2 Physical Methods Part (iii) X-Ray Crystallography By M. B. HURSTHOUSE Chemistry Department Queen Mary College Mile End Road London El 4NS and S. NEIDLE Department of Sioph ysics Kings College Drury Lane London WC2 5RL 1 Introduction As in previous years the large volume of published work in the field of organic chemical crystallography necessitates that this survey be highly selective. In choosing the papers for inclusion attempts have been made not only to highlight those structures which are novel or relevant to theoretical and mechanistic studies currently in progress but also to give a broad indication of the distribution of crystallographic effort amongst the various areas of organic chemistry. Fortunately the gaps left by the restricted coverage are adequately filled by two other surveys.Volume 1 of the Specialist Periodical Report ‘Molecular Structure by Diffraction Methods’ published in late 1973 covers the literature (for both organic and inorganic structures) for the period from January 1971 to March 1972. Four volumes of the compilation ‘Molecular Structures and Dimensions’ have now appeared giving virtually complete coverage of the litera- ture up to mid-1972. Volume 2 of the Specialist Report and Volume 5 of ‘Mole- cular Structures and Dimensions’ should both have appeared by the time this Report is in print. 2 BondingStudies The structures of two stable free radicals have been reported. The verdazyl ring in 2,4,6-triphenylverdayl (1) is non-planar the methylene carbon atom being displaced by 0.59 A from the plane of the four nitrogen atoms.’ An earlier e.s.r.study had indicated approximately equal sharing of the unpaired electron among the four nitrogen atoms and this is confirmed by the delocalization in the N,CN portion of the ring. The crystal structures of the free radical (2) and its diamagnetic analogue (3) have been compared.2 The approximate molecular * D. E. Williams Acta Cryst. 1973 B29 96. ’ W. Wong and S. F. Watkins J.C.S. Chem. Comm. 1973 888. 46 Physical Methods-Part (iii) X-Ray Crystallography symmetry of both is C,and in neither is the five-membered ring planar probably owing to the steric effect of the methyl groups. The most striking difference between the two molecules is in the dihedral angle between the phenyl group and the mean plane of the five-membered ring.This is 29" in (2) and 71" in (3) although it is not clear how the cause of this difference is apportioned between packing and electronic effects. Three papers describe structures containing carbanions stabilized by bis- N-ligand co-ordinated Li+ ions. In both triphenylmethyl-lithium tetramethyl- ethylenediamine3" and fluorenyl-lithium bis-quin~clidine~~ the position of the lithium ion which presumably is an indication of the location of the charge on the anions is asymmetric with respect to the symmetry ofthe organic species. In a third structure which contains the naphthalene dianion the lithium ions sit over the long bisector of the naphthalene skeleton (one above and one below the plane each associated with one ring only) and are moved out towards the C-24-3 (or C-6-C-7) bond which has a length of 1.34A only.3' The structures of ylides continue to attract attention.Triphenylphosphonium cyclopentadienylide is quite unreactive for an ylide. The structure analysis shows considerable n-delocalization within the cyclopentadiene ring and a rather small amount (ca. 20 %) of double-bond character in the P-C bond which has a length of 1.718A.4 The predominant structure is thus considered to be (4). In 2,2,3,3 4,4-hex~uoro(triphenylphosphoranylidene)cyclob~tane,~ the C-P distance (1.713A)is very similar to that in compound (4),and the results generally imply that the structure is predominantly (5). (a)J . J. Brookes and G.D. Stucky. J. Amer. Chem. Soc. 1972,94,7333; (6)J. J. Brooks. W. Rhine and G. D. Stucky ibid. p. 7339; (c)ibid. p. 7347. H. L. Ammon G. L. Wheeler and P. H. Watts jun. J. Amer. Chem. Sor. 1973,95,6158. M. A. Howells R. D. Howells N. C. Baenziger and D. J. Burton J. Amer. Chem. Soc. 1973 95 5366. M. B. Hursthouse and S. Neidle The crystal-structure analysis of the Wittig salt (3,7-dimethylocta-2,6-dienyl)-triphenylphosphonium bromide shows the P-C bond to the octadienyl chain to have a length of 1.805A slightly less than for a normal P-C(sp3) single bond.6 Crystallographic studies have been combined with CND0/2 calculations to investigate the structures and charge separations in 4,4-dicyano-2,3-diphenyl-triafulvene and 2,3-diphenyl~yclopropenone.