Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2024 Jul 18;63(30):14216–14230. doi: 10.1021/acs.inorgchem.4c02386

Supramolecular Binding of Phosphonate Dianions by Nanojars and Nanojar Clamshells

Pooja Singh †, Matthias Zeller ‡, Gellert Mezei †,*
PMCID: PMC11289757  PMID: 39023277

Abstract

graphic file with name ic4c02386_0015.jpg

Despite the widespread use of phosphonates (RPO32–) in various agricultural, industrial, and household applications and the ensuing eutrophication of polluted water bodies, the capture of phosphonate ions by molecular receptors has been scarcely studied. Herein, we describe a novel approach to phosphonate binding using chemically and thermally robust supramolecular coordination assemblies of the formula [RPO3⊂{cis-CuII(μ-OH)(μ-pz)}n]2– (Cun; n = 27–31; pz = pyrazolate ion, C3H3N2–; R = aliphatic or aromatic group). The neutral receptors, termed nanojars, strongly bind phosphonate anions by a multitude of hydrogen bonds within their highly hydrophilic cavities. These nanojars can be synthesized either directly from their constituents or by depolymerization of [trans-CuII(μ-OH)(μ-pz)]∞ induced by phosphonate anions. Electrospray-ionization mass spectrometry, UV–vis and variable-temperature, paramagnetic 1H and 31P NMR spectroscopy, single-crystal X-ray diffraction, along with chemical stability studies toward NH3 and Ba2+ ions, and thermal stability studies in solution are employed to explore the binding of various phosphonate ions by nanojars. Crystallographic studies of 12 different nanojars offer unprecedented structural characterization of host–guest complexes with doubly charged RPO32– ions and reveal a new motif in nanojar chemistry, nanojar clamshells, which consist of phosphonate anion-bridged pairs of nanojars and double the phosphonate-binding capacity of nanojars.

Short abstract

The binding of phosphonate anions bearing different aliphatic and aromatic groups by nanojars and nanojar clamshells has been investigated using mass spectrometry, UV−vis and variable-temperature, paramagnetic 1H and 31P NMR spectroscopy, X-ray crystallography, and chemical stability studies.

Introduction

Phosphonates (RPO32–) are organophosphorus compounds derived from the phosphate ion (PO43–) by replacement of one O atom with an organic group (R).1 They can also be viewed as derivatives of deprotonated phosphonic (phosphorous) acid, HPO32–, although the oxidation state of the phosphorus atom in the latter is 3+ as opposed to 5+ in phosphonates and phosphates. Phosphonates are encountered in nature and are involved in physiological processes, the global phosphorus cycle and the biogeochemical generation of methane.2,3 A large variety of industrial, agricultural and household activities also employ phosphonates, in applications including pesticides (e.g., glyphosate), plant growth regulators, detergents, bleach stabilizers, water softening and desalination agents, corrosion inhibitors, antiscalants, concrete retarders, metal extracting agents, as well as inhibitors of enzymes that utilize phosphate or pyrophosphate as a substrate (treatment of osteoporosis, antibiotic and antiviral medications).4−8 Phosphonates are also actively studied for different novel applications, such as porous materials for gas storage and catalysis, magnetic and luminescent materials, proton conduction, and prodrugs.9−13 Phosphonate esters, such as dimethyl methylphosphonate, are used as flame retardants among various other applications.14

Phosphonates are less hydrophilic than phosphate. Nevertheless, they are water-soluble, and despite their low toxicity to aquatic organisms,15 their accumulation in water bodies has harmful effects on the environment and contributes to eutrophication.16,17 Therefore, the removal of phosphonates from aqueous environments is desirable.18,19 Furthermore, phosphorus is a strategic element (phosphate rock is a nonrenewable, finite resource), and its recovery is becoming increasingly important.20−23 While several supramolecular receptors for phosphates have been developed,24−28 the supramolecular binding of phosphonates is much less studied and only a few receptors are known. These include a tripodal 8-aminochromenone-2-carboxamide-based receptor,29 a generation 5 poly(amidoamine) (PAMAM) dendrimer-based sensing array30 and fluorescent sensors based on a tripodal receptor with H-bond donor thiourea and/or pyrrole-2-yl-amide appendages.31 Receptors for singly charged hydrogenphosphonates, including bis(hydrogenphosphonates) based on aminocoumarin,32 anthracene- and 1,8-diphenylnaphthalene-diamidine or -diguanidine,33 and cyanostar macrocycles have also been reported,34 as well as receptors for neutral phosphonate esters.35

Structural characterization of supramolecular host–guest complexes of monophosphonates is even scarcer. In fact, no crystal structure of a supramolecular (without metal-phosphonate coordination) host–guest complex of a fully deprotonated, simple phosphonate anion is known to date. The only reported crystal structure of a monophosphonate supramolecular host–guest complex in the Cambridge Structural Database (CSD)36 is for protonated phenylphosphonate in an amide-functionalized cyclodextrin receptor.37

We recently provided the first examples of supramolecular binding of phosphite (HPO32–) based exclusively on H-bonding, using copper pyrazolate/hydroxide coordination complexes termed nanojars, [HPO3⊂{cis-CuII(μ-OH)(μ-pz)}n]2– (CunHPO3; n = 27–32), as receptors.38 Herein, we describe the synthesis of phosphonate nanojars with the formula [RPO3⊂{cis-CuII(μ-OH)(μ-pz)}n]2– (CunRPO3; n = 27–31; R = Me, Et, nPr, nBu, n-dodecyl (nC12), Bn, Ph; Figure 1) and report the first structural characterizations of monophosphonate dianions bound to a neutral supramolecular receptor exclusively by H-bonding, based on single-crystal X-ray diffraction. The solid-state crystallographic studies are complemented by solution-phase electrospray ionization mass spectrometric (ESI-MS), variable-temperature (VT), paramagnetic nuclear magnetic resonance (1H and 31P NMR), and UV–vis spectroscopic studies, along with chemical stability studies toward NH3 and Ba2+ ions and thermal stability studies in solution. We also report the crystal structures of the first nanojar clamshells, a novel motif in nanojar chemistry, which consist of two nanojar units tethered by two adjacent phosphonate units.

Figure 1.

Figure 1

Open in a new tab

Ball-and-stick representation of the structure of methylphosphonate (MePO32–), ethylphosphonate (EtPO32–), n-propylphosphonate (nPrPO32–), n-butylphosphonate (nBuPO32–), n-dodecylphosphonate (nC12PO32–), benzylphosphonate (BnPO32–), and phenylphosphonate (PhPO32–) anions. Color code: red, O; purple, P; orange, C; white, H.

Results and Discussion

Synthesis and Mass Spectrometric Studies

Two different methods were employed for the synthesis of phosphonate nanojars. The first method involves a direct synthesis using Cu(NO3)2·2.5H2O, pyrazole, NaOH, Bu4NOH, and phosphonic acid (H2RPO3; R = Me, Et, nPr, nBu, tBu, n-dodecyl, Bn, Ph) in a 1:1:4:1 molar ratio in tetrahydrofuran (THF) at ambient temperature under an N2 atmosphere. Except in the case of R = tBu, a mixture of nanojars of the formula (Bu4N)2[RPO3⊂{cis-CuII(μ-OH)(μ-pz)}n] (CunRPO3; n = 27–31) was obtained, with varying distributions of the different nanojar sizes. ESI-MS of these mixtures indicates that varying amounts of carbonate nanojars, (Bu4N)2[CO3⊂{cis-CuII(μ-OH)(μ-pz)}n] (CunCO3; n = 27–31) are also present, except in the case of the nanojars with methylphosphonate. This is due to the fact that nanojars prefer to bind small anions with large hydration energies, such as CO32– (ΔGh° = −1324 kJ/mol), over larger, less hydrophilic anions. Although the reactions were carried out under a CO2-free atmosphere, small amounts of carbonate are present in the bases (Bu4NOH or NaOH) used for the synthesis. The use of excess phosphonate anion to minimize the interference of carbonate worked only in the case of the phosphonate with the smallest R substituent. Therefore, a different method was chosen for the synthesis of carbonate-free nanojars with larger phosphonates. This second method, which requires no base, relies on the depolymerization of [trans-CuII(μ-OH)(μ-pz)]∞ in the presence of (Bu4N)2RPO3 in refluxing toluene (Figure 2). No reaction was observed with (Bu4N)2tBuPO3. In contrast with nanojars based on other anions, the CunRPO3 mixtures are not affected by NH3 treatment (which converts CunCO3 to Cu27CO3 and CunSO4 to Cu31SO4).39

Figure 2.

Figure 2

Open in a new tab

Schematic representation of the depolymerization of [trans-Cu(OH)(pz)]∞ into phosphonate-incarcerating nanojars [RPO3⊂{cis-Cu(OH)(pz)}n]2– (R = Me, Et, nPr, nBu, nC12, Bn, Ph; n = 27–31). Color code for n = 31 shown: persimmon, Cu8 ring; green, Cu14 ring; cyan, Cu9 ring.

Figure 3 shows the ESI-MS(−) spectra of the nanojar mixtures obtained with different phosphonates with the corresponding m/z values. Occasionally, substituted nanojar derivatives also form. For example, phosphonate-substituted species (phosphonate substituting one OH and one pz ligand) are present in the case of the n-butyl-, n-dodecyl-, and phenylphosphonate anions, most prominently in the case of the n-dodecylphosphonate nanojars: [nC12PO3⊂{Cun(OH)n−y(pz)n−y(nC12PO3)y}]2– (n = 27: y = 1, m/z 2199; y = 2, m/z 2281; n = 29: y = 1, m/z 2347). In the case of the phenylphosphonate nanojars, different phosphonate-substituted species are also observed (phosphonate substituting two OH ligands), [PhPO3⊂{Cun(OH)n−2y(pz)n(PhPO3)y}]2– (n = 27: y = 1, m/z 2155; y = 2, m/z 2232; n = 29: y = 1, m/z 2347). It is noteworthy that despite the large excess (∼30-fold) of phosphonates used during synthesis, only minor amounts of phosphonate-substituted nanojars were obtained, although phosphonates are good ligands for copper and several multinuclear copper(II) phosphonate/pyrazolate complexes are known.40 Small amounts of formate-substituted analogues, in which one or more pz– moieties are substituted by HCOO–, are also observed in most spectra at 11 m/z units less than the parent peak. Nanojars are extremely sensitive to even traces of formic acid, which is either present in the mass spectrometer due to its common use as an additive to improve peak shapes and to promote ionization by producing [M + H]+ ions or could result from the degradation of formaldehyde-based resins used in vial caps.41,42

Figure 3.

Figure 3

Open in a new tab

ESI-MS(−) spectra in CH3CN of the phosphonate-incarcerating nanojar mixtures [RPO3⊂{Cu(OH)(pz)}n]2– (CunRPO3; n = 27–31; R = Me, Et, nPr, nBu, nC12, Bn, Ph) obtained by depolymerization (except uppermost left: CunMePO3 obtained by direct reaction). Detailed isotopic distributions are shown in Figure S1. Nanojar derivatives with pairs of pz– and OH– moieties or pairs of two OH– moieties substituted by phosphonate and with one or more pz– moieties substituted by HCOO– are indicated by blue, green, and black asterisks, respectively.

Structural Analysis by X-ray Crystallography

Nanojars consist of three stacked [cis-CuII(OH)(pz)]x metallamacrocyclic rings, with one larger central ring (x = 12–14) and two smaller side rings (x = 6–10). The rings are held together by axial Cu···O interactions and hydrogen bonds between the central ring and the side rings and by hydrogen bonds between the entrapped anion and the two side rings. The anion guest appears to act as an essential “glue” between the three rings, as an empty nanojar host has never been detected.

X-ray diffraction-quality single crystals of nanojars are difficult to obtain. Due to their toroidal shape, the close-packing of nanojars in 3-D lattices leaves large voids, which are only partially filled by the Bu4N+ counterions. The remaining voids accommodate several, usually highly disordered, solvent molecules, which are often disordered with the counterions as well. To successfully grow single crystals of the phosphonate nanojars, various uncommon solvents including 1-methylnaphthalene, methoxybenzene (anisole), and different isomers of dimethoxybenzene as the nanojar-dissolving solvent and different isomers of butanol as the precipitating solvent were considered in addition to more frequently used ones. Out of the over 1.3 million structures in the CSD only 83 contain anisole, 13 contain 1,3-dimethoxybenzene, 31 contain 1-methylnaphthalene, 224 contain n-butanol, and 107 contain t-butanol as a solvate molecule, compared to 18450 structures with toluene and 29266 with methanol.36 Crystal growing efforts using various combinations of solvents led to 12 crystal structures with the different phosphonate anions, including Cu6+12+9, Cu6+12+10, Cu7+13+9, and Cu8+14+9 nanojars: (Bu4N)2[MePO3⊂{Cu(OH)(pz)}6+12+9] (1, from 1,2-dichlorobenzene/n-heptane), (Bu4N)2[MePO3⊂{Cu(OH)(pz)}7+13+9] (2, from 1,2-dichlorobenzene/n-heptane), (Bu4N)2[MePO3⊂{Cu(OH)(pz)}8+14+9] (3, from anisole/n-butanol), (Bu4N)2[EtPO3⊂{Cu(OH)(pz)}8+14+9] (4, from 1-methylnaphthalene/hexanes), (Bu4N)2[EtPO3⊂{Cu(OH)(pz)}8+14+9]0.95[EtPO3⊂{Cu(OH)(pz)}7+13+9]0.05 (5, from anisole/n-butanol), (Bu4N)2[nBuPO3⊂{Cu(OH)(pz)}7+13+9] (6, from chlorobenzene/n-heptane), (Bu4N)2[nC12PO3⊂{Cu(OH)(pz)}8+14+9] (7, from 1,2-dichlorobenzene/n-heptane), (Bu4N)2[BnPO3⊂{Cu(OH)(pz)}8+14+9] (8, from 1,2-dichlorobenzene/n-heptane), (Bu4N)1.53(Bn3MeN)0.47[PhPO3⊂{Cu(OH)(pz)}6+12+10]0.87[PhPO3⊂{Cu(OH)(pz)}6+12+9]0.13 (9, from 1,3-dimethoxybenzene/n-heptane), (Bu4N)2[PhPO3⊂{Cu(OH)(pz)}8+14+9] (10, from 1,3-dimethoxybenzene/n-heptane), (Bu4N)4[EtPO3⊂{cis-CuII(μ-OH)27(μ-pz)25(μ4-EtPO3)}]2 (11, from 1-methylnaphthalene/t-butanol), and (Bu4N)4[nPrPO3⊂{cis-CuII(μ-OH)27(μ-pz)25(μ4-nPrPO3)}]2 (12, from anisole/t-butanol). Structure 9 contains an additional counterion, Bn3MeN+, which was used as an additive in the form of its nitrate salt during crystallization. In all structures, the nanojar unit is located in a general position within the crystal lattice. Tables S1−S53 and Figures S2−S13 (thermal ellipsoid plots) contain details of the structures.

With MePO32–, three different nanojar sizes, Cu27MePO3 (1), Cu29MePO3 (2), and Cu31MePO3 (3) were crystallized, with no disorder in the position of the bound anion (Figure 4). The three structures have a Cu9 side ring in common, with Cu12, Cu13, and Cu14 rings as the central ring and Cu6, Cu7, and Cu8 rings as the other side ring. In all three structures, the methyl group of the MePO32– anion points outward through the Cu9 ring, while the three O atoms point toward the Cu6, Cu7, or Cu8 rings. In the case of the Cu6+12+9 nanojar, this is in stark contrast with the orientation of the SO42– anion in the similar Cu6+12+10 nanojar (Cu6+12+9SO4 has not been crystallized yet).43 In Cu6+12+10SO4, three O atoms of the sulfate ion point toward the larger Cu10 side ring, and the fourth O atom (equivalent to the methyl group in MePO32–) points toward the smaller Cu6 side ring (Figure 5). This is undoubtedly a result of the larger size of the methyl group compared to the O atom and the inability of the former to form hydrogen bonds with the nanojar host. In Cu8+14+9SO4, the sulfate ion does have the same orientation as MePO32– in Cu8+14+9MePO3, with three O atoms pointing toward the smaller Cu8 ring and the fourth O atom pointing toward the larger Cu9 ring (Figure S14). No comparison can be made in the case of the Cu7+13+9 nanojar since it could not be crystallized with sulfate (or with any other tetrahedral anion). Nevertheless, crystals of the Cu7+13+9 nanojar were also obtained with EtPO32–, which is found cocrystallized with Cu8+14+9EtPO3 in a 5/95 ratio in 5 (wherein the Cu7/Cu8 and Cu13/Cu14 rings are disordered with each other and the nondisordered Cu9 ring is shared by the Cu29 and Cu31 nanojars; the whole EtPO32– ion is disordered over three positions in a 62/33/5 ratio) and nBuPO32– (6; the PO3 moiety is disordered over two positions in a 82/18 ratio, by a ∼ 60° rotation approximately around the P–C bond), which show similar binding of the anion (Figure S15). However, as the size of the substituent increases, the number of hydrogen bonds to the phosphonate group shorter than 3.2 Å decreases from 13/12 for Me (two crystallographically independent nanojar moieties in the asymmetric unit; longest O···O distance: 3.172(8)/3.160(7) Å) to 12 for Et (longest O···O distance: 3.05(4) Å) and 11 for nBu (longest O···O distance: 2.943(6) Å) (Tables S5 and S6).

Figure 4.

Figure 4

Open in a new tab

Ball-and-stick representation of the crystal structures of 1–3 (side and top views). Green and blue dotted lines indicate hydrogen bonds and axial Cu···O interactions, respectively. Counterions, lattice solvent molecules, and C–H bond H atoms are omitted for clarity, and only the major component is shown for disordered moieties. For 2, only one of the two crystallographically independent nanojar moieties from the asymmetric unit is shown.

Figure 5.

Figure 5

Open in a new tab

Comparison of the crystal structures of Cu6+12+9MePO3 and Cu6+12+10SO4, illustrating the binding of the anions in opposite orientations. Color code: cyan, Cu6 ring; blue, Cu12 ring; green, Cu9 (upper) or Cu10 ring (lower).

