Eco-friendly microwave synthesis of Mg(II) phenoxy carboxylic acid coordination compounds with specific motifs driven by multiple hydrogen bonding

Abstract

Four novel Mg(II) coordination compounds [Mg(CPA)₂(H₂O)₄] 1, [Mg(H₂O)₆]·(MCPA)₂ 2, [Mg(DCPA)(H₂O)₅]·(DCPA) 3 and [Mg(TCPA)₂(H₂O)₄]·2H₂O 4 were designed by the strategy of hydrogen bonding driven self-assembly (HCPA = 2-chlorophenoxyacetic acid, HMCPA = 2-methyl-4-chlorophenoxyacetic acid, HDCPA = 2,3-dichlorophenoxyacetic acid and HTCPA = 2,4,6-trichlorophenoxyacetic acid). Single-crystal X-ray diffraction demonstrated that each coordination compound had its own specific motif driven by multiple hydrogen bonding: rectangle motif (compound 1), butterfly motif (compound 2), head and tail connected double heart-rectangle motif (compound 3) and alternative one side shared hexagon-rectangle motif (compound 4). The hydrogen bonding effects (O–H⋯Cl and O–H⋯O) on thermostability of the title compounds were well proved by TG analysis. The antibacterial activity of 1–4 has been explored, and all of them showed antibacterial efficiency against colibacillus at certain concentrations. The number of chlorides in the ligands seemed to affect the efficiency of the complexes while their position had no obvious effect. In addition, all the title compounds were evaluated for plant growth regulation by using the lettuce seedling growth bioassay. The results showed that these kinds of compounds had potential to serve as natural plant growth regulators because of their higher allelopathic activity. This research firstly applied the effective microwave method to synthesize Mg(II) phenoxy carboxylic acid coordination compounds. The design strategy of hydrogen-bonding-driven self-assembly presented here gave a novel insight into the further structural prediction and biofunction achievement of Mg(II) supramolecular assemblies.

---

Introduction

Metal–organic supramolecular assemblies (MOSAs) controlled through crystal engineering have emerged as a hot topic of functional materials in application areas like drug delivery,¹ molecular sensing,² light-emitting devices,³ gas storage,⁴ and semiconductors.⁵ Weak interactions such as hydrogen bonding (H-bonding), π–π stacking, and van der Waals force have been used to construct a large number of highly ordered supramolecular architectures.^6–8 The H-bonding, firstly proposed by Moore and Winmill in 1912,⁹ has been proved to be one of the most widely applied noncovalent interactions to assemble supramolecular compounds.^10–12 Although a great variety of H-bonding-driven MOSAs with fascinating structures have been reported so far,^13–15 the prediction of desired H-bonding frameworks still remains a far-reaching challenge because of the broad ranges of H-bonding lengths and angles.¹⁶ To better understand the packing patterns of MOSAs governed by multiple H-bonding, more novel supramolecular compounds with H-bonding interactions should be designed in crystal engineering.

In 2009, Bharadwaj's group synthesized two Mg(II) coordination polymers based on pyridine-2,4,6-tricarboxylic acid (ptcH₃).¹⁷ It is reported that after hydrothermal reaction at 140 °C for 72 h, Mg(NO₃)₂·6H₂O (1 mmol) reacted with ptcH₃ (1 mmol) in 5 mL of water, providing {Mg₃(ptc)₂·8H₂O}_n directly from filtration, and {[Mg(ptc)(H₂O)₂]·1/2[Mg(H₂O)₆]·H₂O}_n by further evaporation of filtrate. We are interested in phenomenon that from the same starting materials, just a little difference in reaction condition, two totally different products were obtained in this paper. It seems that the intricate H-bonding arrayed in the two compounds plays an important role in extending supramolecular assemblies with different structures. However, if we explore these compounds in deeper, we would find that the nature of Mg(II) contributed largely to the H-bonding driven of these kinds of MOSAs. As the smallest group 2 ion with the second highest electronegativity among group elements, Mg(II) has a considerably strong interaction with neutral water molecules, so it forms more stable Mg–OH₂ bonds than other metals.^18,19 The strong affinity between water and Mg(II) causes the coexistence of water molecules (including coordination water or lattice water) and ligands, therefore the competition reaction to win the center metal Mg(II) between the ligands and coordination water molecules occurs in water-containing solvent synthesis. More diverse H-bonding driven MOSAs would be obtained by applying different synthesis methods with various external stimuli such as temperature, pH, or solvent effect to influence the equilibrium.^20–22 In fact, beside the strong Mg–OH₂ interactions facilitated to multiple H-bonding supramolecular assembles, divalent Mg(II) cation also possesses other advantages such as low-cost, non-toxic, low molecular weight, biocompatibility, large abundances in the earth's crust,^23–26 and possible ability to improve the solubility of drugs.^19,27 These merits would greatly improve the functionality of Mg(II) MOSAs and enhance their application in various areas.

With the aim to design functional MOSAs directly governed by H-bonding, we picked up four phenoxyacetic acids of HCPA, HMCPA, HDCPA and HTCPA to construct MOSAs with Mg(II) centers in aqueous solution. The presence of flexible –OCH₂– spacers together with –Cl substituent groups can cooperate with water molecules Mg(II) to facilitate supramolecular assemblies through multiple H-bonding (O–H⋯O or O–H⋯Cl). The various coordination modes of –COOH, even more the changes of the number or position of –Cl in ligands are beneficial to structural diversity in crystal engineering. What's more, the good antimicrobial efficiency²⁸ and herbicidal activity²⁹ of phenoxyacetic acids can provide synergistic effects with metal centre to attain the biofunctionality of the target complexes. In 2001, Kessissoglou's group gave us the first example about the study of antibacterial activity of coordination compounds constructed by the copper(II) centres and a series of phenoxyalkanoic acids. It's interesting that five principle factors such as the chelate effect influenced on the antimicrobial activity of these copper complexes.³⁰ Subsequently, they continued to report the manganese metallacrowns complexes based on auxin herbicides 2,4,5-trichlorophenoxyacetic acid, 2-methyl-4-chlorophenoxyacetic acid, and 2,3-dichlorophenoxyacetic acid. Manganese metallacrown was picked up as the highest antibacterial compound in this literature. The important information obtained from Kessissoglou's group is that phenoxyacetic acids served as a perfect building block to improve the antimicrobial efficiency of the metal coordination compounds. The number and position changes of –Cl substituent groups in ligands seemed to have certain effect on antimicrobial efficiency.³¹ Promoted by mentions above, we are interested in designing novel H-bonding driven magnesium MOSAs functionalized by bioactive phenoxyacetic acids with different –Cl substituent groups and further exploring their application as antimicrobial agents and herbicides. It is expected that more functional H-bonding frameworks could be predicted by making a deeper research on the relationship between structure and bioactivity.

In this work, four novel Mg(II) MOSAs [Mg(CPA)₂(H₂O)₄] 1, [Mg(H₂O)₆]·(MCPA)₂ 2, [Mg(DCPA)(H₂O)₅]·(DCPA) 3 and [Mg(TCPA)₂(H₂O)₄]·2H₂O 4 with specific motifs driven by multiple H-bonding were reported. The synthesis method used here was microwave synthesis, which was firstly applied in Mg(II) phenoxy carboxylic acids coordination compounds, with eco-friendly advantages of shorter reaction time, lower-energy consumption, and higher product yield.^32–37 Using this method instead of the common used hydrothermal synthesis would influence the equilibrium by different external stimuli, which provides more opportunities to achieve structural diverse Mg(II) coordination polymers. The H-bonding effects (O–H⋯Cl and O–H⋯O) influence the thermostability of the title colorless compounds, which was well proved by the TG analysis. The biological activities referring to antimicrobial efficiency and plant growth regulation were also evaluated. The design strategy presented here gave novel insight into the further structural prediction and biofunction achievement of novel Mg(II) MOSAs driven by H-bonding.

Experimental section

Materials and physical measurements

2-Chlorophenol, 4-chloro-2-methylphenol, 2,3-dichlorophenol, 2,4,6-trichlorophenol and ethyl chloroacetate were purchased from Aladdin Reagent Co. Ltd. (Shanghai, China). Microwave synthesis was implemented using a CEM Discover SP single-mode microwave synthesis reactor. Elemental analyses (C and H) were performed on a Vario MACRO cube elemental analyzer. IR spectra was recorded on a BRUKER TENSOR 27 spectrophotometer within 400–4000 cm⁻¹ using the samples prepared as pellets with KBr. TG analysis was performed on a SHIMADZU TA-60ws from 25 to 1000 °C at a heating rate of 10 °C min⁻¹ in N₂ atmosphere. Viability was assessed by absorbance at 600 nm using an Perlong DNM-9062 microplate reader.

Synthesis of the ligand HCPA

The ligands were prepared by the microwave synthesis in the following route (Scheme 1). In a microwave tube, 2-chlorophenol (5 mmol, 642.80 mg) was dissolved in dry N,N-dimethylformamide (2 mL) under stirring followed by addition of 2 mL aqueous solution of K₂CO₃ (691.05 mg, 5 mmol) and KI (166 mg, 1 mmol). The reaction mixture was allowed to irradiate for 4 min at 100 °C under 200 W. Then 5 mL of NaOH (2 M) was added to the mixture and irradiated for 5 min at 500 W. The resulting solution was acidified (pH = 6) by adding of a 1 M HCl solution under vigorous stirring after mixture was cooled to room temperature. Column chromatography provided the title compound as white solids. Yield: 88%. Found: ¹H NMR (DMSO/TMS, 500 MHz, ppm): δ 13.21 (s, 1H), 7.55–7.49 (m, 1H), 7.37 (dd, J = 11.4, 4.4 Hz, 1H), 7.12 (d, J = 8.3 Hz, 1H), 7.06 (t, J = 7.6 Hz, 1H), 4.90 (s, 2H).

Scheme 1 Microwave synthesis of the ligands. HCPA: X₁ = Cl, HMCPA: X₁ = CH₃, X₃ = Cl, HDCPA: X₁ = X₂ = Cl, HTCPA: X₁ = X₃ = X₄ = Cl.

Synthesis of the ligand HMCPA

The procedure was similar to that of HCPA except 4-chloro-2-methylphenol was used instead of 2-chlorophenol. Column chromatography provided white solids. Yield: 86%. Found: ¹H NMR (DMSO/TMS, 500 MHz, ppm): δ 13.14 (s, 1H), 7.32 (d, J = 2.2 Hz, 1H), 7.29–7.22 (m, 1H), 6.93 (d, J = 8.8 Hz, 1H), 4.80 (d, J = 12.1 Hz, 2H), 2.28 (s, 3H).

Synthesis of the ligand HDCPA

The procedure was similar to that of HCPA except 2,3-dichlorophenol was used instead of 2-chlorophenol. Column chromatography provided white solids. Yield: 91%. Found: ¹H NMR (DMSO/TMS, 500 MHz, ppm): δ 13.29 (s, 1H), 7.39 (t, J = 8.2 Hz, 1H), 7.31 (dd, J = 8.1, 1.3 Hz, 1H), 7.13 (dd, J = 8.4, 1.2 Hz, 1H), 4.95 (s, 2H).

Synthesis of the ligand HTCPA

The procedure was similar to that of HCPA except 2,4,6-trichlorophenol was used instead of 2-chlorophenol. Column chromatography provided white solids. Yield: 83%. Found: ¹H NMR (DMSO/TMS, 500 MHz, ppm): δ 13.25 (s, 1H), 7.81 (s, 2H), 4.68 (s, 2H).

Synthesis of [Mg(CPA)₂(H₂O)₄] 1

An appropriate amount of 0.2 mmol NaOH solution was added dropwise to HCPA ethanol solution under stirring to adjust the pH value to approximately 7. Then the MgCl₂·6H₂O (0.1 mmol, 20.33 mg) aqueous solution was added into this mixture under stirring with the molar ratio of Mg²⁺ : HCPA = 1 : 2. Finally, the mixture was transferred into a 25 mL microwave tube. The mixture was heated by microwave at 100 °C for 1 h at 120 W and then cooled to room temperature. The colorless, block crystals suitable for the X-ray diffraction were obtained directly on the filter paper. Anal. calc. for C₁₆H₂₀MgCl₂O₁₀: C, 41.06; H, 4.27%. Found: C, 41.07; H, 4.28%. IR data (cm⁻¹): 3391(b), 1638(s), 1488(s), 1431(s), 1135(s), 1074(s), 685(s).

Synthesis of [Mg(H₂O)₆]·(MCPA)₂ 2

A mixture containing HMCPA (0.2 mmol, 0.04 g), MgCl₂·6H₂O (0.1 mmol, 20.33 mg), NaOH (0.2 mmol, 8 mg) and deionized water (8 mL) was placed in a microwave reaction tube. The mixed solution was sealed and heated at 100 °C under 140 W for 1 h in a microwave reactor. After cooling to room temperature, colorless, hexagon crystals of coordination compound 2 were obtained directly on the filter paper. Anal. calc. for C₁₈H₂₈MgCl₂O₁₂: C, 40.63; H, 5.26%. Found: C, 40.64; H, 5.27%. IR data (cm⁻¹): 3311(b), 1617(s), 1457(s), 1430(s), 1136(s), 1056(s), 652(m).

Synthesis of [Mg(DCPA)(H₂O)₅]·(DCPA) 3

The same synthetic procedure as that for 3 was used except that the HCPA was replaced by HDCPA and the power for microwave reaction is 110 W. Colorless, block crystals of coordination compound 3 were obtained after slowly evaporating the filtrate at room temperature. Anal. calc. for C₁₆H₂₀MgCl₄O₁₁: C, 34.63; H, 3.60%. Found: C, 34.64; H, 3.61%. IR data (cm⁻¹): 3368(b), 1578(s), 1458(s), 1430(s), 1189(s), 1071(s), 698(m).

Synthesis of [Mg(TCPA)₂(H₂O)₄]·2H₂O 4

The procedure was similar to that of 1 except that HTCPA was used instead of HCPA. The power of microwave reaction is 150 W. Colorless, block crystals of coordination compound 4 were obtained after slowly evaporating the filtrate at room temperature. Anal. calc. for C₁₆H₂₀MgCl₆O₁₂: C, 29.94; H, 3.12%. Found: C, 29.93; H, 3.13%. IR data (cm⁻¹): 3279(b), 1734(s), 1709(s), 1461(s), 1425(s), 1141(s), 1084(s), 672(m) (Fig. 1).

Fig. 1 Synthesis of the coordination compounds 1–4.

Single-crystal X-ray structure determinations

All crystallographic data were collected on a Bruker SMART CCD area-detector diffractometer at 298(2) K with graphite monochromatized Mo-Kα radiation (λ = 0.71073 Å) in ω scan mode, and the data reduction was performed using Bruker SAINT. The crystal structures were resolved by direct methods and refined with full-matrix least-squares refinement on F² with anisotropic displacement parameters for non-H atoms using SHELXTL.³⁸ All non-hydrogen atoms and hydrogen atoms were refined geometrically. Structural plots were generated with diamond. A summary of crystallographic data for 1–4 is summarized in Table S2.† Bond lengths (Å) and angles (°) of hydrogen-bond for 1–4 are given in Table S3.† Selected bond lengths and angles for 1–4 are given in Table S4.†

Determination of biological activity

The biological activity was determined against cultures of colibacillus strains. Briefly, the colibacillus suspension (90 μL per well) was prepared in Luria–Bertani culture medium, which was dispensed into 96-well plates. Then the bacteria were dispensed at a volume of 10 μL in triplicate with serial concentrations (2, 4, 6, 8, 10 mg mL⁻¹) of four ligands (HCPA, HMCPA, HDCPA, HTCPA) and coordination compounds 1–4. After incubationed in a carbon dioxide incubator at 37 °C in a 5% CO₂ atmosphere, the optical density (OD) value of each well at 600 nm was measured by a microplate reader after 15 h.

Allelopathic bioassay

The seeds of lettuce (Lactuca sativa) which belongs of herbaceous plants were used for the bioassay. The procedure was conducted according to the reported protocol.³⁹ The lettuce seeds was washed with running water for 2 h, soaked in 0.5% KMnO₄ for 15 min, and flushed until they were colorless. The compounds, positive control and blank solvent acetone, were added to 12-well plates with filter paper to final concentrations of 500, 250, 150 and 50 ppm. After the evaporation of acetone, the lettuce seeds were sown in the microdishes of 12-well plates and irrigated with deionized water. The plates were then incubated at 25 °C for 90 h, and the allelopathic effects [response index (RI)] were calculated according to the following equations:

If T > C, then RI = 1 − C/T; if T < C, then RI = T/C − 1

where T is the length of the treatment, C is the length of the blank control, and RI is the response index. gp stands for glyphosate and ck for the blank.

Results and discussion

The crystal structure of [Mg(CPA)₂(H₂O)₄] 1

Single-crystal X-ray diffraction analysis reveals that 1 crystallizes in monoclinic crystal system, P2(1) space group. The coordination environment and coordination polyhedron of Mg(II) is presented with atom numbering scheme (Fig. 2). The asymmetric unit consists of one crystallographically independent Mg(II) ion, two CPA ligands and four coordinated water molecules. The Mg(II) ion is six-coordinated with a slightly distorted octahedral arrangements, formed by two oxygen atoms of carboxylate groups (O1, O4) from two CPA ligands with a monodentate bridging mode, and four oxygen atoms from coordinated water molecules (O1W, O2W, O3W, O4W). The axial positions of the octahedron are occupied by O1W and O3W with O1W–Mg–O3W bond angles of 179.44(9)°, and the Mg–O bond distance ranging from 2.0363(16) to 2.1048(18) Å. The bond angles and bond distances are similar to the Mg(II) coordination compounds reported previously.⁴⁰

Fig. 2 The coordination environment and coordination polyhedron of Mg(II) ion in 1.

As is shown in Fig. 3, all the carboxylate groups show the same coordination mode of monodentate bridging in 1 and further crosslink the neighboring Mg(II) with coordination water molecules via O1W–H1WB⋯O2 and O3W–H3WB⋯O5 hydrogen bonds, forming the 1D chains with the rectangle-shaped assembly repeat units. The carboxyl group oxygen atoms of O2 and O5 serve as acceptors, and the oxygen atoms of O1W and O3W from coordinated water molecules play a role of donors in the O–H⋯O hydrogen bonds for detail. Notably, the 2D layers have been assembled by countless 1D chains, such as the 2D layer on ac plane extended by the parallel blue, yellow, and green 1D chains along a axis in the Fig. 4.

Fig. 3 A view of the 1D chain along a axis in 1.

Fig. 4 A view of the 2D layer on ac plane in 1.

The crystal structure of [Mg(H₂O)₆]·(MCPA)₂ 2

X-ray diffraction reveals that 2 is a co-crystal structure and crystallizes in orthorhombic space group C2/c. The asymmetric unit contains one crystallographically independent Mg(II) ion, six coordinated water molecules, and two uncoordinated MCPA ligands. One Mg(II) ion and six water molecules (O1W, O1WA, O2W, O2WA, O3W, O3WA) form the Mg(II)–water clusters [Mg(H₂O)₆]²⁺ with the Mg–O distance ranging from 2.031(4) to 2.073(3) Å (Fig. 5), which are consistent with the literature reported.⁴¹ The co-crystal structure formed by MCPA and [Mg(H₂O)₆]²⁺ is in ratio of 2 : 1. To the best of our knowledge, such a co-crystal structure involving both Mg(II)–water clusters and phenoxyacetic acid has not been reported.

Fig. 5 The coordination environment and coordination polyhedron of Mg(II) ion in 2.

Each of Mg(II)–water cluster in 2 further connected the neighboring deprotonated MCPA ligands by multiple hydrogen-bond to generate a neutral 1D chain along the c axis with a non-bonding Mg⋯Mg distance of 7.460 Å. The most striking feature of 1D chain is butterfly-shaped metal–ligand extending crosslinked by hydrogen-bonds, while O2W–H2WB⋯O3 and O3W–H3WA⋯O3 form a pair of bigger wings, O1W–H1WB⋯O1 and O3W–H3WB⋯O2 form the two smaller wings (Fig. 6). Furthermore, adjacent chains stack together to form 2D layers, as is shown in Fig. 7, three parallel chains such as the blue, yellow, and green chains self-assembled to the 2D layer on ac plane.

Fig. 6 A view of the 1D chain along c axis in 2.

Fig. 7 A view of the 2D layer on ac plane in 2.

The crystal structure of [Mg(DCPA)(H₂O)₅]·(DCPA) 3

Coordination compound 3 crystallizes in the triclinic crystal system with the space group of P1̄. There are one crystallographically independent Mg(II) ion, one DCPA ligand, five coordinated water molecules and one uncoordinated ligand in the asymmetric unit. The coordination environment and coordination polyhedron of Mg(II) ion in 3 is shown in Fig. 8. The Mg(II) ions are also six-coordinated with a distorted octahedron geometry where the axial positions were occupied by O3W and O1W, and Mg(II) ions were coordinated by one oxygen atom from one monodentate carboxylate group (O1) and five oxygen atoms (O1W, O2W, O3W, O4W, O5W) from the coordinated water molecules. Bond distances around the central Mg(II) ions vary from 2.035(5) to 2.098(6) Å, and the average distance is 2.074 Å which is in agreement with other coordination compounds containing Mg(II) centers.⁴²

Fig. 8 The coordination environment and coordination polyhedron of Mg(II) ion in 3.

An interesting structural feature for 3 is the formation of 1D chain by hydrogen-bond interactions involving the uncoordinated DCPA molecules and coordinated water molecules. One DCPA ligand act as monodentate bridging ligand (Fig. 9, the yellow part) to coordinate with the Mg(II) center, while another DCPA ligand (Fig. 9, the dark blue part) has no coordination interactions with metals. The uncoordinated DCPA molecules play an important role of balancing the charge, and they also bridge the neighboring Mg(II) by the formation of hydrogen-bonds (O2W–H2WB⋯O4, O3W–H3WA⋯O4, O3W–H3WB⋯O4W) between coordinated water molecules and carboxyl oxygen atoms. It's worth noting that, coordinated water molecules (O2W) can donate free hydrogen atoms (H2WA) to chlorine atoms of DCPA (the o-position chlorine atoms), then form O2W–H2WA⋯Cl3 hydrogen-bond with a distance of 3.472 Å, which has not been found in 1 and 2. The metal–ligand extending of 1D chain is head and tail connected double heart-rectangle-shaped assembly consisted of two Mg(II) centers, a part of uncoordinated DCPA, hydrogen-bonds of O–H⋯O and O–H⋯Cl in the red ring of Fig. 9. Adjacent two parallel 1D chains are further held together to form a 2D layer, as is shown in Fig. 10, two parallel chains such as the green, yellow chains inserted with the uncoordinated DCPA self-assembled to the 2D layer on ab plane.

Fig. 9 A view of the 1D chain in 3.

Fig. 10 A view of the 2D layer on ab plane in 3.

The crystal structure of [Mg(TCPA)₂(H₂O)₄]·2H₂O 4

Coordination compound 4 crystallizes in the monoclinic space group P2(1)/c. The asymmetric unit contains one crystallographically independent Mg(II) ion, two TCPA ligands, four coordinated water molecules and two lattice water molecules. The Mg(II) ions are six-coordinated by monodentate carboxylate groups (O4, O1) and four oxygen atoms (O1W, O2W, O3W, O4W) from the coordinated water molecules. The average Mg–O bond distance is 2.076 Å as literature reported,⁴³ and the central Mg(II) ion shows a distorted octahedron geometry where the axial positions occupied by O3W and O2W (Fig. 11). The carboxylic groups of TCPA are totally deprotonated and displayed the monodentate bridging coordination modes.

Fig. 11 The coordination environment and coordination polyhedron of Mg(II) ion in 4.

The neighboring [Mg(TCPA)₂(H₂O)₄]_n repeat units further connected with each other by hydrogen-bond (O1W–H1WA⋯O2, O1W–H1WB⋯O3W, O4W–H4WB⋯O1, O4W–H4WB⋯O1W) to generate a neutral 1D chain with the feature of one side shared hexagon-rectangle-shaped extending along the a axis in Fig. 12. Moreover, the 1D chains such as the green, yellow and blue one further extend into the stable 2D layer represented in Fig. 13.

Fig. 12 A view showing the 1D chain along the a axis in 4.

Fig. 13 A view of the 2D layer in 4.

