An interwoven Fe3L3 trigonal metallamacrocycle from an in situ ligand hydrolysis reaction

Qing-Fu Zhang ab, Fei-Long Jiang a, You-Gui Huang a, Wei Wei ab, Qiang Gao ab, Ming Yang ab, Ke-Cai Xiong ab and Mao-Chun Hong *a
aState Key Laboratory of Structural Chemistry, Fujian Institute of Research on the Structure of Matter, Chinese Academy of Sciences, Fuzhou, Fujian 350002, China. E-mail: hmc@fjirsm.ac.cn; Fax: +86 591 8371 4946; Tel: +86 591 8379 2460
bGraduate School of the Chinese Academy of Sciences, Beijing, China

Received 13th October 2008 , Accepted 8th December 2008

First published on 24th December 2008


Abstract

Through an in situhydrolysis reaction of the bishydrazone ligand H4L1 [H4L1 = (HOC6H4)CH[double bond, length as m-dash]NNHCO(C5H3N)CONHN[double bond, length as m-dash]CH(C6H4OH)], in the presence of Fe(III) ions, an interwoven trigonal metallamacrocycle [Fe3L3(H2O)3]·9H2O (1) [H3L = (HOC6H4)CH[double bond, length as m-dash]NNHCO(C5H3N)CO2H] containing unusually double-layered capsule-like water hexamers was obtained and characterized by elemental analysis, FT-IR, TGA, ESI-MS and X-ray crystallography.


Metallamacrocycles of the polygon type have received much attention in the past decade owing to their aesthetically fascinating structures, as well as their wide applications as molecular magnets, sensor materials, catalysts and host systems for various guest recognition.1–4 These discrete species that contain metal ions as ring constituents are mostly constructed from coordination-driven self-assembly with well-designed bridging ligands, because the unique shapes and functionalities of these polygonal entities are hardly accessible by other common synthetic means. Among the reported polygonal metallamacrocyclic structures, two distinct structural types have been developed based upon two different self-assembly strategies.4a In the first type, all of the ligands and the metal centers lie on the same plane, and both the metal coordination geometry and the predetermined angle between the binding units located at both ends of the organic ligand can be used to direct the metallamacrocyclic structure (ESI, Fig. S1). Metallamacrocylic compounds with this structural type have been nicely designed and successfully prepared by Fujita, Stang and many others.1,2 In the other type, the organic ligands and metal centers are not co-planar, and bridging ditopic ligands coordinate naked metal ions in an intertwined situation. In this structural type, the organic ligands often possess interwoven (helical) or parallel (grid-type), two different, arrangements depending whether on the coordinating sites of the ditopic ligands adopt a cis or trans orientation.3, 4 Both above mentioned metallamacrocycles are usually prepared with relatively rigid ditopic bridging ligands bearing nitrogen donor atoms, while the examples with heteroditopic chelate ligands are scarcely reported. These hybrid metal–organic analogues of organic macrocycles are being vigorously pursued and recent successes include some very interesting cone-shaped trigonal host systems akin to cyclotriveratrylene (CTV).5,6 Dolphin et al. have prepared two Zn(II)- and Co(II)-based metal–organic cyclotriveratrylenes (MOCTVs),5a Lippert et al. have also prepared several Pt(II)- and Pd(II)-based MOCTVs, [[enM(bpz-N4,N4′)]3]6+ (M = Pt, Pd; en = ethylenediamine; bpz = bipyrazine), which can host many different anions.5b And particularly, Keller and co-workers have reported a double bowl-shaped Cu(I)–phenanthroline MOCTV that can even contain two large Keggin anions, PM12O403−, as non-coordinated units in the two cavities.6

We have recently become interested in easy-to-prepare hydrazone ligands because some of them, such as pyridoxal isonicotinoyl hydrazone, have been identified as effective chelators for therapy of iron overload, both in vivo and in vitro,7 and also because some of them have been successfully employed in the self-assembly of several intriguing architectures, such as square,4d inorganic [2]catenane,8a metallacrown,8bnanocage,8c and [n×n] grids8detc. To continue our research work in molecular polygons and polyhedra,9 a rigid bishydrazone ligand H4L1 was designed and prepared in high yield from a Schiff-base reaction of salicylaldehyde and pyridine-2,5-dicarbohydrazide in ethanol (see ESI), and we have unexpectedly discovered that a trigonal metallamacrocyclic compound was obtained through an in situhydrolysis reaction of the bishydrazone H4L1 in the presence of Fe(ClO4)3·6H2O in a methanolic solution (Scheme 1).



