COMPUTATIONAL STUDY OF IRON(II) SPIN TRANSITION COMPLEXES

• Main Author: T Male, Yusthinus
• Format: Dissertations
• Language: Indonesia
• Subjects: Kimia
• Online Access: https://digilib.itb.ac.id/gdl/view/35569
• Institution: Institut Teknologi Bandung

The change of two different spin electronic states under influence of temperature, pressure, light irradiation and magnetic field was known as spin transition phenomena. Spin transition (ST) exist on octahedrally complexes with intermediate ligand field. ST complexes were intesively investigated because they are highly potential candidates for electronic materials. Most of ST complexes consist of iron(II) core which are surrounded of bidentate diimin ligand with nitrogen atoms acting as electron donor. By experimental method, many new ST complexes has been produced but their properties could not be predicted before they were synthesized. Many theoretical studies have been done to explain the ST phenomena but methods that can predict accurately the ST potential of complexes have not been obtained. In this research, computational chemistry has been used to study theoretical aspect of ST systems. Computational chemistry is very useful to predict structure, mechanism and reaction thermodynamics so that chemists will be able to determine the ST properties of compound that have not been synthesized yet. An alternative ab initio method, i.e. Density Functional Theory-DFT, was chosen in this research using Gaussian 03 software. Reliable computation of optimized geometries, energies, and electronic structures of transition metal containing system is known to require extensive consideration of electron correlation effects. The Hartree-Fock (HF) method without configuration interaction turns out to be inadequate and a treatment of electron correlation by multiconfigurational procedure is computationally expensive. DFT methods are promising alternatives to the traditional ab initio methods for characterizing these systems because they include electron correlation in the exchange-correlation functional. DFT method has been proven that it is efficient enough to calculate macromolecules properties. In this research, computational method has been used to predict structure, energy, electronic spectrum and ST potency for iron(II) ST complexes with various ligands with nitrogen atoms acting as electron donor. Computational calculations of a simple complex [Fe(en)2(NCS)2] in cis and trans isomers have been done using Hartree-Fock (HF), Becke’s exchange with LYP functional (BLYP) and Becke’s three-parameter hybrid functional (B3LYP) with 6-31G(d) basis set. Calculation results show that B3LYP method gives more accurate calculating ?Eel rather than HF and BLYP methods for cis- [Fe(en)2(NCS)2]. These results also shows that methanol as a polar solvent tends to stabilize cis-[Fe(en)2(NCS)2] complex. Thus, B3LYP/6-31G(d) method was chosen because it gives more precise in ?Eel and that it accurately determined stabilized isomers To study the influence of steric effects on the structure and properties of ?Eel on ST complex, calculation has been done with B3LYP/6-31G(d) method for unsubtituted complex, [Fe(bp)3]2+; complex with a bp ligand was substituted by methyl, [Fe(mbp)3]2+; and [Fe(pq)3]2+ with fuzed benzene ring as a substituent. Computational results show that subtituents induced a steric barrier and as a consequence, considerable distortion on the complex cation was observed. The Fe-N bond distance increases in the order of Fe-N(bpy) < Fe-N(bpym) < Fe-N(pyq) while the ?Eel decreases in the opossite direction, ?Eel Fe(bp) > ?EelFe(mbp) > ?EelFe(pq). B3LYP/6-31G(d) method yields a good accuracy with geometry, thermochemistry and vibrational spectrum but gives poor computational results in the calculation of ?Eel. Reparameterized B3LYP with a = 0.15 gives reasonable ?Eel, denoted by B3LYP*/6-31G(d) model and this method was recommended for DFT calculations on transition metal complexes. The main atomic orbitals of HOMO (highest occupied molecular orbital) and LUMO (lowest unoccupied molecular orbital) have been analyzed and proven that the electronic ground bands and the next ground bands are assigned to metal-to-ligand charge-transfer (MLCT) transitions. With calculation based on electronic structure theory, it can be shown here that incorporation of methyl and benzene ring substituent created red shift in the UV spectra of the complex to visible region. A good agreement between computational results and experimental data have shown the capabilities of this method to predict the properties of new complexes. The potency of mononuclear ST complexes, i.e. [Fe(dpa)2(NCS)2] in cis and trans isomers have been predicted using B3LYP*/6-31G(d) computational method. Computational results showed that in vacuum and methanol, cis-[Fe(dpa)2(NCS)2] isomer gave reasonable value of ?Eel for ST. Formation of hydrogen bonding with methanol solvent stabilizing the total electronic energy of cis-[Fe(dpa)2(NCS)2] complexes by 3,41 kJ mol-1 than non hydrogen bonding formation. This results showed that if [Fe(dpa)2(NCS)2] complexes were synthesized, only cis isomer has ST properties. Analysis of main atomic orbitals populations showed that the electronic ground bands and the next ground bands are assigned to singlet ligand- to-ligand charge-transfer (1LLCT) transitions because of the charge transfer from NCS- ligand to the main ligand. It can be concluded that computational method B3LYP*/6-31G(d) gives more predictive power to mononuclear ST complex. Prediction of ST potency were performed too on the Fe(II)-Fe(II) and Fe(II)-Ni(II) complexes by B3LYP*/6-31G(d) method. These results have strengthened the fact that on mononuclear ST complex systems, substitution of one of the bidentate ligand with counter-ion NCS- will reduce ligand field energy in a significant way. By computational calculations, we can predict structures and ST properties of various candidates of spin-transition complexes that have not been synthesized