Nanoporous carbohydrate frameworks and the sequestration and detection of molecules using the same

Abstract

Disclosed are cyclodextrin-based metal organic frameworks comprising a metal cation and cyclodextrin or a cyclodextrin derivative. These metal organic frameworks are permanently porous and capable of molecule storage.

Claims

1. A method of making a cyclodextrin-based metal organic framework comprising a metal cation and a -cyclodextrin or a -cyclodextrin derivative, the method comprising: a) dissolving the -cyclodextrin or the -cyclodextrin derivative and an alkali metal salt in a first solvent to provide a solution thereof; and b) allowing vapor diffusion, into the solution, of a second solvent in which either of the -cyclodextrin or the -cyclodextrin, or the alkali metal salt has poor solubility to provide a cyclo-dextrin-based metal organic framework having a crystal structure with an I432 space group, where the -cyclodextrin or -cyclodextrin derivative is a compound of a formula ##STR00002## wherein n=3; and R is selected from the group consisting of OH; NRR; S(O).sub.2R; S(O)OR; S(O)R; C(O)OR; CN; C(O)R; SR, NN.sup.+N.sup.; NO.sub.2, OSO.sup.2R; C(O)OR, O(S)SR, P(O)(OR).sub.2; OP(O)(OR).sub.2; P(O)(OR)R; NRR; NRP(OR)(OR); OC(O)NRR; C.sub.1-C.sub.18 alkyl optionally substituted with one, two, three, four or five R.sup.1 groups; C.sub.2-C.sub.18 alkenyl optionally substituted with one, two, three, four or five R.sup.1 groups; C.sub.2-C.sub.18 alkynyl optionally substituted with one, two, three, four or five R.sup.1 groups; C.sub.1-C.sub.18 alkoxy optionally substituted with one, two, three, four or five R.sup.1 groups; aryl optionally substituted with one, two, three, four or five R.sup.2 groups; heteroaryl optionally substituted with one, two, three, four or five groups independently selected from R.sup.2 groups; and cycloalkyl optionally substituted with one, two, three, four or five groups independently selected from R.sup.2 groups; wherein each R.sup.1 group is independently selected from hydroxyl, halo, lower alkoxy, NRR, S(O).sub.2R, S(O)OR, S(O)R, C(O)OR, CN, C(O)R, NN.sup.+N.sup., SR, NO.sub.2, OSO.sup.2R, C(O)OR, O(S)SR, P(O)(OR).sub.2, OP(O)(OR).sub.2; P(O)(OR)R, NRR, NRP(OR)(OR), OC(O)NRR, aryl optionally substituted with one, two, three, four or five R groups; heteroaryl optionally substituted with one, two, three, four or five groups independently selected from R groups; and cycloalkyl optionally substituted with one, two, three, four or five groups independently selected from R groups; each R.sup.2 group is independently selected from lower alkyl, lower alkyenyl, lower alkynyl, hydroxyl, halo, lower alkoxy, NRR, S(O).sub.2R, S(O)OR, S(O)R, C(O)OR, CN, C(O)R, NN.sup.+N.sup., SR, NO.sub.2, OSO.sup.2R, C(O)OR, O(S)SR, P(O)(OR).sub.2, OP(O)(OR).sub.2; P(O)(OR)R; NRR; NRP(OR)(OR); OC(O)NRR, aryl optionally substituted with one, two, three, four or five R groups; heteroaryl optionally substituted with one, two, three, four or five groups independently selected from R groups; and cycloalkyl optionally substituted with one, two, three, four or five groups independently selected from R groups; and each R and R are independently selected from the group consisting of H, lower alkyl and aryl.

2. The method according to claim 1 wherein the metal cation is selected from the group consisting of Li.sup.+, K.sup.+, Rb.sup.+, Cs.sup.+, Na.sup.+, Mg.sup.2+, Cd.sup.2+, Sn.sup.2+, Ag.sup.+, Yb.sup.+, Ba.sup.2+, Sr.sup.2+, Ca.sup.2+, Pb.sup.2+, and La.sup.3+.

3. The method according to claim 2 wherein the metal cation is selected from the group consisting of K.sup.+, Rb.sup.+, Cs.sup.+ and Na.sup.+.

4. The method according to claim 1 wherein the counterion is organic or inorganic.

5. The method according to claim 4 wherein the counterion is inorganic and selected from the group consisting of OH and CO.sub.3.sup.2.

6. The method according to claim 4 wherein the counterion is organic and selected from the group consisting of benzoate anion and azobenzene-4,4-dicarboxylate dianion.

7. The method according to claim 1 forming a tetragonal or tetrahedral single crystal.

8. The method according to claim 7 wherein the tetragonal or tetrahedral crystal has a unit cell edge of approximately 31 .

9. The method according to claim 7 wherein the tetragonal or tetrahedral crystal consists of six -cylodextrin rings.

10. The method according to claim 9 wherein the six -cylodextrin rings form a central pore.

11. The method according to claim 10 wherein the central pore has a diameter of approximately 1.7 nm.

12. The method according to claim 11 further comprising a plurality of smaller triangular pores with diameters of approximately 0.4 nm.

13. The method according to claim 12 wherein counterions fill the pores and channels of the metal organic framework and are disordered throughout the crystal lattice.

14. The method according to claim 13 wherein solvent molecules fill the pores and channels of the metal organic framework and are disordered throughout the crystal lattice.

15. The method according to claim 12 wherein particles fill the pores and channels of the metal organic framework.

16. The method according to claim 15 wherein the particles are selected from the group consisting of quantum dots and nanoparticles.

17. The method according to claim 1, wherein the alkali metal salt is selected from KOH, RbOH and CsOH.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 depicts a space-filling representation of the orange single crystals containing azobenzene dicarboxylic acid, CD and K.sup.+ ions (CD-MOF-1). Views of (a) one nanocontainer and (b) a cluster of five nanocontainers taken from the packing diagram in the solid state.

(2) FIG. 2 depicts (a) an individual cube-shaped unit consisting of six CD rings (CD).sub.6, one occupying each face of the cube; (b) body-centered cubic crystal packing diagram; (c) one single CD ring showing the K.sup.+ ions bound to the primary hydroxyl group and ring oxygen atom at residues 1, 3, 5, and 7 and to the C2 and C3 hydroxyl groups at residues 2, 4, 6, and 8; and (d) cube-shaped unit with front and back -CDs removed for clarity. Crystal data of CD-MOF-1: cubic, space group I432, a=b=c=31.006(8) , V=29807(14) .sup.3, Z=12.

(3) FIG. 3 is a graphical representation illustrating the different voids in the crystal structure of CD-MOF-1.

(4) FIG. 4 shows the different nanotubes present in the crystal structure of the polymorph of CD-MOF-3.

(5) FIG. 5 is a classification of the different porous domains in metal-organic frameworks.

(6) FIG. 6 is a schematic encapsulation of a gold nanoparticle in a single nanocontainer of CD-MOF-1.

