This work is oriented to investigate the physical properties of thin film, especially elastic and plastic properties. The elastic modulus, yield strength, tensile strength was investigated by two main mechanical measurement technique and those was compared with theoretical mechanical properties, based on the texture analysis and the Hall-Petch relationship. The microstructure of Cu thin film was quantitatively characterized by the scanning electron microscope analysis and the X-ray diffraction analysis.
The quantitative microstructure of thin film is important to verify the effect of microstructure on mechanical properties. The microstructure of Cu thin film was characterized by high resolution scanning electron microscope and the grain size of Cu thin film was conductedThe grain size was characterized by the computerized image analysis program and the average and the distribution of grain size was obtained from the high resolution SEM picture. The grain size increased with increasing thickness of Cu thin film. The shape of grain was an elongated structure in perpendicular direction to surface of Cu thin film. The aspect ratio of grain was ranging from 12 to 30. The aspect ratio was also increased with increasing thickness of Cu thin film. The grain size and the dislocation density of thin film were characterized by the other method, X-ray diffraction analysis. The modified Williamson-Hall method was used to analyze the result of X-ray diffraction analysis, which was derived by Ungar. The grain size measured by modified WH method showed similar or slightly low value as that measured by microstructure analysis of scanning electron microscope. The dislocation density of thin film was not common. The sputtered Cu thin film showed high dislocation density relative to the electroplated Cu thin film. The dislocation density of severely deformed Cu materials was similar as that of sputtered Cu thin film. The residual stress was characterized by $sin^{2}＼psi$ method of X-ray diffractometer. The residual stress was ranging from 11MPa to 53MPa, which was small in comparison to the yield strength of the thin film.
The texture was characterized by using the pole-figures of X-ray diffraction. The textures of sputtered and electroplated Cu thin film were different. The sputtered Cu thin film showed random texture with strong to <111> direction, and <233> texture increased with increasing the thickness of sputtered Cu thin film. The electroplated Cu thin film showed strong <011> texture, and the texture changed to <311> direction with increasing the thickness of sputtered Cu thin film. The theoretical elastic moduli were estimated base on the texture analysis by using Voigt’s model and Hill’s model in lateral and perpendicular directions to the surface of Cu thin film.
The elastic modulus, the yield strength, the ultimate tensile strength and the elongation of Cu thin film were characterized by using micro-tensile test. The micro-tensile test was performed by employing an Instron 8848 model, which was enhanced capability to measure low force. The enhancement to load resolution was achieved by using a special filter with each load cell. The resolution of load measurement was less than 0.1mN. The elastic modulus of sputtered Cu thin film slightly increased with increasing the thickness of Cu thin film, while that of electrodeposited Cu thin film decreased with increasing the thickness of Cu thin film. The elastic modulus was compared with the theoretical elastic modulus based on the texture analysis in lateral direction to the surface of Cu thin film. The elastic modulus of Cu thin film, measured by micro-tensile test, showed similar value as the theoretical elastic modulus in lateral direction to the surface of Cu thin film. This result indicates that the elastic modulus measured by the micro-tensile test represents the elastic modulus in lateral direction to the surface of Cu thin film. The yield strength and the ultimate tensile strength of Cu thin film decreased with increasing the thickness of Cu thin film. The yield strength of Cu thin film showed good linear correlation with inverse square root of the grain size of Cu thin film. This result indicates that the strength of Cu thin film follows the Hall-Petch type relationship in spite of the small grain size of more than 200nm. The elongation of Cu thin film was smaller than that of severely deformed Cu materials, and the strain-hardening rate was higher than of severely deformed Cu materials with same strength. The surface of fracture was investigated. The fracture of Cu thin film was initiated at the position between the nucleated grains on substrate and the elongated grains on surface, while the fracture of severely deformed Cu was initiated at the center of specimen. This result indicates that the small elongation and the high strain-hardening rate were originated by requirement of geometrically necessary dislocations.
The elastic modulus and the hardness of Cu thin film were characterized by using nanoindentation test. The nanoindentation test was performed by employing a MTS Nanoindenter XP and a Hysitron Triboscope on Seiko SPA 400 AFM. The results of two nanoindenter machine was same in the indentation depth of 100~300nm. The surface effect disappeared after the indentation depth of 100nm, which was more than three times of the roughness of Cu thin film. The substrate effect did not appeared before the indentation depth of 300~500nm, that was less than 10% of the thickness of Cu thin film. The elastic modulus was measured by obtaining average from 100nm to 200nm. The elastic modulus of sputtered Cu thin film increased with increasing the thickness of Cu thin film, while that of electrodeposited Cu thin film decreased with increasing the thickness of Cu thin film. The elastic modulus was compared with the theoretical elastic modulus based on the texture analysis in perpendicular direction to the surface of Cu thin film. The elastic modulus of Cu thin film, measured by the nanoindentation test, showed similar value as the theoretical elastic modulus in perpendicular direction to the surface of Cu thin film. This result indicates that the elastic modulus measured by the nanoindentation test represents the elastic modulus in perpendicular direction to the surface of Cu thin film. The hardness of Cu thin film in same indentation depth decreased with increasing the thickness of Cu thin film. The hardness in same indentation depth did not show linear correlation with the yield strength measured by the micro-tensile test. The absolute hardness was estimated to eliminate an indentation size effect by using Nix and Gao relationship. The absolute hardness was calculated by using the data from 100nm to 500nm to be independent on the surface effect and the substrate effect. The absolute hardness of Cu thin film, estimated by Nix and Gao equation, showed good linear correlation with the yield strength, measured by the micro-tensile test. The absolute hardness of Cu thin film also showed good linear correlation with inverse square root of the grain size of Cu thin film. This result indicates that the absolute hardness of Cu thin film represents the strength of Cu thin film and follows the Hall-Petch type relationship.
