- Email: [email protected]
NMR experiments on acetals—56
NMR EXPERIMENTS ON ACETALS-56 CONFORMATIONAL STUDY OF 1,3-DIOXENE R. CAMERLYNCK and M. ANTEUNIS* Laboratory for NMR Spectroscopy, Department of Organic Chemistry State University of Gent, Krijgslaan 271 (S4bis) B-9000Gent, Belgium (Receioedin UK 17 July 1974; Acceptedforpublication
24 March 1975)
Absbnct-From the ‘H-NMR parameters, extracted from the spectra of 1,3dioxene, S-Me-1,3diox4ene, J,SdiMe1,3diox4ene and of 2,4J-triMe-I,3diox_42ne, it follows that the preferred conformations are the Qdiplanar (sofa) C1form (Fig. IB). The barrier to ring reversal in S-Me-1.3dioxene was found to be 7.2 to.2 kcal/mole at -123” (in freon-21).
RJlswrs
INl'RODlJCl'ION In 1941 Lister proposed’ that the main conformation of
We have measured the ‘H-NMR parameters at cyclohexene is the half-chair (monoplanar form), and 3OOMHz of a series of four l$dioxenes, and the Pitzer’s first calculations’-’ have confbmed his view. extracted data are gathered in Table 1. 2,4J-TriMe-1,3Others&’ have recalculated the energies of a series of diox4ene is believed to exist preponderantly in an form, where the 2-Me has a possible geometric forms, and they all come to the same anancomeric conclusion. Experimental evidence, obtained from (quasi)equatorial 0rientation.t Because ‘J6 in this model compound is found to be ‘H-NMR: X-ray,9 microwave,” Raman and IR” and electrondiffraction,” are convincing criteria in this re- -15.0 Hz, whereas in 1Sdioxane it is only cu. -(ll-12) spect, so that relatively precise molecular data have Hz,” the dihedral angles between the r-lobes and the a-uC_” bonds are very near to 30’(330”) and 210”, become available.“.” Thus, the dihedral angles between according to the BartieldGrant effect.‘” This statement the r-lobes and the gC_” bonds of the a-methylene accords only with a sofa form. In both 4,5-diMe- and groups are resp. 225” and 345” and in the ring-inversed 2,4,5-triMe-1,3diox4ene a homoallylic coupling of conformation they are resp. 195”and 315”.” The next-most stable characteristic conformer is the 2eOHz is found between both H-6 protons and Me-L 12diplanar (sofa) form, less stable by some 1.26- 1.8 Recent INDO-MO reconsiderations’9.c’Imabout the quankcal/mole.’ The characteristic dihedral angles between the titative values for homoalIylic coupling, involving a free a-lobes and a--(~~_~bonds are 210”and 330”and between rotating Me group, suggest that for a transoid coupling, as the r-lobes and a’--~~_~ bonds they are 237”and 357”or in the present case, the 4’s must be close to 120”and 240”. This does not agree with a half-chair form (4 (idealized): 183”and 303”.” The sofa, however, is not a minimum energy conforma- 255”(2.8 Hz) and 135”(1.5 Hz)) nor with a half-boat form tion and neither is the 1,3diplanar (boat or half-boat) form (where the dvalues would give rise to 2.3 and 0.2 Hz), but (having 5.1 to 5.3 kcal/mole strai&’ with respect to the only with a C2 sofa-form+ as depicted in Fig. IB. Here the plane going through O&-C& bisects the CH,6 half-chair). moiety, giving rise to &values of 120”and 240“. The 0, Both of these conformations lie on the topomerization path, the itinerary being characterized by a unique top of sofa-form (Fig. 1A) on the contrary is characterized w about -5.3 kcal/mole, as ascertained by ‘H-NMR experi- &values of 147”and 267”, resulting in valuesI 0.94 and ments,* thus the critical conformation is very similar to 3.15 Hz, respectively, for homoallylic coupling. The O,-sofa furthermore would show up a value for ‘J6 of the l$diplanar form.” -12.5 to -13*0Hz,‘* which is not substantiated by our findings. Finally the fact that we observe a value -6.0 Hz *In I ,3dioxanes a 2-Me group is a “biasing” (anancomerizing) for ‘52 in 5-Me-1,3diox4ene favours decisively the its conformational energy amounts to group, as &-sofa. The normal value in 1,3-dioxanes being *J2= 4.1-4.3 kcal/mole.” This tendency to occupy exclusively the 6.1 Hz, this geminal coupling constant in a XCH,Y moiety equatorial position is believed to be so pronounced, that in whatever form dioxene may occur, this group will still prefer to a is dominated by the electron density and the direction of the p-X and p-Y lobes, and emperical rules with respect to great extent the (quasi)equatorial position. This is especially true idealized orientations have been developed:’ e.g. as a if one notes that even in CMe-c.hexene the conformational energy of a Me substituent amounts to 1.0-1.1 kcal/mole.6.‘6 function of the torsion angles between X-CH, and *In other compounds where 4 amounts to 1200 and 240”. CHrY. Taking as a good guess the torsion angles experimental values near 2.0 Hz have indeed been reported? observed in the different forms in cyclohexene, then we e.g. I-propylidene-indane (1% Hz), C&-2-Me-cpentenone predict8 as ‘J2 for a half-chair a value of -5.1 Hz,“*’ for the (2;OHzjand propylidene cpropane (2.0 Hz). Also in-6-Me-3.4 &sofa a value of -3.5 Hz, but as present data shows, dihvdro-2H-pyrane. ‘J(Me..,) = 2 Hz” (vide infra for a discussion of {his com&nd). 8Because the electron density of the O-3 lobes may have been decreased through mesomeric overlap with the double bond, in fact the reported values may become slightly more negative.du Thus in 4-oxo-1.3dioxane (half-chair”) ‘J2 = - (5.2-5.6) Hz.
almost a normal value5 for the Cl-sofa form. DISCUSION The postulation that in 1,3dioxenes the sofa form
would be stabilized with respect to the halfxhair, seems
1837
(1)
4.7%
-1.7 -
4$-Dimethyl-l$dioxeoe 5-Methyl-1Jdioxene (2) (3) 2.4.5-Trimethvl-Udioxeae (4)
-
-
2,5,
‘JH,&
-
2; 2.0 & 2.0
-
“IMe&
-
-
-
4.77,
H,
a Approximated value obtained at -WY in freon-21 at 100MHz.
-1.9
*J&H*
l$Dioxene (1)
_
J-values (in Hz)
4.85,
(4)
6.276
642,
H,
~~~~~~ ~~~
4%l,
2,4,5-Trime~yl-IJ~xene
_
(in ppm)
43.Dimethyl-13dioxene (3)
5-Methyl-1Jdioxene (2)
1,3-Dioxene
H*
&vaks
0;5 1
-
‘JMe,.Me,
4*05,3*%
392,
402
4.1%
Ha
-1.2 -0.95 -1.4
‘JMt,&
1.30,
-
c
Me,
-I.s -
‘Me,&
1.70,
1.72,
-
Me,
-
1 -15.0
I -
3Hs 6.3s
‘JH,,H,
1.49.
1.50,
1.51,
-
Me,
Table 1. ‘H-NMR oarameters of 13dioxenes obtaiwd at 300 MHz (5 ~01% in CCL, TMS internal)
5.1
1
-
‘JMez&
- -6.W
71
1839
NMRexperimentson acetals-pti 56
(I31 Calculated
1Saetext):
Calculated
YI*+~e)-O.9L
l-text
Hz 5Jik(e-~.~~. 3.15 Hz
SJIMe-~.61=2.2~z *56 v -15.5 Hz
‘J6
r2J2s., -6.2Hzl
d-13.0
12J2u-3.5
HZ Hz1
1
Fig. I.
reasonable. From inspection of ‘J, ‘J and ‘J values in 3,6 dihydro-2H-pyrane, the main conformation here also seems to be” the l&liplanar form. The main reason may reside in the fact that overlap of the p-0 r-system is maximal in this conformation.* However, other considerations agree with the expectation that in 1,3dioxene the sofa form becomes more stabilized in comparison with the half-chair form, in contrast with the situation in cyclohexene. In the latter, a sofa form suffers from some strain by an approach down to 2.55 A of the 3J H nuclei. In the t&sofa O-atoms are put into these positions (a similar situation can be traced in the ZH-pyrane nucleus). l&Dioxene is believed to exist in a half-chair form,” but here two ring O-atoms are equivalent to the n-orbital, and this may result in an “in-between” maximal stabilisation where both O-atoms physically indeed become equivalent in a static fashion, and not in a dynamic fashion (e.g. by equilibration between two equivalent sofa forms). Basic conformation and topomerization phenomena. As pointed out, the basic conformation in cyclohexene is the half-chair, while the 1,2diplanar form lies on the route of
*It may be noted that the &sofa possesses parallelp-orbitals for bothoxygen atomswhile the O,-sofa dots not. Thisparalellity, occasionally called “rabbitcar effecP was for some time believed to cause strain.It is not at all clearwhetherthis is indeed the case. IJ-Dioxane is, however,less stablethan I &dioxane?’ Yn 1.4dioxene (a half-chair conformation),” one might advancea stabilizationpe.roxygen of fcI.35.3) = 1 kcal/molefor the ground state. One would then expect a AG’ value of 5.3 + I.0 = 6.3 kcal/mole for 3,4dihydro-ZH-pyranein the halfchair form. 11is 6.6 kud/mole as the basic conformationis the sofa. Takingthe values for chexene as an approximatemeasure of the intrinsicdestabilisationbetween sofa and half-chair(with MG’ = 1.2- 1.8 kcal/mo1e6.‘),the overall stabiition by the pO-n overlap should thereforeamountto some 1.0t 0.3 t 1.2 B I.8 = 2.5 a 3.2 kcallmole. This is a very reasonablevalue and in fact is almost exactly the dilTertncein iotationalbarrierbetween that observed in propenyl methyl ether (3.83kcal/moleY and cis-butcne (0.65kcal/mole).”
