Rearrangement Studies with 14C. XXXVI. Solvolysis of l-14c-2-phenylethyl Tosylate in Aqueous Acetic Acid with or without Added Sodium Azide

Size: px

Start display at page:

Download "Rearrangement Studies with 14C. XXXVI. Solvolysis of l-14c-2-phenylethyl Tosylate in Aqueous Acetic Acid with or without Added Sodium Azide"

Noah Anderson
6 years ago
Views:

1 Rearrangement Studies with 14C. XXXVI. Solvolysis of l-14c-2-phenylethyl Tosylate in Aqueous Acetic Acid with or without Added Sodium Azide C. C. LEE, D. UNGER, AND SHARON VASSIE Deparrn~enr of Chernislry and Chemical Engineering, Uniuersily of Saskalchewan, Saska~oon. Received September 3, 1971 Srrskalchewrm Solvolyses of l-14c-2-phenylethyl tosylate (I-OTS-I-~~C) were carried out in HOAc and H,O solvent systems with or without added NaN,. In the absence of added azide ion, the products obtained from a given solvent mixture of HOAc-H,O, 1-OAc-'"C, and 1-OH-14C, showed essentially the same extent of rearrangement of the 14C-label from C-1 to -2. In the presence of added azide ion, the rearrangement in the 1-N,-14C was less than that found in the I-OAC-'~C or 1-OH-14C. The results are interpreted in terms of concurrent direct displacement to give product and product formation from the ethylenephenonium ion. Les solvolyses du I4C-1 phknyl-2 ethyl tosylate (OTS-I-~~C-I ) ont &ti. menees dans des systemes AcOH-H,O, avec ou sans addition de NaN,. En I'absence de l'ion azoture, les produits obtenus a partir d'un melange donne AcOH-H,O, c'est-a-dire -0Ac-1 I4C et OH-1 14C, montrent essentiellement le m&me pourcentage de rearrangement du 14C marque du C-1 au -2. En presence de N;, le rearrangement de N,-1 14C est inferieur celui trouve pour OAc-1 I4C ou OH-1 14C. Les resultats sont interpret& par une competition entre un deplacement direct conduisant au produit et la formation de ce dernier par I'intermediaire de I'ion Cthylene phenonium. Canadian Journal of Chemistry, (1972) The acetolysis of 1-14C-2-phenylethyl tosylate (1-OTs-1-14C) has been extensively investigated (1 and references given therein) and the results have generally been attributed to two competing processes, namely, a direct displacement (solvent assisted or k, pathway) and reaction via the ethylenephenonium ion (2) (anchimerically assisted or k, pathway) (1-3). On the other hand, the work of Sneen and Larsen (4, 5) suggested the possibility that such solvolyses could proceed solely through an ion-pair mechanism. Very recently however, Schleyer and coworkers (6, 7) have utilized the azide ion as a probe for the k, process and have interpreted their results as in support of the concurrent mechanism. In the present work, solvolyses of 1-OTs-1-14C in H20-HOAc with or without added NaN, were carried out, and the isotope position rearrangements in the different products, 1-OAc-14C, 1-OH-14C, and 1-N,-14C, were ascertained in an attempt to obtain data that might be pertinent in differentiating between the concurrent k, and k, processes and the single ion-pair mechanism. Moreover, since in all previous solvolytic studies with 1-OTs- 1 -I4C, only a single product from a given reaction has been analyzed for isotopic scrambling, it is of interest to determine the extents of scrambling in the different products obtained from a solvolysis under conditions that could give rise to more than one product. The solvolyses of 1-OTs-1-14C in water and acetic acid solvent systems were carried out by heating under reflux 0.05 mol of the substrate per 100 ml of solvent for 24 h. The products were analyzed and separated by v.p.c. and then oxidized to benzoic acid whose 14C-activity gave the extent of isotope position rearrangement from C-1 to -2. The results are summarized in Table 1. In the case of the solvolysis in pure water, because of the low solubility of the 1-OTs-1-14C, the reaction mixture was not homogeneous. The unreacted tosylate was recovered and degraded by conversion to 1-Br-14C followed by oxidation to PhCOOH. It was found that in both runs in water, the recovered tosylate showed only 0.4% rearrangement of the 14Clabel from C-1 to -2. Since it is known that ionpair return is suppressed in a highly ionizing solvent (8), the small amount of rearrangement observed in the recovered tosylate would suggest that ion-pair return in this solvent was not extensive, although the possibility that solvolysis of 1-OTs-1-14C in H20 may not involve the symmetrical ethylenephenonium tosylate ionpair could not be excluded. The data in Table 1 indicate that when aqueous acetic acid was used as solvent, the resulting 1-OH-14C and 1-OAc-14C showed essentially the same amount of isotope position rearrangement. Moreover, the yields of 1-OH and 1-OAc were almost directly related to the

