A data analytic materials informatics methodology is proposed after applying different data mining techniques on some datasets of particular domain in order to discover and model certain patterns, trends and behavior related to that domain. In essence, it is proposed to develop an information mining tool for vapor-grown carbon nanofiber (VGCNF)/vinyl ester (VE) nanocomposites as a case study. Formulation and processing factors (VGCNF type, use of a dispersing agent, mixing method, and VGCNF weight fraction) and testing temperature were utilized as inputs and the storage modulus, loss modulus, and tan delta were selected as outputs or responses. The data mining and knowledge discovery algorithms and techniques included self-organizing maps (SOMs) and clustering techniques. SOMs demonstrated that temperature had the most significant effect on the output responses followed by VGCNF weight fraction. A clustering technique, i.e., fuzzy C-means (FCM) algorithm, was also applied to discover certain patterns in nanocomposite behavior after using principal component analysis (PCA) as a dimensionality reduction technique. Particularly, these techniques were able to separate the nanocomposite specimens into different clusters based on temperature and tan delta features as well as to place the neat VE specimens in separate clusters. In addition, an artificial neural network (ANN) model was used to explore the VGCNF/VE dataset. The ANN was able to predict/model the VGCNF/VE responses with minimal mean square error (MSE) using the resubstitution and 3olds cross validation (CV) techniques. Furthermore, the proposed methodology was employed to acquire new information and mechanical and physical patterns and trends about not only viscoelastic VGCNF/VE nanocomposites, but also about flexural and impact strengths properties for VGCNF/ VE nanocomposites. Formulation and processing factors (curing environment, use or absence of dispersing agent, mixing method, VGCNF fiber loading, VGCNF type, high shear mixing time, sonication time) and testing temperature were utilized as inputs and the true ultimate strength, true yield strength, engineering elastic modulus, engineering ultimate strength, flexural modulus, flexural strength, storage modulus, loss modulus, and tan delta were selected as outputs. This work highlights the significance and utility of data mining and knowledge discovery techniques in the context of materials informatics.

The use of nanoreinforcements in automotive structural composites has provided promising improvements in their mechanical properties. For the first time, a robust statistical design of experiments approach was undertaken to demonstrate how key formulation and processing factors (nanofiber type, use of dispersing agent, mixing method, nanofiber weight fraction, and temperature) affected the dynamic mechanical properties of vapor-grown carbon nanofiber (VGCNF)/vinyl ester (VE) nanocomposites. Statistical response surface models were developed to predict nanocomposite storage and loss moduli as functions of significant factors. Only ~0.50 parts of nanofiber per hundred parts resin produced a roughly 20% increase in the storage modulus versus that of the neat VE at room temperature. Optimized nanocomposite properties were predicted as a function of design factors employing this methodology. For example, the use of highshear mixing (one of the mixing methods in the design) with the oxidized VGCNFs in the absence of dispersing agent or arbitrarily with pristine VGCNFs in the presence of dispersing agent was found to maximize the predicted storage modulus over the entire temperature range (30-120 °C). To study the key concept of interphase in thermoset nanocomposites, molecular dynamics simulations were performed to investigate liquid VE resin monomer interactions with the surface of a pristine VGCNF. A liquid resin having a mole ratio of styrene to bisphenol A-diglycidyl dimethacrylate monomers consistent with a 33 wt% styrene VE resin was placed in contact with both sides of pristine graphene sheets, overlapped like shingles, to represent the outer surface of a pristine VGCNF. The relative monomer concentrations were calculated in a direction progressively away from the surface of the graphene sheets. At equilibrium, the styrene/VE monomer ratio was higher in a 5 Å thick region adjacent to the nanofiber surface than in the remaining liquid volume. The elevated styrene concentration near the nanofiber surface suggests that a styrene-rich interphase region, with a lower crosslink density than the bulk matrix, could be formed upon curing. Furthermore, styrene accumulation in the immediate vicinity of the nanofiber surface might, after curing, improve the nanofiber-matrix interfacial adhesion compared to the case where the monomers were uniformly distributed throughout the matrix.

The effects of material/fabrication parameters on vapor-grown carbon nanofiber (VGCNF) reinforced vinyl ester (VE) nanocomposite flexural moduli and strengths were investigated. Statistically reliable empirical response surface models were developed to quantify the effects of VGCNF type, use of dispersing agent, mixing method, and VGCNF loading on flexural properties. Optimal nanocomposite formulation and processing (0.74 phr oxidized VGCNFs, dispersing agent, and high-shear mixing) resulted in predicted flexural modulus and strength values 1.18 and 1.26 times those of the neat resin. Additional flexural, tensile, and compressive tests were performed for optimally configured nanocomposites cured in a nitrogen environment. While flexural and tensile moduli significantly increased with increasing VGCNF loading, the corresponding strengths fell below those of the neat resin. In contrast, nanocomposite ultimate compressive strengths significantly exceeded the neat resin strengths. Nanocomposites prepared using aggressive high-shear mixing displayed improved elastic moduli and substantially increased strengths relative to nanocomposites prepared using baseline methods.