摘要:

L e t t e r What are the primary products of reactions between anionic hydrides and HCyCPh? Brandon Reinhart and Dmitry G. Gusev* Department of Chemistry, W ilfrid L aurier University, W aterloo, ON, Canada N2L 3C5. E-mail : FAX: ]1 519 746 0677 dgoussev=mach1.wlu.ca; (in Montpellier, France) 16th September 1998, Accepted 5th October 1998 Receiøed Reactions of phenylacetylene with [L is the [RuH2(CO)L]ó pincer ligand C6H3-2,6-(CH2PBu2 t )2 ] , [RuH3(CO)(PPr3 i )2 ]ó and have aÜorded the new anionic r- [ReH3(NO)(PPr3 i )2 ]ó alkynyl and vinyl complexes trans- [RuH(CyCPh)(CO)L]ó, cis-[RuH2(CyCPh)(CO)(PPr3 i )2 ]ó, cis- [RuH2(CPhxCH2)- and (CO)(PPr3 i )2 ]ó cis- [ReH2(CPhxCH2)(NO)(PPr3 i )2 ]ó, all as the [K(18-crown-6) ]ë salts.Formation of the acetylides does not involve unsaturated species and takes place via vinyl intermediates.[MwC(Ph)xCH2 ] Chemistry of anionic transition metal hydrides has attracted little attention in contrast to the related neutral and cationic species.1 We decided to investigate the reactivity of trans- (1)2a [L is the pincer ligand [RuH2(CO)L]~ C6H3-2,6- (2)2b and (CH2PBu2 t )2], [RuH3(CO)(PPr3 i )2]~ [ReH3(NO)- (3)2b Mall are [K(18-crown-6)]` saltsN in a typical (PPr3 i )2]~ reaction of metal hydrides»insertion of alkynes into the MwH bond.Three primary organometallic products could be anticipated with HCyCPh (Scheme 1). Product a can be formed by replacement after protonation of H2/PhC2~ by HCyCPh.3 Species b and c are insertion pro- [HMLn]~ ducts with vinyl ligands attached via the a or b-carbon atoms, respectively.This communication reports the –nding that c is the primary product with complexes 1»3. Scheme 1 A fast NMR tube reaction of (1) and trans-[RuH2(CO)L]~ two equivalents of HCyCPh in cleanly aÜorded a pyridine-d5 ruthenium product and one equivalent of The H2CxCHPh. product could be isolated from pyridine or THF and characterized as trans-[RuH(CCPh)(CO)L]~ (4) by 1H, 31P and 13C NMR, IR spectroscopy and elemental analysis.The trans- HwRuwCyCPh disposition was con–rmed by the diÜerence 1H NOE (nuclear Overhauser eÜect) spectra (Fig. 1). The anion 4 is not a very strong base and no protonation was observed with 2»4 equivalents of methanol in THF. A second anionic complex, (2), [RuH3(CO)(PPr3 i )2]~ reacted with phenylacetylene to aÜord an isolated mixture of two products in a 1 : 2 ratio.The minor product is of the a type, (5), and was identi- cis-[RuH2(CyCPh)(CO)(PPr3 i )2]~ –ed by NMR and IR spectroscopy. Complexes 4 and 5 have very similar wCyCPh 13C NMR and IR features (Table 1). The major component of the mixture is the vinyl complex cis- (6) according to the 1H [RuH2(CPhxCH2)(CO)(PPr3 i )2]~ and 13C NMR data.Fig. 1 (A) Regular 1H NMR and (B, C) diÜerence NOE spectra of the acetylide 4. The RuH and fragments are trans wCyCwC6H5 since their enhancements in B and C originate from inequivalent CH3 groups on the opposite sides of the PwCwP plane in 4. Solvent resonances (THF- are shown by stars, the one at d 3.58 is overlapped d8) with the intense line of [K(18-crown-6)]`.Table 1 Selected spectroscopic data for complexes 4»7 Complex d(MwH) d(MwCy) d(yCPh) d(Cipso) 4a [9.67 151.59 113.88 134.91 5b [11.10,[8.27 148.24 113.24 135.30 d(MwH) d(MwCPh) d(xCH2) d(Cipso) 6 [12.39, [7.52 191.96 121.62 167.82 7 [5.45, [3.78 182.06 120.30 165.75 cm~1. cm~1. a mC.C\2065 b mC.C\2058 New J. Chem., 1999, 1»3 1Scheme 2 Finally, when rhenium trihydride, (3), [ReH3(NO)(PPr3 i )2]~ was reacted with phenylacetylene, slow and clean formation of (7) was complete in 6 h.cis-[ReH2(CPhxCH2)(NO)(PPr3 i )2]~ No intermediates could be detected in the reaction solution by 31P NMR. The isolated product was characterized by 1H, 31P and 13C NMR, IR spectroscopy, and elemental analysis. Facile protonation of 7 cleanly aÜorded a stable styrene complex, which will be ReH2(CH2xCHPh)(NO)(PPr3 i )2 , reported elsewhere. The results of three diÜerent reactions of the ruthenium and rhenium complexes allow mechanistic considerations.The interaction that brings together coordinatively saturated, thermally stable hydrides 1»3 and an alkyne molecule could be the MwHd~… … …d`HwC hydrogen (also termed ìdihydrogenœ) bonding shown in Scheme 2.4 The experimental evidence for a reversible formation of the transient [LnM(l-H2)CCPh]~ species is represented by the reported H/D scrambling in a system.5 Scheme 2 shows further FeH2(dippe)2/PhCyCD feasible rearrangements of which are [LnM(l-H2)CCPh]~, either protonation of the metal fragment or hydride transfer to the atom of phenylacetylene.An alternative bond Ca CbwH formation (not shown) appears less likely for geometric reasons.In agreement with this, no b-type product has been detected in the reactions of complexes 1»3. The acetylides 4 and 5 apparently do not result from protonation of 1 and 2 by phenylacetylene. For example, complex 4 is not a primary organometallic product since it is formed along with styrene. Furthermore, for 4, the interpretation of Scheme 2 invoking loss and a 16-electron intermediate is H2 implausible in neat pyridine where the intermediate would aÜord a pyridine complex.A deuterium-labeling experiment with 1 and phenylacetylene-d clearly indicated that formation of the acetylide 4 involved a c-type primary product. transand two equivalents of DCyCPh reacted on [RuH2(CO)L]~ mixing in an NMR tube, cleanly aÜording the monohydride 4 and one equivalent of HCDxCDPh (Scheme 3).The deuteriated styrene showed two terminal CH triplets in the 1H NMR spectrum in a 3.5 : 1 ratio at d 5.68 Hz) and (3JHvD\2.7 Scheme 3 d 5.12 Hz), respectively. Thus, only one hydride (3JHvD\1.6 took part in the reaction and it was transferred to the atom Ca of phenylacetylene, aÜording the vinyl intermediate shown in Scheme 3.All of the above indicate that generally (a) reactions of terminal acetylenes with metal hydrides may not require formation of unsaturated intermediates, (b) r-acetylide complexes may not be intermediate species in hydrogenation of acetylenes, but (c) their formation might take place via intermediate c-type vinyl complexes. Experimental All reactions and sample preparations were carried out in dry solvents under puri–ed nitrogen in an Innovative Technology glovebox equipped with a vacuum line and a [35 °C refrigerator.Throughout this paper, the NMR data are reported with the apparent coupling of virtual triplets (vt) denoted as vJ. Preparation of trans- [RuH(CyCPh)(CO){C6H3-2,6- (4) (CH2PBu2 t )2 } ] [K(18-crown-6)] PhCyCH (74 mg, 0.72 mmol) was added to a solution of (300 [RuH2(CO)MC6H3-2,6-(CH2PBu2 t)2N][K(18-crown-6)] mg, 0.36 mmol) in THF (3 mL) or pyridine (1.5 mL).The mixture was stirred for 1 h. Addition of 12 mL of hexane caused the product to crystallize out as a white solid. It was isolated by –ltration, washed with 3]3 mL of hexane and dried under vacuum. Yield from THF: 278 mg (ca. 82%). Yield from pyridine : 316 mg (ca. 94%). Anal. calcd for (928.195) : C, 58.23 ; H, 7.93. Found: C, C45H73KO7P2Ru 57.83 ; H, 8.05. IR (Nujol) : 2065 cm~1, 1852 cm~1, mC.C mCO 1583 cm~1. 1H NMR (200 MHz, d [9.67 (t, mRuH THF-d8) : Hz, 1H, RuH), 1.22, 1.50 (vt, Hz, 2JHhP\21.7 vJHhP\5.6 36H, 3.07 (dvt, Hz, Hz, 2H, CH3), 2JHhH\14.9 vJHhP\3.7 3.51 (s, 24H, 3.55 (overlapped with the CH2), crown-CH2), crown resonance, 2H, 6.31 (m, 1H, 6.57 (m, 3H, CH2), C6H3), Ar), 6.86 (m, 4H, 31PM1HN NMR (80 MHz, d C6H5).THF-d8) : 107.4. 13CM1HN NMR (50 MHz, d 30.23, 29.88 (vt, THF-d8) : Hz, 35.55 (vt, Hz, PC), 37.61 (vt, vJCvP \2.2 CH3), vJChP\8.1 Hz, PC), 42.04 (vt, Hz, 71.12 vJChP\3.4 vJChP\10.0 PCH2), (s, 113.88 (s, yCPh), 118.41 (vt, Hz, crown-CH2), vJChP\7.6 CH, RuwAr), 120.13 (s, CH, RuwAr), 120.23, 127.41, 130.68 (s, CH, 134.91 (s, 149.04 (vt, C6H5), Cipso, C6H5), vJChP\10.3 Hz, C, RuwAr), 151.59 (t, Hz, RuwCy), 188.57 2JChP\10.7 (t, Hz, RuC), 212.46 (t, Hz, CO). 2JChP\7.6 2JChP\10.1 Assignment of the 13C signals was con–rmed by gateddecoupled 13C NMR. Hydride-coupled 13C NMR: the carbonyl and ligand metal-bound carbon atoms both showed small couplings of 5.2 and ca. 5 Hz, respectively. The 2JChH wCyCPh carbons showed two- and three-bond couplings of 14.5 and 8.9 Hz, respectively. Isolation of a mixture of cis- [RuH2(CyCPh)(CO)(PPr3 i )2 ] - [K(18-crown-6) ] (5) and cis- [RuH2(CPhxCH2)(CO)(PPr3 i )2 ] - [K(18-crown-6)] (6) PhCyCH (338 mg, 3.31 mmol) was added to a cold ([35 °C) solution of (1000 mg, [RuH3(CO)(PPr3 i )2][K(18-crown-6)] 2 New J.Chem., 1999, 1»31.32 mmol) in THF (3 mL). The mixture was left at [35 °C for 5 h. Addition of 12 mL of cold ([35 °C) hexane precipitated a white solid. It was isolated by –ltration, washed with 3]3 mL of hexane and dried under vacuum. Yield : 684 mg (60%). IR (Nujol) : 2058, and 1852, or/and mC.C mCO mRhH mRuH 1597 cm~1. 5: 1H NMR (200 MHz, d [11.10 mC/C THF-d8) : (td, Hz, 1H, RuH), [8.27 (td, 2JHhP\22.3 2JHhH\7.0, Hz, 1H, RuH), 1.30 (dvt, Hz, 2JHhP\26.1 vJHhP\6.3 Hz, 36H, 2.21 (m, 6H, CH), 3.55 (s, 24H, 3JHhH\6.0 CH3), 6.5»7.4 (m, 5H, 31PM1HN NMR (80 MHz, crown-CH2), C6H5).d 84.0. 13CM1HN NMR (50 MHz, d 21.45, THF-d8) : THF-d8) : 21.61 (s, 28.79 (vt, Hz, PCH), 71.12 (s, CH3), vJChP\9.1 113.24 (s, yCPh), 119.84, 127.64, 130.86 (s, CH, crown-CH2), Ph), 135.30 (s, Ph), 148.24 (t, Hz, Cipso , 2JChP\13.3 RuwCy), 212.35 (t, Hz, RuwCO). 6: 1H NMR 2JChP\9.5 (200 MHz, d [12.39 (td, Hz, THF-d8) : 2JHhP\25.7 2JHhH\ 1H, RuH), [7.52 (td, Hz, 1H, RuH), 1.17 7.0, 2JHhP\27.8 (dvt, Hz, Hz, 36H, 2.12 (m, 6H, vJHhP\6.3 3JHhH\6.0 CH3), CH), 3.55 (s, 24H, 5.20, 5.55 (d, Hz, crown-CH2), 2JHhH\7.6 2H, 6.5»7.4 (m, 5H, 31PM1HN NMR (80 MHz, xCH2) C6H5). d 81.2. 13CM1HN NMR (50 MHz, d 21.05, THF-d8) : THF-d8) : 21.97 (s, 28.92 (vt, Hz, PCH), 71.12 (s, CH3), vJChP\8.4 121.62 (t, Hz, 120.40, 125.79, crown-CH2), 3JChP\2.1 xCH2), 128.93 (s, CH, Ph), 167.82 (s, Ph), 191.96 (t, Cipso , 2JChP\9.3 Hz, RuwCPh), 212.59 (t, Hz, RuwCO). Assign- 2JChP\10.5 ment of the 13C signals is con–rmed by an APT (attached proton test) experiment and comparison to the 13C spectra of 4 and 7.Preparation of cis- [ReH2(CCPh)(NO)(PPr3 i )2 ] [K(18-crown- (7) 6) ] PhCyCH (145 mg, 1.42 mmol) was added to a solution of (600 mg, 0.71 mmol) in [ReH3(NO)(PPr3 i )2][K(18-crown-6)] THF (6 mL). The mixture was stirred for 6 h. Addition of 20 mL of hexane precipitated a yellow solid. It was isolated by –ltration, washed with 2]5 mL of hexane and dried under vacuum.Yield : 500 mg (75%). Anal. calcd for (945.28) : C, 48.28 ; H, 8.00 ; N, 1.48. C38H75KNO7P2Re Found: C, 47.89 ; H, 7.85 ; N, 1.54. IR (Nujol) : 1790, mReH mReH or/and 1597, 1500 cm~1. 1H NMR (200 MHz, mC/C mNO THFd [5.45 (td, Hz, 1H, ReH), d8) : 2JHhP\22.1 2JHhH\8.3, [3.78 (td, Hz, 1H, ReH), 1.15, 1.19 (dvt, 2JHhP\31.4 vJHhP\ Hz, Hz, 36H, 2.36 (m, 6H, CH), 3.57 (s, 6.3 3JHhH\5.3 CH3), 24H, 4.81, 5.83 (d, Hz, 2H, crown-CH2), 2JHhH\7.1 xCH2) 6.60, 6.86, 7.40 (m, 5H, 31PM1HN NMR (80 MHz, C6H5).THFd 42.7. 13CM1HN NMR (50 MHz, d 20.78, 21.64 d8) : THF-d8) : (s, 29.51 (vt, Hz, PCH), 71.12 (s, CH3), vJChP\10.1 crown- 120.30 (s, 121.18, 126.07, 128.96 (s, CH, Ph), CH2), xCH2), 165.75 (s, Ph), 182.06 (t, Hz, RewCPh). Cipso , 2JChP\5.0 Assignment of the 13C signals is con–rmed by a gateddecoupled 13C experiment: d 120.30 (dd, 145 Hz, 1JChH\142, xCH2).Acknowledgements Laurier University supported this research in the form Wilfrid of a start-up grant to D.G.G. We thank the Natural Sciences and Engineering Research Council (NSERC) of Canada for the undergraduate student research award to B.R. The competent technical assistance of Ms. Margaret Szymanska is gratefully acknowledged. References 1 M. Y. Darensbourg and C. E. Ash, Adv. Organomet. Chem., 1987, 27, 1. 2 (a) B. Reinhart and D. G. Gusev, submitted. (b) D. G. Gusev, A. J. Lough and R. H. Morris, J. Am. Chem. Soc., in press. 3 Complexes 1»3 aÜord neutral species upon protonation, all unstable by loss. H2 4 Recent reviews : (a) R. H. Crabtree, P. E. M. Siegbahn, O. Eisenstein and A. L. Rheingold, Acc. Chem. Res., 1996, 29, 348; (b) R. H. Crabtree, J. Organomet. Chem., 1998, 577, 111. 5 L. D. Field, A.V. George, T. W. Hambley, E. Y.Malouf and D. J. Young, J. Chem. Soc., Chem. Commun., 1990, 931. L etter 8/07554I New J. Chem., 1999, 1»3 3

ISSN:1144-0546    

DOI:10.1039/a807554i

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

L e t t e r First tungsten complexes with 2º-pyridyl alcoholate ligands : synthesis, structure, and application as novel epoxidation catalysts Wolfgang A. Herrmann,* Joé rg Fridgen, Gerhard M. Lobmaier and Michael Spiegler Anorganisch-chemisches Institut der T echnischen 4, Universitaé t Mué nchen, L ichtenbergstraêe 85747 Garching bei Germany. Mué nchen, E-mail :lit=arthur.anorg.chemie.tu-muenchen.de 10th September 1998, Accepted 26th September 1998 Receiøed Novel tungsten(VI) complexes of the type and WOCl3L [L= 2º-pyridylalcoholate ] are formed from WO2L2 WOCl4 or respectively, and 2º-pyridyl alcohols ; the WO2Cl2 , dioxotungsten(VI) complexes (characterised by X-ray crystallography) can also be obtained from by WO2(acac)2 ligand exchange; they represent catalysts with remarkable activity and excellent product selectivity in ole–n epoxidation. Recently, we reported the synthesis of new dioxomolybdenum( VI) complexes with 2@-pyridyl alcoholate ligands as catalysts for ole–n epoxidation.1 The objective of further research was to develop an easy and straightforward synthesis of analogous dioxotungsten(VI) complexes starting from easily available materials. The stronger Lewis acidity and the greater thermodynamic barrier to the reduction of WVI vs.MoVI should be bene–cial in catalysis. Tungsten(VI) compounds are known for their catalytic activity in ole–n oxidation reactions since Milas reported the metal-catalysed dihydroxylation of ole–ns with and in the 1930s.2 In recent years, H2O2 WO3 tungstates or tungsten-based heteropoly acids were employed as epoxidation catalysts.3,4 Several adducts and chlorine substitution products of have been known since the 1960s.5 Yamanouchi and WOCl4 Yamada reported six-coordinate trichlorooxotungsten(VI) complexes with bidentate SchiÜ-base ligands obtained from salicylaldehyde and various amines.6 To open the –eld of tungsten complexes with 2@-pyridyl alcoholate ligands, WOCl4 was reacted with either alkyl- or aryl-substituted pyridyl alcohols.§ We applied 2-(2@-pyridyl)propan-2-ol as an example for an alkyl-substituted and 9-(2@-pyridyl)—uoren-9-ol for an aryl-substituted ligand.The experiment indicated that the polymeric structure of is broken down easily by these WOCl4 pyridyl alcohols in methylene chloride under mild conditions.Using equimolar amounts of and the respective WOCl4 pyridyl alcohol complexes of the type are formed in WOCl3L almost quantitative yields. The coloured compounds are moisture sensitive solids. The two complexes trichlorooxo[2- (2@-pyridyl)propan-2-olato-N,O]tungsten(VI) 1 (yellow) and trichlorooxo[9-(2@-pyridyl)—uoren-9-olato-N,O]tungsten(VI) 2 (red) were characterised by 1H and 13C NMR spectra.The chemical shift values of the NMR signals of the protons in ortho position to the pyridine nitrogen and the quaternary acarbon atom related to the hydroxy function are useful indicators for the electronic characteristics of the ligand (Table 1). When the pyridyl alcohol is bound to the the WOCl3-unit, signal of the ortho proton is shifted more than 0.5 ppm and the signal of the a-carbon atom is shifted up to 30 ppm to lower –eld.This eÜect can be explained by a shift of electron density towards the tungsten atom. The substitution of further chlorine ligands by a two- or four-fold excess of pyridyl alcohol was not observed under these conditions, obviously due to the lack of a coordination site for an additional nitrogen ligand at the six-coordinate complex Besides the complexes of the type WOCl3L.WOCl3L only the hydrochlorides of the corresponding pyridyl alcohols were formed due to the reaction with HCl. The reason for the little number of known dioxotungsten(VI) complexes is the poor availability of suitable starting materials. Literature-known synthetic routes starting from usually include adducts or ligand exchange reac- WO2Cl2 tions.7h11 McDonell et al.also reported the substitution of chlorine with N,N-dialkylhydroxylamines.9 In our research we performed the direct synthesis of a dioxotungsten(VI) complex with two 2@-pyridyl alcoholate ligands using as start- WO2Cl2 ing material.î The reaction of with two equivalents WO2Cl2 of 2-(2@-pyridyl)propan-2-ol in re—uxing THF aÜorded dioxobis[2-(2@-pyridyl)propan-2-olato-N,O]tungsten(VI) 3 in almost quantitative yield.Since acetylacetonate complexes have proven to be useful in ligand exchange reactions,1,9h13 we chose as a WO2(acac)2 moisture-resistant precursor for further complex synthesis with various pyridyl alcohols. The literature-known Table 1 Representative NMR spectroscopic data Ligand/complex d of (ortho H) d of (a-C) Pyridylcarbinol 8.32 64.11 2-(2@-Pyridyl)propan-2-ol 8.50 71.64 Di[4A,4”-di(methoxy)phenyl](2@-pyridyl)methanol 8.55 80.20 9-(2@-Pyridyl)—uoren-9-ol 8.59 82.68 5-(2@-Pyridyl)-10,11-dihydrodibenzo[a,d]cycloheptan-5-ol 8.51 80.03 Trichlorooxo[2-(2@-pyridyl)propan-2-olato-N,O]tungsten(VI) 1 9.08 99.80 Trichlorooxo[9-(2@-pyridyl)—uoren-9-olato-N,O]tungsten(VI) 2 9.26 100.32 Dioxobis[2-(2@-pyridyl)propan-2-olato-N,O]tungsten(VI) 3 8.72 85.17 Dioxobis[(2@-pyridyl)methanolato-N,O]tungsten(VI) 4 8.56 75.66 DioxobisMdi[4A,4”-di(methoxy)phenyl](2@-pyridyl)methanolato-N,ONtungsten(VI) 5 7.67 94.23 Dioxobis[9-(2@-pyridyl)—uoren-9-olato-N,O]tungsten(VI) 6 9.15 96.76 Dioxobis[5-(2@-pyridyl)-10,11-dihydrodibenzo[a,d]cycloheptan-5-olato-N,O]tungsten(VI) 7 7.57 96.55 New J.Chem., 1999, 5»7 5synthesis9,10 to obtain from was WO2(acac)2 WO2Cl2 improved by using toluene instead of benzene as solvent and –ltering right out of the reaction —ask with a drain tube while still re—uxing. Thus, the yield reported by Yu and Holm10 was increased from 44 to 77% and the reaction time was reduced from 48 to 9 h. We obtained dioxotungsten(VI) complexes of the type with pyridyl alcoholate ligands from WO2L2 by ligand exchange with two equivalents of WO2(acac)2 pyridyl alcohol in methanol.° Depending on the ligand the products precipitate immediately or after reducing the volume of the solvent and yields between 70 and 99% were achieved.We have applied –ve diÜerent pyridyl alcohols and characterised the respective complexes: dioxobis[2-(2@-pyridyl)propan- 2-olato-N,O]tungsten(VI) 3, dioxobis[(2@-pyridyl)methanolato- N,O]tungsten(VI) 4, dioxobisMdi[4A, 4”-di(methoxy)phenyl]- (2@-pyridyl)methanolato-N,ONtungsten(VI) 5, dioxobis[9-(2@- pyridyl)—uoren-9-olato-N,O]tungsten(VI) 6, dioxobis[5-(2@-pyridyl)- 10,11 - dihydrodibenzo[a,d]cycloheptan - 5 - olato -N,O]- tungsten(VI) 7.These complexes form stable –ve-membered ring chelates, which was con–rmed by 1H and 13C NMR spectra (Table 1).The signals of the a-carbon atoms of the bound ligands are shifted about 14 ppm to lower –eld compared to the free pyridyl alcohols. The signals of the ortho protons are shifted to lower –eld in complexes 3, 4 and 6 and to the opposite direction in complexes 5 and 7, which might occur due to an electronic interaction between the pyridine ring and the aromatic substituent at the ligand bound to the dioxotungsten(VI) unit.The IR spectra of 3-7 show two peaks at ca. 900 and 935 cm~1, re—ecting the asymmetric cis-dioxo structure, which is described in the literature.6h11,14 The other peaks correspond to the vibrational signals of the ligands. Crystals of compound 3 suitable for X-ray diÜraction were obtained by adding hexane to a saturated methanol solution.The coordination geometry around the metal centre is similar to analogous molybdenum complexes1 and can be described best as a distorted octahedron (Fig. 1).“ The two oxo ligands form a cis-dioxo unit whereas each nitrogen atom of the pyridine ring is positioned trans to the oxo ligand. The alkoxy functions of the ligands are placed perpendicular to the plane de–ned by the dioxotungsten(VI) core.Fig. 1 Structure of 3 with key atoms labelled. Selected bond lengths and angles (°) : W»O(1) 1.934(3), W»O(2) 1.924(2), W»O(3) 1.725(3), (”) W»O(4) 1.740(3), W»N(1) 2.342(3), W»N(2) 2.361(3), O(1)»C(1) 1.426(4), O(2)»C(2) 1.423(4), N(1)»C(11) 1.342(5), N(2)»C(21) 1.338(4), C(1)»C(11) 1.515(6), C(2)»C(21) 1.519(5) ; O(1)»W»O(2) 149.4(1), O(1)» W»O(3) 103.1(1), 0(1)»W»O(4) 94.8(1), O(1)»W»N(1) 72.0(1), O(1)»W» N(2) 84.2(1), O(2)»W»O(3) 95.7(1), O(2)»W»O(4) 102.9(1), O(2)»W»N(1) 84.6(1), O(2)»W»N(2) 71.5(1), O(3)»W»O(4) 106.2(1), O(3)»W»N(1) 88.9(1), O(3)»W»N(2) 162.2(1), O(4)»W»N(1) 162.1(1), O(4)»W»N(2) 89.1(1), N(1)»W»N(2) 77.8(1), W»O(1)»C(1) 125.0(2), W» O(2)»C(2) 126.5(2), W»N(1)»C(11) 113.2(2), W»N(2)»C(21) 112.9(2), O(1)»C(1)»C(11) 107.4(3), O(2)»C(2)»C(21) 107.7(3), N(1)»C(11)»C(1) 114.4(3), N(2)»C(21)»C(2) 114.5(3).We have selected complexes 3, 5, 6 and 7 and tested them as catalysts (1 mol%) in ole–n epoxidation with tert-butyl hydroperoxide (oxidant) and cis-cyclooctene (substrate) at 70 °C without any addition of solvent.Compound 3 bearing methyl substituents on the ligand achieved the highest yield and quantitative conversion after 60 h whereas compounds 5»7 reached 50»65% conversion. Using 0.1 mol% of 3 62% conversion was achieved. In all cases 100% selectivity towards the epoxide was obtained. Raising the temperature to 85 °C did not signi–cantly increase the catalytic activities of the complexes.At 60 °C the yields were remarkably lower. We consider bulky ligands and a strong tungsten alkoxy bond in the complex to have a negative in—uence on the catalytic activity. The strength of the tungsten»alkoxy bond can be estimated by the NMR data of the quaternary a-carbon atom (Table 1). A series of NMR experiments with rising temperature con–rms that the tungsten»alkoxy bond is broken during catalysis : 3 and tert-butyl hydroperoxide was added to a mixture of and in an NMR tube.The tube was DMSO-d6, C6D6 CDCl3 heated slowly and the 1H NMR spectra showed both signals of the methylene protons starting to collapse at 70 °C, while the signals of the pyridine ring remain unchanged. Further studies to optimise the catalytic conditions including the application of molecular oxygen as oxidant, which has successfully been demonstrated for analogous molybdenum complexes,1 and the development of a synthetic route to dioxotungsten(VI) pyridyl alcoholate complexes from sodium tungstate are in progress. Notes and references § General method for preparation of complexes of the type WOCl3L: was prepared as reported by Gibson et al.15 0.73 mmol of the WOCl4 respective pyridyl alcohol were added to a suspension of (0.25 WOCl4 g, 0.73 mmol) in 20 ml under nitrogen.The resulting coloured CH2Cl2 solution was heated carefully and stirred for a few minutes. Evaporating the solvent aÜorded a coloured moisture-sensitive solid, which was dried in vacuo. Yield : 0.32 g of 1 (98%), 0.41 g of 2 (98%). Satisfactory elemental analyses (C, H, N, W) were obtained.Spectroscopic data for 1: 9.08 (d, 1H, 5.5 dH (CD2Cl2) H6{, 3JH6{, H5{ Hz), 8.15 (dd, 1H, 7.3 Hz, 8.0 Hz), 7.63 (dd, 1H, H4{, 3JH4{, H5{ 3JH4{, H3{ 5.5 Hz, 7.3 Hz), 7.52 (d, 1H, 8.0 H5{, 3JH5{, H6{ 3JH5{, H4{ H3{, 3JH3{, H4{ Hz), 1.78 (s, 6H, H1,3) ; 166.30 148.14 141.67 dc (CD2Cl2) (C2{), (C6{), 125.52 121.11 99.80 (C2), 26.51 (C1,3). (C4{), (C5{), (C3{), î Preparation of 3 from 2.00 mmol (0.27 g) of 2-(2@-pyridyl) WO2Cl2 : propan-2-ol were added to a suspension of 1.00 mmol (0.29 g) of in 20 ml THF under nitrogen.After re—uxing for a few WO2Cl2 minutes 3 started to precipitate. The white solid was –ltered oÜ, washed with a small amount of cold MeOH and dried in vacuo. Yield : 0.44 g of 3 (91%). A satisfactory elemental anlaysis (C, H, N, W) was obtained. ° General method for preparation of complexes of the type WO2L2 from 2.00 mmol of the respective pyridyl alcohol were WO2(acac)2 : added to a solution of (0.41 g, 1.00 mmol) in 25 ml WO2(acac)2 MeOH.Reducing the volume of the solvent to 5 ml aÜords a white precipitate, which was –ltered oÜ, washed with a small amount of cold MeOH and dried in vacuo.Yield : 0.48 g of 3 (98%), 0.30 g of 4 (70%), 0.63 g of 5 (76%), 0.73 g of 6 (99%), 0.75 g of 7 (95%). Satisfactory elemental analyses (C, H, N, W) were obtained. Spectroscopic data for 3: 8.72 (d, 1H, 5.5 dH (CDCl3) H6{, 3JH6{, H5{ Hz), 7.81 (dd, 1H, 7.5 Hz, 8.0 Hz), 7.32 (d, 1H, H4{, 3JH4{, H5{ 3JH4{, H3{ 8.0 Hz), 7.27 (dd, 1H, 5.5 Hz, 7.5 H3{, 3JH3{, H4{ H5{, 3JH5{, H6{ 3JH5{, H4{ Hz), 1.85 (s, 3H, H1,3), 1.73 (s, 3H, H1,3) ; 168.64 (s, dC (CDCl3) C2{), 146.35 (d, 182.5 Hz), 139.88 (d, 163.6 Hz), 123.22 (d, C6{, JCH C4{, JCH 166.1 Hz), 120.39 (d, 157.5 Hz), 85.17 (s, C2), 30.62 (q, C3{, JCH C5{, JCH C1,3, 126.4 Hz), 27.97 (q, C1,3, 127.0 Hz); (KBr pellet) : JCH JCH l(WO2) 932, 893 cm~1; Mass spectrum (CI), m/z 488 ([M]`), 472 ([WO3N2C16H20]).Spectroscopic data for further compounds can be requested from the author.“ Crystal data for 3: M\488.19, orthorhombic, C16H20N2O4W, space group Pbca, a\14.6650(5), b\14.5373(7), c\17.1010(6) ”, V \3645.8(3) Z\8, g cm~3, k(Mo-Ka)\6.36 ”3, Dc\1.779 mm~1, F(000)\1888; all crystallographic measurements were made at 193(1) K using a STOE Image Plate Detecting System. A total of 6 New J.Chem., 1999, 5»724596 re—ections were collected ; 3410 independent re—ections (Rint\ were used in structure re–nement; 0.054) R1\0.0328, wR2\0.0539, GOF\0.954 for 3410 re—ections and 288 parameters. The contribution of a disordered solvent molecule to the re—ection data has been eliminated by the ìì calc squeezeœœ option (program: PLATON).16h19 CCDC reference number 440/075.See http ://www.rsc.org/suppdata/ njc/1999/5 for crystallographic –les in cif format. 1 W. A. Herrmann, G. M. Lobmaier, T. Priermeier, M. R. Mattner and B. Scharbert, J. Mol. Catal. A, 1997, 117, 455. 2 N. Milas, J. Am. Chem. Soc., 1937, 22, 2342. 3 K. Sato, M. Aoki, M. Ogawa, T. Hashimoto and R. Noyori, J. Org. Chem., 1996; 61, 8310; C. Aubry, G. Chottard, N. Platzer, J.-M.Bregeault, R. Thouvenot, F. Chauveau, C. Huet and H. Ledon, Inorg. Chem., 1991, 30, 4409; C. Venturello and R. DœAloisio, J. Org. Chem., 1988, 53, 1553; C. Venturello, E. Alneri and M. Ricci, J. Org. Chem., 1983, 48, 3831; J. Prandi, H. B. Kagan and H. Mimoun, T etrahedron L ett., 1986, 27, 2617. 4 R. Neumann and M. Gara, J. Am. Chem. Soc., 1995, 117, 5066; 1994, 116, 5509; Y. Ishii, K.Yamawaki, T. Ura, H. Yamada, T. Yoshida and M. Ogawa, J. Org. Chem., 1988, 53, 3587; D. C. Duncan, R. C. Chambers, E. Hecht and C. L. Hill, J. Am. Chem. Soc., 1995, 117, 681; A. C. Dengel, W. P. Griffith and B. C. Parkin, J. Chem. Soc., Dalton T rans., 1993, 2683. 5 H. Funk, W. Weiss and G. Mohaupt, Z. Anorg. Chem., 1960, 304, 238; H. Funk and G. Mohaupt, Z. Anorg. Chem., 1962, 315, 204; G. W.A. Fowles and J. L. Frost, J. Chem., Soc. A, 1967, 671; 1966, 1631. 6 K. Yamanouchi and S. Yamada, Inorg. Chim. Acta, 1975, 12, 9. 7 K. Yamanouchi and S. Yamada, Inorg. Chim. Acta, 1974, 11, 223. 8 C. A. Rice, P. M. H. Kroneck and J. T. Spence, Inorg. Chem., 1981, 20, 1996. 9 A. C. McDonell, S. G. Vasudevan, M. J. OœConnor and A. G. Wedd, Aust. J. Chem., 1985, 38, 1017. 10 S. Yu and R. H. Holm, Inorg. Chem., 1989, 28, 4385. 11 M. R. Maurya, D. C. Anthony, S. Gopinathan, V. G. Puranik, S. S. Tavale, C. Gopinathan and R. C. Maurya, Bull. Chem. Soc. Jpn., 1995, 68, 2847. 12 R. L. Dutta and A. K. Pal, Indian J. Chem., Sect. A, 1985, 24, 76. 13 A. Syamal and M. R. Maurya, Coord. Chem. Rev., 1989, 95, 183; M. R. Maurya, S. Gopinathan, C. Gopinathan, and R. C. Maurya, Polyhedron, 1993, 12, 159; M. R. Maurya and C. Gopinathan, Polyhedron, 1993, 12, 1039; M. R. Maurya and C. Gopinathan, Bull. Chem. Soc. Jpn., 1993, 66, 1979. 14 B. J. Brisdon, Inorg. Chem., 1967, 6, 1791. 15 V. C. Gibson, T. P. Kee and A. Shaw, Polyhedron, 1988, 7, 579. 16 IPDS Operating System Version 2.87 Stoe&Cie. GmbH Darmstadt, Germany, 1997. 17 A. Altomare, G. Cascarano, C. Giacovazzo, A. Guagliardi, M. C. Burla, G. Polidori and M. Camalli, SIR-92, University Bari, Italy, 1992. 18 G. M. Sheldrick, SHELXL-93, Crystallographic Computing 3, ed. G. M. Sheldrick, C. Krué ger and R. Goddard, Oxford University Press, England, 1993, pp. 175»189. 19 A. L. Spek, Acta Crystallogr., Sect. A., 1990, 46, C34. L etter 8/08343F New J. Chem., 1999, 5»7 7

