Structural Diversity of Lanthanum–Organic Frameworks Based on 1,4

Jan 4, 2013 - On the basis of the high coordination numbers, typical of lanthanide centers, those isotypical structures exhibited a new trinodal topol...
0 downloads 9 Views 3MB Size
Article pubs.acs.org/crystal

Structural Diversity of Lanthanum−Organic Frameworks Based on 1,4-Phenylenebis(methylene)diphosphonic Acid Sérgio M. F. Vilela,†,‡ Ricardo F. Mendes,† Patrícia Silva,† José A. Fernandes,† Joaõ P. C. Tomé,‡ and Filipe A. Almeida Paz*,† †

Department of Chemistry, CICECO, University of Aveiro, 3810-193 Aveiro, Portugal Department of Chemistry, QOPNA, University of Aveiro, 3810-193 Aveiro, Portugal



S Supporting Information *

ABSTRACT: The preparation of five different compounds, [La2(H2pmd)3(H2O)12] (1: 1D), [La2(H2pmd)(pmd)(H2O)2] (2: 3D), [La(H3pmd)(H2pmd)(H2O)] (3: 3D), [La2(H2pmd)3(H2O)2] (4: 3D), and [La2(H2pmd)(pmd)(H2O)4] (5: 3D), as crystalline materials from the reaction of 1,4-phenylenebis(methylene)diphosphonic acid (H4pmd) with lanthanum chloride is reported. Two different, fast, and economically viable synthetic approaches were employed with their various parameters being probed and, when possible, optimized to increase yields and purity: microwave-assisted synthesis (MWAS, in ca. 1 min) and a bench procedure using standard ambient conditions (one-pot, ca. 10 min). Compounds 1 and 2 were isolated as phase-pure crystalline materials. Compounds 3 and 4 were characterized by single-crystal X-ray diffraction from mixtures, and compound 5 was identified by powder X-ray diffraction studies (also from a physical mixture with 2). Structural details for all compounds were investigated by using in tandem X-ray diffraction (singlecrystal and powder), electron microscopy (SEM and EDS), and FT-IR spectroscopy. Topological studies were also performed for all 3D networks. The conversion of compound 1 (1D) into 3D networks was investigated using hydrothermal, microwave, and one-pot methods: 1 was totally converted into phase-pure 2 via the hydrothermal method and one-pot method.

1. INTRODUCTION Metal−organic frameworks (MOFs), a class of crystalline and well-ordered compounds also known as coordination polymers, remain one of most worldwide-investigated class of materials,1 being the result of the self-assembly of carefully selected or designed multitopic organic linkers2 with metallic centers, which frequently result in intriguing structural architectures and topologies, many of which are capable of reticular design to, for example, expand pore size.3 Moreover, because of their unique and interesting properties, MOFs also have the ability to be employed in distinct fields, such as gas storage and separation (including small vapor molecules),3,4 catalysis,5 ion exchange,6 magnetism,7 as imaging contrast agents,8 and optical sensors,9 among many others. Despite the large number of MOFs described almost on a daily basis in the literature, there remain several underexploited niches within this field of research, many of which may address important technological applications. One of them is the study of the type of heating method applied in the synthesis of these materials. Undoubtedly, the most common way to prepare MOFs is the hydro(solvo)thermal procedure, which requires large amounts of time and energy, usually leading to a substantial amount of chemical waste.10 Novel, cheaper, and more environmentally friendly concurrent approaches are thus in demand, which may then be chosen by the synthetic chemist according to the needs or the availability of materials and © XXXX American Chemical Society

conditions. Some alternative synthetic procedures have already been reported,1d,10a,c,11 such as the use of microwaves.10a,12 Such technology is hardly new in synthetic chemistry as it has been widely used in several branches of organic chemistry and in the synthesis of microporous and nanoporous inorganic materials,13 but it still remains seen as a specialized area in the vast field of MOF synthesis.14 This method is, nevertheless, coined by the fast kinetics of crystal nucleation and growth, typically leading to microcrystalline powder. However, the very short periods of irradiation and the ability to isolate intermediary products make microwave heating a very appealing synthetic methodology for scientists interested in materials science.10a,12,15 Both hydro(solvo)thermal and microwave-assisted approaches have been widely employed by our research group. The experimental reaction conditions (e.g., temperature and time) are, nevertheless, demanding. Having this in mind, we are directing our focus to the use of synthetic approaches that allow the preparation of MOFs in a much simple way, under milder or even ambient conditions. We note that other synthetic methods based on electrochemical procedures, atmospheric pressure and reflux conditions, ultrasonic irradiation, and Received: August 3, 2012 Revised: October 21, 2012

A

dx.doi.org/10.1021/cg301112k | Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design

Article

Scheme 1. Structural Diversity (from 1D to 3D) for the La3+/H4pmd System When Using Different Synthetic Approaches: Microwave Heating and One-Pot Methoda

a

The conversion pathways between the 1D isolated material and the two distinct 3D MOFs are also highlighted.

mechanochemical mixing have also been employed, but many of these remain almost as isolated reports in the literature.10a,b,11,16 The easiest experimental procedure is probably the one-pot method, consisting of stirring the reactants at normal temperature and pressure.17 This “shake and bake” procedure has gained a large interest mainly because of its simplicity and low energy consumption.18 The ability to synthesize different MOFs at room temperature in short periods of time, with particle sizes ranging from a few nanometers to high fractions of millimeters is also of utmost importance.18a,19 Our research group has been employing multipodal phosphonate-based organic linkers for the construction of MOFs with, mainly, rare-earth cations.20 These chelating groups constitute suitable rigid building blocks for the construction of robust multidimensional networks:1a,21 on the one hand, the existence of three oxygen atoms placed at the vertices of a tetrahedron mimics well the basic building unit of zeolites and zeotype materials; on the other, these units are, in general, structurally and thermally more robust than carboxylates, also creating stronger connections with larger cations, such as lanthanides. Several families of structures based on commercial polyphosphonate reactants have been investigated to date in our group, but we are now directing our focus to the preparation of materials based on more bulky, less flexible, and, in several cases, predesigned organic precursors. Indeed, by chemically modifying the structure of the organic ligand, we intend to simultaneously modify the network connectivity while promoting significant improvements on the observed chemical and physical properties, in particular, concerning photoluminescence and catalysis.20c,d,22 Indeed, the quest for new structural features and the systematic probing for new topologies may have a significant impact in such properties as they strongly rely on the metal coordination environment and, lately, on the surface reactivity or quenching effects of the crystallites.

The bipodal organic linker 1,4-phenylenebis(methylene)diphosphonic acid (H4pmd) has been widely employed in the preparation of inorganic−organic hybrid materials. These studies have been focused on the self-assembly of structures, typically under hydro(solvo)thermal conditions, which vary from 1D to 3D networks by combining H4−xpmd−x residues with transition (Ni2+, Mn2+, Cd2+, Co2+, Zn2+, and Cu2+)23 and p-block (Ga3+, Sn2+, and Pb2+)24 metals. Back in 2008, our research group reported the isolation of the first lanthanideorganic frameworks (LnOFs) having residues of H4pmd, [Ln(Hpmd)(H2O)] (where Ln3+ = Ce3+ and Pr3+). On the basis of the high coordination numbers, typical of lanthanide centers, those isotypical structures exhibited a new trinodal topology and high thermal stability.25 That work motivated us to further explore this system concerning: (i) new and more simple synthetic strategies to direct the self-assembly of novel multidimensional materials and (ii) the transposition of the acquired knowledge and optimal experimental conditions to other lanthanide ions with the final objective to isolate materials for technological applications (mostly for photoluminescent and catalytic purposes). While investigating the La3+/H4pmd system using microwave-assisted synthesis (MWAS) and a bench procedure coined as the one-pot method, we discovered five novel MOF structures (Scheme 1): large crystals of [La2(H2pmd)3(H2O)12] (1) and [La2(H2pmd)(pmd)(H2O)2] (2) (isolated from the reaction vials via filtration from the reaction products of both MWAS and the one-pot approaches), small crystals of [La(H3pmd)(H2pmd)(H2O)] (3) (as impurities in some MWAS experiments), [La2(H2pmd)3(H2O)2] (4), and powders of [La2(H2pmd)(pmd)(H2O)4] (5) (from one-pot reactions). These mild synthetic approaches may represent a clear advantage in future studies of conversion of low dimensionality materials (e.g., 1D and 2D networks) into robust 3D MOFs. This possibility is not unprecedented in this class of compounds and has been already reported by our research B

dx.doi.org/10.1021/cg301112k | Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design

Article

group.20d,26 Interconversion studies in MOF-related structures27 are usually based on acidification28 or changes in the crystalline structure promoted by dehydration/rehydration methods.29 For the particular case herein reported, it is shown that the 1D compound [La2(H2pmd)3(H2O)12] (1) could be employed as a starting material for the synthesis of higher dimensionality compounds. In that sense, we successfully used simple experimental conditions based on the one-pot, microwave, and hydrothermal approaches, which were previously optimized in order to obtain each of the compounds and to evaluate conversion pathways between the achieved products.

[Ln(Hpmd)(H2O)] (Ln3+ = Ce3+ and Pr3+), which was prepared under hydro-ionothermal conditions.25 2.2. Microwave-Assisted Hydrothermal Synthesis. Hydrothermal synthesis has been routinely employed by our group as a daily synthetic approach of MOFs. In this context, phase-pure [La2(H2pmd)(pmd)(H2O)2] (2) was first isolated under typical hydrothermal conditions, using convection heating at 180 °C during 72 h (see the Materials and Methods for details). The crystals obtained presented a platelike morphology of dozens of micrometers. Considering that mild and faster conditions are usually preferred, we first transposed the preparation of the aforementioned material to MWAS. The presence of three distinct phases, as unveiled by powder X-ray diffraction (PXRD), was unequivocally confirmed while inspecting the influence of the reaction parameters on the purity and crystal habit (see Scheme 3). Microwave heating was found to be an

2. RESULTS AND DISCUSSION 2.1. Synthetic Strategy of MOFs and Preparation of the Organic Linker. Microwave-assisted synthesis (MWAS) and the one-pot method have distinct principles, and a direct transposition of the experimental conditions from one method to another is difficult. While MWAS permits the reduction of reaction time and the possibility to control the morphology of the materials,20d,30 the one-pot method usually allows growing of crystals for single-crystal X-ray diffraction studies while maintaining small reaction times.19a,b A systematic variation of the experimental conditions is, thus, pertinent in order to obtain (i) phase-pure and (ii) highly crystalline materials while (iii) exhibiting a uniform crystal habit and size distribution. To quickly screen the achieved products in terms of purity and morphology, a combination of X-ray diffraction (powder and single-crystal) and scanning electron microscopy (SEM) was used. Replication of the synthetic conditions was also performed to test the reproducibility of the present reactive system, and the results summarized in the following sections result from these reproducibility tests. The use of residues of 1,4-phenylenebis(methylene)diphosphonic acid (H4pmd) for the synthesis of MOFs has already been reported by some of us.25 In those studies, H4pmd was generated in situ in a synthetic route involving ionic liquids departing from its tetraethyl ester derivative. The method, though novel, proved difficult to perform and to extend to other lanthanide centers. Therefore, and considering the intriguing architectures and peculiar topologies that were found at that time, we wished to overpass that labored synthesis by preparing H4pmd through a direct acid hydrolysis of the commercially available tetraethyl 1,4-phenylenebis(methylene)bisphosphonate (Scheme 2). Alongside with the

Scheme 3. Optimization of the Microwave-Assisted Hydrothermal Synthesis of [La2(H2pmd)3(H2O)12] (1) and [La2(H2pmd)(pmd)(H2O)2] (2)a

a

Legend: (purple) phase-pure 1; (green) phase-pure 2; (brown) [La(H3pmd)(H2pmd)(H2O)] (3); (yellow) indicates the presence of unreacted organic linker, which can be easily removed by washing the products with distilled water. The metal-to-ligand molar ratio used in all syntheses is ca. 1:1.