ring reversal.” The fact that in 3,44hydro-ZH-pyrane and 1.3dioxene the basic conformation has now become the 1.2diplanar form, has an influence on the barrier to topomerization. For c.hexene AG’ = 5.3 kcal/mole,’ in 3,4dihydro-2H-pyrane it is increased to 6.6 kcal/mole,n in Mdioxene it is still higher, i.e. 7.3 kcallmole.P We have now found for l$dioxene a coalescence temperature of -123”* 1, corresponding to AG # - 7.2 k&mole. The increase in topomerization barrier may be rationalized, if one accepts that in the critical conformation during inversion the proposed p.O-R stabilisation would be lost. As the relative orientation between the p and ~-lobes must change during the process for reversal then the increase in AG # might be accounted for an extra stabilization of the ground state. It is nevertheless possible that this stabilization is apparent and that in the real situation a destabilization phenomenon occurs in the transition state.t EXPEUWWAL. Samples for ‘H-NMR spectm were purifiedpreviouslyby GC on QF, andCarbowax.‘H-NMRspectrawere obtainedat 18”with a VarianHR-300spectrometer,the doubleresonanceand INDOR with a Varian HA-100,equippedwith Muirhead-WiiD %B decade oscillators. Low temperature studies were done in freon-21with a VarianXL-100(12 in.). The tempswere measured with a calibrated Iron-Constantanthermocouple. All spectral analyses were checked afterwards by simulationson a Varian 620-i computerwith the aid of the SIMEQ 16/Hprogram.” I$-Dioxene.(a) Glycerolwas acetalatedwith formaldehydein acidic mediumaccordingto the methodof Hibbertand Tritser.)3 The mixture of S-HO-13dioxane and QHOCH,-1,3dioxolane was equilibratedwith a trace of p-toluenesulfonicacid at 70”for 24 hr. The crudemixturewas tosylatedin pyridmeconventionally and the tosylated mixture dissolved in a minimum of boiling benzene. After cooling the crystallii 5-TsO-13dioxane was collected (yield 40%). (b) Wnation of TsOH was performedin DMSOwith the aid of KOt-Buaccordingto the procedureof Snyderand Soto.y The volatile fractionwas collected and redistilled,b.p. 78’, yield 74%. (Found C 56.0, H 6.8. UC.: C 55.85,H 698%). 4$-ðyl-l$-dioxcne (3) and 2,4~-trimethyl-l$-dioxene (4). These compoundswere obtained as by-productsduringthe reductionstudy of 4 - Me - 5 - methylidene- 13 - dioxaneand 2,4 -
1840
R. CAMERLYNCK and M. ANIXUNIS
diMe - 5 - methylideae - I,3 - dioxane.)’ The isolation of the reaction mixture” was performed by GC on Carbowax, yield” of the crude mixture: 16% (3) and 19% (4). 5 -Me -I,3 -Diox -I -enc. 5 - Me - 5 -I- 1.3- dioxane” (1.5 g) was retluxed for 30 min in lOml EtOH, after 0.3 g Na had- been dissolved. Fractionation of the crude mixture gave a fraction (63O-81”)that was subsequently purified by GC (Gm Carbowax, 80°C).Two fractions (61: 39; respectively 5 - Me - I,3 - diox - 4 - ene and 5 - methylidene - I,3 - dioxane) were obtained. (Found: (main fraction): C 60.2,H 7.8,; Calc. C 60.1,, H 7.5%). (minor fraction) C 60.3,%, H 7.71%. Acknowledgements-We thank the Ministry for Scientific Programming for funds for the continuation of our research using a superconducting 300MHz spectrometer. The IWONL is thanked for a grant to one of us (R.C.). REFERENCES
‘M. W. Lister, 1. Am. Chem. Sot. 6.3, 143 (1941). *K. S. Pitzer, Science 101, 672 (1945). ‘C. W. Beckett, N. K. Freeman and K. S. Pitzer, 1. Am. Chem. Sot. 70, 4227 (1948). ‘J. Boeseken and J. Stuurman, Proc. Acad. Sci. 39, 2 (1936). ‘E. J. Corev and R. A. Sneen. /. Am. Chem. Sot. 77.250511955). 6R. Bucoui and D. Hainaui, BUN.Sot. Chim. Fr: 1366‘(1%$. ‘G. Favini, G. Buemi and M. Raimondi, J. Mol. Smrct. 2, 137 (1968). “F. A. L. Anet and M. Z. Haq, J. Am. Chem. Sot. 87,3147(1965). 9R. A. Pastemak, Acta Cry&.4,316 (1951);M. A. Loshem, Ibid. 5, 593 (1972). ‘OT.Ogata and K. Kozina, Bull. Chem. Sot. Japan 42,1263(1%9). “K. Sakashita, Nippon KO&I Zusshi 77, 1094(1956). ‘7. F. Chiang and $1H. Bat&, 1. Am. Chem. Sot. 91,1898 (1969); H. Geise and H. Buvs. Rec. Trao. Chim. 89. 114711970). “For a review see R. Bucourt, Topics in Sterewhemky, ‘Edited by E. Eliel and N. L. Allinger. Wiley-Interscience, New York (1974). “R. Bucourt, Bull. Sot. Chim. Fr. 2080 (1964). “F. W. Nader and E. L. Eliel, 1. Am. Chem. Sot. 92,305O(1970); cf. E. L. Eliel and Sr. M. Knoeber, 1. Am. Chem. Sot. 88.5347
(1966); K. Pihlaja and S. Luoma, ACU Chem. Scand. 22,240l (1968). 16C.B. Rickbom and S. Livo, 1. Org. Chem. 30, 2212 (1%5). “M Anteunis, D. Tavemier and F. Borremans, Bull. Sot. Chim. Belg0 75, 3% (1966). “M. BarIieldand D. M. Grant, 1. Am. Chem. Sot. ES,1899(1%3). ‘M. Barfield and S. Stemhell, Ibid., 94, 1905(1972). ‘OM.Barlield and D. hf. Grant, Adu. hfagn. Res. 1, 149 (1965). “M. Anteunis. G. Swaelens and J. Gelan, Terrnhedron 27, 1917 (1971). “M. Anteunis, G. Swaelens, F. Anteunis-De Ketelaere and P. Dirinck, Bull. Sot. Chim. Beiges 80, 409 (1971). “J. C. Martin. Bull. Sot. Chim. Fr. 277 11970). =E. L. Eliel, iemisk Tidskrifi 81.22(1964);R.‘O. Hutchins, L. D. Kopp and E. L. Eliel, 1. Am. Chem. Sot. 90,7174 (1968);E. L. Eliel, Accounts Chem. Res. 3, 1 (1970). =A. Snelson and H. A. Skinner, Trans. Faraday Sot.. 57, 2125 (I%]); cf. K. Pihlajaand J. Heikkila, Acta Chem. Stand. 21,239O (1967). ‘%. Piundt and S. Farid, Tetrahedron 22.2237 (1966):R. C. Lord, T. C. Rounds and T. Ueda, 1. Chem. Phys. 57, 2572 (1972);K. Pihlaja, Acfa Chem. Stand. 25, 451 (1971). =C. H. Bushweller and J. W. O’Neil. Tekzhedron Uters 4713 (1969). =R. H. Larkin and R. C. Lord, 1. Am. Chem. Sot. 95,5129(1973). 9. Gagnaire, E. Payo-Subiza and A. Rousseau, Bull. Sot. Chim. Fr. 2633 (1%3). )“J. P. Lorve, Progress in Physical Organic Chemisrry(Editedby A. Streitwieser and R. W. Taft) Vol. 6: Interscience. New York
VW.
“J. F. Kilpatrick and K. S. Pitzer, J. Res. Natl. Bur. Std 37, 163 (1946). “We thank Dr. de Bie (Un. Utrecht, Holland) for the SIMEQ 16/H-software. “H. Hibbert and S. Trister, Can. 1. Res. 14B, 415 (1936). “C. H. Snyder and A. R. Soto, /. 0~. Chem. 29, 742 (1%4). “M. Anteunis and R. Camerlynck, Tefrahedron, subsequent paper. %F. Borremans, Ph.D. Thesis 1975 Gent; M. Anteunis and F. BoFemans, in preparation. “P. A@, Suomen Kemistihehti B46, 151 (1973).