2 1372 CANADIAN JOURNAL OF CHEMISTRY. VOL TABLE 1. Data from solvolyses of I-14C-2-phenylethyl tosylate (I-OTS-I-~~C) in water and acetic acid solvent systems Product yield (%)t Specific activity (c.p.m./mmol) Rearrangement to C-2 (%) Solvent (H,O:HOAc)* 1-OH 1-OAC 1-OH$ PhCOOH 1-OAcg PhCOOH 1-OH 1-OAc *By volume.?obtained by v.p.c. analysis. $Assayed as the a-naphthylurethane, SHydrolyzed to 1-OH and then assayed as the a-naphthylurethane. TThe 1-OTs-I-"C was not completely dissolved in these runs. relative amounts of H,O and HOAc in the solvent mixture. These findings thus failed to provide evidence for a difference in k, contributions by the two components of the solvent mixture. A possible explanation of the identical amounts of rearrangement in the two products might be that the products arose from a single intermediate, such as an ion-pair (4, 5). Alternatively, the nucleophilic characters of H,O and HOAc might not be sufficiently different tocause any observable difference in the k, contributions, or the ion-pair might be insensitive to the nucleophilic differences between H,O and HOAc. Since the azide ion is a strong nucleophile and since it has been used as a mechanistic probe for the k, process (6, 7), the work was extended to include the solvolyses of I-OTS-~-'~C in HOAc or 50: 50 (by volume) H,O-HOAc in the presence of added NaN,. The results, recorded in Table 2, showed that the rearrangement of the 14C-label to C-2 was indeed lower in the 1-N3-14C than in the 1-OAc-14C or 1-OH-14C, in accord with the anticipation of a greater k, contribution in the formation of 1-N3-14C. However, the variations in the extents of rearrangement, shown in both Tables 1 and 2, require further interpretations. Utilizing the concurrent mechanism (1, 6, 7), when two or more products are formed in a given solvolysis of ~-OTS-I-'~C, the isotope position rearrangement in each product is the composite result of product formation from the phenonium ion 2 and from the direct displacement involving rearranged starting material that resulted from ion-pair returns. As an illustration, Scheme 1 shows the possible processes involved in the acetolysis of ~-OTS-I-'~C in the presence of added azide ion. Analogous to Scheme 1, if the two nucleophiles were HOAc and H20, and if the pseudo-first order rates for the k, processes in the reaction with HOAc and with H,O, k, and k,, were not significantly different, for a solvolysis in a given H20-HOAc mixture, there would be no significant difference in the extents of rearrangement of the label in the two products, 1-OH-I4C and I-OAc-14C, as was observed (Table 1). It may also be noted from Table 1 that the amount of rearrangement in the products increases with increasing proportions of HOAc in the solvent. This could be attributable to a greater tendency for ion-pair return when the solvent contains more HOAc. It has been pointed out earlier that for solvolysis of 1-OTs-1-14C in H,O the recovered starting material was only 0.4x rearranged. On the other hand, Coke and coworkers (1 ) have found extensive ion-pair returns in the solvolysis of 1-OTs-1-14C in unbuffered HOAc; for example, after a reaction time of 22 h at 11 5", the recovered tosylate was 26.1 :d rearranged. The greater extent of ion-pair return would result in a greater overall rearrangement since the k, process would give rise to more rearranged product besides the rearrangement arising from the Fk, process. For the solvolyses in the presence of added azide ion, the k, process involving azide ion, or kn[n;], apparently could compete with fnk,

