**1. Introduction**

Laser Shock Processing (LSP) is based on focusing a pulse of a high energy laser (I > 10<sup>9</sup> <sup>W</sup>/cm2, τ < 50 ns) over a piece of metal. It results in the instantaneous vaporization of the piece's surface and the generation of a high temperature and density plasma composed of the di fferent ionized species of elements present in the piece and in the atmosphere. The high pressure of the plasma generates a shock wave that propagates to the piece, which a ffects its mechanical characteristics. Today, LSP is a consolidated alternative for improving the surface properties of metal alloys. This technique has been studied since the 1960s, when Askar'yan and Moroz [1] discovered that a high-energy laser pulse produces backpressures in the surface of the metal material on which the laser is focused (target). In Fairand et al. [2] it was found that laser shocking induced a tangled dislocation substructure similar to explosively shocked aluminum. The stress waves were studied using the piezoelectric response of X-cut quartz-crystal disks by Yang [3]. Since then, numerous works have aimed at finding the best conditions in which the technique should be applied. In 1990, Fabbro et al. [4] quantitatively described the evolution of the plasma in a LSP experiment under confined geometry using di fferent characteristic plasma-dynamics stages, including plasma pressure buildup, plasma development under the laser irradiation, and final plasma expansion until the pressure decreases and it is too low to cause plastic deformation in the material. An explanation of this phenomenon was given by Berthe et al. [5].

In 1997, Sano et al. [6] confirmed that the underwater shock processing with YAG was feasible to improve residual stress in metal. There are different studies about several methods to confine the plasma (see, e.g., Morales et al. 2009 [7]), indicating the crucial role of a plasma-confining layer in withstanding both the laser irradiation (with minimum internal breakdown) and the plasma pressure needed to transmit to the shocked target. The use of glass or quartz plates as confining layers are theoretically very suitable, but are generally only valid for a few laser shots and in measurements made in the laboratory, which are far from the needs imposed by industrial application of the technique, so purified water is normally used as the LSP confining medium.

The observation of the plasma shock wave through the confining medium has always been a subject of much interest in order to characterize the actual impulsion transmitted to the shocked material, a key point regarding the systematic monitoring and control of LSP processes in an industrial environment (see, e.g., Berthe et al. [5] and Ocaña et al [8]).

In 2009, Martí-Lopez et al. [9] performed interferometric measurements in order to record hemispherical shock fronts, cylindrical shock fronts, plane shock fronts, cavitation bubbles, and phase disturbance tracks. A summary of the different results obtained in the LSP experiments are presented by Ocaña et al. [8] along with some conclusions about the LSP technology as a profitable industrial method.

Recently, works on the effect of the LSP process have also been presented by some of the authors of this study. A comparison between experimental values and numerical predictions of the inside propagation of the residual stresses in plates of Al2024-T351 of 2 mm thickness was presented in 2015 by Ocaña et al. [10]. Also in 2015, a study on the influence of the randomness of the superposition of pulses on the final anisotropy of the residual stresses induced in the treated material was completed by Correa et al. [11].

Several theoretical works try to model the LSP process with the purpose of finding the practical criteria for its optimization. A self-closed thermal model for LSP was described by Wu and Shin [12]. Unlike most of the existing laser peening models, there are no free parameters in this model, and all the variables are calculated based on related physics theories. Later, a model-based systematization of process-optimization criteria and a practical assessment on the real possibilities of the technique along with practical results at a laboratory scale on the application of LSP to characteristic high-elastic limit metallic alloys were presented in 2008 by Morales et al. [13]. A calculation model conceived for the analysis of the problem of laser shock wave generation and propagation was presented in 2013 by Ocaña et al. [14]. However, these works were not able to predict the best conditions of application of the LSP technique with changes in the target materials. This is due in part to the large number of physical processes involved. As an example, although these models provide, as indicated, good predictions about the shock wave on the surface of the sample, their forecasts on the parameters of the plasma are too far from the experimental values, even when the plasma is confined in air. There is no experimental information on these properties in the case of plasma that is confined by water in flow.

