CIDNP results

On the other hand, CIDNP results obtained with diazonium salts in water-free systems have only a limited value for the interpretation of mechanisms in aqueous systems, because of the reasons mentioned at the beginning of this section. 

The kinetics of the decomposition of the 4-chlorobenzenediazonium ion under strict exclusion of oxygen (< 5 ppb 02, Schwarz and Zollinger, 1981) are compatible with the CIDNP results, subject to the reservation mentioned already, namely that CIDNP as a probe does not necessarily give results for all pathways, whereas kinetic measurements are normally related to the sum of all competitive mechanisms. The first reaction observable with conventional kinetic methods is the formation of the (E )-diazoate (t1/2 ca. 200 min), but it is also first-order with respect to the diazonium ion concentration. 

Other possibilities (e.g., Overhauser effects) exist for the complication of the patterns of polarization predicted by the simple theory. Interesting transienf. splittings have been observed in polarized F-spectra (Bethell et al, 1972a, b). Nevertheless, straightforward application of the simple rules seems to yield reliable conclusions in most cases. Clearly, however, it is unwise to rely on CIDNP results alone in studies of organic reaction mechanisms other information is invariably necessary to ensure that the correct interpretation is chosen from among the several possibilities which CIDNP may suggest. 

Such treatment of CIDNP results produced serious objections. Lippmaa et al. (1973), investigating the same reaction, revealed a strong N, C, and CIDNP effect. The C nuclei in the phenoxyl part of the azo dye were not polarized. At the same time, polarization of N nuclei of the azo bond and C nuclei at positions 1 and 2 of the free-of-hydroxyl phenyl ring connected with... 

Exposure of several methyl-substituted derivatives to y-radiolysis at 77 K in cryogenic matrices gave rise to a family of radical cations of the same structure type, some of which had been previously identified on the basis of CIDNP results. We begin with a discussion of the CIDNP investigations, since they preceded the ESR studies of all species but the prototype. The first CIDNP results attributed to a cyclopropane radical cation were observed during the photoreaction between 1,4-dicyanonaphthalene and cis-l,2-diphenylcyclopropane. However, the nature of the cyclopropane radical cation was characterized by CIDNP effects observed during the reaction of chloranil with cis- and /rans-l,2-diphenylcyclo-propane. ... 

For radical cations of norcaradiene and derivatives, the interaction of the cyclopropane in-plane e orbitals with the butadiene frontier MO favors the type B structure. The assignments are based on ab-initio calculations, CIDNP results, and the ET photochemistry. The norcaradiene radical cation (lla ) has a electronic ground state (Cj symmetry). The Cl—C6 bond is shortened on ionization (—3.4 pm) while the lateral bonds are lengthened (+2.8 pm). The delocalization of spin density to C7 (py = 0.246 py 5 = 0.359) and the hyperfine coupling constants of the cyclopropane moiety a e = 1.36 mT oysyn = —0.057 mT flvanti = —0.063 mT) support a type B structure. 

The third intra-pair reaction to be discussed involves bond formation between radical anion and cation without intervening transfer both singlet and triplet radical ion pairs can couple. For example, the bifunctional radical cation 24 generates two chloranil adducts, most likely via zwitterions (e.g., 74 and 75 ), initiated by forming a C O bond. The CIDNP results indicate that 74 and 75 are formed from a singlet radical ion pair. Adduct 75 is a minor product, as the major spin density of 24 + is located in the allyl function which, therefore, is expected to be the principal site of coupling. 

Such treatment of the CIDNP results produced serious objections. Lippmaa et al. (1973), investigating the same reaction, revealed a strong 15N, 13C, and H CIDNP effect. The 13C nuclei in the phenoxyl C6-ring of the azo dye were not polarized. At the same time, the polarization of 15N nuclei of the azo bond and 13C nuclei at positions 1 and 2 of the phenyl ring connected with the diazo link was an exact replica of the polarization of the same nuclei in the diazonium salt. This has led to the conclusion that the diazo component polarizes as a result of the side reactions and that it is the diazo component that brings it to the azo dye. Thus, the CIDNP effect does not support the ion radical mechanism presented earlier. Several explanations for the observed CIDNP effect have been proposed. We want to discuss one of them here because it seems to explain a whole range of interactions of diazonium salts with oxyanions, an interaction that is accompanied by a pronounced polarization of nuclei. 

In view of the significant conclusions that are derived on the basis of the experimental methods delineated above, it is appropriate to discuss the reliability of these methods as well as potential shortcomings and sources of misinterpretation. In this context, we will also comment on the significance of direct observation. We begin with potential problems with the interpretation of CIDNP results. 

In summary, these results constitute strong evidence for the two-step reaction sequence. They require that the deprotonation of the aminium radical cation be competitive on the CIDNP timescale i.e. surprisingly fast since it involves a carbon acid. The results delineate the fate of the amine derived intermediates with particular clarity, since they are observed directly for amine derived products. The conclusions based on the above CIDNP results were confirmed by time resolved optical spectroscopy in a variety of systems [179-182]. However, in essentially all these systems the reaction progress is monitored by following the complementary spectra of the acceptor derived radical intermediates, such as ketyl, semiquinone, stilbene, or thioindigo radical anions. 

Radical cations of the same general structure type as those derived from cis- and tratw-diphenylcyclopropane have been established for numerous cyclopropane derivatives, including the parent, 1,2-di-, 1,1,2-tri- and 1,1,2,2-tetramethylcyclo-propane (Table 3). Two of these systems provide a direct comparison between the results of CIDNP and ESR experiments. In both cases, the ESR spectra observed by Williams and coworkers following pulse radiolysis in frozen solutions [293, 296, 297] show splitting patterns supporting the presence of spin density on two carbon centers, thus confirming the structure type (102) assigned on the basis of CIDNP results. 

The structure of these ions rests on the following results (Table 4) detailed CIDNP spectra delineating the hyperfine patterns of both ions [320, 321] ab initio calculations with a 6-31 G basis set for both ions [322] and ESR [323, 324] as well as ENDOR data [324] for the norbornadiene radical cation. The CIDNP results indicate the absolute signs and relative magnitude of the hyperfine coupling constants a comparison with the calculated values (Fig. 22) shows satisfactory... 

The unique bonding in the bicyclobutane system, which is reflected in its unusual chemistry, has also spurred an interest in its radical cation. For example, the photoinduced electron transfer chemistry of bicyclobutane derivatives has been investigated in detail (cf. Sect. 4.4). The product distribution obtained under a variety of reaction conditions suggests a radical cation structure in which the transannular bond is substantially weakened. This structure type has been confirmed by CIDNP results for several derivatives [248, 249] and by ESR results for the parent system [340]. 

One might argue that the results least subject to ambiguity are those with the shortest delay between the generation of the radical cation and its observation. In this respect, the time-resolved ODMR results of Trifunac and Qin (Fig. 24) [368] and time resolved CIDNP results observed in the author s laboratory (Fig. 25) [380], may provide the least distorted view of the species in question. Of course, neither of these experiments qualifies as the coveted direct observation. Thus, the direct observation of the elusive hexamethyl-Dewar benzene radical cation must await further scrutiny. 

The 2Bj state was found to be the ground state and its predicted hfcs are fully compatible with the observed CIDNP results, whereas the hfcs calculated for the 2A, state show irreconcilable differences with the experimental findings for every type of proton. 

In summary, the radical cations of the benzene valence isomers show several interesting structures. Although most of their structural features can be rationalized by considering the HOMOs of the precursor molecules, some of the species show substantial changes in individual bond lengths. Accordingly, species such as the 2B2 and 2A1 Dewar benzene radical cations, the 2Bj and 2At benzvalene radical cations, or the 2Bt prismane radical cation cannot be expected to qualify as Koopmans radical cations. To date, most of the information available in this series is based on CIDNP results and ab initio calculations. It is safe to predict increasing involvement of ESR spectroscopy in this area. 

In contrast, CIDNP results indicate that the radical cations of barbaralane (157, X = C = 0) and semibullvalene (157, X = —) correspond to the elusive structure type with a single minimum [391, 424]. The spin density resides primarily on the termini (C-2,4,6,8) of the twin allyl moieties, whereas the remaining (internal) carbons of the 5 jr-electron perimeter have negative spin density. This spin density distribution reflects the coefficients of orbital 158, the HOMO of a bis-homoaromatic structure (Fig. 32) [424], More recently, ESR results have confirmed this assignment [392, 393],... 

---