(7) FIG. 7 depicts the deep red cubic crystals of CD-MOF-1 with encapsulated Rhodamine B guest molecules.

(8) FIG. 8 is an illustration of possible guest molecules that could be incorporated into the major void spaces in CD-MOF-1.

(9) FIG. 9 is a schematic diagram showing the photolinkage of stilbene-substituted octahedral moieties, for example, hexafunctionalized fullerenes, within the CD-MOF framework.

(10) FIG. 10 depicts octahedral fullerene derivatives containing stilbene groups that can be photocrosslinked in the solid state encapsulated by the nanocontainers in CD-MOF-1.

(11) FIG. 11 is a .sup.1H NMR Spectrum of CD-MOF-1 identifying the presence of counterions and thereby confirming CD is not deprotonated.

(12) FIG. 12 shows the robustness of CD-MOF-1 and CD-MOF-2 from PXRD patterns (simulated and experimental) upon evacuation of solvent.

(13) FIG. 13 N.sub.2 thermal isotherm of CD-MOF-1 and CD-MOF-2 to determine BET surface area, Langmuir surface area and pore volume.

(14) FIG. 14 .sup.1H NMR Spectrum of redissolved CD-MOF-1 crystals grown in the presence of Rhodamine B dye in D.sub.2O.

(15) FIG. 15 CP/MAS NMR Spectrum of CD-MOF-2.

(16) FIG. 16 Thermogravimetric analysis traces of CD-MOF-1 and CD-MOF-2; 16a-b confirm the retention of solvent in both CD-MOF-1 and CD-MOF-2, respectively; 16c-d show the stability of activated CD-MOF-1 and CD-MOF-2, respectively, to heating.

(17) FIG. 17 .sup.1H NMR spectrum (500 MHz) in D.sub.2O of redissolved CD-MOF crystals prepared from potassium benzoate and CD, referenced to the H.sub.2O peak (&=4.79).

(18) FIG. 18 CP/MAS NMR spectrum of ground samples of CD-MOF-2.

(19) FIG. 19 Powder x-ray diffraction of a ground sample of CD-MOF-2.

DETAILED DESCRIPTION OF THE INVENTION

(20) Preparation and Characterization of CD-MOFs

(21) Accordingly, the present invention includes cyclodextrin-based metal organic framework (CD-MOFs) comprising at least one metal cation and cyclodextrin or a cyclodextrin derivative. Suitable metal cations included Group I metals, Group II metals and transition metals, preferably Group I metals more preferably Na.sup.+, K.sup.+, Rb.sup.+ or Cs.sup.+. Suitable cyclodextrins are, for example, -, - and -cyclodextrins. Suitable cyclodextrin derivatives are those depicted below for Formula I.

(22) Illustrating certain non-limiting aspects and embodiments of this invention, CD-MOFs are prepared. Cyclodextrins (Nepogodiev, S. A. et al., Chem. Rev. 1998, 98, 1959-1976) and their synthetic analogs (Nepogodiev, S. A. et al., Chem. Rev. 1998, 98, 1919-1958) have been used in a variety of procedures (Jones, J. K. N. et al., Can. J. Chem. 1969, 47, 3213-3215; Khashab, N. M. et al., Eur. J. Org. Chem. 2009, 1669-1673). Many of such cyclodextrins have been predominately alpha or beta. Previously described solid-state structures of complexed (Lindner, K. et al., Biochem. Biophys. Res. Commun. 1980, 92, 933-938; Kamitori, S. et al., Bull. Chem. Soc. Jpn. 1988, 61, 3825-3830) and uncomplexed hydrated (Harata, K., Chem. Lett. 1984, 641-644; Harata, K., Bull. Chem. Soc. Jpn. 1987, 60, 2763-2767) CD reveal the commoner garden cage-type and channel-type packing that has prevailed in the solid-state superstructures in the case of -CD and -CD for decades of close investigation. It does appear, however, that the four-fold symmetry of CD with its C.sub.8 point group that has led to the formation of nanocapsules (MacGillivray, L. R. et al., Nature 1997, 389, 469-472) by calixarenes, thanks to self-assembly during crystallization driven largely by hydrogen bonding, has delivered a somewhat similar superstructure in the case of -CD, thanks to electrostatic interactions provided by appropriate metal ions. In his 1989 review on Complexes of Metal Cations with Carbohydrates in Solution in Adventures in Carbohydrate Chemistry and Biochemistry, Stephen Angyal began by noting that complex formation between salts and carbohydrates is not a new subject, and states that crystalline adducts of sugars with inorganic salts have been studied since 1825 (Angyal, S. J., Adv. Carb. Chem. Biochem. 1989, 47, 1-43). In his review, Angyal points out that, substrate variation taken into consideration, metal cations can be arrayed according to their increasing tendency to form complexes with carbohydrates roughly as follows: Li.sup.+, K.sup.+, Rb.sup.+, Na.sup.+, Mg.sup.2+, Cd.sup.2+, Sn.sup.2+, Ag.sup.+, Yb.sup.+, Ba.sup.2+, Sr.sup.2+, Ca.sup.2+, Pb.sup.2+, and La.sup.3+. In more recent times, Geisselmann et al. (Geisselmann, A. et al., Angew. Chem. Int. Ed. 2003, 44, 924-927) have investigated how carbohydrate-metal interactions are shaped by supramolecular assembly, with references to - and -CDs in particular (Klfers, P. et al., Chem. Eur. J. 1997, 3, 601-608; Benner, K. et al., Angew. Chem. Int. Ed. 2006, 45, 5818-5822; Fuchs, R. et al., Angew. Chem. Int. Ed. Engl. 1993, 32, 852-854). They point out that both iron and manganese, at least in their higher oxidation states, are sufficiently Lewis acidic to stabilize multiply deprotonated carbohydrate ligands, forming complexes of very high stabilities and pointing out that deprotonated carbohydrate ligands are present in the active centers of carbohydrate-directed metal enzymes such as xylose isomerase. This consideration is one that has to be at least contemplated in the case of the new CD-MOFs.

(23) Generally, CD-MOFs are prepared by dissolution of both the cyclodextrin component and the metal salt component in any solvent in which both have solubility. Isolation of CD-MOFs is done by addition of a solvent in which either of the components has poor solubility, including, but not limited to, C.sub.1-C.sub.18 alcohols, acetone, tetrahydrofuran, dioxane, acetonitrile, as well as other common organic solvents miscible with water, or any mixtures thereof. As such, in a specific non-limiting embodiment of the invention, methanol is allowed to vapor diffuse into an aqueous solution containing K.sub.2abdc and CD in a molar ratio of 2:1. Orange cubic single crystals (ca. 1 mm.sup.3) are obtained and are subjected to X-ray crystallography. The crystal structure (FIG. 1) with its I432 space group is unique by comparison with all other known fully-solved, solid-state structures incorporating CD in the literature. Moreover, the unit cell of ca. 30,000 .sup.3 is much larger than anything that has been observed previously incorporating CD, with one exception (Bonacchi, D. et al., Chem. Mater. 2004, 16, 2016-2020).

