Candidate Materials f or High Temperature Tribological Applications of Solid Lubricants Jonathan Liddy Dr. Simon Phillpot Dr. Susan Sinnott Eric Bucholz
Back g round : Increased demand for stronger, longer lasting and more sophisticated materials systems is adding pressure to the materials engineering industry to decrease the time span from materials research to materials deployment. Some tools that are typically used to accomplish this task are Trend Analysis, Data mining, Materials Informatics, and Computational Materials Science These methods c an provide effective ways to reveal suitable candidate materials for high temperatur e tribological applications Objectives: The first objective is to d evelop databases for selecting suitable candidate materials for high temperature tribological applications and begin to populate these databases with materials property data The next ob jective is to p erform Trend Analysis, data mining and informatics on materials property data to determine which properties have an impact on tribological behavior. Finally, we will select suitable candidate materials based on trends and property correlations observed in the data and then compare them with experimental results. Discussion /Results :
An experimental database was constructed using File Make Pro for the purposes of selecting su itable candidate materials for tribological applications. The database currently contains over 1500 materials which can be sorted by various specific materials properties. Examples of these properties include, but are not limited to, space group, d ensity, hardness, melting temperature chemical formula, l attice parameter, Shear Modulus Elastic Modulus, Bulk Modulus, and cleavage planes. File Maker Pro provides a great deal flexibility as to what type and how much information can be added for each materia l. Documents, pictures, Quick Time files, Crystal Maker files, and websites can all be added for each material entry. File Maker Pro also allows for manipulation of the graphical user interface (GUI) which enables the database to be designed to meet spec ific needs, as well as capabilities to sort by materials properties. Figure 1: Sample material data from File Maker Pro database. Trend analysis betwe en materials properties was performed to determine which properties have an influence on tribological behavior. Three structure types are investigated
which include lamellar having secondary bonding between layers, layered having primary bonding between layers, and not layered. It was found that the majority of materials with a lamellar structure have lo w hardness values, materials with layered structure have a range of hardness values from low to high, and non layered materials show a tendency towards higher hardness values. Also, lamellar structures and materials with lower hardness values appear to sh ow lower coefficients of friction than materials with higher hardness (Figure 2) Figure 2: Trend analysis of the influence of structure on the density and hardness of each material (left) and frictional properties of certain materials available in literature as a function of hardness and structure (right) From Literature, we found that as Ionic Potential ( /r ) increases, the coefficient of friction decreases for binary oxides.
Once it is determined which materials properties have the most in fluence on the frictional behavior of materials, principles of data mining and materials informatics will be used such that new candidates for solid lubricants can be obtained.
Figure 3 : Comparison of experimental and empirically derived friction coefficients for oxides, sulfides, and selenides Table 1: Comparison of coefficients of friction found both experimentally using a pin on disk tribometer and with our empirical formula
Figure 4: Predicted friction coefficients for many materials from our database
Figure 5: Loads plot from PCA 2 A correlation matrix for performed using Bivariate Correlations which computes Pearson correlation coefficients. Correlations measure how different variables rank orders are related and it is a measure of a linear association between variables. The cor relation matrix (Table 3) shows what variables are correlated to friction coefficient. Variables that are most closely linearly related to friction coefficient are poisson ratio, melting temperature, C 44 Elastic Modulus, and Shear Modulus. Negative values of coefficients for Coefficient of Thermal Expansion (CTE) show s that Coefficient of friction is increasing in value as the other variable decreases.
Table 4. Rotated Component Matrix Rotated Component Matrix a Component 1 2 Ionic Potential (Z/r) .999 .034 Ave Hardness .983 .184 Debye Temp .976 .219 Melting Temp (K) .995 .096 Electronegativity Difference .994 .108 Ave u .741 .671 C44 (Gpa) .996 .090 E(Gpa) .996 .087 G(Gpa) .997 .083 B(Gpa) .986 .166 Poisson Ratio .796 .606 Thermal Conductivity(W/mK) .924 .382 Coefficient of Themal Expansion (1/K) .608 .794 Refreactive index .996 .086 Density .038 .999 Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization. a. Rotation converged in 3 iterations.
Table 5. Total varia nce explained Total Variance Explained Component Initial Eigenvalues Extraction Sums of Squared Loadings Rotation Sums of Squared Loadings Total % of Variance Cumulative % Total % of Variance Cumulative % Total % of Variance Cumulative % 1 12.360 82.398 82.398 12.360 82.398 82.398 12.246 81.643 81.643 2 2.640 17.602 100.000 2.640 17.602 100.000 2.754 18.357 100.000 3 1.541E 15 1.027E 14 100.000 4 8.721E 16 5.814E 15 100.000 5 5.847E 16 3.898E 15 100.000 6 4.265E 16 2.844E 15 100.000 7 2.443E 16 1.629E 15 100.000 8 1.173E 16 7.823E 16 100.000 9 9.551E 17 6.367E 16 100.000 10 1.544E 16 1.029E 15 100.000 11 2.851E 16 1.900E 15 100.000 12 3.716E 16 2.477E 15 100.000 13 4.822E 16 3.215E 15 100.000 14 8.659E 16 5.773E 15 100.000 15 1.865E 15 1.243E 14 100.000 Extraction Method: Principal Component Analysis.
References: (1) Erdemir, Tribology Letters, 97 102 2000 (2) Rajan, Macromolecular Communications, 972 976, 2007 (3) Rajan, The Annual Review of Materials Research, 299 322, 2008
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E3XVC933B_64VX5N INGEST_TIME 2017-06-01T18:38:12Z PACKAGE AA00057776_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC