Concise template syntheses of gallium phosphates driven by in situ direct alkylation of aliphatic and aromatic precursors by methanol

Abstract

A family of open-framework gallium phosphates (GaPOs), namely [dmdabco]_0.5[Ga(HPO₄)₂] (1, dmdabco = N,N′-dimethyl-1,4-diazabicyclo[2,2,2]octane), [tmpip]_0.5[Ga₃(OH)(PO₄)₃]·(H₂O)_0.25 (2, tmpip = N,N,N′,N′-tetramethyl-piperazinium), [mpy][Ga₃F(PO₄)₃]·(H₂O)_0.25 (3, mpy = N-methyl-pyridine), [dmbpy]_0.5[Ga₂(HPO₄)₂(PO₄)] (4, dmbpy = N,N′-dimethyl-4,4′-bipyridine), and [Hmpip]₂[Ga₇(OH)(PO₄)₆(HPO₄)₂] (5, mpip = N-methylpiperazine) have been fabricated under hydro(solvo)thermal conditions and structurally characterized. The in situ-template-synthesis strategy is firstly used to construct the gallium phosphates. The in situ generated dmdabco²⁺, tmpip²⁺, mpy⁺ and dmbpy²⁺ templates were derived from the methylation of two distinct types of organic-amine precursors under methanol media: aliphatic 1,4-diazabicyclo[2,2,2]octane (dabco) in 1, 1,4-dimethylpiperazine or 1-methylpiperazine or piperazine (pip) in 2, and aromatic pyridine (py) in 3 and 4,4′-bipyridine (bpy) in 4. Such a unique in situ methylation feature is different from classic Eschweiler–Clarke methylation in which excessive formic acid and formaldehyde is needed. Compound 1 exhibits infinite inorganic chains connected by hydrogen bonds to generate a 3D supramolecular framework; compounds 2 and 3 are isomorphous and constructed from a Ga₆P₆ secondary building unit (SBUs), forming a 3D pcu architecture with intersecting 8-membered channels; compound 4 possesses an inorganic layered structure and compound 5 features an interesting 3D open-framework architecture with helical channels.

---

Introduction

Crystalline microporous materials with an open framework have been the subject of continuous interest because of their enormously abundant structural chemistry and potential applications in the domains of catalysis, absorption, ion-exchange and separation, etc.¹ As a significant branch of the various inorganic materials, metal phosphates have experienced extraordinary expansion due to their structural and compositional diversities.^2–5 Therein, gallium phosphates (GaPOs) are of particular interest since gallium is in the same group as aluminum and the forming GaPO framework topologies in some cases are analogous to their corresponding aluminophosphates. Furthermore, the larger radius of metal gallium centre gives opportunity to adopt higher coordination numbers and lead to novel structures. Ga atom exhibits flexible structural coordination as GaO₄ tetrahedron, GaO₅ trigonal bipyramid, and GaO₆ octahedron in many of gallophosphate structures. And numerous new GaPOs with zero- (0D, cluster), one- (1D, chain or ladder), two- (2D, layer), and three-dimensional (3D) architectures have been documented in literature.⁶ Notable examples are NTHU-1 with 24-ring channels,⁷ ICL-1 and cloverite with 20-ring channels,⁸ MIL-31 and MIL-50 with 18-ring channels,⁹ organic–inorganic hybrid yellow phosphor NTHU-6,¹⁰ and chiral gallium phosphate [(1S,2S)-H₂DACH][Ga₂(1S,2S-DACH)(HPO₄)(PO₄)₂] possessing both chiral ligand and chiral template.¹¹

It is of great significance to develop new synthetic routes or strategies for the preparation of crystalline materials. The addition of fluorides in reaction medium, for example, may greatly improve the mineralization process and favor many fluorinated gallium phosphate/phosphite networks;¹² some porous materials with new topological structures including alumino- and gallophosphates, metal phosphites, metal oxalato-phosphonates and metal–organic frameworks (MOF), have also been prepared by ionothermal or solvent-free synthesis methods recently.^13,14 These efforts encourage us to explore new synthetic strategies to construct novel metal phosphates. It has been demonstrated that organic species have been extensively used under hydro(solvo)thermal conditions because of their ability to affect the crystallization processes by acting as coordinating agents to metals, structure-directing agents (SDAs), and charge-balancing agents. However, almost all organic templates used in such syntheses are commercially available and of direct use. By contrast, little attention has been paid to the modification or in situ generation of organic components (especially those that are unavailable by conventional preparation) in the synthesis of porous materials. In a few cases, new smaller organic amines were generated in situ by the decomposition of initially added large amines. For example, the high symmetric gismondine framework can be formed from large amines including N,N,N′,N′′,N′′-pentamethyldiethylenetriamine [((CH₃)₂NCH₂CH₂)₂NCH₃];¹⁵ in situ generation of alkylamines by hydrolysis of alkylformamides could be used to prepare alumino- and gallophosphates;¹⁶ controlling the in situ decomposition of chain-type polyamines during crystallization procedure was useful for directing the preparation of heterometallic cobalt–zinc phosphates,¹⁷ etc. Inspired by the great success achieved in the crystal engineering of coordination polymers containing in situ ligand synthesis over the past several years,¹⁸ we have recently initiated a systematic investigation on the hydrothermal modification of organic templates typically used in porous materials. Using cyclic aliphatic diamines such as 1,4-diazabicyclo[2.2.2]octane and piperazine as the staring precursors, we have successfully isolated a series of zinc phosphate/phosphite phases containing in situ generated new templates.¹⁹ As an extension of our work aiming to further investigate the in situ template generation route for the construction of novel porous materials, five new organically templated GaPOs: [dmdabco]_0.5[Ga(HPO₄)₂] (1), [tmpip]_0.5[Ga₃(OH)(PO₄)₃]·(H₂O)_0.25 (2), [mpy][Ga₃F(PO₄)₃]·(H₂O)_0.25 (3), [dmbpy]_0.5[Ga₂(HPO₄)₂(PO₄)] (4) and [Hmpip]₂[Ga₇(OH)(PO₄)₆(HPO₄)₂] (5), have been successfully prepared and structurally characterized. The inorganic parts of these compounds have different dimensions: infinite chain in 1, 3D network with multinational intersecting 8-ring channels in 2 and 3, 2D distinct layer in 4 and 3D architecture with interesting helical channels in 5. It is worth noting that the in situ generated dmdabco²⁺, tmpip²⁺, mpy⁺ and dmbpy²⁺ templates were derived from the direct methylation of amine precursors with methanol solvent.