          Self-assembly of the trigonal metallamacrocycle through an in situ ligand hydrolysis reaction.
Scheme 1 Self-assembly of the trigonal metallamacrocycle through an in situ ligand hydrolysis reaction.

When treatment of H4L1 and Fe(ClO4)3·6H2O in a 1 : 1 molar ratio in methanol, a black solution was obtained and, after several days, gave black block crystals of compound 1, whose structure was confirmed to be a [Fe3L3(H2O)3]·9H2O metallomacrocycle by X-ray crystallography. Under very similar reaction conditions, if the above experiment is carried out with an excess of Fe(ClO4)3·6H2O or an excess of H4L1, only the black amorphous products can be obtained, indicating that the formation of the monohydrolysis product is extremely sensitive to the stoichiometry between the two starting materials in this reaction system, although further studies are needed to better understand this phenomenon. The in situ generated carboxylate ligands via the hydrolysis of heterocycle,10aα,β-diketone,10bcyano,10c,10dester,10e and C[double bond, length as m-dash]C group10f under hydro(solvo)thermal conditions have been widely reported, nevertheless the heteroditopic bridging ligands containing monocarboxylate and other function groups were rarely obtained through in situhydrolysis of hydrazone compounds under mild reaction conditions.11 Such a C–N hydrolytic reaction of vulnerable hydrazone ligands can be promoted in the presence of metal centers (Lewis acid) and revealed from structural studies on the reaction products. Therefore, elemental analysis, FT-IR, TGA and ESI-MS spectra have been carried out to further confirm the formation of compound 1. The electrospray ionization mass (ESI-MS) spectrum of compound 1 in DMF–MeOH solution exhibits three main peaks appearing at m/z 357.2, 535.5 and 1069.3, with the isotopic distribution patterns being separated by ca. 1 Dalton (ESI, Fig. S3). These signals can be reasonably assigned to [Fe3L3(H2O)3 + 3H]3+, [Fe3L3(H2O)3 + 2H]2+ and [Fe3L3(H2O)3 + H]+, respectively, indicating that the trigonal metallamacrocycle is even stable in solution.

Single-crystal X-ray analysis reveals that compound 1 crystallizes in the P[3 with combining macron] space group and consists of a three-nuclear eighteen-membered trigonal metallamacrocyclic structure. As shown in Fig. 1, each Fe(III) ion is hexa-coordinated by two O atoms and one N atom of the tridentate chelate O2N donor group from one L3− ligand, by one pyridyl N atom and one carboxylate O atom from another L3− ligand and by one coordinated water molecule. The sum of equatorial angles is about 364.97°, and the corresponding axial angle is 163.66(10)° (O5–Fe1–N3A), suggesting a marked distortion from an ideal octahedral geometry. The C–O (C8–O2, 1.295(4) Å) and C–N (C8-N2, 1.307(4) Å) bond distances in compound 1 indicate that the C(O)NH amide group exists as deprotonated enol form, whereas the two C–O bond distances of the carboxylate group (C14–O3 1.234(4) Å and C14–O4 1.289(4) Å) suggest it functions in a monodentate manner. In addition of the deprotonated phenolate group, the ligand L3− in compound 1 totally can be considered as a triply negative pentadentate bridging ligand.