Comparison of Mg(II) coordination compound 1–4 structures

First of all, all of the Mg(II) ions in 1–4 are six-coordinated and show distorted octahedron geometry. Secondly, as shown in molecule formulas [Mg(CPA)₂(H₂O)₄] 1, [Mg(H₂O)₆]·(MCPA)₂ 2, [Mg(DCPA)(H₂O)₅]·(DCPA) 3 and [Mg(TCPA)₂(H₂O)₄]·2H₂O 4, both of HCPA and HTCPA ligands have coordination effect with Mg(II) centers in 1 and 4, while HMCPA ligand in 2 has no coordination effect with Mg(II) centers, only electrostatic force could be found among [Mg(H₂O)₆]²⁺ units and MCPA units. Moreover, HDCPA ligand in 3 produces not only the coordination style but also the lattice style. It could be noticed that whether the ligands coordinate to the metal centers or not, all of them have the H-bonding effects with water molecules, which are beneficial to the H-bonding-driven MOSAs. The O–H⋯O hydrogen-bonds exist in all the title compounds, however, only compound 3 has O–H⋯Cl hydrogen-bond, the existence of uncoordinated ligands HDCPA in 3 provided o-position chlorine atoms as acceptor to form the O–H⋯Cl hydrogen-bond with coordinated water molecules, which achieves the H-bonding controlling by perfect steric effects. Finally, four different supramolecular 1D chains with specific motifs (Fig. 14) were found in 1–4. They are the rectangle-shaped motif for 1, butterfly-shaped motif for 2, head and tail connected double heart-rectangle-shaped motif for 3 and one side shared hexagon-rectangle-shaped motif for 4. All the motifs are governed by multiple H-bonding, which displays successful controlling of H-bonding driven MOSAs in this work.

Fig. 14 The different metal–ligand extending of 1D chains in 1–4.

Thermogravimetric analysis of coordination compounds 1–4

The TG analysis was performed in N₂ atmosphere at a heating rate of 10 °C min⁻¹ on powder samples of 1–4, and the TG curves are shown in Fig. 15. The TG results reveal that all of the four compounds have three weight-loss stages. The first step can be attributed to the removal of coordinated water and lattice water molecules. Upon further heating, the following two weight losses can be ascribed to the collapse of the frameworks. The remaining residue corresponds to formation of MgO. The data of thermal analysis for 1–4 are given in Table S1.† The final decomposition temperature is in the order of 730.86 °C for 3 > 706.18 °C for 4 > 581.33 °C for 1 > 569.61 °C for 2, which corresponds to the thermostability of 3 > 4 > 1 > 2. These results agree quite well with the structural feature. From the structural analysis above, we can clearly see that even the O–H⋯O hydrogen-bonds exist in all the four title compounds, the O–H⋯Cl hydrogen-bonds only coexist with O–H⋯O hydrogen-bonds in compound 3, resulting in the more stable structure of 3 than 1, 2 and 4. The O–H⋯O hydrogen-bonds in 4 are more complex than 1, so the thermostability of 4 is higher than 1. Mg(II) ions in 2 have not coordinated with MCPA ligand, and the weak interaction forces between them resulting in the lowest thermostability of 2.

Fig. 15 TG curves of the coordination compounds 1–4.

Antibacterial activity

Antibacterial assays against colibacillus showed that the four coordination compounds 1–4 exhibited a moderate activity. As is shown in Fig. 16, the x-axis represents the concentration of drugs range from 2–10 mg mL⁻¹. The y-axis represents OD value while the shorter bars correspond to higher activity. The z-axis represents ligands and their coordination compounds use eight different colours. The results are listed as following: (1) both ligands and coordination compounds have certain antibacterial activity against colibacillus, with the obvious rising trend along with the increase of drug's concentration. (2) The antibacterial efficiency of coordination compound 4 is superior to HTCPA ligand, while this rule is not clear in other three groups. (3) The total antibacterial efficiency order for 1–4 and the corresponding ligands is: 4 > HTCPA > HDCPA > 3 > HMCPA > 2 > HCPA > 1, which reveals that the number of the chlorides in the ligands seemed to affect the efficiency of the complexes while their position had no obvious effect on it.

Fig. 16 Antibacterial activity of the coordination compounds 1–4 and corresponding ligands.

Evaluation of allelopathic activity

All isolated compounds were evaluated for allelopathic activity against lettuce (Lactuca sativa) seeds by determining the seedling growth (root elongation) with respect to the control of glyphosate, which is a broad-spectrum systemic herbicide. The RI was selected as an evaluation indicator, which ranges from −1 to 1, with positive values indicating stimulation by the treatments and negative values indicating inhibition by them. As shown in Table 1, all of the RI values for the ligands and compounds 1–4 (RI values range of −0.42 to −0.99) were negative, which showed higher inhibit ability referencing to the positive control glyphosate (RI values of −0.40 at 50 ppm and −0.90 at 500 ppm). For compound 1, at high concentrations, it can cause significant roots elongation which is almost equal to HCPA ligands. Compound 2 showed the prominent inhibition of shoot root elongation (−0.81 to −0.91) which is equivalent to HMCPA ligand. Compounds 3 and 4 have obvious inhibiting ability for lettuce, which are higher than their ligands. These results revealed that compound 1–4 could be identified as potential natural plant growth regulator.