The Cu6+12+9 nanojar could also be crystallized with PhPO32– (Figure S16, Table S3). In 9, Cu6+12+9PhPO3 is found cocrystallized with Cu6+12+10PhPO3 in a 18/82 and 9/91 ratio in the two crystallographically independent nanojar units, wherein the Cu9/Cu10 rings are disordered with each other and the nondisordered Cu6 and Cu12 rings are shared by the Cu27 and Cu28 nanojars. Similarly to the methyl derivative and in contrast to SO42–, the three O atoms of the PhPO32– moieties point toward the Cu6 ring while the phenyl group points outward through the Cu9/Cu10 ring. The O atoms of the PO3 moieties in the two nanojar units are disordered over two positions in a 88/12 and 86/14 ratio, by a 51.2(9) and 50.2(10)° rotation around the P–C bond, respectively. No aromatic interactions between the phenyl group of the PhPO32– anion and pz moieties are observed in 9.

The Cu8+14+9 nanojar was crystallized with MePO32– (3), EtPO32– (4; also, cocrystallized with Cu7+13+9EtPO3 in 5), n-dodecylPO32– (7), BnPO32– (8), and PhPO32– (10) (Figure 6). The anion is found in a single orientation in 3, 7, and 9. In 4 and 7, although the PO3 group is also in a unique orientation, the alkyl group is disordered over two positions (0.57/0.43 and 0.70/0.30 occupancy for Et and n-dodecyl, respectively). As with other nanojar sizes, the three O atoms of the phosphonate moieties point toward the smaller Cu8 side ring, whereas the substituents point outward through the larger Cu9 side ring. Aromatic interactions between the phenyl group of the phosphonate anion and a pz ring are observed in the case of 8 and 10. In 8, the pz moiety involved in the aromatic interaction is disordered over two positions in a 56/44 ratio, with dihedral angles between Ph/pz mean planes of 12.7(6) and 13.2(7)°, centroid-centroid distances of 4.343(9) and 4.758(10) Å, and shortest H atom to plane centroid distances of 3.180(7) and 3.673(9) Å, respectively. In 10, the phenyl group of the PhPO32– anion has interactions with two different pz moieties, with dihedral angles between Ph/pz mean planes of 31.3(2) and 65.6(2)°, centroid-centroid distances of 5.435(3) and 6.061(2) Å, and shortest H atom to plane centroid distances of 3.996(2) and 4.213(2) Å, respectively.

Figure 6.

Figure 6

Open in a new tab

Ball-and-stick representation of the crystal structures of 4, 7, 8, and 10 (side and top views). Green and blue dotted lines indicate hydrogen bonds and axial Cu···O interactions, respectively. Counterions, lattice solvent molecules, and C–H-bond H atoms are omitted for clarity, and only the major component is shown for disordered moieties. For 7, only one of the two crystallographically independent nanojar moieties from the asymmetric unit is shown.

Disorder in the positions of some pyrazolate ligands within the nanojar is also common. Except in 2 and 4, one or more pz moieties are disordered over two positions. In 8, a whole segment of the Cu9 ring including six Cu(pz)(OH) units is disordered over two positions with a 0.56/0.44 occupancy.

Despite the different bound anions and varying ring sizes, the structural parameters within the Cux rings in 1–10 (excluding the minor disordered nanojar components in 5 and 9, which offer less accurate values) are consistent, with average Cu–O bond lengths of 1.920(6)–1.928(4) Å, average Cu–N bond lengths of 1.970(6)–1.981(4) Å, average trans and cis N–Cu–O angles of 169.4(2)–172.2(2)° and 85.2(1)–86.0(1)°, respectively, and average Cu···Cu distances of 3.277(2)–3.327(1) Å (Tables S5–S8). The noncovalent interactions between individual Cux rings are comparable, with average axial Cu···O distances of 2.483(5)–2.540(5) Å and average H-bonded O···O distances of 2.744(5)–2.868(6) Å. The H-bonding parameters between Cux rings and the bound phosphonate anion are also similar, with average O···O distances ranging from 2.836(5) Å (11 H-bonds <3.2 Å) in 6 to 2.965(13) Å (12 H-bonds <3.2 Å) in 4.

Differences in structural details of the various Cux rings in the Cu27–Cu31 nanojars were analyzed using the dihedral angles and the component fold and twist angles between the mean planes of pyrazolate moieties and adjacent Cu–O–Cu units (as defined earlier).42,44Tables S10 and S11 illustrate that in 1–10 the average fold angles in the larger, flatter central rings (Cu12–Cu14) are quite consistent at 41.0(6)–43.2(2)°, whereas in the two smaller, more puckered side rings, the angles vary more widely: 37.9(2)–51.8(5)° (Cu9 or Cu10) and 45.6(4)–56.4(2)° (Cu6–Cu8). The corresponding average twist angles are 3.4(2)–6.2(4)° in the central rings (Cu12–Cu14), and 2.1(2)–5.2(3)° (Cu9 or Cu10) and 0.9(2)–5.2(2)° (Cu6 or Cu8) in the two side rings. Despite the pronounced variations of 78 and 18° between the individual fold (0.31(18)–78.4(3)°) and twist angles (0.0(4)–17.5(2)°), only small variations of <14 and <5° are observed for the fold and twist angle averages, respectively.

Herein, we introduce a new structural descriptor for Cux ring geometry, namely, the dihedral angles and the component fold and twist angles between the mean planes of adjacent pyrazolate moieties. Because they are placed next to each other in each cis-Cu(pz)2(OH)2 fragment, the pz moieties avoid steric hindrance by twisting and/or folding away from a coplanar geometry (Figure 7). In the large, rather flat Cu12–Cu14 central rings, the favored distortion is by twisting, as the pyrazolate units can adopt an alternating up/down orientation relative to the mean plane of the Cu atoms. In contrast, in the small, bowl-shaped Cu6 side-ring, the favored distortion is by folding. Thus, the average twist and fold angles in 1–10 are 55.4(4)–57.8(4) and 12.3(3)–31.9(2)° for the Cu12–Cu14 rings, whereas for the Cu6 ring, the corresponding average values are reversed at 9.9(2)–22.0(3) and 27.1(2)–42.1(3)°. For the Cu7 ring, an approximately equal preference for twisting (21.1(4)–24.9(3)°) and folding (23.8(3)–28.1(4)°) is observed. As the ring becomes larger, distortion by twisting is increasingly preferred (Tables S12–S14).

Figure 7.

Figure 7

Open in a new tab

Examples of different pz–pz mean-plane dihedral angles based on (a) small twist and fold angles, (b) small twist angle and large fold angle, (c) large twist angle and small fold angle, and (d) large twist and fold angles. Blue and red spheres represent Cu atoms and OH groups, respectively.

A relationship of the centroid–centroid distances between adjacent pz mean planes with the corresponding dihedral angles is also noteworthy. Although they are not directly proportional, dihedral angles larger than ∼35° are consistently (with very few exceptions) correlated with centroid–centroid distances shorter than 5.0 Å, whereas dihedral angles smaller than ∼35° are linked with centroid–centroid distances longer than 5.0 Å (Tables S12–S14).

The four- or five-coordinate geometry of the Cu atoms in different Cux rings in 1–12 was analyzed using the coordination geometry indexes τ4 and τ5 (Table S15).45,46 In the case of τ4, a value of 1.00 corresponds to a perfect tetrahedral geometry and a value of zero corresponds to a perfect square planar geometry, whereas in the case of τ5, 1.00 and zero indicate a perfect trigonal bipyramidal and perfect square pyramidal geometry, respectively. The τ4 values in 1–12 range from 0.03 to 0.40 with an average of 0.13, whereas the τ5 values range from 0.00 to 0.49 with an average of 0.08, indicating that most of the Cu atoms have coordination geometries very close to square planar or square pyramidal. As seen in Tables S5–S9, 12–14 Cu atoms of nanojars (Cu27–Cu31) are five-coordinated with axial Cu···O distances shorter than the sum of the van der Waals radii of Cu and O (2.92 Å). All Cu atoms of the larger central Cux ring (x = 12–14) in nanojars are four-coordinated lacking a nearby axial O atom, whereas in the smaller side-rings (x = 6–10), the number of five-coordinate Cu atoms is approximately equally distributed between the two rings. No correlation is observed between the τ4 (or τ5) values and the dihedral (or component fold and twist) angles between the two pz moieties bound to a given Cu atom.

In terms of coplanarity of the Cu atoms in the eight different Cux rings of the Cu27–Cu31 nanojars, the Cu6 ring (overall bowl-shaped with all pz groups on one side of the ring and all OH groups on the opposite side) is closest to planar with average deviations of only 0.028–0.098 Å from the Cu6 mean plane (largest deviation: 0.049 Å in 1 and 0.155 Å in 9). This is in contrast with the Cu12 ring, which is overall flat with pz groups alternating on opposite sides of the ring, but with large average deviations of 0.227–0.342 Å (largest deviation: 0.439 Å in 1 and 0.584 Å in 9) of its Cu atoms from the Cu12 mean plane. The largest average deviations from coplanarity are observed in the case of the largest Cu14 ring, ranging from 0.631 to 0.689 Å (largest deviation: 1.493 Å in 3, 1.283 Å in 4, 1.390 Å in 5, 1.366 Å in 7, 1.427 Å in 8, and 1.424 Å in 10). The average deviations in the Cu7 and Cu10 rings (0.194–0.299 Å) are similar to the ones of the Cu12 ring, whereas in the Cu8, Cu9, and Cu13 rings the values are 0.331–0.648 Å (Table S16).