Experimental

Materials and instrumentation

All starting materials were commercially available, and used as purchased without further purification. IR spectra were performed on an ABB Bomen MB 102 series FTIR spectrophotometer. Elemental analyses were measured on Perkin-Elmer 240C analyzer. Powder X-ray diffraction (PXRD) data were recorded on a Philips X'Pert-MPD diffractometer with Cu-Kα₁ radiation (λ = 1.54076 Å). Thermogravimetric analysis (TGA) was performed on a Mettler Toledo TGA/SDTA 851e analyzer with a heating rate of 10 °C min⁻¹ from 40 to 800 °C.

Syntheses

[dmdabco]_0.5[Ga(HPO₄)₂] (1). A mixture of Ga₂O₃ (0.148 g), H₃PO₄ (0.23 mL), 1,4-diazabicyclo[2,2,2]octane (0.262 g), oxalic acid (0.072 g), methanol (3.00 mL) and H₂O (2.00 mL) was sealed in a 25 mL Teflon-lined autoclave and heated at 170 °C for 6 days.

[tmpip]_0.5[Ga₃(OH)(PO₄)₃]·(H₂O)_0.25 (2). A mixture of Ga₂O₃ (0.148 g), H₃PO₄ (0.23 mL), 1,4-dimethylpiperazine (0.34 mL), methanol (3.00 mL) and H₂O (3.00 mL) was sealed in a 25 mL Teflon-lined autoclave and heated at 160 °C for 6 days.

[mpy][Ga₃F(PO₄)₃]·(H₂O)_0.25 (3). A mixture of Ga₂O₃ (0.131 g), H₃PO₃ (0.245 g), pyridine (0.40 mL), methanol (1.50 mL), HF (0.15 mL) and H₂O (4.50 mL) was sealed in a 25 mL Teflon-lined autoclave and heated at 160 °C for 8 days.

[dmbpy]_0.5[Ga₂(HPO₄)₂(PO₄)] (4). A mixture of Ga₂O₃ (0.115 g), H₃PO₄ (0.39 mL), 4,4′-bipyridine (0.62 g), methanol (1.00 mL) and H₂O (5.00 mL) was sealed in a 25 mL Teflon-lined autoclave and heated at 160 °C for 6 days.

[Hmpip]₂[Ga₇(OH)(PO₄)₆(HPO₄)₂] (5). A mixture of Ga₂O₃ (0.085 g), H₃PO₄ (0.35 mL), 1-methylpiperazine (0.18 mL), and H₂O (5.00 mL) was sealed in a 25 mL Teflon-lined autoclave and heated at 170 °C for 7 days.

After cooling to room temperature, the resulting product, containing colorless crystals of 1–5, was recovered by filtration, washed with distilled water, and dried in air. The yield based on gallium is 82.4%, 83.6%, 72.2%, 76.5%and 58.2% for 1–5, respectively. Attempts to replace 1,4-dimethylpiperazine by 1-methylpiperazine or piperazine under similar methanol-aqueous conditions also results in the formation of 2.

Crystallographic data collection and refinement

Single-crystal X-ray diffraction data for compounds 1–5 were collected on a Siemens SMART CCD diffractometer at 293 K with graphite-monochromated Mo Kα (λ = 0.71073 Å) in the ω and φ scanning mode. All absorption corrections were performed using the multiscan program. All structures were solved by direct methods and refined by full-matrix least-squares methods on F² with the SHELXL-97 program package.²⁰ The gallium and phosphorus atoms were first located, whereas the carbon, nitrogen, oxygen and fluorine atoms were found in the successive Fourier difference maps. The hydrogen atoms associated with the terminal P–O groups and the organic template cations were placed geometrically and refined using a riding model, except the hydrogen atoms of dmdabco in compound 1 and those of mpip in 5 were not fully dealt with due to the disorder of the organic ligands. The occupation factor of O1W in compounds 2 and 3 was determined as 0.25 respectively according to the results of CHN analysis. No hydrogen atoms associated with the water molecules were located from the difference Fourier map. All of the non-hydrogen atoms were refined anisotropically. Crystallographic data and structure refinements for compounds 1–5 are given in Table 1.†

Table 1 Crystal data and structure refinement parameters for compounds 1–5

	1	2	3	4	5
a R1 = ∑||Fo| − |Fc||/∑|Fo|.b wR2 = {∑[w(Fo^2 − Fc^2)^2]/∑w(Fo^2)^2}^1/2.
Formula	C4NH11GaP2O8	C4NH10.5Ga3P2O12.25	C6NH8.5Ga3P3O12.25	C6NH9Ga2P3O12	C10H30N4Ga7P8O34
Fw	332.80	586.70	611.71	519.49	1486.18
T/K	295(2)	295(2)	295(2)	295(2)	295(2)
λ/Å	0.71073	0.71073	0.71073	0.71073	0.71073
Crystal system	Triclinic	Triclinic	Triclinic	Monoclinic	Monoclinic
Space group	P1̄	P1̄	P1̄	P21/c	C2/c
a/Å	8.3010(7)	8.6814(7)	9.3128(7)	8.5770(5)	13.9877(8)
b/Å	8.3046(7)	9.2316(8)	9.3227(5)	20.4760(4)	11.9198(4)
c/Å	8.9569(8)	9.5849(3)	9.3497(3)	8.9999(8)	23.0454(6)
α/°	66.510(3)	100.270(3)	103.170(3)	90	90
β/°	65.630(5)	94.600(4)	93.600(4)	111.013(3)	104.269(3)
γ/°	76.000(2)	92.960(6)	90.140(5)	90	90
V/Å^3	513.61(8)	751.72(9)	788.72(8)	1475.48(16)	3723.8(3)
D/g cm^−3	2.152	2.592	2.576	2.339	2.651
Z	2	2	2	4	4
F(000)	334	571	593	1020	2908
μ/mm^−1	3.019	5.729	5.471	4.043	5.456
Collected reflections	5127	7034	7581	15 445	18 109
Unique reflections	2129	2956	3106	3366	3869
R(int)	0.0318	0.0710	0.0339	0.0691	0.0721
GOF on F^2	1.117	1.136	1.012	1.095	1.135
R1^a/wR2^b [I > 2σ(I)]	0.0368/0.0983	0.0452/0.1242	0.0325/0.0632	0.0437/0.0806	0.0444/0.0752

---

Results

Elemental analytical data correspond to: C₄H₁₁NO₈P₂Ga (1): C, 14.43; H, 3.33; N, 4.21. Found: C, 14.32; H, 3.19; N, 4.13. C₄NH_10.5Ga₃P₂O_12.25 (2): C, 8.18; H, 1.80; N, 2.38. Found: C, 8.04; H, 1.56; N, 2.23. C₆H_8.5FNO_12.25P₃Ga₃ (3): C, 11.78; H, 1.40; N, 2.28. Found: C, 11.69; H, 1.52; N, 2.19. C₆H₉NO₁₂P₃Ga₂ (4): C, 13.87; H, 1.75; N, 2.69. Found: C, 13.76; H, 1.83; N, 2.60. C₁₀H₃₀N₄O₃₄P₈Ga₇ (5): C, 8.08; H, 2.03; N, 3.77. Found: C, 7.86; H, 1.92; N, 3.58.

FT-IR (KBr pellets, cm⁻¹): for 1: 3443(m), 3029(w), 2920(w), 1643(s), 1476(s), 1384(m), 1154(s), 1010(s), 610(s), 515(w), 475(m); for 2: 3416(s), 3022(w), 2981(w), 2926(w), 1629(m), 1466(m), 1384(m), 1093(s), 1025(s), 685(m), 638(m), 610(m), 475(m); for 3: 3449(s), 3090(w), 1636(m), 1500(m), 1446(m), 1398(m), 1100(s), 1045(s), 776(m), 658(m), 583(w), 468(m), 449(m); for 4: 3446(s), 1643(s), 1514(w), 1459(m), 1391(m), 1093(s), 1025(s), 965(s), 827(w), 705(w), 610(m), 515(m), 468(m); for 5: 3456(s), 1636(s), 1466(w), 1378(m), 1086(s), 964(m), 916(m), 665(w), 614(m), 577(m), 475(m).

Discussion

Synthesis and in situ template generation strategy considerations

Hydrothermal methods have been successfully utilized to produce a great deal of crystalline materials, because different solubilities of organic and inorganic precursors can be achieved. Notably, this method has witnessed increasing success in making coordination complexes recently with in situ synthesized ligands.¹⁸ There is of course no reason why this in situ strategy cannot be used to design and make new organic templates suitable for the preparation of porous materials. Since Morris and co-workers developed ionothermal synthesis method for preparing molecular sieves, there are many examples in the literature of ionic liquids being used as not only non-conventional replacement solvents, but also structural templates to the resultant frameworks.^13a Considering that the chemical structures of the organic cations of most ionic liquids feature alkyl-substituted groups, one idea would be to design more alternatives as SDAs for the hydrothermal synthesis of zeolites and other porous materials. Therefore, in situ alkylation reactions of organic species are one such option. Here, several cyclic aliphatic and aromatic diamines are chosen as precursors to explore the in situ-template-synthesis of gallium phosphates. As expected, 1,4-diazabicyclo[2,2,2]octane, piperazine, 1 methylpiperazine and 1,4-dimethylpiperazine aliphatic molecules are in situ transformed into structure-related N,N′-dimethyl-1,4-diazabicyclo[2,2,2]octane (dmdabco²⁺) (in 1) and N,N,N′,N′-tetramethyl-piperazinium (tmpip²⁺) (in 2); pyridine and 4,4′-bipyridine aromatic species are in situ changed into N-methyl-pyridine (mpy⁺) (in 3) and N,N′-dimethyl-4,4′-bipyridine (dmpy²⁺) (in 4). Undoubtedly, hydrothermal in situ methylation occurred between CH₃OH and corresponding organic precursors, i.e. dabco in 1, piperazine, 1-methylpiperazine or 1,4-dimethylpiperazine in 2, pyridine in 3, and 4,4′-bipyridine in 4 (Scheme 1). By contrast, the 1-methylpiperazine was found to be protonated and serve as template in pure aqueous media (in 5). Actually, the hydrothermal in situ alkylation reaction was first observed in the preparation of [methylpyridinium]₂[Zn₂(ox)₃] by Evans and Lin, in which two in situ ligand reactions (i.e. methylation and oxidative coupling of methanol to oxalic acid) were involved.²¹ Similar in situ alkylation transformation occurring directly between methanol solvent and organic precursors, is unique and scarcely observed in porous metal phosphates/phosphites family.^19,22 The in situ formation of methyl viologen from 4,4′-bipyridine, for example, was recently reported in [C₆H₁₄N₂][Zn₆(PO₄)₄(HPO₄)(H₂O)₂], an open-framework zinc phosphate with reversible photovoltaic activity and photoluminescence properties.²² Compared with the traditional method of directly using noxious alkyl halides as raw material, such alkylation reaction is carried out in a sealed black casket and can be described as a green one. In addition, the present in situ methylation process is different from typical Eschweiler–Clarke methylation,²³ in which the methylation of the primary or secondary amine occurs in the presence of excess formic acid and formaldehyde.

Scheme 1 In situ formation of new templates dmdabco²⁺ in 1 (a); tmpip²⁺ in 2 (b); mpy⁺ in 3 (c) and dmbpy²⁺ in 4 (d).

The templating and the structure directing effects are important issues in the synthesis of inorganic porous materials. Protonated organic amines are perhaps the most popular SDAs residing either between the chains, layers or inside pores of the host framework, and their templating abilities may be evaluated in terms of the charge-density matching, space matching and nonbonding interactions of the host–guest systems. However, the specific role of SDAs in complex assembly to highly ordered open architectures still remains unclear, and the crystallization process is kinetically controlled and highly sensitive to many variables such as solvent, starting gel composition, pH value, temperature and time, etc. Given that the methylated SADs have chemical structures similar to those of commonly used organic amines, it is not surprising that they can be used as alternative SDAs functioning as charge compensators and space-fillers. One of the typical examples is the syntheses of zinc phosphites H₂tmdp·Zn₃(HPO₃)₄ and (H₂bpp)₂·Zn₆(HPO₃)₈ with layered structures^14a,b analogous to that of [pmpip][Zn₃(HPO₃)₄],^19a in which the methylated pmpip²⁺ (N,N,N′,N′-tetramethyl-2-methylpiperazinium) SDAs instead of diprotonated organic amines provide positive charge to balance the residual negative charges of the zinc phosphite sheets. Clearly, in situ-template-generation strategy may be a promising synthesis route for the preparations of porous materials.

Crystal structures of GaPOs

Compound 1. The asymmetric unit of 1 has one gallium atom and two phosphorus atoms. Ga(1) atom is tetrahedrally coordinated by oxygen atoms with the Ga–O bond lengths in the range of 1.814(3)–1.822(2) Å and O–Ga–O angles lying between 105.7(2)°and 113.4(2)°. Each phosphorus atom only makes two P–O–Ga linkages and possesses two terminal P–O bonds. The P–O distances are in the range 1.482(3)–1.560(3) Å (av. 1.532 Å), and the O–P–O bond angles span from 103.8(2) to 113.8(2)° (av. 109.4°). Bond valence sum values²⁴ imply that the P(1)–O(3) and P(2)–O(8) linkages with relatively longer distances of 1.555(3) and 1.560(3) Å are terminal P–OH bonds, while the remaining P(1)–O(4) and P(2)–O(7) linkages with relatively shorter distances of 1.482(3) and 1.486(3) Å are P=O double bonds. Considering the usual valence of Ga, P, O and H to be +3, +5, −2 and +1, the composition of [Ga(HPO₄)₂] generates a net charge of −1, which can be compensated by half of one dmdabco²⁺ cation.

The alternative linkage of GaO₄ and HPO₄ tetrahedral groups forms typical four-membered ring (4MR), which is further connected with each other to form 1D chains along the c axis (Fig. 1). Considered the Aufbau principle of high-dimensional frameworks from that of low-dimensional units, the presence of low-dimensional structures is of particular importance. Those noncoordinated P=O and P–OH groups protrude outside of the chains, which hinders the expansion of such chains into a higher-dimensional framework. It is well known that H-bond interactions play a vital role in the formation and stability of low-dimensional structures. In the present case, adjacent inorganic chains are linked with each other via strong O–H⋯O hydrogen-bonds [O(3)⋯O(7), 2.496(4) Å; O(8)⋯O(4), 2.530(5) Å], forming a 3D supramolecular architecture with regular channels in multinational directions (Fig. S6†). The guest dmdabco²⁺ cations, residing at the center of the above mentioned channels, compensate the negative charges and simultaneously interact with the inorganic framework through C–H⋯O hydrogen-bonding interactions (Table S1†).²⁵ PLATON calculations suggest that the accessible volume is 40.8% of the unit cell volume, which is occupied by extra-framework species.²⁵

Fig. 1 The infinite chain formed by the connectivity of corner-sharing 4-rings in 1.

Compounds 2 and 3. Compounds 2 and 3 both crystallize in the triclinic P1̄ space group, and they exhibit similar 3D inorganic frameworks with intersecting 8-MR channels. Each asymmetric unit of compounds 2 and 3 consists of three gallium and three phosphorus atoms (Fig. S7†). In 2, two types of coordination are observed for the three gallium atoms. Ga(1) atom bonds to four bridging O atoms from the adjacent P atoms with Ga–O bond lengths in the range of 1.809(3)–1.839(4) Å (av. 1.826 Å). The remaining Ga(2) and Ga(3) atoms are both five-coordinated to form GaO₄(OH) trigonal bipyramids and each shares four O atoms with neighboring four PO₄ tetrahedron and one independent hydroxyl oxygen atom (O(13)) respectively. It is worth mentioning that the hydroxyl unit is in bridging positions and makes typical Ga(2)–O(13)H–Ga(3) linkages. The slight distortion is observed for both GaO₄(OH) trigonal bipyramid, and the Ga–O bond lengths vary from 1.854(3) to 1.977(3) Å, and the O–Ga–O bond angles are in the range 86.7(2)–177.3(2)°. In 3, however, another two different types of coordination environments are observed for the three distinct gallium atoms. One gallium atom (Ga(1)) exhibits distorted octahedral geometry coordinated by four phosphate oxygen atoms with the distance of Ga(1)–O(2), Ga(1)–O(4), Ga(1)–O(5), and Ga(1)–O(9) bond lengths being 1.946(3), 1.934(3), 1.893(3) and 1.907(3) Å, respectively, and two F atoms with relatively longer Ga–F bond length being 1.973(2) and 1.983(2) Å, respectively. It is crystallographically difficult to identify F⁻ and OH⁻ bridges; however, refinement of the bridging atom as oxygen leads to nonpositive defined anisotropic thermal parameter. In addition, bond valence sum calculations support the presence of fluorination we propose (Table S2†). For Ga(1) polyhedron, two F atoms are in bridging positions corresponding to the F–Ga–F angle being 80.2(2)°. Another two Ga atoms, Ga(2) and Ga(3), are coordinated to four O atoms from phosphate with typical tetrahedral Ga–O distances range from 1.800(3) to 1.818(3) Å. Each of the three phosphorus atoms in 2 and 3 shares four O atoms with neighboring Ga atoms with P–O bond lengths in the scope of 1.508(4)–1.567(4) Å (compound 2) and 1.498(3)–1.555(3) Å (compound 3); the O–P–O bond angles are distributed in the range of 104.1(2)–113.4(2)° and 104.5(2)–116.8(2)° for 2 and 3 respectively. The inorganic framework compositions of [Ga₃(OH)(PO₄)₃] in 2 and [Ga₃F(PO₄)₃] in 3 both generate a net charge of −1, which can be compensated by half of one tmpip²⁺ cation and one mpy⁺ respectively.

The 3D anionic frameworks of 2 and 3 are isomorphous and constructed from similar secondary building units (SBUs), Ga₆P₆. As shown in Fig. 2, the Ga₆P₆ caged-structural cluster in 2 is built from two Ga(1)O₄ tetrahedra, two Ga(2)O₄(OH) and two Ga(3)O₄(OH) trigonal bipyramids, and six PO₄ tetrahedra via bridging oxygen atoms, while the similar Ga₆P₆ caged-structure in 3 is formed from the connectivity between two Ga(1)O₄F₂ octahedra, two Ga(2)O₄, two Ga(3)O₄ and six PO₄ tetrahedra. Interestingly, the hydroxyl groups (–O(13)H) in 2 only participate in the fusion of Ga(2) and Ga(3) polyhedral units within each Ga₆P₆ cluster, while the fluorine vertices within each Ga₆P₆ cluster in 3 serve as bridges to connect adjacent neighbors. In both compounds, each Ga₆P₆ cluster links six neighboring clusters to yield a 3D framework with pcu topology (Fig. 3 and 4). Intersecting 8-membered channels locate along the [100], [010], [001], [101] and [110] directions in 2 (Fig. S8†), and along the [100], [010], [001] and [111] directions in 3 (Fig. S9†). The extra-framework cations, i.e. tmpip²⁺ in 2 and mpy⁺ in 3, are orderly located in the middle region of the 8-ring channels in 2 and near the wall of the 8-ring channels in 3 respectively, balancing the negative charges and simultaneously interacting with the inorganic networks via C–H⋯O H-bonding interactions. PLATON calculations suggest that the accessible volume is 38.0% and 40.6% of per unit cell volume for 2 and 3, which is occupied by tmpip²⁺ in 2 and mpy⁺ species in 3.

Fig. 2 Distinct structures of the Ga₆P₆ cages in 2 (a) and 3 (b).

Fig. 3 View of the 3D structure of 2, showing 8-ring channels occupied by the tmpip²⁺ cations.

Fig. 4 View of the 3D structure of 3, showing 8-ring channels occupied by the mpy⁺ cations.

Compound 4. The asymmetric unit has 25 non-hydrogen atoms, of which two gallium atoms and three phosphorus atoms are crystallographically independent. Ga(1) has a typical tetrahedral coordination with four oxygen atoms, and Ga(2) is five coordinated to form a Ga(2)O₅ trigonal bipyramid geometry. The Ga–O distances are in the range of 1.791(3)–1.951(3) Å (av. 1.851 Å), and the O–Ga–O bond angles are in the scope of 87.6(2)–178.5(2)°. Of the three P atoms, P(1) only forms two P–O–Ga bonds and possess two free P–O linkages (P(1)–O(1) and P(1)–O(2)); P(2) forms three P–O–Ga linkages with one free P–O bond (P(2)–O(8)); P(3) produces four P–O–Ga bonds with free P=O or P–OH bonds. The P–O distances are in the range of 1.495(3)–1.556(3) Å, and the O–P–O bond angles are in the scope of 104.4(4)–115.0(2)°. Bond valence sum values²⁴ imply that the P(1)–O(1) and P(2)–O(8) linkages with relatively longer distances of 1.545(3) and 1.556(3) Å are P–OH bonds, while the P(1)–O(2) linkage with relatively shorter distance of 1.494(3) Å should be P=O bond. The composition of [Ga₂(HPO₄)₂(PO₄)] generates a net charge of −1, which can be compensated by half of one dmbpy²⁺ cation.

The corner-sharing assembly of these Ga(1)O₄ tetrahedra, Ga(2)O₅ trigonal bipyramids, three-connected P(2)O₃(OH) and four-connected P(3)O₄ units makes up an inorganic layer in the ac plane (Fig. 5a). Those two-connected P(1)O₃(OH) tetrahedral units, however, are grafted alternatively up and down on to this layer giving rise to a complex anionic layer with the formula [Ga₂(HPO₄)₂(PO₄)]¹⁻. Obviously, those free P(1)=O and P(1)–OH groups pointing into the interlayer space, are non-coordinated, thus hindering the further expansion of such layers into a 3D architecture (Fig. S10†). Adjacent inorganic layers are further stacked exactly one over the other in an –ABAB– fashion, and interconnected with each other via strong O–H⋯O hydrogen-bonds [O(1)⋯O(2), 2.521(4) Å; O(8)⋯O(2), 2.527(4) Å], giving rise to a 3D supramolecular network with ellipse-like channels (Fig. 5b). The well-ordered dmbpy²⁺ guest templates, balancing the negative charges of inorganic network, locate in the middle of these channels and interact with the host framework via extensive C–H⋯O type H-bonds. PLATON calculations suggest that the accessible volume is 40.6% of the unit cell volume, which is occupied by extra-framework organic cations.

Fig. 5 (a) Polyhedral view the inorganic layer along the [001] direction with 6-ring windows in 4; (b) polyhedral view along the [010] direction showing the hydrogen-bonded 3D network with pseudo-one-dimensional channels occupied by the dmbpy²⁺ cations. Color code: gallium polyhedra, green; phosphate tetrahedra, purple.

Compound 5. The asymmetric unit contains four gallium atoms and four phosphorus atoms, of which Ga(4) occupies a special position with a site occupancy of 0.5. The coordination of the four distinct gallium atoms is divided into three types. The first type of Ga(1) atom is six coordinated to form a Ga(1)O₆ octahedron and share five bridging oxygen atoms with the adjacent P atoms [the average Ga(1)–O bond: 1.967 Å], and one bridging oxygen atom O(6) with Ga(2) [Ga(1)–O(6) = 1.990(3) Å]. The second type of Ga(2) atom adopts five-coordination geometry by five oxygen atoms, all of which are bridging to the adjacent phosphate groups [Ga(2)–O(7) = 1.903(4) Å, Ga(2)–O(8) = 1.967(4) Å, Ga(2)–O(9) = 1.874(4) Å, Ga(2)–O(12) = 1.880(4) Å, Ga(2)–O(6) = 1.990(3) Å] except for the O(6) bridging to Ga(1) [Ga(2)–O(6) = 1.855(4) Å]. The remaining Ga(3) and Ga(4) atoms are both four-coordinated by oxygen atoms to form typical tetrahedral coordination with distances of Ga–O ranging from 1.810(4) to 1.842(4) Å. All P atoms adopt tetrahedral coordination geometry. Atoms P(1), P(3) and P(4) each forms four P–O–Ga linkages, while P(2) atom forms three P–O–Ga linkages and possesses one free P–O bond. Bond valence sum values²⁴ indicate that the terminal P(2)–O(11) linkage with a distance of 1.541(4) Å is a P–OH bond, and this assignment also corresponds well with the proton positions observed near the oxygen atoms in the difference Fourier maps. The P–O bond lengths vary from 1.487(4) to 1.582(4) Å, and the O–P–O bond angles are in the range 102.1(2)–117.4(2)°.

The structure of 5 is 3D, composed of vertex-sharing GaO₄, GaO₅, GaO₆, PO₄ and HPO₄ units. Interestingly, this 3D open-framework structure could be comprehended as the assembly of two kinds of SBUs, Ga₆P₈ and Ga(4)O₄ groups. The Ga₆P₈ cluster is built from two Ga(1)O₅(OH) octahedra, two Ga(2)O₄(OH) trigonal bipyramids, two Ga(3)O₄ tetrahedra, two P(1)O₄, two HP(2)O₄, two P(3)O₄ and two P(4)O₄ units (Fig. S11†). Such Ga₆P₈ SBUs are interconnected to give rise to a 2D layer with 8-membered apertures (Fig. 6a). Adjacent layers are further interconnected by sharing vertex-oxygen atoms with Ga(4)O₄ tetrahedra to generate the final 3D open-framework. The structure appears to exhibit elliptical 8-ring channels along the [010] direction. Actually, the actual channels along this direction are helical and the pitch of the helix is 11.92 Å. The oxygen-to-oxygen distance across the helical channels is 7.9 Å × 8.1 Å. It should be noted that this compound does not crystallize in a chiral space group, so both left- and ring-handed helices coexist in the structure with a ratio of 50 : 50. The right-handed and left-handed helical chains are constructed from the infinite connection of –Ga(4)–O(17)–P(4)–O(8)–Ga(2)–O(9)–P(2)–O(10)– and –Ga(4)–O(10)–P(2)–O(9)–Ga(2)–O(8)–P(4)–O(17)–, respectively (Fig. 6b). The protonated 1-methylpiperazinecationsare orderly arranged in the free thread of the helical chains, balancing the negative charges of the inorganic network. Interestingly, the amino groups of the organic molecules interact with the bridging oxygen atoms of the helical chains via strong H-bonds, making these templates arranged around 2₁ screw axes along b axis. To our knowledge, the occurrence of hydrogen bonded helices in host–guest systems is still rare hitherto, and only a few porous materials possess similar structural feature, including layered aluminophosphates (C₂H₈N)₂[Al₂(HPO₄)(PO₄)₂] and [C₆N₃H₁₇][Al₂(HPO₄)(PO₄)₂], germinate JLG-2 and gallium phosphite Ga(HPO₃)F₃·(trans-C₆N₂H₁₆).²⁶ In addition, large 10-ring window is located along the [110] direction (Fig. 6c). It was composed of one GaO₆ octahedra, one GaO₅ trigonal bipyramids, three GaO₃ and five PO₄ tetrahedral groups, with a diameter size of ca. 6.6 × 9.9 Å. PLATON calculations suggest that the accessible volume is 35.6% of per unit cell volume, which is occupied by extra-framework organic cations.