ISSN:1144-0546    

DOI:10.1039/a808343f

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

L e t t e r N N N O NHR2 N O OH OH HO O O HO O HO O P O P HO O O OH OH R1 NH NH HN O OH O O OHO HO NH2 OH O Guanofosfocin A : R1=Me, R2=H B : R1=H, R2=H C : R1=H, R2= 8 5' 1" Synthesis of novel nucleosideñcarbohydrate hybrids Kevin Stans–eld, Hideyuki Kanamori and Hideyuki Sugimura*§ T he Noguchi Institute, Kaga 1-8-1, Itabashi-Ku, T okyo 173, Japan. E-mail : sugimura=ed.ynu.ac.jp 30th September 1998, Accepted 28th October 1998 Receiøed A substitution approach for the synthesis of 8-(mannosyloxy)- adenosines, novel nucleosideñcarbohydrate hybrids found in a new family of chitin synthase inhibitor»guanofosfocins, has been investigated.When an N6-benzoyl-8-bromoadenosine derivative was exposed to the sodium alkoxides derived from di-O-isopropylidene-â-mannoses or cyclohexanols, substitution occurred at room temperature to yield 8-(mannosyloxy)- adenosines and their 8-cyclohexyloxy derivatives.The 8-mannosyloxy products were isolated solely as a anomers and adopted a syn conformation with respect to the anomeric CwN bond. Whilst the hybrid compounds were obtainable in good yields, the new glycosidic linkage was acid sensitive to furnish 8-oxo derivatives under deprotecting conditions for the dimethoxytrityl group.The synthesis of structurally unique nucleoside analogs is of considerable importance in the development of antiviral, antibiotic, antitumor and antifungal agents.1 Naturally occurring compounds often provide inspiration for the creation of such analogs and a recently isolated family of chitin synthase inhibitor, guanofosfocins, stimulated our interest.The guanofosfocins, isolated from the fermentation broths of Streptomyces sp. and T richoderma sp., have a unique cyclic structure containing a glycosidic bond between the 8-position of guanosine and a D-mannose moiety (Fig. 1).2 Despite their promising therapeutic eÜects against fungal diseases, further investigation of these fascinating molecules has been hindered by their low stability.3 Since the ethereal bond between a glycosyl moiety and the 8-position of a purine nucleoside found in guanofosfocin is previously unknown, we planned to develop methodology for the efficient construction of this Fig. 1 Structure of guanofosfocin A»C. linkage in order to synthesise novel nucleoside»carbohydrate hybrids for screening and to evaluate their stability. One possible strategy for the synthesis of an 8-(glycosyloxy) purine nucleoside would be to substitute a protected carbohydrate-type nucleophile into a purine substrate bearing a leaving group at the 8-position.To date, simple 8- alkoxypurine nucleosides (8-methoxy and 8-benzyloxy substituted derivatives) have been prepared by such a method.4h6 However, the reaction conditions (e.g.non-protected nucleosides are treated with excess alcohol and sodium metal in DMSO at 65 °C or higher) are unsuitable for sensitive carbohydrate-type nucleophiles. Therefore, prior to the introduction of a carbohydrate moiety, we endeavored to establish a general substitution procedure using an easily accessible purine nucleoside substrate, 8-bromoadenosine, and a simple secondary alcohol, cyclohexanol.As the substitution reaction conditions will be basic, the protecting groups employed on 8-bromoadenosine must be base stable. We –rst prepared a 2@,3@-O-isopropylidene-5@-O, N6-ditrityl substrate 1a, which was treated with sodium cyclohexyloxide generated in situ from cyclohexanol and sodium hydride in DMF.î Whilst no reaction occurred at room temperature, on heating the mixture at 70 °C the reaction proceeded and, after 24 h, the starting nucleoside was almost consumed to provide 8-(cyclohexyloxy)adenosine derivative 2a in 65% yield along with a 22% yield of 8-oxoadenosine 4 (run 1 in Table 1).A similar tendency was observed in the reaction of N6 free substrate 1b (run 2). By contrast, the reaction of New J.Chem., 1999, 9»11 9R1O N N N N R2HN O O O Br O Hydrolysis 3 4 Na R1O N N N HN NHR2 O O O O N6-benzoylated substrate 1c was completed at room temperature in 4 h to aÜord 8-cyclohexyloxy product 2c in 75% yield (run 3). It has been reported that an N2-acyl group on guanosine derivatives enhances the rate of intramolecular cyclization at the 8-position.7 The benzoyl group at N6 of adenosine increases the rate of substitution by a similar electronic eÜect.The formation of 8-oxo byproduct 4 is due to hydrolysis of an addition intermediate 3 (Fig. 2) as attempts to form 4 directly by treating the substrates with sodium hydride alone in wet DMF were unsuccessful. Prolonged reaction, however, didnœt lead to improvement in the yield of 2.Since selective deprotection at the 5@-position is required for further studies, other protecting groups at this position were brie—y examined. 5@-O-tert-Butyldiphenylsilyl-protected substrate 1d underwent desilylation under the reaction conditions to yield 5@-hydroxyl product 2d (run 4). 5@-O-Dimethoxytrityl (DMTr) derivative 1e, in which the DMTr group can be removed more readily than the Tr group,8 reacted with cyclohexanol to furnish the desired product 2e in good yield (run 5).Several other base»solvent combinations were also investigated (DMSO as an alternative solvent and ButOK/18-crown- 6, as an alternative base), but sodium hydride in DMF Et3N gave the most satisfactory results. With a suitable substrate in hand, we turned our attention to the substitution reaction of carbohydrate-type nucleophiles (Table 2).When 2,3,4,6-tetra-O-benzylmannopyranose (5) was allowed to react with 1e, 8-(benzyloxy)adenosine 9 was isolated (run 1). From this result, it was deduced that base catalyzed ring opening of 5 was occurring and subsequent b-elimination of benzyloxide resulting in the formation of the 8-benzyloxy product. We anticipated that protection of the 2- and 3-hydroxy groups as a cyclic acetal would restrain this process, allowing the desired reaction to proceed.Indeed, when 2,3 ;4,6-di-O-isopropylidene-D-mannopyranose (6), prepared by a known route,9 was used as a nucleophile, 8- (mannopyranosyloxy)adenosine derivative 11 was successfully obtained in good yield (run 2). Similarly, the corresponding 8-(mannofuranosyloxy)adenosine derivative 12 and the 8- (mannopyranosylthio)adenosine derivative 13 have been prepared using 2,3 ;5,6-di-O-isopropylidene-D-mannofuranose (7) and 2,3 ;4,6-di-O-isopropylidene-1-thio-D-mannopyranose (8), respectively (runs 3 and 4).The mannose»adenosine hybrids obtained here were formed as single isomers. Generally, the con–guration of glycosidic linkages in pyranoses can be determined from 13C NMR spectra by measuring the coupling at C-1.10 In hybrid 11, 1JCH the coupling constant is 178 Hz, showing the new linkage to be exclusively a10,11 which is consistent with the mannosyl bond in guanofosfocins.In addition to the new carbohydrate»nucleoside hybrids, a substrate derived form myo-inositol was employed in the reaction to create a novel 8-inositol substituted adenosine compound 14 (run 5).Fig. 2 A plausible mechanism for the formation of 8-oxo byproduct. Next, we brie—y examined deprotection at the 5@-position. Dimethoxyltrityl group can be removed by trichloroacetic acid in nitromethane»methanol solvent system with little or no concomitant depurination.12 However, when these conditions were applied to the nucleoside»carbohydrate hybrids 11»13, the glycosidic bond was cleaved to give the 8-oxo or 8-thioxo product 17 (runs 1»3 in Table 3).The 8- cyclohexyloxy derivatives 2e and 14 were acid stable, so that the dimethoxytrityl group could be removed without difficulty (runs 4 and 5). EÜorts to investigate the selective deprotection conditions, not aÜecting the glycosidic bond, are currently underway.One intriguing feature of the guanofosfocins (Fig. 1) is that they clearly adopt an anti conformation about the anomeric CwN bond though purine nucleosides with bulky substituent at C-8 are known to have a syn conformation.13h15 It is well documented that the chemical shifts of H(2@) protons in syn and anti adenosines diÜer through changes in the torsion angle around the glycoside bond.13h15 It is therefore possible to predict the glycosyl conformation of the hybrids by comparing their shifts with compounds of known orientation. 8- Bromoadenosines are known to adopt syn conformations with the H(2@) signal characteristically located 0.5»0.6 ppm down- –eld from the H(3@) signal.13 In this respect, hybrid compounds 11 and 12 show patterns similar to 1e, indicating a syn orientation.The 8-(cyclohexyloxy)adenosine 2e also shows a typical syn pattern. However, on removal of the bulky dimethoxytrityl group, the H(2@) signal of compound 15 shifted 10 New J. Chem., 1999, 9»11DMTrO N N N N NHBz O O O Nu HO N N N N NHBz O O O Nu HO N N N HN NHBz O O O X O OBn O OBn O O O 3% CCl3CO2H MeOH:MeNO2 r.t., 15 min 15, 16 17 Run Compound Product Nu Yield (%) 1 11 17 (X=O) — 63 2 12 17 (X=O) — 64 4 2e 15 95 5 14 16 88 Table 3 Deprotection of DMTr group 3 13 17 (X=S) — 75 up–eld to give a 1H NMR spectra consistent with an anti relationship. This –nding is signi–cantly important because an anti relationship is bene–cial for future attempts to prepare guanofosfocin type compounds.In summary, by successfully forming the purine» carbohydrate bond found in guanofosfocin we have prepared a new type of nucleoside»carbohydrate hybrid.The approach will be compatible with the use of a variety of nucleophiles. The glycosidic linkage is extremely acid labile and this may contribute to the instability of the natural products. Further study on the application of this approach to the preparation of a wide variety of nucleoside hybrid molecules, including the use of guanosine substrates, is in progress.Acknowledgement wish to thank the JSPS (Japan Society for the Promotion We of Science) for the award of a Postdoctoral Fellowship to K. S. Notes and references § Current address : Faculty of Education and Human Science, Yokohama National University, Yokohama 240-8501, Japan. î The following general procedure was used for the substitution reaction : an alcohol (0.2 mmol) was added to a suspension of sodium hydride (60% suspension in oil, 0.2 mmol) in DMF (3 ml) at room temperature and the mixture stirred for 15 minutes.An 8- bromoadenosine derivative (0.1 mmol) was added in a single portion to the stirred solution. The reaction was quenched with water (5 ml).The mixture was extracted with ethyl acetate (3]15 ml). The organic layer was washed with a saturated solution of sodium hydrogen carbonate (30 ml) and then dried The solvent was concentrated (MgSO4). in vacuo. Puri–cation by column chromatography aÜorded the corresponding 8-alkoxyadenosine derivative. 1 For relevant reviews see : C. Peç rigaud, G. Gosselin and J.-L.Imbach, Nucleosides Nucleotides, 1992, 11, 903; D. M. Huryn and M. Okabe, Chem. Rev., 1992, 92, 1745. 2 H. Katoh, M. Yamada, K. Lida, M. Aoki, Y. Itezono, N. Nakayama, Y. Suzuki, M. Watanabe, H. Shimada, H. Fujimari, N. Nagata, S. Ohshima, J. Watanabe and T. Kamiyama, Abstract of the 38th Symposium on the Chemistry of Natural Products, Sendai, Japan, 1996, pp. 115»120. (Chem. Abstr., 1996, 125, 322575). 3 Guanofosfocins are extremely unstable to oxygen, on heating and are completely decomposed in an aqueous solution at room temperature within 24 h [T. Kamiyama, Nippon Nogeikagaku Kaishi, 1997, 71, 535 (in Japanese)]. 4 R. E. Holmes and R. K. Robins, J. Am. Chem. Soc., 1965, 87, 1772. 5 G. S. Buenger and V. Nair, Synthesis, 1990, 962. 6 J. A. Secrist III, A. Shortnacy-Fowler, L. L. Bennett, Jr. and J. A. Montgomery, Nucleosides Nucleotides, 1994, 13, 1017. 7 K. Kameyama, M. Sako, K. Hirota and Y. Maki, J. Chem. Soc., Chem. Commun., 1984, 1658. 8 M. Smith, D. H. Rammler, I. H. Goldberg and H. G. Khorana, J. Am. Chem. Soc., 1962, 84, 430. 9 J. Gelas and D. Horton, Carbohydr. Res., 1978, 67, 371. 10 K. Bock and C. Pedersen, J. Chem. Soc., Perkin T rans. 2, 1974, 293. 11 F. Barresi and O. Hindsgaul, Can. J. Chem., 1994, 72, 1447. 12 H. Takaku, K. Morita and T. Sumiuchi, Chem. L ett., 1983, 1661. 13 R. H. Sarma, C.-H. Lee, F. E. Evans, N. Yathindra and M. Sundaralingam, J. Am. Chem. Soc., 1974, 96, 7337. 14 M. Ikehara, S. Uesugi and K. Yoshida, Biochemistry, 1972, 11, 830. 15 F. E. Evans and N. O. Kaplan, J. Biol. Chem., 1976, 251, 6791. L etter 8/07594H New J. Chem., 1999, 9»11 11

ISSN:1144-0546    

DOI:10.1039/a807594h

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

L e t t e r Sawhorse connections in a Ag(I)-nitrite coordination network: { [Ag(pyrazine) ]NO2}= Alexander J. Blake,a Neil R. Champness,*a Marcello Crewa and Simon Parsonsb a School of Chemistry, T he University of Nottingham, University Park, Nottingham, UK NG7 2RD. E-mail : pcznc=unix.ccc.nottingham.ac.uk b Department of Chemistry, T he University of Edinburgh, W est Mains Road, Edinburgh, UK EH9 3JJ (in Cambridge, UK) 15th October 1998, Accepted 5th November 1998 Receiøed The Ag(é) coordination networks, and { [Ag(pyrazine) ]NO2 }= have been constructed in order to { [Ag(4,4º-bipy) ]NO2 }= investigate the eÜect of anion upon network topology ; an unusual sawhorse connection is observed in the structure of with the nitrite anion acting to ìblockœ { [Ag(pyrazine) ]NO2 }= cis coordination sites.The preparation of extended networks using inorganic coordination polymers has become an area of increasing study in recent years.1 One of the reasons that this interest has arisen is because the synthetic procedure used to construct these materials allows a high degree of design. Ultimately this may lead to the development of materials with tuneable properties including structures with host»guest properties similar to those observed with zeolites,2 and compounds with interesting electronic or magnetic properties. The high degree of design arises from the coupling of the well understood coordination properties of individual metal ions and highly-developed ligand syntheses with the newer areas of supramolecular chemistry and crystal engineering.3 We have been studying the eÜect of individual building-blocks upon network structure.This has included the control of network topology and interpenetration in adamantoid networks via ligand design4 and studies on the eÜect of variation of solvent of crystallisation upon network structure.5 The nature of the counter-anion has also been shown to have a dramatic eÜect upon network topology and this is particularly noticeable in Ag(I) chemistry.6,7 Recently we have demonstrated that replacement of or with results in a fun- AgBF4 AgPF6 AgNO3 damental change of the extended structure of a coordination polymer with the ligand 3,6-di-4-pyridyl-1,2,4,5-tetrazine due to interactions between the Ag(I) centre and the anion NO3~ to give a ì helical staircase œ structure.6 We now report the extension of these investigations to the use of and AgNO2 , report an unusual example of a sawhorse connection within an extended network.(pyz\pyrazine) and M[Ag(4,4@-bipy)] M[Ag(pyz)]NO2N= (4,4@-bipy\4,4@-bipyridyl) were prepared as colourless NO2N= microcrystalline samples by adding a solution of the appropriate ligand in EtOH to a solution of in In order AgNO2 H2O.§ to assess the eÜect of the nitrite counter-anion upon network topology single crystals of both complexes were grown by slow diÜusion between an aqueous solution of and an AgNO2 ethanolic solution of the ligand.exists as a M[Ag(pyz)]NO2N=î three-dimensional network in which each Ag(I) ion adopts a distorted octahedral environment (Fig. 1) : each Ag(I) centre is coordinated by two pyrazine ligands, Ag»N 2.277(5) which Aé , bridge adjacent Ag(I) ions, and by the chelating nitrite counter-anion, Ag»O 2.487(6) in the equatorial sites.The Aé , two remaining axial coordination sites are occupied by weak Ag… … …Ag interactions (Fig. 1). The Ag… … …Ag separation of 3.2168(3) is a typical value for Ag… … …Ag interactions unsup- Aé ported by ligands.8 Ag… … …Ag interactions have been found to be signi–cant in the extended structures of inorganic supramolecular networks.This is perhaps best illustrated by the formation of short interactions [Ag… … …Ag 2.970(2) in the Aé ] extended three-dimensional network formed by M[Ag(4,4@- bipy)] in which each Ag(I) centre is coordinated by NO3N=,9 one N-donor from each of two 4,4@-bipy ligands and participates in one Ag… … …Ag interaction to give a T-shaped motif.9 It can be seen that in each Ag(I) ion acts M[Ag(pyz)]NO2N= as a sawhorse junction in the network, with the nitrite blocking two cis sites of the junction (Fig. 2). To our knowledge this represents the –rst example of such a junction within a coordination polymer array. Sawhorse junctions are extremely rare in inorganic framework structures with the most notable examples being and Therefore the overall IrF4 , RhF4 PtF4 .10 network topology (Fig. 2) can be thought of as being related to the solid-state structure of which has been described IrF4 as octahedra which share 4 F atoms, each with one ìMIrF6N other group leaving a pair of cis vertices MIrF6N Fig. 1 View of the Ag(I) coordination environment in Selected bond lengths and angles (°) : Ag(1)» M[Ag(pyz)]NO2N=.(Aé ) N(1) 2.277(5), Ag(1)»O(5) 2.487(6), Ag(1)… … …Ag(1iii), Ag(1)… … …Ag(1iv) 3.2168(3), Ag(1iii)»Ag(1)»Ag(1iv) 158.59(5), Ag(1iv)»Ag(1)»N(1) 110.1(1), Ag(1iii)»Ag(1)»N(1) 79.9(1), Ag(1iv)»Ag(1)»O(5ii) 66.5(1), Ag(1iii)»Ag(1)» O(5ii) 93.5(1), N(1)»Ag(1)»O(5ii) 140.7(2), N(1ii)»Ag(1)»O(5) 92.8(2), O(5)»Ag(1)»O(5ii) 50.1(2), N(1)»Ag(1)»N(1ii) 126.0(3) (symmetry codes: 1[z; ii\[x, y, iii\[x [y, [z ; i\12 [x, 12 [y, 12 [z ; iv\[x, [y, 1[z; v\x[12 , 12 [y, z[12 ).New J. Chem., 1999, 13»15 13Fig. 2 The overall network structure of (a) view M[Ag(pyz)]NO2N=: down the c-axis, (b) view illustrating the development of the threedimensional lattice through Ag… … …Ag contacts and showing the sawhorse connections (nitrite anions are removed for clarity) (Ag, cross-hatched ; O, hatched; N, dotted, C, plain).unsharedœ.10,11 Similarly the network of is M[Ag(pyz)]NO2N= built from octahedra with cis vertices unshared. The structure of contrasts with that M[Ag(pyz)]NO2N= observed in the corresponding salt, NO3~ Extended chains of alternating Ag(I) ions M[Ag(pyz)]NO3N=.12 and pyrazine ligands are observed in and M[Ag(pyz)]NO3N=12 signi–cantly the nitrate anion is non-coordinating, in contrast to the behaviour of the nitrite anion observed in M[Ag(pyz)]NO2N=.Single crystal X-ray studies of M[Ag(4,4@-bipy)]NO2N=î reveal that a diÜerent network structure is adopted to that observed for The structure consists of M[Ag(pyz)]NO2N=. slightly distorted linear chains of alternating Ag(I) ions and 4,4@-bipy ligands, N»Ag»N 171.98(10)° (Fig. 3) with the NO2~ anions sitting between adjacent M[Ag(4,4@-bipy)] chains so `N= that each Ag(I) centre forms two weak Ag… … …O interactions of 2.667(2) and one weak Ag… … …N interaction of 2.978(3) Aé Aé . Signi–cantly, no Ag… … …Ag interactions are observed in M[Ag(4, 4@-bipy)] in contrast to The bridg- NO2N=, M[Ag(pyz)]NO2N=.ing 4,4@-bipy ligands in M[Ag(4,4@-bipy)] adopt a very NO2N= twisted arrangement, with a dihedral angle between the Fig. 3 View of the structure of M[Ag(4,4@-bipy)] showing the NO2N= linear chains of alternating Ag(I) ions and 4,4@-bipy ligands. Selected bond lengths and angles (°) : Ag(1)»N(1) 2.208(3) ; N(1)»Ag(1)»N(1i) (Aé ) 171.98(10) (symmetry codes: y, i\[x]12 , [z]12 ).pyridyl rings of 41.3°. This value compares with observed values of 4.3° in [MCu(cnge)2N2(l-4,4@-bipy)][BF4]213 (cnge\2-cyanoguanidine), and 28.0, 30.0° observed in [Cu(4, 4@-bipy)(MeCN) Both twisted and —at 4,4@-bipy mol- 2]BF4 .13 ecules are incorporated in [Cu(l-4,4@-bipy)- with a dihedral angle of 9.29° (H2O)2(FBF3)2] … 4,4@-bipy, being observed for the coordinated 4,4@bipy ligands.In contrast the non-coordinated 4,4@-bipy ligands are constrained to be ideally planar by crystallographic symmetry.14 The IR spectra of the two complexes and M[Ag(pyz)]NO2N= M[Ag(4,4@-bipy)] are consistent with the non- NO2N= coordinating nitrite anion in M[Ag(4,4@-bipy)]NO2N= cm~1] and the chelating mode of coordi- [lsym(NO2)\1243 nation for the anion in M[Ag(pyz)]NO2N= [lsym(NO2)\1269 cm~1].17 IR spectra of both the precipitated (microcrystalline) products and single crystals were found to be identical con- –rming the crystal structures to be representative of the bulk.Of the few previously reported examples of structurally characterised complexes both strongly coordinated15 AgNO2 and uncoordinated/weakly interacting16 have been NO2~ reported. In the former Ag»O bond lengths are comparable to those observed here.15 Current work is aimed at studying the wider application of the nitrite anion as a fundamental building-block of extended coordination polymers and investigating the use of anions as a controlling factor in Ag(I) supramolecular networks. Acknowledgements thank the EPSRC, Nuffield Foundation and the Uni- We versity of Nottingham for support.Notes and references to a solution of (100 mg, 0.65 mmol) in § M[Ag(pyz)]NO2N=: AgNO2 water (10 cm3) heated to 50 °C was added a solution of pyrazine (52 mg, 0.65 mmol) in EtOH (10 cm3). A colourless crystalline precipitate formed over the period of 5 minutes and the suspension was then stirred for a further 5 min.The product was –ltered oÜ and washed with diethyl ether and then dried in vacuo. Yield 99 mg, 65%. (Found: C, 20.26 ; H, 1.59 ; N, 17.39. Calc. for C, 20.51 ; H, C4H4Ag1N3O2: 1.71 ; N, 17.95%). IR (KBr)/cm~1: 2928w, 2857w, 1507w, 1411w, 1269s, 1143w, 1095w, 1024w, 800s. M[Ag(4,4@-bipy)] was prepared similarly as a colourless pre- NO2N= cipitate in 75% yield. (Found: C, 38.32 ; H, 2.37 ; N, 13.38.Calc. for C, 38.71 ; H, 2.58 ; N, 13.55%). IR (KBr)/cm~1: C10H8Ag1N3O2: 3051w, 3027w, 1599s, 1526s, 1487m, 1410m, 1314m, 1243s, 1222s, 1091w, 1070m, 1039w, 1000m, 992m, 963w, 880w, 846m, 804s, 732w, 667w, 621s, 565m, 510s. î Crystal data : M\233.96, M[Ag(pyz)]NO2N=: C4H4AgN3O2 , monoclinic, space group C2/c (no. 15), a\12.378(5), b\7.934(3), c\ b\99.07(3)°, U\613.04 Z\4, g cm~3, 6.322(3) Aé , Aé 3, Dc\2.53 14 New J.Chem., 1999, 13»15k(Cu-Ka)\26.42 mm~1. A colourless plate (0.31]0.21]0.06 mm) was used for data collection on a Stoe Stadi-4 four-circle diÜractometer (graphite monochromated Cu-Ka radiation, u»h scans, hmax\70°) equipped with an Oxford Cryosystems low-temperature device operating at 220(2) K.18 Absorption corrections utilised t-scan data (lamina procedure, Of 554 unique re—ec- Tmin\0.203, Tmax\0.785).tions 480 with IP2p(I) were used in all calculations. The (Rint\0.10) structure was solved using Patterson methods19 and all non-H atoms were located in a subsequent diÜerence-Fourier map.20 The structure was re–ned against F with all non-H atoms modelled with anisotropic displacement parameters and H-atoms placed in calculated positions.At –nal convergence20 [I[2p(I)]\0.0443, for 48 R1 Rw\0.0459 parameters, S\1.05, (D/p)max\0.02, *omax\ 1.65 e [located 0.9 from Ag(1)], e Aé ~3 Aé *omin\[1.28 Aé ~3. M[Ag(4,4@-bipy)] M\310.06, monoclinic, NO2N=: C10H8AgN3O2 , space group P2/n (no. 13, alt. 2), a\7.2931(12), b\6.08(2), c\11.481(3) b\102.817(17)°, U\496.8 Z\2, g Aé , Aé 3, Dc\2.073 cm~3, k(Cu-Ka)\16.195 mm~1.A colourless needle (0.44]0.07]0.02 mm) was used. Data were collected as for Absorption corrections were carried out by M[Ag(pyz)]NO2N=. Gaussian integration following re–nement of the crystal size and morphology against a set of t-scans,21 880 (Tmin\0.198, Tmax\0.720), unique re—ections of which 838 had IP2p(I), were (Rint\0.014), used in all calculations. The structure was solved using direct methods22 and all non-H atoms were located in a subsequent diÜerence-Fourier map.23 The structure was re–ned against F2 with all non-H atoms modelled with anisotropic displacement parameters; H-atoms were placed in calculated positions and were allowed to ride on their parent atoms.The weighting scheme w~1\[p2(Fo2) ](0.0335P)2]0.0485P], was adopted and P\[max(Fo2 , 0)]2Fc2]/3 the extinction coefficient re–ned to 0.0010(2). At –nal convergence23 [I[2p(I)]\0.0198, (all data)\0.0503 for 75 parameters, R1 wR2 S\1.081, in –nal cycle\\0.001, e (D/p)max *omax\0.50 Aé ~3, e *omax\[0.42 Aé ~3.CCDC reference number 440/081. See http ://www.rsc.org/ suppdata/njc/1998/13/ for crystallographic –les in cif format. 1 S. R. Batten and R. Robson, Angew. Chem., Int. Ed. Engl., 1998, 37, 1460; A. J. Blake, N. R. Champness, P. Hubberstey, W-S. Li, M. Schroé der and M. A. Withersby, Coord. Chem. Rev., 1998, in press ; N. R. Champness and M. Schroé der, Curr. Opin. Solid State Mater. Sci., 1998, 3, 419. M. Munakata, L. P. Wu and T. Kuroda-Sowa, Bull. Chem. Soc. Jpn., 1997, 70, 1727. 2 C. Janiak, Angew.Chem., Int. Ed. Engl., 1997, 36, 1431. 3 G. R. Desiraju in Crystal Engineering : Design of Organic Solids, Elsevier, Amsterdam, 1989; T he Crystal as a Supramolecular Entity, ed. G. R. Desiraju, Wiley, 1995; G. R. Desiraju, Angew. Chem., Int. Ed. Engl., 1995, 34, 2311; C. B. Aakeroy, Acta Crystallogr., Sect. B, 1997, 53, 569. 4 A. J. Blake, N. R. Champness, S. S. M. Chung, W-S. Li and M.Schroé der, Chem. Commun., 1997, 1005; A. J. Blake, N. R. Champness, A. N. Khlobystov, D. A. Lemenovski, W-S. Li and M. Schroé der, Chem. Commun., 1997, 1339. 5 M. A. Withersby, A. J. Blake, N. R. Champness, P. Hubberstey, W-S. Li and M. Schroé der, Inorg. Chem., submitted. 6 M. A. Withersby, A. J. Blake, N. R. Champness, P. Hubberstey, W-S. Li and M. Schroé der, Angew. Chem., Int.Ed. Engl., 1997, 36, 2327. 7 L. Carlucci, G. Ciani, D. M. Proserpio and A. Sironi, Angew. Chem., Int. Ed. Engl., 1995, 34, 1895; L. Carlucci, G. Ciani, D. M. Proserpio and A. Sironi, J. Am. Chem. Soc., 1995, 117, 4562; D. Venkataraman, S. Lee, J. S. Moore, P. Zhang, K. A. Hirsch, G. B. Gardner, A. C. Covey and C. L. Prentice, Chem. Mater., 1996, 8, 2030; K. A. Hirsch, S.R. Wilson and J. S. Moore, Inorg. Chem., 1997, 36, 2960. 8 P. Pyykkoé , Chem. Rev., 1997, 97, 597; A. D. Burrows, M. F. Mahon and M. T. Palmer, J. Chem. Soc., Dalton T rans., 1998, 1941; M. A. Omary, T. R. Webb, Z. Assefa, G. E. Shankle and H. H. Patterson, Inorg. Chem., 1998, 37, 1380. 9 F. Robinson and M. J. Zaworotko, J. Chem. Soc., Chem. Commun., 1995, 2413; O. M. Yaghi and H.Li, J. Am. Chem. Soc., 1996, 118, 295. 10 A. F. Wells, Structural Inorganic Chemistry, Oxford University Press, 5th edn., 1983. 11 N. N. Greenwood and A. Earnshaw, Chemistry of the Elements, Pergamon Press, 1984; N. Bartlett and A. Tressaud, Compt. Rend., 1974, C278, 1501. 12 R. G. Vranka and E. L. Amma, Inorg. Chem., 1966, 5, 1020. 13 A. S. Batsanov, M. J. Begley, P. Hubberstey and J.Stroud, J. Chem. Soc., Dalton T rans., 1996, 1947. 14 A. J. Blake, S. J. Hill, P. Hubberstey and W-S. Li, J. Chem. Soc., Dalton T rans., 1997, 913. 15 H. Lang, M. Herres and L. Zsolnai, Organometallics, 1993, 12, 5008. 16 R. H. Benno and C. J. Fritchie, Jr., Acta Crystallogr., Sect. B, 1973, 29, 2493. 17 K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordination Compounds, Wiley, New York, 4th edn., 1986. 18 J. Cosier and A. M. Glazer, J. Appl. Crystallogr., 1986, 19, 105. 19 P. T. Beurskens, G. Beurskens, W. P. Bosman, R. de Gelder, S. Garcia-Granda, R. O. Gould, R. Israel and J. M. M. Smits. The DIRDIF-96 program system. Crystallography Laboratory, University of Nijmegen, The Netherlands, 1996. 20 D. J. Watkin, C. K. Prout, R. J. Carruthers and P. Betteridge, CRYSTALS Issue 10, Chemical Crystallography Laboratory, Oxford, UK, 1996. 21 X-SHAPE»crystal optimisation for absorption correction, Stoe and Cie, Darmstadt, Germany, 1996. 22 A. Altomare, G. Cascarano, G. Giacovazzo, A. Guagliardi, M. C. Burla, G. Polidori and M. Camalli. SIR-92, J. Appl. Crystallogr., 1994, 27, 435. 23 G. M. Sheldrick, SHELXL-97, University of Goé ttingen, Germany, 1997. L etter 8/08025I New J. Chem., 1999, 13»15 15

