Designing New Lattice Inclusion Hosts Roger Bishop School of Chemistry, The University of New South Wales, Sydney 2052, Australia 1 Introduction Organic inclusion compounds may be classified, in the majority of cases, into two distinct categories. The more widespread group comprises unimolecular compounds where one host molecule interacts with one guest species.’ Familiar examples include host types such as the cyclodextrins, crown ethers, cryptands, carcerands and their related derivatives. The synthesis of new materials of this type has become an extremely active and highly sophisticated branch of chemistry, partly because of their inherent 1: 1 stoichiometry and because molecular modelling often permits reliable prediction.2 On the other hand, there are many other inclusion systems where the arrangement of molecules comprising the host lattice itself results in the observed host-guest properties.* Synthesis of new examples of these multimolecular or lattice inclusion compounds is problematical because lattice packing arrangements (even for simple organic molecules) cannot yet be predicted by computa- tional methods -despite innovative approaches toward this end.7 How, therefore, can new lattice inclusion hosts be discovered? The traditional answer to this question used to be ‘by fortunate acci- dent’: but in this article I shall describe how systematic approaches involving crystal engineering5 can now help us towards overcom- ing these formidable synthetic difficulties.The black comedy Cut’s Cradle by Kurt Vonnegut6 relates how Dr Felix Hoenikker succeeds in synthesising the polymorph ice- nine.In the hands of his dysfunctional offspring, the existence of this close-packed and thermodynamically stable substance leads to an all-too-inevitable outcome -but how was this molecular lattice designed? This strikingly successful fictional example of crystal engineering was inspired by analysis of stacking arrangements of the cannon balls decorating court-house lawns and through model- ling using the permutations of a cat’s cradle -techniques not so far removed, perhaps, from current methodology. The concept of ionic close-packing is widely used to explain the structure of simple lattices such as that of sodium chloride.Similarly, one of the basic tenets of practical organic chemistry is that recrystallisation of a crude reaction product will yield crys- talline material of high purity. Both these everyday illustrations presume that simple ions or molecules will pack together efficiently to yield high density structures, without void spaces which might contain guest species.’ Roger Bishop was educated at George Heriot’s School in Edinburgh, the University of St. Andrews (BSc)and the University of Cumbridge (PhD). In 1974 he took up a Lectureship at the University of New South Wales in Sydney, Australia, where he is currently an Associate Professor in the Department of Organic Chemistry. He has also held visiting research posi- tions at Ohio State and Durham Universities, and was the 1993 Olle‘ Prize winner of the Royal Australian Chemical Institute.His principal research inter- rests lie in the areas of alicyclic and supramolecular chemistry, and in the synthesis and appli- cation of new inclusion systems. In contrast, crystallisation of the extremely simple molecule H,O results in formation of a rather open lattice structure accompanied by the remarkable density decrease from 0.9998 to 0.9168 g cm-3. This is caused, of course, by formation of the highly directional hydrogen-bonding network present in ice. Furthermore, a crystallo- grapher would not be surprised if protein crystals for X-ray struc- tural determination contained associated components such as water molecules, sodium ions, and the crystallisation solvent itself -sometimes in large quantities.Hence, it is also possible that appropriate intermolecular forces combined with ‘awkward’ molecular shapes can lead to difficulties in close-packing, which might be alleviated by guest inclusion. Successful approaches to new lattice inclusion hosts, therefore, must take into account factors such as molecular size, shape and symmetry; and must also seek a fine balance between the intermol- ecular repulsive forces and attractive forces present. In supramole-cular synthesis, specific types of intermolecular non-covalent attraction are akin to the synthons of conventional synthetic chem- istryex We must learn to recognise these motifs, discover their influ-ence on molecular packing, and develop the ability to use such arrangements to our ad~antage.~The subtle interplay of these diverse factors is illustrated here for a selection of synthetic approaches to recently reported lattice inclusion systems.2 Propeller Blades, Wheels and Axles Since the initial report of the triphenylmethane-benzene compound by KekulC in 1872 it has become recognised that inclusion proper- ties are frequently associated with polyaryl systems that can adopt propeller-shaped arrangements. This is because the aryl-aryl inter- actions8 that predominate in such structures tend to result in forma-tion of void spaces suitable for guest entrapment. An interesting example is tris(5-acetyl-3-thienyl)methane1, reported to be a versatile host by bin Din and Meth-Cohn in 1977, but only now receiving proper structural attention.’O In the A-form of its 2:l cyclohexane compound (Fig.1) the near planar acetyl- thienyl groups are the propeller blades, which interact through two distinct offset face-face intermolecular mechanisms. One pair of blades overlaps thienyl.. .thienyl, while the other two pairs overlap carbonyl-*. thienyl. For all three the interplanar separation is about 3.6 A -a characteristic value for this type of interaction. The het- eroaromatic groups of 1function just as effectively as the benzenoid rings traditionally associated with polyaryl propeller-shaped hosts, indicating considerable potential for further host design in this area. Direct connection of two polyaryl groups by a linear spacer leads to the wheel and axle compounds first recognised by Toda and since developed into a major class of inclusion hosts.” Hence, in struc-ture 2, the wheels are tris(m-toly1)phosphine groups and the axle comprises a gold acetylide moiety.In the benzene compound of 2 the guest molecules occupy barrel-shaped cavities created by neigh- bouring m-tolyl rings (Fig. 2). Three m-tolyl groups describe the top, another three the bottom, and six others define the staves, of each barrel.I2 Inclusion properties of both the above categories may be modi- fied (and frequently enhanced) by incorporating hydrogen-bonding groups. Typical structures such as 3 and 4 suffer moderate crowd- ing around their hydroxy groups, which can inhibit intermolecular hydrogen bonding.Consequently, polar guests able to act as bridg- ing links between separate host molecules now tend to be incl~ded.~~ Inclusion systems stabilised by varying proportions of complex- ation and lattice dispersion forces have been termed coordinato- clathrates by Weber, and these often exhibit novel guest selectivity 31 I 3 12 CHEMICAL SOCIETY REVIEWS, 1996 co I CH3 I 2 Figure 1 Crystal packing in l2*(cyclohexane) viewed along a For clarity the tris(5 acetyl 3 thieny1)methane hosts are drawn as molecular frame works, cyclohexane carbons are shown as filled spheres, and all hydrogen atoms are omitted Thienyl thienyl interactions are edge on, and the carbonyl thienyl interactions are inclined, in this view The cyclohexane guests occupy channels along a arising from the balance between these factors l3 For example, the acetylenic alcohol 3 is able to remove ethanol from aqueous solu-tions through formation of the complex 3-(ethanol) Heating of this solid liberates the pure guest, thereby providing an unusual and valuable isolation procedure l4 The crystal structure of 3-(ethanol) (Fig 3) reveals that two mol ecules of each type are hydrogen bonded together through the cyclic Figure 2 Benzene in a barrel For clarity only the P(m tolyl), partial struc ture of compound 2, which IS directly involved in the host-guest con struction, is drawn here Hydrogen atoms are omitted and phosphorus atoms are hatched The six peripheral edge on m tolyl groups define the walls, while the other SIX define the ends, of the barrel shaped cavity present in 2.(benzene) 4 5 6 array (-0-H), which is a well-known supramolecular motif amongst alcohols I5 In this instance, this is the dominant interaction present The parallel aryl groups visible in Fig 3 are actually about 5 5 8, apart and do not interact, but inter-complex stabillsation IS provided through =C-CH3 aryl interactions 3 Inclined Planes and Rigid Spacers Another key strategy is the targeting of molecules the structures of which involve inclined planar sub-structures For example, two planar units may be joined edge-dge to give scissor-shaped mole- cules, or edge-face to give either T-shaped or roof-shaped mole- cules As demonstrated most convincingly by WeberI6 significant numbers of such compounds exhibit inclusion properties These characteristics often persist in the extended structures obtained through addition of further planar sub-structures and/or rigid spacer groups as shown here DESIGNING NEW LATTICE INCLUSION HOSTS -R.BISHOP Figure 3 The unit cell of 3-(ethanol) viewed along a and with the host-guest hydrogen bonded cycles (-0-H), represented by dashed lines. For simplicity all molecules are reproduced as framework drawings and all hydrogens are omitted. The hydrogen bonded units are linked through =C-CH 3.. .aryl interactions. Figure 4 The filled molecular cleft compound 5-(1,3$-trinitrobenzene). Oxygen atoms are lightly stippled, nitrogens are heavily stippled, and hydrogen atoms are omitted.Three planar units joined edge-face-edge as a U-shaped cleft can result in a unimolecular compound, as illustrated in Fig. 4for the combination of 5 and 1,3>-trinitrobenzene. In this structure the sep- aration between the dibenzofuran units and the aromatic guest is about 3.3 A. However, if the depth of the cleft is insufficient, as in the related molecule 6, then multimolecular inclusion compounds result instead. I The inclusion properties of tetraarylporphyrin derivatives have been developed extensively by Strouse and his colleagues. Indeed, these porphyrin sponges arguably now constitute the most compre-hensive family of lattice clathrate hosts.18 The peripheral aryl groups of these hosts are attached in edge-on fashion to the central planar ring and, consequently, these molecules pack as corrugated sheets, which stack to produce parallel guest-containing channels.This is illustrated in Fig. 5 by the host zinc tetraphenylporphyrin 7 as its inclusion compound 7.(naphthacenequinone). Extension of the wheel and axle concept by replacing the central linear spacer by a rigid planar group is typified by the polythiophene derivatives8 and 9 reported by Kobayashi.I9 Both compounds form Figure 5 A typical porphyrin sponge inclusion structure (zinc tetraphenyl- porphyrin 7)-(naphthacenequinone), showing the characteristic guest- filled channel arrangement. Hydrogen atoms are omitted for clarity.7 8 fi hydrogen-bonded complexes with Me,SO, but the stoichiometry and structural type alternate with the number of fused thiophene units in the central planar group. When this is odd then hydrogen- bonded host dimer formation is possible resulting in a compound such as 8;(Me2SO),, but if even then materials like 9.(Me,SO), are produced. The crystal structures of these closely related compounds are compared in Fig. 6. Further implications of this particular design concept are explored later in Sections 5 and 6. 3 14 CHEMICAL SOCIETY REVIEWS, 1996 Figure 6 Variation of inclusion structure with the number of fused thio phene rings in the central planar spacer group Top odd number of units yielding S2 (Me,SO), Bottom even number of units giving 9-(Me,SO), 4 Hydrogen Bonded Lattices Inspired by Hydroquinone Hydroquinone 10 is such a small and highly symmetrical molecule that one would expect it to pack in a trivial lattice arrangement Surprisingly, therefore, its most stable crystal form at room tem- perature (ahydroquinone) has a highly convoluted structure con- taining an amazing 54 molecules per unit cell 2o As for the case of ice, the demands imposed by intermolecular hydrogen bonding and symmetry dominate other contributing factors such as molecular size and shape Hence, hydroquinone 10 exhibits molecular sim plicity but supramolecular complexity Because of these remarkable properties it has played a central role in the development of inclu sion chemistry Packing difficulties can be overcome, in part, by adopting the P-hydroquinone lattice (Fig 7) and by including small guest mole- cules Such materials were first noted by Wohler in 1849 but their exact nature was not proved till nearly a hundred years later through the work of Powell who largely founded the modern era of struc- ture-based inclusion chemistry The P-hydroquinone structure is a superlattice constructed from two identical interpenetrating (but unconnected) simple cubic sublattices Small voids between these sublattices imprison the guest molecules within the superlattice The design of new organic superlattices built up from two