Ab Initio Calculations DOI: 10.1002/ange.200701617 Water Dissociation in the Presence of Metal Ions** Orkid Coskuner,* Emily A. A. Jarvis, and Thomas C. Allison Metal-ion solution chemistry is of fundamental importance and has practical implications in a wide variety of chemical, biological, and environmental processes. [1] In particular, events such as water dissociation play a central role in determining solution pH value and chemical reactivity. Though there are many studies of metal-ion solutions, all have focused on the structure of the solvated metal ion. Herein, water dissociation, hydroxylation of metal ions, and proton transfer in aqueous solutions of metal ions are studied by ab initio molecular dynamics simulations coupled with sampling techniques. The quest for understanding the behavior of the proton in water has a long history. In 1806, Grotthuss [2] introduced the notion of “structural diffusion” for which the detailed “Grotthuss mechanism” has only recently been demonstrated to involve the motion of a proton in water along a network of hydrogen bonds. Important aspects of this mechanism include the structures and energetics of the hydrated proton. In the mid-20th century, Eigen proposed a H 9 O 4 + complex [3] in which a hydronium ion is bound to three water molecules, and Zundel proposed a H 5 O 2 + complex [4] in which a proton is bound to two water molecules. These concepts dominate our modern understanding of the proton in water. A mechanism is presented for water dissociation in the presence of Cr 3+ and Fe 3+ derived from Car–Parrinello molecular dynamics (CPMD) simulations with transition path sampling (TPS). These techniques provide unprece- dented insight into dynamical events that are difficult to observe experimentally. The initial trajectories used in TPS were obtained from CPMD simulations. We observe a slightly elongated OÀH bond (over 1.1 ; Figure 1 A) in the first solvation shell. In the following 80 fs (TPS calculations), the proton moves and leaves the transition-metal ion hydroxylated (Figure 1 B). Once the proton moves beyond the first shell, Zundel and Eigen complex formation may be observed. Within 30 fs of the proton moving to the second shell (Figure 1 C), hydrogen- bond fluctuations lead to the formation of a H 5 O 2 + (Zundel) complex. In the following 50 fs, a H 9 O 4 + (Eigen) complex forms through proton transfer between the second and third shells (Figure 1D). No immediate return path to the undis- sociated state of water is observed once the water ions form and are separated by the breaking of the chain-like hydrogen bonds as previously observed. [5] The system rapidly intercon- verts between Eigen and Zundel complexes as has been observed in simulations of a proton in water. [6–8] The Zundel complex forms in the second shell (1.5  closer to the metal ion), and the Eigen complex forms between the second and thirds shells. Geometric parameters of the Zundel (H + À O bond lengths of 1.16  and 1.19  and an O À O separation of 2.2 ) and Eigen complexes (H + À O bond length of 1.59  and an O À O separation of 2.61 ) are similar for solutions of Fe 3+ and Cr 3+ ions. The asymmetric binding of the proton within the Zundel complex has been found in theoretical [9, 10, 24] and experimental [11] studies. Fur- thermore, the OÀO distances are 0.6  (H 5 O 2 + ) and 0.2  (H 9 O 4 + ) shorter than for the water dimer. [12] The H + À O bond lengths of the Zundel and Eigen complexes differ from those of pure water by 1–3 %. [10] In Figure 2, the free-energy changes for proton transfer between Zundel and Eigen complexes in the vicinity of Cr 3+ and Fe 3+ ions are presented. The plots show minima for Figure 1. Illustration of proton transfer around metal ions from CPMD/TPS simulations (gray metal, red O, white H, H-bonds given by yellow dashed lines). A) t = 6 fs, B) t = 80 fs, C) t = 110 fs, D) t = 160 fs. [*] Dr. O. Coskuner, Dr. E. A.A. Jarvis, Dr. T.C. Allison National Institute of Standards and Technology 100 Bureau Drive, Stop 8380, Gaithersburg, MD 20899-8380 (USA) Fax: (+ 1) 301-869-4020 E-mail: orkid.coskuner@nist.gov Dr. O. Coskuner Synchrotron Radiation Laboratories Stanford University Palo Alto, CA 94025 (USA) [**] The authors thank R. Mountain, C. Gonzalez, D. Chandler, K. Rosso, and E. Bylaska for helpful discussions. This research was supported by National Science Foundation grant CHE-0431425. E.A.A.J. acknowledges funding through an NRC-NIST Fellowship. Supporting information for this article is available on the WWW under http://www.angewandte.org or from the author. 7999 Angew. Chem. 2007, 119, 7999 –8001  2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim