Nickel(II) com­plexes based on L-amino-acid-derived ligands: synthesis, characterization and study of the role of the supramolecular structure in carbon dioxide capture

The formation of the symmetrical μ₃-carbonate-bridged self-assembled tri­nuclear Ni^II com­plex Na₂{[Ni(LO)₂(H₂O)]₃(μ₃-CO₃)} (LO is the carboxyl­ate anion of a L-tyrosine derivative), involves atmospheric CO₂ uptake. The asym­metric unit of the com­plex com­prises an octahedral coordination for the Ni^II with two L-tyrosine-based ligands, a water molecule and one O atom of the carbonate bridge. The Ni₃–μ₃-CO₃ core in this com­pound is the first reported of this kind according to the Cambridge Structural Database (CSD). The supramolecular structure is mainly sustained by hydrogen bonds developed by the phenolic functionality of the L-tyrosine moiety of one ligand and the carboxyl­ate group of a neighbouring ligand. The crystal packing is then characterized by three interpenetrated supramolecular helices associated with a diastereoisomer of the type R-^supP, which is essential for the assembly process. Magnetic susceptibility and magnetization data support weak ferromagnetic exchange interactions within the novel Ni₃–μ₃-CO₃ core. The Ni^II com­plex obtained under the same synthetic conditions but using the analogous ligand derived from the amino acid L-phenyl­alanine instead of L-tyrosine gives rise to to a mononuclear octahedral system. The results obtained for the different com­plexes demonstrate the role of the supramolecular structure regarding the CO₂ uptake property for these Ni^II–amino-acid-based systems.

1. Introduction

The fixation of CO₂ by nickel ions in basic solutions and the subsequent generation of a carbonate com­plex is relevant to the bioinorganic, environmental, structural and materials chemistry fields. The metallocarbonate system has been studied extensively due to its central role in the reversible hydration of CO₂ and the dehydration of bicarbonate pro­cesses catalyzed by carbonic anhydrase (Bertini et al., 1987; Christianson & Fierke, 1996). Thus, the syntheses of many metal–carbonate models have been performed with the idea of contributing to the elucidation of the mechanism related to the catalytic activity of this enzyme (Palmer & Van Eldik, 1983; Lipscomb & Sträter, 1996). There is also considerable interest in the fixation of CO₂ as metal com­plexes in order to explore possible ways of mitigating global warming by providing sustainable sources of C-products of higher value (Liu et al., 2015; Leitner, 1996; Belli et al., 2003; Zevenhoven et al., 2006; Dickie et al., 2008).

The carbonate ion is a very versatile ligand, since each O atom may act in a mono- or bidentate manner, and it is thus able to produce bi-, tri-, tetra-, penta- and hexanuclear coordination systems. This characteristic makes the carbonate ligand an interesting source for the design and synthesis of structurally rich com­plexes. The incorporation of more than one carbonate or in fact other bridging ligands can generate polynuclear com­plexes or even clusters which exhibit remark­able magnetostructural properties and, consequently, potential applications (Halmann, 2018). Considering that the three carbonate O atoms can participate in coordination to a metal centre and that each can exhibit a maximum of two bonds, seven different coordination modes are proposed. An additional mode, which is in fact a combination of two different modes, is observed in the Cambridge Structural Database [CSD; Groom et al., 2016; representative examples of structures discussed in the text are identified by their CSD refcodes and a list of the associated references is available as supporting information] (Fig. 1, Mode 7; Anderson et al., 2009) giving a total of eight modes (Fig. 1). Despite the versatility of the carbonate bridge, the majority of reported examples are Cu^II-based com­plexes (see supporting information). An updated descriptive statistical analysis using the hits obtained from the CSD showed that every coordination mode is not equally found, modes 1 and 3 being the most common having 24 and 19% of the total, respectively (Fig. 1 and supporting information). They are followed by mode 8 (bridging μ₃-carbonate) with 17% and mode 2 with 15% of the total. The remaining modes showed less than 10% of occurrence. For copper systems, the most frequently observed mode is mode 8 (supporting information).

Figure 1 The coordination modes for CO3^2− acting as a bridging ligand in Ni^II carbonate-bridged systems, adapted from Anderson et al. (2009) using updated CSD results.

Aromatic amino acids [tryptophan (Trp), tyrosine (Tyr) and phenyl­alanine (Phe)] are related to brain function. Several studies correlated the self-assembly of Phe with phenyl­ketonuria and β-amyloid-based neurodegenerative conditions, such as Alzheimer's and Huntington's diseases (Perween et al., 2013; Adler-Abramovich et al., 2012; Singh et al., 2014; Do et al., 2015) and atrial amyloidosis (Stefani, 2004). On the other hand, it is suggested also that the Tyr residue in amyloidogenic proteins acts as a key motif in the self-assembly process (Anjana et al., 2012). In this context, a number of peptide-based building blocks related to these amino acids have been designed and developed to create supramolecular structures (Zhou et al., 2017; Lee et al., 2018). Moreover, the amino acids Phe and Tyr have been shown to be capable of self-assembling into supramolecular nanostructures (Adler-Abramovich et al., 2012; Singh et al., 2014; Ménard-Moyon et al., 2015). Even though it is feasible for both Tyr and Phe-based fragments to be part of self-assembly processes, the structural difference associated with the phenolic OH group, as a side-chain functional group in Tyr, seems to be critical for the structural features and properties of the resultant entity. In this sense, the design and study of biomimetic analogous systems that help the understanding of the structural differences and properties among them could be considered important.

In our group, we investigated the synthesis of com­pounds derived from amino acids and explored their use as ligands for the construction of structurally diverse coordination com­plexes. In this work, we report the direct and reproducible synthesis of a trinuclear Ni^II–carbonate com­plex exhibiting a μ₃-CO₃²⁻ coordination mode (Fig. 1, Mode 8). It was obtained under mild conditions using an N,O-based chiral ligand derived from the amino acid L-tyrosine (L-Tyr) and the aldehyde piperonal. To the best of our knowledge, this com­pound is the first amino-acid-based com­plex exhibiting the capability to self-assembly through CO₂ fixation, a process that was not expected. Besides, the resultant symmetrical Ni₃–μ₃-CO₃ core would be the first reported of this kind. Although there are a few examples in the literature of Ni^II–carbonate com­plexes containing Schiff base ligands (Mukherjee et al., 2008; Schmitz et al., 2016), none of them are related to the amino-acid-based skeleton or the carbonate-bridging mode exhibited in our com­plex. Besides, overall, there is a limited number of examples of Ni^II com­plexes bearing L-Tyr or L-Tyr-based ligands (Groom et al., 2016). On the other hand, the Ni^II com­plex synthesized using an analogous ligand obtained from L-phenyl­alanine (L-Phe) instead of L-Tyr and the same aldehyde produced an octahedral mononuclear com­pound. This information suggests that the functionalities of the ligands could be responsible for their structural features and behaviour against CO₂ uptake which is, therefore, related to the supramolecular properties that sustain the self-assembly process. These easy-to-synthesize complexes could be con­sidered as a simple example to illustrate self-assembly properties along Tyr and Phe analog systems which have a close relationship with many cases in biological systems. Finally, the magnetic behaviour of the trinuclear Ni^II L-Tyr-based com­plex is also described as a com­plement of the study of the structural, supramolecular and physical properties.

2. Materials and methods

2.1. General considerations

UV–Vis spectra were recorded using a Hewlett–Packard 8453 diode array spectrometer in 10 mm optical path quartz cuvettes. Elemental analysis was carried out in a Carlo Erba CHNS EA-1108 microanalyzer using atropine as the standard. Mass spectra were recorded on a Xevo G2S Q-TOF (Waters Corporation) instrument using an electrospray ionization source and a quadrupole time-of-flight analyzer in methanol as the solvent. Differential scanning calorimetric (DSC) studies were performed in a Shimatzu DSC-50 with an aluminium pan under an N₂ atmosphere. Thermogravimetric analysis (TGA) was performed in a Shimatzu TGA-50 with an aluminium pan. NMR spectra were recorded using a Bruker AM500 equipped with a broadband probe. ¹H shifts are reported relative to protic solvent in these solvents. ¹³C shifts are reported relative to DMSO-d₆ (δ) 39.52 ppm. IR spectra were recorded using a Nicolet Avatar 320 FT–IR spectrometer with a Spectra Tech cell for KBr pellets. Light micrographs using polarized light microscopy (PLM) were taken with a Nikon SMZ-745 T stereoscopic trinocular micro­scope that includes a Nikon Ni-150 lighting system. Images were processed using the program Micrometrics SE Premium. Scanning electron microscopy (SEM) images were produced using a Carl Zeiss NTS SUPRA 40. All samples were purified and dried before being placed over carbon tape strands in aluminium pin stubs.

2.2. Materials and general procedures

All chemicals and solvents used for synthesis were obtained from commercial sources and were used as received, without further purification. Methanol was distilled before use. All reactions were carried out under aerobic conditions. The ligands 1–3 were synthesized following previously reported procedures for similar com­pounds with some modifications (Singh et al., 2017; Kumar et al., 2015).

2.3. Synthesis of the ligands

2.3.1. Ligand 1

To a solution of L-Tyr (110 mg, 0.6 mmol) and KOH (34 mg, 0.6 mmol) dissolved in distilled methanol (2.5 ml) were added piperonal (100 mg, 0.7 mmol) dissolved in of distilled methanol (0.5 ml). The resulting mixture was refluxed for 4 h. The yellow reaction mixture was then brought to room temperature and NaBH₄ (46 mg, 1.2 mmol) was added under stirring at the same temperature. The resulting solution was refluxed for 20 min until the colour disappeared. The pH of the solution was adjusted to 5 using hydro­chloric acid and was then stirred for 1 h. The obtained white precipitate (1) was filtered off, washed with water and methanol, and dried under vacuum (yield: 150 mg, 78.4%). EI/MS: m/z calculated for [M]⁺ 315, found 315, other signals: 270, 208, 135, 107, 77, 51. CHN elemental analysis for C₁₇H₁₇NO₅ calculated (%): C 64.8, N 4.4, H 5.4; found: C 64.3, N 4.4, H 5.7. NMR ¹H (500 MHz, DMSO-d₆): δ 7.04–6.95 (d, J = 8.4 Hz, 2H), 6.86–6.78 (m, 2H), 6.74–6.67 (d, J = 7.7 Hz, 1H), 6.69–6.59 (d, J = 8.4 Hz, 2H), 6.04–5.93 (s, 2H), 3.73–3.65 (d, J = 13.4 Hz, 1H), 3.61–3.50 (d, J = 13.4 Hz, 1H), 3.26–3.19 (t, J = 6.6 Hz, 1H), 3.20–3.10 (s, 1H), 2.88–2.80 (dd, J = 13.8, 6.2 Hz, 1H), 2.79–2.69 (m, 1H). ¹³C NMR (126 MHz, DMSO-d₆): δ 155.69, 130.23, 130.14, 121.52, 115.13, 114.80, 108.67, 107.84, 100.76, 62.32, 50.42, 37.40. Selected FT–IR peaks (KBr, cm⁻¹): ν(N—H) = 3180.2, ν(C=O) = 1577.56. Selected UV–Vis bands (in H₂O/NaOH solution, nm): 239, 287. Spectra are included in the supporting information.

2.3.2. Ligand 2

Ligand 2 was prepared following the procedure described for the preparation of ligand 1, except that 120 mg (0.7 mmol) of L-Phe were used instead of L-Tyr, and after the addition of NaBH₄, the reaction mixture was stirred at room temperature and not under reflux (yield: 175.5 mg 80.3%). EI/MS: m/z calculated for [M]⁺ 299.0, found 299; other signals: 208, 135. CHN elemental analysis for C₁₇H₁₇NO₄ calculated (%): C 68.2, N 4.7, H 5.7; found: C 67.4, N 4.7, H 5.7. NMR ¹H (500 MHz, DMSO-d₆): δ 7.27–7.18 (m, 12H), 6.80–6.78 (m, 2H), 6.69–6.67 (m, 1H), 5.97 (m, 2H), 3.70–3.67 (m), 3.58–3.53 (m), 3,27 (d), 2.97–2.93 (m), 2.85–2.82 (m). ¹³C NMR (126 MHz, DMSO-d₆): δ 190.52, 147.10, 129.32, 127.89, 125.92, 121.28, 108.53, 107.80, 100.71, 62.32, 50.60, 38.51. Selected FT–IR peaks (KBr, cm⁻¹): ν(O—H) = 3446.3, ν(C=O) = 1612.3. Selected UV–Vis bands (in H₂O/NaOH solution, nm): 233, 284. Spectra are included in the supporting information.

2.4. Synthesis of the com­plexes

2.4.1. Synthesis of com­plex 1Ni starting from Ni(NO₃)₂·6H₂O

In a 5 ml vial, Ni(NO₃)₂·6H₂O (4.6 mg, 0.016 mmol) was dissolved in methanol (1 ml) and a solution of the sodium salt of ligand 1 was then added. The solution of the ligand was prepared by dissolving 1 (10 mg, 0.032 mmol) and NaOH (2.5 mg, 0.063 mmol) in methanol (1 ml). The resulting turquoise solution of the com­plex was stirred overnight and, after removing the cap of the vial and covering with parafilm foil pierced with small holes, it was opened to the air and left for slow evaporation of the solvent at room temperature. After ca 3 d, green prismatic single crystals (1Ni-SC) were obtained and preserved in their mother liquors. After removing the crystals from the solution and leaving them under atmospheric conditions, the quality of the crystals was clearly affected; opacity and some fractures were observed. A polycrystalline material (1Ni-PM) was obtained after careful removal of the remaining supernatant and leaving it open to the air to dry. When the resultant crystalline material was dried under vacuum, a light-green amorphous powder was obtained (1Ni-A). The single crystals and the polycrystalline material were characterized by single-crystal and powder XRD, UV–Vis and FT–IR spectroscopy, and the magnetic behaviour was studied by EPR and SQUID. The solvent contents in 1Ni-SC and 1Ni-PM were quantified by thermal techniques and elemental analysis (yield: 95%). CHNS elemental analysis (%) for 1Ni-SC with the chemical formula Na₂{[Ni(LO)₂(H₂O)]₃CO₃}·11H₂O·16MeOH, C₁₁₉H₁₈₈N₆Na₂Ni₃O₆₃ (M_r = 2932.83, solvent confirmed by TGA) calculated (%): C 48.7, H 6.5, N 2.9; found: C 48.0, H 6.2, N 2.8. CHNS elemental analysis (%) for 1Ni-PM with the chemical formula Na₂{[Ni(LO)₂(H₂O)]₃CO₃}·6H₂O·3MeOH, C₁₀₆H₁₂₆N₆Na₂Ni₃O₄₅ (M_r = 2426.23) calculated (%): C 52.5, H 5.2, N 3.5; found: C 52.0, H 4.9, N 4.0. Selected FT–IR peaks (KBr, cm⁻¹): ν(O—H) = 3600–3300 (s), ν(N—H) = 3278 (s), ν(C=O) = 1608 (s), ν(C=C) = 1585 (s), ν(C—O) = 1515, 1444 (s), ν(C—H) = 1398, 1384 (s), ν(C—O) = 1243 (m), ν(C—O) = 931 (m). UV–Vis in methanol [λ_max, nm (ɛ, l mol⁻¹ cm⁻¹)]: 624 (21), 382 (54), 286 (33153), 227 (81045). Spectra and additional figures are included in the supporting information.

2.4.2. Synthesis of com­plex 1Ni starting from other Ni^II salts

The same procedure as described above was followed using other Ni^II salts: NiSO₄·6H₂O (4.2 mg, 0.016 mmol), NiCl₂·6H₂O (3.8 mg, 0.016 mmol) and NiAc₂·4H₂O (4.0 mg, 0.016 mmol). In all cases, the results were the same as described for the crystals obtained using Ni(NO₃)₂·6H₂O, in agreement with the discussion developed in the corresponding section. The single crystals and crystalline material obtained using these salts were studied by FT–IR and powder and single-crystal XRD (see supporting information).

2.4.3. Synthesis of com­plex 2Ni starting form Ni(NO₃)₂·6H₂O

In a 5 ml vial, Ni(NO₃)₂·6H₂O (4.9 mg, 0.017 mmol) was dissolved in methanol (1 ml) and a solution of the sodium salt of ligand 2 was then added. The solution of the ligand was prepared by dissolving 2 (10 mg, 0.033 mmol) and NaOH (2.7 mg, 0.066 mmol) in methanol (1 ml). The resulting turquoise solution was left unaltered in the same vial and, after covering with parafilm foil with small holes, it was opened to the air and left for slow evaporation of the solvent at room temperature. After ca 3 d, light-blue sphere-like crystalline aggregates (2Ni-CA) were obtained. The crystalline material was then washed with water and methanol, and air dried (yield: 40%). CHNS elemental analysis (%) for 2Ni-CA with the chemical formula [Ni(LO)₂(H₂O)₂]·2H₂O·0.5MeOH or C₃₄H₃₆N₂NiO₁₀·2H₂O·0.5MeOH (M_r = 743.4, solvent confirmed by TGA) calculated (%): C 55.7, H 5.7, N 3.8; found: C 55.3, H 5.0, N 3.8. Selected FT–IR peaks (KBr, cm⁻¹): ν(O—H) = 3608 (s), ν(N—H) = 3268 (s), ν(C=O) = 1602 (s), ν(C=C) = 1490 (s), ν(C—O) = 1438 (s), ν(C—H) = 1403 (s), ν(C—O) = 1247 (s), ν(C—O) = 925 (m). UV–Vis in DMSO [λ_max, nm (ɛ, l mol⁻¹ cm⁻¹)]: 622 (9), 379 (168). Details are included in the supporting information. Crystals of 2Ni suitable for single-crystal studies (2Ni-SC) were obtained after slow evaporation of the solvent from a dimethylformamide (DMF) concentrated solution of the com­plex.

2.4.4. Synthesis of com­plex 2Ni starting from other Ni^II salts

The same procedure as described above was followed using other Ni^II salts: NiSO₄·6H₂O (4.2 mg, 0.016 mmol), NiCl₂·6H₂O (3.8 mg, 0.016 mmol) and NiAc₂·4H₂O (4.0 mg, 0.016 mmol). In all cases, the results were the same as described above, i.e. light-blue sphere-like crystalline aggregates. The crystalline material obtained using the tested salts was studied by FT–IR, diffuse reflectance spectra, SEM and powder XRD (see supporting information).

2.5. Powder X-ray diffraction (PXRD) studies

Data were recorded on a PANalytical Empyrean diffrac­tometer equipped with a 4 kW Cu Kα sealed X-ray tube (generator power settings: 60 kV and 100 mA) and a PIXcel^3D area detector using parallel beam geometry (– mm slits, 15 mm incident mask). Samples were packed on a silicon monocrystal sample holder that was then placed on the sample holder attachment. For all PXRD experiments, data were collected over an angle range 5–90° with a scanning speed of 23 s per step and a 0.026° step.

2.6. Single-crystal X-ray diffraction (SC-XRD) measurements

For the single crystals of 1Ni-SC and 2Ni-SC, good-quality results were obtained at the W01A-MX2 beamline of the Synchrotron National Laboratory (LNLS, Campinas, Brazil) using wavelengths λ = 0.79983 Å and 0.82610 Å respectively. Data were collected using a PILATUS2M area detector (Dectris). The measurements were performed at 100 K and data reduction was done using MXCube software (Gabadinho, 2010). Using OLEX2 (Dolomanov et al., 2009), the structures were solved by intrinsic phasing employing ShelXT (Sheldrick, 2015a) and refined with the ShelXL (Sheldrick, 2015b) package, using least-squares minimization. Non-H atoms were refined anisotropically. H atoms were mostly included at geometrically calculated positions with displacement parameters derived from the parent atoms. H atoms attached to the coordinated water molecules or to groups suitable for forming hydrogen bonds were located on Fourier maps, and refined using isotropic displacement parameters depending on the parent atoms.

For 1Ni-SC, residual electron density associated with the solvent molecules was detected. Thus, data were treated with the SQUEEZE procedure (Spek, 2015) from PLATON (Spek, 2020). The volume occupied by the solvent was 3315 ų and the number of electrons per unit cell deduced by SQUEEZE was 854 electrons per void; these values could be interpretable as approximately 32 water and 29 methanol molecules. The presence of solvent in the structure is mostly in agreement with the results obtained by TGA/DSC measurements and experimental observations described previously. Additional crystallographic information and tables are included in the supporting information

For 2Ni-SC, one of the coordinated water molecules (O10) is disordered over two sets of sites with occupancies of 0.49 and 0.51, and atom O1 of one of the carboxyl­ate groups of the ligands shows positional disorder over two sites with occupancies of 0.48 and 0.52 (Table S9 in the supporting information). Disorder was not modelled for the solvent molecules and is represented through large displacement ellipsoids. Additional crystallographic information and tables are included in the supporting information.

Crystals of com­plex 1Ni obtained from different Ni^II salts were studied by single-crystal XRD using an Oxford Diffraction Gemini E lab diffractometer with Mo Kα (λ = 0.71073 Å) radiation. After unit-cell determination it was ob­served that in all cases the obtained parameters and suggested space groups were com­parable with those of 1Ni-SC synthesized using the nitrate salt. Due to the fact that experiments could be performed in ca 10 min, the crystal features and loss of crystallinity did not affect the data collection and quality (supporting information). Full data collection was performed at a synchrotron beamline as described.

2.7. Magnetic measurements

Magnetic measurements were performed with a Quantum Design MPMS XL-7 SQUID magnetometer. All experimental magnetic data were corrected for the diamagnetism of the sample holders and of the constituent atoms (M_r/2 × 10⁻⁶ cm³ mol⁻¹ formula). DC measurements were conducted from 4 to 300 K at 1 kOe and at 4 K in the range 1–70 kOe. Variable-temperature X-band CW–EPR measurements were performed on a Bruker EMX-Plus spectrometer equipped with a nitro­gen continuous-flow cryostat (room temperature to 100 K) and a rectangular cavity with 100 kHz field modulation. The X-band CW–EPR spectra of the oriented single crystals or polycrystalline sample of 1Ni were obtained as explained elsewhere (Schveigkardt et al., 2002).

2.8. Quantum chemical calculations

For com­putation of the exchange interaction J parameter, the ORCA (Neese, 2012) program package was employed. Single-point calculations for the high-spin (HS) and broken symmetry (BS) states at the X-ray geometry were carried out at the B3LYP level of density functional theory (DFT), employing the def2-TZVP Ahlrichs basis set for all atoms and taking advantage of the RI (Resolution of Identity) approximation. The SCF calculations were of the spin-polarized type and were tightly converged (10⁻⁷ Eh in energy, 10⁻⁶ in the density change and 10⁻⁶ in the maximum element of the DIIS error vector).

The methodology applied here relies on the BS formalism, originally developed by Noodleman for SCF methods (Noodleman, 1981), which involves a variational treatment within the restrictions of a single spin-unrestricted Slater determinant built upon using different orbitals for different spins. This approach was later applied within the framework of DFT (Baerends & Noodleman, 1984). The HS and BS energies were then combined to estimate the exchange coupling parameter J involved in the widely used Heisenberg–Dirac–van Vleck Hamiltonian. We used the method proposed by Ruiz and co-workers (Ruiz et al., 1999, 2003), where the following equation is applied [equation (1)]:

We have calculated the different spin topologies of BS nature by alternately flipping the spin on the different metal sites. The exchange coupling constants J_i are obtained after considering the individual pair-like com­ponent spin interactions involved in the description of the different BS states by solving a set of linear equations. We have also employed the BS-type spin unrestricted solutions after a corresponding orbital transformation (COT) as a means to visualize the interacting magnetic orbitals (Neese, 2004).

3. Result and discussion

3.1. Synthesis of new amino-acid-based Ni^II com­plexes

The molecules used as ligands (1–3) were obtained from the N-derivatization of L-α-amino acids through reaction with aldehydes and a subsequent reduction. Compounds 1 and 2 were synthesized using piperonal and the amino acids L-Tyr and L-Phe, respectively, and 3 used benzaldehyde and L-Tyr (Fig. 2 and supporting information). The reaction of the ligands with Ni^II salts gave coloured solids. Based on spectroscopic characterization and elemental analysis, the products were assigned as the corresponding coordination com­pounds but only after single-crystal X-ray diffraction (XRD) experiments were their structures unequivocally confirmed (Fig. 3). The results obtained for 1 and 2 are presented in detail, but those for 3 and the corresponding com­plex are included in order to study the effect of the piperonal moiety in the crystalline structure.

Figure 2 Ligands 1 and 2 obtained from the derivatization of the amino acids L-Tyr and L-Phe, respectively.

Figure 3 (Top) Com­plex 1Ni obtained from the reaction of Ni^II salts with ligand 1 (L-Tyr derivative). (Bottom) Com­plex 2Ni obtained when using ligand 2 (L-Phe derivative).

Different Ni^II salts (acetate, nitrate, chloride and sulfate) were used as the starting materials for the synthesis of the com­plexes and, in all cases, the results were the same. This observation was a preliminary indication that the anions of the starting Ni^II salts were not part of the resulting com­plex. After mixing the starting materials in a 2:1 ligand-to-metal ion ratio, the resultant clear solutions were left for ca 12 h under stirring for ligand 1 but unaltered for 2. After slow partial evaporation of the solvent, a coloured crystalline material was obtained in all cases. The reaction using ligand 1 gave large green–blue single crystals denoted 1Ni-SC. Single-crystal XRD results showed a structure com­prising a trinuclear Ni^II com­plex with a carbonate ion as the bridging central ligand of the three metal ions (Figs. 4 and 5). The carbonate anion in the system is derived from the CO₂ present in the atmosphere. Once it is spontaneously absorbed by the reaction mixture, it self-assembles into the trinuclear system. The com­plex is templated around the carbonate ligand, which is displayed in a triply bridging coordination mode, which is not the most frequent, as previously introduced.

Figure 4 The structure of com­plex 1Ni determined by single-crystal X-ray diffraction (1Ni-SC). For clarity, only the asymmetric unit and the com­plete carbonate ligand are shown as displacement ellipsoids at the 30% probability level; the rest of the com­plex is presented using grayscale wireframe. Sodium ions have been omitted.

Figure 5 View of the trinuclear com­plex along the crystallographic ab plane, showing the central carbonate ligand with displacement ellipsoids at the 50% probability level and the rest of the atoms using a stick model. Colour code: O atoms red, C gray, N violet–blue and Ni green. H atoms, Na ions and disorder have not been included for clarity.

The same synthetic procedure was followed using ligand 2, except for the stirring, but instead of large single crystals, light-blue sphere-like crystalline aggregates constructed from nano- and microcrystals (2Ni-CA) were obtained (Figs. S9 and S10). They proved to be unsuitable for single-crystal XRD studies but, based on the spectroscopic characterization, the structure of 2Ni-CA was suggested as an octahedral Ni^II com­plex bearing two deprotonated L-Phe-based ligands and two molecules of water. Crystalline material 2Ni-CA was soluble only in DMF and dimethyl sulfoxide (DMSO). Concentrated DMF solutions of 2Ni-CA were left in a vial open to the atmosphere and, after about two months, larger light-blue–green crystals were observed (denoted 2Ni-SC). Single-crystal XRD experiments confirmed the previously proposed structure (Figs. 3 and 6). Taking into account that the synthetic procedures were very similar, the solubility and the coordination sites provided by both ligands were also equivalent, a priori it was suggested that a possible source for the different structural features and the property regarding the CO₂ uptake could be directly associated with the structural variance provided by the phenol group of the Tyr residue and the supramolecular structure developed by each system.

Figure 6 The structure of com­plex 2Ni determined by single-crystal X-ray diffraction (2Ni-SC), viewed along the bc plane. Some H atoms, water and DMF solvent molecules, and disordered moieties have been omitted for clarity. Displacement ellipsoids are drawn at the 50% probability level. Colour code: O atoms red, C gray, N violet–blue and Ni green.

3.2. Molecular and supramolecular structure of the trinuclear nickel–carbonate system 1Ni

Even though crystals of the L-Tyr-based com­plex 1Ni-SC were of good quality, single-crystal XRD studies were difficult to perform because of the large amount of solvent included in the structure. Once the crystals had been removed from the mother liquor, they had to be frozen or embedded in oil to prevent solvent loss from the structure and thus alteration of the crystallinity. Suitable measurements were nevertheless obtained using a Synchrotron beamline and a temperature of 100 K. Complex 1Ni crystallized in the noncentrosymmetric space group R3 with a hexagonal unit cell [a = b = 20.020 (2), c = 31.13 (1) Å]. Three trinuclear com­plexes are observed per unit cell and there was a solvent-accessible volume of 3315 ų per unit cell, representing about the 30% of the total. The structure of the com­plex showed an unusual symmetry dictated by the μ₃-CO₃²⁻ ligand acting in a tridentate bridging mode with a C₃ rotational axis located at the C atom. Each Ni^II centre possess a distorted octahedral geometry; every metal ion bonds to an O atom of CO₃²⁻, two L-Tyr derivatives acting as bidentate chelating ligands binding through the carboxyl­ate group in a κO mode and the secondary amine group, and finally a water molecule to com­plete the sixth position (Fig. 5). The total charge of the com­plex was determined as −2 and thus two sodium ions, coming from the NaOH used to generate the salt of the ligands, were also present in the structure. The three Ni—O—C angles are about 132.7° and the Ni⋯Ni distances are 5.332 (1) Å. The arrangement adopted by the carbonate ligand and the three Ni centres is planar and shows an average O—C—O bond angle of 119.99°. Crystallographic tables and com­plementary analysis are included in the supporting information.

Although the CO₂ fixation and subsequent incorporation in the crystal structure as a carbonate ligand has been documented before for a few systems (Groom et al., 2016), none of them were constructed using amino acids or their derivatives as ligands. What is more, the examples which are most closely related to the present com­plex are scarce. The results obtained after searching the CSD (Groom et al., 2016) for polynuclear com­plexes having a CO₃²⁻ ligand acting in a central bridging manner bonded to at least a minimum of three Ni^II centres, showed just 11 hits. None contain amino-acid-based molecules as part of their structure and only one includes an N,O-type ligand, but the N atom is part of a pyridine ring (Graham et al., 2000). Only four hits correspond to com­plexes showing the μ₃-CO₃²⁻ arrangement of mode 8 as in 1Ni, but they show more than three metal ions, confirming what was stated regarding the peculiarity of this binding mode (Fig. 1 and supporting information). When analyzing the structures of these com­plexes in detail, it is observed that all of them bear the same TMEDA (N,N,N′,N′-tetra­methyl­ethane-1,2-di­amine) ligand as a multi-chelating moiety which indeed provides extra stability (Fig. S4; Anderson et al., 2009; Mustapha et al., 2008; Miyamoto et al., 2008). Another aspect of note is that 1Ni is the only metallocarbonate system showing a symmetrical triangular array of a total of three nickel ions bearing a central μ₃-CO₃²⁻ bridge. Only five structures match with the core Ni₃–carbonate, but with other coordination modes, making our com­plex the first of its kind (Fig. S5; Groom et al., 2016). Regarding this type of symmetrical arrangement, Zn₃ (Mohapatra et al., 2019) and Cu₃ (Mukherjee et al., 2008; Kolks et al., 1980; Escuer et al., 1996) systems seem to be the most frequent and, concordantly, their structural and magnetic properties (just for copper) have been extensively studied (supporting information). When focusing on the amino acids, according to the CSD (Groom et al., 2016), there are ten Ni^II com­plexes with unsubstituted L-Tyr and eight with L-Phe. In those com­pounds, the amino acids show typical chelating coordination via the N atom of the amino group and one of the O atoms of the carboxyl­ate group, as in 1Ni-SC and 2Ni-SC. Even though some of these com­plexes, as well as 2Ni, were obtained under a similar synthetic procedure to 1Ni, neither resulted in assemblies where CO₂ uptake and its incorporation as carbonate are developed. All these observations reinforce the fact that com­plex 1Ni is a rather unique system.

The presence of COO⁻, OH and NH groups and the central carbonate ligand results in several intra- and intermolecular hydrogen bonds and other short contacts. Fig. 7 shows in detail the intramolecular interactions developed among the ligands and between them and CO₃²⁻. It is interesting to point out that each O atom of the carbonate interacts with one of the H atoms of the water molecule coordinated to each Ni^II centre, giving a planar supramolecular moiety, which probably contributes to the stability of the trinuclear com­plex (Figs. 7b and 7c). When the other four related structures based on the TMEDA ligand were inspected, a closely similar supramolecular hydrogen-bonded network around the central bridging ligand was observed. Taking into account that the related systems exhibiting mainly μ₅- or μ₆-binding modes for the carbonate are more frequent, it could be suggested that the stabilization of the structures showing μ₃-CO₃²⁻ could be associated mainly with the presence of other short contacts, such as the mentioned hydrogen bonds, related to the central bridging ligand (Fig. S6).

Figure 7 The main intramolecular contacts in 1Ni-SC. For clarity, one asymmetric unit is displayed in displacement ellipsoid style and the rest in wireframe in gray. H atoms not involved in short contacts have been omitted. (a) π–π interactions and CH⋯O and OH⋯O contacts. (b, c) Hydrogen-bond sets between the coordinated water molecules and the carbonate bridging ligand along the ab and bc crystallographic planes, respectively.

Several intermolecular interactions involving the ligands take place in 1Ni-SC (Figs. 8 and S28). Among all the interactions, the hydrogen bonds developed by the phenolic functionality of the L-Tyr group of one ligand and the carboxyl­ate group of a neighbouring ligand could be considered as the driving force of the crystal packing [H⋯O distances are 1.819 and 1.857 Å]. The supramolecular structure of amino acids and their derivatives is usually described by the intermolecular electrostatic contacts defined by the amino/ammonium and carb­oxyl/carboxyl­ate groups, apart from other noncovalent interactions, such as hydrogen bonding, π–π stacking and hydro­phobic interactions (Bera et al., 2018). The reason why the amino group in 1Ni-SC does not participate in the supramolecular structure could be associated with the fact that the N-atom lone electron pair is mainly com­promised in the coordination bond and also steric effects. Looking closely to the structure of 1Ni-SC, it is observed that each of the two ligands of each nickel centre of the trimeric unit shows a different hydrogen-bond network. In one of them, the carboxyl­ate and phenol OH groups interact with the corresponding hydrogen-bond donor and acceptor, respectively, of two molecules coordinated to two different nickel ions, but belonging to the same trinuclear com­plex. The other ligand of the same centre interacts with two molecules belonging to two different trinuclear units (Figs. 8 and 9). These networks give rise to the development of three different graph sets, but all of them are of the type R₂²(27) (Fig. S30).

Figure 8 The OH⋯O contacts in 1Ni-SC com­prising the carboxyl­ate and OH groups of L-Tyr. For clarity, only these moieties are depicted as coloured balls and ticks; the rest of the system is depicted in wireframe style in gray.

Figure 9 (Top) A trimeric unit with the hanging contacts (red dashed lines) starting from the O—H and COO^− hydrogen-bond donors and acceptors. (Bottom) The extended supramolecular structure, showing three trinuclear com­plexes above and below the starting unit; the expanded contacts are indicated in gray and the hanging contact is in red. In the images, the counter-ions, the piperonal moieties and H atoms which do not participate in the short contacts have been omitted for clarity.

Considering the trinuclear com­plex as a unit, an analysis shows that each of them interacts with another six units through the mentioned O—H⋯O contacts, three above and the other three below. The ligands of these six units then interact with others through the pendent hydrogen-bond acceptors and donors, resulting in three interpenetrated supramolecular helices. Another strategy to visualize the helices is through different chains that can be rationalized using the Mercury `Graph Set' function (Macrae et al., 2020). The voids containing the solvent molecules are located between these extended arrangements (Figs. 9 and 10).

Figure 10 The extended C2^2(34) infinite chain in 1Ni-SC. (Top) The chain (in balls) along the crystallographic bc plane. (Bottom) View allowing visualization of the anticlockwise helix (space-filled pink spheres). In the images, the counter-ions, the piperonal moieties and H atoms which do not participate of the short contacts have been omitted for clarity.

On the other hand, since enantiomerically pure L-amino acids were used for the synthesis, ligands 1 and 2 retained their S configuration. Besides, considering each asymmetric unit of the com­plex as a cis-bis­(bidentate) octahedral system and the right-handed helix described by the ligands, they could be assigned with the Δ absolute configuration (Connelly et al., 2005). The chiral nature of the trinuclear com­plex was then also expressed in the chiral space group R3 and with a Flack parameter of 0.035 (3). During the process of self-assembly in the crystalline solids of chiral molecules, supramolecular diastereoisomers can be generated. In the case of supramolecular helices, the resulting assembly can be described as R or S, depending on the configuration provided by the chiral molecule, followed by the term ^supP if the produced helix is developed counterclockwise or ^supM if it is clockwise (Tsang et al., 2015; Sasaki et al., 2013). Since com­plex 1Ni-SC presents an R configuration (or Δ) and each helix has an anticlockwise orientation, the supramolecular structure can be clearly associated with a diastereoisomer of the type R-^supP (Fig. 10).

3.3. Comparative structural analysis of the amino-acid-based Ni^II com­plexes

The L-Phe-based com­plex 2Ni was studied with the aim of exploring the role of the amino acid in CO₂ uptake. Although differences in the supramolecular behaviour of L-Phe and L-Tyr are widely discussed in the literature (Adler-Abramovich et al., 2012; Ménard-Moyon et al., 2015), their implication in biologically relevant systems is a subject still under investigation (Yamagata et al., 1998). As previously discussed, the central role of the phenol OH group in the molecular and supramolecular structure is clear. This functionality is provided by the L-Tyr moiety and is absent in the L-Phe derivative. Nevertheless, before the synthesis it was not clear how the behaviour of this system was going to affect CO₂ uptake, since the synthetic procedure (identical for both com­plexes) using basic media could have been the driving force for carbonate inclusion in the structure, and not necessarily the supramolecular properties and the self-assembly. Then, as mentioned above, experimental data confirmed a mononuclear octahedral structure for the L-Phe derivative com­plex. This com­pound crystallized as a DMF and water solvate in the noncentrosymmetric space group P2₁2₁2, with an orthorhombic unit cell [a = 14.910 (3), b = 33.6200 (9), c = 7.370 (2) Å, see supporting information]. Solvent molecules are related to the com­plex through different hydrogen-bond motifs (Fig. S32). Each ligand is coordinated through the N atom of the amino group and one carboxyl­ate O atom, and they are not crystallographically equivalent. Two water molecules com­plete the coordination sphere in the equatorial positions, giving place to a cis-bis­(bidentate) octahedral system, as was observed for each asymmetric unit of the trinuclear com­plex 1Ni-SC. The bond lengths and angles exhibited in 2Ni-SC are similar to those observed for 1Ni-SC. The configuration for 2Ni corresponds also to the Δ enantiomer (Figs. 4 and 6). These results support the fact that the coordination environment for the metal centre can be considered as equivalent for both com­plexes and the main difference between them lies in the self-assembly process adopted in the L-Tyr com­plex due to carbon dioxide uptake. All of these originate from the presence of the phenolic OH group. The supramolecular structure observed for com­plex 2Ni is com­pletely influenced by the presence of the solvent molecules because they act as excellent hydrogen-bond donors and acceptors. The crystal packing of 2Ni describes a three-dimensional (3D) hydrogen-bond network sustained by interactions involving the ordinated water molecules, the amino and carboxyl­ate functionalities of the ligands, and the solvent molecules (Fig. 11 and Fig. S33 in the supporting information).

Figure 11 The hydrogen-bond 3D network in 2Ni-SC.

Finally, and in order to cover all the possible key factors acting in the supramolecular structure of trinuclear com­plex 1Ni, the same reaction was performed but using ligand 3, the L-Tyr-benzaldehyde analog (supporting information). The idea was to test the apparently negligible effect of the piperonal ring versus a phenyl group. The synthesis of com­plex 3Ni gave rise to light-blue crystalline aggregates with low solubility and very similar features to what was observed for 2Ni (Fig. S11). So far, single crystals of this com­plex have not been obtained and as a preliminary conclusion it could be inferred that the behaviour of this ligand is not equivalent to 1. The piperonal moiety is involved in the supramolecular structure of com­plexes, not only in the structure of 1Ni, but also in that of 2Ni. Unfortunately, we cannot confirm our suggestion since we do not have single-crystal X-ray diffraction data for com­plex 3Ni. Finally, it should be mentioned that even though the supramolecular architecture described in 1Ni is mainly sustained by O—H⋯O=C hydrogen bonds involving the phenol and carboxyl­ate groups (Fig. 8), the O—CH₂—O group of piperonal, through the development of C–H⋯O contacts, also influences the stability of the crystal packing.

3.4. Magnetic properties

The magnetic behaviour of the trinuclear Ni^II system was investigated by performing direct current (DC) magnetic susceptibility measurements on ground single crystals of com­pound 1Ni, as well as magnetization measurements at 4 K (Fig. 12). The χT value of 3.107 cm³ K mol⁻¹ at room temperature is in good agreement with the expected value for three independent Ni^II ions of 3.00 cm³ K mol⁻¹ (three S = 1 with g = 2; Kahn, 1993). With decreasing temperature, χT remains almost constant until 50 K, when it increases sharply to a maximum value of 6.798 cm³ K mol⁻¹ at ca 7 K; it finally drops to a value of 5.388 cm³ K mol⁻¹ at 4 K. This behaviour suggests the presence of ferromagnetic exchange interactions within the Ni^II ions. In fact, the expected value for χT for an isolated S = 3 ground state (arising from com­plete spin alignments due to ferromagnetic dominant exchange) is 6.00 cm³ K mol⁻¹ (g = 2; Mohapatra et al., 2019) and is in agreement with the observed value at the maximum of χT versus T data profile. The χT lowering below 7 K is most probably attributable to the onset of zero-field splitting (ZFS) arising from local Ni^II ions. A magnetization plot at 4 K does not show saturation in agreement with ZFS onset, reaching a maximum value of 4.20 Nμ_B at 7 T.

Figure 12 The pattern of exchange interactions in the 1Ni com­plex.

In order to further understand the magnetic behaviour, we performed different experimental data fitting with the PHI package (Chilton et al., 2013). The arrangement of the Ni^II ions with the unique carbonate bridge corresponds to a regular triangular array of sites with S = 1 (Fig. 12) due to the crystallographically imposed C₃ axis.

This arrangement implies a unique exchange interaction parameter J for the three possible Ni^II⋯Ni^II exchange pathways. Equation (2) shows the spin Hamiltonian suitable for this spin topology, including an axial local ZFS term on each equivalent Ni^II site.

If the data fitting is performed employing only magnetic susceptibility data (χT versus T plot), a satisfactory result can be obtained (Fig. 13) only if an intermolecular mean-field corrections is added (if not, low-temperature data cannot be reproduced), as shown in equation (3).

Figure 13 χT versus T plot at 1 kOe external magnetic field of com­pound 1Ni. Open symbols represent the experimental data, full lines represent the simulated data with best-fit parameters, black represents χT data only and blue represents simultaneous χT and M data.

The low-temperature data fitting is impossible when removing either the ZFS or the zJ′ terms. The best fitting parameters obtained are: g = 2.0, J = 1.3 cm⁻¹, D = 18.5 cm⁻¹ and zJ′ = 0.46 cm⁻¹. On the other hand, if a simultaneous fitting of magnetic susceptibility and magnetization data is performed, a poorer agreement is obtained for the χT versus T profile, than if only magnetic susceptibility is considered for data fitting (Fig. 13), it gives rise to a model which is in very very good agreement with the magnetization plot (Fig. 13). It does not happen if one tries to model both magnetic susceptibility and magnetization data simultaneously. Nevertheless, the best fitting parameters obtained within this approach are com­parable to those previously shown for the magnetic susceptibility only data fitting: g = 2.0; J = 2.6 cm⁻¹; D = 16.1 cm⁻¹ and zJ′= 0.18 cm⁻¹.

Finally, even with a ZFS contribution larger than the exchange interaction, it is possible to fit the magnetization data at 4 K with an S = 3 isolated spin model having an axial ZFS contribution [Equation (4))], in agreement with the observed ferromagnetic J parameter for the carbonate-mediated Ni^II⋯Ni^II exchange pathways (Fig. 14). The best fitting parameters in this case are: g = 2.17 and D = 13.9 cm⁻¹. The simulated plot is indistinguishable from that obtained with Hamiltonian equation (2).

Figure 14 M versus H plot at 4 K of com­pound 1Ni. Open symbols represent experimental data, full lines represent simulated data with best-fit parameters, black represents Hamiltonian equation (2) (three coupled S = 1) and blue represents Hamiltonian equation (4) (isolated S = 3).

Regarding the parameters obtained from this DC magnetic data analysis, the dominant ZFS contribution is explained in terms of the small values of the exchange interaction, as the obtained values for the D parameter are frequently observed in Ni^II systems (Atanasov et al., 2012). No reasonable agreement with the experimental data is reached if a negative D parameter is employed.

With respect to the found small ferromagnetic exchange coupling parameter, a precise determination of its value is not trivial for com­plex 1Ni, as from magnetic susceptibility data fitting a strong correlation with the zJ′ parameter is observed (supporting information). Considering that for the mean-field approximation validity J > zJ′, a range of 1.0–2.5 cm⁻¹ can be established for the J parameter with possible values of zJ′ in the range 0.2–0.5 cm⁻¹. The existence of quite a strong intermolecular exchange interaction can be rationalized in terms of the one-dimensional nature of the supramolecular arrangement of the Ni₃ com­plexes (cf. structural discussion).

Finally, the observed small ferromagnetic Ni^II⋯Ni^II exchange interaction through the carbonate bridge deserves special consideration. To the best of our knowledge, there are no structurally characterized reported examples of trinuclear Ni^II com­plexes where a carbonate bridge adopts the μ₃–syn–anti mode (Fig. 1, Mode 8) as the unique bridging ligand. In this sense, this becomes the first report of magnetic behaviour of this type for an Ni₃-μ₃-CO₃ system. From studies of related Cu com­plexes, it has been stated that this bridging mode favours ferromagnetic Cu^II⋯Cu^II interactions (Dussart et al., 2002; Newell et al., 2005; Martínez-Prieto et al., 2013; Yoo & Lee, 2016; Jonasson et al., 2018; Gerwien et al., 2018). However, the existence of an additional magnetic orbital in Ni^II ions, together with low-symmetry environments, makes a direct com­parison unreliable. A few dinuclear Ni^II com­plexes bearing a carbonate bridge as the unique exchange pathway mediator can be found in the CSD (Groom et al., 2016). All of them show a μ₂–anti–anti bridging mode, and magnetic studies have been performed for only one of them, highlighting a small antiferromagnetic interaction (Dussart et al., 2002). On the other hand, there are also a few reported Ni^II trinuclear com­plexes where the carbonate bridge appears as the unique ligand mediating the exchange interaction (Fig. S6; Dussart et al., 2002; Newell et al., 2005; Martínez-Prieto et al., 2013; Yoo & Lee, 2016; Jonasson et al., 2018; Gerwien et al., 2018). In these systems, it is observed that the Ni^II⋯Ni^II exchange interaction is through the Ni—O—C—O—Ni pathway and it is at most 10 cm⁻¹, clearly weaker than the interactions through the Ni—O_carbonate—Ni pathways (e.g. modes 2 to 7 in Fig. 1). These results agree with those observed for com­plex 1Ni. However, predicting the nature (ferro or antiferro) of the exchange interaction in Ni^II carbonate-bridged com­plexes is still not obvious and no magneto-structural correlations are available.

As 1Ni has an unprecedented carbonate-bridging mode for trinuclear com­plexes, we performed broken symmetry (BS) DFT calculations on the X-ray geometry in order to support the experimental magnetic data. Due to the C₃ axis, the unique BS state (Fig. S35) was sufficient to obtain a quantum com­puted value for the Ni^II⋯Ni^II exchange interaction. The obtained value for J of 0.19 cm⁻¹ is in good agreement with the experimental results and supports the ferromagnetic nature of the exchange interaction. When looking at the magnetic orbitals, it can be observed that overlaps are negligible (Fig. S36), thus explaining the observed ferromagnetic exchange. More examples like this Ni₃ carbonate-bridged com­plex are needed to further understand its role in the exchange interaction propagation.

Crystalline samples of 1Ni were also investigated using X-band EPR spectroscopy. Neither the polycrystalline samples (1Ni-PM) nor the single-crystal samples of the trinuclear com­plex (1Ni-SC) measured in the range from ambient temperature to 100 K gave EPR signals. Hence, we suggest that com­plex 1Ni is EPR silent under these experimental conditions, supporting the sizeable Ni^II local ZFS inferred from the magnetic susceptibility and magnetization data (Krzystek et al., 2002).

4. Conclusions

We observed that CO₂ air fixation is a critical tool in the assembly of an L-Tyr-based trinuclear Ni^II com­plex. The higher nuclearity of this com­pound in com­parison with the mononuclear L-Phe analog is provided, on one hand, by the multiple bridging mode of the incorporated carbonate ligand and, on the other hand, by the identity of the amino acid. It is suggested also that the self-assembly process and the resulting structure is governed by the nature of the amino acid derivative through the supramolecular structure that is developed. The L-Tyr system showed a helix-like supramolecular assembly mainly sustained by hydrogen bonds provided by the carboxyl­ate and phenol functionalities. For the L-Phe analog com­plex, where such interactions cannot take place due to the absence of the phenol group, the supramolecular structure and thus the self-assembly followed a com­pletely different scheme. On the other hand, for the stabilization of the packing, the moiety provided by the aldehyde used for the synthesis of the ligands seems to also be important; in the absence of the piperonal skeleton in the com­plex obtained using a L-Tyr ligand derivatized with benzaldehyde, the results were com­pletely different. As seen in biological systems, strong differences in the supramolecular hierarchy are generated for different amino acids. In this sense, this work could be considered as a simple example to demonstrate the direct effect of the amino acid identity on a specific property and also how such behaviour is highly influential on the self-assembly and supramolecular structure.

Finally, the metallocarbonate core exhibited in the trinuclear Ni^II com­plex proved to be rather unique since, to our knowledge, there is no information in the literature of other Ni^II systems showing this exact structure. Magnetic susceptibility and magnetization data support weak ferromagnetic exchange interactions within this novel symmetrical Ni₃–μ₃-CO₃ core.

Supporting information

Computing details

For both structures, data collection: MXCuBE (Gabadinho et al., 2010); cell refinement: XDS (Kabsch, 2010); data reduction: XDS (Kabsch, 2010). Program(s) used to solve structure: olex2.solve (Bourhis et al., 2015) for 2ni-sc. Program(s) used to refine structure: SHELXL 2018/3 (Sheldrick, 2015) for 1ni-sc; SHELXL (Sheldrick, 2015) for 2ni-sc. Molecular graphics: Olex2 1.3 (Dolomanov et al., 2009) for 1ni-sc; Olex2 (Dolomanov et al., 2009) for 2ni-sc. Software used to prepare material for publication: Olex2 1.3 (Dolomanov et al., 2009) for 1ni-sc; Olex2 (Dolomanov et al., 2009) for 2ni-sc.

1Ni (1ni-sc)

Crystal data

1(C103H102N6Na2Ni3O36)	Dx = 1.024 Mg m^−3
Mr = 2222.03	Synchrotron radiation, λ = 0.79983 Å
Trigonal, R3	Cell parameters from 11715 reflections
a = 20.020 (2) Å	θ = 1.5–31.5°
c = 31.13 (1) Å	µ = 0.63 mm^−1
V = 10805 (4) Å^3	T = 100 K
Z = 3.0	Cube, green
F(000) = 3468	1 × 0.5 × 0.5 mm

Data collection

MX2_definiton_file diffractometer	9751 independent reflections
Radiation source: LNLS-MX2	9583 reflections with I > 2σ(I)
Detector resolution: 5.814 pixels mm^-1	Rint = 0.030
shuterless φ scans	θmax = 30.0°, θmin = 2.0°
Absorption correction: empirical (using intensity measurements) XDS (Kabsch, 2010)	h = −25→25
Tmin = 0.757, Tmax = 1	k = −25→25
287004 measured reflections	l = −38→38

Refinement

Refinement on F^2	Hydrogen site location: mixed
Least-squares matrix: full	H-atom parameters constrained
R[F^2 > 2σ(F^2)] = 0.071	w = 1/[σ^2(Fo^2) + (0.1601P)^2 + 1.9907P] where P = (Fo^2 + 2Fc^2)/3
wR(F^2) = 0.186	(Δ/σ)max < 0.001
S = 1.11	Δρmax = 1.04 e Å^−3
9751 reflections	Δρmin = −0.47 e Å^−3
466 parameters	Absolute structure: Flack x determined using 4664 quotients [(I+)-(I-)]/[(I+)+(I-)] (Parsons, Flack and Wagner, Acta Cryst. B69 (2013) 249-259).
1 restraint	Absolute structure parameter: 0.035 (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.

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

	x	y	z	Uiso*/Ueq	Occ. (<1)
Ni1	0.50440 (3)	0.71093 (3)	0.50952 (2)	0.04490 (19)
O2	0.40545 (17)	0.71814 (17)	0.51016 (10)	0.0466 (6)
O21	0.7064 (2)	0.7459 (3)	0.45864 (13)	0.0670 (10)
O3	0.5558 (2)	0.8206 (2)	0.53391 (12)	0.0556 (7)
O4	0.44046 (19)	0.59588 (18)	0.49020 (11)	0.0526 (7)
H4A	0.442912	0.590248	0.461770	0.079*
H4B	0.390209	0.578459	0.494629	0.079*
O5	0.60346 (19)	0.7056 (2)	0.49986 (12)	0.0553 (7)
O78	0.5916 (4)	1.0578 (3)	0.34582 (18)	0.0901 (15)
H78	0.607999	1.060818	0.320638	0.135*
N6	0.5321 (2)	0.7567 (2)	0.44671 (12)	0.0476 (7)
H6	0.525504	0.803065	0.445573	0.057*
O58	0.4836 (3)	0.8734 (3)	0.29758 (14)	0.0749 (11)
O81	0.6788 (4)	0.5647 (4)	0.7148 (2)	0.0966 (17)
H81	0.696630	0.587331	0.738261	0.145*
O20	0.5975 (3)	0.8876 (3)	0.59493 (16)	0.0782 (11)
N7	0.5084 (2)	0.6831 (3)	0.57435 (13)	0.0551 (9)
H7	0.530964	0.648419	0.574822	0.066*
O72	0.5156 (5)	0.8118 (4)	0.24570 (15)	0.0933 (17)
C15	0.6141 (3)	0.7826 (3)	0.43856 (14)	0.0537 (10)
H15	0.620814	0.772504	0.407847	0.064*
C13	0.6424 (3)	0.7401 (3)	0.46782 (15)	0.0536 (9)
O76	0.4658 (8)	0.5344 (12)	0.7566 (4)	0.196 (7)
O67	0.4773 (5)	0.4736 (5)	0.6965 (3)	0.131 (3)
C42	0.6692 (4)	0.9343 (4)	0.3779 (2)	0.0705 (14)
H42	0.698571	0.913495	0.366021	0.085*
C35	0.4778 (3)	0.7909 (3)	0.35861 (16)	0.0548 (10)
H35	0.462905	0.815357	0.379599	0.066*
C86	0.6461 (3)	0.9198 (3)	0.42077 (18)	0.0618 (12)
C24	0.4902 (3)	0.7292 (3)	0.36886 (15)	0.0520 (9)
C16	0.4784 (3)	0.6998 (3)	0.41477 (14)	0.0510 (9)
H16A	0.424839	0.683307	0.423490	0.061*
H16B	0.484600	0.653724	0.415585	0.061*
C73	0.6647 (5)	0.6117 (4)	0.6879 (2)	0.0813 (17)
C18	0.5637 (3)	0.7548 (4)	0.59717 (17)	0.0622 (12)
H18	0.544204	0.753635	0.626885	0.075*
C23	0.6657 (3)	0.8700 (3)	0.44821 (17)	0.0616 (12)
H23A	0.659888	0.879375	0.478840	0.074*
H23B	0.720353	0.885037	0.443386	0.074*
C43	0.6037 (4)	0.9517 (3)	0.43722 (18)	0.0694 (13)
H43	0.587175	0.942348	0.466307	0.083*
C37	0.6485 (3)	0.7081 (4)	0.6301 (2)	0.0724 (15)
C68	0.6096 (4)	1.0116 (4)	0.3698 (2)	0.0771 (16)
C45	0.4888 (4)	0.8132 (3)	0.31607 (16)	0.0608 (11)
C8	0.333333	0.666667	0.50971 (19)	0.0394 (12)
C9	0.5713 (3)	0.8256 (3)	0.57333 (17)	0.0603 (11)
C28	0.6449 (3)	0.7644 (4)	0.6000 (2)	0.0711 (15)
H28A	0.660977	0.757626	0.570998	0.085*
H28B	0.681812	0.817476	0.609678	0.085*
C65	0.4515 (6)	0.5639 (10)	0.7203 (3)	0.125 (5)
C48	0.6401 (4)	0.7146 (4)	0.6749 (2)	0.0716 (14)
H48	0.627485	0.751440	0.685469	0.086*
C29	0.4373 (3)	0.6127 (4)	0.64046 (19)	0.0681 (14)
C57	0.5853 (5)	0.9966 (4)	0.4120 (2)	0.0784 (16)
H57	0.555668	1.017330	0.423617	0.094*
C55	0.6497 (4)	0.9787 (4)	0.3525 (2)	0.0788 (17)
H55	0.664004	0.986511	0.323020	0.095*
C40	0.4313 (5)	0.6465 (6)	0.6776 (2)	0.090 (2)
H40	0.423651	0.689469	0.674834	0.108*
C41	0.4531 (4)	0.5507 (4)	0.6441 (2)	0.0812 (19)
H41	0.458707	0.525207	0.619811	0.097*
C62	0.6502 (4)	0.6673 (4)	0.7032 (2)	0.0722 (14)
H62	0.647127	0.673339	0.733243	0.087*
C71	0.4959 (7)	0.8697 (6)	0.2532 (2)	0.098 (3)
H71A	0.538126	0.920157	0.243091	0.118*
H71B	0.448582	0.857509	0.237036	0.118*
C59	0.5091 (5)	0.7765 (4)	0.28509 (18)	0.0747 (15)
C53	0.4595 (5)	0.5312 (6)	0.6857 (3)	0.102 (3)
C51	0.4355 (7)	0.6224 (10)	0.7186 (3)	0.128 (4)
H51	0.427831	0.644872	0.743631	0.153*
C19	0.4327 (3)	0.6411 (3)	0.59622 (16)	0.0601 (11)
H19A	0.397072	0.596449	0.578330	0.072*
H19B	0.410754	0.675660	0.598706	0.072*
C36	0.5100 (4)	0.6927 (4)	0.3376 (2)	0.0698 (14)
H36	0.517156	0.650865	0.345400	0.084*
C49	0.6629 (5)	0.6528 (5)	0.6163 (2)	0.088 (2)
H49	0.668530	0.648094	0.586357	0.106*
C47	0.5196 (6)	0.7168 (5)	0.2946 (2)	0.088 (2)
H47	0.533049	0.692010	0.272999	0.105*
C64	0.6700 (6)	0.6018 (6)	0.6445 (3)	0.096 (2)
H64	0.678220	0.561892	0.634016	0.116*
C77	0.4734 (8)	0.4748 (11)	0.7437 (5)	0.170 (10)
H77A	0.428942	0.426019	0.754007	0.204*
H77B	0.520965	0.479272	0.756143	0.204*
Na32	0.6271 (6)	1.0207 (5)	0.5641 (5)	0.108 (4)	0.3333
Na12	0.4949 (5)	0.5206 (5)	0.5363 (3)	0.077 (2)	0.3333

Atomic displacement parameters (Ų)

	U^11	U^22	U^33	U^12	U^13	U^23
Ni1	0.0454 (3)	0.0496 (3)	0.0392 (3)	0.0234 (2)	−0.00398 (18)	0.00021 (18)
O2	0.0470 (14)	0.0490 (14)	0.0430 (14)	0.0234 (12)	−0.0029 (11)	−0.0057 (11)
O21	0.0583 (18)	0.101 (3)	0.0534 (19)	0.048 (2)	−0.0079 (15)	−0.0112 (18)
O3	0.0538 (16)	0.0558 (16)	0.0498 (18)	0.0218 (13)	−0.0103 (13)	−0.0028 (13)
O4	0.0522 (15)	0.0495 (15)	0.0560 (18)	0.0254 (13)	0.0010 (13)	−0.0011 (12)
O5	0.0511 (16)	0.0641 (18)	0.0544 (18)	0.0315 (14)	−0.0052 (13)	0.0025 (14)
O78	0.122 (4)	0.084 (3)	0.067 (3)	0.054 (3)	−0.013 (3)	0.016 (2)
N6	0.0457 (17)	0.0531 (18)	0.0425 (17)	0.0237 (14)	−0.0029 (13)	−0.0009 (14)
O58	0.109 (3)	0.081 (3)	0.049 (2)	0.059 (2)	0.001 (2)	0.0068 (18)
O81	0.126 (5)	0.085 (3)	0.076 (3)	0.051 (3)	−0.033 (3)	−0.014 (2)
O20	0.081 (3)	0.075 (2)	0.065 (2)	0.029 (2)	−0.020 (2)	−0.024 (2)
N7	0.054 (2)	0.066 (2)	0.0419 (19)	0.0277 (17)	−0.0043 (14)	0.0064 (16)
O72	0.148 (5)	0.106 (4)	0.046 (2)	0.079 (4)	0.006 (3)	0.006 (2)
C15	0.047 (2)	0.069 (3)	0.041 (2)	0.0256 (19)	−0.0024 (15)	0.0012 (18)
C13	0.050 (2)	0.062 (2)	0.048 (2)	0.0278 (19)	−0.0065 (17)	−0.0063 (18)
O76	0.186 (10)	0.290 (18)	0.094 (6)	0.105 (12)	0.000 (6)	0.109 (10)
O67	0.108 (4)	0.117 (5)	0.127 (6)	0.026 (4)	−0.028 (4)	0.064 (5)
C42	0.065 (3)	0.080 (3)	0.057 (3)	0.030 (3)	0.005 (2)	0.018 (2)
C35	0.061 (2)	0.064 (2)	0.045 (2)	0.036 (2)	−0.0055 (18)	−0.0047 (18)
C86	0.052 (2)	0.067 (3)	0.051 (2)	0.019 (2)	−0.0019 (18)	0.014 (2)
C24	0.057 (2)	0.063 (2)	0.042 (2)	0.034 (2)	−0.0088 (17)	−0.0038 (18)
C16	0.051 (2)	0.057 (2)	0.043 (2)	0.0261 (17)	−0.0070 (17)	−0.0034 (16)
C73	0.097 (5)	0.070 (3)	0.066 (4)	0.034 (3)	−0.021 (3)	−0.010 (3)
C18	0.055 (2)	0.077 (3)	0.045 (2)	0.025 (2)	−0.0115 (18)	−0.004 (2)
C23	0.053 (2)	0.063 (3)	0.052 (2)	0.016 (2)	−0.0085 (18)	0.011 (2)
C43	0.086 (4)	0.065 (3)	0.046 (2)	0.029 (3)	−0.006 (2)	0.005 (2)
C37	0.059 (3)	0.093 (4)	0.058 (3)	0.033 (3)	−0.018 (2)	0.001 (3)
C68	0.092 (4)	0.071 (3)	0.057 (3)	0.033 (3)	−0.013 (3)	0.015 (2)
C45	0.077 (3)	0.068 (3)	0.043 (2)	0.040 (2)	−0.002 (2)	0.0023 (19)
C8	0.0445 (18)	0.0445 (18)	0.029 (3)	0.0222 (9)	0.000	0.000
C9	0.054 (2)	0.067 (3)	0.050 (2)	0.022 (2)	−0.0096 (19)	−0.007 (2)
C28	0.051 (2)	0.093 (4)	0.055 (3)	0.025 (2)	−0.014 (2)	0.005 (3)
C65	0.101 (6)	0.183 (11)	0.064 (5)	0.050 (7)	0.000 (4)	0.055 (6)
C48	0.064 (3)	0.084 (3)	0.062 (3)	0.034 (3)	−0.017 (2)	−0.002 (3)
C29	0.057 (2)	0.088 (4)	0.051 (3)	0.030 (2)	−0.003 (2)	0.016 (2)
C57	0.099 (5)	0.083 (4)	0.059 (3)	0.051 (4)	−0.010 (3)	0.004 (3)
C55	0.075 (3)	0.092 (4)	0.053 (3)	0.029 (3)	0.002 (2)	0.026 (3)
C40	0.080 (4)	0.128 (7)	0.053 (3)	0.045 (4)	0.008 (3)	0.013 (3)
C41	0.070 (3)	0.079 (4)	0.067 (3)	0.017 (3)	−0.015 (3)	0.024 (3)
C62	0.073 (3)	0.083 (4)	0.053 (3)	0.033 (3)	−0.018 (2)	−0.011 (2)
C71	0.166 (9)	0.100 (5)	0.045 (3)	0.078 (6)	−0.001 (4)	0.004 (3)
C59	0.104 (5)	0.084 (4)	0.042 (2)	0.052 (3)	0.002 (3)	−0.001 (2)
C53	0.084 (4)	0.103 (5)	0.082 (5)	0.019 (4)	−0.016 (4)	0.048 (4)
C51	0.123 (8)	0.189 (13)	0.051 (4)	0.063 (8)	0.003 (4)	0.010 (6)
C19	0.055 (2)	0.077 (3)	0.044 (2)	0.030 (2)	−0.0009 (18)	0.012 (2)
C36	0.092 (4)	0.080 (3)	0.055 (3)	0.056 (3)	−0.007 (2)	−0.010 (2)
C49	0.102 (5)	0.107 (5)	0.062 (4)	0.056 (4)	−0.029 (3)	−0.019 (3)
C47	0.141 (6)	0.105 (5)	0.051 (3)	0.087 (5)	−0.003 (3)	−0.015 (3)
C64	0.128 (7)	0.105 (5)	0.068 (4)	0.067 (5)	−0.027 (4)	−0.024 (4)
C77	0.123 (8)	0.167 (12)	0.130 (11)	0.004 (8)	−0.052 (8)	0.097 (11)
Na32	0.092 (6)	0.059 (4)	0.154 (10)	0.023 (4)	−0.057 (7)	−0.018 (5)
Na12	0.083 (4)	0.088 (4)	0.084 (5)	0.059 (4)	−0.006 (3)	0.019 (4)

Geometric parameters (Å, º)