or more equivalent interpenetrating sublattices is a fascinating current aspect of crystal engineering and is frequently associated with guest inclusion Readers are directed to the excellent account in this Journal by Zaworotko 22 MacNicol noted the characteristic hydrogen-bonding pattern present in P hydroquinone and realised that it was also present in inclusion compounds formed by Dianin's compound 11 and related 'OH 10 11 ArAr Ar ArI I Ar Ar d * Ar Ar Ar Figure 7 Top Diagrammatic representation of one p hydroquinone sublat tice emphasising the network characteristics of this part structure The solid tapering rods represent hydroquinone molecules 10,and the dashed lines indicate hydrogen bond cycles (-0-H), Bottom left Representation of one hydrogen bonded hexamer motif which is the supramolecular core of the above structure Bottom right Diagrammatic representation of the hexa host principle, showing aryl (Ar) and linking (L) groups, and comparing the interatomic separation (d') with that present (d)in the (-O-H), core of p hydro quinone derivatives In a brilliant piece of thinking he realised that the 'arms' of a hexa-substituted benzene would adopt a closely related geom- etry, which also would be favourable for guest inclusion (Fig 7) Hence, for the first time an entire family of new lattice inclusion hexa hosts such as 12, without a direct molecular relationship to previously known compounds, were successfully designed and syn- thesised 23 This striking outcome represents the first direct applica- tion of what we would now describe as the supramolecular synthesis of new inclusion hosts A more recent strategy by the MacNicol group affords novel inclusion hosts termed Piedfort assemblies24 where the central aro- matic core is doubled in thickness (Fig 8) This is achieved by assembly of two stacked tri-substituted aromatic units, which replace the previous hexa-substituted molecule but retain its three- fold symmetry Hence, for example, 2,4,6-tris[4-(2-phenylpropan-2 yl)phenoxy]-l,3,5 triazine 13 forms the highly crystalline 1 2 adduct with dioxane illustrated in Fig 9 More recently, inspection of the P-hydroquinone superlattice suggested to Ermer25 that the dimensions of [ 60lfullerene closely matched those of a single sublattice unit Furthermore, he reasoned that aryl aryl interactions between the electron-acceptor fullerene and the electron-donor hydroquinone rings should provide consid- erable stabilisation if a host-guest compound were produced On testing this outstanding idea it was indeed found that crystallisation of a mixture of the two components afforded black crystals of com- position lO;(C,,) where the fullerene was embedded within just one P-hydroquinone sublattice as illustrated in Fig 10 5 Other Hydrogen-bonded Network Structures Of course, hydroquinone is by no means the only molecule capable of providing a strong hydrogen bonded host network containing guests Indeed, allusions to interpenetrating lattice structures22 and Dianin's compound 1127have already been made Hydrogen bonding is usually stronger, and often more directional, than the other intermolecular attractions present in a lattice inclusion struc- ture Therefore, the requirement here is for the host to exert a DESIGNING NEW LATTICE INCLUSION HOSTS-R.BISHOP Figure 8 The Piedfort unit of 2,4,6-tris(4-(2-phenylpropan-2-yl)phenoxy1-1,3,5-triazine molecules 13 in its inclusion compound 13.(dioxane),. emphasising its symmetry, the double thickness aromatic core, and the six pendant arms. Non-hydrogen atoms of one molecule of 13 are represented by open spheres, and those of the other by shading. The inter-triazine aryt...aryt separation is ca. 3.5 A. SPh I (332I SPh 12 13 dominant effect on the molecular combination by providing a strong intermolecular network of low density.Three quite different cases of this phenomenon are described here. The steroidal bile acids are an important family of compounds involving a flexible side chain with a terminal carboxylic acid group, attached to a rigid fused-ring steroidal skeleton whose @-face is hydrophobic and a-face hydrophilic owing to the presence of hydroxy substituents. These materials frequently self-associate by means of convergent intermolecular hydrogen bonding, thereby producing head-tail bilayers. Crystal packing results in flexible par- allel hydrophobic channels, which frequently enclose less polar guest molecules .