2. Part (iv). MASS SPECTROSCOPY By John M.Wilson (Department of Chemistry University of Manchester Manchester 13) General Methods of Interpretation.-The most interesting feature of publica- tions in mass spectrometry in the past year has been the tendency among many of the groups in the field to concentrate on much more detailed work on a narrower range of systems. Substituent effects as described in last year's Report have been widely used and give interesting results even if linear Hammett plots cannot always be obtained. In the case of ions which bear the substituent a linear relationship can be found between intensity ratios and o-values only at low electron energies,' i.e. when th; intensity ratio is a measure of kl and not of k,. The hypothesis of charge localisation in the molecular ion is supported by two types of correlation RC,H4COCH3+' kl,RC6H4+' + -COCH k4Decomposition products firstly between ionisation potentials and fragmentations of bifunctional mole- cules.2 The dominant fragmentations of nitrogen- and sulphur-containing carboxylic acids are those typical of compounds bearing the functional group with the lowest ionisation potential.In molecules of the type (1) there is a (M -C2H4)'. ion of appreciable abundance except when R = NR,. In the latter compound the nitrogen-bearing ring carries the positive charge and the typical ketone fragmentation is s~ppressed.~ R Metastable ions. Several authors have reported decompositions of meta- stable ions that cannot be single proce~ses,~ for example the elimination of two molecules of water from 3,17-dihydro~y-steroids.~" McLafferty has developed his method of identifying the structures of ions by examining the peak shapes and intensities of their metastable ions.Ions of the same structure M.M. Bursey and F. W. McLafferty J. Aiiicv-. Cheril. Soc.. 1967 89 1. H. J. S ec and G. Junk J. Amer. Chem. SOL..,1967,89 790. 'T. J. Wachs and F. W. McLafferty J. Amer. Chem. SOC.,1967,8!3 5044. (a) E. Caspi J. Wicha and A. Mandelbaum Chem. Comm. 1967 1162; (b)J. Seibl Helu. Chim. Acrri 1967 50 263; H. Budzikiewicz F. v. d. Haar and H. H. Inhoffen. Annalen 1967,701 23. John M.Wilson from different molecules will decompose through metastable ions which will have the same shape and the same intensity relative to each other although not relative to the precursor ion.The intensity of metastable ions relative to their precursor ion is found to be a function of the number of vibrational degrees of freedom in the original molec~le.~ This is in agreement with the quasi- equilibrium theory. Measurements of appearance potentials and of metastable ions in benzonitrile show a very slow rise in k with electron energy.6 Measure- ments of half-lives for the process C,HSN+. -+ C6H4+.+ HCN suggest that the maximum value of k is 5 x lo-*. ’ Theoretical calculations have been published of peak shapes for metastable ions decomposing in all regions of a double-focussing mass spectrometer.8 Values of kinetic-energy release for processes in a large number of aromatic hydrocarbons have been obtained.’ AIthough there are a few exceptions,’* triply charged ions are generally of very low abundance but examination of the metastable ions of 9,lO-diphenylanthracene suggests that a considerable proportion of the ions in the spectrum may have a triply charged ion as their origin.l1 Conventional mass spectra of isomeric alkanes and alkenes are often very similar.In many cases where they are indistinguishable or barely dis- tinguishable the metastable spectra are quite different.12 This does not apply to cases such as the isomeric xylenes where decomposition takes place through a common intermediate. Fragmentation Mechanisms.