Article pubs.acs.org/crystal
Discrete and Polymeric Cu(II) Complexes Derived from in Situ Generated Pyridyl-Functionalized Bis(amido)phosphate Ligands, [PO2(NHPy)2]− Arvind K. Gupta, Anant Kumar Srivastava, Indra Kumar Mahawar, and Ramamoorthy Boomishankar* Department of Chemistry, Mendeleev Block, Indian Institute of Science Education and Research, Pune, Dr. Homi Bhabha Road, Pune 411008, Maharashtra, India S Supporting Information *
ABSTRACT: New examples of Cu(II) complexes containing in sit u gener at ed bis(am ido )ph osph ate lig an ds, [PO2(NHPy)2]−, have been synthesized [(L1)−, where Py = 2-pyridyl, and (L2)−, where Py = 4-pyridyl). These anionic ligands were obtained by the P−N bond hydrolysis reaction of the corresponding phosphonium salts or phosphoric triamides in the presence of Lewis acidic Cu(II) ions in polar solvents. A discrete mononuclear complex {Cu[PO2(NH2Py)2]2} (3) was obtained for the in situ ligand featuring 2-pyridyl substituents, whereas a 1D coordination polymer {[Cu(PO2(NH4Py)2)2(H2O)2]·2DMF·2H2O}∞ (4) was obtained for the ligand containing 4-pyridyl functionalities. The structural analyses show that in 3 the ligand (L1)− is bonded to the metal ion in a chelating tridentate N,N,O coordination and the coordinated P−O group is oriented in a syn fashion with respect to the pyridylamino groups. However, in 4 the ligand (L2)− is bonded in a bridging fashion to two Cu(II) ions via its pyridyl nitrogen donors. The two phosphorus-bound oxygen atoms in (L2)− are anti-oriented and are involved in intermolecular H-bonding interactions with the metal-bound water hydrogen atoms to yield a 3D network. Topology analysis of the H-bonded assembly in 4 shows the presence of a (4-c)2, 8-c binodal scu net as represented by the Schläfli symbol of (44·62)2(416·612) [44·62 for the (L2)− ligand and 416·612 for the Cu atoms]. An interesting 2D Cu(I) coordination polymer {[Cu(PO(NH4Py)3)(PO2(NH4Py)2)]·DMF·2H2O}∞ (5) in uninodal hcb topology was obtained as a minor product along with 4. Formation of the 2D assembly in 5 is mediated by the tridentate coordination of the phosphoric triamide ligand [PO(NH4Py)3]. The role of the anionic (L2)− ligand in 5 is to merely restore the charge balance in the assembly and provide the fourth coordination to the tetrahedral Cu(I) ion. Solvent uptake studies on the activated sample of 4b shows a preferential adsorption of water vapors over aliphatic alcohols.
■
INTRODUCTION Rational design of functional framework materials such as metal−organic frameworks (MOFs) and porous coordination polymers (PCPs) has been the focus area of interest for several research groups.1,2 This is attributed to both the novel structural topologies exhibited by these materials and their utility in the areas of separation, storage, enantioselective catalysis, sensors, drug delivery, etc.3,4 These materials are traditionally synthesized by the use of various organic linkers in combination with metal cluster nodes in the form of their oxide or hydroxide building units. Typical organic linkers include neutral and/or anionic ligands such as 4,4-bipyridine and/or 1,4-benzenedicarboxylate and their analogues, which are shown to form families of porous framework materials with tunable pore dimensions.5 Over the years, several ligand systems have been tried as organic linkers in order to establish the relationship between various framework compositions and the observed functionalities.6,7 In this regard, in situ generated ligand moieties offer an attractive pathway in crystal engineering to obtain novel coordination assemblies.8 By this approach, it is possible to generate new kinds of multimetallic assemblies © 2014 American Chemical Society
that are either inaccessible or involve cumbersome synthetic pathways.9 Earlier literature reports show that in situ ligand synthesis in carbon chemistry can be achieved in reaction pathways, viz., hydrolysis of -CN/-COOR groups, reduction of carboxylates (-COO−), C−C bond formation facilitated by oxidative or reductive coupling, and hydroxylation.10 Recently, pyridyl-functionalized amino P(V) ligands have been shown to offer a flexible platform to specifically obtain multimetallic assemblies in cage, cluster, and framework architectures.11 Moreover, these ligands can be subjected to controlled metalassisted P−N bond hydrolysis reactions to generate interesting examples of in situ generated ligands.12 Thus, from a tetra(amido)phosphonium cation, ligand scaffolds such as phosphoric tri-, di-, and monoamide ligands containing both N- and O-donor functionalities can be obtained in a facile manner as their corresponding complexes (Figure 1). We have recently shown the utility of an in situ generated bis(amido)Received: December 11, 2013 Revised: February 13, 2014 Published: February 17, 2014 1701
dx.doi.org/10.1021/cg401847w | Cryst. Growth Des. 2014, 14, 1701−1709
Crystal Growth & Design
Article
Figure 1. P−N bond cleavage routes to access the pyridylamino-functionalized in situ generated ligands L− and L2−.
