摘要:

L e t t e r Engineering diversity into dynamic combinatorial libraries by use of a small —exible building block Stuart J. Rowan, Philip S. Lukeman, Dominic J. Reynolds and Jeremy K. M. Sanders* Cambridge Centre for Molecular Recognition, University Chemical L aboratory, L ens–eld Road, Cambridge, CB2 1EW , UK Oligomerization of preorganized cinchona and xanthene building blocks to produce dynamic combinatorial libraries results in libraries with only a few components. Diversity is generated by the addition of semi-—exible ephedrine and cholate building blocks.We report here a signi–cant step forward in the generation of diverse macrocyclic dynamic combinatorial libraries (DCL); such libraries provide a potential ì selection œ approach to biomimetic catalysis that is conceptually related to molecularly imprinted polymers,1 catalytic antibodies and ribozymes.2 A combinatorial library consists of many members … … … (M1 Mn), each of which contains two or more building blocks (A, B, C, … … …) arranged in a particular way.In a traditional combinatorial library3 the covalent bonds between building blocks are –xed using irreversible chemistry during synthesis so there can be no post-synthetic interconversion between members such as ABC and ACB.In a dynamic combinatorial library the connections between building blocks are reversible and in —ux, continuously being made and broken; these connections may be covalent bonds such as imine,4 ester,5h7 disul –de,8 borate9 or alkene linkages, or they may be non-covalent, utilizing metal»ligand10 or hydrogen bonding11 interactions.The composition of a dynamic library will be dependent on its environment (Fig. 1) : addition of a template that selectively binds one member will bias the equilibrium T1 towards that member, while a diÜerent template will bias T2 the composition in a diÜerent direction. Two complementary types of DCL have been proposed: libraries of small ligands for large receptors [Fig. 2(a)],3 and libraries of large receptors for small ligands [Fig. 2(b)].6h10,12 In practice, however, no dynamic libraries of large receptors have yet been reported to contain more than a very few members. We now show why attempts to create dynamic libraries with relatively rigid building blocks can easily fail, giving self-sorted homo-oligomers rather than the desired diversity,13 and demonstrate that the use of small semi-—exible building blocks induces precisely the kind of mixing and diversity required.Fig. 1 Changing the equilibrium in a dynamic combinatorial library of members … … … using thermodynamic templates M1 Mn Tm * E-mail: jkms=cam.ac.uk DCLs in this laboratory are currently produced using basecatalysed transesteri–cation of the building block monomers shown in Fig. 3. Building blocks 1»7 possess recognition features and spectroscopic tags as part of an overall concave geometry that encourages macrocyclization ; each also carries a hydroxymethylester functionality for transesteri–cation. 1, 2 and 6 have been described before5,13 while the new monomers 3»5 were prepared by analogous methods.The new ephedrinederived monomer 7 is discussed in more detail below. As reported earlier,5,7 the quinine- and cinchonidine-derived monomers 1 (HOwCqwOMe and HOwCcwOMe, respectively) yielded cyclic trimers or in more than Cq3 Cc3 95% yield when subjected to transesteri–cation at 5 mM concentration. In order to explore the eÜect of a radically diÜerent shape within the context of otherwise identical chemistry and components, we prepared the diastereomeric quinidinederived monomer 3 (HOwCdwOMe); this yields, under the same reaction conditions, more than 95% cyclic dimer Cd2 .Addition of an extension arm to give 2 (HOwCewOMe) and 4 (HOwCawOMe) would be expected to increase —exibility and so to expand the range of cyclic oligomers produced at equilibrium.This is indeed what is observed for 2,5 but to our surprise the new extended quinidine 4 still cyclized to more than 90% dimer. These results are summarized in Table 1. When mixtures of two of the alkaloid building blocks 1»4 were subjected to transesteri–cation, the outcome ranged from eÜective mixing (e.g., 1 and 2) to almost perfect ì self-sorting œ (Table 1). The two most striking examples of self-sorting are 3 and 4, which yielded around 90% of homo-dimers when transesteri –ed together, and 1b and 3, which gave almost exclusively a mixture of and Such self-sorting»which Cc3 Cd2 . occurs despite the fact that many mixed intermediates must be Fig. 2 The use of dynamic combinatorial libraries to generate (a) the optimum ligand for a large biological receptor and (b) the optimum receptor for a tethered ligand New J.Chem., 1998, Pages 1015»1018 1015Fig. 3 Building block monomers explored for potential production of dynamic combinatorial libraries. Each monomer is identi–ed by a unique two-letter code and with HO/OMe labels to indicate that this is the hydroxymethylester series formed in the course of the reaction»is a unique property of thermodynamically controlled chemistry.14 Some of its origin can be understood by considering the reaction between two monomers of the same shape but diÜerent ì bite size œ [Fig. 4(a)] : the heterodimer will exhibit some bond or angle strain [Fig. 4(b)] and so will be destabilized relative to the self-sorted homo-dimers [Fig. 4(c)]. Because all the reactions are Fig. 4 How incompatible ì bite size œ can lead to self-sorting of building blocks in thermodynamically controlled chemistry : (a) a mixture of large and small building blocks, (b) mixed dimer exhibiting bond and/or angle strain, (c) self-sorted mixture of strain-free dimers and (d) the use of a small diversi–er to facilitate formation of mixed oligomers reversible, the product distribution will move to minimize the concentration of strained (i.e., mixed) products.15 An argument based on bite angle or relative conformational stability of the monomer once incorporated in a cyclic oligomer, which could lead to trimers or tetramers, would be just as eÜective, and it seems likely that the self-sorting observed here is due to a combination of size, angle and conformational eÜects. Of course, two complementary monomers might lead efficiently to a single hetero-dimer but that does not improve diversity in any general way.One way to increase diversity and avoid self-sorting would be to use only small and very —exible building blocks but the likely cavity collapse that would result might minimize substrate binding and selectivity. We chose, therefore, to explore the eÜect of adding a small, semi-—exible building block to mixtures of the larger monomers: we reasoned that the smaller unit should act in situ as a diversi–er to facilitate mixing of monomers with incompatible bite sizes, [Fig. 4(d)]. This approach should allow –ne-tuning of macrocyclic cavity size while at the same time giving access to relatively small oligomers containing large monomers in low-energy, unstrained conformations.(1R, 2S)-(»)-Ephedrine adopts a similar conformation to quinine,16 but is more —exible ; in order to impose some rigidity and give the required hydroxymethylester functionality, ephedrine was re—uxed with methyl bromomethyl benzoate in acetonitrile to produce the new monomer 7 (HOwEbwOMe).17 When cyclized alone at 5 mM under thermodynamic conditions, 7 produced a distribution of macrocycles, from the cyclic dimer through to the Eb2 cyclic heptamer con–rming that the monomer is relaxed Eb7 , in the desired way.When the quinine monomer 1a and ephedrine monomer 7 were subjected to transesteri–cation together, we obtained a library of macrocycles that included all possible compositions Table 1 Outcome of transesteri–cation reactions of the alkaloids 1»4 HOwCcwOMe HOwCewOMe HOwCdwOMe HOwCawOMe HOwCcwOMe [95% trimer Good mixing Self-sorting Good mixing Quinine HOwCewOMe Dimer\trimer Mainly dimers, » Extended quinine [tetramer including mixed HOwCdwOMe [95% dimer Self-sorting, Quinidine \5% mixing HOwCawOMe [90% dimer Extended quinidine 1016 New J.Chem., 1998, Pages 1015»1018Fig. 5 Part of the ESI-MS spectra of the reaction products resulting from two transesteri–cation reactions. In both cases the dominant Cd2 peak has been truncated. (a) A mixture of cinchonidine 1b and quinidine 3 monomers at initial concentrations of 5 mM each: the only visible mixed product is a trace of (b) The same monomers Cc2Cd. together with 5 mM ephedrine monomer 7. Insets show the trimer and tetramer region of the spectrum, expanded for clarity of mixed cyclic dimers, trimers, tetramers and even some mixed pentamers.18 A more severe test was the eÜect of 7 on the transesteri–cation of a mixture of the cinchonidine19 1b and quinidine 3 monomers, which self-sort as described above and as illustrated in Fig. 5(a). The eÜect of 7, as observed by Fig. 6 Part of the ESI-MS spectrum of the reaction products resulting from transesteri–cation of 2.5 mM xanthene monomer 6 with 2.5 mM ephedrine monomer 7.The high-mass region is expanded to allow labelling of key peaks ESI-MS [electrospray ionization mass spectrometry, Fig. 5(b)], is a library of cyclic oligomers that ranges up to pentamers. Both ESI-MS and high pressure liquid chromatography indicate that is still the most favoured product, but Cd2 quinidine»ephedrine and cinchonidine»ephedrine conjugates and, most importantly, mixed macrocycles containing all three monomers (e.g., EbCcCd, are Eb2CcCd, EbCc2Cd, Eb2CcCd) obtained; not surprisingly perhaps, the most abundant pentamers (not shown) contain relatively large proportions of the ephedrine monomer.The eÜect of ephedrine monomer 7 on a pair of components that already mix well, for example, the quinine and extended quinine monomers 1a and 2, was to produce a large library that contains all possible trimers and some of the possible tetramers.The almost complete absence of pentamers suggests that the lower mixed oligomers are more stable than they are in the CcCdEb mixed reaction. This should not be surprising : in the absence of any strain, entropy arguments tend to favour formation of small oligomers because that leads to the largest number of independent molecules,20 and the use of —exible building blocks inevitably reduces strain.When four diÜerent monomers (1a, 1b, 5 and 7) were subjected to transesteri –cation together the result was a library containing over 30 diÜerent compositions of species, many of which undoubtedly disguise the presence of several diÜerent isomers.21 All mixed cyclic trimer compositions are observed along with some tetramers ; the most abundant cyclic tetramer observed by ESI-MS is the statistically most favourable composition containing all four monomers, EbCcCqSp.The most severe test of 7 as a diversi–er was the xanthene monomer 6 (HOwXawOMe): this monomer is so strongly pre-organized to give cyclic dimers that it self-sorts almost exclusively, even in the presence of the extended quinine 2 and cholate 5.However, when 6 and 7 were transesteri–ed together, small but signi–cant quantities of mixed trimers and tetramers were obtained (Fig. 6). It is important to note that for a dynamic combinatorial library to be eÜective in the sense implied by Figs. 1 and 2 it is not necessary for all members to be formed in equal, or even comparable, quantities : we need only to know that a receptor is thermodynamically accessible, that is, formed in trace quantities. If it is selected by thermodynamic templating, then its concentration will increase and it will be possible to isolate and identify it.Absolute quantititative conclusions cannot be drawn from the mass spectra shown in Figs. 5 and 6 because the intrinsic detectability of each species depends on its component building blocks, on its degree of oligomerization and on the presence of multiply charged ions ; however, relative intensities within diÜerent spectra containing the same components do correlate directly with relative concentrations.22 The larger proportion of relative to visible in Fig. 5(b) by Eb3 Eb2 comparison with Fig. 6 is consistent with the higher total monomer concentration used in the three-component experiment: larger oligomers are favoured by higher concentrations. In summary, we have shown (i) why the counter-intuitive phenomenon of covalent self-sorting is common when preorganized building blocks are subjected to thermodynamically equilibrating conditions23 and (ii) that a small —exible building block can facilitate mixing and produce diversity in dynamic combinatorial libraries.We now have to turn to the more challenging task of selecting interesting members of such libraries through the use of tethered templates. Experimental Deoxycholate 7 was prepared similarly to the p-phenylbenzyl derivative6 except that p-bromobenzyl-2,2,2-trichloroacetimidate was used instead of p-phenylbenzyl-2,2,2-trichloroacetimidate. The p-bromobenzyl cholate ether was then subjected to a Stille coupling using 4-trimethylstannyl pyri- New J.Chem., 1998, Pages 1015»1018 1017dine and deprotected as before.6 All new monomers were characterized by 1H and 13C NMR and accurate FAB-MS (fast atom bombardment).Transesteri–cation conditions and preparation of samples for ESI-MS were as described before.7 Monitoring of combinatorial libraries was mainly carried out using positive-ion ESI mass spectrometry. Spectra were obtained on a VG BioQ triple quadrupole apparatus with a m/z range up to 4000 (VG Bio Tech Ltd, Altrincham, UK). The electrospray source was heated to 70 °C, with 80 V sampling cone voltage The samples were introduced into the (Vc).mass spectrometer source with an LC pump (Shimadzu LC-9A LC pump) at a —ow rate of 4 lL min~1 of acetonitrile» water (1 : 1). The data system was operated as a multichannel analyzer and several scans were summed to obtain the –nal spectrum. Acknowledgements thank the EPSRC for generous –nancial support, David We Sanders for numerical simulations and Dr Darren Hamilton for the xanthene sample.References and Notes 1 G. WulÜ, T. Gross and R. Schoé feld, Angew. Chem., Int. Ed. Engl., 1997, 36, 1962. 2 P. A. Brady and J. K. M. Sanders, Chem. Sov. Rev., 1997, 26, 327. 3 S. R. Wilson, A. W. Czarnik, Combinatorial Chemistry. Synthesis and Application, Wiley, New York, 1997. 4 I. Huc and J.-M. Lehn, Proc. Natl. Acad. Sci. USA, 1997, 94, 2106. 5 S. J. Rowan and J. K. M. Sanders, Chem. Commun., 1997, 1407. 6 P. A. Brady and J. K. M. Sanders, J. Chem. Soc., Perkin T rans. 1, 1997, 3237. 7 S. J. Rowan and J. K. M. Sanders, J. Org. Chem., 1998, 63, 1536. 8 H. Hioki and W. C. Still, J. Org. Chem., 1998, 63, 904. 9 T. Giger, M.Wigger, S. Audeç tat and S. A. Benner, Synlett., 1998, 688. 10 B. Klekota, M. H. Hammond and B. L. Miller, T etrahedron L ett., 1997, 38, 8639. 11 M. C. Calama, R. Hulst, R. Fokkens, N. M. M. Nibbering, P. Timmermans and D. N. Reinhoudt, Chem. Commun., 1998, 1021. 12 A. V. Eliseev and M. I. Nelen, Chem. Eur. J., 1998, 4, 825. 13 S. J. Rowan, D. G. Hamilton, P. A. Brady and J.K. M. Sanders, J. Am. Chem. Soc., 1997, 119, 2578. 14 Similar eÜects can be seen in the self-sorting of metal-coordinated helices : B. Hasenknopf, J. M. Lehn, N. Boumediene, A. Dupont- Gervais, A. Van Dorsselaer, B. Kneisel and D. Fenske, J. Am. Chem. Soc., 1997, 119, 10956. 15 Entropy considerations favour mixed oligomers over self-sorting, but even a small degree of *H strain will be magni–ed into a biased distribution. 16 G.D. H. Dijkstra, R. M. Kellogg and H. Wynberg, Recl. T rav. Chim. Pays Bas, 1989, 108, 195. 17 Details will be described elsewhere : S. J. Rowan, S.-A. Poulsen and J. K. M. Sanders, to be submitted. 18 The terms trimer, tetramer, pentamer are used loosely to mean a library member containing three, four or –ve monomer units. When applied to the products of mixed reactions they cover both homo-oligomers and mixed products. 19 The chemistry of 1a and 1b is identical for the purposes of this work, but 1a is isomeric with 3 and so would be indistinguishable by mass spectrometry. 20 X. L. Chi, A. J. Guerin, R. A. Haycock, C. A. Hunter and L. D. Sarson, J. Chem. Soc., Chem. Commun., 1995, 2563. 21 Mass spectrometry cannot distinguish isomeric mixed oligomers, but for good receptors isolated using the selection scheme of Fig. 2(b) the ordering of components within the structure will be possible using heteronuclear NMR techniques or via unambiguous linear synthesis.7 22 P. A. Brady and J. K. M. Sanders, New J. Chem., 1998, 22, 411. 23 As the number of building blocks in a mixture increases, the number of possible members of the dynamic library increases much more rapidly. However, this does not signi–cantly increase the time requirement for achieving equilibrium since the number of available reaction pathways increases even more rapidly : (a) D. P. Sanders, unpublished simulations ; (b) S. A. KauÜman, T he Origins of Order, Oxford University Press, New York, 1993. Received in Montpellier, France, 25th May 1998; L etter 8/03935F 1018 New J. Chem., 1998, Pages 1015»1018