~ The structures can be represented by (6) and (7) respectively with a slightly larger negative charge on 0 in the cyclopropenone than on the C(CN) group in the fulvene.0-Ph h, PhA Ph Charge-transfer complexes continue to attract attention. Crystals of the com- plex [(riboflavin)(quinol)(HBr),(H,O),] are black and the quinol molecules and bromide ions both donate charge to the riboflavin molecules which are probably in the semiquinone state.' It is considered that the interaction with the donors is not confined to a specific region of the isoalloxazine nuclei. A number of papers are concerned with studies on the aromaticity of non-benzenoid n-systems. Crystal-structure analyses of a number of substituted 1,2,4-triazoles show that the aromaticity is decreased as the H atoms are replaced the least aromatic systems being thiols such as (8),in which form (9) is a significant contributor.' The tropone molecule studied at -60 "C is approximately planar but shows pronounced bond alternations." The lengths of the formally single and double bonds fall into the ranges 1.415-1.459 and 1.327-1.365 A respec-tively.The increased accuracy of crystal-structure determination arising from the use of diffractometers and study of non-heavy-atom derivatives is being exploited to investigate the distribution of bonding electron density in molecules. One particularly neat use of this facility has been made in the accurate analysis of the ' J. Hjortas Acfa Crysf. 1973 B29 767.' H. L. Ammon J. Amer. Chem. Soc. 1973,95 7093. C. A. Bear J. M. Waters and T. N. Waters J.C.S. Perkin II 1973 1266. R. C. Seccombe and C. H. L. Kennard J.C.S. Perkin IZ 1973 1. lo M. J. Barrow 0.S. Mills and G. Filippini J.C.S. Chem. Comm. 1973.66. Physical Methods-Part (iii) X-Ray Crystallography 49 structures of 2,2’-anhydro-1-/3-arabino-furanosyluracil (10) and 2,5’-anhydro- 2’,3’-isopropylidenecyclouridine(1 l).’ The analysis has not only provided evidence of a shift of electron density from two bonding orbitals on N-3 of the uracil ring in (10) to the lone-pair orbital when a hydrogen bond is formed but also an explanation for two anomalous coupling constants in the I3C n.m.r. spectrum of (1 1). H 3 Strained Molecules and Conformational Studies Intramolecular strain can arise either from excessive steric repulsions or the formation of small rings.1,2,3,3-Tetrachloro-4,5-dimethylspiro[2,3]hexa-1,4-diene molecules contain spiro-linked cyclobutene and cyclopropene rings (12). ’ The double bonds are both localized and in fact are a little shorter than normal (1.302 and 1.31 1 A).However all the bond lengths in the ring are slightly shorter than in related compounds and it is suggested that the rings contain some degree of aromaticity. Other highly strained molecules which contain cyclobutyl rings are the photodimer (14) of ‘basketene’ (13) in which the C-C bond lengths are’ 1.52-1.57 8 and 2,3,6,7,7,8-hexamethyl-( **-1,5-diphenyl)tetracyclo[3,3,0,0’ 03*6]octan-4-one (15).14 In this molecule the C-3-C-6 and C-6-C-5 bonds are unusually long (1.602 and 1.608A)for a small ring and it is one of these bonds which shifts in the rearrangement reactions which this type of ring system under- goes.Another feature of interest in this structure is that the C-4 atom is only 2.145 8 away from the mid-point of the bond C-1-C-2 compared with 2.33 8 in related compounds. This also is believed to be significant in the rearrangement mechanisms. L. T. J. Delbaere M. N. G. James and R. Lemieux J. Amer. Chem. SOC.,1973,95,7866. ** R.J. Guttormson and B. E. Robertson Acta Cryst. 1973 B29 173. N.J. Jones W. D. Deadman and E. LeGoff Terrahedron Letters 1973 2087. l4 C.G.Biefeld H. A. Eick and H. Hart Tetrahedron Letters 1973,4507. M.B. Hursthouse and S.Neidle The effects on bond lengths and angles of molecular overcrowding are also the subject of many investigations. A redetermination of the structure of octachloro- pentafulvene (16) shows that the central C=C bond length 1.365A is not c1 CI strongly affected by the strain arising from theCl- -C1 repulsions this beingrelieved by the twist of 37" about the central bond." Another molecule showing over- crowding is 2-methylhexahelicene.' A number of unusually short C-C bonds are found on the periphery of the helix whilst those in the helix core are lengthened a pattern found in other fused aromatics. The medial rings are also significantly distorted into a boat conformation. A boat distortion of a benzene ring is also found in the molecule of 4-~arboxy[8]paracyclophane(17).' In the molecular structure of hexa-o-phenylene (18) the benzene rings are slightly distorted in a twist form.'' Within the twelve-membered ring all the angles are enlarged to 129.3' because of intramolecular repulsion.A neutron diffraction study of the ten-membered ring compound cyclodecane- 1,6-trms-diol has provided ex-perimental verification of short transannular H--H contacts (ca. 1.95A) in the ring.Ig These interactions cause significant deviations of the methylene groups involved from local CZvsymmetry. In the fifteen-membered ring compound cyclopentasarcosyl(19) the ring contains a succession of three cis amide groups and another succession of two trans amide groups.2o Three of the groups deviate significantly from planarity.Variable temperature n.m.r. studies indicate that the conformation found in the crystal is also predominant in solution. (18) H. L. Ammon G. L. Wheeler and I. Agranat Tetruhedron 1973,29 2695. I6 G. W. Frank D. T. Hefelfinger and D. A. Lightner Actu Cryst. 1973 B29 223. I7 M.G. Newton T. J. Walter and N. L. Allinger J. Amer. Chem. SOC.,1973 95 5652. H. Irngartinger Actu Cryst. 1973 B29 894. l9 0.Ermer J. D. Dunitz and I. Bernal Actu Cryst. 1973 B29 2278. *O K.Titlestad P. Groth and J. Dale J.C.S. Chem. Comm. 1973,646. Physical Methods-Part (iii) X-Ray Crystallography 4 Heterocyclic Systems Perhydropyrido-oxazepines (20)are formed from condensations of substituted 3-(2-piperidyl)propan-l-olswith formaldehyde.trans-Isomers have been found to undergo an unusual dimerization on crystallization from the liquid state which is reversed on dissolution.21 The dimer (22)contains a fourteen-membered ring and its formation is believed to involve an intermediate such as (21).Another novel heterocyclic system (25)involving a bridged ring has been formed by the reaction of the hindered diketo-sulphide (23) with hydrazine.22 Compound (24) is stable in the crystal but in benzene in the presence of oxygen it undergoes a remarkable transformation to (25). A number of other sulphur-containing heterocyclics have been studied. The S-0 distances in (26)are very short 1.875 and 1.878A,23 and are equivalent to the Se-0 distances in the analogous 21 D. A. Whiting R. Cahill and T.A. Crabb I.C.S. Chem. Comm. 1972 1307. '' E. Cuthbertson A. D. U. Hardy and D. D. MacNicol I.C.S. Chem. Comm. 1973,597 23 E. C. Llaguno and I. C. Paul Tetrahedron Letters 1973 1565. M. B. Hursthouse and S. Neidle selenium compound. The interactions are somewhat greater than similar systems where the N atoms are replaced by carbon atoms. The reaction of tetramethyl- cyclobutane-1,3-dione with P,S produces a number of sulphur-containing species. One of the minor products for which the crystal-structure analysis has been reported is the spiro-compound (27).," Another report provides the first detailed structural information on a three-membered ring composed entirely of heteroatoms which is present in the molecule bis-( I ,1,3,3-tetramethylbutyI)-thiadiaziridine 1,l-dioxide (28).25 The S-N bond is quite short (1.62A) but the N-N bond is longer than in any analogous distance.A bridging SO group is also found in the compound (29) formed by reaction of