(24) Single crystals of -Fe.sub.2O.sub.3/CD are obtained by a procedure described by this reference. Ferrous chloride is added to a solution obtained by dissolving CD in DMF and the solution is stirred under argon for two hours. An ethanolic solution of NaOH is then added and the resulting solution is stirred and exposed to air. Cubic-shaped reddish-orange single crystals separated out after a few weeks from the filtered solution under ethanol diffusion. From the full diffraction set collected at 100 K, a cubic unit cell with a 30.217 edge is determined. The symmetry and systematic absences of the reciprocal lattice are consistent with I432, a space group that has never been observed for CD or any of its complexes.

(25) Since it is only possible to identify clearly CD units and K.sup.+ ions in the X-ray crystal structure depicted in FIG. 1, a range of much simpler potassium salts (other than K.sub.2abdc) is examined for their abilities to form crystals in aqueous methanol. Although colorless cubic crystals are obtained using both KOH and K.sub.2CO.sub.3, the crystallization process is very much slower for the carbonate than for the hydroxide. This observation has led to the proposition that the OH.sup. ions (and CO.sub.3.sup.2 ions) deprotonate each CD ring (twice) during the crystallization process. In the case of K.sub.2abdc, it is presumably the conjugate base that deprotonates the CD ring. However, .sup.1H NMR Spectroscopy has identified the presence of counterions throughout the crystal lattice, thereby confirming that CD is indeed not deprotonated (FIG. 11). These conditions involve mixing 1 equivalent of CD with 8 equivalents of KOH or K.sub.2CO.sub.3 in water, followed by slow diffusion of methanol into the solution during 2-7 days. The crystal structure obtained using KOH is presented in FIG. 2.

(26) The structure in FIG. 2, CD-MOF-1, is isostructural with the structure of FIG. 1, obtained when the source of the K.sup.+ ions is K.sub.2abdc. CD-MOF-1 is a body-centered cubic structure with each cube consisting of six CD rings, or (CD).sub.6, occupying the faces of the cube with the primary hydroxyl groups of the CD facing into the interior of the cube. The (CD).sub.6 is held together by K.sup.+ ions bonded to alternating -d-glucopyranosyl residues, i.e., residues 1, 3, 5, and 7. Each cubic unit is attached to another one at each secondary CD face through four K.sup.+ ions. The K.sup.+ ions are coordinated to the C2 and C3 hydroxyl groups at the 2, 4, 6, and 8 -d-glucopyranosyl residues. The structure is a highly porous one with a regular array of large voids (FIG. 3) created by the six CD cubic units.

(27) Perfect alignment of the CDs in the structure leads to infinite channels in the x, y, and z directions with diameters equating to the inner diameter of the CD rings, specifically 0.9 nm. The spherical pore, or cavity, contained within the six CD cubes (CD).sub.6 has a diameter of 1.7 nm. Smaller triangular pores with diameters of about 0.4 nm are present along the plane. Not surprisingly, this highly symmetrical arrangement can not be reproduced either with - or with -CDs, presumably since each of the eight -d-glucopyranosyl residues in CD is involved in binding to K.sup.+ ions. CD is known to crystallize frequently in unusual, high symmetry space groups, e.g., P42.sub.12. Counterions and solvent molecules fill the cavities and are disordered throughout the crystal lattice. The solvent is removed from the extended structure by evacuation at room temperature (20% w/w solvent by thermogravimetric analysis) leaving robust CD-MOFs as revealed by powder x-ray diffraction patterns (FIG. 12). The BET surface area for CD-MOF-1 is 1020 m.sup.2 g.sup.1 (adsorption isotherm of CO.sub.2 measured for N.sub.2 on CD-MOF-1; FIG. 13), while the Langmuir surface area is 1320 m.sup.2 g.sup.1 and the pore volume is 0.47 cm.sup.3 g.sup.1. In comparison, the surface areas of other MOFs are: a) Basolite A100=1100 to 1500 m.sup.2 g.sup.1; ZIF-95=1240 m.sup.2 g.sup.1; MOF-200=10,000 m.sup.2 g.sup.1; and porous polymers=800-1000 m.sup.2 g.sup.1.

(28) Crystal growing experiments with NaOH/Na.sub.2CO.sub.3, RbOH/Rb.sub.2CO.sub.3, and CsOH/Cs.sub.2CO.sub.3 yield colorless, cubic-shaped crystals, which are examined by X-ray crystallography. The rubidium structure (CD-MOF-2) is isostructural with the potassium one, and the outcome is similar for the sodium structure. In the case of cesium, it appears that at least two different morphologies of crystals exist. One is cubic (CD-MOF-3) and is isostructural with the potassium and rubidium structures. Another batch of crystals (polymorph of CD-MOF-3), however, are needlelike and it transpires that they also have an extended MOF-like structure but, on this occasion, the cavities are oriented in a series of parallel channels; in one case defined by nanotubes of CD rings linked by Cs.sup.+ ions, and in the other by the space left between any four CD channels (FIG. 4).

(29) The use of long organic struts (2 nm) incorporating 34- and 36-membered macrocyclic polyethers as recognition modules in the construction of several crystalline primitive cubic frameworks that behave in a manner beyond open reticulated geometries (BORGs) is performed (Li, Q. et al., Science 2009 Aug. 14 Issue). The first MOF in this BORG series, MOF-1001, is capable of docking the paraquat (methyl viologen) dication within the macrocycles in a stereoelectronically controlled fashion.

(30) The vast majority of porous MOFs prepared by the methods of the invention can be regarded as having two important architectural domains: (i) the pore aperture, which is responsible for the shape- and size-selective binding of incoming molecules, and (ii) the internal surface of the pores, onto which gases or small molecules can be compacted and distributed with simple interaction sites covering the struts and joints (FIG. 5). These two domains are called the sorting domain (Kitagawa, S. et al., Angew. Chem. Int. Ed. 2004, 43, 2334-2375) and the coverage domain (Wang, Z. et al., Chem. Soc. Rev. 2009, 38, 1315-1329). In the Li et al. article, it is shown how molecular recognition components, much used in supramolecular chemistry, can be integrated in a modular fashion into the struts of MOFs, thereby creating recognition sites into which incoming guests will dock in a highly specific manner with stereoelectronic control. This third architectural domain, the active domain, combines shape, size, and electronic elements in the recognition of incoming guests and brings order to otherwise highly disordered guests in conventional MOFs.

(31) In still another embodiment of the invention, chemical modifications of the primary hydroxyl groups on alternating glucopyranosyl residues are explored, since only one half of the primary hydroxyl groups on CD, and likewise only one half of the secondary hydroxyl groups on C2 and C3 of the glucopyranosyl residues, are involved in coordination to the metal cations on MOF formation. This goal has been achieved (Boger, J. et al., J. Am. Chem. Soc. 1979, 7630-7631) with -CD where selective tritylation of the CD torus has been demonstrated (Ling, C. C. et al., Carbohydrate Res. 1992, 287-291). As such, functionalizing -CD with post-assembly modification (fixing) of its CD-MOF superstructure is preferred. This is pursued in tandem with molecular modeling to establish the feasibility or otherwise of stabilizing the array of supramolecular nanocapsules by introducing covalent bonds between the CD components, both within and beyond single nanocapsules. CD-MOF formation does not have to be perturbed by introducing (presumably relatively small) active functional groups onto some or all of the free hydroxyl groups.