*6 An illustration of how this behaviour can be modified and con- trolled is provided by cholanamide, the amide derivative 14 of the natural product cholic acid.In its inclusion compounds one amide hydrogen atom participates in the host-host hydrogen-bonding 3 15 w Figure 9 Lattice packing in 13.(dioxane), showing four Piedfort units and their associated guest molecules. For clarity, all molecules are shown in framework representation, all hydrogens are omitted, and no distinctions are made between carbon and heteroatoms. Each triazine core is shown as a solid hexagon to emphasise the Piedfort units in this structure. / (I 1 \ \L--A L--d Figure 10 Diagrammatic representation of (hydroquinone),. (CHI) viewed along c. The small solid hexagons represent the aryl rings, and the large dashed hexagons the (-0-H), cycles. network. However, the other hydrogen protrudes from the channel wall and is now available as a molecular hook for polar hydrogen- bond acceptor groups like alcohols and ethers, as shown in Fig. 1 I for 14-(dioxane).26 Our own work has uncovered a remarkable family of inclusion hosts related to the simple diol 15 where the hydroxy groups assem- ble around a threefold screw axis producing a hydrogen-bonded 'spine'. The alicyclic framework functions as a rigid spacer group and results in these diols crystallising as lattices containing large parallel canals (Fig. 12) in which guests are trapped on a size and shape basis.This spine motif is another example of a supramolecular synthon* and CHEMICAL SOCIETY REVIEWS, I996 14 15 16 17 can be transplanted into other alicyclic systems thereby providing a family of hosts termed the helical tubulands.These have exactly the same crystal packing but differ considerably in their canal dimensions and resulting inclusion proper tie^.^^ The cutaway view of 15; (chlorobenzene) shown in Fig. I3 reveals the head-tail guest packing arrangement along one such canal of this helical tubulate compound. The third example illustrating network hydrogen-bonding hosts is Aoyama's phenol 16,2*which combines this property with those of the inclined plane systems discussed in Section 3. In the inclu- sion compound 16.(benzophenone),, the four phenol groups of 16 hydrogen bond to their neighbours creating a rather open layer structure in which the protruding anthracene groups are orthogonal to the phenolic rings.This results in the construction of a series of open cages with two benzophenone guest molecules occupying each cage. Each guest is stabilised by its carbonyl oxygen accept- ing a phenolic hydrogen bond and through aryl. . . aryl interaction with the anthracene groups (Fig. 14). Offset stacking of these layers affords a series of tubes formed by placing the open cages on top of each other. These tubes can be envisioned by imagining an infinite number of Fig. 14 units directly stacked on top of each other in a fully eclipsed manner. Consequently, the benzophenone guests occupy parallel tubes running through the crystal lattice. It is particularly noteworthy that, despite the considerable dif- ferences in molecular structure between the building blocks 15 and 16 and also in the construction of their inclusion networks, both compounds behave rather similarly.Both include guests in parallel tubes, their networks can survive as apohosts on removal of guest species, and their inclusion structures can be regenerated on exposure to guest vapour. These properties are remarkably similar to some of those exhibited by inorganic zeolitic struc- tures. Figure 11 Structure of (cholanamide 14).(dioxane) viewed along the b axis. Oxygens are shown as solid spheres, nitrogen atoms are striped, and hydrogen bonds are indicated by dashed lines. All hydrogen atoms are omitted. Figure 12 Projection in the ab plane of a section through the helical tubu- land lattice of diol 15 showing the parallel canals of triangular cross- section.The helical characteristics of each canal are largely masked in this representation. Oxygen atoms of the helical hydrogen bonded spine motifs are stippled. DESIGNING NEW LATTICE INCLUSION HOSTS-R. BISHOP Figure 13 View of one canal of 15,.(chlorobenzene) with one column of canal wall diol molecules removed to show the guest arrangement. In this instance head-tail packing is adopted through utilisation of a 2.4 8, Ar-H..-CI interaction between neighbouring guests. 