endo-2-chloro-sulphane with potassium t-butoxide in the presence of lY3-diphenylisobenzo- furan.2 Ph I 5 Reaction Mechanisms and Products In this section a few examples are given of the use of X-ray analysis to characterize a reaction product (structure and stereochemistry) or as an aid to the interpreta- tion of the mechanism of a particular reaction. Thus the analysis of the product of bromination of semibullvalene (30) shows that stereoselective cis,exo-l,4 addition has o~curred,~ in contrast to bullvalene itself where kinetically controlled trans-1,4addition takes place.Similarly analysis of the product from the cycloaddition of 2-ethyl-3-methylisoquinolinium perchlorate and cyclopentadiene shows that the reaction is also stereospecific.28 24 C. D. Shirrell and D. E. Williams Acta Crysf. 1973 B29,2128. 25 L. M. Trefonas and L. D. Cheung J. Amer. Chem. SOC.,1973,95 636. 26 C. B. Quinn J. R. Wiseman and J. C. Calabrese J. Amer. Chem. SOC.,1973. 95 6121. 2' L. A. Paquette G. H. Birnberg J. Clardy and B. Parkinson J.C.S. Chem. Comm. 1973 129. C. K. Bradsher F. H. Day A. T. McPhail and P. S. Wong J.C.S. Chem. Comm. 1973 156. 53 Physical Methods-Part (iii)X-Ray Crystallography Direct photolysis of 5-acetyl-3,3-dimethyl-3H-pyrazole (31) yields 7-acetyl- 3,5,5,9,9-pentamethy1-1,6-diazabicyclo[4,3,O]nona-3,7-dien-2-one (32).*' The re- action is believed to occur via formation of a vinylketen which then reacts further 0 with the starting material.The irradiation of exo-3-aza-4-ketobenzotricyclo-[4,2,1,02-5]non-7-ene (33) in methanol gives exo-2-methoxy-3-aza-4-keto-7,8-benzobicyclo[4,2,1]nonene (34) the reaction involving ring-expansion of the fi-lactam moiety in (33) and the incorporation of a molecule of rnethan01.~' The structure of the product was determined by use of n.m.r. lanthanide shift reagents and confirmed by X-ray analysis. (33) An example of the use of X-ray analysis to study a solid-state photochemical reaction is provided by work on 1,l'-trimethylenebisthymine (35).3 This molecule forms a compound (36) intramolecularly and the structure adopted by (35) does not lead immediately to an explanation for the geometry of (36) which is cis-syn.29 A. C. Day A. N. McDonald B. F. Anderson T. J. Bartezak and 0. J. R. Hodder J.C.S. Chem. Comm. 1973 247. 30 H. L. Ammon P. H. Mazzocchi W. J. Kopecky jun. H. J. Tarnburin and P. H. Watts J. Amer. Chem. SOC.,1973 95 1968. 31 J. K. Frank and I. C. Paul J. Amer. Chem. SOC.,1973 95 2324. M. B. Hursthouse and S. Neidle The great enhancement of the reaction rate of a substrate promoted by an enzyme suggests that the latter may impose restrictions on the conformation of the former to produce a favourable geometry. Similar effects may occur if 0 0 PJ0 0 cffo Q0 0 0X-NH the conformation of a compound is restricted by ‘over-methylation’.Such a situation occurs with the reaction rate of lactonization in o-hydroxyhydro- cinnamic acid when methyl groups are substituted both on the ring and the side- chain and the structure analysis of the over-methylated lactone pentamethyl- hydrocoumarin and the alcohol analogue of pentamethyl-o-hydroxyhydro-cinnamic acid shows that the conformation of the acid has been restricted by the presence of a ‘trialkyl lock’ similar to that in the la~tone.~~ Finally the structure analysis of the pyrolysis product (38) of ethylene-1,l- bis(triphenylphosphonium)-2,2-bis(phenylamide) (37) shows that an unprece-dented phenyl migration has occurred from a four-co-ordinate phosphorus to a two-co-ordinate nitrogen atom.33 Ph,P \/ $Ph p\c/ PPh I I C c NH \N-Ny ‘N-Ph I I I I Ph Ph Ph Ph 6 Natural Products and Biologically Active Molecules Many natural products studied have absolute configurational assignments made on the basis of Bijvoet anomalous scattering differences and these often serve to provide key assignments in whole families of compounds.A disquieting report34 that a rigorous exciton analysis of the c.d. spectra of 1,s-disubstituted 32 J. M. Karle and I. L. Karle J. Amer. Chem. Soc. 1972 94 9182. 