The molecular structure of compound 1 (all hydrogen atoms and solvent water molecules are omitted for clarity). Selected bond lengths [Å] and bond angles [°]: Fe1–O1 1.910(2), Fe1–O2 1.997(2), Fe1–N1 2.088(3), Fe1–O5 2.018(2), Fe1–O4A 1.978(2), Fe1–N3A 2.196(3); N1–Fe1–O4A 168.20(10), O1–Fe1–O5 90.34(10), O1–Fe1–N3A 96.79(10), O2–Fe1–O5 88.31(10), O2–Fe1–N3A 89.53(10). Symmetry codes: A −x + y + 1, −x + 1, z; B 2 −y + 1, x−y, z.
Fig. 1 The molecular structure of compound 1 (all hydrogen atoms and solvent water molecules are omitted for clarity). Selected bond lengths [Å] and bond angles [°]: Fe1–O1 1.910(2), Fe1–O2 1.997(2), Fe1–N1 2.088(3), Fe1–O5 2.018(2), Fe1–O4A 1.978(2), Fe1–N3A 2.196(3); N1–Fe1–O4A 168.20(10), O1–Fe1–O5 90.34(10), O1–Fe1–N3A 96.79(10), O2–Fe1–O5 88.31(10), O2–Fe1–N3A 89.53(10). Symmetry codes: A −x + y + 1, −x + 1, z; B 2 −y + 1, xy, z.

As shown in Fig. 2, three Fe(III) ions are linked by three heteroditopic trans-L3− to form an interwoven [Fe3L3(H2O)3] trigonal metallamacrocycle. In compound 1, each L3− ligand occurs obvious distortion by virtue of the rotation of the Cpy–Camide bond to adjust itself to accommodate the Fe(III) octahedral coordination, and the corresponding dihedral angle between pyridyl ring and phenyl ring is about 29.36°. One of the most striking features of compound 1 is that the three sloping pyridyl rings form a cone-shaped structure with three uncoordinated carboxylates (O3, O3A, O3B) in the upper rim and three uncoordinated amide nitrogen atoms (N2, N2A, N2B) in the lower rim. The dimensions of the cavity in compound 1 (upper rim 10.612 Å (O3⋯O3A), lower rim 5.183 Å (N2⋯N2A) and depth 4.507 Å) are even larger than that observed in the cyclotricatechylene12a (e.g., upper rim 9.490 Å, lower rim 3.092 Å and depth 3.168 Å) and calix[4]arene12b (e.g., upper rim 2.688 Å, lower rim 5.977 Å and depth 3.429 Å), indicating it can host many guest molecules and be potentially used as a metallamacrocyclic host in the guest inclusion and recognition. Interestingly, the three phenyl groups of three L3− ligands are located at the same side of the lower rim of the cone with a windmill-like arrangement to form a more hydrophobic environment, so the metallamacrocyle in compound 1 can be considered as an interestingly amphiphilic host molecule containing both hydrophilic carboxylate O atoms and hydrophobic phenyl groups.


Trigonal metallamacrocycle of compound 1 is formed by three Fe(iii) ions and three trans-L3− (coloured separately) with an interwoven fashion: (a) side view with stick model (except three carboxylate O atoms in the upper rim and three amide N in the lower rim) and (b) top view with space filling model. Symmetry codes: A −x + y + 1, −x + 1, z; B 2 −y + 1, x−y, z.
Fig. 2 Trigonal metallamacrocycle of compound 1 is formed by three Fe(III) ions and three trans-L3− (coloured separately) with an interwoven fashion: (a) side view with stick model (except three carboxylate O atoms in the upper rim and three amide N in the lower rim) and (b) top view with space filling model. Symmetry codes: A −x + y + 1, −x + 1, z; B 2 −y + 1, xy, z.

These trigonal metallamacrocyclic molecules are further organized in a typical graphite-like hexagonal packing in the ab plane. As shown in Fig. 3a, each six neighbouring trigonal metallamacrocyclic molecules are linked through six intermolecular face-to-face π⋯π stacking interactions between adjacent pyridyl rings to form a hexagonal structure. The other of the most striking feature of compound 1 is that there is a double-layered capsule-like hexameric water cluster in the center of each hexagon. As shown in Fig. 3b, three lattice water molecules (O6, O6A, O6B) and three coordinated water molecules (O5, O5A, O5B) in compound 1 form an unusual conic water hexamer. To our knowledge, the conic water hexamer has not been theoretically predicted and experimentally found heretofore, although the five low-energy structures (prism, cage, book, boat, cyclic) of water hexamer have been widely investigated.13 Interestingly, each two adjacent water hexamers are further packed into a double-layered capsule-like structure (Fig. 3c). The O⋯O distances in each water hexamer range from 2.609–2.872 Å, whereas the closest O⋯O contact between the two water hexamer in the “capsule” is about 3.507 Å, indicating not any significant H-bonding interactions between them. Obviously, the O–H⋯O H-bonding interactions among the lattice water molecules, the coordinated water molecules and the uncoordinated carboxylate O atoms also play very important roles in the stabilization of hexagonal packing structure in ab plane (Table S2). These graphite-like layers are further packed along c axis to form a 3D framework, and the average distance between each two neighbouring layers is about 12.054 Å (Fig. S4). The triangular tubes are established along the c axis through intermolecular C–H⋯O H-bonding interactions between uncoordinated carboxylate O atoms and phenyl C atoms.


(a) Packing diagram of compound 1 in the ab plane, the doubled-layered hexameric water clusters are shown with space-filling model; (b) view of conic water hexamer; (c) view of the double-layered capsule-like water hexamers and (d) schematic graphite-like packing of compound 1 in the ab plane. Symmetry codes: A −x + y + 1, −x + 1, z; B 2 −y + 1, x−y, z.
Fig. 3 (a) Packing diagram of compound 1 in the ab plane, the doubled-layered hexameric water clusters are shown with space-filling model; (b) view of conic water hexamer; (c) view of the double-layered capsule-like water hexamers and (d) schematic graphite-like packing of compound 1 in the ab plane. Symmetry codes: A −x + y + 1, −x + 1, z; B 2 −y + 1, xy, z.

Thermogravimetric analysis (TGA) was carried out for polycrystalline samples of compound 1 in the temperature range 30–1000 °C under N2 atmosphere (Fig. S5). The first weight loss from 30–100 °C (13.39%) corresponds to 9 lattice water molecules (13.21% calcd), and the second weight loss from 100–190 °C (4.83%) corresponds to 3 coordinated water molecules (4.40% calcd). No further weight loss was observed until it reached 300 °C, at which the decomposition of Fe3L3 framework occurred.

Temperature-dependent magnetic susceptibility measurements for compound 1 were performed on a polycrystalline sample in the temperature range 2–300 K under a 5 KOe external field (Fig. 4). The value of χmT at room temperature (4.74 cm3 mol−1 K) is slightly higher than the expected value of high-spin Fe(III) compound (4.37 cm3 mol−1 K).14 Upon cooling the temperature, the χmT value first decreases slightly until reaching 4.55 cm3 mol−1 K at 28 K, and then decreases rapidly below this temperature, until reaching 2.19 cm3 mol−1 K at 2 K. The continuous decrease of the χmT value suggests antiferromagnetic interaction between the Fe(III) ions. Furthermore, the Curie–Weiss law fitting gives a negative Weiss constant, θ, of −3.2 K, indicating the presence of antiferromagnetic interaction between the Fe(III) ions too.


Variable-temperature magnetic studies for compound 1.
Fig. 4 Variable-temperature magnetic studies for compound 1.

The spin system for compound 1 can be considered as an equilateral triangle, and the spin Hamiltonian is given as H = −2J[S1S2 + S1S3 + S2S3] based on the HDVV model,15 where J is the exchange integral between the adjacent metal centers. The best fitting parameters for 1 are g = 2.01, J = −0.11 cm−1, and the agreement factor R = ∑(χobs−χcald)2/χ2obs] = 6.1 × 10−3. The small value of J (−0.11 cm−1) in compound 1, is indicative of rather weak antiferromagnetic coupling interactions between neighbouring S = 5/2 spins, and full agreement with the large Fe–Fe distance (8.44 Å) imposed by L3− as revealed by structural analysis.