ISSN:1144-0546    

DOI:10.1039/a808025i

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Phthalic acid, a versatile building block in organic-organometallic crystal engineering§ Dario Braga,*a Alessandro Angeloni,a Lucia Maini,a Andreas W. Go é tza and Fabrizia Grepioni*b a Dipartimento di Chimica G. Ciamician, di Bologna, V ia Selmi 2, 40126 Bologna, Universitaç Italy. E-mail : dbraga=ciam.unibo.it;homepage:http ://catullo.ciam.unibo.it E-mail : grepioni=ssmain.uniss.it b Dipartimento di Chimica, di Sassari, V ia V ienna 2, 07100, Sassari, Italy Universita` (in Montpellier, France) 17th August 1998, Accepted 2nd 1998 Receiøed Noøember Phthalic acid and terephthalic acid have been reacted [C6H4-1,2-(COOH)2, H2PA] [C6H4-1,4-(COOH)2, H2TPA] with aqueous solutions of the hydroxides [(g5- and [(g6- produced in situ C5H5)2Co]`[OH]~ C6H6)2Cr]`[OH]~ by oxidation of the parent neutral molecules.The acid»base reaction leads to self-assembly of the deprotonated acid anions into honeycomb superstructures held together by hydrogen-bonding interactions of the OwH… … …O and charged OwH… … …O~ types. The superanions accommodate the [(g5- and the paramagnetic [(g6- C5H5)2Co]` organometallic cations via charge-assisted CwHd`… … …Od~ hydrogen bonds.Four novel C6H6)2Cr]` organic»organometallic cocrystals, namely M[(g5- (1), [(g6- C5H5)2Co]`N4M[HPA]~N2[PA]2~ … 4H2O C6H6)2Cr]`- (2), M[(g5- (3) and M[(g6- (4) have [HPA]~[H2PA] C5H5)2Co]`N2[TPA]2~ … 6H2O C6H6)2Cr]`N2[TPA]2~ … 6H2O been isolated and structurally characterized by low-temperature X-ray diÜraction measurements. It is shown that phthalic acid is a very versatile building block in the formation of hydrogen-bonded networks and unprecedented superanionic architectures. The role played by water molecules in the stabilization of the crystal structures in the absence of all or almost all acidic protons is discussed. Organometallic crystal engineering is an attractive –eld of research ;1 early attempts to make crystals based on organometallic components have been mainly directed towards obtaining charge transfer and magnetic systems.2 Much of the current interest stems from the potential inherent to the utilization of crystal construction strategies developed in the neighboring –eld of organic crystal engineering3 to assemble organometallic molecules or ions in a predesigned way.The ultimate goal is that of preparing novel materials in which the characteristics of transition metal coordination chemistry (e.g., variable valence, oxidation and spin states of the metal atoms) are brought to the crystals.This objective can be achieved, for example, by controlling the non-covalent interactions established between the ligands, which are most commonly organic in nature, or by using the organometallic moieties to template aggregation of organic components.The intelligent utilization of non-covalent interactions to obtain aggregates that function diÜerently from the separate components is the paradigm of supramolecular chemistry.4 Hence, crystal engineering, and organometallic crystal engineering of course, may be regarded as being at the crossing point of supramolecular and materials chemistry.A distinction should be made, however, between the engineering of molecular crystals and that of coordination networks or of covalent solids.5 The distinction is, in essence, in the energetic scale of the interactions : while molecular crystal engineering implies the utilization of intermolecular bonds with energies of the order of a few tens of kJ mol~1, coordination polymers and covalent solids are constructed on the energetic scale of dative ligandto- metal coordinative bonds and covalent bonds, that is of the order of several hundreds of kJ mol~1.The principal non-covalent interaction in molecular crystal engineering is the hydrogen bond.6 The strength and direc- § Part 5 of the series ìOrganic»organometallic crystal synthesisœ.For part 4 see ref. 7c. tionality of this three-center four-electron interaction can be tuned by varying the nature of the acceptors and donors and/or the polarity of the groups involved. The classical OwH… … …O hydrogen bonds formed by wCOOH and wOH groups are among the strongest neutral bonds. It is well known, however, that the OwH… … …O bond can be further strengthened if the polarity of the acceptor systems is increased via deprotonation.Negatively charged OwH… … …O~ bonds have been shown to possess dissociation energies in the range 60»120 kJ mol~1.6g,h These interactions are sufficiently strong to control recognition and self-assembly of carboxylic acid/carboxylate anions in robust three-dimensional superstructures. Furthermore, the utilization of polycarboxylic acids permits the simultaneous use of neutral OwH… … …O and negatively charged OwH… … …O~ bonding interactions in the construction of complex organic superstructures as shown in the case of D-, L- and L-tartaric acid,7a and of trimesic acid.7b The same strategy has been successfully applied also to the synthesis of mixed-metal and mixed-valence organometallic crystals from organometallic acids7c and to the preparation of charge-transfer salts based on the utilization of squaric acid.7d A similar approach has been adopted in the organic crystal engineering –eld to prepare ionic materials by using L-malic acid and substituted benzylamines8a,b and host»guest clathrates based on guanidinium and organosulfonate ions.8c,d In this paper we report the synthesis and structural characterization of four novel organic»organometallic crystalline materials, namely M[(g5-C5H5)2Co]`N4M[HPA]~N2- (1), [(g6- (2), [PA]2~ … 4H2O C6H6)2Cr]`[HPA]~[H2PA] M[(g5- (3) and M[(g6- C5H5)2Co]`N2[TPA]2~ … 6H2O (4). Crystalline 1»4 are C6H6)2Cr]`N2[TPA]2~ … 6H2O obtained by reacting the organometallic hydroxides [(g6- and [(g5- produced by in C6H6)Cr][OH] C5H5)2Co][OH],7b situ oxidation of the neutral complexes [(g6- and C6H6)Cr] [(g5- in water or THF, with ortho-phthalic acid C5H5)2Co] (1 and 2) or terephthalic acid (3 and 4).The (H2PA) (H2TPA) New J. Chem., 1999, 17»24 17organometallic cations produced by the redox process are very stable so that the arene and cyclopentadienyl ligands cannot be displaced and metal coordination by the carboxylate groups is not possible (but see Experimental).Details of the redox processes and of the eÜect of the solvent choice have been discussed in previous papers of this series.7 It is useful to stress that this seemingly simple reaction has been devised to take advantage of the complementary roles of weak and strong hydrogen bonds. The abundance, in organometallic species, of cyclopentadienyl and arene ligands carrying a large number of polarized CwH systems makes CwH… … …O hydrogen bonding very important in organometallic crystals.9 As in the case of OwH… … …O bonds, CwH… … …O bonds can be reinforced by ììaddingœœ polarization to the system, which can be easily achieved if the CwH donors belong to cations and the available acceptor sites belong to anions.10 In the cases discussed here CwH… … …O bonds are formed between the organometallic cations and the O atoms belonging to wCOOH, wCOO~ and to water molecules.Such weak, but charge-assisted and numerous, hydrogen bonds provide a web of interactions that are used to mold the anionic organic frameworks around the organometallic cations.11 One of the rules of thumb of hydrogen-bond-based crystal engineering is that strong acceptor sites preferentially accept hydrogen bonds from strong donors, leaving all unused lone pairs to the weak donors as a second choice. In this respect, one should note that water molecules may be brought into the crystal if the reaction, or the subsequent crystallization, is carried out in water.When complete deprotonation occurs, water molecules may ììcompensate the loss œœ of carboxylic hydrogens, providing alternative strong donors to guarantee crystal cohesion.7b In this context, the surprising topological analogy between the crystal structures of 4 and that of the hydrated hydroxide [(g5-C5H5)2Co]` previously determined, will also be discussed.[OH]~… 4H2O7b Experimental Crystal synthesis As in the cases discussed in Part 4 and preceding papers,7 the synthetic aspect of this work is related to the synthesis and crystallization of solid materials.It should be stressed that all usual spectroscopical tools for the characterization of chemical products in solution cannot be used in the context of a crystal synthesis. The products of the synthesis exist only in the condensed phase for which diÜraction techniques are essential. Cobaltocene, and phthalic and terephthalic acid were purchased from Aldrich, bis(benzene)chromium from Strem.Preparation of the hydroxides and [(C5H5)2Co]ë[OH]ó The brown powder of or [(C6H6)2Cr]ë[OH]ó. (C5H5)2Co (100 mg, 0.53 or 0.48 mmol) was suspended in 20 (C6H6)2Cr ml of bidistilled water under stirring at room temperature.Oxygen was bubbled until a clear solution of bright yellow or (pH of the [(C5H5)2Co]`[OH]~ [(C6H6)2Cr]`[OH]~ solution [10) was obtained. Synthesis of crystalline { [(g5-C5H5)2Co]ë} 4 { [HPA]ó} 2- (1), (2), [PA]2ó Æ 4H2O [ (g6-C6H6)2Cr]ë[HPA]ó[H2PA] (3) and { [(g5-C5H5)2Co]ë} 2 [TPA]2ó Æ 6H2O { [(g6- (4). Acid-to-hydroxide stoi- C6H6)2Cr]ë} 2 [TPA]2ó Æ 6H2O chiometric ratios of 1 : 1 and 2 : 1 were used for the reaction of and with cobaltocenium and bis(benzene) chro- H2PA H2TPA mium hydroxides.The white powder of phthalic and terephthalic acid (21.6 mg, 13 mmol and 43.2 mg, 26 mmol for the stoichiometric ratios 1 : 1 and 1 : 2, respectively) was added to 5 ml (0.13 and 0.12 mmol for the cobalt and the chromium hydroxides, respectively) of the hydroxide solution.Powders of 1»4, were obtained by evaporation of the –ltered aqueous solution. The powders were then dissolved in and CH3NO2 crystals suitable for X-ray diÜraction were obtained by slow evaporation at room temperature on a watch glass. Yields were almost quantitative. When bis(benzene)chromium hydroxide was used the formation of undesired green material, probably resulting from overoxidation of chromium, was Table 1 Crystal data and details of measurements for 1»4a 1 2 3 4 Formula C64H62Co4O16 C28H23CrO8 C28H36Co2O10 C32H40Cr2O10 Mol wt 1261.72 539.46 650.43 688.64 T /K 213(2) 223(2) 223(2) 223(2) Crystal system Orthorhombic Triclinic Triclinic Monoclinic Space group Pbcn P1 6 P1 6 C2/c a/Aé 28.534(8) 7.952(2) 8.937(6) 25.066(4) b/Aé 15.244(8) 11.592(3) 14.084(8) 9.615(3) c/Aé 12.982(8) 13.970(10) 23.760(10) 13.945(8) a/° 90 67.77(4) 99.27(4) 90 b/° 90 88.13(4) 100.71(5) 112.69(3) c/° 90 84.30(2) 90.21(5) 90 U/Aé 3 5647(5) 1186(1) 2898(3) 3101(2) Z 4 2 4 4 F(000) 2728 558 1352 1440 Min. and max.transmission 0.76»1.00 0.85»1.00 0.80»1.00 0.84»1.00 k(MoKa) mm~1 1.228 0.536 1.198 0.758 Measd re—ect 7178 5195 10089 4106 Unique re—ect 4939 4834 9822 3773 Unique re—ect [I[2p(I)] 2698 2483 7019 1623 Re–ned parameters 313 407 667 186 GOF on F2 0.988 0.957 0.864 0.964 R1 [on F, I[2p(I)] 0.0844 0.0404 0.0557 0.0513 wR2 (on F2, all data) 0.2938 0.1340 0.1755 0.1818 a Common to all compounds: MoKa radiation, j\0.71069 graphite monochromator.Aé , 18 New J. Chem., 1999, 17»24sometimes observed.In these cases, the preparation was repeated. The correspondence between the structure of the bulk material and that of the crystalline materials subjected to single crystal diÜraction experiments (see below) has been checked by measuring powder spectra and by comparing the spectra with those calculated on the basis of the single crystal structures. Crystal structure characterization All X-ray diÜraction data collections were carried out on a Nonius CAD-4 diÜractometer equipped with an Oxford Cryostream liquid device.Crystal data and details of measure- N2 ments are reported in Table 1. DiÜraction data were corrected for absorption by azimuthal scanning of high-s re—ections. SHELX8612a and SHELXL9212b were used for structure solution and re–nement based on F2.Fractional atomic coordinates and anisotropic displacement parameters are available as Supporting Information. SCHAKAL9712c was used for the graphical representation of the results. All non-H atoms, except for the disordered and the atoms in 1, were CCp Owater re–ned anisotropically. The positions of the hydrogen HCOOH atoms in 1 and 2 and of 13 water hydrogens in 3 have been observed in the Fourier maps.The remaining H atoms bound to C atoms were added in calculated positions in all compounds. The computer program PLATON12d was used to analyze the geometry of the hydrogen-bonding patterns. In order to evaluate CwH… … …O bonds the CwH distances were normalized to the neutron-derived value of 1.08 DiÜraction Aé . data have all been measured at 223 K.The Cp ligands bound to Co(2) in 1 and one of the Cp rings bound to Co(3) in 3 have been found to be disordered over two sites with occupancy factors of 0.70 : 0.30, 0.50 : 0.50 and 0.50 : 0.50 for the three rings, respectively. One of the water molecules [O(9)] in 1 is also aÜected by disorder, and has been re–ned over two distinct positions with occupancy factors of 0.60 and 0.40.CCDC reference number 440/080. See http ://www.rsc.org/suppdata/ njc/1999/17/for crystallographic data in .cif format. Results and discussion Since the focus of this paper is on the supramolecular features of the four crystalline materials, details of the structures of the ions will not be described. Table 2 collects some important average structural parameters, whereas Table 3 collects all relevant hydrogen bonding parameters.OwH… … …O and OwH… … …O~ hydrogen-bonding distances are comparable to those formed by other polycarboxylic acidates discussed previously, 7 with the notable exception of the intramolecular O… … …O separation in one of the phthalates in crystalline 2, which is aÜected by disorder (see below). The eÜect of charge is notable on all interactions. Waterwwater hydrogen bonds are, in general, longer than waterwcarboxylate bonds.The CwHd`… … …Od~ hydrogen bond interactions between the cations and the organic framework and between the cations and the water molecules follow the same general trend : the former are shorter, on average, than the latter distances. CwH… … …O distances are also shorter when the completely deprotonated dianions in 1, 3, and 4, are involved.The crystalline edi–ce of { [(g5-C5H5)2Co]ë} 4 { [HPA]ó} 2 (1) [PA]2ó Æ 4H2O The species is obtained when cobaltocenium hydroxide and are reacted in a 1 : 1 stoichiometric ratio. The crystal H2PA architecture is remarkable: the hydrated organic superstructure recalls a brick wall, with large rectangular channels extending along the c axis [see Fig. 1(a)]. The channels are occupied by pairs of columns of [(g5- cations C5H5)2Co]` [see Fig. 1(b)]. The ììsmallœœ and ììlargeœœ walls of the rectangular channels have diÜerent chemical compositions. The bc Fig. 1 (a) Space-–lling representation of the brick-wall type anionic organic superstructure present in crystalline 1. (b) The cations are arranged in pairs of columns along the c axis.H atoms are omitted for clarity. Line-shaded spheres are the oxygen atoms. Table 2 Relevant average intramolecular bonding parameters (in for 1»3 ”) 1 2 3 4 Mean MwC (M\Cr, Co) 2.03320 2.1349 2.02819 2.1317 Mean CCpwCCp/CBzwCBz 1.42b 1.3967 1.42b 1.39b Mean CringwCring 1.391 1.381 1.39b 1.39b Mean CringwCCOO 1.4994 1.501 1.5143 1.526(6) Mean CwO 1.2916 1.292 » » Mean CwO 1.2056 1.211 » » Mean Cw~ ~ ~O (dianion)c 1.241 » 1.25010 1.24711 a e.s.d.s on the mean values are given as subscripts.b Cp, Bz or phthalate rings re–ned as rigid groups. c CwO and CxO groups are not distinguishable on the basis of bond distances. New J. Chem., 1999, 17»24 19Table 3 Relevant intermolecular hydrogen-bonding parameters in crystalline 1»4 (distances in angles in degrees) ”, Interaction type 1 2 3 4 OCOOH… … …OCOO~ » 2.605, 2.652 » » OCOOH~… … …OCOO~ 2.493 2.356 (intra)a » » OCOO~… … …OW 2.753, 2.591 » 2.774, 2.763 2.741, 2.744 2.828, 2.831 2.738 2.778, 2.743 2.745, 2.770 2.728, 2.736 2.830, 2.817 2.767, 2.772 OW… … …OW 2.876, 2.762 » 2.839, 2.788 2.767, 2.739 2.773, 2.737 2.764 2.770, 2.745 2.792 (C)Hd`… … …O\2.6 » 2.354, 2.207 » » CwHd`… … …O » 172.54, 158.28 » » (C)Hd`… … …Od~\2.6 dianion monoanion dianion dianion 2.308, 2.099 2.351, 2.446 2.579, 2.563 2.384, 2.442 2.141, 2.451 2.486, 2.573 2.225, 2.229 monoanion 2.498, 2.504 2.365, 2.381 2.325, 2.229 2.362, 2.383 2.437, 2.129 2.559 CwHd`… … …Od~ dianion dianion dianion dianion 163.89, 173.30 165.83, 132.26 137.80, 138.87 156.31, 133.17 155.96, 177.79 124.05, 119.65 165.14, 164.56 monoanion 125.65, 124.24 139.86, 139.97 158.22, 171.80 139.90, 139.4 111.79, 145.49 113.39 (C)Hd`… … …OW\2.6 2.576, 2.589 » 2.158, 2.173 2.402 2.163, 2.156 2.509, 2.437 2.453, 2.251 2,300, 2.293 2.275, 2.368 2.535 CwHd`… … …OW 133.62, 130.89 » 169.95, 167.46 164.24 169.08, 167.46 141.60, 136.20 168.30, 167.86 162.06, 161.61 167.94, 148.58 154.50 a See text for a discussion of the disorder aÜecting the atoms involved in this bond.planes are formed by fully deprotonated PA2~ dianions interlinked by OwH… … …O hydrogen bonds [see Fig. 2(a)] involving the water molecules as donors, whereas the ac planes are formed by mono-deprotonated HPA~ anions linked by negatively charged OwH… … …O~ hydrogen-bonding interactions [see Fig. 2(b)]. Although the acceptor oxygen atoms belong exclusively to the anions, it is worth noting that the bonds are appreciably longer than the OwH… … …OW OwH… … …O~ ones but shorter than in ice (see Table 2).13 In summary, 1 contains two well distinct building blocks, resulting from complete and partial deprotonation of H2PA that play diÜerent structural roles in the construction of the crystal edi–ce.The two hydrogen bond arrangements in 1 represent two ììoptionsœœ for phthalic acid. A third one will be seen in the structure of 2. The crystalline edi–ce of [ (g6-C6H6)2Cr]ë[HPA]ó[H2PA] (2) When bis(benzene) chromium is employed, a diÜerent arrangement of the acceptor and donor groups derived from is H2PA observed. The crystal of 2 contains chains formed by an alternation of mono-deprotonated anions and of neutral H2PA acid molecules (see Fig. 3). Interestingly, the wOH group of the HPA~ anion is used in intramolecular hydrogen bonding, so that the two outer oxygen atoms of the monoanion can accept hydrogen bond donation on both sides by two molecules of neutral Neutral molecules and monoanions H2PA. form chevron-type patterns that ììembraceœœ pairs of bis(benzene) chromium cations as shown in Fig. 4(a). We have found an interesting structural analogy between this packing motif and the arrangement observed previously in the case of the derivatives obtained from cyclohexane-1,3-dione [(g6- [compare Fig. 4(a) and (b)]. C6H6)2Cr]`(CHD)2~… (CHD)21a As in the case of 1 there are several short charge-assisted CwHd`… … …Od~ interactions between the benzene ligands and the O atoms (see Table 3).The overall packing arrangement in 2 is shown in Fig. 5. There is a crystallographic detail that needs some attention. Judging from the data in Table 2, the intramolecular O… … …O distance of 2.356(2) in the mono-deprotonated anion HPA~ Aé may be regarded as very short, even shorter than in the case of lithium hydrogen phthalate (2.40 which has been subjected Aé ), to a neutron diÜraction study.14 Fig. 6 shows the orientation of the anisotropic displacement parameters (adps) for the monoanion. It is apparent that the O atoms of the two carboxylic groups possess an extensive additional motion, roughly perpendicular to the ring plane, which is indicative of some degree of orientational disorder. A reasonable model for 20 New J.Chem., 1999, 17»24Fig. 2 (a) The bc planes in crystalline 1 are formed by fully deprotonated PA2~ dianions interlinked by OwH… … …O hydrogen bonds involving the water molecules as donors. (b) The ac planes are formed by mono-deprotonated HPA~ anions linked by negatively charged OwH… … …O~ hydrogen-bonding interactions. the disorder implies that the two wCOO groups are displaced alternatively above and below the plane of the ring with the H atom (located from the Fourier map) almost in the center of the system.An intramolecular O… … …O separation of 2.4 in Aé , agreement with the neutron data, can be easily accommodated by oxygen atoms occupying two alternate sites along the adps. Unfortunately the positions of the O atoms of the same wCOO groups are too close to be treated separately in the re–nement so that the disorder results in the large anisotropic displacement shown in Fig. 6. In summary, there are as many as four diÜerent structural arrangements of phthalic acid or its deprotonation products. 1 contains one fully deprotonated PA2~ unit that accepts hydrogen-bond formation from water molecules, and one mono-deprotonated HPA~ unit that forms chains via OwH… … …O~ hydrogen-bonding interactions. 2 contains one neutral molecule acting as a bridge between mono- H2PA deprotonated HPA~ units.These monoanions, however, diÜer from those in 1 because of the presence of an intramolecular OwH… … …O hydrogen bond. In this way the HPA~ unit in 2 is more likely the fully deprotonated unit in 1 since it may accept twin donation of OwH… … …O bonds from neutral H2PA Fig. 3 Chains formed by an alternation of mono-deprotonated anions and of neutral molecules in crystalline 2. Note how the H2PA wOH group of the HPA~ anion is used in intramolecular hydrogen bonding, so that the two outer oxygen atoms of the monoanion can accept hydrogen bond donation on both sides by two molecules of neutral H atoms were observed from the Fourier maps (see H2PA.Experimental). Fig. 4 Comparison between the ììcation embraceœœ in (a) crystalline 2 and in (b) derivatives obtained from cyclohexane-1,3-dione [(g6- Note how, in both cases, a pair of C6H6)2Cr]`(CHD)2~ … (CHD)2. bis(benzene)chromium cations are embraced by the superanions. Available O atom lone pairs are directed towards the benzene ligands to optimize CwH… … …O interactions.as it does in 1 with respect to water. There is an additional diÜerence, though. The intramolecular bond makes possible an ìì all-—at œœ geometry for the monoanion, whereas the two wCOO~ groups in 1 repel each other and generate a twist geometry. The crystalline edi–ce of { [(g5-C5H5)2Co]ë} 2 [TPA]2ó (3) Æ 6H2O Though chemically similar, phthalic and terephthalic acid diÜer substantially in topology.The position of the two wCOOH groups on opposite sides of the benzene moiety, beside making impossible the intramolecular utilization of hydrogen bond donor and acceptor sites, renders the terephthalate building block longer than wider with respect to the phthalate unit. This diÜerence in shape is well-re—ected in the packing patterns in which the TPA2~ unit can take part.As in the cases of 1 and 2 discussed above, when terephthalic acid is used with the cobaltocenium and bis(benzene)- chromium hydroxides, two diÜerent crystalline products are obtained. Crystalline 3 is constituted of TPA2~ anions, resulting from complete deprotonation of the acid, joined by hydro- H2TPA gen bonds with the six water molecules of the unit formula.The water-bridged TPA2~ system forms a fascinating anion superstructure [see Fig. 7(a)] that can be described as constituted of large, almost rectangular, channels delimited on the left and right sides (with respect to the orientation in Fig. 7) by water molecules only and above and below (—oor and ceiling) by the terephthalate anions.The large-channeled New J. Chem., 1999, 17»24 21Fig. 5 Space-–lling representation of the packing motif in crystalline 2. Projections in the (a) ab and (b) bc planes. structure accommodates the M[(g5- cations [see C5H5)2Co]` Fig. 7(b)]. A view of the TPA2~ layers is shown in Fig. 7(c). It can be appreciated how the TPA2~ anions are linked by two diÜerent bridging systems formed by water molecules.The crystalline edi–ce of { [(g6-C6H6)2Cr]ë} 2 [TPA]2ó (4) Æ 6H2O Crystalline 4 bears some resemblance with the structure of 3. As in the case of 3 the acid is completely deprotonated and the hydrogen bond scaÜolding is provided by water molecules joining the TPA2~ anions. Moreover, the channels have a Fig. 6 ORTEP representation of the anisotropic displacement parameters of the monoanion in crystalline 2.Note how the O(6) and O(7) atoms show extensive displacement perpendicular to the ring plane. Fig. 7 The anionic superstructure in crystalline 3. (a) The channels are delimited on both sides by water molecules only and above and below by the terephthalate anions. (b) The large channeled structure accommodates the [(g5- cations.H atoms are omitted for C5H5)2Co]` clarity. Line-shaded spheres are the oxygen atoms. (c) A view of the TPA2~ layers showing how the anions are linked via water bridges. shape similar to those in 3 with the organic anions forming the —oor and the ceiling [see Fig. 8(a)]. The [(g6-C6H6)2Cr]` cations occupy two diÜerent sites, one along the channel and one in between the water bridges.One way to look at the relationship between the crystals 3 and 4 is that of ideally expanding with another row of TPA2~ anions the width of the channels. The conjunction between TPA2~ anions is provided by hexamers of water molecules [Fig. 8(b)], which form the steps of a kind of staircase. Quite surprisingly, the packing organization of crystalline 4 very strongly resembles that observed in the crystal structure of a chemically very diÜerent system, namely the hydrated hydroxide [(g5- As a matter of C5H5)2Co]`[OH]~… 4H2O.7b fact, the two crystals are almost isomorphous [compare the unit cell of the hydroxide: a\22.28(2), b\8.805(2), c\13.81(1) b\101.55(9)°, U\2654 space group Aé , Aé 3, C2/c, with the data for 4 in Table 1].7b In crystalline [(g5- the OH~ groups are obviously C5H5)2Co]`[OH]~… 4H2O indistinguishable from the water molecules.The phthalate anions in 4 take the place of the oxygen hexagonal rings in the hydroxide species [compare Fig. 8(a) with Fig. 9]. It is reasonable to attribute this very peculiar case of pseudopolymorphism to the analogy in shape between the —at elongated TPA2~ anions with its four protruding O atoms and a hydrogen-bonded water hexamer.In other words, the two hydrogen bond networks, though chemically diÜerent, are, topologically, extremely similar. While in 4, as in all other crystals, the interaction between hydrated organic networks and the cations is provided by a profusion of weak, chargeassisted, CwH… … …O interactions, in the crystalline hydroxide these are replaced by interactions with the CwH… … …Owater water/hydroxyl network.This analogy, together with those discussed previously for 1 and 3, and for 3 and 4, provides evidence of the interplay, at the stage of crystal formation and growth of these peculiar salts, between the need to optimize hydrogen bonding, which has precise geometrical requirements, and the templating eÜect of the cylindrical cations. 22 New J.Chem., 1999, 17»24Fig. 8 (a) Space-–lling representation of the terephthalate/water network in crystalline 4. Line-shaded spheres are the oxygenatoms. The hydrogen-bonded superstructure hosts the bis(benzene)chromium cations. (b) The conjunction between TPA2~ anions is provided by hexamers of water molecules, which form the steps of a kind of staircase.Conclusions With this paper we have shown that phthalic acid is a very versatile building block for the construction of hydrogenbonded anionic organic frameworks. Contrary to previously used tartaric acid, which is —exible,7a phthalic acid adds some degree of rigidity to the crystal construction process. Whether neutral, mono- or diprotonated the phthalic acid structure is substantially —at and, thanks to its anchoring points, can be used to pave channeled structures.While phthalic acid tends to give mono-deprotonation products, both compounds obtained from terephthalic acid are the result of complete deprotonation. This behavior matches the pKs of the two acids (H2PA: pK1\2.3, pK2\5.5 : H2TPA: pK1\3.9, pK2\ Terephthalic acid is less acidic than with respect to 4.3).H2PA –rst deprotonation, but the loss of its second proton is easier. It is thus not surprising that, when reacting with organometallic bases, behaves as a strong diprotic H2TPA acid. Complete deprotonation is not inconsequential for our crystal engineering strategy. The strongly basic TPA2~ dianions bring into the crystal a large number of water mol- Fig. 9 The OH~»water network in the hydrated hydroxide [(g5- (compare with Fig. 8(a)]. C5H5)2Co]` [OH]~… 4H2O ecules (6 water molecules per cation unit in 3 and 4). These water molecules have a twofold role. They provide the ììmissingœœ hydrogens for stabilization via hydrogen bonding and act as ììspace –llers œœ between the anions, thus reducing inter-anion electrostatic repulsions, which are certainly more relevant in the case of the dianions in 3 and 4.A similar relationship between full deprotonation and high hydration had been observed previously in the crystalline material M[(g5- (L-BTA\L-bisbenzoyltar- C5H5)2Co]`N2[L-BTA]2~ … 11H2O taric acid)7a as well as in the case of the mixed-metal crystalline system M[(g5-C5H5)2Co]`N2[(g5-C5H4COO)2Fe]2~ The formation of highly hydrated species is an 7.75H2O.7d intriguing aspect of crystal engineering, which may have interesting consequences in the development of new design strategies.We have observed previously that the polar crystal of the hydroxide [(g6- possesses a C6H6)2Cr]`[OH]~… 3H2O layered architecture in which the layers have [OH]~… 3H2O the structure of hexagonal ice.16 More recently, the role of a water decamer which possesses an ice-like molecular (H2O)10 , arrangement within a large supramolecular complex in the solid state, has been discussed.16 We are learning that water itself can be used, in a somewhat predictable way (i.e., within the design strategy), to generate intermolecular scaÜolding and to link strongly nucleophilic building blocks.This also points to the role of ionic charges.The eÜect of the Coulombic –eld generated by the ionic charges on the strength of the various types of hydrogen-bonding interactions is difficult to evaluate. It is worth mentioning, in this respect, that theoretical ab initio calculations have demonstrated that the shortest CwH… … …O interaction in crystalline [(g6-C6H6)2Cr]` cannot hold together anions because of [CrO3(OCH3)]~ strong inter-anion Coulombic repulsions.17 In general, however, a positive charge carried by the cation decreases the shielding of the proton on the donor CwH groups and makes it more acidic, while the presence of a negative charge on the acceptor increases its nucleophilicity resulting in a strengthening of the weak bonds.This is the situation commonly observed when the organometallic cations interact with supramolecular anions as in crystalline 1»4.Work is in progress to isolate and characterize other crystalline aggregates and to investigate the hydrogen bonding by means of spectroscopic, diÜraction, and theoretical tools. Acknowledgements support by M.U.R.S.T. (Supramolecular Devices Financial project) and by the University of Bologna (Innovative Materials project) is acknowledged.References 1 (a) D. Braga and F. Grepioni, Chem. Commun., 1996, 571; (b) D. Braga, F. Grepioni and G. R. Desiraju, Chem. Rev., 1998, 98, 1375; (c) D. Braga and F. Grepioni, Coord. Chem. Rev., in press. 2 (a) J. S. Miller, A. J. Epstein and W. M. ReiÜ, Acc. Chem. Res., 1988, 21, 11; (b) P. J. Fagan, M. D.Ward and J. C. Calabrese, J. Am. Chem. Soc., 1989, 111, 1698; (c) M. D. Ward and M. D. Hollingsworth, Chem. Mater., 1994, 6, 1087; (d) J. S. Miller and A. J. Epstein, Chem. Commun., 1998, 1319; (e) For a general survey see also P. J. Fagan and M. D. Ward, in T he Crystal as a Supramolecular Entity. Perspectives in Supramolecular Chemistry, ed. G. R. Desiraju, Wiley, Chichester, 1996, vol. 2, p. 107. 3 (a) G. R. Desiraju, Angew. Chem., Int. Ed. Engl., 1995, 34, 2311; (b) C. B. Aakeroé y, Acta Crystallogr., Sect. B, 1997, 53, 569. 4 J. M. Lehn, Supramolecular Chemistry: Concepts and Perspectives ; VCH, Weinheim, 1995. 5 See for example: (a) B. F. Abrahams, B. F. Hoskins, D. M. Michail and R. Robson, Nature, 1994, 369, 727; (b) O. M. Yaghi, C. E. Davis, G. Li and H.Li, J. Am. Chem. Soc., 1997, 119, 2861; (c) R. E. Melendez, C. V. K. Sharma, M. J. Zawarotko, C. Bauer and New J. Chem., 1999, 17»24 23R. D. Rogers, Angew. Chem., Int. Ed. Engl., 1996, 35, 2231; (d) C. L. Bowes and G. A. Ozin, Adv. Mater., 1996, 8, 13; (e) G. A. Ozin, Acc. Chem. Res., 1997, 30, 17. 6 (a) G. A. JeÜrey, An Introduction to Hydrogen Bonding, Oxford University Press, New York, 1997; (b) L.Brammer, D. Zhao, F. T. Ladipo, J. Braddock-Wilking, Acta Crystallogr., Sect. B, 1995, 51, 632; (c) C. B. Aakeroé y and K. R. Seddon, Chem. Soc. Rev., 1993, 397; (d) D. Braga, F. Grepioni, E. Tedesco, K. Biradha and G. R. Desiraju, Organometallics, 1997, 16, 1846 and references therein ; (e) G. Aullon, D. Bellamy, L. Brammer, E. A. Bruton and A. G. Orpen, Chem.Commun., 1998, 653; ( f ) D. Braga, F. Grepioni and E. Tedesco, Organometallics, 1998, 17, 2669; (g) M. Meot-Ner (Mautner), J. Am. Chem. Soc., 1984, 106, 1257; (h) M. Meot-Ner (Mautner) and L. W. Sieck, J. Am. Chem. Soc., 1986, 108, 7525. 7 (a) D. Braga, A. Angeloni, F. Grepioni and E. Tagliavini, Organometallics, 1997, 16, 5478; (b) D. Braga, A. Angeloni, F. Grepioni and E. Tagliavini, J. Chem. Soc,. Dalton T rans., 1998, 1961; (c) D. Braga, L. Maini and F. Grepioni, Angew. Chem., Int. Ed. Engl., 1998, 37, 2240; (d) D. Braga and F. Grepioni, Chem. Commun., 1998, 911. 8 (a) C. B. Aakeroé y and M. Nieuwenhuyzen, J. Am. Chem. Soc., 1994, 116, 10983; (b) C. B. Aakeroé y and M. Nieuwenhuyzen, J. Mol. Struct., 1996, 374, 223; (c) V. A. Russell, C. C. Evans, W. Li and M. D. Ward, Science, 1997, 276, 575; (d) J. A. Swift, V. A. Russell and M. D. Ward, Adv. Mater., 1997, 9, 1183. 9 D. Braga and F. Grepioni, Acc. Chem. Res., 1997, 30, 81. 10 (a) F. Grepioni, G. Cojazzi, S. M. Draper, N. Scully and D. Braga, Organometallics, 1998, 17, 296; (b) D. Braga and F. Grepioni, in Current Challenges on L arge Supramolecular Assemblies, ed. G. Tsoucaris, Kluwer Academic Publishers, Dordrecht, 1998. 11 D. Braga, A. Angeloni, F. Grepioni and E. Tagliavini, Chem. Commun., 1997, 1447. 12 (a) G. M. Sheldrick, Acta Crystallogr., Sect. A, 1990, 46, 467; (b) G. M. Sheldrick, SHELXL92, Program for Crystal Satructure Determination, University of Goé ttingen, Goé ttingen, Germany, 1992; (c) E. Keller, SCHAKAL97, Graphical Representation of Molecular Models, University of Freiburg, Germany, 1997; (d) A. L. Spek, Acta Crystallogr., Sect. A, 1990, 46, C31. 13 W ater A Comprehensive T reatise, ed. F. Franks, Plenum Press, New York, 1973, vol. 2, p. 55. 14 H. Kué ppers, F. Takusagawa and T. F. Koetzle, J. Chem. Phys., 1985, 82, 5636. 15 D. Braga, A. L. Costa, F. Grepioni, L. Scaccianoce and E. Tagliavini, Organometallics, 1996, 15, 1084. 16 L. J. Barbour, G. W. Orr and J. L. Atwood, Nature, 1998, 393, 671. 17 D. Braga, F. Grepioni, E. Tagliavini, J. J. Novoa and F. Mota, New J. Chem., 1998, 22, 755. Paper 8/06501B 24 New J. Chem., 1999, 17»24

ISSN:1144-0546    

DOI:10.1039/a806501b

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Systematic analysis of the probabilities of formation of bimolecular hydrogen-bonded ring motifs in organic crystal structures Frank H. Allen,*a W. D. Samuel Motherwell,a Paul R. Raithby,b Gregory P. Shieldsa,b and Robin Taylora a Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge, UK CB2 1EZ. E-mail : allen=ccdc.cam.ac.uk b Department of Chemistry, University of Cambridge, L ens–eld Road, Cambridge, UK CB2 1EW (in Cambridge, UK) 15th September 1998, Accepted 5th 1998 Receiøed Noøember A methodology has been developed for characterising hydrogen-bonded ring motifs formed between two organic molecules without any prior knowledge of the topology or chemical constitution of the motifs. The method has been implemented by modifying the current Cambridge Structural Database (CSD) System programs.All intermolecular ring motifs comprising O20 atoms formed with N»H… … …N, N»H… … …O, O»H… … …N and O»H… … …O hydrogen bonds in organic structures in the CSD have been classi–ed. The 75 bimolecular motifs occurring in [12 structures in the CSD are described in terms of their graph sets and chemical functionalities. Motifs are ranked according to their frequency of occurrence and according to their probabilities of formation, i.e. their frequency relative to the number of possible motifs which could have formed.These probabilities provide insights into the relative robustness of known and potential supramolecular synthons. Introduction It is well known that the stronger hydrogen-bond motifs found in organic systems1 can be used to direct the synthesis of supramolecular complexes, e.g.in crystal engineering.2h6 The conceptual relationship between crystal engineering and organic synthesis has led to the term supramolecular synthon5 being proposed for structure-directing motifs involving non-covalent bonds. Many of the synthons identi–ed so far involve N»H… … …N, N»H… … …O, O»H… … …N and O»H… … …O Hbonds, 5h7 which confer a degree of robustness, hence reproducibility, in supramolecular retrosynthesis, while knowledge of these common patterns is also important in other applications, e.g.the modelling of protein»ligand interactions, and in ab initio crystal structure prediction. A number of early studies explored the nature of H-bonded motifs in classes of compounds with particular functional groups, e.g.carbohydrates, 8 carboxylic acids9 and amides.10 Similar motifs had been recognised previously in inorganic systems.11 Other studies have examined the eÜect of H-bond cooperativity and resonance-assistance on the robustness (strengths) of Hbonded systems.12h14 In order to describe the topology of H-bonded motifs and networks systematically, a graph-set approach has been suggested. 15,16 This provides a description of H-bonding patterns in terms of chains (C), rings (R), –nite complexes (D) and intramolecular H-bonds (S). The degree of the pattern (n, the number of atoms comprising the pattern), together with the number of donors (d) and the number of acceptors (a), are combined to form the graph-set designator This Xd a (n).16 description does not distinguish between patterns of the same degree that have diÜerent numbers of bonds in the constituent fragments or diÜerent arrangements of the donors and acceptors.However, these purely topological descriptors have proved useful in decoding diÜerences between the packings adopted in polymorphic systems, e.g. in L-glutamic acid17 and iminodiacetic acid.18 Patterns are distinguished on the basis of their level, the –rst level describing patterns involving crystallographically equivalent hydrogen bonds (if any), the second level involving two such H-bonds, and similarly for higher levels.Chemically equivalent patterns may become apparent at diÜerent levels depending on the presence or otherwise of crystallographic symmetry. The graph set nomenclature provides a basic description of H-bonded synthons and can aid the identi–cation of preferred motifs.Recently a systematic general search for motifs has been performed, to explore R2 2 (8) the chemical diversity of the functional groups which adopt this topology.19 The aim of our present work is to perform a computerised analysis of the non-covalent motifs that occur in the [160 000 crystal structures in the Cambridge Structural Database (CSD, October 1996).20 This is a data-driven approach which is more general than any previous analysis, and is designed to establish the topologies, chemical constitutions and numbers of occurrence of non-covalent motifs in an objective (Nobs) manner.While values are interesting, their interpretation Nobs is complicated by the fact that they depend on the number of times that the various donor and acceptor groups occur in the CSD.Thus, the carboxylic acid cyclic dimer motif may have a high simply because carboxylic acids are common in the Nobs CSD. To correct for this eÜect, we determine the probability of occurrence, through which for each motif is related to Nobs the number of times the motif could possibly occur.Nposs , Since we can assume that the probability of formation of a motif across many crystallographic instances is related to its robustness5 (i.e. how reliably a motif may be exchanged from one network structure to another), then the analysis has the potential to reveal motifs that may act as novel supramolecular synthons in crystal engineering applications.Because of the broad scope of the overall analysis, we have subdivided the work. Thus, our initial investigations have concentrated on the identi–cation of intermolecular cyclic motifs formed between pairs of molecules by medium to strong H-bonds involving N»H and O»H donors and N or O acceptors. New J. Chem., 1999, 25»34 25These cyclic patterns encompass many of the supramolecular synthons already identi–ed.5 In performing this study, we have also extended the computer methodology so as to generate graph-set descriptors for the motifs which were located.A preliminary account of this work, summarising the 24 most signi–cant motifs, has appeared elsewhere.21 As with other CSD investigations, there is the possibility of bias in data selection, i.e.the CSD is not a random sample of crystal structures and may not provide a representative picture of the typical environments of the fragments comprising the ring motifs. In particular, there is no guarantee that the degree of competition between diÜerent motifs will be the same for all motifs studied. For certain motifs used extensively in crystal design, a particular eÜort may have been made to avoid competing motifs.Alternatively, attempts may have been made to investigate structures where the expected motif did not occur, e.g. the carboxylic acid dimer. However, these considerations are likely to apply to a relatively small number of structures, although certain motifs may be particularly aÜected. For fragments occurring in a large number of diverse molecules represented in the CSD, as in this study, such bias should not be a signi–cant issue. Methodology Probabilities of motif formation To assess the probability of formation of a particular motif across the complete CSD, it is necessary to count the (Pm) total number of motifs that actually occur, and then to (Nobs) compute the total number of motifs that could have occurred in all structures recorded in the CSD (computation of (Nposs) is discussed below).Thus: Nposs Pm\Nobs/Nposs (1) Alternatively, we can compute a structural probability If (Ps). is the number of structures that contain a particular Sobs motif, and is the number of structures that contain the Sposs component functional groups (the donor and acceptor groups that comprise the motif), then: Ps\Sobs/Sposs (2) The quantity is easier to compute than hence Sposs Nposs , Ps provides a useful check on the computation of since Pm , Pm should be equal to if is independent of the number of Ps Pm motifs which could form in a structure.It should be Nposs remembered that both and are based solely on the Ps Pm sample of structures represented in the CSD. is de–ned as the percentage of chemically symmetric Psymm motifs which are also crystallographically symmetric.The principal steps in deriving the probabilities were to : (i) Identify potential donors and acceptors in each structure. (ii) Search for intermolecular H-bonds between the donors and acceptors. (iii) Find the shortest intramolecular (covalent) paths between donors and acceptors and hence locate intermolecular ring motifs.(iv) Count ring motifs and classify them in terms of their sizes, topologies and chemical constitutions. (v) Identify the chemical constitutions of the donor and acceptor fragments comprising the most frequently occurring motifs. (vi) Perform substructure searches for the component donor and acceptor fragments. (vii) Combine results for donor/acceptor fragments and derive the number of possible occurrences of the motifs.(viii) Calculate probabilities based on actual and possible frequencies of occurrence. Identi–cation of potential donors and acceptors The ability to locate intermolecular fragments of unde–ned topology (i.e. with variable intramolecular bond path lengths between H-bond donor and acceptor atoms) is not available in the released CSD System.Hence, the search program QUEST 3D was modi–ed in order to provide this additional functionality. Potential H-bond donors and acceptors were identi–ed on the basis of atom type and the present analysis was restricted to N»H and O»H donors and N and O acceptors. Deuterium was treated as being equivalent to H for the purpose of motif recognition.The October 1996 release of the CSD (ca. 160 000 structure determinations) was used throughout.21 The search was restricted to structures which were fully matched, error-free at the 0.02 level, had R-factor O10%, contained ” the elements C, H, D, N, O, S, P, B, Si and halogens only, and for which there was a perfect match between the chemical and crystallographic connectivity descriptions.Only structures for which H coordinates are stored were included in the analysis, avoiding any ambiguity as to the protonation state of N or O donor and acceptor atoms. Intermolecular contact search Each donor and acceptor combination was considered as a potential H-bond and contacts X»H… … …A»Y were found using a modi–ed version of the existing contact search routine, which is based on the algorithm described by Rollett.22 To permit meaningful comparison of the H… … …A distances it was necessary to normalise the X»H distances to ideal values derived from neutron diÜraction studies, placing H atoms along the existing X»H vector but at a standard distance of 1.009 and 0.983 for N»H and O»H bonds respectively.23 ” Additionally, the condition that the X»H… … …A angle must be greater than 90° was imposed.Whilst it is difficult to establish distance criteria for Hbonding, since the principal attractive (electrostatic) terms vary as r~1 and r~3, working distance limits are required for motif recognition based on crystallographic data. These were established using standard QUEST 3D non-bonded searches for intermolecular N»H… … …N, N»H… … …O, O»H… … …N and O»H… … …O contacts for which the angle X»H… … …A[90°.Symmetry-equivalent fragments were rejected and contacts up to the sum of the appropriate van der Waals radii24 were accepted. These searches yielded more than 10 000 fragments for N»H… … …O and O»H… … …O contacts and a limit of RO4% was imposed for these, to permit examination of geometrical distributions in the CSD program VISTA.Visual inspection of the minima in the frequency distributions (Fig. 1) gave limits of 2.30 (N»H… … …N), 2.25 (N»H… … …O) and 2.20 (O» ” H… … …N, O»H… … …O). H-bond distance distributions are dependent on the chemical environment of the donor and acceptor, not the element type alone. Thus, ether and carbonyl O atoms are not equally good acceptors, and the histograms in Fig. 1 represent the superposition of several chemically-independent distributions. However, since the composite distributions are unimodal, the generalised distance limits above were deemed appropriate for this study.The contact search was modi–ed where one or more of the starting molecules possessed crystallographic symmetry. In such cases, additional symmetry-generated atoms are included in the CSD, so that the crystallographic connectivity description represents complete molecules.These complete molecules are related to their symmetry-generated neighbours in the extended crystal structure by more than one symmetry operation. When generating symmetry-related molecules in the intermolecular contact search, the redundant operations were not applied to avoid generating duplicate contacts.For each X»H… … …A contact located, i.e. an H-bond from H in an original molecule to an acceptor A in a symmetry- 26 New J. Chem., 1999, 25»34Fig. 1 Histograms of (a) N»H… … …N, (b) O»H… … …N, (c) N»H… … …O and (d) O»H… … …O contact distances. related or crystallographically-independent neighbouring molecule, the symmetry operation S applied to the acceptor molecule was stored.H-bonds from acceptors in the starting molecule(s) to donors in the neighbouring molecules were derived by considering the inverse symmetry operation S~1 relating the acceptor molecule back to its original position for each X»H… … …A contact found. Having established the H-bonds present, the intramolecular bond paths between all pairs of donors or acceptors in each starting molecule were derived.No attempt was made to account for duplicate paths in the recognition process ; for small rings such paths are rare and, where they do occur, they are commonly chemically equivalent. Location of intermolecular rings Each pair of starting molecules was considered in terms of its hydrogen bond contacts.If two molecules were joined by a single H-bond, the motif was taken as being a –nite discrete motif and was neglected in the analysis. Where two molecules were joined by two or more H-bonds, each of the possible pairwise combinations of H-bonds constitutes a ring motif and was recorded separately. Bifurcated patterns were included, although they involve only one donor or acceptor in one of the molecules.No attempt was made to characterise explicitly motifs involving more than two H-bonds, by ring analysis or any other method, and only the individual components have been considered. Where some molecules possess internal symmetry, symmetry-equivalent rings were retained. This is equivalent to treating the structure as if it were in a sub-group for which the internal symmetry operator of the higher symmetry molecule was absent.If a ring is formed between a molecule which has internal symmetry and a molecule which does not, the rings were also counted more than once, since the lower symmetry molecules would be crystallographically independent in the sub-group. This facilitates a more straightforward comparison with the number of possible motifs, [eqn.(1)], in any Nposs structure than if symmetry-equivalent motifs were rejected. Generation of graph set descriptors16 At this stage, the topology of a cyclic motif is now fully identi- –ed in terms of its ring size R(n) and the numbers of acceptors (a) and donors (d) so that the appropriate descriptor, of the form can be assigned. For individual structures, the full Rd a (n), New J.Chem., 1999, 25»34 27set of R-type descriptors (describing all cyclic motifs located in the extended structure) can be displayed. For our purposes, however, it was more efficient to generate the R-descriptors for the most frequently occurring motifs after completing the chemical classi–cation and counting exercise described in the next section, i.e.they were generated from the chemical constitution keys used there. The generation of graph-set descriptors for individual structures has recently been generalised to encompass the C, D and S-type descriptors16 noted in the Introduction. This code has been incorporated in the CSD System structure visualiser, PLUTO, and displays graph sets up to the second level, subject to some restrictions for molecules which possess internal crystallographic symmetry.Full details of this implementation will be published shortly.25 Classi–cation and counting of motifs Ring motifs were characterised on the basis of atom types (element and coordination number) and bond types (CSD bond type and cyclicity). Two amendments were made to the CSD bond type conventions to overcome ambiguities in connectivity coding: a ìguanidiniumœ bond type was introduced for cationic carbon bound to 3 three-coordinate nitrogen atoms (in the conventional CSD description two C»N bonds are described as single and one as double although they are equivalent chemically). A ìterminal oxygenœ bond type for any mono-coordinate oxygen for which the bonds were described as single, double or delocalised was also introduced. This provides a consistent description of the bonding in carboxylate, phosphate and related anions.Similar ambiguities exist in the de–nitions of certain aromatic nitrogen moieties which may be coded with alternate single and double bonds; no attempt was made to take the alternative descriptions into account for such systems.For chemically symmetric motifs, a record was also made of whether or not the ring was formed about a crystallographic symmetry element. An upper limit of 20 atoms was placed on the ring size. For each structure, every motif identi–ed was compared with those already recorded in previous structures. A running total was kept of the number of occurrences of each discrete motif and the number of structures in which it was found.For every structure containing an H-bond motif, the motifs identi- –ed in that structure were logged in a –le in motif-number order. Rather than comparing the explicit atom and bond properties of a new motif with those for all pre-existing motifs, atom and bond type integer keys were derived for each motif by packing the number of atoms and bonds of particular types.To increase the efficiency of searching on these two integer keys, they were arranged as nodes in a binary tree with no duplicate keys. A full list was made of the atom and bond properties for each motif corresponding to a particular key and these were compared in full once a key had been matched. By convention, a motif was de–ned as starting with an H atom and an H-bond.This leaves an ambiguity as to which H atom should be chosen if more than one donor is involved, making it necessary to test both possibilities when comparing motifs, reversing the atom and bond order where appropriate. In motifs involving a bifurcated hydrogen, the donor atom X does not comprise part of the ring ; however, its identity is required if the chemical nature of the ring is to be de–ned and therefore it was included in the motif description.To minimise any bias in the sample, motifs were recorded only once if more than one determination of a structure occurred in the CSD, only the –rst determination which exhibited a motif being retained. After the search, all discrete motifs were sorted both chemically (in terms of atom and bond keys) and according to the frequency with which they occurred.For each ring, the total numbers of motifs which were (a) associated with a symmetry element only pos- (Nsym: sible if the motif is symmetric chemically) or (b) structurally asymmetric were also derived. (Nasym), Identi–cation and counting of constituent covalent fragments The total number of motifs [eqn. (1)] which could be Nposs formed is dependent on the number of covalent fragments comprising the motif which are present in a structure.The –rst step in establishing involves identifying the fragments Nposs comprising the motifs and performing standard QUEST 3D substructure searches for them. The motif recognition program generated an output –le which described each motif in terms of its constituent atom and bond types, and which served as input to a stand-alone fragment recognition program. The motifs were sorted in terms of the number of structures in which they occurred and only those found in [12 structures were considered for the fragment identi–cation process.Each motif was decomposed into its constituent covalent fragments, which were compared with those which had already been identi–ed. The same atom and bond keys used to classify the motifs were used to de–ne the constituent fragments and the fragments were described in full with the same atom and bond types.Where necessary, the fragments were reversed to achieve a match. This program produced a –le containing the motif identi–ers, motif statistics and the identi- –ers for the fragments comprising the motif.For each fragment, a QUEST 3D connectivity search query was written automatically, comprising the required atoms (with total coordination number de–ned) and bonds. The QUEST 3D symmetry-check facility was set to retain symmetry-equivalent fragments, for consistency with the approach adopted in motif recognition. The same secondary search criteria and associated element tests used for motif recognition were adopted here, in order to search an identical subset of the CSD. Manual editing of the substructural queries was only necessary for the ìguanidiniumœ bonds, there being no equivalent in the CSD.Bonds to mono-coordinate oxygen were de–ned in substructural queries with a variable bond type setting (single, double or delocalised).Each query comprised a QUEST 3D instruction –le, and the searches were run consecutively using a suitable script. For each query, a tables –le was produced containing the (CSD) atom labels for each fragment and the structure identi–er. Derivation of the number of possible motifs The values were derived, using a separate program, from Nposs the covalent fragments listed in these tables –les.Duplicate determinations of the same crystal structure were omitted. Where some molecules in a structure contained symmetrygenerated atoms, the structure was treated as if it were in a lower symmetry sub-group in which the operators generating the symmetry-atoms were absent, to be consistent with the method used to count the observed motifs. The main difficulty in deriving lies in determining Nposs which overlapping fragments could simultaneously form motifs.The lower limit is given by excluding all overlap and the upper limit by including all overlapping fragments. The problem is not merely topological, however, being complicated by the need to consider the three-dimensional geometric nature of the motif. Bifurcation, as well as other multi-centre H-bonds and H-bonds which are shared between two rings, must also be taken into account.A set of fragment combination rules (Fig. 2) was derived, which takes into account molecular symmetry, fragment overlap and the possibility of multi-centre H-bonding. These rules express which fragments cannot simultaneously form motifs, i.e. are mutually exclusive. For example, only one fragment is available from an amide (Y same for both Hœs), re—ecting the impossi- RC(xO)NH2 bility of both Hœs being cis to the carbonyl oxygen, and from a sulfonamide (X same for both A atoms), since RS(xO)2NHR 28 New J.Chem., 1999, 25»34C N C O O H N C O H N C O H H N C N H O C N C O O H (a) (b) Fig. 2 Combination rules for covalent fragments comprising motifs.the H cannot be cis to both oxygen acceptors. Similarly, there are only three HNCNH fragments in a cation, [C(NH2)3]` with H atoms mutually cis, which may simultaneously form motifs. The maximum number of chemically-equivalent fragments which could simultaneously form motifs was determined by considering all the fragment combinations permitted by these rules. No attempt was made to ensure that all the allowed fragments could be assembled into motifs ; in some instances this is not possible because of the relative geometric disposition of the functional groups.Consider amide fragments having complementary functionality of two donors plus one acceptor and one donor plus two acceptors respectively [Fig. 3(a)]. In this case, it is not possible to assemble them with three H-bonds to form two cis amide eight-membered rings : the individual H»N»CxO fragments may not all form cis amide dimers unless the amide H makes two H-bonds [Fig. 3(b)], which also may not be geometrically possible. Similarly, it was assumed that the molecule was sufficiently —exible to adopt a conformation which permits motif formation, although this might not be the case if the conformation was constrained by ring closure (e.g.hydroxyl substituents of a pyranose sugar). As a result, could be an over-estimate, Nposs albeit a more realistic value than would be obtained if all overlapping fragments were included. For motifs composed of two fragments which were chemically identical, is simply equal to the number of frag- Nposs ments which may simultaneously form motifs.A molecule containing two chemically identical fragments P and Q, may form either P»P@, Q»Q@ or P»Q@ and Q»P@ motifs (where a prime denotes a symmetry-related fragment in a neighbouring molecule). In both cases, two motifs would have been counted Fig. 3 Complementary amide fragments (a) and bifurcated amide fragments (b). in the motif search process, once for each fragment in the starting molecules, and the possible motifs are counted in the same manner.An overall count was made of the total number of possible motifs in all structures and of the number of Nposs structures that could possibly contain a motif. Sposs Motifs which are not symmetric in terms of their chemical constitution were treated similarly, and the respective covalent fragments were read from two tables –les concurrently. A structure was only considered further when it contained both chemically-distinct fragments.The number of covalent fragments of each chemical constitution which could form motifs simultaneously was derived. is determined by the smaller Nposs of these two values : the number of fragments of the other chemical constitution which could form motifs has no eÜect on Although the proportion by which one value Nposs .exceeds the other could aÜect the probability of forming the motif, this was not taken into account in the derivation of Pm . Results The 75 motifs having were ordered on their Sobs[12 Pmand are depicted in Fig. 4; the motif numbers indicate values the probability ranking of the motif. Relevant statistics for these motifs are given in Table 1, and graph sets16 are also given for the motifs, in order to identify the more important topologies.Motifs 1 and 2, which have the highest probabilities of formation, are components of the triply hydrogen-bonded pattern shown in Fig. 5, which has some analogies with nucleotide recognition in DNA26 and whose robustness in crystal engineering applications has been recognised by a number of authors.27 Here, the individual rings are not independent and the motif might better be considered as the ensemble of the two ring motifs.The chemically-asymmetric motif 2 occurs in more than 80% of possible structures R2 2 (8) and 90% of fragments. That is higher than suggests a Ps Pm degree of cooperativity, i.e. a second motif is more likely to form given the existence of one motif, although the diÜerence may not be statistically signi–cant.Whilst there are ca. 200 such fragments, these occur only in 29 structures. The R2 2 (12) motif 1 represents the envelope of these rings ; it occurs in every structure in which the respective fragments exist and Pm is ca. 97%. These structures include many for which the planned use of this synthon in crystal engineering has been successful. It is proposed that the 1 : 1 complex of cyanuric acid and melamine contains in–nite sheets connected by this motif28 and the complex with 3(HCl) contains dimers linked by such units.29 Its eÜectiveness can be attributed to the complementarity of the functional groups and the additional robustness engendered by three rather than two H-bonds.5 In terms of only 25 motifs have a probability greater Pm , than 30% (Table 1).The only motifs, in addition to 1 and 2, with are the carboxylic acid»oxime motif 3, Pm[75% carboxylate»HNCNH motif 4 and carboxylic acid»HNCN ring 5. None of these occur in more than 30 structures and all are chemically asymmetric. The HNC`NH fragment may be associated with carboxylate (8) and sulfate-type (7) functional groups and such motifs are formed in ca. 0.66 of the possible structures and for more than 50% of possible motifs. Motif 8 is commonly formed in biological systems between arginine and aspartic and glutamic acids.26 The unusual motif 9, incorporating two cyano-N… … …H-bonds, is formed with a probability of nearly 50% and is generally crystallographically symmetric (B90%).Motifs 6 and 10, components of crown ether host»guest motifs are two of the most probable. However, whilst 6 occurs in almost 85% of possible structures, is only 58%. This may be due to the manner in which the Pm fragments are counted: although a cation has six [NH4]` H»N»H combinations (allowing overlap) and the crown ether six (R)OCCOCCO(R) units, each ammonium ion may only New J.Chem., 1999, 25»34 29O O O N+ H H H N C+ N H O S O O O O O H H N H P O P O N H N O H O H O H N N H O O H N C+ N H O O O H O N N H N H N O N H C N C O O H N N N H 11 10 6 7 H 12 N H P O P O N H 3 5 2 4 8 1 N C N C H H N N 9 N -O O N+ H N -O O N H O O O H O H N+ O H H O C O N O N O H H N H N N N H O O O O H H N O N O H H N H N N N H N H N N N H 15 14 O H O O O H 13 16 20 17 18 19 21 22 23 N+ P O N+ P O H H 24 O O O N H H H O N N O H N H N N N H N N H H N N N H O O N H O H P O P O O H H N C N H O N H S O S O N H H N C N H O O N O N+ H O O N+ H 27 25 29 30 31 N N H O H 33 26 34 32 36 N H N H N N 37 28 35 Fig. 4 75 H-bonded ring motifs occurring in [12 structures in the CSD. form three H-bonds to a crown ether molecule (Fig. 6: the count will be given correctly as three if one of the substituents on the cationic N is not H). In these host»guest complexes, molecular recognition involves the three H atoms and the O atoms of the cyclic ether30 and this pattern is best regarded as the H-bonding motif. Similar complexes with trivalent nitrogen (25) or water (10) are not as likely to occur as 6, although they still exist in over half the number of possible structures.The four most frequently occurring motifs include the cyclic amide and carboxylic acid dimers. Perhaps surprisingly, R2 2 (8) the carboxylic acid dimer 23 only has a probability of ca. 0.33 of occurring, and the proportion of structures in which it Ps occurs is similar. It has been suggested that aliphatic and aromatic carboxylic acids generally form the dimer motif,9 although in some cases they form chains and a number of other patterns have been identi–ed.5 The relatively low overall probability may suggest that the motif is less prevalent than might have been thought, and this is due principally to the presence of competing functional groups, e.g. solvent water or carboxylate groups in partially deprotonated polycarboxylic acids.When restricted to those structures containing only C, O and H atoms, with no competing N, O acceptors or NH, OH donors in addition to carboxylic acid groups, the probabilities of formation are signi–cantly higher (Table 2). For structures containing any number of molecules, each possessing only one COOH group, and are ca. 95% for the Pm Ps motif, indicating a strong preference for the formation of R2 2 (8) intermolecular rings rather than chains.is somewhat lower Pm for dicarboxylic acids (ca. 85%); this is due in part to the alternative possibility of forming an intramolecular S(7) ring in 1,2-dicarboxylic acids. Dicarboxylic acids have a lower probability of forming rings across a crystallographic symmetry element than monoacids (60 vs. 75%); in monoacids, the number of crystallographically independent molecules must be [1 if an asymmetric motif is to be formed, whereas such a motif may form between two COOH groups of the same diacid molecule provided it does not itself lie on a crystallographic symmetry element. The lactam dimer 29, which has the largest value of Nobs , occurs in a similar proportion of structures to 23 although (Ps) the –gure is smaller (ca. 25%). This may suggest that some Pm fragments are geometrically incompatible in certain structures, perhaps due to fragment overlap as discussed above (Fig. 3). The H donors are constrained to adopt the appropriate cis conformation if the ring size is sufficiently small, although the trans conformation is commonly adopted in larger rings (P9 atoms) and this is incompatible with an motif.In struc- R2 2 (8) tures comprising only C, N, O and H atoms, with no competing H-bond donor/acceptor groups, the overall probability of 44% is less than half of for 23 (86%). The values for Pm Pm the pattern are higher in mono- and di-amides where the R2 2 (8) HNCxO groups do not overlap (note that ca. 17% of the 30 New J. Chem., 1999, 25»34O O O O H H H O O H O C O O O N+ H O O O O H H 56 57 N O H H N N O 58 59 60 H O O H O O 61 O N O N H O O O H N H N N N H O O O O H H O O O O H H H N N H O C N C O H N+ O H O C O H O O H O O H O H O O O H H O N H O O O O H H 62 67 65 63 64 68 O O O O H H 70 72 73 69 66 O O H H 75 74 71 N O N O H H O O O O H H N H N N N H H N C N H O O O N+ H N+ O N+ O H H H N+ H O O H N C+ N H O N O O H O O H O H O O N H 38 39 N H O O N H 40 41 43 42 44 45 H N N O 46 47 48 H N N O 49 N+ O N+ O H H N O N O H H H O O H O C O O H O O O H 50 52 53 51 H N C N H O O O O O H H 54 55 C N C O O H N N N H H O O O O O O NR + H H H Fig. 4 Continued amide groups are trans in each case). For ureas HNC(xO)NH, the probabilities of formation are signi–cantly higher based on the sample of eight structures ; the motifs tend to be crystallographically symmetric and the molecules typically lie on a crystallographic symmetry element.In contrast, Fig. 5 Motif formed with two fused rings of 2. Fig. 6 Ammonium/1,4,7,10,13,16-hexaoxacyclooctadecane (18-crown- 6) motif. is only ca. 40% for OxCN(H)CxO molecules, since it is Pm less favourable to adopt a bifurcated con–guration about an NH donor than a CxO acceptor group.is rather higher Ps (ca. 60%), since the diamide fragment may form one but not two rings in the absence of bifurcation at the donor. As a consequence, molecular symmetry is less likely to be re—ected in the crystal due to the lower symmetry of the molecular environment. The acyclic amide dimer 48 is much less likely to occur which may re—ect an increased tendency to form (Pm\10%) chain motifs when the conformation of the amide is not constrained by cyclisation. is only 16% when the subset is Pm restricted to molecules containing C, N, O and H with no competing donor groups.In secondary amides, the HNCxO groups are usually trans for steric and electronic reasons and form chains [the only cis example, which has a sterically undemanding H atom as the other CxO substituent, does form the motif].In contrast, ring motif is much more R2 2 (8) R2 2 (8) prevalent in primary amides, since one of the donor H atoms is necessarily cis, and the probabilities and are in the Pm Ps range 80»95%, i.e. comparable to those for the carboxylic acid dimers 23 in the absence of strong competing H-bonding groups.The overall proportions of motifs 23, 29 and 48 which form across a symmetry element are similar (each ca. 66%). In contrast, the cyclic sulfonamide dimer 32 (which has a \20% New J. Chem., 1999, 25»34 31Table 1 Statistics for the 75 motifs with SpossP12 Rank in Motif Graph set Pm(%) Nobs Nposs Sobs Ps(%) Sobs Sposs Psymm(%) 1 R2,2(12) 97 93 96 16 100 25 25 » 2 R2,2(8) 91 199 218 7 83 29 35 » 3 R2,2(7) 90 36 40 48 92 12 13 » 4 R2,2(8) 82 36 44 49 81 17 21 » 5 R2,2(8) 76 62 82 29 71 20 28 » 6 R2,2(10) 58 206 354 5 84 42 50 » 7 R2,2(8) 54 158 290 9 63 26 41 » 8 R2,2(8) 51 79 154 21 67 26 39 » 9 R2,2(12) 47 21 45 70 47 18 38 91 10 R2,2(10) 45 86 192 18 62 34 55 » 11 R2,2(8) 43 20 46 74 59 17 29 80 12 R2,2(8) 41 38 92 47 44 30 68 68 13 R2,2(20) 39 15 38 75 76 13 17 100 14 R2,1(6) 39 45 114 40 36 14 39 » 15 R2,1(6) 39 47 120 38 38 17 45 » 16 R2,2(9) 37 44 118 43 42 20 48 » 17 R2,2(8) 37 204 556 6 37 159 435 75 18 R2,2(10) 36 58 159 32 38 47 123 62 19 R2,2(10) 36 20 55 73 34 16 47 60 20 R2,2(10) 35 39 111 45 35 30 86 64 21 R2,2(8) 35 29 83 57 40 25 63 79 22 R2,2(8) 33 27 81 63 37 21 57 63 23 R2,2(8) 33 847 2541 2 32 596 1873 65 24 R2,2(10) 32 23 71 66 47 17 36 57 25 R2,2(10) 32 99 306 13 55 31 56 » 26 R2,2(6) 27 93 341 17 28 72 256 76 27 R2,2(8) 26 172 660 8 32 126 390 64 28 R2,2(9) 24 50 206 35 26 23 89 » 29 R2,2(8) 24 876 3687 1 34 627 1796 62 30 R2,2(8) 21 84 404 19 29 59 205 64 31 R1,2(6) 18 54 296 33 19 21 113 » 32 R2,2(8) 17 61 350 30 21 53 253 87 33 R1,2(6) 17 153 904 10 18 63 344 » 34 R2,1(4) 16 74 450 25 14 21 150 » 35 R2,1(5) 16 44 268 41 16 15 92 » 36 R2,2(7) 16 30 192 56 18 14 80 » 37 R2,2(10) 16 22 141 68 15 18 119 73 38 R2,2(10) 13 26 196 65 15 20 132 62 39 R2,2(10) 12 59 488 31 14 54 390 83 40 R2,2(8) 11 28 258 59 13 17 136 43 41 R2,2(6) 10 31 297 54 9 21 223 74 42 R2,1(5) 10 32 312 51 13 16 122 » 43 R2,2(10) 10 78 781 23 13 55 431 54 44 R2,2(7) 10 48 486 36 22 17 79 » 45 R1,2(6) 10 78 790 22 19 24 126 » 46 R2,2(12) 10 74 773 27 19 37 193 11 47 R2,2(8) 10 67 702 28 9 26 282 » 48 R2,2(8) 8 361 4310 4 10 268 2614 66 49 R2,2(14) 8 48 637 37 9 29 330 33 50 R2,2(10) 8 21 279 69 7 14 207 71 51 R2,2(10) 7 101 1362 12 9 62 726 33 52 R2,2(12) 7 32 434 52 8 28 344 94 53 R2,2(8) 7 28 402 61 7 24 349 86 54 R1,2(6) 6 31 532 53 7 13 189 » 55 R2,2(10) 5.4 28 519 60 6 21 355 64 56 R2,2(10) 5.4 80 1490 20 10 48 483 30 57 R2,2(12) 5.1 21 412 71 7 20 282 100 58 R2,1(4) 5.1 96 1892 15 6 42 708 » 59 R2,2(10) 4.6 26 570 64 6 17 275 54 60 R2,2(12) 4.4 54 1234 34 7 23 314 » 61 R2,2(12) 4.3 32 750 50 8 13 156 » 62 R2,1(4) 3.8 74 1936 26 4.3 30 691 » 63 R2,1(5) 3.8 44 1160 42 4.9 20 407 » 64 R2,2(8) 3.7 21 567 72 4.5 20 442 100 65 R2,2(10) 3.7 135 3702 11 7 79 1121 48 66 R2,2(10) 3.2 97 3002 14 5.5 47 861 » 67 R2,2(12) 3.1 38 1232 46 5.7 18 316 » 68 R2,2(9) 2.9 40 1378 44 3.9 19 488 » 69 R2,2(10) 2.2 28 1256 58 3.1 14 458 » 70 R2,2(4) 1.9 408 21824 3 1.9 203 10694 14 71 R2,2(7) 1.1 27 2466 62 1.5 14 907 » 72 R2,1(5) 1.0 46 4644 39 1.9 22 1173 » 73 R2,2(4) 1.0 75 7682 24 1.2 32 2768 » 74 R2,2(10) 0.8 22 2671 67 1.3 18 1369 64 75 R2,2(7) 0.4 30 7048 55 1.2 13 1121 » 32 New J.Chem., 1999, 25»34Table 2 Statistics for motifs 23, 29 and 48 in structures with no competing donors or acceptors Motif Pm(%) Nobs Nposs Ps(%) Sobs Sposs Psymm(%) 23 All 86 311 361 91 181 198 64 Mono 96 128 134 95 105 111 75 Di 85 139 164 88 63 72 60 29 All 44 136 308 64 72 114 74 Mono 52 17 33 52 15 29 88 Di, no overlap 57 33 58 59 17 29 70 Di, shared NH 86 19 22 100 8 8 90 Di, shared CxO 40 15 38 62 8 13 73 48 All, 1y and 2y 16 50 318 22 35 158 84 Mono, 1y 83 19 23 90 17 19 79 Di, 1y 95 19 20 90 9 10 100 Mono, 2y 2.1 1 47 2.4 1 41 » Di, 2y 0 0 52 0 0 46 » probability of occurring) forms across a symmetry element in ca. 87% of cases. Interestingly, the mixed amide/carboxylic acid dimer 47 is only as likely to occur ca. 9%) as the chemically- (Pm symmetric acyclic amide motif 48, again re—ecting the tendency of acyclic amides to adopt the trans conformation. The dimers 17 and 27 are similar chemically, with N donors R2 2 (8) and acceptors, diÜering only in whether the cyclic C»N bond is double or aromatic. For some structures, the precise identi- –cation of the motif as 17 or 27 will depend on the particular manner in which the structure has been coded in the CSD. 17 has a slightly higher probability of occurring than the R2 2 (8) dimers 23, 29 and 48 whereas 27 has closer to 0.33. 17 has Pm a greater proportion of occurrences associated with a symmetry element (ca. 75% compared to 62»66% for 23, 27, 29 and 48). As with motif 29, 27 is found in a lower proportion of fragments than structures which may be due in part (Pm) (Ps) to a lack of fragment complementarity.Interestingly, the cyclic (11) and acyclic (12) phosphoramide dimers are more likely to form than either their amide R2 2 (8) or carboxylic acid analogues 11 is particularly (Pm[40%); likely to be associated with a crystallographic symmetry element The most probable motifs involving (Psymm\80%).bifurcated H atoms are rings in which an ammonium R1 2 (6) (15), or amine (14), H is chelated by a nitro and aryloxy oxygen atom; these would appear to be signi–cantly more probable than the more common motifs 34 and 62 in which the H is chelated by both nitro oxygen atoms. The motif containing a bifurcated H which occurs in the largest number of structures comprises an ammonium cation and a carboxylate anion 58 (e.g.in amino acid zwitterions) ; this motif has only a ca. 5% probability of being formed. The analogous motif with carboxylate replaced by nitro 34 is R1 2 (4) more than twice as likely to occur. However, if the ammonium is also replaced by tri-coordinate nitrogen the probability of forming 62 (e.g.in nitroanilines2) is only ca. 4%. However, it is possible that the distance limits used were too short for the longer contact to be found for 34 and 62 in cases where the chelation was more asymmetric. Of those in which two donor atoms chelate an acceptor, 33 occurs most frequently although is less than 20%. This motif occurs typically in the Pm R2 1 (6) ìhead-to-tail œ packing of diarylurea molecules.31 The motif is formed with a similar probability if one of the C»N bonds is cyclic (31) although it is much less likely to occur if the oxygen atom is two-coordinate (54).Motif 45 (analogous to 33 but with a cationic three-coordinate C atom) occurs in the same proportion of structures as 33 although it only forms for on average 50% of the possible motifs in any structure in which it occurs, indicating that it is not possible to use all fragments simultaneously. In contrast to the cyclic and acyclic amides, the chemicallysymmetric HNCCxO dimer 51 is crystallographicallysymmetric in only 0.33% of the motifs although it occurs with a similar probability to 48 If the amino group is (Pm\8%).replaced by cationic nitrogen, the motif (43) is somewhat more probable and is almost equally likely to be sym- (PmB10%) metric crystallographically as not.Alternatively, if the C»N bonds are cyclic (20) the H donor is constrained to a cis conformation and the ring motif occurs in a comparable proportion of structures to that of the cyclic amide 29 and is (Ps) more likely to be symmetric crystallographically. The unit may form either the symmetric HOC(R)2C(R)2OH motif 65 or asymmetric motif 66 (where the C»C bond is cyclic, e.g.in pyranose and furanose sugars) due to the ability of the O-hydroxyl atoms to act as both donor and acceptor ; these motifs may also be adopted if some of the atoms are O-ether rather than O-hydroxyl. Interestingly, the probability of the chemically-symmetric motif 65 being crystallographically symmetric is close to 50%.The statistics for these motifs are based on the assumption that O-hydroxyl may act as both donor and acceptor [i.e. there are two fragments associated with a unit] ; whilst this does HOC(R)2C(R)2OH occur, it is relatively uncommon so that is about half the Pm value of (which is less than 10% for both 65 and 66). The Ps analogue of 65 in which the C»C bonds are intramolecularly acyclic (56) is slightly more probable, perhaps because the conformation is not restricted by the constraints of ring closure.Motif 70 represents a tandem RO»H/H»OR hydrogen R2 2 (4) bond con–guration; the large number of occurrences are due to the large number of RO»H fragments rather than any strong tendency to form the motif The majority of (Pm\2%).these motifs ([85%) are not associated with a crystallographic symmetry element. Calculations on the water dimer have shown that the tandem con–guration is energetically disfavoured with respect to the terminal con–guration by ca. 4 Cs kJ mol~1.32 The dimer 73 has an even lower R2N»H/H»OR probability of occurring (ca. 1%). Conclusions A new methodology has been used to identify the most common rings involving two H-bonds which are formed between pairs of organic molecules in the CSD, and to calculate their relative probabilities of occurrence.Whilst the derivation of might be re–ned further, comparison with the Nposs structural probability indicates that the values (Ps) Pm obtained are reasonable (at least for the 75 motifs with Sobs[ 12), and discrepancies may be rationalised on the basis of the geometrical characteristics of the constituent fragments.Some motifs are less likely to occur in the presence of competing intermolecular interactions than might have been thought [e.g. carboxylic acid and cis-amide dimers], a R2 2 (8) result that has clear consequences for crystal design. However, some of the motifs which have been applied speci–cally in crystal engineering (e.g.components of motifs comprising New J. Chem., 1999, 25»34 33three H-bonds), and host»guest complexes of crown ethers, have a high as expected. The survey has not revealed Pm , many other motifs with although those with Pm[50%, Pm[ 30% could be of practical use in the absence of strong competing interactions. These results have important implications, not only for crystal engineering, but also for molecular modelling and crystal structure prediction, and investigations are continuing on the eÜects of competing interactions on motif formation.The methodology can be applied to systems involving weaker H-bonds and other non-covalent interactions, including those in organometallic and inorganic systems.Work in these areas is in progress. Acknowledgements The authors wish to thank Dr. R. Scott Rowland (CCDC) for the non-bonded search routine, Dr. Jason Cole (CCDC) for the script to run the fragment searches and Profs. Gautam Desiraju (Hyderabad), Joel Bernstein (Beer Sheva) and Ray Davis (Austin) for valuable discussions. References 1 G. A. JeÜrey, An Introduction to Hydrogen Bonding, Oxford University Press, New York, NY, 1997. 2 T. W. Panunto, Z. Urbaç nczyk-Lipkowska, R. B. Johnson and M. C. Etter, J. Am. Chem. Soc., 1987, 109, 7786; M. C. Etter and G. M. Frankenbach, Materials, 1989, 1, 10; M. C. Etter, J. Phys. Chem., 1991, 95, 4601. 3 W. Jones, V. R. Pedireddi, A. P. Chorlton and R. Docherty, Chem. Commun., 1996, 997. 4 F. Garcia-Tellado, S.J. Geib, S. Goswani and A. D. Hamilton, J. Am. Chem. Soc., 1991, 113, 9265; J. Bernstein, M. C. Etter and L. Leiserowitz, in Structure Correlation, ed. H.-B. Bué rgi and J. D. Dunitz, VCH, Weinheim, 1994, vol. 2. 5 G. R. Desiraju, Crystal Engineering»T he Design of Organic Solids, Elsevier, Amsterdam, 1989; Angew. Chem., Int. Ed. Engl., 1995, 34, 2311; Chem. Commun., 1997, 1475. 6 C.B. Aakeroé y and K. R. Seddon, Chem. Soc. Rev., 1993, 22, 397. 7 V. R. Thalladi, B. S. Goud, V. J. Hoy, F. H. Allen, J. A. K. Howard and G. R. Desiraju, Chem. Commun., 1996, 401. 8 G. A. JeÜrey and S. Takagi, Acc. Chem. Res., 1978, 11, 264. 9 L. Leiserowitz and G. M. J. Schmidt, J. Chem. Soc. A, 1969, 2372. 10 L. Leiserowitz, Acta Crystallogra., Sect. B, 1976, 32, 775; L. Leiserowitz and M.Tuval, Acta Crystallog., Sect. B, 1978, 34, 1230. 11 A. F. Wells, Structural Inorganic Chemistry, Oxford University Press, Oxford, 1962. 12 G. Gilli, F. Belluci, V. Ferretti and V. Bertolasi, J. Am. Chem. Soc., 1989, 111, 1023. 13 W. F. Bailinger, P. v. R. Schleyer, T. S. S. R. Murty and L. Robinson, T etrahedron, 1964, 20, 1635. 14 G. A. JeÜrey, M. E. Gress and S.Takagi, J. Am. Chem. Soc., 1977, 99, 609. 15 L. N. Kuleshova and P. M. Zorkii, Acta Crystallogr., Sect. B, 1980, 36, 2113; P. M. Zorkii and L. N. Kuleshova, Zh. Strukt. Khim., 1980, 22, 153. 16 M. C. Etter, J. C. MacDonald and J. Bernstein, Acta Crystallogr., Sect. B, 1990, 46, 256; M. C. Etter, Acc. Chem. Res., 1990, 23, 120; J. Bernstein, R. E. Davis, L. Shimoni and N.-L. Chang, Angew. Chem., Int. Ed. Engl., 1995, 34, 1555. 17 J. Bernstein, Acta Crystallogr., Sect. B, 1991, 47, 1004. 18 J. Bernstein, M. C. Etter and J. M. MacDonald, J. Chem. Soc., Perkin T rans. 2, 1990, 695. 19 J. Bernstein and R. E. Davis, Graph Set Analysis of Hydrogen Bond Motifs, in Implications of Molecular and Materials Structure for New T echnologies, ed. J. A. K. Howard and F. H. Allen, Kluwer Academic Publishers, Dordrecht, to be published. 20 F. H. Allen and O. Kennard, Chem. Des. Automat. News, 1993, 8, 1; 31. 21 F. H. Allen, P. R. Raithby, G. P. Shields and R. Taylor, Chem. Commun., 1998, 1043. 22 J. S. Rollett, in Computing Methods in Crystallography, ed. J. S. Rollett, Pergamon, Oxford, 1965. 23 F. H. Allen, O. Kennard, D. G. Watson, L. Brammer, A. G. Orpen and R. Taylor, J. Chem. Soc., Perkin T rans. 2, 1987, S1. 24 A. Bondi, J. Phys. Chem., 1964, 68, 441. 25 F. H. Allen, W. D. S. Motherwell and G. P. Shields, Acta Crystallogr., Sect. B, submitted. 26 W. Saenger, Principles of Nucleic Acid Structure, Springer-Verlag, New York, 1984, ch. 6; G. A. JeÜrey and W. Saenger, Hydrogen Bonding in Biological Structures, Springer-Verlag, Berlin, 1991. 27 J.-M. Lehn, M. Mascal, A. DeCian and J. Fischer, J. Chem. Soc., Chem. Commun., 1990, 479; J. A. Zerkowski, C. T. Seto, D. A. Wierda and G. M. Whitesides, J. Am. Chem. Soc., 1990, 112, 9025; N. Kimizuka, T. Kawasaki and T. Kunitake, J. Am. Chem. Soc., 1993, 115, 4367; R. Ahuja, P.-L. Caruso, D. Moé bius, W. Paulus, H. Ringsdorf and G. Wildberg, Angew. Chem., Int. Ed. Engl., 1993, 32, 1033; D. A. Bell and E. V. Anslyn, J. Org. Chem., 1994, 59, 512; G. M. Whitesides, E. E. Simanek, J. P. Mathais, C. T. Seto, D. N. Chin, M. Mammen and D. M. Gordon, Acc. Chem. Res., 1995, 28, 37. 28 C. T. Seto and G. M. Whitesides, J. Am. Chem. Soc., 1993, 115, 905. 29 Y. Wang, B. Wei and Q. Wang, J. Cryst. Spectrosc. Res., 1990, 20, 79. 30 M. S. Fonar, Yu. A. Simonov, A. A. Dvorkin, T. I. Malinovskii, E. V. Ganin, S. Kotlyar and V. F. Makarov, J. Incl. Phenom., 1992, 12, 3291; F. Seel. N. Klein, B. Krebs, M. Dartmann and G. Henkel, Z. Anorg. Allg. Chem., 1985, 524, 95. 31 M. C. Etter and T. W. Panunto, J. Am. Chem. Soc., 1988, 110, 5896. 32 M. J. Frisch, J. A. Pople and J. E. Del Bene, J. Phys. Chem., 1985, 89, 3664. Paper 8/07212D 34 New J. Chem., 1999, 25»34