efficient method to prepare different phases by solely changing the reaction temperature: [La2(H2pmd)3(H2O)12] (1) is formed at low temperatures (below 40 °C); [La2(H2pmd)(pmd)(H2O)2] (2) is mainly obtained at high temperatures, typically above 90 °C, in line with the results of the static hydrothermal conditions. For a midrange temperature, mixtures of both compounds were always observed alongside with the presence of a small amount of [La(H3pmd)(H2pmd)(H2O)] (3). Unreacted organic ligand may be found in all prepared materials (except for high temperatures). Nevertheless, it does not represent an experimental hindrance because it is easily removed from the final product by washing with distilled water. We note that grinding the reactant prior to the synthetic procedure can drastically reduce the presence of unreacted ligand. The reaction time does not seem to have a significant influence on the preparation of 2. The same does not happen for the employed irradiation power. A strict control of the reaction temperature is, thus, crucial. It is also noted that, using

Scheme 2. Synthetic Route to Prepare 1,4Phenylenebis(methylene)diphosphonic Acid (H4pmd)

simplification of the previous ionic liquid approach, the reactive routes herein investigated further allowed us to control more easily the amount of product obtained as well as the reactive parameters. Full characterization of H4pmd by 1H, 13C, and 31P nuclear magnetic resonance (NMR) was done and is given in the Supporting Information (Figures S1, S2, and S3). Noteworthy, none of the framework structures presented herein correspond to that previously reported by us, C

dx.doi.org/10.1021/cg301112k | Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design

Article

our experimental setup, power input may have a great influence on the outcome because it is difficult to maintain stable low temperatures while using high irradiation power and vice versa. Besides 1 and 2, the optimization stage also revealed the presence of minor quantities of compound 3 as an impurity. Optimization of the experimental conditions to isolate phasepure 3 proved unsuccessful. Nevertheless, from the obtained physical mixtures and taking into account the considerable differences in particle size and morphology, it was possible to manually isolate a crystal representative of each phase. These crystals allowed a full structural characterization using singlecrystal X-ray diffraction studies (see crystallographic details in section 2.5 and in the Supporting Information, section 2). 2.3. One-Pot Synthesis. The one-pot methodology is substantially different from MWAS: milder temperatures are typically used alongside with no (autogenous) pressure. For each pair of temperature/time, the metal-to-ligand molar ratios were varied at the proportions of ca. 1:1 and 1:2. A second 1:1 trial was performed, but diluted to the double, hereafter defined as 1:1*. As in MWAS, phase-pure products [La2(H2pmd)3(H2O)12] (1) and [La2(H2pmd)(pmd)(H2O)2] (2) may be readily obtained under various reaction conditions. Additionally, two new compounds were found, namely, [La2(H2pmd)3(H2O)2] (4) and [La2(H2pmd)(pmd)(H2O)4] (5). All the materials are crystalline and show particle sizes in the micrometer range, as observed for the samples prepared by MWAS. We note that this method has the drawback of introducing some randomness in the prediction of the phases to be obtained (Scheme 4). In opposition to the MWAS, where a clear evolution of a phase into the other with increasing temperature (the most significant reaction parameter) is observed, in the one-pot method, the same logical behavior is not present. As for MWAS, the reaction time does not seem to have a strong influence on the obtained compound and the most important parameter is, once again, the temperature. Above room temperature, the composition of the reactive mixtures is also a factor to be considered. Despite a handful of exceptions at 40 °C, some conclusions can be inferred:

Scheme 4. Optimization of the One-Pot Synthesis of [La2(H2pmd)3(H2O)12] (1) and [La2(H2pmd)(pmd)(H2O)2] (2)a

a Legend: (purple) phase-pure 1; (green) phase-pure 2; (blue) [La2(H2pmd)3(H2O)2] (4); (red) [La2(H2pmd)(pmd)(H2O)4] (5). “∗” refers to a concentration of metal and ligand that is half of those used for the other syntheses. The ratio is represented as the metal-toligand molar ratio. Phase 1 is predominant at room temperature, and 2 is formed preferentially at higher temperatures. These two compounds can be isolated as pure phases in a variety of experimental conditions. Compounds 4 and 5 are only formed at high temperatures (80 °C) and could not be isolated as pure phases.

(i) Compound 1 predominates at room temperature, whereas compound 2 is formed especially at high temperatures (Scheme 4). (ii) In general, compound 5 appears essentially in the 1:1* trials; at 80 °C, this compound is even the major product. (iii) Higher temperatures and ligand-to-metal molar ratios favor the appearance of compound 4. (iv) Compounds 4 and 5 can only be prepared (but not isolated as pure-phase) at high temperatures (80 °C). In short, we can assume that 1 is the kinetic compound because it is formed at milder conditions. On the other hand, 2 seems to be the thermodynamic phase. A higher quantity of H4pmd favors the formation of 4 due to its 1:1.5 La:H2pmd stoichiometry, with this tendency being enhanced at higher temperatures. 2.4. Crystal Morphology. Crystal morphology varies according to the obtained phase for the two heating methods. Crystals of [La2(H2pmd)3(H2O)12] (1) exhibit a needle-like morphology for all tested reaction conditions, with individual crystallites being stacked into large aggregates with varying shapes and sizes. This was also observed for compound

[La2(H2pmd)(pmd)(H2O)2] (2), with the exception of the crystallite habit, being plates. Comparing the two methods, MWAS originates needles of 1 with lengths typically smaller than those obtained by the onepot methodology. On the other hand, the platelike crystallites of 2 have similar sizes for the two approaches. Concerning crystallite size, when MWAS is employed, the size of needles of 1 is generally smaller than the platelike crystallites of 2. In opposition, it is not possible to unequivocally determine the relation of the size of crystallites of the two compounds for the one-pot method. Because the temperature is of crucial importance to isolate phase-pure products, it is, thus, pertinent to look into more detail at the variation of the average particle size and thickness, when this parameter is changed. For MWAS, at lower temperatures (e.g., 30 and 40 °C) and irradiation power (10 W), phase-pure 1 is obtained (after removing the excess of the organic linker) with particle sizes ranging from a few to tens of micrometers. At 50, 60, and 70 °C, mixtures of 1 and 2 are isolated (for irradiation powers of 25 and 50 W). Crystals of phase-pure 2 begin to appear at 50 °C, becoming predominant above 70 °C, as observed from the combined PXRD and SEM D

dx.doi.org/10.1021/cg301112k | Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design

Article

Figure 1. Powder X-ray diffraction patterns and SEM images of bulk materials prepared using MWAS at different temperatures during 5 min of reaction: phase-pure 1, as needle-like crystals, is isolated at low temperatures (30 and 40 °C), whereas phase-pure 2 comprises platelike crystals at 150 °C (please note: at 90 and 120 °C, compound 2 is contaminated with small amounts of 3). Intermediate temperatures (50, 60, and 70 °C) promote the preparation of mixtures composed of both phases 1 and 2.

studies given in Figure 1. In this case, it is not possible to conclude on a trend for the variation of the average crystal size. Only at 150 °C, a significant increase in the crystallite size is observed. However, the emergence of [La(H3pmd)(H2pmd)(H2O)] (3) in the reactions performed at 90 and 120 °C may, somehow, inhibit crystal growth. Hence, only at 150 °C, there is no interference of contaminant species (i.e., free organic ligand or compound 3), and therefore, the increase in particle size is clearly noticeable. For the materials obtained at 150 °C, but using different microwave irradiation periods, it seems that reaction time does not have a significant role in the final particle morphology and size (Figure 2). When using the one-pot method and for a metal-to-ligand molar ratio of ca. 1:1, at mild temperatures in which compound 1 is predominant (room temperature and 40 °C), the increase of the reaction time decreases, in general, the length of the needles. Furthermore, the difference in size for these two temperatures is negligible. For higher temperatures (60 and 80 °C), platelike crystallites of 2 are observed, and the reaction time does not affect the particle size. The rise in temperature increases the aggregation of particles and the size distribution for all sample ranges from a few micrometers (ca. 2−3 μm) to dozens of micrometers (ca. 20−30 μm) (Figure 3). 2.5. Crystal Structure Elucidation and Description. Because of the variety of compounds herein obtained using two different synthetic methods, and our main goal to compare

those products concerning their purity, crystallinity, morphology, and particle size, in this section, we shall only discuss their general structural elucidation. Compounds 1 through 4 were characterized by single-crystal X-ray diffraction, whereas compound 5 was identified by ab initio powder X-ray diffraction studies. For a detailed description of the collection of data and crystal structure resolution, see section 4.6 (for compounds 1− 4) and section 4.7 (for compound 5). The present crystallographic description has comparative purposes, and for a more detailed discussion on each of the compounds, please see section 2 in the Supporting Information, which includes a more detailed description of the coordination environment around the metal centers, figures depicting the asymmetric units and the crystal packing viewed along the [100] direction of the unit cell for compounds 1 through 4 (Figures S4−S8), and along [010] for 5 (Figure S9), as well as crystallographic tables presenting selected bond lengths and angles and the hydrogen-bonding geometry (Tables S1−S9). Compounds [La2(H2pmd)3(H2O)12] (1) and [La2(H2pmd)(pmd)(H2O)2] (2), which were isolated as phase-pure products, were also characterized in detail using thermogravimmetric studies, FT-IR, and solid-state nuclear magnetic resonance (see sections 6−8, respectively, in the Supporting Information, which also includes analysis of the collected data). Five different crystal structures were unveiled in the course of this work, namely, [La2(H2pmd)3(H2O)12] (1), [La2(H2pmd)E

dx.doi.org/10.1021/cg301112k | Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design

Article

Figure 3. SEM pictures of the materials obtained when a ca. 1:1 metalto-ligand molar ratio is employed, depicting the variation of temperature and reaction time. Variation in temperature produces changes in particle morphology and size. Despite the variation in the reaction time, significant modifications are not observable: particle size and morphology remain approximately identical at the same temperature. The white bars represent 10 μm. Legend: (purple) phase-pure [La 2 (H 2 pmd) 3 (H 2 O) 12 ] (1); (green) phase-pure [La2(H2pmd)(pmd)(H2O)2] (2); (blue) [La2(H2pmd)3(H2O)2] (4); (red) La2(H2pmd)(pmd)(H2O)4] (5).

distorted octahedron (Figure 5; Figure S6 in section 2 of the Supporting Information). The H4−xpmd−x residues appear in two different protonation degrees and hapticities, one being of the type μ2-O,μ2-O′,κ1-O″-phosphonate and the other of the type μ2-O,O′-hydrogenophosphonate. In 3 and 4, the metal center is heptacoordinated to six nonchelating phosphonic residues and to one molecule of water. The large number of bulky ligands prevented chelation, reducing, in this way, the coordination number to a considerably low number for the La3+ cation. Remarkably, while in 3, the metal center is almost rotosymmetric with the coordination environment resembling a distorted pentagonal bipyramid, in 4, the metal center resembles a monocapped trigonal prism (Figures 6 and 7; Figures S7 and S8 in the Supporting Information). The refinement of the crystal structure of 3 did not permit a clear determination of the protonation sites. Excluding the assignment of protonation degree, it is possible to say that there are two different hapticities for the phosphonate moiety residues, one being μ2-O,O′ and another κ1-O. In 4, all the ligand moieties act as μ2-O,O′-hydrogenophosphonates. In compound 5, the metal center is octacoordinated. As it happens with compounds 3 and 4, the metal center is, in 5, coordinated to six organic residues, which, given their large volume, prevent chelation. However, the high number of organic ligands does not limit the number of water molecules, and in this case, the metal center is connected to two molecules of water. The coordination geometry around the metal center resembles a square antiprism (Figure 8; Figure S9 in the Supporting Information). Given that the model of the crystal structure of 5 was obtained by ab initio solution (Figure 9) from powder Xray data, a clear determination of the proton sites was not possible. Excluding the location of the protons, the organic ligands appear as two centrosymmetric phosphonic residues comprising a μ3-O,O′,O″-coordination of the two crystallographically independent phosphonic residues.

Figure 2. SEM images of phase-pure [La2(H2pmd)(pmd)(H2O)2] (2) prepared under microwave irradiation at 150 °C using different reaction periods. Despite the variation of reaction time (from 1 to 10 min), significant variations in crystallite habit are not observable: the particle size and platelike morphology remain approximately identical.

