past research work

In the beginning...

My scientific activity in the field of computational chemistry started in 1994/95. In that period, under the supervision of Dr. Alexandre Magalhães, I've carried out semi-empirical calculations with the aim of obtaining the hidration environment of neighboring protein Arg-Arg residues, simulated by two guanidinium cations.

Hidration environment of a system Arg-Arg

In this work, the semi-empirical Austin Method 1, AM1, was used to study the local environment of two Arginine fragments. These fragments are known to be in the surface of some proteins and usually they are close to each other, interestingly enough since their chain terminations are positively charged. Everybody knows that charged particles with the same sign are thrusted away from each other! Thus, the solvent around these protein terminations should be responsible for the stabilization of these equally charged particles. To reach our purposes, each protein Arg fragment was simulated by a guanidinium cation, Gu+, and several systems Gu+Gu+.nH2O, n=[1-36], were studied. The Gu+Gu+ system is stabilized more efficiently when the number n of water molecules is 24. The stabilization in energy is of about 215 kcal/mol and the C-C distance between the two Gu+ fragments is 5.40 Å.

Then, DFT came up...

Within DFT theory, I've done an extensive research on the methanol decomposition reaction taking place at copper, silver and gold surfaces. The B3LYP hybrid method was used since previous work carried out by Anna Ignaczak, also in the FCUP's Theoretical Chemistry group, show that this approach yields excellent at a relative low computational cost. This was the main topic of my Ph.D. work performed under the orientation of Prof. Dr. José A. N. Ferreira Gomes.

Chemical reactions at metal surfaces

My Ph.D. work was devoted to the oxidation mechanism of methanol catalyzed by transition metal surfaces. The importance of the interaction of methanol with transition metal surfaces is due to its presence both as a reactant or product in the deoxygenation and in partial oxidation of methanol. Methanol is also used nowadays in fuel cells and prototypes of this kind of batteries are used for example in space missions or celular phones. Due to its simple structure, reactions involving methanol are interesting in the understanding of other heterogeneously catalyzed reactions.

The phenomena which occur near the metal surface involves only a small number of surface atoms, which possibilitates the use of theoretical chemistry based methods. The Density Functional Theory based hybrid B3LYP method was used in this work. The B3LYP yields very good results in systems where transtion metal atoms are present and it uses an exchange functional proposed by A. D. Becke and adds the functional correlation proposed by Lee, Yang and Parr based in a previous work of Colle and Salvetti. Also, throughout our work, the cluster approach was used to model the catalyic metal surfaces.

The main objectives of this work were:

# Find which adsorption sites and geometries that stabilized more efficiently the adsorption of the species experimentally proposed to appear during the oxidation of methanol in the catalyst surface. These were: CH3OH, CH3O, H2CO, H2CO2, HCO2, HCO, CO2, O, H, OH, H2O.
# Find if the dioxymethylene species, H2CO2, is (or not) the intermediate between formaldehyde, H2CO, and formate, HCO2.
# Find partial and global reaction routes.

Main conclusions from:

# Adsorption of methoxy on Cu(111), Ag(111) and Au(111).
Geometry: O-bonded with CO axis quasi-perpendicular to the metal surface
Site: 3-fold fcc hollow site
Stability: Cu>Ag>Au

# Adsorption of formate on Cu(100), Cu(110), Cu(111).
Geometry: Bridge-bonded species with molecular plane perpendicular to the metal surface
Site: Short-bridge
Stability: Similar

# Adsorption of dioxymethylene on Cu(111).
Geometry: Bridge-bonded species with molecular plane perpendicular to the metal surface
Site: Cross-bridge
Possible intermediate in the methoxy to formate reaction

# Adsorption of formyl on Cu(111), Au(111), Pt(111).
Geometry: C-bonded with the HCO plane normal to the surface
Site: Bridge
Stability: Pt>Cu>Au

# Decomposition of methoxy on Cu(111) and OH-covered Cu(111).
Preferred decomposition route found is for the hydroxylated Cu(111). The hydroxy species adsorbed on a top site site captures the hydrogen atom of the methoxy radical adsorbed on a fcc-hollow site. However, when compared with the clean surface, the energetic differences are of a few kJ/mol.

# Oxidation of methanol on Cu(111).
Full oxidation to carbon dioxide occurs at a zero cost in energy! Energetic barriers were not computed in this work. Formaldehyde is weakly adsorbed on the Cu(111) surface and formate is easily obtained from formaldehyde possibly by the dioxymethylene intermediate. The methanol oxidation to formaldehyde is in competition with the oxidation to formate.

# Oxidation of formaldehyde on clean and atomic oxygen covered Cu(111).
Formaldehyde is weakly adsorbed on clean Cu(111). The presence of atomic oxygen on the surface increases the adsorption energy by 50%. The calculated energy barrier for the reaction of formaldehyde yielding the dioxymethylene species is of about 36 kJ/mol. This low value confirms that dioxymethylene is the reaction intermediate in the oxidation to formaldehyde