ISSN:1144-0546    

DOI:10.1039/a803935f

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

L e t t e r Reversible synthesis of p-associated [2] catenanes by ring-closing metathesis : towards dynamic combinatorial libraries of catenanes Darren G. Hamilton,a Neil Feeder,b Simon J. Teatc and Jeremy K. M. Sanders*,a a Cambridge Centre for Molecular Recognition, University Chemical L aboratory, L ens–eld Road, Cambridge, UK CB2 1EW b University Chemical L aboratory, L ens–eld Road, Cambridge, UK CB2 1EW c CL RC Daresbury L aboratory, Daresbury, W arrington, UK W A4 4AD Ring-closing metathesis of aromatic diimides substituted with ole–n-terminated alkyl chains in the presence of a dinaphtho crown ether aÜords neutral [2]catenanes.Templating p-donor/p-acceptor interactions are exploited to drive mechanical interlocking and favour production of the more thermodynamically stable catenane products.Structure proof was provided by single-crystal X-ray synchrotron diÜraction of a representative catenane. In this letter we describe how catalytic ring-closing metathesis (RCM) may be used to prepare neutral p-associated [2]catenanes under thermodynamic control.1 The approach employs our well-established kinetically controlled catenane assembly process based on the electronic complementarity of p-de–cient aromatic diimides and p-rich aromatic diethers.2 We,3 and others,4 have highlighted the need to engineer reversibility into syntheses of complex molecular systems in order that ìincorrectœ bond formation can be proof-read and repaired.This capability may be programmed into a molecular system in the form of non-covalent interactions favouring a particular supramolecular geometry; for example, interactions between the component rings of a catenane may be exploited to favour production of the interlocked ring compound over the free macrocycles.This approach has been demonstrated with two ìmagic ringœ catenane syntheses, that is, the mutual interlocking of two preformed rings, directed and driven by hydrophobic and hydrogen-bonding interactions, respectively.5,6 The concept of product selection via the expression of a thermodynamic preference has appeared in various contexts in recent years.Lehn and coworkers have demonstrated various modes of self-sorting and selection in a variety of linear and circular inorganic helicates.7h9 Progression to selfsorting at the covalent level has been achieved for mixtures of building blocks where a thermodynamic driving force is provided by the conformational stability of particular macrocyclic products.10 Thermodynamic templating has been demonstrated for an equilibrating mixture of cyclocholate receptors where the addition of alkali metals is observed to alter the distribution of cyclic oligomers,11 and, very recently, Hioki and Still have reported the substrate-promoted chemical ampli–cation of a macrocyclic receptor containing reversible covalent linkages.12 Finally, we have shown that reversible [2]catenane formation may be quantitatively driven by donor»acceptor interactions in a system utilizing reversible zinc(II)-bipyridyl ligation.13 To progress to wholly covalent systems we required a reversible bond-forming process compatible with the rather weak nature of donor-acceptor interactions.Catalytic RCM operates in chlorinated organic solvents at room temperature and has previously been applied to many complex macrocycle syntheses, including catenanes14 and knots.15 Accordingly, we * E-mail: jkms=cam.ac.uk equipped the electron de–cient diimide components of our system with ole–n-terminated alkyl substituents. Mitsunobu alkylation16 of the parent diimides proved an efficient and versatile procedure and provided diole–ns 1»4 in good yield from the corresponding unsaturated alcohols (Scheme 1).§ Addition of two molar equivalents of diimide derivatives 2 or 4 to a 5 mMsolution of crown ether 5 in chloroform established familiar orange- and purple-coloured solutions respectively, the result of formation of donor»acceptor complexes.Subsequent addition of Grubbsœ catalyst 6 (0.05»0.10 equiv.) led to smooth formation of [2]catenanes 7 and 9; periodic LC-MS analyses of the reaction mixtures revealed the presence of the catenanes at m/z 1358 and 1458 ([M]NH4]`), respectively ; little change in the product distributions was Scheme 1 Preparation of ole–nic aromatic diimides via Mitsonobu alkylation § Example diole–n synthesis, 2: Pyromellitic diimide (200 mg, 0.93 mmol) was suspended in 15 mL of dry THF containing 5-hexen-1-ol (185 mg, 222 lL, 1.9 mmol) and (485 mg, 1.9 mmol).Ph3P Diethylazodicarboxylate (322 mg, 291 lL, 1.9 mmol) was added dropwise via syringe under argon and the clear yellow solution was stirred for 15 min.Flash column chromatography gave 2 (SiO2; CHCl3) as a pale yellow solid (232 mg, 66%): 1H NMR (250 MHz, (RfB0.6) d 8.25 (s, 2H), 5.84»5.68 (m, 1H), 5.04»4.92 (m, 2H), 3.74 (t, CDCl3) J\7 Hz, 2H), 2.10 (q, J\7 Hz, 2H), 1.73 (quintet, J\7 Hz, 2H), 1.45 (quintet, J\7 Hz, 2H); 13C NMR (100 MHz, d 166.28, CDCl3) 138.04, 137.26, 118.15, 115.08, 38.56, 33.14, 27.85, 26.05.New J. Chem., 1998, Pages 1019»1021 1019observed after 3 days. Preparative TLC gave around 20% yields of [2]catenane 7 as a mixture of three –nely separated alkene isomers (cis/cis, cis/trans, trans/trans). Hydrogenation of the mixture Pd»C) gave a single saturated interlocked (H2 , product 8 in near quantitative yield.î° Isolated yields of around 50% (post-hydrogenation) were obtained for [2]catenane 11, consistent with the stronger electron-accepting nature of naphthalene diimides ; this increase parallels the results of kinetic syntheses of related systems.2 As with related catenanes in this series only very tiny crystals could be obtained (from However, the solid- DMSO-d6).state structure of a representative example has been obtained from one such crystal using X-ray synchrotron diÜraction.The structure of 11 reveals a familiar arrangement of mutually interlocked complementary macrocycles with the planes of the electron-rich and electron-de–cient subunits adopting an alternating parallel stacked arrangement with interplanar spacings of around 3.5 These intramolecular donor» Aé . acceptor stacking interactions are perpetuated intermolecuî Example catenane synthesis, 8: Grubbsœ catalyst (1 mg, 1.25 lmol; Strem Chemicals, Royston, UK) was added to a solution of diole–n 2 (19 mg, 50 lmol) and crown 5 (16 mg, 25 lmol) in dry (5 mL, CHCl3 B5 mM) under argon. After three days stirring at room temperature LC-MS analysis revealed the presence of catenane 7 at m/z 1358 [M for competition experiments this material could be iso- ]NH4]`; lated by repeated preparative TLC 1:99) to (SiO2; MeOH»CHCl3 , aÜord pure 7 as a mixture of three isomers (7 mg, 21%).Alternatively, the metathesis mixture could be hydrogenated Pd»C) to (CHCl3; H2 , aÜord 8 as the single saturated interlocked product (LC-MS; m/z 1362, after preparative TLC separation [M]NH4]`) (SiO2; 0.5 :99.5).MeOH»CHCl3 , ° Crystal data for 11: monoclinic, C84H92N4O18, Mr\1445.6, a\12.741(2), b\28.768(4), c\21.035(3) b\106.620(10)°, Aé , U\7387.9(19) space group Z\4, g cmv3. Aé 3, P21/n, ocalcd\1.30 Crystal size 0.17]0.06]0.01 mm. Bruker SMART CCD diÜractometer on the single-crystal diÜraction station 9.8 at the Daresbury Laboratory Synchrotron Radiation Source (UK): 150 K, k\0.6883 h range for data collection 3.46 to 25.00° ; corrections were applied Aé .for incident beam decay. The structure was solved by direct methods and re–ned by full-matrix least-squares analysis on F2; 38373 re—ections measured of which 14127 were observed as unique (Rint\0.08). Despite substantial disorder in one of the chains the structure C10 re–ned to R\0.075, Atomic coordinates, bond lengths Rw\0.145.and angles, and thermal parameters have been deposited at the Cambridge Crystallographic Data Centre (CCDC). Any request to the CCDC for this material should quote the full literature citation and the reference number 102181. larly, individual [2]catenanes stacking one upon another to continue the alternating pattern (Fig. 1). However, whilst the aromatic subunits within individual catenanes are essentially parallel individual catenanes are tilted by 6»7° with respect to each other, producing a pronounced ìwaveœ in the extended donor»acceptor stack.The overlap orientation of the sandwiched naphthalene diimide unit is similar to that previously observed in a donor-acceptor co-crystal comprised of the components of 11, the long axes being disposed in a parallel fashion.17 The external diimide is twisted relative to the included subunit, presumably as a consequence of the structural organization needed to accommodate the lengthy tetraethylene glycol and decyl linker chains.The substantial twist angle (B40°) between the long axes of the diimide components renders this macrocycle, and hence individual [2]catenane molecules, chiral.The extended donor»acceptor stacks of chiral [2]catenanes in crystals of 11 are built of alternating enantiomers (Fig. 1), in contrast to our earliest system where individual columns of catenanes are comprised of a single enantiomer.2 Omitting the crown from the initial RCM reaction of 4 prompted the formation of a white precipitate, presumably comprised of a variety of linear and cyclic oligomeric species.“ Fig. 1 Solid-state structure of 11 displayed using Cerius Molecular Simulations software (alternating enantiomers in the stacks have different shading) “ The composition of the precipitate has proved difficult to ascertain, though FAB-MS analysis does reveal the presence of a cyclic dimer: m/z 805, [M]H]`. 1020 New J. Chem., 1998, Pages 1019»1021After three days of stirring, half a molar equivalent of crown 5 and additional catalyst were added.The production of [2]catenane product could be monitored by periodic LC-MS analysis ; after three further days stirring hydrogenation and separation as described before aÜorded similar isolated yields of 11 as obtained from direct RCM of 4 in the presence of 5. The reversible nature of the ole–n bond-forming reaction endows the system with the ability to recover from an energetically unfavourable state and drive molecular interlocking. This experiment is identical in concept to Grubbs and coworkersœ alkali metal-templated conversion of an ole–nic ethylene»glycol polymer to discrete unsaturated crown macrocycles. 18 For an unambiguous demonstration of reversibility in these reactions we performed competition experiments wherein a preformed [2]catenane is challenged with the addition of an alternative diimide derivative. Catalytic RCM of catenane 7 with an excess of diole–n 4 yielded some of the predicted more thermodynamically stable interlocked product 9, but only after protracted reaction.Similarly, addition of an excess of the shorter –ve-carbon chain naphthalene diimide derivative 3 to [2]catenane 9 also indicated slow incorporation of the new diole–n to aÜord small amounts of [2]catenane 10.It is clear that, at these solubility-limited concentrations, slow reaction kinetics militate against attainment of ideal statistical distributions. It is also likely that the current systems lack sufficient thermodynamic gradients between closely related derivatives for chemical evolution to be observed on a reasonable time scale.Perhaps an additional problem lies in attempting to maintain several competing species simultaneously in solution. Nevertheless, a 1 : 2 : 2 ratio of crown 5 to diole–ns 3 and 4 at 5 mM crown concentration gave around 70% isolated yield of and interlocked products 11 and 12 after C10- C9-linked hydrogenation; no catenane was observed.The C8-linked and products were isolated in an approximate C10- C9-linked ratio of 1 : 5, perhaps indicating a thermodynamic preference for the shorter linked derivative 10 (at least prior to hydrogenation) arising from a tighter mutual –t of the component macrocycles. These results show that it is possible to create a range of catenanes under reversible conditions and that hydrogenation provides a simple method to lock irreversibly the resulting rings in place.They also bring closer the creation of dynamic combinatorial libraries19 of catenanes from mixtures of p-rich and p-poor components, especially when coupled with a method for reversibly controlling the ability of these components to interact.20 Acknowledgements thank the Engineering and Physical Sciences Research We Council (UK) for generous –nancial support.References 1 For a comprehensive account of catenane chemistry see : D. B. Amabilino and J. F. Stoddart, Chem. Rev., 1995, 95, 2725. 2 D. G. Hamilton, J. E. Davies, L. Prodi and J. K. M. Sanders, Chem. Eur. J., 1998, 4, 608. 3 D. G. Hamilton, N.Feeder, L. Prodi, S. J. Teat, W. Clegg and J. K. M. Sanders, J. Am. Chem. Soc., 1998, 120, 1096. 4 D. B. Amabilino, P. R. Ashton, L. Peç rez-Garcïç a and J. F. Stoddart, Angew. Chem., Int. Ed. Engl., 1995, 34, 2378. 5 M. Fujita, F. Ibukuro, H. Hagihara and K. Ogura, Nature (L ondon), 1994, 367, 720. 6 D. A. Leigh, personal communication. 7 J.-M. Lehn, A. Rigault, J. Siegel, J.Harrow–eld, B. Chevrier and D. Moras, Proc. Natl. Acad. Sci. USA, 1987, 84, 2565. 8 B. Hasenknopf, J. M. Lehn, B. O. Kneisel, G. Baum and D. Fenske, Angew. Chem., Int. Ed. Engl., 1996, 35, 1838. 9 B. Hasenknopf, J.-M. Lehn, N. Boumediene, A. Dupont-Gervais, A. Van Dorsselaer, B. Kneisel and D. Fenske, J. Am. Chem. Soc., 1997, 119, 10956. 10 S. J. Rowan, D. G. Hamilton, P. A. Brady and J. K. M. Sanders, J. Am. Chem. Soc., 1997, 119, 2578. 11 P. A. Brady and J. K. M. Sanders, J. Chem. Soc., Perkin T rans. 1, 1997, 3237. 12 H. Hioki and W. C. Still, J. Org. Chem., 1998, 63, 904. 13 A. C. Try, M. M. Harding, D. G. Hamilton and J. K. M. Sanders, Chem. Commun., 1998, 723. 14 B. Mohr, M. Weck, J.-P. Sauvage and R. H. Grubbs, Angew. Chem., Int. Ed. Engl., 1997, 36, 1308. 15 C. Dietrich-Buchecker, G. N. Rapenne and J.-P. Sauvage, Chem. Commun., 1997, 2053. 16 O. Mitsunobu, W. Makato and T. Sano, J. Am. Chem. Soc., 1972, 94, 679. 17 R. S. Lokey and B. L. Iverson, Nature (L ondon), 1995, 375, 303. 18 M. J. Marsella, H. D. Maynard and R. H. Grubbs, Angew. Chem., Int. Ed. Engl., 1997, 36, 1101. 19 S. J. Rowan, P. S. Lukeman, D. J. Reynolds, J. K. M. Sanders, New J. Chem., 1998, 22, 1015. 20 D. G. Hamilton and J. K. M. Sanders, Chem. Commun., 1998, 1749. Received in Montpellier, France, 13th July 1998; L etter 8/05505J New J. Chem., 1998, Pages 1019»1021 1021