Table 1 Allelopathic effects on lettuce (Lactuca sativa)

Ligands	50 ppm	150 ppm	250 ppm	500 ppm	Compounds	50 ppm	150 ppm	250 ppm	500 ppm
CPA	−0.52	−0.68	−0.73	−0.89	1	−0.42	−0.55	−0.75	−0.90
MCPA	−0.81	−0.91	−0.91	−0.92	2	−0.81	−0.87	−0.89	−0.91
DCPA	−0.62	−0.82	−0.9	−0.92	3	−0.86	−0.9	−0.95	−0.99
TCPA	−0.62	−0.73	−0.78	−0.89	4	−0.91	−0.92	−0.94	−0.96
gp	−0.4	−0.5	−0.64	−0.9	ck	0	−0.02	−0.05	−0.06

---

Conclusions

Four novel Mg(II) MOSAs based on phenoxyacetic acids were designed by the strategy of H-bonding driven self-assembly. Specific motifs driven by multiple H-bonding dominate greatly to the formation of 1D chains in 1–4, that is: the rectangle-shaped motif in 1, the butterfly-shaped motif in 2, the head and tail connected double heart-rectangle-shaped motif in 3 and the one side shared hexagon-rectangle-shaped motif appearing alternatively in 4. The co-crystal structure involving both Mg(II)–water clusters [Mg(H₂O)₆]²⁺ and MCPA in 2 was firstly reported in this work. From the structural analysis, it can be clearly seen that even the O–H⋯O hydrogen-bonds exist in all the four title compounds, the O–H⋯Cl hydrogen-bonds only coexist with O–H⋯O hydrogen-bonds in compound 3, these H-bonding (O–H⋯Cl and O–H⋯O) influence the thermostability of the title compounds, which is well proved by the TG analysis. Moreover, the total antibacterial efficiency order against colibacillus for 1–4 and the corresponding ligands is: 4 > HTCPA > HDCPA > 3 > HMCPA > 2 > HCPA > 1, which would be affected by the number of the chlorides in the ligands. Allelopathic bioassay test proved that compound 1–4 can be identified as potential natural plant growth regulators because they can inhibit the growth of lettuce (Lactuca sativa). Notably, the synthesis method used here was eco-friendly microwave synthesis, which was for the first applied in Mg(II) phenoxy carboxylic acid coordination compounds. Considering that more metal ions can be used by this method, the present microwave synthesis could be further applied in construction transition metal and rare earth metal coordination compounds. The design and synthetic strategies in this report presented a novel insight into the further structural prediction of novel Mg(II) supramolecular assemblies driven by H-bonding.

Acknowledgements

The authors gratefully acknowledge the financial supports from the National Natural Science Foundation of China (Grant No. 21473136, Grant No. 21103140), and the PhD. Programs Foundation of Ministry of Education of China (Grant No. 20110204120037).

