Kinetics of the Deamination of Amides by Nitrous Acid$ Khawla Al-Mallah and Geoffrey Stedman* Chemistry Department, University of Wales Swansea, Singleton Park, Swansea SA2 8PP, UK The kinetic profile of the rate constant for the nitrous acid¡¾amide reaction in sulfuric acid as a function of acidity for a range of aliphatic and aromatic primary amides has been interpreted in terms of the HNO2/NOa and the amide/ amide Ha equilibria. The mechanism of the deamination of primary amides by nitrous acid, eqn.(1), was originally thought1¡¾3 to involve a rate-determining N-nitrosation reaction, as had been estab- lished4 for primary amines. Primary aliphatic and aromatic amides are much weaker bases than the corresponding primary amines, and in order to get the reaction to proceed at a reasonable rate much higher concentrations of mineral acid are necessary. In these concentrated solutions activity e€ects are important. Attempts to account quantitatively for the pro¢çle of rate constant versus acid concentration in terms of a single mechanism1 were unsuccessful.Subsequent work by a number of groups5 has established that with the presence of a strongly electron attracting group such as the carbonyl function the initial nitrosation is reversible, and the rate-determining stage involves proton loss. Furthermore the site of the initial nitrosation may be the oxygen of the carbonyl group; this is certainly the site for protonation. The mechanism is summarised in eqns. (2)¡¾(6).RCONH2 a HNO24RCO2H a N2 a H2O O1U NOa a H2O N HNO2 a Ha KNO O2U RCONH2Ha N RCONH2 a Ha KA O3U RCONH2 a NOa N RCONH2 NOa KE O4U RCONH2 NOa a B4RCONHNO a BHa k3 slow O5U RCONHNO4RCO2H a N2 fast O6U We summarise in this paper the results of kinetic studies on a group of aliphatic and aromatic amides reacting with nitrous acid over the range [H2SO4] a 2.8 to 10.5 mol dm¢§3, at 25 8C. Reactions were run with a large excess of amide over nitrous acid, and gave good ¢çrst order plots of ln[nitrite] versus time, k1/min¢§1.Values of k1 were directly proportional to [amide], yielding second order rate con- stants k2/dm3 mol¢§1 min¢§1 in terms of stoichiometric concen- trations. ¢§danitritea=dt a k2aamideaanitritea O7U All amides studied showed a similar dependence of k2 on [H2SO4]; k2 increased with acidity and passed through a sharp maximum close to 8 mol dm¢§3 sulfuric acid, and then decreased sharply. A typical plot is shown in Fig. 1. The decrease in k2 at higher acidities was always markedly steeper than the increase at lower acid concentrations. In order to check that there was no di€erence in mechanism for reaction at acidities on each side of the rate maximum, Arrhenius activation energies were measured on each side of the maximum for two of the aliphatic amides, giving the following results at the molar concentrations of sulfuric acid speci¢çed in parentheses, E/kJ mol¢§1: Ra (CH3)3C, 74.221.4 (4.5); 76.924.8 (9.3); R a CNCH2, 60.124.4 (6.8), 64.022.4 (8.0).The activation energies are constant to within one standard deviation. The results for trimethyla- cetamide were used to correct the rate pro¢çle data measured at 0 to 25 8C, in order to compare the results with those for the other amides studied. The rate maximum is due to the opposing e€ects of acidity on equilibria (2) and (3), combined with the e€ect of sulfuric acid concentration on KE and k3.Equilibrium (2) is described formally by the HR acidity function for which DHR/D[H2SO4] is on average ca. ¢§1.05 dm3 mol¢§1. The pKNO for (2) determined spectro- photometrically using the HR function is ¢§8.11. At [H2SO4] a 8 mol dm¢§3 HR=¢§ 7.59, and thus at higher acidities conversion of nitrite into NOa is virtually com- plete. Equilibrium (3) is described by the HA acidity func- tion for which DHA/D[H2SO4] is only ca. ¢§0.28 dm3 mol¢§1. Thus at lower acidities the increase in [NOa] dominates over the decrease in [RCONH2], while as we approach the con- ditions where HR becomes close to pKNO the fraction of NOa, hR/(KNOahR), levels o€ and tends to 1, while the fraction of free