摘要:

Chemical Society Reviews Vol 12 No 2 1983 Page Chemistry of Bis(dipheny1phosphino)methane By R. J. Puddephatt 99 4-Dialkylaminopyndines: Super Acylation and Alkylation Catalysts By Eric F. V. Scriven 129 Current Aspects of Unimolecular Reactions By K. A. Holbrook 163 Anionic Cydization of Phenols By William S. Murphy and Sompong Wattanasin 213 The Royal Society of ChemistryLondon Chemical Society Reviews EDITORIAL BOARD Professor K. W. Bagnall, B.Sc., Ph.D., D.Sc., C.Chem., F.R.S.C. Professor K. R. Jennings, M.A., D.Phi1, C.Chem., F.R.S.C. Professor G. W. Kirby, M.A., Ph.D., Sc.D.,*F.R.S.E., C.Chem., F.R.S.C. Professor G. Pattenden, Ph.D., C.Chem., F.R.S.C. Professor B. L. Shaw, B.Sc., Ph.D., F.R.S. Professor P.A. H. Wyatt, B.Sc., Ph.D., C.Chem., F.R.S.C. (Chairman) Editor: K. J. Wilkinson, B.Sc., M.Phi1. Chemical Society Reviews appears quarterly and comprises approximately 20 articles (ca. 500 pp) per annum. It is intended that each review article shall be of interest to chemists in general, and not merely to those with a specialist interest in the subject under review. The articles range over the whole of chemistry and its interfaces with other disciplines. Although the majority of articles are intended to be specially commissioned, the Society is always prepared to consider offers of articles for publication. In such cases a short synopsis, rather than the completed article, should be submit- ted to the Managing Editor, Books and Reviews Section, The Royal Society of Chemistry, Burlington House, Piccadilly, London, W1V OBN.Members of the Royal Society of Chemistry may subscribe to Chemical Society Reviews at f14.00 per annum; they should place their orders on the Annual Subscription renewal forms in the usual way. All other orders accompanied with payment should be sent directly to The Royal Society of Chemistry, The Distribution Centre, Blackhorse Road, Letch- worth, Herts. SG6 1HN England. 1983 annual subscription rate U.K. f39.50, Rest of World f42.00, U.S.A. $85.00. Air freight and mailing in the U.S.A. by Publications Expediting Inc., 200 Meacham Avenue, Elmont, New York 11003. U.S.A. Postmaster: Send address changes to Chemical Society Reviews, Publi- cations Expediting Inc., 200 Meacham Avenue, Elmont, New York 11003. Second class postage paid at Jamaica, New York 11431. All other despatches outside the U.K. by Bulk Airmail within Europe, Accelerated Surface Post outside Europe. @ Copyright reserved by The Royal Society of Chemistry 1983 ISSN 0306-001 2 Published by The Royal Society of Chemistry, Burlington House, London, W1V OBN Printed in England by Eyre & Spottiswoode Ltd, Thanet Press, Margate.

ISSN:0306-0012    

DOI:10.1039/CS98312FP003

出版商:RSC    

年代:1983

数据来源: RSC

ISSN:0306-0012    

DOI:10.1039/CS98312FX005

出版商:RSC    

年代:1983

数据来源: RSC

ISSN:0306-0012    

DOI:10.1039/CS98312BX007

出版商:RSC    

年代:1983

数据来源: RSC

摘要:

Chemistry of Bis(dipheny1phosphino)methane By R.J. Puddephatt DEPARTMENT OF CHEMISTRY, UNIVERSITY OF WESTERN ONTARIO, LONDON, CANADA N6A 5B7 1 Introduction Tertiary phosphine ligands have played a major role in modern co-ordination chemistry.' These ligands are easy to synthesize, and the electronic and steric properties of the ligands can be varied in a systematic way by varying the substituents on phosphorus' and by varying the backbone length (see below). The ligands bind strongly to many transition metals in low oxidation states, and are commonly used to stabilize organometallic and hydride derivatives of the elements, either in isolated compounds or as intermediates in homogeneous catalysis. In forming chelate complexes, the optimum ring size for a metal having natural bond angles at 90" to one another is five and it has long been known that bis(diphenylphosphino)ethane, dppe, is an excellent chelate ligand.Bis(dipheny1- phosphino)methane, dppm, can chelate but the four-membered ring so formed is strained, and the ligand has a greater tendency to act either as a monodentate ligand or as a bridging bidentate ligand. The chelating tendency also decreases as the chain length increases, so that for the ligands Ph2P(CH2),PPh2 the tendency to chelation is greatest for n = 2. A particularly good example is seen in the complexes [RhCl(CO)(Ph2P(CHz),PPhz}],which are dimers, (l),when n = 1, 3, or 4 but a monomer, (2), when n = 2.3 It is the ability ofdppm to form bridged binuclear complexes such as (1)* that has led to the recent interest in this and related ligands.Because metal-phosphorus bonds are often very strong, the bridging diphosphine ligand can lock together two metal atoms in close proximity and hence promote organometallic reactions involving two metal centres. Such reactions are often invoked in heterogeneous catalysis or in homogeneous catalysis using binuclear or cluster complex catalysts, and many useful models for these reactions can be developed using these locked-in binuclear phosphine-bridged complexes. The role of the phosphine is then to prevent dissociation of dimer to monomer, to promote bridging by other groups, and to promote binuclear reactions involving formation and cleavage of metal- metal bonds.This area of endeavour will form a major part of this review. There *In structural formulae, dppm and related ligands will generally be drawn without the phenyl substituents, e.g. p A p = dppm. ' C. A. McAuliffe and W. Levason, 'Phosphine, Arsine and Stibine Complexes of the Transition Elements', Elsevier, Amsterdam, 1979. C. A. Tolman, Chem. Rev., 1977, 77, 313. A. R. Sanger, J. Chem. SOC., Chem. Commun., 1975, 893. Chemistry of Bis(diphenyZphosphin0)methane have already been a number of excellent reviews including work on dppm complexes, but none cover the range of chemistry described here.4- * 2 Some Related Ligands For the sake of simplicity, this account will describe dppm complexes only, but there are many other ligands having similar properties and a brief account of a few of them is included here.One simple change is to replace phosphorus by other Group 5 elements, arsenic and antimony. The ligands Ph2AsCH2AsPh2 and Ph2SbCH2SbPh2 have been studied in some depth.9* lo For these ligands, the longer metal-ligand bonds lead to greater ring strain in chelate complexes, and there is an even stronger tendency for them to act as monodentate or bridging ligands. Thus, for example, dppm reacts with [Mo(CO),(diene)], diene = cyclo-octa- 1,5-diene or norbornadiene, to give monomeric (3) whereas bis(dipheny1stibino)methane gives dimeric (4). co rn Another simple modification involves changing the substituents R in A. L. Balch, in ‘Homogeneous Catalysis with Metal Phosphine Complexes’, ed.L. Pignolet, Plenum, New York, in press. M. P. Brown, J. R. Fisher, S. J. Franklin, R. J. Puddephatt, and M. A. Thomson, in ‘Catalytic Aspects of Metal Phosphine Complexes’, ed. E. C. Alyea and D. W. Meek, ACS Advances in Chemistry Series, 1982, 196, 231. A. L.Balch, in ref. 5, p. 243.’A. L. Balch, in ‘Reactivity of Metal-Metal Bonds’, ed. M. H. Chisholm, ACS Symposium Series, 1981, 155, 167. R.J. Puddephatt, in ref. 7, p. 187. R.Colton, Aust. J. Chem., 1976, 29, 1833. lo R. Okawara and Y.Matsumura, Adv. Organomet. Chem., 1976, 14, 187. ’’ K. K. Cheung, T. F. Lai, and K. S. Mok, J. Chem. SOC. (A), 1971, 1644. ’’ T. Fukumoto, Y.Matsumura, and R.Okawara, Inorg. Nucl. Chem. Lett., 1973,9, 711. Puddephatt R2PCH2PR2, This can easily be accomplished using the precursor C12PCH2PC12, and a number of derivatives (e.g.R = Me or Pr’) have already been prepared and studied.13*14As well as the ability to modify electronic properties of the ligands, it is likely that steric effects will have a major influence on the bridging us. chelating behaviour of these ligands. As a general rule, bulky substituents favour small-ring formation and it is therefore expected that more chelate complexes will be formed when R = But, and less when R = Me, compared to the case when R = Ph.” There is little evidence yet to substantiate this prediction, but it seems that for complexes [PtMe2(R2PCH2PR2)] the monomeric form (5) is the more stable for R = Ph and the dimeric form (6) for R = Me.’6 R2 Me Me ’‘PMe ‘P P’ ‘Me R2 R2V R2 Steric effects also influence reactivity to a major extent.For example, (6), R = Me, is reactive in oxidative addition with reagents like Me1 or I2 but, when R = Ph, Me1 fails to react and I2 cleaves a methylplatinum bond.16 Molecular models demonstrate that, in the dppm derivative, access of these reagents to the metal centre is blocked by the phenyl substituents in (6), R = Ph. Substitution at the methylene group appears to have less effect, and the ligands dppm, Ph2PCHMePPh2, and Ph2PCMe2PPh2 appear to behave in very similar ways, for example in gold complexes.’ Other ligands showing similar properties are those containing POP, PSP, or PN(R)P groupings. Of the POP ligands, (EtO),POP(OEt), ,18- 2o (CF3)2POP(CF3)2,21 -23 and (HO)zPOP(OH)224 are perhaps the most significant and they do not act as chelate ligands.This is apparently because the open POP angle of 120-150” does not allow the ligand conformation needed for chelation. When the substituents are small it is possible for four such l3 Z. S. Novikova, A. A. Prishcenko, and I. F. Lutsenko, J. Gen. Chem. USSR, 1977, 47, 707. l4 H. H. Karsch, Angew. Chem., Int. Ed. Engl., 1982, 21, 311. B. L. Shaw, J. Organomet. Chem., 1980, 200, 307. l6 R. J. Puddephatt, M. A. Thomson, Lj. Manojlovic-Muir, K. W. Muir, A. A. Frew, and M. P. Brown, J. Chem. SOC., Chem. Commun., 1981, 805; S. M. Ling and R. J. Puddephatt, Inorg. Chim. Acta Lett., in press. H. Schmfdbaur, Angew. Chem., ht.Ed. Engl., 1976, 15, 728. R. J. Haines, A. Pidcock, and M. Safari, J. Chem. SOC.,Dalton Trans., 1977, 830. l9 F. A. Cotton, R. J. Haines, B. E. Hanson, and J. C. Sekutowski, Inorg. Chem., 1978, 17, 2010. 2o A. L. du Preez, I. L. Marais, R. J. Haines, A. Pidcock, and M. Safari, J. Chem. SOC., Dalton Trans., 1981. 1918. 21 A. B. Burg and R. A. Sinclair, J. Am. Chem. SOC., 1966, 88, 5354. 22 R. A. Sinclair and A. B. Burg, Inorg. Chem., 1968, 7, 2160. 23 H. Einspahr and J. Donohue, Inorg. Chem., 1974, 13, 1839. 24 M. A. F. D. R. Pinto, P. J. Sadler, S. Neilde, M. R. Sanderson, A. Subbiah, and R. Kuroda, J. Chem. SOC.,Chem. Commun., 1980, 13. Chemistry of Bis( dip heny lphosp hino)methane ligands to bridge between two metal atoms, as in the fluorescent platinum(I1) derivative [Pt,(p-( HO)(0)POP(O)(OH)}4]4 -of structure (7).The ligands F,PN(R)PF, ,R = Me or Ph, have been studied extensively,2s and can act as chelate, monodentate, or bridging ligands. Again the small substituents allow three or four bridging ligands as illustrated by the derivative of the unknown [Moz(CO) [Mo, (CO),(p-CO)(p-F, PNPhPF,)J, and the mu1 ti ply bonded molybdenum derivative [Mo,C~,(~-F,PNM~PF,),].~~ ligandThe (PhO),PNEtP(OPh), is unusual in forming a bridging complex with trans stereochemistry at one metal but cis stereochemistry at the other, in [Rh,Cl,(CO~p-(PhO),PNEtP(OPh)~},]of structure (8),'* as well as many other bridged complexes.2 The above discussion illustrates how the ligands X2PYPX, can be modified considerably in both electronic and steric properties by systematically changing the substituents X and Y.The following discussion of dppm complexes, although substantial in itself, forms only a small part of this general field of phosphine bridged binuclear complexes.Finally, we note that an even greater number of unsymmetrical ligands can be envisaged, though this field is in its infancy. One good example is 2-diphenyl- 25 R. B. King, Acc. Chem. Res., 1980, 13, 243. 26 M. G. Newton, R. B. King, T. W. Lee, L. Norskov-Lauritzen, and V. Kumar, J. Chem. SOC., Chem. Commun., 1982, 201. ''F. A. Cotton, W. H. Ilsley, and W. Kaim, J. Am. Chem. Soc., 1980, 102, 1918.*' R. J. Haines, E. Meintjies, and M. Laing, Inorg. Chim. Acta, 1979, 36, L.403.29 G. de Leeuw, J. S. Field, R. J. Haines, B. McCulloch, E. Meintjes, C. Monberg, K. G. Moodley, G. M. Oliver, C. N. Sampson, and N. D. Steen, J. Organomet. Chem., 1982, 228, C66. Puddephatt phosphinopyridine, which is particularly useful for bridging between unlike metal atoms, as in complex P -,I/" I C1-Rh -Pd -C1 N vp 3 Complexes with Monodentate dppm Complexes with monodentate dppm are formed with metal halides, metal carbonyls, and several organometallic derivatives. Metal halide or pseudohalide derivatives include tran~-[Ni(NCS),(q'-dppm)~]and the less stable [NiCl,(q'-dppm),], and the cations [MOCI2(q1-dppm),]+, M = Mo or W.31-33 Metal carbonyl derivatives include [Cr(CO)5(ql-d~pm)]~~ [Fe( CO),( ql-and d~pm)],~~as well as more complex examples containing both monodentate and bidentate dppm such as mer andfuc isomers of [M~(CO)~(q~-dpprn)(q'-dppm)],~~ (10)and (ll),identified by the characteristic 31Pn.m.r.spectra. The complexes with monodentate d~pm,~ [Ru(TPP)(q'-dPPm),I7*38 (TPP= [tetraphenylporphyrin]'-) and truns-[Pd(B~'NC),(q'-dpprn)~]~+,(12), and with chelating and monodentate d~pm,~' [MoC1,(CO),(q2-dppm) (q'-dppm)], or bridging and monodentate d~pm,~' [Pt,H(p-dppm),(q'-dppm)]+, 30 J. P. Farr, M. M. Olmstead, and A. L. Balch, J. Am. Chem. SOC., 1980, 102, 6654. 31 C. Ercolani, J. V. Quagliano, and L. M. Vallarino, Inorg. Chim. Acta, 1973, 7, 413. 32 K. K. Chow and C. A. McAuliffe, Inorg. Chim. Acta, 1974, 10, 195. 33 W.Levason, C. A, McAuliffe, and F. P. McCullough, Inorg. Chem., 1977, 16, 2911. 34 J. A. Connor, J. P. Day, E. M. Jones, and G. K. McEwen, J. Chem. SOC.,Dalton Trans., 1973, 347. 35 P. A. Wegner, L. E. Evans, and J. Haddock, Inorg. Chem., 1975, 14, 192. 36 E. E. Isaacs and W. A. G. Graham, Inorg. Chem., 1975, 14, 2560. 37 R. G. Ball, G. Domazetis, D. Dolphin, B. R. James, and J. Trotter, Inorg. Chem., 1981, 20, 1556. 38 M. M. Olmstead, C. L. Lee, and A. L. Balch, Inorg. Chem., 1982, 21, 2712. 39 M. G. B. Drew, A. P. Wolters, and I. B.Thomkins, J. Chem. SOC., Dalton Trans., 1977, 974. 40 Lj. Manojlovic-Muir and K. W. Muir, J. Organomet. Chem., 1981, 219, 129. 103 Chemistry of Bis(diphenylphosphino)methane (13), have been characterized crystallographically and two structures are shown in Figure 1.P C +Figure 1 Structures of trans-[Pd(Bu'NC), (q1-dppm),12 and [Pt, H(p-dppm), (q'-dppm)] + Complexes (12), (13), and tran~-[Pt(C=CPh)~(q~-dppm)~],(14), are fluxional on the n.m.r. time scale at ambient temperat~re,4~*~~ probably as the result of an intramolecular displacement of co-ordinated phosphorus by free phosphorus in the ql-dppm ligands. In complex (13) the p-dppm ligands are not involved in the fluxional process, and the exchange rate is, as expected, greater for the palladium(1r) derivative, (12), than for the platinum(u) derivatives (13) and (14).38941742The structures (12) and (13) (Figure 1) show how the free phosphorus atom of the ql-dppm ligands may be expected to displace the co-ordinated phosphorus by intramolecular co-ordination.The free phosphorus atom in q'-dppm complexes may be oxidized to the phosphine oxide,42 quaternized by reaction with Me1 or Me30+, or may be used as a ligand to the same or a different metal atom. Two examples to illustrate the latter reactions are given in equations 1 and 2.34,43 P"'\i>'A Cr I 'co TO 'I'POC'co hcat OC\pP> OC co Ph " M. P. Brown, J. R. Fisher, R. H. Hill, R. J. Puddephatt, and K. R. Seddon, Inorg. Chem., 1981, 20,3516. "P. G. Pringle and B. L. Shaw, J. Chem Soc., Chem. Commun., 1982, 581. A3 F. Sato, T. Uemura, and M. Sato, J. Organomet. Chem., 1973,56, C27. Puddephatt The interconversion between monodentate and chelating dppm is also'.32illustrated by the rea~tion.~ [NiC12(q2-dppm)]+ dppm t[NiC12(q '-dppm),] There is some evidence that steric effects may limit the ability of dppm to bridge in some cases.Thus, [Fe(CO)2(q'-Ph2P(CH2)nPPh2)(q-C,H,)1'is a good ligand for cobalt(i1) through the free phosphine when n = 2 or 3, but not for n = 4 Complexes with Chelate dppm When dppm acts as a chelate ligand the MP2C skeleton is essentially planar, as seen for example in trans-[RhHCl(dpprn),]+, (19, whose structure is given in Figure 2.45The strain in these chelate complexes is seen by the low PMP angles HlCl Figure 2 The structure of trans-[RhHCl(dppm),] + of 67-74" and the PCP angles of about 95", compared to normal bond angles of 90"and 109" respectively.Even greater strain would be expected for metal ions preferring tetrahedral or trigonal co-ordination geometries, and no such complexes have been characterized unambiguously, although there is some evidence for monomeric pseudotetrahedral [CoCl,(dppm)] in solution.46 In [Fe(CO),(dppm)] the chelating dppm spans an axial and an equatorial site, with a natural bond angle of 90",rather than two equatorial sites.47 There are a great number of complexes with chelating dppm, some of which have already been mentioned, and no attempt at comprehensive coverage will be 44 M. L. Brown, J. L. Cramer, J. A. Ferguson, T. J. Meyer, and N. Winterton, J. Am. Chem. Soc., 1972,!M, 8707. 45 M. Cowie and S. K. Dwight, Inorg. Chem., 1979, 18, 1209. 46 K. K.Chow and C. A. McAuliffe, Inorg. Chim. Acta, 1975,14, 121. 47 F. A. Cotton, K. I. Hardcastle, and G. A. Rusholme, J. Coord. Chem., 1973, 2, 217. Chemistry of Bis(dipheny1phosphino)methune made. The series of Group 6 carbonyl derivatives is noteworthy in its completeness; for example, ~is-[Mo(CO)~(dppm)], ci~-[Mo(CO),(dppm)~] and [Mo(dppm),] are all kn~wn."~~~-~~ Several platinum complexes [PtX,(dppm)]; X = Cl, Br, I, Me, Ph, 4-tolyl, as well as [Pt(dppm)2]2+, and the unsymmetrical complexes [PtClR(dppm)], R = Me, 4-tolyl, are also known.5 -56 The latter complexes isomerize in solution to the ionic binuclear derivatives [Pt2R2(p-C1) (p-dppm)2]C1.53956It is not clear what factors influence the relative stabilities of the monomeric and dimeric isomers, but it may be relevant that complexes [PtClRL,] have a strong preference for the trans stereochemistry, which is possible for the dimer but not for monomeric [PtClR(dppm)].Steric effects also are likely to promote chelation. For example with dppm: Rh ratio of 2:l in (15) and in [Rh(dppm),]+, dppm acts as a chelate whereas with dppm: Rh ratio of 1:l bridging dppm is observed (see later).57 It is likely that the Rh,(p-dppm), skeleton would be very congested and hence is not formed, whereas the Rh,(p-dppm), skeleton is not too crowded. 5 Complexes with Bridging dppm A. Compounds without Metal-Metal Bonds or other Bridging Groups.-In this class there are compounds with one, two, or three p-dppm ligands. A good example of a complex with one p-dppm ligand is [(CO)4FePPh2CH2PPh2Fe(C0)4],(16). On photolysis a carbonyl group is lost and a metal-metal bonded complex [Fe,(C0)6(p-CO)(p-dppm)], (17), is formed.It is likely that this reaction is facilitated by the proximity of the two metal centres in (16).35*58Another interest- ing case is the ruthenium(I1) derivative [(R~Cl(bipy),>,(p-dpprn)]~+,which can be oxidized to the RU'~,RU'" and then the RUIII,RU~'~ analogue without disruption of the p-dppm ligand. Complexes with two trans p-dppm ligands are most common in rhodium(1) chemistry and have been thoroughly investigated.60- " A typical structure of 48 A. M. Bond, R. Colton, and J. J. Jackowski, Inorg. Chem., 1975, 14, 274. 49 M. Hidai, K. Tominari, and Y. Uchida, J.Am. Chem. SOC., 1972, 94, 110. M. W. Anker, J. Chatt, G. J. Leigh, and A. G. Wedd, J. Chem. SOC.,Dalton Trans., 1975, 2639. 51 M. P. Brown, R. J. Puddephatt, M. Rashidi, and K. R. Seddon, J. Chem. SOC., Dalton Trans., 1977, 951. "T. G. Appleton, M. A. Bennett, and I. B. Tomkins, J. Chem. SOC., Dalton Trans., 1976, 439. "S. J. Cooper, M. P. Brown, and R. J. Puddephatt, Inorg. Chem., 1981, 20, 1374. s4 P. S. Braterman, R. J. Cross, Lj. Manojlovic-Muir, K. W. Muir, and G. B. Young, J. Organomet. Chem., 1975, 84, C40. "R. J. Puddephatt and M. A. Thomson, J. Organomet. Chem., 1982,238, 231. 56 G. K. Anderson, H. C. Clark, and J. A. Davies, J. Organomet. Chem., 1981, 210, 135. 5' B. R. James and D. Mahajan, Can. J. Chem., 1979, 57, 180. F.A. Cotton and J. M. Troup, J. Am. Chem. SOC., 1974, %, 4422. 59 B. P. Sullivan and T. J. Meyer, Inorg. Chem., 1980, 19, 752. 6o J. T. Mague, Inorg. Chem., 1969, 8, 1975. 61 M. Cowie and S. K. Dwight, Inorg. Chem., 1980, 19, 2500. 62 J. T. Mague and J. P. Mitchener, Inorg. Chem., 1969,8, 119. 63 A. L. Balch, J. Am. Chem. SOC., 1976,98, 8049. 64 A. L. Balch and B. Tulyathan, Inorg. Chem., 1977, 16, 2840. A. L. Balch, J. W. Labadie, and G. Delker, Inorg. Chem., 1979, 18, 1224. 66 J. T. Mague and S. H. DeVries, Znorg. Chem., 1980, 19, 3743. 67 A. R. Sanger, J. Chem. SOC.,Dalton Trans., 1981, 228, and references therein. Puddephatt C(13) Figure 3 The structure of the face-to-face dimer [Rh,C1,(CO)z(p-dppm),] [Rh2Cl,(CO)2(p-dppm),], (18), is shown in Figure 3.Similar complexes with chloro and carbonyl ligands replaced by ligands such as isocyanides, giving for example [Rh2(MeNC),(p-dppm),I2 +, are readily ~repared.~~~~~ These compounds are usually described as face-to-face complexes for obvious reasons (Figure 3), and there is evidence of weak metal-metal interactions, although no formal metal-metal bond is present.63 -65 The complexes will undergo binuclear oxidative addition reactions and will add the unsaturated molecules CO and SO2between the metal centres, in each case producing rhodium-rhodium bonded complexes (Scheme l).63-66 A further development is the synthesis of face-to-face complexes with square pyramidal rhodium(1) centres as in complex ( 19).67 P -P I I/co /coNC-Rh Rh-CN P \/p 107 Chemistry of Bis(dipheny1phosphino)methane 2+ P-P I LLI., /l A I-Rh-Rh-I L = McNC Lq 1’1 P I 2+ P -P Scheme 1 The face-to-face dimers are much more elusive with other metals, and the first platinum(I1) complexes of formula [Ptz(C=CR)4(p-dppm)2], R = Me, CF3, Ph, or 4-toly1, (20) were prepared only recently.42i ” P -P Complexes [M,Cl,(p-dppr~)~], M = Pd or Pt, may be generated but they rapidly rearrange to the monomeric forms [MC12(dppm)].68* The complex 69 [Pt2C1,Mez(p-dppm)Z]s3 rearranges rapidly to the ionic form [Pt,Me,(p-Cl) (p-dp~m)~]Cl.Other face-to-face compounds have been proposed as reaction C. T.Hunt and A. L. Balch, Inorg. Chem., 1981, 20, 2267. 69 R. J.Puddephatt, unpublished work. Puddephatt intermediates but have not been directly ~haracterized.~' -72 The factors affecting the relative stabilities of these isomers are not understood. Complexes with two cis bridging dppm ligands are much less common, and are limited to the dimethylplatinum(n) derivative (6)' and the mallylrhodium(1) complex [Rh2(q3-CH2CMeCH2)2(p-dppm)2].73The former complex has a twist saddle conformation (Figure 4), and undergoes two fluxional processes involving a twisting motion and a saddle inversion process.16 n Figure 4 The structure of[Pt,Me,(p-dppm),], showing a view along the Pt-Pt axis The only compounds with three p-dppm ligands are [M2(p-dppm),], M = Pd or Pt.747 75 There is evidence from electronic spectroscopy of weak metal- metal interaction and the platinum derivative has been shown to have the 'manxane' structure (21), each platinum having trigonal planar stereochemistry. PA P PP \/ \P Pt Pt These compounds have a very rich chemistry as shown by the reactions of Scheme 2.4,68,7 1,75,76 It can be seen that oxidative addition can give either oxidation of each metal 'O C.L. Lee, C. T. Hunt, and A. L. Balch, Organometallics, 1982, 1, 824.'' A. L. Balch, C. T. Hunt, C. L. Lee, M. M. Olmstead, and J. P. Farr, J. Am. Chem. Soc., 1981, 103, 3764. 'Iz R. H. Hill and R. J. Puddephatt, Inorg. Chim. Acta, 1981, 54, L277. l3 M. D. Fryzuk,Inorg. Chim. Acta, 1981, 54, L265. 74 E. W. Stern and P. K. Maples, J. Catal., 1972, 27, 120.''M.C. Grossel, M. P. Brown, C. D. Nelson, A. Yavari, E. Kallas, R. P. Moulding, and K. R. Seddon, J. Organomet. Chem., 1982,232, C13; Lj. ManojloviOMuir, personal communication. 76 K. A. Azam, R. J. Puddephatt, M. P. Brown, and A. Yavari, J. Organomet. Chem., 1982, 234, C31. Chemistry of Bis(dipheny1phosphino)methane P -PII jt H-Pt -A-PI I)P Pvp VP P -P P-P Pd PdI A1 II Me-Pt -Pt 7 P P -P I-i1 -Pd -Pd -I PII -Scheme 2 by one unit with formation of a metal-metal bond or oxidation of each metal by two units, with partial (M = Pt) or complete (M = Pd) dissociation of one dppm ligand. B. Compounds with Metal-Metal Bonds but no other Bridging Groups.-There are a number of compounds with multiple metal-metal bonds in this class, derived from the quadruply bonded [Mo,Clsl4- and [Re,Cl,]’-.The structures of the quadruply bonded [Mo,Cl,(p-dpprn),], (22), and the paramagnetic P-P Mo Mo Puddephatt P-P 17' I /c'C1-Re -Re (23) [Re,Cl,(p-dppm),], (23), have been determined crystallographically, and the complexes fRe,C16(p-dppm),] and [Re2C14(p-dppm)2] are also known,77- 8o but few reactions of the products have been studied. The best studied compounds of this kind are the singly metal-metal bonded derivatives [MM1X2(p-dppm)2], M,M1 = Pd,Pt, of which the mixed Pd-Pt derivative was prepared very recently.81-86 A typical structure of [Pt,Cl,(p-dppm),] is given in Figure 5.84 These complexes display high Figure 5 The structure of [Pt,Cl,(p-dppm)2] 77 S.A. Best, T. J. Smith, and R. A. Walton, Inorg. Chem., 1978, 17,99. 78 E. H. Abbott, K. S. Bose, F. A. Cotton, W. T. Hall, and J. C. Sekutowski, Inorg. Chem., 1978, 17, 3240. 79 J. R. Ebner, D. R. Tyler, and R. A. Walton, Inorg. Chem., 1976, 15. 833. F. A. Cotton, L. W. Shive, and B. R. Stults, Inorg. Chem., 1976, 15, 2239. R. G. Holloway, B. R. Penfold, R. Colton, and M. J. McCormick, J. Chem. Soc., Chem. Commun., 1976, 485. M. P. Brown, R. J. Puddephatt, and M. Rashidi, Inorg. Chim. Acta, 1976, 19, L33. M. P. Brown, R. J. Puddephatt, M. Rashidi, and K. R. Seddon, J. Chem. SOC., Dalton Trans., 1977,951. e4 Lj. Manojlovic-Muir, K. W.Muir, and T. Solomun, Actacrystallogr., 1979, 835, 1237. R. Colton, M. J. McCormick, and C.D. Pannan, Aust. J. Chem., 1978, 31, 1425. 86 P. G. Pringle and B. L. Shaw. J. Chem. SOC., Chem. Commun., 1982, 81. Chemistry of Bis(dipheny1phosphino)methane reactivity either by displacement of terminal chloride ligands by anionic or neutral ligands or insertion of SnCl, into the metal-chloride bonds (Scheme 2+ P-P IIPh3P -M -M ’-PPh3 II P vp P-P M = M’=Pt 2 SnC12I P-P II C13Sn -M -M -SnC13 II pvp Scheme 3 However, the addition of small molecules to the metal-metal bond is of greatest interest. Some examples of these reactions are shown in Scheme 4.85*86*90-100In most of these reactions the metal-metal bond is acting as a ”M.M. Olmstead, L. S. Benner, H. Hope,and A. L. Balch, Inorg. Chim. Acta, 1979, 32, 193.M. P. Brown, S. J. Franklin, R. J. Puddephatt, M. A. Thomson, and K. R. Seddon, J. Organomet. Chem., 1979, 178, 281. 89 M.C. Grossel, R. P. Moulding, and K. R. Seddon, Inorg. Chim. Acta Lett., 1982, 64, L275. 90 M. P. Brown, R. J. Puddephatt, M.Rashidi, and K. R. Seddon, J. Chem. SOC., Dalton Trans., 1978, 1540. 91 M. P. Brown, A. N. Keith, Lj. Manojlovit-Muir, K. W. Muir, R. J. Puddephatt, and K. R. Seddon, Inorg. Chim. Acta, 1979,34, L223. 92 M. P. Brown, J. R. Fisher, R. J. Puddephatt, and K. R. Seddon, Inorg. Chem., 1979,18, 2808. Puddephatt P -P P -P MI AM'I x'( (\h VPP #p P P -P P-PI rI AM I IM I t1+M-M'-x 'x PI I/PVP J S M ,C--S\ MI I !Scheme 4 nucleophile, as is most easily seen in the reactions with cationic reagent^^^.^^ H+ and PhN2+.It can be seen that single atom bridges (e.g. CH2, CO, SO,, or S)85*86*90-95or double atom bridges (CS2 or RC=CR)86.96-98.'00are 93 M. M. Olmstead, H. Hope, L. S. Benner, and A. L. Balch, J. Am. Chem. SOC., 1977, 99, 5502. 94 L. S. Benner, M. M. Olmstead, H. Hope, and A. L. Balch, J. Organomet. Chem., 1978, 153, C31. 95 A. D. Rattray and D. Sutton, Inorg. Chim.Acta, 1978, 27, L85. 96 A. L. Balch, C. L. Lee, C. H. Lindsay, and M. M. Olmstead, J. Organomet. Chem., 1979, 177, C22. 97 C. L. Lee, C. T. Hunt, and A. L. Balch, Inorg. Chem., 1981, 20, 2498. 98 R. J. Puddephatt and M. A. Thomson, Inorg. Chem., 1982, 21, 725. 99 M. P. Brown, R. J. Puddephatt, M. Rashidi, and K. R. Seddon, J.Chem. SOC.,Dalton Trans.. 1978, 516. loo T. S. Cameron, P. A. Gardner, and K. R. Grundy, J. Organomet. Chem., 1981, 212, C19. Chemistry of Bis(dipheny1phosphino)methane possible. The alkynes must have electronegative substituents (R = CF, or C0,Me) in order to undergo reaction. Several of the reactions are reversible (CO, SO,, and H+ additi~n),*~*~~*~~and when M = Pd, M' = Pt, the reversible addition of CO is catalysed by Figure 6 The structure of [Pt2C12(p-CHz)(p-dppm),] Of particular interest in these reactions is the formation of unique bonding types. For example, these constitute almost unique examples of CO and CH, bridging between two metal atoms in which there is no metal-metal bond. The absence of such metal-metal bonding has been established in several cases by X-ray structure determination^,^^^^'^^^^^^*^^' and an example is shown in Figure 6, [Pt,C1,(p-CH,)(p-dppm),].'olThe Pt-Pt distance of 3.16 A and the L (PtCPt) of 102"provide strong evidence that there is no Pt-Pt bond.In some cases, insertion may occur into both metal-metal and metal-ligand bonds, and an example is given in equation 3.'" P-P P-P PtI /"\ Pt1 'm ,PPh,II (3) P VP Lj. Manojlovic-Muir, personal communication. lo* K. A. Azam, A. A. Frew, B. R. Lloyd, Lj. Manojlovic-Muir, K. W. Muir, and R. J. Puddephatt, J. Chem. SOC.,Chem. Commun., 1982, 614. Puddephatt In other cases a ligand may add to a metal-metal bond or displace a terminal ligand (equation 4).90*919 lo3 -P P P -P I-i co C1-Pt I-i-Pt-co c1-co Cl-Pt -Pt-c1 p PtIIAPtI (4) ‘Cl P PIvp II -I PVP a’ \Ip-1 Finally, further insertion reactions can lead to catalytic trimerization of alkynes using [Pd2C12(p-dppm)2].97 The bridging dppm ligands are able to stabilize a donor-acceptor metal-metal bond in one case.The complex cation [Pt2Me3(p-dppm),]+ has structure (24), + I’ -P Me I \/Me -Pt uPt -PI /\Mevp (24) with the electron-rich PtMe, centre donating electrons from the filled 5d22 orbital to the monomethylplatinum centre.lo4 This complex is also unusual in having cis stereochemistry at one metal centre but trans stereochemistry at the other. From the above account it should be apparent that the p-dppm ligands can promote new chemistry by holding two metal atoms in close proximity so that they can be bridged by other more weakly bonded groups even when there is no (or only very weak) metal-metal bonding.C. Compounds with other Bridging Groups.-There is no clearcut distinction between this section and the last, which contained many examples of compounds with bridging groups (e.g. Scheme 4), but the emphasis here is different. Many of these complexes have been termed ‘A-frame’ complexes, and examples have been prepared according to equation 5.”’-lo* lo’ Lj. Manojlovic-Muir, K. W. Muir, and T. Solomun, J. Organornet. Chem., 1979, 179, 479. lo’ M. P. Brown, S. J. Cooper, A. A. Frew, Lj. Manojlovic-Muir, K. W. Kuir, R. J. Puddephatt, K. R. Seddon, and M.A. Thomson, Inorg. Chem., 1981, 20, 1500. C. P. Kubiak and R. Eisenberg, J. Am. Chem. Soc., 1977, 99, 6129. lo6 M. Cowie and S. K. Dwight, Inorg. Chem., 1979, 18, 2700. lo’ C. P. Kubiak and R. Eisenberg, Inorg. Chem., 1980, 19, 2726. lo8 M. Cowie and S. K. Dwight, Inorg. Chem., 1980, 19, 2508. Chemistry of Bis(dipheny1phosphino)methane P-P P-P 17 I S*-Rh Rh ___, OC’I c1’I P P P vp v1”-\ PA + O P P-P I ARhIRh -Rh Rh As can be seen from equation 5 and Scheme 4, the A-frame complexes may or may not have a metal-metal bond as well as a bridging ligand. The presence or absence of a metal-metal bond is predictable in terms of the normal electron- counting procedures, and a detailed molecular orbital treatment has been published which describes the bonding in a particularly lucid manner.log Of especial interest is the reversible addition of small molecules to the vacant site between the two metal centres, as illustrated in Schemes 5-7.66p679105-11* As can be seen, the addition or loss of these small molecules occurs with making or breaking a metal-metal bond, in a similar way as shown in Scheme 4, and many further examples have been studied. In several cases, however, the incoming ligand does not attack the ‘pocket’ site between the metal atoms directly. Instead, attack occurs at a terminal site and a terminal carbonyl group moves to the bridging position, as illustrated in equation 6.lI5 I.1+r rP-P P-Pb\I P-P IARhI*CO Rh R NCRh-Rh c--+ oc’ I \( I ‘*co oc’ I I ‘cOP P --\Ip.PVP Io9 D. M. Hoffman and R. Hoffmann. Inora. Chem.. 1981. 20. 543. ‘lo C. P. Kubiak and R. Eisenberg, J: Am. ?hem. Soc., 1980, ’102, 3637. M. Cowie and S. K. Dwight, Inorg. Chem., 1980, 19, 209. M. Cowie, Inorg. Chem., 1979, 18, 286. M. M. Olmstead, C. H. Lindsay, L. S. Benner, and A. L. Balch, J. Organomet. Chem., 1979,179,289. M. Cowie, J. T. Mague, and A. R. Sanger, J. Am. Chem. SOC., 1978, 100, 3628. J. T. Mague and A. R. Sanger, Inorg. Chem., 1979, 18, 2060. Puddephatt PH’I Pvp P -P PAP c1 \ P P-P vp Scheme 5 + P-P P -P I A1I I/c1\ I RhRh oc -Rli--Rh-Co OC’I ‘co P -VP -co,c1-I P-P P -Pr + I A I oc -Rh -Rh-co I ‘i’ I POP P -v wp Scheme 6 Chemistry of Bis(dipheny1phosphino)methane P-P r P-P IOC-Rh-Rh-CO I I IPVP ic0 P -P r P-P !Scheme 7 This topic has been illustrated with examples from rhodium chemistry, but an extensive, related chemistry of iridium is also l15* l1 In this case 149 hydrogen reacts with [Ir2(p-S)(C0)2(p-dppm)2]or with [I~-~(p-cl)(cO)~ (p-dppm),]+ to give hydridoiridium(r1) complexes.’ 9*120 This binuclear activation of small molecules has potential application in catalysis.’ The A-frame complexes of palladium and platinum do not add small molecules in the ‘pocket’ position, but insertion reactions can occur in some cases (equation 7).70 ‘16 M.Cowie and T. G. Southern, Inorg. Chem., 1982, 21, 246. 11’ M. Cowie and R.S. Dickson, Inorg. Chem., 1981, u),2682. ll8 M. Cowie and S. K. Dwight, J. Organornet. Chem., 1981, 214, 233. ‘19 C. P. Kubiak, C. Woodcock, and R. Eisenberg, Inorg. Chern., 1980, 19, 2733. lZo M. Cowie, personal communication. Puddephatt The complex [PtzHz(p-H)(p-dppm)z]+ undergoes reductive elimination of hydrogen on treatment with ligands such as phosphines or carbon monoxide by an associative mechanism, but by a dissociative mechanism on photolysis (equation 8). The photolysis occurs with high quantum yield; hydrogen loss + t P-P P-P L -H -Pt-II Pt -LIt P VP P VP is intramolecular, and reaction occurs from a singlet excited state. The reaction is noteworthy because almost all other binuclear and polynuclear hydrides are inert to 122 This type of binuclear reductive elimination can be reversed in some cases,91~1z1*123and a similar reductive elimination of Hz from [Pt2MeH(p-H) (p-dppm)2]+ or of CH4 from [Pt2Mez(p-H)(p-dppm)z]+ can also be achieved.76* The reductive elimination of hydrogen is thought to be involved as an initial step in reactions of [PtZH2(p-H)(p-dppm),]+ with alkynes and with methanethiol and diphenylphosphine (Scheme 8).92998 The A-frame complexes are fluxional in some cases, with apparent rapid inversion of the A-frame structure.This reaction may occur by direct inversion"' as in [Pt2H2(p-H)(p-dppm)z]+ or it may occur by more complex mechanisms as shown in equations 9 and 10. In the former case, the reversible formation of the face-to-face complex t P -P P -PI ll\ c1-.I Pd Pd 7 C1' h P P v p \/p H.C. Foley, R. H. Morris, T. S. Targos, and G. L. Geoffroy, J. Am. Chem. SOC.,1981, 103, 7337. R. H. Hill, P. de Mayo, and R. J. Puddephatt, Znorg. Chem., 1982,21,3642. 123 J. R. Fisher, A. J. Mills, S. Sumner, M. P. Brown, M. A. Thomson, R. J. Puddephatt, A. A. Frew, Lj. ManojloviGMuir, and K. W. Muir, Organometallics, 1982, 1, 1421. 124 M. P. Brown, S. J. Cooper, A. A. Frew, Lj. Manojlovic-Muir, K. W. Muir, R. J. Puddephatt, and M. A. Thomson, J. Chern. SOC.,Dalton Trans., 1982, 299. R. J. Puddephatt, K. A. Azam, R. H. Hill, M. P. Brown, C. D. Nelson, R. P. Moulding, K. R. Seddon, and M. C. Grossel, J. Am. Chem. SOC., in press. Chemistry of Bis (dip hen ylp hosp hino)methane r p-P + C1-Pt P -P I I I I -Pt-C1 vpP I1 c -.I c1-Pt PI-i PI IVP -Pt -c1 + c1’ - -PP PtI /“\I PtI I ‘C! Pvpd leads to apparent inversion70* 71 whereas, in the latter case, intermolecular transfer of H+ to a catalytic amount of [Pt,Cl,(p-dppm),] leads to the same result.125 There is an interesting intramolecular binuclear #?-elimination reaction leading to an A-frame ethyl(p-hydrido)platinum complex, but the reverse insertion of alkenes into the Pt2(p-H) linkage is not observed (equation 11).’26 ’+ P-P In most A-frame complexes the methylene groups of the dppm ligands are displaced from the M2P, plane towards the apex bridging group (Figure 7).This conformation allows the phenyl substituents of dppm to bend away from the bridging group and so minimizes steric effects.The overall conformation of the 8-membered ring is then an extended boat. However, in [Pt,Me,(p-H) (p-dp~rn)~]+ , the conformation is an extended chair (Figure 7) perhaps because steric effects are less with the small p-hydrido ligand. 126 K. A. Azam, M.P.Brown, S. J. Cooper, and R.J. Puddephatt, Organometallies, 1982, 1, 1183. Puddephatt t P -P Pt’lHI\Pt H’l I‘H P VP t + P -P P A P L = McSH - H-Pt I-Pt I-L L=RC=C‘K H’I . I\H vp.P I I P VP CH3CN L = Ph2PIl.1 + P -P H’l I‘ P VP Scheme 8 The reactions of dppm with metal carbonyls have been particularly significant for the manganese and iron group elements. Thus [Fe2(CO)9], with three p-CO groups, reacts with dppm to give, amongst other products, [Fez (CO),(p-CO)( p-dppm)] and then [Fe, (CO),(p-CO)( p-dppm),], each having only one p-CO gro~p.~~~~~,~~ Interesting derivatives of [Fe,(CO),(Cp),] are also known, and the reaction of equation 12 giving a complex with an unsymmetrical bridging methyl group is of particular significance. 27 + P -P Fe Fe Fe -Fe1 /i\/ \ CII-,1’/\/\cp cp II ---c CP G.M. Dawkins, M. Green, A. G. Orpen, and F. G. A. Stone, J. Chem. Soc., Chem. Commun., 1982, 41. Chemistry of Bis(d iphen ylp hosphino)methane c A C Figure 7 The conformations of dppm ligands in [Rh2(p-H)(C0),(p-CO)(p-dppm),]+ and in [PtzMez(~--H)(~-dPPm)*I+ Another very unusual bridging group is found in the derivative [Mn,(CO),(p-CO)(p-dppm),], (25).In this complex the p-CO group acts as a P-P P-P II (CO), Mn -Mn(CO), (CO)*Mn Mn(CO)2I P RP v 4-electron ligand, the unusual feature being the donation of two electrons from a CEO z-bond.'28-'30 A similar complex, (26), R = 4-tolyl, with a 4-electron bridging isocyanide ligand is also known.I3 ' lZ8C. J. Commons and B. F. Hoskins, Aust. J. Chem., 1975, 28, 1663. lZ9R. Colton and C. J. Commons, Aust. J. Chem., 1975, 28, 1673. 130 K. G. Caulton and P. Adair, J. Organomet. Chem., 1976, 114, C11. 13' A. L. Balch and L. S. Benner, J. Organomet. Chem., 1977, 135, 339. Puddephatt P -P -P 1’ coI I II (CO),Mn -Mn(CO), f--(C0)2Mn -Mn(CO)2 % (CO), Mn -H-Mn(CO),Cl slow I I I ‘%;I II P PvP VP P -P P-P (C0)3Mn-H-Mn(C0)2(CN)II P \/pScheme 9 Complex (25) has an interesting chemistry as shown in Scheme 9.It can be seen that ligand addition occurs slowly to convert the 4-electron carbonyl group into a normal terminal carbonyl gr~up,~~O-~~* and that these reactions are greatly accelerated by protonation of the Mn-Mn bond.’33 The Mn-H-Mn linkage is formulated as being linear, and this would be unique if correct. Addition of diazomethane gives an unusual dimetallacycle by coupling with one of the carbonyl ligand~.’~~ Another interesting complex is [Rez(p-H)z(C0)6(p-dppm)]having a Re=Re double bond. Again this complex shows high reactivity with small molecules, giving L.S. Benner, M. M. Olmstead, and A. L. Balch, J. Organomet. Chem., 1978, 159, 289. 133 H. C. Aspinall and A. J. Deeming, J. Chem. SOC.,Chem. Commun., 1981, 724. 134 G. Ferguson, W. J. Laws,M. Parvez, and R. J. Puddephatt, Organometallics, 1983, 2, 276. Chemistry of Bis(dipheny1phosphino)methane an adduct [Re2H(p-H)(CO)6(P(OMe)3}(p-dppm)]with trimethylphosphite and insertion products [Re2(p-H)(p-RN=CH)(C0)6(p-dppm)] and [Re2(p-H) (p-MeCH=N)(CO),(p-dppm)] with alkyl isocyanides and with methyl cyanide respectively. A further rich chemistry of this complex can be predicted.l3' 6 Cluster Complexes with dppm The use of dppm to stabilize dimers is better known, but stabilization of cluster complexes is also known. For example, heating [Rh,(CO),,] with dppm gives [Rh6(p3-C0)4(C0)6(p-dppm)3],which contains an octahedron of rhodium atoms with dppm bridges between adjacent rhodium centres.'36 Similarly, [RWO)121 gives [Ru3(CO),o(P-dPPm)I.l3 An example of a cluster being formed from a dimer is seen in the reaction of [Pt2H2(p-H)(p-dppm)2]+ with CO-H20-02 to give the tetranuclear [Pt4(p-C0)2(p-dppm)3(Ph2PCH,POPh2)].This has a butterfly structure (27) with one partially oxidized dppm ligand.I3* O( .'O This area should be capable of very significant future development. 7 Complexes with Deprotonated dppm It has been known for several years that dppm can be deprotonated by strong bases to give the anion [(Ph,P),CH]- and that this anion is itself a good ligand.139*140 Many complexes of this ligand are oligomeric and sparingly soluble, but more tractable derivatives have been prepared recently (equation 13; M = Ni, Pd, or Pt).14' 135 M.J. Mays, D. W. Prest, and P.R. Raithby, J. Chem. Commun.,1980, 171. 13' A. Ceriotti, G. Ciani, L. Garlaschelli, U. Sartorelli, and A. Sironi, J. Organomet. Chem., 1982, 229, C9. 13' F. A. Cotton and B. E.Hanson, Znorg. Chem., 1977, 16, 3369. 138 A. A. Frew, R. H. Hill, Lj. Manojlovic-Muir, K. W. Muir, and R. J. Puddephatt, J. Chem. SOC., Chem. Commun., 1982, 198. K. Issleib and H. P. Abicht, J. Prakt. Chem., 1970,312,456. K. Issleib, H. P. Abicht, and H. Winkelmann, Z. Anorg. A&. Chem., 1972, 388,89. 14' J. M. Bassett, J. R. Mandl, and H. Schmidbaur, Chem. Ber., 1980, 113, 1145.Puddephatt 1’11 1 (13) Co-ordinated dppm can also be deprotonated in some cases and this can be a particularly useful synthetic procedure (equation 14).142 t Ph 9p2 E*>R< > P Ph 2 Bridging dppm as well as chelating dppm can be deprotonated in these reactions as illustrated in equation 15. Here the carbon atom as well as the phosphorus atoms co-ordinates in the product.143 A significant chemistry derived from the carbanion can be developed (Scheme 10).144*14s A number of derivatives of the elements copper and gold are known. These may be prepared by planned syntheses such as shown in equation (16).146 - Ph2 [AuCl(PMe3)] 1. LiCH(PPh2)2 2. Et2MeP=CH2 Ph2 14* J. Browning, G. W. Bushnell, and K. R. Dixon, J.Organomet. Chem., 1980, 198, C11. 143 G. M. Dawkins, M. Green, J. C. Jeffery, and F. G. A. Stone, J. Chem. SOC., Chem. Commun., 1980, 1120. 144 S. Al-Jibori and B. L. Shaw, J. Chem. SOC., Chem. Commun., 1982, 286. 14’S. Al-Jibori, W. S. McDonald, and B. L. Shaw, J. Chem. SOC., Chem. Commun., 1982, 287. 146 H. Schmidbaur and J. R. Mandl, Angew. Chem., Znt. Ed. Engl., 1977, 16, 640. Chemistry of Bis(diphenylphosphino)methane P ph*p>< >,CO,,H /PhZPCI P H Scheme 10 However, in other cases the deprotonation of dppm has occurred in a more serendipitous way, as illustrated in equations 17-19.147-149 atomic Au + dppm -, [A~~(p-dppm),(,u~-~~-Ph~PCHPPh,)]+(17) PhCu + dppm -+ [Cu3(,u2-Ph2PCHPPh2)(,u3-Ph2PCHPPh2),](18) Ph 2 I’h 2 P-Au-P [Au,3CIZ(PMe*Ph),[J + dppm + H(/(-),(19) ‘P-Au--P Ph2 Ph2 Again there is obviously great potential for expansion of this area of chemistry, based on the very recent planned syntheses described above.8 Catalysis with dppm Complexes The exceptional ability of complexes with the M2(p-dppm), skeleton to activate small molecules by co-ordination in the bridging position has led to several attempts to develop useful catalysis by these compounds. The complexes [Rh2(CO),(p-dppm)2] and [Pd2C1,(p-dppm),] will each ‘add alkynes in the bridging position, and they act as catalysts for hydrogenation of acetylene to ethylene and for cyclotrimerization of alkynes re~pectively.’~* lo Similarly 14’ J. W. A. van der Velden, J.J. Bour, F. A. Vollenbroek, P. T. Beurskens, and J. M. M. Smits, J. Chem. SOC., Chem. Commun., 1979, 1162. 14‘ A. Camus, N. Marsich, G. Nardin, and L. Randaccio, J. Organomet. Chem., 1973, 60,C39. C. E.Briant, K. P.Hall, and D. M.P.Mingos, J. Organomet. Chem., 1982, 229, C5. Puddephatt [Rh2(p-S)(CO), (P-dPP42 I, Pr2(P-SW0)2 (P-dPPm ),I, and [Ira H* (P-S)(CO12 (p-dppm),] are catalysts for hydrogenation of acetylene and ethylene.' '' There is a correlation between hydrogenation activity and the ability of rhodium A-frame complexes to bind small molecules in the 'pocket' position,"' but no complexes can be said to have exceptionally high activity. The complex cations [Rh ,(p-H)(CO),(dppm),] and [Pt, H, (p-H )(dppm),] ++ are catalysts for the water-gas shift reaction under mild condition^.'^**'^^ In each case it is considered probable that binuclear activation is involved but the detailed mechanisms are not certain.The rhodium catalyst will also act as a hydroformylation catalyst using CO-H20 as the source of hydrogen via the concurrent catalysis of the water-gas shift reaction."' The complex cation [Pt2H2(p-H)(p-dppm),]+ is also a shortlived catalyst for reduction of CO to CH4 using sodium borohydride as reducing agent. Again the mechanism is unknown. These studies represent only the first preliminary survey of catalytic activity of dppm complexes, and it is too early to tell if the demonstrated ability of dppm bridged complexes to stabilize unusual bonding modes of small molecules like carbon monoxide or alkynes will form the basis for useful catalysis.Acknowledgments. It is a pleasure to acknowledge the help of researchers in this area, A. L. Balch, M. P. Brown, M. Cowie, Lj. Manojlovik-Muir, R. Hoffmann, J. T. Mague, A. R. Sanger, and K. R. Seddon, who kindly sent reprints and preprints of their recent work and, in several cases, communicated unpublished results, and B. L. Shaw who suggested that this article be written. 150A. R. Sanger, personal communication. 151 C. P. Kubiak, C. Woodcock, and R. Eisenberg, Inorg. Chem., 1982, 21, 2119.