3’ F. K. Ross Lj. Manojlovic-Muir W. C. Hamilton F. Ramirez and J. F. Pilot J. Amer. Chem. SCC.,1972,94 8738. 34 J. Tanaka C. Katayama F. Ogura H. Tatemitsu. and M. Nakagawa J.C.S. Chem. Comm. 1973 21. Physical Methods-Part (iii) X-Ray Crystallography 55 9,lO-bridged 9,lO-dihydroanthracenes does not agree with their Bijvoet X-ray assignments suggested that the whole basis of absolute configurational deter- minations by X-ray methods might be in error with serious implications for much of organic chemistry.However it has been pointed o~t~~.~~ that flaws in the exciton analysis make the above conclusions no longer tenable and that there is no discrepancy between the X-ray and the c.d. method. A number of sesquiterpenoids have been examined. The y-lactone ring in parthemollin (39) of previously uncertain stereochemistry has been found to be cis ;this has established the complete stereochemistry or the related xanthano- lides. The stereochemistry of the biological divinyloxiran-dihydro-oxepin rearrangement has been defined with reference to the structures of miscandenin (40)and dihydromikanolide (41).38 The trans double bond in (41) has a torsion angle of -163" a considerable distortion from ideality.Other germacranolides reported include enh~drin,~ with a cis,trans configuration of double bond and epoxide ring relative to the ten-membered macrocycle and the potent cytotoxic agent molephantin (42).40 Molephantin has a novel unsaturation pattern and 0 (39) 0 is the first germacranolide found to contain a dienone ring system. The structure of euparotin4' has been determined and the conformation of the cycloheptane 35 S. F. Mason J.C.S. Chem. Cornrn. 1973 239. 36 A. M. F. Hezemans and M. P. Groenewege Tetrahedron 1973,29 1223.'' P. Sundararaman R. S. McEwen and W. Herz Tetrahedron Letters 1973 3809. '' P. J. Cox G. A. Sim J. S. Roberts and W. Herz J.C.S. Chem. Cornrn. 1973,428. 39 G. Kartha K. T. Go and B. S. Joshi J.C.S. Chem. Cornrn. 1972 1327. 40 K. H. Lee H. Furukawa M. Kosuka H. C. Huang P. A. Luhan and A. T. McPhail J.C.S. Chem. Comm. 1973 476. A. T. McPhail and G. A. Sim Tetrahedron 1973,29 1751. 56 M. B. Hursthouse and S. Neidle ring in this and other perhydroazulene sesquiterpenes has been surveyed; in general the ring can be described as closer to a twist chair (C,) than a chair (C,) form. (-)-Arktolone (43)42 has a fused cyclopropane ring adjacent to the car- bony1 group; this results in a hyperconjugative effect between the two groups in spite of the geometry here deviating from ideality.For maximum interaction it is considered that the plane containing the carbonyl group and the adjacent ring carbons should bisect the three-membered ring. In spite of the carbonyl oxygen deviating by 29.6" from the ideal orientation the conjugation still shortens the linking C(sp3)-C(sp2)bond to 1.436 A. (+)-2,5-Diepi-fi-cedrene (44)43has the two five-membered rings trans-fused whereas in P-cedrene itself they are cis. This has induced considerable internal strain. The diterpene alcohol pachydictyol A (49 isolated from a species of marine algae also possesses the perhydroazulene ring previously unknown among diterpene~."~It has been suggested that pachydictyol A has been derived biogenetically as a sesquiterpene to which an isoprene unit has been added.Amongst higher terpenoids investigated is the diterpene barbatusin (46)"' which has the unusual feature of a methylcyclopropyl ring at C-13. Ring c is qo (43) H' (44) (45) described as in a half-chair conformation with considerable distortions from planarity about the conjugated 1,4-diketone group ; torsion angles here are as 42 F. H. Allen 0.Kennard and J. Trotter Acta Cryst. 1973 B29 1451. 43 B. Karlsson A. M. Pilotti and A. C. Wiehager Acru Cryst. 1973 B29 1710. 44 D. R. Hirschfeld W. Fenical G. H. Y. Lin R. M. Wing P. Radlick and J. J. Sims J. Amer. Chem. SOC.,1973 95 4049. 43 A. H. J. Wang I. C. Paul R. Zelnik I(.Mizuta and D. Lavie J. Amer.Chem. SOC. 1973 95 598. Physical Methods-Par t (iii) X-Ray Crystallography 57 large as 32.4". In view of the reactivity of the cyclopropane molecule barbatusin may be considered to be an intermediate in the biosynthesis of some naturally occurring quinoid diterpenes. New skeletal classes of triterpenes continue to be reported ; baccharis oxide (47)46has an oxide bridge between C-3 and C-10 in ring A. Short-range steric strain due to close angular substituents as well as ring-junction strain is con- sidered to be responsible for considerable deviations from ideality of many of the bond lengths and angles. Repulsions between 1,3-diaxial substituents are also concluded to be responsible for the pronounced flattening and twisting observed for the trans-fused rings c and D.The crystal structure of the physiologically important hormonal steroid cortisol$' as its methano solvate has been established and the structure com- pared4' with five other corticosteroids of known conformation in order to attempt a correlation between conformational differences and biological activity. It was noted that the degree of bowing towards the a-face is related to the varia- tion in anti-inflammatory activity in these compounds. The ring-D-bridged steroid 14a 17a-etheno-15,16-di(trifluoromethyl)pregna-4,15-diene-3,20-dione (48)"' has a conformation in which ring A is said to be intermediate between a half-chair and a sofa and the geometry of the bicyclic system of ring D is very similar to that of norbornadiene.Other steroidal derivatives analysed include (23R)-23-hydroxy-3a,5a-cycloergost-7-en-6-one (49)," with a cis A/B ring junction and the 17-spirobicyclomethyI-steroidNic 11 (50).' (47) @H ..-c OH **(j 46 F. Mo Acta Cryst. 1973 B29 1796. 4' P. J. Roberts J. Coppola N. W. Isaacs and 0. Kennard J.C.S. Perkin II 1973 774. 48 C. M. Weeks W. L. Duax and M. E. Wolff J. Amer. Chem. SOC.,1973,95 2865. 49 G. I. Birnbaum Acta Cryst. 1973 B29 54. M. B. Hursthouse and S. Neidle J.C.S. Perkin II 1973 781. 51 M. Begley L. Crombie P. J. Ham and D. A. Whiting J.C.S. Chem. Comm. 1973,821. M. B. Hursthouse and S. Neidle Among the many alkaloids investigated have been a number of novel structures. The anticholinergic agent dihydroisohistrionicotoxin (5 1y2 has the unique combination of a spiro-alkaloid moiety and two unsaturated side-chains one with an allene group and the other a vinylacetylene.The cis-butenynyl chain is quite planar and the C-13-C-14 single bond is of length 1.43A. The allene moiety deviates slightly from linearity. The neuromuscular blocking compound d-tubocuranine (52),53 which is a curare alkaloid has been found to have only one quaternary nitrogen atom. The molecular conformation is thought to be consistent with a one-site attach- ment to a receptor site thus blocking the possible binding of acetylcholine molecules. The large hydrophobic grouping would appear to be an effective block. 12s-Tetrahydroaustamide (53)54 and cytochalasin E (54)" are both fungal metabolites.The former has the diketopiperazine ring in a boat con- formation. -3C I1 CH Several studies have served to interrelate various members of alkaloid families. Thus the knowledge of the structure and absolute configuration of (+)-coron- aridine (55)56 has enabled configurational assignments to be made for a number '* I. L. Karle J. Amer. Chem. SOC.,1973 95 4036. 53 P. W. Codding and M. N. G. James Acta Cryst. 1973 B29 935. " J. Coetzer and P. S. Steyn Acta Cryst. 1973 B29 685. " G. Biichi Y. Kitaura S. S. Yuan H. E. Wright J. Clardy A. L. Demain T. Glinsukon N. Hunt and G. N. Wogan J. Amer. Chem. SOC.,1973,95 5423. " J. P. Kutney K. Fuji A. M. Treasurywala,J. Fayos J. Clardy A. I. Scott and C. C. Wei J. Amer. Chem. SOC.,1973 95 5407. Physical Methods-Part (iii) X-Ray Crystallography 59 of the Iboga alkaloids.Lythrumine (56)57 has an unusual skeleton which has now been found in lythraceae alkaloids from both North America and Japan. Among the sugar analyses reported are those of 2,6-anhydro-/?-~-fructo-furanose (57)58 and 1,2:4,5-di-0-isopropylidene-/?