1. Introduction
The phenomenon of hydrodynamic erosion has gained increasing attention due to its severe effects on the hydraulic mechanical components such as impellers, valves, fittings, pumps, and hydrofoil surfaces [
1,
2]. The erosion initiated by impact loading, cyclic loading in the form of a hammer effect, and lateral jetting causes elastic-plastic deformation during the interaction of the jet with the material [
3]. Fundamentally, the impact of a water jet or droplet on a solid surface generates stresses that are categorized into two stages. The initial stage is the water hammer effect, which causes compressive stress for a short duration, followed by the shearing action of the material during lateral jetting [
4]. The compressive and shear stresses are distributed within the surface and sub-surface layers in the form of shock waves, which are responsible for enhancing the strength and surface integrity of the material. The conclusions inferred from previous experiments suggest that the erosion effect is the consequence of fatigue failure caused by the impact of a jet [
5]. The energy of a jet imparted on a solid surface is utilized for treatment applications by exploring the early stages of erosion (incubation). However, at the advanced stages of erosion, disintegration of the material occurs.
In view of the above, the erosion phenomenon (early and advanced stages of erosion) has been investigated using a pulsating (PWJ) or continuous (CWJ) water jet on various ductile materials. The water droplet erosion (WDE) mechanism has been investigated on Ti–Al alloys (fully lamellar) [
6], during which the test was interrupted at three stages of erosion (i.e., the incubation, maximum erosion, and steady-state stages); the microstructural analysis revealed that the erosion was initiated through inhomogeneous and localized material flow, followed by crack generation on the surface. Moreover, the advanced stages of erosion were directed by the periodic roughening and formation of deep craters. The WDE (early and advanced stages) has also been studied on Ti–6Al–4V surfaces [
7], in which the influence of the impact speed on erosion behavior in terms of mass loss was explored. It was observed that at impact speed (350 m/s), the maximum erosion rate increases as the initiation time becomes faster. The microstructural analysis revealed that the early erosion was limited to microcracks and isolated pits; however, substantial material removal (sub-surface cracking, upheaving) was observed in the advanced stage of erosion. Additionally, the effect of a water jet has also been explored for surface treatment application on various ductile materials, such as aluminum alloys [
8], in which the effect of various process parameters (e.g., pressure, number of passes, standoff distance, and feed rate) on the surface roughness and microhardness were investigated. The experimental analysis revealed that an increase in the number of passes (1 to 3), pressure (50–150 MPa), and standoff distance (20–60 mm) results in higher roughness and hardness values; however, a reverse effect was observed upon increasing the feed rate. Another study [
9] investigated the effect of the number of passes, the pressure, and the feed rate on the surface integrity of stainless steel. An enhancement in erosion and roughness was revealed upon increasing the number of passes and the pressure; however, the feed rate showed a reverse effect on the surface erosion and roughness.
Under the action of varying loads (impact and lateral jetting) on the material, modulated water jets are more suitable for material processing in comparison to CWJs [
3]. Modulated jets can be generated using different methods, such as self-resonating nozzles, interrupting continuous jets by a rotating disc with an orifice, or using vibrating mechanical devices (e.g., needle or ultrasonic cylinder) inside the pressure chamber. The impact of modulated jets generated using the above methods induces cyclic loading on the material surface [
10]. However, the related drawbacks, such as the short life of the moving components and design complexities, are overcome by using ultrasonic generator methods of pulse generation. This technology utilizes a repeated hammer effect produced by the periodic motion of the sonotrode and is advantageous in terms of the lifetime and reliability of the system [
10].
The interaction of a PWJ in terms of parametric influence (i.e., pressure, traverse speed, and standoff distance) in the early and advanced stages of erosion [
11,
12,
13,
14] and the flow field characteristics [
15] has been explored previously. Moreover, to eliminate the tensile stresses induced in AISI 304 tungsten inert gas (TIG) welded joints [
16], the weld surface has been exposed to a PWJ under variable parametric levels [
17]. However, the effect of a variable interaction time and an excitation frequency at the same volumetric flow rate on welded joints has still not been effectively determined. Therefore, in the present study, the effect of different interaction times obtained by varying the nozzle traverse speed (1–4 mm/s) with different excitation frequencies (20, 40 kHz) on the surface integrity of welded AISI 304 was studied. The surface integrity was studied using residual stress, microhardness, and surface roughness measurements of the sample after PWJ treatment.
The outline of the study is divided into following sections: The introduction is followed by
Section 2 (Materials and Methods) which illustrates the sample preparation and the experimental conditions along with the description of the setup used for the experiments and subsequent analysis
Section 3 includes the results and the detailed discussion of the results obtained and
Section 4 summarizes the primary conclusions of the investigation.