Ni1—O2	2.057 (3)	C16—H16B	0.9900
Ni1—O3	2.049 (4)	C73—C62	1.368 (11)
Ni1—O4	2.087 (3)	C73—C64	1.376 (11)
Ni1—O5	2.061 (3)	C18—H18	1.0000
Ni1—N6	2.113 (4)	C18—C9	1.538 (9)
Ni1—N7	2.106 (4)	C18—C28	1.540 (8)
O2—C8	1.288 (3)	C23—H23A	0.9900
O21—C13	1.260 (6)	C23—H23B	0.9900
O3—C9	1.258 (7)	C43—H43	0.9500
O4—H4A	0.8966	C43—C57	1.376 (9)
O4—H4B	0.8955	C37—C28	1.495 (9)
O4—Na12	2.676 (7)	C37—C48	1.419 (10)
O5—C13	1.242 (6)	C37—C49	1.346 (12)
O78—H78	0.8400	C68—C57	1.380 (10)
O78—C68	1.368 (8)	C68—C55	1.378 (12)
N6—H6	1.0000	C45—C59	1.390 (8)
N6—C15	1.477 (6)	C28—H28A	0.9900
N6—C16	1.489 (6)	C28—H28B	0.9900
O58—C45	1.387 (7)	C65—C53	1.31 (2)
O58—C71	1.412 (8)	C65—C51	1.36 (2)
O81—H81	0.8400	C48—H48	0.9500
O81—C73	1.391 (10)	C48—C62	1.379 (10)
O20—C9	1.272 (7)	C29—C40	1.375 (12)
O20—Na32	2.606 (12)	C29—C41	1.428 (12)
N7—H7	1.0000	C29—C19	1.511 (7)
N7—C18	1.484 (7)	C57—H57	0.9500
N7—C19	1.482 (7)	C55—H55	0.9500
O72—C71	1.416 (11)	C40—H40	0.9500
O72—C59	1.388 (8)	C40—C51	1.383 (13)
C15—H15	1.0000	C41—H41	0.9500
C15—C13	1.534 (7)	C41—C53	1.378 (9)
C15—C23	1.552 (8)	C62—H62	0.9500
O76—C65	1.368 (12)	C71—H71A	0.9900
O76—C77	1.34 (3)	C71—H71B	0.9900
O67—C53	1.406 (14)	C59—C47	1.347 (10)
O67—C77	1.47 (2)	C51—H51	0.9500
C42—H42	0.9500	C19—H19A	0.9900
C42—C86	1.394 (8)	C19—H19B	0.9900
C42—C55	1.384 (9)	C36—H36	0.9500
C35—H35	0.9500	C36—C47	1.403 (10)
C35—C24	1.411 (7)	C49—H49	0.9500
C35—C45	1.380 (7)	C49—C64	1.408 (13)
C86—C23	1.506 (7)	C47—H47	0.9500
C86—C43	1.391 (10)	C64—H64	0.9500
C24—C16	1.519 (6)	C77—H77A	0.9900
C24—C36	1.390 (8)	C77—H77B	0.9900
C16—H16A	0.9900	Na32—Na12^i	2.758 (13)
O2—Ni1—O4	89.97 (13)	C57—C43—H43	119.5
O2—Ni1—O5	172.12 (13)	C48—C37—C28	119.7 (7)
O2—Ni1—N6	91.78 (13)	C49—C37—C28	122.1 (7)
O2—Ni1—N7	100.07 (15)	C49—C37—C48	118.2 (7)
O3—Ni1—O2	84.63 (13)	O78—C68—C57	118.9 (8)
O3—Ni1—O4	172.62 (14)	O78—C68—C55	121.5 (6)
O3—Ni1—O5	97.42 (15)	C55—C68—C57	119.6 (6)
O3—Ni1—N6	89.73 (15)	O58—C45—C59	109.9 (5)
O3—Ni1—N7	82.13 (16)	C35—C45—O58	127.7 (5)
O4—Ni1—N6	95.48 (14)	C35—C45—C59	122.5 (5)
O4—Ni1—N7	93.90 (16)	O2^i—C8—O2^ii	119.990 (15)
O5—Ni1—O4	88.61 (14)	O2—C8—O2^ii	119.988 (14)
O5—Ni1—N6	80.64 (14)	O2^i—C8—O2	119.987 (15)
O5—Ni1—N7	87.76 (15)	O3—C9—O20	123.9 (6)
N7—Ni1—N6	164.88 (15)	O3—C9—C18	119.5 (5)
C8—O2—Ni1	132.6 (2)	O20—C9—C18	116.5 (5)
C9—O3—Ni1	114.7 (4)	C18—C28—H28A	109.0
Ni1—O4—H4A	112.4	C18—C28—H28B	109.0
Ni1—O4—H4B	109.6	C37—C28—C18	112.7 (5)
Ni1—O4—Na12	105.6 (3)	C37—C28—H28A	109.0
H4A—O4—H4B	103.0	C37—C28—H28B	109.0
Na12—O4—H4A	113.3	H28A—C28—H28B	107.8
Na12—O4—H4B	113.1	C53—C65—O76	111.0 (17)
C13—O5—Ni1	116.2 (3)	C53—C65—C51	122.5 (8)
C68—O78—H78	109.5	C51—C65—O76	126.4 (16)
Ni1—N6—H6	107.9	C37—C48—H48	120.0
C15—N6—Ni1	108.4 (3)	C62—C48—C37	120.0 (7)
C15—N6—H6	107.9	C62—C48—H48	120.0
C15—N6—C16	113.8 (4)	C40—C29—C41	118.2 (6)
C16—N6—Ni1	110.7 (3)	C40—C29—C19	122.9 (7)
C16—N6—H6	107.9	C41—C29—C19	118.8 (6)
C45—O58—C71	105.6 (5)	C43—C57—C68	120.4 (7)
C73—O81—H81	109.5	C43—C57—H57	119.8
C9—O20—Na32	125.2 (5)	C68—C57—H57	119.8
Ni1—N7—H7	106.7	C42—C55—H55	119.9
C18—N7—Ni1	108.2 (3)	C68—C55—C42	120.1 (6)
C18—N7—H7	106.7	C68—C55—H55	119.9
C19—N7—Ni1	115.2 (3)	C29—C40—H40	117.7
C19—N7—H7	106.7	C29—C40—C51	124.7 (12)
C19—N7—C18	112.7 (4)	C51—C40—H40	117.7
C59—O72—C71	106.1 (5)	C29—C41—H41	122.8
N6—C15—H15	109.4	C53—C41—C29	114.4 (9)
N6—C15—C13	111.0 (4)	C53—C41—H41	122.8
N6—C15—C23	111.0 (4)	C73—C62—C48	120.0 (6)
C13—C15—H15	109.4	C73—C62—H62	120.0
C13—C15—C23	106.7 (4)	C48—C62—H62	120.0
C23—C15—H15	109.4	O58—C71—O72	109.4 (6)
O21—C13—C15	117.1 (5)	O58—C71—H71A	109.8
O5—C13—O21	124.0 (5)	O58—C71—H71B	109.8
O5—C13—C15	118.8 (4)	O72—C71—H71A	109.8
C77—O76—C65	106.2 (16)	O72—C71—H71B	109.8
C53—O67—C77	100.9 (13)	H71A—C71—H71B	108.2
C86—C42—H42	119.6	O72—C59—C45	108.8 (6)
C55—C42—H42	119.6	C47—C59—O72	129.0 (6)
C55—C42—C86	120.8 (7)	C47—C59—C45	122.3 (6)
C24—C35—H35	122.2	C65—C53—O67	110.9 (9)
C45—C35—H35	122.2	C65—C53—C41	125.4 (11)
C45—C35—C24	115.5 (5)	C41—C53—O67	123.7 (12)
C42—C86—C23	120.8 (6)	C65—C51—C40	114.7 (12)
C43—C86—C42	118.1 (5)	C65—C51—H51	122.7
C43—C86—C23	121.1 (5)	C40—C51—H51	122.7
C35—C24—C16	119.4 (4)	N7—C19—C29	113.2 (4)
C36—C24—C35	121.5 (5)	N7—C19—H19A	108.9
C36—C24—C16	119.0 (5)	N7—C19—H19B	108.9
N6—C16—C24	114.8 (4)	C29—C19—H19A	108.9
N6—C16—H16A	108.6	C29—C19—H19B	108.9
N6—C16—H16B	108.6	H19A—C19—H19B	107.7
C24—C16—H16A	108.6	C24—C36—H36	119.5
C24—C16—H16B	108.6	C24—C36—C47	121.0 (6)
H16A—C16—H16B	107.6	C47—C36—H36	119.5
C62—C73—O81	122.6 (7)	C37—C49—H49	118.7
C62—C73—C64	121.4 (7)	C37—C49—C64	122.6 (7)
C64—C73—O81	115.8 (8)	C64—C49—H49	118.7
N7—C18—H18	109.0	C59—C47—C36	117.3 (6)
N7—C18—C9	110.0 (4)	C59—C47—H47	121.4
N7—C18—C28	112.2 (5)	C36—C47—H47	121.4
C9—C18—H18	109.0	C73—C64—C49	117.6 (8)
C9—C18—C28	107.5 (5)	C73—C64—H64	121.2
C28—C18—H18	109.0	C49—C64—H64	121.2
C15—C23—H23A	109.0	O76—C77—O67	110.1 (10)
C15—C23—H23B	109.0	O76—C77—H77A	109.6
C86—C23—C15	112.9 (4)	O76—C77—H77B	109.6
C86—C23—H23A	109.0	O67—C77—H77A	109.6
C86—C23—H23B	109.0	O67—C77—H77B	109.6
H23A—C23—H23B	107.8	H77A—C77—H77B	108.2
C86—C43—H43	119.5	O20—Na32—Na12^i	96.5 (4)
C57—C43—C86	121.0 (6)	O4—Na12—Na32^ii	119.0 (4)
Ni1—O2—C8—O2^ii	2.4 (9)	C37—C48—C62—C73	−3.1 (10)
Ni1—O2—C8—O2^i	−179.8 (2)	C37—C49—C64—C73	−2.5 (14)
Ni1—O3—C9—O20	−174.1 (5)	C45—O58—C71—O72	−5.4 (11)
Ni1—O3—C9—C18	10.0 (6)	C45—C35—C24—C16	−179.2 (5)
Ni1—O5—C13—O21	176.1 (4)	C45—C35—C24—C36	−1.7 (8)
Ni1—O5—C13—C15	−0.4 (6)	C45—C59—C47—C36	−0.6 (14)
Ni1—N6—C15—C13	23.3 (4)	C9—C18—C28—C37	168.2 (5)
Ni1—N6—C15—C23	−95.1 (4)	C28—C18—C9—O3	98.4 (6)
Ni1—N6—C16—C24	176.1 (3)	C28—C18—C9—O20	−77.9 (7)
Ni1—N7—C18—C9	24.3 (5)	C28—C37—C48—C62	−174.1 (5)
Ni1—N7—C18—C28	−95.3 (4)	C28—C37—C49—C64	176.9 (8)
Ni1—N7—C19—C29	171.8 (4)	C65—O76—C77—O67	9.9 (16)
O78—C68—C57—C43	−178.7 (7)	C48—C37—C28—C18	−68.8 (7)
O78—C68—C55—C42	177.9 (7)	C48—C37—C49—C64	−0.1 (12)
N6—C15—C13—O21	166.9 (4)	C29—C40—C51—C65	4.1 (16)
N6—C15—C13—O5	−16.4 (6)	C29—C41—C53—O67	−177.9 (7)
N6—C15—C23—C86	−60.6 (6)	C29—C41—C53—C65	0.7 (12)
O58—C45—C59—O72	−1.4 (9)	C57—C68—C55—C42	−3.5 (11)
O58—C45—C59—C47	178.6 (8)	C55—C42—C86—C23	178.6 (6)
O81—C73—C62—C48	177.0 (7)	C55—C42—C86—C43	−0.5 (9)
O81—C73—C64—C49	−174.5 (8)	C55—C68—C57—C43	2.7 (11)
N7—C18—C9—O3	−24.1 (7)	C40—C29—C41—C53	0.9 (9)
N7—C18—C9—O20	159.7 (5)	C40—C29—C19—N7	105.7 (7)
N7—C18—C28—C37	−70.7 (7)	C41—C29—C40—C51	−3.5 (12)
O72—C59—C47—C36	179.4 (8)	C41—C29—C19—N7	−70.5 (7)
C15—N6—C16—C24	−61.6 (5)	C62—C73—C64—C49	2.3 (14)
C13—C15—C23—C86	178.4 (5)	C71—O58—C45—C35	−177.1 (7)
O76—C65—C53—O67	2.3 (13)	C71—O58—C45—C59	4.1 (9)
O76—C65—C53—C41	−176.4 (10)	C71—O72—C59—C45	−2.0 (10)
O76—C65—C51—C40	173.6 (11)	C71—O72—C59—C47	178.0 (10)
C42—C86—C23—C15	−76.5 (6)	C59—O72—C71—O58	4.6 (11)
C42—C86—C43—C57	−0.3 (9)	C53—O67—C77—O76	−8.4 (14)
C35—C24—C16—N6	−63.7 (6)	C53—C65—C51—C40	−2.4 (17)
C35—C24—C36—C47	1.0 (10)	C51—C65—C53—O67	178.9 (11)
C35—C45—C59—O72	179.8 (6)	C51—C65—C53—C41	0.2 (16)
C35—C45—C59—C47	−0.2 (12)	C19—N7—C18—C9	−104.3 (5)
C86—C42—C55—C68	2.5 (10)	C19—N7—C18—C28	136.1 (5)
C86—C43—C57—C68	−0.8 (11)	C19—C29—C40—C51	−179.7 (9)
C24—C35—C45—O58	−177.3 (6)	C19—C29—C41—C53	177.3 (6)
C24—C35—C45—C59	1.3 (9)	C36—C24—C16—N6	118.7 (5)
C24—C36—C47—C59	0.2 (13)	C49—C37—C28—C18	114.2 (8)
C16—N6—C15—C13	−100.3 (4)	C49—C37—C48—C62	2.9 (10)
C16—N6—C15—C23	141.3 (4)	C64—C73—C62—C48	0.5 (12)
C16—C24—C36—C47	178.5 (7)	C77—O76—C65—C53	−7.7 (16)
C18—N7—C19—C29	−63.3 (7)	C77—O76—C65—C51	175.9 (13)
C23—C15—C13—O21	−72.0 (6)	C77—O67—C53—C65	3.5 (10)
C23—C15—C13—O5	104.7 (5)	C77—O67—C53—C41	−177.7 (8)
C23—C86—C43—C57	−179.4 (6)	Na32—O20—C9—O3	−0.5 (9)
C43—C86—C23—C15	102.6 (6)	Na32—O20—C9—C18	175.6 (5)

Symmetry codes: (i) −y+1, x−y+1, z; (ii) −x+y, −x+1, z.

Hydrogen-bond geometry (Å, º)

D—H···A	D—H	H···A	D···A	D—H···A
O4—H4B···O2^ii	0.90	1.80	2.619 (5)	151
O78—H78···O20^iii	0.84	1.86	2.681 (8)	166
O81—H81···O21^iv	0.84	1.82	2.636 (7)	164

Symmetry codes: (ii) −x+y, −x+1, z; (iii) −x+y+1/3, −x+5/3, z−1/3; (iv) −x+y+2/3, −x+4/3, z+1/3.

2Ni (2ni-sc)

Crystal data

C34H36N2NiO10·C3H7NO·2(H2O)	Dx = 1.439 Mg m^−3
Mr = 800.48	Synchrotron radiation, λ = 0.8261 Å
Orthorhombic, P21212	Cell parameters from 5935 reflections
a = 14.910 (3) Å	θ = 1.4–28.5°
b = 33.6200 (9) Å	µ = 0.89 mm^−1
c = 7.370 (2) Å	T = 100 K
V = 3694.4 (12) Å^3	Prism, blue-green
Z = 4	0.1 × 0.1 × 0.05 mm
F(000) = 1688

Data collection

MX2_LNLS diffractometer	5935 independent reflections
Radiation source: LNLS-MX2	5905 reflections with I > 2σ(I)
Detector resolution: 5.814 pixels mm^-1	Rint = 0.038
shuterless φ scans	θmax = 28.5°, θmin = 1.4°
Absorption correction: empirical (using intensity measurements) XDS (Kabsch, 2010)	h = −16→16
Tmin = 0.929, Tmax = 1	k = −38→38
113805 measured reflections	l = −8→8

Refinement

Refinement on F^2	H-atom parameters constrained
Least-squares matrix: full	w = 1/[σ^2(Fo^2) + (0.0241P)^2 + 5.5869P] where P = (Fo^2 + 2Fc^2)/3
R[F^2 > 2σ(F^2)] = 0.034	(Δ/σ)max < 0.001
wR(F^2) = 0.079	Δρmax = 0.57 e Å^−3
S = 1.05	Δρmin = −0.57 e Å^−3
5935 reflections	Extinction correction: SHELXL-2014/7 (Sheldrick 2014, Fc^*=kFc[1+0.001xFc^2λ^3/sin(2θ)]^-1/4
517 parameters	Extinction coefficient: 0.00110 (19)
0 restraints	Absolute structure: Flack x determined using 2509 quotients [(I+)-(I-)]/[(I+)+(I-)] (Parsons, Flack and Wagner, Acta Cryst. B69 (2013) 249-259).
Primary atom site location: iterative	Absolute structure parameter: 0.022 (2)
Hydrogen site location: mixed

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.

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

	x	y	z	Uiso*/Ueq	Occ. (<1)
Ni1	0.41514 (3)	0.74733 (2)	0.40203 (6)	0.01521 (13)
O5	0.53948 (17)	0.74095 (8)	0.5150 (4)	0.0246 (6)
O6	0.63832 (19)	0.77379 (8)	0.6853 (4)	0.0256 (6)
O3	0.3086 (2)	0.50326 (8)	0.2853 (4)	0.0286 (7)
O4	0.42554 (19)	0.53700 (8)	0.1512 (4)	0.0262 (6)
N1	0.3885 (2)	0.68704 (9)	0.4589 (4)	0.0154 (7)
H1	0.4443	0.6718	0.4293	0.019*
N2	0.4620 (2)	0.80761 (9)	0.3915 (4)	0.0153 (6)
H2	0.4101	0.8262	0.4061	0.018*
O7	0.6875 (2)	0.96074 (9)	0.1211 (5)	0.0379 (8)
O8	0.5492 (2)	0.96261 (8)	0.2598 (5)	0.0378 (8)
O2	0.3492 (2)	0.75330 (9)	0.6418 (4)	0.0358 (8)
O9	0.3002 (2)	0.76153 (10)	0.2652 (6)	0.0630 (13)
H9A	0.2829	0.7879	0.2778	0.094*
H9B	0.2498	0.7486	0.3095	0.094*
C28	0.5578 (2)	0.85355 (11)	0.1944 (5)	0.0159 (8)
C31	0.6530 (3)	0.92260 (12)	0.1325 (5)	0.0234 (9)
C33	0.6422 (3)	0.85290 (11)	0.1145 (5)	0.0192 (8)
H33	0.6675	0.8280	0.0815	0.023*
C13	0.3696 (3)	0.56399 (11)	0.2376 (5)	0.0206 (8)
C11	0.3117 (3)	0.62496 (11)	0.3439 (5)	0.0168 (8)
C15	0.2347 (3)	0.56297 (11)	0.4152 (6)	0.0236 (8)
H15	0.1872	0.5490	0.4727	0.028*
C19	0.5235 (3)	0.81186 (11)	0.5481 (5)	0.0171 (8)
H19	0.5703	0.8318	0.5143	0.021*
C20	0.4756 (3)	0.82722 (11)	0.7193 (5)	0.0195 (8)
H20A	0.4214	0.8109	0.7412	0.023*
H20B	0.5159	0.8240	0.8249	0.023*
C12	0.3784 (3)	0.60421 (11)	0.2474 (5)	0.0186 (8)
H12	0.4272	0.6176	0.1917	0.022*
C26	0.5112 (3)	0.90009 (12)	0.7389 (5)	0.0236 (9)
H26	0.5700	0.8929	0.7765	0.028*
C16	0.2425 (2)	0.60441 (11)	0.4265 (5)	0.0197 (8)
H16	0.1987	0.6189	0.4930	0.024*
C18	0.5709 (2)	0.77252 (11)	0.5856 (5)	0.0195 (8)
C29	0.5204 (3)	0.89038 (11)	0.2478 (6)	0.0211 (8)
H29	0.4632	0.8918	0.3044	0.025*
C21	0.4483 (3)	0.87038 (12)	0.7046 (5)	0.0207 (9)
C27	0.5066 (3)	0.81512 (11)	0.2143 (5)	0.0194 (8)
H27A	0.4602	0.8143	0.1183	0.023*
H27B	0.5487	0.7929	0.1908	0.023*
N3	0.2167 (3)	0.88494 (12)	0.0539 (6)	0.0404 (10)
C14	0.2992 (3)	0.54373 (11)	0.3169 (6)	0.0215 (8)
C32	0.6914 (3)	0.88766 (12)	0.0810 (6)	0.0230 (8)
H32	0.7487	0.8869	0.0250	0.028*
C30	0.5697 (3)	0.92366 (11)	0.2145 (5)	0.0231 (9)
C2	0.3730 (3)	0.68253 (12)	0.6558 (5)	0.0288 (10)
H2A	0.3234	0.6629	0.6742	0.035*
C9	0.4374 (3)	0.59324 (12)	0.7376 (6)	0.0281 (10)
H9	0.3844	0.5958	0.8083	0.034*
O10'	0.4409 (5)	0.7325 (2)	0.1408 (10)	0.0205 (17)	0.49
H10A	0.4969	0.7378	0.1158	0.031*	0.49
H10B	0.4105	0.7477	0.0674	0.031*	0.49
C22	0.3630 (3)	0.88178 (13)	0.6510 (6)	0.0291 (10)
H22	0.3184	0.8621	0.6294	0.035*
C4	0.4861 (4)	0.62701 (14)	0.6929 (7)	0.0391 (13)
C10	0.3152 (3)	0.66989 (11)	0.3471 (5)	0.0189 (8)
H10C	0.3220	0.6796	0.2210	0.023*
H10D	0.2572	0.6800	0.3935	0.023*
C17	0.3733 (3)	0.50124 (12)	0.1419 (6)	0.0306 (10)
H17A	0.3429	0.4993	0.0229	0.037*
H17B	0.4123	0.4777	0.1576	0.037*
C25	0.4898 (3)	0.93964 (13)	0.7192 (7)	0.0307 (10)
H25	0.5335	0.9594	0.7448	0.037*
C23	0.3423 (3)	0.92174 (14)	0.6285 (7)	0.0383 (12)
H23	0.2839	0.9291	0.5894	0.046*
C24	0.4050 (3)	0.95074 (13)	0.6622 (7)	0.0363 (11)
H24	0.3904	0.9780	0.6467	0.044*
C34	0.6150 (4)	0.98625 (13)	0.1714 (7)	0.0379 (12)
H34A	0.6366	1.0074	0.2541	0.045*
H34B	0.5890	0.9990	0.0623	0.045*
C8	0.4648 (4)	0.55589 (14)	0.6809 (7)	0.0372 (12)
H8	0.4311	0.5330	0.7134	0.045*
C1	0.3446 (4)	0.72228 (13)	0.7380 (6)	0.0427 (14)
C3	0.4562 (5)	0.66737 (14)	0.7549 (8)	0.0600 (19)
H3A	0.5058	0.6865	0.7362	0.072*
H3B	0.4433	0.6663	0.8866	0.072*
C37	0.2577 (4)	0.92218 (15)	0.1040 (9)	0.0571 (16)
H37A	0.2773	0.9362	−0.0056	0.086*
H37B	0.3097	0.9171	0.1821	0.086*
H37C	0.2141	0.9386	0.1696	0.086*
C5	0.5630 (3)	0.6221 (2)	0.5881 (11)	0.078 (3)
H5	0.5975	0.6447	0.5543	0.094*
C36	0.1421 (4)	0.8867 (2)	−0.0689 (9)	0.0649 (18)
H36A	0.0941	0.9027	−0.0150	0.097*
H36B	0.1200	0.8597	−0.0922	0.097*
H36C	0.1614	0.8989	−0.1833	0.097*
C7	0.5411 (5)	0.5521 (2)	0.5774 (9)	0.074 (2)
H7	0.5601	0.5266	0.5369	0.089*
C35	0.2430 (5)	0.85120 (17)	0.1233 (12)	0.079 (3)
H35	0.2143	0.8277	0.0818	0.095*
C6	0.5890 (4)	0.5851 (3)	0.5338 (12)	0.104 (4)
H6	0.6421	0.5824	0.4634	0.125*
O1	0.2965 (5)	0.7257 (2)	0.8789 (9)	0.0186 (15)	0.48
O1'	0.3387 (5)	0.71750 (19)	0.9173 (9)	0.0258 (15)	0.52
O13	0.6211 (4)	0.73246 (17)	1.0067 (6)	0.101 (2)
H13A	0.6707	0.7248	1.0580	0.151*
H13B	0.6384	0.7471	0.9159	0.151*
O12	0.1904 (3)	0.80210 (17)	0.6551 (9)	0.098 (2)
H12A	0.2342	0.7853	0.6436	0.147*
H12B	0.1418	0.7921	0.6080	0.147*
O11	0.3010 (5)	0.8476 (2)	0.2359 (9)	0.135 (3)
O10	0.4801 (5)	0.7220 (2)	0.1656 (9)	0.0216 (15)	0.51
H10E	0.4409	0.7181	0.0784	0.032*	0.51
H10F	0.4998	0.6980	0.1911	0.032*	0.51

Atomic displacement parameters (Ų)

	U^11	U^22	U^33	U^12	U^13	U^23
Ni1	0.0165 (2)	0.0158 (2)	0.0133 (2)	−0.00411 (19)	−0.00139 (18)	0.0006 (2)
O5	0.0177 (13)	0.0191 (15)	0.0369 (16)	0.0017 (11)	−0.0039 (12)	−0.0099 (12)
O6	0.0198 (15)	0.0283 (15)	0.0286 (15)	−0.0033 (12)	−0.0066 (13)	0.0020 (12)
O3	0.0314 (16)	0.0145 (13)	0.0399 (18)	−0.0019 (12)	−0.0009 (14)	−0.0036 (13)
O4	0.0234 (15)	0.0211 (14)	0.0342 (16)	0.0018 (12)	0.0021 (13)	−0.0069 (12)
N1	0.0176 (17)	0.0163 (16)	0.0125 (15)	0.0000 (12)	−0.0014 (12)	−0.0032 (12)
N2	0.0168 (15)	0.0182 (15)	0.0110 (14)	−0.0021 (12)	0.0022 (14)	−0.0007 (13)
O7	0.0488 (19)	0.0285 (16)	0.0365 (18)	−0.0215 (15)	0.0083 (16)	−0.0003 (15)
O8	0.053 (2)	0.0154 (14)	0.045 (2)	−0.0029 (13)	0.0148 (17)	0.0004 (14)
O2	0.0439 (17)	0.0178 (14)	0.0459 (18)	−0.0096 (14)	0.0304 (14)	−0.0081 (15)
O9	0.045 (2)	0.0270 (19)	0.117 (4)	−0.0224 (16)	−0.054 (2)	0.036 (2)
C28	0.0153 (19)	0.0203 (19)	0.0120 (18)	−0.0004 (14)	0.0014 (14)	0.0018 (15)
C31	0.029 (2)	0.024 (2)	0.017 (2)	−0.0111 (17)	−0.0012 (17)	0.0027 (16)
C33	0.022 (2)	0.0222 (19)	0.0137 (18)	0.0008 (15)	0.0007 (17)	−0.0005 (16)
C13	0.0181 (19)	0.022 (2)	0.022 (2)	0.0042 (16)	−0.0042 (17)	−0.0030 (16)
C11	0.0144 (18)	0.0179 (19)	0.0182 (19)	−0.0005 (15)	−0.0071 (15)	0.0007 (14)
C15	0.020 (2)	0.025 (2)	0.026 (2)	−0.0070 (16)	0.0002 (18)	0.0045 (18)
C19	0.0169 (19)	0.0193 (19)	0.0151 (19)	−0.0074 (15)	0.0007 (15)	−0.0003 (15)
C20	0.024 (2)	0.022 (2)	0.0125 (19)	−0.0041 (16)	0.0033 (16)	−0.0020 (16)
C12	0.0168 (19)	0.0218 (19)	0.0174 (19)	−0.0028 (15)	−0.0030 (16)	−0.0009 (16)
C26	0.025 (2)	0.027 (2)	0.019 (2)	−0.0050 (17)	−0.0019 (17)	−0.0038 (17)
C16	0.0162 (19)	0.0237 (19)	0.019 (2)	−0.0021 (15)	−0.0003 (17)	−0.0025 (16)
C18	0.016 (2)	0.0235 (19)	0.0192 (19)	−0.0060 (15)	0.0049 (17)	−0.0024 (17)
C29	0.020 (2)	0.021 (2)	0.023 (2)	−0.0006 (16)	0.0067 (17)	0.0013 (17)
C21	0.024 (2)	0.025 (2)	0.0134 (19)	−0.0052 (16)	0.0074 (16)	−0.0056 (16)
C27	0.027 (2)	0.0178 (19)	0.0138 (19)	−0.0033 (16)	0.0027 (16)	−0.0020 (16)
N3	0.054 (3)	0.030 (2)	0.037 (2)	−0.0005 (18)	0.006 (2)	0.0025 (18)
C14	0.023 (2)	0.0157 (19)	0.025 (2)	−0.0035 (16)	−0.0059 (17)	0.0001 (16)
C32	0.017 (2)	0.033 (2)	0.019 (2)	−0.0048 (16)	0.0006 (18)	0.0001 (18)
C30	0.034 (2)	0.0159 (18)	0.020 (2)	0.0003 (16)	0.0019 (18)	0.0013 (15)
C2	0.057 (3)	0.017 (2)	0.0125 (19)	−0.010 (2)	0.0022 (19)	0.0018 (15)
C9	0.039 (3)	0.024 (2)	0.021 (2)	0.0006 (18)	0.0033 (19)	0.0046 (17)
O10'	0.024 (5)	0.021 (4)	0.016 (4)	0.006 (3)	−0.002 (3)	−0.002 (3)
C22	0.021 (2)	0.033 (2)	0.034 (2)	−0.0044 (18)	0.0036 (18)	−0.0124 (19)
C4	0.047 (3)	0.033 (3)	0.037 (3)	−0.012 (2)	−0.027 (2)	0.018 (2)
C10	0.0166 (19)	0.0183 (19)	0.022 (2)	−0.0018 (15)	−0.0061 (16)	0.0019 (15)
C17	0.030 (2)	0.0184 (19)	0.043 (3)	−0.0001 (18)	−0.005 (2)	−0.0085 (19)
C25	0.035 (3)	0.025 (2)	0.032 (3)	−0.0098 (19)	0.003 (2)	−0.0067 (19)
C23	0.022 (2)	0.036 (2)	0.057 (3)	0.0063 (19)	0.004 (2)	−0.008 (2)
C24	0.038 (3)	0.023 (2)	0.048 (3)	0.003 (2)	0.011 (2)	−0.0051 (19)
C34	0.058 (3)	0.021 (2)	0.034 (2)	−0.012 (2)	0.006 (2)	0.0048 (19)
C8	0.053 (3)	0.029 (2)	0.029 (2)	0.011 (2)	0.009 (2)	0.008 (2)
C1	0.075 (4)	0.024 (2)	0.029 (3)	−0.023 (2)	0.028 (3)	−0.012 (2)
C3	0.121 (5)	0.022 (2)	0.038 (3)	−0.019 (3)	−0.049 (3)	0.010 (2)
C37	0.072 (4)	0.035 (3)	0.064 (4)	0.009 (3)	−0.030 (4)	−0.006 (3)
C5	0.022 (3)	0.101 (5)	0.111 (6)	−0.013 (3)	−0.023 (3)	0.082 (5)
C36	0.047 (3)	0.091 (5)	0.056 (4)	−0.020 (3)	0.010 (3)	−0.025 (4)
C7	0.086 (5)	0.079 (4)	0.056 (4)	0.062 (4)	0.034 (4)	0.042 (4)
C35	0.089 (5)	0.033 (3)	0.117 (7)	0.028 (3)	0.065 (5)	0.032 (4)
C6	0.042 (4)	0.144 (7)	0.126 (7)	0.057 (4)	0.044 (4)	0.109 (6)
O1	0.021 (4)	0.023 (4)	0.012 (3)	0.002 (3)	0.002 (3)	0.002 (3)
O1'	0.041 (5)	0.020 (3)	0.016 (3)	−0.004 (3)	0.006 (3)	0.000 (3)
O13	0.146 (5)	0.121 (4)	0.036 (2)	0.104 (4)	0.036 (3)	0.026 (2)
O12	0.031 (2)	0.119 (4)	0.145 (5)	0.000 (2)	0.001 (3)	−0.103 (4)
O11	0.146 (6)	0.150 (6)	0.108 (5)	0.115 (5)	0.061 (5)	0.086 (5)
O10	0.030 (5)	0.015 (3)	0.020 (3)	−0.003 (3)	0.002 (3)	0.002 (3)

Geometric parameters (Å, º)

Ni1—O5	2.043 (3)	C27—H27A	0.9900
Ni1—N1	2.107 (3)	C27—H27B	0.9900
Ni1—N2	2.145 (3)	N3—C37	1.442 (7)
Ni1—O2	2.032 (3)	N3—C36	1.434 (8)
Ni1—O9	2.045 (3)	N3—C35	1.305 (7)
Ni1—O10'	2.026 (7)	C32—H32	0.9500
Ni1—O10	2.168 (7)	C2—H2A	1.0000
O5—C18	1.271 (4)	C2—C1	1.527 (6)
O6—C18	1.246 (5)	C2—C3	1.527 (7)
O3—C14	1.387 (5)	C9—H9	0.9500
O3—C17	1.433 (6)	C9—C4	1.388 (6)
O4—C13	1.388 (5)	C9—C8	1.385 (6)
O4—C17	1.434 (5)	O10'—H10A	0.8726
N1—H1	1.0000	O10'—H10B	0.8712
N1—C2	1.477 (5)	C22—H22	0.9500
N1—C10	1.485 (5)	C22—C23	1.389 (6)
N2—H2	1.0000	C4—C3	1.500 (8)
N2—C19	1.481 (5)	C4—C5	1.392 (9)
N2—C27	1.488 (5)	C10—H10C	0.9900
O7—C31	1.384 (5)	C10—H10D	0.9900
O7—C34	1.429 (6)	C17—H17A	0.9900
O8—C30	1.386 (5)	C17—H17B	0.9900
O8—C34	1.421 (5)	C25—H25	0.9500
O2—C1	1.263 (6)	C25—C24	1.383 (7)
O9—H9A	0.9290	C23—H23	0.9500
O9—H9B	0.9267	C23—C24	1.374 (7)
C28—C33	1.389 (5)	C24—H24	0.9500
C28—C29	1.414 (5)	C34—H34A	0.9900
C28—C27	1.507 (5)	C34—H34B	0.9900
C31—C32	1.361 (6)	C8—H8	0.9500
C31—C30	1.381 (6)	C8—C7	1.375 (8)
C33—H33	0.9500	C1—O1	1.267 (8)
C33—C32	1.402 (5)	C1—O1'	1.334 (8)
C13—C12	1.360 (5)	C3—H3A	0.9900
C13—C14	1.380 (6)	C3—H3B	0.9900
C11—C12	1.407 (6)	C37—H37A	0.9800
C11—C16	1.384 (5)	C37—H37B	0.9800
C11—C10	1.511 (5)	C37—H37C	0.9800
C15—H15	0.9500	C5—H5	0.9500
C15—C16	1.400 (5)	C5—C6	1.362 (12)
C15—C14	1.367 (6)	C36—H36A	0.9800
C19—H19	1.0000	C36—H36B	0.9800
C19—C20	1.539 (5)	C36—H36C	0.9800
C19—C18	1.525 (5)	C7—H7	0.9500
C20—H20A	0.9900	C7—C6	1.360 (11)
C20—H20B	0.9900	C35—H35	0.9500
C20—C21	1.511 (6)	C35—O11	1.205 (11)
C12—H12	0.9500	C6—H6	0.9500
C26—H26	0.9500	O13—H13A	0.8694
C26—C21	1.394 (6)	O13—H13B	0.8701
C26—C25	1.375 (6)	O12—H12A	0.8679
C16—H16	0.9500	O12—H12B	0.8709
C29—H29	0.9500	O10—H10E	0.8786
C29—C30	1.362 (6)	O10—H10F	0.8783
C21—C22	1.386 (6)
O5—Ni1—N1	89.37 (11)	C13—C14—O3	109.7 (3)
O5—Ni1—N2	79.55 (11)	C15—C14—O3	128.6 (4)
O5—Ni1—O9	170.28 (13)	C15—C14—C13	121.7 (4)
O5—Ni1—O10	83.2 (2)	C31—C32—C33	116.8 (4)
N1—Ni1—N2	167.75 (12)	C31—C32—H32	121.6
N1—Ni1—O10	82.3 (2)	C33—C32—H32	121.6
N2—Ni1—O10	101.4 (2)	C31—C30—O8	109.2 (3)
O2—Ni1—O5	95.45 (13)	C29—C30—O8	127.9 (4)
O2—Ni1—N1	80.26 (12)	C29—C30—C31	122.9 (4)
O2—Ni1—N2	95.48 (12)	N1—C2—H2A	108.5
O2—Ni1—O9	90.03 (18)	N1—C2—C1	110.1 (3)
O2—Ni1—O10	162.5 (2)	N1—C2—C3	112.2 (4)
O9—Ni1—N1	99.49 (12)	C1—C2—H2A	108.5
O9—Ni1—N2	91.96 (12)	C3—C2—H2A	108.5
O9—Ni1—O10	94.0 (3)	C3—C2—C1	109.2 (4)
O10'—Ni1—O5	100.9 (2)	C4—C9—H9	119.5
O10'—Ni1—N1	89.3 (2)	C8—C9—H9	119.5
O10'—Ni1—N2	97.8 (2)	C8—C9—C4	121.0 (5)
O10'—Ni1—O2	160.5 (2)	Ni1—O10'—H10A	109.4
O10'—Ni1—O9	75.4 (3)	Ni1—O10'—H10B	110.3
C18—O5—Ni1	114.4 (2)	H10A—O10'—H10B	104.4
C14—O3—C17	103.8 (3)	C21—C22—H22	119.8
C13—O4—C17	104.1 (3)	C21—C22—C23	120.4 (4)
Ni1—N1—H1	107.0	C23—C22—H22	119.8
C2—N1—Ni1	108.9 (2)	C9—C4—C3	120.8 (5)
C2—N1—H1	107.0	C9—C4—C5	117.8 (5)
C2—N1—C10	113.0 (3)	C5—C4—C3	121.4 (5)
C10—N1—Ni1	113.7 (2)	N1—C10—C11	115.0 (3)
C10—N1—H1	107.0	N1—C10—H10C	108.5
Ni1—N2—H2	109.6	N1—C10—H10D	108.5
C19—N2—Ni1	105.3 (2)	C11—C10—H10C	108.5
C19—N2—H2	109.6	C11—C10—H10D	108.5
C19—N2—C27	113.0 (3)	H10C—C10—H10D	107.5
C27—N2—Ni1	109.7 (2)	O3—C17—O4	106.9 (3)
C27—N2—H2	109.6	O3—C17—H17A	110.3
C31—O7—C34	105.0 (3)	O3—C17—H17B	110.3
C30—O8—C34	105.4 (3)	O4—C17—H17A	110.3
C1—O2—Ni1	115.7 (3)	O4—C17—H17B	110.3
Ni1—O9—H9A	114.0	H17A—C17—H17B	108.6
Ni1—O9—H9B	113.4	C26—C25—H25	119.8
H9A—O9—H9B	100.7	C26—C25—C24	120.3 (4)
C33—C28—C29	119.3 (3)	C24—C25—H25	119.8
C33—C28—C27	119.1 (3)	C22—C23—H23	119.6
C29—C28—C27	121.6 (3)	C24—C23—C22	120.9 (4)
C32—C31—O7	128.8 (4)	C24—C23—H23	119.6
C32—C31—C30	121.5 (4)	C25—C24—H24	120.5
C30—C31—O7	109.7 (4)	C23—C24—C25	119.1 (4)
C28—C33—H33	118.8	C23—C24—H24	120.5
C28—C33—C32	122.4 (4)	O7—C34—H34A	110.1
C32—C33—H33	118.8	O7—C34—H34B	110.1
C12—C13—O4	128.0 (4)	O8—C34—O7	107.8 (3)
C12—C13—C14	122.8 (4)	O8—C34—H34A	110.1
C14—C13—O4	109.2 (3)	O8—C34—H34B	110.1
C12—C11—C10	118.6 (3)	H34A—C34—H34B	108.5
C16—C11—C12	120.1 (4)	C9—C8—H8	120.1
C16—C11—C10	121.2 (3)	C7—C8—C9	119.8 (5)
C16—C15—H15	121.8	C7—C8—H8	120.1
C14—C15—H15	121.8	O2—C1—C2	119.0 (4)
C14—C15—C16	116.4 (4)	O2—C1—O1	114.6 (5)
N2—C19—H19	107.7	O2—C1—O1'	131.2 (5)
N2—C19—C20	112.6 (3)	O1—C1—C2	124.2 (5)
N2—C19—C18	110.2 (3)	O1'—C1—C2	107.8 (5)
C20—C19—H19	107.7	C2—C3—H3A	108.9
C18—C19—H19	107.7	C2—C3—H3B	108.9
C18—C19—C20	110.9 (3)	C4—C3—C2	113.5 (4)
C19—C20—H20A	109.0	C4—C3—H3A	108.9
C19—C20—H20B	109.0	C4—C3—H3B	108.9
H20A—C20—H20B	107.8	H3A—C3—H3B	107.7
C21—C20—C19	112.9 (3)	N3—C37—H37A	109.5
C21—C20—H20A	109.0	N3—C37—H37B	109.5
C21—C20—H20B	109.0	N3—C37—H37C	109.5
C13—C12—C11	116.8 (4)	H37A—C37—H37B	109.5
C13—C12—H12	121.6	H37A—C37—H37C	109.5
C11—C12—H12	121.6	H37B—C37—H37C	109.5
C21—C26—H26	119.4	C4—C5—H5	119.8
C25—C26—H26	119.4	C6—C5—C4	120.4 (5)
C25—C26—C21	121.1 (4)	C6—C5—H5	119.8
C11—C16—C15	122.1 (4)	N3—C36—H36A	109.5
C11—C16—H16	118.9	N3—C36—H36B	109.5
C15—C16—H16	118.9	N3—C36—H36C	109.5
O5—C18—C19	118.6 (3)	H36A—C36—H36B	109.5
O6—C18—O5	124.6 (3)	H36A—C36—H36C	109.5
O6—C18—C19	116.8 (3)	H36B—C36—H36C	109.5
C28—C29—H29	121.4	C8—C7—H7	120.4
C30—C29—C28	117.1 (4)	C6—C7—C8	119.3 (6)
C30—C29—H29	121.4	C6—C7—H7	120.4
C26—C21—C20	119.6 (4)	N3—C35—H35	117.5
C22—C21—C20	122.2 (4)	O11—C35—N3	125.0 (8)
C22—C21—C26	118.1 (4)	O11—C35—H35	117.5
N2—C27—C28	117.2 (3)	C5—C6—H6	119.1
N2—C27—H27A	108.0	C7—C6—C5	121.8 (6)
N2—C27—H27B	108.0	C7—C6—H6	119.1
C28—C27—H27A	108.0	H13A—O13—H13B	104.5
C28—C27—H27B	108.0	H12A—O12—H12B	109.5
H27A—C27—H27B	107.2	Ni1—O10—H10E	110.5
C36—N3—C37	117.0 (4)	Ni1—O10—H10F	109.8
C35—N3—C37	121.8 (6)	H10E—O10—H10F	104.0
C35—N3—C36	121.1 (6)

Hydrogen-bond geometry (Å, º)

D—H···A	D—H	H···A	D···A	D—H···A
N2—H2···O11	1.00	2.18	2.980 (7)	136
O9—H9A···O11	0.93	2.05	2.903 (7)	152
O9—H9B···O6^i	0.93	1.83	2.715 (4)	160
O10′—H10A···O13^ii	0.87	2.03	2.863 (9)	160
O13—H13A···O12^iii	0.87	2.32	2.938 (9)	128
O13—H13B···O6	0.87	1.92	2.758 (5)	161
O12—H12A···O2	0.87	2.02	2.882 (6)	170
O12—H12B···O5^i	0.87	2.09	2.955 (5)	170

Symmetry codes: (i) x−1/2, −y+3/2, −z+1; (ii) x, y, z−1; (iii) x+1/2, −y+3/2, −z+2.

 

Footnotes

‡These authors contributed equally to this work