Diazepines. Part XX. The properties of 2,3-dihydro-1,4-diazepinium perchlorate

摘要

1260 J.C.S. Perkin IDiazepines. Part XX.l The Properties of 2,3-Dihydro-l,4-diazepiniumPerchlorateBy Douglas Lloyd," and Hamish McNab, Department of Chemistry, Purdie Building, University of St. Andrews,Donald R. Marshall, School of Physical and Molecular Sciences, University College of North Wales, Bangor,The unsubstituted 2,3-dihydro-l,4-diazepinium ion shows a number of differences from the substituted derivativesstudied hitherto. Although it undergoes halogenation at position 6, it cannot be nitrated by nitric acid and isdecomposed by mineral acids. The ring is also opened by the action of nucleophiles a t the 5- and 7-positions. Atransdiazepination reaction occurs with NN'-dimethylethylenediamine to give the NN'-dimethyldihydrodiazepiniumsalt. 6-Halogeno-derivatives do not react a t the 6-position with nucleophiles.St.Andrews, FifeCaernarvonshireTHE properties of 2,3-dihydro-l,4-amp;azepinium salts havebeen studied extensively but the chemistry of the parentsalt (I) has hitherto been uninvestigated, principallybecause the standard method of preparation, from a1,2-diamine and a 1,3-dicarbonyl compound gave only avery low yield.3 An alternative mode of preparation:HCLOL'(Elwhich provides this salt in good yield, has made it pos-sible to study its properties, which differ in a number ofrespects from those of its substituted derivatives, prob-ably for steric reasons.The m.p. of compound (I) had earlier been reported as79-80"; the much higher value now found (ca. 250",with violent decomp.) presumably reflects the difficultiesinvolved in isolation and purification by the oldermethod.The earlier reported value for the U.V.extinction co-efficient at A,, 331 nm (E 11,200) is lower than the valuecalculated empirically (E 13,500) from the values for othersubstituted dihydrodiazepinium salts? but measurementson a carefully purified sample now indicate a value of14,300 amp;- 200. This high value may reflect the absenceof distortion of the x-system caused by substituents (forwhich there is spectroscopic evidence 6,7) and/or an1 Part XIX, C. Barnett, D. R. Marshall, and D. Lloyd, J.C.S.Perkin 11, 1976, 325.a D. Lloyd, H. P. Cleghorn, and D. R. Marshall, Adv. Hetero-cyclic Chem., 1974, 17, 1.3 C. Barnett, D. R. Marshall, and D.Lloyd, J . Chem. SOC. (B),1968,1638.4 D. Lloyd, H. McNab, and D. R. Marshall, Synthesis, 1973,791.6 C. Barnett, H. P. Cleghorn, G. E. Cross, D. Lloyd, and D. R.Marshall, J . Chem. SOC. ( C ) , 1966, 93.6 A. M. Gorringe, D. Lloyd, and D. R. Marshall, J . Chem. SOG.( C ) , 1969, 1081.extremely low concentration of the non-absorbing bis-imine tautomer (11) ; a small contribution from this formmay arise in the case of substituted dihydrodiazepiniumsalts because some relief of crowding by substituents atthe 5-,6-, and 7-positions is introduced, although at theexpense of the loss of the conjugated system.The n.m.r. spectrum of the salt (I) in 2H6acetone 7presents the 5- and 7-H signals as a broad multiplet, butaddition of a small quantity of trifluoroacetic acid pro-vides sharp signals.Coupling constants J5amp;7) andJ4,5(1,7) are respectively 7.8 and 7.2 Hz; their near iden-tity is evidence of the almost complete delocalisation ofthe electrons in the conjugated portion of the cation.There is also a small meta-coupling, J1(4),6 1.5 Hz. Themethylene groups appear as a singlet at room temperature, indicating fast inversion of this portion of the ring(cf. ref. 7).Dihydrodiazepinium salts undergo electrophilic substi-tution at position 6,*-1deg; and the ratio of the reactivitiesof positions 6 and 5(7) towards deuteriation in deuterio-acids has been shown l1 to be at least lo9 : 1. This resultwas obtained from the study of salts in which both posi-tions 5 and 7 or position 6 were unsubstituted, but nocomparison of these sites in the same molecule hashitherto been recorded.When the salt (I) was dissolvedin trifluoroacetic 2H acid, no signals due to the I-, 4-, or6-protons were seen, owing to their rapid exchange withthe solvent, but after 9 days in this solution no decompo-sition had occurred and the ratio of the 5- and 7-Hintegral to that of the methylene 2- and 3-protons wasunchanged, demonstrating again the low reactivity of the5- and 7-positions towards electrophiles.D. Lloyd, R. K. Mackie, H. McNab, and D. R. Marshall,J.C.S. Perkin 11, 1973, 1729.8 D. Lloyd and D. R. Marshall, J . Chem. SOC., 1968, 118.9 A. M. Gorringe, D. Lloyd, F. I. Wasson, D. R. Marshall, andP. A. Duffield, J .Chem. SOG. ( C ) , 1969, 1449.10 C. Barnett, Chem. Comm., 1967, 637; J . Chem. SOC. (C),1967, 2436; A. M. Gorringe, D. Lloyd, D. R. Marshall, and L. A.Mulligan, Chem. and Ind., 1968, 130; A. M. Gorringe, D. Lloyd,and D. R. Marshall, J . Chem. SOC. ( C ) , 1970, 617.11 A. R. Butler, D. Lloyd, and D. R. Marshall, J . Chem. SOC.(B), 1971, 7961975 1261Like other dihydrodiazepinium salts without 6-substi-tuents the salt (I) is readily brominated at the 6-positionby bromine in methanol. 6-Chloro- and 6-iodo-deriv-atives of (I) were also obtained in excellent yield by theaction, respectively, of N-chloro- and N-iodo-succin-imide. Like other 6-brominated dihydrodiazepiniumsalts, (111) shows a bathochromic shift in its principalabsorption maximum but the shift is greater in this case(29 nm; c j .24 nm for the 5,7-dimethyl derivative, 16 nmfor the 5,7-diphenyl derivative).The 6-iodo-derivative of (I) resembles other 6-iodo-dihydrodiazepinium salts in being protodeiodinated insolution in acid; in accord with the trend for protodehalo-genation to take place more readily with increase in sizeof 5- and 7-substituents,12 the present iodo-compound isprotodeiodinated much more slowly than its 5,7-dimethylor 5,7-diphenyl derivative.Whereas 5,7-disubstituted dihydrodiazepiniuni saltsare readily nitrated by nitric acid to give 6-nitro-derivatives,1deg; attempts failed to prepare the 6-nitro-derivative of (I) under analogous conditions. In con-trast to the substituted salts, the parent salt (I) is appar-ently decomposed irreversibly by mineral acids.HHcm, c101-The 6-unsubstituted dihydrodiazepines studied hither-to are inert towards nucleophiles.The parent compound(I) reacts readily with piperidine, however, to give thedipiperidinopropenylium salt (IV) l3 in high yield. Akinetic study of this reaction showed two isosbesticpoints, indicating that there is no stable intermediate.This reaction indicates clearly the reactivity of the 5- and7-positions towards nucleophiles, which is not generallyobserved in the cases of 5,7-substituted derivatives. Forexample, 5,7-dimethyldihydrodiazepinium perchloratewas 83 unchanged when heated with piperidine underreflux for 12 h, and when a solution of this dihydrodiaze-pinium salt and piperidine in methanol was kept for 7days at 25 "C the U.V.spectrum showed that 95 ofthe salt remained. The methyl groups thus have a stronginhibiting effect, presumably mainly steric, on the reac-tivity of the 5- and 7-positions, and kinetic studies showedthat they lower the reactivity by a factor of at least20,000. 1 ,4-Dimet h y ldih y dro diaze pinium perc hlorat e ,also lacking 5- and 7-blocking substituents, reacted simi-larly with piperidine in 2H$acetone solution to give theproduct (IV) .1,4-Dimethyldihydrodiazepinium perchlorate was itselfprepared by the reaction of compound (I) with a nucleo-phile, NN'-dimethylet hylenediamine, but unf or tuiiat elythis transdiazepination reaction was not of general appli-cation; for example reaction of (I) with 2-methylpro-l2 E.M. Grant, D. Lloyd, and D. R. Marshall, Chem. and Ind.,1974, 525.pane-1,Z-diamine did not, as judged from spectra, provideany dihydrodiazepine.6-Halogenodihydrodiazepines react readily with nu-cleophiles, either (a) by normal substitution, or else (b) togive the protodehalogenated product, the latter coursebeing favoured by increasing size of the nucleophile andof 5- and 7-substit~ents.~~~J~ It was expected, therefore,that the bromo-compound (111) would rapidly form the6-methoxy-derivative on exposure to methoxide ion, butin fact it was unchanged after being heated with sodiummethoxide in refluxing methanol for 30 min. Spectro-scopic examination (u.v., n.m.r.) of the resultant solutionshowed that the only product was the corresponding6-bromodihydrodiazepine base ; in particular the n.m.r.spectrum confirmed that there had been no attack by thenucleophile at the 5- or 7-position.Since the 6-position in dihydrodiazepines is itselfnucleophilic, the reactions of other 6-halogenodihydro-diazepines (V) with nucleophiles have been rationalised interms of the presence of a small amount of the bis-iminetautomer (VI), which would be expected to be highlyreactive towards nu~leophiles.~*12 As discussed aboveformation of the tautomer (VI) is disfavoured because theconjugation of form (V) is thereby lost; indeed tautomer(VI) must be present in only a small amount in anysimple dihydrodiazepine since it is not observed spectro-scopically. However there will be some slight compen-sating energetic gain if R1 and R2 are large, since in form(V) R1, R2, and Hal must be distorted from coplanarityowing to crowding, but this crowding is relieved whenposition 6 becomes tetrahedral as in tautomer (VI).In theNu-HL I(UI 1Hal = halogen, Nu = nucleophilecase where R1= R2 = H there will be no crowding in (V)and hence no compensation in forming the tautomer (VI).Thus there may be a vanishingly small contribution fromthe bis-imine form which would explain the unreactivityof the bromo-compound (111) towards nucleophilicsubstitution.Likewise, whereas 6-bromo-5,7-dimethyldihydrodiaze-pinium bromide underwent protodebromination whenheated in ethanolic thiourea, compound (111) was un-changed after 23 h under the same conditions.The 6-iodo-analogue of (111) was however convertedinto (I) when heated with ethanolic thiourea or ethanolalone.5,7-Disubstituted 6-bromodihydrodiazepine basesreact with neat piperidine to give either the 6-piperidino-or 6-unsubstituted analogue^,^ but the bromo-compound(111) reacted with piperidine in methanol with destructionl3 G.Scheibe, W. Seiffert, H. Wengenmayr, and C . Jutz, Ber.Bunsengesellschaft Phys. Chem., 1963, 67, 560J.C.S. Perkin Iof the seven-membered ring. No tractable products wereisolated save piperidinium bromide.Thus two surprising properties of the dihydrodiazepinesdescribed previously, namely their ready reactivitytowards nucleophiles at position 6 but not at positions 5and 7, are shown not to be inherent properties of the2,3-dihydro-l,4dazepine system itself, but rather aconsequence of substituent effects.EXPERIMENTALElectronic spectra were recorded for methanolic solutions,and i.r.spectra for Nujol mulls.2,3-Dihydro- 1,4-diazepinium perchlorate (I) was preparedas previously de~cribed.~6-Bromo-2,3-dihydro-l,4-diazepinium Perchhate (111) .-Bromine (0.1 6 g, 1 mmol) in methanol (3 ml) was added drop-wise to compound (I) (0.2 g , 1 mmol) in methanol (10 ml).Addition of ether precipitated the bromo-compound (0.1 2 g,ca. 72) , which was recrystallised from ethanolic perchloricacid to give the pevchlwate, m.p. 154-l55OJ Lk 360 and268 nm (E 10,200 and 3000), vmX 3300, 1630, 1540, 1330,1250, 1100, and 920 cm-l, T (CD,),SO -0-4br, 1.95 (ZH, s),and 6.36 (4H, s), J1,, (CDJ,SO-CF,*CO,H 7.8 Hz (Found:C, 22.05; H, 3-1; N, 10.1.C,B,BrClN,O, requires C, 21-8;H, 2-9; N, 10deg;15y0).6-Chloro-2,3-dihydro- 1,4-diazepinium Perchlorate.-A solu-tion of the unsubstituted dihydrodiazepinium salt (I) (0.4 g,2 mmol) and N-chlorosuccinimide (0.27 g , 2 mmol) in aceticacid (8 ml) was heated under reflux for 2 min. Addition ofether to the cooled solution precipitated the chlwodihydro-diaxepinium Pevchlwate (0.38 g, 83) , m.p. 121-121-5deg;{from ethanol), A,, 359 nm (E 9800), vmx. 3300, 1640, 1650,1320, 1240, 1100, and 930 crn-l, t (CD,),SO -0-42br, 1.96(2H, s), and 6.34 (4H, s), J1,, (CD,),SO-CF,CO,H 8-4Hz (Found: C, 26.15; H. 3.45; N, 12.4.C,H,CI,N,O,requires C, 25-95; H, 3-45; N, 12.1y0).2,3-Dihydro- 6-iodo- 1,4-diazepinium Perchlorate.-A solu-tion of the unsubstituted dihydrodiazepinium salt (I) (0-4g, 2 mmol) and N-iodosuccinimide (0.45 g, 2 mmol) in aceticacid (8 ml) was heated under reflux for 2 min. When thesolution was cooled the iodo-compound separated (0.58 g,90yo), m.p. 236235" (from ethanol), LX 369 and 314 nm(E 7700 and 1350), v,, 3300, 1630, 1550, 1330, 1100, and930 cm-1, 7 (CD,),SO -@25br, 1.98 (233, s), and 6.34 (4H,H, 2.6; N, 8.8. C,H,C11N,04 requires C, 18.6; H, 2.5; N,8.7 yo).Reaction of the Dihydrodiazepinium Salt (I) with Pipedine.-A solution of the salt (I) (0.2 g, 1 mmol) and piperidine(0-45 g, 10 mmol) in methanol (12 ml) was heated underreflux for 20 min.Solvent was evaporated in vacuo, andaddition of ether to the residue caused crystallisation of 1,3-dipiperidinopropenylium perchlorate (IV) (0.26 g, 8Zy0),m.p. 129-131O (Iit.,l* 130-131deg;), mixed m.p. 129-130deg;,u.v., ix., and n.m.r. spectra identical with those of anauthentic sample.Attempted Reaction of 2,3-Dihydro-5,7-dimethyl- 1,4-diaze-pinium Perchlorate with Piperidine. The dihydrodiazepin-ium salt (0.23 g, 1 mmol) and piperidine (0-45 g, 10 mmol) inmethanol (12 ml) were heated under reflux for 12 h. Afterevaporation of solvent and addition of ether the diazepiniumsalt was recovered (0.19 g, 83), m.p. 140--141.6deg;.Kinetic Studies of the Reactions with Pifieridine.-Stndieswere carried out in solutions in methanol (reagent grade) ats), J1.7 (CDJ,SO-CF,*CO,H 8.4 Hz (Found: C, 18.65;26 O C , and the reaction was followed over 1 half-life at 295nm.Piperidine concentration was ca. 1-5-2.0 mol 1-l.The dihydrodiazepinium salt (I) concentration was 5.1 x10-3 moll-I, and the reaction in this case followed first-orderkinetics k (2.55 f 0.09) x 103 min-l. In the case of the5,7-dimethyl analogue of (I) (0.25 mol l-l), no reaction wasobserved in 7 days, and the salt was unchanged (mixedm.p.).2, 3-Dihydro-l,4-dimethyl- I, 4-diazepinium Perchlorate.