686 J.C.S. Perkin I1The Preparation and Photolysis of 3-Aryl-3WDiazirinesBy Richard A. G. Smith and Jeremy R. Knowles,*t The Dyson Perrins Laboratory, University of Oxford,The preparation of a number of 3-aryl-3H-diazirines is reported. On irradiation, these materials undergo bothphotolytic fragmentation to the arylcarbene and photoisomerisation to the linear diazo-compound which is thenitself photolysed. The existence of a second intermediate is also apparent from the spectral changes observed.Oxford OX1 3QYOUR interest in chemically stable precursors of non-rearranging carbenes for the photoaffinity labelling ofbiological receptor sites has led us to explore syntheticroutes to, and the photochemistry of, 3-aryl-3H-di-azirines. Unlike the 3-aryl-3-halogenodiazirines, the3N-compounds have not been well-studied and only onemember of the series has been previously synthesised.2We report here a number of synthetic approaches tothese compounds and evidence for two intermediates inthe photochemical generation of carbcnes from them.A preliminary account of some of this work has beenp~blished.~RESULTS AND DISCUSSIONPre$aration of 3-Aryl-3H-diazirines.-Two generalapproaches to the 3H-diazirine skeleton have beenreported.In the first, the reaction of aliphatic alde-hydes with ammonia and chloramine results in theformation of 1,3,5-triazabicyclo[3.1 .O] hexane~,~ whicharise from the condensation of first-formed diaziridinewith excess of aldehyde and ammonia [reaction (l)].Formation of the diazirine is achieved by partial acidhydrolysis of the bicyclic compound to the diaziridinewhich is trapped in situ by oxidation.Hydroxylamine-0-sulphonic acid can be used with advantage to replacethe unstable ~hloramine.~ The only reports of aromaticketones or aldehydes being used with aminating agentsand ammonia concern acetophenone (from which thediaziridine is formed in very low yield) and benz-aldehyde (from which the triazabicyclohexane isproduced in 30% yield).In attempts to produce the triaryltriazabicyclo-hexanes from a variety of p-substituted benzaldehydes(H, 9-Cl, 9-CH,, $-OCH3) with an equimolar quantity ofhydroxylaniine-0-sulphonic acid in methanolic ammoniaat or below O", only the ammonium salts of the corre-sponding 0-sulphonyl-oximes were produced along withtraces of the azines.These oximes are inert to furtherreaction with methanolic or liquid ammonia at tem-peratures at which any diaziridine formed could bestable. This behaviour is in contrast to the O-tosyl-oximes of perfluoroalkyl ketones which have beenreported to react with ammonia to give diaziridines inhigh yield.8 However, 0-tosylbenzaldoxime was foundnot to react with ammonia to yield diaziridine, and theoriginal chloramine method was investigated. Whenchloramine is used as the aminating reagent, ring closureof the N-chloro-gem-diamine [Scheme 1 ; (I)] becomesArCHO 4- RNHz + NHzX 9 ArCH,NHX 1 "HRlo x i d a t i o nA rR('Iy 1SCHEME 1 Preparation of 3-aryl-3H-diazirinescompetitive with elimination of amine [to (11)] and theresulting 3-aryl-3H-diaziridines (I I I) undergo furthercondensation with excess of aldehyde and ammonia togive 2,4,6-triaryl-1,3,5-triazabicyclo[3.1 .O]hexanes ( T V )(see Table 1).The acid-catalysed partial cleavage of alkyltriaza-bicyclohexanes to diaziridines, followed by oxidationirt sit% with chromic acid, is a well established routeto 3-alkyl-3H-diazirines.4 Aryl substitution, however,renders opening of the N-protonated diaziridine ringU .>.A.H. J. Abendroth, Angew. Chem., 1961,7S, 67.E. Schmitz and R. Ohme, Chem. Ber., 1961, 94, 2166.E. Schmitz, Chem. Ber., 1962, 95, 688.J. R. Knowles, Accoztnts Ckem. Res., 1972, 5, 156.W. A. Graham, J . Amer. Chem. SOC., 1966, 88, 4677.R. A.G. Smith and J. R. Knowles, J . Amer. Chela. Soc., B. L. Dyatkin, K. N. Makarov, and T. L. Knunyants,1973, 94, 5072. Tetrahedron. 1971, 27, 611975competitive with the hydrolysis of the remainder ofthe triazabicyclohexane system, resulting in a low yieldof diaziridine (Table 2) even if a large excess of di-chromate with minimal amounts of acid are used.TABLE 1Reactions of aromatic aldehydes with aminating reagentsand aminesYield 7Ar * R * X * Product Characterisation (%)a H OSO,H (IIa) n.m.r., C . H, N, m.s. 41b H OSO,H (IIb) n.m.r.,C, H,N,m.s. 37c H OSO,H (IIc) n.m.r., C, H, N, m.s. 35d H OSO,H (IId) n.m.r.,C, H, N, S,C1, 45e H OS0,H (IIe) n.m.r., C, H, N, S 26m.s.a H C1 (IVa) n.m.r., C, H, N, 25b H C1 (IVb) n.m.r., C, H, N, 28c H C1 (IVc) n.m.r., C, H,N, 27i.r., m.s.i.r., m.s.i.r., m.s.j H C1 Isolation of n.m.r., i.r.nitrileg But C1 (IIg) n.m.r., C, H, N, C1,i.r., m.s.j But OSO,H (IIIj) n.m.r., HI oxidationg But OSO,H Hydrazone n.m.r., i.r.j Ally1 OS0,H (IXIi) n.m.r., HI oxidation* As in Scheme I : a, phenyl; b, p-tolyl; c, p-methoxy-phenyl ; d, p-chlorophenyl; e, p-carboxyphenyl; f, p-carbosy-methoxyphenyl ; g, p-nitrophenyl ; h, m-nitrophenyl ; i, 4-pyridyl; j, 3-pyridyl.7 Based on aldehyde.TABLE 2Oxidation of aryldiaziridines andStartingmaterial * R * Oxidant t Product(IVa) (H) iii + iv (Va)11 (Vali (Vali CVb)( I W $(IIIb) $ H( I I I C ) 1 I3 i (VC)i (Vf)(III!) $ Rut ii (Vi)(IJIj) 5 Rut ii (VJ)UVa) (HH)IE2): &t ii (Vh)* As in Scheme 1 and Table 1.triazabic yclohexanesYieldCharacterisation (%)n.m.r., u.v., i.r., 4n.rn.r., u.v., i.r.48U.V. 1n.m.r., u.v., i.r., 3.6n.m.r., u.v., i.r. 1.9n.m.r., u.v., i.r. 1.0n.m.r., u.v., i.r. 3.3n.m.r., u.v., i.r. 16n.m.r., u.v., i.r. 8.4t i, HaO; ii, ButOC1:m.s.ms.iii, H+; iv, H2Cr,0,. $ Generated in &tu from aldehyde.Other oxidants that were tried in the hope of trappingthe incipient diaziridine more effectively were H202-Cu2+(cf. ref. lo), lead tetra-acetate, N-bromosuccinimide,and peracetic acid. Direct oxidative cleavage of thetriphenyltriazabicyclohexane was achieved, however,with t-butyl hypochlorite in methanol at O", yielding3-phenyl-3H-diazirine in 48% yield. Unfortunately,this appears not to be a general method since para-substituted arylbicyclohexanes (e.g.fi-CH,,ap-OCH3) giveno diazirine with this reagent. The sensitivity of thisreaction to the ring substituent can be attributed to therelative stabilities of the cations (VI) and (VII) formedby loss of chloride ion from N-3.A. T. Nielsen, R. L. Atkins, D. W. Moore, D. Mallory, andlo S. Hunig, H.