The presence of hydrogen in the plasma is proof that chemical attacks on the surface of the sample can occur during the process. The interaction between the laser, the plasma, and the surface of the sample goes beyond the mechanical impact of the shock wave on the target. The relative presence of different ions in the plasma and its chemical interaction with the surface cannot be established without the knowledge of the plasma electron density number and its electronic temperature.

The purpose of this work was to experimentally estimate, by means of a practical (ideally industrially applicable) procedure, the electron density and the electronic temperature of the plasma in water-flow conditions to study chemical interactions between plasma and the surface of the sample and improve the theoretical treatments of the LSP process.

These studies of electron density number and plasma temperature determination have been performed for years using the technique known as Laser Induced Breakdown Spectroscopy (LIBS). The LIBS technique is based on the study of the emission of the different atomic and ionic species present in plasma. An excellent compilation of experiments and applications of the LIBS technique can be found in the literature (see, e.g., Musazzi and Perini [15]). This technique has been used in several scientific applications by the authors signing this work. In 2006, Colón and Alonso-Medina [16] measured Stark broadening of several Pb II spectral lines, and in 2011, Alonso-Medina [17] measured the broadening Stark of Pb III spectral lines. The LIBS technique also stands out as an analytical technique in different industrial applications (see, e.g., Noll et al. [18]).

In 2006, El Sherbini et al. [19] used the Hα-line to measure electron density in an experiment of the laser-induced breakdown of plasma in air. In 2010, Parigger and Oks [20] presented a review about plasma diagnostics based on Stark broadening of hydrogen Balmer lines in laser-induced breakdown of plasma. Later, in 2012, the Hα-line was used again by El Sherbini et al. [21] to determine the electron density in aluminum plasma and to correct the self-absorption in the Mg I and Mg II spectral lines present in this plasma. In 2004, De Giacomo et al. [22] completed several experiments of laser induced breakdown spectroscopy in aqueous solution. Nath and Khare [23] studied laser-induced breakdown experiments in water in 2010 using the emission bands of different molecular species.

As mentioned in a previous work by Moreno-Díaz et al. [24], the flow of water in LSP conditions weakens the emission of all species present in the plasma, with some exceptions. In the present work, in addition to the emission of hydrogen, already mentioned above, a weak emission of Mg II in the 279.5 nm zone was targeted. As in the previous work, the electron density was measured using the Stark width of the Hα-line. Now, the temperature can be estimated directly under LSP conditions using the emission of the spectral lines of Mg II.

The current work used the Stark width of the Hα-line on a sample of aluminum alloy (Al2024-T351) in LSP conditions (with water as the confining environment). The novelty of this study is in the measurement of the electronic temperature, which was performed directly in the LSP experiments using a Boltzmann plot with measurements of the intensities of the spectral lines of species present in the plasma in the water flow, not in air as was used in the earlier work of Moreno-Diaz et al. [24]. Measurements were taken with different delay times from the laser pulse (2–5 μs), while the plasma cooled adiabatically, allowing us to obtain the width and shift of the Hα-line in all cases. In order to obtain the best conditions for measurements, different gate times of 100 ns, 200 ns, 300 ns, 500 ns, and 1000 ns were used.

In order to verify the suitability of the proposed one-determination diagnosis procedure, a comparison of the obtained temperatures with those estimated from the shift of the Hα-line (two determinations needed) was successfully performed. This diagnosis with a single experiment would allow for the much-needed real-time monitoring of plasma behavior as described in Ocaña et al. [25] and Takata et al. [26].

In this paper, in Section 2 (Materials and Methods) the equipment and the experimental procedures are described. In Section 3 (Results) the analysis of the results is presented by comparing the values of the plasma temperature achieved by two different methods. Finally, in the Sections 4 and 5 (Discussions and Conclusions, respectively) we present the possibilities of plasma diagnosis through a single experiment and through a second experiment without using the Mg II lines.
