YC N,m,r. Investigation on the First and Second Nitrogen Protonation in the Diazanaphthalenes

Recently we have reported a 13C n.m.r. investigation on the site of protonation in the diazanaphthalenes.' This group of N-heterocycles consists of ten compounds: cinnoline (1,2-diazanaphthalene, l), quinazoline (1,3-diazanaphthalene, 2), quinoxaline (1,4-diazanaphthalene, 3), 1 ,5-naphthyridine (1,5diazanaphthalene,4), 1,6-naphthyridine (1,6-diazanaphthalene, 3, 1,7-naphthyridine (1,7-diazanaphthalene, 6) , 1 &naphthyridine (l$-diazanaphthalene, 7) , phthalazine (2,3-diazanaphthalene, S), 2,6-naphthyridine (2,6-diazanaphthalene, 9) and 2,7-naphthyridine (2,7-diazanaphthalene, 10). In order to establish the site of protonation we recorded the pH dependence of the chemical shift of each separate carbon atom in all isomers. From the titration curves thus obtained we simultaneously derived the pK values of the first nitrogen protonation (pK,) by application of the HendersonHasselbach equatiom2 These pK1 values correlated very well with those values from the literature that were derived by conventional titration methods. However, the pK1 values determined with the n.m.r. technique proved to be systematically smaller than the literature values. The same has been noticed for other aza-aromatics by Breitmaier and S p ~ h n . ~ The most obvious explanation of this systematic deviation is that in the 13C n.m.r. experiment the concentration of the titrant is much higher than in a normal titration procedure. As a consequence, the activity coefficients of the participating species will deviate seriously from unity. In 13C n.m.r. determinations of the second nitrogen protonation constants (pK2) this factor gains even more importance because of the still higher ionic strength of the solutions involved. In this contribution we apply a procedure that takes into account these activity coefficients. Since this method proved to be successful for the pKl values a similar approach was used for the determination of the pK, values.