PSI - Issue 23

A. Eremin et al. / Procedia Structural Integrity 23 (2019) 233–238 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

235

3

For the static tests, the specimens had a dog-bone shape with a gauge length of 14 mm and the cross-section of 1×3 mm. These specimens were subjected to deformation with the strain rate of 2.4×10 -4 s -1 . The data on the deformation and fracture processes were additionally collected in situ implementing:  Digital image correlation technique, which utilizes the metallographic optical microscope Altami MB0670D with mounted CCD-camera Basler piA2400-17gc 5MP. Covered area was 11.3  9.5 mm (scale 4.6 µm/pixel) in order to observe the almost whole specimen gauge area. Image processing for obtaining the displacement and strain fields were done via commercially available VIC-2D software (Correlation Solutions, USA).  Acoustic emission, which utilizes analog-to-digital board LA-n150-14PCI with 14-bit resolution and sampling frequency of 3 MHz (Rudnev-Shilyaev, Russia), acoustic emission transducer WD (Physical Acoustics Corporation, USA), and analog signal amplifier (ISPMS SB RAS, Russia) with the gain of 40 dB. The specimens for fatigue tests had a rectangular shape with a single edge notch of 0.5 mm length. The specimens were subjected to cyclic loading under maximum load P max =0.6 kN and loading ratio R=0.1 which corresponds to maximum stress intensity factor (SIF) K max ≈ 16 MPa · √m and SIF range ΔK≈ 14.7 MPa at the beginning of the test. The image acquisition of the crack tip at high magnification was performed for the further DIC computation of displacement and strain fields and also the estimation of the local strains in the vicinity of the crack tip. The image capture was done under fixed maximum load ( P max ) every 2000 cycles before crack initiation and every 500 cycles during the stage of macrocrack growth.

4. Experimental results

4.1. Static tests results

Fig. 2 represents the loading diagrams for two specimen types (CG and UFG Ti). Acoustic emission data presented only for the UFG specimen, while the AE signals for CG specimens were not registered under chosen conditions (signal magnification, sensor sensitivity, thresholds, etc.). This could be explained due to high plasticity (easy glide) of the CG material and thus low energy accumulation which leads to acoustic wave emission with amplitudes lower than the registration threshold. However, loading of the UFG Ti results in the extensive generation of acoustic waves which are more likely attributed to dislocation bunch formation on the boundaries. It also should be noted, that the acoustic emission signals start approximately jointly with the plastic flow.

Fig 2. Stress-strain curves for CG and UFG titanium (solid line), counts of AE signals for UFG Ti (red circles).

As it was derived from the tensile tests the ultimate tensile stress was enhanced from 419 ± 6 MPa for CG Ti to 863 ± 16 MPa for UFG Ti, at the same time the strain at fracture was reduced from 32 ± 2.7 % to 15 ± 2.3 % correspondingly. The linear parts of the stress-strain curves for UFG and CG specimens have the same slope and thus they have equal Young moduli. Furthermore, grain refinement changes the plastic flow and strain-hardening behavior of the material.