8 Organophosphorus Chemistry By S. TRIPPETT Department of Chemistry The University Leicester LEI 7RH The significant advances in organophosphorus chemistry are reviewed annually in the Specialist Periodical Reports series.’ This article seeks to answer the question ‘What has been happening in the past three years?’. The answer must be subjective and is doubtless biased by the interests of the Reporter. However most organophosphorus chemists would agree that the major publishing event has been the start to publication of the new ‘Kosolapoff ’2 which promises to be as indispensable as its predecessor. Chemically the most significant advances have been in the general area of quinquecovalent phosphoranes. The importance of these as reactive intermediates3 has made an understanding of their properties both chemical and fluxional essential to further developments in mechanism and stereochemistry.1 Structural Aspects Numerous ab initio and semi-empirical MO calculations have been rep~rted.~ In general they agree that the inclusion of d-orbitals in the basis set gives a better bonding picture but the improvement in some cases is small. The calculations on simple trigonal-bipyramidal phosphoranes confirm that the more electro- negative substituents will prefer to occupy apical positions and that atoms having a lone-pair of electrons will prefer to be bonded at the equatorial position. They further predict4b that groups having a low-lying vacant orbital will be more stable in the apical position and that an equatorial substituent with a single donor orbital will prefer to have that orbital in the equatorial plane.This preferred orientation can give rise to a considerable barrier to rotation round equatorial bonds which is well established experimentally in the case of P-N,5 ‘Organophosphorus Chemistry’ ed. S. Trippett (Specialist Periodical Reports) The Chemical Society London 197&-1974 Vols. 1-5. ‘Organic Phosphorus Compounds’ ed. G. M. Kosolapoff and L. Maier Wiley- Interscience New York 1973 Vols. 1-4. ’ P. Gillespie F. Ramirez I. Ugi and D. Marquarding Angew. Chem. Internat. Edn. 1973 12 91. E.g. (a) A. Rauk L. C. Allen and K. Mislow J. Amer. Chem. Soc. 1972 94 3035; J. B. Florey and L. C. Cusachs ibid. 3040; A. Strich and A. Veillard ibid.1973 95 5574; (6) R. Hoffman J. M. Howells and E. L. Muetterties ibid. 1972 94 3047; (c) P. Gillespie P. Hoffman H. Klusacek D. Marquarding S. Pfohl F. Ramirez E. A. Tsolis and I. Ugi Angew. Chem. Innternat. Edn. 1971 10 687. E. L. Muetterties P. Meakin and R. Hoffman J. Amer. Chem. Soc. 1973 95 974 and earlier papers cited therein. 268 Organophosphorus Chemistry and P-S,6 but not P-0,' bonds. Ab initio calculations' on methylenephos- phorane H,P=CH, show no barrier to rotation round the C-P bond and this distinction between four- and five-co-ordinate phosphorus could be of considerable significance. The generally assumed trigonal-bipyramidal geometry has been confirmed in X-ray analysis of numerous phosphoranes among them the 1,2-oxaphosphetan (1),9 the dioxyphosphorane (2)," and the spirophosphorane (3).' However the tetrathiospirophosphorane (4)12has a geometry intermediate between trigonal bipyramidal and square pyramidal while the spirophosphorane (5)13 is essentially square pyramidal with CPO bond angles of 154 and 148".Ph Ph' 1 Ph Ph. ..;) ;Yo C,H,Br-p (3) 2 Five- and Six-Co-ordinate Species Pseudorotation Processes-The exact route by which one trigonal-bipyramidal phosphorane (7) is transformed into its isomers is still contr~versial.~,~~ The two favoured mechanisms both of which satisfy the requirements of the famous Whitesides and Mitchell experiment,14 are Berry pseudorotation (BPR) which involves a square-pyramidal intermediate or transition state (8) and turnstile rotation (TR) which proceeds uia a 30" (2 + 3) species (6) shown in Newman S.C. Peake and R. Schmutzler J. Chem. SOC. (A) 1970 1049. ' S. C. Peake M. Fild M. J. C. Hewson and R. Schmutzler Znorg. Chem. 1971,10,2723; D. U. Robert D. J. Costa and J. G. Riess J.C.S. Chem. Comm. 1973 745. I. Absar and J. R. Van Wazer J. Amer. Chem. SOC.,1972 94 2382. Mazhar-ul-Haque. C. N. Caughlan F. Ramirez J. F. Pilot and C. P. Smith J. Amer. Chem. SOC.,1971,93 5229. lo D. D. Swank C. N. Caughlan F. Ramirez and J. F. Pilot J. Amer. Chem. SOC.,1971 93. 5236. I' M. Sanchez J. Ferekh J. F. Brazier A. Munoz and R. Wolf Roczniki Chem. 1971 131. '' M. Eisenhut R. Schmutzler and W. S. Sheldrick J.C.S. Chem. Comm. 1973 144. l3 J.A. Howard D. R. Russell and S. Trippett J.C.S. Chem. Comm. 1973 856. l4 G. M. Whitesides and H. L. Mitchell J. Amer. Chem. SOC.,1969 91 5384. 270 S. Trippett projection. Numerous calculations (e.g.refs. 4a b) suggest that in simple sym- metrical phosphoranes BPR is the easier route but of course this may not be so in less symmetrical and particularly in highly strained systems. The geometries and energies of the two species (6) and (8) are so similar that for the experimental chemist distinction between BPR and TR processes is probably academic. 3 (7) The pseudorotations of many stable phosphoranes have been studied using dynamic n.m.r. techniques and some have been interpreted in terms of the varying apicophilicities of the groups moving between apical and equatorial positions and the variation in strain as small-membered rings move from apical- equatorial to diequatorial positions.The dangers inherent in such experiments have been emphasized by the ob~ervation'~ that the fluorine exchange in Ph,PF is intramolecular when monitored by n.m.r. in Teflon tubes but intermolecular and of lower energy in Pyrex. The intermolecular processes observed here and in other cases16 could be due to impurities caused by chemical reaction with the glass and it may be that all previous data on fluorophosphoranes are suspect for the same reason. Only a few of the d.n.m.r. studies on phosphoranes can be mentioned. The 31Pand 'H spectra of the difluorophosphorane (9) at -100 "C show clearly17 that the phosphorane is a 2.3 :1 mixture of (9a) and (Sc) equilibration via the high-energy (9b) being slow on the n.m.r.time-scale at this temperature. The apicophilicity of fluorine is balancing the increased strain involved in placing the four-membered ring diequatorial. The 19F d.n.m.r. spectra of the spiro- phosphoranes (10) give data on the relative apicophilicities of the groups R.18 The pseudorotation that can be followed is (10)S(11) and AG*varies as the apicophilicity of R. The results give an order Ph < CH=CMe < Pr' < Me < Me2N < PhO < H. A large difference in apicophilicity between R2N and PhO is supported" by studies on the adducts (12) and data on the hexafluorobiacetyl adducts (13) give information on the apicophilicities of R relative to phenyl.,' Is C.G. Moreland G. 0.Doak and L. B. Littlefield J. Amer. Chem. SOC.,1973 95 255. l6 T. A. Furtsch D. S. Dierdorf and A. H. Cowley J. Amer. Chem. Soc. 1970,92 5759. " N. J. De'ath D. Z. Denney and D. B. Denney J.C.S. Chem. Comm. 1972 272. '' R. K. Oram and S. Trippett J.C.S. Perkin I 1973 1300. l9 S. Trippett and P. J. Whittle J.C.S. Perkin I 1973 2302. J. I. Dickstein and S. Trippett Tetrahedron Letters 1973 2203. Organophosphorus Chemistry 27 1 F (94 R OPh PhO.. I The spirophosphoranes derived from the various ephedrines provide particu- larly intriguing exercises in following pseudorotation processes.* Of those formed from (-)-ephedrine one (14) is obtained pure by crystallization and equilibration with its isomer (15) can be followed polarimetrically in solution.The resulting activation parameters agree with those previously obtained from n.m.r. studies. Isomers (14) and (15) can equilibrate either by routes involving five successive pseudorotations on which the highest energy tbps are of type (16) or by routes involving seven pseudorotations on which the highest energy tbps are of type (17). One would expect the latter to be the preferred pathways. With the accumulation of data on the energetics of pseudorotation processes it should soon be possible to predict the relative stabilities of isomeric quinque- covalent phosphoranes and to apply this knowledge to prediction of the pathways of reations at phosphorus involving such intermediates. The first kinetic evidence for the involvement of a quinquecovalent intermediate in the alkaline hydrolysis of a phosphate ester has been reported.22 Hydrolysis of methyl di-isopropyl- phosphinate shows an induction period and the overall kinetics are consistent with Scheme 1 where k N k N O.lk-A similar long-lived intermediate has been invoked23 to account for the racemization observed in the alkaline hydrolysis *' A.Klaebe J. F. Brazier F. Mathis and R. Wolf Tetrahedron Letters 1972 4367. '' R. D. Cook,P. C. Turley C. E. Diebert A. H. Fierman and P. Haake J. Amer. Chem. SOC.,1972 94 9260. '' L. P. Rieff L. J. Szafraniec and H. S. Aaron Chem. Comm. 1971 366. 272 S. Trippett Ph MGo 0 \ Ph Ph of the ester (18) and the loss of stereospecificity observed during demethylation of the methoxyphosphonium salt (20) has been ascribed24 to racemization (by repeated pseudorotation) of the phosphoranes (19) formed in parasitic equilibria.The Arbuzov reaction” is among others in which the reversible formation of intermediate quinquecovalent phosphoranes has been postulated. Chemistry of Phosphoranes.-Fragmentation of the cis-3-phospholen (21) is 99 % stereospecific26 and probably a concerted disrotatory process. Hoffman showed4* that the concerted reactions (22) e(23) are symmetry allowed for Scheme 1 l4 K. E. DeBruin and S. Chandrasekaran J. Amer. Chem. SOC..1973,95 974. 25 C. L. Bodkin and P. Simpson J.C.S. Perkin II 1972 2049. *‘ C. D. Hail J. D. Brarnblett and F. F. S. Lin J. Amer.Chem. SOC.,1972 94 9264. 273 0rganophosphorus Chemistry Ph OMenthyl Ph OMenthyl Ph OMenthyl \I \+ / \/ P-OMee P +N-+ +MeN /\ Me/I N Me OMe Me (19) (20) apical-apical or equatorial-equatorial loss or addition. Similar reasoning pre- dicts that the allowed process for loss of diene from (21) is from an apical- equatorial position. The 1,3-dipoles (25) formed on thermolysis or photolysis of the 1,3,5-oxaza- phospholens (24) have been trapped27 with a wide variety of dipolarophiles including olefins acetylenes nitriles isocyanides and carbonyl compounds. qR1 (CF3),C=NCOR’ +(R20),P -(CF& (,O *(CF3),C-N=6R‘ +(R20),po P (OR2)3 (25) (24) The possibility that six-co-ordinate species might be intermediates in reactions at phosphorus has long been recognized.Firm kinetic evidence has now been obtained for such an intermediate or transition state in the hydrolysis of penta- phenoxyphosphorane,28 and the base-catalysed exchange of alkoxy-groups in the oxaphosphetan (26) probably proceeds via attack of nucleophile in the ’’K. Burger and K. Einhellig Chem. Ber. 1973 106 3421 and earlier papers. 28 W. C. Archie jun. and F. Westheimer J. Amer. Chem. SOC.,1973 95 5955. 274 S. Trippett equatorial plane.29 A growing number of stable six-co-ordinate phosphorus- containing anions e.g.(27) have been pre~ared.~'.~ ' Although these are doubt- less stabilized by the presence of small rings their very existence suggests that similar anions may be more important as intermediates than has hitherto been appreciated.The triethylammonium salt of (27) on heating' gave the phos- phorane (28) and hydrogen! 1 Among other substitution reactions of five-co-ordinate phosphorus reported are those shown in Scheme 2.32 Nothing is known so far about the stereo- chemistry of such reactions. The equilibria between the tetrasubstituted phosphoranes (29) and the cor- responding P'll species (30) have been studied for X = Y = 0,33for X = 0 Y = N,34for X = Y = N,34 and for X = 0,Y = S.35 In general substitution of the rings favours the Pv form so that for example there is no evidence for the Pi''form in solutions of (31). However when Y = S the compounds exist entirely as (30; Y = S). 29 F. Ramirez G. V. Loewengart E. A. Tsolis and K.Tasaka J. Arner. Chern. SOC. 1972 94 353 1. 30 E.g. B. C. Chang D. B. Denney R. L. Powell and D. W. White Chern. Cornrn. 1971 1070; L. Lopez M. T. Boisdon and J. Barrans Cornpt. rend. 1972 275 C 295; R. Burgada D. Bernard and C. Laurenco ibid. 1973 276 C 297. 31 M. Wieber and K. Foroughi Angew. Chern. Internat. Edn. 1973 12 419. 32 D. Bernard and R. Burgada Tetrahedron Letters 1973 3455. 33 D. Bernard C. Laurenco and R. Burgada J. Organornetallic Chern. 1973,47 113. 34 C. Laurenco and R. Burgada Cornpt. rend. 1972 275 C 237. '' D. Bernard P. Savignac and R. Burgada Bull. SOC.chirn. France 1972 1657. Organophosphorus Chemistry NMe2 OCOPh 00 OMe Scheme 2 H Me Phosphoranes containing at least two alkoxy-groups undergo exchange re- actions with 1,2- and 1,3-glycols e.g.(32)-+ (33).36 The synthesis of phosphoranes using dialkyl peroxides has been extended to include the use of the dioxetan (34)37 and the dithiet (35).38 cy/OEt P(OEt) + HOCH2CH20H -+ 0-P (32) 01 + Ph3P benzene Ph3P 0 (34) '' B.C. Chang W. E. Conrad D. B. Denney D. Z. Denney R. Edelmann R. L. Powell and D. W. White J. Amer. Chem. SOC.,1971 93 4004. 37 P. D. Bartlett A. L. Baumstark and M. E. Landis J. Amer. Chem. Soc. 1973,95,6486. 38 N. J. De'ath and D. B. Denney J.C.S. Chem. Comm. 1972 395. 276 S. Trippett + -@h (35) 3 Stereochemistryand Mechanism The barrier to inversion in phosphines normally increases with the electro-negativity of the ligand~.~'The lower barrier in PhPr'PSi(OMe) than in PhPr'PSiMe is ascribed to negative hyperconjugation in the former.40 Data on the alkaline hydrolysis of phosphonium salts which it is not at present possible to rationalize continue to accumulate.Hydrolysis of the salts (36) with loss of R proceeds with partial inversion or retention at phosphorus de-pending on the nature of R and on whether the reaction is carried out under homogeneous or heterogeneous condition^.^^ Hydrolysis of the alkoxy(alky1thio) salts (37) with formation of thiol involves retention at phosph~rus.~~ Among hydrolyses involving interesting rearrangements are those of (38)43 and (39).44 The stereochemistry of the products from the hydrolysis of the phosphetanium salts (40; X =RO Me,N MeS or C1) has been accounted for in terms of the varying apicophilicity of X.45 Me .+ Ph -,P -R Bu' (36) R = PhCH ,p-F,CC,H,CH Ph,CH or CH,=CHCH Ph \+/ OMenthyloH-Ph\p/OMenthyl OMenthyl OH \A/P + Ph\D/oMenthyl /\\ Me SMe Me / \\o PhaPh 5 Phmph P P /\ Ph CH,I P Ph u A (38) 39 R.D. Baechler and K. Mislow J. Amer. Chem. Soc. 1971 93 773. 40 R. D. Baechler and K. Mislow J.C.S. Chem. Comm. 1972 185. 41 R. Luckenbach Phosphorus 1972 1 223 229,293. 42 N. J. De'ath K. Ellis D. J. H. Smith and S. Trippett Chem. Comm. 1971 714. 43 A. N. Hughes and C. Srivanavit Canad. J. Chem. 1971,49 879. 44 F. Mathey Tetrahedron 1972 28 4171; 1973 29 707. 45 K. E. DeBruin A. G. Padilla and M.-T. Campbell J. Amer. Chem. Soc.1973,95,4681. Organophosphorus Chemistry 277 Whereas methanolysis of (41) with displacement of thiolate ion involves inversion at the reaction of (42) with phenylmagnesium bromide also with loss of thiolate ion gives retenti~n.~’ The latter observation which is not at present understood required the revision of several previously accepted configurational assignments. MexMe ,SMe / SMe O=P. O=P-Me i-* OM en thy1 \ OPr’ Me P Me Ph RO# kx SbCI (40) (41) (42) The reactivities of cyclic tervalent phosphorus compounds relative to their acyclic analogues have been discussed48 in terms of the changes in ring strain between ground state and transition state. An S,1 (P)mechanism has been sug- gested for the aqueous solvolysis of di-t-butylphosphinic ~hloride.~’ The stereochemistry of nucleophilic displacements on phosphorus is being intensively studied using cyclic compounds such as (43)” and (44)” in which the configuration at phosphorus can be deduced from n.m.r.measurements. The outcome depends critically upon the nucleophile and is affected markedly by the presence of added salts. In an area where strain and apicophilicity factors are finely balanced it is difficult to rationalize the results satisfactorily. Studies continue on intramolecular catalysis in the solvolysis of phosphate esters. General acid catalysis has been identified in the hydrolysis of the dianions (45)52 and (46),53 and among examples of intramolecular nucleophilic catalysis are the solvolyses of the phosphonylated hydroxamic acids (47)where the rates CI CH,CI I 04pyz-?A C1 Me0 OMeOMe (43) “ W.B. Farnham K. Mislow N. Mandell and J. Donahue J.C.S. Chem. Comm. 1972 ’’ J. 120. Donahue N. Mandell W. B. Farnham R. K. Murray K. Mislow and H. P. Benschop J. Arner. Chem. Soc. 1971 93 3792; G. R. Van den Berg D. J. H. M.Platenburg and H. P. Benschop Rec. Trao. chim. 1972 91 929. R. Greenhalgh and R. F. Hudson Phosphorus 1972 2 1. 49 P. Haake and P. S. Ossip J. Amer. Chem. SOC.,1971 93 6919. 50 W. S. Wadsworth jun. J. Org. Chem. 1973,38 2921. 51 T. D. Inch and G. J. Lewis Tetrahedron Letters 1973 2187. ’’ Y. Murakami J. Sunamoto and H. Ishizo Bull. Chem. SOC.Japan 1972 45 590. 53 R. H. Bromilow and A. J. Kirby J.C.S. Perkin If 1972 123. 278 S.Trippett Pop EtO I R’ \o I HON=CAr (45) 0 (47) (46) are relatively insensitive to steric hindran~e.’~ Metaphosphates’ and meta- pho~phorimidates’~ continue to be postulated as reactive intermediates in order to account for kinetic and product data. Micelle formation can hinder or assist the hydrolysis of phosphate esters. The elimination of p-nitrophenate anion from (48) in base is strongly catalysed by micelles of long-chain quaternized ethanolamines ;57 in contrast the alkaline hydrolysis of bis-(2,4-dinitrophenyl) phosphate is inhibited by micelles of an uncharged detergent possibly by adsorption of the ~ubstrate.’~ Gels of yttrium hydroxide catalyse hydrolysis of (49) with the hydroxide perhaps acting as both general acid and nu~leophile.’~ 0 No II MeP-0 -C H,NO 2-p (Ph0)2P\ I OC6H4NOZ-p 0-(48) (49) MO calculations suggest that the lowest energy pathway for the Wittig olefin synthesis is via a 1,2-oxaphosphetan which undergoes P-C bond cleavage considerably in advance of P-0 cleavage.60 Kinetic data have been variously interpreted in terms of betaine formation6’ and of direct 1,2-oxaphosphetan formation via a four-centred transition state of low polarity.62 The intermediates in Wittig reactions have been detected at low temperature^^^ by Fourier transform ‘P n.m.r.;from their chemical shifts they are undoubtedly 1,2-oxaphosphetans and it is suggested that they are formed directly in (z2a + n2s) reactions involving orthogonal approach of ylide and carbonyl compound.The isolation of the aminotetraoxyphosphorane (52) from the reaction of the nitro-compound (50) with trimethyl phosphite provides evidence for the 54 J. I. G. Cadogan and D. T. Eastlick J.C.S. Chem. Comm. 1973 238. 55 E.g. D. G. Gorenstein J. Amer. Chem. SOC.,1972 94 2523. 56 E.g. M. A. Fahmy A. Khasaninah and T. R. Fukuto J. Org. Chem. 1972 37 617. 57 C. A. Bunton and L. G. Ionescu J. Amer. Chem. SOC.,1973,95 2912. C. A. Bunton A. Kamego and L. Sepulveda J. Org. Chem. 1971,36 2566. 59 F. McBlewett and P. Watts J. Chem. SOC.(B) 1971 881. 6o C. Trindle J.-T. Hwang and F. A. Carey J. Org. Chem. 1973 38 2664. 61 I. F. Wilson and J. C. Tebby J.C.S. Perkin I 1972 271 3. 62 G. Aksnes and F. Y. Khali Phosphorus 1972 2 105; P. Frcayen Acta Chem.Scand. 1972 26 2163. 63 E. Vedejs and K. A. J. Snoble J. Amer. Chem. SOC.,1973 95 5778. Organophosphorus Chemistry formation of spirodienyl intermediates eg. (51) in deoxygenations of nitro-compounds involving rearrangement^.^^ Phosphinidenesand Related Species.-Phosphinidenes :,have been postulated as intermediates in the thermal decomposition of cyclopolyphosphines65 and of the anhydride (53).66 They have been trapped with dienes diphenylacetylene biphenylene ally1 ethyl sulphide and benzil. Phenylphosphinidene sulphide PhPS formed from phenylphosphonothioic dichloride and magne~ium,~~ and the disulphides RPS, formed from the anhydrides (54),68 have been trapped with similar reagents. The oxide PhPO has been generated by pyrolysis of the phosphine oxide (55).69 Ph (Phi),O 5 PhP(OH) + PhP:b%' II 0 0 64 J.I. G. Cadogan D. S. B. Grace P. K. K. Lim and B. S. Tait J.C.S. Chem. Comm. 1972 520. 65 A. Ecker and U. Schmidt Chem. Ber. 1973 106 1453. 66 M. J. Gallagher and I. D. Jenkins J. Chem. SOC.(0,1971 593. 67 S. Nakayama M. Yoshifuji R. Okazaki and N. Inarnoto Chem. Comm. 1971 1186. 68 H. Ecker I. Boie and U. Schmidt Angew. Chem. Infernat. Edn. 1970 10 191. 69 J. K. Stille J. L. Eichelberger J. Higgins and M. E. Freeburger J. Amer. Chem. SOC. 1972 94 476 1. 280 S. Trippett Reactive Centres a and to Phosphorus.-Carbene or carbenoid centres adjacent to phosphoryl groups have been generated from the corresponding diazo- compounds by photolysis7' or by thermolysis in the presence of ~opper.~ They add to olefins to give cyclopropanes and undergo the expected insertions and rearrangements as in Scheme 3.Irradiation of the aide (56) in methanol gave the esters (57) and (58) consistent with the formation of a nitrene inter- mediate.72 I OMe Scheme 3 Me Me Me Me2 P n r '\ Me (56) (57) (59) Carefully designed experiments on the generation of carbonium centres p to a diphenylphosphinyl group show that this group and methyl migrate com- petitively with the Ph,PO migration slightly pred~minating.~~ 'O H. Scherer A. Hartmanu M. Regitz B. D. Tunggal and H. Gunther Chern. Ber. 1972 105 3357. D. Seyferth and R. S. Marmer J. Org. Chern. 1971 36 128. 72 M. J. P. Harger Chern. Comm. 1971 442." D. Howells and S. Warren J.C.S. Perkin II 1973 1472. Organophosphorus Chemistry 28 1 ?-Irradiation of phosphonium salts gives radicals by loss of a hydrogen atom from an a-~arbon.~~ These from their e.s.r. spectra show little interaction be- tween the radical centre and phosphorus. In contrast when the radical centre is fl to phosphorus there is strong interaction which is believed to be hypercon- jugative and at a maximum for the conformation shown in (59)." Phosphorus Radicals.-Phosphinyl radicals e.g. (60) have been shown to be configurationally stable.76 A large number of phosphoranyl radicals R,P* have been generated by addition of radicals to phosphines or by hydrogen ab- straction from tetraoxyphosphoranes. The pseudorotation processes of some have been followed by e.s.r.spectroscopy; the barrier between (61) and (62) is 16-21 kJ mol-' with an energy difference between the two of 2.9 kJ mol-'. It is not possible to harmonize the results on phosphoranyl radicals with those on phosphoranes and clearly there is a fundamental difference between the two. 0 0. 0 II 11 H,C=CH 11 hv P-Ph *ip\H Ph-1 0 ph-f\ Et EtO EtO EtO The a-and /3-scission reactions (see Scheme 4)of phosphoranyl radicals have been extensively studied.78 a-Scission is thought to occur preferentially from an apical position and p-scission preferentially from an equatorial. These preferences lead to an intimate interplay between the rates of pseudorotation processes and scissions. Thus the radicals (63) are comparatively stable since both a-and /3-fission require a highly unfavourable pseudorotation to (64).78 /3-Scission leading to the formation of a P=O bond would normally be expected to be thermodynamically more favourable than a-scission but kinetically this l4 A.R. Lyons G. W. Neilson and M. C. R. Symons J.C.S. Faruday II 1972 68 807. l5 A. R. Lyons and M. C. R. Symons J.C.S. Faraday II 1972 68 622; A. G. Davies D. Griller and B. P. Roberts J. Arner. Chern. Soc. 1972 94 1782; A. L. J. Beckwith Aitsrral. J. Chern. 1972 25 1887; B. C. Gilbert J. P. Larkin R. 0. C. Norman and P. M. Storey J.C.S. Perkin II 1972 1508. l6 G. R. Van den Berg D. H. J. M. Platenburg and H. D. Benschop Chem. Cornrn. 1971 606. l7 R. W. Dennis and B. P. Roberts J.Organornetallic Chern. 1973 47 C8. E.g. A. G. Davies D. Griller and B. P. Roberts J.C.S. Perkin II 1972 2224. 282 S. Trippett is not always so. It may be that the transition state for /3-scission has little P=O character." /p\ OR2 Scheme 4 4 OrganophosphorusCompounds in Synthesis The halogenophosphonium ions (65) formed from phosphines and carbon tetrahalides have found extensive application as sophisticated and more con- venient forms of phosphorus pentahalides e.g. in the conversion of alcohols into halides" and in a wide variety of dehydration reactions.8' The ylides formed from subsequent reaction of the anions (66) with more phosphine can be used in olefin synthesis82 and the anions can also be trapped with carbonyl compounds.Thus the anions from trichloroacetic esters lead to the glycidic esters (67).83 - R3P + CX R36X + CX,- R,P=CX (65) (66) - 0- R3P + Cl,CCO,Et CI,CCO,Et R1R2d-CC1.C02Et I 1 Cl0 /\R 1R2C-CCl. C0,Et (67) '9 W. G. Bentrude E. R. Hansen W. A. Khan T. B. Min and P. E. Rogers J. Amer. Chem. SOC.,1973,95 2286. 'O E.g. E. I. Snyder J. Org. Chem. 1972 37 1466. " R. Appel K. Warning and K.-D. Ziehn Chem. Ber. 1973 106 3450. a2 E.g. E. J. Corey and P. L. Fuchs Tetrahedron Letrers 1972 3769. 83 J. Villieras G. Lavielle and J.-C. Combret Bull. Soc. chim. France 1971 898. Organophosphorus Chemistry Condensations between alcohols and active-hydrogen compounds have been achievedg4 using the complex between triphenylphosphine -and diethyl azo- dicarboxylate (Scheme 5) when HX = phthalimide MeCOCH,CO,Et CH,(CN), (RO),PO,H and benzoic acid.85 The last is useful for the epi- merization of alcohols.86 Carboxylic acids have also been activated particularly in relation to peptide synthesis with trisdimethylarninoph~sphine-CCl~ ,8 triphenylphosphine-bis-(2-pyridyl) disulphide (Scheme 6),88 phosphites and pyridine in the presence of mercuric chloride (Scheme 7),8 diphenyl phosphoryl a~ide,~' and diethyl phosphoryl ~yanide.~ Ph,P + (Et0,CN l2 + Ph3hN-NC02Et Ph3 ;N-NHCO,Et I 1 C0,Et C0,Et X-J.0.Ph,PO + RX Ph,GOR + (EtOJNH) Scheme 5 Ph,P + RSSR -+ Ph,;SR SR lXOH Ph,PO + XY 3 Ph,;OX + RSH X = RCO or ROP0,H; Y = RO ROPO,H or RNH Scheme 6 Scheme 7 Thiirans (68 ;X = S) are obtained from oxirans and phosphine sulphides in the presence of trifhoroacetic acid.92 The same reaction using triphenylphosphine selenide gives the corresponding olefins stereo~pecfically.~~ A synthesis of allylic alcohols involves cleavage of the sulphenate esters (70) formed by a [2,3]-sigmatropic shift from the sulphoxides (69).'" 84 M.Wada and 0. Mitsunobu Tetrahedron Letters 1972 1279. 85 G. Alfredsson and P. J. Garegg Arta Chem. Scand. 1973 27 724. 8h A. K. Bose B. Lal W. A. Hoffman and M. S. Manhas Tetrahedron Letters 1973 1619. '' S. Yamada and Y. Takeuchi Tetrahedron Letters 1971 3595; T. Wreiand and A. Seeliger Chem. Ber. 1971 104 3992. 88 T. Mukaiyarna and M. Hashimoto J. Amer. Chem. Soc. 1972 94 8528. *' N. Yamazaki and F.Higashi Bull. Chem. Sou. Japan 1973.46 1235 1239. 90 T. Shioiri K. Ninomiya and S. Yamada J. Amer. Chem. SOC.,1972 94 6203. 91 S. Yamada Y. Kasai and T. Shioiri Tetrahedron Letters 1973 1595. 92 T. H. Chan and J. R. Finkenbine J. Amer. Chem. Sor. 1972 94 2880. 93 D. L. J. Clive and C. V. Denyer J.C.S. Chem. Comm. 1973 253. 94 D. A. Evans G. C. Andrew T. T. Fujimoto and D. Wells Tetrahedron Letters 1973 1385 1389. 284 S. Trippett (70) The Wittig olefin synthesis has been extensively developed but mention can be made only of 'one-pot' syntheses in which alkyl halide phosphine and car- bony1 compound are allowed to react together in the presence of an epoxide as the source of base.95 The cyclic acylphosphate (71)is both highly selective and very reactive towards hydroxylic nu~leophiles.~~ The resulting acetoin esters are readily hydrolysed to give phosphate diesters.RO 0 ROH I MeCOpf -MeCOCHMe,OR % \p/ 0 P/o -co 0-P \ MeO' \OH OH 'OMe OMe (71) Finally reference must be made to Khorana's heroic synthesis of the structural gene of tRNAA'" from yeast a double-stranded DNA by a combination of chemical and enzymic methods." '' E.g. G. P. 2 132 032 (Chem.Abs. 1972 76 99 874). '' F. Ramirez S. Glaser P. Stern P. Gillespie and 1. Ugi Angew. Chem. Internat. Edn. 1973 12 66. '' H. G. Khorana K. L. Agarwal H. Biichi M. H. Caruthers N. K. Gupta K. Kleppe A. Kumar E. Ohtsuka U. L. RajBhandary J. H. van de Sande V. Sgaramella T. Terao H. Weber and T. Yamada J.Mol. Biol. 1972 72 209.