amide, KA/(KAahA), continues to decrease.However the decrease in k2 is very much faster than the decrease in 1/hA and the rate of proton loss is a sensitive function of sulfuric acid concentration. This is presumably because of variation in the activity of the various proton acceptors such as SO4 2 ¢§, H2O and possibly HSO4¢§.In seeking for an empirical measure of this e€ect we used the H0 acidity function, with the idea that as the proton donating power measured by h0 increased so the ability of the medium to remove a proton would decrease. The mechanism in eqns. (2)¡¾(6) requires the rate law (8). rate a k3KEaNOaaaRCONH2a O8U To relate this to eqn. (2) we write a a the fraction of nitrite present as NOa, and b a the fraction of amide present as RCONH2, where a and b have been de¢çned in J.Chem. Research (S), 1998, 670¡¾671$ Fig. 1 Variation of k2 with sulfuric acid concentration for trimethylacetamide at 0 8C $This is a Short Paper as de¢çned in the Instructions for Authors, Section 5.0 [see J. Chem. Research (S), 1998, Issue 1]; there is there- fore no corresponding material in J. Chem. Research (M). *To receive any correspondence. 670 J. CHEM. RESEARCH (S), 1998terms of KNO, hR, KA and hA in the previous paragraph.To allow for the variation of k3KE with 1/h0 we write k3KE=k4/h0. With these substitutions, and the assumption that there is only a small fraction of the amide present as RCONH2NOa, we obtain eqn. (9). logOk2=abU a log k4 a H0 O9U Plots of this type yielded good straight lines with slopes between 0.92 and 1.27 over the range of acidities speci¢çed in Table 1. Values of pKA for aliphatic amides are due to Liler6 and for aromatic amides come from the review7 by Boyd.The value for (CH3)3CONH2 comes from our own work8 based upon NMR measurements, and is very close to Liler's value. Although our model gives a reasonably satisfactory description of the data the choice of H0 was arbitrary. However most acidity functions vary in an approximately linear manner with [H2SO4] in the higher range of acid concentrations, and so the choice of other acidity functions to describe the variation of KEk3 would still have yielded linear plots, though with di€erent slopes.Since of the use of 1/h0 is a purely empirical measure of the ability of the medium to accept a proton we do not o€er any comment on the variation of the slopes. There must be considerable di€erences in the solvation of the aliphatic amides in view of the di€erences in R; for the aromatic amides which all have a phenyl ring the slopes are quite similar. An alternative explanation of the dependence of k2/ab upon 1/h0 would be an acid¡¾base equilibrium for (5) followed by a rate-determining rearrangement as in eqn.(6). However as Williams5 points out the evidence for a rate- determining loss of a proton is very strong, so we reject this alternative. Although the present model gives a reasonable description of the data in terms of the generally accepted mechanism, it is important to realise its limitations. One concerns the value used for pKNO. Various values have been suggested,9,10 and in his interpretation of the kinetics of diazotisation in dilute acid Ridd used11 a value of ca.¢§6.5, which yields limiting rates in good agreement with the calcu- lated value of the encounter rate between NOa and ArNH2. As we are concerned with data over a wide range of sulfuric acid concentration, using the HR data to describe (2) we prefer to use the more negative ¢çgure of ¢§8.11 deduced by Deno et al.12 when ¢çtting spectrophotometric estimates of [NOa]/[HNO2] to his HR data obtained by the ionisation of arylcarbinol indicators. At lower [H2SO4] changing the value of pKNO merely displaces the line of log (k2/ab) versus H0 without signi¢çcantly changing the slope.As HR approaches pKNO the choice is more important. We tried using pKNO=¢§ 6.5, but found deviations at sulfuric acid concentrations greater than 6.5 mol dm¢§3, whereas with Deno's value linearity was observed up to the speci¢çed con- centrations shown in Table 1. We have also extended a few measurements to higher acidities, [H2SO4] a 13.1 mol dm¢§3, and there are deviations from eqn.(9); values of log (k2/ab) are greater than predicted from the equations in Table 1 for the aliphatic amides. The measurements for the aromatic amides were not extended to such high acidities. It is poss- ible that an extra pathway comes into play at high acidities. Finally we turn to a consideration of the di€ering reactivities of the amides studied. Since there are variations in the slopes of the plots of eqn.(9) it is simplest to make comparisons at a given sulfuric acid concentration by substi- tuting the appropriate value of H0. The same reactivity sequence is obtained over our range of acidities. For the aliphatic amides it is (CH3)3C>ClCH2>CNCH2>Cl2CH, which is the same as the sequence of pKA values. For the aromatic amides the sequence is p-CH3O>p-CH3>p-Br> p-Cl>p-NO2, again the same sequence as the pKA values, except for the interchange of positions of chlorine and bromine which are very close together anyway. Clearly the greater the electron releasing power of R in RCONH2 the higher is the reactivity.The addition of NOa to the free amide may be expected to vary with R in the same sense as the addition of Ha, and this is undoubtedly favoured by electron release by R. Step (5) however involving proton loss will be reduced in rate by increase in electron release by R. We conclude therefore that the reactivity in the reaction of nitrous acid with primary amides is controlled by the basicity of the amide, by favouring the formation of an amide NOa species, and that this is more important than the e€ect on the rate of proton loss.Experimental Materials.�¢The amides used were either commercially available materials (Aldrich, BDH, Koch-Light) or were prepared from the acid chlorides by addition to stirred, ice-cold 0.88 ammonia. In some cases the acid chlorides were prepared by reaction of thionyl chloride with the carboxylic acid.The amides were recrystallised from water to constant melting point. Kinetic Methods.�¢Some reactions were followed by colorimetric analysis for nitrous acid involving diazotisation of sulfanilic acid and coupling with alkaline 2-hydroxynaphthol-3,6-disulfonic acid. Other reactions were followed by direct UV spectrophotometry at wavelengths where nitrite absorbed. For slow runs a blank exper- iment was carried out to correct for the self-decomposition of nitrous acid.Thanks are due to the University of Mosul for study leave (to K. A.-M.) Received, 17th June 1998; Accepted, 10th July 1998 Paper E/8/04600J References 1 M. N. Hughes and G. Stedman, J. Chem. Soc., 1964, 5840. 2 M. L. Bender and H. Ladenheim, J. Am. Chem. Soc., 1960, 82, 1895. 3 J. Jaz and A. Bruylants, Bull. Soc. Chim. Belges, 1961, 70, 99. 4 J. H. Ridd, Q. Rev. Chem. Soc., 1961, 15, 418. 5 D. L. H. Williams, in Nitrosation, Cambridge University Press, Cambridge, 1988, p. 101. 6 M. Liler, J. Chem. Soc. B, 1969, 385. 7 J. H. Boyd, in Solute¡¾Solvent Interactions, ed. J. F. Coetzee and C. D. Ritchie, Marcel Dekker, New York, 1969, vol. 1, ch. 3, p. 97. 8 K. Al-Mallah, Ph.D. Thesis, University of Wales, 1974. 9 Stability Constants, Special Publication No. 17, The Chemical Society, London, 1964, p. 163. 10 Stability Constants, Special Publication No. 25, The Chemical Society, London, 1971, p. 91. 11 J. H. Ridd, Adv. Phys. Org. Chem., 1978, 16, 1. 12 N. C. Deno, H. E. Berkheimer, W. L. Evans and H. J. Peterson, J. Am. Chem. Soc., 1959, 81, 2344. Table 1 Parameters used for plots of eqn. (9) R Slope Intercept [H2SO4]/ mol dm¢§3 pKA (CH3)3C* 1.2120.02 7.2220.08 2.8 to 9.3 ¢§ 1.40 ClCH2 1.0120.03 5.8420.12 2.8 to 9.3 ¢§ 2.74 CNCH2 0.9220.04 5.2120.16 3.4 to 9.1 ¢§ 3.73 Cl2CH 1.0320.06 5.1220.23 2.8 to 9.3 ¢§ 4.18 p-MeC6H4 1.2220.04 7.4020.16 5.5 to 10.4 ¢§ 1.67 p-BrC6H4 1.2920.07 7.4720.27 5.5 to 10.5 ¢§ 2.02 p-ClC6H4 1.3220.04 7.5420.13 3.2 to 10.5 ¢§ 1.97 p-MeOC6H4 1.2820.02 7.9020.07 3.1 to 9.1 ¢§ 1.54 p-NO2C6H4 1.2520.05 6.9920.19 5.5 to 10.2 ¢§ 2.70 *Corrected from 0 to 25 8C. J. CHEM. RESEARCH (S), 1998