April 5, 1958
1571
SPECTRA OF hfETHYLCYCLOPENTANE AND ? v ~ E T H Y L - C ' ~ - C Y C L O P E N T A N E [CONTRIBUTION FROM
THE SHELL
DEVELOPMENT COMPANY]
The Mass Spectra of Methylcyclopentane and Methyl-C13-cyclopentane BY D. P. STEVENSON RECEIVEDSEPTEMBER 23, 1957 The mass spectra, as excited by single electron impact by 75 volt electrons, of methylcyclopentane and methyl-ClJ-cyclopentane are described and briefly discussed. Certain conclusions with respect to the modes of dissociation of the metliylcyclopentane ion into fragments are presented.
I n connection with the analysis of reaction products arising in the course of the study of the mechanism of the aluminum bromide catalyzed isomerito cyclohexanezation of ~nethyl-C~~-cyclopentane C13, we have had occasion to compare critically the mass spectra of methylcyclopentane (natural carbon isotope abundance) and the C13-labeled, methyl-C13-cyclopentane. Since the comparison yields facts that are of interest with respect to the nature of the mechanisms of the unimolecular dissociation reactions of molecule-ions formed by single electron impact that are pertinent to the general theory of such dissociation reactions2 and to the interpretation of the appearance potentials of fragment ions in the mass spectra of cyclanes,3we present herewith the results of this study. The mass spectra were obtained with a modified Westinghouse Type LV instrument.4 The mass spectra were recorded by scanning the magnetic analyzing field while keeping the ion accelerating potential at - 1000 volts. The ions were withdrawn from the ground potential ion source by application of a 5 volt ion repeller potential, equivalent to a field of 13 volts/cm. The energy of the ionizing electrons obtained from a carburized tungsten cathode was a nominal 75 volts, and the ion source temperature was 125 f 5". The methylcyclopentane with natural carbon isotope abundances was a Phillips Research Grade sample that had been treated with 5 2 0 4 and then percolated through a column packed with Ascarite to remove excess NpOa and then silica gel for the removal of the nitro-olefins and benzene. After this treatment the methylcyclopentane had an optical density (1 cm. path against distilled water in the referen$e cell) less than 0.2 a t all wave lengths greater than 2000 -4.aad less than 0.02 a t all wave lengths greater than 2200 A. The intensity of the ions of mass to charge ratio 86 and 71 in the mass spectrum of the portion used indicated the hexane content of the sample to be less than 0.057, m. The methyl-C13-cyclopentane was prepared5 by the reaction of cyclopentanone with methylmagnesium iodide (ca. 50% C13) to form the tertiary alcohol, methyl-C13-cyclopentanol, that was dehydrated by heating with oxalic acid to the olefin. This was hydrogenated to methyl-C13-cyclopentane over Raney nickel. The product was treated with small portions of concentrated sulfuric acid t o remove residual olefin and ether.
The mass spectrum of monoisotopic (CI2) methylcyclopentane was deduced from that of the "natural" methylcyclopentane by correcting for the C13containing ions under the assumption of negligible isotope effects on the fragmentation pattern, emindiploying the apparent C13 abundance, 1.01~o,6 cated by the intensity of the ion of m/q = 85 (C6C12-C:3) relative to that of the ion of m/q = 84 (1) 0 . Beeck, J. W. Otoos, D. P. Stevenson and C. D. Wagner, t o he submitted for publication. (2) H. M. Rosenstock, M. B. Wallenstein, A. L. Wahrhaftig and H. Eyring, PYOC. .Vat. A c a d . S c i . ( U S A ) ,38, 667 (1952). (3) (a) J. L. Franklin and H. E. Lumpkln, J. Chem. P h y s . , 19, 1073 (1951); (h) J. L. Franklin and F. H. Field, i b i d . , 21, 550 (1953). (4) D. P. Stevenson and C. D. Wagner, ibid., 18, 11 (1950). (5) Preparation and purification b y Dr. C. D. Wagner. ( 6 ) This value is in good agreement with t h a t previously found in butanes and propane, 1.01%, see ref. 4 above.
(CaH12-C',3). This mass spectrum is shown in the second column of Table I. I n this table of mass spectra the intensities of the ions are recorded as the fraction (times 100) the intensities the ions constitute of the total mass spectral intensity. In computing and tabulating the mass spectra doubly charged ions totaling ca. 0.3% of the total intensity were neglected. TABLE I THE MASS SPECTRA
OF
METHYL-C'~A N D h'IETHYL-c'3-
CYCLOPENTANES
75 volt ionizing electrons, ion source temperature 125 zt 5" m/a
85 84 83 82 81 80 79 78 77 70 69 68 67 66 65 57 56 55 54 53 52 51 50 44 43 42 41 40 39 38 37 30 29 28 27 26 25 16 15 14
MelZ-CP
8.16 0.29
.01 .07 .00 .12 .01 .12 .00 12.09 1.38 0.69 .06 .20 .00 31.0 7.07 0.90 .88 .16 .44 .32 . 00 2.73 6.41 14.30 0.92 4.21 0.36 .13 .00 1.79 1.36 3.30 0.29
Mela-CP
% 'Cor.
8.16 0.29 .01
.07
.00 .12 ,Ol ,12 .00 5.84 5.41 1.30 0.55
27.9 8.04 2.89 0.79 0.40
Assumed
i 0.05
13
.1 f .03 i .01
fi 3
=t
.3 .3 C .06 =t .02 i .02
52 64
4
A
Z t
79
70 El3 88
1.94 =t ,052 4.19 i .05 10.2 i .7
7 . 5 5 i .1 2.09 i .03 2 . 6 5 +C .05
0.73 3= 1.75 i 1.96 2.03 i 0.23 =k
+
.O1
.03 .03 .04 .01
70 50 4: 1
57 54
-00
.00 .20 .01
0.057 i: 0.005 .122 =k 0.005 .01
12 ti2 0
1372
D. P. STEVENSON
VOl.
The gross isotopic constitution of the “labeled” methylcyclopentane was computed from the intensities of the ions of m/p = 86, 85 and 84 in the mass spectrum, by use of the relative intensities of the ions of m/q = S4,83 and 82 in the methylcyclopentane-Cy mass spectrum, assuming negligible isotope effects on the relative parent and fragment ion sensitivities of the various isotopic species. The quality of the agreement between the total specific intensity of CS ions in the mass spectrum of the natural and the labeled material ( 2%) provides reasonable assurance that the neglect of isotope effects is warranted. The gross isotopic composition thus found was C6H12-C013= 44.65Y0, C6H12C1: = 52.65% and c6H1a-C;~= 2.70% (all i O . l % ) . This composition corresponds to a 53.0 =t 0.1% C13-content of the E. K. Co. methyl iodide. The concentrations of the individual isotopic isomers calculated from the gross isotopic composition and the assumption that the C13-content of the cyclopentanone was the same as that of the natural methylcyclopentane are : methyl-Cla-cyc1opentane = 5O.4y0,, methylcyclopentane-Cy = 2.2%, methyl-C13-cyclopentane-CY = 2.6% and methyl~yclopentane-C~~ =
so
per cent. of the originally observed intensity) the magnitudes of the corrections. Fifty per cent. retention of the methyl carbon by Cs-fragment ions in the methylcyclopentane mass spectrum would lead to expected intensities of m/p = 70 and 69 of 6.09 and 6.78, respectively, in the methyl-C13-cyclopentane mass spectrum. The observed intensities, 5.5 and 6.4, respectively, are in essential agreement with the “calculated” intensities. This finding of methyl carbon retention by the Cg ionic fragments is contrary to a priori expectation which was that there should be no such retention. The retention indicates that the structure of the C&9+ ion (m/p = 69 in the ordinary M C P mass spectrum) is probably not that of a cyclopentyl ion and that those methylcyclopentane ions that dissociate to (26 ionic fragments largely undergo prior rearrangement to normal hexene ions that subsequently lose a terminal carbon atom. T o the extent that this behavior of methylcyclopentane ions is a general one and is representative of the behavior of alkyl-cyclane ions, it is apparent that appearance potentials of ions of the type, CmHzm- I + , in the mass spectra of R-C,He, - cyalanes, cannot be employed to deduce energetic 0.1%. properties of cycloalkyl radicals or ions. It will The mass spectrum of the methyl-C 13-cyclopen- be noted that this behavior of the methylcyclopentane was deduced from that of the labeled sample tane ion is in marked contrast to that of such alkane in the following manner. First the observed spec- ions as have been studied. In the case of the altrum was corrected for C&&’3 using the concen- kanes, the dissociation of a methyl or ethyl radical tration 44.65y0 and the mass spectrum given for by the alkane ion is not preceded by rearrangethis substance in Table I. The residual mass spec- men t. trum indicated that among the C6 ionic fragments The intensity of the ion of m/q -- 57 in the ca. joyo retain the methyl carbon while ca. 100yo methyl-C 3-cyclopentane mass spectrum, = 27.0, of the C4 ionic fragments retain the methyl carbon. vs. 31.0 for that of m/p = 56 in the M C P - C Omass ~~ 1Iass spectra for the minor isotopic isomers, hle- spectrum indicates that there is no unique process CP-C113and Me-C13-CP-C13 were computed from leading to loss of an ethylene fragment by the C6that of C6H12-C013 using these approximate methyl- Hlz+ ion. A conceptually simple mechanism that carbon retention figures along with the further as- ~vouldlead to 10070 retention of the methyl carbon sumption that among the C3 ionic fragments there by the C4H6+ fragment would be is SOYo methyl carbon retention, while among the C c C2 ionic fragments there is complete randomness with respect to the carbons. Except in the cases of the ion of mass to charge ratio 71 and 58 the corrections for contributions of the minor iso0 b C topic species constitute less than 10% of the total correction and thus even quite large errors in where n and c would give riormal butene ions and the minor species corrections would not introduce b an isobutene ion, if the fission process were ncsignificant error in the final methyl C13-cyclopen- companied by simultaneous proton shift in the C4unit from the c6tane mass spectrum as far as the discussion and con- Ilsf residue. If the loss of a clusions we shall reach are concerned. I n the case Hlz+ ion were preceded by rearrangement to a of the ion, m/q = 71, only C6H12-C2l3 contributesand normal hexene ion, as was found to be the case for the calculated intensity agreed with that obscrved loss of a C,unit, it w d d bc expected that there to within 370. Again in the case of the ion, m/q = would be 50yoretention of the methyl carbon atom 58,only C&1&13 contributes and in this case the by the Cq ionic fragment. If the order of SO% of calculated intensity was 10% less than the observed the C4Hs+ fragments were formed by processes intensity On the basis of these agreements we like a,b and c and 20% by prior rearrangement, the conclude that the accuracies of the intensities de- intensities of the ions m / q = 57 and 56 in the duced for the fragment ions of the ilIe-C13-CP rnethyl-C13-cyclopentane mass spectrum would be ni:m spectrum are controlled by the accuracy of expected to be 27.9 anid 8.5, respectively. Apthe corrections for CGIT~?-CO~’. The repeatability o f proximate comparison of the initial portions of the r e I : ~ t i ~inass e spectral intensities is i1yi) of the ionizatioll efficiency curves uf the ioiis of m / q = S4 iucliviciual iiitensities. This figure has been used and 5(i in the ordinary metliylcyclopentaiie mass i n co~nputitigthe uncertainties of the relative ill- spectrum iiidicate that C2H4is the neutral fraginerit tensities of the ions in the Me-Ci3-CP mass spec- that accompanies the C4Hs + fragment ion. The greater than 50% retention of the methyl trum shown in the third column of Table I. There are shown in the fourth column of this table (;is csrbon by tlic ion C3H7’- ( m / q = 41 In the inethyl-
c:!