ISSN:0306-0012    

DOI:10.1039/CS9831200099

出版商:RSC    

年代:1983

数据来源: RSC

摘要:

4-Dialky laminopyridines :Super Acylation and Alkylation Catalysts By Eric F. V. Scriven REILLY TAR & CHEMICAL CORPORATION, 1510 MARKET SQUARE CENTER, 151 NORTH DELAWARE STREET, INDIANAPOLIS, IN 46204, U.S.A. 1 Introduction Treatment of amines, alcohols, and phenols with acetic anhydride’ (or acetyl chloride2) in the presence of pyridine has provided a general3 acetylation method since the turn of the century. However, this procedure often proves to be unsatisfactory for the acetylation of deactivated substrates. It was not until the late 1960’s that certain 4-dialkylaminopyridines were found, independently by two research to be much superior to pyridine as catalysts for difficult acylations.* 1-Methylcyclohexanol, a sterically hindered alcohol, gave an 86 yield of acetate (1) on reaction with acetic anhydride in the presence of a catalytic amount of 4-dimethylaminopyridine (DMAP) and 1 equivalent of TEA.5 Less than a 5% yield of (1) was obtained by the traditional method (Scheme 1).(1) Reagents: i, Ac,O, pyridine, 14h, r.t. ( <5 % yield); ii, Ac,O, DMAP (4 mol %), TEA, 14h, r.t. (86% yield) Scheme 1 The superiority of DMAP over comparable bases as a catalyst for the benzoylation of m-chloroaniline and benzyl alcohol is apparent from kinetic studies6 (Table 1). Namely, catalytic activity is not directly related to pK,, and it is reduced by the attachment of ambstituent CI to the pyridine nitrogen. * Acylation is used in its generic sense, acetylation refers to XH -+XAc.A. Verley and F. Bolsing, Ber., 1901, 34, 3354.’A. Einhorn and F. Hollandt, Liebigs Ann. Chem., 1898,301, 95. A. 0.Fitton and J. Hill, ‘Selected Derivatives of Organic Compounds’, Chapman Hall, London, 1970. L. M.Litvinenko and A. I. Kirichenko, Dokl. Akad. Nauk. SSSR, Ser. Khirn., 1967, 176, 97; Chem. Abstr., 1968, 68, 68 325. W. Steglich and G. Hofle, Angew. Chem., Int. Ed. Engl., 1969, 8, 981. A. I. Kirichenko, L. M. Litvinenko, I. N. Dotsenko, N. G. Kotenko, E. Nikkel’sen, and V. D. Berestetskaya, Dokl. Akad. Nauk. SSSR, Ser. Khim., 1979, 244, 1125; Chem. Abstr., 1979, 90, 157601. 4-Dialkylaminopyridines :Super Acylation and Alkylation Catalysts Table 1 Eflect of various bases on the relative rates of benzoylation of m-chloroaniline and benzyl alcohol Relative rate' Catalyst PK, m-Chloroanilineb Benzyl alcohol' 3-Pyridinecarbonitrile 1.34 14 12 Quinoline 4.87 138 545 Pyridine 5.23 568 9.29 x 103 Isoquinoline 5.40 2.62 x 103 3.39 x 103 2-Methylpyridine 5.96 29 43 5 3-Methylpyridine 5.63 1.12 x 103 2.29 x 104 4-Meth ylpyridine 6.02 2.96 x 103 3.98 x 104 4-Phenox yp yridine 6.25 4.80 x 103 7.98 x 104 2,6-Dimethylpyridine 6.72 8 115 DMAP 9.70 3.14 x lo6 3.45 x lo8 TEA 1Q.65 21 - rate of catalysed reaction (a) Relative rate = (6)A.I. Kirichenko, L. M. Litvinenko, I. N. Dotsenko,rate of uncatalysed reaction' N. G. Kotenko, E. Nikkel'sen, and V. D. Berestetskaya, Dokl. Akad. Nauk SSSR, 1979, 244, 1125; Chem.Abstr., 1979, 90, 157601~.(c) L. I. Bondarenko, A. I. Kirichenko, L. M. Litvinenko, L. N. Dmitrenko, and V. D. Kobets, J. Org. Chem. (USSR), 1982, 2310. 4-Dialkylaminopyridines were soon found to have general applicability for catalysis of acylations and related reactions, and the subject was reviewed7 in 1978. That review coupled with the commercial availability of DMAP in large quantities, for the first time, stimulated great interest in its use as a catalyst in organic, polymer, analytical, and biochemistry. This review is principally concerned with the developments in the applications of DMAP that have taken place in the last five years. 2 Preparation and Physical Properties DMAP and other 4-dialkylaminopyridines may be prepared fairly easily from suitable 4-substituted pyridines (Scheme 2).However, these methods do not provide viable commercial routes because of the relatively high cost of starting materials and the nature of the reaction conditions. Two industrial processes have been One, by Schering AG,' exploits a modification of an old reaction, that of a pyridylpyridinium salt with nucleophiles" (e.g.DMA). DMF is used in the process instead of DMA for the preparation of DMAP, which appears to avoid the formation of 4-aminopyridine G. Hofle, W. Steglich, and H. Vorbruggen, Angew. Chem., Int. Ed. Engl., 1978, 17, 569.* H. Vorbruggen, Schering AG, U.S.P., 4 140853 (1979); Ger. Offen., 2 517774 (1976); Chem. Abstr., 1977,86, 55 293. ''T. D. Bailey and C. K. McGill, U.S.P., 4 158093 (1979); Chem. Abstr., 1979, 91, 123 636.lo E. Koenigs and H. Greiner, Ber., 1931, 64, 1049. Scriuen 0 H 52 -51% 48% high Reagents: i, HMPT, 2-4h, 220-230°C (ref. a); ii, DMA, ZnCI,, 2h, 150-160°C (ref. b); iii, DMA, HgCl,, 48h, 120°C (ref. c); iv, DMA, EtOH, 2h, 100°C; 12h, 115-130°C (sealed tube) (ref. d); v, DMA.HBr, lh, 200-210°C (ref. e) (a) H. Vorbriiggen, Angew. Chem., Znt. Ed. Engl., 1973, 12, 301. (b) Y. Suzuki, Yakugaku Zasshi, 1961, 81, 1146. (c) H. Vorbriiggen, Angew. Chem., Int. Ed. Engl., 1972, 11, 305. (d) J. W. Wibaut and F. W. Brockman, Recl. Trav. Chim. Pays-Bas, 1961, 80, 309. (e) A. F. Vompe, N. V. Moniih, N. F. Turitsyna, and L. V. Ivanova, Dokl. Akad. Nauk SSSR, 1957,114,1235;Chem. Abstr., 1958,52, 3803.Scheme 2 as a by-product (Scheme 3). The other,g developed by Reilly Tar & Chemical Corporation, involves a novel reaction sequence (Scheme 4). In the first step, 4-cyanopyridine is quaternized with 2-vinylpyridine, which activates the 4-position towards attack by DMA to give (2). This quaternary salt is readily depyridethylated by sodium hydroxide to yield DMAP, and 2-vinylpyridine Scheme 3 4-Dialkylaminopyridines: Super Acylation and Alkylation Catalysts that is available for recycling. Recently, 2-and 4-vinylpyridines have found use as protecting groups in the alkylation of benzimidazoles.' The most popular catalyst for laboratory and industrial use has proved to be DMAP (3). 4-Pyrrolidinopyridine (PPY) (4) is slightly superior to DMAP as a catalyst but this advantage is counterbalanced by its higher cost and lack of availability. A new liquid catalyst (MPP) (5) that is as effective as DMAP has been reported'* recently.Some physical constants for (3)--(5) are given in Scheme 5. Me0N (3) (4) (5) m.p. 112 -113 OC m.p. 57 -58 OC b.p. 148OC/4 mmHg [ 1 122-58-31 [2456-81-71 [ 80965-30-61 Scheme 5 3 Mechanism of Catalysis The study of catalysis of acylations and related reactions has a long history.I3 Indeed, Scheele rep~rted'~ the effect of acid and alkali on esterification and l1 M.Ichikawa, C. Yamamoto, and T. Hisano, Chem. Pharm. Bull.. 1981, 29, 3042. l2 Reilly Report 5, 'DMAP Update', Reilly Tar & Chem. Corp., Indianapolis.1982.'' M.L. Bender, Chem. Rev., 1960, 60, 53. l4 J. W. Baker and E. Rothstek-in 'Handbuch der Katalyse', ed. G. M. Schwab, Springer, Vienna, 1940, Vol. 2, p. 4b. Scriuen hydrolysis in 1792. As this review is concerned mainly with applications, only a few important points about the mechanisms of catalysis will be outlined. Pyridines may act as nucleophilic or general base catalysts. A. Nucleophilic Catalysis.-The hydrolysis of acetic anhydride (acetylation of water) in the presence of pyridine has been shown to proceed by nucleophilic catalysis, and the unstable acetylpyridinium ion (6) was proposed as an intermediate. The mechanism (Scheme 6) was formulated on the basis of kinetic analysis." H20 3 @ + HOAc Ac H Scheme 6 Subsequently, the acetylpyridinium ion was observed' spectrophotometrically [k272nm (E ca.4.4 x lo3) and 1225nm (7 x lo3)]in the above reaction, and in the pyridine catalysed acetylation of anilines. Pyridinolysis of 2,4-dinitrophenyl methyl carbonate,' 79'8 2,4-dinitrophenyl chloroformate2' by a series of pyridines has been found to give curved Brsnsted plots. These plots show a large dependence of rate constants on pK, (/3-0.8-1.0) for most pyridines but it becomes small (/3-0.1-4.3) for the most basic ones. Curvature has been attributed to a change over from rate determining breakdown to formation of the tetrahedral intermediate (7).21This is e~emplified'~ by the Brsnsted plot (Figure 1)for pyridinolysis of acetic anhydride in aqueous solution by a series of pyridines of increasing basicity (Scheme 7).The intermediate salt (8) is most readily formed in the case of the most basic pyridine in the series, DMAP. Unlike pyridine, such salts of DMAP may be isolated and are often quite stable (see below). Hydrolysis of intermediate salt (8) is subject to general base catalysis. The mechanism of breakdown of salts of the type (8) depends upon the nature of R22 and the N-s~bstituent.'~ Hydrolysis, by water of a series of 1-methoxycarbonylpyridinium ions was found22 to be general base-catalysed and formation of the tetrahedral intermediate was rate determining (Scheme 8). A bent Brsnsted plot was obtained which could not be explained by a change in the rate determining step, as (9) was l5 A.R. Butler and V. Gold, J. Chem. SOC., 1961, 4362. l6 A. R. Fersht and W. P. Jencks, J. Am. Chem. SOC., 1970, 92, 5432. l7 E. A. Castro and F. J. Gil, J. Am. Chem. Soc., 1977, 99, 7611. l8 E. A. Castro and M. Freudenberg, J. Org. Chem., 1980,45, 906. l9 C. Castro and E. A. Castro, J. Org. Chem., 1981,46, 2939. 2o P. M. Bond, E. A. Castro, and R. B. Moodie, J. Chem. SOC., Perkin Trans. 2, 1976, 68. '' M.J. Gresser and W. P. Jencks. J. Am. Chem. SOC.,1977, 99, 6970. 22 P. J. Battye, E. M.Ihsan, and R. B. Moodie, J. Chem. Soc., Perkin Trans. 2, 1980, 741. 23 V. A. Saveiova, I. A. Belousova, and L. M.Litvinenko, J. Org. Chem. (USSR), 1982, 1333. 4-Dialkylaminopyridines:Super Acylation and Alkylation Catalysts PKa Figure 1 Br~nsted plot for the pyridinolysis of acetic anhydride by a series of pyridines in aqueous solution at 25 "C (Reproduced by permission from J.Org. Chew., 1981,46, 2942) R HO productsk,_ @R -%@+ Ac20 2, k-1 Me?-0-MeCO OAc-I OAc (7) Scheme 7 ruled out as an intermediate. Curvature was tentatively ascribed to a stabilization of the ions (10) that had electron donating substituents attached at the 4-position. Such stabilization is not available to the other pyridine salts in the series. Therefore, catalysis is not truly reflected by consideration of the pK, alone. The DMAP salt (10) (R = NMe,) provides the prime example of this effect. General base catalysis takes place because the pyridine may take the place of the terminal water molecule in the transition state.The effect of varying the N-substituent, has been studied for the reactions of a series of N-arylsulphonyl-4-dimethylaminopyridiniumsalts with 3-nitro- aniline in di~hloromethane.~ The reactivity of these salts, which exist as ion-pairs in dichloromethane, depends upon the nature of the anion, and it decreases in the order benzenesulphonate > bromide > chloride. Three indepen- dent pathways have been delineated for this reaction; viz. bimolecular reaction of the salt with 3-nitroaniline, trimolecular reaction involving catalysis by a second molecule of salt, and trimolecular reaction with catalysis by a second molecule of DMAP. Some of the complexities of nucleophilic catalysis by pyridines have been outlined above.However, nucleophilic catalysis by 4-dialkylaminopyridines Scriven R R A 7 I A @MeOC =0 I MeO-C -0 -I OH TS products R+6 II ii Me0 -C -0‘ I Me0 C \o-Scheme 8 may be looked at from the ‘opposite’ point of view (Scheme 9). In this simplistic scheme, DMAP and its cogeners are considered as ‘catalytic supports’ for an electrophile (E) that promote reaction with a substrate (Nu-H). Examples of electrophiles that contain the elements C, Si, P, and S attached at N-1 are common (see below), but N-analogues do not seem to have been reported. A few points of practical importance about Scheme 9 are worthy of note. In the first step, DMAP, MPP, and PPY form N-acylpyridinium salts much E = electrophile A = counterion or its precursor Nu = nucleophile B = strong base Scheme 9 4-Dialkylaminopyridines:Super Acylation and Alkylation Catalysts more rapidly than other pyridines (Figure 1).Once formed these salts are stabilized by resonance (lo), and in a number of cases can be isolated, e.g. (11),24 (12), ’’and (13).” They exist as loose ion-pairs in non-polar solvents which facilites nucleophilic attack at the acyi group. The second step is subject to general base catalysis by the counter-ion. Thus acid anhydrides are better acylating agents than the corresponding acid chlorides under these conditions. In practice, a strong base such as TEA is often used to speed up the second step, and if a stoicheiometric amount is used it also has the advantage of preventing protonation of the catalytic amount of DMAP by acid formed in the reaction.As a consequence of these features, acylations in the presence of DMAP are usually carried out using acid anhydrides, a catalytic amount of DMAP (ca. 5-20 mole %), and a stoicheiometric quantity of an auxiliary base (q.TEA) in a nonpolar solvent (e.g. dichloromethane). B. General Base Catalysis.-DMAP has been found to catalyse the formation of polyurethanes in the reaction of isocyanates with alcohols and phenols.26 Recently, the pyridine-catalysis of a similar reaction, that of methanol with phenyl isocyanate in tetrachloromethane was studied.27 A Brransted plot with a slope /?of 0.49 was obtained for catalysis by a series of 3-and 4-substituted pyridines.6b’ tj2BF4-(3CN c1-c1-S02C6H4Me-P CPh 3 m.p. 124 OC (dec.) m.p. 120 OC m.p. 128 -129 OC (1 1) (1 2) (13) NMe NPh NMe 2 52 [lp Meo4]+ 0 y I NPh t H II Me02CNHPh ‘0% Me’ II 0 Scheme 10 24 E. Guibe-Jampel, M. Wakselman, and D. Raulais, J. Chem. SOC., Chem. Commun., 1980, 993. 25 M. Wakselman, E. Guibe-Jampel, A. Raoult, and W. D. Busse, J. Chem. SOC., Chem. Commun., 1976, 21. 2b H. J. Twitchett, B. P., 990637 (1965); Chem. Abstr., 1965, 63, 766. 27 R. B. Moodie and P. J. Sanson, J. Chem. SOC., Perkin Trans. 2, 1981, 664. Scriven A general base catalysis mechanism (Scheme 10) was proposed for this reaction from the /I-value, the steric effect of 2,4,6-trimethylpyridine deuterium isotope, and solvent effects.This reaction provides one of the few that has been shown to involve general base catalysis by DMAP. Almost all the important synthetic applications of DMAP to be discussed probably occur by nucleophilic catalysis, but not all individual examples have been examined closely. 4 Applications in Synthetic Organic Chemistry A. Typesof Reaction Catalysed.-(i) Acylation. DMAP is now commonly used to facilitate the acylations of sterically hindered and other deactivated alcohols. Several studies have employed this reaction to compare the catalytic ability of a number of 4-dialkylaminopyridines and related compounds. The acetylation of 1,l-diphenylethanol to (14) with 2 molar equivalents of acetic anhydride and TEA in the presence of 0.1 molar equivalents of catalyst was stopped before completion, and the progress of the acetylation was determined by n.m.r.or g.c.** The relative effects of the catalysts compared to the best one, PPY, are given (Scheme 11). This order follows the ordei-of reactivity of the Ac20 PhzMeC-OH PhzMeC -OAc (1.0) (0.9) (0.63) (0.12) PPY DMAP OPh, PPh2, SCHzPh, NH2, R= I n Scheme 11 28 A. Hassner, L. R. Krepski, and V. Alexanian, Tetrahedron, 1978, 34, 2069. 4-Dialkylaminopyridines :Super Acylation and Alkylation Catalysts enamines of the respective amines towards ele~trophiles.~' A recent study of the acetylation of t-butyl alcohol, carried out in a similar way, also indicates that PPY (1.0) is the best cataly~t.~' However, DMAP (0.9) and MPP (0.75), the new liquid catalyst, are almost as good; but N-methylimidazole (0.1) is much inferior. The acetylation of mesitol is easier than those described above, and in this reaction DMAP and MPP are again about equally as effectiveI2 (Scheme 12).OH OAc ____)i Me 0"' "'0"' Me Me (0.1mol) Reagents: i, Ac20 (0.16mol), 2h, r.t., CH2CI2, alone (0%) +DMAP (94%) +MPP (91%) Scheme 12 Acetylation of the t-alkynol (15) is sufficiently easy in the presence of DMAP that the acid labile acetal function survives (Scheme 13).3' DMAP has found use in the preparation of (propargyl acetate)dicobalt hexacarbonyl complexes which dimerize readily in the presence of trimethylal~minium.~ Me I Me-C-CCC-C-H OMe I Me I1Me-C-C=C-C-H OMe I I OH I OMe I OAc I OMe (15)(18mmol) (93%) Reagents: i, Ac20 (25 mmol), DMAP (30 mmol), CH2C12, 20 min, r.t.Scheme 13 Sterically hindered phenols are very difficult to formylate by the conventional Reimer-Tiemann, Gattermann, and Duff procedures. The sequence (Scheme 14), which depends upon DMAP catalysis in the first step, provides a high yield route to sterically hindered o-hydro~ybenzaldehydes.~~ The value of isatoic anhydrides as an o-aminobenzoylating agent for primary amines and alcohols is enhanced when it is used with DMAP, as the competing reaction (Path B) is suppressed (Scheme 15).34 A DMAP salt (16) was suggested as an intermediate. 29 S. F. Dyke, in 'Chemistry of the Enamines', Cambridge University Press, Lpndon, 1973.30 G. L. Goe, L. M. Huckstep, and E. F. V. Scriven, Chem. Ind. (London), 1982, 722. 31 G. Hofle and W. Steglich, Synthesis, 1972, 619. 32 S. Padmanabhan and K. M. Nicholas, J. Orgunmet. Chem., 1981, 212, 115. 33 D. 3. Zwanenburg and W. A. P. Reynen, Synthesis, 1976, 624. 34 M. C. Venuti, Synthesis, 1982, 266. Scriven (0.1 mol) (0.22 mol) iii, iv i But But 80% Reagents: i, DMAP (4.1 mmol), CHC13, 10h, reflux; ii, AIC13 ;iii, LAH; iv KIO,. Scheme 14 0 (CONHR)-0:: H \ 0 Path A or ROH 0NHCOR Path (NHCONHR) Yields: NHR; R = Bun (93%), PhCH2 (97%),Bu'. (73%) OR; R = Ph (98%), PhCH, (90%),PhzCH (81%) Scheme 15 C-and 0-acylation of enolates are known (equations 1-3).35-37 The presence of just a catalytic amount of DMAP proves useful for the N-acetylation of indoles and related heterocycles (equation 4).38 Acylation and esterification using preformed mixed anhydrides constitutes a well established method, but preparation of such compounds is not always 35 T.J. Cousineau, S. L. Cook, and J. A. Secrist, 111, Synth. Comrnun., 1979, 9, 157. 36 D. H. R. Barton, E. Buschrnann, J. Haiisler, C. W. Holzapfel, T. Sheradsky, and D. A. Taylor, J. Chern. SOC., Perkin Trans. 1, 1977, 1107. 3' H. Hofmann, H.-J. Haberstroh, B. Appler, B. Meyer, and H. Herterich, Chem. Ber., 1975, 108, 3596. 38 K.Nickisch, W. Klose, and F. Bohlmann, Chem. Ber., 1980, 113, 2036. 4-Dialkylaminopyridines :Super Acylation and Alkylation Catalysts H (0.2 rnol) (79%) ii COCOzEt(MeO)2CHCxyo v (MeO)2CHCXjoH S S (64%) Meo Ph a 0->@OAc (11.5 mmol) (70%) H Ac (89%) Reagents: i, DMAP (0.02 mol), TEA (0.4 mol), Ac20 (1.0 mol), 12-14h, 50 "C; ii, CICOC02Et, DMAP, pyridine, CH2C12, 3h, reflux; iii, Ac20, DMAP, TEA, MeCN, -10 "C, 4 days; iv, Ac20 (1.2 mmol), DMAP (1 mmol), TEA (1.2 mmol), CH2C12, 24h, r.t.easy.39 Use of DMAP allows the high-yield generation of mixed anhydrides in situ (Scheme 16). This may be achieved by the slow addition of acetic anhydride to formic acid at low temperature.' The reaction mixture is then warmed slowly to room temperature to ensure that the alcohol reacts yO+j -..., =0OCH0111 (72%) 1 11. t.. HC02H Reagents: i, DMAP, TEA, CH2Cll, -40°C; ii, Ac,O added over 30min, -35°C; iii, warmed to r.t.over lh Scheme 16 39 E. Haslam, Tetrahedron, 1980, 2409. Scriven selectively at the formyl group of the anhydride. Maintenance of strict temperature control, and the mode of addition are particularly important in the first step to avoid self condensation of the acetic anhydride (Scheme 17). The mixed anhydride approach also avoids the use of halides, which tend to give poorer yields than anhydrides. 0 AczO DMA', MeQo OH\ COMe + Me 6COMe + 4MeMeoC I I Me Me 0 0 (16%) (3 70) (0.3%) Scheme 17 Treatment of the mixed anhydride lactone (17) with acetic anhydride and DMAP in pyridine results in acetylation with epimerization at C-3 (Scheme 18).40 Presumably,, epimerization occurs first with relief of steric crowding at C-5 and this permits acetylation. The methyl ester (18), that has a less acidic proton attached to C-3, does not undergo reaction.DMAP has been employed to catalyse many other types of acylation, including pho~phorylations,~ sulphonyla-ti on^,^' and s~lphinylations.~~ (17) R = Ac R = Ac (18) R=Me Scheme 18 (ii) Tritylation. A cis-trans mixture of 4-t-butylcyclohexano1 gives almost exclusively the trans (equatorial) ether on reaction with trityl chloride and DMAP (Scheme 19).43This result was attributed to a kinetic factor origmating from the large steric requirement of the alkylating agent, the N-trityl-4-dimethylaminopyridinium salt (19). Selective tritylation of a primary alcohol 40 R.M. Carman and S. S. Smith, Aust. J. Chem., 1981, 34, 1285. 41 E. GuiM-Jarnpel, M. Wakselman, and D. Raulais, J. Chem. Soc., Chem. Commun., 1980, 993. 42 P. W. Henniger and J. K. Van Der Drift, Ger. Offen., 2235390 (1973); Chem. Abstr., 1973, 78, 124608. "S. K. Chaudhary and 0.Hernandez, Tetrahedron Lett., 1979, 95. 141 4-Dialkylaminopyridines :Super Acylation and Alkylation Catalysts 97% 3% 70% Conversion Reagents: i, Ph,CCl, DMAP, CH2C12, 4045"C, 18-24h Scheme 19 in the presence of a secondary one may be achieved with this salt (equation 5).44 MPP is as effective as DMAP in catalysing the tritylation of benzyl alcohol. l2 NMe0LL o 1 3 t L 0 C P h 3 (5) C1- CPh 3 (100%) (19) tDMAP Reagents: i, CH,Cl,, 25 "C, 16h (iii) Silylation.DMAP is very much superior to imidazole for promoting the selective silylation of primary alcohols to TBDMS ethers (equations 6 and 7).45 Furthermore, only a catalytic amount of DMAP is required (cf: imidazole, equation 6), and use of DMF as a solvent can be avoided. Diols OH Ph L O R + Ph L O H + PhLORPh Reagents: i, TBDMCS, DMAP (4 mol%), 95% 0 5% CH2Cl,, TEA, 8h, r.t.; (4)ii, TBDMCS, imidazole 59% 11% 30% (220 mol%) DMF, 8h, r.t. (98%) (trace) Reagents: i, TBDMCS, DMA (4 mol%), CH,C12, TEA, 8h, r.t.; TBDMCS = t-butyldimethylchlorosi-lane, R = t-butyldimethylsilyl 44 0. Hernandez, S. K. Chaudhary, R. H. COX, and J. Porter, Tetrahedron Lett., 1981, 1491. 45 S. K. Chaudhary and 0.Hernandez, Tetrahedron Lett., 1979, 99. 142 Scriven may be derivatized readily with di-t-butyldichlorosilaneand DMAP in acetonitrile (equation 8). Di-t-butylsilylene derivatives are more stable towards Lewis acids than the corresponding isopropylidene and diphenylsilylene derivatives. They have the added advantage of readily undergoing cleavage by pyridine hydr~fluoride.~~ (1 equiv.) (65%) Reagents: i, iv, DMAP (2 equiv.), TEA (4 equiv.), lh, 70 "C (iv) Esterification and Lactonization. Esterification carried out by treating an alcohol with an anhydride is merely an example of acylation [see section 4.A(i)]. Nevertheless, one example where the use of DMAP is invaluable is worth mentioning. The acid sensitive alcohol (20) affords the ester (21) (also acid sensitive) in virtually quantitative yield when it reacts with propionic anhydride, TEA, and DMAP (Scheme 20).47 Me u SiMe 3 SiMe3 Me ToOH 0 Reagents: i, TEA, DMAP, 25 "C Scheme 20 Extension of the above method to the esterification of carboxylic acids is less successful, as only half of the starting acid is esterified even under ideal conditions.Furthermore, one equivalent of base (TEA) is required and it is often necessary to preform the anhydride. These difficulties were overcome by three groups of workers at about the same time (Scheme 21).48-s0 The beauty of this method lies in the fact that use of 1.1 equivalents of DCC allows 46 B. M. Trost and C. G. Caldwell, Tetrahedron Lett., 1981, 4999.47 S. R. Wilson and M. F. Price, J. Am. Chem. Soc., 1982, 104, 1124. 48 B. Neises and W. Steglich, Angew. Chem., Int. Ed. Engl., 1978, 17, 522. 49 A. Hassner and V. Alexanian, Tetrahedron Lett., 1978, 4475. F. E. Ziegler and G. D. Berger, Synth. Commun., 1979, 539. 4-Dialkylaminopyridines:Super Acylation and Alkylation Catalysts DCC urea(1.1 mmc ph+; PhC02H I 94% 10% (in absence of DMAP) (in absence of DCC) Scheme 21 almost complete conversion of acid into ester, and the other disadvantages are removed. The method is well-suited to esterification of highly hindered alcohols and N-protected amino-acids; and it may be used for the introduction of bulky protecting groups (e.g. benzyl and t-butyl). Thiol esters may be prepared by this approach (equation 9).” It also provides an excellent cyclization method.The example chosen illustrates two cyclization~,~ which both yield a potentially important new orally active antihypotensive agent (22) (Scheme 22).52 The diastereomer (23) exhibits no such activity. COzH COSBu‘ (10 mmol) (91%) Reagents: i, DMAP (5 molx), CH2C12,then DCC at 0°C for 5 min, then 3h at 30°C The mixed anhydride approach incorporating DMAP catalysis provides an excellent method for the preparation of sterically hindered macrocyclic lactones (Scheme 23),53 and thiol esters.54 The employment of 2,4,6-trichloro-benzoyl chloride to form the mixed anhydride has a double advantage. Closure 51 D. H. Kim, J. Heterocycl. Chem., 1980, 17, 1647. 52 M.A. Ondetti, U.S.P., 1980, 4192945. 53 J. Inanaga, K. Hirata, H. Saeki, T. Katsuki, and M. Yamaguchi, Bull. Chem. SOC. Jpn., 1979, 52, 1989. 54 Y. Kawanami, Y. Dainobu, J. Inanaga, T. Katsuki, and M. Yamaguchi, Bull. Chem. SOC. Jpn., 1981, 54, 943. Scriven r (k?* DCC bCO2H 0JJy Go+ SH (22) (70%) H Me DCC-9 94% 4 DMAP *(22)0nlY O I G Me (23) (30%) Scheme 22 Meo(CH,),CO,H + EtCHO 0 0 Me (46%) Reagents: i, TEA, THF, r.t., then 2,4,6-trichlorobenzoyl chloride, 2h, r.t.; ii, DMAP, PhH, reflux, 40h Scheme 23 takes place exclusively at the less hindered carbonyl group of the mixed anhydride, and trichlorobenzoate is an excellent leaving group. Similar advantages apply to the use of FTNB with DMAP for similar (v) Conversion of 1,2-Diols into AIkenes.An improved mild procedure appeared recently that utilizes DMAP and thiophosgene for the stereospecific synthesis of alkenes from 1,Zdiols (Scheme 24).58 The advantage of this process is 55 S. Kim and S. Yang, Chem. Lett., 1981, 133. 56 S. Kim and S. Yang, Synth. Commun., 1981, 11, 121. 5’ S. Kim, K. H. Ahn, and S. Yang, Bull. Korean Chern. SOC., 1982, 3, 70. 58 E. J. Corey and P. B. Hopkins, Tetrahedron Lett., 1982, 1979. 4-Dialkylaminopyridines :Super Acylation and Alkylation Catalysts H OTBDMS OTBDMS n-csHl++oH ' ,n-CsH11yJy0 I H H 6H (1 equiv.) (97%) Ph \pHS OTBDMS MeN"NMe + co2 + Hu H Reagents: i, DMAP (2.4 equiv.), CI,CS (1.2 equiv.), CH,Cl,, 0 "C, lh Scheme 24 that the intermediate thionocarbonate can be formed at 0°C in one hour, whereas use of thiocarbonylimidazole requires heating at reflux in toluene or xylene.(vi) Conversion of Nitrimines into Alkynes and Alkenes. Nitrimines, which may be prepared from ketoximes and nitrous acid, undergo fragmentation to alkynes and/or allenes on acetylation with acetic anhydride and DMAP (equation 81 19 (76%) (vii) Oxidation. 4-(Dirnethylamino)pyridinium chlorochromate (24) is a mild selective oxidizing agent suitable for the conversion of allylic and benzylic alcohols into the corresponding carbonyl compounds.60 The selectivity of the reagent was demonstrated in the oxidation of the diol (25) (equation ll), pyridinium chlorochromate gives roughly an equal amount of both products.4-(Dimethy1amino)pyridine 1-oxide has been found to convert active halides into aldehydes and ketones via DMAP salts which were then treated with the super base DBU (Scheme 25).61 59 G. Biichi and H. Wuest, J. Org. Chem., 1979,44, 4116. 6o F. S. Guziec, jun., and F. A. Luzzio, J. Otg. Chem., 1982, 47, 1787. 61 S. Mukaiyama, J. Inanaga, and M. Yamaguchi, Bull. Chem. SOC.Jpn., 1981,54, 2221. Scriven -i L O H L O H (25) (62%) ( <2%) NMe 7 I Br -I 0-OCHqPh H202,H' ($jJ PhCHOI I (98%) Scheme 25 (viii) Formation of Heterocycles. 4-Methyl-6-hydroxy-2-pyrone (a cyclic anhydride) dimerizes when heated with a catalytic amount of DMAP in xylene. This reaction does not occur in the absence of DMAP.The so-formed dimer subsequently undergoes decarboxylation to afford a coumarochromanone (Scheme 26).62 Condensations of b-enaminoesters with formaldehyde are catalysed by DMAP (Hantzch Synthesis) (Scheme 27).63 Aryl isocyanates trimerize to triazinones on heatng with DMAP in ethyl acetate.' Annelation of carbonyl sulphide to cyclic o-aminonitriles is much easier in the presence of DMAP (equation 12).64 62 S. D. Burke, J. 0.Saunders, and C. W. Murtiashaw, J. Org. Chem., 1981, 46, 2425. H. Wamhoff, G. Hendrikx, and M. Ertas, Liebigs Ann. Chem., 1982, 489. 64 M. A. Hernandez, F.-L. Chung, R. A. Earl, and L. B. Townsend, J. Org. Chem., 1981, 46, 3941. 4-Dialkylaminopyridines: Super Acylation and Alkylation Catalysts Me Me Me Me 0 -c02 Me H (42%) Reagents: i, DMAP, xylene, reflux Scheme 26fl::Me02C DMAP M~ Me NH2 H2N H (52%) Scheme 27 95 OC,62h (74%) + DMAP, 95 OC, 37h (100%) Scriuen B.Applications in Natural Products Chemistry.--6Dialkylaminopyridines are now used so commonly in natural products chemistry that only a selective account can be given here. Other examples may be found in the two reviews Use of DMAP-DCC facilitates the penultimate step in a new synthesis of some macrocyclic spermidine alkaloids, e.g. lunarine (26) (Scheme 28).65 nn Scheme 28 DMAP has also found application in the synthesis of racemic tenellin,66 ap~vincamine,~’and spermidines; 68 the resolution of a 9-oxaergoline; 69 and the acylation of 11,12-dihydroglaziovine.70Treatment of (27) with acetic anhydride in pyridine results in mono-0-acetylation to give (28), acetic anhydride-DMAP-TEA is required for di-0-acetylation (Scheme 29).65 E. Fujita, Pure Appf. Chem., 1981, 53, 1141. 66 D. R. Williams and S.-Y. Sit, J. Org. Chem., 1982, 47, 2846. 67 B. Danieli, G. Lesma, and G. Palmisano, Gazr. Chim. Ital., 1981, 111, 257. 68 H. Yamamoto and K. Maruoka, J. Am. Chem. SOC., 1981, 103, 6133. 69 P. S. Anderson, J. J. Baldwin, D. E. McClure, G. F. Lundell and J. H. Jones, J. Urg. Chem., 1982,47, 2184. ‘O J. S. Bindra and A. Grodski, J. Org. Chem., 1977,42, 910. 71 A. S. Mesentsev and V. V. Kuljaeva, Tetrahedron Lett., 1973, 2225. 4-Dialkylarninopyridines:Super Acylation and AIkylation Catalysts OAc Me \AcMeNwoMeMe OH OH (28) Me OH (27) AcMeN Me OAc Reagents: i, Ac,O, pyridine; ii, Ac,O, DMAP, TEA Scheme 29 The diol (29) (an intermediate in the total synthesis of leucogenenol) is resistant to acetylation by acetic anhydride-pyridine, but it gives a 95 % yield of (30) when DMAP is added (equation 13).72 %OH -WoAcIC02Me C02Me (29) (30)(95%) The hydroxy-groups indicated in the steroids listed may be acylated under DMAP or PPY catalysis; hydroxycholesterol (la),73 methyl cholate (3a, 7a, (31) (1 1/3),75 testosterone (17/3).76 Dialkylaminopyridine catalysts have been applied widely7 to assist in the acylation of terpene and related alcohols, e.g.(32),77 (33),78 (34),79 (35).*O /3-Lactam ring formation has been achieved using the DCC-DMAP method, but the yield is poor and epimerization takes place (equation 14).81 72 R. G. Salomon, M. F. Salomon, M. G. Zagorski, J. M. Reuter, D. J. Coughlin, J. Am. Chem. SOC., 1982, 104, 1008. 73 D. H. R. Barton, R. H. Hesse, M. M. Pechet, and E. Rizzardo, J. Am. Chem. SOC.,1973, 95, 2748. 74 G. Hofle and W. Steglich, Synthesis, 1972, 619. 75 W. Steglich and G. Hofle, Ger. Offen., 1969, 1958954; Chem. Abstr., 1971, 75, 34673. 76 J. Miiller and J. E. Herz, Steroids, 1979, 793. 77 L. P. J. Burton and J. D. White, J. Am. Chem. SOC.,1981, 103, 3226. 78 K. Yamakawa and T. Satoh, Chem. Pharm. Bull., 1981, 29, 3474. 79 J. P. Genet and F. Piau, J. Org.Chem., 1981, 46, 2414. ‘O P. A. Grieco, P. A. Tuthill, and H. L. Sham, J. Org. Chem., 1981, 46, 5005. ” T. Kametani, T. Nagahara, Y. Suzuki, S. Yokohama, S.-P. Huang, and M. Ihara, Tetrahedron, 1981, 37, 715. Scriven CH20AC 0@LO (33) (85%) (Ac20, pyridine, DMAP) @FHO HO OH (32) (81%) (Ac20, pyridine, DMAP) (34) (65%) (Ac20, TEA, CH2C12, DMAP) (16.5%) 4-Dialkylaminopyridines:Super Acylation and Alkylation Catalysts DMAP catalysis has proved helpful in the phosphorylation of the oleyltetra- hydrofuran (36) (a Darmstofl analogue) (equation 15).82 OH Reagents: i, POC13, DMAP, pyridine, CH,Cl,, r.t., 6h; ii, H20; iii, Dowex-50 (Na’) 5 Uses in Analytical Chemistry and Biochemistry These two fields are combined under one heading as many of the analytical methods described have biochemical applications.A. Determination of Hydroxy-groups.-The determination of hydroxy-groups in alcohols, phenols, glycols, and sugars is an old analytical problem. This was traditionally carried out by heating the alcohol under reflux in an acetic anhydride-pyridine mixture for 1 hour, or more. The mixture was then titrated,83 or analysed by gas chr~matography.’~ Addition of DMAP to the acetylation mixture permits total conversion into the ester in 5-10 minutes at 54°C for typical primary and secondary alcohols.83 This method has been applied to the analysis of clindamycin palmitate hydr~chloride.~~ (i) Urinary Monohydric and Dihydric Phenols. A procedure for the extraction of mono- and di-hydric phenols from urine has been described.” This involves treating a dilute solution of phenols (viz.p-cresol, 4-methylcatechol, resorcinol, and catechol) with acetic anhydride in the presence of DMAP. The acetate derivatives were then analysed by gas chromatography (OV-7 column). This method is particularly suited for the quantitative estimation of dihydric phenols, which is sometimes difficult. Concurrent azlactone formation, from norleucine and glycine, was noted in the above procedure (see below). (ii) Glycolipids and Diglycerides. Glycosphingolipids react rapidly with benzoic anhydride and DMAP in pyridine to give benzoyl derivatives that can be analysed by h.p.l.c.86 Rapid benzoylation with benzoic anhydride avoids unwanted formation of N-benzoyl derivatives, which occurs when benzoyl chloride is used.DMAP has been used to assist in the attachment of a p-nitrobenzoate chromophore to diglycerides in order to increase the ease of detection of such compounds by h.p.l.~.~’ (iii) Phenolic Groups in Tyrosine Residues. The 1-tosyl and 1-dansyl salts of DMAP react specifically with Tyr residues of proteins in aqueous media to give 0-arylsulphonates (equation 16).88 The tosyl reagent is specific for 82 R. A. Wiley, W. Harris, C. Brungardt, and M. Marx, J. Med. Chem., 1982, 25, 121. K. A. Connors and K. S. Albert, J. Pharm. Sci., 1973, 62, 845; S.-F. Lin and K. A. Connors, J. Pharm. Sci., 1981, 70, 235. 84 E. L. Rowe and S. M. Machkovech, J. Pharm. Sci., 1977, 66, 273.8s V. Fell and C. R. Lee, J. Chromatogr., 1976, 121, 41. 86 S. K. Gross and R. H. McCluer, Anal. Biochem., 1980, 102, 429.*’ M. Batley, N. H. Packer, and J. W. Redmond, J. Chromatogr., 1980, 198, 520. E. Guibe-Jampel, M. Wakselman, and D. Raulais, J. Chem. Soc., Chem. Commun., 1980, 993. Scriven -NHCHCO --NHCHCO -I I NMe 26 (J -6+ tDMAP (16) OH Cl'S02Ar OSOzAr (1 1) (Ar = p-MeC6H4) Reagents: i, pH 10-11.5, Li2C03, then (11) in three portions, 20 min. 5 "C TyrAIg of oxidized bovine insulin, and it has proved superior to the corresponding derivative of N-methylimidazole for this purpose. It has also been found effective for the determination of Tyr in calcitonins. The dansyl reagent has been used for the modification of two Tyr residues in prolactin, and four Tyr residues in bovine serum albumin. B.Amino-acid and Protein Chemistry.-DMAP and some of its salts have found many uses in amino-acid and protein chemistry, apart from those mentioned in Section S.A(iii). (i) Dakin-West Reaction. Reaction of an a-amino-acid with acetic anhydride in pyridine to form a-acetylamino-ketones is known as the Dakin-West reaction.*' The original authors suggestedg0 that the transformation involves an intermediate oxazolone (37), and this was confirmed subsequently by kinetic and mechanistic studies.g1 One of the first reports about 4-dialkylaminopyridine catalysis showed that the Dakin-West reaction, which usually requires vigorous conditions (e.g. 100--130"C, 15h), may be carried out at room temperature, and in better yield, when DMAP is added (Scheme 30). DMAP catalyses both the 4-acetylation of oxazolone (37), and the decarboxylative ring fission.COMe MeCHCOMe HOAc c---NHCO 2CH2Ph -CO2 NQ OCH7 Ph (86%) Reagents: i, Ac20, DMAP, TEA, 25 "C Scheme 30 89 H. D. Dakin and R. West, J. Biol. Chem., 1928, 78, 91. 90 H. D. Dakin and R. West, J. Biol. Chem., 1928, 78, 745. 91 N. L. Allinger, G. L. Wang, and B. B. Dewhurst, J. Org. Chem., 1974, 39, 1730. 4-Dialkylaminopyridines:Super Acylation and Alkylation Catalysts There are many examples of the application of the modified Dakk-West procedure, and several of these (e.g. determination of C-terminal amino-acids in peptides, formation of trifluoroalanine, and conversion of glutamic acid into a pyrrolidone) have been discussed previ~usly.~ More recently, it has been employed for the conversion of N-acetylamino-acids into enamides (Scheme 3 1),92 and in the preparationg3 of a series of novel acylamino-a-keto-esters which are valuable precursors for the synthesis of imidazo-triazinones.NOH PhCH2 -CHCO2H PhCH2CHCOMe ii IIPhCH2CH -CMe -I NHCOMe NHCOMe I NHCOMe iii1 fs.H N-OTS I r'll PhCH2-C-CMe (81%) [E: 2,73 : 27) Reagents: i, Ac20, DMAP, TEA, 20-40 "C; ii, NH,OH, NaOAc, H20,2h, 60-80 "C; iii, TsCl, TEA, MeCN, 3h, -10 to 20°C Scheme 31 (ii) Depsipeptide Synthesis. The DCC-DMAP method is effective for the coupling of N-t-butoxycarbonylamino-acidsand carboxy-protected hydroxy- acids (equation 17).94 MeCHCOzH MeCHC02CH2Ph .MeCHC02CHMe I +I LI INHBoc OH NH30c C02CH2Ph(1 mol) (1 mol) (98%) Reagents: i, DMAP (1 mol), DCC (1.1 rnol), CH2CI,, 25 "C (iii) Selective N-Protection. N-Trityl-4-dimethylaminopyridiniumchloride (38) has been used for the selective N-tritylation of serine methyl ester (equation 18).44 (iv) Reagents for Sulphydro-groups in Cysteine Residues. 1-Cyano-4-dimethylaminopyridinium perchlorate (12) reacts readily with thiols in neutral 92 U. Redeker, N. Engel, and W. Steglich, Tetrahedron Lett., 1981, 4263. 93 I. Charles, D. W. S. Latham, D. Hartley, A. W. Oxford, and D. I. C. Scopes, J. Chem. SOC., Perkin I. 1980, 1139. 94 C. Gilon, Y.Klausner, and A. Hassner, Tetrahedron Lett., 1979, 3811.Scriven HOCH2CHCO2 Me HOCH2CHC02 Me II + Nfi2L NHCPh3+NH3Cl-CI-CPh3 (60%) (38) Reagents: i, Pr'; EtN, DMF, 24h, 25 "C or acidic media. It has the advantage of being a water soluble reagent, thus avoiding the need for organic solvents that can denature proteins. The Cys residues numbers 7 and 19 of the reduced B-chain of bovine insulin are quantitatively cyanylated at pH 3.5 by this reagent.25 (v) Solid Phase Peptide Synthesis. Addition of DMAP improves the efficiency of the DCC mediated reaction for the anchoring of the first amino-acid residue onto hydroxymethylpolystyrene supports used for solid phase peptide synthesis.95-97 The coupling efficiencies of DCC, DCC-HOBT (l-hydroxy-benzotriazole), and DCC-DMAP were compared for automated solid phase synthesis on a hydroxymethyl resin (styrene-1 % DVB ~opolymer).~~The protected heptapeptide-resin samples so-formed were analysed for their amino- acid content (Table 2).The DCC-DMAP method was found to be clearly Table 2 Amino-acid composition of Boc-Ala-Cle-lle- Val- Pro-Arg( Tos)-Gly-OCH,-resin prepared by the methods indicated Amino-acid DCC" DCC-HOBT" DCC- D M AP" GlY 1.oo 1.oo 1.oo Arg 0.8 1 0.94 0.96 Pro 1.02 0.92 1.34 Val 0.88 0.74 1.01 Heb 0.83 0.73 0.99 Cle' 0.65 0.65 0.84 Ala 0.89 0.64 1.01 (a) Hydrolysis was carried out in 12N-propionic acid (l:l), 130°C, 6h. (6) Includes Dalloisoleucine produced during acid hydrolysis. (c) Cycloleucine. 95 B. W.Erickson and R. B. Merrifield, in 'The Proteins', ed. H.Neurath and R. L. Hill, 3rd Edn., Academic, N.Y.,1976, Vol. 2, p. 255. 96 S. S. Wang, C. C. Yang, I. D. Kulesha, M. Sonenberg, and R. B. Merrifield, Int. J. Peptide Protein Res., 1974, 6, 103. 97 G. Barany and R. B. Merrifield, in 'The Peptides: Analysis, Synthesis, Biology', Academic, N.Y., 1980, Vol. 2. 98 S. S. Wang, J. P. Tam, B. S. H. Wan& and R. B. Merrifield, Int. J. Peptide Protein Rex, 1981, 18, 459. 4-Dialkylaminopyridines:Super Acylation and Alkylation Catalysts the best, especially for the coupling of sterically hindered amino-acids in the Cle-Ile-Val region. Furthermore, h.p.1.c. analysis of the crude protected hepta-peptide amide Boc-Ala-Cle-Ile-Val-Pro-Arg(Tos)-Gly-NH,prepared by the DCC-DMAP method showed it to be purer than those prepared by the other methods.Racemization has been observed during DCC-DMAP catalysed coupling of Boc-PheOH with HGlu(OBzl)OCH2-resin.g* Other instances of racemization have been reported also in depsipeptide and solid phase peptide synthesis.99-loo In the latter case, racemization of urethane-protected amino-acids was attributed to 2-alkoxyoxazole formation [Section 5.B(i)].'00 This difficulty with the DCC-DMAP method may be overcomelor in the above example by using 2-nitrophenylsulphenyl (Nps) as the protecting group. Thus, Nps-Phe-OH can be converted into Nps-Phe-OMe in 94% yield on treatment with MeOH-DCC-DMAP (5 min at 0 "C, and 3h at 25 "C). C. Nucleotides and Nucleosidesu-Phosphonylation is a most important reaction in nucleic acid chemistry as it provides the main method for formation of internucleotide bonds in oligonucleotide synthesis.A major problem is side reactions of the moiety with a condensing agent at the 0-6 0x0-group. A new method has appeared that involves protecting 0-6 by reaction with di-n-butylthioxophosphoranyl bromide and DMAP which is virtually quantitative, and regiospecific (Scheme 32).'02 The condensing agent may then be attached to the 3'-hydroxy-group in 94% yield, compared with 15% when guanosine is unprotected at 0-6. Dialkyl and diary1 phosphoryl halides, phosphinothioyl halides, arenesulphonyl chlorides, and trialkylsilyl chlorides have been introduced successfully at 0-6 with DMAP cata1y~is.l~~The 5'-hydroxy-group of thymidine may be phosphorylated easily in DMAP-~yridine."~DMAP has found use recently to assist in the attachment of a tri-isopropylbenzene sulphonyl group 07CR C1 Mf 0Meo;x.MeMecJ(g---oI CH2 CH~HMe Me0 OH CHOH CHOH (39) R = H or C,-,, CHOH CH20H (40) 99 N. L. Benoiton and F. M.F. Chen, J. Chem. SOC.,Chem. Commun., 1981, 1225. loo E. Atherton, N. L. Benoiton, E. Brown, R. C. Sheppard, and B. J. Williams, J. Chem. SOC., Chem. Commun., 1981, 336. lo' B. Neises, T. Andries, and W. Steglich, J. Chem. SOC.,Chem. Commun., 1982, 1132. M.Sekine, J-i. Matsuzaki, M.Satoh, and T. Hata, J. Org. Chem., 1982, 47, 571. lo3 H. P. Daskalov, M.Sekine, and T. Hata, Tetrahedron Lett., 1980, 3899.lo4 H. A. Kellner, R. G. K. Schneiderwind, H. Eckert, and I. K. Ugi, Angew. Chem., Int. Ed. Engl., 1981, 20, 577. Scriven S 0 (96%) S O=PSPh I OCHZCCI, (94%) Reagents: i, Bu;P(S)Br, TEA, DMAP, CH2Cl,, 4h, rl.; 0 u ii, PhSPOCH2CC13/TPS, pyridine. 7h; I O-C~HI~~H~ MMTrCl = monomethoxytrityl chloride Scheme 32 at 0-6'" a benzoyl group at 2'-hydroxy, '06 and to facilitate 'transient protection',''' and semi-automated oligonucleotide synthesis.'08 Glycosylthio- carboxamides are versatile intermediates for the synthesis of C-nucleosides and they can be prepared in high yield from a ribofuranosyl cyanide on reaction with hydrogen sulphide and DMAP in a pressure vessel (equation 19).lo9 B. L. Gaffney and R.A. Jones, Tetrahedron Lett., 1982. 2257. lo6 E. Ohtsuka, H. Morisawa, and M. Ikehara, Chem. Pharm. Bull., 1982, 30,874. lo' G. S.Ti, B.L. Gaffney, and R.A. Jones, J. Am. Chem. SOC.,1982, 104, 1316. lo' E. Ohtsuka, H. Takashima, and M. Ikehara, Tetrahedron Lett., 1982, 3081. log M. V. Pickering, P. C. Srivastava, J. T. Witkowski, and R. K. Robins, Nucleic Acid Chem., 1978, 145. 4-Dialkylaminopyridines :Super Acylation and Alkylation Catalysts 5 TlOCH 2 TlOCH2 11 H2S(liq.) DMAP, 20h, bomb 0T1 0t1v * PNH2 (85%) The cyano-sugar is recovered unchanged when this procedure is carried out in the absence of DMAP. D. Lipids.-DMAP is a valuable catalyst for the derivatization of glycosphino-golipids' lo (per-o-benzoyl) and diglycerides' ' (p-nitrobenzoates) prior to analysis by h.p.1.c.Some carbamoyl analogues of phosphatidylcholines, which do not undergo degradation with phospholipase A,, have been prepared by a DMAP catalysed reaction (equation 20).", CHZOH I CHOH 0 I II t CH20 -P -O(CH2)2NMe3I0-Me(CH2)14NCO DMAP CHC13I I II t CH20 -P -O(CH&NMe3I 0-(5 5 -70%) 'lo S. K. Gross and R. H. McCluer, Anal. Biochem., 1980, 102,429. l'' M. Batley, N. H. Packer, and J. W. Redmond, J. Chromarogr., 1980, 19s, 520.'" C. M. Gupta and A. Bali, Biochim. Biophys. Acta, 1981,663, 506. Scriven 6 Industrial Applications A. Organic Chemistry.-Many patent claims have appeared in recent years that exploit the catalytic r61e of DMAP.These applications are based on principles discussed already, therefore they will be outlined briefly. Maytansinoids of the type (39), which have antimitotic, antitumor, antifungal and protozoacidal activity, may be prepared by the DMAP-DCC method.'I3 Riboflavin forms a tetranicotinate (40) with nicotinic anhydride in DMAP-HMPT.'I4 DMAP has been employed to catalyse the formation of arylacetamido-'I5 and sulphamido-"6 derivatives of penicillanic acids, and phosphorothioates and phenyl phosphono- thioates" of pyrimidones that have insecticidal and acaricidal activity. B. Polymer Chemistry.-DMAP catalyses the formation of polyurethanes,' l8 polyurethane coatings and foams,' l9 and isocyanate oligomers having intra- molecular uretidinedione rings.'O All of these reactions probably involve general base catalysis by DMAP (see 3.B). DMAP has been used for the incorporation of an aqueous dispersion of titanium dioxide and bis-tertiary amines into poly(hexamethy1eneadipamide) for use as delustering agents."' Other uses include catalysis of the reaction of poly(ethyleneterephtha1ate)with an epoxy- compound, e.g. N-(2,3-epoxypropyl)benzamideto give a heat-resistant polymer that is suitable for the manufacture of tyre cords."' Processes for the preparation of polyhydric alcohols'23 and poly~arbonates'~~have been reported. 1-Sulphonyl-4-dimethylaminopyridiniumsalts function as quick-acting hardeners for protein-containing layers.125 DMAP has been found effective as an accelerator for the curing of composite restorative resins in dental practice.26 Low molecular weight polystyrene has been grafted onto cellulose acetate in a N. Hashimoto and T. Kishi, Ger. Offen., 2911 248 (1979), (to Takeda Ind. Ltd.); Chem. Abstr., 1980, 92,94449. 114 Y. Kuroyanagi, M. Ban, and K. Suzuki, Japan Kokai, 77, 62297 (to Sanwa Chem. Lab.); Chem. Abstr.. 1977, 87, 184898."'Koninklijke Nederlandsche Gisten Spiritusfabriek N.V., Ger. Offen., 2 155 152; Chem. Abstr., 1972, 77, 88491. 'I6 P. W. Henniger and J. K. van der Drift, Ger. Offen., 2235390; Chem. Abstr., 1973, 78, 124608. H. H. Freedman, S. D. McGregor, M. Yoshimine, and L. M. Kroposki, Belg. P., 832047 (to Dow Chem. Co.); Chem. Abstr., 1977, 86, 43811. H. J. Twitchett, B.P., 990637 (1965); Chem.Abstr., 1965, 63, 766; J. H. Wild and F. E. G. Tate, U.S.P., 3 144452 (1964); Chem. Abstr., 1964, 61, 16068. H. J. Twitchett, B.P., 990635 (1965), Chem. Abstr., 1965, 63, 4483. 12'A. Nishikawa, H. Yokono, and J. Mukai, Japan Kokai, 76262 (1976); Chem. Abstr., 1977, 87, 135 294. V. Mathews, B.P., 1208 691 (1970); Chem. Abstr., 1971,74, 14 116. 122 S. D. Lazarus and K. Chakravarti, U.S.P., 4 130541 (1978); Chem. Abstr., 1979,90, 138985. 123 L. Kaplan, Ger. Offen., 2643913; Chem. Abstr., 1978,89, 214898. D. B. G. Jaquiss, V. Mark, and L. C. Mitchell, U.S.P., 4286085. 12' P. Bergthaller, W. Himmelmann, and L. Rosenhahn, Ger. Offen., 2 547 589 (1977); Chem. Abstr., 1977,87, 76374. G. M. Brauer, D. M. Dulik, J. M. Antonucci, D. J. Termini, and H.Argentar, J. Dent. Res., 1979.9. 1994. 4-Dialkylaminopyridines :Super Acylation and Alkylation Catalysts homogeneous solution with the aid of DMAP.'27 Grafting yields of 83% have been obtained using DMAP which compares with 63% (pyridine as catalyst) and 1% (no catalyst). DMAP is a better acetylation catalyst than N-methylimi- dazole for the determination of the hydroxyl number of polyether alcohols, particularly for those containing secondary hydroxy-groups. 28 Transesterifica-tions of aromatic carbonates, that are of significance in polymer chemistry, are catalysed by DMAP. The effect is most spectacular in the case of the reaction of phenol with o-nitrophenyl carbonate which gives a quantitative yield of products with added auxiliary base (Scheme 33).'*' A specially DMAP PhOH J7 -(41) Scheme 33 stabilized intermediate (41) was proposed to explain this effect, as the p-nitrophenyl isomer does not react as easily under similar conditions.7 Other Dialkylaminopyridine Catalysts A. Polymeric Catalysts.-4-Dialkylaminopyridine carboxylic acids of the type (42) have been attached to poly(ethy1eneimines) to form modified polymers (43) that catalyse the hydrolysis of nitrophenyl esters. 130-131 A new polymer-supported acylation catalyst has been reported recently.' 32 Polymer bound 4-(N-benzyl-N-methy1amino)pyridinecan be prepared by polymerization (Scheme 34). This polymer has been used sucessfully to catalyse the acetylation of linalool. It may be recovered by filtration and regenerated by washing with sodium hydroxide, acetone, and vacuum drying.127 P. MBnsson and L. Westfelt, J. Polym. Sci., Pofym. Chem. Ed., 1981, 19, 1509. lz8 R. Gnauck and R. Algeier, Plaste Kautsch., 1982, 29, 274; Chem. Abstr., 1982, 97,128 182. 129 D. J. Brunelle, Tetrahedron Lett., 1982, 1739. M. A. Hierl, E. P. Gamson, and I. M. Klotz, J. Am. Chem. SOC.,1979, 101, 6020. E. J. Delaney, L. E. Wood, and I. M. Klotz, J. Am. Chem. SOC.,1982, 104, 799. 13' M. Tomoi, Y. Akada, and H. Kakiuchi, Makromoi. Chem., Rapid Commun., 1982, 3, 537. Scriven INJC02H b MeNNa DM t:+ 0__3 CH2Cl i, Suspension copolymerization Scheme 34 B. Crown-ether Catalysts.-Analogues of DMAP [e.g. (44)] have been prepared with crown ethers incorporated at the 4-po~ition.'~~ Complexes of these with various metals (e.g.sodium, potassium, lithium, and rubidium) are better catalysts than the ligand alone for transacylation (Scheme 35). t t CO,HI Scheme 35 133 J. P. Dix, A. Wittenbrink-Dix, and F. Vogtle, Naturwissenschafen, 1980, 67, 91.

ISSN:0306-0012    

DOI:10.1039/CS9831200129

出版商:RSC    

年代:1983

数据来源: RSC

摘要:

Current Aspects of Unimolecular Reactions By K. A. Holbrook DEPARTMENT OF CHEMISTRY, THE UNIVERSITY OF HULL, HULL, HU6 7RX 1 Introduction Although a unimolecular reaction is in principle the simplest kind of elementary reaction, work on such reactions both experimental and theoretical, continues to expand at an ever-increasing rate. The RRKM theory, which was the main subject of two books on unimolecular reactions’*2 published ten years ago, continues to dominate the field largely because of the relative ease of application compared with other theories, although doubts about its applicability in certain cases have been e~pressed.~ Modifications to RRKM theory have also proved to be necessary among other things to interpret experiments using crossed molecular beams where dynamical features need to be considered, and to incorporate tunnelling effect^.^ Reviews covering both the experimenta15v6 and aspects of unimolecular reactions have appeared during the last ten years as well as reviews on unimolecular ion decompositions”.and laser-induced unimolecular reactions.’**l3 The present review is concerned more with experiment than with theory and in keeping with the declared policy of Chemical Society Reviews is intended for the non-specialist as well as the specialist in this field. The coverage is selective and examples are chosen to illustrate particular aspects of current research. This inevitably reflects the author’s personal interests and for this reason the coverage of thermally induced unimolecular reactions in their high-pressure regions is fairly comprehensive and includes most of the papers P.J. Robinson and K. A. Holbrook, ‘Unimolecular Reactions’, Wiley, N. York, 1972.’W. Forst, ‘Theory of Unimolecular Reactions’, Academic Press, N. York, 1973. See for example, D. B. Olson and W. C. Gardiner, Jr., J. Phys. Chem., 1979,83, 922. (a) L. Holmlid and K. Rynefors, Chem. Phys., 1981, 60, 393 and refs. cited therein. (b) W. H. Miller, J. Am. Chem. SOC., 1979, 101, 6810. P. J. Robinson in ‘Reaction Kinetics’, A Specialist Periodical Report, ed. P. G. Ashmore, The Chemical Society, London, 1975, Vol. 1, p. 93. H.M.Ftey and R.Walsh in.‘Gas Kinetics and Energy Transfer’, A Specialist Periodical Report, ed. P. G. Ashmore and R.J. Donovan, The Chemical Society, London, 1978, Vol. 3, p. 1.’M. Quack and J. Troe in ‘Gas Kinetics and Energy Transfer’, A Specialist Periodical Report, ed. P. G. Ashmore and R. J. Donovan, The Chemical Society, London, 1977, Vol. 2, p. 175. J. Troe, Nova Acta Leopold., 1979, 49, 63. W. L. Hase in ‘Potential Energy Surfaces and Dynamics Calculations’, ed. D. G. Truhlar, Plenum, N. York, 1981, p. 1. lo A. G. Brenton, R.P. Morgan, and J. H. Beynon, Ann. Rev. Phys. Chem., 1979,30, 51. J. L. Franklin in ‘Gas Phase Ion Chemistry’, ed. M.T. Bowers, Academic Press, N. York, 1979, Vol. 1, p. 273. N. Bloembergen and E. Yablonovitch in ‘Laser-induced Processes in Molecules’, ed. K. L. Kompa and S.D. Smith, Springer-Verlag, Berlin, 1979, p. 117.l3 R. G. Harrison and S. R.Butcher, Cont. Phys., 1980, 21, 19. Current Aspects of Unirnolecular ReEations appearing between 1978 (covered by the review of Frey and Walsh') and about the middle of 1981. The format of this review follows that of the earlier reviews on this subject, which are cited above, and like those is restricted largely to reactions of neutral gas phase species. 2 Theoretical Aspects An energized molecule A* can decompose unimolecularly (reaction 1) or be de-energized by collision with another molecule M (where M may be a reactant molecule or molecule of an added unreactive gas) (reaction 2). A* + MLA (2)+ M Energized molecules A* may be produced thermally (with a Boltzmann distribution of energies) or by non-thermal methods such as chemical activation, photoactivation, and more recently multiphoton (laser) excitation. For a thermally energized reaction the overall unimolecular rate constant kuni is given by the expression where Eo is the critical energy for the reaction, f(E)dE is the thermal Boltzmann distribution function given by N(E)e-EiRT d E f(E)d E =so" N(E)ePEPTdE (4) and the microscopic rate constant k(E) at energy E may be calculated from the RRKM expression in which L" is reaction path degeneracy, ZP(EL) is the sum of the vibrational- rotational states of the activated complex, N(E) is the density of such states for the energized molecule, and F is a correction factor for adiabatic rotation^.'^ Computer programs exist which permit the evaluation of the energy quantities 2=P(E,+,)and N(E) provided that a prior assignment of vibration frequencies to the activated complex and the energized molecule can be made.15 In a chemically activated system, an energized molecule is produced with l4 See ref.1, Chapter 4. l5 See ref. 1, Chapter 6 for details, also W. L. Hase and D. L. Bunker, Quantum Chemistry Program Exchange Cat. No. QCPE-234; S. E. Stein and B. S. Rabinovitch, .I.Chem. Phys., 1973, 58, 2438; S. E. Stein and B. S. Rabinovitch, Chem. Phys. Lett., 1977, 49, 183. Holbrook sufficient energy to enable it to undergo subsequent unimolecular isomerization or decomposition, for example by an exothermic recombination of two radicals Decomposition (or Isomerization)-9 R + R'eA* \ Stabilization where o is the gas-kinetic collision frequency.RRKM treatment of the reaction scheme above, including the postulation of an activated complex for the process of producing A*, leads to an expression permitting the calculation of the energy distribution function f{E)dE. The results show that chemically activated molecules are produced with a very much narrower spread of energies than those produced by thermal energization. l6 Recently, many unimolecular reactions have been studied in which the energized reactant molecules have been produced by multiphoton absorption of infrared laser radiation.17 There is at present a lack of agreement about the form of the distribution function applicable to the molecules energized in this way.Thus, Jenson, Steinfeld, and Levine" have concluded that a model based on maximal entropy can produce either a Poisson or a Boltzmann distribution of vibrational energy following multiple photon excitation and Colussi, Benson, Hwang, and Tiee have found a very narrow energy distribution in the multiphoton dissociation of CH2DCHzC1." As an alternative to the RRKM formulation of k(E) in equation (5), Forst2' showed how this rate constant could be recovered by the process of deconvolution from the high pressure rate constant k,. This leads to equation (7), where A, and E, are the high pressure Arrhenius parameters. This equation has proved to be extremely useful in enabling the fall-off curves to be predicted without any assumed model of the activated complex and Forst has recently2' listed a number of papers including applications of his method.The method is only strictly valid in the upper regions of the fall-off curve and requires accurate calculation of the energy densities involved. In its original form it also assumes that A, and E, are independent of temperature, which in principle they are not, but in practice the deviation may not be important. Yau and Pritchard22 have tested the Forst procedure by comparing the predicted l6 See ref. 1, p. 274. l7 R. V. Ambartzumian and V. S. Letokhov 'Chemical and Biochemical Applications of Lasers', ed. C. B. Moore, Academic Press, N. York, 1977, vol. 3. C. C. Jenson, J. I. Steinfeld, and R. D. Levine, J. Chem. Phys., 1978, 69, 1432.l9 A. J. Colussi, S.W. Benson, R. J. Hwang,and J. J. Tiee, Chem. Phys. Lett., 1977, 52, 349. 2o W. Forst, J. Phys. Chem.,1972, 76, 342. W. Forst, J. Phys. Chem., 1979, 83, 100. 22 A. W. Yau and H. 0.Pritchard, Can. J. Chem., 1979, 57, 2458. Current Aspects of Unimolecular Reactions reaction rates with theoretical calculations for the reactions N,O -+ N2 + 0 and COz+CO + 0. For these reactions, which adhere well to the Arrhenius law, the approximations in the Forst procedure were considered to be reasonable. The situation may be different at high temperatures such as those involved in shock tube experiments. Forst” has taken some data of T~ang~~ for the decomposition of n-butane into two ethyl radicals and fitted them to the expression k, = A’,(kT)”e-EA/kT (8) In a more recent paper, Forst and T~rre11~~ have fitted the same data to the expression k = A’ eBkTe-E‘m/kT (9)m03 where B is a constant, found to be negative. The exponential expression leads to a larger temperature dependence of A& and EL and calculations for a temperature of 2500 K (actually well above the range of Tsang’s experiments) produces a k, value smaller by a factor of 4.5 than that from the normal Arrhenius equation.The implications for the fall-off curves, i.e. log (kuni/km) versus log pressure, deduced from such temperature-dependent parameters are discussed. Use of expression (9) produces more curvature in the fall-off plot and a bigger decline of activation energy with pressure.Forst has drawn attention to the need for consideration of these points in extrapolating values of high pressure Arrhenius parameters from measurements made in the fall-off region. Many such examples are quoted later in this review. Much recent work on unimolecular reactions has involved the formulation of the processes involved in terms of the so-called master equation. This involves writing the steady state concentration of molecules in the ithenergy level (n,)as the equation _-dni -Ri + CjzjPijnj-CjziPjini-kini = 0 (10)dt where the first two terms on the right-hand side refer to the rate of non-collisional input and collisional input respectively into level i, and the last two terms represent collisional transfer from level i to level j and unimolecular reaction with the microscopic rate constant ki. The complete master equation represents a series of equations of type (10) one for each energy level i which must hold simultaneously subject to the overall conditions c.p..= 1 (11)J 11 and 23 W. Tsang, Int. J. Chern. Kinet., 1978, 10, 821. 24 W. Forst and S. Turrell, Int. J. Chem. Kinet., 1981, 13, 283. Holbrook The quantity Pji represents the collisional transfer probability from level i into levelj and hence equation (11) represents the condition that all transitions from level i must end in some other level j and equation (12) represents the condition for equilibrium between a given pair of levels i and j for which g represents degeneracy and E energy of the appropriate levels.In principle, the complete master equation can be solved by matrix methods and the steady state populations nl ,n2,.. . ,n, and the rate of reaction can be calculated. In practice, to simplify the solution of the large matrices involved, the energy levels are ‘grained’ into blocks and are often solved by iterative methods for different models of the transition probabilities. This approach has proved to be necessary in particular to take into account ‘weak collisions’. Such collisions may be defined as those in which a colliding partner removes an amount of energy less than k T, and these are particularly important for highly energized molecules produced by non-thermal energtzation methods where many collisions.may be needed to reduce the energy of the molecule to a value below the critical energy E,. For further details the reader is referred to reference 1 (Chapter 10 and the references cited therein). In addition to the necessity of formulating the master equation in order to consider weak collisions, there are many other applications where this approach is necessary, particularly for non-thermal energization processes. Recent treatments of multiphoton absorption processes are for example similarly formulated in terms of rate equations which in this case incorporate absorption and stimulated emission processes between the various level^.^ -” Yau and Pritchard have now provided some analytical solutions to their master equation version of unimolecular reaction for particular transition probability modelsJo and Singh and Pritchard3 have recently derived an improved equation for the unimolecular rate constant which should prove easier to use than the original version.Troe has extended his reformulation of the statistical adiabatic channel model of unimolecular reactions to cover thermal reactions at high This simplified version of the earlier theory presented by Quack and Troe33-35 is based on a two-parameter characterization of the potential energy surface, and calculations for the high-pressure reverse-recombination rate constants are sbown to be in reasonable agreement both with experiment and with calculations based on the exact theory. To facilitate further application of the theory, full details of the calculations for the CzH6e2CH3 system are given.25 E. R. Grant, P. A. Schulz, Aa. S. Sudbo, and Y. T. Lee, Phys. Rev. Lett., 1978, 40,115 26 M. Quack, J. Chem. Phys., 1978,69, 1282. ”M. Quack, Ber. Bunsenges. Phys. Chem., 1979,83, 757. A. W. Yau and H. 0.Pritchard, Can. J. Chem., 1978,s. 1389. 29 A. W. Yau and H. 0.Pritchard, Can. J. Chem., 1980,!58, 626. 30 A. W. Yau and H. 0.Pritchard, Can. J. Chem., 1979, 57, 1723. ”S. R. Singh and H. 0.Pritchard, Chem. Phys. Lett., 1980, 73, 191. 32 J. Troe, J. Chem. Phys., 1981,75, 226. 33 M. Quack and J. Troe, Ber. Bunsenges. Phys. Chem., 1974, 78, 240. 34 M. Quack and J. Troe, Ber. Bunsenges. Phys. Chem., 1975, 79, 170. 35 M. Quack and J. Troe, Ber. Bunsenges. Phys. Chem., 1975, 79, 469.Current Aspects of Unimolecular Reactions This theory is particularly relevant to bond-fission reactions and the reverse radical recombinations. Tr~e~~has also presented a simplified treatment of reactions in the fall-off region between high and low pressure limits in terms of reduced fall-off curves of k/k, versus ko/k, where k is the general-pressure first-order rate constant and ko and k, are its limiting forms at low and high pressures. According to Lindemann-Hinshelwood theory, k/km is a function of ko/k,, i.e. Fall-off curves predicted by the Kassel theory are broader (i.e. show more gradual change of curvature) than those from the simple expression above and can be represented by introduction of a broadening factor assuming strong collisions, Fsc(ko/k,) (see Figure 5 of reference 36).A further broadening is caused by weak collisions represented by the factor Fwc(ko/k,). Hence Troe has given expressions for Fsc(ko/k,) and Fwc(ko/k,) which can be derived from experimentally accessible parameters such as the high and low pressure activation energies and the collision efficiency /Ic(see Section 4). A further correction can be made in order to represent the shape expected from a full RRKM fall-off calculation. Illustrative examples which show the ease with which such calculations can be done are given in reference 36. This method has also been used in a recent compilation of rate constants for combination/ dissociation proce~ses.~' Theoretical calculations of potential energy surfaces, activation energy barriers, and transition state structures have continued to be of interest to quantum chemists.New techniques are being devised to locate the saddle point on the potential energy ~urface~'.~~ and some progress is claimed in the estimation of vibration frequencies from ab initio calculations of the activated complex in the isomerization of methyl i~ocyanide.~' calculation^^^ of a similar kind for the decomposition of formaldehyde coupled with experimental measurements of the rate of decomposition of photoexcited formaldehyde* have shown that decomposition occurs at energies about 20-40 kJ mol- below the most accurate ab initio calculation of the activation barrier, implying that the reaction occurs almost entirely by tunnelling.Tunnelling has also been studied theoretically in the isomerization of HNC to HCN.42 Kato and Morokuma have carried * HZCO + H2 + CO 36 J. Troe, J. Phys. Chem., 1979,83, 114. 3' D. L. Baulch, R. A. Cox, R. F. Hampson, Jr., J. A. Kerr, J. Troe, and R. T. Watson, J. Phys. Chem. Re$ Data, 1980, 9, 295. 38 P. Pulay in 'Application of Electronic Structure Theory' (Modern Theoretical Chemistry, Vol. 4, ed. H. F. Schaefer), Plenum, N. York, 1977, p. 153. 39 M.J. Rothman and L. Lohr, Jr., Chem. Phys. Lett., 1980, 70, 405. 40 P. Saxe, Y. Yarnaguchi, P. Pulay, and H. F. Schaefer 111, J. Am. Chem. SOC., 1980, 102, 3718. 41 S. K. Gray, W. H. Miller, Y. Yamaguchi, and H. F. Schaefer 111, J. Am. Chem. SOC., 1981,103.1900 42 S. K. tiray, W. H. Miller, Y. Yamaguchi, and H. F. Schaefer III, J. Chern. Phys., 1980, 73, 2733. Holbrook out a series of extensive studies using ab initio molecular orbital methods to determine the geometries of reactants, products, and transition states. Normal mode analysis of these, when allied to RRKM calculations, enabled predictions of rate constants and energy partitioning among the products. In this way they have examined the decompositions of ethyl fluoride,43 vinyl fluoride,44 and the fluoroethyl radical.45 Although it is clear that quantum mechanical calculations of the complete multidimensional potential energy surface for a polyatomic molecule and precise determinations of activation energies are still a long way off, the methods employed by these authors in concentrating upon particularly important regions of the potential energy surface are clearly yielding some valuable results.A recent selective review of ab initio methods of calculation of potential barriers in unimolecular rearrangements highlights some of the current problems.46 Recent work on the theory of unimolecular reactions can be broadly classified into attempts to modify the RRKM theory or to produce forms applicable to particular purposes, and attempts to devise new theories. Apart from those RRKM modifications already mentioned, Doll47 has described a method for making anharmonic corrections to RRKM theory using Monte Carlo techniques and has also4’ used Monte Carlo methods to calculate unimolecular rate constants by a method reminiscent of early Slater theory49 although without the restrictive conditions.Pritchard et al.” have interestingly gone back to a fore-runner of Slater theory, namely the dynamical theory of Polanyi and Wigner” and recast it in quantum terms. In this theory, reaction occurs when a critical amount of energy accumulates in a particular oscillator; if it does so, then decomposition will occur within the vibrational period of that oscillator. For either a harmonic or a Morse oscillator this is given by T = (2cvr)-(15) where c is the velocity of light and v, the wave number for the particular oscillator. The microscopic rate constant k(E) is then given by k(E) = 2cvra(E) (16) where a(E) is the fraction of molecular states decomposing within the period z.This fraction can be calculated statistically from the energy densities of the molecule and the ‘sub-molecule’ with oscillator r removed. This simple theory can be ’extended to s oscillators and Pritchard et al. have 43 S. Kato and K. Morokuma, J. Chem. Phys., 1980, 73, 3900. 44 S. Kato and K. Morokuma, J. Chem. Phys., 1981,74, 6285. 45 S. Kato and K. Morokuma, J. Chem. Phys., 1980, 72, 206. 46 C. E. Dykstra, Ann. Reo. Phys. Chem., 1981,32, 25. 47 J. D. Doll, Chem. Phys. Lett., 1980, 72, 139. 48 J. D. Doll, J. Chem. Phys., 1981, 74, 1074. 49 See ref. 1, Chapter 2. H. 0.Pritchard, G. M.Diker, and A. W. Yau,Can. J. Chem., 1980, 58, 1516.’’M. Polanyi and E. Wigner, 2.Phys. Chem., 1928, 139, 439.Current Aspects of Unimolecular Reactions shown that k(E) values thus calculated can agree well with those predicted by more elaborate calculations for the methyl isocyanide isomerization. Calculations using the present procedure are very sensitive to the choice of the reactive oscillator or oscillators and, as realized by the authors, this can be both an advantage and a disadvantage. Clearly more work on the factors governing this choice is necessary if the method is to have wider application. The quantum version of RRK theory has also received some attention lately, particularly in connection with the interpretation of multiphoton dissociation experiments where the experiments do not permit a distinction between RRK and RRKM prediction^.^^^^^ Other contributions to unimolecular theory have been.made by Lindenberg et aLS4and by Kay5’ and the absolute-rate theory (or transition-state theory) has also been re-assessed. 56 3 Intramolecular Energy Transfer A key assumption of statistical unimolecular theories such as the RRKM theory is that following excitation, rapid intramolecular vibrational and rotational energy redistribution occurs in a time which is short compared with the time for dissociation or isomerization. This assumption has been tested in the past by chemical activation experiments which produce initially a non-random distribution of energy. The subsequent rate of randomization (or relaxation) was then measured (in carefully designed experiments) relative to the rate of de-energization by collision.In such experiments for example, Rabinovitch and co-workers5’ -59 by assuming an exponential rate of energy randomization found a relaxation constant of about 10l2s-and it is now customary to assume that randomization normally occurs within 10-l1 to 10-l2 The question of whether intramolecular energization processes are rapid or not nevertheless continues to be raised and probed by a variety of experimental techniques. An excellent review of recent experimental work has been given by Oref and Rabinovitch6* who have concluded that little evidence for non-statistical behaviour exists, particularly for polyatomic species with moderate to high levels of excitation. A molecule which obeys the assumption of RRKM theory stated above is variously described as an RRKM-type molecule or a ‘statistical’ molecule and its behaviour loosely as RRKM-like, statistical, random, stochastic, or ergodic ’* E.Thiele, J. Stone, and M.F. Goodman, Chem. Phys. Lett., 1980, 76, 579. ”J. Stone, E. Thiele, M. F. Goodman, J. C. Stephenson, and D. S. King, J. Chem. Phys., 1980, 73, 2259. 54 V. Seshadri, B. J. West, and K. Lindenberg, J. Chem. Phys., 1980, 72, 1145. ”K. G. Kay, J. Chem. Phys., 1978,6!l, 434. 56 D. M. Golden, J. Phys. Chem., 1979, 83, 108. 57 J. D. Rynbrandt and B. S. Rabinovitch, J. Phys. Chem., 1971, 75, 2164. B. S. Rabinovitch, J. F. Meagher, K. J. Chao, and J. R. Barker, J. Chem. Phys., 1974, 60, 2932. 59 A. N. KO, B. S. Rabinovitch, and K.J. Chao, J. Chem. Phys., 1977, 66, 1374. 6o I. Oref and B. S. Rabinovitch, Acc. Chem. Res., 1979, 12, 166. Holbrook although the precise meanings of these terms differ. Hase and Bunker6’ originally defined non-RRKM-behaviour as either apparent or intrinsic. Apparent non-RRKM-behaviour can arise when the reaction time is particularly short, but such a molecule might well show RRKM-behaviour towards a different reaction channel if one were available. Intrinsic non-RRKM-behaviour on the other hand can never lead to RRKM-behaviour since there is an inherent weak coupling between the oscillators in the molecule (or, in phase-space parlance, a ‘bottleneck’ to the energy flow). Some of the early discussion on whether CH3NC is a non-RRKM molecule or not has now been resolved.60 Richardson and Simons62 have described experiments on energy randomization in a molecule they regard as ‘more typical’ than that studied by Rynbrandt and Rabinovitch.” Methylene (CH2’A was made to react with methyl [’H,]silane to give [ZHl]methyl methyl [2H,]silane (Scheme 1).Because of its mode of ~ CH3SiD2CH2D CH3 SiD2 CH D* DCH3 I\ ;‘\ HH /CH3‘si ---c 6CH2D2 +:Si D’ 4\H \D \ CH2D Scheme 1 formation, excess energy initially would be expected to be located in the CH,D part of the molecule. At high pressures, assuming strong collisions, excess energy would be removed before energy randomization could occur and any product would result from elimination of CH2D leading to CH2D2.Randomization of energy would also produce elimination of CH3 to give CH3D. The ratio of CH3D to CH,D, was found to be reasonably constant with pressures up to 4 atmospheres, corresponding to a lifetime of 1.7 x 10-seconds, thus indicating virtual complete randomization within this time. In contrast to this, work on another type of reaction, the 1-5 H-atom shift in the isomerization of cis-3-methyl-[ 1,2-2H2]penta-1,3-diene has provided some evidence for non-randomization with a relaxation constant of about 4 x lo-” seconds.63 This molecule was generated from the cross-combination of 61 W. L. Hase and D. L. Bunker, J. Chern. Phys., 1973, 59, 4621. 62 T. H. Richardson and J. W. Simons, Int. J. Chern. Kinet., 1978, 10, 1055. 63 T.Ibuki and S. Sugita, J. Chern. Phys., 1979, 70, 3989. Current Aspects of Unimoleeular Reactions CH3CH =CCH3 and CD-CHD radicals which produces the cis-isomer having about 512 kJ mol-' excitation energy. The rate constant for isomerization to was~is-3-methyl-[4,5-~H~]penta-1,3-diene measured relative to the rate of collisional stabilization at pressures up to 2006 Torr. At low pressures (-216 Torr) the rate constants agreed with RRKM calculations of k(E) based on known high pressure Arrhenius parameters, but at high pressures a fall in rate constant for isomerization was observed. It was argued that at these collision times the excitation energy cannot reach the isomerizing C-H bond before collisional stabilization occurs. Oref and Rabinovitch6' have pointed out that non-randomization is more likely to be observed at lower levels of excitation when the relaxation constant is lower and may no longer compete with the rate of unimolecular reaction. In this connection it is interesting to note the recent experiments of Bauer and True64 on a low energy process (E, 'v 42 kJ mol- I), the conformational syn anti isomerization of methyl nitrite.This process was followed by n.m.r. line-width measurements at pressures up to 100 Torr. Under these conditions the reaction is unimolecular in its second-order region as predicted by theory. The rate determining step is the rate of energization and experimental values for the rate constant of this process were compared with the predictions of RRKM and Slater theories.Agreement was obtained with the RRKM caicula- tions which could be taken to imply rapid randomization of energy even for molecules with low energy barriers and low densities of states. Some specific evidence for non-randomization of energy has come from experiments using non-thermal energization. Hamer and H~ber~~ for example have recently reported evidence that photoactivated cyclic aliphatic ketones such as cyclopentanone and cycloheptanone undergo dissociation from higher vibrational levels of the lowest triplet-state faster than relaxation occurs from this state. Reddy and Berry66 have photoactivated ally1 isocyanide at selective sites in the molecule using a CW dye laser producing CH stretch overtone excitations. Rate constants (determined from the slopes of Stern-Volmer plots) showed deviation from RRKM calculated values at the appropriate energy and depended on the mode of excitation.This last example is a rare example of 'mode-selective' behaviour which has been the motivation for much experimental work in the field of multiphoton infrared excitation. Other possible examples of mode-specific chemistry have been observed by Zare and co-workers6' and by Hall and Kaldor.68 The criteria governing mode specificity and the related problem of intra- molecular energy randomization have been clearly explained by Waite and Miller in a recent paper.69 In this paper they show that the question of intramolecular 64 S. H. Bauer and N. S. True, J. Phys.Chem., 1980,84, 2507. 65 E. Hamer and J. R. Huber, Chem. Phys. Lett., 1978, 55, 543. 66 K. V. Reddy and M. J. Berry, Chem. Phys. Lett., 1979, 66,223. 67 R. Naaman, D. M. Lubman, and R. N. Zare, J. Chem. Phys., 1979, 71,4192. 68 R. B. Hall and A. Kaldor, J. Chem. Phys., 1979,70,4027. 69 B. A. Waite and W. H. Miller, J. Chem. Phys., 1980, 73, 3713. Holbrook energy randomization as such, is not the prime factor of interest and state that ‘what matters for mode specificity is the rate of intramolecular energy transfer compared to the rate of the chemistry of interest, not the question of ergodicity in an infinite time limit’. In a subsequent paper7* the same authors have calculated energies and lifetimes by quantum mechanical methods for metastable states of a particular potential energy surface (the Henon-Heiles P.E.surface). This is a surface for which the classical dynamics is known to change from quasi-periodic at low energies to ergodic at high energies. Despite this it was found that rate constants for unimolecular decay were well represented by RRKM calculations over the entire energy range, confirming the view that mode specificity is not solely determined by the degree of ergodicity of the intra- molecular dynamical behaviour. Dynamical behaviour has, however, been the subject of much study of classical trajectories on various potential energy surfaces. For example, Wolf and Hase’l have used classical trajectories to investigate the internal dynamics of the model molecule H -C -C.Two different anharmonic potential energy surfaces were studied which exhibited intrinsic non-RRKM lifetimes and gave a large number of quasi-periodic trajectories at energies above the threshold for H-C-C +H + C-C. Trajectory studies have also been used to study energy partitioning, in this case the distribution of translational energy among the product fragments. Ha~e’~ found that for classical trajectory calculations on the molecule Cl-CZC-H it was possible with random or non-random excitation of either HCC or ClCC modes to produce random (i.e. RRKM) or non-RRKM lifetime distribution^.^^ Despite this, the average translational energies were the same, irrespective of the type of excitation, indicating that observation of statistical relative translational energy distributions does not necessarily imply complete intramolecular vibrational energy transfer.Relatively few quantum dynamical studies of energy transfer have been made compared with classical studies. One problem is the definition of ergodicity from a quantum standpoint, and this has received some attention in recent papers.74- 76 In conclusion, the generally confusing picture of the intramolecular energy- transfer process which has arisen from a variety of experimental techniques over the last few years is gradually being clarified. This is an area in which careful planning of experiments is particularly necessary in order to avoid ambiguity of interpretation. The role of collisions in assisting intramolecular energy randomization is still unclear and may be circumvented by excitation under suitable collision-free conditions, e.g.in laser-induced reactions at sufficiently 70 B. A. Waite and W. H. Miller, J. Chem. Phys., 1981, 74, 3910. R. J. Wolf and W. L. Hase, J. Chem. Phys., 1980, 73, 3779. 72 W. L. Hase, Chem. Phys. Lett., 1979, 67, 263. 73 For an excellent illustration of the difference, see the review by W. J. Chesnavitch and M. T. Bowers in ‘Gas Phase Ion Chemistry’, ed. M. T. Bowers, Academic Press, N. York, 1979, Vol. 1, p. 119. 74 E. J. Heller, J. Chem. Phys., 1980, 72, 1337. 7s D. W. Noid, M. L. Koszykowski, M. Tabor, and R. A. Marcus, J. Chem. Phys., 1980, 72, 6169. 76 D. W. Noid, M. L. Koszykowski, and R. A. Marcus, Ann.Rev. Phys. Chem., 1981,32, 267. Current Aspects of Unimolecular Reactions low pressures, or in molecular beam experiments. Energy distribution in the unimolecular decompositions of ions has been recently reviewed," as has the production of ions in selected internal energy states, by the method of photoion-photoelectron c~incidence,~~ and some non-statistical energy distribu- tions have been reported. It is to be hoped that as more data on measured relaxation constants are obtained, it will be possible to establish more clearly their dependence on the energy content and structure of the excited molecules as Oref and Rabinovitch6' have done with the limited data available at present. 4 Intermolecular Energy Transfer The rate-determining step in a unim~lecular reaction at low pressures with thermal energization is the second-order collisional process A + M+A* + M (-2) Direct measurements of the rate of reaction in this region enable the collisional efficiencies of various bath gas molecules M to be determined. The collisional efficiency per unit collision, /?,,can then be defined by the equation where ko is the limiting value of the first-order rate constant kuniat low pressures, and kOSCis the 'strong collision' value of this rate constant which is obtained from equation (3) by writing k,[M] 6k(E) and k,[M] = collision frequency w, hence kOSC= f(E)dE (18)E=Eo where f(E)dE is given as previously by equation (4).To avoid calculation of kOsCfrom equation (18), which requires knowledge of collision diameters and other molecular parameters, relative collisional efficiencies are often measured assuming that the reactant molecule A is a strong collider (/Ic= 1).For weak colliders Bc < 1 and from the values of /?, it is possible to obtain values of (dEd) i.e. the average energy transferred per deactivating collision. Tardy and Rabin~vitch~~have shown that a 'quasi-universal' relationship exists between /?, and (dE,)/(E+) where (E+) is the Boltzmann average energy of the reacting molecules. The exact relationship depends, according to this treatment, on the model assumed for the transition probabilities in the weak collision master equation, i.e.whether step-ladder, exponential, or some other distribution such as Gaussian or Poisson.In an alternative treatment of weak collisions, Tr~e'~**~developed an 77 T. Baer in ref. 73, p. 153. D. C. Tardy and B. S. Rabinovitch, J. Chem. Phys., 1968, 48, 1282.''J. Troe, J. Chem. Phys., 1977, 66,4745. J. Troe, J. Chem. Phys., 1977, 66,4758. Holbrook analytical solution of the master equation for an exponential model of collisional transition probabilities. This leads to equation (19). This involves a different energy quantity uiz. (LIE) which is defined as the average energy transferred per activating or deactivating collision, and FE is defined by the equation where f(E)dE is given by equation (4). The relationships between the energy quantities (A&) and (LIE) and their implications for the expected temperature-dependence of (LIE,), (LIE), and bc have now been discussed in several reviews6* 7.78 In general #Ic is found experimentally to have a slight inverse temperature dependence and (A&) and (LIE) also usually decline with increase in temperature.When studied over a sufficiently large temperature range, e.g. in shock-tube experiments combined with lower temperature thermal studies, the decline of bcwith temperature could in principle lead to a reduced observed activation energy at the higher temperatures. The results of some recent studies of intermolecular energy transfer in unimolecular reactions are collected in Table 1. Many of the data have been obtained from chemical activation work when /I,is calculated from the observed ratio of decomposition (D)and stabilization (S)products (equation 21).k, = B,O(WS) (21) From the chemical activation results in Table 1 it will be seen that there is general agreement with previous expectations5* 78 that LIE values vary from about 2 kJ mol-I for the smallest bath-gas molecules to about 4-60 kJ mol-’ for the large polyatomic molecules, and that the former are fitted better by an exponential and the latter by a stepladder distribution. There is less agreement at present about whether collisional energy transfer depends upon the molecular parameters of the reactant (substrate) molecules. For example, McCluskey and Carr’ have presented some tentative evidence, based on their work on chemically activated alkylcyclopropanes, which shows a possible decrease in (A&) with increasing size of the reactant molecule.On the other hand, Richmond and Setser,” in comparing recent results for chemically activated C2H5F and C2H,F2 with previous results on CH3CF3 have found that differences in the efficiencies of N2 and C02 as deactivators are not due to differences in internal properties of the excited molecules but are more likely due to differences in interaction potentials between these various molecules and the bath gas. This explanation is in accord with the idea of a collision complex A-M* proposed by Lin and Rabin~vitch.~~ R. J. McCluskey and R. W. Cam, Jr., J. Phys. Chem., 1978,82, 2637. 82 G. Richmond and D. W. Setsea, J. Phys. Chem., 1980,84,2699. Y.N. Lin and B. S.Rabinovitch, J. Phys. Chem., 1970, 74, 3151. Table 1 Studies of intermolecular energy transfer Energized species (E*)/lcJ mol-(E,)/lcJ mol-' Bath gas dE/kJ mol-' Bc TIK Comments and Ref: Chemical Activation C3H8 426.7 CH2N2-C2H6-SiMe4-02 20.9-83.7 Step-ladder model (S.L.) a C4H10 431.8 CH2N2-C3H8-SiMe4-02 20.9-83.7 S.L. a 441.4 248.2 Cy clobu tane 16.7 s.L. b Cy clobu tane 42.0 _+ 8.0 S.L. 4dMI MeCF, 426.8 He 4.2 (Exp) co2 8.4 (Exp) MeCl 16.8 (SL) 300 C C2F6 25.1 (SL) n-C8F18 42.0 (SL) 355.6 230.1 0.3 393 d 0.1 386-400 233 C3F8 30-60 S.L. ec121c1 Cyclobutane 41.8 He 2.1 N2 8.4 S.L. Pco2 20.9A 436.8 35.9 s.L. L M e 432.6 16.7 or 26.4 s.L. 438.5 But-1-ene 30.9 S.L. EtF 380.7 238.5 He 4.2 0.19 300 Exponential Model (EXP) N2 10.5 0.55 s.L.9 co2 8.4 0.47 s.L. SF, 21.0 1.o S.L. 1,2-C2H,F2 387.0 259.4 He 4.2 0.19 EXP N2 10.5 0.85 S.L. 9 co2 10.5 0.58 S.L. SF6 21.0 1.o S.L. Vinyl radical 167.4 He 0.078 Modified RRKM including tunnelling h 0.21 RRKM Thermal Energization in Shock Waves Ar 0.05 1038-1208 1 He 0.04 CD, Ar 0.25 500 S.L. 0.01 2500 MeCl Ar -0.70 0.02 1800-2 100 k Me1 Ar 0.29 0.02 1050-1500 1 NO2 Ar 3.0 0.12 1800 ClNO Ar 1.2 0.09 lo00 m Ar 0.16 0.019 80003 N2O Ar 0.51 0.022 2000 Thermal Energization at Low Pressures He 4.6 0.075 770 nn 2.2 0.01 973 He 2.27 & 0.59 0.01 975 (Measured at k/k, = 0.015) (EXPIHe 1.92 & 0.59 0.006 1175 (EXP) H2 1.79 & 0.72 0.009 975 EXP 1.07 0.002 1175 EXP 0 N2 5.39 k1.79 0.02 975 S.L.5.39 k1.79 EXP 4.19 k1.79 0.013 1175 S.L. Table 1-continued Energized species (E*)/kJ mol-' (E,)/kJ mol-' Bath gas dE/kJ mol-' Bc T/K Comments and Re$ 13.2 1.92 0.09 975 S.L. 5.39 & 1.79 0.022 1175 S.L. Wall-Energization 47.9 0.65 775 r 40.7 0.50 975 14.1 0.35 1175 Cycloheptatriene (35.9) (0.5) 800 53.8 0.4 1100 59.9 0.8 930 41.9 0.4 1150dC' a R. J. Wolf and W. L. Hase, J. Phys. Chem., 1978, 82, 1850. T. H. Richardson and J. W. Simons, J. Am. Chem. SOC., 1978, 100, 1062. P. J. Marcoux and D. W. Setser, J. Phys. Chem., 1978, 82, 97. Representative values only given. See ref. for a comprehensive list of 17 bath gases at 300 K and 5 bath gases at 195 K. G.0. Pritchard and J. P. Gute, Int. J. Chem. Kinet., 1978, 10, 759. values relative to #?-fluorocyclohexane = 1. Ref. 106. The parameters quoted relate to the major channel leading to 2,3-dichloropropene. A concurrent process produces 1,l-dichloropropene (-5% total products). Ref. 81. Ethyl cyclopropane values uncertain due to lack of precise thermal Arrhenius parameters. Representative values of the data given. Ref. 82. K. Kowari, K. Sugawara, S. Sato, and S. Nagase, Bull. Chem. SOC. Jpn., 1981, 54, 1222. D. K. Lewis, S. E. Giesler, and M. S. Brown, Int. J. Chem. Kinet., 1978, 10, 277. j C. C. Chiang, J. A. Barker, and G. B. Skinner, J. Phys. Chem., 1980, 84, 939. Ir 0. Kondo, K. Saito, and I. Murakami: Bull. Chem. SOC.Jpn., 1980, 53, 2133. K. Saito, H.Tahara, 0. Kondo, T. Yokubo, T. Higashihara, and I. Murakami, Bull. Chem. SOC.Jpn., 1980,53, 1335. Ref. 86. See this reference for fl, and AE values also for nine other bath gases. " Ref. 84. Values are relative to fl, (A) = 1.0 AE >42 kJ mol- (770 K) and fl,(A) = 0.5 AE >21.8 kJmol-' (973 K). 'Ref. 85. M. B. Callahan and L. D. SDicer. J. Phys. Chem., 1979, 83. 1013. W. S. Kolln, M. Johnson, D. E. Peebles, and J. W. Simons, Chem. Phys. Lett., 1979, 65, 85. 'Ref. 100. Data for Adz were recalculated from Ref. 88. Data for cycloheptatriene based on earlier work of D. C. Astholtz, J. Troe, and W. Wieters. J. Chem. Phys., 1979, 70, 5107. Values of BWand AE for this compound at the lower temperature are uncertain. Holbrook Evidence for the decline of fl, and (LIE,) with temperature has come from work on thermally energized cycl~propane.~~~~~ In one of these studiesB5 the new technique known as the diffusion cloud method in which the reactant diffuses from a point source at low pressure into a flowing gas stream was used.The concentration of reactant at some point distant from the source depends on the diffusion constant, the velocity profile, and the unimolecular rate constant. Although absolute rate constant measurements would be difficult by this technique, the ‘calibration’ of many unknown parameters by the use of a reactant such as cyclopropane with a well known thermal isomerization rate constant makes it possible to obtain relative collisional efficiencies for different flow gases.The method is particularly valuable in yielding data for these at temperatures close to those attainable in shock tubes. Shock tube studies often involve small molecules since large polyatomics are too easily decomposed at the high temperatures involved. Care is needed in comparing the values of (LIE) and pc for small molecules with those for large polyatomic molecules although the same temperature dependence can often be observed. Endo, Glanzer, and TroeS6 have drawn attention to the particularly low value of (LIE) observed for the O3 molecule which could be related to its low density of vibrational states. Stace and MurrellB7 have confirmed the temperature dependence of flc for the ozone molecule by some classical trajectory studies of O3 collisions with He, Ar, and Xe in the range 300-2500 K. Calculations on a model system approximating to the non steady-state conditions applicable to cyclopropane isomerization in a shock tube have been recently described by Mallins and Tardy.88 An important topic which has received much attention in the last few years is that of collisional energization by the wall of the reaction vessel.This is particularly important for very low-pressure pyrolysis experiments (VLPP) where until recently it was assumed that the collisional efficiency for wall energization BW was unity.89 An ingenious method for the direct study of gas-wall vibrational energy transfer known as the Variable Encounter Method has been devised by Rabinovitch and co-workers.90-99 This method allows the study of vibrational energy transfer from the hot walls of a reactor to an 84 B.S. Rabinovitch and I. E. Klein, J. Phys. Chem., 1978, 82, 243. E. Kamaratos, J. F. Burkhalter, D. G. Keil, and B. S. Rabinovitch, J. Phys. Chem., 1979, 83, 984. 86 H. Endo, K. Gliinzer, and J. Troe, J. Phys. Chem., 1979, 83, 2083. A. J. Stace and J. N. Murrell, J. Chem. Phys., 1978, 68, 3028. R. J. Mallins and D. C. Tardy, Int. J. Chem. Kinet., 1979, 11, 1007. 89 K. D. King, Int. J. Chem. Kinet., 1981, 13, 273. 90 D. F. Kelley, B. D. Barton, L. Zalotai, and B. S. Rabinovitch, J. Chem. Phys., 1979, 71, 538. 91 D. F. Kelley, L. Zalotai, and B. S. Rabinovitch, Chem. Phys., 1980, 46, 379. 92 M. CIFlowers, F. C. Wolters, B. D. Barton, and B.S. Rabinovitch, Chern. Phys., 1980, 47, 189. 93 B. D. Barton, D. F. Kelley, and B. S. Rabinovitch, J. Phys. Chem., 1980, 84, 1299. 94 M. C. Flowers, F. C. Wolters, D. F. Kelley, and B. S. Rabinovitch, Chem. Phys. Lett., 1980, 69, 543. 95 D. F. Kelley, T. Kasai, and B. S. Rabinovitch, J. Chem. Phys., 1980, 73, 5611. 96 F. C. Wolters, M. C. Flowers, and B. S. Rabinovitch, J. Phys. Chem., 1981, 85, 589. 97 M. C. Flowers, F. C. Wolters, D. F. Kelley, and B. S. Rabinovitch, J. Phys. Chem., 1981, 85, 849. 98 D. F. Kelley, T. Kasai, and B. S. Rabinovitch, J. Phys. Chem., 1981, 85, 1100. 99 T. Kasai, D. F. Kelley, and B. S. Rabinovitch, Chem. Phys. Lett., 1981, 81, 126. Current Aspects of Unimolecular Reactions initially cold gas i.e. during the transient region of a unimolecular reaction.The reaction (for example the isomerization of cyclopropane to propene) can be followed by sampling after a variable number of collisions between the gas molecules and the reactor wall. The average probability of reaction per collision is then compared with calculations based on various distributions of energy transfer probabilities. The results have shown that whereas gas-wall collisions are more efficient than gas-gas collisions and fiw = 1 for low temperatures (<400K), at higher temperatures fiw and (LIE,) decline with temperature as found for homogeneous gas phase intermolecular energy transfer. VLPP experiments have been carried out by Gilbert and King"' on the decomposition of chlorocyclobutane which occurs via two different reaction channels (see ref.e, Table 2). It was found possible to derive data relating both to gas-gas and gas-wall energy transfer processes from the pressure and temperature variation of the rate constant for the two channels. The results showed a decline of fiw and (dEd) from 930 K to 1150 K in agreement with previous work on single channel decompositions of cycloheptatriene and [2H2]cyclopropane. Interest in non-thermal unimolecular reactions such as those induced by laser irradiation has stimulated some recent model calculations involving solution of a master equation for non-equilibrium conditions and a range of collisional energy transfer models." 5 Other Non-thermal Energization Studies In addition to the chemical activation, photoactivation, and laser-induced unimolecular reaction experiments already described in connection with intra- and inter-molecular energy transfer, many others have been reported where the principal aims have been to obtain other kinds of information.Among examples of recently studied chemically activated systems is a comprehensive studyio2 of 1-, 2-, and 3-methyl chlorocyclobutanes prepared by the insertion of singlet methylene (CH, 'A,) into chlorocyclobutane. Chlorine abstraction can also produce chloromethyl and cyclobutyl radicals, which subsequently combine to produce chemically activated chloromethylcyclobutane (Scheme 2). All these chemically activated species can undergo both HCl elimination and ring rupture reactions.Estimated A factors for the relevant reactions have been used to assign vibration frequencies and hence perform standard RRKM calculations with a view to determining threshold energies for the various processes by matching calculated and observed kE values. It was found that reasonable agreement between these k, values was possible for the methylchlorocyclobutanes if a value of dHfo(CH2'A,) of 422 kJ mol- was assumed. This value is within the error limits of the most recently a~cepted''~~~'~ value of 414 k8 kJ mol- '. loo R. G. Gilbert and K. D. King, Chem. Phys., 1980,49, 367. lo' R.J. Mallins and D. C. Tardy, J. Phys. Chem., 1979, 83, 1017. lo2 B. E. Holmes and D. W. Setser, J. Phys. Chem., 1978, 82, 2450. R.K. Lengel and R.N. Zare, J. Am. Chem. SOC., 1978, 100, 7495. Io4 L. B. Harding and W. A. Goddard, Chem. Phys. Lett., 1978, 55, 217. Holbrook cis + trails ci., +-trans Scbeme 2 In a subsequent paper'05 Holmes and Setser have studied vibrational energy partitioning in the 1- and 3-methylcyclobutenes produced from the corresponding methylcyclobutanes by HCl elimination. These molecules are formed with sufficient energy to undergo further rearrangement to butadienes. The conclusions are that the methylcyclobutenes acquire a relatively low percentage (32%) of the potential energy available. This type of study is complementary to other methods of measuring energy disposal such as i.r. chemiluminescence. Further chemical activation studies reported include work on 1, l-dichloro- cyclopropane,lo6 methylamine,'07 and thiirane.Io8 Chemically activated species sometimes arise as a result of photoactivation, thus Tschuikow-Roux and Yano109 have applied RRKM theory to halogenoethyl radicals produced by chlorine-atom. addition to olefins in secondary reactions following the vacuum U.V. photolysis of 1,1,2-trichloro-2,2-difluoroethane.The photoactivation of butene isomers at similar wavelengths was studied by Collin and Wieckowski' lo who measured the production of allene from vibrationally excited ally1 radicals produced by the sequence of reactions C4Hs + hv + C4H8**+C3H5* + CH3 C3H5* -+ CH2=C=CHz + H Rate constants were measured for three different photon energies and the results when compared with a simple RRK treatment indicate a non-random distribution of the excess photon energy with the majority of the energy residing in the heavier fragment (C3H5* radical).In an extension of this work to various C4 and C5 olefins'll it was found that RRKM calculations on a strong-collision basis compared with experimental data led to similar conclusions regarding the vibrational energy content of butenyl radicals. B. E. Holmes and D. W. Setser, J. Phys. Chem., 1978,82, 2461. K. Eichler and H. Heydtmann, Int. J. Chem. Kinet., 1981, 13, 1107. lo' K. J. Chao, C. L. Lin, M. Hsu, and S. Y. Ho, J. Phys. Chem., 1979,83, 1241. A. G. Sherwood, I. Safarik, B. Verkoczy, G. Almadi, H. A. Wiebe, and 0.P. Strausz, J. Am. Chem. Soc., 1979, 101, 3000.T. Yano and E. Tschuikow-Roux, J. Chem. Phys., 1980, 72, 3401. G. J. Collin and A. Wieckowski, Can. J. Chem., 1978, 56, 2630. G. J. Collin and H. Deslauriers, Znt. J. Chem. Kinet., 1980, 12, 17. Current Aspects of Unimolecular Reactions Energy partitioning has also been studied in photoactivated trimethylene sulphoxide' l2 which produces vibrationally excited cyclopropane and sulphur monoxide with an initial non-random distribution of energy. 6 Infrared Laser-induced Unimolecular Reactions The study of chemical reactions induced by infrared multiphoton absorption (IRMPA) is a rapidly growing area spanning many established research disciplines. The implications of this work for the chemical kinetics of unimolecular reactions are obvious but even a restricted survey of the numerous papers which have appeared in this field in the last few years is beyond the scope of the present review.The highly selective nature of the present section is therefore intended to draw attention to some of the current topics of interest. Since the photon of infrared radiation of a typical frequency of 3 x s-' corresponds to about 12 kJ mol- ',dissociation of a molecule with an activation energy of 240 kJ mol- ' would require absorption of 20 photons. The mechanism by which this may be achieved via the quasi-continuum of absorbing states of a molecule is dealt with in many recent reviews."3~'3~''4-''6 Particular attention to chemical problems in infrared multiphoton dissociation (IRMPD) experiments of relevance to unimolecular reactions is given in the reviews of Ashfold and Hancock'" and Danen and Jang."' The motivation for much early research into the use of lasers to promote unimolecular reactions was the hope of achieving mode-selective chemistry i.e.the hope that energy initially deposited by a specific exciting wavelength might remain isolated in a particular mode which was closely related to the reaction coordinate in the Slater sense. This topic has been mentioned earlier in connection with intramolecular energy transfer since rapid intramolecular energy relaxation would make mode specificity impossible. The whole question has become one of degree, i.e. how fast or how slow can intramolecular energy transfer be made in relation to the rate of particular unimolecular reactions? Particular examples continue to promote discussion.One such example is the work of Hall and Kaldor68 on the irradiation of cyclopropane using two lasers with widely different frequencies. Two possible reaction channels are a high energy fragmentation into ethene and methylene and a lower energy isomerization to propene. A 4CH2xCH2 + :CH, A --* C3H6 ''' F. H.Dorer and K. E. Salomon, J. Phys. Chem., 1980, 84, 3024. 'I3 N. Bloembergen and E. Yablonovitch, Physics Today, 1978, 31(5), 23. H. W. Galbraith and J. R. Ackerhdt in 'Laser-induced Chemical Processes', ed. J. I. Steinfeld, Plenum, N. York, 1981, p.1. ''' M. N. R.Ashford and G. Hancock in 'Gas Kinetics and Energy Transfer', A Specialist Periodical Report, ed.P.G. Ashmore and R. J. Donovan, Royal Society of Chemistry, London, 1980, VOl. 4, p. 73. P. A. Schulz, Aa. S. Sudb4, D. J. Krajnovich, H. S. Kwok, Y. R. Shen, and Y. T. Lee, Ann. Rev. Phys. Chern., 1979, 30,379. 'I7 W. C. Danen and J. C. Jang in 'Laser-induced Chemical Processes', ed. J. I. Steinfeld, Plenum, N. York, 1981, p. 45. Holbrook Irradiation causing excitation of the C -H asymmetric stretch (3.22 pm) produced propene with almost no fragmentation whereas excitation of the CH, wag (9.50 pm) produced roughly equal yields of C3H6 and fragmentation products. The authors interpreted their results in terms of a selective mechanism. Argon was found to increase the yield of fragmentation products when 3.22pm radiation was used and this was thought to assist the intra- molecular energy transfer process.Thiele, Goodman, and Stone"' have developed a theory of intramolecular vibrational relaxation and unimolecular decay in which the former process occurs at a restricted rate. It was concluded that a low relaxation rate (possibly two orders of magnitude less than the lo', s-' deduced from the experiments of Rabinovitch and co-workers) could well account for Hall and Kaldor's observations. An alternative view of these experiments is taken by Ashfold and Hancock'l' who point out the possibility that the higher intensity of the CO, laser pulse (9.50 pm) could be responsible for the access to the higher energy reaction channel rather than any mode selectivity. The need for careful reporting of laser parameters such as pressure, waveIength, pulse shape, fluence variation within the cell and mode quality of the laser beam has recently been emphasized by Jang and Setser.lig Without this, comparison between work in different laboratories is difficult to make. These authors found as an example that inert gas effects on CO, laser-induced elimination of HF from CH3CH2F and CH3CF3 were related to laser pulse length.Brenner',' earlier found that product branching ratios for two different unimolecular channels for the decomposition of ethyl vinyl ether could be varied by changing the pulse duration. The higher energy channel was favoured by shorter more intense pulses. In contrast, Danen, Koster, and ZitterI2' attempted to force the laser-induced reaction of cis-3,4-dichlorocyclobutene to follow the non-allowed channel (2) instead of the channel (1) predicted by the Woodward-Hofmann rules of conservation of orbital symmetry for a thermally-allowed ground-state conrotatory ring-opening reaction (Scheme 3). Despite the use of relatively short laser pulses the non-allowed process (2) could not be made competitive and cis,trans-1,6dichlorobuta- 1,3-diene was the sole product.This system has recently been re-investigated by Gordon and co-workers12 who have found increased amounts of the 'symmetry-forbidden' products at fluences greater than those employed by Danen and co-workers. It is not however considered likely that the disrotatory concerted process (2) occurs on the ground electronic surface contrary to the Woodward-Hofmann rules.A more likely explanation '18 (a) E. Thiele, M. F. Goodman, and J. Stone, Chem. Phys. Lett., 1980, 69, 18. (b) E. Thiele, M. F. Goodman, and J. Stone, Opt. Eng., 1980, 19, 10. 'I9 J. C. Jang and D. W. Setser, J. Phys. Chem., 1979,83, 2809. lZo D. M. Brenner, Chem. Phys. Lett., 1978, 57, 357. lZ1 W. C. Danen, D. F. Koster, and R. N. Zitter, J. Am. Chem. Soc., 1979, 101, 4281. Chung-Rei Mao, N. Presser, Lian-Shun John, R. M. Moriarty, and R. J. Gordon, J. Am. Chem. Soc., 1981, 103, 2105. Current Aspects of Unimolecular Reactions CI c1 Cl Cl Scheme 3 is the direct formation of these products from a biradical intermediate or by secondary isomerization of the vibrationally hot product from reaction (1).Other examples of systems involving competing reaction channels have been the decompositions of methanol123 and of cycl~butanone.'~~* In the former, product analysis suggested that the primary process is CH30H +CH3 + OH irrespective of whether the initial excitation is of a CO stretch (9.7 pm C02 laser) or OH stretch (2.7pm HF laser). In the C02 laser-induced decomposition of cyclobutanone, a low energy pathway produces ethene and a higher energy pathway produces cyclopropane. Conflicting results have been reported with one set of worker^"^ finding the high energy path enhanced and another12' finding the low energy path enhanced at low pressures. Higher pressures would normally be expected to favour thermal equilibration and a Boltzmann-type energy distribution.Under these conditions it is possible in principle to use the RRKM theory to calculate specific rate constants for the unimolecular reactions of the energized species and hence to compare calculated and experimental rate constants for different forms of the energy distribution function. Calculations of this kind have been made for example by Steinfeld and co-workers126 for the decompositions of halogenated ethylenes and by Benson and co-w~rkers'~ for the decomposition of ethyl chloride. Although the data are fitted in general by RRKM-theory, in both cases the form of the distribution function cannot be determined unambiguously without independent measurements of absorbed energy.Despite the difficulties of interpretation, experiments on laser-induced reactions are clearly increasing and theoretical papers (not referred to here) are appearing in increasing numbers. In a purely practical sense, although mode selectivity may only be rarely possible, laser initiation may for other reasons often provide a lZ3 R. Bhatnagar, P. E. Dyer, and G. A. Oldershaw, Chem. Phys. Lett., 1979, 61, 339. lZ4 M. H. Back and R. A: Back, Can. J. Chem., 1979, 57, 1511. 12' R. G. Harrison, H. L. Hawkins, R. M. Leo,and P. John, Chem. Phys. Lett., 1980, 70, 555. C. Reiser, F. M. Lussier, C. C. Jenson, and J. I. Steinfeld, J. Am. Chem. SOC., 1979, 101, 350. Holbrook preferred route to a particular product and may often produce entirely different products from the comparable thermal reaction.The laser-induced decomposition of vinyl chloride, for example,'26 is a concerted elimination to give HCl and acetylene, whereas thermal studies in a flow system at 500-600 "C give dimerization to chloroprene. D. M. Golden and co-worker~'~' have used a CW infrared laser to heat an unreactive bath gas which then transfers energy by collision to the substrate. This technique -'laser powered homogeneous pyrolysis' -pioneered by Shaub and Bauer'28 is useful in the study of reactions which under normal conditions follow a rapid heterogeneous path. It was also claimed that laser-induced, SiF,-sensitized retro-Diels-Alder reactions carried out at room temperature by 'cold pyrolysis' enabled these reactions to proceed with fewer side-products than under normal thermal energization.29 7 Thermal Unimolecular Reactions at High Pressures This section includes a fairly comprehensive compilation of the Arrhenius parameters for the major types of unimolecular reactions considered in the book by Robinson and Holbrook' and in subsequent reviews. The author is well aware that this is not an 'evaluation' in the sense described by Baulch and Montague,' 30 although comments are made on some values where appropriate and some of the more interesting mechanistic aspects of recent work are discussed. An excellent contribution to a Symposium on the current status of the kinetics of elementary gas reactions by Cvetanovic, Singleton, and Paraskevopoulos' deals with recommended methods for the evaluation of temperature coefficients of rate constants.Although a common procedure for reporting the reproducibility of measured Arrhenius parameters does not exist and the procedure used by particular authors is frequently not stated, error limits are usually calculated from a least-squares treatment of the logarithmic form of the Arrhenius equation and most often correspond to the 95% confidence limits. In the tables which follow, the error limits quoted by the authors are listed and reference to the original papers must be made if their precise significance is important. The high pressure Arrhenius parameters A, and E, given in the tables must be assessed in relation to the type of unimolecular reaction involved.In many cases it is helpful to make comparisons with the parameters for the 'parent compound' such as cyclopropane or cyclobutane for small-ring alicyclic compounds, cyclobutene for the isomerizations of cyclic olefins, and ethyl chloride for four-centred decomposition reactions. Absolute rate-theory predicts a value of 10'3-1014 s-l for the pre-exponential factor of a unimolecular reaction whose activated complex resembles the initial state. Higher values can often be attributed to 'loose' activated complexes resembling more the product state, for example reactions with biradical intermediates and bond fissions '*'K. E. Lewis, D. F. McMillen, and D. M. Golden, J. Phys. Chem., 1980, 84, 226.12' W. M. Shaub and S. H.Bauer, Int. J. Chem. Kinet., 1975, 7, 509. D. Garcia and P. M. Keehn, J. Am. Chem.Sac., 1978, 100, 6111. ''O D. L. Baulch and D. C. Montague, J. Phys. Chem., 1979, 83, 42.'" R. J. Cvetanovic, D. L. Singleton, and G. Paraskevopoulos, J. Phys. Chem., 1979, 83, 50. Current Aspects of Unimolecular Reactions producing complex radicals. Activation energies are not so easy to predict but much progress has been made by the judicious application of Benson's rules to the estimation of the enthalpies of reactants and activated complexes. Those readers unfamiliar with the details of such calculations should consult 'Thermochemical Kinetics' by S. W. Benson (J. Wiley, N. York, 2nd Edn. 1976). A. Small AIicyclic Compounds and their Derivatives (Table 2).-Work on substituted cyclopropanes has clarified some of the details of earlier reports of these reactions.Heydtmann and KQrbit~er'~~ have re-investigated the thermolysis of 1,l-dichlorocyclopropane and established that a small fraction of the isomerization occurs to give 1,l-dichloropropeme, a product undetected by earlier workers. Eichler and Heydtrnann'O6 have reported slightly different Arrhenius parameters, for formation of this product, which fit well all the previous thermal and chemical activation studies of 1,l-dichlorocyclopropane. Robinson and Waller'33 have confirmed the strong accelerating effect of methyl substitution into chlorinated cyclopropanes found by Holbrook and co-worker~'~~in a comprehensive study of cis- and trans-l-chloro-2,3-dimethyl-cyclopropanes.Isomerization reactions occur to give cis- or trans-Cchloropent- 2-ene, as well as a 6-centred elimination (from the cis compound) producing pentadiene and HCl. Fluorine substitution in the cyclopropane ring is known to produce a decrease in activation energy and an enhancement of the rate of isomerization. gem-Difluoro-substitution has been studied by Ferrero and Staric~o'~ in 42-bis( trifluoromethy1)- 1,2,3,3- tetra-fluorocy clopropane and found to lead to elimination of CF2 instead of isomerization to propenes. The evidence appears to favour a concerted process for this elimination although a biradical path cannot be excluded. The analogous elimination of a methylene radical from a substituted cyclopropane has been observed in photochemical studies of alkyl-substituted cycl~propanes.'~~ have supported the view Theoretical calculations by Ro~si'~ that this process occurs by a two-step mechanism via the biradical (Scheme 4).Scheme 4 Some of the uncertainty in the postulation of biradical pathways for cyclopropane and other alicyclic systems arises from lack of knowledge of barriers to internal rotation in biradicals. Molecular orbital calculations by Lipscomb and co-13' H.Heydtmann and B. Korbitzer, Z. Phys. Chem. (Frankfurt), 1981, 125,255. 133 P. J. Robinson and M.J. Waller, Int. J. Chem. Kinet., 1979, 11, 937. 134 (u) K. A. Holbrook and K. A. W. Parry, J. Chem. SOC.,1970, 1019. (b) R.P. Clifford and K.A. Holbrook, J. Chem. SOC., Perkin 2, 1972, 1972. J. C. Ferrero and E. H. Staricco, Int. J. Chem. Kinet., 1979, 11, 1287. 136 J. C. Ferrero and E. H. Staricco, J. Am. Chem. SOC., 1978, 100, 7089. 13' A. Rossi, J. Phys. Chem., 1979, 83, 2554. Holbrook workers' 38 for the methylene cyclopropane rearrangement have predicted a low energy barrier to ring-closure of 13.8kJ mol- ' for the biradical intermediate involved. This value is, however, comparable with the difference between the theoretically predicted and observed activation energies for this isomerization. Doubts expressed by Berson about the validity of thermochemical estimates underlying the biradical mechanism have been amplified in a recent review.'39 Despite numerous experimental attempts to distinguish whether rotation a1 the biradical centres occurs simultaneously or independently, Berson concludes that questions still remain to be resolved which will require new experimental approaches.Direct measurements on the relative rates of rotation, cleakdge, and closure of tetramethylene biradicals which are relevant to the thermolysis of cyclobutane derivatives have been reported by Dervan and Santilli.140 Deuterated tetra- methylene radicals were generated by the thermal decomposition of [cis-3,4-'H,]-3,4,5,6-tetrahydropyridazine and these underwent rotation, cleavage, and closure reactions as in Scheme 5. From the results it was found that D D /D?fDc lo s11rc Df; N Df. ci,v D Scheme 5 kcleavage/kclosure = 2.2 0.2 and krotation/kclosure = 12 & 3,hence rotation is much faster than closure or cleavage under these conditions (439"C).The rate constants for closure and cleavage were taken to be the same for the cis and trans reactants. Comparison with the results of other workers shows that increasing methyl substitution at the biradical centre slows the rate of rotation relative to cleavage as would be expected from the increased moments of inertia. 13' D. A. Dixon, R. Foster, T. A. Halgren, and W. N. Lipscomb, J. Am. Chem. SOC., 1978, 100, 1359. 139 J. A. Berson in 'Rearrangements in Ground and Excited States', ed. P. de Mayo, Academic Press, N. York, 1980, p. 311. P. B. Dervan and D. S. Santilli, J. Am. Chem. SOC., 1980, 102, 3863.14' Current Aspects of Unimolecular Reactions High-pressure Arrhenius parameters for the decompositions of chloro-cyclobutane by alternative paths to vinyl chloride and ethene or buta- 1,3-diene and HCI were obtained by King and co-worker~'~~ from VLPP data using RRKM theory and assuming that gas-wall collisions are strong. In a more recent publication, Gilbert and King'" have shown that this assumption needs revision for the higher temperatures employed when Do deviates considerably from unity (See Section 4). Bromocyclobutane was also studied by these authors'42 who found no evidence for the expected path leading to vinyl bromide but only the path producing elimination of HBr. Among other cyclobutane derivatives studied, the decomposition of vinyl cyclobutane is of interest. Frey and P~ttinger'~~ found that this compound undergoes a ring expansion reaction (as does vinyl cyclopropane) in addition to decomposition to ethene and buta-1,3-diene.Both reaction paths were inter- preted in terms of biradical intermediates which are stabilized by allylic resonance. The unimolecular reactions of substituted cyclopentanes and higher alicyclics are receiving more attention. Among mechanistic studies are those of Gajewski onand Sala~arl~~ the ring-opening reactions of some methyl-substituted 1,3-dimethylenecyclopentanes.The results suggest that in these cases the biradical intermediates formed may reclose faster than they undergo rotation, with the consequence of partial stereospecificity in the reaction products.B. Polycyclic Systems(Table 3).-The relatively few studies of polycyclic systems yielding kinetic parameters in this review period are listed in Table 3. The effects observed are generally those predicted from the similar reactions in comparable monocyclic systems. Two gem-fluorine substituents in a cyclopropane ring are found to produce considerable rate enhancement and this is also true for substituted ~piropentane.'~' The observations have been rationalized in terms of biradical intermediates. For 1,1,2,2-tetrafluorospiropentane,a CF, extrusion reaction occurs with similar Arrhenius parameters to that of the comparable reaction for 1,1,2,2-tetrafluorocyclopropane. Huybrechts and co-~orkers'~~ have studied the Diels-Alder addition of ethene and cyclohexa- 1,3-diene to bicyclo[2.2.2]oct-2-ene and the reverse thermolysis at 548-632K. The Arrhenius parameters are in good agreement with those found by Cocks and Frey14' for the reaction at higher temperatures (649-718 K) and can be explained in terms of a biradical intermediate with little or no activation energy to ring closure.141 K. D. King, B. J. Gaynor, and R. G. Gilbert, Int. J. Chem. Kinet., 1979, 11, 11. 14' K. D. king and R. G. Gilbert, fnt. J. Chem. Kinet., 1980, 12, 339. 143 H.M. Frey and R.Pottinger, J. Chem. SOC.,Faraday Trans. 1, 1978,74, 1827. 144 J. J. Gajewski and J. Salazar, J. Am. Chem. Soc., 1979, 101, 2739, 2740. 145 W. R. Dolbier, Jr., S. F. Sellers, B. H.Al-Sader, and T. H. Fuller, J. Am. Chem. Soc., 1981, 103, 717. 14' G. Huybrechts, D. Rigaux, J. Vankeerberghen, and B. van der Mele, Int. J. Chem. Kinst., 1980, 12, 253. 14' A. T. Cocks and H. M. Frey, J. Chem. SOC. (A), 1971, 1661. Table 2 High-pressure Arrhenius parameters for small alicyclic compounds and their derivatives Reactant Product(s) log,,(A,/s- ') E,/kJ mol-' Ref: C 1C1 ~CHZClCCl=CHz 15.13* kO.1 241.9*f1.3 MeCH=CCI, 14.50f0.38 250 f5 a HCl+ /\\/\\ 13.92L-0.08 199.6L0.9 b MeAMe c1 HCl+ 13.8f0.4 190.2f4 bIPe HX, 14.6k0.4 199.5f4 b Me c1 12.6f0.5 123.8L-4F2 F2 D=F2 F2P FxCF3 FXF 15.15 k0.16 186.2L-1.7 d F3C F F3C CF3 F ;F3 'c =c + CF, 15.14k0.28 195.4k2.9 d /\ F3C-F F S F FxCF3 15.35f0.19 181.6f1.7 d F3C CF3 F,G F FF\ /c=c + CF2 15.39f0.26 194.9k 2.5 d / \ F3C CF3 CZH, + CH,-CHCl 14.8f0.3 255.6k4.2 ed" 13.6L-0.3 233.0k4.2 e +HBr 13.6L-0.3 217.6k4.2 fdBr 2 CH,-CHCF 16.0k0.3 221.3k3.3 CN Current Aspects of Unimolecular Reactions Table 2-continued Reactant Em/kJ mol-' C2H4 + CH2=CF2 15.60 k0.13 289.7 i-1.8cr: 15.34 k0.05 292.0 t-0.8d;F 15.27 k0.06 308.0 & 0.9 14.87 k0.07 2 12.2 f0.8CJ-K-ltC2H4 13.86 k0.13 203.5 k1.515.22 f0.15 219.02 f1.770 d" 13.76 k0.20 204.8 1 ? 2.3 1 0 16.1 354.9 16.25 397.6 12.8* k0.3 274.5* f5.4CN 12.8 i-0.3 283.7' f 10.5 16.0* f0.3 334.7* f4.6 16.3* f 0.3 335.13 f8.4 Bu"CH=CH, 16.7 369.I m a Ref. 132. For a better fit to both thermal and chemical activation data see ref.106. *Values taken from previous work,see Trans. Faraday SOC., 1970, 66,869. Ref. 133. W. R. Dolbier, S. F. Sellers, B. H.Al-Sader, and B. E. Smart, J. Am. Chem. SOC., 1980,102,5398. * Ref. 135. 'Ref. 141. Ref. 142. K. D. King and R. D. Goddard, Int. J. Chem. Kinet., 1978, 10, 453. H. M.Frey and R. T. Conlin, J. Chem. SOC., Faraday Trans. I, 1979, 75, 2556. ' H. M.Frey and R. T. Conlin, J. Chem. SOC., Faraday Trans. I, 1980, 76, 322. Ref. 143. 'W. Tsang, Int. J. Chem. Kinet., 1978, 10, 599; see also B. L. Kalra, S. A. Feinstein, and D. K. Lewis, Can. J. Chem., 1979, 57, 1324. ' Ref. 175. *Stirred-flow reactor, 'VLPP extrapolated by RRKM, 'parameters for (k, + k3) from VLPP. W. Tsang, Int.J. Chem. Kinet., 1978, 10, 1119. Holbrook Table 3 High-pressure Arrhenius parameters for reactions of polycyclic compounds Reactant Product log,,(A,/s-’) E,/kJ mol-’ Ref: 16.1& 0.2 242.7 k2.1 amF2+ dCF2 F2 13.8 k0.1 189.5 f 1.3 b DaF2 F2 BCF t CF2 15.2 k0.2 203.3 1.7 14.8 +_ 0.2 216.3 k2.5PaF2 F2 0+ 11 15.12f0.04 239.7 k0.4 C W. R. Dolbier, Jr., S. F. Sellers, B. H. Al-Sader, and S. Elaheimer, J. Am. Chem. SOC., 1981, 103, 715. * Ref. 145. Ref. 146. C. Cyclic Olefins (Table 4).-Some recent kinetic results on reactions of cyclic olefins are given in Table 4. The elimination of hydrogen from cyclopentene to give pentadiene is believed, from previous work, to be a unimolecular process occurring via a concerted symmetry-allowed 1’4-transition state.King14* has re-investigated this compound by VLPP and on the basis of a 1,4-H2 elimination transition state and an assumed A, value based on the results of previous workers he derived the E, value 255.2 kJ mol- by the use of RRKM theory. Lewis’49 and co-workers have carried out some shock-tube experiments using deuterium-labelled cyclopentene and have shown that at high temperatures (1 100-1300 K) both the ‘symmetry allowed’ [1,4] and ‘symmetry disallowed‘ [1,2] eliminations occur, although the former is favoured by a ratio of 2:l. The results lead to an apparent difference in activation energies of ‘v 34 kJ mol- for the two paths. The authors point out the need for laser-induced experiments purporting to produce preferential ‘disallowed’ products to be compared with thermally activated systems at comparable temperatures.Rather similar experiments have been carried out by Tardy’ 50 and co-workers K. D. King, Int. J. Chem. Kinet., 1978, 10, 117. 149 D. K. Lewis, M.Greaney, and E. L. Sibert, J. Phys. Chem., 1981, 85, 1783. lS0 D. C. Tardy, R. Ireton, and A. S. Gordon, J. Am. Chem. SOC., 1979, 101, 1508. 191 Current Aspects of Unimolecular Reactions Table 4 High-pressure Arrhenius parameters for reactions of cyclic oleJins Reactant Product (s) log,,(A,/s-’) E,/kJ mol-Ref: 14.1 277 aQ +D2 D D DQ tHD 12.8 243 a D D D2C =CH--CH=CD2 \/+ C=C 14.93f0.79 272.8k11.7 b / \ D ‘D f D20 12.63k0.76 257.7k 11.3 b 15.57 290.9 c=Y 13.35 255.2 d0 Q +H2 a Ref.149. Ref. 150. Also minor amounts of symmetry-forbidden products. Only 3 temperatures studied. J. M.Simmie, Int. J. Chem. Kinet., 1978, 10, 227. (Retro-Diels-Alder.) Ref. 148. on deuterium-labelled cyclohexene in a conventional static thermolysis apparatus at lower temperatures (750-800 K). Here the activation energy difference between ‘allowed’ and ‘disallowed’ D, or (HD + H,) elimination reactions was (25 f12)kJ mol-’. The primary reaction for this compound however is the retro Diels-Alder reaction to ethene and buta- 1,3-diene. D. Heterocyclic Compounds(Table 5).-The high-pressure Arrhenius parameters for some recently studied thermolyses of heterocyclic compounds are shown in Table 5.Among the oxetanes studied, the effect of a vinyl group in the ring in cis-2,4-dimethyl-trans-3-vinyloxetane’ produced the expected lowering of the 51 activation energy for reaction to penta- 1,3-diene and ethanal, presumably by allylic stabilization of the biradical intermediate. Although symmetrically-substituted oxetanes show Arrhenius parameters similar to those of oxetane itself, it has previously been ob~erved”~ that alkyl substituents in the 2-position lead to lower A-factors and activation energies. H.A. J. Carless, A. K. Maitra, R. Pottinger, and H. M.Frey, J. Chem. SOC., Faraday Trans. 1, 1980,76, 1849. 152 M.J. Clarke and K. A. Holbrook, J. Chem. SOC., Faraday Trans. I, 1977, 73, 1890. 192 Holbrook In the case of the recently studied thermolysis of 2,2-dirnethylo~etane'~~ this lowering occurs to such an extent for the path leading to formation of the more highly alkylated olefin (isobutene) that a biradical mechanism does not appear able to account for the experimental observations.Imai and Nishida'" likewise concluded that their observations on the solution thermolyses of some 3-alkyl-2-phenyloxetanes,in particular the trans isomers could best be explained by concurrent concerted fragmentation and a biradical process. The concerted process for these [2 + 21 eliminations would necessarily involve a twisted transition state [02,+ 02,]from orbital symmetry considerations. This is normally excluded on energetic grounds but such a mechanism would explain the preference for the more highly alkylated olefinic product found both in the case of 2,2-dimethyloxetane and of the 3-alkyl-2-phenyloxetanes. Further work is in progress in the author's laboratory to extend these ideas to other substituted oxetanes.E. Alkyl Halides (Table 6).-The unimolecular decompositions of alkyl halides are known to occur via 4-centred transition states and recent work has been involved with confirming the model proposed initially by Maccoll and Thomas' and quantitatively treated by Benson, Bose, and Haugen.ls6. ' 57 Maccoll and co-~orkers~~~,~~~have recently investigated the involvement of the C1 atom in the transition state for ethyl chloride and some secondary and tertiary chloride decompositions by measurement of the relative k(3'Cl) and k(37C1) ratios and their temperature dependence.Chuchani and co-workers160 on the other hand have studied the enhancement of rate due to stabilization of the polar transition state produced by anchimeric assistance of polarizable groups substituted in the /3 position. Neopentyl halides are unable to decompose molecularly via 4-centred transition states since they lack a B hydrogen atom. Recent work by Shapiro and co-workers'61* 162 has, however, shown that molecular elimination can occur via 3-or 5-membered transition states when neopentyl chloride or bromide is decomposed in the presence of cyclohexene inhibitor. The elimination reactions which occur can be explained by formation of an ion-pair type of transition state followed by highly polar cyclic transition states in which the halogen atom removes an a-or y-hydrogen atom.F. Esters (Table 7).-The unimolecular decompositions of esters are known to occur uia a 6-centred transition state. These decompositions have recently been lS3 P.Hammonds and K. A. Holbrook, J. Chem. SOC.,Faraday Trans. I, 1982,78, 2195. 154 T.Imaiand S. Nishida, Chem. Lett. (Japan), 1980, 1, 41. lS5 A. Maccoll and P. J. Thomas, Nature, 1955, 176, 392. S. W. Benson and A. N. Bose, J. Chem. Phys., 1963,39, 3463. '"S. W. Benson and G. R.Haugen, J. Am. Chem. SOC., 1965,87, 4036. A. Maccoll and M. N. Mruzek, J. Chem. SOC., Faraday Trans. 1, 1978, 74, 2714. lS9 A. Maccoll, M. N. Mruzek, and M. A. Baldwin, J. Chem. SOC.,Faraday Trans. 1, 1980, 76, 838.ltio G. Chuchani, I. Martin, G. Martin, and D. Bigley, Int. J. Chem. Kinet., 1979, 11, 109. R.L. Failes, Y.Mollah, and J. S. Shapiro, lnt. J. Chem. Kinet., 1981, 13, 7. Iti2 R.L. Failes, Y.Mollah, and J. S. Shapiro, Int. J. Chem. Kinet., 1979, 11, 1271. Table 5 High-pressure Arrhenius parameters for the thermolysis of some heterocyclic compounds Reactant Product(s) log1ri(A,/s -CF,COMe 13.15 f0.57 OLCF, CFjCHzCHO 14.49k0.68 Me Me,CCOMe 13.57 f0.46 MeC( =CH,)OCHMe (+ Me2C0 + MeCH=CH,) 13.55k0.34 MeC( =CH2)C(OH)Me2 11.81 4 0.94 MeCHO + MeCH-CHCH-CH2 13.42k0.60 Me MeCH=CH, + CH20 14.64f0.36 MeCHO + C2H4 14.53& 0.36Cl"^ Me2CH=CH2 + CH20 13.48f0.25 Me2C0 + C2H4 15.56f0.34 E,/kJ mol-239.7 _+ 7.6 259.0f9.4 233.4k6.2 35.0f4.5 209.8 k12.5 200.3 f7.5 249.8f4.7 249.8k4.7 222.1 f2.9 270.6 _+ 4.4 Ref: a b d d R' Vc=c//\ R3 -I-co, R* R4 12.89 138.7 12.42 126.5 12.79 142.0 e 12.92 133.6 12.46 147.9 12.81 135.3 Me Me I CHz=CCH=-CH, + CHzO 14.62k0.03 209.5 k0.4 94 CH,=CHCH=CHz + CH20 14.31 k0.14 208.1 k1.7 h MeCHO + MeCH=CHCH=CH2 13.91 k0.1 196.3k1.2 1 Me nMe "M.C.Flowers and M. R. Honeyman, J. Chem. SOC., Faraday Trans. I, 1980, 76, 2290. 'M. C. Flowers and M. R. Honeyman, J. Chem. Sac., Faraday Trans. 1, 1981, 77, 1923. Ref. 151. Ref. 153. T. Imai and S. Nishida, J. Org. Chem., 1980, 45, 2354. Arrhenius parameters calc from AH *, AS * values at 150 "C. Reactions studied in tetramethylene diamine solvent. H.M. Frey and H. P. Watts, Int. J. Chem. Kinet., 1981, 13, 729. H. M. Frey and S. P. Lodge, J. Chem. Sac., Perkin Trans. 2, 1979, 1463. @ H. M. Frey, R. Pottinger, H. A. J. Carless, and D. J. Lingley, J. Chem. SOC.,Perkin Trans. 2, 1979. 1460. Table 6 High-pressure Arrhenius parameters for some alkyl halide eliminations Reactant froduct(s) EJkJ mol-' Re$ EtCl 241.8 f4.2 a+ Hcll 13.84 i-0.2 CHzDCHzCl CHD=CH2 + HC1 13.33fO.l 239.2 f2.09 bC2H4 + DCl EtF C2H4 + HF 13.65 k0.2 248.9 f4.2 EtBr C2H4 + HBr 12.5 216.3 P~"cI C,H, + HCI 13.49 f0.24 227.6 f4.6 13.44 f0.28 229.3 f5.4 fl-substituted ethyl chlorides X'X2CHCH2CI X'X2C=CH2 + HCl X1=H X2=Me2N 13.22f0.17 203.7 f2.1 e X1 = H X2 = Me0 14.06 k0.53 244.7 f7.1 e Xi = H X2 = Me2CH 14.12 f0.05 235.3 k0.7 XT=H XL=Me3C 13.08 f0.19 218.8 f2.5 X'=H X2=CN 13.20 f0.3 241.0 f4.0 Xi = Me X2 = CN 13.49 k0.25 241.2 f 3.5 X'= H X2 = Ph 13.07k0.35 220.9 f4.6 X1 = H X2 = PhOMe 13.8 1f0.34 228.4 i-4.6 MeC(Me)ClEt Me2C=CHMe + HCI 13.77 f0.25 184.1f2.6 MeC(Me)C1CHMe2 Me,C=CMe, + HCI 13.33 i-0.18 175.3 f1.9 R' R3 'c=c'-/ \ + co, 1 R4 12.89 138.7 12.42 126.5 12.79 142.0 e 12.92 133.6 12.46 147.9 12.81 135.3 Me Me I CHZ=CCH=CH2 + CH20 14.62 i-0.03 209.5& 0.4 96 CHZ=CHCH=CH, + CH20 14.31 +_ 0.14 208.1 & 1.7 h MeCHO + MeCH=CHCH=CH, 13.91 & 0.1 196.3k1.2 1 Me AxMe M.C. Flowers and M. R. Honeyman, J. Chem. SOC., Faraday Trans. 1, 1980, 76, 2290. 'M. C. Flowers and M.R. Honeyman, J. Chem. SOC., Faradoj Trans. I, 1981, 77, 1923. Ref. 151. Ref. 153. T. Imai and S. Nishida, J. Org. Chem., 1980, 45, 2354. Arrhenius parameters calc from AH *, AS * values at 150 "C. Reactions studied in tetramethylene diamine solvent. H. M. Frey and H. P. Watts, Int. J. Chem. Kinet., 1981, 13, 729. I, H. M. Frey and S. P. Lodge, J. Chern. SOC., Perkin Trans. 2, 1979, 1463. H. M. Frey, R. Pottinger, H. A. J. Carless, and D. J. Lingley, J. Chem. SOC.,Perkin Trans. 2, 1979, 1460. CMe,CH,Br t } t HBr 14.2 0.3 247 f5 k CMe,CH2CI *2 MB-1 + }+HCI 13.78 & 0.6 258.7 f8.4 **2 MB-2 Me,C =CH(CH2), C1 C6HIo +HCI 13.43f0.3 2 15.0& 3.7 m CH2 =CH(CH2)2Cl CH2=CHCH=CH, +HCI 13.79 k0.17 223.8 f2.1 n CH2=CH(CH2),CI CHZ=CHCHZCH=CH2 +HCl 14.25& 1.20 238.4 k12.7 n CH2 =CH(CH2).+C1 CsHlo + HCI 12.38f 0.22 209.6 f2.9 n * 2-methylbut-1-ene. ** 2-methylbut-2-ene.a P. J. Evans, T. Ichimura, and E. Tschuikow-Roux, Int. J. Chem. Kinet., 1978, 10, 855. P. J. Papagiannakopoulos and S. W. Benson, Int. J. Chem. Kinet., 1982, 14, 63. Overall parameters. K. Okada, E. Tschuikow-Roux, and P. J. Evans, J. Phys. Chem., 1980, 84, 467. T. J. Park and K. H. Jung, Bull. Korean Chem. SOC., 1980, 1, 30. Ref. 160. /G. Chuchani, J. Phys. Chem., 1978, 82, 2767. #K. D. King, J. Chem. SOC., Farday Trans. 1, 1978, 74, 912.'G. Chuchani and A. Rotinov, React. Kinet. Catal. Lett., 1979, 12, 333. 'J. A. Hernandez A and G. Chuchani, Int. J. Chem. Kinet., 1978, 10, 923; see also G. Chuchani, J. D. Medina, I.Martin, and J. A. Hernandez A, J. Phys. Chem., 1981, 85, 3900. 'G. Chuchani and I. Martin, J. Phys. Chem., 1980, 84, 3188. Ref. 161. Maximum inhibition by cyclohexene. 'Ref. 162. Experiments with [*H,]neopentyl chloride have confirmed the molecular eliminations occurring for maximal inhibition by cyclohexene. G. Chuchani, I. Martin, M. E. Alonso, and P. Jano, Int. J. Chem. Kinet., 1981, 13, 1. " G. Chuchani, J. A. Hernandez A, and I. Martin, Int. J. Chem. Kinet., 1979, 11, 1279. For some other examples of neighbouring-group participation see G. Chuchani, A. Rotinov, and R. M. Dominguez, Int. J. Chem. Kinet., 1982, 14, 381. Substitution at C-3 H H H H H H Pr 12.73f0.29 202.5 f3.8 9 H H H H H H But 12.34f0.35 194.1 f4.2 9 H H H H H H BUS 13.62+_ 0.09 211.9 f1.2 k H H H H H H Bu' 12.82+_ 0.05 203.1 f0.6 h H H H H H H CC6H1 1 13.20i-0.26 207.4f3.2 h H H H H H H GH9 13.30f0.28 208.1 f3.4 h H H H H H H Me2N 13.90f0.30 220.4 f3.8 1 H H H H H H Me0 12.04+_ 0.24 203.7 .t2.9 1 H H H H H H MeS 11.27 f0.39 179.0 f4.6 i H H H H H H c1 12.14f0.66 202.0f8.4 i H H H H H H F 12.68f0.60 211.2 f7.1 1 H H H H H H CN 11.51 f0.13 171.9f1.7 i H H H H Me H Bu' 12.87f 0.31 181.2f 3.4 jH H H H H H CH=CH2 13.20-t 0.17 200.8 & 2.1 k H H H H H H CH2CH-CH2 12.81f0.36 204.0f4.5 k H H H H H H (CH2)*CH=CH2 12.43f0.14 197.5 f1.8 k H H H H Me H CH-CH2 12.34f0.25 178.2 f2.9 k H H H Me Me H CH-CH2 13.59.t0.30 169.9f2.9 k H H H H H H Me,C=CH 13.21 f0.14 199.6f1.7 1 H H H H H H O-MeC6H4-12.47 f0.28 192.5 f3.5 m H H H H H H o-MeOC,H4-12.49f0.20 193.0f 2.4 m a G.Chuchani, I. Martin, G. Fraile, 0. Lingstuyl, and M. J. Diaz, Int. J. Chem. Kinet., 1978, 10, 893. * M. A. G. Sarmiento, R. M. Dominguez, and G. Chuchani, J. Phys. Chem., 1980,84, 2531. G. Chuchani, J. L. Triana, A. Rotinov, D. F. Caraballo, J. Phys. Chem., 1981, 85, 1243. P. G. Blake and B. F. Sharydeh, Int. J. Kinet., 1981, 13, 463. 'G. Chuchani and J. L. Triana, React. Kinet. Catal. Lett., 1981, 18, 433. I. Martin, G. Chuchani, I. Avila, A. Rotinov, and R. Olmos, J. Phys. Chem., 1980, 84, 9. G. Chuchani, I. Martin, and I. Avila, Int. J. Chem. Kinet., 1979, 11, 561. '1. Martin and G. Chuchani, J. Phys. Chem., 1981, 85, 3902. G. Chuchani, I. Martin, J. A. Hernindez A, A. Rotinov, and G.Fraile, and D. B. Bigley, J. Phys. Chem., 1980, 84, 944. jG. Chuchani and R. M. Dominguez, Int. J. Chem. Kinet., 1981, 13, 577. Individual Arrhenius parameters for the formation of various pentene products are given. 'I. Martin, J. A. Hernandez A, A. Rotinov, and G. Chuchani, J. Phys. Chem., 1979, 83, 3070. ' G. Chuchani, I. Martin, and M. E. Alonso, J. Phys. Chem., 1981, 85, 1241. Reaction inhibited by propene. G. Chuchani, A. Rotinov, D. F. Caraballo, and J. D. Medina, React. Kinet. CataI. Lett., 1980, 13, 173. Reactions in presence of cyclohexene inhibitor. Current Aspects of Unimolecular Reactions reviewed by Taylor.' The table summarizes recent experimental work (largely by Chuchani and co-workers) concerning the decompositions of esters substituted at the C-1, C-2, or C-3 positions (see Table 7) in the parent molecule.The principal conclusions are that for substitution at C-1, electron-withdrawing substituents enhance the rate of elimination whereas electron-releasing sub- stituents decrease it; many substituents at C-2 enhance the rate by steric acceleration; at C-3, branched alkyl-substituents cause slight steric acceleration and anchimeric assistance appears not to occur. The results confirm, in general, the semi-polar semi-concerted transition state which is considerably less heterolytic than that for alkyl halides. The Arrhenius parameters reported in Table 7 are in line with those found in earlier work. G. Other Six-centred Elimination Reactions (Table 8).-Arrhenius parameters for some other six-centred elimination reactions are listed in Table 8.Included are the molecular retro-ene decompositions of a number of unsaturated hydro-carbons. When studied under normal static thermoiysis conditions these are often accompanied by free radical processes, as for example was found by Richard, Scacchi, and who estimated Arrhenius parameters for 3-methyl- pentene decomposition from measured rate constants for the reverse addition of ethene to but-2-ene and the known equilibrium constant. King'65 and co- workers have used the technique of VLPP to study the competing processes of retro-ene reaction and C-C bond fission for a number of alkenes and alkynes. For alkynes it is found that the retro-ene pathway is faster relative to bond fission than for alkenes.For hex-1-yne, the A factor found by Tsang'66 for the molecular retro-ene reaction was assumed and Em found from the VLPP data by RRKM calculations. The gas-wall collisional energy transfer efficiency pw was assumed to vary with temperature in the manner found for chloro- cyclobu tane. All of the A factors shown in Table 8 lie within the limits of 10".5*'.5 predicted by O'Neal and Benson 167 for reactions with six-centred transition states. H. Bond Fission Reactions (Table 9).-Probably the most-studied unimolecular bond fission reaction in recent years has been the decomposition of ethane into two methyl radicals. Two recent determinations of the high pressure Arrhenius parameters are quoted in Table 9.From a survey of the experimental data on this reaction and the reverse methyl radical recombination, Baulch and Duxbury'68 have given the following recommended rate expressions : log(k&,,mp/S-') = (16.38 f0.5) -(44010 3170/2.303[T/K]) R. Taylor in 'The Chemistry of the Functional Groups' Suppl. Vol. B 'Acid Derivatives', ed. S. Patai, Wiley, London, 1979, p. 859. 164 C. Richard, G. Scacchi, and M.H. Back, Int. J. Chem. Kinet., 1978, 10, 307. 165 K. D. King, Int. J. Chem. Kinet., 1979, 11, 1071, and other refs. following Table 8. 166 W. Tsang, Int. J. Chem. Kinet., 1978, 10, 687. H. E. ONeal and S. W. Benson, J. Phys. Chem., 1967, 71, 2903. D. L. Baulch and J. Duxbury, Combustion and Flame, 1980,37, 313. 4u 2: cn u5 2 +I +I22 N N N N +I 2% dN +I t-' 3 z N 0 +I 2 0 +I oo r- II + II + f / -c 20 1 s J: .-2 2 2: +I +I +I2:+I N r(h rc)---.? 22v1 0 0 t +I 2 +I +I +I v Zd0" 1M-m 03: 0Y+ 3:Y5 + +K II r-OY-0t 0 L \== L Carbamates Me(H)NCO,Et MeNH, + C02 + C,H4 12.47 192.5 MeNCO + EtOH 12.39 197.6 Sulphides Me 'S MeCH=CH, + CH2=S 11.23f0.25 160L-3 I CH2=CHCH2 (CH2 ECHCH,)~S CH,=CHCH=S + C3H6 11.01 f0.06 138.2k0.7 PhCH, PhCH=S + C3H6 10.93f0.18 141 f2.0 k CH, =CHCH2 Bu" PrCH=S + C3Hs 11.42f0.28 155 i-3 CHZ =CHCHZ "Ref. 164. Ref. 165. W. Tsang, lnt. J. Chem. Kine?., 1978, 10, 1119. K. D. King J. Phys. Chem., 1980, 84, 2517. 'K. D. King, In?. J. Chem. Kinet., 1981, 13, 245.I Ref. 166. K. D. King, Int. J. Chem. Kinet., 1981, 13, 273. M. Rossi and D. M. Golden, Int. J. Chem. Kinet., 1979, 11, 715. ' M. C. Flowers and M. R. Honeyman, J. Chem. Soc., Faraday Trans. I, 1981, 77, 1921. N. J. Daly and F.,Ziolkowski, Int. J Chem. Kinet., 1980, 12, 241. 'G. Martin, M. Ropero, and R. Avila, Phosphorus and Sulfur, 1982, 13, 213. ' G. Martin, A. Drayer, M. Ropero, and M. E. Alonso, Int. J. Chem. Kinet., 1982, 14. 131. Current Aspects of Unimolecular Reactions for the range 750-1500 K and l~g(k~~,,~~/cm~mol-'s-') = (13.38 k0.08) for the range 250-420 K. A computer simulation of ethane decomposition under shock-tube conditions at high temperatures has been re~0rted.l~~ The Arrhenius parameters for bond fission reactions at high pressures are expected to give activation energies close to the bond-dissociation energies and high A factors characteristic of loose transition states.The difficulty of applying transition-state theory to such reactions has been discussed by Golden56 and some examples of the application of RRKM theory to a modified .Gorin model have been discussed by Baldwin, Lewis, and Golden.'70 8 Thermal Reactions in the Low-pressure and Fall-off Regions (Table 10) Relatively few conventional studies of unimolecular reactions of moderately complex molecules in their fall-off regions have been reported recently. Bailey and Frey' 71 have studied the decomposition of 1,1,2,2-tetrafluorocyclobutane at pressures down to 0.03 Torr and have measured the ratio of rate constants for the two decomposition channels.The results were explained using a step- ladder energy transfer model with LIE = 12 kJ mol-'. Flowers'72 has re-examined the RRKM treatment of the fall-off for the various decomposition channels of fluorocyclopropane. Although there has been some criticism of the original data, it seems that the experiments are consistent with RRKM theory despite an earlier report to the contrary. Marta and co-~orkers'~~ have recently examined the fall-off behaviour of oxetane and [2,2-2Hz]oxetane and have refined earlier RRKM calculations for the former.'74 The technique of very low pressure pyrolysis (VLPP) has been extensively used especially by King and co-~orkers'~~ and the data obtained extrapolated by RRKM theory to obtain k, values.These data are reported in the appropriate tables in Section 7. Most of the data given in Table 10 refer to small molecules, the decomposi- tions of which are in, or close to, their second-order regions. Extrapolations can often be made with the aid of theory to obtain the rate constants kbim. Studies in which the prime objective has been to obtain vibrational energy transfer parameters are listed in Table 1, and some fall-off studies concerned with radical decompositions will be referred to in Section 9. 169 W. M.Lee and C. T. Yeh, J. Phys. Chem., 1979,83, 771. "O A. C. Baldwin, K. E. Lewis, and D. M.Golden, Int. J. Chem. Kinet., 1979, 11, 529. 17' I. M.Bailey and H. M.Frey, J.Chem. SOC., Faraday Trans, 1, 1981, 77, 709. 172 M.C. Flowers, Can. J. Chem., 1978, 56, 29. 173 L. Zalotai, Zs. Hunyadi-Zoltan, T. Berms, and F. Marta, Int. J. Chem. Kinet. Submitted for publication. 174 K. A. Holbrook and R.A. Scott, J. Chem. SOC.,Faraday Trans. I, 1975, 71, 1849. 17' See for example K. D. King and R. D. Goddard, Int. J. Chem. Kinet., 1981, 13, 755. Holbrook Table 9 High-pressure Arrhenius parameters for some bond-jssion reactions Reactant Product(s) log,,(A,/s-') E,/kJ mol-Re$ Carbon-carbon bonds C2H6 2Me 16.85 377.2 a 16.72 k0.17 371.7 k2.8 b i-C4HI, Pr' + Me 15.92 357.4 c CMe, But + Me 17.3 338.1 d Pr"CH2CH=CH2 CH,=-CHCH; + Pr" 15.9k0.2 296.2 k4.2 e 15.9 295.9 f CH2=CHCH=CHCH2Me CH2=CHCH=CHCH; + Me 15.92 k0.17 277.7 +3.5 g MeCH2 CI-CH HCLCCH; + Me 15.5i-0.3 310.4 k8.4 h MeC( Me)2CsCH HC=CCMe2 + Me 15.8 296.2 i PhCH2CH2CECH HCECtH, + PhCH2 14.6 i-0.4 252.3 k8.4 j MeC=CCHMe, MeCECcHMe + Me 16.2 k0.3 31 1.3 i-6.3 k MeC=CCMe3 MeC=CCMe2 + Me 16.4 k0.3 298.7 f6.3 k Pr"CH,C=CH CHECCH, + Pr" 15.9 ic0.3 295.8 k8.4 I PhCH2Me PhCH2 + Me 15.3 304.2 m 15.85 312.5 n PhCH2Et PhqH2 + Et 15.30 291.2 0 PhCHMe, PhCHMe + Me 15.8 298.3 m PhCH,Pr' PhCH, + Pr' 15.6 283.7 0 PhCMe, PhCMe, + Me 15.9 289.1 m PhCH2But PhCH, + Bu' 15.5 269.0 0 Ph (C6H4)CHz Me Ph(C,H,)CH, + Me 15.65 300.4 n Ph(C,H,),CH,Me Ph(C6H4)2CH,+ Me 15.60 278.2 n EtCMe,NM, Et + CMe2NH, 16.5 320.1 P o-C,H,NCH, Me o-C,H,NCH; + Me 15.0 307+12 qm-C,H,NCH,Me rn-C,H,NCH; + Me 15.0 300k11 q p-C,H,NCH, Me p-c,H,NCH; + Me 15.0 303k11 q Other C-X bonds CH2 CH2 II I', -MeCOMe Me-CL'O + Me 15.8 277.4 r PhCH20CH=CH, PhCH2 + %H,CHO 16.6 k0.3 224.8 k4.2 s MeCOCH,Br MeCOCH; + Br 16.0 261.5 r MeCl Me + C1 13.86 383 t SiMe, SiMe, + Me 14.1 k0.2 301.2f2.9 u GeMe, GeMe, + Me 15.1* 288.7* U 12.1** 213.4** u SnMe, SnMe, + Me 13.9k0.4 231.8 k5.9 u 17.3 291.6 d GaEt, GaEt, + Et 15.7 k0.2 194.6k2.4 L' t-CSHi iNH2 t-C,H,, + NHZ 15.9 330.1 P PhN(Me)H PhNH + Me 15.1 k0.3 279.1 W PhNMe, PhNMe + Me 15.1 k0.3 270.7 W MeNH, Me + NH, 10.84 201.5 X Current Aspects of Unimolecular Reactions Table 9-continued Reactant Product(s) log,,(A,/s-') E,/kJ mol-' Ref: Azo-compounds RN-NR' R + R' + N, R = R' = Et.15.8 kO.1 205.1f1.5 yR = R'= Pr' 16.2f0.3 196.8 f2.6 yR = R'= Ph 12.6 223.4+ 12 z D. B.Olson and W. C. Gardiner, Jr., J. Phys. Chem., 1979, 83, 922; see also D. B. Olson, T. Tanzawa, and W. C. Gardiner, Int. J. Chem. Kinet., 1979, 11, 23. 'A. B. Trenwith, J. Chem. SOC.,Faraday Trans. 1, 1979, 75, 614. 'G. Pratt and D. Rogers, J. Chem. SOC., Faraday Trans. 1, 1980, 76, 1694. (Wall-less reactor pyrolysis.) Ref. 170; A factors estimated. K. D. King, Int. J. Chem. Kinet., 1979, 11, 1071. W. Tsang Int. J. Chem. Kinet., 1978, 10, 1119. #A. B. Trenwith, J. Chem. SOC., Faraday Trans. 1, 1980, 76, 266. K. D. King, Int. J.-Chem. Kinet., 1978, 10, 545. ' K. D. King, Int. J. Chem. Kinet., 1977, 9, 907. K. D. King and T. T. Nguyen, J. Phvs. Chem., 1979, 83, 1940. 'K.D. King, Znt. J. Chem. Kinet., 1981, 13, 255. K. D. King, Int. J. Chem. Kinet., 1981, 13, 273. D. A. Robaugh and S. E. Stein, !nt. J. Chem. Kinet., 1981, 13, 445. " D. F. McMillen, P. L. Trevor, and D. M.Golden, J. Am. Chem. SOC., 1980, 102, 7400. E value for ethyl benzene assumed. A values for l-ethylnaphthalene and 9-ethylanthracene derived by adjustment of the value found for ethylbenzene. 'D. A. Robaugh, B. D. Barton, and S. E. Stein, J. Phys. Chem., 1981,85, 2378. A factor for neopentylbenzene estimated. W. Tsang, Int. J. Chem. Kinet., 1978, 10, 40. 4B. D. Barton and S. E. Stein, J. Chem. SOC., Faraday Trans. 1, 1981, 77, 1755. A factors assumed. 'F. Zabel, S. W. Benson, and D. M. Golden, Int. J. Chem. Kinet., 1978, 10,295. * M. Rossi and D.M. Golden, Int. J. Chem. Kinet., 1979, 11, 715. '0. Kondo, K. Saito, and I. Murakami, Bull. Chem. SOC. Jpn., 1980, 53, 2133. Extrapolated by RRKM from shock-tube data (See Table 10). " J. E. Taylor and T. S. Milazzo, J. Phys. Chem., 1978, 82, 847. *above 710°C; **below 710°C. " M. C. Paputa and S. J. W. Price, Can. J. Chem., 1979, 57, 3178. A. J. Colussi and S. W. Benson, Int. J. Chem. Kinet., 1978, 10, 1139. E. A. Dorko, N. R. Pchelkin, J. C. Wert 111, and G. W. Mueller, J.Phys. Chem.,1979,83,297. Extrapolated by H-L theory from shock-tube data (See Table 10). G. Acs, A. Peter, and P. Huhn, Int. J. Chpm. Kinet., 1980, 12, 992. A, Leiba and 1. Oref, J. Chem. SOC.,Faraday Trans. 1, 1979, 75, 2694. 9 Radical Decompositions (Table 11) A useful review has appeared listing free-radical reactions occurring during the low-temperature thermolysis of n-alkanes.76 Included are many recommended values of Arrhenius parameters for alkyl radical decompositions from a survey of the published data. These are too numerous to list in Table 11, which includes some more recently published data for alkyl radicals and also the extensive results of Batt and co-workers 77 on alkoxy-radical decompositions. The latter have been recently assessed by Choo and Benson.l7* Attention is drawn to the work on the t-butoxy-radical decomposition in which the fall-off curve has been obtained for the first time. A detailed study of the fall-off curve for the decomposition of the 1,2-dichloroethyl radical has been made by Ashmore, Owen, and Robins~n.'~' RRKM calculations incorporat- 176 D.L.Allara and R. Shaw, J. Phys. Chem. Ref: Data, 1980, 9, 523. 177 L. Batt, Int. J. Chem. Kinet., 1979, 11, 977, and refs. cited therein. 178 K. Y.Choo and S. W. Benson, Int. J. Chem. Kinet., 1981, 13, 833. P.G. Ashmore, A. J. Owen, and P. J. Robinson, J. Chem. SOC., Faraday Trans. 1, 1982,78, 677. 206 Table 10 Unimolecular reactions in the low pressure and fall-ofl regions Reactant Product(s) Kinetic data* Theory used Ref. MeCl (+ M) Me +C1(+ M) log kbim = 15.56 -24710 [M =Ar] RRKM Me1 (+ M) Me +I (+ M) log kbim = 15.40 -178.1/8 [M =Ar] CHd+M) Me +H (+M) log kbim = 17.00 -358.9/8 [M =Ar] CDd+M) CD3 +D (+M) log kbim= 16.32 -355.2/8 [M =Ar] RRKM HN3 (+M) NH3(%-) +N2 (+ M) log kbim = 13.77 -128.0/0 [M =Ar] RRKM MeNH, (+ M) Me+NH2(+M) log kbim= 13.50 -147.0/8 [M =Kr] H-L C2H6 (+M) 2Me (+M) log kbim = 11 1.29 -25.26 log T -667.818 RRK [M =C2H6 or Ar] * ‘bgkbi,’ ~log1,(kbi~cm3mol~’s~’~ 8= 19.147 x T/K.0. Kondo, K. Saito, and I. Murakami, Bull. Chem. SOC. Jpn., 1980, 53, 2133. (See also Table 9).* K. Saito, H. Tahara, 0. Kondo, T. Yokubo, T. Higashihara, and I. Murakami, Bull. Chem. SOC. Jpn., 1980, 53, 1335. (See also Table 9) ‘K. Tabayahi and S. H.Bauer, Combustion und Flume, 1979, 34, 63. dC. C. Chiang, J. A. Baker, and G. B. Skinner, J. Phys. Chem., 1980, 84, 939. ‘0. Kajimoto, T. Yamamoto, and T. Fueno, J. Phys. Chem., 1979, 83, 429. J. E. Dove and W. S.Nip, Can. J. Chem., 1979, 57, 689. D. B. Olson, T. Tanzawa, and W.C. Gardiner, Int. J. Chem. Kinet., 1979, 11, 23. Current Aspects of Unimolecular Reactions Table 11 High-pressure Arrhenius parameters for some decompositions of radicals Radical Product(s) log,,(A,/s-l) E,/kJ mol-Ref. Et C2H4 + H 13.5 f0.7 175 f12 a Bu' Me 4-C3H6 -136.5 f4.8 b Bu' i-C4HB+ H 14.67 164.8 c Bu'CH( Me)CH, Bu'CH-CH, + Me 13.8 f0.5 130 L-7 d Bu'CH( Me)CH, C3H, + Bu' 13.8 L-0.5 111 +7 d Me2C(CH2)Pr' CH2=C(Me)Pr' + Me 13.8 f0.5 121 +7 d Me, C( C H )Pr' Me2C=CH, + Pr' 13.8 f0.5 107 + 7 d Bu'CMe, i-C4H8 + Pr' 13.8 k0.5 114+7 d Me0 CH2O + H 14.2 & 0.5 112.3 f4.2 r EtO CH,O + Me 15.0f0.5 90.4 +4.2 e Pr"0 CH2O + Et 13.7 65.3 f Pr'O Me + MeCHO 14.6 f0.5 72.0 f4.2 e 13.8 70.3 f Bu'O CH,O + Pr' 13.7 51.9 f Bu'O Me,CO + Me 15.5f0.5 71.1 k4.2 e 14.1 64.0 f BuSO Et + MeCHO 14.9 f0.5 64.0 f4.2 e 13.6 56.5 f Pr' + MeCHO 13.7 43.1 f Me2C0 + Et 14.8 +0.5 57.7 L-4.2 e 13.6 51.9 f MeEtCO + Me 13.8 67.4 f ' Me,CO + Pr' 13.6 40.6 f C2HjCl+ C1 14.3 84.0 9 "G.Pratt and D. Rogers, J. Chem. Soc., Faraday Trans. I, 1979, 75, 1089. Deduced from ethane pyrolysis in wall-less reactor 941-1073 K. G. McKay and J. M. C. Turner, Int. J. Chem. Kinet., 1978, 10, 89. C. E. Canosa and R. M. Marshall, Int. J. Chem. Kinet., 1981, 13,303. R. R. Baldwin, R. W. Walker, and R. W. Walker, J. Chem. SOC., Faraday Trans. 1, 1981, 77, 2157. Ref. 177. f Ref. 178; A factors estimated. Activation energies derived from A factors and experimental rate-constants from ref.e and other sources. Ref. 179. ing a centrifugal effect are in good agreement with the experimental data of Ashmore and co-workers180 and of Huybrechts and co-workers.lgl 10 Radical Recombination and Addition Reactions (Table 12) The reverse of bond dissociation is the recombination of radicals to give molecules which is a process equally treated by unimolecular theory. The change in order for bond-dissociation from first to second at low pressures is paralleled by a change from second to third order for the corresponding radical recombination. If the process A+B + M+AB+ M P.G. Ashmore, J. W. Gardner, A. J. Owen, B. Smith, and P. R. Sutton, J. Chem. SOC., Faraday Trans. 1, 1982, 78, 657. G. Huybrechts, J. Katihabwa, G. Martens, M. Nejszaten, and J.Olbregts, Bull. SOC. Chim. Belg., 1972, 81, 65. 208 Table 12 Radical recombination and addition reactions React ion TIK Kinetic data* Comments and Ref: CH,02 + NO2 + M -,CH302N02+ M 298 krec,o = (4.32& 0.22) x 10" (M = He) Rate constants extrapolated by krec,o = (8.45& 0.29) x 10" (M = N2) approximate Troe method a krec.0 = (1.85& 0.05)X 1012 (M = SF6) krec,m= (4.82k0.60) x lo9 298 krec= (7.22 f1.80) x lo8 (50 Torr Ar/CH4) krec= (9.60f 1.80) x lo8 (540 Torr N2) b 2CF3 + M + C2F6+ M 1250-140 krec,m= (1.90Ifr 0.9) x 10" Rate constants extrapolated by krec.0 = 1.1 x 1014 (M = Ar) approximate Troe method 930-1110 krec,m= 5 x lo9 -5 x 10" Extrapolated from fall-off data VLPP conditions d F +NO + M-+ FNO + M 195-288 krec,O = (2.99& 0.18) x 10" (M = He) krec.0 = (4.5k4) x 10" exp( +0.88 kJ mol-'/RT) (M=NO) e OH + NO2 + N2 4 HN03 + N2 225-389 krec.0 = 8.34 x 10' ' (T/298)-2.9 Rate constants extrapolated by amroximate Troe method f Table 12-continued Reaction TIK Kinetic data* Comments and Ref.C10 + NO2 + M +ClON02 + M 298 k,, = 5.44 x 10" (M = N2) 274-339 krec= (1.56 k0.43) x lo9 (M = N2) Measured at 50 Torr 0 + NO + M +NO2 + M 217-500 krec,o = 2.80 x lOI4 (M = He) krec,o = 1.00 x 1015 T-1.86 (M = Ar) Application of Troe Theory krec,o = 1.38 x 1015 T -1.82 (M= N2) C1+ NO + M +NOCl + M 200-400 krec.0 = 1.39 x lOI5T-'.'' (M = N2) 298 kreco = (1.49k0.07)x 10" (M = He) Application of Troe Theory krec,o = (1.59k0.10) x 10" (M = Ar) H + CO + M+ HCO + M 298 krec,o= (3.6k0.9) x lo7 (M = CO) krec,o = (5.8kO.6) x lo7 (M = CH4) krec,o= (3.8 k0.4) x lo7 (M = H2) * Units of k,,,,, and k,,, are dm3 rnol-'s-'.Units of krcc,,,are drn6 rnol-'s-'. 'S. P. Sanders and R. T. Watson, J. Phys. Chem., 1980, 84, 1664. R. A. Cox and G. S. Tyndall, Chem. Phys. Lett., 1979, 65, 357. K. Glanzer, M. Maier, and J. Troe, J. Phys. Chem., 1980, 84, 1681. 'M. Rossi and D. M. Golden, Int. J. Chem. Kinet., 1979, 11, 775. P. Kim. D. I. McLean, and W. G. Valance, J. Phys. Chem.. 1980, 84. 1806. L. G. Anderson, J. Phys. Chem., 1980, 84, 2152. M. J. Molina, L. T. Molina, and T. Ishiwata, J. Phys. Chem.. 1980, 84, 3100. " R. A. Cox and R. Lewis, J. Chem. Soc., Furaday Trans. I, 1979, 75, 2649. Ref. 182. j C. J. Hochanadel, T. J. Sworski, and P.J. Ogren, J. Phys. Chem., 1980, 84, 231. Holbrook is described by the second-order rate constant krec defined by then the limiting values of krec are krec,m at high pressures which is independent of [MI, and krec,O at low pressures which is given by krec.0 = krec[M] At intermediate pressures (the fall-off region), krec can be expressed in terms of krec,o, krec,, and a broadening factor Fc which expresses the deviation of the fall-off curve from the Lindemann-Hinshelwood expression. This convention which is due to Tr~e~~ has been adopted in a compilation of rate data prepared for the CODATA Task Group on chemical kinetics and which includes data on recombination reactions involved in atmospheric chemistry published up to December 1978? Table 12 contains some more recent data and comments on radical recombination and addition reactions, particularly those in which some unimolecular theory calculations have been applied.A particularly comprehensive study is that of Michael and Lee'82 for the addition reactions of 0 and C1 atoms with NO which is reported in the Table. Values of the collisional deactivation efficiency Bc were derived by comparison of krec with the theoretical strong-collision values calculated by the Troe theory. Some anomalies remain concerning the temperature dependence of Bc, but in general it was possible to reconcile both dissociation and recombination data on these reactions with the theoretical calculations. J. V. Michael and J. H. Lee, J. Phys. Chem., 1979,83, 10. 211

