Multiple-pulse Laser-induced breakdown spectroscopy for monitoring the femtosecond laser micromachining process of glass

Optics and Laser Technology 111 (2019) 295–302

Contents lists available at ScienceDirect

Optics and Laser Technology journal homepage: www.elsevier.com/locate/optlastec

Full length article

Multiple-pulse Laser-induced breakdown spectroscopy for monitoring the femtosecond laser micromachining process of glass

T

⁎

Julius Skruibis, Ona Balachninaite , Simas Butkus, Virgilijus Vaicaitis, Valdas Sirutkaitis Vilnius University, Faculty of Physics, Laser Research Center, Saulėtekio Ave. 10, Vilnius 10223, Lithuania

H I GH L IG H T S

breakdown spectroscopy applied for laser micromachining monitoring. • Laser-induced schema allows burst micromachining and process monitoring simultaneously. • Proposed • Multi-pulse excitation improves the analytical performance of laser spectroscopy.

A R T I C LE I N FO

A B S T R A C T

Keywords: Femtosecond laser micromachining Multi-pulse laser-induced breakdown spectroscopy (MP-LIBS) Filament processing Glass cutting Process monitoring

Laser-induced breakdown spectroscopy (LIBS) can be applied for laser micromachining not only as a tool for determining the chemical composition of the sample but also for monitoring the process itself. Multiple pulses in comparison to a single pulse improve the analytical performance of LIBS, enable special applications and increase the ablation rate in laser material processing. In this report, we present results on the application of multiple-pulse LIBS in monitoring the micromachining of soda-lime glass with femtosecond high repetition rate pulses.

1. Introduction Ultrafast laser micromachining is an advanced technology for highprecision and quality material microprocessing. This procedure provides many advantages over other conventional processes [1]. The threshold fluency of damage and ablation is orders of magnitude less than for traditional nanosecond laser machining. Compared with longpulse lasers, femtosecond laser pulses allow for extremely high peak power densities to be achieved with low pulse energies. The short pulse duration enables strong multiphoton and electron avalanche mechanisms that provide benefits such as a small heat-affected zone [2], settled threshold for precision microprocessing [3,4], and minimized mechanical and thermal damage [5]. At high power densities, nonlinear absorption resulting from multiphoton ionization enables excitation of electrons in high bandgap materials such as ceramics and glasses, which are hard to process with conventional industrial lasers [6,7]. Thus, ultrafast lasers are widely applied in micromachining dielectrics [8], glasses [9,10], forming optical circuits in glasses [11], and filament scribing of glass panels [12]. While the reduced heat-affected zone is a major advantage of the ultrafast laser-material interaction, the rapid energy dissipation can

⁎

cause a strong shock and a quick heating and cooling cycle. It can produce such harmful effects as microexplosions [13] and microcracks [14]. One promising method to reduce these effects is to use a burst of laser pulses at high repetition rate where residual thermal energy does not diffuse out of the laser interaction zone before the next laser pulse arrives. In this case, a thermal modification zone can be built to temporally modify the material property in this focal zone and redirect the overall laser interaction to provide a more beneficial outcome from the burst laser pulses [15]. Recently, high average power femtosecond laser systems have become more applicable due to an increase in processing throughput, however, maintaining the efficiency of the process comparable to lower power systems without introducing additional thermal damage is a great challenge [16–17]. A new technique – burst mode processing was demonstrated for micromachining materials with better efficiency than the conventional systems [18–20]. By using this technique the single pulse is separated into multiple pulses (burst) by a certain time delay (varying from ∼100 ps to ∼10 ns). In this case, the relatively large pulse energy can be divided into a sequence of pulses with pulse energy values that are closer to the optimal ablation rate setting while the integrated energy of the burst remains high.

Corresponding author. E-mail address: ona.balachninaite@ff.vu.lt (O. Balachninaite).

https://doi.org/10.1016/j.optlastec.2018.10.005 Received 27 March 2018; Received in revised form 24 August 2018; Accepted 4 October 2018 0030-3992/ © 2018 Elsevier Ltd. All rights reserved.

Optics and Laser Technology 111 (2019) 295–302

J. Skruibis et al.

In this paper, we present femtosecond laser micro-machining of soda-lime glass with bursts of multiple pulses. In our experiment we formed a sequence of multiple femtosecond laser pulses by specially constructed optical schema. At the same time for online monitoring of the process, multiple pulse laser-induced breakdown spectroscopy (MPLIBS) was used. LIBS technology makes possible the extended capability for monitoring laser processes and getting actual information back to the laser manufacturing process. This allows online process parameter supervision for laser processing [21–28]. During laser-induced material breakdown, the intense laser pulses initiate an expanding plasma plume, which emits light from the atomic, ionized and molecular constituents of the sample [29]. The spectra of plasma light emissions can provide analytical information about the elemental composition of the material being ablated [29–31]. LIBS has several attractive features for sample analysis: an easy, fast, real-time and in-situ measurement due to minimal sample preparation, and multiple element detection capability regardless of the physical form and aggregation state of the material. This technique is suitable for a wide range of applications such as: depth profiling, surgical selective tissue removal, authentication of artworks, monitoring and control of laser material processing etc. [21–28,32–34]. Using dual-pulse LIBS, atomic emission intensities, signal-to-noise ratio, precision, and detection limits can be improved [35]. Dual-pulse LIBS configuration consists of a sequence of two laser pulses temporally spaced in the order of hundreds of picoseconds or microseconds (depending on the laser pulse duration). These two pulses can ablate the same area and create two temporally spaced plasmas, or the second pulse can reheat the plasma induced by the first pulse. The dominating mechanisms of double pulse intensity enhancement in femtosecond laser-induced breakdown spectroscopy (fs-LIBS) according Ref. [36] are the increased number of atoms in plasma due to the larger volume of double pulse ablation craters and the higher plasma temperature. The atomization of nanoparticles by laser plasma-particle interaction contributes much less. The intensity enhancement is highest at lowest energy of the first pulse. Fs double pulse LIBS may thus enable elemental imaging with improved lateral resolution, higher sensitivity and increased elemental contrast. Another approach is multi-pulse LIBS (MP-LIBS) [37–38]. With this method, more than two pulses are used in a collinear mode. MP-LIBS approaches demonstrated by several research groups [39–41] showed that it is analytically advantageous to use more than two laser pulses in LIBS, as they provide increased material ablation and enhanced signal emission [42–44]. LIBS has been used in monitoring and control of laser material processing mainly by using nanosecond pulses [21–28]. It's worth mentioning the T. Sibillano's sensor, based on LIPS technology designed for monitoring and controlling the laser welding process [25]. T. Tong adapted the LIPS technology to control laser micro-processing of the multilayer elements by recording and comparing spectra [23]. D.D. Vallejo applied LIPS technology to observe and evaluate the focal plane position of the specimen and identify different layers of material [26–28]. However, these authors used nanosecond or two-pulse systems, and we decided to explore the possibilities of using multiple femtosecond pulse systems. LIBS systems of the ultrashort pulses are distinguished by their stability and precision, due to the specific shortpulse interactions with the material. With the help of multiple pulse LIBS systems, the LIBS signal is amplified, and the processes efficiency and quality are enhanced as well. The aim of our present work was to apply femtosecond MP-LIBS in monitoring the micromachining of soda-lime glass by a burst of femtosecond high repetition-rate pulses and to investigate the main influences of different processing parameters.

Fig. 1. Experimental setup of the multi-pulse LIBS-monitored femtosecond laser processing system. Optical components include beamsplitter (BS), mirror (M), coupling optics. TS-translation stage.

Fig. 1. The experiments were performed using the Carbide® Yb:KGW femtosecond laser system (Light Conversion Ltd) operating at 1030 nm (pulse width 280 fs, average power up to 5 W, 60 kHz). The maximum laser energy of a single femtosecond pulse was approximately 80 μJ. The burst of multiple pulses was formed by dividing the beam with the beam splitter BS1, which transmitted 70% and reflected 30% of the incident light. The transmitted part of the laser radiation, undergoing multiple reflections by high reflective (99.95%) dielectric multi-layer mirrors M1-M3, formed a sequence of endless multiple pulses with decreasing energy (Fig. 2). Thus, if the energy of the primary single pulse was 80 μJ, the energies of the multiple pulses in the sequence were ∼24 μJ; 39.2 μJ; 11.8 μJ; 3.5 μJ etc., respectively. Optical delay line consisted of M1-M3 mirrors was used to form the delayed pulses. By choosing the length of the delay line equal to 15 cm, the 500 ps delay time was formed and kept constant during the whole experiment. This value for the delay between the pulses was chosen because the range of the delay time 200–800 ps has been determined to be optimal for obtaining high-intensity LIBS signals using double-pulse femtosecond excitation [45]. The two axis galvanometric scanners (ScanLab Inc.), controlled by SCA fabrication software (Altechna Ltd), were used for

2. Experimental setup Fig. 2. Single-pulse (up) and multiple-pulse (burst- down) regimes used in the experiment.

A schematic representation of the experimental setup is shown in 296

Optics and Laser Technology 111 (2019) 295–302

J. Skruibis et al.

12000

fast and precise positioning of mirrors to deflect laser beams. After the laser beam passed through the scanner, it was focused by an F-Theta lens (f = 75 mm) in the normal incidence onto the sample (focal spot diameter ∼20 μm), which was mounted on a three-dimensional motorized positioning stage (Standa Inc.) for precise positioning of the sample at the start. Galvanometric scanners in combination with F- theta lenses can produce large scanning fields and scanning rates of the order of m/s or greater. Micromachining was carried out on sodalime glass of 1-millimeter thickness (Thermo Scientific Inc.). The sample was processed under a thin (∼0.8 mm thick) water layer and without water. The particular water layer thickness was chosen based on our previous experiments on glass cutting with high repetition-rate femtosecond laser pulses using a similar experimental setup [46]. Tap water was used for the experiments, and the desired thickness of the layer was formed by submerging the glass sample in a reservoir. The thickness of the water layer was kept constant. By applying a thin water layer on top of the glass sample and using a low numerical aperture (NA) objective, filaments can be created within the water layer and can be used for micromachining purposes. This results in high micromachining quality due to the spatial shaping of ultrashort pulses, and the cooling and cleaning features of the covering water [47]. The plasma emission during laser ablation, which occurs above the sample surface at ambient temperature and pressure, is initially collected by the F-Theta lens, then passes through the galvanometer scanner and goes to the beam splitter BS2, which has a high reflection coefficient for 1030 nm but is transparent for wide spectra (400–700 nm) LIBS radiation. In our case, we mainly used a 500–700 nm spectral range for LIBS investigations. After the beam splitter BS2, the transmitted LIBS radiation was collected by the coupling lenses (f = −50 mm; f = +100 mm and f = +50 mm, respectively) and directed to the optical fiber (with NA of 0.22) that was connected to the multichannel spectrometer (AvaSpec USB2-DT, Avantes Inc.). The plasma was imaged on the focal spot of the coupling lens and nearly the entire plasma was coupled to the entrance of the optical fiber. The position of the fiber and the lens was optimized and kept constant during the whole experiment. The spectrometer was synchronized with the laser system. The integration time was 100 ms. In all cases, mainly one channel of the spectrometer in the spectral range 550–660 nm was used for our experiments, since few strongest lines of Na I which manifest in soda-lime glass are located near 589 nm. This channel has a 2048-element CCD array and provides an optical resolution equal to ∼ 0.1 nm. It is known that in LIBS measurements the collection geometry is very important, and that the plasma expands mostly normal to the sample surface. The on-axis plasma light collection is less sensitive to any changes in the distance between the plasma plume and the collecting F-Theta lens that occur when a groove is processed [48]. In our experiment, the plasma radiation collection angle was chosen to be perpendicular to the sample surface. Various scanning speeds were employed for laser cutting and online LIBS signal observation. A scanning algorithm was chosen a “snake” form. The beam was scanned in straight lines: 8 mm positive y axis direction, then 0.5 mm positive x axis direction, again 8 mm y axis just negative direction, and further 0.5 mm positive x axis direction. This path was repeated for 8 times. Microgroove geometry was selected because of the convenience in measuring the depth and width of the groove. The separation between the lines was 0.5 mm. For a scanning speed of 100 mm/s, the time interval between LIBS measurements was equal to 1.36 s, for a scanning speed of 200 mm/s was equal to 0.68 s, and for a scanning speed of 300 mm/s was equal to 0.45 s. Laser microprocessing and corresponding LIBS signal registration were performed by using single-pulse and multi-pulse excitation. The total pulse energy was set constant and equal to ∼80 μJ for both configurations.

Intensity, a.u.

10000

Na D (589 nm)

8000 6000 4000 2000 0 540

560

580

600

620

640

660

680

Wavelength, nm

(a)

(b) Fig. 3. Laser-induced breakdown spectra of the soda-lime glass (a) and LIBS spectres of soda-lime glass at different distances to the nominal focus positions (b).

3. Results and discussions The spectra of plasma excited by single and multiple femtosecond laser pulses were analyzed in order to observe and control laser microprocessing of a transparent media. By varying the parameters of the process, their influences on the strength of LIBS signal were monitored. Experiments were performed while the sample was under water and in a dry environment. Processing quality, speed and other characteristics were also evaluated. 3.1. Focus positioning Before starting the microprocessing, it is important to determine the position of the sample in relation to the focusing lens, where the most effective material ablation is observed. For this purpose, the sample placed on the motorized (z-axis) table was moved to the lowest position. Plasma was excited by single pulse. The plasma spectrum was recorded after a single scanning (forming one microgroove on the surface of the sample), after which the sample was lifted upwards by 25 μm and again the plasma was induced and recorded on the fresh surface of the sample (Fig. 3(a)). In this way, the position of the sample in relation to the focusing lens was determined when the intensity of the Na I 589 nm line (one of the Na D lines) in the recorded plasma spectrum was the greatest. The Na I emission line (589 nm) (one of the Na D lines) in the spectrum of soda-lime glass (Fig. 3 (b)) was monitored for different 297

Optics and Laser Technology 111 (2019) 295–302

J. Skruibis et al.

Scanning speed 100 mm/s single pulse multiple pulses 0,6

0,5

Intensity, a.u.

Intensity, arb. units

1,0

0,0 -200

-100

0

100

200

Distance to nominal focus position, μm

0,4

0,2

Fig. 4. Intensity of Na I emission line (589 nm) in dependence on the sample distance to the nominal focus position. 0,0

focusing distances. This experimental part was performed by using a single-pulse excitation regime. The intensity of Na I emission line (589 nm) dependence on the sample position is shown in Fig. 4. Each data point shown in Fig. 4 is an average value obtained from three separate measurements. The maximum LIBS intensity was reached when the nominal focus position was on the sample surface. Microgrooves formed on the sample surface were analyzed with a microscope. The microgrooves formed when the signal of the plasma was the greatest are seen to be the deepest and narrowest, which corresponds to the highest volume of removed material. In this way, the LIBS signal can be used to determine the optimal distance between the sample and the focusing lens. Since the focusing F-Theta lens was not changed during the experiment, the sample position was kept the same during all the measurements.

0

5

10

15

Number of scans Scanning speed 200 mm/s single pulse multiple pulses

1,0

Intensity, a. u.

0,8

0,6

0,4

3.2. The dependence of the LIBS signal on the number of scans at different scanning speeds by performing micro processing on the dry glass sample

0,2

While using the above-mentioned scanning algorithm, the LIBS signal was recorded at different scanning speeds. Each point on the graph (Fig. 5) is the average value of the three corresponding measurements. Repeated measurements under the same conditions were performed on a new surface of the sample. The LIBS signal measurement was synchronized with the start point of the scan and was measured before the scan was completed. This methodology for recording the LIBS spectrum has been used in all experiments. The variation of the intensity of Na I emission line (589 nm) induced in glass depending on the number of scans at different scan rates is shown in Fig. 5. The intensities were all normalized to the maximum value of the Na I emission line signal. These scans were performed on a dry glass sample. In all cases, it can be seen that the signal using multiple pulses is about 2.5 times stronger than in the case of single-pulse excitation. This can be explained by the fact that plasma heating by the second and subsequent pulses takes place, which excites (or re-excites) particles generated by the first pulse and entrained in the plasma [49]. It can also be seen from these graphs that, for multiple pulses, plasma radiation is observed for a longer period of time. It can be assumed that conditions for the ablation of the material are maintained longer. Fig. 6 shows the variation in the intensity of the plasma depending on the number of scans at different scanning speeds for single and multiple-pulse excitations. At single pulse and at the lowest speed (100 mm/s), the Na I spectral line intensity increases sharply and then decreases steadily, and at higher velocities (300 mm/s and 200 mm/s), the intensity increases at the beginning and then decreases slowly. Such a change in the plasma signal can be explained by the different number of laser pulses per unit area on the sample surface. For slower scanning speeds, there are many pulse overlaps for each ablation spot, so the material is removed faster and the plasma radiation declines faster. At a scanning speed of 100 mm/s and pulse repetition rate

0,0 0

5

10

15

20

25

30

Number of scans

Scanning speed 300 mm/s 0,8

single pulse multiple pulses

Intensity, a. u.

0,6

0,4

0,2

0,0

0

5

10

15

20

25

30

Number of scans Fig. 5. Evolution of Na I emission line (589 nm) intensity with the number of scans at single and multi-pulse excitation for different scanning speeds (100 mm/s; 200 mm/s; 300 mm/s). Total pulse energy is 80 μJ; pulse repetition rate 60 kHz. The sample is dry.

298

Optics and Laser Technology 111 (2019) 295–302

J. Skruibis et al.

Single pulse 0,2

Intensity, a. u.

3.3. The dependence of the LIBS signal induced by single and multiple pulses on the number of scans at different scanning speeds, while performing microprocessing on the glass sample immersed in water

100 mm/s 200 mm/s 300 mm/s

Another part of the experimental work was laser micro-processing of glass plates immersed in water and analysis of the corresponding LIBS signal. The purpose of this experiment was to investigate the effect of a thin (∼0.8 mm) water layer on the surface of a glass plate on the microprocessing process, by analyzing the corresponding plasma signal variations. The variation of the intensity of the Na I spectra line induced in glass, depending on the number of scans at different scan rates, is shown in Fig. 6. The emission intensity is slightly lower than the intensity of plasma induced on the surface of a dry sample. This difference can be explained by the absorption of laser radiation in water. Also, the emission of plasma declines faster in a water than in a dry environment. This is caused by the fact that the ablation in water is faster and the maximum depth of the grooves is reached within a shorter time. The water layer on the surface of the treated glass is used as a buffering layer in which the filaments appear. Apart from reducing the thermal effects and removing the sediments from the surface, the use of this water layer increases the rate of ablation caused by laser-induced plasma confinement via the water layer, which results in better plasmasample coupling [51,52] and the interaction of the plasma with the glass is being prolonged. For the multi-pulse excitation regime, the plasma radiation time is longer compared to the single-pulse regime, allowing longer and more accurate monitoring of the process (Fig. 6).

0,1

0,0

0

5

10

15

20

25

30

Number of scans

Multiple pulses 1,0

100 mm/s 200 mm/s 300 mm/s

Intensity, a. u.

0,8 0,6 0,4 0,2 0,0 0

5

10

15

20

25

30

35

Number of scans Fig. 6. Evolution of Na I emission line intensity with the number of scans at three scanning speeds (100 mm/s; 200 mm/s; 300 mm/s) for single and multipulse excitation. Total pulse energy is 80 μJ; pulse repetition rate 60 kHz. The sample is immersed in water.

3.4. Comparison of microprocessing in water and in a dry environment From Fig. 7, it can be seen that the formation of grooves in water with identical parameters to those formed without water, gives a more regular groove structure. The surrounding surface is less damaged; the structures formed are less polluted by sediments from the process. Comparing the variation in LIB signal intensity when the sample is under water and without water (Fig. 5 and Fig. 6), no significant change is observed with single-pulse configuration. At 100 mm/s scanning speed and using a single-pulse configuration, LIBS signal curves depending on the number of scans for both the dry sample and immersed in water, are very similar. At low scanning speeds, a large number of pulses per unit area can cause the water to evaporate from the surface of the sample [53], and ablation takes place in the same way as in a dry environment. Meanwhile, using multi-pulse configuration, the energy of individual pulses is lower, so it is likely that the water layer is not evaporated and the difference between the curves is more noticeable than by using single pulses. For plasma excitation by multiple pulses, the LIBS intensity decreases faster in water than on the dry sample, possibly because the signal is quenched by the water.

60 kHz, at the spot equal to the focused beam diameter 20 μm there is an overlap of 12 pulses, at a scanning speed of 200 mm/s there is an overlap of 6 pulses, and at a scanning speed of 300 mm/s there is an overlap of only 4 pulses. With a higher scanning speed, when the number of pulses per unit area of the sample is smaller, the ablation time for creation of the similar number of atoms in plasma is longer. In the case of single-pulse excitation (Fig. 5), there is a noticeable increase in signal strength with the first scans of the sample. At a higher scan speed, the LIBS signal grows to its maximum value just after eight scans, while at 100 mm/s scanning speed, the LIBS signal reaches the maximum value and starts to decrease after three scans. Such a variation in the plasma signal can be explained by the interaction between the laser radiation and the glass sample. After the first few tens of laser pulses have reached the sample, the glass is almost transparent and minimal ablation of the material is observed. Each subsequent pulse from the sequence arrives to a slightly modified material as compared to the previous one. Thus, a gradual accumulation of the various intrinsic defect states in transparent glass with larger excitation crosssections makes excitation easier for the subsequent pulses [50]. The more pulses reach the glass sample, the sooner nonlinear absorption takes place in the glass, leading to stronger ablation. Thus, in order to achieve the ablation process, a certain number of “preparation” pulses are required. At a lower scanning speed, this number of pulses per unit area is achieved faster, which results in a lower number of scans needed to reach the maximum value of plasma intensity. It is noteworthy that when using a multi-pulse sequence for excitation of plasma, the LIB signal increases slightly after the first scans (Fig. 5), and after 4–5 scans the ablation mode starts.

Fig. 7. The microgrooves formed on a dry sample (left) and when the sample was under a thin water layer (right). Scanning speed 200 mm/s; 60 kHz repetition rate. Separation between the grooves is 500 μm. 299

Optics and Laser Technology 111 (2019) 295–302

J. Skruibis et al. 1,0 0,8

Intensity, a. u.

reheating effect. By optimising the delay time between the pulses and the energy distribution within the burst, even higher enhancement factors could be achieved.

single pulse multiple pulses

0,6

3.6. Analysis of the depths of the processed grooves

0,4

Several tests with different scanning parameters were performed to compare the groove depths formed by single and multiple laser pulses. Scanning speed and number of scans were changed. All tests were repeated with single and multiple pulse configurations. Groove depths were analyzed with a bright-field microscope. In order to evaluate the uncertainty in depth measurement, the depth was measured at three different cross-sections along the length of the groove. As already mentioned, the intensity of the LIBS signal depends on the scanning speed due to the different number of pulses per unit area. It can be assumed that the depth of the formed grooves will maintain the same dependence on the scanning speed. Groove depth in relation to the speed of the scans was analyzed when grooves are formed by multiple and single pulses underwater. In all cases, a higher groove depth is recorded at a lower scanning speed (100 mm/s). A similar trend is observed for the formation of grooves without a water layer. However, no significant difference was observed in the depths of the grooves formed by single and multiple pulses (Fig. 9). Thus, the enhancement of the femtosecond MP-LIBS signal produced by a sequence of pulses is caused by excitation of the particles entrained in the plasma produced by the first pulse rather than by increased ablation of the surface by the second pulse. It corresponds to the second mechanism of signal enhancement in double pulse LIBS to plasma heating according Ref. [36].

0,2 0,0 0

10

20

30

40

50

60

70

80

90

70

80

90

Energy, μJ/pulse

Enhancement factor IMP/ISP

a) 2,5

2,0

1,5

1,0

0

10

20

30

40

50

60

Energy, μJ/pulse

b) Fig. 8. (a) The LIBS signal of Na I (589 nm) spectral line as a function of the laser pulse energy in both multi-pulse and single pulse configurations. (b) Multiple pulse LIBS: Enhancement factor IMP/ISP. Laser pulse repetition rate 60 kHz; Scanning speed 200 mm/s; signal registered after 5 scans.

4. Conclusions In this work, a sequence of multiple femtosecond laser pulses (burst train) could be obtained in the proposed MP-LIBS technology by splitting a single beam generated by one laser. Thus, the delay time between the multiple laser pulses reaching the sample surface was determined only by the optical path delay. We presented a constructed MP-LIBS system as a monitoring tool for femtosecond micro-machining of sodalime glass immersed in water and without water at high repetition-rate pulses. The method of determining the focal position was investigated by measuring the intensity of laser-induced plasma. Using this method, an optimum sample position can be evaluated in relation to the focusing element or its focal plane. Determination of LIBS signal dependence on focal position performed in these experiments equips laser micromachining with the ability to do online corrections of focal position empowering continuous micromachining of particularly complicated surface shapes. During microprocessing with multiple (burst train) laser pulses

3.5. Comparison of the LIBS signal induced by single and multiple laser pulses depending on the laser pulse energy The evolution of the plasma induced by single and multiple laser pulses depending on the laser energy per pulse was evaluated. The total pulse energy was changed in the range from 15 μJ to 80 μJ and the respective LIBS signals were registered. The scanning speed was kept constant and equal to 200 mm/s. The measurements were performed on the dry sample surface after five scans. By single-pulse excitation the LIBS intensity increases linearly whereas in the case of multi-pulse excitation some nonlinearity in the increase of the LIBS signal is observed (Fig. 8(a)). By increasing the pulse energy up to 80 μJ, the LIBS signal induced by multiple pulses is more intensive than in the case of single-pulse excitation. The nonlinearity observed in the multiple-pulse LIBS regime depending on the laser pulse energy could be explained by the fact that by increasing laser pulse energy more pulses from the endless sequence of the burst train contribute to the formation of the plasma, whereas at the low laser pulse energy only the first pulses are involved in the process. Investigations at higher excitation pulse energies are needed to ground these arguments. The MP-LIBS spectrum shows a very clear enhancement of emission intensity for all lines compared to the spectra obtained with SP of equal energy. Enhancement of a spectral line (IMP/ ISP) was defined as the ratio of the MP over the SP-LIBS line intensity at the same total energy. This ratio directly indicates how many times the LIBS signal induced by multiple-pulse is higher than the LIBS signal induced by using single pulses. At maximum 80 μJ pulse energy more than ∼2.5 times signal enhancement was observed (Fig. 8(b)). The enhancement factor increases with pulse energy, implying that more energy is coupled to the plume produced by the first pulse for the

40

Single pulse Multi-pulse

Groove depth, μm

35 30 25 20 15 10 5 0

2

4

6

8

10

12

Number of scans Fig. 9. Crater depth as a function of the number of scans for both SP and MP configurations. 300

Optics and Laser Technology 111 (2019) 295–302

J. Skruibis et al.

(with total pulse energy ∼80 μJ), at different scanning speeds, in water and on a dry surface of the glass sample, the MP-LIB signal was always stronger and observed for a longer period of time than by using singlepulse excitation with the same pulse energy. By using SP excitation and increasing laser energy, the intensity of the LIB signal increases linearly. In the case of multiple-pulse excitation, a nonlinear growth of LIBS intensity is observed. This can be explained by the fact that, with increasing excitation laser pulse energy in the sequence of decreasing energy pulses, new and additional pulses are contributing to plasma excitation. The maximum increase in MP-LIBS signal intensity (for Na I spectral line at 589 nm) compared to excitation with single pulses is ∼2.5 times (at 80 μJ pulse energy). However, no significant difference was observed in the depths of the grooves formed by single and multiple pulses. Soda-lime glass was chosen as the target for several reasons. Soda lime glass targets allows us to quantitatively investigate properties of transparent materials in laser micromachining. After evaluating the results, any determined micromachining parameters could be applied for different types of transparent materials with minimal corrections. The results achieved with soda lime glass were briefly compared to the results achieved with copper and steel. Primary evaluations indicate that the enhancement factor of LIBS signal for the mentioned materials keeps up the same trend. Detailed investigations have to be fulfilled to confirm multiple pulse LIBS applications for different materials and will be discussed in future publications. Determination of LIBS signal dependence on focal position performed in these experiments equips laser micromachining with the ability to do online corrections of focal position empowering continuous micromachining of particularly complicated surface shapes. In conclusion, the present experimental set-up can be used not only for monitoring the laser microprocessing of materials, but also for controlling the cutting of layered heterogeneous materials, thin film scribing, thin layer removing and for sensitive material analysis using MP-LIBS methodology.