phosphate ligand, [PO2(NH4Py)2]− (where 4Py = 4-pyridyl), for the formation of a two-dimensional Zn(II) coordination polymer and its functional studies.13 Herein, we report the synthesis of two Cu(II) complexes of in situ generated bis(amido)phosphate ligand [PO2(NHPy)2]− featuring pyridyl (Py) substituents. A discrete complex of composition {Cu[PO2(NH2Py)2]2} (3) was obtained for 2-pyridyl (2Py) substituents, and a one-dimensional (1D) polymeric coordination polymer of composition {[Cu(PO2(NH4Py)2)2(H2O)2]· 2DMF·2H2O}∞ (4) was obtained for the ligand containing 4pyridyl (4Py) substituents. Electron paramagnetic resonance (EPR) and magnetic measurements on 4 show the paramagnetic nature of the complex. In addition, vapor sorption studies on the activated sample of 4 reveals the selective adsorption of water vapor over aliphatic alcohol. Use of MOFs and PCPs for the selective separation of water and alcohols has gained a lot of recent attention14 with aspirations in the purification of bioethanol industrially.15 Removal of the last 4% azeotropic mixture of water from bioethanol is very difficult by usual distillation procedures, and thus microporous MOF materials offer an attractive platform to effect this separation.
blue blocklike crystalline material in good yield. Single-crystal X-ray analysis of the material revealed the presence of the anionic ligand [PO2(NH4Py)2]−, (L2)−, as anticipated, forming a 1D macrocyclic chain polymer of composition [Cu(L2)2(H2O)2] (Scheme S1, Supporting Information). The EPR spectrum of crystals of 4 at room temperature showed a single line spectrum typical of a Cu(II) ion with a g value of 1.973 (Figure S2, Supporting Information). Magnetic measurements on the microcrystalline solid of 4 show paramagnetic behavior consistent with an isolated metal ion with S = 1/2 spin (Figures S3 and S4, Supporting Information). The same Cu(II) complexes 3 and 4 were obtained when the reactions were performed with phosphoramide precursors [PO(NH2Py)3] (6) and [PO(NH4Py)3] (7), respectively, instead of the corresponding phosphonium salts. In an effort to deliberately synthesize these in situ ligands (L1)− and (L2)−, we treated DMF solutions of the respective phosphonium salts 1 and 2 with 1 M solution of HCl in water at 90 °C for 72 h and the solutions were evaporated to dryness on a hot plate to yield solid samples. MALDI-TOF mass spectra of these samples gave prominent peaks at m/z = 288.98 and 288.97 pertaining to the species [L1 + K] and [L2 + K], respectively, in addition to peaks due to the corresponding aminopyridine hydrochloride byproducts (Figures S5 and S6, Supporting Information).16 The 31P NMR spectra of these solids show peaks at δ −9.34 and −12.62 ppm, respectively, which are considerably upfield-shifted from those of the corresponding phosphonium precursors (Figures S9 and S10, Supporting Information).11a,13 Furthermore, recrystallization experiments yield single crystals in one of the instances where formation of the hydrochloride [L2 + 2H]Cl was confirmed by single-crystal X-ray diffraction. It is observed