (s, 4H, FcH), 4.19 (s, 10H, FcH), 3.97 (t, J = 6.5 Hz, 4H, OCH₂), 1.81–1.77 (m, 4H, CH₂), 1.50–1.43 (m, 4H, CH₂), 1.42–1.23 (m, 40H, CH₂), 0.88 (t, J = 7.0 Hz, 6H, CH₃). ¹³C NMR (125 MHz, CDCl₃, δ ppm): 192.04, 183.80, 163.77, 161.57, 132.41, 129.32, 124.41, 122.60, 121.09, 114.41, 91.05, 73.88, 73.37, 71.64, 70.78, 70.62, 70.05, 69.49, 31.82, 29.57, 29.01, 26.00, 22.67, 14.12. ESI-MS m/z: calcd 1151.45 for [M + H]⁺; found: 1151.40. Anal. calcd for C₆₆H₉₁Fe₂NO₇Zn: C, 66.75; H, 7.72; N, 1.18%, found: C, 66.76; H, 7.67; N, 1.05%.

Complex Fp8Cu. Yield 81.9%. IR (KBr, ν, cm⁻¹): 3079, 2947, 2925, 2854, 1604, 1582, 1524, 1499, 1469, 11 233, 840. ESI-MS m/z: calcd 1062.32 for [M + H]⁺; found: 1062.32. Anal. calcd for C₆₀H₇₀CuFe₂O₆: C, 67.83; H, 6.64%, found: C, 67.81; H, 6.68%.

Complex Fp10Cu. Yield 78.3%. IR (KBr, ν, cm⁻¹): 3075, 2924, 2852, 1605, 1585, 1529, 1502, 1468, 1120, 844. ESI-MS m/z: calcd 1118.39 for [M + H]⁺; found: 1118.33. Anal. calcd for C₆₄H₇₈CuFe₂O₆: C, 68.72; H, 7.03%, found: C, 68.52; H, 6.96%.

Complex Fp12Cu. Yield 80.5%. IR (KBr, ν, cm⁻¹): 3121, 2923, 2852, 1605, 1584, 1526, 1501, 1468, 1123, 840. ESI-MS m/z: calcd 1174.45 for [M + H]⁺; found: 1174.25. Anal. calcd for C₆₈H₈₆CuFe₂O₆: C, 69.53; H, 7.38%, found: C, 69.23; H, 7.37%.

Complex Fp14Cu. Yield 81.5%. IR (KBr, ν, cm⁻¹): 3075, 2923, 2852, 1605, 1584, 1529, 1501, 1468, 1120, 844. ESI-MS m/z: calcd 1230.51 for [M + H]⁺; found: 1230.48. Anal. calcd for C₇₂H₉₄CuFe₂O₆: C, 70.26; H, 7.70%, found: C, 70.34; H, 7.75%.

Complex Fp16Cu. Yield 79.6%. IR (KBr, ν, cm⁻¹): 3075, 2924, 2852, 1605, 1585, 1529, 1502, 1468, 1120, 844. ESI-MS m/z: calcd 1286.57 for [M + H]⁺; found: 1286.42. Anal. calcd for C₇₆H₁₀₂CuFe₂O₆: C, 70.93; H, 7.99%, found: C, 70.83; H, 8.04%.

Complex Fp8Zn(NH₃·H₂O). Yield 78.3%. IR (KBr, ν, cm⁻¹): 3077, 2926, 2853, 1605, 1546, 1524, 1117, 841. ¹H NMR (500 MHz, CDCl₃, δ ppm): 7.90 (d, J = 9.0 Hz, 4H, ArH), 6.89 (d, J = 9.0 Hz, 4H, ArH), 6.26 (s, 2H, CH), 4.74–4.71 (d, 4H, FcH), 4.29 (s, 2H, FcH), 4.27 (s, 2H, FcH), 4.22 (s, 2H, FcH), 3.99 (d, J = 6.5 Hz, 4H, OCH₂), 3.97 (s, 2H, FcH), 3.86 (s, 2H, FcH), 2.09–1.91 (m, 12H, CH₂), 1.83–1.76 (m, 4H, CH₂), 1.50–1.41 (m, 4H, CH₂), 1.42–1.24 (m, 16H, CH₂), 0.89 (t, J = 7.0 Hz, 6H, CH₃). ¹³C NMR (125 MHz, CDCl₃, δ ppm): 192.63, 183.26, 162.19, 161.45, 132.69, 131.56, 129.13, 128.55, 114.39, 113.95, 93.78, 92.54, 88.74, 86.19, 83.11, 74.34, 72.80, 70.93, 70.29, 69.81, 69.05, 68.13, 35.03, 31.82, 29.47, 29.25, 29.08, 26.04, 24.53, 24.13, 22.67, 14.12. ESI-MS m/z: calcd 1063.32 for [M + H]⁺; found: 1063.25. Anal. calcd for C₆₀H₇₅Fe₂NO₇Zn: C, 65.55; H, 6.88; N, 1.27%, found: C, 65.55; H, 7.00; N, 1.11%.

Complex Fp10Zn(NH₃·H₂O). Yield 75.6%. IR (KBr, ν, cm⁻¹): 3078, 2924, 2852, 1605, 1549, 1524, 1120, 841. ¹H NMR (500 MHz, CDCl₃, δ ppm): 7.91 (d, J = 8.5 Hz, 4H, ArH), 6.89 (d, J = 8.5 Hz, 4H, ArH), 6.27 (s, 2H, CH), 4.73 (d, 4H, FcH), 4.29 (d, 4H, FcH), 4.23 (s, 2H, FcH), 4.00–3.98 (m, 6H, OCH₂, FcH), 3.86 (s, 2H, FcH), 2.06–1.90 (m, 12H, CH₂), 1.82–1.73 (m, 4H, CH₂), 1.47–1.42 (m, 4H, CH₂), 1.31–1.26 (m, 24H), 0.88 (t, J = 6.5 Hz, 6H, CH₃). ¹³C NMR (125 MHz, CDCl₃, δ ppm): 192.44, 183.21, 161.39, 132.81, 129.10, 113.93, 93.68, 88.66, 86.15, 74.31, 72.72, 70.90, 70.24, 69.74, 69.02, 68.12, 35.03, 31.91, 29.39, 26.04, 24.60, 24.20, 22.70, 14.14. ESI-MS m/z: calcd 1120.39 for [M + H]⁺; found: 1120.25. Anal. calcd for C₆₄H₈₃Fe₂NO₇Zn: C, 66.53; H, 7.24; N, 1.21%, found: C, 66.12; H, 6.78; N, 1.19%.

Complex Fp12Zn(NH₃·H₂O). Yield 82.5%. IR (KBr, ν, cm⁻¹): 3078, 2926, 2853, 1605, 1587, 1549, 1524, 1304, 842. ¹H NMR (500 MHz, CDCl₃, δ ppm): 7.89 (d, J = 8.5 Hz, 4H, ArH), 6.87 (d, J = 8.5 Hz, 4H, ArH), 6.22 (s, 2H, CH), 4.71 (s, 2H, FcH), 4.69 (s, 2H, FcH), 4.25 (s, 2H, FcH), 4.22 (s, 2H, FcH), 4.18 (s, 2H, FcH), 3.94–3.92 (m, 6H, OCH₂, FcH), 3.84 (s, 2H, FcH), 2.05–1.91 (m, 12H, CH₂), 1.83–1.74 (m, 4H, CH₂), 1.50–1.40 (m, 4H, CH₂), 1.39–1.24 (m, 32H, CH₂), 0.88 (t, J = 6.8 Hz, 6H, CH₃). ¹³C NMR (125 MHz, CDCl₃, δ ppm): 191.84, 183.08, 161.19, 133.18, 128.99, 113.87, 93.43, 88.37, 86.01, 83.62, 74.16, 72.45, 70.82, 70.07, 69.62, 68.96, 68.06, 53.43, 35.03, 31.93, 29.68, 29.62, 29.36, 26.04, 24.60, 24.20, 22.70, 14.14. ESI-MS m/z: calcd 1175.45 for [M + H]⁺; found: 1175.49. Anal. calcd for C₆₈H₉₁Fe₂NO₇Zn: C, 67.41; H, 7.57; N, 1.16%, found: C, 67.65; H, 7.57; N, 1.13%.

Complex Fp14Zn(NH₃·H₂O). Yield 79.7%. IR (KBr, ν, cm⁻¹): 3076, 2923, 2851, 1065, 1548, 1524, 1120, 840. ¹H NMR (500 MHz, CDCl₃, δ ppm): 7.90 (d, J = 9.0 Hz, 4H, ArH), 6.88 (d, J = 9.0 Hz, 4H, ArH), 6.25 (s, 2H, CH), 4.74 (s, 2H, FcH), 4.71 (s, 2H, FcH), 4.28 (s, 2H, FcH), 4.26 (s, 2H, FcH), 4.21 (s, 2H, FcH), 3.99–3.96 (m, 6H, OCH₂, FcH), 3.86 (s, 2H, FcH), 2.08–1.91 (m, 12H, CH₂), 1.84–1.75 (m, 4H, CH₂), 1.46–1.43 (m, 4H, CH₂), 1.36–1.22 (m, 40H, CH₂), 0.88 (t, J = 7.0 Hz, 6H, CH₃). ¹³C NMR (125 MHz, CDCl₃, δ ppm): 192.24, 183.17, 161.33, 132.95, 129.07, 113.92, 93.59, 88.56, 86.11, 83.38, 74.26, 72.62, 70.88, 70.19, 69.72, 68.99, 68.11, 35.04, 31.94, 29.80, 29.62, 29.56, 29.35, 26.05, 24.61, 24.21, 22.71, 14.15. ESI-MS m/z: calcd 1231.51 for [M + H]⁺; found: 1231.34. Anal. calcd for C₇₂H₉₉Fe₂NO₇Zn: C, 68.22; H, 7.87; N, 1.10%, found: C, 68.25; H, 7.87; N, 1.05%.

Complex Fp16Zn(NH₃·H₂O). Yield 81.2%. IR (KBr, ν, cm⁻¹): 3127, 2923, 2852, 1605, 1543, 1524, 1121, 841. ¹H NMR (500 MHz, CDCl₃, δ ppm): 7.90 (d, J = 8.5 Hz, 4H, ArH), 6.88 (d, J = 8.5 Hz, 4H, ArH), 6.25 (s, 2H, CH), 4.73 (s, 2H, FpH), 4.71 (s, 2H, FcH), 4.28 (s, 2H, FcH), 4.26 (s, 2H, FcH), 4.21 (s, 2H, FcH), 4.02–3.96 (m, 6H, OCH₂, FcH), 3.86 (s, 2H, FcH), 2.09–1.88 (m, 12H, CH₂), 1.86–1.78 (m, 4H, CH₂), 1.50–1.41 (m, 4H, CH₂), 1.35–1.45 (m, 48H, CH₂), 0.88 (t, J = 7.0 Hz, 6H, CH₃). ¹³C NMR (125 MHz, CDCl₃, δ ppm): 193.13, 179.19, 162.19, 161.73, 131.55, 129.12, 128.55, 127.50, 114.39, 86.62, 73.99, 70.96, 69.85, 68.25, 53.44, 35.01, 31.94, 29.81, 29.38, 29.16, 26.01, 24.47, 24.07, 22.71, 14.14. ESI-MS m/z: calcd 1287.57 for [M + H]⁺; found: 1287.65. Anal. calcd for C₇₆H₁₀₇Fe₂NO₇Zn: C, 68.96; H, 8.15; N, 1.06%, found: C, 68.96; H, 8.07; N, 0.97%.

3 Results and discussion

3.1 Synthesis and characterization

Two kinds of β-diketone ligands Fc8–Fc16 and Fp8–Fp16 were prepared. The molecular structures are shown in Fig. 1. Detailed procedures of synthesis and characterization of these β-diketone ligands were described in ESI.† From the ¹H NMR study, the percentage of enolized tautomers in CDCl₃ of the prepared β-diketones was established (see the Experimental section) by comparing the relative intensities of suitable keto–enol signal pairs. It was found that the percentage of enol isomer in ferrocene-containing β-diketones was above 80%, but that in [3]ferrocenophane-containing β-diketones was only near 69%. The lower percentage of enol isomer in the latter may be due to the effect of the trimethylene group as an electron donor. The vinylic C–H singlet at 6.33 ppm (1H) (Fc8–Fc16) and 6.25 ppm (Fp8–Fp16) for the enol isomer gave rise to a sharp resonance at 4.30 ppm (2H) (Fc8–Fc16) and 4.21 ppm (Fp8–Fp16) for the β-diketone isomer. The downfield shifted O–H enol proton, located at the region of 16.64–16.74 ppm in CDCl₃ for Fc8 and Fp14, disappeared for other β-diketones due to the rapid exchange between active hydrogen atoms.

