Molecular Acoustics Part 8.-Conformational Changes in Heterocyclic Molecules Studied by the Ultrasonic Method G. ECCLESTON AND BY (MISS)V. M. GITTINS EVANWYN-JONES W. J. ORVILLE-THOMAS Dept. of Chemistry and Applied Chemistry University of Salford Salford M5 4WT Lancs. Received 21st December 1971 Ultrasonic relaxations observed in some 2-and 4-substituted tetrahydropyrans have been attri- buted to the perturbation of conformational equilibria involving chair to chair inversion and also internal rotation about the exocyclic bond. In symmetrically substituted 1,3-dioxans in which the above equilibria are inactive in ultrasonic relaxation studies a further relaxation associated with the equilibrium between chair and twist boat conformers has been observed.The dynamics of intramolecular conformational changes in liquids can be studied by the ultrasonic relaxation technique ; this paper reports such measurements on heterocyclic molecules. In order to illustrate the potential of the technique in investigating different types of intramolecular motions which can occur in any one system we have chosen as examples substituted tetrahydropyrans and 1,3-dioxans. The former molecules have been chosen for this study because they are model com- pounds for pyranose sugars. Because of its structure the tetrahydropyran molecule lacks the advantageous properties of 1,3-dioxan which lends itself ideally to the study of configuration and conformation using ultrasonic equilibration and n .m.r. techniques. Furthermore compared to 1,3-dioxans substituted tetrahydropyrans are not readily available commercially and in our experience have proved inconvenient to synthesize.Con-sequently there is very little information available on the dynamics of conformational changes in these molecules. ULTRASONIC THEORY When a sound wave is propagated through a system comprising an n-step con- formational equilibrium the changes in the concentration of the various conformers can be described by n -1 linearly independent differential equations. These may be solved to give the relaxation time spectrum of the system. Because of the coupling between the various states that are in equilibrium the relaxation times are functions of all the rate constants within the system. An ultrasonic relaxation in this system will be accompanied by a decrease in the absorption parameter ctlf2 with frequency in accordance with eqn (l) +B = n-c1 (1 +2nfzi)2 106 GITTINS ECCLESTON WYN-JONES AND ORVILLE-THOMAS 107 where a is the sound absorption coefficient at frequency5 Ai is the amplitude para- meter for step i zi is the ith relaxation time and B represents contributions from processes having constant values for alf2.The conformational mechanism can be differentiated from any other mole- cular mechanism because the relaxation times are independent of concentration. EXPERIMENTAL ULTRASONICS The ultrasonic measurements were carried out on a pulse apparatus which has been described previously.2 Both pure liquids and 60 % v/v acetone solutions were measured in the frequency range 25-105 MHz and the temperature range -40" to +60°C.INFRA-RED The infra-red spectra were measured on a Perkin-Elmer model 521 spectro-meter. PREPARATION OF COMPOUNDS The compounds studied in this work were I tetrahydro- pyran ; 11 2-methyl-tetrahydropyran ; 111 4-methyl-tetrahydropyran ; IV 2-methoxy- tetrahydropyran ; V 2-chloromethyl-tetrahydropyran; VI 2-bromomethyl-tetrahydro-pyran ; VII 2-iodomethyl-tetrahydropyran; VIII 2-chloromethyl-6-methyl-tetrahydro-pyran. Compounds I 11 111 and V were bought commercially. Compounds IV VI VII and VIII were prepared according to methods described in the literat~re.~-~ The purity of all the compounds was checked by boiling point g.1.c. and elemental analysis.RESULTS AND DISCUSSION In the present study no relaxation was observed in the parent molecule tetra- hydropyran (I) in the temperature and frequency ranges studied. One relaxation was observed at high temperatures in 2-methyl-(II) 4-methyl-(III) and 2-methoxy-(IV) tetrahydropyrans. Two well-defined relaxations were observed in 2-chloromethyl- (V) 2-bromomethyl-(VI) and 2-iodomethyl-(VII) tetrahydropyrans whereas in 2- chlorometh yl-6-methyltetrahydropyran (VIII) only one relaxation was observed and this was at low temperatures. All the separate relaxation processes observed were consistent with a single relaxation mechanism and the data were analyzed using i = 1 n = 2 in eqn (1). Solution studies showed that the molecular origin of the relaxations must be associated with conformational changes.Physico-chemical evidence shows that tetrahydropyran exists in the chair conformation ; thus the equilibrium set up by the process of ring inversion between the two chair forms is isodynamic and as such is therefore acoustically inactive. On the other hand the two chair forms of compounds 11 111and IV have different energies due to the axial or equatorial orientation which the substituent can adopt (fig. 1) ; thus the high temperature relaxation observed in these compounds has been attributed to the perturbation of the equilibrium shown in fig. 1. The conforma- tional energies of methyl substituents in the tetrahydropyran molecule can be regarded as additive; for the 4-methyl compound (111) the conformational energy is assumed to be approximately equal to that of a methyl group in cyclohexane,* viz 7.5 kJ mol-l in favour of the equatorial conformation.Similarly the free energy difference between the two chair isomers of 2-methyl-tetrahydropyran is assumed to be quite close to the free energy difference in 4-methyl-1,3-dio~an,~ viz 12.1 kJ moll-l in favour of the equatorial isomer. Conversely it was found experimentally that the methoxy group in 2-methoxy-tetrahydropyran is more stable in the axial position to the extent of 1.5 kJ mot1 lo; this is due to the anomeric effect which has been explained by Eliel l1 in terms of the "rabbit-ear effect ". Briefly this means that in a fragment of the type -X-CH2-Y- where X and Y are oxygen or nitrogen those con- formations will be preferred in which the number of 1,3- interactions between lone MOLECULAR ACOUSTICS pair electrons is a minimum.The anomeric effect has been defined l2 as the greater preference of an electron-withdrawing group for the axial position when it is located adjacent to a heteroatom in a ring than when it is located elsewhere. I X=Y=Z= H 11 X=Me,Y=Z=H 111 X = Z = H Y = Me IV X=OMe,Y=Z=H V X = CHzCl Y = 2 == H VI X = CH2Br Y = Z = H VII X = CHJ Y = Z = H VIII X = CHZCI 2 = CHj Y = H In the 2-halomethyl compounds both ring inversion and internal rotation about the exocyclic C-C bond can occur resulting in a six-stage coupled equilibria process as shown in eqn (2). eP + ePo + eP 11 It It aP + aP + aP where a and e refer to the axial and equatorial isomers respectively and Px is the usual nomenclature for primary (P) halides; X is the atom which is in the trans position with respect to the halogen when viewed along the exocyclic C-C bond.The kinetics of scheme (2) are described by five relaxation times but in practice only two discrete single relaxations were observed one at high temperature and one at low temperature. The relaxation times are well separated at a given temperature; thus for the purpose of this discussion the relaxations are treated independently. By comparison with the relaxational behaviour of compounds 11,111 and IV the high temperature relaxation has been attributed to a chair-chair inversion and the low temperature relaxation to internal rotation of the halomethyl group.In the absence of any additional conformation data on tetrahydropyrans the preference of the halomethyl group for the axial or equatorial position (see fig. 1) must first be considered in order to obtain information about the conformational analysis of the 2-halomethyl-tetrahydropyrans. This is not straightforward since two opposing intramolecular effects steric and anomeric can occur. The anomeric effect can arise in this case from the parallel orientation of an unshared electron pair on the oxygen atom with the electronegative halogen. Ultrasonic relaxation data cannot be used directly to predict the more stable conformational isomer ; however the axial or equatorial preference of the halomethyl group can be inferred from the ultrasonic behaviour of 2-chloromethyl-6-methyl tetrahydropyran (VIII) in relation to the behaviour of the 2-halomethyl derivatives (V-VII).The final stage of the preparation of compound VIII