-Asolution of the dihydrodiazepinium salt (I) (0.2 g, 1 mmol)and NN'-dimethylethylenediamine (0.46 g, 5 mmol) inethanol (15 ml) was heated under reflux for 1 h. Solventwas removed and ether was added to give an oil which afterdissolution in ethanol, followed by cooling of this solution,gave the lJ4-dimethyldihydrodiazepinium perchlorate (0.15g, 67), m.p.and mixed m.p. 95-96', i.r. and U.V. spectraidentical with those of an authentic sample,6 T (CF,*CO,H)2.6 (2H, d), 4.95 (lfl, t), 6.15 (4H, s), and 6.5 (6H, s), Ja(,),aAttempted Reaction of 6-Bvonzo-2,3-dihydro- 1 ,4-diazefiniunzPerchlorate (111) with Sodium amp;Iethoxide.-The bromo-compound (111) (0.14 g, ca. 0.5 mmol) was dissolved in asolution of sodium methoxide (from sodium (0.13 g ) inmethanol. The U.V. absorption shifted to Am= 337 nm anddid not change after the solution had been heated underreflux for 30 min. The solvent was then evaporated off invacuo, water (5 ml) was added, and the solution was extrac-ted with ether (3 x 10 ml).The extract was dried (Na,-SO,) and evaporated. Addition of perchloric acid (70;0.1 ml) caused the slow crystallisation of the bromo-compound (111) (0.05 g, 36y0), m.p. 148-150" (from ethan-ol), mixed m.p. 149-151", U.V. and i.r. spectra identicalwith those of an authentic sample. The low recovery wasprobably due to the necessarily harsh work-up procedure.The product of A- 337 nm was identified as the dihydrodi-azepine base by dissolving the bromo-compound (111)in C2H4methanol containing sodium (2H,methoxide in ann.m.r. tube z 2.57 (2H, s) and 6.40 (4H, s).Attempted Reaction of the Brorno-compound (111) withThiourea.-A solution of the bromo-compound (111) (0.14 g,0.5 mmol) and thiourea (0.04 g, 0.5 mmol) in ethanol (5 ml)was heated under reflux for 23 h; the U.V.spectrum indi-cated that no reaction had taken place. Ether was added tothe cooled solution and precipitated starting material (111)(0.1 g, 72), m.p. 146-14Q0, mixed m.p. 14amp;-150", identi-cal i.r. and U.V. spectra. (The low m.p. was due to a smallquantity of thiourea which was not separated by recrystal-lisation.)Reaction of 6-Bromo-2,3-diJzydro-5,7-dimethyl- 1,4-diazepin-ium Bvomide with Thiourea.-A solution of the dihydrodi-azepinium bromide (0.86 g, 3 mmol) and thiourea (0.21 g,3 mmol) in ethanol (15 ml) was heated under reflux for 3 h.Addition of ether to the cooled solution gave 2,3-dihydro-5,7-dirnethyl-l,kdiazepinium bromide (0.58 g, 95), identi-fied by mixed m.p. and spectra.Protodeiadindion of 2,s-Dihydro- 6-iodo- 1,4-diazepiniumPerchlorate.-A solution of this iodo-compound (0.16 g, 0-5mmol) and thiourea (0-04 g, 0-5 mmol) in ethanol (5 ml) washeated under reflux for 24 h. Addition of ether to thecooled solution precipitated 2,3-dihydro- 1,4-diazepiniumperchlorate (I) (0.07 g , 70y0), identical (i.r. and U.V. spectra)with an authentic specimen. The same result was obtainedwhen a rnethanolic solution of the iodo-compound alone washeated under reflux.Reaction of 6-Bromo-2,3-dihydro- 1,4-diaxePinium Salt with7.8 Hz1975 1263Piperidine.-A solution of an unpurified sample of bromin-ated dihydrodiazepinium salt (anion for the most partbromide but some perchlorate) (0.28 g, ca. 1 mmol) and piper-m.p. (of a sample purified by solution in chloroform andreprecipitation by ether) 237--239", mixed m.p. 237-238'.idine (0.86 g, 10 mmol) in methanol (16 amp;) was heated underreflux for 20 min. After solvent had been evaporated off, tively, for and n.m*r. spectra. WeWe thank J. and ''Ii'' M* Pocn-iardowska, respec-thank the S.R.C. for a grant (to H. McN.). addition of ether to the residue gave a solid (0.18 g) whosen1.p. and i.r. spectrum showed it to be piperidinium bromide, 4/2312 Received, 7th November, 1974