Fig. 6 (a) View of the 2D layered structure formed by the connectivity of corner-sharing Ga₆P₈ clusters in 5; (b) a space-filling view of left- and right-handed helical chains with organic cations following the 2₁ screw axis; (c) polyhedral view of the structure along the [110] direction with 10-ring channels. Color code: gallium polyhedra, green; phosphate tetrahedra, purple.

PXRD analyses

The X-ray powder diffraction profiles of compounds 1–5 are in agreement with the simulated patterns from single-crystal X-ray structure data, indicating the purity of the as-synthesized samples (Fig. S1–S5†). The difference in reflection intensities between the simulated and experimental patterns was due to the variation in preferred orientation of the powder sample during collection of the experimental XRD data.

IR, thermogravimetric analysis

In IR spectra (Fig. S12–S16†), the broad bands at 3456–3416 cm⁻¹ can be attributable to the stretching vibrations of the O–H bond. The bands at 3130–2869 cm⁻¹ are caused by the stretching vibrations of C–H bonds, and the C–H bending vibrations are observed at around 1643–1378 cm⁻¹, which agrees with the participation of organic templates. The stretching vibrations of P–O bonds are responsible for the bands at 1154–964 cm⁻¹, where as the bands at 685–449 cm⁻¹ are assigned to the bending vibrations of P–O bonds.

Initial thermogravimetric analyses of the powder samples of 1–5 were carried out to study their thermal stabilities (Fig. 7). The structure of 1 can be stable before 340 °C and then underwent a continuous two-step weight loss, attributing to the decomposition and removal of the organic amines. However, the experimental total weight loss (17.70%) was a little lower than the theoretical value (21.36%), and the lower reduction in this stage was ascribed to the retention of carbon in the solid residue (gray in color). Similar phenomenon was also observed in other reported gallium phosphates/phosphites.²⁷ For compounds 2 and 3, the weight loss of 0.63% in the range 50–130 °C, and 0.80% in the range 56–165 °C were firstly observed, which can be attributable to the release of crystallizing H₂O molecule (calcd: 0.77% for 2 and 0.74% for 3). And these values also match with the results from structure refinements. On further heating, sharp weight losses of 12.72% and 16.07% were observed in the temperature region 260–745 °C for 2 and 365–495 °C for 3, which agree well with the removal of the organic amines (calcd: 12.27% for 2 and 15.38% for 3). For 4, the structure can be stable up to 385 °C, and then underwent a continuous weight loss in which the organic molecules began to decompose. As observed in compound 2, the observed value for mbpy templates (9.82%) was much lower than the expected value (17.91), because of the retention of carbon in the solid residue (black in color). For 5, a total weight loss of 14.13% between 185 and 720 °C was observed, assigning to the combustion of the organic molecules (calcd: 13.46%).

Fig. 7 TG curves of compounds 1–5.

Conclusions

In summary, five new organically templated gallium phosphates [dmdabco]_0.5[Ga(HPO₄)₂] 1, [tmpip]_0.5[Ga₃(OH)(PO₄)₃]·(H₂O)_0.25 2, [mpy][Ga₃F(PO₄)₃]·(H₂O)_0.25 3, [dmbpy]_0.5[Ga₂(HPO₄)₂(PO₄)] 4, and [Hmpip]₂[Ga₇(OH)(PO₄)₆(HPO₄)₂] 5, have been hydrothermally prepared. Employing in situ-template-synthesis synthetic route, four new templating agents dmdabco²⁺, tmpip²⁺, mpy⁺ and dmbpy²⁺ were generated from the direct methylation reactions between CH₃OH solvent and the corresponding organic-amine precursors. Such hydrothermal in situ methylation reactions on aliphatic and aromatic amine precursors are noteworthy, and to the best of our knowledge, firstly observed in gallium-containing compounds. Compared to the conventional Eschweiler–Clarke methylation, the present methylation transformation direct from methanol molecules is simpler and more environmentally friendly. Under different N-alkyl substituted templates, the inorganic frameworks of compounds 1–5 possess different dimensions: infinite chain in 1, 3D pcu framework with multinational intersecting 8-ring channels in 2 and 3, 2D distinct layer structure in 4 and 3D architecture with interesting helical channels in 5. This study may provide fresh opportunity to in situ design and synthesize versatile N-alkylated derivatives used as templates in the preparation of crystalline porous materials. Further work on this subject is in progress.

Acknowledgements

This work was supported by the Natural Science Foundation of China (20901043), A Project of Shandong Province Higher Educational Science and Technology Program (J13LD18), the Young Scientist Foundation of Shandong Province (BS2009CL041), the development project of Qingdao science and technology (13-1-4-187-jch) and Taishan Scholar Program.

