Using support vector machine for materials design

Materials design is the most important and fundamental work on the background of materials genome initiative for global competitiveness proposed by the National Science and Technology Council of America. As far as the methodologies of materials design, besides the thermodynamic and kinetic methods combing databases, both deductive approaches so-called the first principle methods and inductive approaches based on data mining methods are gaining great progress because of their suc- cessful applications in materials design. In this paper, support vector machine （SVM）, including support vector classification （SVC） and support vector regression （SVR） based on the statistical learning theory （SLT） proposed by Vapnik, is introduced as a relatively new data mining method to meet the different tasks of materials design in our lab. The advantage of using SVM for materials design is discussed based on the applications in the formability of perovskite or BaNiO3 structure, the prediction of energy gaps of binary compounds, the prediction of sintered cold modulus of sialon-corundum castable, the optimization of electric resistances of VPTC semiconductors and the thickness control of In203 semiconductor film preparation. The results presented indicate that SVM is an effective modeling tool for the small sizes of sample sets with great potential applications in materials design.

Hydrogen generation from methanol reforming under unprecedented mild conditions

A homogeneous catalyst [Cp＊Rh（NH3）（H2O）2]-（3＋） has been found for the clean conversion of methanol and water to hydrogen and carbon dioxide. The simple and easily available reaction steps can circumvent the formation of CO, therefore, making it possible to avoid inactivating catalysts and contaminating the hydrogen fuel. Different from conventional reforming method for hydrogen production, no additional alkaline or organic substances are required in this method. Valuable hydrogen can be obtained under ambient pressure at 70 C, corresponding TOF is 83.2 h 1. This is an unprecedented success in reforming methanol to hydrogen. Effects of reaction conditions, such as reaction temperature, initial methanol concentration and the initial p H value of buffer solution on the hydrogen evolution are all systematically investigated. In a certain range, higher reaction temperature will accelerate reaction rate. The slightly acidic condition is conducive to rapid hydrogen production. These findings are of great significance to the present establishment of the carbon-neutral methanol economy.

Review of the first principles calculations and the design of cathode materials for Li-ion batteries

Cathode materials are the most critical chal- lenge for the large scale application of Li-ion batteries in electric vehicles and for the storages of electricity. The first principles calculations play an important role in develop- ment and optimization of novel cathode materials. In this paper, we overview the first principles calculations of energy, volume change, band-gap, phase diagram, and Li- ion transport mechanism of cathode materials with an emphasis on the design of such materials. We also over- view the recent progress of data mining techniques and the high-throughput first principles calculations for the design and development of cathode materials. Finally, we preview the challenges and opportunities of this rapidly developing field.

Hydrogenation of graphene nanoflakes and C-H bond dissociation of hydrogenated graphene nanoflakes： a density functional theory study

The Gibbs free energy change for the hydro- genation of graphene nanoflakes Cn （n = 24, 28, 30 and 32） and the C-H bond dissociation energy of hydrogenated graphene nanoflakes CnHm （n = 24, 28, 30 and 32; and m = 1, 2 and 3） are evaluated using density functional theory calculations. It is concluded that the graphene nanoflakes and hydrogenated graphene nanoflakes accept the orth- aryne structure with peripheral carbon atoms bonded via the most triple bonds and leaving the least unpaired dan- gling electrons. Five-membered rings are formed at the deep bay sites attributing to the stabilization effect from the pairing of dangling electrons. The hydrogenation reactions which eliminate one unpaired dangling electron and thus decrease the overall multiplicity of the graphene nanoflakes or hydrogenated graphene nanoflakes are spontaneous with negative or near zero Gibbs free energy change. And the resulting C-H bonds are stable with bond dissociation energy in the same range as those of aromatic compounds. The other C-H bonds are not as stable attributing to the excessive unpaired dangling electrons being filled into the C-H anti-bond orbital.