ISSN:1144-0546    

DOI:10.1039/a805505j

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

L e t t e r Mechanistic and thermodynamic aspects of methylene transfer from to (M = Ru, Os ; L= tertiary phosphine) : CH2N2 MHCl(CO)L2 non-least motion behavior and extreme dependence on phosphine identity Dejian Huang, Greg J. Spivak and Kenneth G. Caulton* Department of Chemistry, Indiana University, Bloomington, IN 47405-4001, USA Reaction of (M\Ru and Os) with was studied from [78 to 25 °C, revealing MHCl(CO)(PBut2Me)2 CH2N2 –rst the formation of where the carbene ligand occupies what was the open MHCl(CH2)(CO)(PBut2Me)2, CH2 coordination site of which lies trans to the hydride.This intermediate then isomerizes to MHCl(CO)(PBut2Me)2 , below 25 °C for each metal. The analogous reaction of with M(CH3)Cl(CO)(PBut2Me)2 , MHCl(CO)(PPri3)2 does indeed give which then ìdecomposesœ unselectively ; when M\Os, CH2N2 MHCl(CH2)(CO)(PPri3)2 , and are among the products.This extreme phosphine dependence is attributed to the C2H4 OsHCl(CO)(PPri3)2 to isomerization requiring phosphine dissociation ; the smaller fails to dissociate at HwMxCH2 M(CH3) PPri3 a rate competitive with alternative decomposition reactions. It was recently reported1 that an apparently straightforward and logical attempt to insert into the OswH bond of CH2 gave instead the product of OsHCl(CO)L2 (L\PPri3) ìadditionœ of to osmium [eqn.(1)]. CH2 One might argue that the mutually trans location of H and was responsible for the lack of as the CH2 Os(CH3)Cl(CO)L2 product; this logic thus says that the observed product forms under kinetic control and implies that Os(CH3)Cl(CO)L2 might (with patience) still be accessible as the thermodynamic product.The earlier report1 of the reaction of with CH2N2 sought to establish if the ì. . . 16-electron OsHCl(CO)L2 alkylosmium(II) compounds of general composition would be accessible . . . œ and com- Os(CH2R)Cl(CO)(PPri3)2 mented that these ìare still unknown.œ We report evidence that these target alkyls are the more stable isomeric structure for osmium, as they are also for ruthenium.We –rst report results on the analog of eqn. (1). PBut2Me Reaction of with in CH2N2 OsHCl(CO)L2@ (L@\PBut2Me) toluene at 25 °C gives only within Os(CH3)Cl(CO)L2@ 2 minutes. If the reaction is executed at [78 °C and all volatiles, including solvent, removed while still cold, NMR spectra recorded within 15 min of reaction showB10% of the corresponding carbene complex in OsHCl(CH2)(CO)(PBut2Me)2 ,3 addition to predominantly The Os(CH3)Cl(CO)(PBut2Me)2 .carbene in this solution then rapidly isomerizes completely to the methyl complex in benzene solution at 25 °C; the carbene complex is thus an intermediate in forming the unsaturated methyl complex. We next report results on the ruthenium analog.Reaction of with in at RuHCl(CO)L2@ (L@\PBut2Me) CH2N2 Et2O [20 °C for 12 h gives a 90% yield of This Ru(CH3)Cl(CO)L2@ . product in solution shows no hydride or carbene 1H NMR signals, and it does show diasterotopic But methyl groups. The * E-mail: caulton=indiana.edu triplet at 0.92 ppm for has an intensity Ru(CH3)Cl(CO)L2@ consistent with three hydrogens and the chemical shift and values exclude this being due to an acetyl group.The 13C JPH NMR signal of this Ru-methyl carbon is a triplet Hz) (JPC\6 at [11.0 ppm. The m(CO) value, 1898 cm~1, is lower than that of (1910 cm~1 in KBr), in agreement OsHCl(CO)(CH2)L2 with a lower oxidation state for the ruthenium species. This molecule [d(31P)\34.4] is thus analogous to even by comparison of the Ru(Ph)(Cl)(CO)(PBut2Me)2 ,4 m(CO) (1902 cm~1) and d(31P) 34.0) values of the latter.Monitoring of this reaction beginning at [75 °C reveals visible gas evolution and 31PM1HN NMR signals of an intermediate, as well as a small amount of At [75 °C, the intermediate Ru(CH3)Cl(CO)(PBut2Me)2 . appears as a 31PM1HN AB spin system (d\81.1 and 72.6) due to hindered rotation of the phosphines in this crowded sixcoordinate species, with a value of 166 Hz.While this JPP{ magnitude is large enough to be consistent with a transoid PwRuwP structure, this value is considerably smaller than the 250 Hz usually seen in square pyramidal or octahedral species ; this is strong evidence for its assignment as (M\Ru), since the solid-state structure1 MHCl(CH2)(CO)L2@ of has OsHCl[CH(SiMe3)](CO)(PPri3)2 nPwOswP\141°.As the temperature is raised (to [40 °C) on the above reaction solution, the AB 31PM1HN NMR pattern coalesces to one broad signal at 77 ppm (i.e., phosphine rotation accelerates) and the peak due to grows. By Ru(CH3)Cl(CO)(PBut2Me)2 [30 °C, the carbene species has rearranged completely to the methyl isomer. The variable-temperature 1H NMR spectra con–rm the above transformations and show ([75 °C) two inequivalent carbene hydrogens, at 16.1 and 15.3 ppm, and a hydride at [3.6 ppm.While all three of these peaks are broad at [75 °C, due to unresolved coupling to the inequivalent phosphorus nuclei, the chemical shifts are sufficiently close to those of to con–rm their assign- OsHCl(CH2)(CO)(PPri3)2 ments. Can it really be that, in eqn.(1), one intercepts a kinetic product and, if so, why? If true, it should be possible to isomerize the osmium hydrido carbene complex to Remarkably, the analogous chem- Os(CH3)(Cl)(CO)(PPri3)2 . istry with is very diÜerent from that of in PPri3 PBut2Me, New J. Chem., 1998, Pages 1023»1025 1023Scheme 1 (M\Ru and Os) spite of their having identical cone angles.5 A solution C6D6 of the previous reported is com- OsHCl(CH2)(CO)(PPri3)2 pletely decomposed within 24 h at 25 °C, yielding as the main ([50%) product, with no OsHCl(CO)(PPri3)2 formation of any species, as determined by NMR OswCH3 spectroscopy.This decomposition produces free ethylene whose signal at 5.49 ppm is broadened due to exchange in the equilibrium of eqn.(2) OsHCl(CO)(PPri3)2]C2H4¢OsHCl(C2H4)(CO)(PPri3)2 (2) which we have independently characterized by 1H and 31P NMR spectroscopy.6 Also observed are three weak hydride doublet peaks, which suggest phosphine loss, perhaps to form methylene phosphorane coordinated to Os. For (H2CxPPri3) comparison, reacts with with gas RuHCl(CO)(PPri3)2 CH2N2 evolution and color change from orange to pale yellow already at [78 °C to give a species RuHCl(CH2)(CO)(PPri3)2 wholly analogous to its Os analog: spectral features at [75 °C include a 31PM1HN NMR singlet, two doublet of CH2 doublets at 16.6 and 15.9 ppm and a hydride ([3.3 ppm) triplet Hz) of doublets of doublets.The approx- (JPH\26 imate threefold symmetry of avoids the rotamers found PPri3 for Again, analogous to the case, above PBut2Me.Os»PPri3 [20 °C decomposes unappealingly RuHCl(CH2)(CO)(PPri3)2 to six unidenti–ed phosphine-containing products. No is formed. RuMeCl(CO)(PPri3)2 7 We suggest that this surprisingly diÜerent behavior of for and MHCl(CH2)(CO)L2 L\PBut2Me L\PPri3 (summarized in Scheme 1) is due to the fact that migrating H and ligands to mutual cis stereochemistry requires CH2 phosphine loss, and the marginally smaller8 fails to PPri3 dissociate eÜectively from M.As a result, intact has a lifetime long enough for it to OsHCl(CH2)(CO)(PPri3)2 explore other reaction channels and the one involving ìCH2 loss œ, regenerating becomes operative. OsHCl(CO)(PPri3)2 , Indeed, is sufficiently long- OsHCl[CH(SiMe3)](CO)(PPri3)2 lived to allow crystal structure determinations.1 In the more crowded phosphine disso- MHCl(CH2)(CO)(PBut2Me)2 , ciation is evidently sufficient for rapid combination of initially trans H and ligands.The hydrido carbene isomer in eqn. CH2 (1) for is thus metastable because the two ligands that PPri3 might combine to make are mutually trans and there is CH3 evidently no facile unimolecular process within the sixcoordinate species that can bring these into mutually cis sites.Other, slow reaction channels thus operate and these are unselective. Numerous attempts to prove inverse phosphine dependence on the rearrangement rate by lengthening the half-life of by adding free OsHCl(CH2)(CO)(PBut2Me)2 PBut2Me or to the cold solution after reacting PPri3 with gave only an array of OsHCl(CO)(PBut2Me)2 CH2N2 new products.Thus, added phosphine acts not merely to shift equilibrium (3) to the left, but plays a more agressive role, perhaps by attack at the carbene carbon. OsHCl(CH2)(CO)(PBut2Me)2¢ OsHCl(CH2)(CO)(PBut2Me)]PBut2Me (3) These results show an extreme dependence of chemical behavior on ligand identity for two such super–cially similar (i.e., isomeric) phosphines.The (apparently9) bulkier ligand permits selective isomerization of mutually trans PBut2Me hydride and ligands to the methyl alternative, while CH2 gives rise to highly unselective ìdecompositionœ, with no PPri3 high yield fate of the carbene ligand, to regenerate some reagent. The latter case thus frustrates OsHCl(CO)(PPri3)2 establishing the truth that the example reveals : PBut2Me unsaturated is more thermodynamically M(CH3)Cl(CO)L2@ stable than MHCl(CH2)(CO)L2@ .The fact that the saturated, higher valent alternative, with more metal ligand bonds, is not OsHCl(CH2)(CO)L2 , thermodynamically preferred stands in contrast to saturated being more stable than unsaturated HOs(yCR)Cl2L2 as was discovered recently.10 In that work, Os[xC(H)R]Cl2L2 , it was shown10 that and RuCl2[C(H)R]L2 OsCl2H(CR)L2 are the contrasting ground state structures. This was supported by quantum chemical computations and was also rationalized by general trends in 4d vs. 5d metal chemistry. However, these results are fully consistent with the thermodynamics of eqn (4).11 The balance of stability between metal formal oxidation states is thus very subtle among these Ru and Os complexes.These accumulated results at two diÜerent hydrocarbyl ligand oxidation levels (A10 and B below), being a-hydrogen migration reactions that relate any pair of redox isomers, represent a rearrangement that could be generally useful for creating unsaturation in the saturated osmium series [eqn. (5)]. One conclusion from this work is that diazomethane does not ì deliver œ directly to the OswH bond.Instead, –nding CH2 intact trans to hydride in the kinetic product indicates CH2 that the open coordination site of square-pyramidal is the point of attack by OsHCl(CO)L2 CH2N2 . The observed isomerization of hydrido carbene to methyl shows that an unsaturated form can be more stable than its saturated isomer, particular when a strong sigma donor ligand like methyl is trans to the empty site.While the location of in in what was the empty site of CH2 MHCl(CO)(CH2)L2 reinforces the hypothesis that this is a kinetic MHCl(CO)L2 product, it is clear from the results that hydride and PBut2Me can –nd each other rapidly at 25 °C. CH2 The idea that the unfavorable insertion stereochemistry of a ì least motionœ primary product can in—uence its ultimate fate is one that must be considered in catalyst optimization.Consider, for example, that the open coordination site in the ole–n metathesis catalyst C is trans to the mechanistically essential 1024 New J. Chem., 1998, Pages 1023»1025carbene ligand. This explains why one metathesis mechanism12 begins by dissociation of L from C. Supplementary material.Detailed syntheses and characterization data are available from the author. Acknowledgements work was supported by the National Science Founda- This tion. We thank Johnson Matthey/Aesar for material support. References and Notes 1 H. Werner, S. Stué er, M. Laubender, C. Lehmann and R. Herbst- Irmer, Organometallics, 1997, 16, 2236. 2 Selected spectroscopic data for 1H [Os(CH3)Cl(CO)(PBut2Me)2] : NMR (300 MHz, 25 °C): 1.90 (t, Hz, 3H, C6D6 , 3JPH\4.8 13CM1HN NMR (75.3 MHz, 25 °C): [37.88 OswCH3) ; C6D6 , (t, Hz, 182.38 (t, Hz, OswCO; 2JPC\4.1 OswCH3), 2JPC\9.6 31PM1HN NMR (121.4 MHz, 25 °C): 16.78 (s, IR C6D6 , PBut2Me); cm~1) : m(CO)\1883s).(C6D6 , 3 1H NMR spectra 25 °C): 17.75 and 16.63 (br, (C6D6, CHa Hb), [4.04 (t, Hz, OswH); 31PM1HN NMR 25 °C): 2JPH\32 (C6D6 , 48.3 (s, Pbut2Me). 4 D. Huang, W. E. Streib, O. Eisenstein and K. G. Caulton, Angew. Chem., Int. Ed. Engl., 1997, 36, 2004. 5 D. White and N. J. Coville, Adv. Organomet. Chem., 1994, 36, 95. 6 The reported chemical shift for the coordinated ethylene vinyl hydrogens is in error, being in fact that of free ethylene. The coordinated ethylene 1H NMR signal is at 2.93 ppm and integrates correctly (4 : 6) against the Pri methine hydrogens. See M. A. Esteruelas and H. Werner, J. Organomet. Chem., 1986, 303, 221. 7 This compound has been synthesized independently from and MeLi. D. Huang and K. G. Caulton, to be RuCl2(CO)(PPri3)2 submitted. is less bulky than See C. Li, M. Olivaç n, S. P. 8 PPri3 PBut2Me. Nolan and K. G. Caulton, Organometallics, 1997, 16, 4223; C. Li, M. Ogasawara, S. P. Nolan and K. G. Caulton, Organometallics, 1996, 15, 4900. 9 The hindered rotation around the MwP bond observed for at low temperature supports the RuHCl(CH2)(CO)(PBut2Me)2 idea that this phosphine is bulkier than PPri3 . 10 G. J. Spivak, J. N. Coalter, M. Olivaç n, O. Eisenstein and K. G. Caulton, Organometallics, 1998, 17, 999. 11 D. Huang and K. G. Caulton, J. Am. Chem. Soc., 1997, 119, 3185. 12 E. L. Dias, S. T. Nguyen and R. H. Grubbs, J. Am. Chem. Soc., 1997, 119, 3887. Received in New Haven, USA, 29th April 1998; L etter 8/03432J New J. Chem., 1998, Pages 1023»1025 1025

