|
1. |
The thermal decomposition ofcis- andtrans-2-methoxy-4-methyl-3,4-dihydro-2H-pyran |
|
Journal of the Chemical Society, Perkin Transactions 2,
Volume 1,
Issue 1,
1975,
Page 1-3
John F. Collins,
Preview
|
PDF (395KB)
|
|
摘要:
JOURNALOFTHE CHEMICAL SOCIETYPERKIN TRANSACTIONS I tPhysical Organic ChemistryThe Thermal Decomposition of cis- and frans-2-Methoxy-4-methyl-3,4-di hydro-2WpyranBy John F. Collins, Henry M. Frey," and Neil S. Isaacs, Chemistry Department, University of Reading,Whiteknights, Reading, Berks.The thermal decomposition of both cis- and trans-2-methoxy-4-methyl-3.4-dihydro-2H-pyran has been investi-gated in the gas phase from 287 to 345'. In this temperature range both isomers decomposed to give croton-aldehyde and methyl vinyl ether by a homogeneous process that obeyed first-order kinetics. The data for thetrans-isomer yielded the Arrhenius equation (i). Similarly the cis-isomer gave equation (ii). These resultslog k,/s-l = 14.246 f 0.1 06 - (201.46 + 1-20 kJ rnol-l)/RTin 10 (0(ii) log k2/s-l = 13-958 f 0.062 - (1 96.00 f 0.70 kJ m ~ l - ~ ) / R T l n 10are discussed in terms of a unimolecular decomposition occurring by a concerted mechanism.THE thermal decomposition of 3,4-dihydro-ZH-pyran is ahomogeneous first-order process in the gas phase1 Theavailable evidence suggests that the reaction is uni-molecular and concerted. Further support for thismechanism has been obtained from studies on the pyro-lysis of some substituted dihydropyrans , especially thekinetic measurements on the 2-methoxy and 6-methylderivative^.^ This work has allowed more detailedsuggestions to be made about the structure of the acti-vated complex and leads to the conclusion that the com-plex is far from symmetrical in the sense that C-0 bondrupture is appreciably more developed than C-C bondrupture (and, conversely, G O bond formation ' leads 'C=C bond formation).To determine the effects ofstereochemistry on reaction rate and see whether furtherinsight could be gained into the nature of the activatedcomplex, by determining rate parameters for geometricalisomers, we undertook the work reported in this paper.EXPERIMENTALJIateriaZs.-A mixture of cis- and tvans-Z-methoxy-4-niethyl-3,4-dihydro-ZH-pyra t i (MMDP) was prepared byheating equimolar quantities of methyl vinyl ether andcrotonaldehyde in a bomb at 160" overnight. The requiredNMDP was obtained from thc reaction mixture as a paleyellott- liquid by distillation undcr reduced pressure ( 1 5inniHg; 56"). Analysis by g.1.c.on four different columnsshowed the presence of two major components with peakareas in the ratio of 2 : 3 with the smaller peak emergingfirst. We designated these isomers A and B, respectively.The two isomers were separated by preparative g.1.c. usinga Perkin-Elmer 452 instrument with a fluorosilicone oilcolumn and operated at 66". Liquid samples (ca. 60 pl)were the largest that could be injected if reasonable resolu-tion was to be obtained. The samples of the two isomerswere essentially free from impurity save for the presence ofthe other isomer. Thus ' pure ' A contained 6% B andIn the n.m.r. spectra of the two isomers, double resonanceexperiments showed that Hb was coupled to Hfl and Hf, andpure ' B ca.4% A.OMeHbno others. In isomer h the two J values were almost identi-cal at ca. 2.6 Hz whereas in isomer B they were 7.4 and 2-4Hz. Identical couplings would arise with Hb equatorial andHI, and Hf2 axial and equatorial. Non-identical couplingswould result from Hb in the axial conformation. Themethyl group is expected to be $-equatorial and the pre-ferred conformation for thc methoxy-group axial. Thus Awas identified as the trans- and B as the cis-isomer.Both isomers were dried over a sodium mirror and de-gassed before each run. All other compounds used werecommercially available samples.Apparatus.-A conventional ' static ' high vacuumC. A. Wellington, J . Chetn. SOC. ( A ) , 1966, 2684.H. M. Frey, R. G. Hopkins, and N. S. Isaacs, J.C.S.PevkilaC . S . Caton, ,r. Anteif. C h m . SOG., 1969, 91, 7869.II, 1972, 20822 J.C.S. Perkin I1kinetic apparatus eqaipped with greaseless stopcocks wasused which was identical in all important details with thosedescribed earlier.a Pyrolyses were carried out in Pyrexreaction vessels immersed in a high temperature fused saltthermostat .4 Pressure changes were monitored using a Belland Howell model 4-327-0003 transducer whose output wasfed to a potentiometric recorder. Analysis of reaction mix-tures was by g.1.c. using a Perkin-Elmer F11 instrumentfitted with a heated gas sampling valve and a flame ionis-ation detector. Signals from the detector were fed to arecorder fitted with a ball and disc integrator.RESULTSIn preliminary studies both cis- and trans-MMDP werepyrolysed and the reaction mixtures analysed by g.1.c.Both isomers gave the same two products which were identi-fied as methyl vinyl ether and crotonaldehyde with onlytrace amounts of any other compounds.These trace com-ponents which were eluted before the ether on the chromato-gram totalled well under 0.1 yo of the products and no furtherattempt was made to identify them. Several pyrolyseswere required before reproducible results were obtained andthis was attributed to the necessity of ' ageing ' the reactionvessel surface.trans-MMDP.