Modeling the Microhydration of Protonated Alanine Catherine Michaux † and Johan Wouters Laboratoire de Chimie Biologique Structurale, Faculte ´s UniVersitaires Notre-Dame de la Paix, rue de Bruxelles, 61, B-5000 Namur, Belgium Eric A. Perpe `te ‡ and Denis Jacquemin* Laboratoire de Chimie The ´orique Applique ´e, Faculte ´s UniVersitaires Notre-Dame de la Paix, rue de Bruxelles, 61, B-5000 Namur, Belgium ReceiVed: April 22, 2008; ReVised Manuscript ReceiVed: May 28, 2008 The microsolvation of protonated L-alanine with one, two, or three water molecules has been investigated using a MP2/6-311++G(d,p) approach fully accounting for the basis set superposition errors. A conformational analysis for unhydrated AlaH + reveals only three minima which have been characterized and compared to the neutral case. We have built a logical tree for the successive hydration stages. This tree shows that the most stable complexes in each step are related and that a systematic approach can be used to grasp the stepwise hydration process. The addition of extra water molecules to the first or second solvation shells leads to the opposite evolution of the hydrogen-bond stretching mode. Comparisons with experimental enthalpies, entropies, and Gibbs free energies clearly demonstrate the adequacy of the approach. Our results also strongly suggest that several di- and trihydrated complexes should coexist under the experimental conditions. I. Introduction Stepwise solvated molecular complexes have recently at- tracted the attention of a huge number of theoretical and experimental works mainly treating biomolecules or metal ions. 1–11 To accurately determine the thermodynamical param- eters related to the solute-solvent complexation, most measure- ments use mass spectrometry (MS), 10–12 though the lack of precise structural information might constitute a dramatic drawback. To circumvent this limitation, the MS experiments can be coupled to the determination of the vibrational signatures. 8,13 Nevertheless, the resulting spectra might not be straightforwardly interpretable, as several microsolvated struc- tures often coexist. In practice, theoretical calculations are often mandatory or, to say the least, helpful to extract a maximum of information from the experimental data. Broadly speaking microsolvation studies can be classified in to three groups: (i) The evaluation of aqueous effects on metal ions. 1–3 (ii) The investigation of the relationships between stepwise hydration and the consequent modifications of the electronic properties, typically the electroaffinities and ionization potentials, of DNA bases. 5,14–18 For instance, for the adenine-uracil base pair, the vertical detachment energy increases by 40% when going from the unhydrated to the dihydrated compound. 5 (iii) The deter- mination of the relative energies of the nonionized and zwit- terionic forms of amino acids (AA). 6–8,19–29 For the two simplest AA, Gly and Ala, recent investigations featuring from 1 to 8 and 1 to 10 surrounding water molecules have been reported by Aikens and Gordon, 30 and Chuchev and DelBruno. 31 Both concluded that the addition of more and more solvent molecules leads to relatively more stabilized zwitterions. Much fewer works tackled protonated AA, AAH + , 13,19,32–34 although for such systems accurate experimental data are available, 10,12 which make meaningful and straightforward the theory/experiment comparisons of the thermodynamic quantities of the complex- ation process. In the present work, we investigate the micro- hydration of AlaH + with refined ab initio tools, and our results are extensively compared with Wincel’s experimental values. 10 An obvious limitation of all major theoretical approaches is that the number of possible minima dramatically increases with the cluster size. Consequently, most works rely on initial molecular dynamics and/or Monte Carlo steps to generate a large number of probable starting geometries. 13,30 However, the structures eventually obtained in this way are so numerous that a subsequent treatment with state-of-the-art theoretical tools still remains difficult; one often ends up in resorting to semiempiri- cal, low-level ab initio, or even chemical intuition, to perform the screening. Such procedures may look fashionable but cannot be viewed as totally safe. Indeed, using a smaller basis set (for instance) could lead to the disappearance of some valid minima for AAH + . 33 In a recent investigation dealing with the micro- hydration of the protonated glycine, 34 we found that an evolutionary logic is suitable to predict the most stable GlyH + -(H 2 O) n complex on the basis of the energetics and structures of the previous generation, that is, the ensemble of GlyH + -(H 2 O) n-1 compounds. For comparisons with gas-phase experiments, it is valid to build up the complexes following a stepwise approach, as three- (or more-) body collisions seem very unlikely in the actual low-pressure experimental setup. 10,12 For GlyH + , our computational approach yields an excellent accuracy on the successive hydration enthalpies. However, the transferability of this systematic approach to other AAH + is still to be considered. Therefore, we here selected protonated L-alanine, 35 a more complex situation than GlyH + (AlaH + lacks the GlyH + symmetry), to confirm/infirm the efficiency of our procedure and, more importantly, determine the structures * To whom correspondence should be addressed. E-mail: denis.jacquemin@fundp.ac.be. Research Associate of the Belgian National Fund for Scientific Research. † Post-Doctoral Researcher of the Belgian National Fund for Scientific Research. ‡ Research Associate of the Belgian National Fund for Scientific Research. J. Phys. Chem. B 2008, 112, 9896–9902 9896 10.1021/jp803476k CCC: $40.75  2008 American Chemical Society Published on Web 07/23/2008