ISSN:1144-0546    

DOI:10.1039/a807212d

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Dynamic processes in organolithium chemistry : tetrameric and ìopenœ tetrameric chiral a-amino lithium alkoxides David R. Armstrong,a John E. Davies,b Robert P. Davies,b Paul R. Raithby,b Ronald Snaithb and Andrew E. H. Wheatley*b a Department of Pure and Applied Chemistry, University of Strathclyde, Glasgow, UK G1 1XL b Department of Chemistry, University of Cambridge, L ens–eld Road, Cambridge, UK CB2 1EW (in Cambridge, UK) 20th October 1998, Accepted 25th 1998 Receiøed Noøember Lithium N,N,N@-trimethylethylenediamide, 1-Li, reacts with 1 equiv.of benzaldehyde to LiN(Me)(CH2)2NMe2 , aÜord the corresponding chiral a-amino lithium alkoxide 2, which in the solid state is a conventional pseudo-cubane tetramer. Reaction of 1-Li with o-methoxybenzaldehyde aÜords instead a novel ìopenœ pseudo-cubane tetramer 3, wherein the coordinative mode of the potentially bidentate N,N,N@- trimethylethylenediamino moiety can be related to ligand chirality.Employment of p-methoxybenzaldehyde results in the isolation of a tetramer 4, whose solid-state structure is intermediate between those of 2 and 3 and therefore suggests that —uxional processes may operate.Extensive studies of 2»4 in non-donor solution indicate diverse and complex behaviour. Introduction While the extensive use of chiral organolithium reagents in organic synthesis in general has been noted,1 attempts to regiospeci–cally metallate aromatic systems2 have led to the utilisation of chiral a-amino lithium alkoxides for the metallation of arylaldehydes.3 Generally, reaction of a lithium dialkylamide with either a heterocyclic arylaldehyde4 or substituted benzaldehyde3,5 has aÜorded an intermediate which, while it contains a superior aldehyde protecting group, is also capable of directing a second equivalent of lithium to the ortho-ring position.The structural nature of such intermediates has attracted surprisingly little attention in spite of their demonstrable utility, only a few examples of solution studies,6,7 solid-state structures (of b-amino lithium alkoxides)6 and theoretical investigations having been reported. 8 It is this paucity of data which prompted us to investigate chiral a-amino lithium alkoxides in order to establish not only their overall structures but also the coordinative modes adopted by the metal centres in the presence of the amino moiety N-centres.Results and discussion Solid-state studies We report here the solid-state and solution structures of three tetrameric a-amino lithium alkoxide species, aÜorded in good yield by the simple 1 : 1 reaction of lithium N,N,N@-trimethylethylenediamide, 1-Li, with the corresponding benzaldehyde (Scheme 1). The simplest dialkylamino lithium alkoxide to incorporate both the bifunctional N,N,N@-trimethylethylenediamino and Scheme 1 benzoid aromatic moieties, 2, results from the treatment of in situ-generated lithium N,N,N@-trimethylethylenediamide (1-Li) with 1 equivalent of benzaldehyde.X-Ray crystallography shows that the crystals deposited when the yellow solution resulting from re—ux of the reaction mixture is stored at ]80 °C have an apparently simple solid-state tetrameric structure wherein each metal centre is made four-coordinate by virtue of a-N-centre coordination (Fig. 1). To facilitate the discussion, the tetrameric aggregates reported here will be considered to be comprised of four monomeric units wherein each monomer is de–ned as a moiety containing a fourmembered N»C»O»Li ring. While intact pseudo- (LiO)4 cubanes which show intra-monomer N-stabilisation6,9 of the Fig. 1 Structure of the pseudo-cubane core of 2; hydrogen (LiO)4 atoms (except the protected aldehyde hydrogen atoms) have been omitted for clarity and only the ipso-carbon atoms of the aromatic rings are shown. New J. Chem., 1999, 35»41 35metal centres (including two diastereomeric b-amino lithium alkoxides)6 have been reported previously, 2 is rendered interesting by virtue of the fact that in situ protection of the aldehyde renders the consequent a-amino lithium alkoxide monomers chiral.The solid-state structure of 2 can be viewed as a tetrameric racemate comprised of two stacked dimers (Fig. 1, top to bottom), wherein the four-membered N»C»O» Li rings associated with the two (S)-ligands are staggered with respect to the lower (R)-associated ones.Like other N»C»O» Li chelated7a structures, cubane distortion aÜords two (LiO)4 intra-dimer Li»O bond types, wherein (presumably by virtue of ring strain in the four-membered chelates) the mean intrachelate distance is appreciably longer than the mean interchelate one [2.339(7) and 1.908(6) respectively].At a mean Aé of 1.893(7) inter-dimer Li»O interactions are nominally the Aé , shortest in the pseudo-cubane core. It is clear from an analysis of the diÜerent Li»N distances that only the a-N-centres interact signi–cantly with the metal centres [Li»N\2.109(6) and 2.131 mean intra-(S)- and (R)-chelate respectively]. Aé However, while at 2.650(7) (mean) the d-N… … …Li distances Aé are substantially too long to be regarded as having signi–cant bonding relevance, Fig. 1 shows that for any given N,N,N@- trimethylethylenediamino moiety the orientation of the d-Ncentre is towards the same metal as the a-one coordinates, suggesting that the d-N-centres are not completely independent of the metal centres. The reaction of 1-Li in THF at [78 °C with omethoxybenzaldehyde aÜords 3 which, unlike, the conventional pseudo-cube of 2, is revealed by X-ray crystallography to be a novel ìopenœ pseudo-cubane tetramer in the solid state (Fig. 2) : the molecule occupies a crystallographic two-fold axis which passes through the Li(1)O(2)Li(1A)O(2A) and Li(2)O(4)Li(2A)O(4A) planes of the ìopenœ cube. Each metal centre achieves four-fold coordination, the fashion in which this is done being related to the chirality of the ligand at the protected aldehyde carbon atom.10 Coordination of the metal centres by the dialkylamino moieties, which is responsible for the exclusion of THF from the product, is either mono- or bisdepending upon whether they are associated with (S)- or (R)- ligands respectively, with bis-coordination by the (R)-ligands incurring the opening of one end of the pseudo-cubane and necessarily introducing signi–cant variations in core Li»O dis- Fig. 2 Structure of the ìopenœ pseudo-cubane core of 3; (LiO)4 hydrogen atoms (except the protected aldehyde hydrogen atoms) have been omitted for clarity and only the ipso-carbon atoms of the aromatic rings are shown. The symmetry operation which relates original atoms to their ìAœ equivalents is 1[x, y, 12 [z.tances. The solid-state structure of 3 can be viewed as an aggregate in which a dimeric pair of (R)-ligands lies next to a dimeric pair of (S)-ligands (Fig. 2, right to left), the inter-dimer Li»O distances varying between 1.909(8) [Li(2)»O(2)] and Aé 1.852(8) [Li(1A)»O(4)]. The closed face of the pseudo- Aé cubane incorporates the four-membered ring which (LiO)2 results from dimerisation of the two (S)-ligands and is itself composed of two types of Li»O bond, the inter-(S)-monomer ones [i.e. Li(1)»O(2)\1.903(8) being much shorter than Aé ] those in the four-membered intra-(S)-monomer chelate rings [i.e.Li(1A)O(2)C(6)N(1), Li(1A)»O(2)\2.303(8) which Aé ] result from mono-coordination of the a-amino moieties and which constitute anti-geometrical isomers with respect to the C»N interaction.While at 2.101(8) the Li(1A)»N(1) distance Aé , is consistent with distances found in related mono-complexed species,5,6,7c,7d,7f the non-bonding Li(1)… … …N(2A) distance is 2.754 Unlike their (S)-counterparts, the two (R)-ligands bis- Aé . coordinate the metal centres via both a- and d-N-centre donation, two eclipsed intra-monomer Li»O bonds in the pseudo-cubane core having concomitantly cleaved [i.e.Li(2)… … …O(4A)\2.723 The result is that instead of each Aé ]. (R)-ligand incorporating a four-membered Li»O»C»N chelate ring, both (R)-ligands combine with two intact edges of the open pseudo-cubane, aÜording an eight-membered ring in a boat conformation, anti-geometrical (Li»O»C»N)2 isomerism being observed again about the C»N bonds.The absence of signi–cant strain in the eight-membered boat is probably responsible for the observation of Li»O bonds [Li(2)»O(4)] which, at only 1.872(8) are somewhat shorter Aé , than those in the four-membered mono-chelate rings associated with the two (S)-isomers. The a-N»Li distance [Li(2)» N(4A)\2.118(8) is not signi–cantly diÜerent to the Aé ] analogous interactions in the mono-chelated (S)-ligands, and is somewhat shorter than the d-N»Li bond lengths [Li(2)» N(3A)\2.365(9) Aé ].The non-bonded Li(2)… … …O(4A) distance is, at 2.723(8) Aé , rather shorter than that of 3.14 observed in the only other Aé known example of an ìopenœ pseudo-cubane, (LiO)4 The two tetramers further diÜer in (PhOLi … THF)4 … PhOH.11 that opening of only one pseudo-cubane bond (by virtue of the inclusion of a non-lithiated phenol molecule whose hydroxyl group bridges one cubane edge) is demonstrated by the latter structure.Cleavage of a single pseudo-cubane bond has also been reported in the structure of (LiN)4-containing M6lithio(trimethylsilyl)diazomethane … 2lithio[4,5-bis(trimethylsilyl) triazine] … 7diethyl etherN.12 More extensive fragmentation has been recorded in polymeric this species (LiBr … THF)=,13 representing the only previous example of an ìopenœ pseudocubane in which, as for 3, two eclipsed Li»X bonds are missing.Some features in common with 3 are also demonstrated by wherein variations Li6Cl6 …2TMEDA… 4/2TMEDA, in the mode of external solvation incur the cleavage of two Li»Cl bonds between rings at either end of a hexa- (LiCl)3 meric aggregate.14 The reaction of 1-Li can be extended to the remaining mand p-methoxybenzaldehyde isomers, in the latter case diÜraction-quality crystals of 4 being aÜorded with the resultant structure demonstrating features characteristic of both 2 and 3.Super–cially, the solid-state structure of 4 is based (like that of 2) on a complete tetrameric pseudo-cubane racemate composed of a staggered stack of enantiomerically pure (S)- and (R)-dimers (Fig. 3, top to bottom). However, a closer analysis of bond lengths in the aggregate core reveals a more intriuging pattern. There exist, within the core, four (LiO)4 types of Li»O bond, the –rst being the inter-dimer interactions [1.893(5) mean]. Next, within either enantiomerically pure Aé dimer there are inter-monomer bonds which are essentially equivalent [1.908(5) mean] irrespective of ligand chirality.Aé However, intra-monomer Li»O interactions show a pronounced dependence both upon ligand chirality and upon the 36 New J. Chem., 1999, 35»41Fig. 3 Structure of the pseudo-cubane core of 4; hydrogen (LiO)4 atoms (except the protected aldehyde hydrogen atoms) have been omitted for clarity and only the ipso-carbon atoms of the aromatic rings are shown. coordinative mode of the dialkylamino moiety.In common with the solid-state structure of 2, the two (R)-ligands demonstrate mono-coordination of the metal centres via a-N-centre donation [2.133(5) mean], relatively long Li»O interactions Aé [2.162(5) mean] in the consequent (presumably) strained Aé four-membered Li»O»C»N chelates and no d-N-coordination [mean Li… … …N\2.941(5) However, bearing in mind the Aé ].novel structure of 3, the behaviour of the (S),(S)-dimer is particularly salient. Here the question of whether dialkylamine coordination is mono- or bi-dentate is rather vexed. Stabilisation of the lithium centres via a-N-centre coordination is again manifest [2.145(5) mean].However, what should be Aé the consequent Li»O»C»N four-membered rings show unexpectedly long lithium»oxygen distances M2.327(5) [Li(2)»O(5)] and 2.527(5) [Li(4)»O(3)] leading to the conclusion that Aé N, these are best viewed as only partial interactions. It is noticeable, however, that these bond extensions correlate with the observation of corresponding (albeit weak) interactions between the metal centres and the d-N-ones M2.544(5) [Li(2)» N(6)] and 2.435(5) [Li(4)»N(4)] Furthermore, the shorter Aé N.of these d-N»Li partial interactions [Li(4)»N(4)] involves the same metal centre as the longer of the partial Li»O bonds [Li(4)»O(3)] and, conversely, the longer d-N»Li partial interaction [Li(2)»N(6)] involves the same lithium centre as the shorter Li»O one [Li(2)»O(5)].This observation suggests that species such as these, if they can exist in solution without deaggregating signi–cantly, will be rapidly —uxional.16 Furthermore, the solid-state structure of 4 represents an intermediate state in the interconversion of the structural types represented by 2 and 3 (Scheme 2).Solution studies While 2 is only nominally soluble in aromatic solvents, it is possible to record the 1H NMR spectrum. It aÜords the protected aldehyde resonance as a singlet at d 5.87. However, the solution properties of 2 speci–cally (and the a-amino lithium alkoxides reported here in general) are rather complex. Over and above the expected aggregation/deaggregation behaviour (see below, compounds 3 and 4) these species exhibit apparently concentration dependent solution equilibria between the intact a-amino lithium alkoxide and its constituent lithium Scheme 2 dialkylamide and free aldehyde (Scheme 3; for 2, R\H).Hence, for 2, which is only sparingly soluble in non-donor aromatic media, equilibrium is moved towards the right-hand side, aÜording a free aldehyde : protected aldehyde ratio by 1H NMR spectroscopy immediately after sample preparation of 1 : 5.According to multinuclear NMR studies, 3 exhibits extremely complex solution behaviour. In the –rst instance, the solution equilibrium demonstrated by 2 is repeated. However, apparently by virtue of the slightly greater solubility of 3, the position of equilibrium is shifted less signi–cantly towards an aldehyde/1-Li mixture.At ambient temperature 1H NMR spectroscopy in (ca. 2 mg ml~1) [2H6]benzene shows a main protected aldehyde singlet at d 6.48 and –ve much less signi–cant satellites in the range d 6.12»6.01, all of which correlate with the protected aldehyde carbon centre at d 88.6 in the HMQC spectrum. 1H DPFGSE NOE spectroscopy15 s) shows a weak NOE (2.8%) (qm\0.7 between the d 6.46 singlet and the aromatic doublet at d 8.14.More interestingly, salient NOEs are observed to both of the dominant a- and d-N-methyl groups (2.4 and 4.8% respectively). This suggests that, in non-donor solution, not only is bis-chelation of the metal centres favoured but that the group adopts an endo-orientation with respect to the d-NMe2 resultant chelate ring [Scheme 4(a)].It is, nevertheless, apparent that the NOE to the a-NMe group is not consistent with the anti-isomerism demonstrated by the Li»O»C»N chelate rings in the solid-state structure of 3 [Scheme 4(b)]. Tentatively, therefore, it can be suggested that the observation of this NOE is evidence for the existence of small amounts of syn-isomers in solution. It is only possible to speculate at this point about their source, though it is possible that they result Scheme 3 Scheme 4 New J.Chem., 1999, 35»41 37from the recombination of free aldehyde and 1-Li (formed when trace amounts of 3 dissociate in solution). The 1H NMR spectrum at ambient temperature is essentially the same in as it is in and [2H8]toluene [2H6]benzene, in both cases the relative free aldehyde singlet : protected aldehyde multiplet ratio is approximately 1 : 50 immediately after preparation. That this ratio does not change on cooling the probe-head to [80 °C or on returning it to ambient temperature points to the equilibrium being neither temperature nor (in this case ; see discussion for 4, below) time dependent.These observations allow 7Li NMR spectroscopy to be undertaken over the same temperature range and on the same timescale.The –rst and most obvious result to be had from 7Li NMR spectroscopy is that, at the same concentration and at ambient temperature, 3 fails to exhibit two lithium environments in a 1 : 1 ratio and, therefore, on the NMR timescale the (R)- and (S)-monomers do not retain their solid-state diÜerences.This suggests either that the tetramer is rapidly —uxional or that it deaggregates. Cryoscopy in benzene17 favours the latter explanation, results suggesting (Table 1) that over the concentration range 1.3»3.9 mg ml~1 the mean aggregation state varies only from 1.4^0.3 to 1.5^0.2, and therefore that in non-polar solution a monomer»dimer equilibrium is dominant. In fact the 7Li NMR spectroscopic results bear out this view.While at ambient temperature in [2H6]benzene two major features in a 3.8 : 1 ratio (d [0.35 and [0.47) are observed, variable-temperature 7Li NMR spectroscopy in allows these (now observed at d [0.42 and [2H8]toluene [0.55) to be assigned respectively as dimer and monomer. At sub-ambient temperatures the high-–eld signal is absent, the low-–eld one having moved to d [0.33 at [25 °C and d [0.26 at [80 °C (Table 2).Compared to 3, 4 is very soluble in non-donor aromatic media and, furthermore, it tends not to deaggregate as signi–- cantly. Hence by cryoscopy in benzene (Table 3) the average aggregation state increases only from 2.9^0.7 to 3.5^0.1 over a wide concentration range (1.7»30.1 mg ml~1) and n is 3.4^0.1 at 10.2 mg ml~1.This value of n correlates well with 7Li NMR spectroscopy wherein a 10.0 mg ml~1 sample of 4 in at 0 °C shows three lithium signals in roughly [2H8]toluene a 2 : 9 : 1 ratio (Table 4). The dominant d 0.25 resonance, observable at all temperatures from 90 to [95 °C can, by virtue of its being the only signal to remain at low temperatures (below [50 °C), be considered to be the most aggregated species in solution, which, judging from cryoscopic data, is the tetramer.Of the two other signals (d 0.44 and 0.14 at Table 1 Cryoscopic data for 3 in benzene in the concentration range 1.3»3.9 mg ml~1 Conc./mg ml~1 Av. mol. mass n 1.3 345^65 1.4^0.3 2.7 376^33 1.5^0.2 3.9 362^50 1.5^0.2 Table 2 7Li NMR data for 3 in over the range 25 to [2H8]toluene [80 °C.(») indicates that the signal was not observable at that temperature T /°C d 25 [0.42(3.5) [0.55(1) 0 [0.38(4.9) [0.50(1) [25 [0.33(1) » [50 [0.29(1) » [80 [0.26(1) » 0 ppm\PhLi in at 25 °C. Integrals given in parenth- [2H8]toluene eses Table 3 Cryoscopic data for 4 in benzene in the concentration range 1.7»30.1 mg ml~1 Conc./mg ml~1 Av. mol. mass n 1.7 701^224 2.9^0.7 2.6 748^152 3.1^0.6 3.8 762^101 3.1^0.4 6.0 712^77 2.9^0.3 8.2 716^78 2.9^0.3 10.2 840^26 3.4^0.1 20.2 845^26 3.5^0.1 30.1 848^18 3.5^0.1 0 °C) the low-–eld one is nominally favoured at higher temperatures, allowing it to be assigned as the lower aggregation state species, the dimer, the remaining signal being attributable to the trimer. Calculation of the average formula mass and aggregation state at 10.0 mg ml~1 in at 0 °C [2H8]toluene aÜords values which, at 817 and 3.3 respectively, correlate closely with crysocopic data.Indeed, the closeness of this agreement, combined with the observation that the adoption of monomer and dimer respectively as the two remaining species aÜord poor concordance with cryoscopic data, demonstrate that the monomer is not viable in non-donor solution at this temperature and concentration.Nevertheless the slightly tentative nature of the dimer/trimer assignments is demonstrated by the fact that good correlation with cryoscopy can also be had if the two assignments are swapped, leading to a calculated average formula mass and aggregation state of 857 and 3.5 respectively. 1H NMR spectroscopy demonstrates that, like both 2 and 3, 4 is in equilibrium with its constituents at low concentration.The enhanced solubility of 4 aÜords the chance, however, to study this solution behaviour more closely and to this end spectra obtained at low, intermediate and high concentrations (0.5, 2.0 and 10.0 mg ml~1 respectively) both immediately after sample preparation and (for low and high concentrations) after a six-hour delay have been recorded (Table 5).While a p-methoxybenzaldehyde/1-H control spectrum shows a free aldehyde singlet at d 9.69, that of 4 (at 10.0 mg ml~1) has the protected aldehyde as a dominant resonance at d 5.89 (which correlates with the 13C NMR spectroscopy singlet at d 95.5 by HMQC spectroscopy) accompanied by a less signi–cant complex multiplet in the range d 5.86»5.83.The essential stability of 4 at high concentration is demonstrated by the fact that the 1 : 15 free aldehyde : protected aldehyde ratio immediately after preparation decreases only to 1 : 10 after 6 h. At intermediate concentration the initial ratio is 1 : 5, and at low concentration it is 1 : 1.75 with only free aldehyde observable after 6 h.The 7Li NMR spectra also re—ect these changes (Table 6). The high concentration spectrum does not change signi–cantly with time, while at lower concentration a signal Table 4 7Li NMR data for 4 in (10.0 mg ml~1), over [2H8]toluene the range 90 to [95 °C. (») indicates that the signal was not observable at that temperature T /°C d 90 » 0.27*(1) 0.21(1.5) » 80 » 0.33(1.5) 0.21(2) 0.15(1) 60 0.46(1) 0.35(4) 0.21(8) 0.11(2) 40 0.42(1) 0.37(4.5) 0.22(9) 0.11(1) 25 » 0.40(2) 0.23(5) 0.11(1) 0 » 0.44(2) 0.25(9) 0.14*(1) [25 » 0.47*(1) 0.28(7) » [50 » » 0.30(1) » [75 » » 0.31(1) » [95 » » 0.33(1) » 0 ppm\PhLi in at 25 °C.Integrals given in parenth- [2H8]toluene eses. *Shoulder. 38 New J. Chem., 1999, 35»41Table 5 1H NMR data for concentrated (10.0 mg ml~1), intermediate (2.0 mg ml~1) and dilute (0.5 mg ml~1) samples of 4 in at [2H6]benzene 25 °C p-Methoxybenz- 4 (conc.) 4 (conc.) 4 (int.) 4 (dil.) 4 (dil.) Assignment aldehyde]1-H t\0 h t\6 h t\0 h t\0 h t\6 h C(O)H 9.69, s, 1H 9.71, s, 0.07H 9.71, s, 0.1H 9.69, s, 0.2H 9.69, s, 0.6H 9.69, s, 1H Ar* » 7.59, d, 2H 7.59, d, 2H 7.59, d, 2H 7.59, d, 2H » Ar 7.55, d, 2H 7.55, d, 0.13H 7.55, d, 0.2H 7.54, d, 0.4H 7.54, d, 1.14H 7.55, d, 2H Ar* » 7.02, d, 2H 7.02, d, 2H 7.02, d, 2H 7.02, d, 2H » Ar 6.55, d, 2H 6.54, d, 0.13H 6.54, d, 0.2H 6.54, d, 0.4H 6.54, d, 1.14H 6.54, d, 2H ArC(H)(O)N* » 5.89, 5.86»5.83, 5.89,5.86»5.83, 5.89, s, 1H 5.89, s, 1H » m, 1H m, 1H OCH3* » 3.42, s, 3H 3.43, s, 3H 3.42, s, 3H 3.42, s, 3H » N(CH3)CH2* » 3.25, m, 1H 3.26, m, 1H 3.26, m, 1H 3.29, m, 1H » OCH3 3.13, s, 3H 3.09, s, 0.2H 3.09, s, 0.3H 3.08, s, 0.6H 3.08, s, 1.71H 3.08, s, 3H N(CH3)CH2* » 2.63, m, 1H 2.63, m, 1H 2.64, m, 1H 2.63, m, 1H » N(CH3)CH2 2.52, t, 2H 2.53, td, 0.13H 2.53, td, 0.2H 2.53, td, 0.4H 2.53, td, 1.15H 2.53, td, 2H NCH3 2.31, s, 3H 2.31, d, 0.2H 2.32, d, 0.3H 2.32, d, 0.6H 2.32, d, 1.71H 2.32, d, 3H N(CH3)2CH2 2.28, t, 2H 2.30, td, 0.13H 2.30, td, 0.2H 2.30, td, 0.4H 2.30, td, 1.14H 2.30, td, 2H N(CH3)CH2* » 2.25»2.20, m, 1H 2.25»2.20, m, 1H 2.25»2.21, m, 1H 2.25»2.21, m, 1H » NCH3* » 2.18, s, 3H 2.18, s, 3H 2.18, s, 3H 2.18, s, 3H » N(CH3)2 2.04, s, 6H 2.06, s, 0.4H 2.07, s, 0.6H 2.05, s, 1.2H 2.05, s, 3.4H 2.05, s, 6H N(CH3)2* » 2.04, s, 6H 2.04, s, 6H 2.04, s, 6H 2.04, s, 6H » N(CH3)2CH2* » 1.95, dt, 1H 1.95, dt, 1H 1.95, dt, 1H 1.95, dt, 1H » NH 1.22, s, 1H » » » » » The concentrated and dilute samples are re-observed after a time delay (t) of 6 hours.Signals marked * are due to 4 at d 0.23, consistent with 4, is initially dominant, but is replaced by a broad low-–eld signal with time. Conclusion Reactions of 1-Li, with benzaldehyde, LiN(Me)(CH2)2NMe2 , o- and p-methoxybenzaldehyde aÜord complexes 2, 3 and 4 respectively. Their solid-state structures (the –rst of chiral a-amino lithium alkoxides) have, in essence, pseudo-cubane cores.However, variations in the coordination modes (LiO)4 adopted by the bis(amino) moieties incur signi–cant diÜerences in the spatial orientations of the organic peripheries and also in the precise structural features of the cores.In all (LiO)4 three structures the a-N centres of the bis(amino) groups are coordinated to the lithium cations, and in 2 the d-N centres are distanced from these cations so that the cube is (LiO)4 intact. However, in 3 two of the four bis(amino) groups show d-N»Li coordination, causing opening of two eclipsed Li»O edges of the cube. The structure of 4 is intermediate between those of 2 and 3, with two weak d-N»Li interactions leading to signi–cant lengthening of two Li»O cube edges.The solution natures of these three chiral a-amino lithium alkoxides have been studied extensively by variableconcentration cryoscopic relative molecular mass measurements and by linked variable-concentration/-temperature 1H and 7Li NMR spectroscopy. All three species exhibit dissociative equilibria involving their constituents (1-Li]free aldehyde) and also equilibria involving tetrameric and lower aggregates and monomers.The precise conformations of these various solution species will be probed further in an attempt to explain the known speci–city, mainly ortho, of further lithiations of these intermediates. Experimental Synthesis of 2 BunLi (1.88 ml, 1.6 M in hexanes, 3.0 mmol) was added to N,N,N@-trimethylethylenediamine (1-H, 0.38 ml, 3.0 mmol) in toluene (13 ml) at [78 °C under nitrogen. After stirring for 10 min, benzaldehyde (0.30 ml, 3.0 mmol) was added.After a further 10 min at [78 °C the solution was warmed to room temperature yielding a white suspension which dissolved at re—ux. After –ltration, storage of the resultant yellow solution for 12 h at ]80 °C aÜorded colourless blocks of 2, mp 203» 205 °C, yield, 55%.Found: C 66.61, H 8.71, N 12.27. Calc. for C 67.29, H 8.88, N 13.08%. 1H NMR (500.130 C12H19LiN2O: MHz, d 7.61 (d, 2H, 2-Ph, Hz), [2H6]benzene), 3JHH\7.01 7.34 (t, 1H, 4-Ph, Hz), 7.21 (dt, 2H, 3-Ph, 3JHH\7.55 3JHH\ Hz), 5.87 [s, 1H PhC(H)(O)N], 3.24»2.28, (m, 4H, 7.28 CH2), 2.13»2.12 (m, 3H, 2.05»1.91 [m, 6H, 13C NCH3), N(CH3)2].NMR (100.614 MHz, d 129.2, 127.2, 126.4 [2H6]benzene), (Ph), 95.9 [PhC(H)(O)N)], 59.5»57.0 50.2 46.4» (CH2), (CH2), 45.6 [m, 36.9 (m, N(CH3)2], NCH3). Synthesis of 3 BunLi (1.88 ml, 1.6 M in hexanes, 3.0 mmol) was added to 1-H (0.38 ml, 3.0 mmol) in THF (4 ml) at [78 °C under nitrogen. After stirring for 10 min, o-methoxybenzaldehyde (0.36 ml, 3.0 mmol) was added and the resultant colourless solution was stirred for a further 10 min at [78 °C.Warming to room temperature gave a white suspension, which aÜorded a yellow solution on gentle heating. Storage at room temperature for one day resulted in the deposition of colourless, rectangular crystals of 3, mp 160»161 °C, yield, 82%. Found: C 63.65, H Table 6 7Li NMR data for concentrated (10.0 mg ml~1), intermediate (2.0 mg ml~1) and dilute (0.5 mg ml~1) samples of 4 in at [2H6]benzene 25 °C Conc., t\0 h Conc., t\6 h Int., t\0 h Dil., t\0 h Dil., t\6 h 0.42(2) 0.41(2) » » » » » 0.30*(1) 0.45*(1.8) 0.59*(1) 0.22(5) 0.22(5) 0.23(2) 0.23(1) » 0.11(1) 0.11(1) » » » Concentrated and dilute samples re-observed after a time delay (t) of 6 hours. 0 ppm\PhLi in at 25 °C. Integrals given in [2H6]benzene parentheses.*Broad. New J. Chem., 1999, 35»41 39Table 7 Crystallographic data for 2, 3 and 4 2 3 4 Formula C48H76Li4N8O4 C52H84Li4N8O8 C52H84Li4N8O8 Mr 856.92 977.03 976.24 Space group P1 6 C2/c P1 6 a/” 11.132(2) 21.795(6) 12.753(6) b/” 11.120(2) 12.828(4) 20.394(7) c/” 20.726(5) 20.754(7) 11.090(6) a/° 77.81(2) 90 98.18(4) b/° 83.31(2) 105.19(2) 105.19(2) c/° 89.950(14) 90 85.50(4) V /”3 2490.0(9) 5600(3) 2714(2) Z 2 4 2 Dc/g cm~3 1.143 1.159 1.196 Crystal size/mm 0.3]0.3]0.1 0.4]0.3]0.15 0.4]0.3]0.25 Radiation (j/”) Mo-Ka(0.71069) Mo-Ka(0.71069) Mo-Ka(0.71069) k/mm~1 0.072 0.077 0.079 F(000) 844 2112 1056 T /K 180 150 150 Scan mode u»2h u»2h u 2h range/° 8.10»50.02 5.12»39.96 5.00»50.16 Measured re—ections 11154 2707 10001 Unique re—ections 8730 2604 9536 Rint 0.1348 0.0335 0.0922 Re—ections with I[2p(I) 8713 2595 9515 Final R(F), wR(F2) 0.0698, 0.2719 0.0536, 0.1412 0.0571, 0.1585 Goodness-of-–t 1.031 1.045 1.014 Max.peak, hole/e ”~3 0.396, [0.351 0.172, [0.186 0.252, [0.258 8.39, N 11.12. Calc. for C 63.93, H 8.61, N C20H21LiN2O2: 11.48%. 1H NMR (500.130 MHz, d 8.14 (d, [2H6]benzene), 1H, 2-Ar, Hz). 7.18 (d, 1H, 5-Ar, Hz), 3JHH\6.57 3JHH\7.35 7.14 (dd, 1H, 4-Ar, Hz), 6.78 (dd, 1H, 3-Ar, 3JHH\7.45 Hz), 6.48»6.01 [m, 1H, ArC(H)(O)N], 3.72»3.31 3JHH\7.78 (m, 3H, 2.75»2.69, 2.32 (m, 2H, 2.68»2.21 (m, OCH3), CH2), 3H, 2.53»2.49 (m, 2H, 2.41»1.92 [m 6H, NCH3), CH2), 13C NMR (100.614 MHz, d 157.0, N(CH3)2]. [2H6]benzene), 135.8, (1-/2-Ar), 130.6 (2-Ar), 126.8 (5-Ar), 119.9 (4-Ar), 110.0 (3-Ar), 88.6 59.0 54.9 52.1 [ArC(H)(O)NR2], (CH2), (OCH3), 46.7»45.5 [m, 36.9»34.7 (m, 7Li (CH2), N(CH3)2], NCH3).NMR (155.508 MHz, 0 ppm\PhLi in [2H6]benzene, at 25 °C), d [0.35 (s, 3.8Li), [0.47 (s, 1Li). 1H [2H6]benzene DPFGSE NOE (500.130 MHz, irradiation at [2H6]benzene), d 6.46 s) gives a 2.8% NOE at d\8.14, a 2.4% NOE (qm\0.7 at d 2.68»2.21 and a 4.8% NOE at d 2.41»1.92.Synthesis of 4 BunLi (1.88 ml, 1.6 M in hexanes, 3.0 mmol) was added to 1-H (0.38 ml, 3.0 mmol) in 8 : 1 toluene»THF (2 : 0.25 ml) at [78 °C under nitrogen. The resultant yellow solution was stirred for 10 min at [78 °C whereupon pmethoxybenzaldehyde (0.36 ml, 3.0 mmol) was added. After a further 10 min at this temperature the solution was allowed to warm to room temperature, yielding a white suspension which was dissolved by gentle heating.Storage at room temperature for 12 h aÜorded colourless blocks of 4, mp 164»166 °C, yield, 75%. Found: C 63.79, H 8.74, N 11.57. Calc. for C 63.93, H 8.61, N 11.48%. 1H NMR C20H21LiN2O2: (400.134 MHz, d 7.59 (d, 2H, Ar, [2H6]benzene), 3JHH\8.49 Hz), 7.02 (d, 2H, Ar, Hz), 5.89, 5.86, 5.84, 5.83 [m, 3JHH\8.51 1H, ArC(H)(O)N], 3.42 (s, 3H, 3.25 (br, m 1H, OCH3), CH2), 2.63 (dt, 1H, Hz), 2.25»2.20 CH2 , 3JHH\11.80, 3JHH\3.23 (m, 1H, 2.18 (s, 3H, 2.04 [s, 6H, 1.95 CH2), NCH3), N(CH3)2], (dt, 1H, Hz, Hz). 13C NMR CH2 , 3JHH\12.48 3JHH\3.12 (100.614 MHz, d 158.8, 139.4, (Ar-CH) 112.7 [2H6]benzene), (Ar-C), 95.5 58.3 54.7 52.4 [ArC(H)(O)NR2], (CH2), (OCH3), 46.5 33.9 7Li NMR (155.508 (CH2), [N(CH3)2], (NCH3).MHz, 0 ppm\PhLi in at [2H6]benzene, [2H6]benzene 25 °C), d 0.42 (s, 2Li), 0.22 (s, 5Li), 0.11 (1Li). X-Ray crystallography Essential crystallographic details are given in Table 7. Data for 2 were collected on a Stoe-Siemens four-circle diÜractometer with data for 3 and 4 being collected on a Rigaku AFC5R four-circle diÜractometer. All three structures were solved using direct methods18 and subsequent Fourier diÜerence syntheses and re–ned by full-matrix least-squares on F2 with anisotropic displacement parameters for all non-hydrogen atoms.19 A riding model with idealised geometry was employed for H-atom re–nement.CCDC reference number 440/085. See http ://www.rsc.org/suppdata/njc1999/35/for crystallographic –les in .cif format.Acknowledgements wish to thank The Royal Society (J.E.D., P.R.R.) for –nan- We cial support and St. Catharineœs (R.P.D.) and Gonville & Caius (A.E.H.W.) Colleges for Research Fellowships. Notes and references 1 For reviews, see : P. J. Cox and N. S. Simkins, T etrahedron : Asymmetry, 1991, 2, 1; K. Koga, Pure Appl. Chem., 1994, 66, 1487; N. S. Simpkins, Pure Appl.Chem., 1996, 68, 691; N. S. Simpkins, in Advanced Asymmetric Synthesis., ed. G. R. Stephenson, Chapman & Hall, London, 1996, pp. 111»125; P. Beak, A. Basu, D. J. Gallagher, Y. S. Park and S. Thayumanavan, Acc. Chem. Res., 1996, 29, 552. 2 H. W. Gschwend and H. R. Rodriguez, Org. React., 1979, 26, 1; I. Omae, Chem. Rev., 1979, 79, 287; P. Beak and V. Snieckus, Acc. Chem. Res., 1982, 15, 306; N. S.Narasimhan and R. S. Mali, Synthesis, 1983, 957; N. S. Narasimhan and R. S. Mali, T op. Curr. Chem., 1987, 138, 63; V. Snieckus, Chem. Rev., 1990, 90, 879. 3 D. L. Comins and J. D. Brown, J. Org. Chem., 1984, 49, 1078. 4 D. L. Comins and M. O. Killpack, J. Org. Chem., 1987, 52, 104. 5 D. L. Comins and J. D. Brown, T etrahedron L ett., 1981, 22, 4213; D. L. Comins, J.D. Brown and N. B. Mantlo, ibid., 1982, 23, 3979; D. L. Comins and J. D. Brown, J. Org. Chem., 1989, 54, 3730. 6 E. M. Arnett, M. A. Nichols and A. T. McPhail, J. Am. Chem. Soc., 1990, 112, 7059; M. A. Nichols, A. T. McPhail and E. M. Arnett, ibid., 1991, 113, 6222. 7 (a) G. Hilmersson and Davidsson, J. Organomet. Chem., 1995, Oè . 489, 175; (b) G. Hilmersson and Davidsson, Organometallics, Oè . 1995, 14, 912; (c) G. Hilmersson and Davidsson, J. Org. Chem., Oè . 40 New J. Chem., 1999, 35»411995, 60, 7660; (d) H. J. Reich and K. J. Kulicke, J. Am. Chem. Soc., 1995, 117, 6621; (e) J. M. Saaç , G. Martorelli and A. Frontera, J. Org. Chem., 1996, 61, 5194, ( f ) K. Sugasawa, M. Shindo, H. Noguchi and K. Koga, T etrahedron L ett., 1996, 37, 7377. 8 A. Tatsukawa, K.Kawatake, S. Kanemasa and J. M. Rudzinski, J. Chem. Soc., Perkin T rans. 2, 1994, 2525; R. K. Dress, T. Roé lle and R. W. HoÜmann, Chem. Ber., 1995, 128, 673; L. M. Pratt and I. M. Khan, T etrahedron L ett., 1995, 6, 2165; G. Fraenkel, S. Subramanian and A. Chow, J. Am. Chem. Soc., 1995, 117, 6300. 9 O. Graalmann, U. Klingebiel, W. Clegg, M. Hasse and G. M. Sheldrick, Angew. Chem., 1984, 96, 904; Angew. Chem., Int. Ed. Engl., 1984, 23, 891; J. T. B. H. Jastrzebski, G. van Koten, M. J. N. Christophersen and C. H. Stam, J. Organomet. Chem., 1985, 292, 319; G. W. Klumpp, Recl. T rav. Chim. Pays-Bas., 1986, 105, 1; T. Maetzke and D. Seebach, Organometallics, 1990, 9, 3032; S. C. Ball. I. Cragg-Hine, M. G. Davidson, R. P. Davies, M. I. Lopez- Solera, P. R. Raithby, D. Reed, R. Snaith and E. M. Vogl, J. Chem. Soc., Chem. Commun., 1995, 2147. 10 J. E. Davies, P. R. Raithby, R. Snaith and A. E. H. Wheatley, Chem. Commun., 1997, 1721. 11 M. Pink, G. Zahn and J. Sieler, Z. Allorg. Allg. Chem., 1994, 620, 749. 12 G. Boche, K. Harms, M. Marsch and F. Schubert, Chem. Ber., 1994, 127, 2193. 13 A. J. Edwards, M. A. Paver, P. R. Raithby, C. A. Russell and D. S. Wright, J. Chem. Soc., Dalton T rans., 1993, 3265. 14 C. L. Raston, B. W. Skelton, C. R. Whitaker and A. H. White, Aust. J. Chem., 1988, 41, 1925. 15 K. Scott, J. Keeler, Q. N. Van and A. J. Shaka, J. Magn. Reson., 1997, 125, 302. 16 The ability of d-N»Li interactions in a-amino lithium alkoxides of this type to rapidly cleave and reform in non-donor solution has been observed by NMR spectroscopy : W. Clegg, S. T. Liddle, R. Snaith and A. E. H. Wheatley, New J. Chem., 1998, 1323. 17 M. G. Davidson, R. Snaith, D. Stalke and D. S. Wright, J. Org. Chem., 1993, 58, 2810. 18 G. M. Sheldrick, Acta Crystallogr., Sect. A, 1990, 46, 467; SHELXTL-PLUS, Program for Structure Solution and Re–nement, University of Goé ttingen, 1991. 19 G. M. Sheldrick, SHELXL-93, Program for Crystal Structure Re–nement, University of Goé ttingen, 1993. Paper 8/08050J New J. Chem., 1999, 35»41 41

