Double role of metalloporphyrins in catalytic bioinspired supramolecular metal–organic frameworks (SMOFs)

Heterogeneous catalysts are of great interest in many industrial processes for environmental reasons and, during recent years, a great effort has been devoted to obtain metal–organic frameworks (MOFs) with improved catalytic behaviour. Few supramolecular metal–organic frameworks (SMOFs) are stable under ambient conditions and those with anchored catalysts exhibit favourable properties. However, this paper presents an innovative approach that consists of using metal nodes as both structural synthons and catalysts. Regarding the latter, metalloporphyrins are suitable candidates to play both roles simultaneously. In fact, there are a number of papers that report coordination compounds based on metalloporphyrins exhibiting these features. Thus, the aim of this bioinspired work was to obtain stable SMOFs (at room temperature) based on metallo­porphyrins and explore their catalytic activity. This work reports the environmentally friendly microwave-assisted synthesis and characterization of the compound [H(bipy)]₂[(MnTPPS)(H₂O)₂]·2bipy·14H₂O (TPPS = meso-tetra­phenyl­porphine-4,4′,4′′,4′′′-tetra­sulfonic acid and bipy = 4,4′-bi­pyridine). This compound is the first example of an MnTPPS-based SMOF, as far as we are aware, and has been structurally and thermally characterized through single-crystal X-ray diffraction, IR spectroscopy, thermogravimetry and transmission electron microscopy. Additionally, this work explores not only the catalytic activity of this compound but also of the compounds μ-O-[FeTCPP]₂·16DMF and [CoTPPS_0.5(bipy)(H₂O)₂]·6H₂O. The structural features of these supra­molecular materials, with accessible networks and high thermal stability, are responsible for their excellent behaviour as heterogeneous catalysts for different oxidation, condensation (aldol and Knoevenagel) and one-pot cascade reactions.

1. Introduction

During recent years, supramolecular materials and metal–organic frameworks (MOFs) have been thoroughly explored in many fields, such as water reuse, photocatalysis, electrochemistry and gas adsorption (de Lange et al., 2015; Dias & Petit, 2015; Li et al., 2016; Li & Hill, 2017; Wang et al., 2016; Gao et al., 2014). Their structural and chemical properties make them excellent candidates as solid catalysts for many reactions (Dhakshinamoorthy et al., 2012). Moreover, supra­molecular metal–organic frameworks (SMOFs), in which the three-dimensional crystalline network is sustained by hydrogen bonds (Pérez-Aguirre et al., 2016; Reger et al., 2012; Thomas-Gipson et al., 2014), are attracting much interest, and in order to obtain those coordination networks, the use of porphyrins has been rising, since they are organic ligands which present unique properties attached to biochemical, enzymatic and photochemical functions (Kornienko et al., 2015; Spoerke et al., 2017). Biomimetic catalysts, such as metalloporphyrins, have been used as cytochrome analogues because of the similarity between these molecules and the active centres of the enzymes. Oxidation, condensation and hydrolysis reactions are very common in living organisms and many efforts have been made to mimic their catalytic activity by means of metalloporphyrin-based synthetic models (Johnson et al., 2016; Chen et al., 2012; Feng et al., 2012). In fact, metalloporphyrinic supramolecular compounds are appearing as a new class of promising materials in the development of catalytic cascade or one-pot reactions (Hajimohammadi et al., 2012; Shinde et al., 2015; Prasad et al., 2014; Lu et al., 2015). Most of the catalytic reactions in industry use the traditional design of simple catalytic reactions involving expensive catalysts and processes. So, in order to reduce costs and optimize processes, the majority of recent work has focused on homogeneous (Omagari et al., 2016; Bonin et al., 2014; Costentin et al., 2014; Pires et al., 2014) or heterogeneous catalytic activity based on metalloporphyrinic networks (Zhang et al., 2016; Chen et al., 2015; Hod et al., 2015; Ucoski et al., 2015; Wang et al., 2013; Meng et al., 2012). There are some disadvantages of heterogeneous catalysis, including poor understanding of the reaction mechanisms, heat diffusion problems, low reaction selectivity and poorly defined active sites. However, recycling is one of the most important advantages for heterogeneous catalysis and it is based on easy catalyst separation (Moulijn et al., 1993).

In order to achieve heterogeneous catalysis, there are a number of successful approaches such as anchoring the catalyst into the cavities of porous coordination networks (Zhan & Zeng, 2016; Liu et al., 2016), doping the network with the catalyst (Lan et al., 2016) or post functionalizing the network (Andriamitantsoa et al., 2016). In this sense, we have been exploring a new strategy that consists of using porphyrins as structural units in SMOFs and catalytic active centres simultaneously (Fidalgo-Marijuan et al., 2015). This strategy includes the use of first-row transition metals, avoiding commonly used heavy and toxic metals such as Ru, Rh and Ce (Liu et al., 2017). It is also noteworthy that we have performed green syntheses, using preferably non-toxic solvents (water) and fast microwave heating.

Taking into consideration the above-mentioned aspects, we report here the heterogeneous catalysis of three porphyrinic SMOFs: [H(bipy)]₂[(MnTPPS)(H₂O)₂]·2bipy·14H₂O, 1, having MnTPPS-based monomers, the μ-O-[FeTCPP]₂·16DMF dimeric compound, 2, and the [CoTPPS_0.5(bipy)(H₂O)₂]·6H₂O one-dimensional compound, 3, where TPPS = meso-tetra­phenyl­porphine-4,4′,4′′,4′′′-tetra­sulfonic acid, TCPP = meso-tetra(4-carb­oxy­phenyl)­porphine, bipy = 4,4′-bi­pyridine and DMF = N,N′-di­methyl­formamide. It is worth noting that compound 1 is the first Mn TPPS metallo­porphyrinic SMOF reported so far. We have reported the structural features of 2 and 3 elsewhere (Fidalgo-Marijuan et al., 2015, Fidalgo-Marijuan et al., 2013).

Compounds 1, 2 and 3 have been exhaustively characterized by means of single-crystal X-ray diffraction, IR spectroscopy, thermogravimetric analysis and transmission electron microscopy (TEM), after which oxidation, Knoevenagel and aldol condensations, and a one-pot cascade reaction (involving an acetal hydrolysis in the first step and a Knoevenagel condensation in the second) have been successfully tested for these compounds.

2. Experimental

2.1. General

All solvents and reagents including meso-tetra­phenyl­porphine-4,4′,4′′,4′′′-tetra­sulfonic acid tetrasodium salt (TPPS), 4,4′-bi­pyridine (bipy) and Mn(NO₃)₂·xH₂O were purchased from Sigma–Aldrich.

2.2. Synthesis of [H(bipy)]₂[(MnTPPS)(H₂O)₂]·2bipy·14H₂O

TPPS (10.2 mg, 0.01 mmol) and Mn(NO₃)₂·xH₂O (1.0 mg, 0.006 mmol) were dissolved in distilled water (10 ml) and the solution was stirred for 30 min. Then, 4,4′-bi­pyridine (9.4 mg, 0.06 mmol) was dissolved in hot (343 K) distilled water (5 ml) and added to the mixture in a 100 ml CEM EasyPrep microwave vessel. The mixture was heated by microwaves under autogenous pressure at 433 K for 2 h, and then cooled naturally to room temperature, yielding diffraction-quality prismatic dark-red crystals. [For C₄₂H₄₆Mn_0.5N₆O₁₄S₂. Found: C, 53.55 (4); H, 4.85 (2); N, 8.84 (6); O, 22.95 (6); S, 6.89 (4). Calculated: C, 53.08; H, 4.88; N, 8.84; O, 23.57; S, 6.75]. ν_max/cm⁻¹ 3397 (OH), 1636–1413 (CC), 1393 and 1180 (SO), 1340 (CN), 1204 and 1070 (bipy), 1003 (MnTPPS) and 857–634 (CH) (see Fig. S1 in the supporting information).

2.3. Single-crystal X-ray diffraction

Prismatic dark-red single-crystals of compound 1 (dimensions given in Table 4) were selected under a polarizing microscope and mounted on MicroMounts. Single-crystal X-ray diffraction data were collected at 100 K on a SuperNova single source diffractometer with Cu Kα radiation (λ = 1.54184 Å). Data frames were processed [unit-cell determination, intensity-data integration, correction for Lorentz and polarization effects (Yinghua, 1987) and analytical absorption correction] using the CrysAlisPro software package (Agilent Technologies UK Ltd, 2012).

The structure of 1 was solved in the triclinic space group using the SIR92 program (Altomare et al., 1993) which allowed us to determine the position of the Mn atom, as well as some of the O, N, S and C atoms of the porphyrin and bi­pyridine molecules. The refinement of the crystal structure was performed by full matrix least-squares based on F², using the SHELXL97 program (Sheldrick, 2008), obtaining the remaining C, N, O and S atoms of the porphyrin and O atoms of water molecules. Anisotropic displacement parameters (Farrugia, 1997) were used for all non-hydrogen atoms, except for the disordered crystallized water molecules (Fig. S2). All H atoms connected to aromatic rings (C—H = 0.95 Å) were fixed geometrically and refined using a riding model with common isotropic displacements. Four of the crystallized water mol­ecules of compound 1 were disordered over two groups, with half occupancy each, as well as for one of the porphyrin sulfite groups. Crystal data for compound 1 are listed in Table 1. Geometric parameters, atomic coordinates and anisotropic displacement parameters are given in the supporting information, Tables S1, S2 and S3.

Table 1
Crystallographic data for 1

Compound	[H(bipy)]2[(MnTPPS)(H2O)2]·2bipy·14H2O
Formula	C42H46Mn0.5N6O14S2
FW (g mol^−1)	950.44
Crystal system	Triclinic
Space group (No. 2)
a (Å)	9.7187 (4)
b (Å)	11.2496 (5)
c (Å)	21.8708 (7)
α (°)	88.401 (3)
β (°)	83.848 (3)
γ (°)	64.446 (4)
V (Å^3)	2144.4 (2)
Z	2
ρobs, ρcal (g cm^−3)	1.44 (5), 1.472
F(000)	993
μ (mm^−1)	2.923
Crystal size (mm)	0.14 × 0.05 × 0.01
Absorption correction	Analytical
Radiation, λ (Å)	1.54184
Temperature (K)	100.0 (2)
Reflections collected, unique	17468, 8113 (Rint = 0.051)
Limiting indices	−9 ≤h ≤ 11, −13 ≤ k ≤ 13, −26 ≤ l ≤ 26
Refinement method	Full-matrix least-squares on F^2
Final R indices [I > 2σ(I)]†	R1 = 0.0609, wR2 = 0.1516
R indices (all data)†	R1 = 0.0984, wR2 = 0.1742
Goodness of fit on F^2	1.012
Parameters, restraints	599, 4

†R1 = [(|F_o|−|F_c|)]/|F_o|. wR2 = [[w|F_o|²−|F_c|²)²]/[w(|F_o|²)²]^1/2.

Diameter values of the channels for compounds 1, 2 and 3 were calculated using the program TOPOS (available at https://www.topos.ssu.samara.ru).

2.4. Physicochemical characterization techniques

The IR spectra were collected on a JASCO FT/IR-6100 spectrometer at room temperature in the range 4000–400 cm⁻¹, in KBr pellets (1% sample). C, H, N, S and O elemental analyses were measured using a Euro EA 3000 elemental analyser. Thermogravimetric analyses were carried out using a NETZSCH STA 449 F3 thermobalance. A crucible containing approximately 10 mg of sample was heated at 5 K min⁻¹ in the temperature range 303–873 K.

TEM work was done on a Philips SuperTwin CM200 operated at 200 kV and equipped with an LaB₆ filament and EDAX EDS microanalysis system. The samples for TEM analysis were prepared by dispersion into ethanol and keeping the suspension in an ultrasonic bath for 15 min. After that, a drop of the suspension was spread onto a TEM copper grid (300 mesh) covered by a holey carbon film, followed by drying under vacuum.

2.5. Catalytic activity

The oxidation reactions of benzyl alcohol, 1-phenyl­ethanol, 1-hexanol and 1-octanol were carried out at 343 K using aceto­nitrile as the solvent. The catalyst/substrate molar ratio (based on the metal) used for all the reactions was 5:100. Powdered crystals of the catalysts were initially dried at 373 K under vacuum to remove solvent and water adsorbed on the surface.

Before the reactions, approximately 5 mg of dried catalyst was activated by stirring with the oxidizing agent tert-butyl hydro­peroxide (TBHP) in 2 ml of aceto­nitrile for 30 min at 343 K. After this activation stage, the catalyst was separated from the liquid medium by centrifugation. The reactor was then charged with the activated catalyst and the corresponding alcohol in 2 ml of solvent. The mixture was heated to 343 K and then the oxidizing agent was added dropwise (1.5 equivalents of TBHP).

Aldol condensation reactions of benzaldehyde, p-tolu­aldehyde, p-methoxybenzaldehyde and heptanal were carried out at 373 K without solvent. The catalyst/substrate molar ratio (based on the metal) used for all the reactions was 10:100. Powdered crystals of the catalysts were first dried at 373 K under vacuum to remove solvent and water adsorbed on the surface. The reactor was charged with the catalyst (10 mg), acetone (1 ml) and the corresponding substrate, and the mixture was then heated to 373 K.

Knoevenagel condensation reactions of benzaldehyde, p-tolu­aldehyde, p-fluoro­benzaldehyde, 4-chloro­benzaldehyde and 3-nitro­benzaldehyde were carried out at 343 K using toluene as the solvent. The catalyst/substrate molar ratio (based on the metal) used for all the reactions was 5:100. Powdered crystals of the catalysts were first dried at 373 K under vacuum to remove solvent and water adsorbed on the surface. The reactor was charged with the catalyst (5 mg), malono­nitrile (4.6 mg), do­decane as internal standard (2.0 µl) and the corresponding substrate in 2 ml of solvent, and then the mixture was heated to 343 K.

The one-pot cascade reaction was tested for acetal hydrolysis followed by Knoevenagel condensation at 343 K in toluene. The catalyst/substrate molar ratio (based on the metal) used for the reaction was 10:100. Powdered crystals of the catalysts were first dried at 373 K under vacuum to remove solvent and water adsorbed on the surface. The reactor was charged with the catalyst (5.2 mg), toluene (2 ml), benz­aldehyde di­methyl acetal (5.3 µl), malono­nitrile (2.3 mg) and do­decane (2 µl) as internal standard, and then the mixture was heated to 343 K.

Detailed results for the catalytic activity exhibited by compounds 1, 2 and 3 will be described in Sections 3.3 and 3.4.

Reaction samples were taken at regular times and analysed on a Hewlett–Packard 5890 II GC–MS or on a Konik HCGC 5000B gas chromatograph–mass spectrometer. Blank experiments were carried out under the reaction conditions in order to determine the extent of the uncatalysed reaction; for all the blank reactions only traces of the product were found. After the reaction, the catalysts were filtered, dried and characterized by IR spectroscopy, and by TEM microscopy in some cases. The calculations of turnover frequencies (TOF; turnover frequency is mol substrate converted/mol catalyst per hour) were performed in the initial stages of the reaction, when the reaction rates are higher, as usual.

3. Results and discussion

3.1. Crystal structures

The structural features of these materials are of great importance in order to achieve satisfactory conversion rates for catalytic reactions. In this sense, the metalloporphyrinic synthons are intended to act as heterogeneous catalysts and structural building units at the same time. In relation to the latter, accessibility of the guest molecules to catalytic metal centres on the surface of the crystal-network cavities is one of the most important features necessary to consider the compounds as potential catalysts, and was taken into account for compounds 1, 2 and 3.

Compound 1 with the formula [H(bipy)]₂[(MnTPPS)(H₂O)₂]·2bipy·14H₂O is a coordination compound consisting of complex ions. The crystal structure, determined by single-crystal X-ray diffraction, shows [(MnTPPS)(H₂O)₂]²⁻ anionic monomers where TPPS⁴⁻ ligands are present. The Mn^II ion is in an octahedral coordination environment, bonded to the four porphyrin N atoms and with two water molecules in the axial positions.

The [(MnTPPS)(H₂O)₂]²⁻ anions crystallize as shown in Fig. 1. The voids generated between metalloporphyrinic monomers are occupied by [H(bipy)]⁺ cations and crystallized bi­pyridine molecules and, as shown in Fig. S3, those bi­pyridine molecules were pairwise hydrogen-bonded [N5—H1N⋯N4; 2.741 7 Å and 171 9°] and parallel-stacked, giving rise to robust π–π interactions (centroid-to-centroid distances are 3.468 and 3.746 Å) (Soltanzadeh & Morsali, 2009). Additionally, the interstitial voids are occupied by 14 crystallized water molecules per monomer.

Figure 1 Crystal-structure packing for compound 1. Colour code: Mn pink, C grey, N blue, O red and S yellow. H atoms have been omitted for clarity.

The crystal structure is stabilized by an intricate hydrogen-bonded system (Table S4), connecting the [(MnTPPS)(H₂O)₂]²⁻ units along the direction between the axial water molecule (O7) and the sulfonate groups (O3) (Fig. S4). Additionally, the metalloporphyrinic monomers are linked by a hydrogen-bonded chain along the [100] direction. This connection involves the coordinated water molecules (O7) and two crystallized water molecules (O8 and O11). It is worth noting that there is a zigzag chain of water molecules along the [100] direction which stabilizes the structure. This chain is located between the sulfonate groups involving the O12 to O16 crystallized water molecules (Fig. 2).

Figure 2 Representation of the accessibility of guest molecules to the active metal centres for compound 1. Colour code: Mn pink, C grey, N blue, O red and S yellow. The accessible pathway is represented by the green cylinder.

As shown in Fig. 1 and Fig. S5, the [H(bipy)₂]⁺ cations are located on the interporphyrinic voids and linked through O9, O10 and O12 water molecules to the sulfonate groups (O2) and through O8 to the axial water molecules (O7).

Fig. 2 shows the accessibility of the crystal structure to external guest molecules in order to access channels along the x axis. The calculated value of the diameter, obtained using TOPOS (Section 2.3), is 4.3 Å. As shown, the porphyrin units are separated by 9.7187 (4) Å (the a cell parameter), allowing the interaction of potential guest molecules with the active metal centres.

The crystal structure and thermal analysis of compound 2 (μ-O-[FeTCPP]₂·16DMF) have been reported previously by us (Fidalgo-Marijuan et al., 2015). This compound consists of FeTCPP dimers, where monomers are connected by an O-bridge. In this way, the active catalytic centres are exposed to the channels of the framework. The calculated value of the diameter, obtained using TOPOS, is 5.2 Å. As shown in Fig. S6, the dimers are packed forming hydrogen-bonded layers in the xy plane, generating the aforementioned channels along the [001] direction, where potential guest molecules are then distributed to the active centres (Fig. 3).

Figure 3 Representation of the accessibility of guest molecules to the active metal centres for compound 2. Colour code: Fe purple, C grey, N blue and O red. Accessible pathways are represented by the green cylinders.

The crystal structure and thermal analysis for compound 3, with the formula [CoTPPS_0.5(bipy)(H₂O)₂]·6H₂O, have also been reported by us in previous work (Fidalgo-Marijuan et al., 2013). The crystal structure is formed by one-dimensional polymers, where octahedral Co^II ions of CoTPPS units are axially bonded to bipy ligands (Fig. S7). The extension of the one-dimensional polymers consists of a link between alternating Co^II centres along the [001] direction through the bipy ligands in a bipy–CoTPPS–bipy–Co(H₂O)₄ fashion. In this case, the accessibility to active metal centres takes place along the direction where the active metal centres are exposed to the channels of the framework (Fig. 4). The calculated value for the diameter obtained using TOPOS is 6.1 Å.

Figure 4 Representation of the accessibility of guest molecules to the active metal centres for compound 3. Colour code: Co orange, C grey, N blue, O red and S yellow. Accessible pathways are represented by the green cylinders.

3.2. Thermal analysis

Thermal behaviour is crucial in determining the stability and correct catalytic activation of these materials. As observed in Fig. 5, the thermogravimetric decomposition curve of compound 1 shows a continuous mass loss from room temperature to 813 K, where three steps can be distinguished. The first is assigned to an overlapped two-stage step between room temperature and 673 K. The first stage occurs up to 468 K, with a 12.1% weight loss attributed to the coordinated and crystallized water molecules. The second stage, up to 673 K (19.7% weight loss), corresponds to the crystallized bi­pyridine molecules. The second step, occurring between 673 K and 723 K, with a 14.3% weight loss, is attributed to the bi­pyridine molecules formerly present as [H(bipy)]⁺ cations. The latter assignment is based on the fact that these cationic entities are more robustly linked than their crystallized analogues. The last step, between 723 K and 813 K (42.7% weight loss), is the final degradation of the TPPS units. The calcination product was identified by X-ray powder diffraction analysis and it consists of Mn₂O₃ (space group Rc, a = 5.04, c = 14.12 Å, γ = 120°; Lee et al., 1980).

Figure 5 Thermal analysis for compound 1.

Compounds 2 and 3 have been thermally characterized in our previous work (Fidalgo-Marijuan et al., 2015, 2013), showing high thermal stability (compound 2 up to 593 K and 3 up to 633 K). The thermogravimetry/differential scanning calorimetry (TG/DSC) curve for activated compound 2 is shown in Fig. S8.

Additionally, Fig. S9 shows the X-ray thermo-diffraction measurements (XRTD) of powdered single crystals for compounds 1 and 2. The results indicate that both compounds were thermally stable after activation for catalytic purposes.

3.3. Catalytic properties

Synthetic metalloporphyrin complexes have been largely used for a wide variety of catalytic transformations (Chatterjee et al., 2016; Rayati et al., 2016; Sengupta et al., 2016; Zhou et al., 2016), and in this work we have explored the catalytic activity of the monomeric framework [H(bipy)]₂[(MnTPPS)(H₂O)₂]·2bipy·14H₂O, 1, the dimeric framework μ-O-[FeTCPP]₂·16DMF, 2 (Fidalgo-Marijuan et al., 2015), and the one-dimensional framework [CoTPPS_0.5(bipy)(H₂O)₂]·6H₂O, 3 (Fidalgo-Marijuan et al., 2013). As previously shown, these three compounds exhibit features that make them suitable candidates for catalysis of different reactions. Firstly, the metal coordination spheres are either unsaturated or there are water molecules that are easy to remove during the activation stage, as shown by the thermogravimetric analysis. Fig. S10 shows a simplified scheme for 1, 2 and 3, highlighting the catalytic active centres. In addition, the networks are significantly accessible, with mobile DMF or water solvent molecules located in the cavities.

Thus, the catalytic performance was studied for the oxidation of alcohols, aldol and Knoevenagel condensations, and a one-pot cascade reaction for an acetal hydrolysis followed by a C—C Knoevenagel condensation (Scheme 1). The studied substrates for all the reactions are summarized in Table S5.

3.3.1. Oxidation of alcohols

The selective oxidation of alcohols to aldehydes is a relevant transformation in waste recovery and in organic synthesis because of the properties and chemical reactivity of carbonylic compounds that make aldehydes the preferred starting materials in many syntheses (Davis et al., 2013). Although considerable progress has been made using noble-metal nanoparticles such as Au (Corma & Garcia, 2008), it would still be desirable to develop catalysts based on less expensive metals and processes. In this sense, some porphyrinic compounds have been tested (Zou et al., 2013; Machado et al., 2013; Chen & Ma, 2016) in an effort to minimize costs and waste generation.

The reaction conditions were initially set using benzyl alcohol as a model substrate. Based on our previous experience (Fidalgo-Marijuan et al., 2015; Larrea et al., 2011), the reactions were carried out using tert-butyl hydro­peroxide (TBHP) as the oxidizing agent in aceto­nitrile. Using 5% catalyst and 1.5 equivalents of TBHP in 2 ml of solvent at 343 K, total conversions of 50% for 1, 73% for 2 and 71% for 3 were achieved after 7 h of reaction. The scope of the reaction was studied with various alcohols: 1-phenyl­ethanol, 1-hexanol and 1-octanol (Table 2). Figs. S11, S12 and S13 show the kinetic profiles of these oxidation reactions for 1, 2 and 3.

Table 2
Selective oxidation of several alcohols over catalysts 1, 2 and 3

C_T is the total conversion.

		Compound 1			Compound 2			Compound 3
Substrate	Oxidant	TOF (h^−1)	CT (%)	t (h)	TOF (h^−1)	CT (%)	t (h)	TOF (h^−1)	CT (%)	t (h)
Benzyl-alcohol	TBHP	72	70	20	72	73	7	143	77	20
1-Phenyl­ethanol	TBHP	46	44	24	91	73	5	8	44	21
1-Hexanol	TBHP	66	92	24	3	15	6	22	71	24
1-Octanol	TBHP	2	12	24	6	9	18	2	25	24

The best result for 1 is achieved towards the oxidation of 1-hexanol, since 2 is better for 1-phenyl­ethanol. For benzyl alcohol oxidation, the conversion values are quite similar for the three catalysts, though compound 3 clearly shows a higher TOF. The poor conversion values achieved for 1-octanol in all cases could be explained by the fact that this substrate presents more steric hindrance. A comparison of these results with similar porphyrinic catalysts found in the literature indicates a significant reduction in the reaction time (half time) for catalyst 2 (Modak et al., 2013). Moreover, comparing the benzyl alcohol results with classic Rh-, Ru- and Ce-based catalysts showed the conversion rates are slightly higher and have much shorter reaction times (Wusiman & Lu, 2015; Burange et al., 2013; Sarkar et al., 2014).

One of the disadvantages of heterogeneous catalysis is the difficulty of studying the reaction mechanisms; often the involved intermediates are unknown. Even so, in the proposed mechanism for alcohol oxidation the initial stage consists of activation of TBHP by coordination to the unsaturated metal centre in order to obtain the corresponding peroxo species. After coordination, tert-butoxyl radicals are generated, abstracting an H atom from the substrate and leading to the corresponding aldehyde or ketone (Fig. 6) (Orive et al., 2013).

Figure 6 Proposed mechanism for the alcohol oxidations with TBHP over catalyst 2. Colour code: Fe purple, C grey, N blue and O red. Bonds to be removed and created are marked in red and blue, respectively.

3.3.2. Aldol condensation

Aldol condensations are important in organic synthesis, providing a way to form C—C bonds; the identification of catalysts capable of performing C—C bond formation remains a challenge (Scheme 1b) (Thankachan et al., 2015). The main drawbacks of using the usual NaOH or KOH catalysts are corrosion problems in the equipment, separation difficulties and the generation of large amounts of waste. To overcome these disadvantages, efforts have been made to design new catalytic systems with controlled basic properties in order to increase the efficiency of the process. In this context, metalloporphyrinic catalysts 1, 2 and 3 were tested towards the aldol condensation of benz­aldehyde and derivatives with acetone.

The reaction conditions were initially set using benz­aldehyde as the substrate, using 5% catalyst in 1 ml of acetone at 343 K, obtaining poor conversion values (31% for catalyst 1, 38% for 2 and 2.5% for 3). However, increasing the catalyst amount to 10% and the reaction temperature to 373 K, the total conversions were 79% for 2 and 16% for 3 after 38 h of reaction. Unfortunately, at this temperature, 1 shows chemical decomposition caused by the loss of the [H(bipy)]⁺ cations. Thus, the scope of the reaction was studied with p-tolu­aldehyde, p-methoxybenzaldehyde and heptanal using compound 2 as the catalyst. Table 3 and Fig. S14 show the conversion values and kinetic profiles for the aldol condensation, respectively.

Table 3
Aldol condensation with acetone over catalyst 2

Substrate	TOF (h^−1)	CT (%)	t (h)
Benzaldehyde	3.5	79	93
p-Tolu­aldehyde	2.3	92	87
p-Methoxybenzaldehyde	1.4	50	87
Heptanal	152	94	4

As observed, the methyl group in the para position leads to high yields, whereas meth­oxy groups give poor conversion. In the case of heptanal, the reaction evolves much more quickly, giving rise to the heptanal autocondensation product (Fig. S15a) in a minority amount (7%) and two isomers (double-bond position) of the aldol condensation: the dec-3-en-2-one (52%) (Fig. S15b) and the dec-4-en-2-one (35%) (Fig. S15c). Comparing the results with other MOF-type catalysts, we conclude that, for aromatic substrates, it is necessary to increase the reaction time considerably (up to three times) to obtain similar conversion rates (Abedi et al., 2016). Nevertheless, the reaction with heptanal shows excellent activity using four times less catalyst than in similar reported studies (Yadav & Aduri, 2012).

3.3.3. Knoevenagel condensation

Taking into account the previous results, compound 2 was also tested as a catalyst for the Knoevenagel condensation reaction (Scheme 1c) between various substrates and derivatives (Table 4) and malono­nitrile (pK_a = 11.1). As above, the reaction conditions were set using benzaldehyde as the substrate, 5% catalyst, 1.0 equivalent of malono­nitrile in 2 ml of toluene and 2 µl of do­decane as internal standard at 343 K, reaching a total conversion of 79% after 22 h of reaction (Table 4). The scope of the reaction was then studied with p-tolu­aldehyde, p-fluoro­benzaldehyde, 4-chloro­benzaldehyde and 3-nitro­benzaldehyde. Fig. S16 shows the kinetic profiles of the Knoevenagel condensation reactions.

Table 4
Knoevenagel condensation over catalyst 2

Substrate	TOF (h^−1)	CT (%)	t (h)
Benzaldehyde	49	79	22
p-Tolu­aldehyde	53	81	24
p-Fluoro­benzaldehyde	32	69	24
p-Chloro­benzaldehyde	137	100	24
m-Nitro­benzaldehyde	129	100	18

As observed, when introducing substituents into the meta- or para-positions of the benzaldehyde ring, the conversion rates differ as a result of electronic effects, following the order of reactivity: m-nitro­benzaldehyde ≃ p-chloro­benz­aldehyde > p-tolu­aldehyde ≃ benzaldehyde > p-fluoro­benz­aldehyde.

For both aldol and Knoevenagel condensations, the activation of the carbonyl groups is prompted by a Lewis acid (the metal ion). The proximity of the active centres (Fe ions) meant that a mechanism involving two-site adsorption of the donor molecule could be proposed (Fig. 7) (Larrea et al., 2015; Položij et al., 2014). In fact, the distance of 5.656 (1) Å between Fe^III ions allowed the cyanide groups of malono­nitrile [N⋯N distance 4.319 (3) Å] to interact with the Fe active centres. Comparing these results with other MOFs (Larrea et al., 2015), the obtained conversion rates and TOF are among the best using only 5% catalyst under similar reaction conditions.

Figure 7 Proposed mechanism for the benzyl alcohol Knoevenagel condensation over catalyst 2. Colour code: Fe purple, C grey, N blue and O red. Bonds to be removed (red dashed lines) and created (turquoise dashed lines) are shown.

3.3.4. One-pot cascade reaction

Taking into account that compound 2 showed the best results for Knoevenagel condensation, it was tested in a bifunctional one-pot cascade reaction (Scheme 1d). The first step involves an acetal hydrolysis using benzaldehyde di­methyl acetal as the substrate, followed by a second step in which Knoevenagel condensation with malono­nitrile takes place. The results (TOF: 10 h⁻¹, CT: 42% in 44 h) indicate that, although the Knoevenagel condensation reaches a high conversion (79%; Table 4), in this case the rate-limiting step is the reaction involving the acetal hydrolysis. Although in the first few hours the TOF is quite high, after 3 h the conversion rate decelerates, probably because of the inability of the catalyst to overcome the first step of the reaction (Fig. S17).

An excellent strategy to carry out this type of one-pot two-step reaction is to introduce, in the same solid catalyst, both acidic and basic active centres (Merino et al., 2013). Compound 2 shows only Lewis acid centres to catalyse the hydrolysis of benzaldehyde di­methyl­acetal into benzaldehyde and react with malono­nitrile through Knoevenagel condensation. This may be the reason for the lower conversion in the one-pot two-step reaction (42%) compared with the single Knoevenagel condensation (79%) for 2.

3.4. Heterogeneity and recyclability tests

The heterogeneous nature of catalysts 1 and 3 towards the oxidation of alcohols were tested using benzyl alcohol (2 has been tested previously; Fidalgo-Marijuan et al., 2015). For rigorous proof of heterogeneity, a test (Sheldon et al., 1998) was carried out by filtering the catalyst from the reaction mixture at 343 K after 20 min, when conversions of 24% and 47% had been reached for 1 and 3, respectively. The filtrate was allowed to react for up to 7 h. The reaction mixture and the filtrate were analysed afterwards by GC–MS. No significant change in the conversion rate was found for the filtrate (Fig. 8), meaning that the active species does not leach and the observed catalysis is truly heterogeneous in nature.

Figure 8 Kinetic profiles of the oxidation of benzyl alcohol (a) over compound 1 and (b) over compound 3 after hot filtering with TBHP.

The recyclability of catalysts 1 and 3 was also tested for the benzyl alcohol oxidation. The catalyst was recovered after the reaction by centrifugation and washed several times with aceto­nitrile, then dried at 373 K and reused. As shown in Table S6, catalyst 3 maintains activity after five cycles, whereas 1 shows a small decrease.

As was previously done for the oxidation reactions, the heterogeneous nature of the catalyst and recyclability were tested or the aldol and Knoevenagel condensations by hot filtration. As shown in Fig. 9, the experiments reveal that 2 is a truly heterogeneous catalyst for this reaction. Additionally, the catalyst was reused over five cycles and shows a progressive decrease in catalytic activity after the third cycle (Table S7).

Figure 9 Kinetic profile of the condensation of benzaldehyde with malono­nitrile over compound 2 after hot filtering.

After the catalytic reactions, the catalysts were recovered by centrifugation, washed with aceto­nitrile, ethanol or toluene and then characterized by IR spectroscopy. The IR spectra of the recovered catalyst for all reactions showed that the chemical-bond systems remained unchanged (Fig. S18). In fact, the solid shows the same characteristic vibration modes as the original compound. As shown in Fig. S18, the characteristic vibrations of the porphyrin macrocycle are present. Additionally, both the fresh catalyst and the recovered solid after the reaction were studied by TEM, as discussed below.

3.5. Transmision electron microscopy

In order to carry out a deeper characterization of the recovered catalyst, compounds 2 and 3 were analysed by TEM before and after the catalytic reactions. Compound 1 is unstable under an electron beam, so it was not analysed by TEM. TEM analysis shows that compounds 2 and 3 maintain a certain grade of crystallinity following the catalytic reactions. Moreover, the pre- and post-catalysis particles of 2 and 3 keep their morphology: elongated prisms for 2 and curved-edge particles for 3 (Fig. S19).

The crystalline nature and moderate stability of samples 2 and 3 under an electron beam allowed us to measure the lattice spacing for both compounds. The HRTEM images (Fig. 10) of the pristine sample (a) and the recovered residue after the catalytic reaction (b) of compound 2 reveal a lattice spacing of 13.45 Å along the width of the crystal. This observed lattice spacing corresponds to the (101), and (210) set of crystallographic planes. For compound 3, the pristine sample (c) and the recovered residue (d) present a spacing near to 14.02 Å, which corresponds to the (011) set of crystallographic planes. Therefore, both compounds maintain the same lattice spacing before and after the catalytic reactions so, as previously observed by IR spectroscopy, compounds 2 and 3 keep their structural integrity.

Figure 10 HRTEM images for pristine compounds (a) 2 and (c) 3 and the recovered residues after the catalytic reactions for (b) 2 and (d) 3. Lattice spacing is marked by yellow lines and the upper right image corresponds to the Fourier transform of the area inside the blue square.

4. Conclusions

This work explores an innovative approach that consists of using metalloporphyrins as both synthons and catalytic units in solid coordination networks. This strategy is compatible with green synthesis, i.e. using first-row transition metals (Mn, Fe and Co), with water as the preferred solvent and performing syntheses at low temperatures. While most of the studies involving solid coordination networks are focused on three-dimensional covalent MOFs, the results presented here indicate that accessibility of reactants to the metal centres is a significant parameter in achieving good heterogeneous catalytic activity. The studied compounds are perfectly operative under ambient conditions as they exhibit recyclability and thermochemical stability. Furthermore, compounds 2 and 3 are remarkably stable upon heating. An improvement in catalytic performance has been achieved in relation to commonly used toxic metal-based catalysts. In summary, bioinspired metalloporphyrinic SMOFs are promising candidates as heterogeneous and recyclable catalysts.

Supporting information

Computing details

Data collection: CrysAlis PRO, Agilent Technologies, Version 1.171.35.11 (release 16-05-2011 CrysAlis171 .NET) (compiled May 16 2011,17:55:39); cell refinement: CrysAlis PRO, Agilent Technologies, Version 1.171.35.11 (release 16-05-2011 CrysAlis171 .NET) (compiled May 16 2011,17:55:39); data reduction: CrysAlis PRO, Agilent Technologies, Version 1.171.35.11 (release 16-05-2011 CrysAlis171 .NET) (compiled May 16 2011,17:55:39); program(s) used to solve structure: SHELXS97 (Sheldrick, 1990); program(s) used to refine structure: SHELXL97 (Sheldrick, 1997).

(I)

Crystal data

C22H15Mn0.50N2O7S2·(C10H8N2)·(C10H9N2)·7(H2O)	Z = 2
Mr = 950.44	F(000) = 993
Triclinic, P1	Dx = 1.472 Mg m^−3 Dm = 1.446 (5) Mg m^−3 Dm measured by Flotation
a = 9.7187 (4) Å	Cu Kα radiation, λ = 1.54184 Å
b = 11.2496 (5) Å	Cell parameters from 3937 reflections
c = 21.8708 (7) Å	θ = 4.1–74.1°
α = 88.401 (3)°	µ = 2.92 mm^−1
β = 83.848 (3)°	T = 100 K
γ = 64.446 (4)°	Prism, dark red
V = 2144.39 (15) Å^3	0.14 × 0.05 × 0.01 mm

Data collection

SuperNova, Single source at offset), Atlas diffractometer	8113 independent reflections
Radiation source: fine-focus sealed tube	5439 reflections with I > 2σ(I)
Graphite monochromator	Rint = 0.051
Detector resolution: 5.2012 pixels mm^-1	θmax = 70.0°, θmin = 4.1°
ω scans	h = −9→11
Absorption correction: analytical CrysAlisPro, Agilent Technologies, Version 1.171.35.11 (release 16-05-2011 CrysAlis171 .NET) (compiled May 16 2011,17:55:39) Analytical numeric absorption correction using a multifaceted crystal model based on expressions derived by R.C. Clark & J.S. (Clark, R. C. & Reid, J. S. (1995). Acta Cryst. A51, 887-897)	k = −13→13
Tmin = 0.826, Tmax = 0.968	l = −26→26
17468 measured reflections

Refinement

Refinement on F^2	Primary atom site location: structure-invariant direct methods
Least-squares matrix: full	Secondary atom site location: difference Fourier map
R[F^2 > 2σ(F^2)] = 0.061	Hydrogen site location: inferred from neighbouring sites
wR(F^2) = 0.174	H atoms treated by a mixture of independent and constrained refinement
S = 1.01	w = 1/[σ^2(Fo^2) + (0.0799P)^2 + 1.0855P] where P = (Fo^2 + 2Fc^2)/3
8113 reflections	(Δ/σ)max < 0.001
599 parameters	Δρmax = 0.53 e Å^−3
4 restraints	Δρmin = −0.44 e Å^−3

Special details

Geometry. All esds (except the esd in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell esds are taken into account individually in the estimation of esds in distances, angles and torsion angles; correlations between esds in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell esds is used for estimating esds involving l.s. planes.