-General fragmentation processes found in common types of functional group are well covered by two books published this year.13 New work of interest has been mainly concerned with rearrange- ments.Hydrogen rearrangements. Further work has been done on the elimination of small molecules from molecular ions of saturated compounds. H,S loss from C5H1 ,SH is 1,3 and 1,4.14 Although HF elimination from C4H9F is also 1,3 and 1,4 in the case of C6HI3F it is 1,5.” Ions of the type (2) undergo a McLafferty rearrangement to product (3) but this does not rearrange further to (4) and (5). This would suggest that keto+nol tautomerism is not generally a favoured process in the mass spectrometer.16 The case of phenol which is thought to rearrange to the keto-form prior to the extrusion of carbon mon- F. W. McLafferty and W. T. Pike J. Amer.Chem. SOC. 1967,89 5951. I. Herter and C. Ottinger Z. Naturforsch. 1967,22a 40. ’ I. Herter and C. Ottinger Z. Naturforsch. 1967,22a 1141. J. H. Beynon and A. E. Fontaine Z. Naturforsch. 1967,22a 334. M. Barber K. R. Jennings and R. Rhodes 2.Naturforsch. 1967,22a 15. lo M. I. Bruce Chem. Comm.,1967 593. l1 K. R. Jennings and A. F. Whiting Chem. Comm. 1967 820. l2 F. W. McLafferty and T. A. Bryce Chem. Comm.,1967,1215. l3 G. Spiteller. ‘Massenspektrometrische Structuranalyse organischer Verbindungen,’ Veriag Chemie Weinheim/Bergstr. 1966; H. Budzikiewicz C. Djerassi and D. H. Williams ‘The Mass Spectra of Organic Compounds,’ Holden-Day SanFrancisco 1967. l4 A. M. Dufield W. Carpenter and C. Djerassi Chem. Comm. 1967 109. l5 W. Carpenter. A. M. Dufield and C.Djerassi Chem. Comm. 1967 1022. l6 J. K. McLeod J. B. Thomson and C. Djerassi Tetrahedron 1967,23,2095. Part (iv).Mass Spectroscopy +. +. 4.. +. (3) oxide is hardly relevant since this process requires an electron energy of 15 ev. Similar conclusions were reached from an examination of the mass spectra of + P-keto-esters. C3H60 * ions are formed by ionisation of acetone McLafferty rearrangement of alkan-2-ones and double rearrangement of alkan-4- and -5-ones. The metastable ions for the decompositions of these ions all have different intensities,18 which would suggest that all have different structures the latter two being (6) and (7). Ideas about the electronic requirements of the McLafferty rearrangement are contradictory.Djerassi suggested'' that since this process was suppressed in some even-electron systems radical character was a necessity at the hydrogen-accepting site. A study of the spectra of a number of substituted butyrophenones suggests that there are two different substituent effects operating. Rearrangement is facilitated by substituents which increase the positive charge at the reactive centre and also by substitu- ents which increase the radical character at the carbonyl group invoking larger participation from canonical forms such as @)?'The operation of the pro- cesses (9) +(10)and (1 1) +(12) shows that such rearrangements may take place when there is positive charge but no radical character present at the carbonyl group.21 Compounds which show behaviour contrary to the above generalisa- tions are ketones of the type (13) which undergo a McLafferty rearrangement with formation of PhCH=CH2+ .at electron energies lower than that necessary to ionise the carbonyl group.22 Other published work on this rearrangement + + l7 R.I. Reed and V. V. Takhistov Tetrahedron 1967,23,4425. 18 F. W. McLafferty and W. T. Pike. J. Amer. Chem. Soc.. 1967.89 5953. 19 C. Djerassi M. Fischer and J. B. Thomson Chem. Comm. 1966 12. 2o F. W. McLafferty and T. J. Wacks J. Amer. Chem. Sac. 1967,89 5043. " M. Kraft and G. Spiteller Chem. Comm. 1967 943. ''J. L. Occolowitz Austral. J. Chem. 1967,20 2387. John M.Wilson ~cH‘Ii3 ’ H’ includes a full discussion of isotope effects. These are nearly all in the range 0.75-1 with the exception of compound (14; X = NH or 0).In the latter compounds the values are in the region of 0.5 presumably because the non- bonding electrons of the X atom contribute to the transition state to give a more unsymmetrical C-H-C system.23 Where there are two unsaturated systems in the molecule which can both undergo this type of rearrangement a ketonic group will decompose more rapidly than an ester or an aromatic ring.Other types of hydrogen rearrangement have been investigated including the double rearrangement in ethers e.g. C2HS+’OC6Hl3+ C2H,0H2+ + C6H1,. In a large proportion of the product ions both rearranged hydrogen atoms come from C-5,and the labelling results can only be explained in terms of a partial randomisation process involving steps such as those shown in Scheme A.25 There has been a general revival of interest in processes involving randomisa- tion in molecular ions.Double-bond shifts in olefin ions are generally associ- ated with random hydrogen migrations,26 but in cyclohexenes they appear to be specific proce~ses.~’ In the n-propane ion all the hydrogen atoms retain their identity even in long-lived states which give metastable decompositions.28 23 J. K. McLeod and C. Djerassi J. Amer. Chem. SOC. 1967,89 5182. 24 J. K. McLeod and C. Djerassi J. Org. Chem. 1967,32 3485. ’’ W. Carpenter A. M. Duffield and C. Djerassi J. Amer. Chem. SOC. 1967,89 6164. 26 B. J. Millard and D. F. Shaw J. Chem. SOC.(B) 1966,664. ’’ T.-H.Kinstle and R. E. Stark J.Org. Chem. 1967,32 1318. ’* C. Ottinger J. Chem. Phys. 1967,47 1452. Part (iv).Mass Spectroscopy All decompositions of the benzene molecular ion are preceded by complete randomisation of the hydrogen atoms. It has been suggested that this may not be a hydrogen migration but a skeletal reorganisation through prismane (15) and benzvalene (16) intermediate^,^^ but '3C-labelling work necessary to h settle this point has not yet been done. The decomposition of pyridine by loss of HCN is likewise preceded by H-rand~misation,~' but in a similar elimination from [1-2H]quinazoline there is 90% loss of HCN.31 The loss of HCN from the benzonitrile ion appears to take place by two different routes one involving complete hydrogen randomisation the other being a specific 1,2-elimination ;32 it is possible that these are decompositions of different electronic states of the molecular ion.Skeletal rearrangements. This type of process has been found to occur in a large variety of organic compounds and its occurrence has been compre- hensively covered in a review,33 so this report will cover only a few selected examples. There has been a lively controversy about the structure of the ion formed by elimination of carbon monoxide from a-pyrone. This was thought to be cyclised to a f~ran,~~ but there has been evidence to the contrary. If the hypothesis is correct [3-2H]-a-pyrone and [6-2H]-a-pyrone should both yield the same product [2-2H]furan. In the further decomposition C4H40+ . -+ C3H3+ the retention of deuterium in C3H3+ is different in these two cases suggesting that the product cannot be f~ran.~' It has been pointed out however that C3H3+ can be formed by an alternative route in which C-3 and C-6 do not become equivalent (1 7) 4(19),36so the furan theory is not ruled out.Intensities of metastable ions for the decomposition of the C4H40' . ion from pyrone are quite different from those in the spectrum of f~ran,~~ but this is a comparison 29 K. R. Jennings Z. Naturforsch. 1967,22a 455. 30 D. H. Williams and J. Ronayne Chem. Comm. 1967 1129. 31 T. J. Batterham A. C. K.Triffett and J. A. Wunderlich J. Chem. SOC.(B),1967,892. 32 R. G. Cooks R. S. Ward and D. H. Williams Chem. Comm. 1967 850. 33 P. Brown and C. Djerassi Angew. Chem. 1967,79,481.34 C. S. Barnes and J. L. Occolowitz Austral. J. Chem. 1964 17 975. 35 W. H. Pirkle J. Amer. Chem. SOC. 1965,87 3022. P. Brown and M. M. Green J. Org. Chem. 1967,32 1682. 37 F. W. McLaNerty and W. T. Pike J. Amer. Chem. SOC. 1967,89 5955. John M.Wilson of a molecular ion and a fragment ion so a difference in internal energy may be responsible for this effect. Another approach has been to examine the kinetic energy release in the metastable ion for Mf. -+ C,H,O+’. This energy does not vary within the group of compounds (20; R = p-H p-OMe p-Ph) where the charge or radical-stabilising ability varies considerably. The authors consider that this evidence is in favour of an unsymmetrical structure (21) in which the aryl-group n-electrons are orthogonal to the electron-deficient orbitals.38 This reporter feels however that we do not know enough yet about the conditions under which internal energy may be converted into kinetic energy to allow us to make decisions based on this evidence.t Ph Ph(37H4R The rearrangement which results in loss of carbon dioxide from organic carbonates has been studied in detail3’ In the case of aryl methyl carbonates the oxygen of the methoxy-group is lost and the process is promoted by electron- withdrawing substituents. The product ion behaves in its further decomposi- tions like ArOMe’ rather than o-MeC,H,OH+ so a mechanism can be written (22)-423). In the corresponding process in the spectra of diary1 carbonates aryl groups with electron-releasing substituents migrate more readily.A number of migrations of groups containing oxygen or nitrogen can be explained by a general mechanism as shown in Scheme B.,Oa Such a mechanism can rationalise the formation of PhCH OMe+ in the spectra of P-bromo-p-phenylpropionic acid and dimethyl phenylsuccinate and other ions in the spectra of dicarPoxylic Another example is the methoxy-migration found with 4-methoxy- cyclohexanone discussed in last year’s Report this process is found in 38 M. M. Bursey and L. R. Dusold Chem. Comm. 1967 712. 39 P. Brown and C. Djerassi J. Amer. Chem. SOC.,1967,89 2711. ’* (a) R. G. Cooks and D. H. Williams Chem. Comm. 1967 51 ; (b) I. Howe and D. H. Williams Chem. Comm. 1967,733. 41 (a)M. M. Green D. S. Weinberg and C.Djerassi J Amer. Chem. Soc. 1966,88 3883; (b) ibid. 1 967,89 5 190. Part (iv).Mass Spectroscopy Y-x Y-x +Y-x I-I-II Ph-CH-CHZ Ph-CH -CH 2 Ph-CH-CH, + R I +I Ph-CH=Y + CH,=X SCHEME B cyclohexanones and cycloheptanones with hydroxy- or alkoxy-groups at the 3- or 4-po~itions.~’~ In the formation of the ion RCH=CHCO+ from 4,4-dialkylcyclohexenones (24) substituent effects suggest that an R group and an R’ group migrate.42 The same authors report similar findings for benzylidene- cyclohe~anones.~~ The structural requirements for the elimination of keten from the molecular ions of UP-unsaturated ketones are the same as those for the photochemical conversion into bicyclohex[3,l,0]an-2-ones and the elimina- tion should go as in (25).44 In the mass spectra of substituted benzophenones there is a rapid loss of carbon monoxide from the doubly charged molecular ion which is not observed with the singly charged species.This may be due to the coupling of a diradical species such as (26).45 Labelling work demonstrates that the methyl group lost from a-ionone is not one of the gem-dimethyls but the vinylic group. This can be explained by invoking a cyclisation (27)-+(28).46 The decomposition C,H2C1,+ . -+ C2C12+. in the spectrum of 3,S-dichloropyridazine suggests that the former ion may be a cy~lobutadiene.~’ Ions C,H3 from various dienes have a cyclic structure.48 Several examples + 42 R. L. N. Harris F. Komitsky jun. and C. Djerassi J. Arner. Chem. SOC.,1967,89 4765.43 R. L. N. Harris F. Komitsky jun. and C. Djerassi J. Amer. Chem. SOC. 1967,89 4775. 44 A. L. Burlingame C. Fenselau W. J. Richter W. G. Dauben G. W. Shaffer and N. D. Vietmeyer J. Amer. Chem. SOC.,1967,89 3346. 45 F. W. McLafferty and M. M. Bursey Chem. Comm. 1967 533. 46 A. F. Thomas and B. Willhalm Tetrahedron Letters 1967 5129. 47 S. J. Weininger and E. R. Thornton J. Amer. Chem. SOC. 1967,89 2050. 48 J. L. Franklin and A. Mogenis J. Chem. Phys. 1967,71 2820. 62 John M. Wilson have been found of elimination of methylene from large ions. 49 Other systems in which skeletal rearrangements have been found include unsaturated esters," azobenzenes," azoxy-compounds,52 a~ylthiophens,~~ aromatic N-~xides,~~ dimet hylaminopyrimidines N-methyltriflu~roacetanilide,~ thioesters 57 and thi~nylaniline.~~ Miscellaneous.-Both 1,4-59 and 1,2-quinones(jo dehydrogenate in the heated inlets of mass spectrometers.These are presumably surface reactions. Work still continues on methods of determining the positions of double and triple bonds in aliphatic molecules. The field-ionisation mass spectra of alkynes show intense peaks corresponding to alkyl ions formed by field- dissociation at the propargylic bond but the spectra are complicated by ion- molecule reactions taking place in the multilayer at the emitter tip.61 An alternative method involves the conversion of the acetylene in a single reaction into a mixture of two ethylene ketals which will decompose very predictably.62 Two methods have been suggested for olefins.The simplest involves direct mass-spectrometric examination of the epoxide which will give intense peaks due to ~t-cleavage.~~ The alternative method is to hydroxylate the double bond and then methylate the hydroxy-gr~ups.~~ This method has been shown to be useful for compounds with more than one double bond. Long-chain dialkyl ethers have much simpler spectra when the molecules are weakly excited; with a source temperature of 90" and 12 ev electrons the only frag- ments observed from R,O are (RH)'. R+,and ROH,+.(j5 The mass spectra of catenanes are of interest in that a molecular ion cor- responding in mass to the sum of both rings is found. In the spectrum of (29)(j6 the elimination of keten gives rise to ions containing both rings but processes which involve fission of bonds in either of the rings produce ions of the same mass as found in the isolated ring compound.There is one ion which requires a hydrogen-transfer from one ring to the other; this could be called an intra- molecular ion-molecule reaction. 49 B. D. Tilak K. G. Das and H. M. El-Narnaky Experientia 1967,23,609. 50 D. H. Williams R. G. Cooks J. H. Bowie P. Madsen G. Schroll and S. 0.Lawesson Tetra-hedron 1967,23 51 J. H. Bowie G. E. Lewis and R. G. Cooks J. Chem. SOC.(B) 1967 621. 52 J. H. Bowie G. E. Lewis and R. G. Cooks Chem. Comm. 1967 284; Austral. J. Chem. 1967 20 1601. 53 J. H. Bowie R. G. Cooks S. 0.Lawesson and C. Nolde J. Chem. SOC.(B) 1967,616. 54 J. H. Bowie R. G. Cooks N. C. Jamieson and G.E. Lewis Austral. J. Chem. 1967,20 1601. 55 Y.Rahamin J. Sharvit A. Mandelbaum and M. Sprecher J. Org. Chem. 1967 32 3856. 56 R. A. W. Johnstone D. W. Payling and A Prox Chem. Comm. 1967 826. 57 J. H. Bowie S. 0.Lawesson F. Duss P. Madsen and R. G. Cooks Chem. Comm. 1967,346. 58 B. E. Job Chem. Comm. 1967 44. 59 R. T. Aplin and W. T. Pike Chem. and Znd. 1967 2009. 6o K. Ukai K. Hirose A. Tatematsu and T. Goto Tetrahedron Letters 1967 4999. 61 B. C. Patterssn and M. Seakins Trans Faraday SOC.,1967,63 1863. 62 H. E. Audier J. P. Beguk P. Cadiot and M. Fetizon Chem. Comm. 1967 200. 63 R. T. Aplin and L. Coles Chem. Comm. 1967 858. 64 W. Niehaus jun. and R. Ryhage Tetrahedron Letters 1967 4999. 65 M. Spiteller-Friedmann and G. Spiteller Chem. Ber.1967 100 79 66 W. Vetter and G. Schill Tetrahedron 1967,23 3079. Part (iu). Mass Spectroscopy A detailed study has been made of the relative ease of loss of alkyl radicals from ketals. As would be expected the most pronounced difference is between primary secondary and tertiary radicals but the increase in fragmentation probability with size within a homologous series is still ~onsiderable.