ISSN:1144-0546    

DOI:10.1039/a803432j

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

L e t t e r Cation binding acceleration of DielsñAlder reaction of quinocrown ethers with cyclopentadiene Akihiko Tsuda and Takumi Oshima* Department of Applied Chemistry, Faculty of Engineering, Osaka University, Machikaneyama 1»16, T oyonaka, Osaka 560, Japan Diels»Alder reactions of 15»21 membered quinocrown ethers 1a»c with cyclopentadiene have been accelerated by the addition of alkali, alkaline-earth, and scandium perchlorates, re—ecting the selective cation-binding abilities of crown ethers.In the study of biological model systems, crown ethers have generated a tremendous amount of interest1 because of their ability to selectively bind guests by non-covalent interactions. In particular, functionalized crown ethers2 possessing multiple bonds or reaction groups are expected to play a promising role in the design of new host molecules with highly efficient and selective binding properties.These reactive crowns are of interest in view of the possible control of the reactivities by addition of cation species.3 Herein, we report the selective cation-binding eÜects in the kinetics of Diels»Alder reactions of 15»21 membered ring quinocrown ethers 1a»c with cyclopentadiene in comparison with the reaction of reference 2,3-dimethoxy-1,4-benzoquinone 2.§ Quinocrown ethers 1a and 1b are known to react with cyclopentadiene at the outer CxC double bond to give [4]2] endo adducts in 92»98% yield.5 We exploited this clear reaction as a probe for exploring the cation-binding features in the kinetics of crown ether reactions as formulated for the reaction of 1b in the presence of K` [eqn.(1)].î * E-mail: oshima=ch.wani.osaka-u.ac.jp § The quinocrowns 1a,b and 2,3-dimethoxybenzoquinone 2 were synthesized according to the literature.4 This method was applied for the synthesis of quino[21]crown-7 1c. The kinetic experiments were performed in acetonitrile solution containing a quinone (0.20»0.80 mM) and 20 equiv.excess of cyclopentadiene (4.0»16.0 mM) with or without added 4.0 mM of alkali and alkaline-earth metal as well as scandium perchlorate. The rates of reactions were determined by monitoring the disappearance of the absorbances due to the quinones 1a»c and 2 at nm. [The monitored kmax\391»399 wavelength at is as follows : 1a nm, e\131 kmax(e) (kmax\399 M~1 cm~1), 1b (399, 129), 1c (399, 130) and 2 (391, 104).] The rates obeyed pseudo –rst-order kinetics up to at least two half lives and the second-order rate constants were obtained by dividing the observed –rst-order rate constants by the corrected concentration of cyclopentadiene for the consumption of half the amount of quinone.The kinetic data for the addition of alkali-metal perchlorates are collected in Table 1.In the absence of perchlorates, all quinones used provided comparable rate constants of 4.6» 5.3]10~2 M~1 s~1 irrespective of having cyclic or acyclic substituents. However, added salts brought about rateacceleration depending on the combination of host molecules and guest cations.° In contrast, the control reaction of dimethoxyquinone 2 showed no appreciable dependency on added salts.All the kinetic features associated with the selective cation binding can be explicitly visualized in the plots of rate ratios, vs. metal ion radius, where and represent the k2M/k2 0k2M k2 0 respective rate constants for the reactions in the presence and absence of added metal perchlorates (Fig. 1). The rate pro–les of quinocrowns 1a»c are taken as re—ecting the cation-binding properties.Thus, 15-membered 1a gave the î Reaction of 1b (1 mM) with 20 equiv. excess of cyclopentadiene at 25°C for 1 h in acetonitrile solution containing 4 mM of also K(ClO4) led to almost the quantitative formation of [4]2] endo adduct as con–rmed by 1H-NMR spectroscopy. Selected data for 1c: 1H-NMR (270 MHz, d 3.66 (s, 8H), 3.68 CDCl3) (s, 8H), 3.81»3.85 (m, 4H), 4.45»4.49 (m, 4H) and 6.58 (s, 2H); IR(KBr) 2917, 1655, 1591, 1351, 1298, 1180, 1108, 951 and 845 cm~1; MS(EI) m/z 388 (M`, 20%), 298 (18%), 194 (30%), 179 (15%), 166 (100%), 138 (82%), 117 (21%), 89 (36%), 82 (49%), 73 (60%), 45 (98%).Found C, 56.11 ; H, 7.08 ; calcd for C, 55.95 ; H, 6.78%. C18H26O9: ° Recently stimulating role of Li` in diethyl ether has been reported to accelerate Diels»Alder reactions.The source of this eÜect has been discussed in terms of internal high pressure of solvent or Lewis acid catalysis.6 The eÜect of the counter anion was investigated for the cycloaddition of 2 (0.8 mM) with cyclopentadiene (16 mM) by addition of 18-crown-6 ether (0.8 mM) in the acetonitrile solution containing potassium perchlorate (4 mM) at 30 °C.Under these conditions, a rate decrease (9%) occurred as compared with the salt-free reaction. Keeping in mind that added 18-crown-6 alone did not aÜect the rate, this small decline may be ascribed to some weak interaction of free perchlorate with 2 owing to the selective complexation of K` by the added crown ether. New J. Chem., 1998, Pages 1027»1029 1027Table 1 Rate constants s~1) for Diels»Alder reactions of quinocrowns 1a»c, dimethoxyquinone 2 with cyclopentadiene in the presence (k2/M~1 and absence of added metal perchlorates in acetonitrile at 30 °Ca 10~2 k2/M~1 s~1 Quinone Additiveb\ None Li` Na` K` Rb` Cs` 1a 5.23 11.2 13.9 10.0 7.87 6.91 1b 4.60 6.68 9.13 17.1 13.7 9.55 1c 4.76 8.20 7.29 8.27 11.8 12.3 2 5.33 5.51 5.33 5.55 5.53 5.44 a Reactions were carried out under pseudo-–rst-order conditions by using 20 equiv.excess of cyclopentadiene (4.0»16.0 mM) with respect to quinone (0.20»0.80 mM). b Counter anion perchlorate was omitted; 4.0 mM solution of alkali, alkaline-earth metal perchlorate was used. maximum rates for Na` (ion radius 0.95 by analogy with ”) the complexation of comparable 15-crown-5 (cavity radius 0.86»1.1 A much more selective rate-pro–le accompanied ”).7 by the peak shift to K` (1.33 was observed for 18- Aé ) membered 1b.The high K` binding ability of 18-crown-6 (cavity radius 1.3»1.6 is well recognized.7 As found for 21- ”) membered 1c, one more oxyethylene unit enlargement produced somewhat strange plots with a small peak at Li`, although the maximum rate was still attained at the cavity-–tted cation Fig. 1 Plot of rate ratios for vs. the metal ion radius, where k2M/k2 0 k2M and represent the rate constants for the reactions in the presence k2 0 and absence of added perchlorate salts, respectively Fig. 2 Calculated structure of the complex [1c … Sc]3` Rb` (1.48) or Cs` (1.69 The additional peak may be due ”). to the formation of a dinuclear complex [1c … 2Li]2`.In fact, such multi-incorporated complexes are known for dibenzo-24- crowns-8 with Na` or K` cations.8“ With respect to the divalent alkaline-earth cations, a greater rate-acceleration was observed, re—ecting the selective cationbinding abilities. The rate constants of 1b at 30 °C were 6.66 (]10~2 M~1 s~1) for Mg2` (0.65 33.6 for Ca2` (0.99 ”), ”), 56.3 for Sr2` (1.13 and 47.1 for Ba2` (1.35 Thus, 18- ”), ”).membered 1b demonstrated 1.5-, 6-, 12-, and 10-fold acceleration by the addition of 4 mM of alkaline-earth perchlorates. Again, reference quinone 2 did not enjoy such a divalent cation-induced rate acceleration ; k\5.40»5.93]10~2 M~1 s~1. The selective rate-accleration by the smaller sized Sr2` rather than the expected Ba2` is probably due to the decline in the eÜective ring cavity owing to steric repulsion between quinone carbonyls and adjacent oxyethylene units.Such repulsion was anticipated in X-ray analysis9 and PM3 calculations of 1a. Of special interest is that the trivalent salt, Sc(ClO4)3 brought about an astonishing rate-acceleration of 117-fold for 1a, 160 for 1b, and 404 for 1c (only 4.1-fold for 2).The rate constants at 30 °C were 61.2 (]10~1 M~1 s~1) for 1a, 73.7 for 1b, 1930 for 1c, and 2.20 for 2. The order of rate-acceleration in going from 1a to 1c by the smaller Sc3` (0.81 apparently ”) contradicts the size-–tting concept.10 Therefore, the maximum rate for 21-membered 1c may be rationalized by considering a wrapping structure, where the counter anions and solvent molecules can be more eÜectively excluded from the bound Sc3` (Fig. 2).11 Such charge-separated incorporation of Sc3` would result in the unexpected rate acceleration. Bearing in mind that quinocrowns 1a»c exhibit rate pro–les re—ecting their cation-binding abilities as well as the cation valency, it can be easily envisaged that the incorporating guest cation behaves as an electron-withdrawing group on the quinone dienophile, thus lowering its LUMO energy suitable for the frontier orbital interaction with the HOMO of cyclopentadiene.In Diels»Alder reactions it is well known that the stronger the electron-withdrawing ability of the substituents then the more reactive the dienophile due to the diminished HOMO(diene)[LUMO(dienophile) energy diÜerence.12 References 1 (a) J.-M.Lehn, Science, 1985, 227, 849; (b) Y. Inoue and T. Hakushi, J. Chem. Soc., Perkin T rans. 2, 1985, 935; (c) J.-M. Lehn, Angew. Chem., Int. Ed. Engl., 1990, 29, 1304; (d) V. G. Young Jr., H. L. Quiring and A. G. Sykes, J. Am. Chem. Soc., 1997, 119, 12477; (e) C. Gong and H. W. Gibson, J. Am. Chem. Soc., 1997, 119, 8585; ( f ) R. Cacciapaglia, L. Mandolini, R. Arnecke, V.Boé hmer and W. Vogt, J. Chem. Soc., Perkin T rans. 2, 1998, 419. “ As unambiguous evidence for the selective alkali-metal cation binding of quinocrowns 1a»c, we also used ESI-MS which exhibited essentially the similar binding behaviors as the above rate pro–les. Furthermore, specially designed ESI-MS analysis showed the presence of the above-mentioned dinuclear complex [1c … 2Li]2`; details will be described elsewhere. 1028 New J. Chem., 1998, Pages 1027»10292 Y. Inoue and G. W. Gokel, Cation Binding by Macrocycles, Marcel Dekker, New York, 1990, ch. 13, pp. 523»547. 3 (a) V. Gold and C. M. Sghibartz, J. Chem. Soc., Chem. Commun., 1978, 507; (b) D. D. S. Baker, V. Gold and C. M. Sghibartz, J. Chem. Soc., Perkin T rans. 2, 1983, 1121; (c) Y. Inoue, M. Ouchi, H.Hayama and T. Hakushi, Chem. L ett., 1983, 431; (d) T. Oshima and T. Nagai, J. Org. Chem., 1991, 56, 673. 4 F. Dietl, G. Gierer and A. Merz, Synthesis, 1985, 626. 5 K. Hayakawa, K. Kido and K. Kanematu, J. Chem. Soc., Perkin T rans. 1, 1988, 511. 6 (a) P. A. Grieco, J. J. Nunes and M. D. Gaul, J. Am. Chem. Soc., 1990, 112, 4595; (b) H. Waldmann, Angew. Chem., Int. Ed. Engl., 1991, 30, 1306; (c) D.A. Smith and K. N. Houk, T etrahedron L ett., 1991, 32, 1549; (d) M. A. Forman and W. P. Daiely, J. Am. Chem. Soc., 1991, 113, 2761; (e) G. Desimoni, G. Faita, P. P. Righetti and G. Tacconi, T etrahedron, 1991, 47, 8399; ( f ) R. M. Pagni, G. W. Kabalka, S. Bains, M. Plesco, J. Wilson and J. Bartmess, J. Org. Chem., 1993, 58, 3130; (g) P. A. Grieco and J.A. Beck, T etrahedron L ett., 1993, 34, 7367; (h) P. A. Grieco, S. T. Handy and J. A. Beck, T etrahedron L ett., 1994, 35, 2663; (i) P. A. Grieco, M. D. Kaufman, J. F. Daeuble and N. Saito, J. Am. Chem. Soc., 1996, 118, 2095. 7 R. M. Izatt and J. J. Christensene, Synthetic Multidentate Macrocyclic Compounds, Academic Press, New York, 1978, pp. 207»243. 8 (a) M. Mercer and M. R. Truter, J. Chem. Soc., Dalton T rans., 1973, 2469; (b) D. L. Hughes, J. Chem. Soc., Dalton T rans., 1975, 2374. 9 A. Tsuda, T. Kawamoto and T. Oshima, Acta Crystallogr., 1998, in the press. 10 C. J. Pedersen and H. K. FrensdorÜ, Angew. Chem., Int. Ed. Engl., 1972, 11. 11 Molecular structure of [1c … Sc]3` was calculated by MM2, MOPAC Version 94.10 in CAChe, Version 3.7, CAChe Scienti–c, 1994. 12 I. Fleming, Frontier Orbitals and Organic Chemical Reactions, John Wiley, London, 1976. Received in Cambridge, UK, 18th May 1998; L etter 8/05326J New J. Chem., 1998, Pages 1027»1029 1029

ISSN:1144-0546    

DOI:10.1039/a805326j

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

L e t t e r Structure of 8-(diethylborylmethylamino)-3-phenyl-1-azaazulene : characteristics of the non-alternant ligand Yoshikazu Sugihara,*a Toshihiro Murafuji,a Noritaka Abe,a Mitsuhiro Takedaa and Akikazu Kakehib a Department of Chemistry, Faculty of Science, Y amaguchi University, Y amaguchi City 753-8512, Japan b Department of Chemistry and Material Engineering, Faculty of Engineering, Shinshu University, W akasato, Nagano 380-8553, Japan 8-(Diethylborylmethylamino)-3-phenyl-1-azaazulene (5) has been synthesized, in which the 8-methylaminoazaazulene moiety is found to act as a bidentate ligand whose two boron»nitrogen bond lengths are almost equivalent.Non-alternant conjugation is known to endow molecules with a variety of properties that are markedly diÜerent from those given by alternant conjugation.In contrast to alternant conjugation, the resonance stabilization energy in non-alternant conjugation has been indicated to be small.1 Hence, a slight modi–cation of the structure causes a loss of this conjugation, resulting in a signi–cant change of the properties. For example, bathochromic shifts of ca. 200 nm were observed in the UV/VIS spectra by the introduction of the acyl group to the adjacent position of the oxy function in the alternant extended phenalenones (1A, 2A),2 exhibiting the pronounced contribution of the non-alternant hydroxycyclopenta[a]- phenalenes (1B, 2B) in the tautomerism.Despite extensive studies on alternant bidentate ligands such as phenanthroline, the non-alternant ones have rarely been examined.3 Their high electron-donating and -accepting characters4 allow production of coordination molecules showing the new interaction between the ligand and the central metal or between the ligands.In line with our work on conjugated boron heterocycles,5 we have examined a coordination molecule composed of a boron atom and the conjugate base of 8-methylamino-3-phenyl-1-azaazulene (3),6§ and compared it with the structures of both 3 and dibromoboryl- 1,2,3,4-tetrahydro-1,10-phenanthroline (4).7 A solution of 3 (234 mg, 1.00 mmol) in tetrahydrofuran (11 ml) was treated with BuLi (1.20 mmol) at [78 °C.After 15 min, a tetrahydrofuran solution (1.0 M) of diethylmethoxyborane (1.20 mmol) was added dropwise and the mixture was stirred at [78 °C for 2 h.Then the solution was allowed to stand at room temperature overnight, and was re—uxed for 1 d. After extraction with chloroform, the mixture was chromatographed over silica gel with a mixture of chloroform and ethyl acetate (1 :1 v/v) to give 8-(diethylborylmethylamino)-3- phenyl-1-azaazulene (5) as orange prisms (270 mg, 89.3%, mp 93»94 °C). (All the compounds here are identi–ed spectro- § Synthesis of 3: A mixture of 8-chloro-3-phenyl-1-azaazulene6 (4.00 g, 16.8 mmol) and 40% aq.methylamine (20 ml) in ethanol (20 ml) was stirred for 1 d at room temperature. Filtration gave 8- methylamino-3-phenyl-1-azaazulene (3.10 g, 79%), which was recrystallized from chloroform»hexane to give reddish orange prisms, mp 159»161 °C. Scheme 1 New J. Chem., 1998, Pages 1031»1033 1031scopically, and by means of combustion analysis.) Though the coordination compound 5 is stable under the ambient conditions, it decomposes gradually in a protic solvent such as methanol.î UV/VIS spectra of 5 in acetonitrile are similar to those of 1, the long-wavelength end up to 600 nm suggesting the highlying HOMO and low-lying LUMO.° Though each corresponding vicinal coupling constant in 3 and 5 is not markedly changed, all the proton signals of 5 display up–eld shifts in the range 0.49»0.86 ppm, indicating the decrease of the diamagnetic ring current.The distinct down–eld shifts of C6, C7, C8 and C8a by 1.1, 2.6, 3.2 and 1.6 ppm, respectively, together with the slight up–eld shifts of C3 and C3a by 1.6 and 1.8 ppm, respectively, exhibit the electron donation from the seven-membered ring to the –ve-membered ring.The marked up–eld shift of C2 by 16.5 ppm suggests the formation of the coordinate bond between the boron and N1 nitrogen atom.8 The signal of the boron atom at 6.8 ppm, which is similar to the value of 8.5 ppm of 4, indicates this atom to be tetracoordinate. Single-crystal X-ray crystallography of 3 (Fig. 1),“ showed that the amino-hydrogen atom in 3 is located more closely to the amino-nitrogen atom (N2) than the skeletal nitrogen (N1) Fig. 1 Computer-generated thermal ellipsoid of 3. Some important bond distances and angles (°) not given in the text are as follows : (”) N1wC1\1.359(5), C1wC2\1.379(6), C2wC3\1.443(6), C3w C9\1.450(5), C3wC4\1.400(6), C4wC5\1.372(7), C5wC6\ 1.402(6), C6wC7\1.376(6), C7wC8\1.428(7), C8wC9\1.426(6), C9»N1\1.345(6), C9wN1wC1\105.4(4), N1wC1wC2\113.8(4), C1wC2wC3\105.4(4), C2wC3wC9\103.6(4), C3wC9wN1\ 111.8(4), C4wC3wC9\129.1(4), C3wC4wC5\127.8(4), C4wC5w C6\128.3(5), C5wC6wC7\131.1(5), C6wC7wC8\130.0(5), C7wC8wC9\124.1(4), C8wC9wC3\129.4(4), C8wC9wN1\ 118.8(4) î The attempted synthesis using 8-(dimethylamino)-3-phenyl-1-azaazulene as a ligand resulted in failure.° UV/VIS of 5 (acetonitrile) : (log e) 244(4.45), 260(4.34, sh), jmax/nm 324(4.27), 446(3.63), 464(3.64), 494(3.39, sh). “ Data were collected at 293 K on a Rigaku AFC5S diÜractometer with graphite monochromated Mo-Ka radiation (k\0.71069 and ”) a 2 kW stationary anode generator. The structure was solved by direct methods. The non-hydrogen atoms were re–ned anisotropically.Crystal data for 3: M\234.30, brown, prismatic crystal C16H14N2 , (0.24]0.46]0.54 mm), orthorhombic a\18.313(3), P21212 (d18), b\7.320(6), c\9.242(6) V \1239(2) Z\4, g ”, ”3, DC\1.256 cm~3, l\0.70 cm~1, F(000)\496. The –nal cycle of full-matrix least-squares re–nement was based on 967 observed re—ections [I[2.00 r(I)] and 220 variable parameters and converged with unweighted and weighted agreement factors of R\0.050 and Rw\ Crystal data for 5: M\302.23, orange, prismatic 0.052.C20H23BN2 , crystal (0.24]0.48]0.64 mm), triclinic a\9.040(8), P1 (d2), b\12.214(4), c\8.039(5) a\102.98(4), b\92.23(7), ”, c\93.28(5)°, V \862(1) Z\2, g cm~3, l\0.63 ”3, Dc\1.164 cm~1, F(000)\324. The –nal cycle of full-matrix least-squares re–nement was based on 2162 observed re—ections [I[2.00 r(I)] and 301 variable parameters and converged with unweighted and weighted agreement factors of R\0.057 and CCDC reference Rw\0.063.number 440/052. [0.90(4) and 2.79(5) respectively], and that the bond dis- ”, tance of N2wC8 is 1.340(5) which is shortened compared ”, with the correspondng bond distance of the amino-substituted alternant aromatic compound (1.375 These imply that the ”).9 resonance energy of the 1-azaazulene skeleton is moderate, though being smaller than that in alternant conjugation.Study of 5 (Fig. 2) showed the dihedral angle constituted by C1wN2 and B1wN1 bonds through N2wB1 is 175.5(3)°, and furthermore that of B1wN1 and C8wC7 bonds through N1wC8 is 176.4(3)°. That is, the boron atom is almost coplanar with the skeletal atoms of the ligand.In contrast to the amino-hydrogen atom (N2) of 3, the distances between the boron atom and two nitrogen atoms in 5 (N1, N2) are 1.605(4) (B1wN1) and 1.591(4) (B1wN2), respectively. The Aé tetrahedral character of the boron atom is estimated to be 73%.10 The marked structural change of the aromatic ring is evident from comparison of bond distances in 3 and 5.The shortened distance between N1 nitrogen and C8 carbon atoms [1.326(3) would be due to the higher double bond charac- ”] ter between these atoms. The small dihedral angle [1.8(5)°] of C10wN1 and C8wC7 bonds through N1wC8 proves this conjugation as well. Furthermore, upon coordination of the boron atom, the three carbon»carbon bonds (C3wC9, C4wC5, C6wC7) are shortened (0.047, 0.010, and 0.013 ”, respectively) and two bonds (C3wC4, C5wC6) are elongated (0.024 and 0.016 respectively), the 8-azaheptafulvenyl char- ”, acter being pronounced in the seven-membered ring.The equivalent boron»nitrogen bonds and the marked structural change of the skeleton of 5 are attributable to the easy destruction of non-alternant conjugation in the 1-azaazulene skeleton with low aromatic resonance energy.The corresponding bond distances of 4 diÜer by 5.9% (1.584 and 1.490 respectively) ; furthermore, the nitrogen atom incorporated ”, into the aromatic ring forms the longer coordinate bond.8 Fig. 2 Computer-generated thermal ellipsoid of 5. Some important bond distances and angles (°) not given in the text are as follows : (”) N2wC1\1.361(3), C1wC2\1.385(4), C2wC3\1.434(4), C3w C9\1.403(3), C3wC4\1.424(4), C4wC5\1.362(4), C5wC6\ 1.418(4), C6wC7\1.363(4), C7wC8\1.426(4), C8wC9\1.417(4), C9wN2\1.351(3), C9wN2wC1\107.2(2), N2wC1wC2\ 110.8(2), C1wC2wC3\106.1(2), C2wC3wC9\105.1(2), C3wC9w N2\110.8(2), C4wC3wC9\124.3(3), C3wC4wC5\126.0(3), C4wC5wC6\130.2(3), C5wC6wC7\133.4(3), C6wC7wC8\ 125.7(3), C7wC8wC9\122.6(3), C8wC9wC3\137.5(3), C8wC9w N2\111.7(2), N1wB1wC13\109.5(2), N1wB1wC11\111.4(3), C11wB1wC13\116.2(3), N1wB1wN2\94.7(2) 1032 New J.Chem., 1998, Pages 1031»1033In conclusion, this work shows potential utility of nonalternant ligands in sophisticated systems, such as threedimensionally interacting organometallics using both the r and p type coordinations with the metals.Acknowledgements work was supported by The Research Fund Grant-in- This Aid for Exploratory Research (No. 09874134) and that for Scienti–c Research on Priority Areas (A) (No. 10146235) from the Ministry of Education, Science, Sports and Culture, Japan. References 1 M. J. S. Dewar and C. de Llano, J. Am. Chem. Soc., 1965, 91, 789. 2 Y. Sugihara, R. Hashimoto, H. Fujita, N.Abe, H. Yamamoto, T. Sugimura and I. Murata, J. Chem. Soc., Perkin T rans. 1, 1995, 22, 2813. 3 For an example of p-type coordination molecules of azulene, see M. R. Churchill and J. Wormald, J. Chem. Soc., Chem. Commun., 1968, 1033. 4 E. S. Pysh and N. C. Yang. J. Am. Chem. Soc., 1963, 85, 2124. 5 (a) Y. Sugihara, T. Yagi, I. Murata and A. Imamura, J. Am. Chem. Soc., 1992, 114, 1479; (b) Y.Sugihara, R. Miyatake, I. Murata and A. Imamura, J. Chem. Soc., Chem. Commun., 1995, 1250; (c) Y. Sugihara, K. Takakura, T. Murafuji, R. Miyatake, K. Nakasuji, M. Kato and S. Yano, J. Org. Chem., 1996, 61, 6829. 6 (a) K. Yamane, K. Fujimori, J.-K. Sin and T. Nozoe, Bull. Chem. Soc. Jpn., 1977, 50, 1184; (b) N. Abe, Y. Fukazawa, Y. Hirai, T. Sakurai, K. Urushido and A. Kakehi, Bull. Chem. Soc. Jpn., 1992, 65, 1784. 7 G. Klebe and D. Tranqui, Inorg. Chim. Acta, 1984, 81, 1. 8 The increase in the electron density of C2 causes an up–eld shift, but which is beyond the scope of the empirical rule. For the up–eld shift of the a carbons of the nitrogen atom in the protonated pyridines, see R. J. Pugmire and D. M. Grant, J. Am. Chem. Soc., 1968, 90, 699. 9 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»S19. 10 S. Toyota and M. Bull. Chem. Soc. Jpn., 1992, 65, 1832. O1 ki, Received in Cambridge, UK, 1st June 1998; L etter 8/05547E New J. Chem., 1998, Pages 1031»1033 1033

ISSN:1144-0546    

DOI:10.1039/a805547e

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

L e t t e r Reaction of halfsandwich iridium polychalcogenide complexes with dimethyl acetylenedicarboxylate. Molecular structure of Cp*Ir [S2C2(COOMe)2 ] Guo-Xin Jin,a Max Herberhold*,b and Arnold L. Reingoldc a Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, Changchun 130022, P. R. China b L aboratorium Anorganische Chemie, Bayreuth, D-95440 Bayreuth, Germany fué r Universitaé t c Department of Chemistry, University of Delaware, Newark, Delaware 19716, USA The pentamethylcyclopentadienyl iridium complexes (E\S, n\4, 5 or 6; E\Se, n\2 or 4; Cp*Ir(PMe3)(En) E\Te, n\2) react with dimethyl acetylenedicarboxylate to give compounds Cp*Ir(PMe3)[E2C2(COOMe)2] which tend to lose the trimethylphosphine ligand ; the molecular structure of the dithiolene derivative, has been determined.Cp*Ir[S2C2(COOMe)2], The size of the chelating cyclo-oligosul–do ligand in pentamethylcyclopentadienyl iridium halfsandwich complexes, (n\4, 5 or 6), is variable.1,2 Dechalcogena- Cp*Ir(PMe3)(Sn) tion of the sulfur-rich complex with tertiary Cp*Ir(PMe3)(S6) phosphines leads reversibly to the ring-shrink derivatives and (1a), which take up Cp*Ir(PMe3)(S5) Cp*Ir(PMe3)(S4) chalcogen from polysul–des (xB10) to reform (NH4)2Sx (Scheme 1).1 A similar equilibrium has been Cp*Ir(PMe3)(S6) observed between (1b) and Cp*Ir(PMe3)(Se4) Cp*Ir(PMe3)- however, the ditelluride complex, (1c) (Se2) ; Cp*Ir(PMe3)(Te2) did not take up additional tellurium.§ Scheme 1 * E-mail: Max-Herberhold=uni-bayreuth.de § Synthesis and spectral data of 1c.A solution containing (0.31 g, 0.65 mmol) and (0.90 g, 0.80 Cp*Ir(PMe3)Cl2 (Bu4 n N)2Te5 mmol) in 20 ml DMF was kept at 60 °C for 2 h, then the solvent was removed in vacuo. The residue was puri–ed by column chromatography (silica, elution with Crystallization from toluene» CH2Cl2).hexane (1 : 3) aÜorded 0.34 g (79%) black needles of 1c. IR (CsI) cm~1; El-MS (70 eV): 660 (M`, 52%), 584 mPvCH3\950 (M` 100%); 1H NMR (300 MHz, d 1.97 (d, 15H, Cp*, [PMe3 , CDCl3) : Hz), 1.82 (d, 9H, Hz); 13C NMR (75 JPH\1.7 PMe3 , JPH\10.0 MHz, d 11.0 98.0 20.7 (d, CDCl3) : (C5Me5), (C5Me5), PMe3 , JPC\ Hz); 31P NMR (121 MHz, d[43.1. 44.3 CDCl3) : In the presence of an excess of dimethyl acetylenedicarboxylate, DMAD, all available cyclooligochalcogenide complexes, or (1a) ; (1b), Cp*Ir(PMe3)(En) [En \S6, S5 S4 Se4 (1c)] uniformly reacted to give ethene»dichalcogeno- Se2; Te2 late derivatives, [E\S (2a), Cp*Ir(PMe3)[E2C2(COOMe)2] Se (2b) or Te (2c)], either in re—uxing toluene or by adding as a potential chalcogen acceptor.Red crystals of 2a»2c PBu3 n were isolated in good yields after column chromatography on silica.î The stable 18-electron complex 2a lost the two-electron ligand to give 3a° by either sulfur-induced phosphine PMe3 î Synthesis and characterization of 2a»c : A solution containing 0.3 mmol of the polychalcogenide complex [e.g. (1a)] Cp*Ir(PMe3)(S4) and a twofold excess of DMAD in 60 ml of toluene was stirred at 60 °C for 3»5 h.Puri–cation by column chromatography over silica (elution with and recrystallization from (3 : 1) CH2Cl2) hexane»Et2O at [78 °C gave the bis(carbomethoxy)dithiolate complex [e.g.(2a)] in 70»80% yield. 2a (Red crys- Cp*Ir(PMe3)[S2C2(COOMe)2] tals, yield 80%), IR (CsI) : and 1681 cm~1, mCO\1714 mPvCH3 \964 cm~1; EI-MS (70 eV): m/z 534 100%), 503 (M` (M`[PMe3 , 35%); 1H NMR d 1.76 (d, 15H, Cp*, [PMe3[OMe, (CDCl3) : Hz), 1.59 (d, 9H, Hz), 3.72 (s, 6H, J(P, H)\2.0 PMe3 , J(P, H)\10.8 OMe); 13C NMR d 8.6 96.4 14.3 (d, (CDCl3) : (C5Me5) : (C5Me5) ; Hz), 52.5 (OMe); 31P NMR [33.0. 2b PMe3 , J(P, C)\41.2 (CDCl3) : (Red crystals, yield 83%), IR (CsI) : 1713 and 1685 cm~1, mCO mPvCH3\ 962 cm~1; EI-MS (70 eV): m/z 704 (M`, 3%) 628 (M`[PMe3 , 10%), 485 78%), 156 100%); FD-MS: m/z 704 (Cp*IrSe3 `, (SePMe3 `, (M`) ; 1H NMR d 1.77 (d, 15H, Cp*, Hz), 1.68 (CDCl3) : J(P, H)\2.0 (d, 9H, Hz), 3.79 (s, 6H, OMe); 13C NMR PMe3 , J(P, H)\11.3 d 9.3 96.1 16.6 (d, (CDCl3) : (C5Me5), (C5Me5) ; PMe3 , J(P, C)\34.9 Hz), 52.4 (OMe), 144.1 183.3 31P NMR d (C2), (CO2Me); (CDCl3) : [32.2. 2c (Dark-red crystals, yield 73%), IR (CsI) : and mCO\1708 1692 (sh) cm~1, cm~1; EI-MS (70 eV): m/z 802 (M`, mPvCH3\953 30%) 726 18%), 584 100%); 1H NMR (M`[PMe3 , (Cp*IrTe2 `, d 1.88 (d, 15H, Cp*, Hz), 1.74 (d, 9H (CDCl3) : J(P, H)\2.0 PMe3 , Hz), 3.66 (s, 6H, OMe); 13C NMR d 10.2 J(P, H)\10.0 (CDCl3) : 112.3 19.1 (d, Hz), 52.3 (C5Me5), (C 5Me5), PMe3 , J(P, C)\45.9 (OMe); 31P NMR d [42.4.(CDCl3) : ° Characterization of 3a (dark-red prismatic crystals). IR (CsI) : mCO\ 1730 and 1689 cm~1; EI-MS (70 eV): m/z 534 (M`, 100%), 503 (M`[OMe, 19%), 502 (21%), 414 14%); 1H NMR (Cp*IrS2 `, d 1.84 (s 15H, Cp*), 3.76 (s 6H, OMe): 13NMR d (CDCl3) : (CDCl3) : 8.9 93.4 52.2 (OMe).(C5Me5), (C5Me5) ; New J. Chem., 1998, Pages 1035»1036 1035abstraction or prolonged thermolysis in re—uxing toluene (Scheme 2). The phosphine-free iridium»diselenolene and »ditellurolene analogues, [M\Se Cp*Ir[E2C2(COOMe)2] (3b) or Te (3c)], were obtained similarly, although not free of side-products ; the EI-MS spectra contained the molecular ions of 3b and 3c, respectively, as the peaks of highest mass.Cyclopentadienyl metal ethene»dithiolate complexes of the lighter Group 9 homologues, CpM[S2C2(COOMe)2] (M\Co3, Rh4,5) have been synthesized earlier by a one-pot reaction of the cycloocta-1,5-diene precursors, CpM(C8H12), with DMAD and elemental sulfur.Scheme 2 Fig. 1 Molecular geometry of 3a in the crystal. Selected bond lengths (pm) and angles (°) : Ir(1)wS(1) 223.27(18), Ir(1)wS(2) 223.42(15), Ir(1)wCp*(center) 181.9, S(1)wC(11) 173.6(6), S(2)wC(12) 174.4(5), C(11)wC(12) 134.1(7), S(1)wIrwS(2) 87.70(5), Ir(1)wS(1)wC(11) 105.9(2), Ir(1)wS(2)wC(12) 105.91(19) The molecular structure of (3a)“ Cp*Ir[S2C2(COOMe)2] combines two essentially planar –ve-membered rings, Cp* and which are arranged perpendicular to each other (dihe- IrS2C2 , dral angle 87.7°).The IrwS bond lengths (223.35 pm av. in 3a) are signi–cantly shorter than in the 18-electron oligosul–do precursor complexes [n\6, 235.8(2) and Cp*Ir(PMe3)(Sn) 234.5(3) pm;1 n\4 (2a),2 237.6(2) and 237.2(2) pm], and in [238.0(2) and 237.0(2) pm], but compar- Cp*Ir(PMe3)(SH)2 6 able with the corresponding IrwS bond distances in the benzene-1,2-dithiolate [225.3(4) and 224.4(4) Cp*Ir[S2C6H4] pm]7 and in the iridathiabenzene complex Cp*Ir[SC4H2Me2] [220.3(2) pm].8 This indicates that the metal in 3a is integrated into a ììpseudo-aromaticœœ dithiolene ring involving considerable IrwS multiple bonding, probably as a result of sulfur-to-metal electron transfer (as discussed for benzene-1,2- dithiolate complexes7,9,10). The intra-annular SwC and CwC distances in 3a are observed in the range expected for complexes containing ethene-1,2-dithiolate ligands.Intermolecular bonding interactions are absent. “ Crystal structure of 3a: dark red block, 0.20]0.40]0.60 mm, monoclinic, space group a\11.849(2), b\11.491(2), P21/n; c\15.100(3) b\111.86(3)°, U\1908.1(6) Z\4, g ”, ”3, DC\1.858 cm~3, F(000)\1032. Siemens P4 diÜractometer Mo-Ka radiation (k\0.71073 293(2) K, 2h range 4.0»52.0°, index ranges ”), 0OhO14, 0OkO14, [18OlO17; 4149 re—ections collected, 3754 independent 3072 observed [F2[2.06(F2)], 209 (Rint\0.0306), re–ned parameters. Siemens, SHELX97 direct methods, (SHELXS97).Final R indices (observed data), R1\0.0292, wR2\0.0637, goodnessof- –t 1.161 ; max./min.residual electron density 0.80/[0.83 e ”~3. CCDC reference number 440/057. References 1 M. Herberhold, G.-X. Jin and A. L. Rheingold, Chem. Ber., 1991, 124, 2245. 2 M. Herberhold, G.-X. Jin and W. Milius, Chem. Ber., 1995, 128, 557. 3 H. Boennemann, B. Bogdanovic, W. Brijoux, R. Brinkmann, M.Kajitani, R. Mynott, G. S. Natarajan and M. G. Samson, Transition-metal-catalyzed synthesis of heterocyclic compounds, in Catalysis in Organic Reactions, ed. J. R. Kosak, Marcel Dekker, New York 1984, pp. 31»62. 4 M. Kajitani, T. Suetsugu, R. Wakabayashi, A. Igarashi, T. Akiyama and A. Sugimori, J. Organomet. Chem., 1985, 293, C15. 5 M. Kajitani, T. Suetsugu, T. Takagi, T. Akiyama, A. Sugimori, K. Aoki and H. Yamazaki, J. Organomet. Chem., 1995, 487, C8. 6 D. P. Klein, G. M. Kloster and R. G. Bergman, J. Am. Chem. Soc., 1990, 112, 2022. 7 R. Xi, M. Abe, T. Suzuki, T. Nishioka and K. Isobe, J. Organomet. Chem., 1997, 549, 117. 8 J. Chen, L. M. Daniels and R. J. Angelici, J. Am. Chem. Soc., 1990, 112, 199. 9 D. Sellmann, M. Geck, F. Knoch, G. Ritter and J. Dengler, J. Am. Chem. Soc., 1991, 113, 3819; D. Sellmann, M. Geck, F. Knoch and M. Moll, Inorg. Chim. Acta, 1991, 186, 187. 10 M. R. Churchill and J. P. Fennessey, Inorg. Chem., 1968, 7, 1123. Received in Basel, Switzerland, 3rd July 1998; L etter 8/06085A 1036 New J. Chem., 1998, Pages 1035»1036

ISSN:1144-0546    

DOI:10.1039/a806085a

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

L e t t e r A new tetraoxa-tetraaza macrobicyclic anthracenyl luminescent receptor : control of the receptor-substrate stoichiometry Cinzia Di Pietro,a Giovanni Guglielmo,a Sebastiano Campagna,*,a Mario Diotti,b Amedea Manfredib and Silvio Quici*,b a Dipartimento di Chimica Inorganica, Chimica Analitica e Chimica Fisica, di Messina, Universita` I-98166 Messina, Italy b Centro CNR and Dipartimento di Chimica Organica e Industriale, di Milano, I-20133 Universita` Milano, Italy The luminescent receptor 1 incorporating anthracenyl groups as luminophores and a tetraoxa-tetraaza macrobicyclic receptor as the binding subunit has been synthesized and its interactions with CaII, investigated by —uorescence measurements, are reported.The receptor-substrate 1-Ca system is an example of multistability at the molecular level, triggered by a reversible modi–cation of the receptor ability and evidenced by luminescence.The change of the luminescence output of a molecular receptor in the presence of a suitable substrate is an extremely useful tool for investigating receptor-substrate interactions.1 This has stimulated great interest in the design of supramolecular receptors whose luminescence properties can be modi- –ed by guest binding.2 Moreover, such systems have recently received additional interest after it was pointed out that they could indeed open the way, together with other systems in which the luminescence output can be reversibly modi–ed by means of an external trigger, to the development of sensors and logic devices operating at the molecular level.3 A fundamental key towards this direction is the development of new species featuring multistability, that is, capable of being switched, by taking advantage of a perturbation, between two diÜerent stable states.3 Here we present a new tetraoxa-tetraaza macrobicyclic ligand, 1, bearing two anthracenyl groups covalently bonded, through a three-carbon atom chain, on two opposite nitrogen atoms (Fig. 1), and we report its complexation behaviour with Ca2` studied by luminescence spectroscopy and the possibility of controlling the receptor-substrate stoichiometry by an external perturbation. Compound 1 exhibits the typical anthracene luminescence in acetonitrile solution (luminescence quantum yield, '\0.03). The luminescence signi–cantly increases ('\0.11) upon addition of salt (Fig. 2). Titration of Ca(ClO4)2 … 4H2O the luminescence intensity vs. Ca2` concentration is –tted (Fig. 2, inset) by a non-linear equation employed by Valeur Fig. 1 Structural formula and schematic representation of 1. In the schematic representation, the rectangles stand for the coordinating nitrogens and co-workers4 for 1 : 1 receptor-substrate adducts.The same 1 : 1 receptor-substrate adduct ratio was obtained by using the Hyperquad programme.5 The stability constants calculated by the two diÜerent methods are in fair agreement (Valeurœs equation gave 4.5]105 M~1, correlation coefficient\0.997 ; Hyperquad gave 3.0]105 M~1, standard deviation\0.025). This behaviour agrees with a partial quenching of the anthracene luminescence in 1 due to reductive excited-state electron transfer involving the nitrogen lone pairs.§ The experimental results as well as CPK (Corey»Pauling»Koltun) models indicate that calcium ions can be coordinated inside the tetraoxatetraaza macrobicyclic cleft of 1, giving rise to a complex 1 Æ Ca with a 1 : 1 receptor-Ca2` stoichiometry. In this complex, the nitrogen lone pairs are no longer available to quench the anthracene excited state and luminescence intensity increases.The shape of the absorption spectrum of 1 remained unchanged upon cation addition, thus indicating that there is no signi–cant chromophore perturbation in the ground state. Luminescence enhancement of 1 was also observed upon protonation. Titration of 1 by followed by CF3COOH, changes in luminescence intensity (not shown), indicated that one equivalent of acid is enough to quantitatively protonate both of the two nitrogens connected by the 2-benzylpropylene chain of 1 (emission quantum yield of 1 in the presence of one equivalent of 0.06).î Addition of Ca2` to an ace- CF3COOH, tonitrile solution of 1 in the presence of one equivalent of acid leads to a further anthracene luminescence enhancement, but the process cannot be –tted by either Valeurœs equation used for a 1 : 1 adduct or the Hyperquad programme on assuming a 1 : 1 substrate-receptor ratio.This behaviour can be explained by considering that the preferred coordination sphere of Ca2` with crown ethers is usually made by eight donor atoms,8 while the protonated species, 1H has only six binding sites available for complexation.Furthermore, Ca2` tends probably to be pushed out from the cleft by repulsive interaction with H`. Under these conditions, two molecules of § Anthracene emission quenching by amine nitrogens in multicomponent systems is well-documented.2 î Because of their proximity, in fact, these nitrogens cooperate to coordinate a proton. The same result was obtained in a related bis(azacrown) species of this family6 and is reminiscent of the protonation behaviour of preorganized base sites in catenands containing 1,10-phenanthroline units.7 New J.Chem., 1998, Pages 1037»1039 1037Fig. 2 Luminescence enhancement of 1 upon Ca2` addition and (inset) –tting of the luminescence intensity vs. Ca2` concentration by employing Valeurœs equation for 1 : 1 adducts.Solvent, acetonitrile ; concentration of 1, 5]10~5 M; excitation wavelength, 280 nm; emission wavelength, 410 nm 1H are needed to complete the coordination sphere of the calcium ion. Indeed, the titration is perfectly –tted by Hyperquad when a 2 : 1 receptor-substrate ratio is assumed (Ka\ M~2, standard deviation\0.15). It should also be 7.9]1010 noted that under the conditions used in the experiments, [90% of the luminescence enhancement is obtained when 0.5 equiv.of Ca2` ions are added with respect to the receptor. The resulting complex can be now labelled as 1H … Ca … 1H, with the calcium cation playing the role of an assembling chromophore species.° To explore the possibility of moving reversibly from 1 Æ Ca to 1H … Ca … 1H, we alternatively added one equivalent (with respect to 1) of and 1,4-diazabicyclo[2.2.2]octane CF3COOH (DABCO) into an acetonitrile solution containing 1 (5]10~5 M) and Ca2` (3]10~5 M).By independent experiments, it could be noted the solution containing 1 and Ca2` in the above mentioned concentrations exhibits a luminescence intensity output (at a –xed excitation wavelength) equivalent to 70% of the luminescence intensity of the same solution in the presence of 5]10~5 M, (Fig. 3). By taking CF3COOH advantage of the diÜerence in luminescence intensity exhibited by the two adducts under such a situation, it can be seen that alternate addition of one equivalent of acid and base shifts the system from the luminescence value typical of the 1 : 1 adduct Fig. 3 Reversible shift between 1 Æ Ca and 1H … Ca … 1H. Total concentration of 1 is 5]10~5 M; concentration of Ca2` is 3]10~5 M. Square and triangles are the experimental luminescence intensities of 1 in the presence and absence of acid, respectively. A is the luminescence output change upon addition of one equivalent (with respect to 1) of B is the luminescence output change upon suc- CF3COOH. cessive addition of one equivalent of DABCO ° The calculated value of the stability constant for 1H … Ca … 1H should be considered with great care, because of its extremely high value.Fig. 4 Schematic representation of the reversible assembling processes. The grey rectangles represent the protonated (deactivated) nitrogens. The sphere stands for the Ca2` ion to that of the 2 : 1 adduct (Fig. 3), thus demonstrating that the system can be reversibly moved between two arrangements (bistability). The situation is schematized in Fig. 4. The system reported is an example of multistability at the molecular level, triggered by a reversible modi–cation of the receptor ability upon protonation and evidenced by luminescence. The two diÜerent states of the system are the two (1 : 1 and 2 : 1) receptor-substrate adducts which, under suitable concentration conditions, exhibit diÜerent luminescence intensities.“ At the moment we are engaged in the synthesis of receptors bearing chromophores diÜerent from anthracenes (e.g., naphthalenes) to explore the possibility of driving interreceptor photoinduced energy and/or electron transfer in 2 : 1 adducts made of mixed receptors.Experimental The preparation of 1 was easily achieved by condensation of the corresponding tetraoxa-tetraaza macrobicyclic bisamine9a with 9-(3-bromopropyl)anthracene9b carried out in at CH3CN re—ux for 6 days and solid as base. Selected data for Na2CO3 1 are : 1H NMR (300 MHz, d 1.80»2.00 (m, 1H), 2.20» CDCl3) 3.10 (m, 34H), 3.30»3.80 (m, 16H), 7.01»7.22 (m, 5H), 7.30»7.53 (m, 8H), 7.96 (d, 4H, J\8.3 Hz), 8.25 (d, 4H, J\8.6 Hz), 8.30 (s, 2H).FAB-MS; found m/z 937 [M]Na]`, 915 [M]1]`; calcd for m/z 937. Anal. calcd for C60H74N4O4Na`: C, 78.72 ; H, 8.16 ; N, 6.12 ; found: C, 78.10 ; H, C60H74N4O4 : 8.20 ; N, 6.0%. Luminescence quantum yields have been obtained with the optically dilute method,10 using anthracene in deoxygenated ethanol as standard ('\0.27).11 Details on Valeurœs equation employed to –t the data are available as supplementary material from the author.References 1 V. Balzani and F. Scandola, Supramolecular Photochemistry, Ellis Horwood, Chichester, 1991, chap. 8. 2 (a) A. P. de Silva, H. Q. N. Gunaratne and C. P. McCoy, Nature (L ondon), 1993, 42, 364; (b) Fluorescent Chemosensors for Ion and Molecule Recognition, ed.A. W. Czarnik, ACS Symposium Series “ The switching system represented in Fig. 4 could be considered as an example of tetrastability rather than of bistability. Such high levels of stability are very interesting because they can allow us to break out of the stranglehold of binary logic (even though binary logic has a monopoly in current computing). We thank one of the reviewers for having suggested this point. 1038 New J. Chem., 1998, Pages 1037»1039538, American Chemical Society, Washington, DC, 1993; (c) H. Bouas-Laurent, J.-P. Desvergne, F. Fages and P. Marsau, in Frontiers in Supramolecular Organic Chemistry and Photochemistry, eds. H.-J. Schneider and H. Dué rr, VCH, Weinheim, 1991, p. 265; (d) R. Y. Tsien, Chem. Eng. News, 1994, July 18, 34; (e) L.Fabbrizzi and A. Poggi, Chem. Soc. Rev., 1995, 197; ( f ) L. Fabbrizzi, M. Licchelli, P. Pallavicini, A. Perotti, A. Taglietti and D. Sacchi, Chem. Eur. J., 1996, 2, 75; (g) A. P. de Silva, H. Q. N. Gunaratne, T. Gunnlaugsson, A. J. M. Huxley, C. P. McCoy, J. T. Rademacher and T. E. Rice, Chem. Rev., 1997, 97, 1515. 3 (a) J.-M. Lehn, Supramolecular Chemistry, VCH, Weinheim, 1995; (b) S.L. Gilat, S. H. Kawai and J.-M. Lehn, Chem. Eur. J., 1995, 1, 275; (c) J.-P. Collin, P. Gavin8 a and J.-P. Sauvage, Chem. Commun., 1996, 2005; (d) A. Credi, V. Balzani, S. J. Langford and J. F. Stoddart, J. Am. Chem. Soc., 1997, 119, 2679; (e) G. M. Tsivgoulis and J.-M. Lehn, Adv. Mater., 1997, 9, 627; ( f ) A. P. de Silva, T. Gunnlaugsson and C.P. McCoy, J. Chem. Educ., 1997, 74, 53; (g) F. Pina, M. J. Melo, M. Maestri, R. Ballardini and V. Balzani, J. Am. Chem. Soc., 1997, 119, 5556. 4 J. Bourson, J. Pouget and B. Valeur, J. Phys. Chem., 1993, 97, 4552. 5 A. Sabatini, A. Vacca and P. Gans, Coord. Chem. Rev., 1992, 120, 389. 6 S. Quici, A. Manfredi, R. Rossi, S. Campagna, G. Calogero and V. Balzani, Gazz. Chim. Ital., 1997, 127, 107. 7 (a) N. Armaroli, L. De Cola, V. Balzani, J.-P. Sauvage, C. O. Dietrich-Bucherer, J.-M. Kern and A. Bailal, J. Chem. Soc., Dalton T rans., 1993, 3241; (b) N. Armaroli, V. Balzani, L. De Cola, C. Hemmert and J.-P. Sauvage, New J. Chem., 1994, 18, 775. 8 N. S. Poonia and A. V. Bajaj, Chem. Rev., 1979, 79, 389. 9 (a) S. Quici, A. Manfredi and M. Buttafava, J. Org. Chem., 1996, 61, 3870; (b) F. H. C. Stewart, Aust. J. Chem., 1960, 13, 478. 10 J. N. Demas and G. A. Crosby, J. Phys. Chem., 1971, 75, 991. 11 W. R. Dawson and M. W. Windsor, J. Phys. Chem., 1968, 72, 3251. Received in Montpellier, France, 13th February, 1998; Revised m/s received 12th June 1998; L etter 8/01329B New J. Chem., 1998, Pages 1037»1039 1039

ISSN:1144-0546    

DOI:10.1039/a801329b

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

L e t t e r Gold(I) complexes with amine ligands. 3.1 Competition between auriophilic and hydrogen bonding interactions in dimeric species Peter G. Jones* and Birte Ahrens Institut Anorganische und Analytische Chemie, T echnical University of Braunschweig, f ué r Postfach 3329, 38023 Braunschweig, Germany The gold(I) complexes and (pip\piperidine, Cy\cyclohexyl) crystallise as loose (pip)2Au`Cl~ (Cy2NH)AuCl dimers; in the ionic species NwH… … …Cl~ hydrogen bonding is the main secondary interaction, whereas in the neutral complex the NwH… … …Cl hydrogen bonding is weaker but is compensated by an auriophilic interaction. Amine complexes of gold(I) have been little studied and are generally regarded as relatively unstable (in the absence of stabilising ligands as phosphines); this can be rationalized in terms of incompatibility of the soft metal centre with the hard nitrogen donor.Additional stabilisation may, however, be provided by secondary interactions such as auriophilic interactions (recognisable by formally non-bonded gold»gold distances of ca. 2.7»3.4 or NwH… … …X hydrogen bonds. ”)2 The stabilizing eÜect of hydrogen bonds was postulated for the imine complex in which [Au(NHxCMe2)2](CF3SO3),3 the cationic gold moieties, despite disadvantageous electrostatic eÜects, form chains with short Au… … …Au distances [3.1663(5), 3.1705(5) and the tri—ate (tri—uoro- ”], methanesulfonate) anions are involved in NwH… … …O hydrogen bonds.Stabilization through hydrogen bonding can also be assumed for in which pairs of ammine»gold(I) [Au(NH3)2]Br,4 cations are linked through auriophilic interactions [Au… … …Au 3.414(1) with additional NwH… … …Br~ hydrogen bonds, ”] forming a layer structure.In isocyanidegold(I) thiosalicylates,5 hydrogen bonding and auriophilic interactions are observed, but the former does not involve the donor atoms at gold; chains are built up via alternating secondary bonding types.We have begun a systematic study of gold(I) complexes with aliphatic amine ligands, generally bearing an NwH function, as a potential donor group for stabilizing hydrogen bonds. We have not been able to obtain stable compounds with tertiary amine ligands, and attribute this to the lack of hydrogen bonds. Steric eÜects (ìovercrowded nitrogenœ) could represent an alternative explanation (see below), but are difficult to quantify ; similar considerations apply, in the case of ionic complexes, to Coulombic interactions.We recently reported the structure of (pyrr)4Au3Cl36 (pyrr\pyrrolidine), in which the Au… … …Au contacts were rather long and thus presumably weak [3.2041(7), 3.5834(4) whereas the hydrogen bonds were more signi–cant ”], [N… … …Cl~ 3.179, 3.284(6) Here we report the structures of ”].two dimeric species, obtained from (tht)AuCl and the corresponding amine: previously synthesised by us,7 (pip)2Au`Cl~, and (pip\piperidine, Cy\cyclohexyl). These (Cy2NH)AuCl reveal an interesting balance between the two types of secondary interaction. with L\piperidine and LAuCl with L2Au`Cl~ L\dicyclohexylamine: (tht)AuCl (160 mg, 0.5 mmol) was dissolved in neat amine (5 ml).The solution was stirred for 1 h with exclusion of light at room temperature; a white precipitate formed. Light petroleum was added and after cooling for 1 h at [18 °C the precipitate was –ltered oÜ and recrystallized from dichloromethane»light petroleum. Crystals were obtained by diÜusion of light petroleum into dichloromethane solution.NMR spectra were measured in solution, CDCl3 with TMS as internal standard.§ In (Fig. 1) an association of two cations and (pip)2Au`Cl~ two anions about a twofold axis is observed, in which chloride anions accept two hydrogen bonds from the NwH groups [N… … …Cl~ 3.108(6), 3.122(7) The Au… … …Au contact is long ”]. [4.085(2) and can represent at best a very weak interaction ; ”] this may be attributable to the positive charge associated with the gold moieties.The NwAuwN axes are rotated with respect to each other by 39.6(3)°. In the uncharged complex dimers are also (Cy2NH)AuCl observed (Fig. 2), but with inversion symmetry, making the NwAuwCl axes antiparallel to each other. They have shorter and presumably stronger Au… … …Au interactions [3.2676(14) ”], whereas the N… … …Cl distances are longer at 3.391(8) Thus, ”.the reduction of charge of the gold species from ]1 to 0 makes the hydrogen bonding weaker, as expected, but the auriophilic interactions stronger. It remains to be seen if this is a general eÜect, but the same trend can also be recognised in the neutral complex (pip)AuCl 7 [Au… … …Au 3.301(5), N… … …Cl 3.346, 3.580 which forms loose tetramers with secondary ”], § Bis(piperidine)gold(I) chloride. Yield 91%, dec.[86 °C. 1H NMR (200.13 MHz, d 1.73 (m, 4 H, 3.25 (m, 4 H, CDCl3) : b-CH2), a-CH2). 13C NMR (50.32 MHz, 23.41 (b- 26.52 (c- 53.50 CDCl3) CH2 ), CH2), (a- EI-MS, m/z 84 (100%, [M-Au-Cl-H]). Anal. calcd for CH2). (402.72) : C, 29.82 ; H, 5.51 ; N 6.96. Found: C, 29.34 ; C10H22AuClN2 H, 5.35 ; N, 7.06%.Chloro(dicyclohexylamine)gold(I). Yield 75%, dec.[114 °C. 1H NMR (200.13 MHz, 1.26»1.84 (16 H, 2.12, 2.40 [m, 4 CDCl3) CH2 ), H, 2-, (eq.)], 2.99 (m, 2 H, CH). 13C NMR (50.32 MHz, 2@-CH2 25.05 (3-C), 35.50 (4-C), 34.82 (2-C) 58.73 (1-C). EI-MS, CDCl3) m/z\181 (16%, [M-Au-Cl]). IR (KBr) m(AuCl): 345 cm~1 (w). Anal. calcd for (413.74) : C, 34.84 ; H, 5.60 ; N, 3.39. Found: C12H23AuClN C, 34.86 ; H, 5.55 ; N, 3.34%.X-Ray structure determinations. Data were measured using Mo-Ka radiation. Absorption corrections were based on w-scans. Structures were re–ned anisotropically on F2 using all re—ections (program SHELXL-979). Hydrogen atoms were included using a riding model except for those bonded to nitrogen, which were re–ned ì freely œ with restrained NwH bond lengths. Crystal data: (pip)2Au`Cl~: M\402.71, monoclinic, C2/c, a\21.346(5), C10H22AuClN2 , b\6.480(2), c\19.630(5) b\111.54(3)°, V \2525.8(13) ”, ”3, Z\8, l\11.8 mm~1, T \[130 °C, Stoe STADI-4 diÜractometer, 3568 re—ections, 2228 unique, wR2 0.067, R1 0.030.(Cy2NH)AuCl: M\413.73, triclinic, P([1), a\8.306(3), C12H23AuClN, b\8.978(2), c\10.132(3) a\68.24(2), b\85.17(3), c\77.71(3)°, ”, V \685.7(3) Z\2, l\10.9 mm~1, T \[100 °C, Siemens P4 ”3, diÜractometer, 2399 re—ections, 2336 unique, wR2 0.088, R1 0.036.CCDC reference number 440/051. New J. Chem., 1998, Pages 1041»1042 1041Fig. 1 Structure of the dimer of in the crystal. Ellip- (pip)2Au`Cl~ soids represent 50% probability levels ; H atom radii are arbitrary. Bond lengths and angles at gold: AuwN1 2.050(6) AuwN2 ”, 2.055(6) N1wAuwN2 179.0(3)° ”, Fig. 2 Structure of the dimer of in the crystal. Ellip- (Cy2NH)AuCl soids represent 50% probability levels ; H atom radii are arbitrary. Bond lengths and angles at gold: AuwN 2.077(7) AuwCl 2.266(2) ”, NwAuwCl 177.7(2)° ”, bonds similar to those in the dimers; the con- (Cy2NH)AuCl siderably smaller size of the piperidine ligands suggests also that steric eÜects are not of central importance in determining either stability or the strengths of other secondary interactions.A recent theoretical study of copper(I) complexes8 has predicted a structure analogous to for (Cy2NH)AuCl (NH3)CuCl, and presumably also re—ects the relative contributions of attractive CuI… … …CuI interactions and hydrogen bonding. We conclude that the contributions of auriophilic and hydrogen bonding are likely to stabilise gold(I) complexes of many protic amines, and thus to render invalid a rigid application of the hard»soft principle ; the relative contributions will depend on the charge of the gold complex.Further investigations (to be published) have provided preliminary con–rmation of this supposition. Acknowledgements thank the Fonds der Chemischen Industrie (Frankfurt) for We –nancial support.References 1 P. G. Jones and B. Ahrens, Z. Naturforsch. B, 1998, 53, 653. 2 H. Schmidbaur, Gold Bull., 1990, 23, 11; Chem. Soc. Rev., 1995, 391. 3 J. Vicente, M.-T. Chicote, M.-D. Abrisqueta, R. Guerrero and P. G. Jones, Angew. Chem., 1997, 109, 1252. 4 D. M. P. Mingos, J. Yau, S. Mentzer and D. J. Williams, J. Chem. Soc., Dalton T rans., 1995, 319. 5 W. Schneider, A. Bauer and H. Schmidbaur, Organometallics, 1996, 15, 5445. 6 P. G. Jones and B. Ahrens, Chem. Ber./Recueil, 1997, 130, 1813. 7 J. J. Guy, P. G. Jones, M. J. Mays and G. M. Sheldrick, J. Chem. Soc., Dalton T rans., 1977, 8. 8 X.-Y. Liu, F. Mota, P. Alemany, J. J. Novoa and S. Alvarez, Chem. Commun., 1998, 1149. 9 G. M. Sheldrick, SHELXL-97, Program for Re–ning Crystal Structures, University of Goé ttingen, Goé ttingen, 1997. Received in Basel, Switzerland, 15th June 1998; L etter 8/05587D 1042 New J. Chem., 1998, Pages 1041»1042

ISSN:1144-0546    

DOI:10.1039/a805587d

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

L e t t e r High pressure NMR —ow cell for the in situ study of homogeneous catalysis Jonathan A. Iggo,* Darren Shirley and Nicola C. Tong Department of Chemistry, Donnan and Robert Robinson L aboratories, University of L iverpool, PO Box 147, L iverpool, UK L 69 7ZD A novel, convenient high pressure NMR —ow cell for the in situ study of homogeneous catalysis capable of operating over the pressure range 1 to 200 bar and temperature range [40 to 175 °C is described ; importantly, the inlet gas passes through the reaction solution ensuring good gas»solution mixing and maintaining the concentration of reacting gas in the solution thus preventing starvation of the reaction of gaseous reagents ; the cell has been used to study the reaction of a rhodium»ruthenium bimetallic complex with CO.The characterisation of the processes occurring in a homogeneously catalysed reaction can be of great importance both to the efficient operation of existing processes and to the development of new and/or improved processes. It has been recognised that in the in situ spectroscopic study of such reactions not only must the gas over-pressure be controlled but, crucially, the concentration of dissolved gases must be maintained.Thus, for example, the BP»Monsanto reaction will consume most of the dissolved CO in a sealed sample in a few minutes resulting in CO starvation leading to deactivation and decomposition of the catalyst.1 Several in situ high pressure NMR cells have been described and used to obtain valuable insight into organometallic chemistry under high gas pressure.2h8 Of these, the ììsapphire tubeœœ or Roe cell has proved extremely popular since it is simple and convenient to use.3,9,10 However, it suÜers from small sample volumes and gas»solution mixing is only possible by physically shaking the tube containing a high pressure of gas, a process that must be repeated frequently during the experiment if the concentration of dissolved gas is to be maintained.The small gas head space can result in large pressure drops in a reacting system; thus gas starvation of the reaction is an important concern. The problem of gas»solution mixing has been addressed by the elegant designs of Jonas and Vander Velde4 and of Merbach and co-workers5 that use ìì—ip-—opœœ stirrers essentially similar to that initially employed by Whyman and co-workers11 to achieve mixing in high pressure IR cells, and by Rathke et al.6 whose design circulates the reaction solution between a conventional autoclave and a high pressure NMR cell.Whilst these designs are eÜective solutions to the problem of gas»solution mixing, our experience is that high pressure NMR cells built inside metal pressure vessels require great skill in their design, maintenance and operation.2 A simpler system that combines the ease of use of Roeœs sapphire tube with good gas»solution mixing and a large sample volume is shown in Fig. 1 and 2. This cell uses a sapphire tube supported between, and sealed using elastomer O-rings to, two titanium —anges. The maximum operating pressure of this arrangement is determined by the elastomer used, the tube internal diameter and the thickness of the sapphire tube walls ; an i.d. of 1.06 cm and wall thickness of 0.081 cm gives a theoretical burst pressure in excess of 600 bar at 25 °C and has been tested by us to 190 bar.Our design places the —ow cell inside an outer aluminium sleeve (o.d. 7 cm length 42 cm), Fig. 2, protecting the sapphire tube from mechanical damage and allowing the use of relatively thin walled tubes thus maximising sample volume and hence signal-to-noise ratio.The dimensions of the outer sleeve are close to those of a standard wide-bore Bruker NMR probe and allow our cell to be used inside a standard Bruker widebore shim set. A notable feature of the design is that the inlet gas passes through the reaction solution providing a constant stream of gas bubbles through the solution ensuring both good mixing and maintaining the concentration of dissolved gases.Provided that a positive pressure is maintained at the base of the probe with respect to the gas head space, leakage of solvent down the feed lines is not problematic. Fig. 3 shows time lapse series of 1H NMR spectra comparing the dissolution of in H2 containing 5% as an internal standard, for a CDCl3 , CHCl3 sample placed directly inside the high pressure (HP) NMR cell, Fig. 3 (—ow), and for a sample contained in an open topped sample tube placed inside the HPNMR cell, Fig. 3 (contained). In the —ow experiment the pressurizing gas bubbles through the solvent whilst in the contained experiment gas dissolution relies solely on diÜusion into the sample.Fig. 3 (—ow) clearly shows dissolved d ca. 4.7 from the start H2 of the —ow experiment, the intensity of the resonance due to dissolved gas increasing in line with the increase in pressure. Fig. 1 Exploded view of the high pressure in situ NMR —ow cell New J. Chem., 1998, Pages 1043»1045 1043Fig. 2 Assembled view of the high pressure in situ NMR —ow cell In the contained sample, Fig. 3 (contained), however, almost no has dissolved by the end of the pressurisation period.H2 Even after standing 12 h at 90 bar the sample contains only half as much dissolved gas compared with the amount of gas immediately dissolved during pressurisation of the sample placed directly in the —ow cell. The inlet and outlet gas —ow rates are controlled by Brooks mass —ow controllers and are adjusted to ensure a constant pressure inside the cell and to minimise degradation of the resolution.Fig. 4 compares the resolution achieved in the 31P NMR spectrum of a sample of in toluene under 30 P(OMe)3 bar pressure in the absence of gas —ow and with a gas —ow N2 Fig. 3 Time lapse 1H NMR spectra of dissolving in H2 (95 : 5) during pressurisation of the HPNMR —ow cell CDCl3»CHCl3 or contained in an open topped tube inside the —ow cell ; (a)»(b) during pressurisation at ca 1.5 bar min~1 to 90 bar, then (b)»(c) every two hours at 90 bar Fig. 4 The 31P NMR spectrum of a sample of in toluene P(OMe)3 under 30 bar pressure, (a) in the absence of gas —ow and (b) with a N2 gas —ow rate of 10 Nl h~1 rate of 10 Nl h~1 (where 1 Nl\1 l at STP) and shows minimal loss of resolution or signal-to-noise in the sample which has gas bubbling through it during acquisition of the free induction decay.Typical linewidths at half-height are \1 Hz on 2H, \2 Hz on 31P, \4 Hz on 1H. The r.f. circuitry employs conventional tank circuits and the probe achieves 90° pulse lengths, resolution and sensitivities comparable with conventional, commercial broadband probes (using nonspinning samples).One of us has previously reported a high pressure IR study of the reactions of a series of ruthenium»rhodium compounds [(C5H5)Ru(l-CO)2(l-LL)RhX2], MLL\Ph2PCH2PPh2 , or X\Cl or I] with Me2PCH2PMe2 , PPh2C(\CH2)PPh2 , CO.12 These bimetallic complexes were found to fragment to mononuclear species and [(C5H5)Ru(CO)2(g1-LL)]` No intermediates were detected.[Rh(CO)2X2]~. We have now investigated the reaction of [(C5H5)Ru(l- 1, CO)2(l-dcpm)RhCl2], [dcpm\(C6H11)2PCH2P(C6H11)2], with CO using our in situ NMR probe and have found an Fig. 5 In situ 31P NMR spectra of [(C5H5)Ru(l-CO)2(l- 1 reacting with CO. (a) 1 atm (b) 30 atm CO, (c) 90 dcpm)RhCl2] N2 , atm CO, (d) 15 atm CO.See Scheme 1 for assignments 1044 New J. Chem., 1998, Pages 1043»1045Scheme 1 Reaction of 1 with CO as revealed by in situ NMR spectroscopy. The position of equilibrium depends on the applied CO pressure ; cy\C6H11 additional species to those reported previously12 and that the species present in solution (as determined by in situ, high pressure NMR spectroscopy) depend on the applied gas pressure.Fig. 5(a)»(d) shows the in situ 31P NMR spectra of the reaction of 1 in toluene with CO.î The 31P NMR spectrum of 1, Fig. 5(a), shows two doublets of doublets, at d 74.0 due to P1 and at 59.3 due to P2 (see Scheme 1 for the labelling scheme). Under 30 bar of CO new resonances attributable to the metal»metal bond-cleaved dinuclear species 2, P3 and P4 are [M(C5H5)Ru(CO)2N`(l-dcpm)MRh(CO)Cl2N~] visible in addition to those of 1, Fig. 5(b). The resolution achieved in these spectra allowed Hz to be deter- 3JPRh\6 mined. On increasing the CO pressure to 90 bar cleavage of the dinuclear species to mononuclear species 3, P5 and P6, and [(C5H5)Ru(CO)2(g1-dcpm)]` 4 (the latter con–rmed by in situ IR [Rh(CO)2Cl2]~ spectroscopy13) occurs, Fig. 5(c). The reaction is reversible, 3 î 31P NMR (80.3 MHz, spectroscopic data for the complexes: CDCl3) 1, d 74.0 (dd, 47 Hz, 6 [(C5H5)Ru(l-CO)2(l-dcpm)RhCl2] 2JPP 3JRhP Hz), d 59.3 (dd, 47 Hz, 121 Hz); 2JPP 1JRhP [M(C5H5)Ru(CO)2N`(l- 2, d 63.2 (dd, 22 Hz, 6 Hz), d 33.3 dcpm)MRh(CO)Cl2N~] 2JPP 3JRhP (dd, 22 Hz, 123 Hz); 3, d 60.6 2JPP 1JRhP [(C5H5)Ru(CO)2(g1-dcpm)]` (d, 55 Hz), d [8.7 (d, 55 Hz). 2JPP 2JPP and 4 recombine essentially quantitatively to give 1 on reducing the CO pressure to 15 bar, Fig. 5(d), highlighting the importance of studying reactions in situ; an ex situ analysis would have indicated that no change had taken place. This reaction closely follows the scheme proposed by Bearman et al. on the basis of in situ IR measurements for the reaction of with CO although in those [(C5Me5)Ru(l-CO)2(l-LL)RhX2] cases a metal»metal bond-cleaved dinuclear species analogous to 2 could not be identi–ed.12 Acknowledgements thank EPSRC for supporting this work.NCT thanks We EPSRC for a research studentship and DS thanks BP Chemicals Ltd and EPSRC for a CASE studentship. The authors thank Professors B. T. Heaton and A. K. Smith and Doctors A.Poole, D. Taylor, M. J. Taylor and R. Whyman for helpful discussions. Dr. A. Bennett is thanked for experimental assistance. References 1 A. Poole and M. J. Taylor, BP Chemicals Ltd., personal communication. 2 B. T. Heaton, L. Strona, J. Jonas, T. Eguchi and G. A. HoÜman, J. Chem. Soc., Dalton T rans., 1982, 1159; D. T. Brown, T. Eguchi, B. T. Heaton, J. A. Iggo and R. Whyman, J.Chem. Soc., Dalton T rans., 1991, 677. 3 D. C. Roe, J. Magn. Res., 1985, 63, 388. 4 D. G. Vander Velde and J. Jonas, J. Magn. Res., 1985, 71, 480. 5 A. Cusanelli, U. Frey, D. Marek and A. E. Merbach, Spectroscopy Europe, 1997, 9, 22. 6 J. W. Rathke, R. J. Klingler and T. R. Krause, Organometallics, 1991, 10, 1350 and references therein. 7 K. Woelke and J. Bargon, J. Rev. Sci. Instrum., 1992, 63, 3307. 8 C. R. Yonker, T. S. Zemanian, S. L. Wallen, J. C. Linehan, J. A. Franz, J. Magn. Res., Sect. A, 1995, 113, 102. 9 I. T. Horvath and J. M. Millar, Chem. Rev., 1991, 1339. 10 C. J. Elsevier, J. Mol. Cat., 1994, 92, 285. 11 W. Rigby, R. Whyman and K. Wilding, J. Phys. E: Sci. Instrum., 1970, 3, 572; R. Whyman, in L aboratory Methods in Infrared Spectroscopy, eds. R. G. J. Miller and B. C. Stace, Heyden, London, 1972, p. 149; K. A. Hunt, R. W. Page, S. Rigby and R. Whyman, J. Phys. E: Sci. Instrum., 1984, 17, 559. 12 P. S. Bearman, A. K. Smith, N. C. Tong and R. Whyman, Chem. Commun., 1996, 2061. 13 A. K. Smith, N. C. Tong and R. Whyman, unpublished work. Received in Basel, Switzerland, 10th August 1998; L etter 8/06311G New J. Chem., 1998, Pages 1043»1045 1045