-R. Miiller, and W. Thier, Tetrahedron Letten,J. M. Laberge, Tetrahedron Letters, 1973, 1167.1961, 353.More success attended the oxidative trapping of thediaziridine on its way to the bicyclohexane, rather thanas a product of acid-catalysed bicyclohexane hydrolysis.A r C H ,eHArP NA r,CH+N-NIA r c ? ,CHAr'NThus yellow mercuric oxide is sufficiently stable toammonia and chloramine and yet adequately effectiveas an oxidant of diaziridine, to be added directly to thereaction mixture of arenecarbaldehyde-methanolic am-monia-chloramine.The yields of diazirine based onaldehyde are low, but the simplicity of the work-upprocedure gives this reaction some synthetic utility.Oxidative trapping of the diaziridine in this waysupports the view that the diaairidine ring is formedbefore a sywz-hexahydrotriazine ~ystem.~The second diazirine synthesis, due to Graham,llinvolves the reaction of an alkylimine with a dihalogeno-amine, which is followed by the elimination of alkylhalide [reaction (Z)]. Diazirine itself [reaction (2 ;R1 = R2 = H)] can be prepared by this reaction,difluoramine giving better yields than dichloramine.If R1 or R2 is electron-releasing, the stabilisation of thecarbonium ion (VIII) leads to predominant formation ofthe azo-compound (IX) [reaction (3)].3- ($-Nitropheny1)-3H-diazirine was the only knownexample of a 3-aryl-3H-diazirine, and was prepared byreaction (Z), with difluoramine.Since difluoramine is avery hazardous and expensive reagent, dichloramine wasused in the present work. No diazirine resulted fromreactions in which dichloramine [generated from sodiumhypochlorite and ammonia (cf. ref. 12) was extractedinto an organic layer containing aryl-t-butylimine, andthis approach was modified by preforming the 3-aryl-l-t-butyl-3H-diaziridine and following this by N-chlorination [reaction (4)]. 3-Aryl-1-t-butyl-3H-diaziri-dines are thermally unstable, and although the 3- and4-pyridyl compounds were obtained in low yield from thecorresponding aldehydes, extensive decomposition oc-curred during purification.From 9-nitrobenzaldehydet-butylimine and hydroxylamine-0-sulphonic acid, onlythe t-butylhydrazone of p-nitrobenzaldehyde resulted.Since the isolation of these t-butyldiaziridines provedimpractical, it was decided to trap them by N-chlorin-ation in sit%. The putative diaziridines are generatedby reaction of the t-butylimines with hydroxylamine-0-sulphonic acid in ethanol-water-base at 0". After anappropriate time (commonly 2-3 h) t-butyl hypochloriteis added directly at - 1 0 4 " . Preparative t.1.c.suffices to purify the resulting diazirines from thecomplex mixture of products.3-(3-Pyridyl)-3H-, 3-(4-pyridyl) -3H-, and 3- (m-nitrophenyl) -3N-diazirines were11 W. H. Graham, J . Amer. Chem. SOC., 1962, $4, 1063; W. H.Graham, J . Amer. Chem. SOC., 1966,88,4677.W. €3. Graham, J . Oyg. Chem., 1965,30, 2108688 J.C.S. Perkin I1prepared by this methad in yields of 5-15%. Despitethe complexity of the product mixture, the ease ofisolation makes this a practicable route to electron-withdrawing 3-aryl-3H-diazirines.To compare the photochemical behaviour of the 3H-diazirines with the more readily accessible S-halogeno-diazirines (see below), some 3-aryl-3-halogenodiazirineswere prepared. The method followed was that ofhalogenation of amidines in the presence of halide ionmarkedly with the corresponding linear diazo-compoundisomers, and augurs well for their use as reagents for thephotolabelling of biological macromolecules.Photolysis of 3-A ryZ-3H-diazirines.-It is clearly im-portant to understand, at least in outline, the processesconsequent upon the irradiation of a potential photo-labelling reagent.The ease of photolysis, the inter-mediacy of reactive species, and the selectivity of thesespecies are all features that require evaluation.N H 2 X , N B ~ B ~ O C ~ ”But /NA r C H II ( L l‘NArCH I‘NC tA r c H=N B u A r CH,iHfirst reported by Graham l3 [reaction @)I. The 3-phenyl- and 3-~-methoxyphenyl-3-chloro-diazirine wereprepared.Properties of the 3-Aryldiazi~ines.-With the exceptionof the p-carboxymethoxyphenyl compound, all thearyldiazirines prepared were clear yellow oils, of limitedstability in the pure state at or above room temperature.They can, however, be stored conveniently as dilute( < O .~ M ) solutions in inert solvents such as hexane, andare stable in this form for several months at -20”. Thewater-soluble diazirines (p-carboxymethoxyphenyl, and3- and 4-pyridyl) were stable in aqueous solution forhours in 1M-acetic acid or 1;M-NaOH: the half-life of thep-carboxymethoxyphenyl derivative in O.1br-NaOH beingca. 30 h a t 37”. 3-Phenyl-3H-diazirine does not reactwith cyclopentadiene a t room temperature and is onlyslowly affected by dithionite, in contrast to linear azo-compounds. Unlike diazoalkanes and diazoketones,the decomposition of the aryldiazirines is not significantlycatalysed by cupric ion. The position of the long-wavelength U.V.absorption band of the diazirines,characteristic of the azo-chromophore, depends on thesubstituents in the aromatic ring and there is a markedbathochromic shift of this band on protonation of thepyridine ring of the 3- and 4-pyridyl compounds. Theseobservations suggest some interaction of the x-electronsystems of the diazirine group and the aromatic ring.The chemical stability of these diazirines contrastsl3 W. H. Graham, J . Amer. Chem. SOG., 1966, 8’4, 4396.l4 H. M. Frey and I. D. R. Stevens, Proc. Chem. SOL, 1962, 79.M. J. Amrich and J. A. Bell, J. Amer. Chem. SOL, 1964, 86,292.On photolysis of diazirine itself, singlet methylene isproduced,l* and it has been proposed that at least 20%of the diazirine is photoisomerised to diazomethaneen route to meth~1ene.l~ In 3-alkyl-3H, 3,3-dialkyl, andother diazirines with a-hydrogen atoms, photolysisnormally produces olefins,l6 presumably by rapidhydrogen migration following primary carbene f o m -ation.Small amounts of intramolecular carbene in-sertion products have also been observed. Cycloheptyl-and cyclo-octyldiazirines have been reported t o photo-isomerise to the diazo-compounds, photolysis in thepresence of acetic acid yielding significant amounts ofthe cycloalkyl acetates.17When 3-phenyl-3H-diazirine is irradiated in hexanesolution, there is an immediate rapid rise in the U.V.absorbance a t ca.275,362, (this latter being the diazirineAm=) and 480 nm, followed by a fall in intensity of allthese bands (Figures 1 and 2). Successive spectra donot show isosbestic points. Provided that the concen-tration of absorbing species is sufficiently low that theabsorption of the incident radiation by the solution doesnot significantly affect the radiation intensity over thebulk of the sample, the kinetics of the absorbancechanges approximate to that expected for sequentialfirst-order processes (Figure 2). Evidently the inter-mediate that accumulates early in the photolysis is alsophotolabile, although less so than the diazirine itself.The spectral changes are independent of diazirine con-l6 H. M. Frey, Adv. Photochem., 1966, #, 226.l7 G.F. Bradley, Ph.D. Thefis, University of Southampton,1967 (quoted in W. Kirmse, Carbene Chemistry,’ AcademicPress, New York, 1971)1976 689centration and of temperature (-60" and room tem-perature were studied). Other 3-aryl-3H-diazirinesbehave similarly, though the effect is smaller for (e.g.)the P-tolyl compound, and for the +-methoxyphenylderivative no initial increase in absorbance is observed,though there are still marked deviations from first-orderkinetics.?alCU9,v1 n4 00 2 56h f n mFIGURE 1 Successive U.V. spectra of 3-phenyl-3H-diazirine(0.374mM) in hexane on irradiation : a, initial spectrum ;b-d, spectra after 1, 3, and 10 min respectively.cG. cYIrradiation time (s)FIGURE 2 Time dependence of the absorbance a t 362 nmfor the irradiation of 3-phenyl-3H-diazirine (see Figure 1)The new absorption bands that appear at ca.275 and480 nm correspond exactly to those reported for aryldiazo-compounds,l* and two tests were applied to con-firm the intermediacy of the linear diazo-compound indiazirine photolysis. First, in the presence of adequateacetic acid, diazo-compound should be destroyed fasterthan it is formed. For the 3-phenyl and 3-fl-tolylcompounds it was found that at acetic acid concen-trations of 0*1--0-5~, clean first-order kinetics wereobeyed for the disappearance of the diazirine absorptionband (Figure 3), and successive U.V. spectra showedisosbestic points. Secondly, when 3-P-tolyl-3H-diazirinewas photolysed as a liquid film between sodium chloridediscs or as a solution in carbon tetrachloride, thediazirine absorption at 1600 cm-l disappeared con-comitantly with the appearance and subsequent decayof the characteristic N=N stretching absorption of thediazo-compound a t ca.2000 cm-l. These experimentsconfirm that the diazirines are at least partly photo-isomerised to the linear diazo-compound during photo-lysis.Two questions now arise: can diazirine be formedphotochemically from diazo-compound in significant-0.6 j-- 10,[510d-1.6 0 j b 100 I 200Irradiation time ( 5 )FIGURE 3 Time dependence of the absorbance at the diazirineAmax. during the photolysis of 3-phenyl-3H-diazirine ( 0)(362 nm; 0-304m~ in hexane containing 0-lM-acetic acid) ;and of 3-(fi-tolyl)-3H-diazirine (0) (368 nm; 0.69m~ inhexane containing O-S~-acetic acid)amounts, and does the diazirine also eliminate N,directly to the arylcarbene? On irradiation of phenyl-diazomethane under conditions similar to those used forthe diazirine photolysis, no diazirine absorption couldbe detected in the U.V.If a photochemical equilibriumis set up between diazirine and diazo-compound, it isclearly (as expected) in favour of the diazo-compound.To investigate whether the diazirine fragments directly,the products from irradiating the diazirine in hexanecontaining varying amounts of acetic acid, were in-vestigated. The products are almost exclusively thecorresponding benzyl acetate, and a mixture of aryl-heptanes. The ratio of benzyl acetate (from thereaction of diazo-compound with acetic acid) to aryl-heptane (from carbene insertion into solvent) is shownin Figure 4.The product ratios were obtained fromn.m.r. of the crude photolysis reaction mixture afterconcentration. Confirmation of the presence of the18 G. L. Closs and R. A. Moss, J . Amer. Chew. SOL, 1964, 86,4042690 J.C.S. Perkin 11benzyl acetates and arylheptanes was provided by mass The photolytic pathway derived so far (Scheme 2) is,spectrometry and i.r. spectroscopy. It is apparent from however, inadequate to explain the following. Ifsolutions of 3-phenyl-3H-diazirine in hexane are irradi-ated briefly until the maximal increase in absorbance at275, 362, and 480 nm has occurred, and then acetic acidis added in the dark to destroy all the phenyldiazo-methane, the resulting spectrum shows that the bandsat 275 and 480 nm have disappeared, but there is nochange in the increased absorbance at 362 nm.Anotherintermediate, in addition to phenyldiazomethane, istherefore implicated. Since no anomalies are seen inthe photolysis of diazirines in the presence of acid, thissecond intermediate must be photochemically connectedto an acid-labile substance, presumably the diazo-compound.By using relatively large amounts of diazirine at lowconcentration, this second intermediate could begenerated in small though significant quantity. Analysisby t.1.c. showed a component that migrates very closeto the unchanged diazirine. This component had thefollowing properties: A,, (n-hexane) 233 and 355 nrn(no fine structure); v,,, 1604 cm-l (I N=N); T 2.8 and3.1 (4H, m), 4.2br ( l H , s), and 7.1 (lH, s).On irradi-ation it appears from the changes in the U.V. that this/I 0[Acetic acidl / MFIGURE 4 Variation in product composition with added aceticacid, for the photolysis of various 3-aryl-3H-diazirines inhexane : phenyl, 0 ; p-tolyl, 0 ; pmethoxyphenyl, 0 ;4-pyridyl,Figure 4 that up to ca. 0.2M acid the proportion of benzylacetate increases rapidly with increasing acid concen-tration, indicating that low acid concentrations areinsufficient to convert diazo-compound to acetate (byprotonation and nitrogen loss) before it is itself photo-lysed. The observation of plateaux in the curvesindicates an upper limit for conversion of diazirine toacetate via diazo-compound and suggests that aboveca.0-&-acetic acid, the acetate : heptane ratio representsthe true partition ratio of diazirine that photoisomerisesto diazo-compound and diazirine that photolysesdirectly. The plateaux of Figure 4 are not flat (as isshown for the unsubstituted compound) and this ispresumably due to the direct insertion of arylcarbeneinto the 0-H bond of acetic acid. This implies someelectrophilic selectivity of the arylcarbene favouring0-H insertion, since even at 2~-acetic acid, the molarratio of C-H to 0-H bonds is ca. 50: 1. From theseresults we may conclude that the diazirine does indeedfragment to arylcarbene directly, the direct fragment-ation being about as probable as the photoisomerisationprocess (Scheme 2).It can also be seen from Figure 4 that the photo-isomerisation competes more effectively if electron-releasing substituents are present, and it is likely thatthe transition state for the diazirine-diazo-compoundconversion has some dipolar character [reaction (S)].ArCH’!‘Nr l *A r C H + h zI / --- - I / A c O H / - - AcOH //1 h e x a n ecSCHEME 2 Simplified scheme for the photolysis of3-aryl-3H-diazirincsmaterial can be converted photochemically to the diazo-compound, since an acid-sensitive absorption band at273 nm appears at early stages of the photolysis(Figure 5).This second intermediate has spectroscopic propertiesconsistent with its being an azo-compound, and since itis generated from 3-phenyl-323-diazirine and undergoesapparent photolysis to phenyldiazomethane structure(X), 7aH-indazole, is tentatively assigned to it.Noexamples of 7aH-indazoles are known, though thesc:species have been suggested as intermediates in thee t 1975 691pyrolytic generation of carbenes from 1H-inda~oles.~~Despite the high reactivity expected of such a structure,QYH(XIthe number of plausible alternatives is limited. Benzo-nitrile imine ( P h d - R H ) -a+- PhC=A=NH) 2o is a veryreactive 1,3-dipole that has defied isolation and does notgive rise to carbene insertion products. 2-H- and3a-H-Indazoles should be as prone to rearrangement as7aH-indazole, and are not compatible with the n.1n.r.- 0 ' 5 rhe, ucdn L0v) nu0,m0dI;IGURE 5-1.- 2 .Irradiation time (s)Time dependence of the absorbance a t 273 (a) and a t360 nm (0) of the photolysis of the second intermediate (X)(see text) (ca.1mM) in hexane containing no (a) or 0.2n,r-acetic acid (0)data. Polymerisation products would not be expectedto show the observed photolysis behaviour. Wherethis second intermediate lies in Scheme 2 is not clear,but the ultimate formation of arylcarbenes in highchemical yield from the photolysis of 3-aryl-3H-diazirinesis without doubt.In contrast to the photochemical behaviour of the3-aryl-3H-diazirines described, the analogous 3-aryl-3-halogenodiazirines show no spectral peculiarities onirradiation.Acetic acid is without effect, and fii-st-order decay of the U.V. absorption bands with cleanlQ W. P. Crow and M. W. Paddon-Row, Tetvahedvon Letters,2o P. Scheiner and J. F. Dinda, Tetrahedron, 1970, 26, 2619;21 G. L. Closs and J. Coyle, J . Amer. Chew. Sac., 1962, 84, 4360.22 M. T. H. Liu and D. H. T. Chen, J.C.S. Perkin 11, 1974, 937.23 A. E. Ruoho, H. Kiefer, P. E. Roeder, and S. J. Singer,Pvoc. Nut. Acad. Sci. U.S.A., 1973, 70, 2667.24 G. W. J. Fleet, J. R. Knowles, and R. R. Porter, Biochem. J . ,1972, 128, 499; R. -4. G. Smith and J. R. Knowles, ibid., 1974,141, 61.1973, 2217.P. Scheiner, J . Org. Chcna., 1969, 34, 199.isosbestic points are observed. This suggests that forthe 3-halogeno-compounds either no photoisomerisationoccurs (as has been implicitly assumed in previousstudies of these compounds) or that the cr-halogenodiazo-compounds decompose very rapidly.Chlorodiazo-methane is known to be thermally most unstable.%lVery recently, Liu and Chen2a have reported thethermolysis of some 3-aryl-3-chlorodiazirines in cyclo-hexene. Their conclusions concerning the mode ofbreakdown of aryldiazirines are in broad agreementwith the photochemical results presented here.Recent work on the mechanism of photoaffinitylabelling has emphasised the importance of generating areaction intermediate whose lifetime is less than themolecule's residence time within its receptor site.23 Itis likely that the shorter life-times and more indis-criminate reactivity of arylcarbenes may make themmore suitable intermediates for such experiments thanthe arylnitrenes used in earlier work from this labora-tory.24 Furthermore, bonds formed in labelled materialby arylcarbenes are not likely to be labile, as are somethat may be formed in nitrene insertion reactions. Thearyldiazirine-arylcarbene system should also score overthe a-ketodiazo-compound-a-ketocarbene conversionused by a number of investigator^,^^ both in respect ofthe lesser chemical reactivity of the diazirine precursorand of the absence of wasteful molecular rearrangement(to keten) which a-ketocarbenes undergo.EXPERIMENTALlH N.m.r.spectra were obtained on a Perkin-Elmer R14100 MHz spectrometer a t ambient probe temperature withtetramethylsilane as internal reference and CC1, or [2H,]-dimethyl sulphoxide (DMSO) as solvent.Mass spectrawere measured on an A.E.I. MS9 spectrometer. Analyticaland preparative t.1.c. was carried out on silica gel platesusing the following solvent systems : ( a ) light petroleum-acetone 4 : 1 (v/v); (b) light petroleum-chloroform 1 : 1(v/v) ; (c) light petroleum-acetone 2 : 1 (v/v) ; (d) n-hexane;(e) n-hexane-acetone 2 : 1 (v/v) ; (f) n-hexane-chloroform1 : 1 (v/v); (g) benzene-glacial acetic acid 6 : 1 (v/v);(h) n-hexane-acetone 10 : 7 (v/v) ; (i) chloroform; ( j ) lightpetroleum-acetone 5 : 1 (v/v) ; ( K ) benzene-glacial aceticacid 10 : 1 (v/v); (1) light petroleum-acetone 17 : 13 (v/v).N-Benzylideizehydroxylanzine-Q-sulphonic Acid Amrnon-iunz Salt (IIa; X = OSO,H).-Benzaldehyde (21.2 g) wasdissolved in ammonia-saturated methanol (500 ml) andcooled to - 70".Hydroxylamine-O-sulphonic acid (22.6 g)was added over 30 min and the mixture allowed to warm toroom temperature over 4 h. The material was thenevaporated to dryness, washed with ether, and recrystallisedfrom ethanol-methanol 4 : 1 (v/v) to give crystals (17.8 g),1n.p. 156", T (DMSO) 1.73 (lH, s, CH-N), 2-33 (2H, m,aromatic), 2.55 (3H, m, aromatic), and 2.88br (s, exchange-able NH), w / e 103, A,, (ethanol) 206 (E 19,000), 250 (19,400),25 E.g. A. Singh, E. K. Thornton, and F. H. Westheimer, J .Uiol. Chem., 1962, 237, PC3006; J. Shafer, P. Baronowsky,xi. Laursen, F. Finn, and F. H. Westheimer, ibid., 1966, 241, 421;R. W. Rosenstein and F. F.Richards, J . Immunology, 1972, 108,1467; U. Das Gupta and J. S. Rieske, Biochem. Biophys. Res.Comm., 1973, 54, 1247; J. A. Katzenellenbogen, H. J. Johnson,and €3. N. Myers, Biochemistry, 1973, 12, 4086; D. J. Brunswickand B. S. Cooperman, ibid., p. 4074; B. S. Cooperman and D. J.Brunswick, ibid., p. 4079J.C.S. Perkin I1277 (2400), and 288 (1400) nm, vms. (mull) 3250 cm-1 (NH)(Found: C, 37.3; H, 4-7; N, 13-0. C7H,,N204S requiresC , 38.5; H, 4.6; N, 12-8%), approximate M (osmometry)191. The material liberated iodine from HI solutions and,on heating alone to 120°, decomposed rapidly t o benzo-nitrile (i.r.).N-p-MethylbenzylideneJtydroxylamine-O-sulphonic AcidA~~lmoniurn Salt (IIb; X = OSO,H).-This compound wasprepared analogously to the unsubstituted material, yellow$latelets, m.p.154", T (DMSO) 1.75 (lH, s, CH=K), 2.36(2H, d, 2- and 6-H), 2.70 (2H, d, 3- and 5-H), 3.00br (s,exchangeable NH), and 7.67 (3H, s, CH,), 7n/e 117, A,=.(EtOH) 211 (E 27,600), 253 (26,400), 284 (12,600), 293(9400), and 306 (600) nm, v,, (mull) 3200 cm-1 (NH)(Found: C, 42-8; H, 5.5; N, 12.3. C,HI2N,O,S requiresC, 41-4; H, 5.2; N, 12.1%). In a separate experiment thecrude reaction mixture was poured into water, and themixture extracted with ch1o;oform. The combined ex-tracts were dried (Na,SO,), evaporated to dryness, and theresulting solid recrystallised from methanol to give 9-tolualdehyde azine as yellow plates, m.p. 160" (lit.,26 157")(Found: C, 81-2; H, 6.8; N, 11.7. Calc. for C,,H,,N,:C, 81.4; H, 6.8; N, ll*9yo).N-p-Methoxybe.lazylidene~ydroxyl~~i~~e-O-sulp?~o.laic AcidA~zmonium Salt (IIc; X = OSO,H).-This compound wasprepared analogously, yellow crystals, m.p.128-129'(decornp.), T (DMSO) 1.82 (lH, s, CH=N), 2.39 (2H, d,2- and 6-H), 2-93br (s, exchangeable NH), and 3.00 (2H, d,3- and 5-H), methyl resonance obscured by DMSO, rn/e 133,Aim, (ethanol) 205 (E 24,200), 215 (24,000), 270 (29,600),291 (14,000), and 300 (10,000) nm, vmax. (mull) 3250 cm-l(NH) (Found: C, 40-0; H, 5.0; N, 11.2. C,H,,N,O,Srequires C, 38.8; H, 4.8; N, 11.3y0).N-p-Chlorobenzylideneizydvoxylamine-0-sulphonic AcidAmmo~~ium Salt (IId; X = OSO,H).-This compound wasprepared analogously, plates, m.p. 170", T (DMSO) 1.70(lH, s, CHzN), 2.28 (2H, d, 2- and 6-H), 2.49 (ZH, d,3- and 5-H), and 3-00br (s, exchangeable NH), m/e 135,A,, (ethanol) 206 (E 25,000), 217 (lS,OOO), 260 (28,400),284 (4800), and 294 (2800) nm, vmX.(mull) 3240 cm-1(Found: C, 33.0; H, 3.4; N, 11.9; C1, 13.7; S, 12.8.C,H,ClN,O,S requires C, 33-2; H, 3-6; N, 11.1; C1, 14.0;S, 12.7%).p-Hydroxysulpho.nyloxyiminotoluic Acid MonoammoniumSalt (IIe; X = OSO,H).-The reaction mixture from p-formylbenzoic acid (6 g), methanolic ammonia (330 ml), andhydroxylamine-O-sulphonic acid (5.6 g) was evaporated todryness, suspended in ~N-H,SO, (30 ml), filtered, and dried.The product was dissolved in hot ethanol-methanol 1 : 1(v/v), the solution was filtered, and the filtrate partiallyevaporated. The resulting Fvecipitate was filtered and dried(2.2 g ) , m.p.220°, T (DMSO) 2.63 (lH, s, CH=N), 2.96 (2H,d, J 7 Hz, 2- and 6-H), 3-18 (ZH, d, J 7 Hz, 3- and 5-H),and 3.7-4.lbr (envelope, exchangeable NH and CO,H),A,, (ethanol) 204, 265, and 298 nm, v,, (mull) 3240 and1700 cm-l (Found: C, 36.5; H, 3.5; N, 10.4; S, 12-6.C,HI6O,N,S requires C, 36.7; H, 3.8; N, 10.7; S, 12.2y0).1,3,5-Tripl~e~yl-2,4,6-triazabicyclo[3.1 .O]hexane (IVa) .-This compound was prepared as described by Schmitz.?1,3,5-Tri-(p-tolyl~-2,4,6-triaxabicyclo[3.1.0] hexane (IVb) .-Chloramine was prepared by cautious addition of t-butylhypochlorite (6 ml) to methanolic ammonia (100 ml) at-60". p-Tolualdehyde (10 g ) was added and the stirred26 H. E. Zimmerman and S. Somasekhara, J . Amer. Chem. SOL,1960, 82, 6865.solution allowed to warm t o 4" overnight.A solid (2.26 g)was precipitated, and this was recrystallised from ethanol-methanol and light petroleum-ethanol to give needles, m.p.175", RF 0.7 [system ( j ) ] , oxidising to HI, 7 (DMSO) 2.4-2.9 (12H, m, aromatic), 4.47 (lH, t, J 5 Hz, s on additionof 2H,0,CH), 5.13 (lH, t, J 5 Hz, s on addition of 2H,0,CH),5.64 (lH, m, exchangeable NH), 6.41 (IH, s, CH), and 7-70(9H, s, CH,), m/e 236, 208, 192, 149, 145, and 119, A,.(ethanol) 205, 214, 222, 264, and 272 nm (Found: C, 80.8;H, 7.1; N, 11.7. C,,H,,N, requires C, 81.1; H, 7-1; K,1,3,5-Tris-(p-methoxyphenyl) -2,4,6-triazubicyclo[3.1. OJ-hexane (IVc) .-Chloramine was prepared from t-butylhypochlorite (10 ml) and methanolic ammonia (150 ml) a t-60".p-Anisaldehyde (8.35 g ) was added, and themixture allowed to warm to room temperature over 48 h.A white solid (2.36 g ) precipitated and was recrystallisedfrom ethanol to give material A (the triazabicyclohexane),m.p. 120-122", and a yellow solid B, m.p. 169O, RF [systemaromatic), 4.37 (lH, s, CH), 4.77 (1H, s, CEI), 8-15 (9H, s,OCH,), and 6.79 (lH, s, CH), m/e 268, 254, 241, 225, and134, vmx. 3306 and 3290 cm-1 (Found: C, 69-1; H, 6.2;N, 10.4. C,,H,,N,O, requires C, 69.7; H, 6.1; N, 10.27b).and 6-H), 3.04 (2H, d, J 9 Hz, 3- and 5-H), and 6.12 (3H, s,OCH,), vn/e 268, 240, 225, 161, and 134. The spectralproperties of B are consistent with those of the azine (lit.,27m.p, 168").Reactiovz of Pyridine-3-carbaldehyde with Chloramine andA ~nmoniu.-Pyridine-3-carbaldehyde ( 15 g) was added tochloramine [from niethanolic ammonia (250 ml) and t-butylhypochlorite (15 ml)] a t - 60".The mixture was allowedto warm to -20" and maintained at this temperature for16 h. After warming to room temperature, the mixturewas concentrated by rotary evaporation. T.1.c. {system(a)] showed components with RF 0.00, 0.06, 0.16, 0.20, 0.43(aldehyde), and 0-60, of which the first and third showed HIoxidising power which was lost on standing a t room tem-perature. Components with RP 0-60 ( A ) and 0-2 (B) wereisolated by preparative t.1.c. [system ( a ) ] . A 4 : "; (DMSO)J 9 Hz, 4-H), and 2-70 (lH, m, 5-H), vm,+ (CC1,) 2240 cm-l,m.p. 46". A is pyridine-3-carbonitrile (lit.,28 m.p. 48-49').(lH, d, J 7 Hz, 6-H), 1.71 (lH, d, J 9 Hz, 4-H), and 2.46(lH, m, 5-H), m/e 210 (M+, loo%), 182, 163, 156, 132, 120,and 105, m.p.133-136". B is pyridine-3-carbaldehydeazine (lit.,20 m.p. 148").Reaction of p-Nitrobemaldehyde t-Butylimine with Chlor-awzine.-+-Nitrobenzaldehyde was condensed with t-butyl-amine using azeotropic removal of water with benzene,m.p. (from hexane) 75". Imine (3.09 g) in methanol (25 ml)was added to a chloramine solution [from methanolicammonia (60 ml) and t-butyl hypochlorite (15 ml)] a t-40" and allowed t o warm t o room temperature. Theresulting precipitate was recrystallised from ethanol togive pale yellow plates (1.76 g), T 1.1 (lH, s, CH=N) and1.7-2.14 (4H, q, aromatic), A,, (EtOH) 281 (E 20,000) nm,no NH i.r.band, m/e 184/186, 154/156, 149, and 119{Found: C, 45.0; H, 3.0; N, 14-9; C1, 18.8. C7H,ClN,0,requires [for (IIg)] C, 45.5; H, 2.7; N, 15-2; C1, 19.3%).1 1.8yo).(b)] 0.09 ( A ) , 0.46 ( B ) . A : 7 (DMSO) 2.3-3.1 (12H, 111,B : T (DMSO) 1.38 (lH, S, CH=N), 2-20 (2H, d, J 9 Hz, 2-1.21 (lH, S, 2-H), 1.31 (lH, d, J 6 Hz, 6-H), 2-16 (lH, d,B : T (DMSO) 0.95 (lH, S, CH=N), 1.20 (1H, S, 2-H), 1.2827 G. Knopfer, Munatsh., 1909,30, 32.28 0. Fischer, Ber., 1882,15, 63.2o F. J. Allan and G. G. Allan, J . Org. Chem., 1958, 23, 6391975 693Reaction of p-Nitrobenzaldehyde t-Butylimine withHydroxylanai.Pze-O-suEphonic Acid.-Hydroxylamine-O-sul-phonic acid (2-5 g ) in methanol (10 ml) was added top-nitrobenzaldehyde t-butylimine (3.9 g) in methanol(80 ml) a t -60".The solution was allowed to warm toroom temperature and stirred over 18 h. Evaporation ofthe filtered solution and recrystallisation of the residuefrom hexane gave an almost white solid, 1n.p. 130°, z 1.60(lH, s, CH=N), 1.70 (2H, d, J 9 Hz, aromatic), 2.10 (2H, d,J 9 Hz, aromatic), 2-30br (s, NH), and 8.74 (9H, s, t-butyl),vm, 1627 cm-1, not oxidising to HI. This is believed to bethe t-butylhydrazone of P-nitrobenzaldehyde.Reaction of Pyridine-3-carbaldehyde t-Butylimine withHyduoxylami.Pae-O-szlphonic A cid .-Pyridine- 3-carbaldehy de(4-25 g ) was refluxed for 8 h with t-butylamine (15 ml)in ethanol (10 ml) and benzene (40 ml). Solvents andexcess of t-butylamine were removed by rotary evaporationand the resulting oil dissolved in ethanol (50 ml), water(50 ml), and t-butylamine (10 ml).Hydroxylamine-0-sulphonic acid (8.0 g ) was added over 15 min and thetemperature rose to 45'. The mixture was cooled, filtered,and the filtrate concentrated by rotary evaporation. Atenth of this mixture was purified by t.1.c. [system (c)] togive a yellow oil (0.14 g ; RF 0.3) oxidising to HI. Thissubstance rapidly decomposed on standing and was re-purified [t.l.c. system (c)] to give 43 mg of a mixture ofpyridine-3-carbaldehyde (n.m.r., i.r.) and an oxidising oil(IIIj), T (DMSO) 1.44 (lH, s, 2-H), 1.56 (lH, d, J 6 Hz,6-H), 2.40 (lH, m, 4-H), 2.87 (lH, m, 5-H), 6.53 (lH, d,J 7 Hz, diaziridine CH), and 8.95 (9H, s, t-butyl).Reaction of Pyvidine-4-carbaldehyde Allylimine withHydroxylamine-O-sulphonic A cid .-Pyridine-4-carbaldehyde(5.35 g ) was refluxed with allylamine (5.6 g ) , in benzene(40 ml), and ethanol (10 nil) for 3 h.The solvents werethen removed by evaporation, and the residue fractionallydistilled to give pyridine-4-carbaldehyde allylimine as ayellow oil, b.p. 88-90' a t 1 mmHg. Allylamine (11.4 g),ethanol (30 ml), and water (10 ml) were cooled to -5" andhydroxylamine-O-sulphonic acid (2.26 g) added. To thismixture the imine (1.46 g ) in ethanol (10 ml) was addedslowly, the temperature being maintained below 5". Afterstirring a t 0" for 4 h, ethanol was removed by evaporation.The residue showed an oxidising component, RF 0-28[system ( c ) ] . A portion of the mixture was purified usingthe same t.1.c.system, and resulted in the isolation of ayellow oil (40 mg), z (DMSO) 1.50 (2H, d, J 5 Hz, 2- and4.15 (lH, octet, allyl CH), 4.80 (2H, m, allyl CH,), and6.80 (3H, saperimposed d, CH, and diaziridine CH).3-Phenyl-3H-diazirine (Vat .-1,3,5-Tripheny1-2,4,6-ti-i-azabicyclo[3.l.O]hexane (IVa) (15.65 g) was dissolved inacetone (250 ml) and added dropwise a t 0" to a mixture ofsodium dichromate (26-2 g ) , acetone (125 ml), water(125 ml), and concentrated H,SO, (1 ml). After stirringin the dark a t 4" for 12 h, the mixture was poured into asolution of sodium metabisulphite (400 g ) in water (3 1)and the suspension extracted into n-hexane (6 x 50 ml).The combined extracts were dried (Na,SO,) , concentratedby evaporation below 20°, and purified by t.1.c.[system (a)].The component of Rv 0.9 was eluted with ether and thesolvent evaporated to leave a yellow oil (0-14 g ) , T (DMSO)2.95 (3H, irregular t , aryl), 3.32 (ZH, irregular d, aryl),and 8.24 (lH, s, diazirine 3-H), Amax, (hexane) 253 (E 294),259 (314), 265 (371), 272 (274), 276 (220), 344 (179), 354(230), 362 (299), 373 (230), 382 (245) nm, fine structure was6-H), 2.62 (lH, d, J 6 Hz, NH), 2.80 (2H, d, J 5 Hz, 3-H),less apparent in ether solution, vmxa (film) 1580 (N=N) and987 cm-l, m/e 90 (M+ - N,),Very much better yields were obtained by oxidising the1,3,5-triphenyl-2,4,6-triazabicyclo[3.1. Olhexane (IVa) witht-butyl hypochlorite. Triazabicyclohexane (4 g ) was sus-pended in methanol (30 ml) a t 0" and t-butyl hypochlorite(2 ml) in t-butanol (5 ml) added dropwise over 10 min.After stirring a t 5" for 1 h, the mixture was poured intosodium metabisulphite solution [50 g in water (500 ml)]and extracted into n-hexane (4 x 20 ml).T.1.c. [system(a)] showed mainly (Va) (yield based on E,,, IM, 0.72 g).Surprisingly, reaction of 1,3,5-tri-(p-tolyl)-2,4,6-triaza-bicyclo[3.1.O]hexane ( I n ) with t-butyl hypochlorite, didnot yield the tolyldiazirine (Vb) , but largely a-chloro-p-tolualdehyde 9-methylbenzylidenehydrazone.Oxidative trapping of 3-phenyl-3H-diaziridine could beachieved with yellow mercuric oxide. Benzaldehyde (20 g )was added to a solution of chloramine prepared frommethanolic ammonia (150 ml) and t-butyl hypochlorite(25 ml) in t-butanol (25 ml), a t -60". The mixture wasallowed to warm to room temperature over 4 h.Yellowmercuric oxide (20 g) was added and the mixture stirred,more oxidant being added after 12 (5 g) and 30 h (10 g).The mixture was filtered, concentrated, and added to asolution of sodium metabisulphite (150 g ) in water (1.5 1).Extraction into cyclohexane (2 x 50 ml) and purificationof the extracts by t.1.c. [system (a)] gave a low yield ofmaterial identical (t.1.c. and u.v.) with (Va).3-(p-ToZyZ)-SH-diazirine (Vb) .-p-Tolualdehyde (12 g )was added to a solution of chloramine prepared frommethanolic ammonia (200 ml) and t-butyl hypochlorite(30 ml) a t -40" and the mixture allowed to warm to roomtemperature over 2 h. Yellow mercuric oxide (20 g ) wasadded and stirring continued a t 4" in the dark.After 12 h,more oxidant (20 g ) was added. After 8 h a t 4O, the mixturewas filtered, the filtrate concentrated by evaporation, andpoured into ice-water (1 1). The product was extractedinto n-hexane (3 x 50 ml). Concentrated extracts werepurified by t.1.c. [system (d)] to give the diazirine as ayellow oil (0.45 g ) , T (DMSO) 2.97 (ZH, d, J 8 Hz, aromatic),3.18 (2H, d, J 8 Hz, aromatic), 7.68 (3H, s, CH,), and 8.12{lH, s, diazirine 3-H), A,, (hexane) 272 (E 319), 281 (297),350 (183), 368 (311), 380 (232), 389 (253) nm, m / e 104,v,, (CHC1,) 1620 and 1600 cm-1.3- (p-Methoxyphenyl) - 3H-diazirine (Vc) .+-Anisaldehyde(20 g) was added to a solution of chloramine prepared frommethanolic ammonia (250 ml) and t-butyl hypochlorite(45 ml) a t -6O', and the mixture allowed to warm to roomtemperature over 8 h. The solution was filtered, concen-trated by evaporation and yellow mercuric oxide (10 g)added.The mixture was stirred a t room temperature for4 h, methanolic ammonia added (50 ml) and stirringcontinued for a further 4 days. Concentrated HCl (5 ml)was added to the wine red solution and the material pouredinto a solution of sodium metabisulphite (100 g ) in water(2 1). Extraction into n-hexane (200 ml), drying (Na,SO,),and preparative t.1.c. of extracts gave the diazirine as ayellow-green oil (0-32 g), RF 0.5 [system (a)], 7 (DMSO)3-20 (4H, s, aromatic), 6.18 (3H, s, OCH,), and 8.06 (IH, s,diazirine 3-H), A,, (hexane) 359 (E 193), 378 (290), 390(207), 399 (207), vmax (CHCl,) 1615 and 1610 cm-l.p-( Carboxymethoxy~henyl)-3H-diazivine (Vf) .+-For-inylphenoxyacetic acid (14 g) was added to a solutionof chloramine prepared from methanolic ammonia (600 ml) ,t-butyl hypochlorite (120 ml), and t-butyl alcohol (45 mlJ.C.S.Perkin ITa t -440". The mixture was stirred a t -220" for 1 h andyellow mercuric oxide added. The mixture was cautiouslyallowed to warm to 0", stirred at this temperature for 3 11and then a t 4" for 45 h in the dark. The material wasfiltered under suction into ice (1 kg) and then added to4wHC1 (8 1) containing NaCl (500 g) at 0". Extraction intoether (6 x 500 ml) was carried out until the absorbance a t380 nm in both ether and aqueous layers was negligible.The coinbined extracts were dried a t 4" (Na,SO,) andevaporated to a yellow solid.Two thirds of this solid wasdissolved in ethanol and immediately purified bv t.1.c.