2. Materials and Methods
As done previously, double-butt TIG welded joints (
Figure 1) of AISI 304 were fabricated at a current of 97 A with a voltage of 12.1 V for inner butts and 11.8 V for outer butts [
17]. The chemical composition and mechanical properties of AISI 304 are mentioned in
Table 1 and
Table 2 [
17]. The joints were then treated using an ultrasonically generated PWJ machine comprising a Hammelmann HDP 253 plunger pump (Hammelmann GmBH, Oelde, Germany) integrated with an ABB robot IRB 6640-180/2.55 (ABB s.r.o, Ostrava, Moravská Ostrava, Czech Republic) for handling the PWJ head and an ECOSON WJ-UG_630-40 (Ecoson s.r.o, Nové Mesto nad Váhom, Slovakia) sonotrode for the initiation of the PWJ at frequencies of 20 and 40 kHz. The system of the PWJ consisted of an acoustic generator with a cylindrical waveguide located inside the acoustic chamber. In this system, the electric signals are transmitted to the liquid inside the chamber in the form of mechanical vibrations from the sonotrode (via the aid of a piezoelectric transducer). These periodic oscillations of the sonotrode cause pressure fluctuations, which are amplified in the mechanical amplifier and guided through the nozzle exit. At the nozzle exit, the pressure fluctuations are converted to velocity fluctuations, causing the breakage of a continuous jet stream into discrete clusters. The repeated impact of these clusters on the material surface induces cyclic stress, which surpasses the ultimate strength of the material and results in elastic–plastic deformation. These deformations are responsible for improving the surface characteristics of the material [
4].
A Stonage nozzle was selected for the treatment under the parameters stated in
Table 3. Experiments at a frequency of
f = 40 kHz were conducted under the experimental conditions mentioned in
Table 1, after
f = 20 kHz, which had already been reported earlier [
17] (
Figure 1). AISI 304 weld zone were exposed to PWJ starting with a traverse speed
v = 1 mm/s increasing up to 4 mm/s. This variation in the traverse also varies the number of impingements from 20,000 to 4000 impacts per mm. The same sequence was followed with higher excitation frequency
f = 40 kHz, as shown in
Figure 1. Standoff distance and supply pressure for all experimental runs were kept constant at
z = 70 MPa and
p = 40 MPa.
Figure 2a shows the treatment region and welding direction. The treatment was conducted on the surface along the width (50 mm) of the sample with a consecutive spacing of 5 mm between the traces to prevent the overlapping of the treated region. For observing the effect of interaction time of the jet with the AISI welded joint surface, a traverse speed of
v = 1, 2, 3, and 4 mm/s was used.
The surface residual stress (
Figure 2a) was measured using X-Ray Diffraction technique (XRD, PROTO Manufacturing Inc., Taylor, MI, USA). The conventional sin 2
Ψ method, where the d-spacing is given as a function of sin 2
Ψ based on an elliptical regression plot, was used to calculate the residual stress (XRD Win 2.0 software). The Mn-Kα X-ray tube of a 2 mm diameter was used for the measurements. The diffraction shift {311} was recorded at each point at rotation angles of 0°, 45°, and 90° with tilt phi angles at
Ψ = ±30°. The Bragg’s angle was measured as 152.80°. The measurements were conducted on the surface in the three different welded zones (heat affected zone (HAZ), weld, and base) marked in
Figure 2a.
The surface roughness of the treated region (
Figure 2b) was quantified using a contact surface roughness tester (Make: Mitutoyo, Kawasaki, Kanagawa, Japan). A stylus of 5 µm diameter was traversed with a cutoff length of
lc = 0.8 mm. The measured values (repeated five times) are plotted with mean and standard deviation. The roughness was measured only in the HAZ region.
The microhardness measurements were obtained from the sub-surface region of the polished samples (mirror-finished), starting from distance of 20 µm from the surface until a 1000 µm depth using the Vickers Hardness Tester at a load of 10 gf for a 10 s indentation time. The direction and the location of the measurements are shown in
Figure 2b.
The plastic deformation phenomenon in the treated samples was observed using optical microscopy (
Figure 2b). Before conducting the optical microscopy, a cross-section of the samples was polished and then etched using Adler reagent.
4. Conclusions
This study compared the effect of the excitation frequency on the surface integrity of AISI 304 TIG welded joints at varying nozzle traverse speeds. The effects were compared through residual stress, surface roughness, microhardness measurements, and optical analysis of the weldment. The obtained results can be summarized as follows:
The residual stress measurements showed that the initial condition of −122 MPa in HAZ improved to a maximum of −480 MPa at f = 40 kHz and of −319 MPa at f = 20 kHz at a lower traverse speed (v = 1 mm/s). It was also observed that the residual stress (compressive) distribution could be determined through the process parameters and material characteristics.
The surface roughness (Ra) value recorded in the untreated sample was 1.48 µm, which was increased to a maximum of 6.08 µm at the higher frequency of f = 40 kHz at v = 1 mm/s. This was attributed to the coverage time of the interacting jet.
The initial microhardness of the untreated welded joints was 318 HV, which increased to a maximum of 551 HV at f = 40 kHz. Moreover, the maximum increase in microhardness was observed at a lower traverse speed (v = 1 mm/s). A significant increase in hardening up to a depth of 100 µm was obtained at both frequency levels (f = 20 and 40 kHz).
The optical microscopy showed the presence of surface features such as slip bands and twin boundaries predominantly in the sub-surface region in comparison to the untreated samples. This indicates the intensity of the deformation under various experimental conditions. Additionally, the initial grain size of 38.40 µm was refined by 19.86% (30.77 µm) at f = 20 kHz and by 25.7% (28.53 µm) at f = 40 kHz.
The present study revealed the effect of the technological parameters, excitation frequency, and traverse speed of PWJ on TIG welded stainless steel surface. The enhancement in the mechanical properties shows the probability of utilizing the technology for the surface treatment application as it will be helpful in determining the appropriate parameters. However, for the surface treatment application the effect of variation of parameters should be explored in detail to ensure the surface integrity of the treated surface.