ISSN:1144-0546    

DOI:10.1039/a808050j

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Synthesis, characterization, and crystal and molecular structures of caesium N-(carbamoylethyl)iminodiacetatooxoperoxovanadate(V) monohydrate Lœubomïç r Kuchta,a Michal Sivaç k,*a Jaromïç r Marek,b Frantisó ek Pavelcó ï ç ka and Marian C ã asnyç a a Department of Inorganic Chemistry, Faculty of Natural Sciences, Comenius University, 842 15 Bratislava, Slovak Republic; E-mail : sivak=fns.uniba.sk b Department of Inorganic Chemistry, Faculty of Natural Sciences, Masaryk University, 611 37 Brno, Czech Republic (in Montpellier, France) 24th June 1998, Accepted 21st October 1998 Receiøed A new monoperoxo complex of vanadium(V) of composition [where is Cs[VO(O2)(CEIDA)] …H2O H2CEIDA N-(carbamoylethyl)iminodiacetic acid] was prepared and characterized by IR, UV-VIS and 51V NMR spectroscopies.Thermal analysis showed that after crystal water release the anhydrous is Cs[VO(O2)(CEIDA)] formed, which is stable over a relatively large temperature interval and can be isolated by interruption of the dynamic heating in that interval. The X-ray structure analysis of revealed a Cs[VO(O2)(CEIDA)] …H2O mononuclear structure of the complex anion with a typical distorted pentagonal bipyramidal arrangement around vanadium formed by two peroxo oxygens bound in an g2-fashion, two oxygens from the deprotonated carboxylic groups and a nitrogen in the equatorial plane, and an oxo ligand and an oxygen from the carbamoyl group in the apical positions.The heteroligand is thus bound as a tetradentate CEIDA(2[)-N,O,O,O ligand to form one six-membered and two glycinate rings.Determination of the crystal structures of the native and peroxide forms of the vanadium-containing chloroperoxidase from Curvularia inaequalis fungus showed that vanadium atom in the active site of the enzyme is coordinated by four non-protein oxygens and one nitrogen atom from a histidine. 1,2 The peroxide form can thus be regarded as a monoperoxo complex with the peroxo ligand coordinated in an g2-fashion, which is the same as found in structures of all mono- and dinuclear vanadium(V) monoperoxo complexes.The study of such complexes with heteroligands containing one nitrogen and several oxygen atoms completing the coordination sphere of vanadium is important, as these complexes provide useful models for intermediates formed on interaction of the vanadium haloperoxidase with hydrogen peroxide.To date, the monoperoxo complexes of vanadium(V) with nitrilotriacetate(3[) (NTA),3h8 hydroxyethyliminodiacetate( 2[) (HHEIDA)9 and N-(carbamoylmethyl)iminodiacetate( 2[) (ADA)9,10 as tetradentate ligands, all containing an donor set, have been prepared and structurally O3N characterized. We report here the preparation, thermal and spectral properties and the crystal and molecular structures of where CEIDA is N- Cs[VO(O2)(CEIDA)] …H2O, (carbamoylethyl)iminodiacetate(2[).The study of [VO(O2)- (CEIDA)]~ ion formation in aqueous solution (pH 1»4) has been reported earlier by one of us.11 Experimental Chemicals and apparatus was prepared by thermal decomposition of V2O5 NH4VO3 (Loba Chemie).was synthesized from 3- H2CEIDA chloropropionamide and iminodiacetic acid (Aldrich) as described in the literature.12 All other chemicals were supplied by Lachema. Vanadium content was determined by titration with FeSO4 (diphenylamine as indicator) in a sample that was previously annealed in a Pt crucible. C, H and N were determined on a Carlo Erba 1106 analyzer. Caesium was estimated on a isotachophoretic analyzer ZTKI 01 (URVST Slovakia).The IR spectra were measured on a Nicolet Magna 750 FT IR spectrophotometer, the UV-VIS spectra on a Hewlett Packard 8452 spectrometer and 51V NMR spectrum on a Bruker DRX 500 spectrometer with as external stan- VOCl3 dard. The thermal analysis was performed on a Q-1500 Derivatograph (MOM, Budapest) with the following conditions : sample weight, 100 mg; heating rate, 0.6 °C min~1; air atmosphere.Syntheses Synthesis of (1). (0.091 g, Cs[VO(O2)(CEIDA)] ÆH2O V2O5 5 mmol) was dissolved in 2 ml of 30% The solution H2O2 . formed was diluted by 20 ml Solid CsCl (0.168 g, 1 H2O. mmol) and (0.204 g, 1 mmol) were then slowly dis- H2CEIDA solved in the diluted peroxovanadium(V) solution. The crystallization was initiated by dropwise addition of 96% ethanol to the orange-red solution (pH 2.1) until formation of a stable weak turbidity.The microcrystalline orange product, which formed within 2 days, was washed with ethanol and dried at room temperature. The monocrystals for X-ray analysis were obtained by recrystallization from a water»ethanol solution at pH 2.4. Anal. calcd (%) for Cs 29.40, V CsVC7H12N2O9: 11.27, C 18.60, H 2.68, N 6.20 ; found: Cs 29.20, V 10.94, C 18.64, H 2.58, N 6.11. 51V NMR (5]10~3 mol dm~3 in [583.3 ppm. UV-VIS LMCT band H2O): (H2O): (L\O22~) at 438 nm, e\390 mol~1 dm3 cm~1. IR (Nujol, cm~1) : 415vw, 450w, 470m, 495w, 534w, 564s 621s, 670m, [l(VwOp)], 698s, 734m, 800vw, 830m, 916vs 958vs [l(OpwOp)], [l(VxO)], 1026m, 1055vw, 1095sh, 1110m [l(CwNcoord)], 1206m, 1238w, 1265m, 1298s, 1325vs, 1375vs, 1415s, 1600» New J.Chem., 1999, 43»46 43Table 1 Crystal data and structure re–nement for Cs[VO(O2)(CEIDA)] …H2O Empirical formula C7H12CsN2O9V Formula weight 452.04 T /K 150.0(2) j/” 0.71073 Crystal system Monoclinic Space group P2g/n a/” 6.121(4) b/” 8.959(2) c/” 23.532(12) a/° 90 b/° 96.28(5) c/° 90 U/”3 1282.7(11) Z 4 Dc/g cm~3 2.341 k/mm~1 3.617 F(000) 872 Crystal size/mm 0.70]0.20]0.10 h range for data collection/° 1.74»25.08 Index ranges [7OhO0, 0OkO10, [7OlO27 Re—ections collected 2481 Independent re—ections 2238 (Rint\0.0536) Observed re—ections (I[2pI) 1960 Re–nement method Full-matrix least-squares on F2 Data/restraints/parameters 2238/0/211 Goodness-of-–t on F2 1.020 Final R indices (I[2pI)a R1\0.0476, Rw2\0.1273 R indices (all data)a R1\0.0553, Rw2\0.1355 Largest diÜ.peak and hole/e ”~3 1.396 and [1.694 aR1\&pFo o[oFc p/oFo o ; Rw2\M&[w(Fo2[Fc2)2]/&[w(Fo2)2]N1@2. 1665 vs, br [l(CxO)carbamoyl , las(COO), d(NH2), d(H2O)] (Op is peroxo oxygen). Preparation of (2). 2 was prepared by Cs[VO(O2)(CEIDA)] interruption of dynamic heating during the thermal analysis of 1 at 105 °C.It was identi–ed by CHN analysis and its IR spectrum, which still exhibits all characteristic bands of the monoperoxo complex. Anal. calcd (%) for C 19.37, H 2.32, N CsVC7H10N2O8 : 6.45 ; found: C 18.97, H 2.43, N 6.22. IR (Nujol, cm~1) : 455w, 483w, 525m, 550s, 570s 624s, 684w, 735m, 820m, [l(V»Op)], 838m, 924vs 950vs [l(VxO)], 1026m, 1062w, [l(OpwOp)], 1094w, 1126m 1200m, 1239m, 1280s, 1307s, [l(CwNcoord)], 1333s, 1405m, 1630vs, 1653vs [l(CxO)carbamoyl , las(COO), d(NH2)].X-Ray structure determination Low-temperature X-ray measurements of 1 were made on a KUMA KM4 diÜractometer with graphite monochromated radiation using the u»2h scan technique. No signi–- Mo-Ka cant decay of the intensity of three standard re—ections recorded after every 90 re—ections was observed.The data were not corrected for absorption eÜects. 2238 unique re—ections were obtained, of which 1960 with were classi–ed I[2pI as observed. Additional crystal data and re–nement results are given in Table 1. The structure was solved by a combination of Patterson and diÜerence Fourier methods, and re–ned by a full-matrix least-squares procedure (SHELXL-93) on oF2 o.13 Nonhydrogen atoms were re–ned anisotropically.The hydrogen atoms of the and groups were obtained from a dif- CH2 NH2 ference Fourier map and were re–ned isotropically with Ueq equaled to of the atom to which they are bonded. 1.2]Ueq Two hydrogen atoms from the water molecule were not identi –ed. The –nal re–nement was carried out with the weighting scheme where w\1/[(p2(Fo2)](0.0950P)2]6.4342P] P\ and with 2238 independent re—ections.The (Fo2]2Fc2)/3, re–nement converged at and R1(F)\0.0476 Rw2(F2)\0.1273 for the observed re—ections and 211 re–ned parameters. The largest positive and negative peaks in the diÜerence map were 1.40 and [1.69 e respectively. CCDC reference number Aé ~3, 440/076. Results and discussion The IR spectra of 1 and 2 exhibit in the expected regions all characteristic VxO, and stretches, which VwOp OpwOp allow the compounds to be identi–ed as monoperoxo complexes. 3h5,14h17 The IR spectra indicate that the CEIDA ligand is bound in both complexes via the imino nitrogen [shift of the band corresponding to the l(CwN) vibration from 1193 cm~1 in the uncoordinated ligand to 1110 cm~1 in 1 and 1126 cm~1 in 2] and via both deprotonated carboxylate groups [absence of the l(COOH) vibration observed in at 1705 cm~1]. Involvement of the carbamoyl H2CEIDA group in the coordination with vanadium in 1, due to the complexity of the spectrum in the 1600»1665 cm~1 region, could not be identi–ed by IR.The thermal analysis of 1 was performed under the given conditions in the temperature interval 20»180 °C.The weight loss, 3.9%, corresponding to the endothermic process in the interval 50»73 °C agrees with the calculated value, 3.98%, for loss of one water molecule. In the interval 75»155 °C no weight loss was observed; the decomposition product, obviously the anhydrous monoperoxo complex 2, is stable in this temperature interval. A further sudden exothermic decomposition process occurred at 155 °C.The weight loss corresponding to this process, 29%, exceeds the calculated value, 7.08%, for active oxygen release. The oxygen release thus proceeds simultaneously with decomposition of the CEIDA ligand. The crystal structure of 1 is built up of [VO(O2)(CEIDA)]~ anions, Cs` cations and water molecules held together mainly by electrostatic forces and hydrogen bonds.Their arrangement in the unit cell is shown in Fig. 1. The structure of the complex anion is shown in Fig. 2 and selected interatomic distances and angles are given in Table 2. The vanadium atom is seven-coordinated. The coordination polyhedron of vanadium is a distorted pentagonal bipyramid with carboxylic oxygens O(1), O(3), peroxo oxygens O(7), O(8) and imino Fig. 1 Arrangement of the unit cell in the structure of Cs[VO(O2)(CEIDA)] …H2O. 44 New J. Chem., 1999, 43»46Table 2 Selected interatomic distances and bond angles (°) for (”) Cs[VO(O2)(CEIDA) …H2O VwO(1) 2.041(4) O(1)wC(2) 1.279(8) Cs … … … O(5)ii 3.106(5) VwO(3) 2.038(5) O(3)wC(4) 1.301(8) Cs … … … O(9)w 3.205(6) VwO(5) 2.231(5) O(5)wC(7) 1.264(8) Cs … … … O(3)ii 3.248(4) VwO(6) 1.594(4) N(1)wC(1) 1.479(8) Cs … … … O(3)i 3.259(4) VwO(7) 1.881(5) N(1)wC(3) 1.490(8) Cs … … … O(4)iii 3.276(5) VwO(8) 1.870(5) N(1)wC(5) 1.503(8) Cs … … … O(7) 3.279(5) VwN(1) 2.165(5) Cs … … … O(7)i 3.089(5) Cs … … … O(6)i 3.340(4) O(8)wO(7) 1.439(7) Cs … … … O(8) 3.103(5) Cs … … … O(7)ii 3.450(6) O(1)wVwO(3) 149.7(2) O(5)wVwO(7) 83.6(2) O(8)wVwN(1) 153.8(2) O(1)wVwO(5) 82.7(2) O(5)wVwO(8) 83.6(2) O(7)wO(8)wV 67.9(3) O(1)wVwO(6) 95.7(2) O(6)wVwO(7) 102.6(2) O(8)wO(7)wV 67.0(3) O(1)wVwO(7) 124.4(2) O(6)wVwO(8) 102.5(2) C(2)wO(1)wV 118.3(4) O(1)wVwO(8) 79.9(2) O(7)wVwO(8) 45.1(2) C(4)wO(3)wV 117.4(4) O(3)wVwO(5) 83.7(2) O(1)wVwN(1) 75.7(2) C(7)wO(5)wV 123.6(4) O(3)wVwO(6) 94.7(2) O(3)wVwN(1) 76.1(2) C(1)wN(1)wV 104.7(4) O(3)wVwO(7) 80.5(2) O(5)wVwN(1) 84.1(2) C(3)wN(1)wV 105.5(4) O(3)wVwO(8) 125.2(2) O(6)wVwN(1) 89.3(2) C(5)wN(1)wV 117.1(4) O(5)wVwO(6) 173.4(2) O(7)wVwN(1) 154.5(2) Symmetry transformations used to generate equivalent atoms: i[x, [y]1, [z ; ii x]1, y, z ; iii[x[1, [y]1, [z ; w represents oxygen from water molecule.nitrogen N(1) in the equatorial plane, and with the carbamoyl group oxygen O(5) and doubly bonded oxygen O(6) in the apical positions.The –ve atoms in the equatorial plane are nearly coplanar, the maximum deviation from the plane is 0.016(3) The ”. vanadium atom is displaced from this plane by 0.226(2) ” towards the terminal oxygen O(6). The vanadium»apical oxygen O(5) bond length, generally elongated due to the structural trans in—uence,18 is 2.231(5) in the six-membered ” chelate in 1; this value is not signi–cantly longer than in NTA, ADA and HHEIDA monoperoxo complexes with –vemembered chelates (2.108»2.218 The distorted pentagonal Aé ). bipyramid is almost symmetric: the O(5), V, N(1), O(6) plane passes nearly through the midpoint between the O(8) and O(7) peroxo oxygens [deviations of 0.730(7) and [0.709(7) and ”], also the carboxylic oxygen atoms O(1) and O(3) are symmetrically placed with respect to this plane [deviations of 1.967(5) and [1.970(5) The symmetry of the anion with respect to ”].this plane is disturbed by the carbon atoms C(6) and C(7) of the six-membered ring, which deviate from this plane by 1.015(9) and 0.773(8) respectively. The O(7)wO(8) distance ”, found, 1.439(7) is identical with the average value found for Aé , this bond length in eight monoperoxo complexes containing a heteroligand with an donor set.4h10 The O3N V(OpwOp) group is nearly symmetric, the diÜerence in the two VwOp distances is 0.011 The most signi–cant diÜerence in the Aé .structure of 1 when compared with the structures of NTA, ADA and HHEIDA complexes containig a –ve-membered Fig. 2 ORTEP plot of the anion at 50% prob- [VO(O2)(CEIDA)]~ ability level for non-hydrogen atoms.apical chelate is the enlargement of the O(6)wVwO(5) angle from 165.87»167.7° for NTA, ADA and HHEIDA monoperoxo complexes4h8 to 173.4(2)° in 1 (Table 2). Within a distance of approximately 3.50 which corre- ”, sponds to the highest coordination number (12) of caesium surrounded by oxygens,19 the Cs` cation is in close contact with nine peroxo and carboxylic oxygen atoms belonging to four anions in diÜerent equivalent positions, and the O(9)w oxygen from the water molecule (Table 2).The data on crystal structures with coordinated CEIDA ligand are very scarce»only three structures of CoIII, CuII and NiII complexes have been determined so far.12,20,21 The structures of 1, CuII and NiII complexes indicate an oxophilic nature of these central ions : the carbamoyl group is preferably coordinated via oxygen and not the nitrogen atom.In the CoIII complex, CEIDA is coordinated as a tridentate ligand via two carboxylic oxygens and imino nitrogen. The characterized complex is Cs[VO(O2)(CEIDA)] …H2O the fourth complex of vanadium(V) with a heteroligand containing an donor set and the –rst one with a ligand O3N forming two –ve- and one six-membered chelates.The thermal stability study showed that the heteroligands discussed stabilize the group: oxygen release occurs at temperatures VO(O2) above 150 °C. The maintaining of the complex anion structure on dissolution, proved by a single chemical shift in the 51V NMR spectrum, provide a possibility for using this complex in a study of oxidation ability in the reaction with halides, that is, as functional model for haloperoxidase, or in oxygen transfer reactions.Acknowledgements work was supported by the Ministry of Education of the This Slovak Republic (Grant 1/5227/1998), the Ministry of Education of the Czech Republic (Grant VS 96095) and by the Grant Agency of the Czech Republic (Grants 203/95/1190 and 203/96/0111).We are grateful to Dr. K. Gaç plovskaç from the Chemical Institute, Comenius University Bratislava, Slovakia, for performing the isotachophoretic measurements. References 1 A. Messerschmidt and R. Wever, Proc. Natl. Acad. Sci., 1996, 93, 392. 2 A. Messerschmidt, L. Prade and R. Wever, Biol. Chem., 1997, 378, 309. 3 M. Sivaç k, D.Joniakovaç and P. Schwendt, T ransition Met. Chem., 1993, 18, 304. New J. Chem., 1999, 43»46 454 C. Djordjevic, P. L. Wilkins, E. Sinn and R. J. Butcher, Inorg. Chim. Acta, 1995, 230, 241. 5 Lœ. Kuchta, M. Sivaç k and F. Pavelcó ïç k, J. Chem. Res., 1993, (S) 393, (M) 2801. 6 A. E. Lapshin, Y. I. Smolin, Y. F. Shepelev, M. Sivaç k and D. Gyepesova ç , Acta Crystallogr., Sect.C, 1993, 49, 867. 7 Y. G. Wei, G. Q. Huang and M. C. Shao, Polyhedron, 1994, 13, 1587. 8 W. Da-Xu, L. Xiu-Lian, C. Rong and M. Mao-Chun, L iegou Huaxue (J. Struct. Chem.), 1992, 11, 65. 9 G. J. Colpas, B. J. Hamstra, J. W. Kampf and V. L. Pecoraro, J. Am. Chem. Soc., 1996, 118, 3469. 10 M. Sivaç k, J. Tyrsó elovaç , F. Pavelcó ïç k and J. Marek, Polyhedron, 1996, 15, 1057. 11 M. Sivaç k and P. Schwendt, T ransition Met. Chem., 1989, 14, 273. 12 F. Pavelc` ïç k, P. Novomeskyç , J. Soldaç novaç and T. Polynova, Collect. Czech. Chem. Commun., 1988, 53, 1725. 13 G. M. Sheldrich, SHEL XL 93: Program for Re–nement of Crystal Structures, University of Goé ttingen, Germany, 1993. 14 P. Schwendt, Collect. Czech. Chem. Commun., 1983, 48, 248. 15 C. Djordjevic, M. Lee-Renslo and E. Sinn, Inorg. Chim. Acta, 1995, 233, 97. 16 P. Schwendt, K. Lisó có aç k and L. Chem. Pap., 1993, 47, 288. Dã urisó , 17 A. Butler, M. J. Clague and G. E. Meister, Chem. Rev., 1994, 94, 625 and references therein. 18 R. Stomberg, Acta Chem. Scand., Ser. A, 1984, 38, 541 and references therein. 19 International T ables for X-ray Crystallography, Kynoch Press, Birmingham, England, 1968, Vol. III. 20 A. B. Ilyukin, A. L. Poznyak and L. M. Shkolnikowa, Koord. Khim., 1993, 19, 930. 21 L. M. Shkolnikowa, K. D. Suyarow, A. A. Masyuk, A. L. Poznyak and N. M. Dyatlowa, Koord. Khim., 1990, 16, 1096. Paper 8/04838J 46 New J. Chem., 1999, 43»46