[system (g)] ( l ? ~ 0.59). Extraction into methanol, evapor-ation, and trituration with ether and hexane gave a yellowprecipitate. The remaining third was purified directlyusing t.1.c. [system ( A ) ] , eluting with tetrahydrofuran, andtriturating with hexane (total yielcl 0.16 g), T (CD,OD)3-14 (4H, s, aromatic), 5.36 (2H, s, CH,, visible only in thepresence of CF,CO,H), and 7.90 (1H, s, diazirine 3-H),A,, (Tris-HC1 buffer, O . l x ~ , pH 7.9) 269 (E 1607), 277(1460), and 378 (300), vmn\. (mull) 1663 (CO,H), 1615, and1610 (C=C and N=K) c1n-l.3-(m-NitrophenyZ) -3H-diazirine (Vh) .-9%-Nitrobenzalde-hyde (1.5 g) was dissolved in t-butylamine (20 ml), benzene(10 ml) was added, and the mixture allowed to stand for1 h a t room temperature.Solvents were removed byevaporation, and the residual oil was dissolved in ethanol(10 ml) containing triethylamine (5 ml). Hydroxylamine-0-sulphonic acid (2 g) was added in small amounts withstirring at 0". Stirring was continued at 0" for 3 h andt-but>-1 hypochlorite (2 ml) then added dropwise. Afterstirring a t 0" for 30 min, the mixture was poured into water(300 ml) containing sodium metabisulphite (30 g), and thenextracted into methylene chloridc (2 x 30 ml). Theextracts were dried (Xa,SO,), concentrated, and purifiedby t.1.c. systems (e) (diazirine RF 0.52) and (f) (RP 0-6) togive the diazirine as a yellow oil (65 mg), T (DIMSO) 1-86( l H , d, J 8 Hz, aromatic), 2.19 ( l H , s, aromatic), 2.52 ( l H ,t, J 8 Hz, aromatic), 2.97 ( l H , cl, J 8 Hz, aromatic), and7.82 ( l H , s, diazirine 3-H), Amax.(hexane) 301 (c l200), 334(462), 352 (589), 362 (462), 372 (420) nm, vmIy < - . (CHC1,)1625 (C=C), 1586 (XT=S), and 1358 (NO,) cni-l.3- (4-Pyridy Z) -3H-dicrzirine (Vi) .-Pyridine-4-carbalde-hyde (4.28 g) and t-butylamine (10 ml) were mixed at roomtemperature, allowed to stand for 24 h, and excess ofamine removed by evaporation. The resulting oil wasdissolved in water (25 ml) and methanol (15 nil) andhydrated sodium carbonate (28 g) added. Hydroxyl-amine-O-sulphonic acid (5.6 g) was added over 20 min at0".After 3 h, i-butyl hypochlorite (3 ml) in t-butylalcohol (4.5 ml) was added dropwise a t 0". The solutionwas stirred for 2 11 and poured into saturated brine (400 nil)and extracted into chloroforni (4 x 30 nil). The extractswere dried (Na,SO,) , concentrated by evaporation, andpurified by t.1.c. systems (11) (diazirine RF 0.7), (c) (R, 0-S),and (i) (RF 0.25) to give a yellow oil (0.305 g), z (DMSO) 1.38(2H, d, J 5 Hz, 2- and GH), 3.26 (2H, d, J 5 Ha, 3-H),and 7-99 ( l H , s, diazirine 3-H), A,. (hexane) 330 (E 107),345 (191), 355 ( l G i ) , and 364 (164) nm, A, (liesane-0.1~-acetic acid) 331 nm, v,,,~~, (film) 1610 (C=C) and 1590(N=S) crn-1.3-( 3-P~~vicEyZ)-3M-diazirZrre (Vj) .-Pyridsne 3-carbaldehytlc(4.28 g) was dissolved in t-butylamine (7.5 ml) and allowedto stand for 16 11.Addition of benzene (20 ml) followed byevaporation gave a yellow oil which was dissolved inethanol (25 ml), triethylamine (10 ml), and water (10 ml),and cooled to - 10". Hydroxylamine-O-sulphonic acid(6 g) was added over 15 min with vigorous stirring. Thetemperature rose to 0", and stirring was continued a t thistemperature for 2 h. t-Butyl hypochlorite (6 ml) w;tsadded slowly and the yellow solution stirred at 4" for16 h. The product was concentrated by evaporatiori,filtered, and a fifth of the filtrate was purified by t.1.c.[svstem (Z)] (RR~ 0.65) to give a yellow oil (90 nig), 7: (DMSO)1.53 ( l H , d, J 5 Hz, 0-H), 1-66 (lR, s, 2-H), 2.85 ( l H , m,5-H), 3-04 (lH, M, 4 H ) , and 7.98 ( l H , s, diazirine 3-H),A,,,ax. (hexane) 350 ( c 275), 366 (107), and 377 (191) nin,vmaS. (film) 1622 ( G N ) and 1590 (N=X) cm-1.Yliotolysis Experiments.-Spectroscopic investigations o fdiazirine photolysis were carried out with solutions ofdiazirines (0.4-3m~) in n-hexane, in stoppered quartzcuvettes (3 ml) using a medium pressure Hanovia mercuryvapour U.V. lamp in a Pyrex cooling jacket. For thedetermination of product ratios in diazirine photolysis i nthe presence of acetic acid, diazirine (ca. 1OmM) in hesanecontaining acetic acid at varying concentrations (greaterthan the diazirine concentration) was photolysed in closedglass vials (10 ml) under nitrogen, 5 cm from the source ofthe lamp, at 26" for 90 niin. The solutions were washedwith water (2 x 15 ml) and B~-sodiuni carbonate (2 x 16nil), dried (Na,SO,), and evaporated t o oils. The residucswere dissolved in CC1, and n.m.r. spectra showed that a tleast goo/, of the material could be assigned to benzylacetate and thc arylheptanes. Ifass and i.r. spectra t v c ~ eobtained for some of these mixtures (Table 3).TABLE 3Mass and i.r. spectral data for diazirine photolysis mixturesVmsr.m / e vmax. mje (cm-l)Diszirine (ill+) acetate * tat1c.s tanes t acetate (cm-') hep- hep-3-Pheiiyl-3H 1747 29203- (p-Tolyl) -3H 164 1740 190 29503-( $-Methoxyphenyl)-3H 1738 29403- (4-Pyridylj-3H 161 1740 177 2900* In CCl,, C=O stretch. I n CCI,, C-H stretch.Isolatio9i of the Second Intevwzediate.-3-P1ienyl-3II-diazirine (63 mg) was dissolved in nitrogen-purged hesane(200 ml) and irradiated in a cylinder (250 ml) fitted with analuminium foil reflector and mounted 8 cm from the lampsource. After 3 min, the irradiation was stopped andglacial acetic acid (1.2 ml) added. The mixture was stirredfor 10 min, washed with 1N-NaOH (100 ml) and then dried(Na,SO,) . Concentration of the solution was followed byt.1.c. [system ( d ) J , and gave the second intermediate (RP0.6), after extraction into ether and evaporation, as abrown oil (20 mg).[4/1736 Received, 19th A ~ g i ~ s t , 1974