2. Materials and methods

2.1 Test specimens

Fatigue specimens were machined from 2024-T351 aluminium plates. This aluminium alloy is an Al-Cu-Mg alloy which has been solution heat treated, control stretched and naturally aged. The mechanical characteristics of the specimens are presented in Table 2.
Fatigue specimens were “open-hole” T-Type elementary specimens (Fig. 3) machined in such a way that their longitudinal axis was aligned with the rolling direction of the 2024-T351 plate. The specimen width was three times the nominal hole diameter (D) and its length was 200 mm. The holes were obtained by a one-step drilling operation followed by a deburring operation.
To generate potential differences in hole surface integrity, axial and orbital drilling processes corresponding to industrial applications were used. As different tools and cutting parameters are used with these processes from one drilling diameter to another, two drilling diameters were considered in the study: 6.35 mm and 9.53 mm.
The fatigue specimen thickness was chosen according to industrial applications. It was 6.35 mm for the specimens with D = 6.35 mm and 10 mm for the specimens with D = 9.53 mm.
According to Faurie et al.24, the stress intensity factor in the working section (gross section minus hole area) was 3.5 for the specimens with D = 6.35 mm and 3.2 for the specimens withD = 9.53 mm. This ensured that the onset of cracking took place at the hole edge.

2.2 Procedures

2.2.1 Drilling processes

Drilling tests were carried out using a DMG DMU85eVo machining centre. In order to conduct orbital drilling operations, a PRECISE France – ORBIBOT orbital spindle was fixed on the Z -axis of the machining centre. For this drilling process, the machining was done in up-milling (tool rotation and orbital rotation in the same sense).
The cutting parameters, the tools and the lubrication used for each drilling process and for each hole diameter corresponded to industrial applications. They are presented in Table 3.

2.2.2 Fatigue tests

Fatigue tests were performed using a Schenk servo hydraulic machine equipped with a 100 kN load cell, at room temperature. The load was applied in the longitudinal direction of the specimen as a cyclic (sinusoidal) tensile-tensile load with a load ratio of 0.1 and a frequency of 20 Hz. Fatigue tests were carried out for various stress levels in order to obtain S-N curves (Wöhler curves). The maximum stress level applied in the working section varied from 100 MPa to 280 MPa.

2.2.3 Roughness measurements

2D roughness parameters (Ra, Rq, Rz) were measured along the height of the hole with a profilometer. In accordance with [ISO 4288, 1996], these measurements were performed with a cut-off length of 0.8 mm.
In order to complete the roughness results obtained from the analysis of linear profiles, an analysis of the surface roughness was also carried out. 3D roughness parameter measurements (Sa, Sq, Sp, Sv, Sz, Sdq, Ssk, Sku) were made using an ALICONA Infinite Focus device, which is capable of performing non-contact optical surface topology measurements through the focus variation principle. A magnification of 10x and a vertical resolution of 200 nm were used for the measurements.

2.2.4 Hardness measurements

Vickers microhardness measurements were made on the hole surface of the fatigue specimens previously cut in the working section (Fig. 4). The indentations were made with a load of 1 kgf and an indentation time of 15 s. Because of the cylindrical shape of the hole, a corrective factor was applied to the results as recommended by [ASTM E92-82].
In order to complete the microhardness results and to identify the material depth affected by a variation in hardness, nanoindentation tests were also conducted. The indentations were made at mid-thickness of the fatigue specimen in the (x,y) plane. To do this, specimens were cut from fatigue specimens and set in a resin in order to be polished. They were polished manually using emery papers with decreasing grit size (down to 4000 grit), then using diamond paste (1 µm) and finished with OP-S suspension. The applied load and the size of the indentation matrix were determined on the basis of the results of the preliminary numerical study, which predicted an affected material depth of a few tens of microns (Fig. 1). The indentation matrix was a matrix of 5 x 20 indents equally spaced 5 µm apart, with the first column located around 5 µm from the hole edge (Fig. 5). The indentation tests were performed, with a diamond Berkovich indenter, as load-controlled tests with a maximum load of 5 mN. Load-displacement curves obtained were analysed using the Oliver and Pharr method25 to determine the nanohardness HIT.

2.2.5 SEM-EBSD analysis

Previous works had shown that the Electron Back-Scatter Diffraction (EBSD) technique could be used to localize and estimate plastic strain through the analysis of local changes in the crystal orientation26,27. This technique was therefore used in the study to assess the strain hardening state of the hole subsurface. The Grain Reference Orientation Deviation (GROD), which represents the misorientation between a given pixel and the average orientation of the grain, was studied as a plasticity marker.
SEM-EBSD observations were carried out using a Feg Jeol JSM 7100 F Scanning Electron Microscope (SEM) equipped with an Oxford Instruments Nordlys Nano detector (Centre de microcaractérisation Raimond Castaing, CNRS UMS 3623, Toulouse, France). They were performed on a section at mid-thickness of the fatigue specimen in the (x,y) plane, as for the nanoindentation tests. The observed section was prepared with a cross section polisher, to preserve the microstructure and avoid polishing artefacts. For the measurements, the sample was tilted to 70° relative to the incident beam, as in conventional ESBD analysis, and a voltage of 20 kV was used. An area of 600 x 200 µm in the hole edge area was scanned with a step size of 0.5 µm.

2.2.6 Residual stress analysis

Current techniques for residual stress evaluation (X-ray diffraction and incremental hole drilling) were first considered for the study of the residual stress state of the part. However, these techniques were difficult to implement due to the cylindrical shape of the hole, the large grain size of the material and the thinness of the material layer affected by residual stresses in the hole edge area.
Therefore, a novel strategy, the Hole Opening Comparative Technique (HOCT) was set up. This method is based on the splitting method28 which is usually used to assess the residual stresses in thin-walled tubes29. The HOCT consists in introducing a slot along the half of the working section of the fatigue specimen that opened the hole and measuring the specimen deformation induced by this material removal which indicates the level of residual stresses. The greater the residual stress state in the hole edge area, the greater the specimen deformation after the opening of the hole. Thus, the HOCT allows to evaluated the residual stress state in the hole edge area in drilled open-hole T-Type specimen in a qualitative way and can be used as a quick comparative test.
The machining of the slot was done by Electrical Discharge Machining (EDM) which avoids introducing significant levels of residual stress in the specimen. The specimen deformation was evaluated through measurement of a profile shape before and after the material removal. As the residual stresses were assumed to be much greater in the radial and orthoradial directions than in the axial direction, the specimen deformation induced by the HOCT was measured in the (x,y) plane. The measured profile was orientated along the longitudinal direction of the specimen and located at mid-thickness of the specimen on the side opposite the machined slot (Fig. 6). This shape measurement was performed with an ALICONA Infinite Focus device.