摘要:

L e t t e r O O O O Ti Ti OR OR OR OR Br Br B(OH)2 OMe OMe K4 3 K4[L3Ti2] 2 + R = Me (L-Me4) R = H (L-H4) i ii iii Self-assembly of a triple-stranded helicate from a rigid di(catechol) ligand and formation of its dimer in the solid state Markus Albrecht,a* Matthias Schneidera and Roland Froé hlichb a Institut Organische Chemie, Karlsruhe, D-76131 f ué r Universitaé t Richard-W illstaé tter-Allee, Karlsruhe, Germany b Organisch-Chemisches-Institut, 40, D-48149 Universitaé t Mué nster, Corrensstraêe Mué nster, Germany The p-phenylene bridged di(catechol) ligand forms a triple-stranded dinuclear titanium helicate which in L-H4 the solid state incorporates potassium cations and forms a dimeric species. Triple-stranded helicates are formed by self-assembly processes from two (or more) metal ions that prefer an octahedral coordination environment and three linear ligands with two (or more) bidentate ligand units.1 Recently we described the self-assembly of cryptand-type helicates and meso-helicates from alkyl-bridged di- and tri-(catechol) ligands and titanium(IV) ions.Owing to the —exibility of the spacer, the cavity of the metalla-cryptand can adopt diÜerent sizes and thus shows only low selectivity towards binding of diÜerent cations.2 In this paper we describe a rigid p-phenylene bridged di(catechol) ligand which is used for the formation of L-H4 , dinuclear helicate-type complexes with a de–ned cavity size.A rigid di(bipyridine) ligand has already been used by Lehn and co-workers for the formation of a double stranded helicate, 3 and Raymond and co-workers described the triplestranded helicate of a rigid p-phenylene bridged di(catechol) which additionally bears two amide linkages in the spacer.The amide groups allow some —exibility of the system and prevent the binding of counter ions in the interior of the helicate. 4 The ligand was synthesized by Suzuki coupling of 2,3- L-H4 dimethoxyphenylboronic acid5 (2 equiv.) with 1,4-dibromobenzene (1 equiv.) followed by cleavage of the methyl ethers with The coordination compound was pre- BBr3.K4[L3Ti2] pared by dissolving (3 equiv.), (2 equiv.) and L-H4 Ti(OMe)4 (2 equiv.) in methanol. Overnight a red solution was K2CO3 formed. Solvent was evaporated and the residue was dried in vacuum. The 1H NMR spectrum (methanol- of the red d4) compound has signals at d\6.61 (dd, J\1.4, 7.7 Hz, 6 H), 6.50 (pseudo t, J\7.7 Hz, 6 H), and 6.41 (dd, J\1.4, 7.7 Hz, Scheme 1 i, ii, iii, (Ph3P)4Pd; BBr3; Ti(OMe)4, K2CO3 * E-mail: albrecht=ochhades.chemie.uni-karlsruhe.de 6 H) for the ligand units and a singlet at d\7.60 (12 H) for the three spacer p-phenylene groups.Corresponding 13C NMR signals (methanol- are observed at d\160.7 (C), d4) 158.3 (C), 139.3 (C), 129.7 (CH), 125.5 (C), 118.5 (CH), 118.1 (CH), and 111.6 (CH).However, the structure of the complex (e.g., helicate versus meso-helicate) can not be assigned by spectroscopic methods in solution. Therefore we crystallized from dmf»ether to obtain red crystals of composi- K4[L3Ti2] tion methanol which could be K4[L3Ti2] … 8 dmf …H2O… 0.5 investigated by X-ray crystallography.§ In the solid state the tetraanion [Fig. 1(a)] [L3Ti2]4~ adopts a helical structure with a TiwTi separation of 8.404 ”. The rigidity of the ligand L allows no strong turn of the helicate. Thus, the estimated pitch of the helix is ca. 21 nm.1 Owing to diÜerent orientations of the p-phenylene units of the spacers the complex does not possess in the C3-symmetry crystal.The three ligands L and two titanium(IV) ions form a twisted cylinder with a large cavity in its interior. Some of the counter ions are located within this cylinder and these additionally bind to dmf or water molecules. Two dinuclear complexes which possess opposite helicity form a dimer [L3Ti2]4~ in the solid state [Fig. 1(b), Fig. 2]. Just recently the formation of a meso-helicate by dimerization of two oppositely con–gurated dinuclear helicates in the solid state was observed by Constable et al.6 Two potassium cations are bound in the interior of each of the helicates.7 One of the cations (fourcoordinate) coordinates to an internal oxygen atom of one ligand, L, and additionally is bound to two dmf molecules and one water.The water probably is –xed by hydrogen bonding § X-Ray crystal structure analysis of dmf … water … 0.5 K4[L3Ti2] … 8 methanol.Formula C54H30K4O12Ti2…8 C3H7NO…H2O… 0.5 CH3OH, M\1741.79, 0.25]0.20]0.10 mm, a\13.124(1), b\14.790(2), c\23.042(6) a\85.68(2), b\81.42(1), c\73.21(1)°, ”, V \4231.5(13) g cm~3, l\4.60 cm~1, empirical ”3, qcalc.\1.367 absorption correction via u scan data (0.986OCO0.999), Z\2, triclinic, space group (No. 2), k\0.71073 T \223 K, x/2h scans, P1 6 , ”, 11 619 re—ections collected (^h, ]k, ^l), [(sinh)/k]\0.54 ”~1, 11 037 independent and 4205 observed re—ections [IP2 r(I)], 1022 re–ned parameters, R\0.070, wR2\0.158, max. residual electron density 0.79 ([0.75) e hydrogens are calculated and re–ned as ”~3, riding atoms, O141 to C145 re–ned with common O71 to C75 Uiso , taken as a model for SAME, S.O.F.for O161 and C162 0.5, group not re–ned, hydrogen atoms at O151 not found. Data set was collected with an Enraf-Nonius MACH3 diÜractometer. Programs used: data reduction MolEN, structure solution SHELXS-97, structure re–nement SHELXL-97, graphics SCHAKAL- 92. Crystallographic data (excluding structure factors) for the structure reported in this paper have been deposited with the Cambridge Crystallographic Data Centre as supplementary publication CCDC 44/041.New J. Chem., 1998, Pages 753»754 753K O N O N K O O K K O N O N O O N N Fig. 1 Molecular structure of the helicate (a) and of the dimer (b) in the solid state (hydrogens are ML3Ti2N4~ [ML3Ti2NK2(dmf)3(H2O)]4~ omitted) to two internal oxygen atoms of the titana-cryptand and bridges to the second potassium ion which is coordinated to two ligand oxygen atoms.This –ve-coordinated potassium ion forms a four-membered ring with a symmetry equivalent potassium cation by two bridging dmf molecules. Thus a dimer is formed in the solid state. Additionally, four potassium cations (not shown) are exohedrally bound to the helicate and are saturated by further dmf molecules.In this paper we presented the self-assembly of a cylindrical cryptand-type helicate from a p-phenylene bridged di(catechol) ligand and titanium(IV) ions. The tetraanion possesses a well-de–ned cavity size in which pot- [L3Ti2]4~ assium cations can be incorporated ; in the solid state the potassium ions form a dimer with dmf bridges. At the moment we are preparing further rigid di(catechol) ligands to vary the size Fig. 2 Schematic representation of the dimer as found in the solid state. The two dinu- [ML3Ti2NK2(dmf)3(H2O)]4~ clear helicates only are indicated ML3Ti2N4~ of the cavities of the self-assembled dinuclear coordination compounds. Acknowledgements work was supported by the Deutsche Forschungsgemein- This schaft and the Fonds der Chemischen Industrie.References 1 C. Piguet, G. Bernardinelli and G. Hopfgartner, Chem. Rev., 1997, 97, 2005; M. Albrecht, Chem. Soc. Rev., in the press. 2 M. Albrecht and S. Kotila, Angew. Chem., 1995, 107, 2285; Angew. Chem., Int. Ed. Engl., 1995, 34, 2134; M. Albrecht and S. Kotila, Angew. Chem., 1996, 108, 1299; Angew. Chem., Int. Ed. Engl., 1996, 35, 1208; M. Albrecht and M. Schneider, Chem., Commun., 1998, 137. 3 P. N. W. Baxter, J.-M. Lehn and K. Rissanen, Chem. Commun., 1997, 1323. 4 B. Kersting, M. Meyer, R. E. Powers and K. N. Raymond, J. Am. Chem. Soc., 1996, 118, 7221; M. Meyer, B. Kersting, R. E. Powers and K. N. Raymond, Inorg. Chem., 1997, 36, 5179. 5 I. Preç vot-Halter, T. J. Smith and J. Weiss, J. Org. Chem., 1997, 62, 2186. 6 E. C. Constable, M. Neuburger, L. A. Whall and M. Zehnder, New J. Chem., 1998, 22, 219. 7 See for comparison: M. Albrecht, H. Roé ttele and P. Burger, Chem. Eur. J., 1996, 2, 1264. Received in Basel, Switzerland, 17th April 1998; L etter 8/03652G 754 New J. Chem., 1998, Pages 753»754

ISSN:1144-0546    

DOI:10.1039/a803652g

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

L e t t e r CwHÆ Æ ÆO Hydrogen bonds in the mixed-valence salt [ (g6- and the breakdown of the C6H6)2Cr]ë[CrO3(OCH3) ]ó length/strength analogy Dario Braga,*,a Fabrizia Grepioni,a Emilio Tagliavini,a Juan J. Novoa*,b and Fernando Motaa a Dipartimento di Chimica G. Ciamician, di Bologna, V ia Selmi 2, 40126 Bologna, Italy Universitaç b Departament de Facultat de Universitat de Barcelona, Av.Diagonal Quïçmica Fïç sica, Quïç mica, 647, 08028 Barcelona, Spain The theoretical and experimental study of the ion organisation in crystalline [(g6-C6H6)2Cr]`[CrO3(OCH3)]~ aÜords a complete picture of the relative contribution to crystal cohesion of Coulombic interactions, p-stacks, and charge assisted CwH… … …O([) hydrogen bonds while showing the repulsive nature of short CwH… … …O contacts between anions.Crystal engineering is a booming –eld of research encompassing all traditional chemistry subdivisions.1 The principal noncovalent interaction in crystal engineering, as well as in supramolecular chemistry, is the hydrogen bond since it combines strength with directionality.2 We have recently shown that organic»organometallic crystalline materials can be designed and synthesised by exploiting the coexistence of OwH… … …O and CwH… … …O hydrogen bonds.These latter weaker bonds can be reinforced (ìcharge assisted œ) if the donor groups belong to a cation and the acceptors to an anion.3 Here we report the structural characterization,§ and theoretical evaluation of the mixed-valence crystalline salt [(g6- 1 obtained from the crys- C6H6)2Cr]`[CrO3(OCH3)]~ tallisation of the hydroxide [(g6-C6H6)2Cr]`[OH]~… 3H2O reported earlier.5 Compound 1 possesses some peculiar structural features.(i) The crystal is formed of columns of methoxychromate anions and of columns of paramagnetic bis-benzene chromium cations [Fig. 1(a)]. (ii) The cations stack in piles with benzene»benzene distances of ca. 3.50 (iii) The inter- ”.action between cations and anions is based on ìchargeassisted œ CwH… … …O([) hydrogen bonds [twelve H… … …O§ distances in the range 2.40»2.60 four shorter than 2.45 ”, ”, see Fig. 1(b)].6 (iv) The anions are apparently ìlinkedœ along the column via a short CwH… … …O interaction (2.381 ”, CwH… … …O angle 173°) between a methyl hydrogen and a chromate oxygen [see Fig. 1(c)] Such a distance might commonly be taken to be indicative of a relatively strong hydrogen bond.§ Crystals of 1 were obtained as a minor product (ca. 10%) from attempts to crystallise [(g6- from MeOH.4 Crystal C6H6)2Cr][OH] data for [(g6-C6H6)2Cr]`[CrO3(OCH3)]~: C13H15Cr2O4 , T \273(2) K, M\339.25, triclinic, a\10.060(5), b\10.510(7), P16 , c\6.818(2) a\105.82(3), b\109.12(3), c\82.37(4)°, ”, U\654.6(6) Z\2, g cm~3, F(000)\346, l\1.665 ”3, dc\1.721 mm~1, h-range 3.0»25°, 2504 re—ections measured, 2289 of which independent, re–nement on F2 for 217 parameters, wR (F2, all re—s.)\0.1597, [I[2p(I)]\0.0564.MoKa radiation, R1 k\0.71069 monochromator graphite, psi-scan absorption correc- ”, tion. All non-H atoms were re–ned anisotropically. All H atoms were directly located from Fourier maps.The computer programs SHELX864a and SHEXL924b were used for structure solution and re–nement. The computer program SCHAKAL92 was used for all graphical representations.4c In order to evaluate the CwH… … …O bonds that CwH distances were normalised to the neutron-derived value of 1.08 and the program PLATON was used.4d ” CCDC reference number 440/038. Besides the intrinsic interest in the isolation of this –rst example of a mixed-valence methoxychromate salt, the presence within the same crystal architecture of CwH… … …O separations between anions and cations longer than between anions and anions calls for an explanation.To this end we have investigated the crystal packing and ion organisation by means of ab initio unrestricted Hartree»Fock (UHF) calculations7 using a LANL2DZ basis set, of double zeta for the valence electrons, which are core electrons described by the LANL2 eÜective potentials.7 In a –rst approximation, the crystal interaction energy can be expressed as the sum of pairs [Ep\&E(i,j)] where E(i,j) is the intermolecular interaction energy between the i and j atoms or group of atoms.Clearly, the minimum energy arrangement of the molecules or ions in the crystals is a compromise among all available pair energies, that is, a minimum in Ep does not necessarily correspond to a minimum for each E(i,j).Hence, repulsive contacts between atoms i and j may be ìtoleratedœ to some extent if the combined energy of other interactions in the crystal is larger than the repulsive E(i,j) value.This is precisely what happens in 1 as we have been able to demonstrate by means of crystal-packing functionalgroup analysis8 which consists in the energetic analysis of the primary packing patterns (PPP).î PPP are constituted of two or more nearest neighbour molecules within the crystals and are identi–ed by looking at the molecular arrangements which allow overlap of the positive and negative regions of the molecular electrostatic (MEP) maps. The MEP of the anion in 1, computed at [CrO3(OCH3)]~ the HF/LANL2DZ level, is shown in Fig. 2(a). The potential is negative everywhere in the space around the anion,î therefore the anion should not form stable interanionic interactions of the type observed in the crystal. The interaction energy of î Stable crystals are associated with stable PPP.These are formed when the orientation of the functional groups present in the molecules allow stable intermolecular contacts (in hydrogen bonds, for instance, this means that the molecules are oriented in such a way that the acid and basic groups are at short distances and at the right orientation). By identifying the stable PPP one can rationalise the crystal, de–ne all major contributions to crystal packing, and pinpoint the origin of crystal stability.The analysis of the overlap of the MEP maps of neighbouring molecules is more powerful than a simple look at the positive and negative regions of charge localisation typically obtained from a Mulliken population analysis, as illustrated here. New J. Chem., 1998, Pages 755»757 755Fig. 1 (a) Space –lling representation of the ion organization in crystalline [(g6- note how cations and C6H6)2Cr]`[CrO3(OCH3)]~: anions form parallel columns.(b) Network of CwH… … …O hydrogen bonds between anions and cations (broken lines, oxygen atoms represented as –lled spheres). (c) Schematic representation of the columns of [(g6- cations and of anions: the –lled C6H6)2Cr]` [CrO3(OCH3)]~ broken lines mark the short CwH… … …O contact (2.381 note also ”), the bending of the benzene CwH groups towards the O atoms.Some relevant structural parameters: CrwOMe 1.795(5), CrwO 1.596(5), 1.609(5), 1.592(4) OwMe 1.460(8) CrwC 2.127(6)»2.147(6) ”, CrwOwMe 120.6(4)° two anions computed at the HF/LANL2DZ [CrO3(OCH3)]~ level with the observed geometry is repulsive by]58 kcal mol~1.The MEP map in the region of the methyl hydrogens is also negative, even though a Mulliken population analysis indicates positively charged hydrogens (]0.2 e) and a charge of [0.2 e on the methyl carbon. Based on these charges the methyl should be electrostatically attracted by the nearest oxygen of the chromate anion (the charges on chromium, and on the bridging and terminal oxygens are ]1.8, [0.9, and [0.8 e, respectively).However, the electrostatic energy computed taking into account these point charges remains repulsive (]40 kcal mol~1) mainly because of the repulsive interaction (]82 kcal mol~1) between the subunits of CrO4 the two moieties each bearing a combined [CrO3(OCH3)]~ charge of [1.4 e. This result warns, inter alia, against excessive simpli–cation when dealing with electrostatic interactions.Thus far we have shown that CwH… … …O interactions along the columns are not attractive, and yet why [CrO3(OCH3)]~ are they so short ? The model that best rationalises the observed structure is one in which the crystal is seen as ionic, similar in many ways to the NaCl crystal.° The anions and [(g6- cations aggre- [CrO3(OCH3)]~ C6H6)2Cr]` gate in layers [see Fig. 1(a)] : as in the NaCl case, one can distinguish repeating units. However in 1, the A([)2C(])2 layer stack generates A([)………A([) and C(])………C(]) contacts instead of A([)… … …C(]), as in the NaCl case. Although A([)………([) and C(])… … …C(]) interactions, calculated at the HF/LANL2DZ level, are repulsive along the columns (]58 and ]52 kcal mol~1, respectively), the unit is A([)2C(])2 stable by [212 kcal mol~1.In fact, the repulsions are largely compensated by A([)… … …C(]) attractions between adjacent columns of ions ([128 kcal mol~1). This is an admittedly crude estimate of the energy of the PPP but given the size of the molecules more elaborate computations are not accessible. The cation MEP map clearly shows that the C[H groups are positively charged [see Fig. 2(b)], each with ]0.07 e, thus making a grand total of ]0.84 e, while the Cr atom has a positive charge of ]0.22 e. Therefore the CwH groups may well act as acids against the more basic regions of the anions, located on the oxygens, in particular on the bridge oxygen [dark regions in Fig. 2(a)]. The orientation of the A([)… … …C(]) units is such as to make as many CwH… … …O([) bonds possible.A critical point analysis of the electron density of the unit indicates the presence of six intermo- A([)2C(])2 lecular CwH… … …O([) bonds in one of the A([)… … …C(]) dimers and four in the other A([)… … …C(]) dimers. The intriguing case of the mixed-valence crystalline salt 1 can therefore be rationalised as follows.(i) The crystal 1 is ìmainlyœ ionic, i.e. the major cohesive contribution results from Coulombic interactions between the methoxychromate anions and bis-benzene cations. (ii) The interactions between the bis-benzene chromium cations along the stacks are also repulsive due to a Coulombic component (each benzene has a net charge of]0.4 e) ; (iii) The relative orientations of anions and cations are controlled by the need to maximise ìcharge assisted œ CwH… … …O([) bonds between cations and anions which contribute largely to crystal stability.(iv) On the contrary, the inter-anion CwH… … …O interaction is repulsive and the short CwH… … …O distance is not indicative of an attractive interaction. In summary, in providing theoretical evidence for the relevance of weak CwH… … …O bonds in crystal packing,6 we have been able to show that the distance/strength criterion can be misleading when the energetics of the overall intermolecular interactions are not considered or not properly analysed.° The same type of calculations discussed here have been performed for the prototypical NaCl crystal. units present one A([)2C(])2 Na(])… … …Na(]) and one Cl([)… … …Cl([) contact along the diagonals, and four Na(])… … …Cl([) along the sides.Computing the total energy of the Na(])… … …Na(]), Cl([)… … …Cl([) and Na(])… … …Cl([) subunits at the geometry they have in the NaCl crystal multiplied by their occurrence, the unit is found to have a stability of [165 A([)2C(])2 kcal mol~1 at the HF/LANL2DZ level, or [41 kcal mol~1 per Na(])wCl(w) bond.To test if the unit is a good A(w)2C(])2 model to evaluate the stability of the NaCl crystal, we computed the stability of a planar system formed by 12 units Na8Cl8 Na4Cl4 attached by their sides, obtaining a stability of [31 kcal mol~1 per each Na(])wCl([) bond. 756 New J. Chem., 1998, Pages 755»757Fig. 2 The MEP maps computed at the HF/LANL2DZ level for the anion (a) and for the [(g6- cation (b).[CrO3(OCH3)]~ C6H6)2Cr]` The maps are drawn with energy cuts at [160 (dark), [140 (shaded), 30 (light) for the anion and of ]100 kcal mol~1 for the cation ; far from the nuclei the potential is positive everywhere for the cation and negative for the anion The band structure, charge-transfer properties and spin alignment between CrVI and CrI columns and along the paramagnetic CrI columns will be investigated.Acknowledgements support by M.U.R.S.T. and by the University of Financial Bologna (projects : Intelligent Molecules and Molecular Aggregates and Innovative Materials) is acknowledged. Work at Barcelona was supported by DGICYT (project PB95-0848- C02-02) and CIRIT (project GR94-1077). The use of the computer facilities of CESCA is also acknowledged.References 1 (a) G. R. Desiraju, Crystal Engineering : T he Design of Organic Solids, Elsevier, Amsterdam, 1989; (b) D. Braga and F. Grepioni, Chem. Commun., 1996, 571; (c) C. B. Aakeroé y, Acta Crystallogr., Sect. B, 1997, 53, 569; (d) D. Braga and F. Grepioni, Coord. Chem. Rev. in press. 2 (a) C. B. Aakeroé y and K. R. Seddon, Chem.Soc. Rev., 1993, 397; (b) L. Brammer, D. Zhao, F. T. Ladipo and J. Braddock-Wilking, Acta Crystallogr., Sect. B., 1995, 51, 632. 3 (a) D. Braga, F. Grepioni, J. J. Byrne and A. Wolf, J. Chem. Soc., Chem. Commun., 1995, 1023; (b) D. Braga, A. Angeloni, F. Grepioni and E. Tagliavini, Chem. Commun., 1997, 1447. 4 (a) G. M. Sheldrick, Acta Crystallogr., Sect. A, 1990, 46, 467; (b) G. M. Sheldrick, SHELXL92, Program for Crystal Structure Determination; University of Goé ttingen, Goé ttingen, 1992; (c) E. Keller, SCHAKAL92, Graphical Representation of Molecular Models; University of Freiburg, Freiburg, 1992; (d) A. L. Spek, Acta Crystallogr., Sect. A, 1990, 46, C31. 5 D. Braga, A. L. Costa, F. Grepioni, L. Scaccianoce and E. Tagliavini, Organometallics, 1996, 15, 1084. 6 (a) G. R. Desiraju, Acc. Chem. Res., 1996, 29, 441; (b) T. Steiner, Chem. Commun., 1997, 727; (c) D. Braga and F. Grepioni, Acc. Chem. Res., 1997, 30 81. 7 The LANL2DZ basis is documented in P. J. Hay and W. R. Wadt, J. Chem. Phys., 1985, 82, 270. 8 (a) J. J. Novoa and M. Deumal, Mol. Cryst. L iq. Cryst., 1997, 305, 143; (b) J. Veciana, J. J. Novoa, M. Deumal and J. Cirujeda, Magnetic Properties of Organic Materials, ed. P. Lahti, Marcel Dekker, New York, in press. Received in Basel, Switzerland, 24th February 1998; L etter 8/03376E New J. Chem., 1998, Pages 755»757 757