Besides the intended outcome of favoring the crystallization of different nanojar sizes from a mixture, the use of different crystallization solvents also led to surprising new results. While the vapor diffusion of hexanes into a 1-methylnaphthalene solution of the CunEtPO3 nanojars provided crystals of Cu31EtPO3, changing the precipitant solvent to t-butanol led to the unexpected formation of a clamshell-like structure (11; Figures 8 and S17). This novel motif consists of a pair of Cu6+12+9EtPO3 nanojars tethered by two adjacent EtPO32– ions, each replacing two pyrazolate moieties (one from each nanojar unit). The clamshell structure is located on a C2 rotation axis passing in-between the two EtPO3 units. Consequently, only half of the molecule (one nanojar unit) is located within the asymmetric unit of the crystal lattice. The overall structure of the clamshell’s nanojar unit is very similar to the one of the individual nanojars, with the notable exception of the longer average Cu···Cu distances of 3.356(2) Å within the Cu9 rings, compared to 3.311 Å in regular nanojars (Table S9). This is due to the larger binding angle of the phosphonate O atoms compared to the pyrazolate N atoms, which causes two of the Cu···Cu distances within the Cu9 ring to increase to 3.496(2) and 3.555(2) Å.

Figure 8.

Figure 8

Open in a new tab

Ball-and-stick representation of the crystal structure of 11. Green and blue dotted lines indicate hydrogen bonds and axial Cu···O interactions, respectively. Counterions, lattice solvent molecules, and pz H atoms are omitted for clarity.

No individual nanojars could be crystallized with nPrPO32–. Instead, a nanojar clamshell similar to 11 was obtained from an anisole solution of CunnPrPO3 by vapor diffusion of t-butanol (Figure S18). In the crystal lattice of 12, which is also monoclinic (although in the P21/n space group instead of C2/c in the case of 11), the clamshell is not located on a C2 rotation axis. Consequently, the two nanojar units are not symmetry-related. As in 11, a longer average Cu···Cu distance of 3.355(1) is observed within the Cu9 rings of 12, due to longer Cu···Cu distances of 3.475(1)–3.559(1) Å between Cu atoms bridged by phosphonate instead of pyrazolate ligands (Table S9).

In spite of the apparent similarity, there is an important difference between the structures of 11 and 12, likely due to the larger size of the nPr chain compared to Et. In 11, the dihedral angle between mean planes running through the 12 Cu atoms of the Cu12 rings is 46.46(1)° with a centroid···centroid distance of 13.632(1) Å and a P···P separation between centrally bound anions of 9.241(2) Å, whereas the corresponding values in 12 increase to 66.89(1)°, 14.455(2) and 10.432(2) Å (Figures 9 and S19). Moreover, not only is the clamshell opening significantly larger in 12 (illustrated by the larger fold angle of 86.57(1)° compared to 66.30(3)° in 11) but also the twist angle between nanojar units (measured between Cu12 mean planes as defined above), which increases from 0° in 11 (crystallographically imposed) to 31.86° in 12.

Figure 9.

Figure 9

Open in a new tab

Illustration of the different angles of opening in nanojar clamshells 11 and 12.

1H NMR Spectroscopy

To enable VT-NMR experiments at high temperatures, DMSO-d6 was chosen as the solvent. Previous studies have shown that DMSO, despite being a good coordinating solvent, does not interfere with the nanojar framework and does not induce speciation at room temperature.39 While the 1H NMR spectra of the CunRPO3 nanojars (R = Me, Et, nPr, nBu, Bn, and Ph) in DMSO-d6 are similar to the ones of CunCO3 and CunSO4 in that the paramagnetism of the Cu2+ centers leads to drastic downfield and upfield shifts of the pyrazolate and OH proton peaks, respectively, as well as to broadening of the peaks and loss of the J coupling between nuclei, there are significant differences to be pointed out (Figures 10, 11, and S20–S31, Table S54). MePO32– is similar to SO42– considering their tetrahedral shape, but different in terms of charge distribution (formal charge on O atoms is −0.67 in the former and −0.50 in the latter) and number of hydrogen bond acceptor atoms (three vs four). Conversely, MePO32– is different from the trigonal-planar CO32– in terms of shape but identical with regard to the number of hydrogen bond acceptor atoms and their formal charges. Phosphonates (HPO3H2: pKa1 = 1.43, pKa2 = 6.68; MePO3H2: pKa1 = 2.12, pKa2 = 7.29; EtPO3H2: pKa1 = 2.43, pKa2 = 8.05; nPrPO3H2: pKa1 = 2.49, pKa2 = 8.18; BnPO3H2: pKa1 = 2.4, pKa2 = 7.8; PhPO3H2: pKa1 = 1.83, pKa2 = 7.07) are more basic than sulfate (H2SO4: pKa1 ≃ – 3; pKa2 = 1.99) but less basic than carbonate (H2CO3: pKa1 = 3.49; pKa2 = 10.33).47−49

Figure 10.

Figure 10

Open in a new tab

Variable-temperature 1H NMR spectra of the CunMePO3 (n = 27–31) nanojar mixture in DMSO-d6, showing pyrazolate proton signals in the 21–33 ppm window. The temperatures shown are the target temperatures of the probe.

Figure 11.

Figure 11

Open in a new tab

Variable-temperature 1H NMR spectra of the CunMePO3 (n = 27–31) nanojar mixture in DMSO-d6, showing OH proton signals in the −24 to −48 ppm window. The given temperatures are the target temperatures of the probe.

Subsequent discussion will focus on the Cu6+12+9 nanojars, since they could be obtained with CO32–, SO42–, and various phosphonates RPO32– (R = Me, Et, nPr, nBu, Bn, and Ph) as well. At ambient temperature, pyrazolate proton peaks for Cu6+12+9CO3 and Cu6+12+9SO4 are observed in the 37.63–22.43 and 34.83–22.56 ppm windows, respectively, whereas the corresponding OH proton peaks are observed from −29.65 to −68.08 ppm and from −26.18 to −52.95 ppm. For Cu6+12+9MePO3, the corresponding windows are 32.75–22.16 and −26.14 to −45.10 ppm. A closer inspection of the individual chemical shifts (δ) for the different Cux rings in Table S54 shows no consistent correlation with the basicity of the entrapped anions (more deshielding of the H atoms of OH groups hydrogen-bonded to more basic anions is expected). Thus, the variation of chemical shifts on changing the anion is apparently more profoundly influenced by changes in the magnetism of the Cux rings as a consequence of structural changes within the nanojar. In the case of Cu6+12+9CO3 vs Cu6+12+9SO4, the differences are caused by the different shape and orientation of the anions. In Cu6+12+9CO3, the three O atoms of CO32– form H bonds with both the Cu6 and Cu9 rings,39 whereas in Cu6+12+9SO4, one O atom of SO42– forms H bonds only with the Cu6 ring and the other three only with the Cu9 ring (assumed by analogy to Cu6+12+10SO4, as discussed in the Crystallography section).43a

For the different RPO32– anions, an expected deshielding of the H atoms of the Cu6 and Cu9 rings H-bonded to the incarcerated anion is observed on going from MePO32– (−41.30 and −45.10 ppm) to the more basic Et (−41.07 and −43.99 ppm), nPr (–41.0 and –41.0 ppm), nBu (−40.40 and −41.33 ppm), and Bn derivatives (−38.60 and −38.86 ppm). However, an even more pronounced deshielding is observed with the less basic PhPO32– anion (−34.14 and −39.68 pm). This inconsistency must be a result of differences in the structure of the nanojar framework as the orientations of the different RPO32– anions relative to the nanojar framework is identical. Indeed, an analysis of the structures of the Cu6+12+9 nanojars with MePO32– and PhPO32– reveals subtle, yet significant differences in the shape and orientation of the Cux rings relative to each other, caused by the sterically (laterally) more demanding Ph vs Me (or n-alkyl) substituent on the phosphonate moiety (Figure 12). These structural changes are expected to affect the magnetism of the Cux rings, which, in turn, affects the hyperfine shifts of the nanojar protons.

Figure 12.

Figure 12

Open in a new tab

Comparison of the Cu6+12+9 rings in 1 and 9 (only Cu atoms shown; the two symmetry-independent nanojar units of 9 are virtually identical) and the corresponding MePO32– and PhPO32– anions.

The VT 1H NMR measurements of the phosphonate nanojars in DMSO-d6 over the 20–150 °C range also show significant differences compared to the ones of the carbonate and sulfate nanojars (Table S54). For example, in the case of the Cu6+12+9 nanojar, the degree of temperature dependence of the chemical shifts of the Cu9 ring protons is significantly smaller with alkyl- and phenylphosphonates (1.5–3.8 ppm for pz and 0.3–1.9 ppm for OH) than with carbonate (4.7–5.2 ppm for pz and 13.7 ppm for OH) and sulfate (3.8–4.6 ppm for pz and 4.9 ppm for OH). The benzylphosphonate analogue shows a much larger variation of 5.1 ppm for the OH protons and a smaller variation of 1.3–2.0 ppm for the pz protons, compared to the other phosphonates. The Cu6 and Cu12 rings show smaller degrees of variation across all phosphonate nanojars, which are essentially the same as with carbonate and sulfate (Cu6: 0.3–0.9 ppm for pz and 1.1–2.5 ppm for OH; Cu12: 0.1–0.3 ppm for pz and 0.3–1.4 ppm for OH, with an outlier of 2.7 ppm for phenylphosphonate). A Curie behavior (δ ∝ 1/T) is observed for the Cu9 and Cu7 rings, which display large variations with temperature becoming less paramagnetically shifted at higher temperatures (Figures S32–S34). The δ values of all other Cux rings show only slight variations with temperature, except for one of the OH signal of the Cu12 ring in Cu6+12+9PhPO3, which shows an anti-Curie behavior becoming more paramagnetically shifted at higher temperatures.

The thermal stability of phosphonate nanojars is strikingly different from those of the carbonate and sulfate analogues. At ambient temperature, the CunCO3 mixture contains larger amounts of Cu6+12+9CO3 and Cu8+14+9CO3, smaller amounts of Cu7+13+9CO3 and Cu8+13+8CO3, and no Cu6+12+10CO3. On heating in a DMSO-d6 solution, the Cu8+14+9CO3 and Cu7+13+9CO3 nanojars gradually decompose/rearrange and give rise to a mixture of mostly Cu8+13+8CO3 and Cu6+12+9CO3 nanojars at 150 °C.39 The CunSO4 mixture, which contains larger amounts of Cu6+12+10SO4 and Cu8+14+9SO4 and smaller amounts of the other nanojar sizes at ambient temperature, gives rise to a mixture of mostly Cu8+14+9SO4 and small amounts of Cu8+13+8SO4 at 150 °C.43b In contrast, the different nanojars in the CunRPO3 mixtures appear to be equally robust, as the composition of the mixtures shows very little changes on heating to 150 °C (except for the Cu6+12+10PhPO3 nanojar, which decomposes by 60 °C). This observation is consistent with the fact that the CunRPO3 mixtures are not affected by NH3.

31P NMR Spectroscopy

In the case of the (Bu4N)2RPO3 salts, prepared in situ by dissolving stoichiometric amounts of Bu4NOH (1 M in H2O) and the corresponding phosphonic acid in DMSO-d6, the 31P chemical shifts follow a trend that reflects the basicity of the phosphonate ion (Me: 13.4 ppm; Et: 19.1 ppm; nBu: 23.8 ppm; Bn: 16.1 ppm; Ph: 6.5 ppm). In contrast, no correlation with basicity is found for the corresponding hyperfine shifts in the different CunRPO3 nanojar species (Figure 13). For example, in the case of Cu31RPO3, the δ values are 46.9 (Me), 41.6 (Et), 41.7 (nBu), 37.7 (Bn), and 38.9 (Ph) ppm. Yet, the average distance from the P atom to the Cu atoms in Cu31RPO3 (R = Me, Et, Bn, and Ph) is rather consistent (5.809(3), 5.842(3), 5.844(2), and 5.840(1) Å in 3, 4, 8, and 10, respectively) (Table S16). An even more disparate trend is observed in the case of Cu27RPO3, with δ values of 62.7 (Me), 40.0 (Et), 43.4 (nBu), 50.7 (Bn), and 61.8 (Ph) ppm. Therefore, the origin of the observed differences in the δ values must be a combination of the basicity of the phosphonate anion (related to the electron donating or withdrawing ability of its R group) and the different spin densities on the various Cu31RPO3 nanojars caused by structural changes in their corresponding framework (as also indicated by the 1H NMR studies). Nevertheless, there seems to be a correlation between the observed hyperfine shifts and the average P···Cu distance in different nanojars with the same RPO32– anion (except for R = Et). For example, the δ values for Cu27MePO3 (ave. P···Cu: 5.546(2) Å), Cu7+13+9MePO3 (ave. P···Cu: 5.672(2) Å), and Cu31MePO3 (ave. P···Cu: 5.809(3) Å) are 62.7, 52.0, and 46.9 Å, respectively, showing decreasing paramagnetic shift with increasing average P···Cu separation.

Figure 13.

Figure 13

Open in a new tab

31P NMR spectra of the CunRPO3 (Cun; n = 27–31; R = Me, Et, nPr, nBu, Bn, and Ph) nanojar mixtures in DMSO-d6 at ambient temperature. Chemical shift values (ppm) are shown under the nanojar symbols. Color code for the Cu29 nanojars: yellow, Cu7+13+9; magenta, Cu8+13+8. Assignments were made based on correlations with ESI-MS and 1H NMR spectra.

UV–Vis Spectroscopy

The UV–vis spectra of CunRPO3 (n = 27–31; R = Me, Et, nPr, nBu, Bn, and Ph) in THF are virtually identical and display two peaks with absorption maxima at 345–349 and 602–608 nm, corresponding to charge-transfer and d–d transitions, respectively (with extinction coefficients of ε347 nm = 2 × 104 L mol–1 cm–1 and ε605 nm = 2 × 103 L mol–1 cm–1) (Figure 14). The λmax values for the phosphonate nanojars are similar to the ones measured for the analogous nanojar mixtures with carbonate, sulfate, and chromate (λmax = 350–351 and 599–602 nm).42,43a,50 Thus, the entrapped anion has minimal effects on the UV–vis absorption of the nanojar framework.

Figure 14.

Figure 14

Open in a new tab

UV–vis spectra of CunRPO3 (n = 27–31; R = Me, Et, nPr, nBu, Bn, and Ph) in THF (20 μM).

Assessment of Phosphonate Binding Strength by Competitive Anion Binding

Given the inability of using host–guest titration (because a guest-free nanojar host cannot be obtained), the binding strength of phosphonates by nanojars was assessed similarly to other anions by using competitive binding experiments with Ba2+.42,43a,50,51 Neither in the case of heterogeneous conditions, when aqueous Ba(NO3)2 and water-immiscible 2-methyltetrahydrofuran (2-MeTHF) solutions of CunRPO3 nanojars (n = 27–31, R = Me or Bn) were vigorously stirred together, nor in the case of homogeneous conditions using 2-MeTHF-soluble barium dioctyl sulfosuccinate, Ba(DOSS)2, was any precipitate of a barium salt or a nanojar degradation product observed. Furthermore, ESI-MS analyses of the 2-MeTHF solutions after Ba2+ treatment show no nanojar degradation products either nor any significant changes in nanojar composition, confirming the strong binding of phosphonates by the different nanojar species.

Conclusions

In summary, we demonstrated that nanojars can bind not only small inorganic oxoanions but also their organic derivatives bearing aliphatic or aromatic substituents, such as phosphonates (RPO32–). ESI-MS analysis shows that in solution, the favored nanojar sizes are Cu31 and Cu27 followed by Cu29, whereas Cu28 and Cu30 are only observed occasionally and in small amounts. 1H and 31P NMR studies offer further details about structural isomers in the case of Cu29 and indicate that Cu7+13+9 is formed almost exclusively. In contrast to its carbonate- and sulfate-incarcerating analogues, VT 1H NMR studies reveal that the differently sized phosphonate-incarcerating nanojars are much more robust on heating to 150 °C in a DMSO-d6 solution, and with the exception of Cu28, which decomposes by 60 °C, resist conversion to a preferred nanojar size (Cu8+13+8 and Cu8+14+9 in the case of carbonate and sulfate, respectively). Similarly, NH3 treatment has a negligible effect on the CunRPO3 mixtures (except for Cu28), whereas the CunCO3 and CunSO4 mixtures are converted into Cu27CO3 and Cu31SO4.

While 1H NMR spectroscopy offers insight into the structure and magnetism of the nanojar host, 31P NMR spectroscopy probes the incarcerated phosphonate guest. The hyperfine shift of the P atom appears to be very sensitive to small changes in the magnetic environment, as varying the organic substituent leads to very large changes in the corresponding chemical shifts, not correlated with the basicity of the P atom bearing different R groups. Minor differences in the structure of the nanojar framework, which nevertheless cause significant changes in magnetism, must account for these observations. A similar inconsistency was also observed between the expected deshielding of the OH hydrogen atoms of certain Cux rings and the basicity of the different RPO32– anions involved in H-bonding.

Structural investigations by single-crystal X-ray crystallography on 12 different phosphonate nanojars (with Cu6+12+9, Cu6+12+10, Cu7+13+9, and Cu8+14+9 ring combinations, and R = Me, Et, nPr, nBu, nC12, Bn, and Ph) reveal that only the hydrophilic group of phosphonates (PO32–) needs to be buried within the nanojar cavity for efficient binding, while the organic group strings on the outside. Within the OH-lined cavity of the nanojars, the PO32– moiety is oriented toward the smaller Cux side ring (x = 6–8) and is bound by a multitude of hydrogen bonds (10–14 bonds with O···O distances shorter than 3.2 Å). Conversely, the organic substituent pokes out through the larger Cux side ring (x = 9 or 10). The Cu8 ring appears to be unfavorable for accommodating the organic group, as the Cu8+13+8 ring combination is only observed in the case of R = Me, in small amounts. The Cu9 ring, however, accommodates organic moieties as bulky as Ph, but not t-Bu directly bound to the P atom. Despite the larger size of the Cu10 ring, the Cu6+12+10 ring combination is again elusive and is observed in only small amounts in the case of R = Ph.

For the first time, cocrystallization of nanojars of different sizes by positional disorder has been observed. Thus, Cu8+14+9EtPO3 is cocrystallized with Cu7+13+9EtPO3 in a 95/5 ratio in 5 (sharing the Cu9 ring), and Cu6+12+9PhPO3 is cocrystallized with Cu6+12+10PhPO3 in 82/18 and 91/9 ratios in the two crystallographically independent nanojar units of 9 (sharing both the Cu6 and Cu12 rings). In spite of the H-bond donor-rich cavity of nanojars, which can often bind the entrapped anion in different orientations, the PO32– anion is only found disordered in 5 and 6 (Cu7+13+9nBuPO3). In 9, the occupancies of the two different positions of the PO32– anion correlate with the occupancies of the Cu6+12+9 and Cu6+12+10 nanojars.