The β-diketones Fc8–Fc16 and Fp8–Fp16, as free ligands, were converted to the Cu(II) and Zn(II) complexes by reaction with copper acetate or zinc acetate in ethanol. In the IR spectroscopy, the strong bands assignable to stretching vibrations of the skeleton C=C and C=O groups, which were observed at about 1600 and 1550 cm⁻¹ in the ligands, were all red shifted to the region 1530 and 1500 cm⁻¹ in the complexes. This is because increased electron density on a metal centre of complex leads to a decrease in double bond character of the C=O and C=C groups.³⁵ The ESI-MS spectra showed [M + H]⁺ peaks for all these complexes. In the ¹H NMR spectra, upon coordination, the downfield shifted O–H enol proton vanished, and the two phenyl protons near vinylic doublet at 7.87–7.90 ppm in the ligands moved the signal to more downfield (7.90–7.94 ppm in the Zn(II) complexes), but the signal of other phenyl and cyclopentadienyl protons moved the signal to more upfield (by 0.03–0.06 ppm). In addition, slight free ligands were observed in CDCl₃ in the ¹H NMR spectra. A reasonable explanation is in CHCl₃ and CH₂Cl₂ solutions that were not treated with basic alumina to remove any photochemically generated HCl, complexes quickly converted back to the appropriate β-diketones together with Cu(II) or Zn(II) chloride.³⁶ The possible formulas of complexes were Cu(L)₂ and Zn(L)₂(NH₃·H₂O) presumed by elemental analysis data.

Complex Fc8Cu was also characterized by X-ray diffraction methods. The molecular structure of Fc8Cu is shown in Fig. 2. The X-ray crystal structure reveals that the molecule is centrosymmetric. The Cu atom is chelated with two β-diketone ligands in the equatorial plane. The geometry around the Cu centre is square planar with the ferrocenyl moieties being positioned opposite to each other. As shown in Table S2,† the bond lengths of Cu(1)–O(5) (1.889(5) Å) and Cu(1)–O(1) (1.899(5) Å) are almost equal, and a slightly shorter than the bond lengths of Cu(1)–O(2) and Cu(1)–O(6) (1.914(5) Å). The bond angle O(5)–Cu(1)–O(1) (179.2(2)°), as that of O(2)–Cu(1)–O(6) (179.1(2)°), is near 180° indicating the collinearity of three atoms. The other O–Cu–O bonding angles range from 86.4(2)° to 93.2° with the largest deviation from 90° being 0.36° for O(1)–Cu(1)–O(6), and the dihedral angle between two chelating OCCCO semi-rings is 1.57° indicating their coplanarity. Phenyl fragment and square plane are also closer to a coplanar position with cyclopentadiene ring, with dihedral angles of 1.05° (C5) and 4.02° (O1) respectively.

Fig. 2 The molecular structure of Fc8Cu.

3.2 UV-vis absorption spectra

The absorption spectra of the representative compounds (Fc14, Fp14 and their complexes) in CH₂Cl₂ at 10⁻⁵ M are shown in Fig. 3. Peak maxima are summarized in Table S3.† It can be seen that changing the alkyl chain slightly shifts the absorption position of ligands and complexes. As shown in Fig. 3, the UV-vis absorption spectra correspond to the assembled spectra of ferrocene and aryl substitute groups. The absorption bands below 320 nm are assignable to Fe(d)–π* and π–π* electronic transitions of the benzene rings. The absorption band associated with the carbonyl chromophore for the ligands shifted bathochromically to 350 nm due to the conjugational effect of the adjacent aromatic rings, and this band further shifted bathochromically about 3–10 nm and enhanced in the complexes, probably overlapping with the π–π* transitions of the conjugated cyclopentadiene rings. Very broad absorption bands at 450–550 nm belong to d–d type transitions of the electrons of the iron atoms, moreover, these bands for the ferrocene-containing β-diketone ligands and complexes are stronger than those of [3]ferrocenophane-containing β-diketones.³⁷

Fig. 3 UV-vis absorption spectra of Fc14, Fp14 and their complexes taken at 10⁻⁵ M in CH₂Cl₂.

3.3 Electrochemical investigation

It is well known that the electrochemical reaction mechanism of a ferrocene-based BR catalyst is closely linked with its combustion catalytic performance in solid propellants.²⁹ The redox properties of representative compounds (Fc8, Fp8, Fc14, Fp14 and their complexes) were studied by CV using CH₂Cl₂ as the solvent containing 0.1 M TBAPF₆ as a supporting electrolyte. Electrochemical data are shown in Table 1. Selected CV curves are shown in Fig. 4, S1 and S2.†

Table 1 CV data of 10⁻³ M solutions of selected ligands and complexes in CH₂Cl₂ containing 0.1 M TBAPF₆ as supporting electrolyte at 20 °C and 100 mV s⁻¹ scan rate. Potentials are vs. Fc/Fc⁺

