Direct Intramolecular Proton Transfer Reactions

In summary, the spectroscopy of jet-cooled BBXHQ, and of its thiazole analogue BBTHQ, likely reveals two kinds of potentials for the excited state intramolecular proton transfer reaction. The first case is characterized by a small reaction enthalpy and by a small but significant energy barrier for the reaction. In the second case, the reaction enthalpy exceeds 1.5 kcal/mole there is no energy barrier and the excited proton—transferred molecule may be directly reached by optical excitation, albeit with small Franck-Condon factors for non-vertical transitions. 

Porco et al. reported the synthesis of ( ) methyl rocaglate using [3 + 2] dipolar photocycloaddition reaction of an oxidopyrylium betaine derived from excited state intramolecular proton transfer reaction of 3-hydroxyflavin and methyl cinnamate [144]. Methyl rocaglate was obtained by a base-mediated a-ketol rearrangement followed by hydroxy-directed reduction sequence. They subsequently succeeded in the asymmetric synthesis of methyl rocaglate using functionalized TADDOL derivative (34) (Figure 2.30) as a chiral Bronsted acid (Scheme 2.77) [145]. 

The third type of experiment is photolysis, where the product is one of a tautomer pair [2, 7, 75]. Again, almost aU reactions studied are keto-enol tautomerizations where the proton transfer is not direct but in a number of steps via the solvent Since the first step is often an ionization (proton transfer to solvent molecule), which is thought to be diffusion-controlled [67], it does give some insight into proton transfer reactions, but exact elucidation is hard, since often there are numerous possibiHties for reaction mechanisms and roles of solvent molecules and internal vibrations [76, 77]. In view of the lack of understanding of proton transfer reactions, it would be much better to have a simpler and more direct way to initiate intramolecular proton transfer. This possibility is offered by looking at intramolecular proton transfer reactions in the excited state, which can be initiated much faster and followed on a much shorter timescale than ground-state reactions. 

All these reactions are thermodynamically favourable in the direction of proton transfer to hydroxide ion but the rate coefficients are somewhat below the diffusion-limited values. In broad terms, the typical effect of an intramolecular hydrogen bond on the rate coefficient for proton removal is to reduce the rate coefficient by a factor of up to ca 105 below the diffusion limit. Correspondingly the value of the dissociation constant of the acid is usually decreased by a somewhat smaller factor from that of a non-hydrogen-bonded acid. There are exceptions, however. 

DFT calculations on the Mg(L—H)(L) complex reveal how water and acetonitrile can be lost (Scheme 9). Thus intramolecular proton transfer tautomerizes the neutral acetamide ligand in 48 into the hydroxyrmine form in 49, which can then dissociate via another intramolecular proton transfer to yield the four-coordinate adduct 50, which now contains both water and acetonitrile ligands. It is this complex that is the direct precursor to water and acetonitrile loss. Note that the reaction shown in Scheme 9 is a retro-Ritter reaction and involves fragmentation of the neutral rather than the anionic acetamide ligand, which is a bidentate spectator ligand. 

Yates and coworkers have examined the mechanism for photohydration of o-OH-8. The addition of strong acid causes an increase in the rate of quenching of the photochemically excited state of o-OH-8, and in the rate of hydration of o-OH-8 to form l-(o-hydroxyphenyl)ethanol. This provides evidence that quenching by acid is due to protonation of the singlet excited state o-OH-8 to form the quinone methide 9, which then undergoes rapid addition of water.22 Fig. 1 shows that the quantum yields for the photochemical hydration of p-hydroxystyrene (closed circles) and o-hydroxystyrene (open circles) are similar for reactions in acidic solution, but the quantum yield for hydration of o-hydroxystyrene levels off to a pH-independent value at around pH 3, where the yield for hydration of p-hydroxystyrene continues to decrease.25 The quantum yield for the photochemical reaction of o-hydroxystyrene remains pH-independent until pH pAa of 10 for the phenol oxygen, and the photochemical efficiency of the reaction then decreases, as the concentration of the phenol decreases at pH > pAa = 10.25 These data provide strong evidence that the o-hydroxyl substituent of substrate participates directly in the protonation of the alkene double bond of o-OH-8 (kiso, Scheme 7), in a process that has been named excited state intramolecular proton transfer (ESIPT).26... 

In contrast, a Markovnikov addition of water was reported in the irradiation of a variety of o-hydroxystyrenes, again in aqueous acetonitrile, with the formation of 2-(2-hydroxyphenyl)ethanols. In this case, an intramolecular proton transfer from the excited state of the styrene was envisaged as the first step of the reaction [51]. A similar mechanism was postulated in the photohydratation of m-hydroxy-1,1 -diaryl alkenes that gave the corresponding 1,1-diarylethanols, although direct protonation of the P-carbon by water competed in some cases [52]. 

For these systems, direct hydrogen abstraction by benzophenone triplets was observed in benzene whereas in a polar solvent electron transfer and hydrogen-atom abstraction were observed. Electron transfer followed by an intramolecular proton transfer was observed in these systems although such proton-transfer reactions are not observed in unlinked systems of primary and secondary amines. The observed differences between the linked and unlinked systems have been attributed to the dependence of electron transfer, proton transfer, and hydrogen transfer on mutual distance and orientation. In the unlinked systems, rotational and translational motion of two reacting molecules are usually much faster than those in linked systems. 

The conclusion we can draw from all this research is that there is still no coherent picture of intramolecular ground and excited-state proton transfer reactions in tautomers. The topic is complicated from an experimental as well as a theoretical point of view, and many questions remain. Intramolecular ground-state proton transfer is hard to study directly, and although femtosecond pulsed lasers allow initiating and following proton transfers in the excited state on a very short time scale, these methods bring their own complications to the interpretation of the results. ... 

Comparing these observations to the requirements of TST, we can immediately see a number of problems. Even apart from the fact that the mass of the proton requires it to be treated as a quantum mechanical particle, so that even if there were a well-defined barrier, we would still need to take the possibility of tunneling into account. Transfer of the proton is directly coupled, or may even be driven by a redistribution of electron density in the molecule. In excited-state intramolecular proton transfer (ESIPT) reactions, the redistributed charge almost certainly provides the driving force. The generic picture for such is reaction is due to WeUer [7, 8], who was the first to realize that the enormous Stokes shift of about 10 000 cm he observed in the fluorescence of salicylic acid (X = OH) could be a consequence of a rapid proton transfer in the excited state. 

The situation presented in fig. 29 corresponds to the sudden limit, as we have already explained in the previous subsection. Having reached a bend point at the expense of the low-frequency vibration, the particle then cuts straight across the angle between the reactant and product valley, tunneling along the Q-direction. The sudden approximation holds when the vibration frequency (2 is less than the characteristic instanton frequency, which is of the order of In particular, the reactions of proton transfer (see fig. 2), characterised by high intramolecular vibration frequency, are being usually studied in this approximation [Ovchinnikova 1979 Babamov and Marcus 1981]. 

In this region, the equilibrium constant for the proton-transfer step in Scheme 7 has a value K2> 1 and the proton transfer step is strongly favourable thermodynamically in the forward direction. This reaction step is a normal proton transfer between an oxygen acid which does not possess an intramolecular hydrogen bond and a base (B) and will therefore be diffusion-limited with a rate coefficient k2 in the range 1 x 109 to 1 x 1010dm3mol-1 s 1. It follows from (65) that kB will have a value which is... 

---