(pmd)(H 2 O) 2 ] (2), [La(H 3 pmd)(H 2 pmd)(H 2 O)] (3), [La2(H2pmd)3(H2O)2] (4), and [La2(H2pmd)(pmd)(H2O)4] (5) (where H4pmd stands for 1,4-phenylenebis(methylene)diphosphonic acid). As it is shown by the formulas, all compounds appear in three different La/H4−xpmd−x ratios: 1:1 for 2 and 5; 1:1.5 for 1 and 4; and 1:2 for 3. Except for compounds 1 and 5, which have higher contents of water molecules per metal center (6 and 2, respectively), compounds typically have a sole water molecule per metal center. The coordination number of the La3+ centers also varies and is related to the kind and hapticity of the coordinated ligands. In 1, the coordination geometry around the nonacoordinated metal center resembles a tricapped trigonal prism formed by the coordination to six water molecules and three crystallographically independent κ1-O-hydrogenophosphonato moieties (Figure 4; Figure S4 in the Supporting Information). The coordination number of nine is only possible given the presence of a large number of small water ligands and a low number of bulk H4−xpmd−x residues. In compound 2, La3+ is octacoordinated to five phosphonic residues, two of them being O,O′-chelating, and to one molecule of water. If we consider the chelating phosphonates as single coordination sites, the geometry around the metal center can be seen as a highly F

dx.doi.org/10.1021/cg301112k | Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design

Article

Figure 4. (a) Schematic representation of the inorganic layer of [La2(H2pmd)3(H2O)12] (1) perpendicular to the c axis of the unit cell. The {LaO9} cores are represented as green polyhedra. Please note: because phosphonate groups do not act as bridges between La3+ cations, the La···La intermetallic distances are represented by dashed green lines in an offset fashion. (b) Coordination environment around metallic center of 1 showing all non-hydrogen atoms as displacement ellipsoids drawn at the 70% probability level and hydrogen atoms as small spheres with arbitrary radii. For selected bond lengths and angles, see Table S1 and Figure S4 (in the Supporting Information). Symmetry code used to generate equivalent atoms: (i) x, y, −1 + z.

Figure 5. (a) Schematic representation of the inorganic layer of [La2(H2pmd)(pmd)(H2O)2] (2) perpendicular to the c axis of the unit cell. The {LaO8} cores are represented as green polyhedra. Please note: La···La intermetallic distances imposed by bridging phosphonate groups are represented by green lines. (b) Coordination environment around metallic center of 2 showing all non-hydrogen atoms as displacement ellipsoids drawn at the 70% probability level and hydrogen atoms as small spheres with arbitrary radii. For selected bond lengths and angles, see Table S3 and Figure S6 (in the Supporting Information). Symmetry codes used to generate equivalent atoms: (ii) −1 + x, y, z; (iii) 2 − x, 2 − y, −z; (iv) 1 − x, 1 − y, −z.

slightly distorted close hexagonal packing [other La···La distances range between 6.9514(11) and 7.6851(12) Å] (Figures 4 and 11). Only in 2, there is a connection between the {LaO8} polyhedra forming a zigzag chain parallel to the a axis of the unit cell, with the La···La intermetallic distances being 4.1819(12) and 4.5705(12) Å (Figures 5 and 11). While the μ2-O,μ2-O′,κ1-O″-phosphonate moiety is embedded in this chain, contributing to its stability, the μ2-O,O′-hydrogenophosphanate moiety acts as an interconnection between chains, giving the 2D aspect to the inorganic layer. In 3, the isolated {LaO7} polyhedra are arranged in the inorganic layer in a rectangular grid with the La···La distances being 5.717(2) and 8.694(4) Å. In this compound, the inorganic layers are disposed in an ABAB fashion, where each vertex of the grid is aligned with the center of the layer immediately above and below, instead of being completely aligned along the c axis as it

Given its high number of water ligands, compound 1 is a 1D polymer, in which two metallic coordination polyhedra alternate with three bridging H2pmd2− ligands (Figure 10). The remaining compounds are 3D structures. Despite having different stoichiometries, coordination numbers, and dimensionalities, the crystal packing of the five compounds is based on the alternation of inorganic layers, containing the {LaOn} polyhedra, with organic layers comprising the organic linkers. In compounds 1−4, the lanthanide phosphonate layers are disposed perpendicular to the c axis, whereas in 5, it is along the a axis (Figures S5 and S9 in the Supporting Information, respectively). Along the inorganic layer, the geometrical arrangement of {LaOn} polyhedra and their connection with the phosphonate residues are different for the five structures. In 1, the 1D polymer shows two nondirectly bonded metal centers at 6.2410(10) Å from each other, which are arranged in a G

dx.doi.org/10.1021/cg301112k | Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design

Article

Figure 6. (a) Schematic representation of the inorganic layer of [La(H3pmd)(H2pmd)(H2O)] (3) perpendicular to the c axis of the unit cell. The {LaO7} cores are represented as green polyhedra. The water molecule bonded to the metal center is disordered over two symmetry-related positions with half occupancy for each. Please note: La···La intermetallic distances are represented as filled or dashed green lines in the presence or otherwise, respectively, of phosphonate bridges between metallic centers. (b) Coordination environment around metallic center of 3 showing all non-hydrogen atoms as displacement ellipsoids drawn at the 70% probability level and hydrogen atoms as small spheres with arbitrary radii. For selected bond lengths and angles, see Table S5 and Figure S7 (in the Supporting Information). Symmetry codes used to generate equivalent atoms: (v) 1/2 − x, y, 1/2 − z; (vi) −1/2 − x, y, 1/2 − z; (vii) 1 + x, y, z.

Figure 7. (a) Schematic representation of the inorganic layer of [La2(H2pmd)3(H2O)2] (4) perpendicular to the c axis of the unit cell. The {LaO7} cores are represented as green polyhedra. Please note: La···La intermetallic distances imposed by bridging phosphonate groups are represented by green lines. (b) Coordination environment around metallic center of [La2(H2pmd)3(H2O)2] (4) showing all non-hydrogen atoms as displacement ellipsoids drawn at the 50% probability level and hydrogen atoms as small spheres with arbitrary radii. One of the hydrogen atoms attached to O1W is disordered over two positions with half occupancy for each. For selected bond lengths and angles, see Table S6 and Figure S8 (in the Supporting Information). Symmetry codes used to generate equivalent atoms: (viii) 2 − x, 1 − y, 2 − z; (ix) 2 − x, 1 − y, 1 − z; (x) 1 + x, y, −1 + z; (xi) 2 − x, −y, 1 − z.

along the a axis, that is, between each three consecutive inorganic layers. The first pair is connected by the ligand containing atom P1, and the other is connected by the ligand containing atom P2, or vice versa. 2.6. Topological Studies. The four novel three-dimensional [La 2 (H 2 pmd)(pmd)(H 2 O) 2 ] (2), [La(H 3 pmd)(H 2 pmd)(H 2 O)] (3), [La 2 (H 2 pmd) 3 (H 2 O) 2 ] (4), and [La2(H2pmd)(pmd)(H2O)4] (5) frameworks reported in this paper can be more conveniently described by employing a typical topological approach.31 This method consists of reducing each network into simple mathematical objects that structurally could be envisaged as central nodes and connecting bridges (between two or more individual nodes). All

happens with the remaining four compounds (Figures 6 and 11). In 4, the {LaO7} polyhedra are arranged in the inorganic layer in an elongated open hexagonal packing with the hole centered in the unit cell coordinates (1/2, 0, 1/2). This is a hydrophilic hole toward which the coordinated water molecules and the hydroxyl groups from H2pmd2− are directed. The distances between vertices of the hexagonal pattern (the metal centers) range from 4.756(8) to 6.237(10) Å (Figures 7 and 11). In 5, the {LaO8} polyhedra are arranged in a hexagonal fashion along the inorganic layer with La···La distances ranging from ca. 5.46 to 6.34 Å (Figure 8; Figure S9 in the Supporting Information). The two crystallographically independent phosphonate residues are distributed in alternated layers H

dx.doi.org/10.1021/cg301112k | Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design

Article

Figure 8. (a) Schematic representation of the inorganic layer of [La2(H2pmd)(pmd)(H2O)4] (5) perpendicular to the a axis of the unit cell. The {LaO8} cores are represented as green polyhedra. Please note: La···La intermetallic distances imposed by bridging phosphonate groups are represented by green lines. (b) Coordination environment around metallic center of [La2(H2pmd)(pmd)(H2O)4] (5) showing all atoms as small spheres with arbitrary radii. For selected bond lengths and angles, see Table S8 and Figure S9 (in the Supporting Information). Symmetry codes used to generate equivalent atoms: (xii) x, 1 + y, z; (xiii) x, −1/2 − y, −1/2 + z; (xiv) x, −1/2 − y, 1/2 + z.

Figure 9. Final Rietveld plot (powder X-ray diffraction data) of a mixture of [La2(H2pmd)(pmd)(H2O)2] (2) and [La2(H2pmd)(pmd)(H2O)4] (5) (8.4% and 91.6%, respectively, from total integrated intensity). Observed data points are indicated as red circles; the best-fit profile (upper trace) and the difference pattern (lower trace) are drawn as solid black and blue lines, respectively. Green and orange vertical bars indicate the angular positions of the allowed Bragg reflections for 2 and 5, respectively. Refinement details are given in Table 2. Inset: SEM image of a representative portion of the studied material (physical mixture of compounds 2 and 5) showing the platelike crystal habit of compound 2 and the block-type for compound 5.

even a full molecule) of two crystallographically independent H4−xpmd−x residues. In this context, and following the ideas proposed by Alexandrov et al., who considered that an organic ligand is of crucial structural importance in a framework when it establishes links between two or more metallic centers (μn),33 the centers of gravity of each pair of H4−xpmd−x residues were taken as network nodes alongside with the single La3+ metallic center. Most of the compounds are, in this way, trinodal: 2 is a 4,5,6-connected network with a total Schäfli symbol {44.62}{44.64}2{48.66.8}, in which the La3+ is five-connected (Figure

topological studies have been performed using the software package TOPOS.32 From the crystallographic studies performed on compounds 2−5 (see the previous section and the Supporting Information for a more detailed description of each structure), it is possible to discern a number of common structural features: (i) all compounds are assembled from a single crystallographically independent La3+ metallic center; (ii) H4−xpmd−x residues ensure physical connections between La3+ polyhedra; and (iii) the asymmetric unit of each compound contains fragments (or I

dx.doi.org/10.1021/cg301112k | Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design

Article

group.20d,26 In the present study, and as a corollary to the results summarized in the previous sections, we envisaged that the 1D compound [La2(H2pmd)3(H2O)12] (1) could be employed as a starting material for the preparation of higher dimensionality materials, in particular, some of those obtained using slightly distinct conditions. For this purpose, we have explored three distinct experimental conditions: (i) hydrothermal reactions at 180 °C for 3 days; (ii) microwave synthesis at 120 °C for 15 min; and (iii) one-pot approaches at 60 °C for different periods of time (2, 6, and 18 h). We refer the reader to the Materials and Methods section for further details on these investigations. Frameworks [La 2 (H 2 pmd)(pmd)(H 2 O) 2 ] (2) and [La2(H2pmd)(pmd)(H2O)4] (5) were typically formed from the conversion of the 1D compound 1, with compound 2 being even isolated as a phase-pure material (Scheme 5). When using the hydrothermal and the microwave methods (with the exception of the synthesis with hydrochloric acid at hydrothermal conditions with quenching), there is strong evidence of incomplete material conversion (Figures S14 and S15 in the Supporting Information). Moreover, in the hydrothermal approach, the cooling procedure after the reaction has a significant effect in the final product because the quenching of the reactions favors the formation of 2, whereas a slow cooling favors the formation of 5 (Figure S14 in the Supporting Information). On the other hand, the one-pot methodology permits a full conversion of the 1D polymer into a mixture of 2 and 5, with the exception of the synthesis with hydrochloric acid at 18 h of reaction time giving pure 2 (Figure S16 in the Supporting Information). To gain better insight into the 1D → 3D transformation, we have proceeded with detailed investigations on the reaction media. The empirical formula of 1 has an excess of ligand in comparison with 2 and 5, and thus, structural transformation should be accompanied by the release of the extra linkers. Indeed, the solid recovered from the evaporation of the solvent of the transformation study clearly revealed the presence of crystalline H4pmd, as ascertained from powder X-ray diffraction and FT-IR studies (Figures S17 and S18 in the Supporting Information). The one-pot transformation reactions clearly show that there is always the presence of an insoluble material in the vessel (see transformation video provided in the Supporting Information). However, as this transformation occurs with a modification of the empirical formula, we must assume at this stage that the process may occur via a dissolution−precipitation process and not through a singlecrystal-to-single-crystal transformation. 2.7.2. Structural Considerations. From a first approach, it seems that 1 shares some structural features with 2 and 5, namely, the intercalation of inorganic layers comprising {LaOn} polyhedra with layers comprising the ligands, and also the registered intermetallic distances. The three compounds comprise short La···La distances of 6.2410(10), 5.7098(14), and ca. 5.84 Å for 1, 2, and 5, respectively, and long La···La distances of 12.7044(19), 11.822(2), and ca. 12.18 Å, for 1, 2, and 5, respectively. When the structure of 1 is compared in detail with those of 2 and 5, we note that the two structures are highly superimposable, despite their different La:H4−xpmd−x proportions in the empirical formula. From only a structural point of view, 1 can be transformed into 2 or 5 by the formation of new O−P−O bridges and the intercalation of additional La3+ centers between the 1D polymers, while eliminating the excess of water molecules. The need for the

Figure 10. Schematic representation of a portion of the 1D polymer present in compound [La2(H2pmd)3(H2O)12] (1), viewed in perspective along the [100] direction of the unit cell. The 1D polymer is formed by the alternation between two metallic coordination polyhedra and three bridging H2pmd2− ligands.

Figure 11. Crystal packing of compounds 1−4 viewed in perspective along the [001] direction of the unit cell.

S10 in the Supporting Information); 3 is, instead, a 2,4,6connected network (Schäfli symbol: {44.62.89}{44.62}{8}) in which La3+ is six-connected (Figure S11 in the Supporting Information); and 4 appears as a 4,6,6-connected network (Schäfli symbol: {44.62}{45.6}2{47.68}2) with the metallic center being also six-connected (Figure S12 in the Supporting Information). Remarkably, compound 5 is topologically a uninodal framework as all nodes exhibit the same connectivity (6-connected) and point symbol ({412.63}), being a pcu α-Po primitive topology (Figure S13 in the Supporting Information). Compounds 2 and 4 also have known MOF topologies deposited within the TOPOS database (4,5,6T11 and 4,4,6T18, respectively), whereas material 3 corresponds to a completely new topological type. Indeed, searches in the Reticular Chemistry Structure Resource (RCSR)34 and in EPINET35 reveal that the nodal connectivity of this latter framework, to the best of our knowledge, is unprecedented among MOF structures. 2.7. Conversion of 1D into 3D MOFs. 2.7.1. Experimental Studies. The possibility of conversion of low dimensionality networks (e.g., 1D or 2D) into 3D framework architectures has already been explored by our research J

dx.doi.org/10.1021/cg301112k | Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design Scheme 5. 1D Conversion Pathways Using

a

Article

Hydrothermal,

One-Pot Synthetic Methodsa

Microwave, and

Legend: (purple) phase-pure 1; (green) phase-pure 2; (red) [La2(H2pmd)(pmd)(H2O)4] (5).

addition of La3+ is compatible with the higher content of these cations in the crystal structures of 2 and 5 in comparison with 1. Nevertheless, a detailed inspection reveals that 2 and 5 show different structural features. In compound 2, the coordination spheres around the metal centers are fused through a bridging oxygen atom, whereas in 5, all the oxygen atoms from the −PO3 moieties are directly connected to the metal center through single La−O bonds. Hence, we conclude that compound 1 should transform in an irreversibly fashion into 2 and 5, with interconversion between the last two products being structurally difficult. In short, the actual mechanism of the observed transformations remains unknown, and several pathways are possible for the transformation of 1 into 2 or 5. On the basis of the experimental data, we are inclined to the hypothesis of the disassembling of 1 and reaction of the recycled reagents to produce the final 3D networks. However, it is also not possible to exclude, in an unequivocal fashion, the partial disassembling of 1 in order to provide enough La3+ cations to react with the whole polymer of 1. Further investigations in this area are, thus, pertinent.

Compounds [La2(H2pmd)3(H2O)12] (1) and [La2(H2pmd)(pmd)(H2O)2] (2), 1D and 3D architectures, respectively, were successfully isolated as phase-pure compounds using both synthetic approaches and were the subject of detailed structural characterization studies. Crystals of [La(H3pmd)(H2pmd)(H2O)] (3) and [La2(H2pmd)3(H2O)2] (4) (obtained using microwave irradiation and from the one-pot approach, respectively) were manually harvested from physical mixtures, ultimately allowing their full structural elucidation using singlecrystal X-ray diffraction studies. Compound [La2(H2pmd)(pmd)(H2O)4] (5), also obtained in a one-pot approach, could only be isolated as a microcrystalline powder, and its crystal structure was determined instead by powder X-ray diffraction studies. It was found that mild synthetic conditions are the optimal conditions for both synthetic methods to promote the formation of the low dimensional compound 1, whereas the use of a higher temperature favored the formation of the 3D network 2. In that sense, we may look to 1 as the kinetic compound, whereas 2 seems to be the thermodynamic one. Investigations of the modification of the particle size and shape promoted by variation of the reaction temperature and time revealed that 1 crystallizes preferentially as aggregates of needle-like crystals, whereas 2 crystallizes as platelike particles. Furthermore, we concluded that there is no direct relationship between the variations of those parameters and the final average particle size. Compound 3 could only be isolated as a residual product in MWAS, and despite all the efforts, it was not possible to successfully isolate it as a phase-pure material. Likewise, the appearance of 4 and 5 (synthesized using the one-pot approach) was promoted by the use of high values of temperature (typically above 80 °C). Compound 4 was favored by the use of a higher ligand-to-metal proportion, and its

3. CONCLUDING REMARKS Five different 1D and 3D MOF structures based on lanthanum centers and on residues of the bipodal organic linker 1,4phenylenebis(methylene)diphosphonic acid (H4pmd) have been successfully prepared using two different synthetic methodologies: microwave irradiation (MWAS) and one-pot synthesis. For several years, our research group has been mainly focused on the use of hydro(solvo)thermal synthetic approaches and aiming at exploring new, faster, more economically viable, and environmentally friendly methods; we are now using in a systematic fashion the aforementioned approaches. K

dx.doi.org/10.1021/cg301112k | Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design

Article

Table 1. Crystal Data Collection and Structure Refinement Details for Compounds 1−4 formula fw temp/K cryst syst space group a/Å b/Å c/Å α/deg β/deg γ/deg volume/Å3 Z Dc/g cm−3 μ(Mo Kα)/mm−1 cryst size/mm cryst type θ range (deg) index ranges

reflns collected independent reflns completeness final R indices [I > 2σ(I)]a,b final R indices (all data)a,b weighting schemec largest diff. peak and hole CCDC no. a

1

2

3

4

C24H54La2O30P6 1286.31 150(2) triclinic P1̅ 7.6851(11) 11.6395(16) 12.7044(19) 105.665(7) 102.544(7) 93.172(6) 1060.2(3) 1 2.015 2.315 0.09 × 0.02 × 0.02 colorless needle 3.56−29.13 −10 ≤ h ≤ 10 −15 ≤ k ≤ 15 −17 ≤ l ≤ 17 28682 5655 [Rint = 0.0335] 98.9% (to θ = 29.06°) R1 = 0.0261 wR2 = 0.0657 R1 = 0.0310, wR2 = 0.0671 m = 0.0228 n = 2.0451 1.256 and −0.733 e Å−3 893006

C8H11LaO7P2 420.02 150(2) triclinic P1̅ 5.7098(9) 9.6259(14) 11.822(2) 104.874(11) 96.054(11) 105.347(10) 595.02(16) 2 2.344 3.881 0.02 × 0.02 × 0.01 colorless plate 3.63−29.06 −6 ≤ h ≤ 7 −13 ≤ k ≤ 13 −15 ≤ l ≤ 15 8322 2965 [Rint = 0.1007] 93.4% (to θ = 29.06°) R1 = 0.0654 wR2 = 0.1242 R1 = 0.1382, wR2 = 0.1451 m = 0.0654 n = 0.0000 2.031 and −2.855 e Å−3 893007

C16H23LaO13P4 686.13 150(2) monoclinic P2/n 5.717(2) 8.694(3) 22.649(8) 90.00 94.384(16) 90.00 1122.6(7) 2 2.030 2.256 0.06 × 0.06 × 0.02 colorless plate 3.58− 25.34 −6 ≤ h ≤ 6 −10 ≤ k ≤ 10 −27 ≤ l ≤ 26 11489 2019 [Rint = 0.1212] 97.7% (to θ = 25.34°) R1 = 0.1443 wR2 = 0.3267 R1 = 0.1684, wR2 = 0.3368 m = 0.0000 n = 156.2171 5.839 and −3.728 e Å−3 893008

C12H17LaO10P3 553.08 150(2) triclinic P1̅ 9.232(15) 9.263(15) 11.124(18) 90.15(5) 93.08(5) 110.77(5) 888(2) 2 2.069 2.726 0.02 × 0.01 × 0.01 colorless needle 3.67−25.34 −11 ≤ h ≤ 11 −8 ≤ k ≤ 11 −13 ≤ l ≤ 13 7165 3123 [Rint = 0.2430] 96.3% (to θ = 25.34°) R1 = 0.0982 wR2 = 0.1943 R1 = 0.2594, wR2 = 0.2636 m = 0.0975 n = 0.0000 1.967 and −2.194 e Å−3 893009

R1 = ∑∥Fo| − |Fc∥/∑|Fo|. bwR2 = (∑[w(F2o − F2c )2]/∑[w(F2o)])1/2. cw = 1/[σ2(F2o) + (mP)2 + nP, where P = (F2o + 2F2c )/3.

pure compound from employing either hydrothermal or onepot methodologies with hydrochloric acid as an additive. In these conditions, for the one-pot method, the reaction time increases the purity of the obtained phase while promoting a full conversion of the starting material. When the hydrothermal method was employed with hydrochloric acid as an additive, the cooling stage was found to be the most critical step: the slow cooling originates compound 5 with a large amount of starting material (almost no conversion occurs), whereas the quenching favors the isolation of phase-pure 2. Potential technological applications for the MOFs prepared in our laboratories have been the basis of the recent changes in our case study systems and the designing of novel ligands and structures. In that sense, we are currently investigating the isolation of materials isotypical with 1 and 2 using the procedures presented herein, but having different lanthanide centers. Our main goal is focused on the preparation of compounds with optically active lanthanide centers (based on Eu3+ and Tb3+), envisaging photoluminescent applications. At the same time, we are optimizing our synthetic approaches toward the reduction of the crystallite size, which we suppose will not only influence luminescent properties but also extend the range of applications (for catalytic processes, for instance).

formation at higher temperatures could be related to its 1:1.5 La:H4pmd stoichiometry in the reactive gel. MWAS and the one-pot methodologies have considerable advantages over the typical solvo(hydro)thermal synthesis. In terms of reaction time, and as an attempt to solve the typical problem of using long reaction periods in the latter case, we were able to drastically decrease the typical 3 days initially used to prepare compound 2 to just a single minute of reaction under microwave irradiation or, alternatively, to 10 min using the one-pot methodology. Noteworthy, this drastic reduction in the reaction time has a direct and huge influence on the total energy consumed to prepare the MOF compound. For instance, considering the above reaction periods and power input/output values associated with each of the three synthetic methods, we were able to estimate the energy consumption, which ranged from ca. 340 000 kJ (3 days of hydrothermal conditions) to ca. 18.0 kJ (for 1 min in MWAS) and to ca. 8.6 kJ (for 10 min in one-pot). Therefore, these two alternative synthetic approaches to the typical hydrothermal reaction allow us, on one hand, to probe the phase diagram of one given system for other architectures and, on the other hand, may also be considered for a possible implementation of the synthesis of MOFs at an industrial scale as they are cost efficient. We have also investigated the conversion of 1D structures into typical 3D MOF networks. The 1D compound 1 was readily converted into two of the aforementioned compounds (2 and 5), but only compound 2 could be isolated as a phase-