involves the elimination of a molecule of water from 1-chloroheptane-2,6-diol with phosphoric acid. If the distillation is carried out slowly under 50 Torr pressure the most stable isomer will GITTINS ECCLESTON WYN-JONES AND ORVILLE-THOMAS 109 be formed almost exclusively ; thus the 6-methyl substituent will occupy the equatorial position and the 2-chloromethyl group will occupy its more preferential conforma- tion. The absence of an ultrasonic relaxation at high temperatures indicates that ring inversion does not occur to such an extent that it can be detected in this study.The only way to explain the present data therefore is to assume that the chloromethyl group also prefers the equatorial position. The absence of ring inversion can then be explained in terms of the 1,3-diaxial interactions between the large substituents at C2 and C6 (fig. 2) whch will cause the ring to be essentially locked in the diequa- torial conformation. FIG. 2. Consequently it is reasonable to assume that the equatorial conformation of the halomethyl group in compounds V-VII also will be more stable than the axial con- formation. Further evidence that the conformation of the chloromethyl group is the same in compounds V and VIII was obtained from an infra-red study of these compounds. Further implications of the ring inversion process in compounds 11-VII can now be considered.The enthalpies and entropies of activation AH and AS for the backward reaction i.e. the less stable to more stable step can be obtained in the usual way from a plot of loge(Tz(l +K))-l against 1/T. These are listed in table 1 for compounds 11 V VI and VII. The enthalpy difference AH" between the two chair forms of 2-methyl-tetrahydropyran has already been estimated (ca. 12 kJ mol-I) ; thus the enthalpy barrier for the forward step i.e. the more stable to less stable transition (AH:) can be estimated as 45 kJ mol-l. This value is in excellent agree- ment with the enthalpy barrier found for tetrahydropyran which is 42f5 kJ mol-l. TABLE PARAMETERS FOR CHAIR-CHAIR INVERSION 1 .-THERMODYNAMIC AH$ I AS$ I "2 801 AGO/ compound k3 mol-1 J K-1 mol-1 kJ mol-1 kJ mol-1 I1 33.2 -1.9 33.8 12.1 I11 ca 34 7.5 IV ca 34 3.35 V 22.9 -22.5 30.6 VI 26.4 -11.5 30.5 VII 10.4 -59.2 31.3 the error in the above values is f10 % In the temperature range studied the relaxation frequencies (fc) for compounds I11 and IV lie outside the experimental frequency range of 25-105 MHz and only an estimate of the relaxation frequency could be made at higher temperatures.An approximate value of the free energy barrier AGZ was calculated from the relaxation frequencies at 80°C for compounds XI-VII and these are listed in table 1 together with MOLECULAR ACOUSTICS AGO values for 11 I11 and IV. For substituted cyclohexanes and 1,3-dioxans,13 the free energy barrier for the more stable to less stable step AGf+ has a reasonably con- stant value which is independent of the substituent.Thus from the present data a minimum value for the free energy difference between the two chair forms of the 2-halomethyl-tetrahydropyrans can be roughly estimated ; this value is 10 kJ mol-I. The corresponding minimum value for the 2-halomethyl-l,3-dioxans can be esti- mated l4 similarly as ca. 15 kJ mol-l. These free energy differences show that the conformational equilibria in 2-halomethyl-tetrahydropyrans and 1,3-dioxans are biased towards the equatorial conformer showing that the anomeric effect is negligible. It was not possible to make a full quantitative analysis of the ultrasonic spectrum of compound VIII due to the small quantity of sample available.However measure- ments at 25 MHz 45 MHz and 105 MHz indicated that there was no dispersion in the quantity a/f2with frequency in the temperature range 20-90°C but that the ultra- sonic spectrum in the range +20°C to -40°C was the same for compounds VIII and V. Since the conformational equilibrium in 2-halomethyl-tetrahydropyrans is biased towards the equatorial conformer the relaxation observed at lower temperatures can be attributed to the internal rotation of the side chain of the equatorial isomer. In order to gain more information on the number of isomers present an infra-red study of compounds V-VIII was carried out. Although the data for