Notes and references

1. (a) M. E. Davis, Nature, 2002, 417, 813–821; (b) N. Zhang, P. Feng and X. Bu, Acc. Chem. Res., 2005, 38, 293–303; (c) E. R. Parnham and R. E. Morris, Acc. Chem. Res., 2007, 40, 1005–1013; (d) S. Natarajan and S. Mandal, Angew. Chem., Int. Ed., 2008, 47, 4798–4828; (e) J. H. Yu and R. R. Xu, Acc. Chem. Res., 2010, 43, 1195–1204; (f) J. Jiang, J. Yu and A. Corma, Angew. Chem., Int. Ed., 2010, 49, 3120–3145; (g) H. Y. Lin, C. Y. Chin, H. L. Huang, W. Y. Huang, M. J. Sie, L. H. Huang, Y. H. Lee, C. H. Lin, K. H. Lii, X. Bu and S. L. Wang, Science, 2013, 339, 811–813.
2. A. K. Cheetham, G. Férey and T. Loiseau, Angew. Chem., Int. Ed., 1999, 38, 3268–3292.
3. (a) S. T. Wilson, B. M. Lok, C. A. Messina, T. R. Cannan and E. M. Flanigen, J. Am. Chem. Soc., 1982, 104, 1146–1147; (b) Q. Huo, R. Xu, S. Li, Z. Ma, J. M. Thomas, R. H. Jones and A. M. Chippindale, J. Chem. Soc., Chem. Commun., 1992, 875–876; (c) J. H. Yu and R. R. Xu, Chem. Soc. Rev., 2006, 35, 593–604; (d) Y. L. Lai, K. H. Lii and S. L. Wang, J. Am. Chem. Soc., 2007, 129, 5350–5351.
4. (a) G. Y. Yang and S. C. Sevov, J. Am. Chem. Soc., 1999, 121, 8389–8390; (b) Y. Wang, J. Yu, M. Guo and R. Xu, Angew. Chem., Int. Ed., 2003, 42, 4089–4092; (c) W. M. Chang, M. Y. Cheng, Y. C. Liao, M. C. Chang and S. L. Wang, Chem. Mater., 2007, 19, 6114–6119; (d) P. Chen, J. Y. Li, F. Z. Duan, J. H. Yu, R. R. Xu and R. P. Sharma, Inorg. Chem., 2007, 46, 6683–6687; (e) Y. Yang, Y. N. Zhao, J. G. Yu, S. Z. Wu and R. J. Wang, Inorg. Chem., 2008, 47, 769–771; (f) P. C. Jhang, N. T. Chuang and S. L. Wang, Angew. Chem., Int. Ed., 2010, 49, 4200–4204; (g) J. D. Feng, S. W. Tang, K. Z. Shao, R. S. Wang, C. Yao, H. M. Xie and Z. M. Su, CrystEngComm, 2010, 12, 3448–3451; (h) J. Z. Zhang, G. S. Yao, J. Liao, J. C. Dai and Z. Y. Fu, J. Mater. Chem. A, 2013, 1, 4945–4948; (i) J. B. Wu, Y. Yan, B. K. Liu, X. L. Wang, J. Y. Li and J. H. Yu, Chem. Commun., 2013, 49, 4995–4997.
5. (a) N. Guillou, Q. Gao, M. Nogués, R. E. Morris, M. Hervieu, G. Férey and A. K. Cheetham, C. R. Acad. Bulg. Sci., 1999, 2, 387–392; (b) J. M. Li, Y. X. Ke, Y. G. Zhang, G. F. He, Z. Jiang, M. Nishiura and T. Imamoto, J. Am. Chem. Soc., 2000, 122, 6110–6111; (c) Y. Q. Yu, W. Zhan and T. E. Albrecht-Schmitt, Inorg. Chem., 2008, 47, 9050–9054; (d) L. Liu, J. F. Yang, J. P. Li, J. X. Dong, D. Sisak, M. Luzzatto and L. B. McCusker, Angew. Chem., Int. Ed., 2011, 50, 8139–8142.
6. (a) S. Weigel, S. C. Weston, A. K. Cheetham and G. D. Stucky, Chem. Mater., 1997, 9, 1293–1295; (b) S. M. Stalder and A. P. Wilkinson, Chem. Mater., 1997, 9, 2168–2173; (c) M. A. Leech, A. R. Cowley, C. K. Prout and A. M. Chippindale, Chem. Mater., 1998, 10, 451–456; (d) A. M. Chippindale, Chem. Mater., 2000, 12, 818–822; (e) C. Y. Chen, F. R. Lo and K. H. Lii, Chem. Commun., 2000, 1061–1062; (f) C. H. Lin, Y. C. Yang, C. Y. Chen and S. L. Wang, Chem. Mater., 2006, 18, 2095–2101; (g) S. Lohmeier, M. Wiebcke and P. Behrens, Z. Anorg. Allg. Chem., 2008, 634, 147–152; (h) G. P. Zhou, Y. L. Yang, R. Q. Fan, X. R. Liu, H. W. Hong and F. P. Wang, Solid State Sci., 2010, 12, 1103–1106; (i) C. Martineau, T. Loiseau, L. Beitone, G. Férey, B. Bouchevreau and F. Taulelle, Dalton Trans., 2013, 42, 422–431.
7. C. H. Lin, S. L. Wang and K. H. Lii, J. Am. Chem. Soc., 2001, 123, 4649–4650.
8. (a) R. I. Walton, F. Millange, T. Loiseau, D. O'Hare and G. Férey, Angew. Chem., Int. Ed., 2000, 39, 4552–4555; (b) M. Esterman, L. B. McCusker, C. Baerlocher, A. Merrouche and H. Kessler, Nature, 1991, 352, 320–323.
9. (a) C. Sassoye, T. Loiseau, F. Taulelle and G. Férey, Chem. Commun., 2000, 943–944; (b) L. Beitone, J. Marrot, T. Loiseau, G. Férey, M. Henry, C. Huguenard, A. Gansmuller and F. Taulelle, J. Am. Chem. Soc., 2003, 12, 1912–1922.
10. Y. C. Yang and S. L. Wang, J. Am. Chem. Soc., 2008, 130, 1146–1147.
11. C. H. Lin and S. L. Wang, Inorg. Chem., 2001, 40, 2918–2921.
12. (a) C. Sassoye, J. Marrot, T. Loiseau and G. Férey, Chem. Mater., 2002, 14, 1340–1347; (b) L. Wang, T. Y. Song, Y. Fan, Y. Wang, J. N. Xu, S. H. Shi and T. Zhu, J. Solid State Chem., 2006, 179, 824–829; (c) T. Loiseau and G. Férey, J. Fluorine Chem., 2007, 128, 413–422; (d) L. L. Huang, T. Y. Song, L. R. Zhang, Y. L. Chen, J. H. Jiang, J. N. Xu and L. Wang, CrystEngComm, 2010, 12, 2198–2202; (e) P. Ramaswamy, S. Mandal and S. Natarajan, Inorg. Chim. Acta, 2011, 372, 136–144.