4. MATERIALS AND METHODS 4.1. General Instrumentation. SEM (scanning electron microscopy) images were acquired using either a Hitachi S4100 field L

dx.doi.org/10.1021/cg301112k | Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design

Article

Gasses required: carrier, helium; combustion, oxygen; pneumatic, compressed air. Routine powder X-Ray diffraction (PXRD) data for all synthesized materials were collected at ambient temperature on an X′Pert MPD Philips diffractometer (Cu Kα1,2 X-radiation; λ1 = 1.540598 Å, λ2 = 1.544426 Å), equipped with an X′Celerator detector and a flat-plate sample holder in a Bragg−Brentano para-focusing optics configuration (45 kV, 40 mA). Intensity data were collected by the step-counting method (step 0.04°), in continuous mode, in the ca. 3.5 ≤ 2θ ≤ 40° range. 1 H, 13C, and 31P nuclear magnetic resonance (NMR) spectra were recorded with a Bruker AVANCE 300 spectrometer at 300.13, 75.47, and 121.49 MHz, respectively. Deuterated dimethylsulfoxide-d6 (DMSO-d6) was used as the solvent, and tetramethylsilane or H3PO4 (85%) was used as an internal reference. Chemical shifts (δ) are quoted in parts per million and the coupling constants (J) in Hz. 13 C CP MAS and 1H or 31P MAS spectra were recorded at 9.4 T on a Bruker Avance 400 wide-bore spectrometer (DSX model) on a 4 mm BL cross-polarization magic-angle spinning (CPMAS) VTN probe at 100.6, 400.1, and 161.9 MHz, respectively. For the 13C{1H} CP MAS spectra, the Hartmann−Hahn (HH) “sideband” matching condition ν113C = ν11H + nνR (n = ± 1) was carefully matched by calibrating the 1H and the 13C rf field strengths: recycle delay, 5 s; contact time, 2 ms; νR = 8 or 12 kHz. For the 31P HPDEC spectra, a 90° single pulse excitation of 3.0 μs was employed: recycle delay, 60 s; νR= 8 or 12 kHz. Chemical shifts are quoted in parts per million with respect to TMS for the 1H and 13C nuclei, and to an 85% H3PO4 solution for the 31P nucleus. 4.2. Reagents. Chemicals were readily available from commercial sources and were used as received without further purification: lanthanum(III) chloride heptahydrate (LaCl3·7H2O, 99.9%, SigmaAldrich); tetraethyl 1,4-phenylenebis(methylene)bisphosphonate (C16H28O6P2, >95%, TCI Europe); hydrochloric acid (HCl, ≥37%, Sigma-Aldrich); dichloromethane (CH2Cl2, pure, Sigma-Aldrich); dimethyl sulfoxide-d6 (DMSO-d6, 99%, CIL - Cambridge Isotope Laboratories, Inc.); potassium bromide (KBr, for infrared spectroscopy, BDH SpectrosoL); sodium hydroxide (NaOH, ≥98%, SigmaAldrich); and potassium hydroxide (KOH, 85% p.a., Riedel-de Haën). Studies based on the one-pot method also used a stock solution of lanthanum chloride that was prepared using La2O3 (Inframat Advanced Materials, 99.995%): La2O3 was dissolved in a solution of hydrochloric acid for 2 h at 100 °C and diluted with distilled water to the desired concentration (0.04 M). 4.3. Acid Hydrolysis of Tetraethyl 1,4-Phenylenebis(methylene)bisphosphonate. Tetraethyl 1,4-phenylenebis(methylene)bisphosphonate (500 mg, 1.32 mmol) was added to an aqueous solution of HCl (30 mL, 6 M). The reaction mixture was kept under constant magnetic stirring and heated to reflux. The evolution of the reaction was controlled by TLC and stopped after approximately 20 h, when the hydrolysis was complete. Impurities were successfully eliminated via a liquid/liquid extraction with dichloromethane. Water was then removed under reduced pressure, and the final compound, 1,4-phenylenebis(methylene)diphosphonic acid (H4pmd), was obtained in quantitative amounts (yield of about 96%). Elemental analysis. Calcd (in %): C, 36.10; H, 4.54. Found: C, 35.60; H, 4.49. 1 H NMR (300.13 MHz, DMSO-d6) δ: 2.91 (d, 4H, J(1H−31P) = 19.9 Hz, CH2) and 7.15 (s, 4H, Ar-H). 13 C NMR (75.47 MHz, DMSO-d6) δ: 35.0 (d, J(13C−31P) = 132.6 Hz, CH2), 129.4 (Ar-CH) and 131.8 (Ar-C). 31 P NMR (121.49 MHz, DMSO-d6) δ: 21.7 (t, J(31P−1H) = 22.2 Hz, PO3H2). Selected FT-IR data (cm−1): attribution of the most prominent bands of H4pmd was performed from a comparison between the FTIR spectrum of this molecule (collected as a KBr pellet) and values found in the literature.36 H4pmd: ν(POH) = 2329 w, ν(Ph) = 1508 m and 1425 w, δ(CH) = 1401 w, δ(POH) = 1265 vs and 1231 s, ν(PO) = 1115 vs, δ(POH) = 998 vs, γ(CH) = 942 vs, ρ(CH2) + γ(CH) = 803 m, γ(Ph) + ν(PO3) = 558 vs and 427 s, ρ(CH2) + δ(PO3) = 349 m.

Table 2. X-ray Data Collection, Crystal Data, and Structure Refinement Details for [La2(H2pmd)(pmd)(H2O)4] (5) (Present in a Physical Mixture with Compound 2) unit cell C8H5LaO8P2a 429.97a monoclinic P21/c 21.9799(7) 5.4581(2) 10.8462(3) 94.285(2) 1297.57(7) 4 2.201a

formula fw cryst syst space group a/Å b/Å c/Å β/deg volume/Å3 Z Dc/g cm−3 profile parameters profile function Caglioti law params

pseudo-Voigt η = 0.407(9) U = 0.016(4) V = −0.002(3) W = 0.0157(3) 0.062(1) and 0.0413(4) −0.0758(7)

asymmetry params (up to 20° 2θ) zero shift [2θ°] refinement details no. of independent reflns 2214 (2: 1097; 5:1117) no. of global refined params 1 no. of profile refined params 14 no. of intensity-dependent refined params 62 reliability factors for all nonexcluded data points with Bragg contribution (conventional − not corrected for background) Rp 7.14 Rwp 9.23 Rexp 7.33 χ2 1.59 structure reliability factors (only for compound 5) RBragg 11.4 RF 16.3 indexing figures-of-merit M(n) 28.0 (n = 20) a

Hydrogen atoms associated with the water molecules, the aromatic rings, and the protonated POH groups were added to the empirical formula.

emission gun tungsten filament instrument working at 25 kV or a highresolution Hitachi SU-70 working at 4 kV. Samples were prepared by deposition on aluminum sample holders, followed by carbon coating using an Emitech K950X carbon evaporator. EDS (energy-dispersive X-ray spectroscopy) data and SEM mapping images were recorded using the same microscope, a Hitachi SU-70, working at 15 kV and using a Bruker Quantax 400 or a Sprit 1.9 EDS microanalysis system. Thermogravimetric analyses (TGA) were carried out using a Shimadzu TGA 50, from room temperature to ca. 800 °C, with a heating rate of 5 °C/min, under a continuous stream of air with a flow rate of 20 cm3/min. Fourier transform infrared (FT-IR) spectra (in the range of 4000− 400 cm−1) were recorded as KBr pellets (typically 2 mg of sample was mixed in a mortar with 200 mg of KBr) using a Bruker Tensor 27 spectrometer by averaging 256 scans at a maximum resolution of 2 cm−1. Elemental analyses for C and H were performed with a Truspec Micro CHNS 630-200-200 elemental analyzer at the Department of Chemistry, University of Aveiro. Analyses parameters: sample amount between 1 and 2 mg; combustion furnace temperature, 1075 °C; and afterburner temperature, 850 °C. Detection method: carbon, infrared absorption; hydrogen, infrared absorption. Analysis time: 4 min. M

dx.doi.org/10.1021/cg301112k | Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design

Article

Selected FT-IR data (cm−1): ν(POH) = 2363 w, ν(Ph) = 1515 m, δ(CH) = 1407 w, δ(POH) = 1248 vs and 976 s, ν(PO) = 1136 vs, ρ(CH2) + γ(CH) = 798 m, γ(Ph) + ν(PO3) = 563 vs. 4.5. Conversion of 1D to 3D MOFs. 4.5.1. Hydrothermal Synthesis. [La2(H2pmd)3(H2O)12] (1) (30 mg, 0.023 mmol) was suspended in distilled water (ca. 6 mL). KOH (9 mg, 0.13 mmol) or HCl (10 μL, 6 M) was added to some solutions. The reactive mixtures were then left stirring at ambient temperature for approximately 15 min, with the resulting suspensions being then placed inside an MMM Venticell oven. Reactions took place at 180 °C during 72 h. The oven was turned off, and the samples were allowed to cool slowly to ambient temperature (while inside the oven). Some vessels were also quenched by transferring the autoclaves, while at 180 °C, for a recipient with ice and water. After that, the final compounds were recovered by vacuum filtration as white microcrystalline powders, washed with abundant amounts of distilled water, and air-dried. 4.5.2. Microwave-Assisted Synthesis. Mixtures of [La2(H2pmd)3(H2O)12] (1) (30 mg, 0.023 mmol) in distilled water (ca. 6 mL) were prepared inside a 10 mL IntelliVent microwave reactor. KOH (9 mg, 0.125 mmol) or HCl (10 μL, 6 M) was added to some suspensions. The resulting mixtures were stirred at ambient temperature for 1 min and then placed inside a CEM Focused Microwave Synthesis System Discover S-Class equipment and kept under constant magnetic stirring (controlled by the microwave equipment). A constant flow of air (ca. 20−30 psi of pressure) ensured a close control of the temperature inside the reactor. Reactions took place at 120 °C for 15 min using 50 W of irradiation power. The final compounds were recovered, as described in section 4.5.1. 4.5.3. One-Pot Method. [La2(H2pmd)3(H2O)12] (1) (30 mg, 0.023 mmol) and distilled water (ca. 10 mL) were added to a round-bottom flask. KOH (9 mg, 0.13 mmol) or HCl (10 μL, 6 M) was added to some solutions. Reactions took place at 60 °C for different periods of time (2, 6, and 18 h). The final compounds were recovered, as described in section 4.5.1. 4.6. Single-Crystal X-ray Diffraction. Single crystals of compounds [La 2(H 2 pmd) 3 (H2 O) 12 ] (1), [La 2 (H2 pmd)(pmd)(H2O)2] (2), [La(H3pmd)(H2pmd)(H2O)] (3), and [La2(H2pmd)3(H2O)2] (4) were manually harvested from the crystallization vials and immediately immersed in highly viscous FOMBLIN Y perfluoropolyether vacuum oil (LVAC 140/13, SigmaAldrich) to avoid degradation caused by the evaporation of the solvent. Crystals of compounds 3 and 4 were particularly difficult to select because of the presence of other crystalline phases. Attempts were made to isolate the largest and most clear crystals, but those available consisted systematically of highly twinned crystals (i.e., not single) that most likely could also have incrustations of other phases (hence, the observed poor diffraction quality). Crystals were mounted on Hampton Research CryoLoops with the help of a Stemi 2000 stereomicroscope equipped with Carl Zeiss lenses.37 Data were collected on a Bruker X8 Kappa APEX II CCD area-detector diffractometer (Mo Kα graphite-monochromated radiation; λ = 0.71073 Å) controlled by the APEX2 software package38 and equipped with an Oxford Cryosystems Series 700 cryostream monitored remotely using the software interface Cryopad.39 Images were processed using the software package SAINT+,40 and data were corrected for absorption by the multiscan semiempirical method implemented in SADABS.41 Space group determination for compounds 1, 2, and 4 from the systematic absences and the determination of |E2 − 1| was straightforward. For 3, doubts arose between the noncentrosymmetric monoclinic space group Pn and the centrosymmetric space group P2/n because, on the one hand, they exhibit the same systematic absence (h0l) and, on the other hand, the overall poor diffraction quality of the crystal produced a data set with rather poor statistics. Crystal solution was then attempted for both space groups with stable refinements being only possible for the centrosymmetric P2/n one. Structures were solved using the Patterson synthesis algorithm implemented in SHELXS-97,42 which allowed the immediate location of the La3+ centers and most of the heaviest atoms. All remaining non-

4.4. Preparation of Phase-Pure MOFs Using Different Synthetic Methods. 4.4.1. Hydrothermal Synthesis of [La2(H2pmd)3(H2O)12] (2) Using Conventional Convection Heating. A reactive mixture of LaCl3·7H2O (0.4122 g, 1.11 mmol) with 1,4phenylenebis(methylene)diphosphonic acid (H4pmd) (0.2954 g, 1.11 mmol) in distilled water (ca. 12 mL), with a molar ratio of approximately 1:1:600 (La3+/H4pmd/H2O), was kept under constant magnetic stirring in open air at ambient temperature for approximately 30 min. The resulting homogeneous suspension was transferred to a Teflon-lined Parr Instrument reaction vessel and then placed inside an MMM Venticell oven. Reaction took place at 180 °C over a period of 72 h, after which time the oven was turned off and the samples were cooled slowly to ambient temperature (while inside the oven). After that, the final compound was vacuum filtrated, washed with copious amounts of distilled water, and dried at room temperature. The final product was a white crystalline material that was identified as [La2(H2pmd)3(H2O)12] (2). This material was shown to be the expected product for MWAS and one-pot synthesis at high temperature and ca. 1:1 for the La3+/H4pmd molar ratio (Schemes 3 and 4). 4.4.2. Microwave-Assisted Hydrothermal Synthesis. Mixtures of H4pmd (0.1477g, 0.55 mmol) and water (ca. 6 mL) were stirred at ambient temperature inside a 10 mL IntelliVent microwave reactor for ca. 5 min. LaCl3·7H2O (0.2061 g, 0.55 mmol) was then added, obtaining an overall molar ratio of ca. 1:1:600 (La3+/H4pmd/H2O) for the reactive mixture. The resulting homogeneous suspension was placed inside a CEM Focused Microwave Synthesis System Discover S-Class equipment and was kept under constant magnetic stirring (controlled by the microwave equipment). A constant flow of air (ca. 20−30 psi of pressure) ensured a close control of the temperature inside the reactor. Synthetic conditions were investigated by systematically varying the experimental parameters: (i) temperature (30, 40, 50, 60, 70, 90, 120, and 150 °C); (ii) power (10, 25, and 50 W); and (iii) irradiation time (1, 3, 5, and 10 min). After reacting, a white precipitate was obtained, and the final product was recovered by vacuum filtration, followed by washing with abundant amounts of distilled water, and then air-dried. 4.4.3. One-Pot Method. For a typical metal/ligand molar ratio of 1:1, a mixture containing ca. 0.1 mmol of an aqueous LaCl3·xH2O solution (5 mL of ca. 20 mmol/L solution) and ca. 5 mL of an aqueous solution of H 4pmd (0.0254 g, 0.0954 mmol) was homogenized in a round-bottom flask with constant magnetic stirring. The resulting solution was heated to the desired temperature and kept stirring during the required time, leading to the formation of a white precipitate. After cooling, the obtained solid product was centrifuged for 20 min at 3500 rpm, using a SIGMA 2-5 equipment, or filtrated under vacuum, washed with copious amounts of distilled water and ethanol, and allowed to air-dry at room temperature. As performed for the microwave-assisted approach, a systematic variation of the experimental conditions was performed by changing the following parameters: (i) La3+/H4pmd molar ratios (1:1 and 1:2); (ii) temperature (room temperature and 40, 60, and 80 °C); and (iii) reaction time (10, 20, 30, and 60 min). 4.4.4. Characterization Data for the Isolated Phase-Pure MOFs. [La2(H2pmd)3(H2O)12] (1) (MW = 1286.31). Elemental analyses. Calcd (%): C, 22.41; H, 4.23. Found: C, 22.45; H, 4.24. Thermogravimetric analysis (TGA) data (weight losses in %) and derivative thermogravimetric peaks (DTG; in italics inside the parentheses): 25−270 °C, −3.1% (246 °C); 450−640 °C, −4.1%; 640−800 °C, −18.5%. Selected FT-IR data (cm−1): ν(POH) = 2230 w, ν(Ph) = 1511 m, δ(CH) = 1403 w, δ(POH) = 1257 vs and 1014 s, ν(PO) = 1152 vs, ρ(CH2) + γ(CH) = 769 m, γ(Ph) + ν(PO3) = 566 vs. [La2(H2pmd)(pmd)(H2O)2] (2) (MW = 840.04). Elemental analyses. Calcd (%): C, 22.88; H, 2.64. Found: C, 22.85; H, 2.74. Thermogravimetric analysis (TGA) data (weight losses in %) and derivative thermogravimetric peaks (DTG; in italics inside the parentheses): 45−130 °C, −15.4% (91 °C); 200−400 °C, −3.1%; 490−800 °C, −19.4%. N

dx.doi.org/10.1021/cg301112k | Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design

Article

with the software program DICVOL04.45 The first 20 well-resolved reflections (located using the derivative-based peak search algorithm provided with Fullprof.2k)46 solely belonging to phase 5 were isolated (from a direct comparison with the reflections arising from a simulation based on the single-crystal data for 2). A fixed absolute error on each line of 0.03° 2θ was employed for the calculations, with initial unit cell metrics being obtained with reasonable figures-of-merit (see Table 2 for the M47 and F48 values). Analysis of the systematic absences was performed using the software package CHECKCELL,49 which allowed the identification of the monoclinic centrosymmetric P21/c space group as the most suitable to describe the symmetry of the compound. Noteworthy, the ADDSYM routines implemented in the software package PLATON50 indicate a possible change in overall symmetry to the C2/c space group. This option was, however, disregarded because, even though it explains the absence of the (300) reflection at a 2θ of ca. 12.1°, it also implies that the (−302) and (−211) reflections at a 2θ of ca. 19.7° would be absent, which is not the case (see Figure 9). Structure solution in P21/c was carried out using the direct methods of SIRPOW, included in the software package eXpo2004 (version 2.1).51 The presence of phase 2 as an impurity was disregarded at this stage because the small amount in which it appears was not considered a significant limitation in the structure solution process. In fact, eXpo2004 disregards in the Pawley extraction process isolated reflections arising from possible impurities. After subtraction of the background using polynomial functions in selected angular intervals, the intensity of each individual reflection was extracted by employing Pearson profile functions. This strategy allowed the location of all nonhydrogen atoms composing the asymmetric unit after a series of subsequent Rietveld refinements and peak search attempts in eXpo2004. Rietveld structural refinement52 was performed with FullProf.2k46 by applying fixed background points throughout the entire angular range determined by the linear interpolation between consecutive (and manually selected) breakpoints in the powder pattern. Typical pseudoVoigt profile functions to generate the line shapes of the simulated diffraction peaks were considered for both crystalline phases. The atomic coordinates of 2 were taken from the single-crystal X-ray diffraction studies. The initial atomic coordinates for 5 were those derived from eXpo2004. Because of the small amount of 2 in the physical crystalline mixture, most of the typical Rietveld parameters for this phase were heavily restrained (or even nonexisting) during the entire analysis: (i) Atomic coordinates and (ii) unit cell parameters were always fixed. (iii) No asymmetry parameters were considered, and only the W parameter of the Caglioti function correction53 was allowed to vary. For the desired compound 5, a pseudo-Voigt profile function was used, along with two asymmetry correction parameters, and the angular dependence of the individual reflections was taken into account by using the three parameters of the Caglioti function correction.53 The collected powder pattern exhibited a slight preferential orientation along the [200] vector for compound 5. A modified March’s correction function was employed, producing a refined G1 value of about 1.068(2), which is consistent, coherent with the blocktype crystal habit observed for the material (see SEM image as an inset picture in Figure 9). Overall structural refinement for the crystallographic model of compound 5 (as derived from eXpo2004) was performed in consecutive stages to avoid instability and divergence. Zero shift, scale factor, parameters related to peak shape, and unit cell parameters were consecutively added as fully refineable variables upon previous full convergence of the remaining parameters to their optimal values. Fractional atomic coordinates for all (non-hydrogen) atoms were ultimately allowed to refine in conjunction with weighted soft distance constraints solely applied to the organic ligand (C−C, C−P, P−O, O···O, and C···O bond lengths and internuclear distances, respectively). This approach has the advantage to ensure a final chemically feasible geometry for the two crystallographically independent centrosymmetric organic moieties while it allows the rare-earth environment to refine freely. This treatment also notably

hydrogen atoms were located from difference Fourier maps calculated from successive full-matrix least-squares refinement cycles on F2 using SHELXL-97.42b,43 When possible, non-hydrogen atoms were refined using anisotropic displacement parameters. The carbon and oxygen atoms in compound 3 did not refine properly using anisotropic displacement parameters, and for this reason, an ISOR instruction was used in order to better describe the thermal motion of these atoms. Hydrogen atoms bound to carbon were placed at their idealized positions using appropriate HFIX instructions in SHELXL:23 for the −CH2− methylene groups and 43 for those belonging to the aromatic rings. For compounds 1 and 2, all hydrogen atoms bound to oxygen from coordinated water molecules and hydroxyl moieties (from the protonated phosphonate groups) were found directly from difference Fourier maps and included in the final structural models with the O− H and H···H distances restrained to 0.95(1) and 1.55(1) Å, respectively, in order to ensure a chemically reasonable geometry for these moieties. For compound 4, and despite the overall poor quality of the data set, one of the hydrogen atoms attached to the O1W water molecule was found from difference Fourier maps, while the other hydrogen atom was geometrically placed in the two other possible locations (0.5 occupancy factor for each). As for compounds 1 and 2, the O−H and H···H distances have also been restrained using identical geometrical parameters. The remaining hydrogen atoms attached to oxygen atoms were geometrically placed using the HFIX instruction 147. The strategy of placing geometrically hydrogen atoms in the crystal structure of 3 did not produce a stable refinement with acceptable geometries. For this reason, hydrogen atoms have been omitted from the structural model, but they were included in the empirical formula. All hydrogen atoms were included in each structure in subsequent refinement cycles with isotropic thermal displacements parameters (Uiso) fixed at 1.2 × Ueq or 1.5 × Ueq, when attached to carbon or oxygen, respectively. Noteworthy, the crystal structures of compounds 3 and 4 exhibit a number of limitations mainly due to the poor quality of the available crystals. These limitations are individually addressed in the CIF files and the checkCIF reports, and we refer the reader to the Supporting Information for further details. Collection of better data sets for 3 and 4 seems to be only possible if better crystals are isolated (larger and without impurities) and, if size remains, using a synchrotron source. Nevertheless, we wish to note that the structural models derived from the collected data sets clearly support the information derived from other techniques, in particular, powder X-ray diffraction studies that were employed throughout this paper as a routine characterization technique. Crystallographic data collection and structure refinement details are summarized in Table 1. Selected bond lengths and angles for the metallic coordination environments and hydrogen-bonding geometry for compounds 1−4 are presented in Tables S1−S7 in the Supporting Information. Structural drawings have been produced using the software package Crystal Diamond.44 Crystallographic data (including structure factors) for compounds 1−4 have been deposited with the Cambridge Crystallographic Data Centre as supplementary publication Nos. CCDC-893006 (for 1), 893007 (for 2), 893008 (for 3), and 893009 (for 4). Copies of the data can be obtained free of charge on application to CCDC, 12 Union Road, Cambridge CB2 2EZ, U.K. Fax: (+44) 1223 336033. E-mail: [email protected]. 4.7. Structural Determination from Powder X-ray Diffraction of Compound 5. Conventional powder X-ray diffraction data suitable for structural solution and refinement of compound [La2(H2pmd)(pmd)(H2O)4] (5) (from a physical mixture with [La2(H2pmd)(pmd)(H2O)2] (2)) were collected at ambient temperature on a X′Pert MPD Philips diffractometer (Cu Kα1,2 X-radiation; λ1 = 1.540598 Å and λ2 = 1.544426 Å), equipped with an X′Celerator detector and a flat-plate sample holder in a Bragg−Brentano parafocusing optics configuration (45 kV, 40 mA). Intensity data were collected by the step-counting method (step 0.01°; 695 s per step), in continuous mode, in the ca. 5 ≤ 2θ ≤ 90° range. The collected powder X-ray diffraction pattern for the physical mixture of 5 with 2 was indexed by means of the routines provided O

dx.doi.org/10.1021/cg301112k | Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design

Article

leads, nevertheless, to very close proximities between adjacent moieties, as clearly observed from the geometrical parameters of the various possible O−H···O hydrogen-bonding interactions present in the structure, whose internuclear distances range from 2.39(4) to 3.02(4) Å (in several cases being smaller than the sum of the van der Waals radii). Each family of atom type has been refined using common refineable isotropic parameters. No correction was made for absorption effects. Because of the limitations of the structural determination of 5 from powder studies, hydrogen atoms have been omitted from the final structural model. We note, however, the presence of various possible hydrogen-bonding interactions connecting water molecules to neighboring phosphonate groups and also between adjacent phosphonate groups. The most striking interaction is related to a very close proximity between O3 and O4 atoms (2.39(4) Å), both coordinated to lanthanum cations. Indeed, to balance the crystal charge of [La2(H2pmd)(pmd)(H2O)4] (5), one hydrogen atom needs to be bound to a phosphonate group. Because of the close proximity between these oxygen atoms, we assume that a rather strong and directional hydrogen bond has to occur so as to stabilize the crystal structure. The final Rietveld refinements were performed with all relevant parameters for compound 5 allowed to refine freely, ultimately converging to the profile and reliability factors summarized in Table 2. The composition of the physical mixture was refined from the total integrated intensity of the powder pattern to the values given as an inset in Figure 9. Crystallographic data (excluding structure factors) for the powder X-ray diffraction determination of compound 5 have been deposited with the Cambridge Crystallographic Data Centre as supplementary publication No. CCDC-893010. Copies of the data can be obtained free of charge on application to CCDC, 12 Union Road, Cambridge CB2 2EZ, U.K. (Fax: (+44) 1223 336033; E-mail: [email protected]. ac.uk).



single-crystal diffractometer. We are also grateful to FCT for the Ph.D. scholarships, Nos. SFRH/BD/46601/2008 (to P.S.) and SFRH/BD/66371/2009 (to S.M.F.V.), and the postdoctoral grant SFRH/BPD/63736/2009 (to J.A.F.).



(1) (a) Gagnon, K. J.; Perry, H. P.; Clearfield, A. Chem. Rev. 2012, 112, 1034−1054. (b) Gliemann, H.; Wall, C. Mater. Today 2012, 15, 110−116. (c) O’Keeffe, M.; Yaghi, O. M. Chem. Rev. 2012, 112, 675− 702. (d) Stock, N.; Biswas, S. Chem. Rev. 2012, 112, 933−969. (2) Paz, F. A. A.; Klinowski, J.; Vilela, S. M. F.; Tomé, J. P. C.; Cavaleiro, J. A. S.; Rocha, J. Chem. Soc. Rev. 2012, 41, 1088−1110. (3) Deng, H. X.; Grunder, S.; Cordova, K. E.; Valente, C.; Furukawa, H.; Hmadeh, M.; Gándara, F.; Whalley, A. C.; Liu, Z.; Asahina, S.; Kazumori, H.; O’Keeffe, M.; Terasaki, O.; Stoddart, J. F.; Yaghi, O. M. Science 2012, 336, 1018−1023. (4) (a) Senkovska, I.; Barea, E.; Navarro, J. A. R.; Kaskel, S. Microporous Mesoporous Mater. 2012, 156, 115−120. (b) Wei, W.; Chen, S.; Wei, Q.; Xie, G.; Yang, Q.; Gao, S. Microporous Mesoporous Mater. 2012, 156, 202−208. (c) Yuan, B. Z.; Ma, D. Y.; Wang, X.; Li, Z.; Li, Y. W.; Liu, H. M.; He, D. H. Chem. Commun. 2012, 48, 1135− 1137. (d) Senkovska, I.; Barea, E.; Navarro, J. A. R.; Kaskel, S. Microporous Mesoporous Mater. 2012, 156, 115−120. (e) Montoro, C.; Linares, F.; Procopio, E. Q.; Senkovska, I.; Kaskel, S.; Galli, S.; Masciocchi, N.; Barea, E.; Navarro, J. A. R. J. Am. Chem. Soc. 2011, 133, 11888−11891. (f) Procopio, E. Q.; Linares, F.; Montoro, C.; Colombo, V.; Maspero, A.; Barea, E.; Navarro, J. A. R. Angew. Chem., Int. Ed. 2010, 49, 7308−7311. (g) Galli, S.; Masciocchi, N.; Colombo, V.; Maspero, A.; Palmisano, G.; Lopez-Garzon, F. J.; Domingo-Garcia, M.; Fernandez-Morales, I.; Barea, E.; Navarro, J. A. R. Chem. Mater. 2010, 22, 1664−1672. (5) (a) Dang, D. B.; Wu, P. Y.; He, C.; Xie, Z.; Duan, C. Y. J. Am. Chem. Soc. 2010, 132, 14321−14323. (b) Kim, J.; Cho, H.-Y.; Ahn, W.S. Catal. Surv. Asia 2012, 16, 106−119. (6) Ranjbar, Z. R.; Morsali, A. Inorg. Chim. Acta 2012, 382, 171−176. (7) (a) Dusan, D.; Youcef, H.; Lin, C. L.; Tan, Y.; Kunhao, L.; Jing, L. J. Appl. Phys. 2012, 111, 07E335. (b) Maspoch, D.; Ruiz-Molina, D.; Wurst, K.; Domingo, N.; Cavallini, M.; Biscarini, F.; Tejada, J.; Rovira, C.; Veciana, J. Nat. Mater. 2003, 2, 190−195. (8) (a) Pereira, G. A.; Peters, J. A.; Paz, F. A. A.; Rocha, J.; Geraldes, C. Inorg. Chem. 2010, 49, 2969−2974. (b) Della Rocca, J.; Lin, W. Eur. J. Inorg. Chem. 2010, 2010, 3725−3734. (c) Liu, D. M.; Huxford, R. C.; Lin, W. B. Angew. Chem., Int. Ed. 2011, 50, 3696−3700. (9) (a) Rocha, J.; Carlos, L. D.; Paz, F. A. A.; Ananias, D. Chem. Soc. Rev. 2011, 40, 926−940. (b) Harbuzaru, B. V.; Corma, A.; Rey, F.; Jordá, J. L.; Ananias, D.; Carlos, L. D.; Rocha, J. Angew. Chem., Int. Ed. 2009, 48, 6476−6479. (c) Hinterholzinger, F. M.; Ranft, A.; Feckl, J. M.; Ruhle, B.; Bein, T.; Lotsch, B. V. J. Mater. Chem. 2012, 22, 10356− 10362. (10) (a) Férey, G. Chem. Soc. Rev. 2008, 37, 191−214. (b) Mueller, U.; Schubert, M.; Teich, F.; Puetter, H.; Schierle-Arndt, K.; Pastre, J. J. Mater. Chem. 2006, 16, 626−636. (c) Rowsell, J. L. C.; Yaghi, O. M. Microporous Mesoporous Mater. 2004, 73, 3−14. (11) Forster, P. M.; Thomas, P. M.; Cheetham, A. K. Chem. Mater. 2002, 14, 17−20. (12) (a) Choi, J. S.; Son, W. J.; Kim, J.; Ahn, W. S. Microporous Mesoporous Mater. 2008, 116, 727−731. (b) Choi, J. Y.; Kim, J.; Jhung, S. H.; Kim, H. K.; Chang, J. S.; Chae, H. K. Bull. Korean Chem. Soc. 2006, 27, 1523−1524. (c) Jhung, S. H.; Lee, J. H.; Forster, P. M.; Férey, G.; Cheetham, A. K.; Chang, J. S. Chem.Eur. J. 2006, 12, 7899−7905. (d) Jhung, S. H.; Lee, J. H.; Yoon, J. W.; Serre, C.; Férey, G.; Chang, J. S. Adv. Mater. 2007, 19, 121−124. (e) Wang, X. F.; Zhang, Y. B.; Huang, H.; Zhang, J. P.; Chen, X. M. Cryst. Growth Des. 2008, 8, 4559−4563. (13) (a) Park, S. E.; Chang, J. S.; Hwang, Y. K.; Kim, D. S.; Jhung, S. H.; Hwang, J. S. Catal. Surv. Asia 2004, 8, 91−110. (b) Manjan-Sanz, A.; Sanchez-Sanchez, M.; Sastre, E. Catal. Today 2012, 179, 102−114.

ASSOCIATED CONTENT

S Supporting Information *

Additional experimental details and structural characterization data: 1H, 13C, 31P NMR of H4pmd; electron microscopy studies, including EDS mapping of compounds 1, 2, 4, and 5; thermogravimetric experiments, FT-IR spectra, and solid-state NMR of 1 and 2. Crystallographic information comprising a detailed description of the crystal structures herein reported for compounds 1−5: coordination geometry, bond lengths, and angles for the coordination environments and hydrogenbonding geometry, schematic representations of the asymmetric units, and the crystal packing of compounds 1−4 viewed along [001] and along [010] for 5. Crystallographic information files (CIF) for compounds 1−5. This material is available free of charge via the Internet at http://pubs.acs.org.



REFERENCES

AUTHOR INFORMATION

Corresponding Author

*E-mail: fi[email protected]. Fax: (+351) 234 401470. Phone: (+351) 234 401418. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS We would like to thank Fundaçaõ para a Ciência e a Tecnologia (FCT, Portugal), the European Union, QREN, FEDER, COMPETE, and Laboratório Associado Centro de Investigaçaõ em Materiais Cerâmicos e Compósitos, CICECO (PEst-C/ CTM/LA0011/2011) for their general funding scheme. We further wish to thank FCT for funding the R&D project PTDC/QUI-QUI/098098/2008 (FCOMP-01-0124-FEDER010785), and for specific funding toward the purchase of the P

dx.doi.org/10.1021/cg301112k | Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design