ISSN:1144-0546    

DOI:10.1039/a806311g

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

Stereochemical outcome of McMurry coupling Rusli Daik,a W. James Feast,*,a Andrei S. Batsanovb and Judith A. K. Howardb a Interdisciplinary Research Centre in Polymer Science and T echnology and b Crystallography Group, Durham University, South Road, Durham, UK DH1 3L E McMurry coupling of 4-bromoacetophenone using the product of reaction of lithium aluminium hydride and titanium trichloride in THF gives 2,3-bis(4-bromophenyl)-2-butenes in a cis : trans ratio of approximately 9 : 1.This unexpected result was con–rmed by X-ray crystallography and resolves an inconsistency in earlier literature assignments. Coupling of 4-bromobenzophenone under the same conditions gives cis and trans isomers in an approximately 1 : 1 ratio. 2,3-Bis(4-bromophenyl)-2-butene (1) and 1,2-bis(4-bromophenyl)- 1,2-diphenylethene (2) were required as monomers for the synthesis of poly(arylene vinylene)s via Yamamoto1 and Suzuki2 coupling procedures. Since the objective was to study the in—uence of vinylene geometry on polymer properties, pure cis and trans isomers of the monomers were required.We selected McMurry coupling of the appropriate bromoketones for these syntheses.This is a well established method and product stereochemistries have been assigned on the basis of melting points and spectroscopic analysis. Unfortunately the literature concerning such assignments is inconsistent. In this paper we report the synthesis, separation and puri–cation of these compounds. Isomer structures were assigned unambiguously by means of X-ray crystallography. 2,3-Bis(4-bromophenyl)-2-butenes The synthesis of 2,3-bis(4-bromophenyl)-2-butenes (1) (Scheme 1) was carried out by McMurry coupling of 4-bromoacetophenone. The product was recovered as colourless crystals. The 1H NMR spectrum of this product shows two singlet peaks attributed to methyl hydrogens in cis and trans isomers, at 2.12 and 1.84 ppm, with the former peak corresponding to the major product; however, it is not certain which one is which.Inamoto et al. have discussed the stereochemical assignment of 2,3-bis(4-bromophenyl)-2-butenes.3 In an attempt to solve this assignment problem, the chemical shifts of the methyl hydrogens for a series of substituted aa@- dimethylstilbenes were measured. The shielding eÜect experienced by the methyl groups, which arises from the p electron current in the phenyl groups, varies with isomer geometry.Calculations were made in an attempt to correlate the stereochemistry of the various substituted aa@-dimethylstilbenes with the chemical shifts of the methyl hydrogens. On this basis it was suggested that methyl hydrogens in trans isomers absorb at a higher –eld than those in cis isomers. Also it was found Scheme 1 Synthesis of 2,3-bis(4-bromophenyl)-2-butenes via the McMurry reaction * E-mail: w.j.feast=durham.ac.uk that the isomers assigned trans stereochemistry on this basis had higher melting points than the corresponding isomers assigned cis stereochemistry, which is reasonable and was not unexpected.In the case of 2,3-bis(4-bromophenyl)-2-butene, the melting points of cis and trans isomers were recorded at 82 and 146 °C, respectively.At least one group,4 has supported the assignment made by Inamoto et al.3 However, in 1978 McMurry and co-workers reported that the synthesis of 2,3-diphenyl-2-butene from acetophenone gave a 9 : 1 mixture of isomers.5 In the 1H NMR spectra the methyl hydrogen in major and minor products displayed resonances at 2.14 and 1.87 ppm, respectively.According to the earlier assignments,3,4 this was a mixture of 90% cis and 10% trans isomers; however, McMurry and co-workers believed that these assignments were almost certainly wrong since analogous compounds such as stilbene and stilbestrol [3,4-bis(4- hydroxyphenyl)-3-hexene] were known to prefer a trans geometry. This argument was supported by Richardson in 1981.6 Subsequently, Leimner and Weyerstahl7 concluded that for McMurry coupling of alkylaryl ketones cis isomers predominated for sterically undemanding alkyl groups, whereas bulky alkyl groups favoured the trans isomers.These reasonable and reasonably convincing assignments were based on the analysis of 1H NMR spectra. We have been able to provide an unambiguous resolution of this inconsistency in the literature by obtaining the crystal structure of one of the isomers.Fig. 1 presents the crystal structure of the major product which is the cis isomer (1b). Fig. 1 Two independent molecules of cis-2,3-di(4-bromophenyl)-2- butene 1b (showing 50% displacement ellipsoids). Bond distances/”: C(1)wC(2) 1.329(10), C(1)wC(5) 1.491(1), C(2)wC(11) 1.497(10), C(1)wC(3) 1.513(10), C(2)wC(4) 1.542(1), C(17)wC(18) 1.343(10), C(17)wC(27) 1.493(10), C(18)wC(21) 1.486(10), C(17)wC(19) 1.536(10), C(18)wC(20) 1.507(9), BrwC average 1.901(8) New J.Chem., 1998, Pages 1047»1049 1047This isomer shows a methyl hydrogen resonance at 2.12 ppm. The asymmetric unit of 1b contains two molecules (Fig. 1) with similar geometry and molecular conformation, corresponding to a local approximate symmetry (phenyl rings C2 are inclined with respect to the mean ole–nic plane in the same direction, by 54»58°).Both molecules show an insigni–- cant torsion angle around the double bond (5.2 and 2.9°, cf. 3.9° in the parent cis-2,3-diphenyl-2-butene,8 3). This bond is marginally longer [mean 1.336(10) than the standard CxC ”] distance (from X-ray data, 1.31 and essentially the same as ”)9 in 3 [1.343(4) The 1H and 13C NMR spectra are consis- ”].tent with the assigned structure. Thus, it is con–rmed that the formation of the cis isomer is favoured in the synthesis of 2,3-bis(4-bromophenyl)-2-butenes via the McMurry reaction. The cis isomer was found to constitute about 88% of the recovered product mixture before separation of the isomers.Since the isomer with the methyl hydrogen resonance at 2.12 ppm in 1H NMR spectrum turned out to be the cis isomer, this result con–rms Inamotoœs hypothesis that the methyl hydrogens in the trans isomer of 2,3-bis(4-bromophenyl)-2-butene absorb at higher –eld than those in the cis isomer. The 13C NMR spectrum was consistent with the assigned structure. This work provides unambiguous con–rmation of the assignment –rst made by Inamoto and supported by the work of Leimner and Weyerstahl and proves the rather surprising result that the cis isomer is the major product of McMurry coupling of acetophenones. 1,2-Bis(4-bromophenyl)-1,2-diphenylethenes A similar procedure to that described above was used for the synthesis of 1,2-bis(4-bromophenyl)-1,2-diphenylethene (2) from 4-bromobenzophenone, see Scheme 2.The product was a mixture of cis and trans isomers, in this case the selectivity was less and the ìmajorœ isomer constituted 56% of the yield before separation of the isomers. Both isomers were obtained, with a purity in excess of 98% as measured by 1H NMR spectroscopy, by repeated recrystallisation from a 2 : 3 mixture of toluene and ethanol.The assignments of the cis and trans isomers were made on the basis of the crystal structure obtained for a sample of the ìmajorœ isomer, see Fig. 2. Crystalline molecule 2a has an approximate local sym- C2 metry. The planes of the unsubstituted phenyl rings A and B (Fig. 2) form similar dihedral angles of 54.8 and 56.9°, respectively, with the bonding (sp2) planes of C(1) and C(2).For the bromo-substituted rings C and D, the corresponding dihedral angles are 36.4° and 34.6°, respectively. Owing to steric overcrowding, the torsion angle around the central double bond (11.9°) and stretching of this bond [1.357(7) are larger than ”] in 1b, but similar to those observed in other tetraphenylethene derivatives.10 The only unusual intermolecular contact is Br(1)… … …C(12@) 3.35 which is considerably shorter than the ”, standard distance of 3.63 ”.11 In contrast to the case of 2,3-bis(4-bromophenyl)-2-butene, in which the cis isomer forms the major product, the trans isomer was very slightly favoured in the synthesis of 1,2-bis(4- bromophenyl)-1,2-diphenylethene. Fig. 3 presents the 1H Scheme 2 Synthesis of 1,2-bis(4-bromophenyl)-1,2-diphenylethenes Fig. 2 Molecular structure of the major product trans-1,2-bis(4- bromophenyl)-1,2-diphenylethene 2a (showing 50% displacement ellipsoids). Bond C(1)wC(2) 1.357(7), C(1)wC(3) 1.497(7), distances/”: C(1)wC(9) 1.504(8), C(2)wC(15) 1.505(7), C(2)wC(21) 1.488(8), Br(1)wC(6) 1.903(6), Br(2)wC(18) 1.902(6) NMR spectra of both cis- and trans-1,2-bis(4-bromophenyl)- 1,2-diphenylethene. Experimental 4-Bromobenzophenone, 4-bromoacetophenone, lithium aluminium hydride and titanium trichloride were purchased from Aldrich.THF was purchased from BDH. All reagents were used without further puri–cation except 4-bromoacetophenone which was recrystallised from ethanol twice. THF was dried by distillation from sodium metal/sodium benzophenone ketyl radical prior to use. 1H and 13C NMR spectra were recorded using a Varian 400 MHz spectrometer and were referenced to IR Me4Si. Spectra were recorded using a Perkin»Elmer 1600 series FTIR spectrometer. X-Ray crystallography Single-crystal diÜraction experiments were carried out at room temperature on a Siemens SMART 3-circle diÜractometer, equipped with a CCD area detector.Graphitemonochromated Mo-Ka radiation was used (k\0.71073 ”). A full hemisphere of the reciprocal space up to 2h\50° was scanned by x in 0.3° frames. The integrated intensities were corrected for absorption: in 1b by a semi-empirical method based on Laue equivalents with diÜerent w angles, in 2a by a numerical integration method, based on measured crystal shape and face-indexing.The structures were solved by direct methods and re–ned by full-matrix least-squares against F2 of Fig. 3 1H NMR spectra of cis- (top) and trans-1,2-bis(4-bromophenyl)- 1,2-diphenylethene (bottom) 1048 New J. Chem., 1998, Pages 1047»1049all positive data (all non-hydrogen atoms with anisotropic displacement parameters, H atoms in ìridingœ model), using SHELXTL software.12 Crystal data and experimental details are listed in Table 1, atomic coordinates are deposited at the Cambridge Crystallographic Data Centre (CCDC reference number 440/063).Synthesis of 2,3-bis(4-bromophenyl)-2-butenes Titanium trichloride (11.70 g, 76 mmol) and dry THF (150 ml) were transferred under a dry oxygen free nitrogen atmosphere into a two-necked round-bottomed —ask (250 ml) –tted with a condenser, dry nitrogen inlet and a magnetic stirrer.The dry oxygen-free nitrogen atmosphere was maintained through the experiment until destruction of excess reagents. After immersing the —ask into an ice bath, (1.43 g, 38 mmol) was LiAlH4 slowly added over a period of approximately 30 min while stirring rapidly. The mixture was re—uxed for 1 h. The resultant black slurry was allowed to cool to room temperature and 4-bromoacetophenone (7.55 g, 38 mmol) was added.After a further 20 h at re—ux, the mixture was cooled in an ice bath and dilute HCl (100 ml, 2 M) was added slowly to quench the reaction and destroy excess coupling reagents. The product was then extracted into chloroform (3]100 ml) ; solvent was evaporated from the combined extracts to give a yellow oil which was eluted through neutral alumina using hexane to give as a colourless oil a mixture of cis- and trans-2,3-bis(4- bromophenyl)-2-butene (3.35 g, 51 mol% yield w.r.t. 4-bromoacetophenone). Repeated recrystallisation from hexane gave the pure cis-2,3-bis(4-bromophenyl)-2-butene, mp 80.2»81.8 °C (lit.3 82 °C); found: C, 52.39 ; H, 3.59 ; requires C, C16H14Br2 52.49 ; H, 3.85%. 1H NMR (d, 400 MHz): 7.22 (pseudo CDCl3, d, 4, aromatic CH), 6.81 (pseudo d, 4, aromatic CH) and 2.12 (s, 6, 13C NMR (d, 100 MHz): 21.4 119.7 CH3). CDCl3 , (CH3), (CBr), 130.8 (aromatic CH), 130.9 (aromatic CH), 132.5 (alkylidene C) and 143.1 (aromatic C). FTIR (KBr disc, 2987, 2915, 1584, 1482, 1392, 1074, 1007, 827, 725, mmax/cm~1) : 559, 525, 476. The same procedure was applied to isolate the pure trans-2,3-bis(4-bromophenyl)-2-butene, mp 140.3»142.1 °C (lit.3 146 °C).The product was characterised by mass spec- Table 1 Crystal data and X-ray experiment parameters 1b 2a Formula C16H14Br2 C26H18Br2 M 366.1 490.2 Crystal size/mm 0.35]0.35]0.25 0.28]0.20]0.06 Colour colourless colourless Crystal system monoclinic monoclinic Space group P21/n P21/n a/” 12.226(1) 10.692(1) b/” 15.976(1) 9.191(1) c/” 15.792(1) 22.641(1) b/° 98.38(1) 102.47(1) U/”3 3051.6(4) 2172.4(3) Z 8 4 dcalcd/g cm~3 1.59 1.50 l(Mo-Ka)/cm~1 52.9 37.4 Data total 17 930 12 604 Data used (unique) 4184 3228 Rint before, after abs.corr. 0.134, 0.082 0.166, 0.090 Transmission max,min 0.291, 0.185 0.810, 0.418 Data observed, J[2r(I) 2682 2270 Variables re–ned 326 254 R1 (obs.data) 0.067 0.062 wR2 (obs. data) 0.100 0.092 Goodness of –t 1.14 1.15 *qmax,min/e ”~3 0.46, [0.45 0.40, [0.54 trometry; M]1 ion\365 determined by EI mass spectrometry: 1H NMR (d, 7.49 (pseudo d, 4, aromatic CH), CDCl3), 7.14 (pseudo d, 4, aromatic CH) and 1.84 (s, 6, 13C CH3) ; NMR (d, 100 MHz), 22.3 120.2 (CBr), 130.0 CDCl3 , (CH3), (aromatic CH), 131.4 (aromatic CH), 132.6 (alkylidene C) and 142.9 (aromatic C); FTIR (KBR disc, 2940, 2914, mmax/cm~1), 1583, 1486, 1443, 1391, 1069, 1005, 828, 784, 719, 608, 527.Synthesis of 1,2-bis(4-bromophenyl)-1,2-diphenylethenes Lithium aluminium hydride (1.45 g, 38 mmol) was slowly added to a slurry of titanium trichloride (11.80 g, 76 mmol) in dry THF (150 ml) under a dry nitrogen atmosphere at about 0 °C while stirring rapidly.The mixture was then re—uxed for 1 h. At room temperature, 4-bromobenzophenone (10.00 g, 38 mmol) was added and this mixture was re—uxed for 20 h. The reaction was quenched by adding dilute hydrochloric acid (2 M, 100 ml) into the mixture at room temperature while stirring. The product was extracted into chloroform (3]50 ml), washed with brine and dried over magnesium sulfate. The solvent was evaporated from the combined extracts and the residual yellow oil was reprecipitated into methanol.The precipitate was collected and dried to give a cis/trans mixture of 1,2-bis(4-bromophenyl)-1,2-diphenylethene (7.30 g, 78 mol yield w.r.t. 4-bromobenzophenone). Multiple recrystallisation from a mixture of ethanol and toluene (3 : 2, v/v) yielded the pure cis-1,2-bis(4-bromophenyl)-1,2- diphenylethene, mp 214.3»214.6 °C.The product was characterised by: 1H NMR see Fig. 3; 13C NMR (d, 100 CDCl3 , MHz) 120.8 (CBr), 126.8 (aromatic CH), 127.8 (aromatic CH), 131.1 (aromatic CH), 131.2 (aromatic CH), 132.9 (aromatic CH), 140.2 (alkylidene C), 142.3 (aromatic C) and 142.9 (aromatic C). The same procedure was applied to isolate the pure trans-1,2-bis(4-bromophenyl)-1,2-diphenylethene, mp 214.5»214.9 °C; M]1 ion\490 determined by EI mass spectrometry; found: C, 63.82 ; H, 3.49 ; requires C, C26H18Br2 63.70 ; H, 3.70%; molecular mass, 488 (2Br), 1H NMR see Fig. 3, 13C NMR (d, 100 MHz) 120.6 (CBr), 126.9 CDCl3 , (aromatic CH), 128.0 (aromatic CH), 130.9 (aromatic CH), 131.2 (aromatic CH), 132.8 (aromatic CH), 140.2 (alkylidene C), 142.3 (alkylidene C) and 142.8 (alkylidene C).References 1 T. Yamamoto, Y. Hayashi and A. Yamamoto, Bull. Chem. Soc. Jpn., 1978, 51, 2091. 2 N. Miyaura, T. Yanagi and A. Suzuki, Synth. Commun., 1981, 11, 513. 3 N. Inamoto, S. Masuda, Y. Nagai and O. Simamura, J. Am. Chem. Soc., 1963, 1433. 4 J. R. C. Light and H. H. Zeiss, J. Organometal. Chem., 1970, 21, 517. 5 J. E. McMurry, M. P. Fleming, K. L. Kess and L. R. Krepski, J. Org. Chem., 1978, 43, 3255. 6 W. H. Richardson, Synth. Commun., 1981, 11, 895. 7 J. Leimner and P. Weyerstahl, Chem. Ber., 1982, 115, 3697. 8 F. R. Fronczek, A. M. Swan, J. A. Corkern and R. D. Gandour, Acta Crystallogr. Sect. C, 1984, 40, 1875. 9 F. H. Allen, O. Kennard, D. G. Watson, L. Brammer, A. G. Orpen and R. Taylor, J. Chem. Soc., Perkin T rans. 2, 1987, suppl. 1. 10 C. Nather and H. Bock, Acta Crystallogr., Sect. C, 1997, 53, 231. 11 R.S. Rowland and R. Taylor, J. Phys. Chem., 1996, 100, 7384. 12 G. M. Sheldrick, SHELXTL, Version 5/VMS, 1995, Bruker axs, Analytical X-ray Systems, Madison, WI, USA. Received in Cambridge, UK, 6th July 1998; Paper 8/05208E New J. Chem., 1998, Pages 1047»1049 1049

ISSN:1144-0546    

DOI:10.1039/a805208e

出版商:RSC    

年代:1998

数据来源: RSC