24 March 1975)
Absbnct-From the ‘H-NMR parameters, extracted from the spectra of 1,3dioxene, S-Me-1,3diox4ene, J,SdiMe1,3diox4ene and of 2,4J-triMe-I,3diox_42ne, it follows that the preferred conformations are the Qdiplanar (sofa) C1form (Fig. IB). The barrier to ring reversal in S-Me-1.3dioxene was found to be 7.2 to.2 kcal/mole at -123” (in freon-21).
RJlswrs
INl'RODlJCl'ION In 1941 Lister proposed’ that the main conformation of
We have measured the ‘H-NMR parameters at cyclohexene is the half-chair (monoplanar form), and 3OOMHz of a series of four l$dioxenes, and the Pitzer’s first calculations’-’ have confbmed his view. extracted data are gathered in Table 1. 2,4J-TriMe-1,3Others&’ have recalculated the energies of a series of diox4ene is believed to exist preponderantly in an form, where the 2-Me has a possible geometric forms, and they all come to the same anancomeric conclusion. Experimental evidence, obtained from (quasi)equatorial 0rientation.t Because ‘J6 in this model compound is found to be ‘H-NMR: X-ray,9 microwave,” Raman and IR” and electrondiffraction,” are convincing criteria in this re- -15.0 Hz, whereas in 1Sdioxane it is only cu. -(ll-12) spect, so that relatively precise molecular data have Hz,” the dihedral angles between the r-lobes and the a-uC_” bonds are very near to 30’(330”) and 210”, become available.“.” Thus, the dihedral angles between according to the BartieldGrant effect.‘” This statement the r-lobes and the gC_” bonds of the a-methylene accords only with a sofa form. In both 4,5-diMe- and groups are resp. 225” and 345” and in the ring-inversed 2,4,5-triMe-1,3diox4ene a homoallylic coupling of conformation they are resp. 195”and 315”.” The next-most stable characteristic conformer is the 2eOHz is found between both H-6 protons and Me-L 12diplanar (sofa) form, less stable by some 1.26- 1.8 Recent INDO-MO reconsiderations’9.c’Imabout the quankcal/mole.’ The characteristic dihedral angles between the titative values for homoalIylic coupling, involving a free a-lobes and a--(~~_~bonds are 210”and 330”and between rotating Me group, suggest that for a transoid coupling, as the r-lobes and a’--~~_~ bonds they are 237”and 357”or in the present case, the 4’s must be close to 120”and 240”. This does not agree with a half-chair form (4 (idealized): 183”and 303”.” The sofa, however, is not a minimum energy conforma- 255”(2.8 Hz) and 135”(1.5 Hz)) nor with a half-boat form tion and neither is the 1,3diplanar (boat or half-boat) form (where the dvalues would give rise to 2.3 and 0.2 Hz), but (having 5.1 to 5.3 kcal/mole strai&’ with respect to the only with a C2 sofa-form+ as depicted in Fig. IB. Here the plane going through O&-C& bisects the CH,6 half-chair). moiety, giving rise to &values of 120”and 240“. The 0, Both of these conformations lie on the topomerization path, the itinerary being characterized by a unique top of sofa-form (Fig. 1A) on the contrary is characterized w about -5.3 kcal/mole, as ascertained by ‘H-NMR experi- &values of 147”and 267”, resulting in valuesI 0.94 and ments,* thus the critical conformation is very similar to 3.15 Hz, respectively, for homoallylic coupling. The O,-sofa furthermore would show up a value for ‘J6 of the l$diplanar form.” -12.5 to -13*0Hz,‘* which is not substantiated by our findings. Finally the fact that we observe a value -6.0 Hz *In I ,3dioxanes a 2-Me group is a “biasing” (anancomerizing) for ‘52 in 5-Me-1,3diox4ene favours decisively the its conformational energy amounts to group, as &-sofa. The normal value in 1,3-dioxanes being *J2= 4.1-4.3 kcal/mole.” This tendency to occupy exclusively the 6.1 Hz, this geminal coupling constant in a XCH,Y moiety equatorial position is believed to be so pronounced, that in whatever form dioxene may occur, this group will still prefer to a is dominated by the electron density and the direction of the p-X and p-Y lobes, and emperical rules with respect to great extent the (quasi)equatorial position. This is especially true idealized orientations have been developed:’ e.g. as a if one notes that even in CMe-c.hexene the conformational energy of a Me substituent amounts to 1.0-1.1 kcal/mole.6.‘6 function of the torsion angles between X-CH, and *In other compounds where 4 amounts to 1200 and 240”. CHrY. Taking as a good guess the torsion angles experimental values near 2.0 Hz have indeed been reported? observed in the different forms in cyclohexene, then we e.g. I-propylidene-indane (1% Hz), C&-2-Me-cpentenone predict8 as ‘J2 for a half-chair a value of -5.1 Hz,“*’ for the (2;OHzjand propylidene cpropane (2.0 Hz). Also in-6-Me-3.4 &sofa a value of -3.5 Hz, but as present data shows, dihvdro-2H-pyrane. ‘J(Me..,) = 2 Hz” (vide infra for a discussion of {his com&nd). 8Because the electron density of the O-3 lobes may have been decreased through mesomeric overlap with the double bond, in fact the reported values may become slightly more negative.du Thus in 4-oxo-1.3dioxane (half-chair”) ‘J2 = - (5.2-5.6) Hz.
almost a normal value5 for the Cl-sofa form. DISCUSION The postulation that in 1,3dioxenes the sofa form
would be stabilized with respect to the halfxhair, seems
1837
(1)
4.7%
-1.7 -
4$-Dimethyl-l$dioxeoe 5-Methyl-1Jdioxene (2) (3) 2.4.5-Trimethvl-Udioxeae (4)
-
-
2,5,
‘JH,&
-
2; 2.0 & 2.0
-
“IMe&
-
-
-
4.77,
H,
a Approximated value obtained at -WY in freon-21 at 100MHz.
-1.9
*J&H*
l$Dioxene (1)
_
J-values (in Hz)
4.85,
(4)
6.276
642,
H,
~~~~~~ ~~~
4%l,
2,4,5-Trime~yl-IJ~xene
_
(in ppm)
43.Dimethyl-13dioxene (3)
5-Methyl-1Jdioxene (2)
1,3-Dioxene
H*
&vaks
0;5 1
-
‘JMe,.Me,
4*05,3*%
392,
402
4.1%
Ha
-1.2 -0.95 -1.4
‘JMt,&
1.30,
-
c
Me,
-I.s -
‘Me,&
1.70,
1.72,
-
Me,
-
1 -15.0
I -
3Hs 6.3s
‘JH,,H,
1.49.
1.50,
1.51,
-
Me,
Table 1. ‘H-NMR oarameters of 13dioxenes obtaiwd at 300 MHz (5 ~01% in CCL, TMS internal)
5.1
1
-
‘JMez&
- -6.W
71
1839
NMRexperimentson acetals-pti 56
(I31 Calculated
1Saetext):
Calculated
YI*+~e)-O.9L
l-text
Hz 5Jik(e-~.~~. 3.15 Hz
SJIMe-~.61=2.2~z *56 v -15.5 Hz
‘J6
r2J2s., -6.2Hzl
d-13.0
12J2u-3.5
HZ Hz1
1
Fig. I.
reasonable. From inspection of ‘J, ‘J and ‘J values in 3,6 dihydro-2H-pyrane, the main conformation here also seems to be” the l&liplanar form. The main reason may reside in the fact that overlap of the p-0 r-system is maximal in this conformation.* However, other considerations agree with the expectation that in 1,3dioxene the sofa form becomes more stabilized in comparison with the half-chair form, in contrast with the situation in cyclohexene. In the latter, a sofa form suffers from some strain by an approach down to 2.55 A of the 3J H nuclei. In the t&sofa O-atoms are put into these positions (a similar situation can be traced in the ZH-pyrane nucleus). l&Dioxene is believed to exist in a half-chair form,” but here two ring O-atoms are equivalent to the n-orbital, and this may result in an “in-between” maximal stabilisation where both O-atoms physically indeed become equivalent in a static fashion, and not in a dynamic fashion (e.g. by equilibration between two equivalent sofa forms). Basic conformation and topomerization phenomena. As pointed out, the basic conformation in cyclohexene is the half-chair, while the 1,2diplanar form lies on the route of
*It may be noted that the &sofa possesses parallelp-orbitals for bothoxygen atomswhile the O,-sofa dots not. Thisparalellity, occasionally called “rabbitcar effecP was for some time believed to cause strain.It is not at all clearwhetherthis is indeed the case. IJ-Dioxane is, however,less stablethan I &dioxane?’ Yn 1.4dioxene (a half-chair conformation),” one might advancea stabilizationpe.roxygen of fcI.35.3) = 1 kcal/molefor the ground state. One would then expect a AG’ value of 5.3 + I.0 = 6.3 kcal/mole for 3,4dihydro-ZH-pyranein the halfchair form. 11is 6.6 kud/mole as the basic conformationis the sofa. Takingthe values for chexene as an approximatemeasure of the intrinsicdestabilisationbetween sofa and half-chair(with MG’ = 1.2- 1.8 kcal/mo1e6.‘),the overall stabiition by the pO-n overlap should thereforeamountto some 1.0t 0.3 t 1.2 B I.8 = 2.5 a 3.2 kcallmole. This is a very reasonablevalue and in fact is almost exactly the dilTertncein iotationalbarrierbetween that observed in propenyl methyl ether (3.83kcal/moleY and cis-butcne (0.65kcal/mole).”
