Since this study is mainly concerned with the focusing position of the laser irradiation, it is very important to know exactly where the tight focus position is. In this case, we use a pure copper plate (Rare Metallic, 5N, 1 cm diameter, and thickness of 0.4 mm) as a sample. We detect the emission spectra of typical Cu emission lines of Cu I 510.5 nm, Cu I 515.3 nm, and Cu I 521.8 nm as the laser energy decreases to as low as 0.5 mJ. By adjusting the y-direction of the x-y-z stage in which the sample is positioned, we determined the highest emission intensities of the above three Cu lines. This position is then marked as a tight focus position (zero position). It is important to note that shifting 0.1 mm away from the sample surface from the tight focus position yielding almost no emission of the above three Cu lines. From this result, we conclude that the precision of the focusing apparatus used in this study is 0.1 mm. We also noted by our naked eyes that the tiny green plasma of Cu disappears when the focusing position is shifted 0.1 mm from its tight focus position. This is mainly due to the very small energy of the laser irradiation used to determine the tight focus position yielding a very high accuracy of the tight focus position.
Figure 2 shows the emission spectrum of KCl at various focusing positions, including tight focus, +3 mm, -3 mm, -6 mm, -9 mm, and -12 mm. The negative (–) and positive (+) signs denote a position deviation from tight focus (zero) position in y-axis, respectively. The laser energy is fixed at 21 mJ, and the ICCD's gate delay and width are set to 1 µs and 30 µs, respectively. When laser irradiation is performed at a tight focus, a self-reversing emission spectrum of K I 766.4 nm and K I 769.9 nm is observed, and this effect is exacerbated when the focus position is set to +3 mm. When the focus position is shifted to the negative direction (shorter than the lens's focal length), self-reversal emission is significantly reduced. Negligible self-reversal of the K I 769.9 nm emission line is observed at -3 mm. Both K emission lines exhibit a free self-reversal emission line at –6 and –9 mm. Further defocus to –12 mm results in significant self-reversal emission of both K emission lines, as well as a significant reduction in emission intensity. We observed that the best emission spectra for K are observed at –6 mm and –9 mm defocus positions. A similar spectrum is obtained for a NaCl sample, as shown in Fig. 3. The difference between the NaCl and KCl spectra is minimal, with the best focusing position being only –6 mm. According to the results of Figure 2 and Fig. 3, the next experiment will focus exclusively on the –6 mm defocusing position
The reason why the self-reversal effect is greatly reduced in the negative defocus position could be that in the defocus condition, the larger beam waist and more flatter beam profile of the laser irradiation is responsible for the ablation of the sample yielding homogeneous plasma as opposed to tight focus laser irradiation with a much smaller beam waist on the sample surface. When there is too much negative defocus, such as – 12 mm, the fluence of the laser becomes too small, resulting in a serious self-reversal process. This explanation also applies to + 3 mm defocus because most of the laser energy is used to create air breakdown in front of the sample surface and only a small fraction of the laser energy is used to ablate the sample. Rezaei et al34 provide a comprehensive study on the relationship between plasma homogeneity and the self-reversal process. In this experiment, the beam waist is 25.4 µm for tight focus and 141 µm for defocus at -6 mm. Meanwhile, the fluence is 0.41 GJ/cm2 at tight focus and 0.013 GJ/cm2 for defocus at -6 mm.
The following experiment was carried out to determine the relationship between laser energy and defocus position that produces spectra without self-reversal. This is accomplished by varying the laser energy from 9 mJ, 21 mJ, 33 mJ, and 44 mJ, as shown in Fig. 4. As a sample, a KCl sample is used. Once again, the -6 mm defocus position provides the best spectral results without requiring a self-reversal process for the various energy lasers used. One might wonder why we only use laser energies ranging from 10 to 50 mJ. This is because we hope to use this technique for in-situ analysis soon, and in that case, a small portable Nd:YAG laser with a maximum energy of 50 mJ is easily available on the market.
Various pellet samples with different concentrations of KCl were prepared to determine the linearity of the calibration curve. As a KCl companion, NaCl is used. Fig. 5 shows the KCl calibration curve at 30%, 35%, 40%, 50%, and 55% concentrations which is corresponds to the K concentration of 16.6%, 18.3%, 21%, 26% and 29%. In Fig. 5, the measured K I 766.4 nm and K I 769.9 nm emission intensities are plotted against the associated K concentrations. Each data point in this graph is the average of five data points obtained from 20 successive laser irradiations. Over the dynamical range used in this study, the K concentration and its associated emission intensity show a clear linear relationship with a very high determination coefficient R2 of 0.99 for K I 769.9 nm and 0.98 for K I 766.4 nm. However, we discovered that the calibration curve is not intercepted at 0 points, which can be explained simply by the high concentration used to obtain the calibration curve. It should also be noted that the K concentration used to generate the calibration curve is extremely high, up to 29%. As a result, the likelihood of self-absorption in the plasma is extremely high. However, the presented calibration curve is still linear, demonstrating that our novel defocus laser irradiation technique is effective in completely suppressing self-absorption in LIBS.