Article

(14) Klinowski, J.; Paz, F. A. A.; Silva, P.; Rocha, J. Dalton Trans. 2011, 40, 321−330. (15) Lin, Z. J.; Wragg, D. S.; Morris, R. E. Chem. Commun. 2006, 2021−2023. (16) Schlesinger, M.; Schulze, S.; Hietschold, M.; Mehring, M. Microporous Mesoporous Mater. 2010, 132, 121−127. (17) Sudik, A. C.; Côté, A. P.; Wong-Foy, A. G.; O’Keeffe, M.; Yaghi, O. M. Angew. Chem., Int. Ed. 2006, 45, 2528−2533. (18) (a) Imaz, I.; Hernando, J.; Ruiz-Molina, D.; Maspoch, D. Angew. Chem., Int. Ed. 2009, 121, 2361−2365. (b) Ma, M.; Zacher, D.; Zhang, X.; Fischer, R. A.; Metzler-Nolte, N. Cryst. Growth Des. 2011, 11, 185− 189. (c) Marandi, F.; Morsali, A. Inorg. Chim. Acta 2011, 370, 526− 530. (d) Wang, L.; Lu, S.; Zhou, Y.; Guo, X.; Lu, Y.; He, J.; Evans, D. G. Chem. Commun. 2011, 47, 11002−11004. (e) Zhang, L.; Hu, Y. H. Phys. Lett. A 2011, 375, 1514−1517. (19) (a) Cho, W.; Lee, H. J.; Oh, M. J. Am. Chem. Soc. 2008, 130, 16943−16946. (b) Tachikawa, T.; Choi, J. R.; Fujitsuka, M.; Majima, T. J. Phys. Chem. C 2008, 112, 14090−14101. (c) Imaz, I.; Maspoch, D.; Rodríguez-Blanco, C.; Pérez-Falcón, J. M.; Campo, J.; Ruiz-Molina, D. Angew. Chem., Int. Ed. 2008, 47, 1857−1860. (20) (a) Rocha, J.; Paz, F. A. A.; Shi, F. N.; Ferreira, R. A. S.; Trindade, T.; Carlos, L. D. Eur. J. Inorg. Chem. 2009, 4931−4945. (b) Cunha-Silva, L.; Ananias, D.; Carlos, L. D.; Paz, F. A. A.; Rocha, J. Z. Kristallogr. 2009, 224, 261−272. (c) Cunha-Silva, L.; Lima, S.; Ananias, D.; Silva, P.; Mafra, L.; Carlos, L. D.; Pillinger, M.; Valente, A. A.; Paz, F. A. A.; Rocha, J. J. Mater. Chem. 2009, 19, 2618−2632. (d) Silva, P.; Vieira, F.; Gomes, A. C.; Ananias, D.; Fernandes, J. A.; Bruno, S. M.; Soares, R.; Valente, A. A.; Rocha, J.; Paz, F. A. A. J. Am. Chem. Soc. 2011, 133, 15120−15138. (e) Soares-Santos, P. C. R.; Cunha-Silva, L.; Paz, F. A. A.; Ferreira, R. A. S.; Rocha, J.; Carlos, L. D.; Nogueira, H. I. S. Inorg. Chem. 2010, 49, 3428−3440. (f) Paz, F. A. A.; Rocha, J.; Klinowski, J.; Trindade, T.; Shi, F. N.; Mafra, L. Prog. Solid State Chem. 2005, 33, 113−125. (g) Paz, F. A. A.; Shi, F. N.; Klinowski, J.; Rocha, J.; Trindade, T. Eur. J. Inorg. Chem. 2004, 2759− 2768. (h) Shi, F. N.; Paz, F. A. A.; Girginova, P. I.; Amaral, V. S.; Rocha, J.; Klinowski, J.; Trindade, T. Inorg. Chim. Acta 2006, 359, 1147−1158. (i) Shi, F.-N.; Paz, F. A. A.; Girginova, P.; Rocha, J.; Amaral, V. S.; Klinowski, J.; Trindade, T. J. Mol. Struct. 2006, 789, 200−208. (21) (a) Wharmby, M. T.; Pearce, G. M.; Mowat, J. P. S.; Griffin, J. M.; Ashbrook, S. E.; Wright, P. A.; Schilling, L. H.; Lieb, A.; Stock, N.; Chavan, S.; Bordiga, S.; Garcia, E.; Pirngruber, G. D.; Vreeke, M.; Gora, L. Microporous Mesoporous Mater. 2012, 157, 3−17. (b) Perry, H. P.; Gagnon, K. J.; Law, J.; Teat, S.; Clearfield, A. Dalton Trans. 2012, 41, 3985−3994. (c) Gagnon, K. J.; Prosvirin, A. V.; Dunbar, K. R.; Teat, S. J.; Clearfield, A. Dalton Trans. 2012, 41, 3995−4006. (d) Wharmby, M. T.; Mowat, J. P. S.; Thompson, S. P.; Wright, P. A. J. Am. Chem. Soc. 2011, 133, 1266−1269. (e) Demadis, K. D.; Paspalaki, M.; Theodorou, J. Ind. Eng. Chem. Res. 2011, 50, 5873−5876. (f) Colodrero, R. M. P.; Cabeza, A.; Olivera-Pastor, P.; ChoquesilloLazarte, D.; Garcia-Ruiz, J. M.; Turner, A.; Ilia, G.; Maranescu, B.; Papathanasiou, K. E.; Hix, G. B.; Demadis, K. D.; Aranda, M. A. G. Inorg. Chem. 2011, 50, 11202−11211. (g) Amghouz, Z.; GarciaGranda, S.; Garcia, J. R.; Clearfield, A.; Valiente, R. Cryst. Growth Des. 2011, 11, 5289−5297. (22) Cunha-Silva, L.; Mafra, L.; Ananias, D.; Carlos, L. D.; Rocha, J.; Paz, F. A. A. Chem. Mater. 2007, 19, 3527−3538. (23) (a) Jones, S.; Liu, H.; Schmidtke, K.; O’Connor, C. C.; Zubieta, J. Inorg. Chem. Commun. 2010, 13, 298−301. (b) Stock, N.; Bein, T. J. Solid State Chem. 2002, 167, 330−336. (c) Sun, Y.-Q.; Hu, J.; Zhang, H.-H.; Chen, Y.-P. J. Solid State Chem. 2012, 186, 189−194. (d) Tripuramallu, B. K.; Kishore, R.; Das, S. K. Polyhedron 2010, 29, 2985−2990. (24) (a) Harvey, H. G.; Herve, A. C.; Hailes, H. C.; Attfield, M. P. Chem. Mater. 2004, 16, 3756−3766. (b) Irran, E.; Bein, T.; Stock, N. J. Solid State Chem. 2003, 173, 293−298. (c) Stock, N.; Guillou, N.; Bein, T.; Férey, G. Solid State Sci. 2003, 5, 629−634. (25) Shi, F.-N.; Trindade, T.; Rocha, J.; Paz, F. A. A. Cryst. Growth Des. 2008, 8, 3917−3920.

(26) Shi, F. N.; Cunha-Silva, L.; Mafra, L.; Trindade, T.; Carlos, L. D.; Paz, F. A. A.; Rocha, J. J. Am. Chem. Soc. 2008, 130, 150−167. (27) Vittal, J. J. Coord. Chem. Rev. 2007, 251, 1781−1795. (28) (a) Yamada, T.; Maruta, G.; Takeda, S. Chem. Commun. 2010, 47, 653−655. (b) Fang, H.-C.; Zhu, J.-Q.; Zhou, L.-J.; Jia, H.-Y.; Li, S.S.; Gong, X.; Li, S.-B.; Cai, Y.-P.; Thallapally, P. K.; Liu, J.; Exarhos, G. J. Cryst. Growth Des. 2010, 10, 3277−3284. (29) (a) Mobin, S. M.; Srivastava, A. K.; Mathur, P.; Lahiri, G. K. Dalton Trans. 2010, 39, 8698−8705. (b) Ranford, J. D.; Vittal, J. J.; Wu, D.; Yang, X. Angew. Chem., Int. Ed. 1999, 38, 3498−3501. (c) Wang, C.-C.; Yang, C.-C.; Yeh, C.-T.; Cheng, K.-Y.; Chang, P.-C.; Ho, M.-L.; Lee, G.-H.; Shih, W.-J.; Sheu, H.-S. Inorg. Chem. 2011, 50, 597−603. (30) Silva, P.; Valente, A. A.; Rocha, J.; Paz, F. A. A. Cryst. Growth Des. 2010, 10, 2025−2028. (31) (a) Baburin, I. A.; Blatov, V. A.; Carlucci, L.; Ciani, G.; Proserpio, D. M. J. Solid State Chem. 2005, 178, 2452−2474. (b) Blatov, V. A.; Carlucci, L.; Ciani, G.; Proserpio, D. M. CrystEngComm 2004, 6, 377−395. (c) Carlucci, L.; Ciani, G.; Proserpio, D. M. Coord. Chem. Rev. 2003, 246, 247−289. (d) Carlucci, L.; Ciani, G.; Proserpio, D. M. CrystEngComm 2003, 5, 269−279. (e) Delgado-Friedrichs, O.; Foster, M. D.; O’Keeffe, M.; Proserpio, D. M.; Treacy, M. M. J.; Yaghi, O. M. J. Solid State Chem. 2005, 178, 2533−2554. (32) (a) Blatov, V. A.; Shevchenko, A. P. TOPOS, Version 4.0 Professional (beta evaluation); Samara State University: Samara, Russia, 2006. (b) Blatov, V. A.; Shevchenko, A. P.; Serezhkin, V. N. J. Appl. Crystallogr. 2000, 33, 1193. (33) Alexandrov, E. V.; Blatov, V. A.; Kochetkov, A. V.; Proserpio, D. M. CrystEngComm 2011, 13, 3947−3958. (34) RCSR (Reticular Chemistry Structure Resource). Website: http://rcsr.anu.edu.au/. (35) Hyde, S. T.; Delgado-Friedrichs, O.; Ramsden, S. J.; Robins, V. Solid State Sci. 2006, 8, 740. (36) Socrates, G. Infrared Characteristic Group , 2nd ed.; John Wiley & Sons Ltd: Chichester, U.K., 1994. (37) Kottke, T.; Stalke, D. J. Appl. Crystallogr. 1993, 26, 615−619. (38) APEX2: Data Collection Software, Version 2.1-RC13; Bruker AXS: Delft, The Netherlands, 2006. (39) Cryopad: Remote Monitoring and Control, Version 1.451; Oxford Cryosystems: Oxford, U.K., 2006. (40) SAINT+: Data Integration Engine, v. 7.23a; Bruker AXS: Madison, WI; 1997−2005. (41) Sheldrick, G. M. SADABS: Bruker/Siemens Area Detector Absorption Correction Program, v.2.01; Bruker AXS: Madison, WI, 1998. (42) (a) Sheldrick, G. M. SHELXS-97: Program for Crystal Structure Solution; University of Gö ttingen: Gö ttingen, Germany, 1997. (b) Sheldrick, G. M. Acta Crystallogr., Sect. A 2008, 64, 112−122. (43) Sheldrick, G. M. SHELXL-97: Program for Crystal Structure Refinement; University of Göttingen: Göttingen, Germany, 1997. (44) Brandenburg, K. Crystal Diamond, 3.2g ed.; Crystal Impact: Bonn, Germany, 1997−2011. (45) Boultif, A.; Louer, D. J. Appl. Crystallogr. 2004, 37, 724−731. (46) (a) Rodriguez-Carvajal, J. FULLPROF: A Program for Rietveld Refinement and Pattern Matching Analysis. Abstract of the Satellite Meeting on Powder Diffraction of the XV Congress of the IUCR, Toulouse, France; p 127. (b) Roisnel, T.; Rodriguez-Carvajal, J. WinPLOTR [February 2008]: A Windows Tool for Powder Diffraction Pattern Analysis. In Materials Science Forum, Proceedings of the Seventh European Powder Diffraction Conference (EPDIC 7); Delhez, R., Mittenmeijer, E. J., Eds.; 2000; pp 118−123. (47) Boultif, A.; Louer, D. J. Appl. Crystallogr. 1991, 24, 987−993. (48) Louer, D. Automatic Indexing: Procedures and Applications, Accuracy in Powder Diffraction II; NIST: Gaithersburg, MD, 1992; pp 92−104. (49) Laugier, J.; Bochu, B. CHECKCELL: A Software Performing Automatic Cell/Space Group Determination; Collaborative Computational Project Number 14 (CCP14); Laboratoire des Matériaux et du Q

dx.doi.org/10.1021/cg301112k | Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design

Article

Génie Physique de l’Ecole Supérieure de Physique de Grenoble (INPG): Grenoble, France, 2000. (50) (a) Spek, A. L. Acta Crystallogr., Sect. A 1990, 46, C34. (b) Spek, A. L. J. Appl. Crystallogr. 2003, 36, 7−13. (51) Altomare, A.; Burla, M. C.; Camalli, M.; Carrozzini, B.; Cascarano, G. L.; Giacovazzo, C.; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.; Rizzi, R. J. Appl. Crystallogr. 1999, 32, 339−340. (52) Rietveld, H. M. J. Appl. Crystallogr. 1969, 2, 65−71. (53) Caglioti, G.; Paoletti, A.; Ricci, F. P. Nucl. Instrum. 1958, 3, 223−228.

R

dx.doi.org/10.1021/cg301112k | Cryst. Growth Des. XXXX, XXX, XXX−XXX