In particular, Kononen et al.  studied if the pure titanium is susceptible to stress corrosion cracking in a topical fluoride solution by using a U-shape specimen exposed in different time periods. Throughout the paper it was concluded that topical fluoride solutions can cause SCC of commercially pure titanium. Nakamura et al.  investigated titanium SCC by using two different methods, the Slow Strain Rate Tensile method and the Constant load test. The sample was immersed in is 20% NaCl solution at 90C. The primary cracks was found to be approximately 10 μm. Sanderson et al.  in their paper dealt with U-bent tests and dynamic tensile tests which showed that SCC are formed in titanium alloys in NaCl environment at room temperature. It had been found that the cracks are developed because the chemical polishing creates a layer of very small hydride precipitates. Roy et al.  studied SCC of titanium by using double cantilever-beam technique. The sample was immersed in acidic brine pH 2.7 containing 5wt% NaCl at 90C continuosly. Simbi et al.  investigated the intergranular stress corrosion cracking of pure titanium in methanol hydrochloric acid at room temperature. The surface of the specimen was chemically polished in a solution containing 40% HF and 70% HNO3 concentrated acids. Hsiung et al.  studied the corrosion resistance of pure titanium in a mixture of 1% NaCl and 0% - 1% NaF at a constant pH of 6 under different elastic tensile strains. The results of polarization resistance reveal that the Rp decreased on increasing the tensile strain and increase in NaF concentration.
In the present work, for the first time, the design and experimental apparatus testing was evaluated towards formulating an appropriate set of conditions for creating a tree shape morphology on the titanium surface. In order to achieve the above goal ultrasonic loading was used for promoting crack propagation. In ultrasonic loading, cracks propagate during the tensile part of each cycle, while during the compression part crushing, attrition and dissolution of surface oxide films widen the conduits of the corrosive medium to the crack. All cracking network pictures taken from SEM are processed by the Matlab software to show the percentile amount of cracks.
The morphology of titanium foils was characterized using Scanning Electron Microscope (SEM). The largest stress is found at the point where the sample is bent. Further, an image processing code was created in the Matlab software, example is provided in Figure 3, which processes the image from SEM, and shows the average percentage of the cracked surface that has been created in each image. Then we compare all the results of the experiments in order to come to conclusions about the effect of the different conditions applied on each experiment.
examined in SEM and it was observed that on the sample surface there were some primary cracks. These were created due to the sample machining. All primary cracks are similarly oriented. Several experiments were in which the primary cracks were in a horizontal or vertical position in relation to the electrolyte surfaces conducted.
In the absence of ultrasound stress corrosion it was found that the average length of the cracks (in microns) of the reference sample (Figure 5) was about 4.19%. The average percentage of the reference sample represents the primary cracks found in the sample.
Next objective of this study was to explore the contribution of each one of the important experimental conditions reported above (e.g. temperature, electrolyte concentration, etc.), separately in order to decouple their effect and find out which conditions affect the increase of the size of the primary cracks. For clarification, it should be stated that in all samples below the primary cracks are horizontal in relation with the electrolyte surface.
Air presence: As a next step, another Ti-sample examined under experimental conditions 0/1/0/0/0/0/1. In this case, oxygen was provided in the electrolyte solution. The average percentage of cracks was 4.97%. The chemical corrosion due to air bubbling led to a small increase of cracks.
Electrochemical anodization: Another Ti-sample was tested under experimental conditions 0/0/1/0/0/0/1. A platinum electrode was put at 10mm distance from the sample in order for the electrochemical anodization to take place. In this case the average percentage was found to be 2.9%. One observes decreasing of the cracks compared to the average percentage of the reference sample which was 4.2%. Under conditions of anodization which can be considered a dynamic condition, the rate of oxide development and intrinsic characteristics of the material (e.g. coherency) and at the same time the adherence of the formed oxide with the underlying metal seem to change with time and this could lead to a decrease of cracks .
Based on the previously-presented results, it can be stated that the experimental conditions which result in an increase of the size of primary cracks are ultrasonic loading and provision of air to the electrolyte for the reasons explained above, where mechanically-forced or chemical-assisted corrosion was performed.
