(a) The interior of the PNP vacuum chamber, showing the X, Y, and Z coarse specimen positioning stages as well as the gantry and load head. (b) A close-up of the PNP load head, showing some of the electronics associated with the capacitance gauges, tuning fork sensors, and piezoelectric actuators.
A computer-assisted design rendering of the PNP load head, showing the critical components. Details of PNP operation are presented in the discussion related to Figure 3 .
Schematic representation of the operation of the PNP load head in surface tracking mode. On the left, the tuning fork has sensed the surface through a shift in its resonant frequency, and the indenter tip is positioned just above the specimen (1). On the right, the D-frame in the load head has advanced downward, the indenter tip has been pressed into the specimen, and the springs (2) supporting the indenter tip have deflected, allowing measurement of the applied force via the capacitance transducer (3). As the D-frame advanced, however, feedback to the piezoelectric actuator (4) held the displacement frame (5) fixed with respect to the specimen surface. Therefore, the capacitance gauges (6) make a direct measurement of the motion of the indenter tip with respect to the specimen surface that is not influenced by the compliance of the gantry, the X, Y, and Z axis translation stages, or any aspects of the specimen mounting scheme.
Details of the tuning fork mounting and adjustment mechanism. (a) A close-up of a typical tuning fork. The cylindrical protective can has been cut and approximately two-thirds of it removed. (b) The pivot mechanism used to adjust the height of each tuning fork relative to the height of the indenter tip. Screws pressing against stiff and compliant regions of the adjuster provide both coarse and fine adjustment of tuning fork angle. (c) A photograph showing the indenter tip and tuning forks just above a specimen of single-crystal silicon. Circles are shown to indicate similar regions in (b) and (c).
A block diagram of the PNP control system. The feedback control system for tuning-fork surface tracking operation, consisting of a signal generator, the tuning fork sensors, tracking piezoelectric (PZT) actuators, and part of the real-time controller, is highlighted in red (with shading).
Vibration and noise environment of the PNP. (a) Vibration spectra recorded by the PNP load head in the sealed chamber before and after the implementation of several vibration isolation measures. (b) Typical temperature variations observed inside and outside the sealed PNP chamber. The laboratory temperature is controlled to ∼±0.5 °C. The rms variation in temperature inside the chamber is less than 0.005 °C, with a maximum excursion less than 0.02 °C, over the same time period.
The observed change in the position of a Berkovich indenter tip relative to the specimen surface while a constant indentation force of 30 mN was held for 12 h.
(a) Photograph showing the fused silica specimen, 12 mm × 12 mm × 3 mm thick resting on 3 mm thick rubber gasket material. (b) The stiffness of three different mounts for the silica specimen, as measured with a circular flat punch 10 μm in diameter: silica glued to an Al cylinder with cyanoacrylate adhesive (red solid line), 546 mN/μm; silica resting on 3 mm thick rubber gasket (green dashed line), 7.4 mN/μm; and silica resting on 12 mm thick polystyrene foam (blue dotted line), 0.75 mN/μm.
(a) Force-displacement data from 5 mN Berkovich indentations on the silica specimen shown in Figure 8 . Blue solid lines are for silica glued to an Al mount; red dashed lines are for silica resting on the rubber gasket. Although movement of the D-frame is significantly different for the two mounting methods, the indentation depth measured relative to the tuning-fork surface reference frame is correct even for the much more compliant rubber mount. (b) Force-displacement data for the silica specimen resting on polystyrene foam. Again, both D-frame motion and surface-referenced indentation depth are shown.
(a) Force vs. indentation depth for Berkovich indentation of fused silica, measured with the PNP. Peak forces from 5 mN to 50 mN, in 5 mN steps, were applied sequentially, with the indenter tip fully retracted above the specimen surface prior to each reloading. (b) Young's modulus calculated from power-law fits to the individual unloading slopes in (a), assuming a Poisson's ratio of 0.17 for silica. The values represented by black triangles are calculated assuming a perfectly sharp Berkovich geometry; the red squares assume a rounding of the Berkovich tip with a radius of curvature of ≈50 nm. The dashed line is the literature value of 72 GPa. Error bars represent one standard deviation in the uncertainty of the unloading fits.
Measurement of indentation creep in poly(methyl methacrylate) using a Berkovich indenter tip and a fixed applied force of 5 mN. In (a), time = 0 and depth = 0 correspond to the beginning of initial loading. The maximum force of 5 mN was reached after approximately 10 s (indicated by the circled spot) and the data show a slight noise deflection when the PNP switched to holding a fixed 5 mN force. Figures (b) and (c) show the change in indentation depth under fixed force, where time and depth now correspond to changes relative to the point in (a) where the force first reached 5 mN. The red curve in (b) and (c) is a logarithmic fit to the data over the first 10 min under full applied force, as described in the text, and the green curve in (c) is a logarithmic fit to the data after 60 min.
Article metrics loading...
Full text loading...