ISSN:1144-0546    

DOI:10.1039/a803376e

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

L e t t e r 1—2 Pd Cl 2 Pd Pd I R Pd R R Pd I Pd I R R –HCl RI 1 2 3 R 4 RI 6 5 – –HI 3 i 4 7 Pd–I R R1 Pd R1 I R1 + Pd0 + Hl ii 10 – 11 12 R R Selective aryl coupling via palladacycles : a new route to m-alkylbiphenyls or m-terphenyls Marta Catellani* and Elena Motti Dipartimento di Chimica Organica e Industriale V .le delle Scienze, I-43100 dellœUniversita` , Parma, Italy o-Alkyl- or o-aryl-substituted arylnorbornylpalladium chloride complexes react with aromatic iodides in dimethylformamide (DMF) via palladacycles to give after hydrogenolysis m-alkylbiphenyls or m-terphenyls.Regioselectivity in aromatic substitution is a topic of fundamental interest in organic synthesis. Palladium chemistry has proved to be quite useful to this aim. In particular we recently succeeded in alkylating an aromatic ring of a palladacycle at the two ortho positions.1 As shown in simpli–ed Scheme 1 the reaction involves double alkylation with alkyl iodide RI of the ortho positions of the aromatic ring of complex 1 (cis,exo)2 through metallacycles 2 and 4,3 followed by norbornene deinsertion with formation of an o,o@-dialkylated arylpalladium(II) species.The overall process thus implies selective formation of two sp2»sp3 carbon»carbon bonds.When we tried to utilise the same procedure to obtain terphenyl complexes using aryl iodides in place of the alkyl ones Scheme 1 Scheme 2 Reagents and conditions : DMF, under K2CO3, N2 , room temperature, 6 h the system showed again a strong tendency to form an sp2»sp3 carbon»carbon bond, migration of the aryl group of the aromatic iodide this time occurring selectively onto the norbornyl site of the alkylaromatic palladacycle 2.A clear example is oÜered by the reaction of 2 (formed in situ from 1) with substituted aryl iodides 7, which aÜords compound 9 through complex 8 and subsequent intramolecular aromatic substitution (R1\4- 68% yield) (Scheme 2). We have CO2Me, observed, however, that formation of an sp2»sp2 carbon» carbon bond leading to arylation at the aromatic site of metallacycle 2 can be achieved in the presence of an alkyl or aryl substituent R in the ortho position.Thus complexes 3 (R\alkyl or aryl) were prepared separately as dimers and caused to react via complexes 4 with aromatic iodides 74 in DMF at room temperature in a one-pot reaction.The resulting intermediates 10 spontaneously liberate norbornene aÜording alkylbiphenyl- and terphenyl-palladium complexes 11 (R\alkyl or phenyl), from which biphenyl derivatives 12 (m-alkylbiphenyls or m-terphenyls) were obtained by hydrogenolysis according to Scheme 3. By contrast when m-alkyl or p-alkyl substituents were used in place of the ortho one in 3, attack on the norbornyl site as in the unsubstituted compounds 8 invariably occurred, thus indicating that the eÜect of ortho substituents in directing arylation at the aryl site is essentially steric in origin.While o-alkyl groups R in complexes 3 appear the most eÜective in promoting sp2»sp2 bond formation, aryl groups do not prevent the reaction at the norbornyl site from taking place to a limited extent (when R\Ph, 10% R1\4-CO2Me of a product characterized as cis,exo-3@-M3-(4-methoxycarbonylphenyl)- 2-bicyclo[2.2.1]heptylN-[1,1@:4@,1A-terphenyl]- 4-carboxylic acid methyl ester was obtained). For the success of the procedure it is important that the equilibrium corresponding to the deinsertion step (from 10 to 11) is displaced to the right. Continuous removal of norbornene from the reaction mixture has a bene–cial eÜect on Scheme 3 Reagents and conditions : (i) DMF, under K2CO3, N2 , room temperature, 6 h; (ii) under at room temperature and atmo- H2 spheric pressure New J.Chem., 1998, Pages 759»761 759R Pd I R Pd R1 + Pd0 + Hl R1 3 R Pd I 13 14 – H2 15 16 R 7 3 (R = But) 4 (R = But) + Pd0 + Hl But 17 + 7 (R1 = H) 18 1 2 7 (R1 = H) 8 (R1 = H) Pd Pd I 18 7 (R1 = H) 19 20 – Scheme 4 this aim; moreover it prevents further insertion of norbornene into CwPd bonds of other species present in the reaction solution with formation of by-products such as 16, as shown in Scheme 4.5 We were pleased to observe that removal of norbornene in vacuo results in satisfactory yields.Thus when the dimer of o-(n-butyl)phenylnorbornylpalladium chloride (analogous to iodide 3, R\Bun), 4-methoxycarbonyliodobenzene (7, R1\ in excess to minimise competitive reactions) and 4-CO2Me, Table 1 EÜect of R and R1 on the yield of 12 in the reaction of complexes 3 with aromatic iodides 7 (Scheme 3) at room temperature in DMFa Biphenyl 12 Entry R R1 yield (%)b 1 Me 4-CO2Me 61c 2 Me 3-Me 35c, 68d,e 3 Me 4-Me 36c, 67d,e 4 Bun 4-CO2Me 71, 99d 5 But 4-CO2Me 36f 6 Ph 4-CO2Me 76c,g a Compound 3 as a dimer (0.05 mmol), (0.3 mmol) and aryl K2CO3 iodide (0.5 mmol).b GC yield based on 3; conversions are higher than 95%. c The main by-product for runs 1, 2, 3, 6 is compound 16 with yields ranging from 11 to 15%. d Run under 0.1 mm Hg. e Ref. 6. f The only by-product is compound 17 (50% yield). g Run 6 also leads to cis, exo-3@-3-(4-methoxycarbonylphenyl)-2-bicyclo[2.2.1]heptyl-[1, 1@ : 4@,1A-terphenyl]-4-carboxylic acid methyl ester (10%) resulting from hydrogenolysis of the parent unsubstituted species 20 reported in Scheme 7.Scheme 5 Scheme 6 Reagents and conditions : Pd(OAc)2, K2CO3, Bu4 NBr, DMF, under 60»100 °C N2 , Scheme 7 Proposed mechanism for the formation of compound 18 were reacted in DMF for 6 h at room temperature in K2CO3 vacuo (0.1 mm Hg) and subsequently treated with or H2 (in excess) compound 12 (R\Bun, NaBH4 R1\4-CO2Me) was obtained in almost quantitative yield according to Scheme 3 (entry 4).Working at atmospheric pressure without removing norbornene led to 71% only. Other representative examples are reported in Table 1. While with R\Me and an acceptable R1\4-CO2Me yield was obtained even without working in vacuo (entry 1), with R1\3- or 4-Me yields were low and almost doubled under vacuum (entries 2 and 3).With R\But (entry 5), however, the formation of 12 was accompanied by that of the benzocyclobutene derivative 17, which is obtained from 4 by reductive elimination (Scheme 5). This anomalous behaviour may be attributed to excessive steric crowding generated by the t-butyl group, which makes complex 4 more susceptible to reductive elimination.7 In the case of R\Ph in complex 3, when R1\4-CO2Me the yield of 12 (entry 6) was satisfactory (76%) even without norbornene removal, with compound 16 (Scheme 4) amounting to 11%.When R1 was a meta or para alkyl group yields of 12 were lower (qualitative results not reported in the Table).This suggests that electronic eÜects are also involved in the aryl»aryl coupling. Although at present the reaction is stoichiometric it adds to the existing organometallic methods8 oÜering new perspectives in catalysis.9 The above results also have a bearing as far as the interpretation of other palladium-catalysed arylations is concerned.In particular the reaction shown in Scheme 6 has been described.10 Although a mechanism based on the formation of a coordinated aryne was proposed,10 the reaction course can be explained straightforwardly by the same palladacycle mechanism already proved by us according to Scheme 7. In fact on comparing complex 4 with complex 19 we can observe that the situation is quite similar, this time the R substituent in the ortho positions being an arylnorbornyl group.Is has also been ascertained that the reaction shown in Scheme 7 does not occur in the absence of norbornene. In conclusion we have found a new type of aromatic arylation via palladacycles which is promoted by the presence of ortho alkyl or aryl substitutents. Further study is in progress to ascertain the scope of the method here described.Experimental General procedure for compounds 12 The desired compound 3 prepared as a dimer (0.05 mmol) according to the literature procedure reported for the parent complex,2a and (0.3 mmol) were introduced under K2CO3 nitrogen into a Schlenk-type —ask and dissolved in DMF (4 ml). The appropriate aryl iodide 7 (0.5 mmol) in DMF (2 ml) was then added and the reaction mixture was stirred at room temperature for 6 h.The decomposition of the arylpalladium species 11 and of the unconverted starting compound was carried out either by adding in excess or by placing NaBH4 760 New J. Chem., 1998, Pages 759»761the solution under hydrogen for 2 h at room temperature and atmospheric pressure. After conventional work up the crude product was separated by —ash chromatography using hexane or mixtures of hexane»ethyl acetate as eluents.Reactions under vacuum (0.1 mm Hg) were carried out analogously. Compound 9 was prepared under the same conditions. NMR data of selected compounds 20 °C; COSY, (CDC13 , NOESY, CwH correlation experiments; *: interchangeable assignments). 1,2,3,4,4a,12b-Hexahydro-7-methoxycarbonyl-1, 4-methanotriphenylene (9, 1H NMR: d 8.49 R1\4-CO2CH3). (1H, d, J\1.8 Hz, H8), 7.91 (1H, m, H9), 7.83 (1H, dd, J\8.0, 1.8 Hz, H6), 7.29 (1H, d, J\8.0 Hz, H5), 7.26»7.20 (3H, m, H10, H11, H12), 4.03 (3H, s, 3.35 (2H, AB CO2CH3), system, H4a, H12b), 2.53 (2H, m, H1, H4), 1.98»1.78 (4H, m, 1.53 (1H, d quintets, J\10.0, H2exo, H3exo, H2endo, H3endo), 1.6 Hz, 1.23 (1H, d quintets, J\10.0, 1.5 Hz, H13syn), H13anti) ; 13C NMR: d 167.2, 143.0, 137.5, 131.7, 130.4, 130.3, (C5), 130.2 (C10*), 128.3 (C6), 128.2 (C11*), 128.1 (C7), 126.4 (C12*), 123.6 (C8), 122.4 (C9), 52.05 49.7 (C1**), 49.5 (C4**), (CO2CH3), 46.1 (C4a***), 45.8 (C12b***), 33.2 (C13), 30.3 (C2****), 30.2 (C3****). 3@-(n-Butyl)-(1,1@-biphenyl)-4-carboxylic acid methyl ester (12, R\Bun, 1H NMR: d 8.10 (2H, H3, R1\4-CO2CH3) : H5), 7.66 (2H, H2, H6), 7.47»7.41 (2H, m, H2@, H6@), 7.37 (1H, dd, J\8.2, 7.4 Hz, H5@), 7.22 (1H, dt, J\7.4, 1.5 Hz, H4@), 3.94 (3H, s, 2.69 (2H, ABX system, 1.65 CO2CH3), CH2Ar), (2H, m, 1.39 (2H, sxt, J\7.3 Hz, 0.95 CH2CH2Ar), CH2CH3), (3H, t, J\7.3 Hz, 13C NMR: d 167.0, 145.9, 143.6, CH3) ; 139.9, 130.0 (C3, C5), 128.8 (C5@), 128.7 (Cl), 128.3 (C4@), 127.4 (C2@), 127.1 (C2, C6), 124.6 (C6@), 52.1 35.7 (CO2CH3), 33.7 22.4 13.9 (CH2Ar), (CH2CH2Ar), (CH2CH3), (CH3). 3@-(1,1@-Dimethylethyl) (1,1@-biphenyl)-4-carboxylic acid methyl ester (12, R\But, R1\ 1H NMR: d 4-CO2CH3). 8.10 (2H, H3, H5), 7.66 (2H, H2, H6), 7.62 (1H, br s, H2@), 7.48»7.40 (3H, m, H4@, H5@, H6@), 3.94 (3H, s, 1.38 CO2CH3), (9H, s, 13C NMR: d 167.0, 151.8, 146.3, 139.8, 130.0 3CH3) ; (C3, C5), 128.7 (C1), 128.6 (C4@*), 127.2 (C2, C6), 125.2 (C5@*), 124.5 (C6@*), 124.3 (C2@), 52.1 34.8 31.4 (CO2CH3), [C(CH3)3], (1,1@ : 3@,1A-Terphenyl)-4-carboxylic acid methyl ester, (3CH3). 12 (R\Ph, 1H NMR: d 8.13 (2H, H3, H5), R1\4-CO2CH3). 7.83 (1H, t, J\1.8 Hz, H2@), 7.72 (2H, H2, H6), 7.65 (2H, H2A, H6A), 7.63»7.59 (2H, m, H4@, H6@), 7.54 (1H, t, J\7.5 Hz, H5@), 7.48 (2H, H3A, H5A), 7.39 (1H, tt, J\7.3, 1.3 Hz, H4A), 3.95 (3H, s, 13C NMR: d 167.0, 145.6, 142.0, 140.9, CO2CH3) ; 140.5, 130.1 (C3, C5), 129.3 (C5@), 129.0 (C4), 128.8 (C5A, C3A), 127.5 (C4A), 127.2 (C2A, C6A), 127.1 (C2, C6), 127.0 (C4@*), 126.2 (C2@), 126.1 (C6@*), 52.1 (CO2CH3). 3@-(2-Bicyclo[2.2.1]heptyl)-(1,1@ : 4@,1A-terphenyl)-4-carboxylic acid methyl ester (16, R\Ph, 1H NMR: d R1\4-CO2CH3). 8.12 (2H, H3, H5), 7.70 (2H, H2, H6), 7.62 (1H, d, J\1.9 Hz, H2@), 7.47 (1H, dd, J\7.8, 1.9 Hz, H6@), 7.43 (2H, H3A, H5A), 7.39 (1H, H4A), 7.32 (2H, H2A, H6A), 7.28 (1H, d, J\7.8 Hz, H5@), 3.95 (3H, s, 2.82 (1H, dd, J\9.0, 6.0 Hz, CO2CH3), H2”), 2.37 (1H, m, H1”), 2.33 (1H, m, H4”), 1.76 (1H, d quintets, J\9.8, 1.8 Hz, 1.62 (1H, m, 1.53»1.43 (3H, H7syn ” ), H3exo ” ), m, 1.26 (1H, d further split, J\9.8 Hz, H3endo ” , H5exo ” , H6exo ” ), 1.10 (2H, m, H7anti ” ), H5endo ” , H6endo ” ). 5-(1,1@-Dimethylethyl)-1,2,3,4,4a,8b-hexahydro-1,4-methanobiphenylene (17). 1H NMR: d 7.16 (2H, H6, H7), 6.82 (1H, H8), 3.28 (1H, br d, J\3.9 Hz, H4a), 3.10 (1H, br d, J\3.9 Hz, H8b), 2.41 (1H, m, H4), 2.26 (1H, m, H1), 1.65»1.54 (2H, m, 1.32 (9H, s, 1.24»1.16 (2H, m, H2exo, H3exo), CH3), H2endo , 0.97, 0.92 (2H, AB system further split, 2H9); 13C H3endo), NMR: d 146.8, 146.2, 143.0, 127.5 (C6), 123.9 (C7), 119.0 (C8), 52.6 (C4a), 49.6 (C8b), 37.4 (C4), 36.7 (C1), 34.7 [C(CH3)3], 31.7 (C9), 31.0 27.9 (C2, C3).(3CH3), Acknowledgements thank Ministero Universita` e Ricerca Scienti–ca and We National Research Council for –nancial support. Mass and NMR facilities were provided by Centro Interfacolta` di Misure of the University of Parma.References 1 M. Catellani and M. C. Fagnola, Angew. Chem., Int. Ed. Engl., 1994, 106, 2559. 2 (a) H. Horino, M. Arai and M. Inoue, T etrahedron L ett., 1974, 647; (b) M. Catellani and G. P. Chiusoli, J. Organomet. Chem., 1983, 250, 509; (c) C.-S. Li, S.-H. Cheng, F.-L. Liao and S.-L. Wang, J. Chem.Soc., Chem. Commun., 1991, 710. 3 M. Catellani and G. P. Chiusoli, J. Organomet. Chem., 1988, 346, C27. 4 The process possibly implies oxidative addition of the aryl iodide to the metallacycle with formation of a palladium(IV) species as ascertained for alkyl iodides.1 5 The initial step 3]13 formally consists of a rearrangement probably occurring via a palladacycle as previously reported : G.Bocelli, M. Catellani and G. P. Chiusoli, J. Organomet. Chem., 1985, 279, 225. 6 E. A. Johnson, J. Chem. Soc., 1957, 4155. 7 M. Catellani and L. Ferioli, Synthesis, 1996, 769. 8 See for example: J. K. Stille and D. E. James, J. Am. Chem. Soc., 1975, 97, 674; J. K. Stille and D. E. James, J. Organomet. Chem., 1976, 108, 491; J. K. Stille and R. Divakaruni, J. Am. Chem. Soc., 1978, 100, 1303; J. K. Stille, A. M. Echavarren, R. M. Williams and J. A. Hendrix, Org. Synth., 1992, 71, 97; D. E. Ames and D. Bull, T etrahedron, 1982, 32, 383; D. E. Ames and A. Opalko, T etrahedron, 1984, 40, 1919; A. Suzuki, Pure Appl. Chem., 1994, 66, 213; A. Suzuki, ibid, 1985, 57, 1749; N. Miyaura and A. Suzuki, Chem. Rev., 1995, 95, 2457; A. F. Indolese, T etrahedron L ett., 1997, 38, 3513; J. C. Anderson, H. Namli and C. A. Roberts, T etrahedron, 1997, 53, 15123. 9 A catalytic process was derived from the dialkylation procedure shown in Scheme 1. M. Catellani, F. Frignani and A. Rangoni, Angew. Chem., Int. Ed. Engl., 1997, 36, 119. 10 K. Albrecht, O. Reiser, M. Weber, B. Knieriem and A. de Meijere, T etrahedron, 1994, 50, 383. Received in Cambridge, UK, 23rd February 1998; L etter 8/02846J New J. Chem., 1998, Pages 759»761 761