6 Selective and Stereoselective Inclusion Properties A particularly important area of inclusion chemistry is the design of hosts capable of highly selective inclusion properties, and the fol- lowing three very different cases illustrate this aspect.The roof- shaped host 17 bears em-bromo substituents which reduce efficient awl-aryl packing and encourage neighbouring molecules to asso-ciate in other ways. This results in 17 being a selective host for small polyhalogenated guests that help create a network of halogen-..halogens interactions, as illustrated in Fig. 15 for 17.(CHCI,). Non-halogenated molecules of comparable size and shape are excluded.29 Chirally pure materials are assuming ever-growing importance in organic chemistry, and hence there is considerable interest in devis-ing chiral host systems capable of controlling stereoselective sepa- rations or reactions. Inclusion hosts are especially valuable for such processes since, subject to mechanical losses, the active agent is fully recoverable.These areas have been pioneered especially by Toda and his colleagues who have developed highly original appli- cations of new chiral hosts.3O An illustration is provided here by (S,S)-(-)-18, which can be used to resolve racemic 2-methylpiperi- dine 19 through two distinct procedures. First, if host and racemic guest are crystallised from toluene then a I: 1 complexof(S,S)-( -)-18and(R)-( -)-19isproduced. Distillation under reduced pressure then affords a 67% yield of (R)-(-)-19in 7 1 '31 ee. In this inclusion structure one hydroxy group of 18 is hydrogen bonded to both the piperidine and a second molecule of 18. In contrast, if the materials are crystallised from methanol solu-tion then the quite different 1: 1 :1 complex of (S,S)-( -)-18, methanol, and (S)-(+)-19 is obtained.This time distillation pro- duces a 67% yield of (S)-(+)-19 in 62% ee. This crystal structure differs in having the methanol acting as a hydrogen-bonded link between the two molecules of 18,and creates an environment suited to the alternative enantiomer of 19. In these, and other similar res- olution experiments, repetition of the procedure generates products of extremely high optical purity. K Figure 14 The hydrogen bonding network present in a molecular sheet of lC(benzophenone), showing incorporation of two guests in the supramolecular cavity. Repeated eclipsed stacking of this unit affords an example of the guest-filled zeolite-like tubes which run parallel through- out this structure.Figure 15 The different intermolecular halogen-halogen attractions (3.46-4.10 A) present in the structure 17-(CHCI,)indicated by heavy dashed lines. These form a network allowing specific trapping of small polyhalogenated guests. Brorno and chioro atoms are shown as large filled spheres, and hydrogens as small filled spheres. c1 18 19 OH 20 21 The structure of host 18 represents further development of the wheel and axle and inclined plane design strategies (see Sections 2 and 3),as does its interesting cousin 20 developed by Weber 31 In this case, use of bulky camphor groups as the wheels allows facile intro- duction of chirality from natural sources without any need for resolu-tion This host includes the aromatic epoxide (S)-(+)-21 as a 1 1 compound, but totally excludes the (-)-enantiomer, on crystallisation from solution Design features readily discernible in this inclusion structure (Fig 16) involve both host-host and host-guest hydrogen- bonding, aryl aryl interactions, and coordinatoclathrate attractions 7 Conclusions The design of new lattice inclusion compounds has now left tradi- tional serendipity far behind Synthetic approaches using analogy with known examples, or based on design elements such as planes and spacers, are well-established and frequently successful With our rapidly advancing understanding of intermolecular attractive forces, opportunities for synthesis are entering an excit-ing new phase Lattice inclusion compounds are an excellent choice for developing applications of these supramolecular synthons, since the formation of poorly packed host lattices must depend on impor-tant factors, which are capable of our recognition and exploitation Such knowledge, backed up by strong chemical intuition, eventu- ally will allow design to order of entirely new lattice inclusion systems Materials capable of specific inclusion behaviour will be especially important targets Acknowledgements I thank Dr Marcia Scudder for data retrieval from the