The studies of phosphonate nanojars led to the discovery of a novel motif in nanojar chemistry, nanojar clamshells, which consist of two nanojar units tethered by two adjacent μ4-RPO3 (R = Et or nPr) phosphonate anions. This is a surprising result given that they could not be detected in any significant amounts in solution by ESI-MS. In contrast to regular nanojars, nanojar clamshells produce much higher quality crystal structures with less disorder and smaller anisotropic displacement parameters, suggesting better molecular packing. These more robust crystal lattices might drive the formation of nanojar clamshells from phosphonate-substituted nanojars (which have indeed been observed in small amounts in solution by ESI-MS) upon crystallization. Considering that nanojar clamshells double the phosphonate-binding capacity of nanojars and could also accommodate bisphosphonates in the clamshell cavity, future studies will focus on their rational synthesis and further characterization with an aim at developing liquid–liquid extraction agents for phosphonates.

Experimental Section

General Information

All commercially available chemicals were used as received (solvents are ACS- or HPLC-grade, and THF is inhibited with 250 ppm of BHT). Cu(NO3)2·2.5H2O (ACS reagent, 98%), NaOH (ACS reagent, 97%), and phenylphosphonic acid (98%) were purchased from Sigma-Aldrich, pyrazole (99%) and n-propylphosphonic acid (95%) from Oakwood Chemical, nBu4NOH (HPLC grade, 1.0 M in H2O) and phosphonic acids (methyl 98%, ethyl 98%+, n-butyl 98%, t-butyl 98%, n-dodecyl 95%, benzyl 97%) from Thermo Scientific, and t-butylphosphonic acid (98%) from Acros Organics. Deionized water was freshly boiled and cooled to room temperature under N2(g). (Bn3MeN)NO3 and Ba(DOSS)2 were prepared according to the published procedures.38,51 Gaseous NH3 was generated by gently heating a 30% aqueous NH3 solution. The synthesis and reactions of nanojars were carried out under a N2(g) atmosphere. NMR spectra were collected on a Jeol JNM-ECZS (400 MHz) instrument, and UV–vis measurements were carried out on a Shimadzu UV-1650PC spectrophotometer. 31P NMR spectra (162 MHz) were recorded in NMR tubes with fused, coaxial glass inserts, and the corresponding chemical shifts are referenced to H3PO4 (85% in H2O). Because phosphonate nanojars have only one single P atom per molecule, their phosphorus content is very low (0.6% in the larger Cu31BnPO3 to 0.7% in the smaller Cu27MePO3). Therefore, long acquisition times of up to 3 days are needed for a good signal-to-noise ratio (especially for the less abundant nanojar species).

Synthesis of [trans-CuII(μ-OH)(μ-pz)]∞

Cu(NO3)2·2.5H2O (16.250 g, 0.0700 mol) was dissolved in H2O (300 mL) in a 1 L round-bottom flask. A solution of pyrazole (4.760 g, 0.0700 mol) and NaOH (5.600 g, 0.140 mol) in H2O (300 mL) was added dropwise to this solution under stirring, which resulted in the immediate formation of a dark blue/purple precipitate. The mixture was stirred overnight and filtered the next day. The dark blue/purple solid, which is insoluble in all common solvents, was washed thoroughly with water, followed by drying in air and then in high vacuum. Yield = 9.808 g (95%).

Synthesis of (Bu4N)2[RPO3⊂{Cu(OH)(pz)}n] (CunRPO3; n = 27–31)

For method A, Cu(NO3)2·2.5H2O (200 mg, 0.859 mmol) and pyrazole (58 mg, 0.859 mmol) are dissolved in THF (10 mL). A solution of nBu4NOH (1 M in H2O, 3.439 mL, 3.439 mmol) and the corresponding phosphonic acid (0.859 mmol) in THF (10 mL) is added dropwise under stirring. The resulting deep blue solution is stirred for 10 min and then slowly added to water (300 mL) under stirring. The dark blue precipitate is filtered out, washed with water, and dried under high vacuum. Yields vary between 84 and 99%, except for R = Bn (52%) and R = nC12 (17%). For method B, [trans-CuII(μ-OH)(μ-pz)]∞ (0.2000 g, 1.355 mmol), nBu4NOH (1 M in H2O, 1.355 mL, 1.355 mmol), and the corresponding phosphonic acid (1.355 mmol) were added to toluene (15 mL) and the mixture was refluxed at 105 °C overnight. After cooling, the solid residue was filtered out and rinsed with toluene, yielding a dark blue filtrate. Then, the solvent was evaporated under a vacuum, and the solid product was washed with water to remove the excess nBu4N+ salts. Yields are similar to those obtained using Method A.

Treatment of CunRPO3 with NH3

CunRPO3 (50 mg) was dissolved in THF (10 mL), and gaseous NH3 was slowly bubbled through the resulting solution for 20 min. Then, the flask was stoppered, sealed with Parafilm, and left standing for 7 days. The solution was filtered, and the solvent was evaporated to give a dark blue residue (43–46 mg).

Competitive Anion Binding under Heterogeneous Conditions

A clear, blue solution was prepared by dissolving CunRPO3 nanojars (n = 27–31, R = Me or Bn; ∼0.02 mmol) in 2-MeTHF (20 mL). This solution was then transferred onto a solution of Ba(NO3)2 (0.0104 g, 0.04 mmol) in H2O (20 mL) by using a cannula. The mixture was stirred vigorously for an hour, followed by separation of the two layers and analysis of the 2-MeTHF layer by ESI-MS.

Competitive Anion Binding under Homogeneous Conditions

The experiments were conducted similarly to the heterogeneous setup, using 2-MeTHF (20 mL) solutions for both CunRPO3 and Ba(DOSS)2 (0.0198 g, 0.02 mmol).

Mass Spectrometry

Mass spectrometric analysis of the nanojars was performed with a Waters Synapt G1 HDMS instrument using electrospray ionization (ESI). Solutions (10–4–10–5 M) were prepared in CH3CN using either solids or aliquots taken from solutions. Samples were infused by a syringe pump at 5 μL/min, and nitrogen was supplied as the nebulizing gas at 500 L/h. The electrospray capillary voltage was set to −2.5 or +2.5 kV, with a desolvation temperature of 110 °C. The sampling and extraction cones were maintained at 40 and 4.0 V, respectively, at 80 °C.

X-ray Crystallography

Single-crystals of 1–12 were grown at room temperature by solvent vapor diffusion using the solvents indicated in the Structural Analysis section above. Once removed from the mother liquor, the crystals are extremely sensitive to solvent loss at ambient conditions and were quickly mounted under a cryostream (150 K) to prevent decomposition. X-ray diffraction data were collected from a single-crystal mounted atop a MiTeGen micromesh mount under Fomblin oil with Bruker AXS D8 Quest diffractometers with either Photon III C14 charge-integrating and photon counting pixel array detector (CPAD) using a microfocus X-ray tube with multilayer optics for monochromatization with Cu Kα (λ = 1.54178 Å) radiation or Photon II CPAD detector and graphite-monochromated Mo-Kα (λ = 0.71073 Å) radiation (for 7). The data were collected using APEX4,52 integrated using SAINT,53 and scaled and corrected for absorption and other effects using SADABS.54 The structures were solved by employing direct methods using ShelXS55 or ShelXT56 and refined by full-matrix least-squares on F2 using ShelXL.57 C–H hydrogen atoms were placed in idealized positions and refined by using the riding model. Further refinement details and thermal ellipsoid plots (Figures S2–S13) are provided in the Supporting Information. Supramolecular features (angles and distances) were measured using OLEX2.58

Acknowledgments

Dedicated to Professor Frank T. Edelmann on the occasion of his 70th birthday. This material is based on work supported by the National Science Foundation under grant no. CHE-1808554.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.inorgchem.4c02386.

  • Additional data for mass spectrometry, X-ray crystallography, and 1H NMR spectroscopy (PDF)

The authors declare no competing financial interest.