ISSN:1144-0546    

DOI:10.1039/a802846j

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

L e t t e r S S N N R • Se Se N N R • N S N S R • 1 (1a: R = Ph) 2 (2a: R = Ph) 3 (3a: R = Ph) Isolation of the –rst diselenadiazolyl complex, Pd3 [PhCNSeSeN] 2[PPh3 ] 4 Æ 2PhMe John E. Davies,a Robert J. Less,a Iain May§,b and Jeremy M. Rawson*,a a Department of Chemistry, University of Cambridge, L ens–eld Road, Cambridge, UK CB2 1EW b Department of Chemistry, T he University of Durham, South Road, Durham, UK DH1 3L E The selenium»nitrogen radical PhCNSeSeN is shown to be partially associated in solution ; it undergoes an oxidative addition reaction to via SewSe bond cleavage, and the structure of the –rst Pd(PPh3)4 diselenadiazolyl»metal complex, is reported.Pd3[PhCNSeSeN]2[PPh3]4 The chemistry of the 1,2,3,5-dithiadiazolyl ring system (1) is well established.1 However, it was not until 1989 that the related diselenadiazolyl ring system (2) was prepared.2 Since then a number of derivatives have been reported3,4 by Oakley and co-workers.These materials have been investigated for their potential as low-dimensional molecular conductors and a number of doped forms show electrical conductivity.4 In contrast to the extensive chemistry1 of the sulfur-based ring system, 1, there has, to our knowledge, been only one report5 regarding the chemistry of 2 and we were intrigued to investigate whether derivatives of 2 might exhibit similar, or diÜerent, physical and chemical properties to 1.We now report studies on the solution behaviour of 2a and its reactivity towards the zero-valent group 10 metal complex Pd(PPh3)4 . Derivatives of the dithiadiazolyl radical, 1, are almost exclusively associated in the solid state through a spin-paired dimerisation process at sulfur.1 Nevertheless, the dimerisation energy is relatively weak (ca. 35 kJ mol~1)6 and the derivatives become almost completely dissociated upon dissolution. The monomer»dimer equilibria exhibited by these radicals can be elegantly studied by UV/VIS spectroscopy ; Passmore and Sun7 have shown that solutions of the isomeric radical 3a exhibit two distinct absorption bands (250 and 376 nm) and a weaker band at 480 nm; the intensity of the high-energy band is directly proportional to the concentration of monomeric 3a whereas the intensities of the bands at 376 and 480 nm are proportional to the square of the radical concentration, consistent with the presence of dimers in solution.Our solution UV/VIS studies (10~2 to 10~5 M solutions in THF) show that 1a exhibits absorption maxima at 225, 285, 410 and 580 nm. Variable concentration studies indicate that the absorption at 410 nm corresponds to monomeric 1a, whereas the absorption at 580 nm corresponds to dimeric and this absorption gives rise to the typical purple (1a)2 colouration observed for concentrated solutions of 1a.Previous work has shown that 0.04 M solutions of 1a in CFCl3 are yellow and essentially completely dissociated.7 Solutions of 2a in THF are EPR active (broad singlet, g\2.04)5 and 2a exhibits an intense absorption around 244 nm in its UV/VIS spectrum, corresponding to monomeric 2a. § Present address : B319.4, British Nuclear Fuels Ltd.UK Group, Sella –eld, Seascale, Cumbria, UK, CA20 1PG. * E-mail: jmr31=cus.cam.ac.uk Estimates of the unpaired spin density on selenium (based on EPR studies5) indicate that there is a small increase in unpaired electron density at the chalcogen5 (compared to S) and this might be expected to strengthen this dimerisation process. Even under dilute conditions (10~3»10~4 M), two additional absorption maxima can be observed (375 and 487 nm) which exhibit a linear dependence between the square of the radical concentration and the absorption maxima; indicative of incomplete dissociation (cf. 1a). These absorptions (Fig. 1) give rise to the purple colour of 2a in solution and we infer that the monomer»dimer equilibrium [PhCNSeSeN]2H2PhCNSeSeN~ (1) lies further to the left for 2a in comparison to 1a.The intensity of the EPR spectrum (relative to a standard DPPH solution) indicates that at room temperature the dissociation process is only ca. 70% complete.î Fig. 1 UV/VIS spectrum of a 10~4 M solution of 2a in THF. Inset : linear relationship between absorbance and the square of 2a concentration for the absorption maxima observed at 375 and 487 nm î The width of the solution EPR signal precludes the observation of any hyper–ne structure.The peak width broadens markedly on cooling which makes an accurate determination of peak intensity difficult. This has hindered our attempts to extract the thermodynamic *H and *S parameters for the equilibrium process from variabletemperature EPR studies.New J. Chem., 1998, Pages 763»765 763Se N N Se Pt R PPh3 PPh3 • Ph N N Se PPh3 PPh3 N N Se Ph Ph3P Ph3P Pd Se Pd Se Pd 4 5 One of the most diverse aspects of the chemistry of 1, is its coordination chemistry,8 in which it can provide 2e~, 3e~, 5e~ or 6e~ for metal»ligand bonding. EPR evidence has supported5 the transient formation of the diselenadiazolyl» platinum complex, 4, which rapidly decomposed to EPR inactive products.As a continuation of our work in this area we now provide the –rst solid-state characterisation of a diselenadiazolyl»metal complex, Pd3[PhCNSeSeN]2[PPh3]4 (5). Reaction of with in THF [PhCNSeSeN]2 Pd[PPh3]4 yielded (5) as a deep-red pre- Pd3[PhCNSeSeN]2[PPh3]4 cipitate,° sparingly soluble in most common organic solvents. The FAB mass spectrum of 5 exhibited a multiplet at 830 amu consistent with the frag- MPd3[PhCNSeSeN]2[PPh3]3N2` ment.Crystals of 5 suitable for X-ray diÜraction“ were grown from toluene by slow diÜusion techniques.9 The structure of 5 (Fig. 2) is centrosymmetric and reveals three Pd atoms bridged by two PhCNSeSeN ligands in which the SewSe bond is formally cleaved, the SewSe distance increasing from 2.341(3) in 2a2 to 3.201(1) in 5.The increase in Se………Se ” ” distance is accommodated by an increase in the average hinge angles at C and N [from 128(2)° and 116(1)° in 2a to 134.8(7)° and 124.5(5)° in 5, respectively]. Heterocyclic SewSe bond cleavage has previously been reported by Chivers et al.14 for derivatives of the 1,5- ring system. The terminal Ph4P2N4Se2 Pd atoms take up planar geometries with the PPdP angle [101.47(7)°] somewhat greater than the idealised 90° owing to the strain imposed by the chelating diselenadiazolyl ligand [SePdSe\80.38(3)°].The central Pd atom possesses a planar coordination geometry with the corresponding SePdSe PdSe4 bond angles of 81.56(3)°. The diamagnetic nature of this compound, coupled with the square-planar environment about each Pd is entirely analo- ° A mixture of 2a (80 mg, 0.30 mmol) and (500 mg, 0.43 Pd(PPh3)4 mmol) were stirred in THF (20 ml) for 4 h at room temperature. The resultant red-brown solid was –ltered and washed with THF (3]20 ml) and dried in vacuo.Yield 205 mg, 72%. 31P NMR (CDCl3), d]35.6(s) (the low solubility of 5, precluded observation of Se satellites).Calcd for 5, THF: C, 54.25%; H, 3.92%; N, 2.81%; found: C, 54.25%; H, 4.28%; N, 2.46%. “ Crystal Data for 5, 2PhMe: M\2102.65, C100H86N4P4Pd3Se4 , triclinic, space group a\14.102(4), b\14.242(4), c\11.022(3) P1 6 , ”, a\98.87(2), b\93.82(2), c\79.74(2)°, V \2151 [from 2h values ”3 of 25 re—ections measured at ^x (15O2hO20)], k\0.710 69 ”, Z\1 (molecule lies on an inversion centre), g cm~3, De\1.624 T \150 K, red block, crystal size 0.3]0.2]0.2 mm, l\2.44 mm~1.Data Collection and Processing : Rigaku AFC 5R diÜractometer with Oxford Cryosystems low-temperature device10 graphite monochromated Mo-Ka radiation, x-2h scans. The data were corrected for absorption by means of psi-scans.11 The structure was solved by direct methods using SHELXS-8612 and re–ned using full-matrix least-squares techniques.13 All non-H atoms were re–ned anisotropically and H atoms were added at calculated positions with a –xed thermal parameter.Convergence was obtained at R1[F0[2r(F)]\ (all data)\0.114, S\1.032 for 7538 independent re—ec- 0.051, wR2 tions (5O2hO50°). Atomic coordinates, bond lengths and angles and thermal parameters have been deposited at the Cambridge Crystallographic Data Centre, reference number 440/043.See Information for Authors, Issue No. 1. Fig. 2 Structure of 5 (with solvent molecules removed and only P-bound C atoms of groups shown for clarity). Selected bond PPh3 lengths and angles (°) are : Pd(1)wSe(1) 2.4481(10), Pd(1)wSe(2) (”) 2.4526(9), Pd(2)wP(1) 2.345(2), Pd(2)wP(2) 2.342(2), Pd(2)wSe(1) 2.4698(11), Pd(2)wSe(2) 2.4903(11), Se(1)wN(1) 1.826(6), Se(2)wN(2) 1.815(6), N(1)wC(7) 1.318(9), N(2)wC(7) 1.326(9), Pd(1)… … …Pd(2) 2.9657(9), Se(1)Pd(1)Se(2) 81.56(3), Se(1)Pd(1)Se(2a) 98.44(3), Se(1)Pd(2)Se(2) 80.38(3), P(1)Pd(2)P(2) 101.47(7), P(1)Pd(2)Se(1) 168.70(5), P(1)Pd(2)Se(2) 88.53(5), P(2)Pd(2)Se(1) 89.83(6), P(2)Pd(2)Se(2) 166.34(5).Symmetry transformation to generate equivalent (a) atom: [x, [y, 1[z gous to the sulfur derivative,9 Pd3[PhCNSSN]2[PPh3]4 .Thus, the 16e~ associated with terminal Pd atoms are composed of : 10e~ from the metal, 2e~ from each phosphine donor and 2e~ from the diselenadiazolyl ligand. The bonding requirements for the central Pd atom are made up as follows : 10e~ from the metal centre and 3e~ from each diselenadiazolyl ligand.Thus each diselenadiazolyl ligand provides 5e~ for metal»ligand bonding. These studies illustrate that, in comparison to the sulfur analogue 1a, the diselenadiazolyl radical, 2a, is more strongly associated in solution. Despite these diÜerences in physical properties, the chemical reactivity of 2a towards zero-valent group 10 metal phosphines would appear to be analogous to that observed for dithiadiazolyls, 1.A more detailed investigation of the monomer»dimer equilibrium and substituent eÜects on this process are underway. Further experiments are being carried out to compare the reactivities of 1 and 2. Acknowledgements would like to thank the EPSRC for a studentship (R.J.L.), We BNFL and the Newton Trust for additional –nancial support (R.J.L.) and the Royal Society for an equipment grant (J.M.R.).We are particularly indebted to Dr. E. J. L. McInnes (EPSRC cw EPR Service, Department of Chemistry, University of Manchester) for EPR studies. References 1 J. M. Rawson, A. J. Banister and I. Lavender, Adv. Hetetrocycl. Chem., 1995, 62, 137. 2 P. D. B. Belluz, A. W. Cordes, E. M. Kristof, P. V. Kristof, S. W. Liblong and R. T. Oakley, J. Am.Chem. Soc., 1989, 111, 9276. 3 P. D. Belluz, A. W. Cordes, E. M. Kristof, P. V. Kristof, S. W. Liblong and R. T. Oakley, J. Am. Chem. Soc., 1989, 111, 9276; A. W. Cordes, R. C. Haddon, R. T. Oakley, L. F. Schneemeyer, J. V. Waszczak, K. M. Young and N. M. Zimmerman, J. Am. Chem. Soc., 1991, 113, 582; M. P. Andrews, A. W. Cordes, D. C. Douglas, R. M. Fleming, S. H. Glarum, R.C. Haddon, P. Marsh, R. T. 764 New J. Chem., 1998, Pages 763»765Oakley, T. T. M. Palstra, L. F. Schneemeyer, G. W. Trucks, R. Tycko, J. V. Waszczak, K. M. Young and N. M. Zimmerman, J. Am. Chem. Soc., 1991, 113, 3559; A. W. Cordes, R. C. Haddon, R. G. Hicks, R. T. Oakley and T. T. M. Palstra, Inorg. Chem., 1992, 31, 1802; A. W. Cordes, R. C. Haddon, R. G. Hicks, R. T.Oakley, T. T. M. Palstra, L. F. Schneemeyer and J. V. Waszczak, J. Am. Chem. Soc., 1992, 114, 1729; A. W. Cordes, S. H. Glarum, R. C. Haddon, R. Haliford, R. G. Hicks, D. K. Kennepohl, R. T. Oakley, T. T. M. Palstra and S. R. Scott, J. Chem. Soc., Chem. Commun., 1992, 1265; A. W. Cordes, R. C. Haddon, R. G. Hicks, D. K. Kennepohl, R. T. Oakley, T. T. M. Palstra, L. F. Schneemeyer, S.R. Scott and J. V. Waszczak, Chem. Mater., 1993, 5, 820; W. M. Davis, R. G. Hicks, R. T. Oakley, B. Zhao and N. J. Taylor, Can. J. Chem., 1993, 180; A. W. Cordes, C. D. Bryan, W. M. Davis, R. H. de Laat, S. H. Glarum, J. D. Goddard, R. C. Haddon, R. G. Hicks, D. K. Kennepohl, R. T. Oakley, S. R. Scott and N. P. C. Westwood, J. Am. Chem. Soc., 1995, 115, 7232. 4 C. D. Bryan, A.W. Cordes, R. C. Haddon, R. G. Hicks, R. T. Oakley, T. T. M. Palstra, A. S. Perel and S. R. Scott, Chem. Mater., 1994, 6, 508; A. W. Cordes, R. C. Haddon and R. T. Oakley, Adv. Mater., 1994, 6, 798; C. D. Bryan, A. W. Cordes, N. A. George, R. C. Haddon, C. D. MacKinnon, R. T. Oakley, T. T. M. Palstra and A. S. Perel, Chem. Mater., 1996, 8, 762. 5 J. M. Rawson, A. J. Banister and I.May, Magn. Reson. Chem., 1994, 32, 487. 6 S. A. Fairhurst, K. M. Johnson, L. H. SutcliÜe, K. F. Preston, A. J. Banister, Z. V. Hauptman and J. Passmore, J. Chem. Soc., Dalton T rans., 1986, 1465. 7 J. Passmore and X. Sun, Inorg. Chem., 1996, 35, 1313. 8 A. J. Banister, I. May, J. M. Rawson and J. N. B. Smith, J. Organomet., 1998, 550, 241. 9 A. J. Banister, I. B. Gorrell, J. A. K. Howard, S. E. Lawrence, C. W. Lehman, I. May, J. M. Rawson, B. K. Tanner, C. I. Gregory, A. J. Blake and S. P. Fricker, J. Chem. Soc., Dalton T rans., 1997, 377. 10 J. Cosier and A. M. Glazier, J. Appl. Crystallogr., 1986, 19, 105. 11 A. C. T. North, D. C. Phillips and F. S. Mathews, Acta Crystallogr., Sect. A, 1968, 24, 351. 12 G. M. Sheldrick, Acta Crystallogr., Sect. A 1990, 46, 467. 13 G. M. Sheldrick, SHELXL-93, University of Goé ttingen, 1993. 14 T. Chivers, D. D. Doxsee and R. W. Hilts, Inorg. Chem., 1993, 32, 3244. Received in Basel, Switzerland, 23rd April 1998; L etter 8/04234I New J. Chem., 1998, Pages 763»765 765