Compd	Epa^a/mV	Epc^b/mV	ΔE^c/mV	E1/2^d/mV	ipa/μA	ipa/ipc
a Anodic peak potential.b Cathodic peak potential.c ΔE = Epa − Epc.d E1/2 = 1/2(Epa + Epc).
Fc8	288	125	163	207	32.50	0.95
Fc14	288	120	168	204	28.07	0.94
Fc8Cu	200	61	139	131	27.07	0.95
Fc14Cu	195	59	136	127	27.07	0.93
Fc14Zn	245	102	143	174	17.45	0.95
Fp8	222	56	166	139	32.83	0.95
Fp14	208	54	154	131	29.18	0.95
Fp8Cu	137	7	130	72	26.08	0.96
Fp14Cu	134	2	132	68	21.91	0.95
Fp8Zn	154	7	147	81	28.41	0.96
Fp14Zn	151	10	141	81	18.41	0.94

---

Fig. 4 CV curves of Fc14, Fp14, Fp14Cu, Fp14Zn in CH₂Cl₂, 0.1 M TBAPF₆. Scan rate = 100 mV s⁻¹.

All redox-active ferrocenyl groups exhibit one-electron transfer processes with formal redox potentials E_1/2 = 207, 204 mV for Fc8, Fc14 and E_1/2 = 139, 131 mV for Fp8, Fp14 vs. Fc/Fc⁺, which indicates that lengthening of the alkyl chain exerts little effect on the formal redox potential of ferrocene. However, as compared to ferrocene-containing β-diketone derivatives (Fc8, Fc14), the potentials for Fp8 and Fp14 cathodically shifted about 70 mV thereby suggesting easy oxidation by loss of an electron for [3]ferrocenophane-containing β-diketone.³⁸ Similar phenomena are also observed in the corresponding Cu(II) and Zn(II) complexes. The oxidation potentials change of ferrocene and its derivatives in ways dependent on the substituent: the oxidation potentials of the derivatives with the electron donating substituents, such as alkyl groups, are lower than that of ferrocene. In the series of [3]ferrocenophane-containing β-diketone derivatives, the effect of trimethylene group as a substituent is as an electron donor and, therefore, the oxidation potentials of these derivatives are lower than that of ferrocene itself.³⁹ In addition, compared with the ligands, the negative potential shifts (63–77 mV for Cu(II) complexes, 30–58 mV for Zn(II) complexes) of the E_1/2 values for ferrocene indicate that the coordination decreases the electron withdrawing ability of C=O groups, making the complexes easier to oxidise than the respective free ligands. Moreover, the E_1/2 values for ferrocene in Cu(II) complexes are lower than those of Zn(II) complexes due to the higher stability of Cu(II) complexes. The more stable of the complexes, the weaker is the electron withdrawing ability of C=O group.

Theoretically, electrochemically reversible one-electron transfer processes are characterized with ΔE = 59 mV. However, in practice, the value is up to 90 mV. As shown in Table 1, the i_pc/i_pa ratios approached 1, but the ΔE value was in the range of 130–170 mV at slow (100 mV s⁻¹) scan rates, which showed a quasi-reversible redox step for all compounds.³⁶ In addition, the effect of scan rate was investigated (Fig. 5, S3 and S4†). The result showed that the redox potentials of β-diketone ligands and their complexes were slightly influenced by the scan rate in a range from 10 to 300 mV s⁻¹, and both the anodic and cathodic peak currents are linear to the square root of scan rates, indicating a diffusion-controlled process.²⁵

Fig. 5 CV curves and peak currents of Fp14Cu obtained at different scan rates in CH₂Cl₂.

3.4 Thermalstability

Thermal stability is an essential parameter for a BR catalyst because a higher thermal stability of a catalyst aids the steady combustion of a solid propellant. TG analyses of several representative Cu(II) and Zn(II) complexes were performed in a N₂ atmosphere at a heating rate of 10 °C min⁻¹. The results are shown in Fig. S5.† From the TG curves, we can see the Zn(II) complexes began to lose their weight at about 75 °C and gave rise to weight loss of about 3% before 140 °C, which almost equal to the percentage of one molecular NH₃·H₂O. However, the Cu(II) complex barely lost their weight before 140 °C, indicating no crystal water in the Cu(II) complexes. These results are consistent with the elemental analyses data. In addition, TG curves showed the complexes had high thermal stability. Both Fp8Zn and Fp16Zn started to decompose at about 325 °C, which indicated that lengthening of the alkyl chain has no effect on the stability of complexes, and the Fp16Zn had higher thermal stability compared with the Fp16Cu (decomposition temperature 290 °C). In addition, the thermal stability of complex decreased when the ligand of [3]ferrocenophane-containing β-diketone was replaced by ferrocene-containing β-diketone (Fc14Zn, decomposition temperature 270 °C).