-In a series of runs, trans-MMDP wasadmitted to the heated reaction vessel and the pressurechanges monitored by displaying the output of the trans-ducer using a variable speed potentiometric recorder.Hence the total pressure at any time could be obtained fromchart measurements. For each run a number of pressure,time pairs of datum points were obtained and from these thecorresponding rate constant obtained by a least squarestreatment of the log (2P0 - Pt) against time plot (where Pois the initial pressure and Pt the total pressure a t time t ) .These plots were accurately linear (correlation coefficients>0.9998) and the rate constants were independent of theinitial reactant pressure from 4 to 12 Torr.At each of 10temperatures in the range 291-345" several runs werecarried out and rate constants evaluated from the pressurechanges. The mean values obtained at each temperatureare shown in Table 1.TABLE 1Rate constants for the decomposition of trans-MMDPT/"C 291.65 298.2 304.4 308.7 313.61 04k Js-1 0-395 0-681 1.069 1.47 2.10T/"C 318.9 324.4 332.3 336-0 346.1104k,/s-1 2.96 4.245 7.41 9.37 16.43The results in Table 1 are shown as an Arrhenius plot inthe Figure. From this plot the Arrhenius parameters weredetermined by least squares [equation (l)].The errorlimits are 1 standard deviation.1OghJs-l = 14.246 f 0.106 -(201.46 f 1-20 kJ wol-l/RTln 10) (1)To confirm that pressure changes were an appropriatemethod to monitor composition, a series of runs was carriedout at 312.2" in which the extent of reaction was determinedanalytically by g.1.c. Since it proved difficult to determineall the components of the reaction mixture with the sameprecision, 1-methylcyclohexene was used as an internalstandard and the percentage of trans-MMDP decomposeddetermined by reference to this standard.In this way arate constant of 1.90 x 10-4 s-1 was obtained. The valuecalculated from the Arrhenius equation (based only onpressure data) is 1.86 x 10-4 s-l which indicated the validityof using pressure change to determine rate constants in thissystem.cis-MMDP.-Rate constants for the thermal deconiposi-tion of the cis-isomer were determined in the same iiinnner1Arrhenius plots for A, tram- and B, cis-MMDPas for the trans-compound. The mean rate constants ob-tained a t 13 temperatures in the range 287.4-345.4" aregiven in Table 2. As with the irans-isomer the appropriatefirst-order plots were linear and the rate constants inde-pendent of pressure from 4 to 12 Torr.TABLE 2Rate constants for the decomposition of cis-MMDPT/"C 287.4 293.4 301.65 308.26 310.5104k,/~-1 0.489 0.772 1.40 2.269 2.666T/"C 313.36 313.80 321.2 322-9 328.710*k,/~-~ 3-15 3.276 6.284 5.974 8-94T/"C 334.9 337.15 346.4104k,/~-1 13.37 16-49 26-03While 4 figures are often shown for the rate constants, this is notintended to indicate such a high precision.However, thesewere the values used without rounding in the subsequentleast squares analysis to obtain the Arrhenius parameters.The results in Table 2 are shown in the Figure and theArrhenius equation (2) was obtained from these results.log K,/s-' = 13.958 & 0.062 -(196.00 f 0-70 kJ mol-l/RTln 10) (2)A series of runs was carried out at 313.5" where theextent of reaction was determined by g.1.c.analysis using 1-methylcyclohexene as an internal standard. These runsA. T. Cocks and H. M. Frey, J . Chem. SOC. ( A ) , 1969, 16711975 3yielded a rate constant of 3.40 x 10-4 s-l which comparesreasonably with the value of 3.21 x lob4 s-l obtained fromthe Arrhenius equation.To determine whether there was an appreciable hetero-geneous component of the reaction some pyrolyses of mix-tures of l-methylcyclohexene and cis-MMDP were camiedout in a reaction vessel packed with Pyrex tubes. Thepacked reaction vessel had a surface : volume ratio 14 timesthat of the unpacked vessel. Several pyrolyses were neces-sary before anything approaching reproducible results couldbe obtained, Thus initially large and variable amounts oftwo ' new ' products appeared in the chromatogram andthere was considerable conversion of cis- into trans-MMDP.Even after many pyrolyses some of the ' new ' products werestill