ISSN:1144-0546    

DOI:10.1039/a804234i

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

L e t t e r Sense and nonsense of science citation analyses : comments on the monopoly position of ISI and citation inaccuracies. Risks of possible misuse and biased citation and impact data. Jan Reedijk L eiden Institute of Chemistry, Gorlaeus L aboratories, L eiden University, P.O. Box 9502, 2300 RA L eiden, T he Netherlands Journal editors and publishers, authors of scienti–c papers, research directors, university and research council administrators, and even government officials increasingly make use of so-called ìImpact Factorsœ to evaluate the quality of journals, authors and research groups.These –gures are used in decision-making processes about (dis)continuation of journal subscriptions, selection of journals for submission of papers, ranking of authors and groups of authors, and even for increase and decrease of funding to research groups. All data are based on the counting of citations of the scienti–c papers of authors.Very few users appear to realize that these –gures can be seriously wrong, biased and even manipulated, as a result of : (i) citation habits for authors in diÜerent –elds, (ii) selectivity in (non)citations by authors, (iii) errors made by authors in citation lists at the end of papers, (iv) errors made by ISI in entering publications and citations in databases, and in classifying citations and accrediting them to journals and authors, and (v) incomplete and misleading impact –gures published by ISI.Although quite a few bona –de and competent analysts and organisations specialized in citation analyses exist, the incompetence of many analysts, when using crude ISI data in discussing rankings of journal and/or authors, is an additional factor that makes such analyses often unreliable.This paper reviews some of the current practices in publications and citations for (bio)chemists and (bio)chemistry journals ; critical comments are made with regard to the use and consequences of erroneous and incomplete or too detailed data.A few recent examples are given of the use and misuse of such data, to illustrate and evaluate the (non)sense of current practice. Originally citation databases were developed for use by scientists to –nd out who were the (competing) researchers interested in their work, and to –nd out who had picked up certain methodologies.The use of subject indexes had increasingly become much too time-consuming for this purpose. More recently, citation records of authors, institutes, universities and journals are increasingly considered as quality indicators, not only by authors, librarians and journal publishers, but also by science policy-makers. In a sense the academic world has gradually become obsessed with impact factors.This whole process originally started with journals analyses, but nowadays has extended to cover also topics like countries, universities, institutes and even individual researchers. The present paper will review some of the current practices in publications and citations for (bio)chemists and (bio)chemistry journals. Some critical comments will be presented with regard to the use and consequences of erroneous and incomplete or too detailed data.A few recent examples of the use and misuse of such data will be given, to illustrate and evaluate the (non)sense of current practice and their limited value for the evaluation of (small) research groups. Journal Analyses With the ever-increasing number of journals and decreasing library budgets worldwide it has become almost common use to rank scienti–c journals with the aid of a so-called ìImpact Factorœ (IF).These IF values can be calculated in several ways and may cover several periods. A very commonly used and published IF is the short-term impact IF, de–ned as : Fax: ]31 71 527 4451; e-mail : Reedijk=chem.leidenuniv.nl The number of citations published in year Y to (all) documents published in years Y [1 and Y [2, divided by the number of citeable documents published in years Y [1 and Y [2.(Thus the IF de–ned by ISI for 1997 relates to citations in 1997 of documents published in 1996 and 1995.) It has been proven repeatedly that the Institute for Scienti–c Information (ISI) inaccurately de–nes what a ì citeable documentœ is, and that as a consequence the IF values in the Science Citation Index are inaccurate and deviations up to 40% may occur.1,2 Nevertheless, many libraries take decisions about the discontinuation of journals, and about accepting new journals, based on this often incorrect short-term IF value.Libraries have a duty also to serve future generations, and at least IF values for a period of up to say 10 years (rather than the 2 years now) would be much more useful for those who take decisions about subscribing to or publishing in a certain journal. A recent detailed analysis by Moed et al.3 has shown that the ranking of journals, when periods longer than 2 years are used, can change dramatically.It is possible to get citation data on journals covering longer periods of time from ISI.Unfortunately, it is to be feared that library subscriptions to journals will be increasingly determined by the»so far the only published»short-term impact factors of journals, and certainly so when leading journals even use their IF in advertisements. Analyses of (Groups of) Scientists For a variety of reasons, authors increasingly (are encouraged to) publish in so-called high-impact journals, and even more so with the increasing trend to use journal IF values in the evaluation of research.Whether the published citation –gures New J. Chem., 1998, Pages 767»770 767of that journal are correct, relevant, or maybe not, is rarely considered. Often at best limited attention is given to matters such as : (i) citation habits for authors diÜer for diÜerent research sub–elds; (ii) selectivity in (non)citations by authors (easily available papers are easy to cite) ; (iii) errors made by authors in citation lists at the end of papers (e.g., page, issue and volume numbers); (iv) errors made by ISI in counting and classifying citations and accrediting them to journals and authors; it is crucial to know how ISI inputs publications and citations ; (v) incomplete impact –gures published by ISI ; it is important to know how ISI retrieves citation data; it appears that ISI uses exact matching to analyse the impact of a scientist, so all data that contain even a small error in either input or output will almost certainly get lost.A few recent papers of Seglen4 have addressed several problems associated with the use of impact factors and pointed out clearly that no correlation at all exists between the IF of a journal and the individual papers in that journal.4a Given the fact that citations to scienti–c articles are stored and can be easily retrieved (accurately or not), now also science policy-makers can use these data and the consequences of citation analysis errors (and interpretation errors) can be quite large.Research groups can be discontinued and funding for research can be increased, decreased or even stopped. It appears that nowadays one seems to forget that the primary aim of scienti–c publication is the improvement of science and NOT to generate funds! In some cases funding is directly and proportionally determined by the ìimpactœ of publications in previous years ; for example, in Finland each ìImpact Pointœ is worth $15000 in medicinal research funding,5 incorrectly assuming that the journal impact factor would be automatically valid for ALL its articles.One tends to neglect the fact that even in journals with a high IF, several papers will never get cited ! Citation rates of articles determine the journal IF, but not the reverse ! Tenure decisions and departmental reviews are other common (mis)uses of citation data.Methods used for Counting and Analysing Citations At the least, the following analysis methods of research impact, listed by increasing level of sophistication, can be considered : A. Simply use the NUMBER of citations (in a given period), called CIT, to an institute (or to a group or a person) ; this method is rarely used these days (but see below for a www site doing this).B. Use the number of citations (in a given period) to the papers (as far as known to ISI) published in a certain period by an institution (group, person). Applied parameter: CIT/ PUB. C. The same method as described in B, but now using a threshold value of (say) at least 100 papers by a given author (or group) in a certain period (to avoid eÜects of one or two papers in—uencing the average too much).D. As in method B, but now only the highest cited papers, that is, the top fraction (say 10 or 50%) with respect to their citation scores are considered in the analysis and ranking (to prevent dilution of papers from productive groups). E. As in method B, but the outcome of CIT/PUB is compared to the average of the journals in which a group has published (i.e., the number of expected citations in these journals are considered).So when one publishes in Nature or Science, one should expect more citations then when one has published in a national chemistry journal. From these data it can be calculated whether the institute (group, author) has a value that is above or below the expected journal-package value.F. As in method E, but the outcome of CIT/PUB is now compared with the citation average of all articles in the subareas in which the group is active. These last and quite advanced methods (E, F) are not yet used on a wide scale, but can be quite helpful,6 even though such analyses also have disadvantages ; thus, using method E may, for example, lead to the publication of a top paper (expected to be highly cited) in a low-ranked journal.In fact, each of these methods and parameters has its value and limitations, which will not be discussed here, although the risk of overestimation appears to be large.7 Also, as this process goes on it is not unlikely that some scientists will be challenged to cite preferentially papers of their friends (and vice versa) or otherwise to bias the data by careless citation practices.Such behaviour would require journal editors to add questions about citation behaviour on their checklists for referees. In a recent study the citation habits in many journals have been analyzed and discussed in detail with respect to author and journal behaviour8a and from this analysis it is evident already that care is required in using citation data for evaluating the quality of journals and scientists.In fact, the Royal Society (UK) has recently expressed very critical comments, 8b referring to corruption of the peer-review process and even the promotion of scienti–c misconduct. How useful are such Citation and Impact Analyses? At this point I do feel that two other important questions need to be raised in discussions about citations and their analysis, namely: 1.Are the published citation data accurate enough to be correct ? (Or put another way: how inaccurate are they ?) 2. Are these data correctly interpreted by analysts (and politicians) ? Surprisingly, these questions have rarely been addressed so far in the literature.The general public feeling appears to be quite often : ìWhat a computer produces must be correct and cannot be wrong.œ However, one should realize that the quality of the output is directly determined by the quality of the input. Below I shall show that there are good reasons to assume that the data made available by ISI do contain signi–- cant errors. Up until recently, it was not easy and quite labour intensive to –nd out whether ISI had (correctly) listed and published all citations to a given research paper.One could go to the library-bound volumes and apply manual counting; somewhat later one could consult CD ROMs and databases to perform searches and checks. All of this was and is quite timeconsuming. Quite recently, ISI has started to oÜer to individual scientists (and also to institutes, research councils, etc.) a personal or institutional (or even a university or country) ìpro–leœ with papers and citations (starting from 1981) for a reasonable price.ISI can now provide authors (or anybody who pays) with a database that can be easily analyzed on a PC; it contains all papers (of the scientist or the group of scientists) they have ì collected œ in their database and also all ìcollected œ citations to those papers (full references). Whether or not all data have been correctly included remains to be seen.Nevertheless, this database now allows a direct check between an observed citation and its possible absence/presence in the ISI database. This type of analysis can be done manually by each individual author, but also certain research institutes are doing such detailed analyses, based on information purchased from ISI.Some that can be mentioned are : The Science Policy Research Unit (SPRU) at the University of Sussex, the Information Science and Scientometrics Research Unit (ISSRU) in the Academy of Science Library at Budapest and the Centre for 768 New J.Chem., 1998, Pages 767»770Science and Technology Studies (CWTS) at Leiden University. Given the fact that one can now purchase the abovementioned personal pro–les for any scientist from ISI, it has become relatively easy to check for individual authors (and organisations), whether ISI has correctly and completely listed all their papers in the ISI database and correctly and completely listed all the citations to their papers.Finally, another fashion should be mentioned, introduced in recent years and also to be questioned, it deals with giving ISI (paid) orders such as ìWho is the most cited chemist? œ or ìWho is the most cited physicist ? œ Two interesting (and frequently cited) web sites provide answers to such questions and rank the most cited scientists according to the citations to their papers in chemistry journals (or to be precise : in journals that ISI has classi–ed as belonging to chemistry). This kind of information can give both too low and too high results because of two major limitations.Papers (if any) of the same authors in journals classi–ed by ISI as, for example, biochemistry or physics are NOT considered.Also, scientists with exactly the same names and initials are not discriminated and in fact are seen as just one person. For interested readers the sites for information on ìchemistsœ are : http ://—uo.univ-lemans.fr :8001/1000chimistes.html http ://pcb4122.univ-lemans.fr/cgi-bin/perl.exe ?chimistes.pl http ://—uo.univ-lemans.fr :8001/chimie/chimistes.html It should not need saying that data from such a web site should NOT be used in quantitative analyses ! Such analyses should only be performed by quali–ed experts and institutions, AFTER the necessary corrections for errors and omissions have been made.Critical Comments about the ISI Database Input and Output Wisely, ISI is not claiming 100% coverage, and it is often assumed that a score of about 90% can be reached by ISI. This number might look quite acceptable, but what if your most-cited paper happens to be in their missing 10%? Will your research council understand or realize that when they assess your research work? What happens when your mostcited paper is mis-referred to by others (e.g., one author name missing or misspelled, year or page mistyped) and subsequently other authors quote the wrong citation, so that every journal reader will notice that your work is cited, but ISI will not (or incorrectly) include it in its database, so that it will often not, or not at all, appear in analyses and assessments ? A few examples of often neglected inaccuracies and errors that currently cannot be easily corrected and that may have serious consequences in evaluations will now be given as an illustration.They primarily deal with correct names and volume numbers of journals. It appears that certain publications, like those in the series Metal Ions in Biological Systems, which can be regarded as books or journals, are easily neglected as a source of input or as sources of citations, owing to diÜerent author (and journal) practices to refer to work from this series.Sometimes publishers may decide to combine one or more volumes of a journal, such as in a special issue with volume 39»40. However, this can result is three possible references : to volume 39, to volume 40 and to volume 39»40. ISI will in many cases routinely only accept one of these as the real citation and the others will get lost. Rather recently, J.Chem. Soc. Chem. Commun. modi–ed its name to Chem. Commun., but in practice authors that cite papers published in Chem. Commun. often still use the old name, and as a consequence the citations might get lost for the journal (and the authors), eventually resulting in a lower Impact Factor for Chem. Commun. In this respect, it will be of interest to see how ISI will respond to recent changes in the organic and inorganic chemistry journals of West European countries, where –rst (1997) Recueil des T ravaux Chimiques des Pays Bas merged with Chemische Berichte and L iebegs Annalen der Chemie, followed in 1998 by a further merger with Belgian, French and Italian national journals to generate Eur.J. Inorg. Chem. and Eur. J. Org. Chem. (EurJIC and EurJOC).Browsing through my own publication and citation records over the last 15 years [purchased from ISI in early 1997 (a so-called Personal Citation Report, in a format as they sell it to ìany organisationœ)] and assuming this output to be a representative case for papers published in the period 1981»1995, I found quite a few papers that contain minor and major errors with respect to the citation scores.I just mention here two serious, and I believe representative, cases. 1. The journal (Structure and Bonding) in which one paper was published (1987) is one processed by ISI for inclusion in the Science Citation Index but, for some unclear reason, the reference to it is not included in the ISI database. This paper has been frequently cited during the last decade but, as a consequence of the non-inclusion of the paper in the database, unfortunately it is NOT listed a single time in my personal citation database obtained from ISI. 2. The journal that contains another one of my papers9 is also an ISI-processed journal ; the reference to the paper is listed by ISI as a regular (1992) paper from me; this paper had received over 40 citations in the given period (independently checked from an another source10a,b) in regular journals, whereas ISI lists only 2 in their ìcommercial database.œ It appears that ISI retrieves by exact matches only, and therefore small deviations, such as in ref. 9b and 9c, are neglected ! The origins for this error are to be assigned to ISI and at the moment can only be ascribed to careless input and output procedures at ISI.I have good reasons to assume that similar errors can occur with other papers and citations from me, or from other authors. When confronting ISI with missing citations and errors, their usual reply will be: ìWe have not yet found a way to implement a procedure of systematically identifying and correcting erroneous source or citation data on a paper-bypaper basis in the 13 million record database.œ I conclude, therefore, that analyses of groups or individual authors, especially when short periods are concerned, and/or small groups of papers are involved, should never be taken seriously.The ability of ISI to –nd citations and to store them depends on how they have been cited by others, and as a consequence mis-citations (even minor ones, such as a second initial) may lead to loss of the citation and eventually, even in decreased funding.A warning to all organisations to never use ISI data directly by non-experts is appropriate. In fact, warnings of this type of inaccuracy were already made almost a decade ago.11 Recently, the Second European Report on Science and T echnology Indicators (1998) has appeared, which illustrates beautifully how non-experts can ask the wrong questions of analysts, and subsequently misuse the answers in publications. 12 Conclusions and Recommendations It is beyond any doubt that ISI has done and is doing great work in collecting and classifying citations to scienti–c papers. However, their data should primarily be used for their original purpose: to serve scientists in the progress of research and development (R&D)! In summary, I arrive at four major recommendations: 1.Librarians : Do not take your decisions on journal subscriptions based on 2 year IF values published by ISI ; instead try to collect other ìimpact factors œ covering much longer periods from other sources. Other organisations, such as STN International,13 do allow specialized searches for librarians. New J.Chem., 1998, Pages 767»770 7692. Authors : Do not rank journals too much on published 2 year IF values, and do not rely on such inaccurate numbers in deciding where to publish. It is to be regretted that as long as politicians primarily judge science quality based on the ìcoversœ(\IF) of journals, scientists will choose such journals. 3. Research councils and science policy-makers : Do not rank the quality of scientists based on (often inaccurate) ISI citation statistics only. Consult experts and do not rely on crude ISI data only! If you really want to use citation information in your analyses, you had better do it well, otherwise donœt do it at all !14 4. Journal editors and publishers : Beware of (even minor) name changes in your journals.It may cost you citations and consequently library subscriptions and even contributions of high-level articles. Never combine volume numbers. Give accurate and unambiguous instructions to your readers on how to cite papers in your journal ! Be prepared to give instructions to your referees to check whether the citation list at the end of the paper is well-balanced.References 1 (a) H. F. Moed and Th. N. van Leeuwen, J. Am. Soc. Inform. Sci., 1995, 46, 461; (b) H. F. Moed and Th. N. van Leeuwen, Nature (L ondon), 1996, 381, 186; (c) H. F. Moed, Th. N. van Leeuwen and J. Reedijk, Scientometrics, 1996, 37, 105; (d) T. N. van Leeuwen, H. F. Moed and J. Reedijk, Chem. Intelligencer, 1997, July, 32. 2 (a) A. T. Braun and W. Glaé nzel, Chem. Intelligencer, 1995, January, 31; (b) E. Gar–eld, Curr. Cont., January 12, 1987. 3 H. F. Moed, Th. N. van Leeuwen and J. Reedijk, J. Chem. Docum., 1998, in press. 4 (a) P. O. Seglen, J. Am. Soc. Inform. Sci., 1997, 45, 1; (b) P. O. Seglen, Br. Med. J., 1997, 314, 498. 5 K. Raivio, Eur. Sci. Editing, 1997, 23, 80. 6 H. F. Moed and F. Th. Hesselink, Res. Policy, 1996, 25, 819. 7 J. Reedijk, Chem. Ind., 1993, 690; idem, ibid., 1997, 288. 8 (a) I. A. Williams, Chem. Br., 1996, 32, (February), 31; (b) P. J. Lachmann and J. S. Rowlinson, Sci. Public AÜairs, 1997, Winter, 8. 9 (a) J. Reedijk, Inorg. Chim. Acta, 1992, 198ñ200, 873; (b) J. Reedijk, Inorg. Chim. Acta, 1992, 198, 873; (c) J. Reedijk, Inorg. Chim. Acta, 1992, 200, 873. 10 (a) A. P. Schubert, Inorg. Chim. Acta, 1996, 253, 111; (b) A. P. Schubert and G. A. Schubert, Inorg. Chim. Acta, 1997, 266, 125. 11 H. F. Moed and M. Vriens, J. Inform. Sci. 1989, 15, 95. 12 Second EU Report on Science and T echnology Indicators, European Union Brussels, 1998. 13 C. F. Huber, DATABASE, 1995 (April), 18, 52; ST News, 1988, 14, (May), 12. 14 A. F. J. van Raan, Scientometrics, 1996, 36, 397. Received in Montpellier, France, 6th April 1998; L etter 8/02808G 770 New J. Chem., 1998, Pages 767»770