13. (a) E. R. Cooper, C. D. Andrews, P. S. Weathley, P. B. Webb, P. Wormald and R. E. Morris, Nature, 2004, 430, 1012–1016; (b) E. R. Parnham, P. S. Weathley and R. E. Morris, Chem. Commun., 2006, 380–382; (c) E. R. Parnham and R. E. Morris, J. Am. Chem. Soc., 2006, 128, 2204–2205; (d) E. R. Parnham and R. E. Morris, Chem. Mater., 2006, 18, 4882–4887; (e) E. R. Parnham, E. A. Drylie, P. S. Weathley, A. M. Z. Slawin and R. E. Morris, Angew. Chem., Int. Ed., 2006, 45, 4962–4966; (f) E. A. Drylie, D. S. Wragg, E. R. Parnham, P. S. Wheatley, A. M. Z. Slawin, J. E. Warren and R. E. Morris, Angew. Chem., Int. Ed., 2007, 46, 7839–7843; (g) R. E. Morris, Chem. Commun., 2009, 2990–2998; (h) J. D. Feng, S. W. Tang, K. Z. Shao, R. S. Wang, C. Yao, H. M. Xie and Z. M. Su, CrystEngComm, 2010, 12, 3448–3451; (i) Z. Ma, J. Yu and S. Dai, Adv. Mater., 2010, 22, 261–285; (j) H. Z. Xing, W. T. Yang, T. Su, Y. Li, J. Xu, T. Nakano, J. H. Yu and R. R. Xu, Angew. Chem., Int. Ed., 2010, 49, 2328–2331.
14. (a) Z. Lin, W. Fan, F. Gao, N. Chino, T. Yokoi and T. Okubo, Cryst. Growth Des., 2006, 6, 2435–2437; (b) Z. Lin, H. P. Nayek and S. Dehnen, Inorg. Chem., 2009, 48, 3517–3519; (c) X. C. Luo, M. C. Gong, Y. Q. Chen and Z. E. Lin, Microporous Mesoporous Mater., 2010, 131, 418–422; (d) C. M. Duan, D. B. Luo, H. M. Zeng, M. P. Kang and Z. E. Lin, CrystEngComm, 2012, 14, 5734–5736; (e) P. Ramaswamy, N. N. Hegde, R. Prabhu, V. M. Vidya, A. Datta and S. Natarajan, Inorg. Chem., 2009, 48, 11697–11711; (f) Z. X. Chen, H. Y. Yang, M. L. Deng, Y. Ling, L. H. Weng and Y. M. Zhou, Dalton Trans., 2012, 41, 4079–4083; (g) Z. Y. Fu, D. S. Song, Y. Zeng, S. J. Liao and J. C. Dai, Dalton Trans., 2012, 41, 10910–10912.
15. P. Y. Feng, X. Bu and G. D. Stucky, Nature, 1997, 388, 735–740.
16. (a) L. Lakiss, A. Simon-Masseron and J. Patarin, Microporous Mesoporous Mater., 2005, 84, 50–58; (b) L. Vidal, C. Pray and J. Patarin, Microporous Mesoporous Mater., 2000, 39, 113–116.
17. A. L. Lu, N. Li, Y. F. Ma, H. B. Song, D. P. Li, N. J. Guan, H. G. Wang and S. H. Xiang, Cryst. Growth Des., 2008, 8, 2377–2383.
18. (a) J. Black, N. R. Champness, S. S. M. Chung, W. S. Li and M. Schröder, Chem. Commun., 1997, 1675–1676; (b) N. F. Zheng, X. Bu and P. Y. Feng, J. Am. Chem. Soc., 2002, 124, 9688–9689; (c) D. Li and T. Wu, Inorg. Chem., 2005, 44, 1175–1177; (d) X. M. Zhang, Coord. Chem. Rev., 2005, 249, 1201–1219; (e) X. M. Chen and M. L. Tong, Acc. Chem. Res., 2007, 40, 162–170; (f) H. Zhao, Z. R. Qu, H. Y. Ye and R. G. Xiong, Chem. Soc. Rev., 2008, 37, 84–100; (g) Z. J. Zhang, S. C. Xiang, G. C. Guo, G. Xu, M. S. Wang, J. P. Zou, S. P. Guo and J. S. Huang, Angew. Chem., Int. Ed., 2008, 47, 4149–4152; (h) H. B. Zhu and S. H. Gou, Coord. Chem. Rev., 2011, 255, 318–338; (i) Y. Q. Zhao, M. X. Fang, Z. H. Xu, X. P. Wang, S. N. Wang, L. L. Han, X. Y. Li and D. Sun, CrystEngComm, 2014, 16, 3015–3019.
19. (a) G. M. Wang, J. H. Li, X. Zhang, P. Wang, B. B. Pang, Z. H. Wang, Y. X. Wang, J. H. Lin and C. Y. Pan, Dalton Trans., 2013, 42, 13084–13091; (b) G. M. Wang, Z. G. Ding, J. H. Li, X. Lv, X. Zhang, X. M. Zhao, Z. H. Wang, Y. X. Wang and J. H. Lin, CrystEngComm, 2014, 16, 3296–3304.
20. G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr., 2008, 64, 112–122; G. M. Sheldrick, SHELXS97, Program for Crystal Structure Solution, University of Göttingen, Göttingen, 1997; G. M. Sheldrick, SHELXL97, Program for Crystal Structure Refinement, University of Göttingen, Göttingen, 1997.
21. O. R. Evans and W. B. Lin, Cryst. Growth Des., 2004, 1, 9–11.
22. J. B. Wu, Y. Yan, B. K. Liu, X. L. Wang, J. Y. Li and J. H. Yu, Chem. Commun., 2013, 49, 4995–4997.
23. (a) H. T. Clarke, H. B. Gillespie and S. Z. Weisshaus, J. Am. Chem. Soc., 1933, 55, 4571–4578; (b) A. C. Cope and W. D. Burrows, J. Org. Chem., 1965, 30, 2163–2165.
24. I. D. Brown and D. Aldermatt, Acta Crystallogr., Sect. B: Struct. Sci., 1985, 41, 244–247.
25. A. L. Spek, Acta Crystallogr., Sect. A: Fundam. Crystallogr., 1990, 46, C34.
26. (a) M. Wang, J. Y. Li, J. H. Yu, Q. H. Pan, X. W. Song and R. R. Xu, Inorg. Chem., 2005, 44, 4604–4607; (b) Y. Song, J. H. Yu, Y. Li, G. H. Li and R. R. Xu, Angew. Chem., Int. Ed., 2006, 45, 2546–2548; (c) Q. H. Pan, X. Y. Ren, Y. Xu and W. F. Yan, Inorg. Chem. Commun., 2011, 14, 1842–1845; (d) L. L. Huang, T. Y. Song, L. R. Zhang, Y. L. Chen, J. H. Jiang, J. N. Xu and L. Wang, CrystEngComm, 2010, 12, 2198–2202.
27. (a) Z. E. Lin, J. Zhang, Y. Q. Sun and G. Y. Yang, Inorg. Chem., 2004, 43, 797–801; (b) L. L. Huang, T. Y. Song, Y. Fan, L. R. Zhang, H. Yang, Z. F. Tian and L. Wang, Inorg. Chim. Acta, 2009, 362, 3030–3034; (c) C. X. Li, L. L. Huang, M. D. Zhou, J. Xia, H. W. Ma, S. L. Zang and L. Wang, J. Solid State Chem., 2013, 208, 86–92.

---

Footnote

† Electronic supplementary information (ESI) available: X-ray data in CIF format, simulated and experimental XRPD, related figures, IR spectra, detailed hydrogen bonds information and selected bond lengths and angles. CCDC 1041971–1041975. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c5ra12300c

---