the liquid samples are not conclusive an extensive comparative study of the absorption bands in the region of the carbon-halogen stretching modes shows that the spectra of V and VIII are similar in the liquid phase.At least two and not more than three bands due to carbon-halogen stretching vibrations were observed in all the compounds V-VIII. In the 2-bromomethyl- compound three strong absorption bands characteristic of C-Br stretching vibrations were observed at 670 cm-l 656 cm-l 634 cm-l indi- cating that there is a substantial proportion of each of the three equatorial rotational isomers present in the liquid phase. The observation of a single relaxation for a 3-state model has been interpreted recently by one of us l5; the most probable explanation is that one step in the conformational mechanism is much faster than the other step.From the ultrasonic data the enthalpy barriers to internal rotation AH for compounds V VI and VII were found to be 12.7 kJ mol-l 11.7 kJ mol-1 and 29.5 kJ mol-1 respectively. The corresponding AH; values found for 2-chloro- methyl- and 2-bromomethyl-1,3-dioxansare 12.5 kJ mol-l and 19.7 kJ mol-1 respec- tively. A relaxation which is consistent with an intramolecular mechanism has been observed in the following symmetrically substituted 1,3-dioxans IX 2,2-dimethyl- 1,3-dioxan X 5,5-dimethyl-1,3-dioxan-2-spiro-cyclohexane XI 2,2,5,5-tetramethyl- 1,3-dioxan The chair-chair equilibrium for compounds IX-XI is acoustically inactive as shown for IX in fig. 3. This relaxation has been attributed to the perturbation of the equili- brium between the chair and twist-boat conformers (fig.4). From the relaxation data the activation-free energy barriers (AG&) have been calculated and are listed in table 2 together with the activation-free energy barriers for the chair-chair inversion (AG&> obtained from n.m.r. data. Assuming that both processes proceed via the same transition state the difference between the n.m.r. and ultrasonic activation parameters gives an estimate of the free energy difference AGO (twist-chair) between GITTJNS ECCLESTON WYN-JONES AND ORVILLE-THOMAS the chair and twist-boat conformers in 1,3-dioxans. These values which are also listed in table 2 can be compared with other independent estimates of AGO (23.4_+ 2.5 kJ mol-l) found for 2,2,4,6-tetramethyl-l,3-dioxanand of AHo (23.83.7.9 25.1 +_ 1.2 26.3 _+ 1.2 kJ mol-I) found for 4-t-Bu-6-methy1 -6-isopropyl- arid -6-cyclohexyl- 1,3-dioxan respectively.\ Me Me FIG.3. FIG.4. TABLE X-FREE ENERGY VALUES FOR COMPOUNDS IX-XI AG;?,kJ mol-1 AG;98./kJ mol-1 AGo/kJ mol-1 compound twist-boat chair-chair twist-chair IX 21.2 31.1 9.9 X 24.8 44.8 20.0 XI 25.4 36.9 11.5 the error in the above values is f10 % Research studentships from S.R.C. (to V. M. G.) and Unilever Ltd. (to G. E.) are gratefully acknowledged. The S.R.C. also provided funds for constructing the ultrasonic equipment. J. Lamb in Physical Acoustics vol. 2 part A ed. W. P. Mason (Academic Press New York 1965) p. 23. E. Wyn-Jones and W. J. Orville-Thomas Truns.Furaday SOC.,1968 64 2907. G. F. Woods and D. N. Kramer J. Amer. Chem. SOC.,1947,69,2246. Patent specification 56613 No. 792 163 G.B. S. A. Barker J. S. Brimacombe M. R. Harnden and M. Stacey J. Chem. Soc. 1961 5256. J. Cologne and P. Lasfargues Bull. SOC. Chim. Frunce 1962 177. G. Gatti A. L. Segre and C. Morandi J. Chem. SOC.B 1967 1203. E. L. Eliel N. L. Allinger S. J. Angyal and G. A. Morrison Conformational Analysis (Inter-science New York 1964). E. L. Eliel and Sr. M. C. Knoeber J. Amer. Chem. SOC.,1968,90(13) 3444. lo A. J. de Hoog H. R. Buys C. Altona and E. Havinga Tetrahedron 1969,25 3365. E. L. Eliel Kemisk Tidskrif 1969 81 22. l2 E. L. Eliel and C. A. Giza J. Org. Chem. 1968,33 3754. l3 G. Eccleston Ph.D. Thesis (Salford 1970). l4 G. Eccleston B. Walsh E. Wyn-Jones and (in part) H. Morris Truns. Furuduy SOC.,1971,67 3223. l5 E. Wyn-Jones and J. Rassing to be published. G. Eccleston and E. Wyn-Jones,J. Chem. SOC.B 1971;2469. l7 K. Pihlaja Acta Chem. Scand. 1968,22 716. l8 M. Anteunis and G. Swaelens Org. Mag. Resonance 1970 2 389. l9 H. Friebolin H. G. Schmid S. Kabuss and W. Faisst Org. Mag. Resonance 1969 1 67. 2o V. I. P. Jones and J. A. Ladd Trans. Faruday SOC.,1970,66,2948.