ISSN:1144-0546    

DOI:10.1039/a802808g

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

L e t t e r N H O C O OCnH2 n+1 n-C10H21 = S S S nSzTT Extended mesogen n = 4,8,12 or 18 nSzT Novel supramolecular liquid-crystalline complexes derived from 5-n-decyl-substituted thieno [ 3,2-b] thiophene-2- and thiophene-2-carboxylic acids Hsi-Hwa Tso,*,a Jy-Shih Wang,a Chin-Yi Wub and Hong-Cheu Lin*,a a Institute of Chemistry, Academia Sinica, Nankang, T aipei, T aiwan 115, Republic of China b Institute of Applied Chemistry, Chinese Culture University, T aipei, T aiwan, Republic of China A convenient route to thieno[3,2-b]thiophene-2-carboxyaldehyde (1) for the synthesis of new 5-n-decylthieno[3, 2-b]thiophene-2-carboxylic acid (10) is described.A class of novel supramolecular liquid-crystalline complexes are formed starting from 10 or 5-n-decylthiophene-2-carboxylic acids and stilbazoles, through intermolecular hydrogen bonding.The synthesis and characterization of hydrogen-bonded liquid crystals have recently drawn increasing attention.1 By way of the molecular recognition and self-assembly of suitably functionalized similar or dissimilar molecules through hydrogen bonding, a variety of supramolecular liquid crystals can be constructed simply and spontaneously.Although many supramolecular liquid crystals formed by interaction of various hydrogen donors and acceptors have been reported, the use of heterocyclic carboxylic acid as a donor in this system hitherto remains unexplored. Here we report the preparation and characterization of a novel class of thieno[3,2-b]thiophene- and thiophene-containing supramolecular liquid-crystalline complexes (Fig. 1) where the 5-n-decylthieno[3,2-b]thiophene-2- or 5-n-decylthiophene-2-carboxylic acid2 and trans-4-alkoxy-4@- stilbazoles3 are used as building components. A convenient route to thieno[3,2-b]thiophene-2-carboxyaldehyde (1)4 from the available methyl 3-hydroxy-5,6- dihydrothieno[3,2-b]thiophene-2-carboxylate (2)7 was developed for the synthesis of the required 5-n-decylthieno[3,2-b] thiophene-2-carboxylic acid (10) (Scheme 1).Treatment of 2 (24.4 mmol with methanesulfonyl chloride (MsCl, 29.7 mmol) and triethylamine (64.5 mmol.) in (100 mL) at room CH2Cl2 temperature gave the mesylate 3§ as a solid in 98% yield after puri–cation by —ash chromatography (silica gel, hexane»ethyl acetate, 3 : 1). Reductive removal of the methanesulfonyl group from 3 to aÜord 5 was accomplished by treatment of 3 (7.2 mmol) with sodium iodide (36 mmol) and Zn»Ag (72 Fig. 1 mmol) in anhydrous DMSO (20 mL) under nitrogen at 80 °C for 26 h, presumably via the iodo intermediate 4.8 After –ltration, extraction and puri–cation by —ash chroma- (CHCl3), tography (silica gel, hexane»ethyl acetate, 15 : 1), methyl 5,6- dihydrothieno[3,2-b]thiophene-2-carboxylate (5)§ was obtained as a solid in 80% yield.Reaction of 5 with LiAlH4 (2.5 equiv.) in anhydrous diethyl ether aÜorded the crude alcohol 6§, which was directly treated with an excess of MnO2 (10 equiv.) in by stirring at room temperature for 24 h CH2Cl2 giving the aldehyde 1 in 91% overall yield from 5. A smooth oxidation of the alcohol functionality as well as the dihydrothiophene ring occurred in one —ask.The spectral data of 1 are identical with those previously reported.6 Synthesis of the new hydrogen-bonded donor 10 is straightforward as shown in Scheme 1 by sequential protection, alkylation and deprotection reactions wherein the crude intermediates 7 and 8 were used without puri–cation.9 Thus to a stirred solution of 76 (1.88 mmol) in anhydrous THF (50 mL) and haxamethylphosphoramide (HMPA, 12.2 mmol) at [78 °C was added BunLi (1.98 mmol) dropwise.The reaction mixture was stirred at the same temperature for 1 h prior to the addition of n-iododecane (2.12 mmol). After stirring for another 1 h, the resultant mixture was allowed to warm gradually to [60 °C and quenched with aqueous THF. The crude product 8 was obtained by eluting the reaction mixture through a silica gel column (hexane»ethyl acetate, 5 : 1) to remove HMPA. Further treatment of 8 with a catalytic amount of tosylic acid (0.2 equiv.) in re—uxing acetone for 4 h gave 9§ as a solid after puri–cation by —ash chromatography (silica gel, hexane»ethyl acetate, 35 : 1) (87% overall yield from 1). It is noteworthy that the use of HMPA (6.5 equiv.) as a § Compounds: 3: mp 66»67 °C; 1H NMR d 3.26 (t, 2 H, J\8 Hz), 3.34 (s, 3 H), 3.77 (t, 2 H, J\8 Hz), 3.86 (s, 3 H) ; MS m/z 294 (M`, 100), 216, 215, 187, 184, 144, 100, 71.Exact mass calcd for 293.9690; found: 293.9689; 5: mp 55.5»56 °C; 1H NMR C9H10O5S3 : d 3.23 (t, 2 H, J\8 Hz), 3.77 (t, 2 H, J\8 Hz), 3.86 (s, 3 H), 7.46 (s, 1 H); MS m/z 200 (M`, 100), 169, 141, 140, 97, 69.Exact mass calcd for 199.9966; found: 199.9971; crude 6: 1H NMR d 1.72 (t, 1 C8H8O2S2 : H, J\5.9 Hz), 3.17 (t, 2 H, J\8 Hz), 3.74 (t, 2 H, J\8 Hz), 4.71 (d, 2 H, J\5.9 Hz), 6.72 (s, 1 H) ; 9: mp 40.5»41 °C; 1H NMR d 0.88 (t, 3 H, J\6.8 Hz), 1.28 (br m, 14 H), 1.73. (m, 2 H), 2.91 (t, 3 H, J\7.4 Hz), 7.01 (s, 1 H), 7.84 (s, 1 H), 9.91 (s, 1 H) ; MS m/z 308 (M`), 183, 182, 181 (100), 153.Exact mass calcd for 308.1269; C17H24OS2 : found: 308.1269; 10: 1H NMR d 0.88 (t, 3 H, J\7.1 Hz), 1.28 (br m, 14 H), 1.73. (m, 2 H), 2.91 (t, 3 H, J\7.6 Hz), 6.99 (s, 1 H), 7.99 (s, 1 H); MS m/z 324 (M`), 199, 198, 197 (100), 175, 153, 149, 121, 90, 43, 41. Exact mass calcd for 324.1218; found: 324.1231. C17H24O2S2 : New J. Chem., 1998, Pages 771»773 771S S CO2Me OH S S CO2Me OMs S S CO2Me I S S CO2Me S S CH2OH S S CHO S S S S S S CHO S S CO2H n-C10H21 n-C10H21 n-C10H21 O O O O 2 3 4 1 6 5 7 8 9 10 ( i) (ii) ( iii) ( iv) ( v) ( vi) ( vii) ( viii) Scheme 1 Reagents and conditions (i) MSCl, 25 °C; Et3N, CH2Cl2 , (ii) NaI, Zn»Ag, DMSO, 80 °C; (iii) 25°C;(iv) LiAlH4, Et2O, MnO2 , 25 °C; (v) TsOH, re—ux; (vi) BunLi, CH2Cl2 , HOCH2CH2OH, C6H6 , THF, HMPA, [78»60 °C; (vii) TsOH, acetone, re—ux; n-C10H21I, (viii) O»25 °C Ag2O, Et2OwMeOH, cosolvent for the alkylation of 7 is crucial. In the absence of HMPA, less than 10% of the alkylated product 8 was obtained under the reaction conditions.Oxidation of 9 was carried out by treatment of an ethereal solution of 9 (1.2 mmol in 40 mL ether) with the vigorously stirred semisolid Ag2O10 (prepared from 2.85 mmol aq.and 5.61 mmol aq. AgNO3 NaOH) at 0 °C for 10 min prior to the addition of 20 mL MeOH. The reaction mixture was stirred at room temperature for another 3 h. The silver suspension was removed by –ltration and washed with hot water. The cold –ltrate was acidi –ed with concentrated hydrochloric acid precipitating the acid 10§ (90% yield), which was further puri–ed by recrystallization from 95% EtOH.The two novel classes of hydrogen-bonded liquid-crystalline mesogens nSzTT and nSzT (Fig. 1) containing three diÜerent aromatic rings within the extended rigid core are simply prepared by the slow evaporation of a tetrahydrofuran solution of 10 (TT) or 5-n-decylthiophene-2-carboxylic acid (T) and trans-4-alkoxy-4@-stilbazole in a 1 : 1 molar ratio.Their phasetransition temperature ranges are listed in Table 1. All new hydrogen-bonded complexes nSzTT and nSzT (n\4, 8, 12 or 16) are found to behave as one single component and exhibit diÜerent mesomorphic properties from their original constituents. As indicated in the Table 1, the new hydrogen donor 10 (TT) exhibits both smectic C and nematic (N) phases.11 T rans-4-alkoxy-4@-stilbazoles (nSz, n\4, 8, 12 or 16) are known to have narrow ranges of smectic B and E phases3b while the 5-n-decylthiophene-2-carboxylic acid (T) itself is nonmesogenic.2 However, only the smectic A smectic C and two unidenti–ed smectic X and (SA), X@ and phases are observed for complexes nSzTT and (SX SX{) nSzT, and the nSzTT (n\4, 8, 12 or 16) series possess neither Table 1 Phase-transition temperaturesa of hydrogen-bonded complexes nSzTT and nSzT from a 1 : 1 molar ratio of trans-4-alkoxy-4@-stilbazoles (nSz) with 5-n-decylthieno[3,2-b]thiophene 2-carboxylic acid (TT) or 5-n-decylthiophene-2-carboxylic acid (T) a Phase-transition temperatures (°C) and the corresponding enthalpies (J g~1, in parentheses) were determined by the second heating and cooling scans (heating and cooling rate of 10 °C min~1) of diÜerential scanning calorimetry using a Perkin Elmer DSC-7 calorimeter ; abbreviations : K and K@\crystalline phases, smetic phase, I\isotropic liquid.b The enthalpy was too small to be detected by DSC and the SX\unidenti–ed phase-transition temperature was assigned by polarizing optical microscopy. 772 New J. Chem., 1998, Pages 771»773the nematic nor the smectic B and E phases. In contrast to the pure dimeric acids TT and T, the smectic A phase is apparently introduced in both complexes nSzTT and nSzT by the hydrogen-bonded interaction between the acid and stilbazole components. In addition, this interaction suppresses the nematic phase displayed by the dimeric acid TT.It is noted that the ranges of smectic A phase become wider for both nSzTT and nSzT bearing shorter alkoxy tails of the stilbazole moieties, and the smectic A phase disappears (replaced by the smectic C phase) for those with longer alkoxy chains (12SzTT, 16SzTT, 12SzT and 16SzT). Signi–cantly, the tilted layered (smectic C) phase is stabilized (i.e. with a wider range of the SC phase) by the binary extended core, and the complex 8SzTT has the widest range of the phase (D68 °C) in both series.SC Such a trend is consistent with that of similar benzenoid homologues of thermotropic mesogens.12 It is also interesting that the new supramolecular liquidcystalline complexes nSzT having a kinked molecular structure exhibit stable mesophases similar to those of the more rod-like complexes nSzTT, although with relatively narrow mesophase ranges and lower transition temperatures.This phenomenon is comparable with that of similar benzenoid supramolecular liquid-cystalline complexes having an angular structure.13 It might be attributable to the reduced packing efficiency of the nonlinear con–guration in the nSzT series. In summary, we report herein a convenient synthesis of thieno[3,2-b]thiophene-2-carboxyaldehyde (1) as the precursor to 5-n-decylthieno[3,2-b]thiophene-2-carboxylic acid (10).Novel supramolecular liquid crystals nSzTT and nSzT are readily formed from 10 or 5-n-decylthiophene-2-carboxylic acid and stilbazoles through intermolecular hydrogen bonding. The present results suggest that the use of suitable heterocyclic acids as hydrogen-bonded donors could be of value in the design of new supramolecular liquid crystals.Acknowledgements thank the National Science Council of the Republic of We China for –nancial support (Grant NSC 87-2113-M-001-025). References 1 For a recent review, see C. M. Paleos and D. Tsiourvas, Angew. Chem., Int. Ed. Engl., 1995, 34, 1696. 2 H. Wynberg and A.Logothetis, J. Am. Chem. Soc., 1956, 78, 1958. 3 (a) D. W. Bruce, D. A. Dunmur, E. Lalinde, P. M. Maitlis and P. Styring, L iq. Cryst., 1988, 3, 385. (b) H. C. Lin, L. L. Lai, Y. S. Lin, C. Tsai and R. C. Chen, Mol. Cryst. L iq. Cryst., 1998, in the press. 4 The established routes to 1 involving either formylation of thieno[3,2-b]thiophene5 or intramolecular cyclization of a 2,3-substituted thiophene6 suÜered the limitations of low overall yield and/or delicate synthetic operations. 5 Y. A. L. Golœdfarb, V. P. Litvinov and S. Ozolin, Izv. Akad. Nauk SSSR. Ser. Khim., 1965, 510; V. P. Litvinov and Y. A. L. Golœdfarb, Chemistry of T hienothiophenes, in Advanced Heterocyclic Chemistry, eds. A. R. Katritzky and A. J. Boulton, Academic Press, New York, 1976, 19, 123. 6 J. D. Prugh, G. D. Hartman, P. J. Mallorga, B. M. McKeever, S. R. Michelson, M. A. Murcko, H. Schwam, R. L. Smith, J. M. Sondey, J. P. Springer and M. F. Sugrue, J. Med. Chem., 1991, 34, 1805. 7 R. Donoso, P. J. de Urries and J. Lissavetzky, Synthesis, 1992, 526. 8 Y. Fujimoto and T. Tatsuno, T etrahedron L ett., 1976, 3325. The use of Zn»Ag instead of the zinc powder in our system greatly improved the yield of 5 (from 55% to 80%). 9 Ketals 7 and 8 are stable enough on silica gel and can be isolated by —ash chromatography (silica gel, EtOAc»hexane, 1 : 25 for 7, 1 : 35 for 8). 10 E. Campaigne and M. LeSuer, Org. Syntheses Coll. V ol. IV , 1963, 919. 11 For several carboxylic acids with mesogenic properties due to the dimeric acid association through hydrogen bonding, see ref. 1. 12 L. J. Yu, L iq. Cryst., 1993, 14, 1303. 13 M. Willis, K. J. Price, H. Adams, G. Ungar and D. W. Bruce, J. Mater. Chem., 1995, 5, 2195; D. J. Price, K. Willis, T. Richardson, G. Ungar and D. W. Bruce, J. Mater. Chem., 1997, 7, 883; T. Kato, H. Adachi, A. Fujishima and J. M. J. Frechet, Chem. L ett., 1992, 265; K. Willis, J. E. Luckhurst, D. J. Price, J. M. J. Frechet, J. Kihara, T. Kato, G. Ungar and D. W. Bruce, L iq. Cryst., 1996, 21, 585. Received in Cambridge, UK, 27th February 1998; L etter 8/03315C New J. Chem., 1998, Pages 771»773 773

ISSN:1144-0546    

DOI:10.1039/a803315c

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

L e t t e r R N COOMe R NH COOMe CH2=CHCH2Br,Zn CeCl3•7H2O 1a R = H 1b R = OTs 1c R = OCH2Ph 2a R = H 2b R = OTs 2c R = OCH2Ph Asymmetric reactions on polymers: diastereoselective allylation of polymer-supported chiral imines Shinichi Itsuno,*,a Ashraf A. El-Shehawy,î,a Magdy Y. Abdelaalb and Koichi Itoa a Department of Materials Science, T oyohashi University of T echnology, T oyohashi, 441-8580 Japan b Chemistry Department, Faculty of Science, Mansoura University, Mansoura, Egypt Reaction of allylzinc reagent with enantiopure imine species attached to polystyrene proceeded smoothly with perfect diastereoselectivity and in excellent yield to aÜord polymer-supported chiral homoallylic amines. Polymers having chiral pendant groups have been of great interest owing to their applications in enantiomer separation technology and in polymeric chiral catalysts or reagents in asymmetric synthesis.1 Generally these chiral polymers are prepared by the following two methods.1d One involves a chemical modi–cation of preformed reactive polymer with enantiomerically pure compounds.Copolymerization of enantiopure monomers with some achiral monomers is the alternative method.However, rather straightforward methodology is available if an asymmetric reaction can be applied directly to the prochiral pendant functionality of the polymer to create a new stereogenic center on the polymer. This method has not been studied extensively mainly owing to the requirement of a highly stereoselective and quantitative reaction and a suitable analytical method to evaluate the optical purity of the chiral moieties on the polymer.2 In chemical transformations of polymeric functionality, it is impossible to remove any side product yielded in the polymer.Stereoisomeric impurities also can not be eliminated once the reactions have been completed on the polymer. A highly stereoselective and quantitative reaction is therefore required to achieve this methodology. Another problem pointed out in the literature is that a lowering of the stereoselectivity usually accompanies the polymer reaction compared to the corresponding reaction with the low-molecular-weight counterpart.3 Recent vigorous developments in asymmetric reactions allowed us to apply some highly stereoselective reactions to the direct asymmetric transformation of polymeric functionalities.For example, Umani- Ronchi and co-workers4 have developed an excellent methodology for diastereoselective allylation of chiral imines. Here we have chosen this reaction to investigate diastereo- * Fax: ]81 532 44 6813; E-mail: itsuno=tutms.tut.ac.jp î Current address : Chemistry Department, Faculty of Education, Tanta University, Kafr El-Sheikh, Egypt.selectivity on the polymeric substrate and have shown that polystyrene having diastereomerically pure pendant groups can be prepared by this new method. According to the Umani-Ronchi»Savoia method, benzaldimine 1a was smoothly converted into (S, S)-homoallylamine 2a in quantitative yield with excellent diastereoselectivity (Table 1, run 1) by using the allylzinc reagent prepared from allyl bromide (3-bromopropene), zinc powder and as additive (Scheme 1).4a Since we CeCl3 … 7H2O chose a sulfonate (OTs) or a benzyl ether linkage between the chiral ligand and polystyrene, the corresponding model compounds such as 1b and 1c were prepared to examine their reactivity and selectivity in the allylation reaction.Table 1 shows that allyl reacted with both bromide»Zn»CeCl3 … 7H2O 1b and 1c to give the homoallylic amines 2b and 2c, respectively, with almost perfect diastereoselectivities and excellent isolated yields (run 2, 6).5 Instead of is CeCl3 … 7H2O, SnCl2 another choice of additive in a highly diastereoselective allylation (run 3).The reactions of the prenylzinc reagent formed from prenyl bromide (4-bromo-2-methylbut-2-ene) were also efficient and highly diastereoselective with the same imines (run 4, 5, 7, 8).The above results encouraged us to apply this reaction to the synthesis of a polymer-supported chiral ligand by using direct transformation of the prochiral functionality in the Scheme 1 Diastereoselective allylation of chiral imine Table 1 Diastereoselective allylation of iminesa Run R Allylating agent Time/h Yield/% Diastereoselectivity (de)/%b 1c H CH2 xCHCH2Br»Zn»CeCl3 … 7H2O 0.5 100d [99 2 OTs CH2xCHCH2Br»Zn»CeCl3 … 7H2O 4 93e 100 3 OTs CH2xCHCH2Br»Zn»SnCl2 0.25 96e 100 4 OTs Me2CxCHCH2Br»Zn»CeCl3 … 7H2O 2 91e 100 5 OTs Me2CxCHCH2Br»Zn»SnCl2 3 94e 100 6 OCH2Ph CH2xCHCH2Br»Zn»CeCl3 … 7H2O 12 89e [99 7 OCH2Ph Me2CxCHCH2Br»Zn»CeCl3 … 7H2O 2 93e 100 8 OCH2Ph Me2CxCHCH2Br»Zn»SnCl2 1 94e 100 a Reaction conditions : imine»bromide»Zn»salt\1.0 : 1.2 : 2.0 : 0.15 (mmol), THF, 25 °C.b Determined by 1H NMR analysis. c See ref. 4(a). d Conversion determined by GC. e Isolated yield. New J. Chem., 1998, Pages 775»777 775S O O O N CO2 Me S O O O N CO2 Me + Styrene ( i ) 91% 3 0.90 0.10 ( ii) 97% 4 P O2S O NH CO2Me 5 >99% de P O2S O NH OH 6 98% ( iii) ( iv) 95% P O2S O NH2 HO NH2 ( v) 95% 7 8 >99% ee P : polystyrene N COOMe O + Styrene 9 ( i) 95% N COOMe O P 10 ( ii) NH COOMe O P R R NH COOMe R R HO ( iii) 11a R = H 94%, >99% de 11b R = Me 98%, 100% de 12a R = H 93%, >99% de 12b R = Me 96%, 100% de polymer.Enantiopure imine monomers 3 and 9 were thus prepared and polymerized with styrene under radical polymerization conditions to give 4 and 10, respectively.These polymers were then allylated with the allylzinc reagent. As shown in Scheme 2 and 3, polymers 5 and 11 were obtained in excellent yield with almost perfect diastereoselectivity ([99% de). In these polymer reactions, no lowering of the diastereoselectivity was observed compared to the model reactions described above.It should be also emphasized that the polymers possessing an allylic hydrogen such as 5 and 11a were not successfully synthesized by polymerization of the corresponding chiral monomers, since such monomers caused degradative chain transfer during the radical polymerization. The prenylzinc reagent also reacted with 10 to aÜord 11b in excellent yield and selectivity (Scheme 3).6 The diastereoselectivities in these polymeric reactions could easily be con- –rmed by the 1H NMR spectra of the polymeric products (5, 11).7 In order to obtain more precise information about the chiral pendant group on the polymers 5 and 11, we examined some cleavage reactions.Since various attempts to cleave the Scheme 2 Reagents and conditions : (i) AIBN (2,2@-azobisisobutyronitrile), 48 h; (ii) C6H5Me, 80°C, CH2xCHCH2Br»Zn»CeCl3…7H2O, THF, r.t., 4 h; (iii) THF, \0 °C, 3h; (iv) LiAlH4, H5IO»MeNH2 , MeOH»THF, r.t., 1 h; (v) KOH»DMF Scheme 3 Reagents and conditions : (i) AIBN, 80 °C, 48 h; C6H5Me, (ii) or THF, CH2xCHCH2Br Me2CxCHCH2Br»Zn»CeCl3 … 7H2O, r.t., 4 h; (iii) THF H2/PdwC, sulfonate linkage in 5 gave a mixture of undesired side products, we took the somewhat devious route shown in Scheme 2.This stepwise cleavage, however, proceeded smoothly on the polymer to give homoallylamine 8 which is structurally identical to that derived from the model compound 2b. The enantioselectivity of 8 was determined to be [99% ee by HPLC analysis.8 On the other hand, the cleavage of the benzyl ether linkage in 11 could be achieved in a single hydrogenolysis step to aÜord 12 in high chemical yield (Scheme 3). 1H NMR and HPLC analysis of 12 revealed [99% de and 100% de for 12a and 12b, respectively. These diastereoselectivities of the cleavage products are exactly equal to those obtained in the polymers 5 and 11. In conclusion, we have developed a new method of asymmetric transformation on the polymeric prochiral functionality, which allowed us to synthesize novel polymers having a chiral pendant group.Particularly striking are the excellent results with the allylzinc reagent where the chiral polymers were obtained in quantitative yield with perfect stereoselectivity. The obtained polymer-supported chiral amino acids and amino alcohols are almost free of side products including stereoisomeric structures, and are –nding extensive applications in catalytic asymmetric synthesis.Experimental Asymmetric allylation of the polymeric imine 4 To a stirred suspension of Zn powder (0.13 g, 2 mmol) in THF (5 ml), was added (56 mg, 0.5 mmol) at 0 °C. A CeCl3 … 7H2O THF (10 ml) solution of 4 (1 mmol) and allyl bromide (0.15 g, 1.2 mmol) were then added. After 4 h at room temperature, the mixture was diluted with (50 ml), quenched with CHCl3 NaOH aqueous solution (2M, 10 ml) and stirred for 5 min.The organic layer was separated, washed with brine and concentrated at ca. 20 ml under reduced pressure. The resulting polymer solution was then precipitated into methanol to give a –ne powder of 5. 97% yield, Mn\42 000, Mw : Mn\2.0. 1H NMR (270 MHz): peaks attributed to polystyrene backbone, d 7.2»6.2 (Ar), 2.2»1.2 (CH, peaks attributed to CH2) ; the chiral pendant moiety having (S, S) con–guration, d 5.75 776 New J.Chem., 1998, Pages 775»777(1H, 5.1 (2H, 3.7, (3H, 3.5 CH2xCH), CH2\CH), CO2CH3), (1H, ArwCHwN), 2.7 (1H, CHwCO), 0.95 [6H, (CH3)2CH]. No peak at 3.0 ppm assignable to the (R, S) isomer was detected from the NMR spectrum. Acknowledgements work was partially supported by a Grant-in-Aid for This Scienti–c Research from the Ministry of Education, Science, Sports and Culture, Japan and The Mitsubishi Petrochemical (Mitsubishi Chemical) Foundation.The Yamanouchi Award in Synthetic Organic Chemistry, Japan (to S. I.), is also gratefully acknowledged. References 1 For reviews on polymer-supported reagents and catalysts, see : (a) S.Itsuno, in Polymeric Materials Encyclopedia: Synthesis, Properties and Applications, ed. J. C. Salamone, CRC Press, Boca Raton, FL, 1996, vol. 10, pp. 8078»8087; (b) S. Itsuno, in Macromolecules 1992, ed. J. Kahovec, VSP, Utrecht, 1993, pp. 413»422; (c) E. C. Blossey and W. T. Ford, in Comprehensive Polymer Science, ed. G. Allen, Pergamon, Oxford, 1989, vol. 6, pp. 81»114; (d) Synthesis and Separations Using Functional Polymers, ed. D. C. Sherrington and P. Hodge, Wiley, Chichester, 1988. 2 M. Cernerud, J. A. Reina, J. Tegenfeldt and C. Moberg, T etrahedron: Asymmetry, 1996, 7, 2863. 3 (a) M. Calmes, J. Daunis, R. Jacquier, G. Nkusi, J. Verducci and P. Viallefont, T etrahedron L ett., 1986, 4303; (b) M. Kawana and S. Emoto, T etrahedron L ett., 1972, 4855; (c) A.R. Colwell, L. R. Duckwall, R. Brooks and S. P. McManus, J. Org. Chem., 1981, 46, 3097. 4 (a) T. Basile, A. Bocoum, D. Savoia and A. Umani-Ronchi, J. Org. Chem., 1994, 59, 7766; (b) A. Bocoum, D. Savoia and A. Umani- Ronchi, J. Chem. Soc., Chem. Commun., 1993, 1542; (c) A. Bocoum, C. Boga, D. Savoia and A. Umani-Ronchi, T etrahedron L ett., 1991, 32, 1367. 5 Diastereoselectivity was determined by 1H NMR and HPLC analysis according to the Umani-Ronchi method for the allylation of benzaldimine.4a The con–guration of the obtained homoallylamines 2b and 2c is estimated to be (S, S) from the results obtained from the allylation of benzaldimine. 6 The polymer 11b was alternatively prepared by copolymerization of styrene with the corresponding chiral monomer and showed an identical 1H NMR spectrum to that obtained in Scheme 3. 7 Methine protons a to the tester group in 5, 11a and 11b appear at 2.70, 2.75 and 2.80 ppm, respectively, for the (S, S) isomer. Peaks for the (R, S) isomers appear at around 3.0 ppm. 8 Enantiomeric excesses were determined by HPLC using a chiral stationary phase [Daicel Chiralcel OD-H, hexane»isopropyl alcohol»diethylamine (90 : 10 : 0.1)]. Received in Cambridge, UK, 9th March 1998; L etter 8/03375G New J. Chem., 1998, Pages 775»777 777

ISSN:1144-0546    

DOI:10.1039/a803375g

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

L e t t e r N N N N OH HO HO But But Me Me Me Me OH But 2 1 But Tetraaza [ 1.1.1.1 ]metacyclophane Akihiro Ito,* Yukiharu Ono and Kazuyoshi Tanaka Department of Molecular Engineering, Graduate School of Engineering, Kyoto University, Sakyo-ku, Kyoto 606-8501, Japan The facile preparation of the title compound has been achieved using Pd(0)-catalysed CwN coupling reaction of 3-bromo-N-methylaniline ; X-ray crystallography shows that the molecule adopts an approximate S4 conformation in the crystalline state.In 1944 Zinke and Ziegler reported the preparation of a cyclic tetrameric arene structure (1) from the reaction of parasubstituted phenols and formaldehyde,1 and this assignment has gained acceptance as the result of X-ray crystallographic studies by Andreetti et al.2 Moreover, since the discovery of a facile selective synthesis method by Gutsche et al.,3 a large number of the called calixarenes3 have [1n]metacyclophanes been synthesised and investigated mainly in the –eld of molecular recognition.4 Furthermore, in the last few years, several research groups have reported on the preparation of the heteroatom-bridged In spite of the strong [1n]cyclophanes.5 affinity of amino groups for a variety of metal ions, however, azacalixarenes have not yet been prepared, probably owing to the paucity of existing methods for the preparation of arylamines, and only in which phenol units [3n]metacyclophanes, are connected with dimethylene aza bridges, are so far known.6 Recently, Buchwald and co-workers have reported a simple catalytic method for the conversion of aryl bromides to arylamines, superseding the laborious Ullmann coupling reaction.7 Since then, several complicated oligo- and polyarylamine compounds have been successfully prepared on the basis of this method.8 In this context, this method has been applied to the synthesis of aza-bridged simply using [1n]metacyclophanes bromo-substituted N-methylanilines as a starting material, and this attempt gave an affirmative result.Here we report the –rst synthesis of N-methyl-substituted tetraaza[ 1.1.1.1]metacyclophane (2)§ having the prototypical macrocyclic structure of azacalix[4]arene. Scheme 1 * E-mail: aito=scl.kyoto-u.ac.jp § CAS name for 2: 2,8,14,20-tetramethyl-2,8,14,20- tetraazapentacyclo[19.3.1.13,7.19,13.115,19]octacosa-1(25),3,5,7(28),9,11, 13(27),15,17,19(26),21,23-dodecaene. A mixture of 3-bromo-N-methylaniline (2.0 g, 11 mmol), NaOBut (1.4 g, 15 mmol) and (0.17 g, [PdCl2(P(o-tolyl)3)2] 0.22 mmol) in toluene (90 ml) was heated under an Ar atmosphere at 100 °C for 19 h, following the reported procedure.7 After the usual workup, the crude product was chromatographed on silica gel using (1 : 1) as the CH2Cl2»hexane eluent.Throughout the above treatment, the black intractable polymeric solid was removed. After the –rst fraction (Rf\ of P(o-tolyl) the second fraction aÜorded a 0.85) 3, (Rf\0.65) white solid (0.14 g). In the 13C NMR spectrum, only –ve peaks were observed, indicating a high molecular symmetry of the isolated compound. Moreover, the FAB-mass spectrum showed an M` peak at m/z 420.2313 (calc. 420.2314), corresponding to 2.î The isolated yield was 12.4%. It is interesting to note that the main product of this reaction is cyclic tetramer 2, whilst considering the following : (i) in the preparation of some cyclic oligoarenes including calix[n]arene,9 the tetramer tends to be selectively prepared and (ii) according to the reaction mechanism proposed by Buchwald and co-workers,7 the reaction proceeds via formation of an intermediate PdII complex coordinated with both ends of a linear oligoaniline in the last stage of cyclization.The 1H NMR spectrum (270 MHz, of the third fraction 0.26 g), on the CDCl3) (Rf\0.45 ; other hand, indicated that there was either a mixture of the aza-bridged or a linear oligomer. The [1n]metacyclophanes GC/MS study showed this fraction to contain three major components.Each component had an M` peak assignable to 315, 420 and 525, respectively.° This –nding can be rationalised only by the presence of the aza-bridged (n\3, 4 and 5). However, attempts to [1n]metacyclophanes isolate the aza-bridged (n\3 and 5) from [1n]metacyclophane this mixture have so far met with failure.Colourless plates were obtained by slow evaporation of a dilute solution of 2 at room temperature, and X-ray CH2Cl2 structure analysis unequivocally established the structure of î Selected data for 2: m.p. 280 °C (decomp); FAB-MS (NBA matrix) : m/z calcd for 420.2314; found: 420.2313; (270 MHz, C28H28N4: dH 3.19 (12 H, s), 6.41 (4 H, t, J\2.2 Hz), 6.54 (8 H, dd, J\7.9, CDCl3) 2.2 Hz), 7.26 (4 H, t, J\7.9 Hz); (67.5 MHz, 40.23, 112.40, dC CDCl3) 113.75, 130.74, 149.43 ; FTIR: m(KBr)/cm~1 2927, 2813, 1601, 1572, 1492, 1357, 1240, 1127, 846, 775 and 701; UV/VIS: (log e) 257 (4.60) and 288 (4.47). kmax(cyclohexane)/nm ° The M` peaks (m/z : 630, 735 and 840) of three minor components with longer retention time corresponding to the aza-bridged (n\6»8) were also observed.[1n]metacyclophane New J. Chem., 1998, Pages 779»781 779Fig. 1 (a) Molecular structure of 2 with atomic numbering scheme and (b) the side view. Selected dihedral angles/° : C2wC1wN4wC23 [122.8(6), C4wC5wN1wC25 2.5(9), C5wN1wC7wC8 104.2(7), C10wC11wN2wC26 [6.7(9), C11wN2wC13wC14 [118.0(6), C16wC17wN3wC27 [20(1), C17wN3wC19wC20 102.5(8), C22wC23wN4wC28 14.9(9) 2.“ Fig. 1 shows two views of the conformation of 2. It is noteworthy that 2 adopts an approximate conformation in S4 which each benzene ring alternately inclines upward and downward to the macrocyclic plane de–ned by atoms C6, C12, C18 and C24. The two opposite benzene rings de–ned by atoms C1 and C13 (C7 and C19) make a cavity ; the dihedral angle between the two opposite least-square planes is 79° (66°).Interestingly, each amino group adopts a planar conformation and, furthermore, the least-square planes of the amino groups de–ned by atoms N1, N2, N3 and N4 are almost coplanar with those of the benzene rings de–ned by atoms C1, C7, C13 and C19, respectively, enabling good p-conjugation to be maintained within the N-methylaniline moiety. As shown in Fig. 2, the molecules 2 are stacked so as to occupy each “ Single-crystal structure determinations : C28H28N4 , Mr\420.56, colourless plate 0.57]0.21]0.15 mm, triclinic, space group (no.P1 6 2), a\10.310(5), b\11.443(4), c\10.091(6) a\95.48(4), Aé , b\107.04(4), c\84.41(3)°, U\1130.0(9) Z\2, g Aé 3, DC\1.236 cm~3, l(Mo-Ka)\0.74 cm~1, T \296 K. Rigaku AFC7R diÜractometer, Mo-Ka radiation (k\0.71069 The structure Aé ), 2hmax\55°.was solved by direct methods (SIR88)10 and re–ned by full-matrix least squares on F2 with all non-H atoms assigned anisotropic displacement parameters. All H atoms were located from *F syntheses ; thereafter only isotropic B values were re–ned. R\0.078, Rw\0.085 for 2346 independent observed re—ections [I[2r(I)]. Supplementary data available from the Cambridge Crystallographic Data Centre, reference number 440/042.Fig. 2 Packing diagram of 2 otherœs cavities in the crystal. This suggests that the molecules 2 can form a closed cavity which encapsulates a suitable guest molecule. The cyclic voltammogram (CV) of 2 in MeCN exhibited irreversible three oxidation waves at ]0.38, ]0.44 and ]0.71 V (0.1 mol dm~3 in MeCN, Pt electrode, 25 °C, TBAClO4 scan rate 100 mV s~1, V vs.Fc/Fc`). After several sweeps, a new oxidation wave (]0.83 V) appeared in place of the former three peaks and, furthermore, the intensity of the peak increased throughout the multisweep experiment, indicating polymerization of 2. In summary, we have found that 2 is obtained in a one-pot reaction of 3-bromo-N-methylaniline using Buchwaldœs method.Aza-bridged like 2 could be a [1n]metacyclophanes new class of the host molecules associated with molecular recognition chemistry. We are currently tackling the isolation of cyclic oligomers other than 2 and investigating the molecular recognition properties in these molecular systems. Acknowledgements work is a part of the project of Institute for Fundamental This Chemistry, supported by Japan Society for the Promotion of Science-Research for the Future Program (JSPSRFTF96P00206). References 1 A.Zinke and E. Ziegler, Chem. Ber., 1944, 77, 264. 2 G. D. Andreetti, R. Ungaro and A. Pochini, J. Chem. Soc., Chem. Commun., 1979, 1005. 3 C. D. Gutsche, B. Dhawan, K. H. No and R. Muthukrishnan, J. Am. Chem. Soc., 1981, 103, 3782. 4 R. M.Izatt, J. S. Bradshaw, K. Pawlak, R. L. Bruening and B. J. Tarbet, Chem. Rev., 1992, 92, 1261; S. Shinkai, T etrahedron, 1993, 49, 8933; V. Boé hmer, Angew. Chem., Int. Ed. Engl., 1995, 34, 713; A. Ikeda and S. Shinkai, Chem. Rev., 1997, 97, 1713. 5 J. Nakayama, N. Katano, Y. Sugihara and A. Ishii, Chem. L ett., 1997, 897; H. Kumagai, M. Hasegawa, S. Miyanari, Y. Sugawa, Y. Sato, T.Hori, S. Ueda, H. Kamiyama and S. Miyano, T etrahedron L ett., 1997, 38, 3971; M. Mascal, J. L. Richardson, A. J. Blake and W.-S. Li, T etrahedron L ett., 1997, 38, 7639; A. S. Abd-El-Aziz, C. R. de Denus, M. J. Zaworotko and C. V. K. Sharma, Chem. Commun., 1998, 265; I. Baxter, H. M. Colquhoun, P. Hodge, F. H. Kohnke and D. J. Williams, Chem. Commun., 1998, 283; T. Freund, C.Kué bel, M. Baumgarten, V. Enkelmann, L. Gherghel, R. Reuter and K. Mué llen, Eur. J. Org. Chem., 1998, 555. 6 Y. Urushigawa, T. Inazu and T. Yoshino, Bull. Chem. Soc. Jpn., 1971, 44, 2546; H. Takemura, M. Suenaga, K. Sakai, H. Kawachi, T. Shinmyozu, Y. Miyahara and T. Inazu, J. Inclusion Phenom., 1984, 2, 207. 7 A. S. Guram, R. A. Rennels and S. L. Buchwald, Angew. Chem., Int. Ed.Engl., 1995, 34, 1348. 8 T. Kanbara, K. Izumi, Y. Nakadani, T. Narise and K. Hasegawa, Chem. L ett., 1997, 1185; J. Louie, J. F. Hartwig and A. J. Fry, J. Am. Chem. Soc., 1997, 119, 11695; M.-H. Yang, T.-W. Lin, C.-C. Chou, H.-C. Lee, H.-C. Chang, G.-H. Lee, M.-K. Leung and S.-M. 780 New J. Chem., 1998, Pages 779»781Peng, Chem. Commun., 1997, 2279; R. A. Singer, J. P. Sadighi and S. L. Buchwald, J. Am. Chem. Soc., 1998, 120, 213. 9 P. A. Gale, J. L. Sessler and V. Kraç l, Chem. Commun., 1998, 1; S. Shinoda, M. Tadokoro, H. Tsukube and R. Arakawa, Chem. Commun., 1998, 181; C. D. Gutsche, J. S. Rogers, D. Stewart and K.-A. See, Pure Appl. Chem., 1990, 62, 485. 10 M. C. Burla, M. Camalli, G. Cascarano, C. Giacovazzo, G. Polidori, R. Spagna and D. Viterbo, J. Appl. Crystallogr., 1989, 22, 389. Received in Cambridge, UK, 3rd April 1998; L etter 8/04229B New J. Chem., 1998, Pages 779»781 781