Cambridge Structural Database32 and for generation of the figures illustrating this article Additional artwork was kindly carried out by Martin Dudman We thank the Australian Research Council for financial support of our own research into inclusion chemistry 8 References 1 D D MacNicol, J J McKendrtck and D R Wilson, Chem Soc Rev, l978,7,65 2 'Inclusion Compounds,' ed J L Atwood, J E D Davies and D D MacNicol, Academic Press, London, 1984 Vols 1-3, Oxford CHEMICAL SOCIETY REVIEWS, 1996 Figure 16 The molecular arrangement in 20-[(S)-(+) 211 with hydrogen bonding represented by dashed lines Oxygen atoms are indicated by solid spheres and only hydroxy hydrogens are drawn The stabillsing aryl-aryl interactions between the guest molecules and the planar naphthyl spacer group are clearly apparent University Press, Oxford, 1991 Vols 4-5.'Comprehensive Supramolecular Chemistry,' ed J M Lehn and J L Atwood, J E D Davies. D D MacNicol and F Vogtle, Pergamon Press, Oxford, 1996 Vols 1-11 3 J Maddox, Nature, 1988,335,201, A Gavezzotti,J Am Chem Soc , 1991,113,4622 4 J E D Davies, W Kemula, H M Powell and N 0 Smith, J Inch Phenom , 1983,1,3 5 G R Desiraju, Crystal Engrneerrtig The DeJign of Organic Solids, Materials Science Monographs No 54, Elsevier, Amsterdam, 1989 6 K Vonnegut, Jr , 'Cat's Cradle,' Gollancz, UK,1963 7 A I Kitaigorodskii, 'Molecular Crystals and Molecules,' Academic Press, New York, 1973 8 G R Destraju,Angew Chem , Int Ed Engf , 1995,34,23 1 I 9 M Mascal,Contemp Org Svnrh , 1994,1,31 10 L bin Din and 0 Meth-Cohn,J Chem SOC , Chem Cotnmun , 1977, 741.L Pang and F Brisse, Can J Chem , 1994,72,2318 I1 F Toda and K Akagi, Tetrahedron Lett, 1968, 3695, F Toda, in 'Inclusion Compounds,' ed J L Atwood, J E D Davies and D D MacNicol, Oxford University Press, Oxford, 1991 Vol 4 ch 4,pp 126-187 12 M I Bruce, K R Grundy, M J Liddell, M R Snow and E R T Tiekink,J Organomer Chem , 1988,344, C49 13 E Weber, in 'Inclusion Compounds,' ed J L Atwood, J E D Davies and D D MacNicol, Oxford University Press, Oxford, 1991, Vol 4, ch 5,pp 188-262 4 F Toda, K Tanaka and T C W Mak, Bull Chem Soc Jpn , 1985,58, 2221 5 S C Hawkins,M L Scudder,D C Craig,A D Rae,R B Abdul Raof, R Bishop and I G Dance,J Chem Soc Perkin Trans 2,1990,855 6 E Weber and M Czugler, Top Curr Chem , 1988,149,45 7 M Harmata and C L Barnes, J Am Chern Soc, 1990,112,5655.Tetrahedron Lett . 1990,31, I825 8 M P Byrn,C J Curtis, I Goldberg,Y Hsiou,S I Khan,P A Sawin, S K Tendtck and C E Strouse,J Am Chem Soc , 1991,113,6549 DESIGNING NEW LATTICE INCLUSION HOSTS-R BISHOP 19 Y Mazaki, N Hayashi and K Kobayashi, J Chem Soc Chem Commun ,1992,1381 20 S C Wallwork and H M Powell, J Chem Soc , Perkin Trans 2,1980, 21 H M Powell, in ‘Inclusion Compounds,’ ed J L Atwood, J E D Davies and D D MacNicol ,Academic Press, London, 1984 Vol 1,ch 1,pp 1-28 22 M J Zaworotko, Chem SOC Rev, 1994,23,283 23 D D MacNicol, in ‘Inclusion Compounds,’ ed J L Atwood, J E D Davies and D D MacNicol, Academic Press, London, 1984, Vol 2, ch 5, pp 124-168, D D MacNicol and G A Downing, in ‘Comprehensive Supramolecular Chemistry, Vol 6 Solid State Supramolecular Chemistry Crystal Engineering,’ ed D D MacNicol.F Toda and R Bishop, Pergamon Press,Oxford, 1996,ch 14,pp 421464 24 A S Jessiman, D D MacNicol, P R Mallinson and I Vallance, J Chem Soc ,Chem Commun ,1990,1619 25 0 Ermer, Hefv Chim Actu, 1991,74,1339 26 K Sada, T Kondo. M Miyata, T Tamada and K Miki,J Chem Soc Chem Commun ,1993,753,M Miyata and K Sada, in ‘Comprehensive 3 I9 Supramolecular Chemistry. Vol 6 Solid State Supramolecular Chemistry Crystal Engineering,‘ ed D D MacNicol, F Toda and R Bishop, Pergamon Press, Oxford, 1996, ch 6, pp 147-176 27 A T Ung, D Gizachew, R Bishop, M L Scudder, I G Dance and D D Craig, J Am Chem Soc , 1995.117,8745, R Bishop. D C Craig I G Dance,M L ScudderandA T Ung,Supramol Chem 1993,2. I23 28 K Endo, T Sawaki, M Koyanagi, K Kobayashi, H Masuda and Y Aoyama,J Am Chem Soc , 1995,117,8341 29 C E MarJo,R Bishop,D C Cra1g.A O’Brienand M L Scudder,J Chem SOC, Chem Commun . 1994.251 3 30 F Toda, K Tanaka. I Miyahara, S Akutsu and K Hirotsu. J Chem Soc Chem Commun , 1994, 1795. F Toda. Top Curr Chem . 1988. 149, 21 1 31 P P Korkas. E Weber, M Czugler and G Naray Szabo, J Chem Soc Chem Commun . 1995,2229 32 F H Allen, J E Davies, J J Galloy, 0 Johnson. 0 Kennard, C F Macrae. E M Mitchell. G F Mitchell. J M Smith and D G Watson. J Chem Inf Comp Scr , 1991,31,187