Refinement. Refinement of F² against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F², conventional R-factors R are based on F, with F set to zero for negative F². The threshold expression of F² > 2sigma(F²) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F² are statistically about twice as large as those based on F, and R- factors based on ALL data will be even larger.

Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Ų)

	x	y	z	Uiso*/Ueq	Occ. (<1)
C1	0.9495 (4)	0.0089 (4)	0.86434 (16)	0.0273 (8)
C2	0.8818 (4)	0.1008 (4)	0.81719 (16)	0.0316 (8)
H2	0.8754	0.0795	0.7762	0.038*
C3	0.8290 (4)	0.2230 (4)	0.84204 (16)	0.0324 (8)
H3	0.7784	0.3037	0.8217	0.039*
C4	0.8633 (4)	0.2087 (4)	0.90482 (16)	0.0273 (8)
C5	0.8294 (4)	0.3134 (4)	0.94507 (16)	0.0274 (8)
C6	0.8582 (4)	0.3003 (4)	1.00647 (17)	0.0274 (7)
C7	0.8193 (4)	0.4074 (4)	1.04849 (17)	0.0305 (8)
H7	0.7759	0.4984	1.039	0.037*
C8	0.8555 (5)	0.3562 (4)	1.10404 (17)	0.0306 (8)
H8	0.8408	0.4044	1.141	0.037*
C9	0.9204 (4)	0.2152 (3)	1.09724 (16)	0.0256 (7)
C10	1.0239 (4)	−0.1277 (4)	0.85537 (16)	0.0273 (8)
C11	0.7547 (4)	0.4501 (3)	0.92070 (16)	0.0282 (8)
C12	0.8397 (6)	0.5117 (6)	0.8944 (3)	0.079 (2)
H12	0.9484	0.4683	0.8926	0.094*
C13	0.7704 (6)	0.6359 (6)	0.8703 (4)	0.083 (2)
H13	0.8314	0.6771	0.852	0.099*
C14	0.6130 (4)	0.7001 (3)	0.87284 (16)	0.0264 (7)
C15	0.5276 (5)	0.6409 (4)	0.9002 (2)	0.0445 (11)
H15	0.4188	0.6856	0.9032	0.053*
C16	0.5973 (5)	0.5159 (4)	0.9239 (2)	0.0438 (11)
H16	0.5361	0.4752	0.9425	0.053*
C17	1.0508 (4)	−0.1838 (3)	0.79172 (16)	0.0279 (8)
C18	0.9326 (5)	−0.1859 (4)	0.76185 (18)	0.0338 (9)
H18	0.8309	−0.1498	0.7818	0.041*
C19	0.9626 (5)	−0.2407 (4)	0.70298 (19)	0.0393 (10)
H19	0.8815	−0.2419	0.6826	0.047*
C20	1.1103 (5)	−0.2934 (4)	0.67417 (17)	0.0361 (9)
C21	1.2292 (5)	−0.2910 (4)	0.70291 (18)	0.0373 (9)
H21	1.3304	−0.326	0.6826	0.045*
C22	1.1987 (5)	−0.2366 (4)	0.76196 (17)	0.0337 (9)
H22	1.28	−0.2357	0.7822	0.04*
C23	−0.1755 (13)	−0.4032 (7)	0.2379 (3)	0.114 (4)
H23	−0.1813	−0.4854	0.2407	0.137*
C24	−0.1004 (11)	−0.3722 (6)	0.2801 (3)	0.098 (3)
H24	−0.0515	−0.4351	0.3099	0.118*
C25	−0.0941 (6)	−0.2517 (4)	0.2801 (2)	0.0481 (12)
C26	−0.1697 (6)	−0.1681 (5)	0.2344 (3)	0.0681 (17)
H26	−0.1754	−0.0816	0.2327	0.082*
C27	−0.2355 (6)	−0.2069 (5)	0.1923 (3)	0.0651 (16)
H27	−0.2802	−0.1484	0.1603	0.078*
C28	−0.0207 (6)	−0.2142 (4)	0.3270 (2)	0.0478 (12)
C29	0.0774 (6)	−0.3079 (5)	0.3644 (3)	0.0588 (14)
H29	0.0998	−0.3982	0.3595	0.071*
C30	0.1410 (7)	−0.2700 (5)	0.4081 (3)	0.0644 (15)
H30	0.2063	−0.3353	0.4335	0.077*
C31	−0.0454 (7)	−0.0836 (4)	0.3366 (2)	0.0561 (14)
H31	−0.1103	−0.016	0.312	0.067*
C32	0.0224 (7)	−0.0532 (5)	0.3805 (2)	0.0586 (15)
H32	0.0034	0.0363	0.386	0.07*
C33	0.4985 (7)	0.2878 (6)	0.6435 (3)	0.0697 (17)
H33	0.4908	0.3737	0.6354	0.084*
C34	0.4365 (7)	0.2345 (5)	0.6048 (3)	0.0692 (16)
H34	0.3868	0.2843	0.5713	0.083*
C35	0.4457 (5)	0.1098 (5)	0.6140 (2)	0.0499 (12)
C36	0.5189 (7)	0.0452 (6)	0.6649 (3)	0.0661 (16)
H36	0.5298	−0.0414	0.6736	0.079*
C37	0.5753 (7)	0.1059 (6)	0.7024 (3)	0.0701 (17)
H37	0.6209	0.0603	0.7374	0.084*
C38	0.3791 (5)	0.0507 (5)	0.5743 (2)	0.0518 (13)
C39	0.2766 (6)	0.1258 (5)	0.5332 (2)	0.0556 (13)
H39	0.2497	0.2175	0.5304	0.067*
C40	0.2146 (7)	0.0699 (6)	0.4970 (3)	0.0611 (14)
H40	0.1454	0.1235	0.4692	0.073*
C41	0.3478 (7)	−0.1349 (6)	0.5378 (3)	0.0703 (16)
H41	0.373	−0.2264	0.5393	0.084*
C42	0.4138 (6)	−0.0842 (5)	0.5750 (3)	0.0619 (14)
H42	0.4841	−0.1406	0.6017	0.074*
N1	0.9354 (3)	0.0765 (3)	0.91788 (13)	0.0255 (6)
N2	0.9225 (3)	0.1823 (3)	1.03668 (13)	0.0245 (6)
N3	−0.2400 (6)	−0.3224 (4)	0.1937 (2)	0.0650 (13)
N4	0.1154 (5)	−0.1440 (4)	0.4167 (2)	0.0555 (11)
N5	0.2484 (5)	−0.0594 (5)	0.4993 (2)	0.0601 (12)
N6	0.5694 (5)	0.2248 (5)	0.6921 (2)	0.0619 (12)
O1	0.5987 (4)	0.9340 (3)	0.85989 (16)	0.0468 (8)
O2	0.5578 (4)	0.8331 (3)	0.77288 (12)	0.0438 (8)
O3	0.3623 (3)	0.9093 (3)	0.85953 (12)	0.0353 (6)
O4	1.073 (4)	−0.243 (4)	0.5621 (17)	0.043 (6)	0.5
O5	1.3143 (19)	−0.4228 (17)	0.5860 (7)	0.046 (3)	0.5
O6	1.079 (2)	−0.4466 (10)	0.5964 (6)	0.057 (4)	0.5
O4B	1.109 (4)	−0.245 (4)	0.5559 (18)	0.042 (6)	0.5
O5B	1.306 (3)	−0.447 (2)	0.5844 (9)	0.117 (10)	0.5
O6B	1.052 (3)	−0.422 (2)	0.5937 (9)	0.111 (8)	0.5
O7	1.2327 (3)	−0.0046 (3)	0.97599 (11)	0.0311 (6)
H7A	1.2745	−0.032	0.9413	0.047*
H7B	1.2277	0.0686	0.9823	0.047*
S1	0.52566 (11)	0.85690 (8)	0.83888 (4)	0.0294 (2)
S2	1.14667 (15)	−0.35683 (10)	0.59723 (4)	0.0442 (3)
Mn1	1	0	1	0.0227 (2)
O8	0.5535 (3)	1.1354 (2)	0.93990 (11)	0.0447 (7)
H8A	0.6146	1.1074	0.9659	0.067*
H8B	0.5616	1.0721	0.9191	0.067*
O9	−0.4426 (4)	−0.3526 (3)	0.13067 (17)	0.0741 (12)
H9A	−0.3825	−0.3419	0.1517	0.111*
H9B	−0.4814	−0.2857	0.1099	0.111*
O10	0.5546 (2)	0.43141 (18)	0.76702 (9)	0.0881 (15)
H10A	0.5831	0.4823	0.7476	0.132*
H10B	0.5819	0.3633	0.7461	0.132*
O11	0.2477 (3)	1.2320 (2)	0.98446 (13)	0.159 (4)
H11A	0.3407	1.1949	0.9861	0.239*
H11B	0.2276	1.2548	0.9493	0.239*
O12	0.6479 (2)	0.6008 (2)	0.70792 (9)	0.130 (3)
H12B	0.6624	0.6257	0.6732	0.195*
H12A	0.6128	0.6637	0.7324	0.195*
O13	0.6131 (4)	−0.39526 (18)	0.59422 (15)	0.107 (3)*	0.5
H13B	0.6391	−0.3935	0.5573	0.16*	0.5
H13A	0.656	−0.363	0.6145	0.16*	0.5
O15	0.8695 (3)	−0.4201 (2)	0.51535 (18)	0.077 (2)*	0.5
H15A	0.7977	−0.4326	0.5049	0.115*	0.5
H15B	0.9425	−0.4495	0.4884	0.115*	0.5
O14	0.7365 (4)	−0.33992 (18)	0.59153 (16)	0.070 (2)*	0.5
H14A	0.7218	−0.3823	0.5655	0.106*	0.5
H14B	0.8103	−0.3234	0.5808	0.106*	0.5
O16	0.5871 (4)	−0.45657 (18)	0.52983 (16)	0.092 (3)*	0.5
H16A	0.5231	−0.4781	0.5171	0.137*	0.5
H16B	0.6257	−0.5054	0.5581	0.137*	0.5
H6	0.985 (13)	−0.412 (7)	0.589 (18)	0.137*	0.5
H6B	0.958 (9)	−0.373 (5)	0.590 (17)	0.137*	0.5
H1N	0.193 (10)	−0.107 (9)	0.461 (4)	0.137*