3.5 Catalytic effect on the decomposition of ammonium perchlorate

In general, the catalytic activity of a BR catalyst in solid propellant can be assessed by studying its effect on the thermal degradation of AP by a simultaneous TG-DSC instrument, and the weight percentage of ferrocene-based BR catalysts in AP were generally 2–5%. In this paper, 3 wt% of catalyst was used. The thermal degradation curves of pure AP in an open pan or pan closed with a pierced lid, and mixtures of AP and 3 wt% of the β-diketone ligands or complexes in open pans were shown in Fig. 6. From Fig. 6, we can see the first endothermic peak (241 °C) in the DSC curve of pure AP corresponds to the morphological transition from an orthorhombic to a cubic phase of AP. Further heating of pure AP in an open pan is accompanied by an exothermic peak (294 °C, LTD) followed by an endothermic one (398 °C, HTD). The endothermic peak disappears when decomposition is conducted in a pan closed with a pierced lid and an exothermic peak is seen at ∼438 °C. Analysis of the literature TG-DSC data suggests that the HTD is either exothermic or endothermic process, depending on the competition between sublimation and thermal decomposition.⁴ When the HTD occurs under the conditions that retard the escape of reaction gases from the reaction zone (elevated pressures, closed pans, static atmosphere, etc.), it appears as an exothermic peak (predominated by thermal decomposition). Conversely, under the conditions that facilitate the removal of gaseous products (vacuum, open pans, high flow rates of a carrier gas), the HTD appears as an endothermic peak (predominated by sublimation).⁵

Fig. 6 DSC curves of AP and mixtures of AP with 3 wt% β-diketone ligands or complexes in open pans and AP in lidded pan.

As shown in Fig. 6, after adding 3% of the β-diketone ligands or complexes, the peaks near 241 °C appearing in all samples exhibit no significant changes and the LTD peaks shift upwards by 7–19 °C compared with pure AP. The exothermic processes of the mixtures with peak temperatures in the range 200–220 °C are caused by a partial decomposition of AP, since the exothermic effects in this range are relatively low compared to the exothermic effects after 300 °C.²³ In addition, dramatic changes have been observed at the stage of HTD. The endothermic peaks caused by sublimation disappeared when decomposition is conducted in open pans. The exothermic process suggests that the HTD of the AP with the additives occurs predominantly via the thermal decomposition channel.⁴⁰ This implies that the additives have greater catalytic effects on the HTD of AP than on the beginning stage. After adding Fc14, Fc14Zn and Fp14Zn, the highest thermal decomposition temperatures are 370.3, 361.8 and 367.1 °C respectively, decreased by 68.2, 76.7 and 71.4 °C compared with pure AP, which indicates the additives have obvious catalytic effects on thermal decomposition of AP. From the decomposition heat of AP added Fc14 (−893 J g⁻¹), Fc14Zn (−1086 J g⁻¹) and Fp14Zn (−853 J g⁻¹), we can see that the catalytic activity of Fp14Zn is weaker than that of Fc14Zn and almost the same as the ligand. Moreover, it can be seen that lengthening of the alkyl chain insignificantly changes the catalytic activity from the curves of Fp10Zn and Fp14Zn. Compared with ligands and Zn complexes, the addition of Cu complexes lowers the thermal decomposition temperature of AP more obviously, and the highest thermal decomposition temperature reduces from pure AP 438.5 °C to about 345 °C (shift downwards by about 92 °C), and the decomposition heat of AP added the Cu complexes is about 1300 J g⁻¹. It is worth pointing out that the two exothermic peaks tend to merge into one after adding the Cu complexes. This implies that the complexes can advance the decomposition temperature of AP as well as narrow significantly the decomposition temperature range of AP.²⁵ So the Cu complex is a more effective catalyst than the Zn complex, which may attribute to the enhanced catalytic performance of Cu oxide compared with Zn oxide to the thermal decomposition of AP. These oxides produced by the thermal decomposition of the complexes are the active species during the combustion of propellants. The decomposition of the complex near the surface layer of AP during thermal decomposition may increase the surface and flame temperatures. This may also enhance the BR.^41,42 Generally speaking, transition metal oxide with variable valence has higher catalytic performance than others.⁴³

