1973 1167The Positive and Negative Ion Mass Spectra of Some Nitro- and Polynitro-acenaphthenesBy John F. J. Todd * and Robert B. Turner, University Chemical Laboratory, University of Kent at CanterburyBrian C. Webb and Clifford H. J. Wells," School of Chemical Science and Technology, Kingston PolytechnicKing st on - upon -T ha mes, SurreyThe positive and negative ion 70 eV mass spectra of thirteen nitro- and polynitro-acenaphthenes are reported. Thepostulated fragmentation paths, many of which are supported by metastable ion evidence, are interpreted in terms ofstructural features of the molecules and stabilities of the products formed. The significant similarities and differencesbetween the two types of spectra are rationalised where possible.THE positive ion mass spectra of simple nitroaromatics extent to which the two types of spectra relate both toare now relatively well characterised,l the compounds each other and to the published data on other nitro-studied including nitrobenzene, the mononitrotoluenes, compounds.nitroanilines, and nitrophenols.2 Numerous mono- anddi-nitronaphthalenes have also been in~estigated.~,~ DISCUSS1oNThe negative ion mass spectra of these compounds havereceived considerably less attention, the only significantpublications having appeared within the last two years.( A ) The Positive Ion Mass S+ectra.-The most abund-ant ions in the spectra of the nitroacenaphthenes arelisted in Table 1 along with the value for the percentageTABLE 1Metastable transitions a and percentage total ion current for important fragmentations in the positive ion spectraof nitroacenaphthenesDaughter ionsc c 7[(Af - OH.) - NO-]+*Suiistitution M+* LM -pattern (%) :*3-(1) 14.44-(11) 23.95-(III) 19.6 *3,6-( IV) 13.83,s-(VI) 17.1 *4,5-(VII) 17.64,6-(VIII) 32.5 *5,6-f IX) 14.7 *5.8 *12-2 * 3,5,6-(X)3,5,8-(XI)3,5,6,7-(XII) 5.2 *3,5,6,8-(XIII) 4.4 *3,7-(V) 12.1 *OH.]+ (or [Af -% m * b16.6 (z) 1.53.7 *16.4 (:)15.218.5 *0.23.1 (*)1.83.21.6 (*)0.1 (*)12.0 '2)HONO]+%23.516.616.4ti.7.5 .61.90.01 .o0.10.63.10.20.3*) [PI - NO.]+ [ M - NO,.]+ [(M - NOS.) - NO*]+' [M - O]+* [Af - HJ+' [(M - He) -m* Yo m* yo m* /O m* % % %0.8 * 9.7 0.0 2.4 1.1 6.4 * 0.9 * 17.6 0.1 0.9 0.0 5.9 * 2.3 * 9.1 0.1 0.8 0.1 4.61.4 2.6 0.1 0.70.0 2.4 0.0 0.8* 0.2 1.40.0 1.8 T 0.2 1-1 1.0 2.8 0.2 0.7* 1.5 * 3.8 2.0 0-7 0.1 0.2 * 0.1 * 6.8 18.3 0.3 0.1 0.12-3 0.4 0.4 0.8 * 0.1 0.7 0.3 1.9 3.6 1.20.0 * 2.80.3 1.1 0.5 0.7 2 -7 2.90.9 * 0-6 0.1 0.4 3.4 6.301* 0.1 2.0 * 0.2 0.0 0.0 0.1*NO,.]+a Asterisk in Table indicates that metastable ion observed at appropriate m/e value.b Parentheses indicate metastable ions observed for the transitions [M - OH.)- NO-]+* and [M - HONO]+*. The remaining metastable ions are for the transition [M - OH.) - NO-]+= only.Brown and Weber5 have reported upon the negativeion spectra of m- and (P-dinitrobenzenes obtained with2-20 eV electrons, and Bowie and his co-workers havediscussed the negative ion mass spectra of the isomericnitrophthalic anhydrides and N-substituted nitro-milines,' as well as a series of nitrobenzenes containingsubstituent groups such as formyl and carboxy8 andtrifluor~acetamido.~ Direct comparisons between posi-tive and negative ion mass spectra within the samepublication are extremely rare, although one recent ex-ample is the report by Larkins et al.1° on a series of sub-s titu t ed trinitromethanes .In this account we discuss the positive and negativeion mass spectra of a series of nitro- and polynitro-acenaphthenes, a group of compounds not previouslysubjected to mass spectral analysis, and consider theH.Budzikiewicz, C. Djerassi, and D. H. Williams, 'MassSpectrometry of Organic Compounds, ' Holden-Day, San Fran-cisco, 1967, p.617.J. H. Beynon, R. A. Saund,ers, and A. E. Williams, ' The MassSpectra of Organic Molecules, Elsevier, Amsterdam, 1968, pp.3 Ref. 2, pp. 334-339.E. F. H. Brittain, C. H. J. Wells, H. M. Paisley, and D. J.322-334, 340-342.Stickley, J . Chenz. SOC. (B), 1970, 1714.total ion current carried by each ion. These values whentaken in conjunction with the metastable ion evidence[M-HONOI: [ M - N O : ] + [(M-2H*)-NOz*]+SCHEME 1for specific decay processes suggest that the molecularions fragment in the main by the routes shown in Scheme1. The molecular ions are relatively stable and althoughC. L. Brown and W. P. Weber, J . Amer. Chem. Soc., 1970,T. Blumenthal and J. H. Bowie, Austral. J . Chem., 1971, 24,J.H. Bowie, T. Blumenthal, and I. Walsh, Org. MassJ. H. Bowie, Org. Mass Spectrometry, 1971, 5, 946.J. T. Larkins, J. M. Nicholson, and F. E. Saalfeld, Org.92, 6775.1863.Spectrometry, 1971, 5, 777.B J. H. Bowie, Austral. J . Chem., 1971, 24, 989.Mass Spectrometry, 1971, 5, 2651168 J.C.S. Perkin I1the stability is less for the tri- and tetra-nitroacenaph-thenes than for the mono- and di-nitroacenaphthenes themolecular ion peak is still of significant intensity in thespectra of the former compounds.The resultsshow that the losses of OH- and HONO represent majorfragmentation routes in the decomposition of the majorityof compounds studied. Possible mechanisms for theexpulsion of OH* and HONO from nitroacenaphthenesin which there is a nitro-group ortho to the aliphaticbridge system are shown for 3-nitroacenaphthene (I) inScheme 2.Loss of hydroxyl radical and d r o u s acid.HH6 5H -OH/ 1 I..*SCHEbIE 2The loss of OH- via interaction of a nitro-group with aneighbouring alkyl group as shown in Scheme 2 isanalogous to the fragmentation of the molecular ion ofo-nitrotoluene.ll However, whereas the [M - OH-] +ion formed from o-nitrotoluene fragments further bysequential loss of CO and HCN, where the carbon atomlost as CO is the methyl carbon atom, this route forfurther fragmentation is not available to the [M - OH*]+ions from nitroacenaphthenes since the elimination of COby an analogous mechanism to that operative for o-nitrotoluene would involve rupture of two bonds in thealiphatic bridge.As expected, therefore, the peaks atnz/e values corresponding to the [(M - OH*) - CO]+ions and the [(M - OH-) - CO - HCN]+ ions areinsignificant in the spectra of the nitroacenaphthenes.Elimination of OH* and of HONO also occurs fromthose nitroacenaphthenes which are not substituted a tthe sites ortho to the aliphatic bridge. For such com-pounds the hydrogen abstraction step analogous to thatshown in Scheme 2 will involve the neighbouring ringhydrogen atoms rather than the bridge hydrogen atoms.Previous reports on the loss of OH* from alkyl substitutednitroaromaticsl1-l4 suggest that in the case of 5-sub-stituted derivatives the hydrogen would be preferentiallyabstracted from the ring not containing the nitro-group.Loss of nitrogen dioxide and nitric oxide.The elimin-l1 S. Meyerson, I. Puskas, and E. K. Fields, J . Amer. Chenz.l2 J. Harley-Mason, T. P. Toube, and D. H. Williams, J. Chew.SOC., 1966, 88, 4974.SOC. (B), 1966, 396.ation of NO,* from the molecular ion of the nitro-acenaphthenes can occur by simple bond fission and inthe majority of cases the corresponding metastable ionpeak is observed in the spectra (Table 1). Similarly, theelimination of NO* from the molecular ions, presumablyvia the nitro --+ nitrito rearrangement,15 is indicated inmost cases by the presence of a metastable ion peak.These fragmentation processes do not warrant furthercomment except that insofar as their relative importancedepends upon the structure of the nitroacenaphthene andthe facility of the other competing primary fragmentationprocesses.The sequential loss of NO,. and NO- is ofgreater interest since this process appears to be markedlystructure-dependent as shown by the fact that thepercentage total ion current for the [(M - NO,.) -NO*]+' ion is relatively low for all the compounds exc.ept5,6-dinitroacenaphthene (IX) where the value is veryhigh, uix. 18.3%. This last phenomenon can be rational-ised in terms of the mechanism for sequential loss ofNO,* and NO- shown in Scheme 3.One feature of Scheme 3 is that transfer of an oxygenatom from a nitro-group at the 5-position to C-6 can occurunder favourable circumstances, e.g. a positive chargeresiding at C-6. The observation of a metastable ionpeak in the spectrum of 5,6-dinitroacenaphthene corre-sponding to the loss of CO from the [(M - NO,*) -NO*]+* ion provides support for this oxygen transfer step.Elimination of NO,* followed by NO* does not play aprominent role in the fragmentation of the molecularions of tri- and tetra-nitroacenaphthenes substituted a tIO+ &N: b+SCHEME 3the 5- and 6-positions since in these compounds alter-native fragmentations, such as elimination of OH-,dominate the breakdown pattern.A notable feature of the spectra ofthe nitroacenaphthenes is the increase in intensity of the{M - HJ+* peak as the number of nitro-groups in thecompounds increases (Table 1).The loss of hydrogen is13 J. H. Beynon, B. E. Job, and A. E. Williams, 2.Nat.ur-~ O Y S C ~ . , 1966, 21a, 210.14 G. E. Robinson, C. B. Thomas, and J. M. Vernon, J . Chem.SOC. (B), 1971, 1273.15 J. H. Beynon, R. A. Saunders, and A. E. Williams, I n d .Chirn. belge, 1964, 311.Loss of hydyogen1973 1169likely to occur from the aliphatic bridge and the resultsin Table 1 suggest that the probability of loss increasesas the Lewis acidity of the naphthalene system increasesand the charge density on the aliphatic bridge decreases.This is supported by the results for the percentage totalion current for the [M - HqT* ion from 3,5,6-trinitro-acenapht hene (X) (0.4%) and 3,5,8-trinitroacenaphthene(XI) (3.6%). The Lewis acidity of the naphthalenesystem of (X) will be less than that of (XI) since in theformer compound steric interaction between the nitro-groups at the 5- and 6-positions forces the groups out ofthe plane of the aromatic system and reduces theirelect ron-wi t hdrawing power.Thus the electron densityin the aliphatic bridge of (XI) will be lower than that in(X). Also, the inductive effect of the two ortlzo-nitro-groups in (XI) as compared to that of the one ortho-nitro-group in (X) will be such that the electron densitywill be lower in the aliphatic bridge of the former com-pound. Similarl>-, the electron density in the aliphaticbridge of 3,5 ,GI 8- t e tranitroacenaph t hene (XI1 I) will beless than that in 3,5,6,7-tetranitroacenaphthene (XII)and here the percentage total ion currents carried by the[M - H21+' ions are 3.4 and 2-7 respectively.The smallvalues for the percentage total ion current for the[JI - H&" ion in the spectra of the mononitro- anddinitro-derivatives suggest that the electron density inthe aliphatic bridge in these compounds is not sufficientlylow for the elimination of H, to be an energeticallyfavoured process compared with the other primary frag-men t ation routes.The values given in Table 1tor the percentage total ion current for the [M - O]+' ionshow that the elimination of an oxygen atom is favouredwhen there is a nitro-group ortho to the aliphatic bridge.A similar, though less marked, effect is observed in theypectra of mono nitro toluene^.^^ The mechanism where-by the neighbouring alkyl group in the ortho-substitutednitroacenaphthenes and in o-nitrotoluene enhances theloss of an oxygen atom is open to conjecture.It is of interest that nitro-acenaplithems substituted at the 5-position but not atthe 6-position do not undergo significant loss of CO fromthe molecular ion whereas for nitronaphthalenes sub-stituted at the l-position but not at the 8-position theloss of CO from the molecular ion is a major fragmenta-tion p r o c e ~ s .~ . ~ ~ The elimination of CO from nitro-naphthalenes of this type is dependent upon the transferof an oxygen atom from the nitro-group to C-8. Thedifference between the nitroacenaphthenes and the nitro-naphthalenes could arise because the C-5-C-5a-C-6 bondangle in the acenaphthene skeleton is greater than thecorresponding bond angle in the naphthalene skeleton,l6a factor which has been invoked1' to explain theobservation that the rate of nitration at the 5-positionof acenaphthene is seventeen times greater than that a tthe 4-position in 1,s-dimethylnaphthalene.The in-creased bond angle results in an increased separationbetween the oxygen and the adjacent peri-carbon atomin nitroacenaphthenes relative to that in nitronaphthal-Loss of atomic oxygegz.Loss of cwboPJ monoxide.enes with the effect that oxygen transfer is likely to beinhibited. Nevertheless under favourable conditions(see above) the oxygen transfer step can compete effec-tively with other processes involved in the fragmentationof the positive ions.(B) The Negative Ion Mass Spectva.-Gemval features.The percentage total ion current carried by each of themajor negative ions produced from the nitroacenaph-thenes is given in Table 2, as are the observed metastableion transitions. As can be seen the majority of themolecular anions formed are highly stable and for nineof the compounds studied the molecular ion peaks arealso the base peaks of the spectra.The most prominentroutes for fragmentation of the molecular ions involveloss of hydroxyl radical, nitrous acid, nitric oxide, andnitrite. These primary fragmentations and the majorsubsequent fragmentations are summarised in Scheme 4.u ;[M-HONOI; f--- [ M - N O * ] -[ ( M - N 0 * ) - N 0 ] [IM-NO+NO;~-;SCHEME 4Loss of lzydroxyl radical and ktrozts acid. As can beseen from Table 2 the loss of hydroxyl radical and ofnitrous acid from 4,5-dinitroacenaphthene, 4,G-dinitro-acenaphthene, and 5,6-dinitroacenaphthene is signifi-cantly less than the corresponding losses from the re-mainder of the polynitroacenaphthenes which all have anitro-group positioned ortho to the aliphatic bridge.This indicates that the losses involve abstraction of ahydrogen atom from the aliphatic bridge system by thenitro-group.The presence of a further nitro-group in ananalogous meta-position to the ortho-substituted groupfacilitates the losses by resonance delocalisation of thenegative charge that remains after the hydroxyl radicalor nitrous acid is removed. This is exeniplified for3,8-dinitroacenaphthene in Scheme 5 . It is significantthat the combined yield of [hi? - OH-]- and [M -HONOI-' ions from 3,8-dinitroacenaphthene, in whichthere are two nitro-groups ortho to the aliphatic bridgeis much greater than that from 3,6-dinitroacenaphthenein which the nitro-groups are in analogous nzeta-positionsbut only one group is ortho to the aliphatic bridge.Inthe case of 3,5,8-trinitro- and 3,5,6,8-tetranitro-acenaph-thene there are two and three nitro-groups respectivelywhich are in analogous meta-positions to the nitro-groups ortho to the aliphatic bridge. Also, since thereare two nitro-groups ortho to the bridge in both com-pounds, the likelihood of loss of OH* would be expectedto be high. In fact the peak corresponding to thel6 H. W. W. Ehrlich, Acta Cvyst., 1957, 10, 699.l7 A.Davies and K. D. Warren, J . Chem. SOC. ( R ) , 1969, 873I170 J.C.S. Perkiii I1[M - OH*]- ion is the base peak in the spectrum of eachof these compounds.It has been suggested by Bowieet aL7 that the loss of NO* from nitroaromatic compoundsis facilitated when an ortlzo/para electron-withdrawinggroup is present. This is thought to arise because ofresonance stabilisation of the phenoxide ion whichhere nitro-groups are also in meta-positions to each other.The mixed ortho/para- and gneta-character of this coni-pound is reflected in its negative ion spectrum in that theyield of the [M - NO*]- ion is similar to the combinedyield of the [M - OH*]- and [M - HONOI- ions.Formation of nitrite i o n . Loss of the nitrite ion frointhe molecular anions is a particularly prominent featureLoss of nitric oxide.TABLE 2Metastable transitions a and percentage total ion current for important fragmentations in the negative ion massspectra of nitroacenaphthenesDaughter ionsh 7 r[ ( M - OH.) - I\iO.]- .(or [A4 - [ ( M - NO.) [ ( N - NO.) [(.TI - HZ)Substitution dl-.[JI - OH*]- HONOI-*) [JI - NO.]- - N0.j-O [JI - NO,.]- - NOL.]-* [aI - OJ-• [If - He]- [JI - H2;-* - NO.]- X0,-3-(I) 64.6 0.6 1.6 0.5 1 . 3 16.74-(11) 63.3 0.0 0.1 0.3 1.1 s.5 0 4 18.30.2 3 .O 0 . 3 1.1 I .3 0.3 0-2 49.9CbG 0.6 0.6 3 7 (1.2 0.7 42.3 1.; * 5.3 * 7.4:i,7- f V) 9. d 1.9 * b 5.6 * 18.9 * 3.6 1.4 0.7 0.8 1.3 0.1 0.4 41.70.7 0.7 0.8 (J.9 0.2 0.5 23.54,5-(VII) 15.2 0 c) 0.0 * 16.1 0.0 1 .5 0-0 0.0 0 0 0.0 0.0 6.5.50.0 3 8 4lJ,G-(IX) 26.2 0.0 0-8 * 0.7 * U.1 0.:; 1.4 0.0 0 9 0-0 0.1 61.116.8 * 8.7 1.2 * 4.0 * 0 .G 1.0 * 3.4 0.0 5.4 ().>I 0.7 18.1 1.s * 1.8 3.0 1.8 0.7 1.3 9.10.5 * 8.6 * 3.7 0.8 0.5 1.6 1.:4 2.23.4 1 . 3 3.4 4.0 2.1pattern ("6) m* $it* "A m* 76 m* ob in* ?A fit* 0 0 n ' o / 10 05-(III) 27.4 0.73,6-(IV) 31.2 * 2.93,s (1'1 ) 28.0 * 6.3 2.11 * 3.2 * 0.8<,6-(VIII) 32.6 0.0 0.6 * 1 5 . 1 * l.? 0.5 * 0.3 1.4 11.3 0-37.9 * I!,.; * C 3.6 * 3.6 * 1.0), 6-( X 1>&(XI)~,6,7-(X11) 13.0 * 3.2 2.2 * 6.33,5,6,8-(XIII) 4.7 * 18.8 3.0 * 1.7 0.5 * 1.8 * d 6.7o -4sterisk inTable indicates that metastable ion observed at appropriate in'e value. b Corresponding to [(M - NO.) - OH*]-*. c Corresponding to [(M - OH-) - KO*:-..d Metastable ions indicate [(A4 - NO,.) - X;O.]-* also formed.results from loss of NO*.It is thus to be expected thatthe loss of NO* will be favoured in compounds containinga nitro-group in an effective ortho/pam position relativeSCHEME 6to the nitro-group from which the loss eventually occurs.Presumably this fragmentation is preceded by a nitro----t nitrito rearrangement as is shown for 4,6-dinitro-acenaphthene in Scheme 6. 3,7-Dinitro-, 4,5-dinitro-,and 4,6-dinitro-acenaphthene all have nitro-groupswhich are effectively ortho/para to each other and it canbe seen from Table 2 that the relative abundance of the[M - NO*] ion from these compounds is far higher thanfor any of the other compounds studied. 3,5,6,7-Tetra-nitroacenaphthene is the only other compound in whichnitro-groups are effectively ortho/para substituted butof the negative ion spectra.However the percentagetotal ion carried by this ion is less easy to correlatedirectly with the structure of the parent ion than is thecase for the other major ions. No metastable ion wasobserved for the formation of the nitrite ion and it maybe concluded that it must be produced by a rapid process,e.g. dissociative electron capture.