ISSN:0306-0012    

DOI:10.1039/CS9831200163

出版商:RSC    

年代:1983

数据来源: RSC

摘要:

Anionic Cyclization of Phenols By William S. Murphy* and Sompong Wattanasin? DEPARTMENT OF CHEMISTRY, UNIVERSITY COLLEGE, CORK, IRELAND 1 Introduction This review is concerned with those cyclizations of phenols that are catalysed by base and in which a new carbon to carbon bond is formed. In 1957 Winstein and Baird' demonstrated that under basic conditions suitably substituted phenols undergo intramolecular geminal cyclization via participation of the neighbouring phenoxide ion group to form dienones, e.g. (2) and (4).2-4 Since then this intra- molecular reaction of phenolic compounds has been extensively studied from both mechanistic and synthetic standpoints. This reaction has been extended to the synthesis of fused products, e.g. (5)-+(6)' and has been used to gain understanding OH 0 (3) (4) OH Br03 @BU~OHButOK) (5) (6) * To whom correspondence should be addressed.7' Present address: Sandoz, Inc., Route 10, East Hanover, N.J. 07936, U.S.A. ' S. Winstein and R. Baird, J. Am. Chem. SOC.,1957, 79, 756. R. Baird and S.Winstein, J. Am. Chem. SOC., 1957, 79, 4238. R. Baird and S. Winstein, J. Am. Chem. SOC., 1962, 84, 788. R.Baird and S. Winstein, J. Am. Chem. SOC.,1963,85, 567. 'M.S.Newman and A. B. Mekler, J. Org. Chem., 1961, 26, 336. Anionic Cyclization of Phenols 0 0 0 (10) 0 0 OH C02Me 0 07)roQo+ 0 OH OH 0 2 14 Murphy and Wattanasin of ortho-para-alkylation ratios, reactivity of leaving groups, and the effect of the nature of the chain involved in the ring being formed.A review has covered aspects of these topics to a limited extent.6 Intramolecular alkylation of phenoxide ions has found widespread application to the synthesis of theoretically interesting compounds such as the tetracyclic dienone (7),' the adamantanoid dienones (8),' and the bridged bis-dienone (9),' as well as to the synthesis of natural products such as cedrol (lo),'*lo hinesol (1 l),' p-vetivone (12),12 kaurene (13),13 garryine (14),14 atisine (15),15 the anthra- cyclinone (16), and conicaquinone (17). It is the purpose of this review to discuss the known intramolecular alkylations of phenoxide ions. Discussion will include base-catalysed cyclization reactions of phenolic epoxides, aldehydes, ketones, and related functional groups.An emphasis has been placed on their scope, limitations, and applications in natural product synthesis. Detailed mechanistic studies have not been undertaken in most systems. 2 Mechanism Intramolecular alkylation of phenoxide ions are conveniently classified according to transition ~tates:~ Ar, -n and Ar, -n (Scheme 1). Ar- denotes the partici- pating (rate enhancing) phenoxide ion. The subscript, 1or 2, refers to the position ortho para ortho para Arl--n Arz- -n Scheme 1 of ring closure, and n to the size of the ring formed. Only one product is possible in the course of an Ar, -n cyclization, whereas two regio-isomers are possible in the course of Ar; -n cyclization. It should be noted that these symbols are used to B.Capon, Quart. Rev., 1964,18,45. A. P. Krapcho, Synthesis, 1974,383. R. S. Atkinson and J. E. Miller, J. Chem. SOC.,Perkin Trans. 1, 1979, 3017. E. J. Corey, N. N. Gikotra, and D. T. Mathew, J. Am. Chem. SOC., 1969,91, 1557. lo T. G. Crandall and R. G. Lawton, J. Am. Chem. SOC., 1969,91,2127. l1 J. A. Marshall and S. F. Brady, J. Org. Chem., 1970,35,4068. S. Torii, K. Uneyama, and K. Okamoto, Bull. SOC.Chem. Jpn., 1978,51,3590. l3 S. Masamune, J. Am. Chem. SOC.,1964,86,289. l4 S. Masamune, J. Am. Chem. SOC., 1964, 86, 290. S. Masamune, J. Am. Chem. SOC., 1964,86, 291. l6 K. Krohn, J. Chem. Res. (S),1978,394. l7 J. K. MacLeod, B. R. Worth, and R. J. Wells, Aust. J. Chem., 1978,31, 1533. Anionic Cyclization of Phenols describe the reaction throughout, although the mechanism frequently has not been established.A. Arl -n Cyclizationa-In Ar, -n cyclizations (Scheme 2), intramolecular displacement of the leaving group by the phenolate anion leads to spirodienones. This has been proved spectrophotometrically'~4and by product isolation.' -4*8-Scheme 2 + 0 _fjL) 8U Murphy and Wuttanusin Kinetic data which confirm Ar; -3 and Ar; -5 participation have been provided by Winstein and Baird.'~~*~ le Noble and GabrielsenI8 confirmed both the Ar; -3 and Ar, -5 mechanisms by studying the effect of pressure on the isopropanolysis of the phenols (1) and (3) under basic conditions. Phenoxide participation resulted in rate enhancements of lo6 and 50, respectively.The small effect of hydrostatic pressurel8I' on the rate constants of the solvolysis of (1) and (3) readily distinguished between these concerted processes and mechanisms involving carbo-cations. A detailed examination of phenoxide (18) led to the conclusion that Ar, -4 participation was ineffective compared to Ar; -3 and Ar; -5processes. Other workers20*21 investigating the phenoxides (20) and (22) arrived at the same conclusion as le Noble." 0-0-c1 -C02 H A stereochemical study also lent support to the Ar; -3 mechanism. Deaminative bromination, followed by amination, is a standard method of inverting the configuration of cr-amino-acids.22 However, when applied to 3,5-dichloro-L-tyrosine (26) net retention was observed.23 Koga and co-w~rkers~~ proved that amination of the intermediate (S)-( -)-a-bromo-b-phenylpropionic acid (24)with aqueous ammonium hydroxide occurred with complete retention of configuration. These results are consistent with an Ar, -3 mechanism and an intermediate spirodienone (25).A mechanism involving an a-lactone' was dis- proved in this instance. B. Ari -nCyc1izations.--In Ar; -n cyclization (Scheme 3), intramolecular dis- placement of a leaving group by phenolate anion participation at both the ortho and para ring-positions leads to dienone intermediates. These tautomerize spontaneously, under the reaction conditions, to the more stable phenols when R = H. Thus Ar, -n cyclization gives both ortho and para alkylation products. This mode of cyclization has not been investigated as extensively as Arc -n cyclizations.The first two examples of Ar; -6 cyclization appeared in the same W. J. le Noble and B. Gabrielsen, Tetrahedron Lett., 1971, 3417. l9 W. J. le Noble and B. Gabrielsen, Tetrahedron Lett., 1970, 45. 2o S. Dorling and J. Harley-Mason, Chem. Ind., 1959, 1551. 21 W. S. Murphy and K. P. Raman, unpublished results. 22 N. Izumiya, Bull. Chem. SOC.Jpn., 1951, 72, 26. 23 W. K. Warburton, J. Chem. Soc., 1961,2651. 24 K. Koga, T. M. Juang, and S. Yamada, Chem. Pharm. Bull., 1978,26, 178. 25 P. Brewster, F. Hiron, E. D. Hughes, C. K. Ingold, and P. A. D. S. Rao, Nature. 1950, 166, 179. 217 Anionic Cyclizution of Phenols @+R=H \HO ‘ R R ortho-productbse / X para-product Scheme 3 year.Mande1126 noted that the phenol (27) cyclized to give the two dienones (28) and (29) in the ratio 2 : 1. Only one product (6) was formed when the phenol (5) was treated with base.5 Later, a kinetic investigation2’ of the cyclization of phenol HO OTs 0 (27) (28) (29) (30) revealed that it cyclized within 10 hours under reflux in t-butyl alcohol- tetrahydrofuran with potassium t-butoxide. The phenols (31) and (32) were formed in the ratio 3 : 1. The benzyl ether of (30) did not react under these conditions. A detailed study of the effect of pH on the rate of cyclization confirmed an Ar; -6 mechanism. The rate of cyclization increased 400-fold when the pH was increased from 8 to 12. 93-p HO OTs HO (3 0) (31) (32) 3 Synthetic Applications Anionic cyclization of phenols, first reported by Winstein and Baird,’ -has been applied to the synthesis of a wide range of novel structures, for example, the spiro- dienones (7),’ (9),* and (36).7 26 L.Mandell, D. Caine, and G. E. Kilpatrick, J. Am, Chem. SOC., 1961,84,4457. 2’ P. G. Duggan and W. S. Murphy, J. Chem. SOC.,Perkin Trans. 2, 1975, 1054. 218 Murphy and Wattanasin Br Br (331 BrHoaqeBr \/ 011 (34) OH$\0tl 01I (35) (30) This method has been used to generate compounds containing the bicyclo- [2.2.l]heptane system, e.g. (38),28 and bicyclo[2.2.2]octane systems, e.g. (40).29 OH 0 (3 7) R. Barner, A. S.Dreiding, and H. Schmid, Chem. Ind., 1958, 1437. 29 D.J. Beames and L. M. Mander, Aust. J. Chem., 1971,24,343. Anionic Cyclization of Phenols THPO 0BoT* HO (39) (41) In search of a general route to natural products containing the bicyclo- [3.2.lloctane unit, e.g. phyllocladene (43), Masamune initiated an outstanding series of papers with a study of the base catalysed Arc -5 cyclization of phenols (41)30 which culminated in the formal synthesis of kaurene (13),13 garryine (14),14 and atisine (19.' The key intermediate leading to (13)-( 19, dl-16-keto-10- carboxy- 17,20-bisnorkaurane (44),was derived from (45) using standard methods. The latter was synthesized by an Ar; -5 cyclization of the phenol (46). (44) (45) (46) The neolignan, futoenone (48) was synthesized by Ogiso and his group3 1932 by a base catalysed Ar, -6 cyclization of the phenol (47).30 S. Masamune, J. Am. Chem. Soc., 1961,83, 1009. 31 A. Ogiso, M. Kurabayashi, M. Mishima, and M. C. Woods, Tetrahedron Lett., 1968, 2003. 32 A. Ogiso, M. Kurabayashi, S. Takahashi, H. Mishima, and M. C. Woods, Chem. Pharm. Bull., 1970, 18, 105. Murphy and Wattanasin (:%-*;Me0 \I Me0 ' LOOH (47) (48) Both Corey' and Lawton" and co-workers synthesized cedrol (10) from the common intermediate (51). Corey also synthesized cedrene (52) from this intermediate. Marshall and Brady' 'used Masamune's dienone (53)30in the total synthesis of OH OH Br EtO2C Br C02 Me 221 Anionic Cyclization of Phenols ( k )-hineso! (1 1). They proved, for the first time, the relative stereochemistries of all chiral centres, in the course of their synthesis.Ogiso and co-~orkers~~ prepared 6,10-dimethylspir0[4.5]dec-6-en-2-one(56), by utilizing Arr-cyclization (54455). It was intended to apply this mode of cyclization to the synthesis of spirosesquiterpenoids, e.g. hinesol (11) and /?-vetivone(12).A related approach to /?-vetivone was undertaken in the same year with the successful Arc -5 cyclization of the phenol (57) to the spirodienone (58).34 Torii and co-~orkers,~ however, finally accomplished the synthesis of /?-vetivone (12) using this approach by way of the spirodienone (60). (57) (58) OH 0 0 0 t 33 A. Ogiso, M. Kurabayashi, H. Nagahori, and H. Mishima, Chem. Pharm. Bull., 1970, 18, 1283.34 P. C. Mukharji and P. K. Sen Gupta, Chem. Ind., 1970, 533. 35 S. Torii, K. Uneyama, and K. Okamoto, Bull. Chem. SOC.Jpn., 1978, 51, 3590. 222 Murphy and Wattanasin HOgQ HOa A recent synthesis of pseudoclovene-B (66) is of particular interest.36 The dienone (65) was employed as the key intermediate. Initial attempts centred on the cyclization of the phenol (63). A low yield of (65) was observed. In contrast, cyclization of (64)was effective.* The successful synthesis of the left-hand segment (69) of the anti-tumour agent CC-1065(67)employed the Ar, -3 cyclization ofthe phenolic bromide (68) as the key step.37 This is the first natural product synthesis which has employed an Ar; -3 reaction. r. t., 7070 *Note added in proof: subsequent to the submission of the ms it was noted (J.D. McChesney and R. A. Swanson, J. Org. Chem., 1982,47, 5201) that phenol (i) cyclized to dienone (ii) in 57 % yield when heated with potasslum t-Dutoxide in t-butyl alcohol. HO 36 S. Chatterjee, A. Sarkar, and P. C. Dutta, J. Chem. SOC.,Perkin Trans. 1, 1979, 2914. 37 W. Wierenga, J. Am. Chem. SOC.,1981,103,5621. Anionic Cyclization of Phenols Asymmetric induction was observed38 in the course of Ar; -6 cyclization of phenol (70)using the chiral (+)-camphor-10-sulphonate leaving group (X). The degree of induction depended on the solvent and the metal cation. A maximum of 19%e.e. was obtained for (71) and 13% e.e. for (72). The results were rationalized in terms of diastereomeric transition states involving a co-ordinated metal cation.The predominance of the (Stenantiomer of both (71)and (72)meant that attack on the ortho-and para-positions of (70) occurred predominantly on one prochiral face of the phenoxide ring. Contrary to a recent claim,39 this is the first example of asymmetric induction by a chiral leaving-group during nucleophilic substitution at a saturated carbon centre. .Qc!-@+f& (+) -(70) (S)-(+)-(71) (S)-( ) -(72) Cyclization of phenoxyketones to benzofurans4’ were first observed some 50 years ago. Results were erratic4’ and the reaction unsati~factory~~ until MacLeod and co-workers investigated the area in detail. They applied this cyclization to the synthesis of a wide range of benzofuranoid natural products.The naturally occurring coumarin geiparvarin (73) was found to generate the linear furocoumarin psoralene (74) when treated with aqueous base. The postu- lated intermediate (75) was synthesized by Ma~Leod~~ and when similarly treated, 0 n (75) (76) 38 P. G. Duggan and W. S. Murphy, J. Chem. SOC.,Perkin Trans. 1, 1976,634. 39 J. M. Wilson and D. J. Cram, J. Am. Chem. SOC., 1982, 104, 881; see also Chem. Eng. News, 1982, Feb. 22, p. 34. 40 J. K. MacLeod and B. R. Worth, Tetrahedron Lett., 1972, 237. ”J. N. Ray, S. S. Silooja, and V. R. Vaid, J. Chem. SOC.,1935, 813. 42 R. C. Esse and B. E. Christensen, J. Org. Chem., 1960,25, 1565. F. N. Lahey and J. K. MacLeod, Aust. J. Chem., 1967,u), 1943. Murphy and Wattanasin was converted into psoralene. This conversion, considered to involve an Ar; -5 cyclization of the intermediate (76),was proved by analogy.Thus, when the mono- acetonyl ether of resorcinol was treated with hot aqueous base, 6-hydroxy-3- methylbenzofuran was formed exclu~ively.~~ The Ari -5 mechanism was further evidenced by the fact that the corresponding methyl ether did not react under these conditions. This strikingly facile and regiospecific reaction was then applied to synthesis. The dibenzofuran (79) was synthesized4' by base-catalysed Ar, -5 cyclization of (77) followed by DDQ dehydrogenation. (77) (78) The structure (80), suggested for scabequinone, was confirmed by total ~ynthesis'~via the key intermediate (81). This intermediate was also used in the total synthesis of both cyperaquinone (82)and conicaquinone (17).The synthesis of the intermediate (81)involved the regiospecific Ar; -5 cyclization of the phloro- glucinol derivative (84)to the benzofuran (85).0 Br R' K2 (82) Me CHl (17) Me 0 R' 0 (83) H CH2 225 Anionic Cyclization of Phenols -----+-(81) 0x-g"" joyoH\ OMe OMe (84) (85) An alternative synthesis of (82), (17), and (83) from daphnetin (86) was achieved.44 The introduction of the furan ring was accomplished by Ar; -5 cyclization, (87)+ (88). OH OH OH (87) -0 NMe HO b (89) (90) Me0 HO 44 .I.K. MacLmd, B. R.Worth, and R. J. Wells, Aust. J. Chern., 1978,31, 1545. Murphy and Wattanasin Phenoxide cyclizations have also been involved in a number of biomimetic syntheses and there is some evidence for their involvement in biosynthesis.A biogenetically patterned synthesis of ( k)-cherryline (91), a unique Amaryllidaceae alkaloid, was completed by Schwartz and The key cyclization step was considered to involve Ar; -6 phenoxide-quinone methide coupling. Thus, phenol (89) was converted into cherryline in 79% yield by refluxing in aqueous ammonium hydroxide. However, there remains some doubt about whether the cyclization was acid or base catalysed. This in part was due to the high reactivity of phenol (89). In addition, the fact that none of the regio-isomer (92) was detected must be compared with the phenoxide-quinone methide coupling of (93) where both the ortho-and para-cyclization products (94) and (95) were detected.46 Anionic cyclization of phenols has also been applied to the synthesis of iso-quinoline alkaloids, e.g.petaline (96).47 (96) 45 M. A. Schwartz and S. W. Scott, J. Org. Chem., 1971,36, 1827. 46 W. S. Murphy and S. Wattanasin, J. Chem. SOC.,Perkin Trans. I, 1980, 1567. 47 T. Karnetani, T. Kobari, K. Fukumoto, and M. Fujihara, J. Chem. SOC., (C), 1971, 1796 227 Anionic Cyclization of Phenols 4 Ring Size The case of cyclization and the extent of anchimeric assistance6 by phenoxide depends on the size of the ring being formed and whether the reaction involves an Ar, -n or an Ar; -n transition state. The anchimeric assistance6 in Ar, -3 reactions is considerably greater than that in Ar; -5 reactions. Compound (97; R = H) has been obtained4' only by 0 heating the potassium salt of 5-(p-hydroxyphenyl)pentyl bromide to 170 "C in t-butyl alcohol.This suggests that phenoxide participation in Ar 1 -6 reactions is weaker than in Ar; -5 reactions. However, anomalies exist. The dienone (97; R = But)was isolated in 98%yield by heating the anion of the corresponding tosylate in t-butyl alcohol at reflux temperat~re,~' whereas the dienone (38) was obtained in 8% yield even though vigorous conditions were employed. The dienone (99)has been prepared in 40 "/, yield by heating (98)with potassium t-butoxide in t-butyl alcohol at 180"C.20It is of particular note that no spiro- dienone was obtained when the higher and lower homologues of (98) were subjected to the above conditions.20 Thus, although certain ambiguities still remain, it can be concluded from these results that the relative rates of ring closure in Ar, -n reactions decrease in the order 3 > 5 > 6 $4.Phenol (100) was investigated as a model for Ar, -5 cyclization. Reactions were studied in a variety of bases and solvents at varying temperatures. Cyclization of (loo),to give (101)or (102), was not ob~erved.~' The main product was the dimer (103). A polymer was also occasionally formed. Thus, Ar; -5 participation is an inefficient process with the result that intermolecular reactions compete success- 48 A. S. Dreiding, Helv. Chim. Acta, 1957,40, 1812. 49 J. D. McClure, J.Org. Chem., 1962, 27, 2365. Murphy and Wattanasin OH OTs fully. However, base catalysed Ar; -5 cyclization was observed36 in the case of the phenol (63).The dienone (65)was formed in 11%yield. Although not rigorously proved, the mechanism was almost certainly an example of Ar; -5 cyclization. In the light of these results, the base catalysed Ar; -5 cyclization of the phenoxides derived from the aldehydes and ketones (104), extensively studied by MacLeod and co-w~rkers~~from a synthetic view-po;tlt (Section 3), are of particular note. In all cases, base catalysed intramolecular cyclization gave para products, in good yield under mild conditions. No products corresponding to the possible ortho cyclization were observed (see Section 8).It is clear from a comparison of the reactions of (30), (63), and (100) that the relative rates of ring closure in Ar; -n reactions decrease in the order 6 3 5. Examples of Ar; -3, Ar; -4, and Ar; -7 are not known. Kineticdata for Ar, -n and Ar; -n cyclizations are summarized in the Table (columns C and D) together with related anionic cyclizations for comparison (columns A and B). The ratio k3/k, is normally greater than unity.50 This is observed in Ar, -n reactions and, in the Ar, -3 system, is due5' to the overlap C. J. M.Stirling, Angew. Chem., Znt. Edn. Engl., 1968, 7, 648. 51 C. J. M.Stirling, J. Chem. Educ., 1973,50,844. Anionic Cyclization of Phenols Table Relative rates of cyclization by nucleophilic displacement reactions Aa D 0-0-OT Ring Size Rel.rate Rel. rate Rel. rate Rel. ratee 50 "C 50 "C 25 "C 50 "C 21.7 -1100' 2.4 x 104 2.8 x lo6 2.6 x 104 2.63 x lo6 1.32 x 105 1.O' < 1.Od slowf'g-0.7' C. Galli, G. Illuminati, L. Mandolini, and P. Tamborra, J. Am. Chem. SOC.,1977,99,2591.* G. Illuminati, L. Mandolini, and B. Masci, J. Am. Chem. Soc., 1975, 97, 4960. S. Winstein and R. Baird, J. Am. Chem. SOC., 1957, 79, 756. A. S. Dreiding, Helv. Chim. Acta, 1957, 40, 1812. Rates relative to those in column C. S. Chatterjee, A. .Sarkar, and P. C. Dutta, J. Chem. SOC., Perkin Trans. I, 1979, 2914. P. G. Duggan and W. S. Murphy, J. Chem. SOC., Perkin Trans. 2, 1975, 1054. of the It-orbitals of the aryl ring and the distorted high p-character ring-bonds of the three-membered ring system which lowers the free energy of the transition state.The ratio k5/k, is also usually greater than unity5' (compare columns A, B, and C). It is of note that this is not the case in Ar, -n reactions. This contrast between Ar, -n and Ar, -n reactions is analogous to that observed by Heck and Winstein5, during an investigation of the solvolysis of w-phenylalkyl brosylates. They observed the following: Ar, -5 > Ar, -6 and Ar, -6 > Ar, -5. This comparison suggests that Ar- -n and Ar -n53,54 reactions have transition states of similar geometry. Whereas a measure of the degree of phenyl participation has been determined in Ar, and Ar, reactions from the effects of ring s~bstitution,~~ no comparable studies of phenoxide cyclizations have been undertaken.5 Leaving Groups; Electrophilic Centre Developments of new leaving groups or new functional groups initiating cyclization are relatively few.,' To date the only functional groups used success- fully are sulphonates (OTs and OBS),'~~~ br~mides,~*''*~' chlorides,20 aldehyde^,^'.^^ and ketone^.^' The last two cases are in rather specific systern~.~~~~~ 52 R. Heck and S. Winstein, J. Am. Chem. SOC., 1957, 79, 3114. 53 L. M. Jackman and V. R. Haddon, J. Am. Chem. SOC., 1974,%, 5130. 54 M. Gates, D. L. Frank, and W. C. von Felton, J. Am. Chem. SOC.,1974,%, 5138 55 P. G. Duggan and W. S. Murphy, Chem. Comm., 1972,770. 56 B. Rickborn and M. T. Wuesthoff, J. Am. Chem. SOC., 1970,92,6894.57 K. H. Bell, Tetrahedron Lett., 1968, 3979. 230 Murphy and Wattanasin Forcing conditions are normally required to effect the cyclization of chlorides. 2o Interestingly, whereas cyclization involving Ar; -3 participation of the epoxide (106)58is observed, the others, (109)13and (l1O),l2 do not cyclize under similar basic conditions. However, these results may be attributed to stereoelectronic effects (see Section 6) and to the more effective aryl participation in Ar; -3 reactions than in Ar; -4 or Ar; -5 reactions.6 The phenolic oxetane (111) is stable under basic conditions.12 OH One example of Ar ;-3 participation involving a phenoxide leaving-group has been observed59 in aqueous base at 170°C [(112)+(114)]. Other electrophilic sites have been used in Ar; -type cyclizations.For example, the iminium group functions as electron sink in the base catalysed varient of the Pictet-Spengler reaction. Thus, epinephrine (115) reacts with either formaldehyde (R = H) or acetaldehyde (R = Me) under basic conditions.60 In addition, both ortho-(1 19) (n = 1 or 2) and para-quinone methides (93) and (120)46as well as the benzyne (121)6’ undergo cyclization in basic conditions. 6 StereoelectronicFactors Stereoelectronic factors are particularly significant in cyclization at aromatic centre^.^^^' This is clear from a consideration of the probable transition states involved. In such reactions, perpendicular approach to the aromatic ring is con- sidered ~ornmon.’~ However, the actual angle depends on the extent of aryl participation.’ In addition, strict SN2stereochemistry at the site of the leaving J.Meinwald, H. Nozoki, and G. A. Wiley, J. Am. Chem. SOC.,1957,79, 5579. 59 J. Gierer and I. Pettersson, Con. J. Chem., 1977, 55, 593. 6o H. A. Bates, J. Org. Chem., 1981,46,4931. 61 D. H. Hey, J. A. Leonard, and C. W. Rees,J. Chem. Soc., 1963,5266. 231 Anionic Cyclization of Phenols -I Li OMe OMe 0- 0- (1 13)I OH (1 14) OH r OH 1 OH OH HO @Me + Ho*NMcHO OH R R 232 Murphy and Wattanasin group is assumed. The transition states for Ar; -6 and Ar; -5 participation leading to para-alkylation of the phenolic sulphonates (30) and (100) can be represented by (122) and (123) re~pectively.~~ It is clear from inspection of models of these transition states that there is little, if any, strain in (122) but that the transition state (123)for the formation of a five-membered ring is highly strained.It (1 22) (I 23) is of interest also that the transition state for Ar; -5 participation (124)(' and Ar, -6 (125) are relatively strainless. It is probable that Ar; -6 cyclization is slower than Ar, -5 for entropy reasons.6 These stereoelectronic requirements can be used to explain the failure of the phenol (109),13(126),13 and (127) to cyclize under basic conditions. HO I0 Br (126) Anionic Cyclization of Phenols These considerations form a basis for understanding the facility with which the phenol (128)cyclizes under basic conditions.Applying the conclusions of Burgi and co-workers62 on the stereochemistry of reaction paths at carbonyl centres, a transition state (129)can be envisaged which is not as strained as (123).An alter- native rationale for the difference in reactivity between (123)and (128)is that the SN2component in (123)makes more stereoelectronic demands than nucleophilic attack at the carbonyl centre in (129). In order to explain the failure of the enone (130)to cyclize under basic con- ditions it was concluded63 that the stereoelectronic requirements of the transition states were nucleophilic attack by the phenoxide ring along a trajectory such as that suggested by Biirgi62 and Bald~in,~~ coupled with co-planarity of the enone functional group.From molecular models it was concluded that whereas this transition state was highly strained, higher homologues could cyclize, in the absence of over-riding entropy factors.63 0 7 Steric Effects Subtle steric effects have frequently resulted in dramatic changes in yield. For example, the keto-phenol (126) failedI3 to cyclize whereas the corresponding phenol (41)was converted into dienone (42).30It is likely that entropy losses inhibited Ar, -5 cyclization of the trans-phenol (37) to the dienone (38).28The latter was formed in 8% yield. 62 H. B. Biirgi, J. D. Dunitz, J. M. Lehn, and G. Wipff, Tetrahedron, 1974,30, 1563. 63 W. S. Murphy and S. Wattanasin, J. Chem. SOC.,Perkin Trans. 1, 1980, 1555. 64 J.E. Baldwin, J. Chem. SOC.,Chem. Commun., 1976, 734. 234 Murphy and Wattanasin 011 (37) The ethylene dioxy-group when located c1 to the carbon bearing the leaving group confers neopentyl-like steric requirements and strongly represses reactivity. Attempts to cyclize phenol (131) failed.13 Cyclization of (54) required pyrolysis with potassium t-butoxide.20 Similar conditions were required to effect cyclization of phenol (98).20 The homologous phenol (132) did not cyclize.20 The steric requirements in this series, therefore, are very stringent. 011 OH (131) (132) The gem-dimethyl effect, whereby the presence of a gem-dimethyl substituted carbon in the chain facilitates cyclization, has been discussed elsewhere.6 Although steric in origin, no explanation is universally accepted.65- 67 This effect, however, does not appear to operate in Ar, -5 cyclizations.Thus the phenol (3; R = Me) cyclized in 25% yield6* whereas the analogous phenol (3; R = H) cyclized in 50% yield. However, further study will be required to confirm this point. It is well established that an oxygen heteroatom in a chain facilitates cyclization due to the absence of gauche interactions or transannular hydrogen rep~lsions.~~ The notable facility of the Ar; -5cyclization of the keto-phenol (128)40 is, at least in part, a result of this effect. The bicyclic phenol (41) cyclized, under the conditions employed by Winstein and Baird, to the dienone (60) in 90% yield.30 Masamune3O considered that the efficiency of this reaction was due to an advantageous orientation of the carbon atom carrying the tosylate group and, in addition, that side reactions were 65 M.Harfenist and E. Thorn, J. Org. Chem., 1972, 37, 841. 66 R. T. Borchardt and L. A. Cohen, J. Am. Chem. SOC.,1972,94,9175. 67 R. T. Borchardt and L. A. Cohen, J. Am. Chem. SOC., 1972, 94,9166. 68 D. I. Schuster and W. V. Curran, J. Org. Chem., 1970,354192. 69 B. M. Trost and T. R. Verhoeven, J. Am. Chem. SOC.,1980, 102,4743 and references therein. 235 Anionic Cyclization of Phenols suppressed by the tertiary carbon atom C-6, a structural feature not present in phenols such as (3). However, the dienone (45) was not formed in high yield when phenol (46), as a mixture of diastereomers (a-and 8-),was treated with potassium t-b~toxide.'~Phenol (468) was recovered unchanged. Models (133) and (134) respectively, show that the ethereal oxygen atom of the uncyclized phenol (468) in the transition state (133) interacts severely with a hydrogen atom at position-8.Failure of the ethylene glycolate (131) to cyclize13 can be explained on the same basis. flH HiTHp0 \OTHP* / HO \ Br HO \ Br (467.) i46P) THP OTHP -0' -0-Beames and Mander2' investigated the synthesis of the related dienones (135) and (136). In contrast, Ar, -6 cyclization of a 1 : 1 mixture of diastereomeric* phenols (137) gave a 3 : 1 mixture of dienone ethers (135) and (136) in 50% yield. The yield was not improved by variations in time, temperature, and solvent.THPO ,OTHP HO $'' 0 The transition state for Ar, -6 cyclization of the bromide mixture (137) requires both diastereomers (138) to have the side-chain ether substituent eclipsed with a ring methylene group. This interaction is intermediate in severity between * Diastereoisomerism due to the tetrahydropyranyl (THP)group is not considered. Murphy and Wattana, that for (4601) and (468). The smaller yield of epimer (136) is due to the 1c important, but extra, 1,Cnon-bonded interaction between the ether function ai the syn C-10 hydrogen atom in the transition state leading to this isomer.29 R' R* - OTHP H - (135) H OTHP (136) -6 8 ortho-para Ratio Discussion of the ortho-para ratio is applicable only to Ar; -n reactions.Of the two transition states, that leading to para-substitution is a priori favoured on steric grounds. Whereas simple Hiickel theory predicts equal electron density on the ortho as on the para position, frontier electron density is higher at the para p~sition.~'That these theoretical predictions have had only limited success is to be expected since steric effects, the effect of solvent, metal cations, and temperature variations are also involved. The phenol (70) cyclized at both the ortho- (71) and para- (72) positions in a ratio CQ. 4 : l.38MandellZ6 had noted that the disubstituted phenol (27) also cyclized with an ortho :pura ratio 2 : 1, under basic conditions. When the para-position only was substituted, as in the case of phenol (9,only ortho-alkylation occ~rred.~ None of the dienone (139) was detected.Similarly, only the phenol (142) was 063 (1 39) obtained when the dienone (140) was treated with methanolic sodium hydr~xide.~' A number of anohalies exist. For example, the para-dienone (144) was isolated in 0$-yJ[-oat]* d HO ( 140) 041) (142) 70 I. Fleming, 'Frontier Orbitals and Organic Chemical Reactions', Wiley, London,1976, p. 63. K. H.Bell, Ausr. J. Chem., 1972, 25, 1117. 237 Anionic Cyclization of Phenols 12% yield when the phenol (143) was treated with potassium t-butoxide. Apparently none of the ortho-dienone (145) was detected.72 (143) (1 44)-(145) A detailed investigation of factors affecting the ortho-para ratio in the Ar; -6 cyclization of phenol (30) was reported.73 Results were consistent with two transition states (146) and (147) leading respectively to ortho-alkylation (31) and (1 46) (1 47) para-alkylation (32) products.Chelation of the metal cation facilitated charge transfer to the leaving group in the case of ortho-alkylation. Thus ortho-alkylation invariably predominated in solvents of low polarity. The percentage of ortho- alkylation decreased linearly with dielectric constant of the solvent in the order Bu'OH > Pr'OH > EtOH > MeOH and reflected decreasing chelation of the metal cation. On the other hand, ion aggregation accelerated the relative rate of para-alkylation since charge transfer to the leaving group was facilitated by neigh- bouring metal cations in the ion cluster.Thus a lower ortho-para ratio was observed in toluene. The ionic radius of the metal, a measure of chelating ability, was inversely proportional to the ortho-para ratio. This ratio decreased in the order Li+ > Na+ > K+ > BuZN+. At higher temperatures, metal chelation was less effective. The ortho-para ratio was less than unity in methanol and in water at reflux temperatures. These conclusions are consistent with the results of the investigation of the Ar; -6 cyclization of phenol (148).61 This reaction, in which sodium amide in liquid ammonia was employed, involved a benzyne electrophilic centre (149). The phenolic products (150) and (151) were formed in a 1 : 1 ratio reflecting high solvent polarity and a low degree of metal cation complexation. l2 W.L.Mock and K. A. Rumon, J. Org. Chem., 1972,37,400. 73 P. G. Duggan and W. S. Murphy, J. Chem. SOC.,Perkin Trans. 2,1975, 1291. 238 Murphy and Wattanasin OH 0-b,, OH A dramatic example of the effect of metal cation chelation was noted in the course of an investigation of the Ar, -5 and Ar; -6 cyclizations invohing the ortho- quinone methides (152) and ( 154).46The phenoxide (152) cyclized exclusively at the ortho-position to give phenol (153). In the presence of 18-crown-6, (152) gave no l3r I (1 52) (1 53) cyclized products (see Section 3). The phenoxide (154) was less strongly chelated since both para- and ortho-cyclization occurred. The ortho-para ratio was 88 : 12.Br Anionic Cyclization ofPhenols In the presence of 18-crown-6, (154)cyclized but the ratio of (155) to (156) changed to 38 : 62. The cyclization of phenol (128),"' one of the few known examples of Ar; -5,is deserving offurther comment. A detailed mechanistic study was not undertaken by Ma~Leod,~'although para-cyclization only was detected. No ortho-alkylation was detected. In this instance the ability of phenoxides to alkylate at the para-position is at least partially accentuated by the hardness of the carbonyl The ortho-para ratio (1 17): (1 18)of the Pictet-Spengler cy~lization'~~~~ of (1 15) was pH dependent?O pH 1, 0:p = 3.97; pH 9, 0:p = 34: 66. This effect is suggestiveof two mechanisms-an Ar, -6cyclization at low pH and an Ar, -6 cyclization (1 16)at higher pH values.In the so-called phenol cyclizations, Kametani and co-worker~~~ noted that the phenol (157;R =i H) underwent para-cyclization with 3,4-dimethoxybenzaldehyde in the absence of acid to the isoquinoline (158;R = H). None of the ortho-isomer (159)was detected. However, when the ether (157;R = Me) was treated in the RO RO Me0 Me0 M eO (1 57) \ OMe RO ' OMe OMe OMe same way, no isoquinoline was formed. The corresponding Schiff s base only was isolated. Electron densities at the cyclization positions of (157;R = H) and (157; R = Me) were invoked as e~planation.~~ In electrophilic substitution, it had been established that the hydroxyl is more activating than the hydroxy-group. '* This effect is again highlighted by the comparison between the results of the following two reactions.In the reaction of 1-veratrylnorhydrohyrastinine (160) tetrahydro-+-berberine (161)was obtained exclusively.79 However, it was sub- sequently noteds0 that treatment of tetrahydropapaveroline (162)with formal- dehyde, afforded the ortho- (163)and para- (164)products in 1 : 1 ratio. It has been 74 T.-L. Ho, Chem. Rev., 1975,75, 1. 75 W. M. Whaley and T. R. Govindachari, Org. React., 1951,8, 151. 76 T. Kametani, 'The Total Synthesis of Isoquinoline Alkaloids', ed. J. ApSinlon, Wiley-Interscience, N.Y., 1977, 3, p. 1. 77 T.Kametani, K. Fukomoto, H. Agui, H. Yagi, K. Kigasawa, H. Sugahara, M. Hiiragi, T. Hayasaka, and H. Ishimaru, J.Chem. SOC. (C), 1968,112. 78 C. K. Ingold, 'Structure and Mechanism in Organic chemistry', Cornell University Press, Ithaca, N.Y.,1953, p. 70; P. B. D. de la Mara, 0.M. H. el Dusouqui, T. G. Tillett, and M. Zeltner, J. Chem. SOC.; 1964, 5306; G. Chuchani, H. Diaz, and J. Zabicky, J. Org. Chem., 1966,31, 1573; 2330. 79 R. D. Haworth, W. H. Perkin, and J. Rankin, J. Chem. SOC.,1924, 125, 1686. E. Spath and E. Kruta, Monatsh. Chem., 1928,50,341. Murphy and Wattanasin OMe ~uggested’~that the appropriate free hydroxy-group in the benzyl residue activated the ortho-and para-positions equally. HO 4100 oc OH OH An alternative explanation is feasible :Ar; -6 cyclization, wherein a reasonably high equilibrium concentration of phenoxide (pK, -is provided by the amino-group (pK, -even as it is transformed via the Schiffs base to tetra- hydroisoquinoline.Although the o :p ratio is not always consistent with this mechanism, subtle effects do operate. For example, phenol cyclizations2 of tetra-hydroisoquinoline (165) with fsrmalin in hot ethanol was converted into the homoprotoberberine-type product (166)exclusively. This is the ortho-cyclization product. S. H. Pine, J. B. Hendrickson, D. J. Cram, and G. S. Hammond, ‘Organic Chemistry’, McGraw-Hill, Kogakusha, Tokyo, 1980, p. 200. 82 T. Kametani, T. Terui, A. Ogino, and K. Kukumoto, J. Chem. SOC.(C), 1969, 874. 241 Anionic Cyclization of Phenols OMe Kametani and co-workers have used their theory to explain numerous other examples of phenol cy~lization.~~.~~ Whether or not these reactions are examples of Ar; -n cyclizations remains to be determined.The importance of reaction conditions on the Pictet-Spengler reaction has recently been highlighted by Cook and co-worker~.~~ They found that yields in aprotic media were generally 300-400% better than those in aqueous acid. 9 Reactivity Considerations External factors such as solvent, metal cation, and temperature, affect Ar; -n and Ar, -n cyclizations. Since they can be classified as intramolecular SN2reactions, their efficiency will also depend on the nucleophilicity of the phenoxide, the nature of the leaving group, and reactivity of the electrophilic centre. In addition, as dis- cussed earlier, the nature of the chain8' joining these two centres such as chain length, presence of an oxygen atom in the chain, and effect of gem-dimethylation, strongly affects the ease of cyclization.Hydrogen-deuterium exchange under basic conditions was used to predict the ability of phenoxides to react with citral.86 The basicity so determined was found to correlate with nucleophilicity. Only strongly nucleophilic phenoxides, such as the anion of resacetophenone, react with citral or related enones. It is therefore not surprising that enone (130) and (167) did not cyclize under basic conditions. (167) However, if forcing conditions are used then as Atkinson and Miller8 noted, phenoxide-enone coupling may occur. They synthesized the dienone (169), required for this study, by an Ar; -6 cyclization of (168).Dienone (169) cyclized quantitatively to the adamantanoid dienone (170) when heated at 170 "C with 83 T. Kametani and M.Ihara, J. Chem. SOC.,Perkin Trans. I, 1980,629. 84 D. Soerens, J. Sandrin, F. Ungemach, P. Mokry, G. S. Wu, E. Yamanaka, L. Hutchins, M.Di Pierro, and J. M.Cook, J. Org. Chem., 1979,44,535. 85 M.A. Winnik, Chem. Rev., 1981,81,491. 86 D. G. Clarke, L. Crombie, and D. A. Whiting, J. Chem. SOC.,Perkin Trans. I, 1974, 1007. Murphy and Wattanasin potassium t-butoxide for three hours. Although spectroscopically (8) and (170) could not be differentiated, Dreiding models indicated a clear preference for (8). It is of note that the stereoelectronic requirements clearly absent in the transition state (130) are fulfilled in (169).However, the vigorous conditions employed here were not applied to the enone ( 130).63 OH b f---$yo0 0'-ll I The ortho-quinone methides (152) and (154) cyclized via Ar; -5 and Ar; -6 modes re~pectively.~~ The success of these reactions is without doubt due to the inherently high reactivity of the quinone methide f~nctionality.~ However, without the assistance of Mg" chelation, Ar; -5 cyclization failed whereas Ar; -6 occurred efficiently but with modified regiochemistry. The failure of (152) under these conditions again emphasizes the general reluctance of systems to undergo Ar; -5 cyclization. Ester functional groups such as methoxy- and ethoxy-carbonyl are insufficiently electrophilic to permit Ar; -6 cyclizations.Thus the ester (171) was recovered unchanged after treatment with sodium methoxide.' The same result was observed with ester (172).*' By comparison, the aldehyde function is sufficiently reactive. The aldehyde (141) cyclized smoothly under mildly basic condition^.^ 87 A. Zanarotti, Tetrahedron Lett., 1982,3815; 3963. W. S. Murphy and S. Wattanasin, unpublished results. 243 Anionic Cyclization of Phenols (171) (1 72) Sih and co-workers largely confirmed these observations in the course of their regiospecific synthesis of the anthracyclinone, adriamycinone (173) via base catalysed cyclizati~ns.~~ Retrosynthetic analysis revealed (174) as a plausible precursor. However, the ester (174) resisted cyclization.This failure was attributed n Me0 0 OH Me0 0 OH 0 OM' (173) (1 74) to the electron-withdrawing property of the anthraquinone system. When the aldehyde (175), derived from (174), was treated with basic sodium dithionite, the ethylene ketal of adriamycinone was isolated in 53% yield. By analogy with his earlier results Sih suggested a mechanism which involved the leucoform (176). However, a reasonable alternative is base-catalysed Ar, -6 cyclization of (175). r OII 0 ni glycolate of (173) t--89 F. Suzuki, S. Trenbeath, R. D. Gleim, and S. J. Sih, J. Org. Chem., 1978,43,4159. Murphy and Wattanasin Support for this alternative mechanism is provided by the results of Krohn’sgo synthesis of 9-deoxyanthracyclinones.He isolated (179) as a mixture of epimers OH C02Me 0 OH C02Me \ NaH, ‘3M[ 0 OH 0 OH OH (1 78) (1 79) after heating the ester (178) with enals under basic conditions. The intermediate (NO), although not isolated, most reasonably cyclizes via an Ar; -6 mechanism. This methodology was extended to the synthesis of ( & )-Pdeoxy-s-rhodo-mycinones (181).” A reaction of the phthalide (182) and benzoquinone in the presence of base was attern~ted.~’No anthraquinone (185) was formed. It was concluded that the phenolate (~4)~~ derived from the Michael adduct (183) was insufficiently nucleo- 90 K. Krohn, J. Chem. Res. (S), 1979,318. R. A. Russell and R. N. Warrener, J. Chem. SOC.,Chem. Commun., 1981, 108. 92 W. Trueb and C.H. Eugster, Helu. Chim. Acta, 1972,55,969. Anionic Cyclization of Phenols phiiic. Two other factors, not considered by the authors,” also mitigated against the reaction (184)-,(185), i.e. (a) an unfavourable Ar; -5 transition state, and (b) the low electrophilicity of the lactone functional group. 0 (1 87) The base-catalysed phthalide-benzyne annulation reaction93 has recently been used by Town~end~~ in a synthesis of (f)-averufin (187; R = R’ = H). In contrast to the phenoxide (184) the intermediate (188) readily cyclized since it (a) is an aryl anion, (b)is highly nucleophilic, and (c) has quite different stereoelectronic require- ments from a phenoxide anion, resulting in a transition state much less strained than the Ar, -5 (184).10 Limitations The principal limitation to Ar; -n cyclizations is steric factors. Forcing conditions are frequently used and presumably, required. Side reactions ‘therefore, predominate on occasion.28 Ar; -n cyclizations are limited both by ring size (Section 4) and reactivities of the substrate (Section 5). MasamuneI3 concluded from his investigation of the Ar; -5 cyclization of (46) that one diastereomer failed to react for steric reasons. Beames and Mander29 probed these limitations in the course of their investigation of the Ar; -6 cyclization of the related system (39). However, they could not improve on Masamune’s method. They concluded that steric interactions due to the presence of the ether substituent in the Ar, -6 cyclization of phenol (39) and similar substrates placed a severe restriction on the general utility of phenoxide cyclization for preparing compounds with an oxygenated bridge.In an attempt to solve this g3 D. J. Dodsworth, M.-P.Calcagno, E. U. Ehrmann, B. Devadas, and P. G. Sammes, J. Chem. SOC., Perkin Trans. I, 1981, 2120. 9* C. A. Townsend, S. G. Davis, S. B. Christensen, J. C. Link, and C. P. Lewis, J. Am. Chem. SOC., 1981,103,6885. 246 Murphy and Wattanasin problem they investigated the acid catalysed cyclization of phenolic diazo- ketones.29 They were notably successful. Two examples are included here. The dienone (190) was formed in 100% yield.95 The comparable Ar, -5 cyclization of phenol (126)failed.13 The formation of the dienone (192)96 is notable.All efforts to detect evidence for Ar, -4 cyclization have failed.18,20~21~96 (191) (192) The conditions necessary for reaction are not predictable. Whereas it required4' 170 "Cto effect Ar, -6 cyclization to the dienone (97; R = H), the dienone (22; R = But) was obtained in 98% yield after heating with potassium t-butoxide in t-butyl alcohol at reflux tem~erature.~~ Variable yields have been reported for spirodienones (193) synthesized via Ar; -3 cyclization. Although the yields appear to be affected by the nature of the double-bond substituents, in general they have not been optimized. In addition, R' R2 Yield (5%) /.I? f. 81 97 50 97 59 97 Me But 76 97 (193) Me Me 22 97 H H 2 -4 4,98.99 95 D. J.Beames, T. R. Rose, and L. M. Mander, Aust. J. Chem., 1974,27, 1269. 96 D. J. Bearnes and L. N. Mander, Aust. J. Chem., 1974,27,1257. 97 V. V. Ershov, I. S. Belostotskaya, and V. I. Volod'kina, Zh. Org. Khim., 1967,3, 511. 98 D. I. Schuster and C. J. Polowczyk, J. Am. Chem. SOC., 1966,88, 1722. 99 D. I. Schuster and I. S. Krull, J. Am. Chem. SOC., 1966,88,3456. Anionic Cyclization of Phenols certain dienones are extremely reactive towards electrophilic and nucleophilic reagent^.^^'^^*^^' Spirodienones'" e.g. (3; R = H and R = Me) have a strong tendency to rearrange to phenols in the presence of acid,68 or thermallys6 in the course of distillation. These product characteristics may account for the low and variable yields.However, the successful Ar; -3 synthesis3' of (69) by means of a tri- alkylamine catalyst at room temperature, promises more scope for this reaction. The Ar; -5 transition state is highly strained and usually will. not occur. Although Ar, -6 cyclizations occur readily, they are not regiospecific. Pre- dominant ortho-alkylation is usually observed which contrasts with Ar, -6 cyclization wherein para-alkylation occurs, frequently ex~lusively.~~ This inability to control the o :p ratio has limited the synthetic potential of Ar, -6 cyclizations. For this reason Mande1126 did not pursue the Ar, -6 cyclization of phenol (27) route to the synthesis of santonin. Aryl ring substituents deflect Ar; -6 cyclization to unsubstituted positions. Newman and Mekler' were therefore disappointed to find that none of the dienone (139),a potentially useful intermediate in terpenoid total synthesis,' was formed when the phenol (5) was treated with base.11 Conclusions Although the area of phenoxide cyclizations has been investigated in considerable detail, there is much scope for further development and application, particularly in the area of natural products synthesis. Base catalysis involves protecting groups and tolerance of substituents which differ from those used in the more common cationic processes. The SN2character of the reaction is implicitly more stereo- controlled than analogous cationic processes. With the exceptions of the cyclizations of (24),24 (47),31*32 and (70),38this aspect has not been fully appreciated. Stereoelectronic requirements of Ar; -n cyclizations in particular require further elucidation.In addition, the steric requirements for the formation of rings with n > 6 has not been investigated. More recent developments, such as phenoxide-quinone methide coupling,46 offer scope in the area of biomimetic syntheses, e.g. lignan total synthesis. The effect of metal cations leading to predominant ortho-cyclizati~n~~~~~~ suggests an important alternative to cationic processes. A number of areas remain to be developed such as the anionic Pictet-Spengler reaction,60 intramolecular coupling of phenoxide with quinones, 'O4 quinone m~noacetals,~'and arynes,6f in addition to phenoxide-radical coupling'05 and loo V.V. Ershov, I. S. Belostotskaya, and V. I. Volod'kina, 120. Akad. Nuuk SSR, Ser. Khim., 1967, 930. lo' V. V. Ershov, I. S. Belostotskaya, and V. I. Volod'kina, lzv. Akad. Nauk SSR, Ser. Khim., 1966, 1496. lo* R. S. Ward, Chem. Br., 1973,444. lo3 L. Bolzoni, G. Casiraghi, G. Casnati, and G. Sartori, Angew. Chem., lnt. Ed. Engl., 1978, 17, 684. lo4 H. Muzzo, U. V. Gizychi, U. I. Zahorszky, and D. Bormann, Annalen, 1964,676, 10. lo5 W. A. Waters, J. Chem. SOC.(B),1971,2026. 248 Murphy and Wattanasin photochemically induced' O6 phenoxide cyclizations. The importance of the nature of the leaving group and electrophilic centre needs to be stressed. For example, the low reactivity of common esters resulted in ineffective cyclization.More reactive esters coupled with appropriate metal cation chelation may lead to useful developments in the area of anthracyclone and tetracycline total synthesis. The possible involvement of phenoxide cyclization in biosynthetic pathways has not been investigated. Phenolic cyclization is the most frequently cited'07 mode of isoquinoline ring formation in the course of isoquinoline alkaloid biosynthesis. However, Kametani'" showed that when (+ )-retidine (189)was incubated with rat liver microsomes under slightly basic conditions it was biotransformed uia the iminium salt (195)into ( -)-scoulerine (196) and ( -)coreximine (197),in the ratio 27 : 73, respectively. This ratio is suggestive of an Ar; -6 cyclization of (195). In addition, the ratio of isomeric alkaloids found together is, on occasion, more con- sistent with Ar; -6 cyclization.For example, longimammidine (198) and Me0 M eO '9-OMe OH OH (194) (197) lo6 2.Horii, Y. Nakashita, and C. Iwata, Tetrahedron Lett., 1971, 1167; T. Kametani and Kohno, Tetrahedron Lett., 197 1, 3 155. lo' U. Weiss and J. M. Edwards, 'The Biosynthesis of Aromatic Compounds', Wiley, N.Y., 1980, p. 49; M. Luckner, 'Secondary Metabolism in Plants and Animals', Academic Press, N.Y., 1972, p. 321. lo* T. Kametani, N. Kanaya, Y. Ohta, and M. Ihara, Heterocycles, 1980, 14,963. Anionic Cyclization of PhenoIs HO (198) longimammosine (199) were isolated in a 1 : 1 ratio from the peyote cactus.'0g In the field of lignan biosynthesis' lo phenoxide-quinone methide coupling is a pathway which has yet to be investigated.log R. L. Ranieri and J. L. McLaughlin, J. Org. Chem., 1976,41, 319. 'lo A. J. Birch and A. J. Liepa, 'Chemistry of Lignans', ed. C.B.S. Rao, Andhra University Press, India, 1978, p. 316. 250

ISSN:0306-0012    

DOI:10.1039/CS9831200213

出版商:RSC    

年代:1983

数据来源: RSC