Notes and references

1. A. Mishra, S. C. Kang and K. Chi, Eur. J. Inorg. Chem., 2013, 2013, 5222–5232.
2. D. K. Singha, P. Majee, S. K. Mondal and P. Mahata, Eur. J. Inorg. Chem., 2015, 2015, 1390–1397.
3. Y. E. Cha, X. Li, D. Ma and R. Huo, Eur. J. Inorg. Chem., 2014, 2014, 2969–2975.
4. I. Marin, C. Turta, A. C. Benniston, R. W. Harrington and W. Clegg, Eur. J. Inorg. Chem., 2015, 2015, 786–793.
5. T. Terauchi, S. Sumi, Y. Kobayashi, Y. Matsushita and A. Sato, Cryst. Growth Des., 2014, 14, 1412–1418.
6. I. Boldog, J. Daran, A. N. Chernega, E. B. Rusanov, H. Krautscheid and K. V. Domasevitch, Cryst. Growth Des., 2009, 9, 2895–2905.
7. S. Z. Zhu, C. H. Xing, W. Xu, G. F. Jin and Z. T. Li, Cryst. Growth Des., 2004, 4, 53–56.
8. Z. N. Wu, J. L. Liu, Y. C. Li, Z. Y. Cheng, T. T. Li, H. Zhang, Z. Y. Lu and B. Yang, ACS Nano, 2015, 9, 6315–6323.
9. T. S. Moore and T. F. Winmill, J. Chem. Soc., 1912, 101, 1635–1676.
10. L. L. Yang, X. T. Tan, Z. Q. Wang and X. Zhang, Chem. Rev., 2015, 115, 7196–7239.
11. J. Liu, M. D. Morris, F. C. Macazo, L. R. Schoukroun-Barnes and R. J. White, J. Electroanal. Chem., 2014, 161, H301–H313.
12. J. Liu, S. Wagan, M. Dávila Morris, J. Taylor and R. J. White, Anal. Chem., 2014, 86, 11417–11424.
13. S. Varughese and V. R. Pedireddi, Chem. Commun., 2005, 1824–1826.
14. S. Z. Zhu, C. C. Xing, W. Xu, G. F. Jin and Z. T. Li, Cryst. Growth Des., 2004, 4, 53–56.
15. L. Vella-Zarb, R. E. Dinnebier and U. Baisch, Cryst. Growth Des., 2013, 13, 4402–4410.
16. J. Lü, L. W. Han, N. H. Alsmail, A. J. Blake, W. Lewis, R. Cao and M. Schröder, Cryst. Growth Des., 2015, 15, 4219–4224.
17. M. C. Das, S. K. Ghosh, E. C. Sañudo and P. K. Bharadwaj, Dalton Trans., 2009, 9, 1644–1658.
18. A. M. Jones and M. P. Lord, Macmillan's Chemical and Physical Data, Macmillan Press, London, 1992, p. 486.
19. J. Arlin, A. J. Florence, A. Johnston, A. R. Kennedy, G. J. Miller and K. Patterson, Cryst. Growth Des., 2011, 11, 1318–1327.
20. D. Maier and A. B. Kulakov, Cryst. Growth Des., 2005, 5, 1751–1754.
21. B. P. Ziemba, C. Pilling, V. Calleja, B. Larijani and J. J. Falke, Biochemistry, 2013, 52, 4820–4829.
22. S. Narra and S. Shigeto, J. Phys. Chem. B, 2015, 119, 3808–3814.
23. Z. F. Wu, B. Tan, C. F. Du, M. L. Feng, Z. L. Xie and X. Y. Huang, CrystEngComm, 2015, 17, 4288–4292.
24. A. M. Kirillov, Y. Y. Karabach, M. V. Kirillova, M. Haukka and A. J. L. Pombeiro, Cryst. Growth Des., 2012, 12, 1069–1074.
25. D. Rankine, T. D. Keene, C. J. Sumby and C. J. Doonan, CrystEngComm, 2013, 15, 9722–9728.
26. K. M. N. Burgess, Y. Xu, M. C. Leclerc and D. L. Bryce, J. Phys. Chem. A, 2013, 117, 6561–6570.
27. H. N. Wang, X. Meng, X. L. Wang, G. S. Yang and Z. M. Su, Dalton Trans., 2012, 41, 2231–2233.
28. R. H. Müller, S. Kleinsteuber and W. Babel, Microbiol. Res., 2001, 156, 121–131.
29. H. W. He, H. Peng, T. Wang, C. B. Wang, J. L. Yuan, T. Chen, J. He and X. Tan, J. Agric. Food Chem., 2013, 61, 2479–2488.
30. C. Dendrinou-Samara, G. Psomas, C. P. Raptopoulou and D. P. Kessissoglou, J. Inorg. Biochem., 2001, 83, 7–16.
31. T. Afrati, A. A. Pantazaki, C. Dendrinou-Samara, C. Raptopoulou, A. Terzis and D. P. Kessissoglou, Dalton Trans., 2010, 39, 765–775.
32. G. K. Kantar, N. Baltaş, E. Menteşe and S. Şaşmaz, J. Organomet. Chem., 2015, 787, 8–13.
33. X. Chen, L. Li, T. T. Yi, W. Z. Zhang, X. L. Zhang and L. L. Wang, J. Solid State Chem., 2015, 229, 141–149.
34. N. Barbero, C. Magistris, J. Park, D. Saccone, P. Quagliotto, R. Buscaino, C. Medana, C. Barolo and G. Viscardi, Org. Lett., 2015, 17, 3306–3309.
35. F. Conrad, Y. Zhou, M. Yulikov, K. Hametner, S. Weyeneth, G. Jeschke, D. Günther, J. Grunwaldt and G. R. Patzke, Eur. J. Inorg. Chem., 2010, 2010, 2036–2043.
36. E. G. Duruk, H. Y. Yenilmez, A. Altindal and B. Z. Altuntas, Dalton Trans., 2015, 44, 10060–10068.
37. M. C. Bagley, R. Lunn and X. Xiong, Tetrahedron Lett., 2002, 43, 8331.
38. G. M. Sheldrick, Acta Crystallogr., Sect. E: Struct. Rep. Online, 2008, 64, 112.
39. İ. Gülçin, J. Med. Food, 2011, 14, 975–985.
40. S. Gao, J. Li, L. Huo, J. Liu and C. Gu, Acta Crystallogr., Sect. E: Struct. Rep. Online, 2004, 60, m100–m101.
41. J. Liu, S. Gao, L. Huo, Y. Dong and H. Zhao, Acta Crystallogr., Sect. E: Struct. Rep. Online, 2004, 60, m845–m847.
42. G. Smith, Acta Crystallogr., Sect. E: Struct. Rep. Online, 2012, 68, m1178.
43. A. Shafir, D. Fiedler and J. Arnold, Dalton Trans., 2002, 555–560.

---

Footnote

† Electronic supplementary information (ESI) available: X-ray crystallographic files of 1–4 in CIF format. A summary of crystallographic data for 1–4. Additional selected bond lengths and angles for 1–4. Bond lengths (Å) and angles (°) of hydrogen-bond for 1–4 IR and NMR spectra of 1–4. CCDC number 1411705 for 1, 1411706 for 2, 1411707 for 3 and 1411665 for 4. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c6ra09589e

---