The additives lowering of the thermal decomposition temperature of AP is further proved by the TG experiments. As shown in Fig. 7, after adding 3% of the β-diketone ligands or complexes, the TG curves exhibit that the samples start to decompose at about 200 °C, and cease their weight loss between 353.4 and 381.2 °C, shifting the final decomposition temperature of pure AP (439.9 °C) dramatically. In addition, the TG curve of pure AP exhibits two obvious weight-loss steps, while the TG curve of AP in the presence of the additive only exhibits one weight-loss step. It is also noted that the final decomposition temperatures of AP with Cu complexes as additives are lower obviously than those of the samples containing ligands and Zn complexes, implying that the effects of Cu complexes on the thermal degradation of AP are greater than those of ligands and Zn complexes, in agreement with the decomposition trend of the DSC curves.

Fig. 7 TG curves of AP and mixtures of AP with 3 wt% β-diketone ligands or complexes in open pans and AP in lidded pan.

4 Conclusion

In summary, ferrocene and [3]ferrocenophane-based β-diketones and their Cu(II) and Zn(II) complexes have been synthesized and characterized. All new compounds show quasireversible and diffusion-controlled redox step. The [3]ferrocenophane-containing β-diketone ligands and complexes are easier oxidation by loss of an electron compared with those of ferrocene-containing β-diketone ligands and complexes. The complexes exhibit high thermal stability according to the TG data. The β-diketone ligands and complexes exhibit high catalytic activity in the thermal degradation of AP, and lengthening of the alkyl chain or replacing ferrocene with [3]ferrocenophane insignificantly change the catalytic activity. In addition, compared with β-diketone ligands and Zn(II) complexes, the Cu complexes lower the thermal decomposition temperature of AP (by ∼92 °C) more dramatically. Although the effects of BR catalysts can not only be focused on the influence on the thermal decomposition of AP, the results confirm ferrocene-containing β-diketone ligands and their Cu(II) and Zn(II) complexes are a kind of potentially high BR catalyst.

Acknowledgements

We thank the National Natural Science Foundation of China (NSFC, No. 21562032 and 21262023), Natural Science Foundation of Inner Mongolia of China (No. 2013MS0207) and Research Program of science and technology at Universities of Inner Mongolia (NJZZ001) for their generous financial support.

References

1. M. Kohga and K. Okamoto, Combust. Flame, 2011, 158, 573.
2. G. Tang, S. Tian, Z. Zhou, Y. Wen, A. Pang, Y. Zhang, D. Zeng, H. Li, B. Shan and C. Xie, J. Phys. Chem. C, 2014, 118, 11833.
3. Y. Yuan, W. Jiang, Y. Wang, P. Shen, F. Li, P. Li, F. Zhao and H. Gao, Appl. Surf. Sci., 2014, 303, 354.
4. A. J. Lang and S. Vyazovkin, Combust. Flame, 2006, 145, 779.
5. S. Vyazovkin and C. A. Wight, Chem. Mater., 1999, 11, 3386.
6. W. Zhang, P. Li, H. Xu, R. Sun, P. Qing and Y. Zhang, J. Hazard. Mater., 2014, 268, 273.
7. P. R. Patil, V. N. Krishnamurthy and S. S. Joshi, Propellants, Explos., Pyrotech., 2006, 31, 442.
8. P. R. Patil, V. N. Krishnamurthy and S. S. Joshi, Propellants, Explos., Pyrotech., 2008, 33, 266.
9. A. A. Vargeese, K. Muralidharan and V. N. Krishnamurthy, Propellants, Explos., Pyrotech., 2015, 40, 260.
10. J. Wang, Y. Liu, H. Zhao, J. Liu, D. Wang, Y. Song and Y. Wang, Polym. Degrad. Stab., 2009, 94, 625.
11. G. Singh, I. P. S. Kapoor, S. Dubey and P. F. Siril, Propellants, Explos., Pyrotech., 2009, 34, 72.
12. I. P. S. Kapoor, P. Srivastava and G. Singh, Propellants, Explos., Pyrotech., 2009, 34, 351.
13. N. Dilsiz and A. Ünver, J. Appl. Polym. Sci., 2006, 101, 2538.
14. A. Unver, N. Dilsiz, M. Volkan and G. Akovali, J. Appl. Polym. Sci., 2005, 96, 1654.
15. J. B. Zhuo, H. D. Li, C. X. Lin, L. L. Xie, S. Bai and Y. F. Yuan, J. Mol. Struct., 2014, 1067, 112.
16. J. M. Gao, L. Wang, H. L. Yu, A. G. Xiao and W. B. Ding, Propellants, Explos., Pyrotech., 2011, 36, 404.
17. M. Alfonso, A. Tarraga and P. Molina, Dalton Trans., 2010, 39, 8637.
18. R. Tong, Y. Zhao, L. Wang, H. Yu, F. Ren and M. Saleem, J. Organomet. Chem., 2014, 755, 16.
19. F. Xiao, X. Sun, X. Wu, J. Zhao and Y. Luo, J. Organomet. Chem., 2012, 713, 96.
20. Z. Deng, H. Yu, L. Wang, X. Zhai, Y. Chen and R. Sun, J. Organomet. Chem., 2015, 799–800, 273.
21. D. Saravanakumar, N. Sengottuvelan, V. Narayanan, M. Kandaswamy and T. L. Varghese, J. Appl. Polym. Sci., 2011, 119, 2517.
22. X. Liu, W. Zhang, G. Zhang and Z. Gao, New J. Chem., 2015, 39, 155.
23. X. Liu, J. Li, F. Bi, W. Zhang, G. Zhang and Z. Gao, Eur. J. Inorg. Chem., 2015, 1496.
24. X. Liu, D. Zhao, F. Bi, X. Fan, F. Zhao, G. Zhang, W. Zhang and Z. Gao, J. Organomet. Chem., 2014, 762, 1.
25. J. Zhuo, Z. Ma, C. Lin, L. Xie, S. Bai and Y. Yuan, J. Mol. Struct., 2015, 1085, 13.
26. H. Zhao, L. Guo, S. Chen and Z. Bian, RSC Adv., 2013, 3, 19929.
27. B. Florczak, Cent. Eur. J. Energ. Mater., 2008, 5, 103.
28. J. Zhai, R. Yang and J. Li, Combust. Flame, 2008, 154, 473.
29. C. Wang, J. Li, X. Fan, F. Zhao, W. Zhang, G. Zhang and Z. Gao, Eur. J. Inorg. Chem., 2015, 1012.
30. B. E. Buitendach, A. Gągor and J. C. Swarts, Inorg. Chem., 2013, 52, 10245.
31. H. J. Gericke, A. J. Muller and J. C. Swarts, Inorg. Chem., 2012, 51, 1552.
32. J. C. Swarts, T. G. Vosloo, S. J. Cronje, W. C. Du Plessis, C. E. Van Rensburg, E. Kreft and J. E. Van Lier, Anticancer Res., 2008, 28, 2781.
33. H. Zhao, M. Chen, X. Zhu, S. Chen and Z. Bian, Res. Chem. Intermed., 2015, 41, 3971.
34. G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr., 2008, 64, 112.
35. Z. Li, J. Liu, H. Zhao, B. Li and Z. Bian, Res. Chem. Intermed., 2015, 41, 8545.
36. C. C. Joubert, L. van As, A. Jakob, J. M. Speck, H. Lang and J. C. Swarts, Polyhedron, 2013, 55, 80.
37. H. Zhao, X. Zhu, D. Wang, S. Chen and Z. Bian, Aust. J. Chem., 2015, 68, 1035.
38. S. Toma and E. Solčániová, J. Organomet. Chem., 1985, 288, 331.
39. T. Ogata, K. Oikawa, T. Fujisawa, S. Motoyama, T. Izumi, A. Kasahara and N. Tanaka, Bull. Chem. Soc. Jpn., 1981, 54, 3723.
40. Z. Cheng, G. Zhang, X. Fan, F. Bi, F. Zhao, W. Zhang and Z. Gao, Inorg. Chim. Acta, 2014, 421, 191.
41. G. Singh and D. K. Pandey, Propellants, Explos., Pyrotech., 2003, 28, 231.
42. D. Chen, S. Huang, Q. Zhang, Q. Yu, X. Zhou, H. Li and J. Li, RSC Adv., 2015, 5, 32872.
43. S. M. Vuga, Propellants, Explos., Pyrotech., 1991, 16, 293.

---

Footnote

† Electronic supplementary information (ESI) available: Detailed procedures of synthesis and characterization of ligands, crystal data, structure refinement, bond lengths and angles for Fc8Cu; Tables S1–S3, Fig. S1–S35. CCDC 1000655. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c6ra02588a

---