The near-surface mechanical properties of thin films as well as bulk materials are amongst the key parameters important for their application, and instrumented nanoindentation is a standard technique for determining these mechanical properties. However, it is known that the surface topography of the characterized materials may affect the nanoindentation data when a sharp indenter for small penetration depths (displacements) is used. A thin film of hydrogenated amorphous silicon carbide with a thickness of 1.0 μm was deposited on a silicon wafer by plasma-enhanced chemical vapour deposition. The cyclic nanoindentation was used to construct a depth profile of mechanical properties for the flat surface (0.5 nm roughness) of the thin film, which made it possible to determine its modulus of elasticity of 83 GPa and hardness of 8.6 GPa unaffected by the silicon substrate. Grains with a spherical cap geometry with a typical radius of 0.5 µm and a height of 60 nm are distributed along the flat surface of the film. The grains have the same mechanical properties as the deposited film. Depth profiles of mechanical properties were determined for different types of contact between the Berkovich indenter with a radius of 50 nm and the selected grain (grain top, grain foot, two or three grains); i.e. for these measurements the following applied - the radius of the tip curvature was less than grain radii (R_Berkovich < R_grain). Residual imprints after nanoindentation measurements were carefully observed by atomic force microscopy and scanning electron microscopy. The near-surface mechanical properties were significantly affected by the surface topography, and the determined modulus of elasticity and hardness were crucially under- or overestimated in the range of 50% to 100% compared to the real values. The nature of these deviations was discussed. The solution is to use cyclic nanoindentation performed on the flat surfaces or on the top of grains, followed by extrapolation of the depth profiles to the zero-contact depth (film surface).

薄膜和块状材料的近表面机械性能是对其应用很重要的关键参数之一，仪器化纳米压痕是确定这些机械性能的标准技术。然而，众所周知，当使用用于小穿透深度（位移）的尖锐压头时，表征材料的表面形貌可能会影响纳米压痕数据。通过等离子体增强化学气相沉积在硅片上沉积了厚度为 1.0 μm 的氢化非晶碳化硅薄膜。循环纳米压痕用于构建薄膜平坦表面（0.5 nm 粗糙度）的机械性能深度分布，这使得确定其弹性模量为 83 GPa 和硬度为 8 成为可能。6 GPa 不受硅衬底的影响。具有典型半径为 0.5 µm 和高度为 60 nm 的球形帽几何形状的晶粒沿薄膜的平面分布。晶粒具有与沉积膜相同的机械性能。针对半径为 50 nm 的 Berkovich 压头与选定晶粒（晶粒顶部、晶粒底部、两个或三个晶粒）之间的不同接触类型，确定了机械性能的深度分布；即对于这些测量，应用以下 - 尖端曲率的半径小于晶粒半径（针对半径为 50 nm 的 Berkovich 压头与选定晶粒（晶粒顶部、晶粒底部、两个或三个晶粒）之间的不同接触类型，确定了机械性能的深度分布；即对于这些测量，应用以下 - 尖端曲率的半径小于晶粒半径（针对半径为 50 nm 的 Berkovich 压头与选定晶粒（晶粒顶部、晶粒底部、两个或三个晶粒）之间的不同接触类型，确定了机械性能的深度分布；即对于这些测量，应用以下 - 尖端曲率的半径小于晶粒半径（R _Berkovich < R_谷物）。通过原子力显微镜和扫描电子显微镜仔细观察纳米压痕测量后的残留印记。近表面机械性能受到表面形貌的显着影响，与实际值相比，确定的弹性模量和硬度在 50% 至 100% 的范围内被严重低估或高估。讨论了这些偏差的性质。解决方案是使用在平坦表面或晶粒顶部执行的循环纳米压痕，然后将深度分布外推到零接触深度（薄膜表面）。