In summary, we have demonstrated the self-assembly and structural characterization of an interwoven Fe3L3 trigonal metallamacrocycle with in situ hydrolyzed heteroditopic ligands. The trigonal metallamacrocycle 1, with a large cavity and amphiphilic character, offers potential possibilities for its further application as an interesting macrocyclic host for some specific guest inclusion and recognition. Besides, an unusually double-layered capsule-like conic hexameric water cluster that is firstly found in solid crystal structure is also reported.

Acknowledgements

This work was supported by the 973 program (2006CB932900), the National Nature Science Foundation of China, and the Nature Science Foundation of Fujian Province.

Notes and References

  1. (a) S. Leininger, B. Olenyuk and P. J. Stang, Chem. Rev., 2000, 100, 853 CrossRef CAS; (b) P. J. Stang and B. Olenyuk, Acc. Chem. Res., 1997, 30, 502 CrossRef CAS; (c) M. Fujita, Chem. Soc. Rev., 1998, 27, 417 RSC; (d) M. Fujita, K. Umemoto, M. Yoshizawa, N. Fujita, T. Kusukawa and K. Biradha, Chem. Commun., 2001, 509 RSC.
  2. (a) H. Jiang and W. B. Lin, J. Am. Chem. Soc., 2004, 126, 7426 CrossRef CAS; (b) X.-C. Huang, J.-P. Zhang and X.-M. Chen, J. Am. Chem. Soc., 2004, 126, 13218 CrossRef CAS; (c) G. R. Newkome, T. J. Cho, C. N. Moorefield, R. Cush, P. S. Russo, L. A. Godinez, M. J. Saunders and P. Mohapatra, Chem.–Eur. J., 2002, 8, 2946 CrossRef CAS; (d) F. M. Romero, R. Ziessel, A. Dupont-Gervais and A. van Dorsselaer, Chem. Commun., 1996, 551 RSC.
  3. (a) G. F. Swiegers and T. J. Malefetse, Chem. Rev., 2000, 100, 3483 CrossRef CAS; (b) M. Ruben, J. Rojo, F. J. Romero-Salguero, L. H. Uppadine and J. M. Lehn, Angew. Chem., Int. Ed., 2004, 43, 3644 CrossRef CAS.
  4. (a) E. Coronado, J. R. Galan-Mascaros, P. Gaviña, C. Martí-Gastaldo, F. M. Romero and S. Tatay, Inorg. Chem., 2008, 47, 5197 CrossRef CAS; (b) B. Wisser, A.-C. Chamayou, R. Miller, W. Scherer and C. Janiak, CrystEngComm., 2008, 10, 461 RSC; (c) T. Bark, A. von Zelewsky, D. Rappoport, M. Nenburger, S. Schaffner, J. Lacour and J. Jodry, Chem.–Eur. J., 2004, 10, 4839 CrossRef CAS; (d) Z. He, C. He, Z.-M. Wang, E.-N. Gao, Y. Liu and C.-H. Yan, Dalton Trans., 2004, 502 RSC.
  5. (a) A. Thompson, S. J. Rettig and D. Dolphin, Chem. Commun., 1999, 631 RSC; (b) R.-D. Schnebeck, E. Freisinger, F. Glahe and B. Lippert, J. Am. Chem. Soc., 2000, 122, 1381 CrossRef CAS.
  6. J. M. Knaust, C. Inman and S. W. Keller, Chem. Commun., 2004, 492 RSC.
  7. J. L. Buss and P. Ponka, Biochim. Biophys. Acta, 2003, 1619, 177 CrossRef CAS.
  8. (a) T. S. M. Abedin, L. K. Thompson and D. O. Miller, Chem. Commun., 2005, 5512 RSC; (b) Y. Bai, D.-B. Dang, C.-Y Duan, Y. Song and Q.-J. Meng, Inorg. Chem., 2005, 44, 597; (c) C He, Z. H. Lin, Z He, C. H. Duan, C. H. Xu, Z. M. Wang and C. H. Yan, Angew. Chem., Int. Ed., 2008, 47, 877 CrossRef CAS; (d) L. N. Dawe, T. S. M. Abedin and L. K. Thompson, Dalton Trans., 2008, 1661 RSC.
  9. M. C. Hong, Cryst. Growth. Des., 2007, 7, 10 CrossRef CAS.
  10. (a) Y.-T. Wang, H.-H. Fan, H.-Z. Wang and X.-M. Chen, Inorg. Chem., 2005, 44, 4148 CrossRef CAS; (b) X.-M. Zhang, H.-S. Wu and X.-M. Chen, Eur. J. Inorg. Chem., 2003, 2959 CrossRef CAS; (c) M.-L. Tong, L.-J. Li, K. Mochizuki, H.-C. Chang, X.-M. Chen, Y. Li and S. Kitagawa, Chem. Commun., 2003, 428 RSC; (d) R.-G. Xiong, J.-L. Zuo, X.-Z. You, B. F. Albrahams, Z.-P. Bai, C.-M. Che and H.-K. Fun, Chem. Commun., 2000, 2061–2062 RSC; (e) O. R. Evans and W. B. Lin, Acc. Chem. Res., 2002, 35, 511 CrossRef CAS; (f) R.-H. Wang, M.-C. Hong, J.-H. Luo, R. Cao and J.-B. Weng, Chem. Commun., 2003, 1018 RSC.
  11. V. A. Milway, L. K. Thompson and D. O. Miller, Chem. Commun., 2004, 1790 RSC.
  12. (a) C. J. Sumby and M. J. Hardie, Acta Crystallogr., Sect. E: Struct. Rep. Online, 2007, 63, O1537 CrossRef; (b) H. Akdas, L. Bringer, E. Graf, M. W. Hosseini, G. Mislin, J. Pansanel, A. D. J. Cian and J. Fischer, Tetrahedron Lett., 1998, 39, 2311 CrossRef CAS.
  13. (a) K. Liu, M. G. Brown, C. Carter, R. J. Saykally, J. K. Gregory and D. C. Clary, Nature, 1996, 381, 501 CrossRef CAS; (b) J. Narasimha Moorthy, R. Natarajan and P. Venugopalan, Angew. Chem., Int. Ed., 2002, 41, 3417 CrossRef CAS; (c) B. K. Saha and A. Nangia, Chem. Commun., 2006, 1825 RSC; (d) M. Losada and S. Leutwyler, J. Chem. Phys., 2002, 117, 2003 CrossRef CAS.
  14. D. Armentano, G. D. Munno, J. Faus, F. Lloret and M. Julve, Inorg. Chem., 2001, 40, 655 CrossRef CAS.
  15. H. Adams, D. E. Fenton, L. R. Cummings, P. E. Mchugh, M. Ohba, H. Okawa, H. Sakiyama and T. Shiga, Inorg. Chim. Acta, 2004, 357, 3648 CrossRef CAS.

Footnotes

Electronic supplementary information (ESI) available: Experimental section, ESI-MS, TGA and other complementary tables and drawings. CCDC reference number 705039. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/b817929h
Crystal data for 1: C42H44Fe3N9O24, M = 1226.40, trigonal, a = 17.5191(15), c = 12.0541(17) Å, V = 3204.0(6) Å3, T = 293(2) K, space groupP[3 with combining macron], Z = 2, 19[thin space (1/6-em)]549 reflections measured, 4850 unique (Rint = 0.1012), which were used in all calculations. The final R1 = 0.0572, wR2 = 0.1359 (I > 2σ(I)) and GOF = 1.029.

This journal is © The Royal Society of Chemistry 2009
Click here to see how this site uses Cookies. View our privacy policy here.