~ Ionisation potentials of ally1 radicals are almost identical with those of isomeric non-allylic radicals,68 i.e. allylic stabilisation is the same for radicals as for carbonium ions. Stereochemical effects. Epimeric steroid pairs give distinguishable spectra.69 The stereochemistry of the A/B ring junction in 4,4-dimethylandrostan-6-one affects some of the fragment intensities particularly where there is a difference in the distance between sites of exchange of a rearranging hydr~gen.~' trans-Methyl 2,3,3-trimethylcyclopentylacetatehas a much more intense peak for the McLafferty rearrangement than its ~is-isomer.~' Presumably this process is favoured by the cis-relationship of the carbonyl group and the tertiary hydrogen atom in the trans-isomer.Similarly in the case of the 2-acyl norbornanes the endo-isomer rearranges more rapidly.' In the 2,3-dimethylcyclobutanone series the cis-isomer always decomposes more rapidly than the trans-isomer but the difference shows up in all fragmentation processes not only in those involving the bond between the methyl-substituted carbon atoms.73 In p-lactams such as (30) the effect is more specific.The products of fission a are more abundant from the cis-isomer than from the trans-isomer and the reverse applies to fission b.74 In the spectra of some 2-aryl-3-acylazetidines there is reported the first case of stereospecific process. The compounds in which there are aryl and acyl groups trans to each other show an (M -OH)' peak in their spectra which is completely absent from those of the cis-isomers. The molecular arrangement necessary is thought to be as in (31).75 67 J. T. B. Marshall and D. H. Williams Tetrahedron 1967 23 321. 68 S. Pignataro A. Cassuto and F. P. Losing J. Amer. Chem. SOC. 1967,89 3693. 69 N. S. Wulfson V. I. Zaretskii V. L. Sadovskaya A. V. Zakharychev S. N. Ananchenko and I. V. Torgov Tetrahedron 1967 23 3667; V.I. Zaretskii N. S. Wulfson and V. G. Zaikin ibid. p. 3683. 'O H. E. Audier M. FCtizon and P. Foy Bull. SOC. chim. France 1967 1271. J. Cason and A. I. A. Khodair J. Org. Chem. 1967,32 575. 72 A. F. Thomas and B. WiIIhaIm Helv. Chim. Acta 1967 50 826. 73 H. E. Audier J. M. Conia M. Fktizon and J. Gore Bull. SOC. chim. France 1967 787. 74 H. E. Audier M. FCtizon H. B. Kagan and J. L. Luche Bull. SOC. chim. France. 1967 2297. 75 J. L. Imbach E. Dornes N. H. Cromwell H. E. Baumgarten and R. G. Parker J. Org. Chem. 1967.32. 31 23 John M. Wilson Chemical ionisation. The reactions of s-butyl ions (from n-butane) and t-butyl ions (from isobutane) with olefins have different reaction rates.76 The study of such gas-phase reactions may be useful for the identification of the ions used.In benzene at high pressure C6H6+ is unreactive towards benzene except possibly in a symmetrical charge-transfer process. All fragment ions from benzene react by electron-transfer with C6H6 to form C6H6+.77 Chemical ionisation mass spectra of cycloalkanes show (M -H)+ ions whose abundance can be correlated with the number of hydrogen atoms available for abstrac- ti~n.~~ The reaction leading to the formation of cyclo-olefin ions by elimination of side-chains appears to be unique in this sort of system in being endothermic. In the case of aromatic hydrocarbons the most important processes are proton- ation and alkylation of the aromatic ring.79 Some hydride-abstraction does take place but only from saturated C-H groups in side-chains.In toluene and cycloheptatriene the hydrogen atoms retain their identity.80 76 M. S. B. Munson J. Amer. Chem. SOC. 1967,89 1772. 77 F. H. Field P. Hamlet and W. F. Libby J. Amer Chem. SOC.,1967,89 6035. '13 F. H. Field and M. S. B. Munson J. Amer. Chem. SOC. 1967,89,4272. 79 M. S. B. Munson and F. H. Field J. Amer. Chem. SOC. 1967,89 1047. *' F. H. Field J. Amer. Chem. SOC.,1967,89 5328.