formed and there remained some cis-trans-isomerisationof the reactant.Nevertheless the overall rate of decom-position of the cis-MMDP (corrected for the formation of thetrans-isomer) was close to that obtained in the unpackedvessel. Since the two new products were not observed inthe unpacked vessel we attribute these entirely to a hetero-geneous reaction process, and their absence as indicativethat such processes were not significant in the unpackedvessel. While it did not prove possible to identify theheterogeneous products, the one formed in larger yield (ca.5% of the total product yield) was eluted before the methylvinyl ether and is therefore a low niolecular weight (frag-mentation) product.DISCUSSIONEvidence that the decomposition of 3,4-dihydro-2H-pyran (DHP) as well as some of its derivatives occurs by aconcerted process has already been discussed.2 Further,it has been suggested that the activated complex is asym-metric with respect to the extent of C-0 and C-C break-age but only slightly polar.The data obtained in thepresent work together with relevant previous studies aregiven in Table 3. Inspection of Table 3 makes it im-mediately clear that the free energies of activation for theTABLE 3Kinetic parameters for dihydropyran decompositionsAG~,o,llog(A Is1) E / k J mol-l k J mol-lDHP 14.63 219.4 201-86-Methyl-DHP 14.45 214.2 198-72-Methoxy-DHP 14-42 203-1 187.9trans-MMDP 14-25 201.5 188.2cis-MMDP 13-96 196.0 186.1two compounds reported here are very close to that for2-methoxy-DHP.Thus the overall effect of the methylgroup in the 4-position is small. trans-MMDP can existin the most favourable conformation with the methylgroup in the equatorial and methoxy-group in the axialposition, whereas for cis-MMDP this is not possible.This isomer is presumed to have both the methyl andmethoxy-groups in equatorial positions. For a largenumber of cases in six-membered rings the value forAG(equatorial --+ axial) for the methyl group is ca.7.5k J rnol-l. In the sugars, pyranosides are more stable withthe methoxy-group axial when next to oxygen and AGfor this is ca. 6 kJ mol-l. Thus for trans-MMDP oneexpects virtually all molecules to be in the single con-formation since here AG = 13.5 kJ mol-l. This is incomplete agreement with the observed n.m.r. spectrumof this isomer. For cis-MMDP the difference betweenthe two conformations would only be between 1 and 2 kJmol-l on the basis of the values quoted above. Thiswould yield a ratio of conformers of ca. 2 : 1 and with thenormal fast chair-chair conversion should have led to a lesswell defined n.m.r. spectrum than that observed. Thissuggests that the value for AG corresponding to axial --+equatorial conformations of the methoxy-group is ratherless in MMDP than the 6 k J mol-l found for the sugars.A conformer ratio of 5-10: 1 for cis-MMDP wouldcertainly be consistent with the observed spectrum andthis would correspond to a value for AG of between 2 and4.5 kJ mol-l rather than 6 kJ mol-l.This would implythat AG for trans-MMDP + cis-MMDP is also smalland of this magnitude (i.e. 3.3 & 1.8 kJ mol-l).At 160" the ratio of cis- to trans-MMDP formed fromcrotonaldehyde and methyl vinyl ether was 3 : 2. Thiscorresponds to an activated complex for cis-f ormationwith a free energy of activation ca. 1.5 kJ mol-l less thanthat for the trans-isomer. This preference for cis-form-ation is exactly analogous to the frequent occurrence ofpreferential formation of endo-isomers in Diels-Alderreactions. In the present case stabilisation of theactivated complex by interaction between the highestoccupied molecular orbital of the vinyl ether x systemwith that of the lowest unoccupied molecular orbital ofthe crotonaldehyde leads to the slight cis-isomer prefer-ence. (There may also be a small effect due to coulombicterms.)If the energetic data on the cis- and trans-isomerstabilities which refer to room temperature remained thesame a t 600 K then we would expect the difference in thevalues of AGtm for the two isomers to be 3.3 + 1.5 =4.8 kJ mol-l. In fact, the observed value is only 2-1 kJmol-1. It would appear likely that both the differencein the free energies of the isomers at room temperatureand the stabilisation of the complex decrease with in-creasing temperature.Finally, we note that since trans-MMDP has a value forAGS,, greater than for 2-methoxy-DHP any effect on theactivated complex due to the methyl group stabilising anincipient radical centre is more than offset by stericeffects.We thank Dr. I. D. R. Stevens for valuable discussionconcerning the interpretation of the n.m.r. spectra of theMMDP.[4/1214 Received, 21st June, 1974
ISSN:1472-779X
DOI:10.1039/P29750000001
出版商:RSC
年代:1975
数据来源: RSC
|
|