Atomic displacement parameters (Ų)

	U^11	U^22	U^33	U^12	U^13	U^23
C1	0.0238 (18)	0.0307 (18)	0.0208 (16)	−0.0060 (15)	−0.0018 (13)	0.0080 (14)
C2	0.031 (2)	0.0320 (19)	0.0196 (16)	−0.0026 (16)	−0.0024 (14)	0.0079 (14)
C3	0.032 (2)	0.0309 (19)	0.0230 (17)	−0.0035 (16)	−0.0028 (14)	0.0103 (15)
C4	0.0257 (18)	0.0284 (18)	0.0206 (17)	−0.0061 (15)	0.0006 (13)	0.0098 (14)
C5	0.0246 (18)	0.0254 (17)	0.0245 (17)	−0.0047 (14)	0.0009 (13)	0.0100 (14)
C6	0.0215 (17)	0.0279 (18)	0.0275 (18)	−0.0071 (14)	0.0021 (13)	0.0061 (14)
C7	0.031 (2)	0.0233 (17)	0.0320 (19)	−0.0068 (15)	−0.0070 (15)	0.0102 (14)
C8	0.037 (2)	0.0283 (18)	0.0251 (18)	−0.0127 (16)	−0.0032 (15)	0.0026 (14)
C9	0.0228 (17)	0.0274 (18)	0.0233 (17)	−0.0086 (14)	0.0008 (13)	0.0051 (14)
C10	0.0254 (18)	0.0298 (18)	0.0210 (16)	−0.0073 (15)	−0.0012 (13)	0.0056 (14)
C11	0.0286 (19)	0.0242 (18)	0.0244 (17)	−0.0051 (15)	−0.0012 (14)	0.0049 (14)
C12	0.025 (2)	0.061 (3)	0.138 (6)	−0.010 (2)	−0.010 (3)	0.066 (4)
C13	0.030 (3)	0.059 (3)	0.146 (6)	−0.014 (2)	0.001 (3)	0.065 (4)
C14	0.0280 (18)	0.0222 (16)	0.0220 (16)	−0.0046 (14)	−0.0034 (13)	0.0075 (13)
C15	0.025 (2)	0.033 (2)	0.068 (3)	−0.0065 (17)	−0.0080 (19)	0.022 (2)
C16	0.027 (2)	0.035 (2)	0.069 (3)	−0.0133 (18)	−0.0087 (19)	0.025 (2)
C17	0.032 (2)	0.0245 (17)	0.0215 (17)	−0.0069 (15)	−0.0039 (14)	0.0075 (14)
C18	0.036 (2)	0.034 (2)	0.0290 (19)	−0.0131 (17)	−0.0039 (16)	0.0085 (16)
C19	0.056 (3)	0.035 (2)	0.032 (2)	−0.022 (2)	−0.0188 (19)	0.0110 (17)
C20	0.058 (3)	0.0247 (18)	0.0203 (17)	−0.0132 (18)	−0.0058 (17)	0.0061 (14)
C21	0.046 (2)	0.031 (2)	0.0271 (19)	−0.0097 (18)	−0.0003 (17)	0.0037 (15)
C22	0.037 (2)	0.033 (2)	0.0235 (18)	−0.0082 (17)	−0.0024 (15)	0.0005 (15)
C23	0.230 (11)	0.047 (3)	0.065 (4)	−0.054 (5)	−0.048 (5)	0.018 (3)
C24	0.200 (9)	0.037 (3)	0.052 (3)	−0.040 (4)	−0.038 (4)	0.019 (2)
C25	0.048 (3)	0.032 (2)	0.046 (2)	−0.0062 (19)	0.019 (2)	0.0083 (18)
C26	0.038 (3)	0.032 (2)	0.128 (6)	−0.008 (2)	−0.020 (3)	0.022 (3)
C27	0.046 (3)	0.040 (3)	0.106 (5)	−0.015 (2)	−0.014 (3)	0.022 (3)
C28	0.047 (3)	0.029 (2)	0.051 (3)	−0.0073 (19)	0.020 (2)	0.0051 (19)
C29	0.055 (3)	0.029 (2)	0.079 (4)	−0.008 (2)	0.006 (3)	−0.002 (2)
C30	0.057 (3)	0.037 (3)	0.081 (4)	−0.004 (2)	−0.001 (3)	−0.001 (3)
C31	0.079 (4)	0.030 (2)	0.039 (2)	−0.012 (2)	0.024 (2)	0.0041 (19)
C32	0.084 (4)	0.034 (2)	0.045 (3)	−0.019 (3)	0.023 (3)	0.001 (2)
C33	0.068 (4)	0.044 (3)	0.076 (4)	−0.003 (3)	−0.015 (3)	0.020 (3)
C34	0.072 (4)	0.047 (3)	0.069 (4)	−0.004 (3)	−0.024 (3)	0.023 (3)
C35	0.030 (2)	0.043 (2)	0.055 (3)	0.0002 (19)	0.008 (2)	0.021 (2)
C36	0.059 (3)	0.057 (3)	0.078 (4)	−0.022 (3)	−0.012 (3)	0.035 (3)
C37	0.064 (4)	0.070 (4)	0.076 (4)	−0.027 (3)	−0.020 (3)	0.040 (3)
C38	0.032 (2)	0.051 (3)	0.051 (3)	−0.002 (2)	0.012 (2)	0.015 (2)
C39	0.047 (3)	0.051 (3)	0.055 (3)	−0.011 (2)	0.003 (2)	0.016 (2)
C40	0.055 (3)	0.062 (3)	0.051 (3)	−0.014 (3)	0.006 (2)	0.011 (3)
C41	0.056 (3)	0.054 (3)	0.079 (4)	−0.006 (3)	0.003 (3)	0.001 (3)
C42	0.048 (3)	0.049 (3)	0.069 (3)	−0.004 (2)	−0.001 (3)	0.011 (3)
N1	0.0218 (14)	0.0258 (15)	0.0196 (14)	−0.0026 (12)	0.0008 (11)	0.0056 (11)
N2	0.0247 (15)	0.0226 (14)	0.0186 (14)	−0.0039 (12)	0.0003 (11)	0.0064 (11)
N3	0.079 (3)	0.048 (2)	0.064 (3)	−0.026 (2)	0.001 (2)	0.011 (2)
N4	0.058 (3)	0.041 (2)	0.056 (2)	−0.016 (2)	0.018 (2)	−0.0080 (19)
N5	0.048 (2)	0.060 (3)	0.056 (3)	−0.012 (2)	0.009 (2)	−0.002 (2)
N6	0.046 (2)	0.060 (3)	0.068 (3)	−0.012 (2)	−0.008 (2)	0.021 (2)
O1	0.0454 (18)	0.0292 (15)	0.065 (2)	−0.0161 (13)	0.0007 (15)	−0.0043 (14)
O2	0.0521 (18)	0.0367 (15)	0.0225 (13)	−0.0022 (14)	0.0023 (12)	0.0076 (11)
O3	0.0346 (15)	0.0275 (13)	0.0316 (14)	−0.0033 (11)	0.0011 (11)	0.0073 (11)
O4	0.062 (14)	0.038 (6)	0.026 (11)	−0.019 (8)	−0.006 (9)	0.001 (6)
O5	0.062 (6)	0.040 (5)	0.019 (4)	−0.007 (5)	0.008 (4)	−0.015 (3)
O6	0.132 (11)	0.024 (3)	0.030 (5)	−0.042 (4)	−0.027 (6)	0.002 (3)
O4B	0.064 (15)	0.030 (5)	0.018 (4)	−0.010 (8)	0.005 (8)	0.008 (3)
O5B	0.137 (15)	0.075 (12)	0.053 (8)	0.047 (9)	−0.051 (8)	−0.023 (7)
O6B	0.157 (16)	0.19 (2)	0.055 (10)	−0.145 (17)	0.013 (10)	−0.012 (12)
O7	0.0302 (14)	0.0332 (14)	0.0233 (12)	−0.0092 (12)	0.0041 (10)	0.0014 (10)
S1	0.0335 (5)	0.0205 (4)	0.0247 (4)	−0.0043 (4)	0.0020 (3)	0.0046 (3)
S2	0.0720 (8)	0.0315 (5)	0.0242 (5)	−0.0167 (5)	−0.0102 (5)	0.0031 (4)
Mn1	0.0220 (4)	0.0212 (4)	0.0170 (4)	−0.0023 (3)	−0.0010 (3)	0.0069 (3)
O8	0.0351 (16)	0.0542 (19)	0.0417 (16)	−0.0150 (14)	−0.0092 (12)	−0.0032 (14)
O9	0.063 (3)	0.053 (2)	0.100 (3)	−0.019 (2)	−0.006 (2)	−0.005 (2)
O10	0.119 (4)	0.072 (3)	0.070 (3)	−0.031 (3)	−0.048 (3)	0.027 (2)
O11	0.064 (3)	0.123 (5)	0.296 (9)	−0.061 (3)	0.081 (4)	−0.128 (6)
O12	0.197 (7)	0.125 (5)	0.095 (4)	−0.107 (5)	0.061 (4)	−0.061 (4)