April 5 , 1958
SPECTRA OF METHYLCYCLOPENTANE AND h’fETHYL-C13-CYCLOPENTAKE
1573
potentials of the Cz units than of the complenientary Cq units. Similar evidence that the CZ ionic fragments in the mass spectrum of neopentane arise through a C two stage process, C5 + C4+ C Cz+ Cz +I C1, with essentially random selection of carbons in xc the second stage dissociation has been found by ,A’ I c-c Langer and Johnson’ in studies of the mass spectra d e of two C13-labeledneopentanes. It is well known8 that methyl ions are formed in where the fission is accompanied by simultaneous the mass spectra of the higher hydrocarbons by a t hydrogen atom transfer across the bonds that are least two different processes. One of these, the disbeing broken. The fact that the intensity of m/q = sociation of a doubly charged molecule- (or frag44 in the labeled compound mass spectrum is but ment) ion, leads to a methyl ion with sufficient ex0.7 times that of m/q = 43 in the mass spectrum of tra kinetic energy to cause the nominal m/q = 15 the unlabeled compound indicates that this ion is (C12H3)peak to have a “high mass” satellite. The not formed by a unique process. Other processes manner in which the mass spectra obtained in this that may contribute are (1) prior rearrangement to investigation were recorded, 1000 volt analyzing poa normal hexene ion followed by fission of the cen- tential, did not permit sufficient resolution of the tral bond that would lead to 50y0retention of the satellites of the C12H3+and CI3H3+ peaks for their methyl carbon and (2) secondary dissociation of a separate measurement, and thus we cannot make any CsHs+ ion formed as described above. The latter statements with respect to whether or not the life of process would lead to but 25Y0 methyl carbon re- the doubly charged ions is sufficiently great to permit tention. “rearrangement” prior to dissociation. The inAn essentially random selection of carbon atoms tensities reported in Table I for ions of m/q = 15 from CsH12 to form two carbon ionic fragments and 16 are for the ions with relatively little excess would lead to 33% retention of the methyl carbon kinetic energy, presumably those formed by secondatom by the ionic fragments. If the process of ary or tertiary dissociation of heavier f r a p e n t ions. formation of the two carbon fragments follows the The C13-retention, cu. 33%, by the “normal” sequence methyl ions is a factor of two greater than would be expected if the methyl ions were formed from an cot --.f Cat c2 --+ cz 2cz intermediate state of C6H12+ in which there was we should expect about 40% retention of the methyl randomization of the carbon atoms. There are two conclusions with resFect to the incarbon by the CZionic fragments, since we found above that there is about SOY0 methyl carbon reten- terpretation of the mass spectrum of a substance tion by the C4 ionic fragments. The observed in- that are to be derived from the preceding discustensity of C2H5+-C13, 0.73, in the methyl-C13-cyclo- sion. These are: (1) it is generally not possible to pentane mass spectrum is 40.5% that of CZH5+ in assign a unique mechanism of formation nor a the mass spectrum of the unlabeled methylcyclo- unique structure to a fragment ion in a mass specpentane, The absolute rate theory of Eyring and trum of a formally unsaturated hydrocarbon, and co-workers2 would lead to the expectation that the (2) the process leading to a particular ion may not two carbon ionic fragments should be formed as be the “obvious” one, and thus careful checks secondary dissociation products rather than as should be made with respect to the probable strucprimary dissociatior, products. This is a conse- ture of an ion before energetic conclusions are quence of the greater excitation energy required of based on an appearance potential of an i o i in a a molecule ion, C6H12+,for dissociation to C2H6+(or mass spectrum. C2H4+,etc.) plus C4Hi (or C4&, etc.) than for the EMERYVILLE, CALIF. complementary dissociation to CzH5 (or C2H4) (7) A. Langer and C . Johnson, J . P h y s . Chem., 61, SUI, 1010 (1957). plus C4H7+ (or CdH*+). The higher required ex(8) F. L. Mohler, V. H. Dibeler and R . M. Reese, J. C h e w . P h y s . , citation energy results from the greater ionization 22, 394 flY.54).
C13-cyclopentane mass spectrum) suggests that a major mode of formation of this ion is by the process
+
+
+
+
-
+ +