ISSN:1144-0546    

DOI:10.1039/a804229b

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

L e t t e r P O Si O P O Si O Ph O O Ph P O Si O P O Si O Cl O O Cl Bu2 Bu2 Ph2 Ph2 2 Bu2Si(OH)2 2 Bu2Si(OLi)2 1 3 4 Bu nLi 2 2 PhPOCl2 4 (1) (2) 2 Ph2Si(OH)2 2 Ph2Si(OLi)2 6 7 4 Bu nLi 2 2 POCl3 8 THF – 4 LiCl THF – 4 LiCl 5 9 t t t t Synthesis and structure of new eight-membered SiwOwk5r4wP heterocycles Yann Vaugeois,a Roger De Jaeger,a Joeé lle Levalois-Mitjaville,a Ahmed Mazzah,*,a Michael Woé rleb and Hansjoé rg Grué tzmacherb a L aboratoire de Spectrochimie Infrarouge et Raman (CNRS L P 2631), des Sciences et Universiteç T echnologies de L ille, 59655 V illeneuve dœAscq France ceç dex, b ET H-Zentrum, L aboratorium Anorganische Chemie, 6, 8092 f ué r Universitaé tsstrasse Zué rich, Switzerland The reaction of 1 with or of with in a 1 : 1 molar ratio leads to the Bu2 t Si(OLi)2 PhPOCl2 Ph2Si(OLi)2 POCl3 formation of the 1,3,3,7-tetraoxa-2,6-disila-4,8-k5r4-diphosphacyclooctanes 5 and 9, respectively.An X-ray structure analysis shows 5 to contain an eight-membered ring in the chair conformation. The SiwO Si2O4P2 (1.65 and PwO 1.57 bond lengths in the ring lie within the typical range; the terminal PxO oxygen ”) ” centres are in an axial position and oriented trans with respect to the central plane.Si2O4 Metallasiloxanes, containing a metalwoxygenwsilicon linkage, have been extensively explored,1h23 mainly owing to two reasons : (i) to attempt the synthesis of siloxane polymers with special properties by incorporating diÜerent metal atoms in the siloxane chains and (ii) to compare the properties of metal siloxides with metal alkoxides. Although heterosiloxanes containing phosphorus atoms have been studied already, many of them are intractable polymers.Of the molecular species, which have been described, the large majority are derivatives of the monodentate triorganosiloxy ligands Exceptions to this (wOSiR3).24h26 group are two eight-membered heterocycles with an Si3O4- or skeleton, the syntheses and properties of k5P Si2 O4-k5P2 which were reported by Graalman et al.27 and Kerger and Kohlass,28 respectively.In neither case were the molecules structurally characterised. As part of a study of the bonding characteristics of phosphoruswsiloxane ring systems, we present in this study a simple and convenient synthesis of two new eight-membered heterocycles and report the Si2O4-k5P2 structure of one of them.Syntheses and Characterisation In order to prepare heterocycle 5, (1)29 is –rst Bu2 t Si(OH)2 converted into the lithiumsilanolate 3. This is subsequently reacted with phenyldichlorophosphane, 4, giving 5 in good yields. In the same manner, the related compound 9 is formed according to reaction (2) when the freshly prepared diphenylsilanolate 7 is condensed with phosphorylchloride 8 (Scheme 1).Reaction (2) was followed by 31P NMR spectroscopy. Remarkably, although the reaction was not performed under high dilution conditions, after four hours of stirring at ambient temperature only one singlet at 3.7 ppm assigned to 9 was detected, shifted by about 30 ppm to lower frequencies from the resonance of compound 8 (34.7 ppm).The resonance signal of 5 is observed at [4.9 ppm. After isolation and crystallisation, compounds 5 and 9 were characterised by diÜerent spectroscopic methods. In the infrared spectra, intensive bands at 1041 and 1032 cm~1 are assigned to the SiwOwP stretching vibrations of 5 and 9, respectively.30 A band at 531 cm~1 in the IR spectrum of 9 is assigned to the stretching vibration of the PwCl bond.In the EI mass spectra of 5 and 9 peaks of the molecular ions at m/z 597 and 595 are observed with 10% and 100% intensities, respectively. The base peak at m/z 539 in the mass spectrum of 5 corresponds to [M[But]`. From these data, it was concluded that compounds 5 and 9 correspond to eight-membered heterocycles. In par- Si2O4P2 ticular, the synthesis of 9 is noteworthy as it is easily prepared from readily available chemicals and still contains reactive PwCl groups that may be further derivatised. Crystal Structure of 5 In order to con–rm the structure of compounds 5 and 9 as eight-membered 1,3,3,7-tetraoxa-2,6-disila-4,8-k5r4-diphosphacyclooctanes, we performed an X-ray analysis of 5.Suitable colourless crystals of 5 were obtained by recrystallisation from toluene, n-pentane or diethyl ether at [18 °C.The result is shown in Fig. 1; selected bond lengths and angles are listed in Table 1. To our knowledge structural details of het- Si2O4-k5r4P2 erocycles have not been elucidated before. The Si2O4P2 centrosymmetric ring adopts a chair conformation. The phosphorus atoms deviate by 0.634 from the best plane through ” the two silicon and four oxygen centres [deviations from that plane are : Si 0.021 O(2)[0.020 O(3) 0.019 The dihe- ”, ”, ”].dral angle formed by the intersection of this central Si2O4 plane and the planes is 40.4°. The silicon as well as the PO2 phosphorus centres are slightly distorted from tetrahedral co- Scheme 1 New J. Chem., 1998, Pages 783»785 783Si O M O Si O M O Ge O Si O Si O Ge O rB rA rA rB a d b g g b a d Fig. 1 Molecular structure of 5. Hydrogen atoms have been omitted for clarity. Selected bond lengths and angles are given in Table 1. ordination ; the inner ring OwSiwO [107.1(2)°] and OwPwO [102.8(2)°] angles deviate only a little from the ideal values. The angles at the oxygen centres O(2) and O(3) are, as expected, quite large [148.8(2)° and 146.1(2)°].The phenyl rings at the phosphorus centres bind in an equatorial position while the terminal oxygen centres O(1) and O(1A) bind axially and in a trans position with respect to the central plane. The SiwO bond lengths [1.646(3) are compa- Si2O4 ”] rable to those found in other heterocycles.31,32 The terminal PxO(1) [1.449(3) as well as the other PwO distances ”], [1.571(3) lie within the range expected for phosphates.33 ”], It is interesting to compare the main structural features of rings obtained for comparable heterocycles contain- Si2O4M2 ing boron,32 germanium,32 titanium31 or zirconium31 centres instead of the phosphorus centres in 5 (Scheme 2, Table 2).The heterocycle is folded, as observed in 5 though to Si2O4B2 Table 1 Selected bond and angles/° for 5 lengths/” SiwO(3)a 1.643(3) PwO(1) 1.449(3) SiwO(2) 1.649(3) PwO(2) 1.567(3) SiwC(7) 1.863(5) PwO(3) 1.575(3) SiwC(11) 1.869(4) PwC(1) 1.779(4) O(3)awSiwO(2) 107.1(2) O(1)wPwO(3) 114.6(2) O(3)awSiwC(7) 105.3(2) O(2)wPwO(3) 102.8(2) O(2)wSiwC(7) 110.2(2) O(1)wPwC(1) 113.0(2) O(3)awSiwC(11) 108.9(2) O(2)wPwC(1) 105.3(2) O(2)wSiwC(11) 105.1(2) O(3)wPwC(1) 104.8(2) C(7)wSiwC(11) 119.7(2) PwO(2)wSi 148.8(2) O(1)wPwO(2) 115.2(2) PwO(3)wSia 146.1(2) a Symmetry transformation used to generate equivalent atoms: [x, [y, [z.Scheme 2 a lesser extent (intersection of the and planes is Si2O4 BO2 19.3°), while all other cycles (M\Ge, Ti, Zr) are planar. With the notable exception of the mixed silicon/germanium heterocycle the bond angles at the oxygen centres do not diÜer by more than about 10°, leading to rather symmetrically shaped eight-membered ring systems as expressed by the ratios rA/rB (0.89»0.97).In the germanium compound these angles are quite diÜerent (*\32°), which leads to an elongated ring structure The inner ring angles b Si2O4Ge2 (rA/rB\0.75). and d at the metallic centres show the expected values ; they vary between 103° and 112° for centres in a tetrahedral coordination sphere (M\Ti, Ge, P), 122° for a centre in a trigonal planar coordination sphere (M\B) and 99° for zirconium in an octahedral coordination environment.The diameters rA and of the heterocycles compare well with the cavity of the rB twelve-membered crown ether 12-crown-4 (approximately 3.5 Hence, these heterocycles may be suitable ligands for small ”). ions like Li`.Furthermore, heterocycles 5 and 9 contain terminal PxO groups as further Lewis basic sites. Since the Lewis acidic k5, r4-phosphorus centres, as well as the silicon centres, may undergo expansion of their valence spheres to become penta- or even hexa-coordinated, these heterocycles may serve as multifunctional complexing agents for cations as well as anions.These issues, as well as more in-depth studies to understand the in—uence of silicium centres on the shape of these heterocycles, are under current investigation. Experimental All experiments were carried under an atmosphere of dry oxygen-free nitrogen. Solvents were dried over molecular sieves and distilled prior to use. (1) was prepared Bu2 t Si(OH)2 according to a method described by Weidenbruch et al.,29 diphenylsilanol (6), (4) and (8) were pur- PhPOCl2 POCl3 chased from Aldrich.NMR spectra were recorded on a Brucker WP 300 spectrometer. Chemical shifts are referenced against internal TMS (1H and 13C) or external 85% H3PO4 (31P). IR and Raman spectra were recorded on Bomen MB-100 or Dilor XY multi-canal spectrometers, respectively ; the latter used a 514.32 nm argon ion laser for excitation.Chemical ionisation (CI) mass spectra were run on a Nermag R10-10B spectrometer. Synthesis of 5 BunLi (1.6 M; 7.5 ml, 6.00 mmol) in n-hexane was added dropwise at room temperature to a solution of 1.156 g (6.57 mmol) of in 15 ml of toluene. To complete the lithiation, ButSi(OH)2 the solution was heated at 80 °C for 20 min, then 1.17 g (6 mmol) of in toluene (15 ml) were added dropwise.PhPOCl2 The mixture was kept for one additional hour at 60 °C. After cooling to room temperature, the solution was –ltered through Celite. All volatiles were removed in vacuo to aÜord a white solid, which was puri–ed by crystallisation from pentane to give colourless crystals of 5. Yield : 1.52 g (85%), mp 272 °C. 31PM1HN NMR d\[3.6 (s). 1H NMR (C6D6) : (C6D6) : d\0.9 (s, 36H, 7.0 (m, 10H, IR (KBr): 1261s, CH3), C6H5). 1132s, 1041vs, 938w, 831s, 737w, 664.4s, 545vs cm~1. MS: m/z Table 2 Comparison of inner ring angles and diameters and of heterocycles (M\B, Ge, Ti, Zr, P) rA rB, Si2 O4M2 M a/° b/° c/° d/° rA/” rB/” rA/rB Ba 149.0 122.0 149.8 111.6 3.52 3.62 0.97 Gea 174.8 112.2 142.9 109.3 3.31 4.41 0.75 Tib 156.1 107.9 169.1 106.5 3.65 4.09 0.89 Zrb 169.2 99.6 161.2 109.9 3.84 4.11 0.93 Pc 146.1 102.8 148.8 107.1 3.47 3.74 0.93 a Ref. 31. b Ref. 32. c This work. 784 New J. Chem., 1998, Pages 783»785597 ([M]1]`, 10%), 539 ([M[But]`, 100%). Anal. calcd for C, 56.3 ; H, 7.8 ; P, 10.4 ; found: C, 55.6 ; C28H46O6P2Si2 : H, 7.9 ; P, 9.7. Synthesis of 9 The reaction of (2.83 g, 18.33 mmol) with dilithiated P(O)Cl3 diphenylsilanediol (3.95 g, 18.28 mmol) in THF (30 ml) was performed as described for the synthesis of 5.The dark yellow residue obtained after removal of the volatiles was precipitated in dichloromethane, yielding a yellow solid, and was characterised as compound 9. Yield (3.75 g, 73%), mp 228 °C. 31PM1HN NMR d\[4.9 (s). 1H NMR (C6D6) : (C6D6) : d\7.3 (m, 20H, IR (KBr): 1594w, 1477m, 1130s, C6H5). 1040vs, 737s, 545s, 530 m cm~1. MS: m/z 595 ([M]1]`, 100%). Anal. calcd for C, 48.6 ; H, 3.4 ; C24H20Cl2O6P2Si2 : Cl, 11.9 ; P, 10.4 ; found: C, 50.8 ; H, 3.4 ; Cl, 11.9 ; P, 10.6. X-ray crystal structure determination A colourless single crystal (0.68]0.61]0.49 mm) of 5 M\596.77) was studied at 293 K on a Sot (C28H46O6P2Si2 , IPDS System (l\0.242 mm~1).The re–nement in the triclinic space group P[1 with one molecule in the unit cell [a\8.566(6), b\10.167(8), c\10.542(8) a\102.80(7), ”, b\103.96(6), c\104.74(7)°] converged at R1\0.0535 (based on F) and (based on F2) for 4648 re—ec- wR2\0.1242 tions with I[2r(I). All non-hydrogen atoms were re–ned with anisotropic displacement parameters.All hydrogen atoms were found in the diÜerence Fourier map and re–ned independently with diÜerent isotropic displacement parameters for each group. CCDC reference number 440/039. References 1 H. N. Stokes, Chem. Ber., 1891, 24, 933. 2 (a) A. Ladenburg, Chem. Ber., 1871, 4, 91. (b) A. Ladenburg, Ann. Chem. Pharm., 1872, 164, 300. 3 S. N. Borisov, M. G. Voronkov and E. Y. Lukevits, Organosilicon Heteropolymers and Hetero Compounds, Plenum, New York, 1970. 4 J. C. Saam, in Silicon Based Polymer Science, ed. J. M. Zeigler and F. W. G. Fearon, Advances in Chemistry 224, American Chemical Society, Washington DC, 1990. 5 Y. I. Yermakov, B. N. Kuznestov and V. A. Zakharov, Catalysis by Supported Complexes, Elsevier, New York, 1981. 6 T. Seiyama and K. Tanabe, New Horizons in Catalysis, Elsevier, Amsterdam, 1980. 7 F. R. Hartley, Supported Metal Complexes, Reidel, Boston, 1985; C. L. Thomas, Catalytic Process and Proven Catalysts, Academic, New York, 1970. 8 R. Pearce and W. R. Patterson, Catalysis and Chemical Process, Blackie, Glasgow, 1981. 9 T ailored Metal Catalysts, ed. Y. Iwasawa, Reidel, Boston, 1986. 10 D. E. Leyden, G. H. Luttrell, A. E. Sloan and N.J. De Angelis, Anal. Chim. Acta, 1976, 84, 97. 11 K. Unger, H. Ruppercht, B. Valentin and W. Kircher, Drug Dev. Ind. Pharm., 1983, 9, 69. 12 (a) E. P. Plueddmann, Silane Coupling Agents, Plenum, New York, 982. (b) Silanes and Other Coupling Agents, ed. K. L. Mittal, VSP, Utrecht, 1992. 13 B. Arkles, Chemtech, 1977, 7, 766. 14 H. Schmidbaur, Angew. Chem., 1965, 77, 206; Angew.Chem., Int. Ed. Engl., 1965, 4, 201. 15 F. Schindler and H. Schmidbaur, Angew. Chem., 1967, 79, 697; Angew. Chem., Int. Ed. Engl., 1967, 6, 683. 16 K. A. Andrianov, Inorg. Macromol. Rev., 1970, 1, 33. 17 M. Baier, P. Bissinger, J. Bluué mel and H. Schmidbaur, Chem. Ber., 1993, 126, 947. 18 A. R. Barron, C. C. Landry, L. K. Cheatham and A. N. MacInnes, J. Mater. Chem., 1991, 1, 143. 19 A. K. McMullen, T. D. Tilley, A. L. Rheingold and S. J. Geib, Inorg. Chem., 1989, 28, 3772. 20 G. A. Sigel, R. A. Bartlett, D. Decker, M. M. Olmstead and P. P. Power, Inorg. Chem., 1987, 26, 1773. 21 D. C. Hrncir and G. D. Skiles, J. Mater. Res., 1988, 3, 410. 22 M. G. Voronkov, E. A. Maletina and V. K. Roman, in Heterosiloxanes, ed. M. E. Volœpin and K. Gingild, Soviet Scienti–c Review Supplement, Series Chemistry, Academic Press, London, 1988, vol. 1. 23 R. Murugavel, A. Voigt, M. G. Walawalkar and H. W. Roesky, Chem. Rev., 1996, 96, 2205. 24 M. G. Voronkov, S. N. Borisov and E. Y. Lukevits, in Organosilicon Derivatives of Phosphorus and Silicon, Plenum, New York, 1971. 25 K. A. Andrianov, Izv. Akad. Nauk SSSR, Ser. Khim., 1965, 381; translated in Bull. Acad. Sci. USSR, Div. Chem. Sci., 1965, 366. 26 K. A. Andrianov, Zh. Obshch. Khim., 1972, 42, 850; translated in J. Gen. Chem. USSR, 1972, 42, 840. 27 O. Graalman, M. Meyer and U. Klingebiel, Z. Anorg. Allg. Chem., 1986, 534, 109. 28 K. Kerger and R. Kohlass, Z. Anorg. Allg. Chem., 1967, 354, 44. 29 D. Weidenbruch, H. Pesel and D. Van Hieu, Z. Naturforsch., B, 1980, 35, 31. 30 K. A. Andrianov, T. V. Vasilœeva and L. M. Kananashvili, Izv. An. SSR, Otd. Khim., Nauk, 1961, 1030. 31 A. Haoudi-Mazzah, A. Mazzah, H.-G. Schmidt, M. Noltemeyer and H. W. Roesky, Z. Naturforsch., B, 1991, 46, 587. 32 A. Mazzah, A. Haoudi-Mazzah, M. Noltemeyer and H. W. Roesky, Z. Anorg. Allg. Chem., 1991, 604, 93. 33 International T ables of X-Ray Crystallography, ed. International Union of Crystallography, The Kynoch Press, Birmingham, UK, 1969, vol. I. Received in Montpellier, France, 5th March 1998; L etter 8/03283A New J. Chem., 1998, Pages 783»785 785

ISSN:1144-0546    

DOI:10.1039/a803283a

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

L e t t e r Cl Au PPh3 Os3(CO)12 hn AuPPh3Cl (m-AuPPh3)(m-Cl)Os3(CO)10 2b Photochemistry of with synthesis and Os3(CO)12 AuPPh3Cl: structural characterisation of (l-AuPPh3)(l-Cl)Os3(CO)10 Caroline M. Hay,a Nicholas E. Leadbeater,*,aJack Lewis,a Paul R. Raithbya and Kevin Burgessb a Department of Chemistry, University of Cambridge, L ens–eld Road, Cambridge, UK CB2 1EW b Department of Chemistry, T exas A & M University, College Station, T X 77843-3255, USA Studies showing the potential of photochemistry for the preparation of mixed-metal clusters are presented, the synthesis and molecular structure of (l- being reported. AuPPh3)(l-Cl)Os3(CO)10 The photochemistry of organometallic compounds is a developing research area, being of importance both in synthesis and catalysis as well as proving fascinating in its own right.1,2 Research in our group has focused on harnessing the synthetic potential of photochemistry to generate known and novel compounds efficiently and in high yields.3 The trinuclear clusters [M\Ru (1a), Os (1b)] M3(CO)12 can, in principle, undergo either photofragmentation or photosubstitution reactions.The photochemistry of 1a can be controlled by the solvent in which the reaction is undertaken, solvents such as hexane favouring fragmentation and donor solvents such as ethyl acetate favouring substitution.4 In the case of 1b, although substitution dominates the photochemistry in most media, by using a donor solvent it is possible to stabilise the coordinatively unsaturated fragment [M3(CO)11] formed and, in the presence of a ligand, it is possible to form substitution products.5 To date, we have used photochemistry to attach organic fragments to organometallic fragments. The work reported here represents an extension of that work to look at the photogeneration of polynuclear mixed-metal complexes.Since we have shown that photochemistry oÜers a route to the di-hydrido cluster (l-H) and the 2Ru3(CO)10 hydrogen chloride bridged complex (l-H)(l-Cl) it Ru3(CO)10 ,6 was decided to investigate the photochemistry of M3(CO)12 with the isolobal fragment Since osmium clusters AuPPh3Cl.are often more stable than their ruthenium counterparts, initial studies have focused on the photochemistry of 1b with the gold fragment. Broad-band UV irradiation of an ethyl acetate solution of 1b (30 mg in 100 ml) containing an excess (3 equiv.) of leads to the formation of the bimetallic cluster (l- AuPPh3Cl (2b) (Fig. 1).§ The cluster is not AuPPh3)(l-Cl)Os3(CO)10 photostable and prolonged irradiation of the reaction mixture leads to decomposition. The optimal photolysis time was found to be 30 min, this producing 2b in ca. 60% yield after chromatographic separation.Although 2b has been prepared previously, by re—uxing a xylene solution of 1b with the yields are signi–- AuPPh3Cl, cantly lower than by the presented photochemical routes.7 In addition, in the case of the thermochemical route, a number of § All photochemical reactions were performed in a specially designed glass reaction vessel –tted with a nitrogen bubbler, re—ux condenser and dry-ice cooling –nger.A 125 W mercury arc broad-band UV lamp was used as the irradiation source and re—ectors placed around the reaction vessel to maximise efficiency. Photolysis mixtures were kept at between 2 and [2 °C. Product 2b was separated by thin-layer chromatography (1 : 1 dichloromethane : hexane as eluent). Spectroscopic data for 2b: IR (cyclohexane) : m(CO) 2097 (m), 2045 (vs), 2014 (vs), 2008 (vs), 1984 (m), 1974 (w), 1996 (m) cm~1; 1H NMR (250 MHz, d 7.65 (m, Ph); negative-ion FAB mass spectrum: m/z CDCl3) : 1343 (calc. 1345 based on 190Os and 197Au). by-products are formed, this making separation of the desired product difficult. This is not the case in the photochemical preparation where careful monitoring of the conditions leads to the product being formed with minimal decomposition or by-product formation.The molecular structure of 2b was determined by a singlecrystal X-ray diÜraction study and is shown in Fig. 2 together with selected bond lengths and angles.î The solid-state structure of 2b is consistent with that proposed in solution, the and Cl moieties bridging one AuPPh3 of the OswOs bonds symmetrically.The doubly-bridged Os(l)wOs(2) bond is slightly longer [2.880(2) and the Aé ] unbridged OswOs bonds slightly shorter [mean 2.838(5) Aé ] than the metal»metal bonds in the parent cluster 1b [mean 2.867(3) This compares to a doubly bridged OswOs Aé ].8 Fig. 1 The photochemical reaction of with Os3(CO)12 AuPPh3Cl î Crystal data for (l- 2b: AuPPh3)(l-Cl)Ru3(CO)10 M\1345.42, red blocks, crystal dimensions C28H15O10PClOs3Au, 0.46]0.25]0.51 mm, monoclinic, space group (no. 14), P21/c a\12.892(9), b\16.899(1), c\16.255(3) b\113.75(1)°, Aé , U\3241.46(10) Mg cm~3, Z\4, F(000)\2372, Aé 3, Dcalc\2.76 Mo-Ka radiation, k\0.71069 l(Mo-Ka)\16.379 mm~1, Aé , T \298(2) K. Stoe-Siemens AED diÜractometer, 7019 re—ections collected by the x/h scan method in the range 2.5°OhO25.00°.Cell parameters were obtained by least-squares re–nement on diÜractometer angles from 25 centred re—ections (20\2h\22.5°). An analytical absorption correction based on 12 indexed faces was applied. The structure was solved by direct methods followed by Fourier diÜerence synthesis and re–ned by blocked-cascade least-squares on F (SHELXL-76).11 All Au, Os, P, Cl and O atoms were treated anisotropically.The hydrogen atoms were placed in idealised positions and allowed to ride on the relevant carbon atom. In the –nal cycles of re–nement a weighting scheme was introduced which produced a —at analysis of variance. The phenyl rings on the triphenylphosphine moiety were re–ned as rigid groups with the bond lengths and angles –xed at 1.395 and 120°, respctively. The re–nement converged at Aé R1\0.055 and wR1\0.0450 for 4298 observed data [F[4r(F)].R1\&pFo o[oFcp/&oFo o, wR1\[&w1@2(pFo o[oFcp)/&w1@2 oFo o], The –nal diÜerence synthesis showed no w\1/[r2(Fo)2]0.00080F2]. *q above 2.53 or below [1.91 the major features lying near the eAé ~3, osmium and gold atoms. CCDC reference number 440/040. New J. Chem., 1998, Pages 787»788 787Fig. 2 The molecular structure of (l- (2b) AuPPh3)(l-Cl)Os3(CO)10 showing the atomic numbering scheme; O, phenyl C and H atoms have been omitted for clarity. Selected bond lengths and angles (°) : (Aé ) Os(1)wOs(2) 2.880(2), Os(2)wOs(3) 2.835(2), Os(1)wOs(3) 2.841(2), Au(1)wOs(1) 2.766(2), Au(1)wOs(2) 2.765(2), Au(1)wP(1) 2.309(4), Os(1)wCl(1) 2.450(4), Os(2)wCl(1) 2.459(4), Os(1)wC(11) 1.881(15), Os(1)wC(12) 1.940(21), Os(1)wC(13) 1.919(20), Os(2)wC(21) 1.942(18), Os(2)wC(22) 1.868(14), Os(2)wC(23) 1.905(21), Os(3)wC(31) 1.961(16), Os(3)wC(32) 1.946(21), Os(3)wC(33) 1.911(21), Os(3)wC(34) 1.936(15), Os(1)wOs(2)wOs(3) 59.6(1), Os(2)wOs(1)wOs(3) 59.4(1), Os(1)w Os(3)wOs(2) 61.0(1), Os(1)wOs(2)wCl(1) 53.9(1), Os(2)wOs(1)wCl(1) 54.2(1), Os(3)wOs(2)wCl(1) 87.2(1), Os(3)wOs(1)wCl(1) 87.3(1), Os(3)w Os(1)wCl(1) 87.3(1), Au(1)wOs(1)wOs(2) 58.6(1), Au(1)wOs(2)wOs(3) 96.5(1), Au(1)wOs(1)wOs(3) 96.4(1), Os(1)wAu(1)wOs(2) 62.7(1), Os(1)wAu(1)wP(1) 150.7(1), Os(2)wAu(1)wP(1) 146.6(1), Au(1)w Os(1)wCl(1) 97.1(1), Au(1)wOs(2)wCl(1) 96.9(1).bond length of 2.846(1) in the related cluster (l-H)(l-Cl) Aé (3b)9 and, although the diÜerence in doubly- Os3(CO)10 bridged OswOs bond lengths in 2b and 3b at 0.034 is on Aé the limit of chemical signi–cance, bond lengthening in—uence induced by a bridging gold atom is greater than the related eÜect of a bridging hydride ligand.Comparison between the doubly bridged metal»metal bond lengths in (l-AuPPh3)(l- [2.8742 (6) and (l-H)(l-Cl) Cl)Ru3(CO)10 Aé ]10 Ru3(CO)10 [2.833 (2) reveals a similar trend. Aé ]6b The dihedral angles made by the Os(1)wOs(2)wAu(1) and Os(1)wOs(2)wCl(1) planes with the Os(3) triangle are 120.8° and 110.0°, respectively.As a result, 2b can be considered as consisting of a ìbutter—yœ metal framework with the gold atom adopting a wing-tip position. All the carbonyl ligands in 2b are terminally bound but their arrangement and bonding are of interest.The bridging gold and chlorine groups force the carbonyl ligands C(12)wO(12), C(13)wO(13), C(21)wO(21) and C(23)wO(23) into semi-axial positions. There is a distinct trans in—uence in the metal-carbonyl distances for the carbonyl groups trans to the halogen atom, these being the shortest OswC(O) bonds [Os(1)wC(11)\1.881 (15) Os(2)wC(22)\1.868 (14) Aé , Aé ]. The longest OswC(O) bonds are found to involve the carbonyl ligands on Os(3) that are trans to each other [Os(3)wC(31)\1.961(16) Os(3)wC(32)\1.946 (24) All Aé , Aé ].these bond lengths can be compared to those reported for 1b [OswCO(equatorial)\1.912 (7) OswCO(axial)\1.946(6) Aé , Aé ]. Acknowledgements College Cambridge is thanked for a Research Fellow- Girton ship. This work was funded in part by the UK Engineering and Physical Sciences Research Council and Johnson- Matthey plc is acknowledged for the generous loan of osmium tetroxide.Advice and assistance from G.P. Shields is greatly appreciated. References 1 N. E. Leadbeater, J. Chem. Soc., Dalton T rans., 1995, 2923. 2 (a) R. N. Perutz, Chem. Rev., 1993, 361; (b) N. E. Leadbeater, J. Lewis, P. R. Raithby and G.N. Ward, J. Chem. Soc., Dalton T rans., 1997, 2511; (c) G. R. Haire, N. E. Leadbeater, J. Lewis, P. R. Raithby, A. J. Edwards and E. C. Constable, J. Chem. Soc., Dalton T rans., 1997, 2997. 3 A. J. Edwards, N. E. Leadbeater, J. Lewis and P. R. Raithby, J. Organomet. Chem., 1995, 15, 512. 4 N. E. Leadbeater, J. Organomet. Chem., in the press. 5 D. R. Tyler, M. Atobelli and H. B. Gray, J. Am. Chem. Soc., 1980, 102, 3022. 6 (a) N. E. Leadbeater, J. Lewis and P. R. Raithby, J. Organomet. Chem., 1997, 543, 251; (b) N. E. Leadbeater, J. Lewis, P. R. Raithby and M.-A. Rennie, Polyhedron, 1998, 17, 1755. 7 C. W. Bradford, W. van Bronswijk, R. J. H. Clark and R. S. Nyholm, J. Chem. Soc., 1970, 2889. 8 M. R. Churchill and B. G. De Boer, Inorg. Chem., 1977, 16, 878. 9 M. R. Churchill and R. A. Lashewycz, Inorg. Chem., 1979, 18, 1926. 10 G. Lavigne, F. Papageorgiou and J.-J. Bonnet, Inorg. Chem., 1984, 23, 609. 11 G. M. Sheldrick, SHEL XL -76 program for crystal structure determination, University of Cambridge, UK, 1976. Received in Cambridge, UK, 16th April 1998; L etter 8/02946F 788 New J. Chem., 1998, Pages 787»788

ISSN:1144-0546    

DOI:10.1039/a802946f

出版商:RSC    

年代:1998

数据来源: RSC