Geometric parameters (Å, º)

C1—N1	1.376 (5)	C33—C34	1.374 (9)
C1—C10	1.397 (5)	C33—H33	0.95
C1—C2	1.437 (5)	C34—C35	1.377 (8)
C2—C3	1.350 (6)	C34—H34	0.95
C2—H2	0.95	C35—C36	1.397 (7)
C3—C4	1.436 (5)	C35—C38	1.462 (8)
C3—H3	0.95	C36—C37	1.374 (9)
C4—N1	1.380 (5)	C36—H36	0.95
C4—C5	1.390 (5)	C37—N6	1.328 (7)
C5—C6	1.392 (5)	C37—H37	0.95
C5—C11	1.500 (5)	C38—C39	1.393 (7)
C6—N2	1.383 (5)	C38—C42	1.405 (8)
C6—C7	1.427 (5)	C39—C40	1.357 (8)
C7—C8	1.345 (5)	C39—H39	0.95
C7—H7	0.95	C40—N5	1.346 (7)
C8—C9	1.437 (5)	C40—H40	0.95
C8—H8	0.95	C41—N5	1.338 (8)
C9—N2	1.381 (5)	C41—C42	1.361 (9)
C9—C10^i	1.393 (5)	C41—H41	0.95
C10—C9^i	1.393 (5)	C42—H42	0.95
C10—C17	1.494 (5)	N1—Mn1	2.012 (3)
C11—C12	1.365 (6)	N2—Mn1	2.011 (3)
C11—C16	1.377 (6)	N4—H1N	1.46 (9)
C12—C13	1.381 (7)	N5—H1N	1.28 (9)
C12—H12	0.95	O1—S1	1.446 (3)
C13—C14	1.377 (6)	O2—S1	1.451 (3)
C13—H13	0.95	O3—S1	1.458 (3)
C14—C15	1.356 (6)	O4—S2	1.42 (4)
C14—S1	1.776 (3)	O5—S2	1.465 (16)
C15—C16	1.384 (6)	O6—S2	1.425 (15)
C15—H15	0.95	O6—H6	0.851 (10)
C16—H16	0.95	O6—H6B	1.13 (5)
C17—C22	1.388 (5)	O4B—S2	1.46 (4)
C17—C18	1.391 (6)	O5B—S2	1.44 (2)
C18—C19	1.390 (6)	O6B—S2	1.41 (2)
C18—H18	0.95	O6B—H6	0.62 (4)
C19—C20	1.379 (6)	O6B—H6B	0.851 (10)
C19—H19	0.95	O7—Mn1	2.245 (3)
C20—C21	1.384 (6)	O7—H7A	0.8208
C20—S2	1.785 (4)	O7—H7B	0.8182
C21—C22	1.391 (5)	Mn1—N2^i	2.011 (3)
C21—H21	0.95	Mn1—N1^i	2.012 (3)
C22—H22	0.95	Mn1—O7^i	2.245 (3)
C23—N3	1.321 (8)	O8—H8A	0.8226
C23—C24	1.373 (11)	O8—H8B	0.8258
C23—H23	0.95	O9—H9A	0.8296
C24—C25	1.382 (8)	O9—H9B	0.8297
C24—H24	0.95	O10—H10A	0.8263
C25—C26	1.390 (8)	O10—H10B	0.8274
C25—C28	1.476 (8)	O11—H11A	0.8201
C26—C27	1.356 (9)	O11—H11B	0.8214
C26—H26	0.95	O12—H12B	0.8219
C27—N3	1.318 (7)	O12—H12A	0.8233
C27—H27	0.95	O13—H13B	0.8213
C28—C29	1.394 (7)	O13—H13A	0.8256
C28—C31	1.400 (6)	O13—H14A	1.2319
C29—C30	1.361 (8)	O13—H16B	1.441
C29—H29	0.95	O15—H15A	0.8237
C30—N4	1.345 (7)	O15—H15B	0.8244
C30—H30	0.95	O14—H13A	1.0074
C31—C32	1.350 (8)	O14—H14A	0.8156
C31—H31	0.95	O14—H14B	0.8249
C32—N4	1.342 (7)	O16—H13B	1.2329
C32—H32	0.95	O16—H16A	0.8327
C33—N6	1.342 (7)	O16—H16B	0.8243
N1—C1—C10	126.0 (3)	N6—C33—H33	118.3
N1—C1—C2	109.6 (3)	C34—C33—H33	118.3
C10—C1—C2	124.3 (3)	C33—C34—C35	120.7 (5)
C3—C2—C1	107.2 (3)	C33—C34—H34	119.6
C3—C2—H2	126.4	C35—C34—H34	119.6
C1—C2—H2	126.4	C34—C35—C36	115.5 (6)
C2—C3—C4	107.5 (3)	C34—C35—C38	122.2 (5)
C2—C3—H3	126.3	C36—C35—C38	122.3 (5)
C4—C3—H3	126.3	C37—C36—C35	120.6 (5)
N1—C4—C5	126.3 (3)	C37—C36—H36	119.7
N1—C4—C3	109.4 (3)	C35—C36—H36	119.7
C5—C4—C3	124.4 (3)	N6—C37—C36	123.4 (5)
C4—C5—C6	124.6 (3)	N6—C37—H37	118.3
C4—C5—C11	117.6 (3)	C36—C37—H37	118.3
C6—C5—C11	117.9 (3)	C39—C38—C42	115.8 (6)
N2—C6—C5	125.4 (3)	C39—C38—C35	121.5 (5)
N2—C6—C7	109.7 (3)	C42—C38—C35	122.7 (5)
C5—C6—C7	124.8 (3)	C40—C39—C38	120.9 (5)
C8—C7—C6	107.5 (3)	C40—C39—H39	119.5
C8—C7—H7	126.3	C38—C39—H39	119.5
C6—C7—H7	126.3	N5—C40—C39	121.9 (6)
C7—C8—C9	107.6 (3)	N5—C40—H40	119
C7—C8—H8	126.2	C39—C40—H40	119
C9—C8—H8	126.2	N5—C41—C42	121.8 (6)
N2—C9—C10^i	126.4 (3)	N5—C41—H41	119.1
N2—C9—C8	109.2 (3)	C42—C41—H41	119.1
C10^i—C9—C8	124.4 (3)	C41—C42—C38	120.8 (6)
C9^i—C10—C1	123.2 (3)	C41—C42—H42	119.6
C9^i—C10—C17	117.8 (3)	C38—C42—H42	119.6
C1—C10—C17	119.0 (3)	C1—N1—C4	106.3 (3)
C12—C11—C16	118.4 (4)	C1—N1—Mn1	127.4 (2)
C12—C11—C5	121.4 (4)	C4—N1—Mn1	126.3 (2)
C16—C11—C5	120.2 (4)	C9—N2—C6	106.0 (3)
C11—C12—C13	121.1 (4)	C9—N2—Mn1	127.1 (2)
C11—C12—H12	119.5	C6—N2—Mn1	126.9 (2)
C13—C12—H12	119.5	C27—N3—C23	117.8 (6)
C14—C13—C12	120.1 (4)	C32—N4—C30	117.6 (5)
C14—C13—H13	120	C32—N4—H1N	120 (4)
C12—C13—H13	120	C30—N4—H1N	122 (4)
C15—C14—C13	119.2 (4)	C41—N5—C40	118.8 (6)
C15—C14—S1	121.2 (3)	C41—N5—H1N	123 (4)
C13—C14—S1	119.6 (3)	C40—N5—H1N	118 (4)
C14—C15—C16	120.7 (4)	C37—N6—C33	116.3 (6)
C14—C15—H15	119.7	S2—O6—H6	115 (3)
C16—C15—H15	119.7	S2—O6—H6B	98 (2)
C11—C16—C15	120.5 (4)	S2—O6B—H6	142 (5)
C11—C16—H16	119.7	S2—O6B—H6B	116 (3)
C15—C16—H16	119.7	Mn1—O7—H7A	116.4
C22—C17—C18	119.2 (3)	Mn1—O7—H7B	109.2
C22—C17—C10	118.9 (3)	H7A—O7—H7B	110.6
C18—C17—C10	121.9 (3)	O1—S1—O2	111.6 (2)
C19—C18—C17	120.3 (4)	O1—S1—O3	113.68 (18)
C19—C18—H18	119.9	O2—S1—O3	112.49 (18)
C17—C18—H18	119.9	O1—S1—C14	106.19 (19)
C20—C19—C18	119.8 (4)	O2—S1—C14	106.04 (17)
C20—C19—H19	120.1	O3—S1—C14	106.21 (17)
C18—C19—H19	120.1	O6B—S2—O4	101.7 (19)
C19—C20—C21	120.8 (4)	O4—S2—O6	112.7 (18)
C19—C20—S2	119.6 (3)	O6B—S2—O5B	111.0 (14)
C21—C20—S2	119.6 (3)	O4—S2—O5B	123.1 (16)
C20—C21—C22	119.2 (4)	O6—S2—O5B	100.0 (12)
C20—C21—H21	120.4	O6B—S2—O4B	113 (2)
C22—C21—H21	120.4	O6—S2—O4B	122.5 (18)
C17—C22—C21	120.8 (4)	O5B—S2—O4B	109.6 (16)
C17—C22—H22	119.6	O6B—S2—O5	123.4 (12)
C21—C22—H22	119.6	O4—S2—O5	115.9 (15)