ISSN:1144-0546    

DOI:10.1039/a804838j

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Solvation of praseodymium and cerium chlorides in anhydrous methanol and iso-propanol from ultrasonic velocity measurements Jacek Glin8 ski,* Barbara Keller and Janina Legendziewicz Faculty of Chemistry, University of F. Joliot-Curie 14, 50-383 Poland. W roc°aw, W roc°aw, Fax: ]48 71 328 2348; E-mail : glin=wchuwr.chem.uni.wroc.pl (in Montpellier, France) 28th April 1998, m/s 25th August 1998, Receiøed Reøised receiøed Accepted 9th 1998 Noøember Ultrasonic velocities and densities were measured, and the isentropic compressibility coefficients and solvation numbers for and (anhydrous and hydrated) solutions in methanol and iso-propanol at 25 °C were CeCl3 PrCl3 calculated.The results were interpreted in terms of inner-sphere coordination of chloride anions by lanthanide cations, as well as changing dissociation degree of the solutes with concentration. The steric eÜect of the solvent molecules on modelling the structure and coordination number of metal ions was considered and compared with the X-ray diÜraction of solid solvates.The above studies were correlated with our earlier optical investigations1 and con–rmed by conductivity measurements.For comparison, the same methods were also applied to cesium, calcium and tetraethylammonium chlorides. For over twenty years we have reported the application of spectroscopy to study equilibria in non-aqueous solutions of anhydrous and hydrated chlorides of lanthanides.2h6 Very recently, we have investigated the structure and emission properties of hydrated and anhydrous cerium and praseodymium chlorides dissolved in diÜerent alcohols.1 The aim of this work was to explain how the steric factor (namely the branch chain of the alcohol) aÜects the optical properties of the solution (absorption, excitation and emission), how the equilibrium of diÜerent metal oxidation states coexisting in the solution (M3` and M4`) in—uences the emission intensity, and how the concentration of active ions aÜects the electronic transition probabilities, emission and M3`/M4` equilibrium.The results show that the solutes under investigation are weak electrolytes in both methanol and iso-propanol: even at low concentrations there are two chloride anions incorporated in the inner solvation spheres, while the third Cl~ anion dissociates, but not fully.The inner-sphere coordination of lanthanide cations in ethanol was reported as early as 1966 by Ryan and Jorgensen.7 These results are in accord with our earlier ones in which, on the basis of europium charge-transfer bands, the inner-sphere coordination of halide ions (Cl~ and Br~) was shown to occur and their number depends on the amount of water present in the solution.5,6 The above means that water in—uences the equilibrium between diÜerent halogenocomplexes in solutions.1,2,5,6 In light of the above, it seemed interesting to apply ultrasonic methods, which are very useful for the study of liquid mixtures, especially in determining the solvation numbers.While the results obtained for chlorides of lanthanides were a typical, solutions of chlorides of cesium, calcium and tetraethylammonium were also investigated for comparison and to observe the in—uence of cation size and/or charge density on the solvation.Moreover, the coordinating abilities of these cations are much lower (or non-existent) than those of lanthanide ones, which should also be re—ected in the observed phenomena. Experimental Chemicals and solutions Cerium and praseodymium hydrated and anhydrous chlorides were obtained by the technique described earlier in detail for lanthanide perchlorates.1,8 IR spectra were used to detect water in the anhydrous salts.Cesium chloride (Aldrich 99%) and calcium chloride (Ubichem Ltd., pure for analysis) were dried in vacuum at elevated temperature for a few days. Tetraethylammonium chloride (Fluka, pure) was crystallized from a water»methanol mixture and dried and stored in a vacuum desiccator under Puri–cation of solvents was as report- P2O5 .ed before.2,9,10 Initial solutions (the most concentrated ones) were obtained by weighing. Other solutions were prepared by gradual dilution of the initial ones by volume. All manipulations and measurements were performed in a dry nitrogen atmosphere to avoid any contact of humid air with the solutions. Density and ultrasonic velocity measurements All these measurements were performed at 298.15 K (controlled with an accuracy of ^0.02 K).Density was measured with an accuracy of ^0.1 kg m~3 using a MG-2 (Ecolab, Poland) vibrating tube microprocessor apparatus. Sound velocity was measured using a SA-1000 ììsing-aroundœœ type device with an accuracy of ^0.1 m s~1.The details are given elsewhere.11,12 Conductivity measurements were performed using a MP2 Energopionier (Poland) conductivity meter (accuracy ^0.2 lS). For calibration, a 0.01 M aqueous solution of KCl was used (speci–c conductivity 1.4138]10~3 ohmv1 m~1). Results and discussion From the primary sound velocity (v) and density (o) data, collected in Table 1, we calculated the adiabatic compressibility coefficients (b) of solutions, also included in Table 1, using the Laplace formula b\(v2o)~1. Figs. 1 and 2 show the concentration dependences of the compressibility coefficient of the systems tested. For comparison, Figs. 1 and 2 also show the results obtained for CsCl, and solutes. Note, (C2H5)4Cl CaCl2 however, that solubilities of CsCl, anhydrous and anhy- CeCl3 drous in iso-propanol were too low and the data for PrCl3 these systems are not included in Fig. 2. In both methanol and iso-propanol, all the solutes cause a gradual decrease in the compressibility with increasing salt New J. Chem., 1999, 47»52 47Table 1 The experimental results Adiabatic compressibility Molar Molarity Density Sound velocity coef–cient conductivity c/mol l~1 o/kg m~3 v/m s~1 b/10~9 m2 N~1 Kmol/m2 mol~1 )~1 CsCl in methanol 0.000000 787.67 1081.12 1.0862 » 0.000500 787.78 1081.12 1.0860 0.0245 0.001000 787.89 1081.05 1.0860 0.0154 0.002000 787.94 1081.21 1.0856 0.0139 0.005000 788.42 1081.37 1.0847 0.0123 0.010000 789.28 1081.60 1.0830 0.0107 0.020000 790.63 1081.85 1.0807 0.0100 0.050000 795.30 1082.93 1.0722 0.00894 0.100000 802.97 1084.01 1.0598 0.00787 (C2H5)4NCl in methanol 0.000000 787.67 1081.12 1.0862 » 0.001000 787.65 1081.40 1.0857 0.0150 0.002000 787.73 1081.66 1.0850 0.0136 0.005000 787.90 1082.14 1.0838 0.0124 0.010000 788.31 1082.93 1.0817 0.0111 0.020000 788.72 1084.59 1.0778 0.0102 0.050000 790.48 1089.39 1.0660 0.00930 0.100000 792.82 1095.86 1.0503 0.00787 CaCl2 in methanol 0.000000 787.67 1081.12 1.0862 » 0.002000 787.81 1081.34 1.0856 0.0159 0.006000 788.45 1082.23 1.0829 0.0128 0.010000 789.08 1082.77 1.0809 0.0114 0.020000 790.53 1084.33 1.0759 0.00956 0.050000 794.27 1088.53 1.0625 0.00822 0.100000 800.93 1093.37 1.0444 » CeCl3 (anhydr.) in methanol 0.000000 787.67 1081.12 1.0862 » 0.000768 788.29 1081.28 1.0850 0.0116 0.001920 788.33 1081.34 1.0848 0.00959 0.002855 788.79 1081.53 1.0838 0.00839 0.004800 789.26 1081.66 1.0829 0.00715 0.028550 795.45 1084.27 1.0693 0.00495 0.048000 800.15 1085.41 1.0608 0.00439 PrCl3 (anhydr.) in methanol 0.000000 787.67 1081.12 1.0862 » 0.000726 788.30 1081.47 1.0846 0.0143 0.001814 788.53 1081.56 1.0841 0.0103 0.002702 788.96 1081.66 1.0833 0.0106 0.004535 789.37 1081.75 1.0826 0.00709 0.027020 795.06 1083.57 1.0712 0.00582 0.045350 799.78 1084.78 1.0625 0.00512 CeCl3 … 7H2O in methanol 0.000000 787.67 1081.12 1.0862 » 0.000500 787.81 1081.28 1.0857 0.0179 0.001000 788.20 1081.53 1.0846 » 0.002000 788.50 1081.34 1.0846 0.0111 0.005000 789.45 1082.52 1.0809 0.00894 0.010000 790.94 1084.17 1.0756 0.00787 0.020000 793.73 1085.35 1.0695 0.00697 0.050000 802.97 1092.00 1.0444 0.00572 0.100000 817.43 1100.73 1.0097 » PrCl3 … 6H2O in methanol 0.000000 787.67 1081.12 1.0862 » 0.000050 787.80 1081.63 1.0850 0.100 0.000500 787.87 1081.75 1.0846 0.0161 0.001000 787.98 1081.95 1.0841 0.0127 0.002500 788.48 1082.36 1.0826 0.00994 0.005000 789.32 1083.12 1.0799 0.00822 0.010000 790.70 1084.17 1.0759 0.00712 0.050000 802.36 1091.40 1.0463 0.00472 (C2H5)4NCl in iso-propanol 0.000000 782.03 1118.50 1.0221 » 0.000500 781.90 1118.31 1.0226 0.00211 0.001000 781.90 1118.28 1.0227 0.00182 0.002000 781.98 1118.82 1.0216 0.00147 0.005000 782.20 1119.11 1.0208 0.00111 0.010000 782.45 1120.10 1.0187 0.00091 0.020000 782.89 1121.41 1.0157 0.00072 0.050000 784.55 1124.70 1.0076 0.00058 0.100000 787.25 1131.09 0.9929 » 48 New J.Chem., 1999, 47»52Table 1 (Continued) Adiabatic compressibility Molar Molarity Density Sound velocity coef–cient conductivity c/mol l~1 o/kg m~3 v/m s~1 b/10~9 m2 N~1 Kmol/m2 mol~1 )~1 CaCl2 in iso-propanol 0.000000 782.03 1118.50 1.0221 » 0.001875 782.67 1119.52 1.0194 0.00012 0.003750 782.78 1119.59 1.0192 0.000072 0.007500 783.09 1119.65 1.0186 0.000054 0.015000 783.81 1119.62 1.0178 0.000041 0.030000 784.98 1119.52 1.0164 0.000029 CeCl3 … 7H2O in iso-propanol 0.000000 782.03 1118.50 1.0221 » 0.000938 782.27 1119.14 1.0206 0.00019 0.001875 782.54 1118.88 1.0208 0.00015 0.003750 783.04 1119.62 1.0188 0.00015 0.007500 784.16 1119.78 1.0170 0.00012 0.015000 785.86 1119.90 1.0146 0.00011 0.030000 789.90 1116.33 1.0159 0.00010 PrCl3 … 6H2O in iso-propanol 0.000000 782.03 1118.50 1.0221 » 0.001875 782.36 1118.98 1.0208 0.00019 0.003750 782.86 1118.60 1.0209 0.00013 0.007500 783.77 1119.20 1.0186 0.00012 0.015000 785.53 1118.56 1.0175 0.00010 0.030000 789.31 1115.12 1.0189 0.00089 concentration.In methanol this compressibility decrease is roughly proportional to the number of ions that could dissociate and/or to their charges ; thus, 1 : 1 electrolytes [CsCl and cause the smallest, and (1 : 3 electrolytes), (C2H5)NCl] LnCl3 the strongest compressibility changes. This behaviour is consistent with the electrostrictive character of the interactions of electrolytes with solvent (see below).It seems interesting that the eÜect of hydrated lanthanide chlorides on the compressibility of methanol is much stronger than that of anhydrous ones, suggesting higher solvation numbers of the former.Generally, decreasing compressibility of an ionic solution can be attributed to the increased number of solvent molecules that experience the electric –eld of the ions. While Fig. 1 Adiabatic compressibility vs. salt concentration in methanol. CsCl, (|) (») N(C2H5)4Cl, (K) CaCl2 , (+) CeCl3(anhydr.), ()) CeCl3 … 7H2O, (Ö) PrCl3(anhydr.), (L) PrCl3 … 6H2O.solvent molecules become incompressible in strong electric –elds (electrostriction), increasing concentration of ions results in the engagement of more solvent molecules in incompressible solvation spheres. However, both chloride ions and water molecules (when hydrated salt is used) can replace solvent molecules in the primary, as well as secondary, spheres.It is possible that, even at low concentrations of lanthanide chlorides in alcohol, chloride ions are present in close vicinity to the lanthanide cations. As a result, the lanthanide cation, together with its sphere (containing negatively charged chlorides), has a lower charge density, which means much fewer electrostrictive interactions with its surroundings. This Fig. 2 Adiabatic compressibility vs. salt concentration in isopropanol. For symbols see Fig. 1. Solubilities of CsCl and anhydrous lanthanide chlorides in iso-propanol are very low and determination of the dependence of compressibility on concentration was impossible in these systems. New J. Chem., 1999, 47»52 49process should increase the observed compressibility of the solution with increasing salt content.Two questions remain: how many chloride ions are engaged in the solvation sphere of the lanthanide cation at a given concentration, and what is the composition of the solvation sphere of this cation at in–nite dilution of the salt. The results of spectroscopic investigations (see Introduction) suggest that there are two Cl~ anions in the –rst solvation sphere in solutions.1,2,5 Thus, the observed non- iso-C3H7OH proportionality between the charge of the cation and the decrease in compressibility in iso-propanol (Fig. 2) seems to be caused by replacement of iso-propanol molecules in the outer solvation sphere by the third chloride anion. The resulting complex is not charged and, consequently, aÜects the neighbouring alcohol molecules only weakly.These observations were con–rmed by our conductivity measurements described below. The above is in contrast with the behaviour of aqueous solutions of chlorides. Choppin reported that very weakly basic ligands (like chlorides) favour outer-sphere complexation. 13 However, as proved below, in the case of alcoholic solutions the lanthanide chlorides apparently are not strong electrolytes and the chloride ligands may be present in the inner solvation sphere even at in–nite dilution, that is, they do not dissociate during salt dissolution.In light of this explanation, the methanolic solutions must not exhibit a similar characteristic. Methanol is a somewhat better solvating agent than iso-propanol, mainly because it is small and its dielectric constant is higher.It is possible, however, that addition of chloride to the solvation shell also occurs in methanolic solutions of lanthanide chlorides. Since is small, the interaction of M3` ions reaches more CH3OH easily the second coordination sphere and the observed CN (coordination number) is higher. This is con–rmed by the X-ray structural data, where CN of Nd3` ions decreases in its solvates with iso-propanol down to 7.14,15 Additionally, the exchange of molecules in the initial solvation sphere of H2O hydrated lanthanide chlorides can occur. As a result, the eÜect of the above processes is re—ected in the observed changes of adiabatic compressibility of the systems under investigation.Cations aÜect the solvent structure much more strongly than anions, mainly because the ionic radii of cations are smaller than those of anions.Assuming a purely electrostrictive mechanism of solvation of ions by alcohol molecules, decreasing compressibility can be attributed mainly to the formation of solvation shells of cations. The solvent molecules forming the shell are compressed enough that we can assume their incompressibility.This assumption allowed Pasynski to calculate solvation numbers of cations according to the formula:17 ns\[ U2 V 1 0 b1 0 + n1 n2 A1[ b b1 0 B (1) where is the solvation number, and are the number of ns n1 n2 moles of solvent and solute, respectively, is the molar V 1 0 volume of the solvent, is the apparent molar compress- U2 ibility of the electrolyte, and b and are the compressibility b1 0coefficients of the solution and the pure solvent, respectively.Both and can be easily calculated from sound velocity U2 b1 0 and density data. Another method to calculate the solvation number, although based on similar assumptions, that is, formation of an incompressible solvation shell around ions, was derived by Ernst and coworkers.18,19 Purely electrostrictive solvation should lead to the following relation between the composition of the electrolytic solution and its compressibility : 103b c1 \V 1 0 b10[ns V 1 0 b1 0 M1 103 m2 (2) Table 2 The solvation numbers of the salts in two alcohols ns Equation (1) Equation (2) Solute CH3OH i-C3H7OH CH3OH i-C3H7OH CsCl 6.4^0.8 » 5.8^0.6 » N(C2H5)4Cl 8.8^0.6 3.7^0.6 5.8^1.2 2.0^0.5 CaCl2 10.1^1.5 1.3^0.3 13.5^1.6 1.2^0.9 CeCl3 … 7H2O 18.6^2.5 7^0.5a 16.5^1.6 4.3^1.2 CeCl3 (anhydr.) 12.1^1.0 » 12.0^1.5 » PrCl3 … 6H2O 21.3^3.0 8^0.5a 16.5^1.6 5.0^0.8 PrCl3 (anhydr.) 11.7^1.5 » 11.5^1.2 » a At in–nite dilution, see text.where is the molar concentration of the solvent in the solu- c1 tion, is the molar mass of the solvent, and is the molal- M1 m2 ity of the solute (concentration expressed in moles per kg of the solvent).The solvation numbers of the systems under investigation, calculated using eqn. (1) and (2), are collected in Table 2. Since depends on concentration, Table 2 contains ns solvation numbers calculated using eqn. (2) by extrapolation to c\0, that is, at in–nite dilution. A (roughly) linear character of vs. was V 1 0 b1 0 M1m2 103bc1~1 observed over the entire concentration range in both alcohols.The solvation numbers determined in this way are very high compared to those of other electrolytic solutions. As an example: the hydration numbers of KCl in water, as well as in water»dioxane mixed solvents, do not exceed 611,20 and hardly depend on the experimental method and the method of calculation (for instance, the classical Pasynski method for aqueous solutions of KCl yields the hydration number nh\ The diÜerences in obtained by the two acoustic 5.4).21 ns methods originates from slightly diÜerent assumptions and are a good example of the well-known fact that any comparison of solvation numbers obtained by diÜerent methods may be misleading.The hydration numbers of light lanthanide perchlorates vary from 10 to 13,10 while those of heavy lanthanides are between 11.7 and 13.3 These numbers, obtained by the same method for aqueous lanthanide chlorides, vary from 16.8 to 19.4, depending on the lanthanide.22 The high solvation numbers obtained by us, sometimes even higher than those observed for aqueous solutions of the solutes under investigation, may indicate either strong electrostriction around the chlorine ions and/or that the range of the cation electrostrictive interaction is longer than the –rst solvation shell.The –rst explanation would mean that Pasynskiœs assumption of negligible eÜect of solvation of the anion is not true ; the second one implies that the acoustic method generally yields too high solvation numbers (not only the innershell molecules are compressed). Very high solvation numbers of lanthanide salts in non-aqueous solvents are not surprising : in our recent paper it was about equal to 25 for neodymium perchlorate in methanol.23 Neutron diÜraction studies of lanthanide perchlorates in heavy water also suggest high coordination numbers of the cations (8»9) and the strong radial orientation of water molecules around the ions is interpreted in terms of the high charge of the cations.24 Another striking observation is the fact that solvation numbers of hydrated lanthanide chlorides in iso-propanol, calculated from eqn.(1), are not concentration independent. The dependences of on salt concentration are shown in Fig. 3. ns The changing is excellent con–rmation of the above ns assumption that chloride anions enter the solvation sphere of lanthanide cations, lowering the overall charge of the solvated species (and thus causing a slight nonlinearity of the compressibility vs.concentration dependences). Recently, in our paper mentioned above, very similar behaviour was found for anhydrous in methanol and interpreted in terms of Nd(ClO4)3 50 New J.Chem., 1999, 47»52Fig. 3 Solvation numbers vs. concentration of hydrated praseodymium and cerium chlorides in iso-propanol calculated from eqn. (1). inclusion of the perchlorate ions into the solvation shell of the lanthanide cation.23 Table 2 shows clearly that, in general, the solvation numbers in methanol are higher than in iso-propanol. This is easy to understand in terms of steric reasons : molecules of methanol are smaller than these of iso-propanol.Another feature, the fact that anhydrous lanthanide chlorides in methanol have much lower than the hydrated ones, is not com- ns pletely clear and additional studies are needed. It is clear that water molecules interact more strongly than alcohol molecules with ions and should replace the latter in the solvation shells.On the other hand, the statistical eÜect of the excess of methanol in solution weakens this process. One must also consider the size of water molecules and their role in the easier exchange of chloride ions than by alcohol molecules. As a result, more solvent or water molecules are placed in the –rst solvation sphere and the M3` ion has also its second and higher shells of solvent molecules.This leads to an increase of the solvation number when compared to anhydrous salts. Moreover, interaction of water molecules with the solvent should also decrease compressibility, leading to increased overall solvation number. Our conductivity measurements strongly con–rm the conclusions drawn from the spectroscopic measurements. The systems under investigation cannot be described in terms of simple Debye»Hué ckel theory ; the reason for this could be changing charge and/or number of ions.This idea is consistent with the model of an equilibrium process in which chloride anions are entering the close environment of the lanthanide cation. Thus, it was necessary to assume that the lanthanide chlorides tested are weak electrolytes and that their dissociation could be described by the reaction LnCl3HLnCl3~n `]nCl~ or more precisely by LnCl3(solvent)m]solventHLnCl3~n(solvent)m`1 `]nCl~.In this case, it is easy to derive the equation linking the observed conductivity with molar concentration of the salt : ln K\ln K0] 1 n]1 ln Kdiss[ n n]1 ln c (3) Fig. 4 Plots of ln K vs. ln c for the systems under investigation.For symbols refer to Fig. 1. The thick solid lines are –tted to the data for lanthanide chlorides. where c is the molar concentration of the salt, is the Kdiss dissociation constant of the salt, is the limiting conductiv- K0 ity of the salt, that is extrapolated to in–nite dilution and K is the conductivity of a solution of molar concentration c. Eqn. (3) is a linear relation between the logarithms of conductivity and concentration, whose slope depends on the number of dissociated chloride ions and the intercept with the ln K axis depends on the dissociation constant of the salt.Fig. 4 presents the corresponding dependences for all the systems measured. It is striking that for lanthanide chlorides, independent of the solvent or crystalline water content, all the slopes are equal to [0.5 (within experimental error).Generally, these results con–rm very well our previous conclusions : all the lanthanide chlorides are weak electrolytes and only one chloride anion can be dissociated. The dissociation constants of the chlorides, both hydrated and anhydrous, are of the order of 0.25 in methanol and 6.3]10~5 in iso-propanol, almost independent of the lanthanide. Other chlorides behave diÜerently, most probably because of the formation of ionic aggregates ; it is interesting that lanthanides do not have such a tendency. Conclusions It seems evident that lanthanide chlorides are weak electrolytes in alcohols, much weaker in iso-propanol than in methanol.Steric eÜects can lead to the reduction of the coordination numbers of metal ions in iso-propanol. At low concentrations the –rst solvation sphere of lanthanide contains two chloride anions; at higher concentrations replacement of solvent molecule in the outer solvation sphere by a chloride anion is suggested.The conductivity investigations con–rm the above results. However, the stronger electrostrictive eÜect of hydrated lanthanide chlorides compared to the anhydrous ones remains open to question and needs further investigation.Acknowledgements support from the Polish Committee for Scienti–c Financial Research is acknowledged. New J. Chem., 1999, 47»52 51References 1 B. Keller, J. Legendziewicz and J. Glinski, Spectrochim. Acta, Part A, 1998, in the press. 2 B. Keller, Ph.D. Thesis, University of Poland, 1975.Wroc°aw, 3 J. Legendziewicz, K. Bukietyn8 ska, G. Oczko, S. Ernst and B. Jezowska-Trzebiatowska, Chem. Phys. L ett., 1980, 73, 576. 4 J. Legendziewicz, K. Bukietynska and G. Oczko, J. Inorg. Nucl. Chem., 1981, 43, 2393. 5 J. Legendziewicz, G. Oczko, B. Keller, W. Strek and B. Jezowska- Trzebiatowska, Bull. Pol. Acad. Sci., Chem., 1984, 32, 301. 6 B. Keller, K.Bukietynska and B. Jezowska-Trzebiatowska, Bull. Pol. Acad. Sci., Chem., 1976, 24, 763. 7 J. L. Ryan and C. K. Jorgensen, J. Phys. Chem., 1966, 70, 2845. 8 J. Legendziewicz, G. Oczko and B. Keller, J. Mol. Struct., 1984, 115, 421. 9 G. Oczko, J. Legendziewicz, B. Keller and B. Jezowska- Trzebiatowska, Spectrochim. Acta, Part A, 1989, 55, 945. 10 K. Bukietynska, B. Jezowska-Trzebiatowska and B. Keller, J. Inorg. Nucl. Chem., 1981, 43, 1065. 11 B. Jezowska-Trzebiatowska, S. Ernst, J. Legendziewicz and G. Oczko, Bull. Pol. Acad. Sci., Chem., 1978, 26, 805. 12 S. Ernst, J. Glinski and B. Jezowska-Trzebiatowska, Acta Phys. Pol., Part A, 1979, 55, 501. 13 G. R. Choppin, J. Alloys Compd., 1997, 249, 9. 14 J. Zhongsheng et al., Chem. J. Chin. Univ., 1989, 6, 735. 15 M. Schaé fer, R. Herbst-Irmer, U. Groth and T. Kohler, Acta Crystallogr., Sect. C, 1991, 50, 1256. 16 See, for example: Y. C. Wu, in Structure of W ater and Aqueous Solutions, ed. W. A. P. Luck, Verlag Chemie, Marburg, 1973, ch. II.6, p. 189. 17 A. C. Pasynski, Zh. Fiz. Khim., 1938, 11, 608; A. C. Pasynski, Zh. Fiz. Khim., 1946, 20, 98. 18 S. Ernst and B. Jezowska-Trzebiatowska, J. Phys. Chem., 1975, 79, 2113. 19 S. Ernst and J. Glinski, Mater. Sci., 1977, 3, 69. 20 E. R. Nightingale, Jr., J. Phys. Chem., 1959, 63, 1381. 21 D. S. Allam and W. H. Lee, J. Chem. Soc., 1966, 5, 5. 22 G. Oczko, J. Legendziewicz, B. Jezowska-Trzebiatowska and S. Ernst, Bull. Pol. Acad. Sci., Chem., 1980, 28, 793. In the case of chlorides the authors suggest that the anion solvation number (+6) should be subtracted from the above numbers to obtain those of cations. 23 B. Keller, J. Glinski, K. Orzechowski and J. Legendziewicz, New J. Chem., 1997, 21, 329. 24 C. Cossy, L. Helm, D. H. Powell and A. E. Merbach, New J. Chem., 1995, 19, 27. Paper 8/08331B 52 New J. Chem., 1999, 47»52

ISSN:1144-0546    

DOI:10.1039/a808331b

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Homoleptic and heteroleptic iron(II) and ruthenium(II) complexes of novel 4º-nitro-2,2º : 6º,2/-terpyridines and 4º-amino-2,2º : 6º,2/-terpyridines Reza-Ali Fallahpour,* Markus Neuburger and Magareta Zehnder Institut Anorganische Chemie, Basel, Spitalstrasse 51, CH-4056 Basel, f ué r Universitaé t Switzerland in Germany, 23rd August 1998, Accepted 19th October 1998 Receiøed Goé ttingen, Several series of 4-nitro-6-bromo-2,2@-bipyridines and of symmetrical and unsymmetrical 4@-nitro- and 4@-amino- 2,2@ : 6@,2A-terpyridines have been prepared. The structure of 4@-amino-2,2@ : 6@,2A-terpyridine has been determined by X-ray structure analysis.The unusual internal angles of the two terminal rings with respect to the central one have been rationalized in terms of hydrogen bonding between the amino protons and nitrogen atoms of the terminal pyridine rings. The new ligands have been used in the preparation of homo- and heteroleptic ruthenium(II) and iron(II) complexes and their chemical and electrochemical properties have been investigated.The synthesis and properties of a heteroleptic iron(II) complex with both 4@-nitro- and 4@-amino-2,2@ : 6@,2A-terpyridines are reported for the –rst time.Since the metal-bonded 2,2@ : 6@,2A-terpyridines (tpy) with spacers at C(4@) provide a means of directionality, and thus a means of linear communication, the functionalization of tpy at this position has been of interest to chemists. A number of substituents can be directly inserted by the Kroé hnke methodology. 1 Some 2,2@ : 6@,2A-terpyridines with functionalities directly attached to C(4@) such as 4@-hydroxy-2,2@ : 6@,2A-terpyridine, 2 4@-chloro-2,2@ : 6@,2A-terpyridine,2 4@-bromo-2,2@ : 6@,2Aterpyridine, 3 4@-methylthio-2,2@ : 6@,2A-terpyridine4 and 4@- methanesulfonyl-2,2@ : 6@,2A-terpyridine4 have been reported and have been used in the development of the chemistry of multinucleated complexes.5 The only reported example of a nitrogen-containing 2,2@ : 6@,2A-terpyridine is 4@-dimethylamino- 2,2@ : 6@,2A-terpyridine.6 The literature methods have not permitted the simultaneous introduction of functionalities (substituents) at C(4@) and at the terminal pyridine rings. We have already published the synthesis of dimethylsubstituted 4@-ethoxy- and 4@-hydroxy-2,2@ : 6@,2A-terpyridines.7 Now we report the synthesis of such 2,2@ : 6@,2A-terpyridines bearing nitro and amino groups at C(4@), as well as methyl groups at the terminal pyridine rings, which are precursors to new ligands and heterocycles.8 Results and discussion In studies of functionalized 2,2@ : 6@,2A-terpyridines, we have become interested in 2,2@ : 6@,2A-terpyridines with amino and nitro groups that are directly linked to C(4@) of 2,2@ : 6@,2Aterpyridine, motivated by the electron-donating and -withdrawing properties, respectively, of these groups.Here we report the syntheses of both ligands and of the homo- and heteroleptic iron(II) and ruthenium(II) complexes, whose electronic and electrochemical properties were then compared. Synthesis of tpy ligands The Stille reaction,9 which consists of the reaction of stannyl and bromo compounds in the presence of a catalytic amount of palladium(0), has found wide application in the synthesis of aromatic and heterocyclic compounds.The advantage of this method is that many functionalities, such as nitro groups, do not react under the reaction conditions. The key compound of the Stille coupling here was 2,6- dibromo-4-nitropyridine, 2 (Scheme 1).Commercially available 2,6-dibromopyridine was converted to 2,6- dibromopyridine-N-oxide, which was then reacted with nitric acid in sulfuric acid to give 2,6-dibromo-4-nitropyridine-Noxide. 10 Subsequent deoxygenation with phosphorus trichloride in chloroform11 produced 2 as a yellow microcrystalline compound in 43% overall yield. 2-Bromopyridine, 3, was converted to tributyl(pyridin-2-yl) stannane, 4, upon reaction with butyllithium and tributyltin chloride in tetrahydrofuran.12 Compound 2 was reacted with 1 equiv.of 4 in the presence of 0.01 equiv. of for Pd(PPh3)4 16 h at re—ux in toluene to give 4-nitro-6-bromo-2,2@-bipyridine, 7, in 60% yield as yellow crystals. 2-Bromo-5-methylpyridine, 5,13 was converted to tributyl(5-methylpyridin-2-yl) stannane, 6, in the same manner as in the synthesis of 4.When 2 was reacted under the same conditions with 1 mole equiv. of 6 we obtained 4-nitro-6-bromo-5@-methyl-2,2@-bipyridine, 11, in 65% yield as a pale yellow crystalline solid. However, if 2 was reacted with two equivalents of 4 in the presence of the catalyst under the same conditions, ligand 8 was directly obtained in 68% yield.Alternatively, 2 was reacted with two equivalents of 6 under the same conditions to give 12 in 64% yield. The unsymmetrical tpy ligand 10 was obtained in good yield upon reaction in toluene of bipyridines 7 or 11 with the stannanes 4 or 6, respectively, in the presence of 0.01 equivalent of Pd(PPh3)4 . A doublet in the 1H NMR spectra of bipyridines 7 and 11 was observed at d 9.10 and 9.08, respectively, due to protons H3 and an additional doublet due to the protons H5 was observed at d 8.05 and 8.16 at similar shift to protons H3 of 2 (Table 1).The two symmetrical unsubstituted and substituted terpyridines 8 and 12, respectively, and the unsymmetrical terpyridine 10 are interesting target molecules. In the 1H NMR spectra of 8, 10 and 12 we observed a singlet due to protons at d 9.16, 9.12 and 9.08, respectively, which is fully consis- H3{ tent with the inductive eÜect of the methyl groups. In the unsymmetrical terpyridine 10, while each proton was observed as a separate signal, the proton was also observed at d H5{ 9.12, in other words, the protons and are identical H3{ H5{ New J.Chem., 1999, 53»61 53Scheme 1 (a) 61 °C, 20 h, 73%; (b) THF, [78 °C, n-BuLi, 1 h, 95%; (c) as (b), 97%; (d) 4 (1 equiv.), PCl3 , CHCl3, Bu3 SnCl, Pd(PPh3)4 (0.01 equiv.), toluene, 110 °C, 16 h, 60%; (e) 6 (1 equiv.), as (d), 65%; ( f ) 4 (2 equiv.), as (d), 68%; (g) 6 (2 equiv.), as (d), 64%; (h) 6 (1 equiv.), as (d), 96%; (i) 4 (1 equiv.), as (d), 96%; ( j) as (d), 81%; (k) with 6, as (d), 70%; (l) Pd/C (10%), EtOH, 78 °C, 1 h, 76%; (m) as (l), 69%.H2NNH2 …H2O, (Table 1). All the data of elemental analysis and mass spectra are consistent with the proposed structures. The two nitroterpyridines 8 and 12 had been readily reduced with hydrazine hydrate in the presence of palladium on charcoal in ethanol.14 In the IR spectra of 9 and 13, no bands assigned to nitro groups were observed, but bands attributed to amino groups were observed at about 3400 cm~1.All the data are in accord with the proposed structures. In conclusion, this method permits the synthesis of the functionalised 2,2@-bipyridines 7 and 11, as well as the novel nitroterpyridines 8, 10 and 12 and the aminoterpyridines 9 and 13, which are precursors to new heterocycles and oligopyridines that are under current investigation. Crystal structure of 4º-amino-2,2º : 6º,2/-terpyridine The X-ray crystal structure of 4@-amino-2,2@ : 6@,2A-terpyridine, 9, con–rms the proposed structure and is presented in Fig. 1. The three pyridine rings exhibit transoid con–gurations about the interannular CwC bonds, as it had previously been reported in 4@-dimethylamino-2,2@ : 6@,2A-terpyridine.6 This con–guration minimizes electrostatic interactions between the nitrogen lone pairs and also the van der Waals interactions between the meta protons.The interannular CwC bonds [C(5)wC(6), 1.490(3) are comparable with those of 4@- Aé ] dimethylamino-2,2@ : 6@,2A-terpyridine [1.492(4) In other Aé ].6 terpyridine derivatives, the three pyridine rings are not coplanar and the interplanar angles of the two terminal rings with Fig. 1 Crystal structure of the terpyridine ligand 9. 54 New J. Chem., 1999, 53»61Table 1 1H NMR spectroscopic data for 2,2@-bipyridines and 4@-substituted- 2,2@ : 6@,2A-terpyridines in solutions CDCl3 H3 H4 H5 H6 H3{ H4{ H5{ H6{ H3_ H4_ H5_ H6_ Others 7 9.10 8.05 8.45 8.76 8.76 8.73 d d d ddd ddd d J 1.45 1.50 8.30 8.30 8.30 7.80 7.80 7.80 1.95 1.95 8 8.64 7.91 7.42 8.76 9.16 d ddd ddd d s J 7.80 8.30 8.30 7.80 7.80 7.80 1.95 1.95 9 8.60 7.84 7.32 8.67 7.75 4.33 d ddd ddd d s s J 7.80 8.30 8.30 7.80 NH2 7.80 7.80 1.95 1.95 10 8.52 7.71 8.58 9.12 9.12 8.76 7.91 7.42 8.63 2.46 d dd d s s d ddd ddd d s J 7.80 7.80 1.45 7.80 8.30 8.30 7.80 CH3 1.95 7.80 7.80 1.95 1.95 11 9.08 8.16 8.34 8.67 8.55 2.44 d d d dd d s J 1.45 1.50 7.80 7.80 1.50 CH3 1.95 12 8.51 7.70 8.57 9.08 2.45 d dd d s s J 8.30 8.30 1.50 CH3 1.50 13 8.07 7.72 8.58 7.94 8.75 d dd bs s s J 8.30 8.30 NH2 1.45 the central ring are similar and vary from 5.7° (4@-phenyl- 2,2@ : 6@,2A-terpyridine),15 7.4° (4@-dimethylamino-2,2@ : 6@,2Aterpyridine) 6 to 10.9° (6,6A-dibromo-4@-phenyl-2,2@ : 6@,2A-terpyridine). 16 In the aminoterpyridine 9, however, the interplanar angles of the two terminal rings with the central ring are 11.23° and 20.68°, respectively.This deviation from the expected angles is due to intermolecular hydrogen bond formation. Fig. 2 illustrates that a hydrogen-bonded network extends through the lattice involving amino protons and nitrogen atoms of the terminal pyridine rings. The distances N(1)wH(1) of 2.271 and N(3)wH(2) of 2.333 are in accord Aé Aé with the known values.17,18 The N(4)wC(8) distance of 1.364(3) strongly suggests an Aé sp2 character for the nitrogen atom and a high degree of pconjugation of the amino group with the aromatic ring (Table 2).Preparation and characterization of homo- and heteroleptic iron(II) complexes 2,2@ : 6@,2A-Terpyridines react readily with iron(II) salts at room temperature to yield purple metal complexes; however, 6,6A-disubstituted-2,2@ : 6@,2A-terpyridines react with iron(II) salts only at elevated temperature to give the metal complexes.Nitroterpyridines 8, 10 and 12 were reacted with excess in ethanol at room temperature to give the blue FeCl2 … 4H2O mononucleated iron complexes 14»16 (Scheme 2).In the 1H NMR spectra of complexes 14»16, the resonance was H3{ observed as a singlet at d 9.64 (14), 9.57 (15) and 9.53 (16), which exactly correlates with the electron-releasing methyl groups. Interestingly, in the unsymmetrical complex 15, the signal due to the two protons adjacent to the nitro group were split and were observed at d 9.57 and 9.59 (Table 3). (H3{) (H5{) The nitroterpyridine iron(II) complexes 14 and 16 were easily reduced to the aminoterpyridine iron(II) complexes 17 and 18, respectively, in ethanol in the presence of iron and hydrochloric acid.This reaction is easy to follow due to the colour change from blue to purple. Alternatively, the isolated aminoterpyridines were reacted directly with to FeCl2 . 4H2O give 17 and 18. In the 1H NMR spectra of the complexes the signal was observed as a singlet at d 8.07 (17), or 7.97 (18), H3{ which also correlates with the electron-releasing methyl groups. Our attempt to synthesize the heteroleptic iron(II) complex 19 was successful. Ten milligrams of each ligand 8 and 9 were dissolved in 3 ml ethanol and was added in FeCl2 . 4H2O excess (Scheme 3). The statistical distribution of the two homoleptic complexes 14, 17 and the heteroleptic complex 19 should be 1 : 1 : 2.The three complexes (with chloride as counter ion) were successfully separated by chromatography on aluminium oxide with an eluting solution of acetonitrile» water»ammonia (9 : 1 : 0.2). The purple complex 14 was isolated as the –rst fraction followed by the dark blue complex 19. The blue complex 17 was isolated as the last fraction.This is the –rst example of a heteroleptic iron(II) complex that has been separated from the two homoleptic ones by chromatography. In the electronic spectrum the metal-to-ligand charge transfer (MLCT) absorption of 19 was shifted by about 18 nm to lower energy and was observed at 623 nm, compared to the MLCT absorption of 14 at 605 nm (Fig. 3). The 1H NMR spectrum of the heteroleptic complex 19 shows signi–cant shifts of some signals (Table 3) when compared to the homoleptic complexes 14 and 17. The proton H3{ of the aminoterpyridine moiety of 19 was shifted to low –eld and observed at d 8.23 (*d\0.22) while proton of the H3{ nitroterpyridine moiety was shifted to high –eld and observed at d 9.54 (*d\0.10). More dramatically, the amino protons were shifted to low –eld at d 6.55 (*d\0.42).All other signals belonging to the aminoterpyridine moiety were shifted to low –eld while the protons of the nitroterpyridine moiety were shifted to high –eld. All iron(II) complexes are electrochemically active in acetonitrile solution, each exhibiting a wave corresponding to the New J. Chem., 1999, 53»61 55Table 2 Selected bond lengths and angles (°) for 9 (”) N(1)wC(1) 1.332(3) N(2)wC(10) 1.349(3) N(1)wC(5) 1.348(3) C(6)wC(7) 1.374(3) C(1)wC(2) 1.370(4) C(7)wC(8) 1.395(3) C(2)wC(3) 1.370(4) C(8)wC(9) 1.393(3) C(3)wC(4) 1.381(3) C(8)wN(4) 1.364(3) C(4)wC(5) 1.380(3) N(3)wC(11) 1.345(3) C(5)wC(6) 1.490(3) N(3)wC(15) 1.334(4) N(2)wC(6) 1.347(3) C(1)wN(1)wC(5) 117.2(2) C(5)wC(6)wC(7) 120.0(2) N(1)wC(1)wC(2) 124.2(3) N(2)wC(6)wC(7) 123.9(2) C(1)wC(2)wC(3) 118.3(2) C(7)wC(8)wN(4) 120.9(2) C(2)wC(3)wC(4) 119.0(2) C(9)wC(8)wN(4) 121.8(2) C(3)wC(4)wC(5) 119.3(2) N(2)wC(10)wC(9) 123.5(2) N(1)wC(5)wC(4) 122.0(2) N(2)wC(10)wC(11) 115.5(2) N(1)wC(5)wC(6) 116.4(2) C(11)wN(3)wC(15) 116.7(3) C(4)wC(5)wC(6) 121.6(2) C(10)wC(11)wN(3) 117.0(2) C(6)wN(2)wC(10) 116.3(2) N(3)wC(11)wC(12) 121.8(2) C(5)wC(6)wN(2) 116.0(2) N(3)wC(15)wC(14) 124.5(3) Fig. 2 Hydrogen bonds among the terpyridine ligand 9 in the crystal lattice. The distances are : N(1)wH(1), 2.27 and N(3)wH(2), 2.333 Aé Aé . FeII/FeIII process. As expected, the introduction of the electron-releasing amino and the strong electron-withdrawing nitro substituents has a dramatic eÜect upon the redox couple (Table 4).The redox potential of the aminoterpyridine iron(II) complexes 17 and 18 ranges between 0.308»0.356 V (vs. ferrocene/ferrocenium internal reference). However, the potential of the nitropyridine iron(II) complexes 14»16 ranges between 0.900»0.963 V. This increase is due to the electronwithdrawing nitro group and is –netuned by the methyl groups: the redox potential decreases in the series 14 (unsubstituted)[15 (one methyl group)[16 (two methyl groups).This means that, among a series of substituted tpy ligands, we have synthesized the strongest electronwithdrawing ligand 8 and the corresponding metal complex 14 (E°\0.963 V) with respect to [Fe(MeSO2-tpy)2][PF6]2 (E°\0.904 V). The potential diÜerence between these electron-donor and electron-acceptor complexes ranges between 0.600»0.650 V.The heteroleptic iron(II) complex 19 exhibits a potential of 0.654 V, which is about the average of the potentials of the homoleptic iron(II) complexes 14 and 17. Preparation and characterization of homo- and heteroleptic ruthenium(II) complexes Ruthenium(II) complexes of terpyridines are of interest because of their photochemical and photophysical properties. 19,20 We chose the formation of kinetically inert ruthenium(II) complexes to exemplify the application of these new ligands. The advantage of ruthenium(II) complexes is that not only the homoleptic but also the heteroleptic ones can be formed in good yields (Scheme 4).6 Scheme 2 (a) EtOH, 25 °C, 5 min, 14 (98%), 15 (96%), 16 (94%); (b) Fe powder, conc. HCl, 78 °C, 15 min, 17 (88%), FeCl2 … 4H2O, EtOH»H2O, 18 (82%). 56 New J. Chem., 1999, 53»61Table 3 1H NMR spectroscopic data for acetonitrile solutions of metal(II) complexes X-tpy Y-typ 3 4 5 6 3@ 3A 4A 5A 6A 3” Others 14 8.72 7.97 7.15 7.12 9.64 d ddd ddd d s 15a 8.59 7.04 6.89 9.57 8.69 7.95 7.12 7.77 9.59 2.16 d d bs s d ddd ddd d s s CH3 16 8.57 7.76 6.81 9.53 2.13 d d s s s CH3 17 8.23 7.80 7.05 7.24 8.07 6.13 d ddd ddd d s s NH2 18 8.10 7.80 6.96 7.97 6.01 d ddd s s bs NH2 2.16 s CH3 19 8.29 7.83 6.97 7.26 8.23 8.64 7.92 7.20 6.99 9.54 6.55 d ddd ddd d s d ddd ddd d s s NH2 22 8.73 7.98 7.24 7.37 9.47 d ddd ddd d s 23 8.24 7.83 7.11 7.40 7.91 5.84 d ddd ddd d s bs NH2 24 8.27 7.86 7.33 7.60 7.96 8.70 7.99 7.03 7.13 9.40 6.06 d ddd ddd d s d ddd ddd d s bs NH2 25 8.46 7.85 7.17 7.43 7.91 8.46 7.88 7.08 7.30 8.26 3.45 d ddd ddd d s d ddd ddd d s s NMe2 26 8.25 7.84 7.17 7.48 7.93 8.45 7.88 7.08 7.31 8.30 5.95 d ddd ddd d s d ddd ddd d s bs NH2 appears at d 9.59 and the coupling constants are as follows : J\7.80 (d) ; J\8.30, 7.80, 1.95 (ddd).a H5{ Table 4 Electrode potentials (E°/V) in acetonitrile solutions (vs. ferrocene/ferrocenium) 1st 2nd 3rd M2`/M3` reduction reduction reduction 14 0.963 [0.825 [0.990a [1.490 15 0.931 [0.861a [1.299 16 0.900 [1.011 [1.566 17 0.356 [1.271 [2.284 18 0.308 [2.008 19 0.654 [0.939 [2.114 22 1.114 [0.861 [1.289 23 0.474 [1.266 24 0.740 [0.997a [1.305 25 0.485 [1.270 [1.825a [2.047a 26 0.564 a Irreversible. Initially, hydrated was reacted with one equivalent of RuCl3 the free ligands 8 or 9 at re—ux in ethanol or methanol to obtain the insoluble dark blue or brown ruthenium(III) complexes 20 and 21, respectively.The ruthenium(III) salts 20 and 21 were then reacted at re—ux with one equivalent of the corresponding ligand, 8 or 9, respectively, in Methanol in the presence of the reducing agent N-ethylmorpholine, to obtain the homoleptic ruthenium(II) complexes 22 and 23, respectively.Alternatively, a mixture of hydrated ruthenium trichloride (1 equiv.) and either 8 (2 equiv.) or 9 (2 equiv.) in 5 ml ethylene glycol was heated in a microwave oven for 10 min to yield the homoleptic ruthenium(II) complexes 22 and 23, respectively, as red-orange solutions. The heteroleptic ruthenium(II) complex 24 was obtained by reaction of ruthenium(III) salts 20 and 21 with one equivalent Scheme 3 (a) EtOH, 25 °C, 5 min, 19 (29%).New J. Chem., 1999, 53»61 57Fig. 3 UV/VIS spectra of the homoleptic iron(II) complexes 17 (a), 14 (b) and the heteroleptic Fe(II) complex 19 (c) in acetonitrile solutions. of the complementary ligands 9 or 8, respectively, in methanol in the presence of N-ethylmorpholine. The ruthenium(II) complexes 22»24 were precipitated as their red-orange hexa- —uorophosphate salts and were puri–ed by chromatography followed by recrystallization.In the reactions to obtain the heteroleptic ruthenium(II) complex, a byproduct was identi–ed in which the nitro group was reduced to a hydroxylamine. In order to con–rm this, we synthesized 25 by reaction of 20 [(Me2N-tpy)Ru(HOHN-tpy)] with 4@-dimethylamino-2,2@ : 6@,2A-terpyridine6 in methanol at re—ux.In this series, when ruthenium(III) salt 20 and 9 or ruthenium(III) salt 21 and 8 were reacted in re—uxing methanol or in ethylene glycol under microwave irradiation we obtained the ruthenium(II) complex 26 in good yields. We believe that some trace of metallic ruthenium in the protic solvents is involved in the reduction of the nitro group to the hydroxylamine. 21 The homo- and heteroleptic ruthenium(II) complexes 22»26 exhibit analogous NMR (Table 3) and electrochemical (Table 4) properties as the iron(II) complexes. In conclusion, the Stille coupling reaction has been used for the synthesis of novel 4-nitro-6-bromo-2,2@-bipyridines, 4@- nitro-2,2@ : 6@,2A-terpyridines and 4@-amino-2,2@ : 6@,2A-terpyridines that are precursors for new heterocycles and oligopyridines.The homoleptic and heteroleptic iron(II) and ruthenium(II) complexes have been investigated. In particular, the heteroleptic iron(II) complex is of interest as it can be incorporated into multinucleated metal complexes that are under current investigation. Acknowledgements would like to thank Professor Edwin C.Constable for his We generous support and Professor Catherine Housecroft for critical reading of the manuscript. We should also like to thank the Schweizerischer Nationalfonds zur Foé rderung der wissenschaftlichen Forschung and the University of Basel for support. Experimental All reagents were used as supplied. Silica gel (0.060»0.200 mm) was obtained from Chemie Uetikon and aluminium oxide (type 507 C neutral ; 100»125 mesh) from Fluka.Melting points were measured on a Bué chi 535 apparatus and are not corrected. IR spectra were recorded on a Mattson Genesis Fourier transform spectrophotometer with samples in compressed KBr discs. UV/VIS spectra were measured on a Perkin Elmer Lambda 19. Proton and carbon NMR spectra Scheme 4 (a) EtOH, 78 °C, 2 h, 70%; (b) N-ethylmorpholine, EtOH, 78 °C, 2 h, 22 (33%), 23 (92%), 24 (80%); (c) as (b), 20, RuCl3 … 3H2O, (77%), 20, 9 or 21, 8]26 (48%).Me2N-tpy]25 58 New J. Chem., 1999, 53»61were recorded on a Bruker AM 250 spectrometer and referenced against Matrix-assisted laser desorption ioniza- Me4Si. tion time-of-—ight (MALDI-TOF) spectra were recorded using a PerPespective Biosystems Voyagers-RP Biospectrometry Workstation.Electrochemical measurements were performed with an Ecochemie Autolab PGSTAT 20 potentiostat. Crystal structure determination of 9 Data collection was carried out on a four-circle Enraf- Nonius CAD4 diÜractometer using monochromated Mo-Ka radiation (k\0.71069 T \293 K. Details of the crystal Aé ) ; parameters, data collection and re–nement are listed in Table 5.The structure was solved by direct methods using the program SIR92.22 Anisotropic least squares re–nement was carried out on all non-hydrogen atoms using the program CRYSTALS.23 Scattering factors were taken from the International Tables for X-Ray Crystallography.24 CCDC reference number 440/072. Syntheses 2,6-Dibromo-4-nitropyridine, 2. 2,6-Dibromo-4-nitropyridine- N-oxide, 1 (5.41 g, 0.018 mol), was suspended in 25 ml chloroform and then cooled to 0 °C.Phosphorus trichloride (7.45 g, 4.74 mol) was gradually added and then heated for 20 h at 100 °C. Upon cooling to room temperature the yellow solution was poured into ice water and the crystals were –ltered. The pale yellow crystals were puri–ed by chromatography on aluminium oxide using dichloromethane as solvent.The yield was 3.74 g (73%). mp 125 °C. IR (KBr): 3090w, 1544s, 1348s, 1297m, 1169m, 1141m, 1036m, 881m, 735s. 1H NMR d 8.20 (s, 2H). 13C NMR d 142.37, (CDCl3) : (CDCl3) : 120.50, 116.80. (Found: C, 21.03 ; H, 0.78 ; N, 9.72 ; Br, 57.60%. Calcd for C, 21.30 ; H, 0.72 ; N, 9.94 ; Br, C5H2Br2N2O2 : 56.69%). General procedure for Stille coupling reactions. Bromo compound (1 mol), stannanyl compound (1 or 2 mole equiv.) and (0.01 or 0.02 mole equiv.) were heated under nitro- Pd(PPh3)4 gen in 50 ml toluene for 16 h.After cooling to room temperature, 20 ml saturated ammonium chloride was added and Table 5 Crystal data and data collction parameters for tepyridine ligand 9 Formula C15H12N4 M 248.29 Crystal Monoclinic Space group C2/c a/” 13.807(1) b/” 11.784(1) c/” 16.444(2) a/° 90 b/° 109.818(7) c/° 90 U/”3 2517.0(4) Z 8 F(000) 1040 Dc/g cm~3 1.31 k/mm~1 0.08 Crystal size/mm 0.08]0.18]0.32 T /K 293 Radiation Mo Ka (k\0.710 69) Scan type x/20 0'/° 26.32 Re—ections collected 3736 Independent re—ections 2179 Re—ections in re–nement 1404 Number of variables 180 Final R 0.0545 Final Rw 0.0610 the organic phase separated.The aqueous phase was extracted with toluene (3]20 ml). The combined organic phases were dried and the solvent was removed. Concentrated (MgSO4) hydrochloric acid (30 ml) was added to the residue, followed by extraction with dichloromethane (3]30 ml). The aqueous phase was cautiously neutralized by solid sodium hydroxide. The oligopyridines were then extracted with dichloromethane (3]30 ml) and dried The solvent was removed fol- (MgSO4).lowed by puri–cation on silica gel with dichloromethane. 4-Nitro-6-bromo-2,2º-bipyridine, 7. 2 (0.40 g, 1.42 mmol), tributyl(pyridin-2-yl)stannane, 4 (0.542 g, 1.42 mmol), and (0.020 g, 0.01 mole equiv.) gave 7 (0.240 g, 60%) as Pd(PPh3)4 yellow crystals. mp 78 °C. IR (KBr): 1531s, 1380m, 1347s, 1282m, 1145m, 797m, 748s, 736m.UV/VIS (CH3CN): kmax 280, 326; 302 nm. 13C NMR d 159.56, 156.22, kmin (CDCl3) : 155.66, 149.57, 142.33, 138.18, 125.37, 121.66, 116.91, 112.55. MS (MALDI-TOF): m/z 280. (Found: C, 43.02 ; H, 2.37 ; N, 14.83%. Calcd for C, 42.88 ; H, 2.16 ; N, C11H8BrN3O2: 15.00%). 4º-Nitro-2,2º : 6º,2ºº-terpyridine, 8. 2 (1.17 g, 4.15 mmol), 4 (3.18 g, 8.30 mmol, 2 equiv.) and (0.100 g, 0.02 mole Pd(PPh3)4 equiv.) gave 8 (0.907 g, 68%) as pale yellow needles.mp 177 °C. IR (KBr): 1561m, 1531s, 1467w, 1400m, 1358s, 1338m, 1268w, 1058w, 797w, 748s. UV/VIS 279, 345; (CH3CN): kmax 305 nm. 13C NMR d 158.44, 156.33, 154.05, kmin (CDCl3) : 149.47, 136.97, 124.77, 121.33, 113.33. MS (MALDI-TOF): m/z 278. (Found: C, 64.51 ; H, 3.58 ; N, 20.09%. Calcd for C, 64.74 ; H, 3.62 ; N, 20.13%).C15H10N4O2 : 4º-Amino-2,2º : 6º,2ºº-terpyridine, 9. Under nitrogen, 8 (0.100 g, 0.36 mmol) was heated under re—ux for 1 h in 30 ml ethanol in the presence of 0.100 g of 10% palladium on charcoal. Hydrazine hydrate (4 ml, 95%) was gradually added. TLC control of the solution after 2 min showed only a purple colour upon reaction with an iron(II) solution and no trace of a blue colour, which would indicate the nitropyridine. The resulting solution was –ltered and washed with 30 ml dichloromethane. The solvents were removed, 20 ml water was added and extracted with dichloromethane (3]30 ml).The combined organic phases were then dried –l- (MgSO4), tered and dichloromethane was removed. Chromatographic separation on aluminium oxide with dichloromethane»ethyl acetate (1 : 2) followed by recrystallization from ethanol»ethyl acetate (4 : 1) gave 9 (0.070 g, 76%) as colourless crystals. mp 179»180 °C.IR (KBr): 3226m, 1652s, 1611m, 1586s, 1564s, 1474m, 1458m, 1416m, 987m, 790m. 13C NMR d (CDCl3) : 156.52, 156.26, 154.55, 148.89, 136.76, 123.61, 121.30, 106.77. MS (MALDI-TOF): m/z 248. (Found: C, 72.01 ; H, 4.93 ; N, 22.55%.Calcd for C, 72.56 ; H, 4.87 ; N, 22.57%). C15H12N4: 5-Methyl-4º-nitro-2,2º : 6º,2ºº-terpyridine, 10. From 4-nitro- 6-bromo-5@-methyl-2,2@-bipyridine, 11 (0.150 g, 0.508 mmol), 4 (0.195 g, 0.508 mmol) and (0.010 g, 0.01 mole Pd(PPh3)4 equiv.), we obtained 10 (0.120 g, 81%) as yellow crystals. Upon reaction of 7 (0.100 g, 0.357 mmol), tributyl(5-methylpyridin-2- yl)stannane, 6 (0.150 g, 0.357 mmol), and (0.010 g, Pd(PPh3)4 0.01 mole equiv.), 10 (0.100 g, 96%) was obtained as yellow crystals.mp 222»3 °C. IR (KBr): 1559m, 1536s, 1407m, 1360s, 1265m, 745m. UV/VIS 280, 345; 311 nm. (CH3CN): kmax kmin 13C NMR d 158.68, 158.44, 156.43, 154.27, 151.68, (CDCl3) : 150.03, 149.54, 137.58, 137.08, 134.91, 124.82, 121.45, 120.99, 113.16, 113.11. MS (MALDI-TOF): m/z 292.(Found: C, 66.06 ; H, 3.99 ; N, 19.73%. Calcd for C, 65.75 ; C16H12N4O2 : H, 4.14 ; N, 19.17%). 4-Nitro-6-bromo-5º-methyl-2,2º-bipyridine, 11. 2 (0.500 g, 1.77 mmol), 6 (0.710 g, 1.77 mmol) and g, Pd(PPh3)4 (0.020 New J. Chem., 1999, 53»61 590.01 mole equiv.) gave 11 (0.340 g, 65%) as pale yellow crystals. mp 120 °C. IR (KBr): 1530s, 1482m, 1403m, 1349s, 1140m, 734m.UV/VIS 285, 333; 308 nm. (CH3CN): kmax kmin 13C NMR d 155.69, 150.14, 150.09, 150.01, 142.28, (CDCl3) : 137.61, 135.62, 121.33, 120.12, 112.53, 18.49. MS (MALDITOF): m/z 294. (Found: C, 45.29 ; H, 2.87 ; N, 14.83%. Calcd for C, 44.92 ; H, 2.74 ; N, 14.29%). C11H8BrN3O2 : 4º-Nitro-5,5ºº-dimethyl-2,2º : 6º,2ºº-terpyridine, 12. 2 (1.15 g, 4.08 mmol), 6 (3.26 g, 8.16 mmol, 2 mole equiv.) and g, 0.02 mole equiv.) gave 12 (0.800 g, 64%) Pd(PPh3)4 (0.100 as a pale yellow solid.mp 104 °C. IR (KBr): 1530s, 1384s, 1369s, 1358s, 1294m, 1260m, 739m. UV/VIS (CH3CN): kmax 285, 330; 310 nm. 13C NMR d 158.53, 150.37, kmin (CDCl3) : 149.98, 137.46, 134.76, 124.58, 120.91, 112.70, 18.54. MS (MALDI-TOF): m/z 306. (Found: C, 66.76 ; H, 3.59 ; N, 18.97%. Calcd for C, 66.66 ; H, 4.61 ; N, C17H14N4O2 : 18.29%). 4º-Amino-5,5ºº-dimethyl-2,2º : 6º,2ºº-terpyridine, 13. Under the same conditions as for the reduction of 9 we obtained 13 (0.075 g, 69%) from 12 (0.120 g, 0.392 mmol) as colourless crystals. mp 203 °C. IR (KBr): 3318m, 1655s, 1638m, 1600s, 1574m, 1509m, 1473m, 1071m, 1033m, 1011m, 553m. UV/VIS 269 nm. 13C NMR d 150.10, 150.02, (CH3CN): kmax (CDCl3) : 145.03, 144.02, 138.49, 136.77, 121.29, 105.45.MS (MALDITOF): m/z 276. (Found: C, 74.02 ; H, 5.93 ; N, 20.75%. Calcd for C, 73.89 ; H, 5.85 ; N, 20.27%). C17H16N4: Alternatively, both 4@-aminoterpyridines 9 and 13 were prepared by the cleavage of the corresponding iron(II) complexes. The iron(II) complexes 17 and 18 (0.100 g of each were dissolved in 1 : 1 water»acetonitrile (30 ml) to which potassium hydroxide (0.20 g) has been added.Hydrogen peroxide solution (30%) was added dropwise, while the mixture was stirred at room temperature, until all of the complex had been oxidatively cleaved, giving a brown suspension with no residual purple colour. This suspension was collected on Celite and washed with dichloromethane (30 ml) and methanol (20 ml) to dissolve any precipitated ligands.The solvents were removed and separation by chromatography was carried out as described above. 9 (0.040 g, 66%) and 13 (0.042 g, 67%) were obtained in good yields. General procedure for the synthesis of iron(II) complexes of nitroterpyridines. Nitroterpyridines 8, 10 or 12 were dissolved in 5 ml ethanol and excess iron(II) chloride tetrahydrate was added to yield the blue complexes.The complexes were –ltered over Celite and washed with 50 ml water. The resulting iron(II) complexes were precipitated as their hexa- —uorophosphate salts by the addition of methanolic ammonium hexa—uorophosphate. The complexes were –ltered over Celite, washed with 30 ml water, followed by diethyl ether and then dried. The complexes were then dissolved in acetonitrile and solvent was removed.The blue compounds have been puri–ed on a silica gel column utilizing acetonitrile»ammonia (10 : 0.5) as eluent, followed by recrystallization by diÜusion of diethyl ether into the acetonitrile solution. The yields of these reactions are about 95%. Data of 14. Compound 8 (0.050 g, 0.18 mmol) gave 14 (0.080 g, 98%). IR (KBr): 1534s, 1356s, 1357m, 836s, 559m.UV/VIS 339, 355, 505sh, 605; 345, 421 (CH3CN): kmax kmin nm. 13C NMR d 162.86, 157.37, 154.16, 154.08, (CD3CN): 140.48, 129.30, 126.40, 117.42. MS (MALDI-TOF): m/z 612. (Found: C, 39.49 ; H, 2.33 ; N, 12.26%. Calcd for C, 39.93 ; H, 2.23 ; N, 12.42%). C30H20F12FeN8O4P2 : Data of 15. Compound 10 (0.050 g, 0.16 mmol) gave 15 (0.075 g, 96%). IR (KBr): 1536s, 1437m, 1349s, 1338m, 835s, 558m.UV/VIS 281, 286, 363, 607; 284, (CH3CN): kmax kmin 311, 427 nm. MS (MALDI-TOF): m/z 668. (Found: C, 41.99 ; H, 3.10 ; N, 11.47%. Calcd for C, C34H28F12FeN8O4P2 : 42.61 ; H, 2.94 ; N, 11.69%). Data of 16. Compound 12 (0.050 g, 0.17 mmol) gave 16 (0.075 g, 94%). IR (KBr): 1536s, 1434m, 1353s, 1340s, 835s, 558. UV/VIS 280, 359, 606; 306, 417, 427 (CH3CN): kmax kmin nm.MS (MALDI-TOF): m/z 640. (Found: C, 40.53 ; H, 2.76 ; N, 12.07%. Calcd for C, 41.31 ; H, C32H24F12FeN8O4P2: 2.60 ; N, 12.07%). General procedure for the synthesis of iron(II) complexes of aminoterpyridines. To a mixture of 0.050 g each of nitroterpyridines 8 and 12 in 25 ml ethanol»water (4 : 1) and 0.090 g powdered metallic iron was added 0.5 ml concentrated hydrochloric acid and the mixture was heated for 15 min at 100 °C.The colour of the blue complexes formed changed immediately to purple. Ethanol was removed and the residue was dissolved in 30 ml water. The complexes were –ltered over Celite and worked up as described for complexes 17 and 18. The yields of these reactions are about 85%. Data of 17. Compound 8 (0.060 g, 0.216 mmol) gave 17 (0.080 g, 88%).IR (KBr): 3398m, 1638m, 1621m, 1484m, 1450m, 842s, 558m. UV/VIS 372, 525sh, 566; (CH3CN): kmax 423 nm. 13C NMR d 160.24, 159.49, 157.65, kmin (CD3CN): 154.43, 138.98, 127.80, 123.63, 109.98. MS (MALDI-TOF): m/z 552. (Found: C, 42.46 ; H, 3.14 ; N, 13.48%. Calcd for C, 42.78 ; H, 2.87 ; N, 13.30%). C30H24F12FeN8P2: Data of 18. Compound 12 (0.050 g, 0.163 mmol) gave 18 (0.060 g, 82%).IR (KBr): 3401m, 1638s, 1621s, 1490m, 1458m, 842s, 558m. UV/VIS 283, 320s 368s, 558; (CH3CN): kmax kmin 430 nm. MS (MALDI-TOF): m/z 608. (Found: C, 45.54 ; H, 3.62 ; N, 12.11%. Calcd for C, 45.45 ; H, C34H32F12FeN8P2 : 3.59 ; N, 12.47%). Synthesis of the heteroleptic iron(II) complex 19. Compounds 8 (0.010 g, 0.040 mmol) and 9 (0.011 g, 0.040 mmol) were dissolved in 3 ml ethanol; was added in excess and FeCl2 … 4H2O a blue mixture was obtained.The three complexes were separated on an aluminium oxide column utilizing acetonitrile»ammonia (10 : 0.5) as eluent. As –rst fraction we obtained purple complex 17 and as second fraction the dark blue heteroleptic complex 19. The blue complex 14 was obtained as the last fraction. The solvent mixture with 19 was removed, water was added followed by ammonium hexa- —uorophosphate.The complex was –ltered over celite, washed with 30 ml water, followed by diethyl ether and dried. The complex was then dissolved in acetonitrile and solvent was removed. Upon recrystallization by diÜusion of diethyl ether into the acetonitrile solution, 19 (0.010 g, 29%) was obtained. Data of 19.IR (KBr): 3405m, 1638m, 1526m, 1349m, 1092m, 1066s, 1040m, 980m, 559s. UV/VIS (CH3CN): kmax 271, 391sh, 525, 623; 444, 562 nm. 13C NMR d kmin (CD3CN): 159.21, 158.35, 158.06, 154.35, 153.60, 140.15, 139.60, 129.128, 127.86, 125.56, 124.28, 117.15, 110.47. MS (MALDI-TOF): m/z 582. (Found: C, 41.84 ; H, 2.42 ; N, 12.99%. Calcd for C, 41.31 ; H, 2.54 ; N, 12.85%). C30H22F12FeN8P2: Synthesis of terpyridine ruthenium trichlorides 20 and 21.A mixture of 0.050 g each of terpyridines 8 and 9 and 1 equiv. of ruthenium trichloride trihydrate were heated at re—ux for 2 h in 20 ml methanol. The dark blue 20 and brown 21 insoluble salts were collected by –ltration and washed with methanol (5 ml), diethyl ether (5 ml) and dried. The yield was about 70%. Data of 20.IR (KBr): 1600m, 1538s, 1423m, 1344s, 1279m, 798m, 755m. General procedure for the synthesis of ruthenium(II) complexes. A mixture of 1 mole equiv. of 20 or 21 with 1 mole equiv. terpyridine 8 or 9, respectively, in 20 ml ethanol in the 60 New J. Chem., 1999, 53»61presence of 0.3 ml N-ethylmorpholine was heated under re—ux for 2 h. The solution was then –ltered over Celite and washed with 50 ml water.The resulting ruthenium(II) complexes (22 or 23) were precipitated as their hexa—uorophosphate salts by the addition of methanolic ammonium hexa—urophosphate and worked up as described for iron(II) complexes. Alternatively, the homoleptic complexes 22 and 23 have also been prepared in a microwave oven. A suspension of terpyridines 8 or 9 (1 mol) and (0.5 mol) in eth- RuCl3 … 3H2O ylene glycol (5 ml) was heated in a microwave oven at 600 W for 10 min.The red solution was then poured into water (40 ml). After –ltration upon adding the desired [NH4][PF6] complex was isolated and puri–ed as above. The yields were 20% (22) and 90% (23). Data of 22. Compounds 20 (0.030 g, 0.064 mmol) and 8 (0.020 g, 0.064 mmol) gave 22 (0.020 g, 33%).IR (KBr): 1530s, 1349s, 842s, 558m. UV/VIS 274, 519; 403 (CH3CN): kmax kmin nm. MS (MALDI-TOF): m/z 657. (Found: C, 38.66 ; H, 2.41 ; N, 12.16%. Calcd for C, 38.03 ; H, C30H20F12N8O4P2Ru: 2.13 ; N, 11.83%). Data of 23. Compounds 21 (0.040 g, 0.091 mmol) and 9 (0.025 g, 0.091 mmol) gave 23 (0.074 g, 92%). IR (KBr): 3395m, 1634s, 1619m, 1477m, 1430m, 844s, 787m, 558m. UV/VIS 274, 299sh, 350sh, 492; 395 nm.(CH3CN): kmax kmin MS (MALDI-TOF): m/z 597. (Found: C, 40.05 ; H, 3.12 ; N, 12.13%. Calcd for C, 39.97 ; H, C30H24F12FeN8P2 …H2O: 2.91 ; N, 12.43%). Data of 24. Compounds 21 (0.040 g, 0.091 mmol) or 20 (0.030 g, 0.064 mmol) with 8 (0.025 g, 0.091 mmol) or 9 (0.018 g, 0.064 mmol), respectively, gave 24 (0.067 g, 80%) and (0.050 g, 85%). IR (KBr): 3401m, 1636m, 1528m, 1479m, 1430m, 1346s, 836s, 558s.UV/VIS 273, 348sh, 465sh, (CH3CN): kmax 523; 402 nm. MS (MALDI-TOF): m/z 627. (Found: C, kmin 40.05 ; H, 2.19 ; N, 12.13%. Calcd for C30H22F12FeN8O2P2 : C, 39.27 ; H, 2.42 ; N, 12.21%). Data of 25. Compounds 20 (0.020 g, 0.043 mmol) and 4@- dimethylamino-2,2@ : 6@,2A-terpyridine (0.011 g, 0.043 mmol) gave 25 (0.030 g, 77%). IR (KBr): 3400m, 1618m, 1524m, 1426m, 840s, 558m.UV/VIS 272, 302, 496; (CH3CN): kmax 289, 390 nm. MS (MALDI-TOF): m/z 641. (Found: C, kmin 40.57 ; H, 2.83 ; N, 12.26%. Calcd for C, C32H28F12N8OP2Ru: 41.26 ; H, 3.03 ; N, 12.03%). Data of 26. Compounds 21 (0.030 g, 0.069 mmol) and 8 (0.020 g, 0.069 mmol) gave 26 (0.030 g, 48%). IR (KBr): 3405m, 3333m, 2925m, 1655m, 1637m, 1619m, 1474m, 1432m, 832s, 789m, 559s.UV/VIS 271, 304, 492; (CH3CN): kmax kmin 289, 390 nm. MS (MALDI-TOF): m/z 613. (Found: C, 39.27 ; H, 2.33 ; N, 12.26%. Calcd for C, 39.88 ; C30H24F12N8OP2Ru: H, 2.68 ; N, 12.40%). References 1 F. Kroé hnke, Synthesis, 1976, 1. 2 E. C. Constable and M. D. Ward, J. Chem. Soc., Dalton T rans., 1990, 1405. 3 V. Grosshenny and R. Ziessel, J. Organomet. Chem., 1993, C19ñ22, 453. 4 K. Potts, Bull. Soc. Chim. Belg., 1990, 99, 741. 5 E. C. Constable, Chem. Commun., 1997, 925. 6 E. C. Constable, A. M. W. Cargill Thompson, D. A. Tocher and M. A. M. Daniels, New J. Chem., 1992, 16, 855. 7 R.-A. Fallahpour and E. C. Constable, J. Chem. Soc., Perkin T rans. 1, 1997, 2263. 8 R.-A. Fallahpour, Eur J. Inorg. Chem., 1998, 1205. 9 V. Farina, V. Krishnamurthy and W. J. Scott, Org. React., 1997, 50, 1. 10 U. Neumann and F. Voé gtle, Chem. Ber., 1989, 122, 589. 11 F. Kroé hnke and H. Schaé fer, Chem. Ber., 1962, 95, 1104. 12 C. Bolm, M. Ewald, M. Felder and G. SchlingloÜ, Chem. Ber., 1992, 125, 1169. 13 P.-M. Windscheif and F. Voé gtle, Synthesis, 1994, 87. 14 D. St. C. Black and N. E. Rothnie, Aust. J. Chem., 1983, 36, 1141. 15 E. C. Constable, J. Lewis, M. C. Liptrot and P. R. Raithby, Inorg. Chim. Acta, 1990, 178, 47. 16 E. C. Constable, F. K. Khan, P. R. Raithby and V. E. Marquez, Acta Crystallogr., Sect. C, 1992, 48, 932. 17 G. R. Desiraju, Angew. Chem. Int. Ed. Engl., 1995, 34, 2311. 18 J. Emsley, Chem. Soc. Rev., 1980, 9, 91. 19 V. Balzani, A. Juris, M. Venturi, S. Campagna and S. Serroni, Chem. Rev., 1996, 96, 759. 20 J.-P. Collin, P. Gavin8 a, V. Heitz and J.-P. Sauvage, Eur. J. Inorg. Chem., 1998, 1. 21 J. March, in Advanced Organic Chemistry, Wiley, Chichester, 1985, p. 1104. 22 A. Altomare, G. Cascarano, G. Giacovazzo, A. Guagliardi, M. C. Burla, G. Polidori and M. Camalli, J. Appl. Crystallogr., 1994, 27, 435. 23 D. Watkins, CRY ST AL S, Chemical Crystallography Laboratory, Oxford, 1990, issue 9. 24 International T ables for X-ray Crystallography, eds. J. A. Ibers and W. C. Hamilton, Kynoch Press, Birmingham, England, 1974, vol. IV, tables 2.2B and 2.3.1. Paper 8/06690F New J. Chem., 1999, 53»61 61

ISSN:1144-0546    

DOI:10.1039/a806690f

出版商:RSC    

年代:1999

数据来源: RSC