Supplementary Material

ic4c02386_si_001.pdf (20.9MB, pdf)

References

  1. Freedman L. D.; Doak G. O. The Preparation and Properties of Phosphonic Acids. Chem. Rev. 1957, 57, 479–523. 10.1021/cr50015a003. [DOI] [Google Scholar]
  2. Horsman G. P.; Zechel D. L. Phosphonate Biochemistry. Chem. Rev. 2017, 117, 5704–5783. 10.1021/acs.chemrev.6b00536. [DOI] [PubMed] [Google Scholar]
  3. Ruffolo F.; Dinhof T.; Murray L.; Zangelmi E.; Chin J. P.; Pallitsch K.; Peracchi A. The Microbial Degradation of Natural and Anthropogenic Phosphonates. Molecules 2023, 28, 6863. 10.3390/molecules28196863. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Sevrain C. M.; Berchel M.; Couthon H.; Jaffrès P.-A. Phosphonic Acid: Preparation and Applications. Beilstein J. Org. Chem. 2017, 13, 2186–2213. 10.3762/bjoc.13.219. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Galezowska J.; Gumienna-Kontecka E. Phosphonates, their Complexes and Bio-Applications: A Spectrum of Surprising Diversity. Coord. Chem. Rev. 2012, 256, 105–124. 10.1016/j.ccr.2011.07.002. [DOI] [Google Scholar]
  6. Manghi M. C.; Masiol M.; Calzavara R.; Graziano P. L.; Peruzzi E.; Pavoni B. The Use of Phosphonates in Agriculture. Chemical, Biological Properties and Legislative Issues. Chemosphere 2021, 283, 131187 10.1016/j.chemosphere.2021.131187. [DOI] [PubMed] [Google Scholar]
  7. Demadis K. D.; Papadaki M.; Varouchas D.. Metal–Phosphonate Anticorrosion Coatings, In Green Corrosion Chemistry and Engineering: Opportunities and Challenges, ed. Sharma S. K., Wiley-VCH, 2011. [Google Scholar]
  8. Liu Y.; Zhang P. Review of Phosphorus-Based Polymers for Mineral Scale and Corrosion Control in Oilfield. Polymers 2022, 14, 2673. 10.3390/polym14132673. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Metal Phosphonate Chemistry: From Synthesis to Applications; Clearfield A., Demadis K. D., Eds.; Royal Society of Chemistry, 2012. [Google Scholar]
  10. Bhanja P.; Na J.; Jing T.; Lin J.; Wakihara T.; Bhaumik A.; Yamauchi Y. Nanoarchitectured Metal Phosphates and Phosphonates: A New Material Horizon toward Emerging Applications. Chem. Mater. 2019, 31, 5343–5362. 10.1021/acs.chemmater.9b01742. [DOI] [Google Scholar]
  11. Rathore K.; Jangir R. Insight into Synthesis, Properties and Applications of Metal Phosphonates: Emphasis on Catalytic Activities. Inorg. Chim. Acta 2024, 559, 121804 10.1016/j.ica.2023.121804. [DOI] [Google Scholar]
  12. Metal Phosphates and Phosphonates: Fundamental to Advanced Emerging Applications; Gupta R. K., Ed.; Springer, 2023. [Google Scholar]
  13. Heidel K. M.; Dowd C. S. Phosphonate Prodrugs: An Overview and Recent Advances. Future Med. Chem. 2019, 11, 1625–1643. 10.4155/fmc-2018-0591. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Product Stewardship Summary: Dimethyl Methylphosphonate. https://www.ashland.com/esg/environment/responsible-care/product-stewardship.
  15. Gledhill W. E.; Feijtel T. C. J.. Environmental Properties and Safety Assessment of Organic Phosphonates Used for Detergent and Water Treatment Applications. In Detergents, de Oude N. T., Ed.; Vol. 3, part F (Anthropogenic Compounds), Springer, 1992. [Google Scholar]
  16. Nowack B. Environmental Chemistry of Phosphonates. Water Res. 2003, 37, 2533–2546. 10.1016/S0043-1354(03)00079-4. [DOI] [PubMed] [Google Scholar]
  17. Rott E.; Steinmetz H.; Metzger J. W. Organophosphonates: A Review on Environmental Relevance, Biodegradability and Removal in Wastewater Treatment Plants. Sci. Total Environ. 2018, 615, 1176–1191. 10.1016/j.scitotenv.2017.09.223. [DOI] [PubMed] [Google Scholar]
  18. Schindler D. W.; Carpenter S. R.; Chapra S. C.; Hecky R. E.; Orihel D. M. Reducing Phosphorus to Curb Lake Eutrophication is a Success. Environ. Sci. Technol. 2016, 50, 8923–8929. 10.1021/acs.est.6b02204. [DOI] [PubMed] [Google Scholar]
  19. Altaf R.; Sun B.; Lu H.; Zhao H.; Liu D. Removal and Recovery of Phosphonates from Wastewater via Adsorption. Crit. Rev. Environ. Sci. Technol. 2023, 53, 1032–1058. 10.1080/10643389.2022.2115271. [DOI] [Google Scholar]
  20. Elser J.; Bennett E. A Broken Biogeochemical Cycle. Nature 2011, 478, 29–31. 10.1038/478029a. [DOI] [PubMed] [Google Scholar]
  21. Cordell D.; Rosemarin A.; Schröder J. J.; Smit A. L. Towards Global Phosphorus Security: A Systems Framework for Phosphorus Recovery and Reuse Options. Chemosphere 2011, 84, 747–758. 10.1016/j.chemosphere.2011.02.032. [DOI] [PubMed] [Google Scholar]
  22. Mayer B. K.; Baker L. A.; Boyer T. H.; Drechsel P.; Gifford M.; Hanjra M. A.; Parameswaran P.; Stoltzfus J.; Westerhoff P.; Rittmann B. E. Total Value of Phosphorus Recovery. Environ. Sci. Technol. 2016, 50, 6606–6620. 10.1021/acs.est.6b01239. [DOI] [PubMed] [Google Scholar]
  23. Witek-Krowiak A.; Gorazda K.; Szopa D.; Trzaska K.; Moustakas K.; Chojnacka K. Phosphorus Recovery from Wastewater and Bio-Based Waste: An Overview. Bioengineered 2022, 13, 13474–13506. 10.1080/21655979.2022.2077894. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Pal S.; Ghosh T. K.; Ghosh R.; Mondal S.; Ghosh P. Recent Advances in Recognition, Sensing and Extraction of Phosphates: 2015 onwards. Coord. Chem. Rev. 2020, 405, 213128 10.1016/j.ccr.2019.213128. [DOI] [Google Scholar]
  25. Raju M. V. R.; Harris S. M.; Pierre V. C. Design and Applications of Metal-Based Molecular Receptors and Probes for Inorganic Phosphate. Chem. Soc. Rev. 2020, 49, 1090–1108. 10.1039/C9CS00543A. [DOI] [PubMed] [Google Scholar]
  26. Neal J. F.; Zhao W.; Grooms A. J.; Smeltzer M. A.; Shook B. M.; Flood A. H.; Allen H. C. Interfacial Supramolecular Structures of Amphiphilic Receptors Drive Aqueous Phosphate Recognition. J. Am. Chem. Soc. 2019, 141, 7876–7886. 10.1021/jacs.9b02148. [DOI] [PubMed] [Google Scholar]
  27. Bazzicalupi C.; Bencini A.; Lippolis V. Tailoring Cyclic Polyamines for Inorganic/Organic Phosphate Binding. Chem. Soc. Rev. 2010, 39, 3709–3728. 10.1039/b926161n. [DOI] [PubMed] [Google Scholar]
  28. Chen L.; Berry S. N.; Wu X.; Howe E. N. W.; Gale P. A. Advances in Anion Receptor Chemistry. Chem. 2020, 6, 61–141. 10.1016/j.chempr.2019.12.002. [DOI] [Google Scholar]
  29. Raposo C.; Pérez N.; Almaraz M.; Mussons M. L.; Caballero M. C.; Morán J. R. A Cyclohexane Spacer for Phosphate Receptors. Tetrahedron Lett. 1995, 36, 3255–3258. 10.1016/0040-4039(95)00458-O. [DOI] [Google Scholar]
  30. Liu Y.; Bonizzoni M. A Supramolecular Sensing Array for Qualitative and Quantitative Analysis of Organophosphates in Water. J. Am. Chem. Soc. 2014, 136, 14223–14229. 10.1021/ja507905r. [DOI] [PubMed] [Google Scholar]
  31. Esipenko N. A.; Koutnik P.; Minami T.; Mosca L.; Lynch V. M.; Zyryanova G. V.; Anzenbacher P. Jr. First Supramolecular Sensors for Phosphonate Anions. Chem. Sci. 2013, 4, 3617–3623. 10.1039/c3sc51407b. [DOI] [Google Scholar]
  32. Choi K.; Hamilton A. D. A Dual Channel Fluorescence Chemosensor for Anions Involving Intermolecular Excited State Proton Transfer. Angew. Chem., Int. Ed. 2001, 40, 3912–3915. . [DOI] [PubMed] [Google Scholar]
  33. a Kusukawa T.; Nakaguchi K.; Nishimura S.; Nakajima A. Recognition of Dicarboxylic Acids and Diphosphonic Acids Using Anthracene-Based Diamidine: Formation of Amidinium-Carboxylate and Amidinium-Phosphonate Salt Bridges in a Protic Solvent. Supramol. Chem. 2021, 33, 43–52. 10.1080/10610278.2021.1908545. [DOI] [Google Scholar]; b Kusukawa T.; Mura R.; Ooe M.; Sumida R.; Nakagawa A. Recognition of Carboxylic Acids and Phosphonic Acids Using 1,8-Diphenylnaphthalene-Based Diguanidine. Tetrahedron 2021, 77, 131770 10.1016/j.tet.2020.131770. [DOI] [Google Scholar]; c Kusukawa T.; Mura R.; Ohtagaki Y.; Ooe M. Synthesis of an Anthracene-Based Diguanidine and its Recognition of Carboxylic Acids and Phosphonic Acids. Tetrahedron 2020, 76, 131065 10.1016/j.tet.2020.131065. [DOI] [Google Scholar]; d Kusukawa T.; Nagano H.; Nakaguchi K.; Takeshita S.; Harumoto Y. Turn-On Fluorescence Sensor for Mono- and Di-Phosphonic Acid Derivatives Using Anthracene-Based Diamidine and its Detection of Amidinium-Phosphonate and Amidinium Formation. Tetrahedron 2018, 74, 465–476. 10.1016/j.tet.2017.12.011. [DOI] [Google Scholar]