-~-fructopyranose (58).59 The furanosyl ring in (57)has an unusual conformation with both C-3 and C-4 in the exo position in contrast to other furanosyl rings investigated to date presumably because of the effect of the bicyclic ring system. The ring is said to be in a distinct envelope conformation as is the anhydride ring. The pyranosyl ring in (58) has a chair-distorted half-chair conformation as a result of the fusion of the five- and six-membered rings and each 1,3-dioxolan ring has an envelope conformation.(55) HO' HOw A large number of biologically active molecules have been examined with a view to obtaining some insight into either structural requirements for activity or into the actual mechanism of action of these compounds. Several thyroid hormones have been analysed in an attempt to establish the structural require- ments for maximum hormonal activity. 3'-Isopropyl-3,5-di-iodo-~-thyronine (59),60the most potent known thyromimetic agent has a conformation such that the j3-ring is oriented with the isopropyl group proximal to the a-ring. This is a very similar situation to that observed in the parent compound tri-iodothyro- nine in which iodine replaces the isopropyl moiety and it is suggested that this conformation is the active one in physiological solution.A study of the tri- iodothyroacetic acid N-diethanolamine complex6 ' has however indicated that the distal conformation may be of importance. 57 H. Wright J. Clardy and J. P. Ferris J. Amer. Chem. Soc. 1973,95,6467. 58 W.Dreissig and P. Luger Acfa Crysr. 1973 B29 1409. '' S.Takagi R. Shiono and R. D. Rosenstein Acfa Crysf. 1973 B29 1177. 6o J. K.Fawcett N. Camerman and A. C. Camerman Biochem. Biophys. Res. Comm. 1973,52,407. 6' V. Cody and W. L. Duax Biochem. Biophys. Res. Comm. 1973,52,430. M.B. Hursthouse and S. Neidle Other pharmacologically interesting analyses include that of a novel ring- expansion product (60)62of penicillin V P-sulphoxide and that of tirandamycic acid (61),63 which has enabled the acyltetramic antibiotics tirandamycin and streptolydigin to have complete stereochemical assignments.Methadone (62),64a potent narcotic analgesic has a large degree of conformational stability to and the structure found is characterized by a folding of the carbon chain so as to bring the basic nitrogen into close contact with the carbonyl carbon atom and it is suggested that methadone is effectively transported across the blood- brain barrier because of the hydrophobic nature of the surface of the molecule as revealed in this study. Extensive studies have continued on nucleic acid and protein constituents and their various analogues. 6-Azauridine (63)65 has like the parent uridine an anti conformation about the glycosidic bond and a C-3’-endo sugar pucker but differs considerably in the detailed arrangement of substituent stereochemistry 62 R.Thomas and D. J. Williams J.C.S. Chem. Comm. 1973,226. 63 D. J. Duchamp A. R. Branfman A. C. Button and K. L. Rinehart jun. J. Amer. Chem. SOC.,1973,95,4077. 64 H. B. Burgi J. D. Dunitz and E. Shefter Nature New Biol. 1973 244 186. 65 C. H. Schwalbe and W. Saenger J. Mol. Biol. 1973,75 129. Physical Methods-Part (iii ) X-Ray Crystallography 61 in the sugar ring. The anti-parallel double-helical arrangement of base-pairs in nucleic acids has been observed for the first time in two dinucleoside phosphates adenosyL3’,5’-uridine phosphate66 and guanylyL3’,5’-cytidine ph~sphate.~~ HO’ OH (63) These structures provide detailed information about the precise geometry of both the ribose phosphate linkage and the complementary base-pairing in the genetic code.The C-terminal tripeptide L-Pro-L-Leu-Gly-NH,68 of the hormone oxytocin has been found to have a compact intramolecularly hydrogen-bonded structure in the solid state whose basic feature a ten-membered p-turn structure has previously been postulated as being present in solution on the basis of n.m.r. evidence. One of the peptide bonds is significantly non-planar with a torsion angle of 10”. “ J. M. Rosenberg N. C. Seeman J. J. P. Kim F. L. Suddath H. B. Nicholas and A. Rich Nature 1973 243 150. ‘’ R. 0.Day N. C. Seeman J. M. Rosenberg and A. Rich Proc. Nut.Acad. Sci. U.S.A. 1973 70 849. L. L. Reed and P. L. Johnson J. Amer. Chem. SOC.,1973,95 7523.