In the next group of experiments the ultrasonic loading is applied in all the cases explored. Apart from this, in each experiment there is a different experimental condition applied. In all of the samples below the primary cracks are horizontal in relation to the ultrasonic loading. The first experiment (1/0/0/0/0/0/1) has already been discussed previously in section 3.1, i.e. in which the ultrasonic loading is applied and the percentage of the primary cracks increases (compared to the reference sample).
The next experiment (1/1/0/0/0/0/1) is a combination of ultrasonic loading and provision air to the electrolyte. The percentage of the cracks is increasing by 3.5% in relation to the previous experiments where only the ultrasonic loading was applied. When the sample is exposed to a combination of ultrasonic loading and electrochemical anodization (Figure 7) (1/0/1/0/0/0/1) one observes that there is an increase in the size of primary cracks. The size of the cracks was increased by another 6.5% compared to the experiment that only the ultrasonic loading was applied (1/0/0/0/0/0/1). This can be explained by the fact that an oxide film is probably formed which is more transparent to electrons and the propagation of cracking is kinetically enhanced .
By comparing the experiments 0/0/1/0/0/0/1 (mechanical stress: ultrasonic load) and 1/0/1/0/0/0/1 (mechanical and chemical stress: ultrasonic load in the presence of air), one observes that in the first sample the cracks decrease and in the second sample they increase (quadrupled) compared to the reference. When there is no application of ultrasonic loading in the sample the primary cracks are covered but when there is ultrasonic loading covering of the cracks is not allowed. The combination of ultrasonic loading and electrochemical anodization helps in the increase of primary cracks. This is an expected trend since it is already known from the literature that the chemical surroundings and the mechaniccal forces facilitate the corrosion cracking [17-21].
The next step was a combined exposure of the Tisample to ultrasonic loading (1/0/0/1/0/0/1) and increase of temperature from room temperature up to 50C - 55C. The percentage of the primary cracks increased in rela-
Exposure of the sample to an increased electrolyte concentration (Figure 8) (1/0/0/0/1/0/1) (doubled concentration from 100 mM to 200 mM) led to an increase of the percentage of cracks by 15.7% compared to the results of the experiments 1/0/0/0/0/0/1. It can be stated that the increased chlorine concentration assist the fracture of the protective oxide layer by forming some stable compounds with titanium. This effect combined with the stress applied can lead to the formation of trenches due to local dissolution. At the points where trenches are formed hydrogen adsorption can occur, coming from the solution, and causing the precipitation of brittle hydrides. As it has been reported, this effect enhances the propagation of cracks in titanium . By increasing the exposure time (1/0/0/0/0/1/1) from four to eight hours it was observed that the percentage of the primary cracks has been almost doubled (17.6%) compared with experiment where the duration of the experiment was 4 hours (1/0/0/0/0/0/1). This shows the control of the kinetics over cracks propagation.
that the size of the primary cracks is increased due to the following conditions: ultrasonic loading, provision of oxygen to the electrolyte, electrochemical anodization, increase of concentration of electrolyte and duration of the experiment.
In the next experiment (Figure 9) the ultrasonic loading, electrochemical anodization and the electrolyte solution with increased concentration were chosen since it was found that they enhance the size of the primary cracks. The code of the experiment is 1/0/1/0/1/0/1. The combination of these three conditions has increased the size of cracks (22.6%) by 2.5 times in comparison to the results of experiment 1/0/0/0/0/0/1 implying the synergistic activity in this case.
In the next two experiments the sample is positioned in such a way that the primary cracks are in a vertical position in relation to the ultrasonic loading. Two experiments were conducted, namely 1/0/1/0/0/0/0 and 1/0/0/0/0/1/0 according to the Table 1. One observes that the results of the experiments are decreased compared to the corresponding results when the sample was in a horizontal position in relation to the ultrasonic loading (1/0/1/0/0/0/1, 1/0/0/0/0/1/1). For this reason there was no sequel to these experiments since when the sample is in vertical position these is no increase on the percentage of the primary cracks.
By situating the primary cracks in a vertical position their percentage is not increased because of the inherent difficulty in cracks opening during the tensile part, and as a consequence the electrolyte diffusion met there a barrier to enter the cracks. In contrast, when the cracks are situated in a horizontal position the cracks open during the tensile part and the electrolyte diffuses relatively easily into the cracks and remains there during the compression part. In this way the size and the depth of the cracks is increasing. 153554b96e