N3—C23—C24	122.1 (6)	O6—S2—O5	112.4 (10)
N3—C23—H23	118.9	O4B—S2—O5	102.0 (15)
C24—C23—H23	118.9	O6B—S2—C20	106.1 (8)
C23—C24—C25	121.6 (6)	O4—S2—C20	103.7 (16)
C23—C24—H24	119.2	O6—S2—C20	106.7 (6)
C25—C24—H24	119.2	O5B—S2—C20	109.8 (8)
C24—C25—C26	114.0 (6)	O4B—S2—C20	107.6 (16)
C24—C25—C28	122.3 (5)	O5—S2—C20	104.2 (6)
C26—C25—C28	123.6 (4)	N2—Mn1—N2^i	180.00 (9)
C27—C26—C25	121.6 (5)	N2—Mn1—N1	90.43 (12)
C27—C26—H26	119.2	N2^i—Mn1—N1	89.57 (12)
C25—C26—H26	119.2	N2—Mn1—N1^i	89.57 (12)
N3—C27—C26	122.8 (6)	N2^i—Mn1—N1^i	90.43 (12)
N3—C27—H27	118.6	N1—Mn1—N1^i	180.00 (18)
C26—C27—H27	118.6	N2—Mn1—O7	90.10 (11)
C29—C28—C31	116.3 (5)	N2^i—Mn1—O7	89.90 (11)
C29—C28—C25	121.4 (4)	N1—Mn1—O7	90.98 (11)
C31—C28—C25	122.4 (5)	N1^i—Mn1—O7	89.02 (11)
C30—C29—C28	120.1 (5)	N2—Mn1—O7^i	89.90 (11)
C30—C29—H29	120	N2^i—Mn1—O7^i	90.10 (11)
C28—C29—H29	120	N1—Mn1—O7^i	89.02 (11)
N4—C30—C29	122.8 (6)	N1^i—Mn1—O7^i	90.98 (11)
N4—C30—H30	118.6	O7—Mn1—O7^i	180
C29—C30—H30	118.6	H8A—O8—H8B	108.2
C32—C31—C28	120.5 (5)	H9A—O9—H9B	107.5
C32—C31—H31	119.7	H10A—O10—H10B	108.9
C28—C31—H31	119.7	H11A—O11—H11B	111
N4—C32—C31	122.7 (5)	H12B—O12—H12A	109.8
N4—C32—H32	118.6	H14A—O13—H16B	96.2
C31—C32—H32	118.6	H14A—O14—H14B	112.4
N6—C33—C34	123.5 (6)	H16A—O16—H16B	108.4
N1—C1—C2—C3	0.7 (5)	C38—C35—C36—C37	−177.7 (5)
C10—C1—C2—C3	−176.6 (4)	C35—C36—C37—N6	−2.4 (10)
C1—C2—C3—C4	0.1 (5)	C34—C35—C38—C39	−13.4 (7)
C2—C3—C4—N1	−0.9 (5)	C36—C35—C38—C39	164.7 (5)
C2—C3—C4—C5	178.5 (4)	C34—C35—C38—C42	166.6 (5)
N1—C4—C5—C6	−2.8 (6)	C36—C35—C38—C42	−15.3 (7)
C3—C4—C5—C6	177.9 (4)	C42—C38—C39—C40	0.6 (7)
N1—C4—C5—C11	178.1 (3)	C35—C38—C39—C40	−179.4 (4)
C3—C4—C5—C11	−1.2 (6)	C38—C39—C40—N5	0.3 (8)
C4—C5—C6—N2	−0.8 (6)	N5—C41—C42—C38	0.2 (9)
C11—C5—C6—N2	178.3 (3)	C39—C38—C42—C41	−0.9 (8)
C4—C5—C6—C7	−178.0 (4)	C35—C38—C42—C41	179.2 (5)
C11—C5—C6—C7	1.2 (6)	C10—C1—N1—C4	176.0 (4)
N2—C6—C7—C8	−1.6 (4)	C2—C1—N1—C4	−1.2 (4)
C5—C6—C7—C8	176.0 (4)	C10—C1—N1—Mn1	−5.8 (5)
C6—C7—C8—C9	1.0 (4)	C2—C1—N1—Mn1	176.9 (2)
C7—C8—C9—N2	−0.1 (4)	C5—C4—N1—C1	−178.1 (4)
C7—C8—C9—C10^i	178.0 (4)	C3—C4—N1—C1	1.3 (4)
N1—C1—C10—C9^i	7.6 (6)	C5—C4—N1—Mn1	3.8 (5)
C2—C1—C10—C9^i	−175.6 (4)	C3—C4—N1—Mn1	−176.9 (2)
N1—C1—C10—C17	−170.1 (3)	C10^i—C9—N2—C6	−178.9 (4)
C2—C1—C10—C17	6.7 (6)	C8—C9—N2—C6	−0.9 (4)
C4—C5—C11—C12	−89.9 (6)	C10^i—C9—N2—Mn1	2.0 (5)
C6—C5—C11—C12	90.9 (6)	C8—C9—N2—Mn1	−179.9 (2)
C4—C5—C11—C16	89.4 (5)	C5—C6—N2—C9	−176.0 (4)
C6—C5—C11—C16	−89.8 (5)	C7—C6—N2—C9	1.5 (4)
C16—C11—C12—C13	−1.3 (10)	C5—C6—N2—Mn1	3.0 (5)
C5—C11—C12—C13	178.0 (6)	C7—C6—N2—Mn1	−179.5 (2)
C11—C12—C13—C14	0.2 (12)	C26—C27—N3—C23	−0.9 (10)
C12—C13—C14—C15	1.3 (10)	C24—C23—N3—C27	−2.5 (13)
C12—C13—C14—S1	−177.4 (6)	C31—C32—N4—C30	0.1 (8)
C13—C14—C15—C16	−1.8 (8)	C29—C30—N4—C32	0.3 (8)
S1—C14—C15—C16	176.9 (4)	C42—C41—N5—C40	0.8 (8)
C12—C11—C16—C15	0.8 (8)	C39—C40—N5—C41	−1.0 (8)
C5—C11—C16—C15	−178.5 (4)	C36—C37—N6—C33	2.6 (9)
C14—C15—C16—C11	0.8 (8)	C34—C33—N6—C37	−1.1 (9)
C9^i—C10—C17—C22	−70.6 (5)	C15—C14—S1—O1	133.6 (4)
C1—C10—C17—C22	107.2 (4)	C13—C14—S1—O1	−47.7 (5)
C9^i—C10—C17—C18	108.6 (4)	C15—C14—S1—O2	−107.5 (4)
C1—C10—C17—C18	−73.6 (5)	C13—C14—S1—O2	71.1 (5)
C22—C17—C18—C19	0.2 (6)	C15—C14—S1—O3	12.3 (4)
C10—C17—C18—C19	−179.0 (4)	C13—C14—S1—O3	−169.0 (5)
C17—C18—C19—C20	0.2 (6)	C19—C20—S2—O6B	−39.3 (11)
C18—C19—C20—C21	−0.8 (6)	C21—C20—S2—O6B	144.3 (11)
C18—C19—C20—S2	−177.1 (3)	C19—C20—S2—O4	67.4 (16)
C19—C20—C21—C22	1.1 (6)	C21—C20—S2—O4	−109.0 (16)
S2—C20—C21—C22	177.5 (3)	C19—C20—S2—O6	−51.8 (8)
C18—C17—C22—C21	0.2 (6)	C21—C20—S2—O6	131.8 (8)
C10—C17—C22—C21	179.4 (4)	C19—C20—S2—O5B	−159.4 (11)
C20—C21—C22—C17	−0.8 (6)	C21—C20—S2—O5B	24.2 (11)
N3—C23—C24—C25	2.9 (15)	C19—C20—S2—O4B	81.3 (16)
C23—C24—C25—C26	0.0 (11)	C21—C20—S2—O4B	−95.1 (16)
C23—C24—C25—C28	176.8 (7)	C19—C20—S2—O5	−170.9 (7)
C24—C25—C26—C27	−3.3 (9)	C21—C20—S2—O5	12.7 (7)
C28—C25—C26—C27	180.0 (5)	C9—N2—Mn1—N1	177.1 (3)
C25—C26—C27—N3	3.9 (10)	C6—N2—Mn1—N1	−1.7 (3)
C24—C25—C28—C29	15.6 (8)	C9—N2—Mn1—N1^i	−2.9 (3)
C26—C25—C28—C29	−167.9 (5)	C6—N2—Mn1—N1^i	178.3 (3)
C24—C25—C28—C31	−164.1 (6)	C9—N2—Mn1—O7	−91.9 (3)
C26—C25—C28—C31	12.3 (7)	C6—N2—Mn1—O7	89.3 (3)
C31—C28—C29—C30	1.1 (7)	C9—N2—Mn1—O7^i	88.1 (3)
C25—C28—C29—C30	−178.7 (5)	C6—N2—Mn1—O7^i	−90.7 (3)
C28—C29—C30—N4	−0.9 (9)	C1—N1—Mn1—N2	−179.2 (3)
C29—C28—C31—C32	−0.7 (7)	C4—N1—Mn1—N2	−1.4 (3)
C25—C28—C31—C32	179.1 (4)	C1—N1—Mn1—N2^i	0.8 (3)
C28—C31—C32—N4	0.1 (8)	C4—N1—Mn1—N2^i	178.6 (3)
N6—C33—C34—C35	−0.6 (10)	C1—N1—Mn1—O7	90.7 (3)
C33—C34—C35—C36	0.9 (8)	C4—N1—Mn1—O7	−91.5 (3)
C33—C34—C35—C38	179.1 (5)	C1—N1—Mn1—O7^i	−89.3 (3)
C34—C35—C36—C37	0.5 (8)	C4—N1—Mn1—O7^i	88.5 (3)

Symmetry code: (i) −x+2, −y, −z+2.