(C) Comparison of Spectra-In their study of thenegative and positive ion mass spectra of trinitro-methanes, Larkins et aZ.1° considered that there was littlecorrelation between the peak intensities in the two typesof spectra. In the present work, ionisation with 70 eVI -NO' I VSCHEME 6electrons should place negative ion formation in thedomain of ion-pair formation l8 with zero yield of parentl.3 C.E. Melton in ' Mass Spectrometry of Organic Ions,' ed.F. W. McLafferty, Academic Press, New York and London,1063, ch. 41973 1171ions.lg That this latter effect is clearly not observed inthese spectra is consistent with the suggestion maderecently by McAllister 2o that at 70 eV there is a suffi-ciently high density of slow-moving secondary electrons(accompanying positive ion formation or emitted fromelectrode surfaces) to make electron attachment anddissociative attachment significant processes under theseconditions. On this basis the axiomatic approach to thefragmentation of the primary molecular anions, suggestedrecently by ,Alexander et aL21 should be applicable tothese compounds.If ion-pair formation is of importance one mightexpect to see similar trends in the yields of complemen-tary pairs of ions between the two sets of spectra.Sucha pair might be [ A 1 - NO,*]+ and NO,-. In fact there isno apparent correlation of intensities over the range ofcompounds studied, and furthermore other ions whichcould be considered to be a complementary pair werenot observed to be of importance in the spectra. Themetastable ion evidence in the negative ion spectraclearly points to the fragmentation of the molecularanion, principally by expulsion of neutral species such asOH-, HONO, NO*, and to a lesser extent NO,.. Thesemodes of fragmentation show fairly clear parallels inthe behaviour observed for both sets of ions, e.g.there isthe requirement in both series of spectra that a nitro-group be situated ortho to the aliphatic bridge for losso f hydroxyl radical from the molecular ion to be generallyeasy. On the other hand the balance between the tend-ency to lose NO* or NO,* is quite different: whereas the1 9 C . E. hlelton, ' Principles of Mass Spectrometry and Nega-tive Ions,' Marcel Dekker, New York, 1970, p. 192.20 T. SIcAllister, J.C.S. C h e w Contni., 1972, 245.21 R. G. Alexander, D. B. Bigley, and J . F. J. Todd, Ovg.,?lass Spectvoniefvy, in the press.former is principally expelled in negative ion fragmenta-tions, with the latter being of only minor importance,the reverse situation is observed in the positive ion massspectra. This might suggest that the anionic parentspecies is the more stable insofar as the loss of NO* mustin general be preceeded by the nitro-nitrito re-arrangement.There is some similarity in behaviour between thetwo sets of spectra in the trends of the yields of ionsformed by expulsion of H, from the molecular ion, whereit appears that in both cases an increase in the number ofelectron-withdrawing substituents renders the bridgehydrogens more labile. Similar trends are also observedin the ions formed by sequential losses: [(M - H2) -NO*]- and [(M - H,) - NO,*]+.EXPERIMENTALThe synthesis of the compounds studied has been de-scribed elsewhere.22 All the compounds were isomericallypure, with the exception of 5-nitroacenaphthene from whichit did not prove possible to separate traces of the 3-nitro-The mass spectra were recorded on AEIMS9 and MS902 instruments operating at source pressuresof ca. Torr and temperatures of 150-220 "C. Assign-ment of the negative ion peaks was assisted by the use of anAEI ' Massmaster' modified for use with negative ionmass spectra.24We thank the S.R.C. for a research grant in partialsupport of this work and Rlr. R. G. Alexander for helpfuldiscussions.[2/2109 Receiued, 7th Sepfeinbev, 1972122 B. C. Webb and C. H. J. Wells, J.C.S. Perkin I, 1972, 16623 L. A. Jones, C. J. Joyner, H. I<. Kim, and R. A. I<@,24 D. A. Gallagher and J. F. J. Todd, Intevwat. J . Massand references cited therein.Canad. J . Chem., 1970, 48, 3132.Spcctvometvy Ion Physics, 1971, 7, 336