(32) Accordingly, in still another embodiment, the invention provides a CD-MOF comprising a CD portion and a metal salt portion; wherein the metal salt portion has the formula MN, wherein M is a Group I, Group II metal or transition metal, and N is an organic or inorganic ion; and the CD portion of CD-MOF is a compound of the Formula I:

(33) ##STR00001##
wherein n=0-10; and R is selected from the group consisting of OH; NRR; C.sub.1-C.sub.18 alkyl optionally substituted with one, two, three, four or five R.sup.1 groups; C.sub.2-C.sub.18 alkenyl optionally substituted with one, two, three, four or five R.sup.1 groups; C.sub.2-C.sub.18 alkynyl optionally substituted with one, two, three, four or five R.sup.1 groups; C.sub.1-C.sub.18 alkoxy optionally substituted with one, two, three, four or five R.sup.1 groups; S(O).sub.2R; S(O)OR; S(O)R; C(O) OR; CN; C(O) R; SR, NN.sup.+N.sup.; NO.sub.2, OSO.sup.2R; C(O)OR, O(S)SR, P(O)(OR).sub.2; OP(O)(OR).sub.2; P(O)(OR)R; NRR; NRP(OR)(OR); OC(O)NRR; aryl optionally substituted with one, two, three, four or five R.sup.2 groups; heteroaryl optionally substituted with one, two, three, four or five groups independently selected from R.sup.2 groups; and cycloalkyl optionally substituted with one, two, three, four or five groups independently selected from R.sup.2 groups; wherein each R.sup.1 group is independently selected from hydroxyl, halo, lower alkoxy, NRR, S(O).sub.2R, S(O)OR, S(O)R, C(O)OR, CN, C(O)R, NN.sup.+N.sup., SR, NO.sub.2, OSO.sup.2R.sup.1, C(O)OR, O(S)SR, P(O)(OR).sub.2, OP(O)(OR).sub.2; P(O)(OR)R, NRR, NRP(OR)(OR), OC(O)NRR, aryl optionally substituted with one, two, three, four or five R groups; heteroaryl optionally substituted with one, two, three, four or five groups independently selected from R groups; and cycloalkyl optionally substituted with one, two, three, four or five groups independently selected from R groups; wherein each R.sup.2 group is independently selected from lower alkyl, lower alkyenyl, lower alkynyl, hydroxyl, halo, lower alkoxy, NRR, S(O).sub.2R, S(O)OR, S(O)R, C(O)OR, CN, C(O)R, NN.sup.+N.sup., SR, NO.sub.2, OSO.sup.2R, C(O)OR, O(S)SR, P(O)(OR).sub.2, OP(O) (OR).sub.2; P(O)(OR)R; NRR; NRP(OR)(OR); OC(O)NRR, aryl optionally substituted with one, two, three, four or five R groups; heteroaryl optionally substituted with one, two, three, four or five groups independently selected from R groups; and cycloalkyl optionally substituted with one, two, three, four or five groups independently selected from R groups; and wherein each R and R are independently selected from the group consisting of H, lower alkyl and aryl.

(34) By lower alkyl in the present invention is meant a straight or branched chain alkyl radical having 1-6, and preferably from 1-3, carbon atoms. Examples include but are not limited to methyl, ethyl, propyl, isopropyl, n-butyl, sec-butyl, tert-butyl, pentyl, 2-pentyl, isopentyl, neopentyl, hexyl, 2-hexyl, 3-hexyl, and 3-methylpentyl. Each alkyl group may be optionally substituted with one, two or three substituents such as, for example, a halo, cycloalkyl, aryl, alkenyl or alkoxy group and the like.

(35) By lower alkenyl is meant a straight or branched hydrocarbon radical having from 2 to 6 atoms and one or two double bonds and includes, for example, ethenyl, propenyl, 1-but-3-enyl, 1-pent-3-enyl, 1-hex-5-enyl. The alkenyl group can also be optionally mono-, di-, or trisubstituted with, for example, halo, aryl, cycloalkyl or alkoxy and the like.

(36) By lower alkynyl is meant a straight or branched hydrocarbon radical having from 2 to 6 atoms and one or two triple bonds and includes, for example, propynyl, 1-but-3-ynyl and the like. The alkynyl group can also be optionally mono-, di-, or trisubstituted with, for example, halo, aryl, cycloalkyl or alkoxy and the like.

(37) By lower alkoxy is meant an O-lower alkyl group wherein lower alkyl is as defined above.

(38) By halo or halogen is meant a halogen radical of fluorine, chlorine, bromine or iodine.

(39) By aryl is meant an aromatic carbocylic radical having a single ring (e.g. phenyl), multiple rings (e.g. biphenyl) or multiple fused rings in which at least one is aromatic (e.g. 1,2,3,4-tetrahydronaphthyl).

(40) By heteroaryl is meant one or multiple fused aromatic ring systems of 5-, 6- or 7-membered rings containing at least one and up to four heteroatoms selected from nitrogen, oxygen or sulfur. Examples include but are not limited to furanyl, thienyl, pyridinyl, pyrimidinyl, benzimidazolyl and benzoxazolyl.

(41) By cycloalkyl is meant a carbocylic radical having a single ring (e.g. cyclohexyl), multiple rings (e.g. bicyclohexyl) or multiple fused rings (e.g.). In addition, the cycloalkyl group may have one or more double bonds.

(42) Suitable inorganic counterions are, for example, chloride, fluoride, hydroxide, sulfide, sulfinate, carbonate, chromanate, cynadie, and the like. Suitable organic counterions are, for example, benzoate, azobenzene-4,4-dicarboxylate, acetate, oxalate, and the like.

(43) In a further embodiment of the invention various other solvents can be used, such as, for example, dimethylformamide (DMF), dimethylsulfoxide (Me.sub.2SO), diethylformamide, and any combination thereof, including with water and various low molecular weight alcohols.

(44) In still another embodiment of the invention, the CD-MOFs incorporate molecules such as, for example, carbon dioxide, hydrogen, organophosphates, chemical warfare agents and small molecules, into the cavities and channels of the CD-MOFs, both during their self-assembly (crystallization) and after their formation, or both. In addition, particles, such as, for example, quantum dots and nanoparticles (FIG. 6) can be incorporated into the cavities and channels of CD-MOFs, both during their self-assembly (crystallization) and after their formation, or both. By incorporating these molecules or particles, the CD-MOFs provide for sequestration and/or detection of the same.

(45) In a specific non-limiting example, imprinting small molecules into millimeter-sized crystals of CD-MOFs is performed in much the same way as complex core-and-shell particles are assembled into open-lattice crystals (Wesson, P. J. et al., Adv. Mater. 2009, 21, 1911-1915). To date, no detection of the location of substrates like pyrene and rhodamine B (FIG. 7) in the crystals by X-ray diffraction methods is seen. These small molecules remain invisible to X-rays, yet dissolution of these highly colored CD-MOF crystals in D.sub.2O and recording .sup.1H NMR spectra reveals that approximately four (4) molecules of Rhodamine B are found in each CD-MOF-1 (CD).sub.6 cube (FIG. 14).

(46) FIG. 8 illustrates the matches that some larger molecules represent with respect to the nanochambers present in CD-MOF-1. This illustration can be used as a guide to what might be possible to incorporate (and what might not be possible) at the level of larger molecules during the crystallization of the CD-MOF in question. Some of the larger molecules, including quantum dots, nanoparticles and polyoxometalates, might act as templates for the formation of CD-MOF crystals with the molecular templates ultimately locked up inside the nanochambers of the CD-MOFs formed during the templation process.

(47) Thus, in an embodiment of the invention, it is necessary to make use of the octahedral shape of the nanochambers to trap therein during crystallization octahedral substrates with six arms. An octahedral substrate, such as the one illustrated in FIG. 9, constitutes a suit[6]ane. Assuming that the octahedral substrate's six arms are terminated by stilbene units and that these stilbene units from different neighboring substrates meet as - stacked pairs (Klotz, E. J. F. et al., J. Am. Chem. Soc. 2006, 128, 15374-15375; Agbaria, R. A. et al., J. Phys. Chem. B 1995, 24, 10056-10060) inside the two CDs that are oriented head-to-head in the centers of the channels linking the nanochambers, following wind to join up all the stacked stilbene dimers by photochemical dimerization is possible.

(48) This supramolecular control of reactivity in the solid-state has been developed elegantly by MacGillivray (MacGillivray, L. R. et al., Acc. Chem. Res. 2008, 41, 280-290; MacGillivray, L. R., J. Org. Chem. 2008, 73, 3311-3317) and Garcia-Garibay (Garcia-Garibay, M. A., Acc. Chem. Res. 2003, 36, 491-498) in recent times and has been applied to other supramolecular systems (Amirsakis, D. G. et al., Angew. Chem. Int. Ed. 2001, 40, 4256-4261; Amirsakis, D. G. et al., Angew. Chem. Int. Ed. 2003, 42, 1126-1132) with close to 100% efficiency for the [2+2]cycloadditions. Moreover, there are also examples of photocrosslinking of stilbenes included inside CD in both the solution (Herrmann, W. et al., Chem. Commun. 1997, 1709-1710) and solid (Rao, K. et al., J. Org. Chem. 1999, 64, 8098-8104) states. FIG. 10 illustrates the formation of a covalent organic framework (COF) based on fullerenes inside CD-MOF-1. In still another embodiment of the invention, the making of a MOF within a MOF, where the archetypal [Zn.sub.4O(CO.sub.2).sub.6] cluster is the secondary building unit of the new templated MOF, is performed (FIG. 9). Washing away the alkali metal ions of the CD-MOF generates novel frameworks rotaxanated by aligned CD dimers, whilst enzymatic cleavage of the -CDs generates the COF/MOF templated within the scaffold.

(49) Sequestration and Detection of CO.sub.2

(50) The CD-MOFs as disclosed herein have a high preference for carbon dioxide over other waste gasses such as methane. By pre-incorporation of a pH-indicator, the content of carbon dioxide can be colormetrically monitored. As depicted in Formula I above, a variety of functionalized CD-MOFs, e.g. N.sub.1-CDMOF, can be used to sequester and/or detect molecules such as, for example, CO.sub.2. N.sub.1-CDMOF is crystallographically isomorphic with CD-MOF-1.

(51) Adsorption isotherms of CO.sub.2 and methane are measured for N.sub.2 and CH.sub.4 on both CD-MOF-1 and N.sub.1-C DMOF. From the adsorption data it is clear that there is a substantial preference for carbon dioxide over methane. CP/MAS NMR Spectroscopy of CD-MOF-2 shows only signals for carbon atoms that are polarized by protons on the framework; gaseous CO.sub.2 is not visible using this process (FIG. 15). 2D CP/MAS shows correlation between .sup.13C-enriched, proposed R-CO.sub.3H peak and protons on the CD framework, indicating that the new peak is covalently attached. DP/MAS .sup.13C NMR Spectroscopy shows all carbon atoms in the framework, including the peaks for: free gaseous CO.sub.2, proposed RCO.sub.3H, and another new peak. Also, using a pH-indicator such as Methyl Red, the pH change is visually detected as CO.sub.2 forms carbonic acid at primary OH on the CD.

(52) The gas uptake of CD-MOFs is dependent on crystallinity. Grinding a sample of CD-MOF-1 for ten (10) minutes and pulverizing it into an amorphous powder subsequently reveals no gas uptake in adsorption measurements, and no CO.sub.2 uptake in CP/MAS NMR or colormetric method.

EXAMPLES

(53) All reagents are supplied by Sigma Aldrich and Fisher, while kosher, food-grade CD is obtained as a gift from Wacker. All chemicals and solvents are used without further purification. .sup.1H Nuclear magnetic resonance (1H NMR) spectra were recorded at ambient temperature (unless noted otherwise) on a Bruker Avance 500 spectrometer, with a working frequency of 500 MHz for 1H nuclei. Chemical shifts are reported in ppm relative to the signals corresponding to non-deuterated residual solvents. Low-pressure gas adsorption experiments (up to 850 torr) are carried out on a Quantachrome AUTOSORB-1 automatic volumetric instrument. Ultrahigh-purity-grade N.sub.2 and He gases are used in all adsorption measurements. N.sub.2 (77 K) isotherms are measured using a liquid nitrogen bath (77 K). The pore volume of each material is estimated from the DR model with the assumption that the adsorbate is in the liquid state and that the adsorption involves a pore-filling process. Powder Xray diffraction data are collected using a Bruker D8 Discover -2 diffractometer in reflectance Bragg-Brentano geometry at 40 kV, 40 mA (1,600 W) for Cu-K radiation (=1.5406 ). Single crystal X-ray diffraction data is collected using a Rigaku MM007/Saturn92 diffractometer (confocal optics Cu-K radiation) or a Rigaku MM007/Mercury/Saturn70 diffractometer (confocal optics Mo-K radiation). Thermogravimetric analyses (TGA) is performed using a TA Q500 thermal analysis system at a heating rate of 5 C. min.sub.1 in air. Elemental analyses were performed on a Thermo Flash EA1112 combustion CHNS analyser.

(54) Synthesis

(55) The preparation of CD-MOFs followed the general procedure of dissolving 1.0 equiv of CD and 8.0 equiv of the alkali metal salt in water, filtering the solution, and subsequently allowing slow vapor diffusion of MeOH into the aqueous solution to occur during several (2-7) days. The crystals that are filtered, washed with MeOH and allowed to dry in air.

(56) CD-MOF-1-CD (1.30 g, 1 mmol) and KOH (0.45 g, 8 mmol) are dissolved in H.sub.2O (20 mL). The aqueous solution is filtered and MeOH (ca. 50 mL) is allowed to vapor diffuse into the solution during the period of a week. Colorless cubic crystals (1.20 g, 66%), suitable for X-ray crystallographic analysis, are isolated, filtered and washed with MeOH (230 mL), before being left to dry in air. Elemental analysis (%) calculated for [(C.sub.48H.sub.80O.sub.40) (KOH).sub.2 (H.sub.2O).sub.8(CH.sub.3OH).sub.8]n: C 37.2, H 7.33; found: C 37.2, H 7.24%. This elemental analysis data corresponds to 22% solvent composition by weight, a percentage which is commensurate with thermogravimetric analytical data that shows a weight loss of about 22% at 1008 C. A sample is dried. Elemental analysis (%) calculated for [(C.sub.48H.sub.80O.sub.40)(KOH).sub.2(H.sub.2O).sub.2]n: C 39.9, H 5.80; found: C 39.9, H 6.00.

(57) CD-MOF-2In a specific example, gCD (1.30 g, 1 mmol) and RbOH (0.82 g, 8 mmol) are dissolved in water (20 mL). The aqueous solution is filtered and MeOH (ca. 50 mL) is allowed to vapor diffuse into the solution during the period of a week. Colorless cubic crystals (1.25 g, 71%), suitable for X-ray crystallographic analysis, are isolated, filtered and washed with MeOH (230 mL) before being left to dry in air. Elemental analysis (%) calculated for [(C.sub.48H.sub.8O.sub.40)(RbOH).sub.2(H.sub.2O).sub.11(CH.sub.3OH).sub.2].sub.n: C 34.0, H 6.40; found: C 34.1, H 6.32%. This elemental analysis data corresponds to 15% solvent composition by weight, a percentage which is commensurate with thermogravimetric analytical data that shows a weight loss of about 15% at 100 C. A sample is dried. Elemental analysis (%) calculated for [(C.sub.48H.sub.8O.sub.40)(RbOH).sub.2(CH.sub.2Cl.sub.2)0.5].sub.n: C 37.7, H, 5.42; found: C 37.8, H 5.24.

(58) The synthesis of CD-MOF-3 (CsOH) is complicated by the tendency for it to crystallize alongside a related polymorph during its preparation. Solid-state structure of CD-MOF-3 is obtained by careful selection of appropriate single crystals.

(59) A complete list of alkali metal salts used to form cubic single crystals of space group I432 and with a unit cell edge of approximately 31 is shown in Table 1. It is clear from the data listed in Table 1 that CD-MOF formation is almost ubiquitous amongst the myriad alkali metal salts available commercially.

(60) TABLE-US-00001 TABLE 1 Ratio of Metal Unit Cell Metal Salt Salt to CD Edge/ KOH (CD-MOF-1) 1:8 31.0006 (8) NaOH (CD-MOF-2) 1:8 31.079 (1) CsOH (CD-MOF-3) 1:8 30.868 (10) Na.sub.2CO.sub.3 1:8 30.751 (9) K.sub.2CO.sub.3 1:8 31.186 (6) KF 1:8 30.987 (8) K.sub.2 (azabenzen-4,4- 1:4 31.040 (4) dicarboxylate) KCl 1:8 31.161 (9) KBr 1:8 30.946 (5) NaBPh.sub.4 1:8 30.272 (10)

(61) When CD-MOF-1 is prepared using potassium benzoate as a source of K+ ions, single crystal X-ray diffraction analysis revealed a trigonal space group, R32, with unit cell parameters as follows: a=b=42.6517(3), c=28.4636(5) , ==90, =120. After refinement, the underlying CD framework linked by K+ ions is found to be analogous to that of the CD-MOF-1 structures that are obtained in the cubic space group I432. The presence of 50% of the benzoate anions in the framework is observed. It is believed that the ordering of the observed benzoate counterions causes the cubic symmetry of the unit cell to be commuted, while the underlying MOF structure remains unaffected.

(62) Single Crystal X-Ray Crystallography

(63) Single crystal X-ray diffraction data for all MOF structures are collected at 93 K using a Rigaku MM007/Saturn92 diffractometer (confocal optics Cu-K radiation) for CD-MOF-2 and CD-MOF-3 at 93 K using a Rigaku MM007/Mercury/Saturn70 diffractometer (confocal optics Mo-K radiation) and CD-MOF-1 crystallized from solutions of potassium benzoate (vide infra). The cell dimensions for the remaining samples listed in Table 1 are also obtained from full hemisphere data collections. Intensity data is collected using co steps accumulating area detector frames spanning at least a hemisphere of reciprocal space for all structures. Data is integrated using CrystalClear. All data is corrected for Lorentz, polarization and longterm intensity fluctuations. Absorption effects are corrected on the basis of multiple equivalent reflections. Structures are solved by direct methods and refined by full-matrix least-squares against F.sup.2. Hydrogen atoms are assigned riding isotropic displacement parameters and constrained to idealized geometries.

(64) Data for CD-MOF-1 grown from solutions of potassium benzoate are collected at 100 K using a Bruker d8-APEX II CCD diffractometer (Cu-K radiation). Intensity data is collected using co steps accumulating area detector frames spanning at least a hemisphere of reciprocal space for all structures. Data is integrated using SHELXTL. The cell is indexed as a superstructure, but the superstructure is proved unsolvable. The substructure is subsequently indexed and a solution is derived. Two of eight glucose rings display partial disorder, which is not resolved as a consequence of solving the substructure. This disorder resulted in isolated oxygen atoms bonded to K100 and K102 which are modeled as hydroxide anions. Structures are solved by direct methods and refined by full-matrix least-squares against F.sup.2. Hydrogen atoms are assigned riding isotropic displacement parameters and constrained to idealised geometries.

(65) Thermal Stability and Activation

(66) In order to remove interstitial solvents, as-synthesized samples of CD-MOF-1 and CD-MOF-2 are immersed in CH.sub.2Cl.sub.2 for three days. During the solvent exchange process, the CH.sub.2Cl.sub.2 is refreshed three times. The resulting CH.sub.2Cl.sub.2-exchanged sample of each CD-MOF is transferred as a suspension to a quartz cell and the solvent decanted. The wet sample is then evacuated (10.sup.3 Torr) at room temperature for 10 hours, and then at 45 C. for 12 hours.

(67) The stabilities of as-synthesized and activated samples of CD-MOF-1 and CD-MOF-2 are examined by thermogravimetric analysis (TGA) under oxidative conditions. The retention of solvents by CD-MOF-1 and CD MOF-2 evident in their respective TGA traces (FIGS. 16a and 16b) are commensurate with values obtained by elemental analysis, and they also show thermal stability of the frameworks, after solvent loss, up to temperatures of approximately 175 C. for CD-MOF-1 and 200 C. for CD-MOF-2. The stability of activated CD-MOF-1 to heating is illustrated in its TGA trace (FIG. 16c), which shows retention of mass until approximately 175 C., with a 2.4% loss in mass at approximately 100 C. corresponding to the loss of two H.sub.2O molecules per CD ring. Retention of small amounts of water by CD-MOF-1 through the activation process may occur as a result of CD's great affinity for water, or indeed this small amount of water present in the sample can be explained by deliquescence after the activation process. Thermal degradation occurs at temperatures over 175 C. An analogous experiment (FIG. 16d) with CD-MOF-2 yields similar results, with thermal stability of the sample at temperatures under 200 C. confirmed. A small (0.6%) drop in mass at approximately 100 C. indicates the presence of approximately 0.5 molecules of H.sub.2O per CD ring, again presumably on account of deliquescence.

(68) .sup.1H NMR SpectroscopyDetermination of Counterions

(69) Potassium salts of the benzoate monoanion and the azobenzene-4,4-dicarboxylate dianion yield single crystals, whose CD-MOF framework structures are confirmed by X-ray diffraction. The crystals are then dried, dissolved in D.sub.2O and the 1H NMR spectra of the solutions recorded.

(70) Potassium benzoatecolorless, cubic crystals are grown by dissolving CD (0.26 g, 0.2 mmol) and potassium benzoate (0.256 g, 1.6 mmol) in water (5 mL), filtering the colorless solution and allowing MeOH vapors to diffuse in slowly over approximately 5 days. The crystals are isolated by filtration and washed twice with MeOH (210 mL) to remove excess of potassium benzoate. The crystals are dried in vacuo, dissolved in D.sub.2O and subjected to analysis by .sup.1H NMR spectroscopy (FIG. 17).

(71) When the integral for the anomeric protons (5) of the CD units is set to eight, representing one CD torus, the remaining 48 CD protons (3.5-4.0) integrate to 49.5. The benzoate aromatic signals have a combined integral of 10.7, which corresponds to approximately two benzoate anions (5 protons are present in each molecule). This integral represents a ratio of two benzoate monoanions, and thus, two K.sup.+ cations, to one CD unit, corresponding to the ratio of K.sup.+ cations to CD tori observed in the crystal structure of CD-MOF-1.

(72) Dipotassium Azobenzene-4,4-DicarboxylateOrange, cubic crystals are grown by dissolving CD (0.26 g, 0.2 mmol) and dipotassium azobenzene-4,4-dicarboxylate (278 mg, 0.8 mmol) in water (5 mL), filtering the orange solution and allowing MeOH vapors to diffuse in slowly over approximately 5 days. The crystals are isolated by filtration and washed twice with MeOH (210 mL) to remove excess of dipotassium azobenzene-4,4-dicarboxylate. The crystals are dried in vacuo, dissolved in D.sub.2O and subjected to analysis by .sup.1H NMR spectroscopy (FIG. 11).

(73) When the integral for the anomeric protons (5) of the CD units is set to eight, representing one CD torus, the remaining 48 CD protons (3.5-4.0) integrated to 49.5. Both the cis and trans isomers of azobenzene-4,4-dicarboxylate are observed in the spectrum; the total integral is 8.44 for all the protons in the molecule. Since azobenzene-4,4-dicarboxylate has eight protons in total, this integral represents a ratio of one azobenzene-4,4-dicarboxylate dianion, i.e., two K.sup.+ cations, to one CD unit, corresponding to the ratio of K.sup.+ cations to CD tori observed in the crystal structure of CD-MOF-1. The significance of this ratio is underscored by conducting a second experiment, wherein double the quantity of dipotassium azobenzene-4,4-dicarboxylate is employed in the initial crystallization process. The .sup.1H NMR spectrum of the dissolved crystals once again show the same ratio of anion to CD, despite the doubling of their amount in the starting material. These experiments provide strong evidence that benzoate and azobenzene-4,4-dicarboxylate act as the counterions in their respective CD-MOF-1 frameworks. The these ratios of counterions correspond to 12 benzoate anions or six azobenzene-4,4-dicarboxylate anions per (CD).sub.6 cube.

(74) Small Molecule Co-Crystallization

(75) Rhodamine B is used to study the co-crystallization of dye molecules within the CD-MOF-2 framework. CD (0.26 g, 0.2 mmol) and RbOH (0.164 g, 1.6 mmol) are dissolved in water (5 mL), and Rhodamine B (ca. 0.25 g) is added until the aqueous solution is saturated. After filtration, MeOH vapors are allowed to diffuse into the red solution and deep red crystals form during 5 days. The crystals are isolated by filtration and washed with MeOH (220 mL). Subsequently, the crystals are washed with 20 mL portions of CH.sub.2Cl.sub.2 until no red color is evident in the washings, ensuring that all excess of Rhodamine B is removed prior to analysis. The crystals are dried in vacuo, dissolved in D.sub.2O and examined (FIG. 14) by .sup.1H NMR spectroscopy.

(76) The integral for the anomeric protons (5) of the CD units is set to eight, representing one CD torus. The integral for the characteristic Me protons (=1.15) of the Rhodamine B diethylamino groups also integrate for approximately eight, and so, since each molecule of Rhodamine B has 12 Me protons, this integral suggests a ratio of two Rhodamine B molecules for every three CD molecules. The total integration of the Rhodamine B aromatic protons (6.5-8.0) is 6.6, and considering that each molecule has 10 aromatic protons, the integral once again indicates the 2:3 ratio. The signals for the CH.sub.2 protons of the Rhodamine B diethylamino groups overlap with the remaining signals from the CD ring, resulting in a total integral of 54. Subtracting the expected value of 48 for the CD protons leaves an integral of approximately 6 for the Rhodamine B CH.sub.2 protons, of which there are eight in the molecule. Using the 2:3 ratio of Rhodamine B to CD molecules, an integral of 5.3 is expected.

(77) In broader terms, this 2:3 ratio indicates a loading of four Rhodamine B molecules per (CD).sub.6 cube. Although X-ray crystallography confirms the structure of CD-MOF-2 is unaffected by dye co-crystallization, the arrangement of the Rhodamine B molecules within the solid state structure of CD-MOF-2 can not be determined, because of their disorder within the vast framework.

(78) Small Molecule Adsorption

(79) 4-Phenylazophenol, an orange azobenzene-based dye, is used to observe the uptake of small molecules from CH.sub.2Cl.sub.2 solution by CD-MOF-2 crystals. As-synthesized crystals are activated by replacing the MeOH/H.sub.2O solvent with CH.sub.2Cl.sub.2, in order to exchange all interstitial solvents. This procedure is carried out without exposing the colorless crystals to air in order to minimize cracking. The crystals are soaked in CH.sub.2Cl.sub.2 for three days, after which a saturated solution of 4-phenylazophenol in CH.sub.2Cl.sub.2 is used to replace the original CH.sub.2Cl.sub.2, and the crystals are allowed to soak for a another 24 hours. The crystals are isolated by filtration and washed with CH.sub.2Cl.sub.2 until no color is evident in the washings, indicating no excess of 4-phenylazophenol remaining. The crystals are dried in vacuo, dissolved in D.sub.2O and analyzed by .sup.1H NMR spectroscopy.

(80) Following the protocol used for previous experiments, the integral for the anomeric protons (5) of the CD units is set to eight, representing one CD torus. The total integral of the 4-phenylazophenol aromatic protons (6.5-8.0) is 6.42 when there are nine aromatic protons in each molecule. This observation results in a ratio of approximately 4.3 molecules of 4-phenylazophenol present with respect to each (CD).sub.6 cube, similar to the value of four molecules of Rhodamine B per (CD).sub.6 cube measured from co-crystallization experiments.

(81) Preparation of Edible CD-MOFs

(82) Food grade crystals of CD-MOF-1 are prepared by dissolving commercially available foodgrade potassium benzoate (283 mg, 1.8 mmol) and food-grade CD (2.30 g, 1.8 mmol) in distilled water. The aqueous solution is filtered through cotton wool and, following vapor diffusion of Everclear grain alcohol into the solution over a few days, crystals of CD-MOF-1 are obtained. These crystals represent a MOF that is comprised of entirely food grade reagents.

(83) N.sub.2 Adsorption Isotherms

(84) Isotherms are measured for both CO.sub.2 and CH.sub.4 on CD-MOF-2 at incremental temperatures (FIG. 17). The total uptake of CO.sub.2 in the low pressure region (<0.01 Torr) is clearly unaffected over the temperature range from 273 K to 298 K, remaining at approximately 23 cm.sup.3/g, regardless of sample temperature. Additionally, the steep slope of the isotherm in this region suggests a strong binding event, one that would equate with the formation of a carbon-carbon bond. Notably, the abrupt transition in higher pressure regimes (>1 Torr) becomes much more dependent upon temperature, as indicated by a 30% greater uptake of CO.sub.2 at 273 K compared to 298 K. These observations are consistent with covalent bond formation occurring preferentially at low pressures. Chemisorption gives way to physisorption at elevated pressures with the change in mechanism of uptake occurring at 23 cm.sup.3/g.

(85) CP/MAS NMR

(86) Spectroscopic evidence showing the solid-state reactivity of CD with CO.sub.2 is obtained by cross polarizing magic angle spinning (CP/MAS) NMR spectroscopy. Crystalline samples are activated by exchanging the aqueous methanolic solution with dichloromethane (DCM) before being evacuation and drying at low pressure (<2.010.sup.3 Torr) for two days to remove remaining water. The activated CD MOF-2 is exposed to an atmosphere of dry CO.sub.2 for 10 minutes and transferred into an airtight zirconium solid-state NMR rotor. The .sup.13C NMR spectrum of crystalline samples of activated CD-MOF-2 show (FIG. 17) separate peaks for Cl and Cl, as well as for C4 and C4, as a result of the commuted symmetry induced by the alternating Rb.sup.+ cations on the primary and secondary faces of the CD tori. Upon exposure to CO.sub.2 a new peak emerges, centered on 158 ppm, in the form of a resonance which is known to be consistent with carbonate formation. Furthermore, peaks associated with the Cl and Cl carbons, as well as the C4 and C4 carbons, undergo appreciable changes in their chemical shift, reflecting the fact that a chemical reaction has occurred on the CD tori.

(87) It is speculated that this favorable reactivity arises because the -CD of CD-MOF-2 bear primary hydroxyl groups which define the circumference of a large (1.7 nm diameter) pore. The accessibility of these hydroxyl groups in the crystalline state far exceeds that of the amorphous state, as determined by Langmuir and BET analysis. For example, no evidence of carbonate formation is observed by CP/MAS NMR spectroscopy on pure CD after exposure to CO.sub.2. Indeed, literature reports suggest that the reaction to form carbonic acid from CO.sub.2 with cyclodextrins is not a favorable one. In order to establish that the reactivity of CD-MOF-2 with CO.sub.2 is dependent on crystallinity, samples are ground in a mortar and pestle for 10 minutes, before being exposed to CO.sub.2 and analyzed by CP/MAS NMR spectroscopy. This experiment reveals that neither new resonances nor changes to the original signals are observed, indicating (FIG. 18) that no carbonate formation occurs. Analysis of the ground sample by powder X-ray diffraction shows (FIG. 19) crystallinity is lost after seven minutes of continuous dry grinding, which renders the sample into an amorphous state, yet it retains the same ratios of rubidium salts and CD that are present in the original sample.

(88) Methyl Red pH Indicator

(89) Methyl red, a zwitterionic azobenzene based pH indicator, is diffused into the pores of CD-MOF-2 by suspension in a DCM solution of the dye. The red solution is decanted and the crystals are washed and dried in vacuo, affording brilliant yellow crystals. The yellow color arises from the incorporated methyl red undergoing partial anion metathesis and, consequently, depronation, with the hydroxide counterions in the pore structure. .sup.1H NMR spectroscopy shows that the incorporation of methyl red into CD-MOF-2 is no more than 10% by weight and BET analysis of the activated samples indicates that incorporation of the pH indicator reduces the surface area by 50%. The dried crystals are transferred to a scintillation vial, which is then exposed to both dry (from a tank) and humid (sublimed dry ice) CO.sub.2 vapor. The initial color change from yellow to orange/red occurs quickly, regardless of the CO.sub.2 source, and, after five minutes, no further color change can be discerned by the naked eye. When the source of CO.sub.2 is removed, the crystals revert to a yellow color, indicating that the transient carbonic acid function is reverted to the alcohol, liberating CO.sub.2. This process is repeated many times and no apparent fatigue is observed on the material.

(90) As a control, the same experiment is performed on crystals of CD-MOF-5, which are loaded with the ammonium salt of methyl red. The structure of CD-MOF-5-Zn.sub.4O clusters linked by terephthalate dianionsdoes not contain any free hydroxyl or amino groups capable of forming an acid moeity to elicit a pH-based color change, and therefore would not be expected to demonstrate chemisorption. The presence of CO.sub.2 has no effect on crystal color, indicating that no chemisorption is occurring. To rule out the possibility that the more nucleophilic hydroxide counterion is the reactive agent forming carbonate anions by reaction with CO.sub.2, crystals of so-called edible-CD-MOF, which are isostructural with CD-MOF-2 but are synthesized from potassium benzoate, are used. After activation and removal of solvent, the crystals of edible CD-MOF changed color reversibly within the same time frame. CP/MAS spectroscopy on edible CD-MOF is performed as well, and the carbonate resonance appearing 158 ppm appearsprecisely the same chemical shift found in CD-MOF-2 when exposed to CO.sub.2.

(91) The disclosures of all articles and references, including patents, are incorporated herein by reference. The invention and the manner and process of making and using it are now described in such full, clear, concise and exact terms as to enable any person skilled in the art to which it pertains, to make and use the same. All references cited in this specification are incorporated herein by reference. It is to be understood that the foregoing describes preferred embodiments of the present invention and that modifications may be made therein without departing from the spirit or scope of the present invention.