ring reversal.” The fact that in 3,44hydro-ZH-pyrane and 1.3dioxene the basic conformation has now become the 1.2diplanar form, has an influence on the barrier to topomerization. For c.hexene AG’ = 5.3 kcal/mole,’ in 3,4dihydro-2H-pyrane it is increased to 6.6 kcal/mole,n in Mdioxene it is still higher, i.e. 7.3 kcallmole.P We have now found for l$dioxene a coalescence temperature of -123”* 1, corresponding to AG # - 7.2 k&mole. The increase in topomerization barrier may be rationalized, if one accepts that in the critical conformation during inversion the proposed p.O-R stabilisation would be lost. As the relative orientation between the p and ~-lobes must change during the process for reversal then the increase in AG # might be accounted for an extra stabilization of the ground state. It is nevertheless possible that this stabilization is apparent and that in the real situation a destabilization phenomenon occurs in the transition state.t EXPEUWWAL. Samples for ‘H-NMR spectm were purifiedpreviouslyby GC on QF, andCarbowax.‘H-NMRspectrawere obtainedat 18”with a VarianHR-300spectrometer,the doubleresonanceand INDOR with a Varian HA-100,equippedwith Muirhead-WiiD %B decade oscillators. Low temperature studies were done in freon-21with a VarianXL-100(12 in.). The tempswere measured with a calibrated Iron-Constantanthermocouple. All spectral analyses were checked afterwards by simulationson a Varian 620-i computerwith the aid of the SIMEQ 16/Hprogram.” I$-Dioxene.(a) Glycerolwas acetalatedwith formaldehydein acidic mediumaccordingto the methodof Hibbertand Tritser.)3 The mixture of S-HO-13dioxane and QHOCH,-1,3dioxolane was equilibratedwith a trace of p-toluenesulfonicacid at 70”for 24 hr. The crudemixturewas tosylatedin pyridmeconventionally and the tosylated mixture dissolved in a minimum of boiling benzene. After cooling the crystallii 5-TsO-13dioxane was collected (yield 40%). (b) Wnation of TsOH was performedin DMSOwith the aid of KOt-Buaccordingto the procedureof Snyderand Soto.y The volatile fractionwas collected and redistilled,b.p. 78’, yield 74%. (Found C 56.0, H 6.8. UC.: C 55.85,H 698%). 4$-ðyl-l$-dioxcne (3) and 2,4~-trimethyl-l$-dioxene (4). These compoundswere obtained as by-productsduringthe reductionstudy of 4 - Me - 5 - methylidene- 13 - dioxaneand 2,4 -
1840
R. CAMERLYNCK and M. ANIXUNIS
diMe - 5 - methylideae - I,3 - dioxane.)’ The isolation of the reaction mixture” was performed by GC on Carbowax, yield” of the crude mixture: 16% (3) and 19% (4). 5 -Me -I,3 -Diox -I -enc. 5 - Me - 5 -I- 1.3- dioxane” (1.5 g) was retluxed for 30 min in lOml EtOH, after 0.3 g Na had- been dissolved. Fractionation of the crude mixture gave a fraction (63O-81”)that was subsequently purified by GC (Gm Carbowax, 80°C).Two fractions (61: 39; respectively 5 - Me - I,3 - diox - 4 - ene and 5 - methylidene - I,3 - dioxane) were obtained. (Found: (main fraction): C 60.2,H 7.8,; Calc. C 60.1,, H 7.5%). (minor fraction) C 60.3,%, H 7.71%. Acknowledgements-We thank the Ministry for Scientific Programming for funds for the continuation of our research using a superconducting 300MHz spectrometer. The IWONL is thanked for a grant to one of us (R.C.). REFERENCES
‘M. W. Lister, 1. Am. Chem. Sot. 6.3, 143 (1941). *K. S. Pitzer, Science 101, 672 (1945). ‘C. W. Beckett, N. K. Freeman and K. S. Pitzer, 1. Am. Chem. Sot. 70, 4227 (1948). ‘J. Boeseken and J. Stuurman, Proc. Acad. Sci. 39, 2 (1936). ‘E. J. Corev and R. A. Sneen. /. Am. Chem. Sot. 77.250511955). 6R. Bucoui and D. Hainaui, BUN.Sot. Chim. Fr: 1366‘(1%$. ‘G. Favini, G. Buemi and M. Raimondi, J. Mol. Smrct. 2, 137 (1968). “F. A. L. Anet and M. Z. Haq, J. Am. Chem. Sot. 87,3147(1965). 9R. A. Pastemak, Acta Cry&.4,316 (1951);M. A. Loshem, Ibid. 5, 593 (1972). ‘OT.Ogata and K. Kozina, Bull. Chem. Sot. Japan 42,1263(1%9). “K. Sakashita, Nippon KO&I Zusshi 77, 1094(1956). ‘7. F. Chiang and $1H. Bat&, 1. Am. Chem. Sot. 91,1898 (1969); H. Geise and H. Buvs. Rec. Trao. Chim. 89. 114711970). “For a review see R. Bucourt, Topics in Sterewhemky, ‘Edited by E. Eliel and N. L. Allinger. Wiley-Interscience, New York (1974). “R. Bucourt, Bull. Sot. Chim. Fr. 2080 (1964). “F. W. Nader and E. L. Eliel, 1. Am. Chem. Sot. 92,305O(1970); cf. E. L. Eliel and Sr. M. Knoeber, 1. Am. Chem. Sot. 88.5347
(1966); K. Pihlaja and S. Luoma, ACU Chem. Scand. 22,240l (1968). 16C.B. Rickbom and S. Livo, 1. Org. Chem. 30, 2212 (1%5). “M Anteunis, D. Tavemier and F. Borremans, Bull. Sot. Chim. Belg0 75, 3% (1966). “M. BarIieldand D. M. Grant, 1. Am. Chem. Sot. ES,1899(1%3). ‘M. Barfield and S. Stemhell, Ibid., 94, 1905(1972). ‘OM.Barlield and D. hf. Grant, Adu. hfagn. Res. 1, 149 (1965). “M. Anteunis. G. Swaelens and J. Gelan, Terrnhedron 27, 1917 (1971). “M. Anteunis, G. Swaelens, F. Anteunis-De Ketelaere and P. Dirinck, Bull. Sot. Chim. Beiges 80, 409 (1971). “J. C. Martin. Bull. Sot. Chim. Fr. 277 11970). =E. L. Eliel, iemisk Tidskrifi 81.22(1964);R.‘O. Hutchins, L. D. Kopp and E. L. Eliel, 1. Am. Chem. Sot. 90,7174 (1968);E. L. Eliel, Accounts Chem. Res. 3, 1 (1970). =A. Snelson and H. A. Skinner, Trans. Faraday Sot.. 57, 2125 (I%]); cf. K. Pihlajaand J. Heikkila, Acta Chem. Stand. 21,239O (1967). ‘%. Piundt and S. Farid, Tetrahedron 22.2237 (1966):R. C. Lord, T. C. Rounds and T. Ueda, 1. Chem. Phys. 57, 2572 (1972);K. Pihlaja, Acfa Chem. Stand. 25, 451 (1971). =C. H. Bushweller and J. W. O’Neil. Tekzhedron Uters 4713 (1969). =R. H. Larkin and R. C. Lord, 1. Am. Chem. Sot. 95,5129(1973). 9. Gagnaire, E. Payo-Subiza and A. Rousseau, Bull. Sot. Chim. Fr. 2633 (1%3). )“J. P. Lorve, Progress in Physical Organic Chemisrry(Editedby A. Streitwieser and R. W. Taft) Vol. 6: Interscience. New York
VW.
“J. F. Kilpatrick and K. S. Pitzer, J. Res. Natl. Bur. Std 37, 163 (1946). “We thank Dr. de Bie (Un. Utrecht, Holland) for the SIMEQ 16/H-software. “H. Hibbert and S. Trister, Can. 1. Res. 14B, 415 (1936). “C. H. Snyder and A. R. Soto, /. 0~. Chem. 29, 742 (1%4). “M. Anteunis and R. Camerlynck, Tefrahedron, subsequent paper. %F. Borremans, Ph.D. Thesis 1975 Gent; M. Anteunis and F. BoFemans, in preparation. “P. A@, Suomen Kemistihehti B46, 151 (1973).