from the crystal structure that [L2 + 2H]+ is zwitterionic, in which the two protonated pyridyl N-terminals carry the dipositive charge and the two P−O terminals carry a negative charge, and the overall charge is restored by the presence of a chloride ion (Figure S11, Supporting Information). However, efforts to prepare the neutral derivatives [L1 + H] and [L2 + H] by reacting these hydrochloride salts with bases such as Na2CO3, NaHCO3, (NH4)2CO3, and Et3N did not result in a clean reaction. These observations clearly indicate that in situ formation of (L1)− and (L2)− offers a facile reaction pathway to obtain complexes 3 and 4. During the reaction of Cu(NO3)2 with 2, a few small pale yellow crystals were present in the reaction vessel in addition to the major quantities of the crystals of 4. Single-crystal X-ray analysis of these crystals show the formation of a twodimensional (2D) Cu(I) coordination network 5, {[Cu(PO(NH4Py)3)(PO2(NH4Py)2)]·DMF·2H2O}∞, in which the Cu(I) ion is complexed with both 7 and (L2)−. Formation of complex 5 is due to the mildly reducing nature of DMF causing
■
RESULTS AND DISCUSSION Synthesis of Metal Complexes. The precursor ligands [P(NH2Py)4]Cl (1) and [P(NH4Py)4]Cl (2) were synthesized according to our previously reported procedure.11a,13 In order to obtain metal complexes of the in situ generated ligands of type L− or L2−, we chose to employ salts of Cu(II) ions, which offer a fair amount of acidity to the reaction medium. Thus, reaction of 1 with Cu(NO3)2 at room temperature gave rise to the in situ generated ligand [PO2(NH2Py)2]−, (L1)−, in the corresponding complex 3. Performing the reaction under hydrothermal conditions at 90 °C with the view of obtaining the dianionic phosphonate ligand (L1)2− results in complex 3, as no further cleavage was observed. This might be attributed to the stable arrangement of octahedral Cu(II) ions in the N,N,O coordination of the monoanionic ligand (L1)−. The same complex as a solvated molecule of 3·2DMF was obtained when the reaction was performed with Cu(ClO4)2 at room temperature. The stability of 3·2DMF in solution was confirmed by its matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF) mass spectrum, which gave a peak at m/z = 674.9 corresponding to the species [3·1DMF + K]+ (Figure S1, Supporting Information). Similar crystallization reaction of 2 at room temperature with metal salts of Cu(II) have led to rapid complexation and precipitation. Hence, we attempted these reactions under hydrothermal conditions at moderate temperatures and pressures. Thus, reaction of 2 in N,N-dimethylformamide (DMF) with a solution of Cu(NO3)2 in water at 90 °C under hydrothermal conditions for 3 days gave 4, {[Cu(PO2(NH4Py)2)(H2O)2]·2DMF·2H2O}∞, as a 1702
dx.doi.org/10.1021/cg401847w | Cryst. Growth Des. 2014, 14, 1701−1709
Crystal Growth & Design
Article
Figure 2. Molecular structure and H-bonding interactions in (a) 3 and (c) 3·2DMF, and Connolly contact surface view of a 2 × 2 × 2 packing structure in (b) 3 and (d) 3·2DMF, for a probe radius of 1.2 Å and a grid spacing of 0.2 Å.
of two eight-membered R228 rings. The net effect of these Hbonding interactions gave rise to a three-dimensional (3D) supramolecular network for the structure of 3. The packing diagram of this H-bonded 3D network shows a tightly packed structure along the tetragonal lattice and shows a very narrow pore along the c-axis with a diameter of 3.870 Å (Figure 2b). The calculated void space is around is 284.5 Å3, which amounts to only 3% of the unit-cell volume. In contrast, the molecular structure of 3·2DMF was solved in the monoclinic space group P21/n. As observed before, the core structure as well as the octahedral coordination around the Cu(II) center remains the same. However, the H-bond-assisted supramolecular network in 3·2DMF leads only to a 1D chain structure (Figure 2c). This is due to one of the ligand amino protons in 3·2DMF being engaged in H-bonding with the solvated DMF. This interaction not only prevents the proximity of the adjacent complex but also destroys the bifurcated interaction at the PO group that was present in 3. As a result, a packing diagram of the molecule shows a larger pore with distances ranging from 6.777 to 13.835 Å. The channel volume inside is calculated to be 534.5 Å3 (34.3% of the unit cell volume), which is occupied by the solvate molecules of DMF
the in situ reduction of some of the Cu(II) ions to Cu(I) ions.17 Repeating the reaction with 2 in the absence of added water [a DMF solution of Cu(NO3)2 instead of its aqueous solution] produced pure material 5, albeit in low yields (Scheme S2, Supporting Information). Attempts to synthesize 5 from Cu(I) sources, such as [Cu(MeCN)4]PF6, or from Cu(II) salts in the presence of NH3 as external base were unsuccessful. This is probably due to the acidity of medium being lowered in both these instances, preventing in situ formation of the (L2)− ligand. Crystal Structures of 3, 4, and 5. The molecular structure of 3 was solved in the tetragonal space group I41/acd. The asymmetric unit consists of a Cu(II) ion in half occupancy and one (L1)− ligand moiety. The molecular core consists of a dicationic Cu center and the charge balance is restored by two (L1)− ligands. The coordination geometry around the Cu(II) atom is distorted octahedral featuring four Npyridyl contacts and two O− coordination, derived from the chelating tridentate N,N,O coordination of the two (L1)− ligands (Figure 2a). The uncoordinated phosphoryl oxygens were involved in a bifurcated H-bonding interaction with protons of the pyridyl amino moieties in which every phosphoryl oxygen atom is part 1703
dx.doi.org/10.1021/cg401847w | Cryst. Growth Des. 2014, 14, 1701−1709
Crystal Growth & Design
Article
Figure 3. (a) View of 1D chain polymer in 4 showing H-bonding interactions of orthogonally oriented neighboring chains leading to the 3D network; (b) closer view of 3D topology of 4 along the a-axis; and (c) packing structure of 4 showing microporous channels.
contiguous Cu2(L2)2 square grids. Although such kinds of Cu(II)-based coordination polymers were known in the literature, they were all formed in the presence of neutral Ndonor ligands and the obtained chains were cationic in nature.18 It is interesting to compare the coordination of (L2)− in 4 to that found in its 1:1 zinc complex of formula [Zn(L2)(HCO2)]∞.13 In this 2D network, due to the higher oxophilicity of Zn(II) ions, the ligand was found in a tridentate coordination involving two Npyridyl and one anionic O− coordination, whereas in 4, the ligand is found in only a bidentate coordination to two Cu(II) atoms, resulting in a 1D chain structure. Furthermore, the two coordinated water molecules are both equivalent and engage in hydrogen-bonding interactions with the uncoordinated P−O groups of the adjacent orthogonally oriented 1D chains to form a tightly knit 3D network. While the 24-membered Cu2(L2)2 macrocycles form the core of the 1D chain, a series of H-bonded R4412 rings were involved in the construction of the 3D network (Figure 3). TOPOS analysis19
(Figure 2d). The metric parameters associated with P−N, P− O, Cu−N, and Cu−O in both 3 and 3·2DMF are consistent with the literature reported values. The one-dimensional Cu(II) coordination polymer 4 was crystallized in the orthorhombic space group Cccm. The asymmetric unit of 4 consists of a unique Cu(II) center having a fourth of an occupancy, half of the phosphate ligand (L2)−, and a half-occupied coordinated water molecule. Thus the overall formula of the 1D chain can be represented as [Cu(L2)2(H2O)2]∞ and the dicationic charge of the complex is balanced by the presence of two (L2)− ligands. The solvated DMF and water molecules were found with positional disorder with half occupancies for each of them in the asymmetric unit. The molecular core consists of a distorted octahedral Cu(II) center consisting of four equatorial pyridyl ligands from four (L2)− ligands in a planar fashion and two coordinated water molecules at the axial sites. Each anionic ligand coordinates through only its pyridyl nitrogens and bridges the two adjacent Cu(II) centers in a V-shaped manner with the formation of 1704
dx.doi.org/10.1021/cg401847w | Cryst. Growth Des. 2014, 14, 1701−1709
Crystal Growth & Design
Article
Figure 4. (a) View of the 2D sheet in 5 along the ab-plane. The neutral ligand 7 is shown with light gray bonds, and the anionic ligand (L2)− is shown with dull-green bonds. (b) Zigzag view of the 2D sheet along the a-axis. (c) Offset view of the hexagonal network in 5. Smaller and bigger spheres represent P and Cu nodes, respectively. (d) Formation of H-bonded 3D structure in 5. The individual sheets are colored differently for better viewing, and the representative H-bonding points are indicated as dotted lines.
4a, {[Cu(L2)2(H2O)2]·2CH3OH·2H2O}∞. The unit cell parameters of 4a are similar to those of 4 and are solved in the orthorhombic space group Cccm. Comparison of the channel structures along the c-axis in both these crystals indicate that the void space observed in 4 is retained in 4a as well. Cu(I) coordination polymer 5 with the mixed ligand configuration was solved in monoclinic space group P21/c. At the core of the molecule resides the tetrahedral Cu(I) ion coordinated to four pyridyl donor ligands, three from the ligand 7 and one from the anionic ligand (L2)−. Each neutral ligand (7) in turn is bonded to three tetrahedral Cu(I) center and aids in 2D network propagation (Figure 4a). The cumulative effect of these interactions is the formation of a zigzag 2D layer in which both the Cu(I) center and the neutral ligand 7 act as three-connected nodes (Figure 4a,b). TOPOS software analysis for 5 gave a Schläfli symbol of 63 for a uninodal net with a distorted hcb network topology (Figure 4c).20 The role of (L2)− is merely to provide coordination saturation for the metal center through one of its pyridyl groups and to restore the charge balance in the network. The other pyridyl group of
on 4 shows the presence of a (4-c)2, 8-c binodal scu net as represented by the Schläfli symbol of (44·62)2(416·612) [first symbol for the (L2)− ligand and second one for the Cu atoms].20 Each ligand moiety acts as a four-connected node (pink spheres, with two short and two long contacts), and each Cu atom (green spheres, with four short and four long contacts) serves as an eight-connected node, leading to a 3D Hbonded network (Figure 3b). The metric parameters associated with these H-bonds are very strong: 2.705(3) Å for the donor− acceptor distance (O1···O2), and the pertaining O2−H2···O1 angle at 160.8(3)°. The packing diagram of the hydrogenbonded 3D network in 4 shows the presence of a solvated channel occupied by disordered molecules of DMF and water (Figure 3c). The channel volume calculated by use of Mercury software,21 after exclusion of the solvate molecules, gave 32.6% of the unit cell volume, which measures about 1068 Å3. The two pore diameters inside 4 are measured to be 13.158 and 14.942 Å. In order to obtain a guest-free material, the solvated DMFs in 4 were exchanged with methanol in a single-crystal to single-crystal manner. X-ray analysis of the crystals soaked in methanol for a period of 10 days indicated a complete guest exchange leading to the formation of a new methanol adduct 1705
dx.doi.org/10.1021/cg401847w | Cryst. Growth Des. 2014, 14, 1701−1709
Crystal Growth & Design
Article
(L2)− is uncoordinated and remains alternately above and below the 2D sheet. The packing diagram of the molecule shows formation of a 3D assembly in which the 2D layers are interdigitated onto each other and involved in H-bonding interactions. The zigzag nature (Figure 4b) of the network allows each 2D-sheet to participate in H-bonding interactions with six other sheets (three above and three below). The amino protons of ligand 7 in a given sheet are involved in H-bonding with the PO groups of (L2)− ligands from adjacent sheets (N3−H3···O2). Furthermore, anionic ligands [(L2)−] from neighboring sheets occupy the space inside the channel and are involved in Hbonding interactions among themselves, forming R228 rings (N4−H4···O2) and providing additional support for the 3D assembly (Figure 4d). The remaining amino protons are hydrogen-bonded to solvated (DMF and water) molecules located in the channels. The packing diagram of the molecule shows a narrow pore along the b-axis with a void volume of 274.09 Å3, which amounts to only 8% of the unit cell volume. Due to the smaller void volume and the oxygen-sensitive nature of this material, we were unable to perform guest exchange and sorption studies for 5. Thermal and Sorption Studies of 4. Thermogravimetric analysis (TGA) of the crystalline sample of 4 shows gradual weight loss of about 20% up to a temperature of 230 °C, corresponding to the loss of all solvent molecules present in the structure. Above this temperature the percentage weight loss is high and abrupt, which is attributed to the framework decomposition (Figure 5). The TGA graph of the methanol-
Figure 6. PXRD patterns for various samples of 4.
PXRD pattern of the desolvated sample 4b shows deviations from that of the pristine sample, indicating some changes in the crystal packing owing to the partial or complete loss of coordinated water upon desolvation. Soaking 4b in methanol for 3 days resolvates it and revokes the crystallinity in the sample 4c. This is evidenced by a visible color change of the sample from turquoise to bright blue and reappearance of PXRD patterns due to 4a. TGA analysis of 4c also matches that of 4a, consistent with complete resolvation. Interestingly, resolvation of 4b in water takes place within 30 min to yield a water-resolvated sample 4d, which has a slightly different PXRD pattern than that of 4a, attributed to a large solvent uptake. Although the volume of the packing cavity in 4a is quite suitable for gas uptake studies, due to the narrow dimensions of the pore windows (2.6 Å × 6.7 Å), excluding the van der Waals radii, no appreciable gas uptake characteristics could be observed for this material. The CO2 and N2 adsorption isotherms for the activated sample of 4b measured at 150 and 77 K, respectively, indicate that the observed values are characteristic of a surface adsorption (Figure S12, Supporting Information). In contrast, solvent sorption studies for water, methanol, and ethanol vapors measured on 4b show an interesting trend among these solvents (Figure 7). The adsorption isotherm for water indicates a gradual uptake, and
Figure 5. Thermogravimetric analysis of various samples of 4.
exchanged sample 4a shows an initial weight loss of about 14% below 100 °C, matching with the removal of solvate molecules of methanol and water. A second weight loss of about 5% occurs in the temperature region between 170 and 230 °C suggesting the loss of two coordinated water molecules. Above this temperature a rapid weight loss occurs as before, indicative of framework decomposition. TGA analysis on the desolvated sample 4b, obtained by evacuating 4a at 120 °C overnight, shows no significant weight loss up to 190 °C and complete decomposition above 250 °C, indicating that the framework remained intact up to this temperature. Powder X-ray diffraction (PXRD) patterns of 4 and 4a were closely matching, albeit with slight variation in intensities for some of the observed peaks (Figure 6). On the other hand, the
Figure 7. Solvent sorption isotherms for 4b at 298 K. 1706
dx.doi.org/10.1021/cg401847w | Cryst. Growth Des. 2014, 14, 1701−1709
Crystal Growth & Design
Article
the final value at P/P0 = 1.0 corresponds to 5.5 water molecules/formula unit. The adsorption isotherm for methanol reveals that the uptake was sluggish and the final value corresponds only to 1 methanol/formula unit of the network. The bulkier ethanol did not adsorb at all, suggesting that 4b takes up water vapors preferentially over aliphatic alcohols. This could be attributed both to its smaller pore diameter and to the presence of donor−acceptor-type H-bonding interactions between the coordinated and solvate molecules of water. Although the previously reported Zn(II) polymer is better in terms of water uptake capacity (10 molecules/formula unit), its water adsorption isotherm indicated a minor structural rearrangement.13 However, 4b exhibits better stability as indicated by its gradual uptake isotherm and sharp PXRD profiles (for both 4b and 4d). Furthermore, from the nearly identical solid-state UV−visible spectral profiles and FT-IR peaks, the structural stability of desolvated and water-resolvated samples was confirmed (Figures S13 and S14, Supporting Information).
obtained from a Bruker D8 Advance diffractometer. Thermal analysis data were obtained from a Perkin-Elmer STA-6000 thermogravimetric analyzer. NMR spectra were recorded on a Jeol 400 MHz spectrometer (1H NMR, 400.13 MHz; 31P{1H} NMR, 161.97 MHz). Direct current (dc) magnetic measurements were obtained on a Quantum Design MPMSXL magnetic properties measurements system. Electron paramagnetic resonance (EPR) spectrum was obtained on a Bruker EMX EPR spectrometer with an X-band frequency. Elemental analyses were performed on a Vario-EL cube elemental analyzer. FT-IR spectra were taken on a Thermo-Scientific Nicolet 6700 spectrophotometer with samples prepared as KBr pellets. Melting points were obtained from an Electro thermal melting point apparatus and are uncorrected. Syntheses. Preparation of Cu(II) Complex 3. (a) To a solution of ligand 1 (362 mg, 0.82 mmol) in methanol was added Cu(NO3)2 (100 mg, 0.41 mmol) in methanol, along with a few drops of DMF (N,Ndimethylformamide). The mixture was stirred for 1 h, filtered, and kept for crystallization. Prismatic green crystals were obtained after 15 days. Yield 70% (162 mg, based on Cu). (b) To a solution of ligand 6 (269 mg, 0.82 mmol) in methanol was added Cu(NO3)2 (100 mg, 0.41 mmol) in methanol, along with a few drops of DMF. The mixture was stirred for 1 h, filtered, and kept for crystallization. Prismatic green crystals were obtained after 15 days. Yield 70% (83 mg, based on Cu). Mp 178−180 °C. FT-IR data in KBr pellet (cm−1) 528, 624, 664, 738, 827, 908, 960, 1011, 1071, 1209, 1285, 1330, 1413, 1459, 1504, 1603, 1670, 3175, and 3434. Anal. Calcd for C20H20N8O4P2Cu: C, 42.75; H, 3.59; N, 19.94. Found: C, 42.84; H, 3.56; N, 19.74. Preparation of Cu(II) Complex 3·2DMF. (a) To a solution of ligand 1 (237 mg, 0.6 mmol) in methanol was added Cu(ClO4)2 (100 mg, 0.27 mmol) in methanol, along with a few drops of DMF. The mixture was stirred for 1 h, filtered, and kept for crystallization. Prismatic green crystals were obtained after 10 days. Yield 60% (71 mg, based on Cu). (b) To a solution of ligand 6 (176 mg, 0.54 mmol) in methanol was added Cu(ClO4)2 (100 mg, 0.27 mmol) in methanol, along with a few drops of DMF. The mixture was stirred for 1 h, filtered, and kept for crystallization. Prismatic green crystals were obtained after 10 days. Yield 70% (83 mg, based on Cu). Mp 185−187 °C. FT-IR data in KBr pellet (cm−1) 524, 578, 769, 825, 933, 1017, 1068, 1118, 1160, 1196, 1384, 1460, 1578, 1617, 1664, 3183, and 3438. Anal. Calcd for C26H34N10O6P2Cu: C, 44.10; H, 4.84; N, 19.78. Found: C, 44.34; H, 4.76; N, 19.65. Preparation of Cu(II) Coordination Polymer 4. (a) To a solution of ligand 2 (438 mg, 1.0 mmol) in DMF (10 mL), placed in a screwcapped glass vessel, was added a solution of Cu(NO3)3 (120 mg, 0.5 mmol) in H2O (10 mL), and the final mixture was heated at 90 °C for 3 days. The reaction mixture was then slowly brought back to room temperature over a period of 24 h, at which point dark blue crystals of 4 were obtained. Yield 60% (234 mg based on Cu). (b) To a solution of ligand 7 (327 mg, 1.0 mmol) in DMF (10 mL), placed in a screwcapped glass vessel, was added a solution of Cu(NO3)3 (120 mg, 0.5 mmol) in H2O (10 mL), and the final mixture was heated at 90 °C for 3 days. The reaction mixture was then slowly brought back to room temperature over a period of 24 h, at which point dark blue crystals of 4 were obtained. Yield 40% (254 mg based on Cu). Mp 282−284 °C. FT-IR data in KBr pellet (cm−1) 520, 670, 839, 912, 1022, 1079, 1215, 1319, 1341, 1387, 1413, 1460, 1508, 1617, and 3419. Anal. Calcd. for C20H24N8O6P2Cu: C, 40.17; H, 4.05; N, 18.74. Found: C, 54.54; H, 4.16; N, 18.55. A small number of pale yellow crystals of 5 were also present in the vessel as a minor product in both of these reactions, in