  34. a Chen Y.; Kuvayskaya A.; Pink M.; Sellinger A.; Flood A. H. A Library of Vinyl Phosphonate Anions Dimerize with Cyanostars, Form Supramolecular Polymers and Undergo Statistical Sorting. Chem. Sci. 2024, 15, 389–398. 10.1039/D3SC03685E. [DOI] [PMC free article] [PubMed] [Google Scholar]; b Zhao W.; Qiao B.; Tropp J.; Pink M.; Azoulay J. D.; Flood A. H. Linear Supramolecular Polymers Driven by Anion–Anion Dimerization of Difunctional Phosphonate Monomers Inside Cyanostar Macrocycles. J. Am. Chem. Soc. 2019, 141, 4980–4989. 10.1021/jacs.9b00248. [DOI] [PubMed] [Google Scholar]
  35. a Daly S. M.; Grassi M.; Shenoy D. K.; Ugozzoli F.; Dalcanale E. Supramolecular Surface Plasmon Resonance (SPR) Sensors for Organophosphorus Vapor Detection. J. Mater. Chem. 2007, 17, 1809–1818. 10.1039/b615516b. [DOI] [Google Scholar]; b Liu P.; Wang H.; Zeng H.; Hong X.; Huang F. A [15]Paracyclophenone and its Fluorenone-Containing Derivatives: Syntheses and Binding to Nerve Agent Mimics via Aryl-CH Hydrogen Bonding Interactions. Org. Chem. Front. 2021, 8, 25–31. 10.1039/D0QO00456A. [DOI] [Google Scholar]
  36. Groom C. R.; Bruno I. J.; Lightfoot M. P.; Ward S. C. The Cambridge Structural Database. Acta Crystallogr., Sect. B: Struct. Sci., Cryst. Eng. Mater. 2016, 72, 171–179. 10.1107/S2052520616003954. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Nakamura T.; Yonemura S.; Nabeshima T. Synthesis of Per(5-N-carboxamide-5-dehydroxylmethyl)-β-cyclodextrins and their Selective Recognition Ability Utilizing Multiple Hydrogen Bonds. Chem. Commun. 2019, 55, 3872–3875. 10.1039/C9CC00517J. [DOI] [PubMed] [Google Scholar]
  38. Al Isawi W. A.; Zeller M.; Mezei G. Tribenzyl(methyl)ammonium: A Versatile Counterion for the Crystallization of Nanojars with Incarcerated Selenite and Phosphite Ions and Tethered Pyrazole Ligands. Cryst. Growth Des. 2022, 22, 1398–1411. 10.1021/acs.cgd.1c01361. [DOI] [Google Scholar]
  39. Ahmed B. M.; Szymczyna B. R.; Jianrattanasawat S.; Surmann S. A.; Mezei G. Survival of the Fittest Nanojar: Stepwise Breakdown of Polydisperse Cu27–Cu31 Nanojar Mixtures into Monodisperse Cu27(CO3) and Cu31(SO4) Nanojars. Chem.—Eur. J. 2016, 22, 5499–5503. 10.1002/chem.201600271. [DOI] [PubMed] [Google Scholar]
  40. Goura J.; Chandrasekhar V. Molecular Metal Phosphonates. Chem. Rev. 2015, 115, 6854–6965. 10.1021/acs.chemrev.5b00107. [DOI] [PubMed] [Google Scholar]
  41. Ahmed B. M.; Calco B.; Mezei G. Tuning the Structure and Solubility of Nanojars by Peripheral Ligand Substitution, Leading to Unprecedented Liquid–Liquid Extraction of the Carbonate Ion from Water into Aliphatic Solvents. Dalton Trans. 2016, 45, 8327–8339. 10.1039/C6DT00847J. [DOI] [PubMed] [Google Scholar]
  42. Al Isawi W. A.; Hartman C. K.; Singh P.; Zeller M.; Mezei G. Chromate Incarceration by Nanojars and Its Removal from Water by Liquid–Liquid Extraction. Inorg. Chem. 2023, 62, 5716–5728. 10.1021/acs.inorgchem.3c00262. [DOI] [PubMed] [Google Scholar]
  43. a Fernando I. R.; Surmann S. A.; Urech A. A.; Poulsen A. M.; Mezei G. Selective Total Encapsulation of the Sulfate Anion by Neutral Nano-Jars. Chem. Commun. 2012, 48, 6860–6862. 10.1039/c2cc32074f. [DOI] [PubMed] [Google Scholar]; b Ahmed B. M.; Hartman C. K.; Mezei G. Sulfate-Incarcerating Nanojars: Solution and Solid-State Studies, Sulfate Extraction from Water, and Anion Exchange with Carbonate. Inorg. Chem. 2016, 55, 10666–10679. 10.1021/acs.inorgchem.6b01909. [DOI] [PubMed] [Google Scholar]
  44. Al Isawi W. A.; Zeller M.; Mezei G. Isomorphous but No Dead Ringer: Contrasting the Supramolecular Binding of Tetrafluoroberyllate and Sulfate Ions by Nanojars. Cryst. Growth Des. 2023, 23, 1676–1688. 10.1021/acs.cgd.2c01261. [DOI] [Google Scholar]
  45. Yang L.; Powell D. R.; Houser R. P. Structural Variation in Copper(I) Complexes with Pyridylmethylamide Ligands: Structural Analysis with a New Four-Coordinate Geometry Index, τ4. Dalton Trans. 2007, 955–964. 10.1039/B617136B. [DOI] [PubMed] [Google Scholar]
  46. Addison A. W.; Rao T. N.; Reedijk J.; van Rijn J.; Verschoor G. C. Synthesis, Structure, and Spectroscopic Properties of Copper(II) Compounds Containing Nitrogen–Sulphur Donor Ligands; The Crystal and Molecular Structure of Aqua[1,7-bis(N-methylbenzimidazol-2′-yl)-2,6-dithiaheptane]copper(II) Perchlorate. J. Chem. Soc., Dalton Trans. 1984, 1349–1356. 10.1039/DT9840001349. [DOI] [Google Scholar]
  47. Dean J. A.; Lange N. A.. Lange’s Handbook of Chemistry, 15th ed.; McGraw-Hill: New York, 1999. [Google Scholar]
  48. Pines D.; Ditkovich J.; Mukra T.; Miller Y.; Kiefer P. M.; Daschakraborty S.; Hynes J. T.; Pines E. How Acidic Is Carbonic Acid?. J. Phys. Chem. B 2016, 120, 2440–2451. 10.1021/acs.jpcb.5b12428. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Fonseca E. M. B.; Trivella D. B. B.; Scorsato V.; Dias M. P.; Bazzo N. L.; Mandapati K. R.; de Oliveira F. L.; Ferreira-Halder C. V.; Pilli R. A.; Miranda P. C. M. L.; Aparicio R. Crystal Structures of the apo Form and a Complex of Human LMW-PTP with a Phosphonic Acid Provide New Evidence of a Secondary Site Potentially Related to the Anchorage of Natural Substrate. Bioorg. Med. Chem. 2015, 23, 4462–4471. 10.1016/j.bmc.2015.06.017. [DOI] [PubMed] [Google Scholar]
  50. Al Isawi W. A.; Zeller M.; Mezei G. Capped Nanojars: Synthesis, Solution and Solid-State Characterization, and Atmospheric CO2 Sequestration by Selective Binding of Carbonate. Inorg. Chem. 2021, 60, 13479–13492. 10.1021/acs.inorgchem.1c01826. [DOI] [PubMed] [Google Scholar]
  51. Al Isawi W. A.; Zeller M.; Mezei G. Supramolecular Incarceration and Extraction of Tetrafluoroberyllate from Water by Nanojars. Inorg. Chem. 2022, 61, 8611–8622. 10.1021/acs.inorgchem.2c01198. [DOI] [PubMed] [Google Scholar]
  52. Bruker APEX4. Bruker AXS Inc.: Madison, Wisconsin, USA, 2019. [Google Scholar]
  53. Bruker SAINT v.8.38A. Bruker: Madison, Wisconsin, USA, 2017. [Google Scholar]
  54. Krause L.; Herbst-Irmer R.; Sheldrick G. M.; Stalke D. Comparison of Silver and Molybdenum Microfocus X-Ray Sources for Single-Crystal Structure Determination. J. Appl. Crystallogr. 2015, 48, 3–10. 10.1107/S1600576714022985. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Sheldrick G. M. A Short History of SHELX. Acta Crystallogr., Sect. A: Found. Crystallogr. 2008, A64, 112–122. 10.1107/S0108767307043930. [DOI] [PubMed] [Google Scholar]
  56. Sheldrick G. M. SHELXT – Integrated Space-Group and Crystal-Structure Determination. Acta Crystallogr., Sect. A: Found. Crystallogr. 2015, A71, 3–8. 10.1107/S2053273314026370. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Sheldrick G. M. Crystal Structure Refinement with SHELXL. Acta Crystallogr., Sect. C: Struct. Chem. 2015, C71, 3–8. 10.1107/S2053229614024218. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Dolomanov O. V.; Bourhis L. J.; Gildea R. J.; Howard J. A. K.; Puschmann H. OLEX2: A Complete Structure Solution, Refinement and Analysis Program. J. Appl. Crystallogr. 2009, 42, 339–341. 10.1107/S0021889808042726. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ic4c02386_si_001.pdf (20.9MB, pdf)

Articles from Inorganic Chemistry are provided here courtesy of American Chemical Society

RESOURCES