3 LEE ET AL.: REARRANGEMENTS WITH "C. XXXVI

4 CANADIAN JOURNAL OF CHEMISTRY. VOL more effectively than did k, in competing with J'k, (or k, in competing with &k,), thus resulting in lesser amounts of isotopic scrambling in the 1-N,-14C than in the 1-OAc-14C or 1-OH-14C (Table 2). It is seen from Table 2 that for the reaction in HOAc with added azide ion, only about 4'%, of 1-N3-14C was formed as product, but this product showed nearly 6% rearrangement of the label from C-1 to -2. On the other hand, solvolysis in 50: 50 H20-HOAc gave more than 20% 1-N3-14C as product, which, however, showed only about 2% rearrangement of the isotopic label. The greater yield of 1-N,-14C in H,O-HOAc could be the result of a greater availability of the azide ion for displacement reactions, possibly because NaN, might be more readily ionized in H20- HOAc than in HOAc. The greater amount of rearrangement in the 1-N,-14C obtained from the solvolysis in HOAc is likely due to more ion-pair returns in HOAc than in H,O-HOAc; thus the formation of 1-N,-14C would involve reactions with more extensively rearranged 1-OTs-14C in HOAc than in H,O-HOAc. Finally, a comparison of the data in Tables 1 and 2 shows that the presence of added azide ion has the effect of decreasing the extent of rearrangement in the products, 1-OAc-14C and 1-OH-14C. Such differences in the amounts of rearrangement are not unexpected since the overall observed rearrangement should be highly dependent on reaction conditions (1 ). A factor that might have played an important part in decreasing the rearrangement could be that in the presence of added NaN,, the reaction rate would likely be enhanced (6, 7), and there would be lesser opportunity for the occurrence of many cycles of ion-pair formation and ion-pair return, thus resulting in a lower amount of rearrangement in the reaction product. Thus all the results obtained in the present work could be explained by, and hence lend further support to, the occurrence of competing k, and kt, processes, involving direct displacement to glve product as well

5 LEE ET AL.: REARRANGEMENTS WITH "C. XXXVl 1375 as product formation from the pl~enonium ion, as illustrated in Scheme 1. Experimental Solvolyses The solvolyses were carried out by heating under reflux for 24 h 0.5 M solutions of ~-OTS-I-'~C in the appropriate solvent (HOAc or H,O-HOAc). In the experiments with H20 as solvent, 0.05 mol of 1-OTs- I -14C in I00 ml of H,O was heated, but a homogeneous solution was not obtained and much of the tosylate remained as a molten oil during the period of reflux. When the solvolysis was effected in the presence of added NaN,, the concentration of NaN, was 0.7 M. In some preliminary work, 0.5 M NaN, was utilized, but the higher concentration of 0.7 M was chosen in order to raise somewhat the yields of 1-N,-14C. The following is a typical experiment carried out with added NaN,. A solution of g ( mol) of l-ot~-l-'~c (9) and g ( mol) of NaN, in 25 ml of 50:50 (by volume) H20-HOAc was heated under reflux for 24 h. Appropriate amounts of ordinary 1-N, (lo), 1-OAc, and 1-OH were added as carriers. The mixture was then subjected to vacuum distillation (about 20 mm) to remove most of the solvent. Ice-water was then added, followed by neutralization with 5% NaOH solution. The resulting material was continuously extracted for 24 h with ether. The extract was dried over MgSO,, the ether removed, and the products were recovered by preparative v.p.c., using a Pye Automatic Preparative Chromatograph with a 10-ft, 318-in. copper column packed with 20% silicone DC 710 on mesh Chromosorb W. Knowing the weights of the carriers and the appropriate specific activities, the yields of the three products were obtained by isotope dilution calculations. In the solvolyses without added NaN,, instead of isotopic dilution, the yields were determined by v.p.c. analyses using I-phenylethanol as an internal standard, a preliminary injection of the product mixture from each solvolysis having shown that I-phenylethanol was not present as a product. An aerograph 200 Chromatograph with flame ionization detector fitted with a 254, 118-in. copper column packed with 20% silicone DC 710 on mesh Chromosorb W was utilized for the analyses. Degradations Samples of the v.p.c. purified 1-OH-14C, I-OAC-~~C, and I-N,-14C, diluted further with carriers if necessary, were converted to appropriate solid derivatives for radioactivity assays. 1-OH-I4C was treated with a-naphthyl isocyanate to give the a-naphthylurethane, m.p " (lit. (I I ), m.p. 119"). The ~-OAC-'~C was hydrolyzed by heating with 10% NaOH to give 1-OH-14C which was, in turn, converted to the a-naphthylurethane. The 1-N,-14C was reduced with LiAIH, to give 1-NH,-14C (12) which was converted, by treatment with phenyl isocyanate, to N-phenethy1-N'- phenylurea, m.p " (lit. (13) m.p "). The remaining portions of the l-oh-l4c, I-OAc-14C, and 1-N,-14C, after the derivative preparations, were oxidized directly with alkaline KMnO, (9) to give benzoic acid, whose radioactivity measured the I4C content at the C-2 position of these products. The 1-OTs-14C recovered from the solvolysis in H,O was converted to 1-Br-I4C by treatment with LiBr in dry acetone (14). The 1-Br-14C so obtained was also oxidized to benzoic acid (9) in order to ascertain the extent of rearrangement. The authors wish to express sincere appreciation to Professor P. v. R. Schleyer for valuable comments and for suggesting the use of NaN, in this study. The financial assistancegiven by the National Research Council ofcanada is gratefully acknowledged. 1. J. L. COKE, F. E. MCFARLANE, M. C. MOURNING, and M. G. JONES. J. Am. Chem. Soc. 91, l I54 (1969). 2. P. v. R. SCHLEYER and C. J. LANCELOT. J. Am. Chem. SOC. 91, 4297 (1969). 3. J. M. HARRIS, F. L. SCHADT, P. V. R. SCHLEYER, and C. J. LANCELOT. J. Am. Chem. Soc. 91, 7508 (1969). 4. R. A. SNEEN and J. W. LARSEN. J. Am. Chem. Soc. 91, 362 (1 969). 5. R. A. SNEEN and J. W. LARSEN. J. Am. Chem. Soc. 91, 6031 (1969). 6. D. J. RABER, J. M. HARRIS, R. E. HALL, and P. v. R. SCHLEYER. J. Am. Chem. Soc. 93, 4821 (1971 ). 7. D. J. RABER, J. M. HARRIS, and P. v. R. SCHLEYER. J. Am. Chem. Soc. 93, 4829 (1971 ). 8. W. G. YOUNG, S. WINSTEIN, and H. L. GOERING. J. Am. Chem. Soc. 73, 1958 (1951 ). 9. C. C. LEE, G. P. SLATER, and J. W. T. SPINKS. Can. J. Chem. 35, 1417 (1957). 10. P. A. S. SMITH and B. B. BROWN. J. Am. Chem. Soc. 73, 2435 (1951). 11. M. FRANKEL and S. PATAI. Tables for identification of organic compounds. 2nd ed. The Chemical Rubber Co., Cleveland, p J. H. BOYER. J. Am. Chem. Soc. 73,5865 (1951). 13. W. H. CAROTHERS and G. A. JONES. J. Am. Chem. SOC. 47, 3051 (1925). 14. D. P. THORNBILL and C. C. LEE. Can. J. Chem. 41, 2482 (1963).

Solvolytic displacement reactions

Solvolytic displacement reactions Outline 1. Brief history and current picture of displacement reactions 2. Neighboring group participation (halogens) 3. Salt effects and ion pairs 4. Role of the solvent Key players Key references Ion