Berlin Heidelberg, 2012. [12] S.A. Hosseini P.R. Herman, Method of material processing by laser filamentation, EP patent, EP 2593266 A2 (May 23, 2013). [13] E.N. Glezer, E. Mazur, Ultrafast-laser driven micro-explosions in transparent materials, App. Phys. Lett. 71 (1997) 882–884. [14] P.R. Herman, M. Lapczyna, D. Breitling, H. Schittenhelm, H.W. Tan, Y. Kerachian, R.S. Marjoribanks, Conference on Lasers and Electro-Optics (2000), 580. [15] P.R. Herman, R. Marjoribanks, and A. Oettl, Burst-ultrafast laser machining method, U.S. patents, 6 552 (301) (July 26, 2001). [16] P. Russbueldt, T. Mans, M. Hermans, D. Hoffmann, D. Wortmann, Powerscaling of ultrafast lasers, Laser Technik J. 7 (2) (2010) 41–44. [17] P. Russbueldt, H.-D. Hoffmann, T. Mans, R. Poprawe, 1100 W Yb:YAG femtosecond Innoslab amplifier, Proc. SPIE 7912, 2011, p. 79120R (9 S.). [18] C. Kerse, H. Kalaycıoglu, P. Elahi, B. Cetin, D.K. Kesim, O. Akcaalan, S. Yavas, M.D. Asik, B. Oktem, H. Hoogland, R. Holzwarth, F.O. Ilday, Ablation-cooled material removal with ultrafast bursts of pulses, Nature 537 (2016) 84–88. [19] T. Kramer, B. Neuenschwander, B. Jaggi, S. Remund, U. Hunziker, J. Zurcher, Laser assisted net shape engineering, 9 international conference on photonic technologies, Proceedings of the Lane 2016, 2016, pp. 123–134. [20] C. Gaudiuso, H. Kämmer, F. Dreisow, A. Ancona, A. Tünnermann, S. Nolte, Ablation of silicon with bursts of femtosecond laser pulses, Proc. SPIE 9740, San Francisco, 2016, p. 974017. [21] I. Gobernado-Mitre, A.C. Prieto, V. Zafiropulos, Y. Spetsidou, C. Fotakis, On-line monitoring of laser cleaning of limestone by laser-induced breakdown spectroscopy and laser-induced fluorescence, Appl. Spectrosc. 51 (1997) 1125–1129. [22] T.J. Li, Q.H. Lou, Y.R. Wei, F. Huang, J.X. Dong, J.R. Liu, Laser-induced breakdown spectroscopy for on-line control of selective removal of cobalt binder from tungsten carbide hard metal by pulsed UV laser surface ablation, Appl. Surf. Sci. 181 (2001) 225. [23] T. Tong, J. Li, J.P. Longtin, Real-time control of ultrafast laser micromachining by laser-induced breakdown spectroscopy, Appl. Opt. 43 (9) (2004) 1971–1980. [24] J.O. Connolly, G.J. Beirne, G.M. O’Connor, T.J. Glyn, A.J. Conneely, Optical monitoring of laser-generated plasma during laser welding, Proc. of SPIE 3935 (2000) 132. [25] T. Sibillano, A. Ancona, V. Berardi, P.M. Lugara, A real-time spectroscopic sensor for monitoring laser welding processes, Sensors 9 (2009) 3376–3385. [26] D. Diego-Vallejo, M. Schwagmeier, D. Ashkenasi, G. Illing, H.J. Eichler, Laser-induced plasma investigations during material processing, J. Laser Mi-cro/Nanoeng. 6 (2) (2011) 146–150. [27] D. Diego-Vallejo, D. Ashkenasi, A. Lemke, H.J. Eichler, Selective ablation of CopperIndium-Diselenide solar cells monitored by laser-induced breakdown spectroscopy and classification methods, Spectrochim. Acta B 87 (2013) 92–99. [28] D. Diego-Vallejo, D. Ashkenasi, H.J. Eichler, Lasers in manufacturing WLT, Munich, Conf. Proc. on Phys. Procedia 41 (2013) 904. [29] D.A. Cremers, L.J. Radziemski, J. Wiley, Handbook of Laser-Induced Breakdown Spectroscopy, John Wiley & Sons, New York, 2006. [30] J.M. Vadillo, J.J. Laserna, Laser-induced plasma spectrometry: Truly a surface analytical tool, Spectrochim. Acta B 59 (2004) 147–161. [31] T.A. Labutin, A.M. Popov, V.N. Lednev, N.B. Zorov, Correlation between properties of a solid sample and laser-induced plasma parameters, Spectrochim. Acta B 64 (2009) 938–949. [32] E. Tzamali, D. Anglos, Lasers in the Conservation of Artworks, Springer, Heidelberg, 2007. [33] T.V. Kononenko, D. Walter, V.I. Konov, F. Dausinger, Optical spectroscopy of laser plasma in a deep crater, Quantum Electron. 39 (2009) 328–332. [34] H. Balzer, M. Hoehne, R. Noll, V. Sturm, New approach to online monitoring of the Al depth profile of the hot-dip galvanised sheet steel using LIBS, Anal. Bio-anal. Chem. 385 (2006) 225–233. [35] D.A. Cremers, L.J. Radziemski, T.R. Loree, Spectrochemical analysis of liquids using the laser spark, Appl. Spectrosc. 38 (5) (1984) 721–729. [36] C.M. Ahamer, J.D. Pedarnig, Femtosecond double pulse laser-induced breakdown spectroscopy: investigation of the intensity enhancement, Spectrochimica Acta Part B: Atomic Spectroscopy 148 (2018) 23–30. [37] E.H. Piepmeier, H.V. Malmstadt, Q-switched laser energy absorption in the plume of an aluminum alloy, Anal. Chem. 41 (1969) 700–707. [38] R.H. Scott, A. Strasheim, Time-resolved direct-reading spectrochemical analysis using a laser source with medium pulse-repetition rate, Spectrochim. Acta 26B (1971) 707–719. [39] V. Sturm, L. Peter, R. Noll, Steel analysis with laser induced breakdown spectrometry in the vacuum ultraviolet, Appl. Spectrosc. 54 (2000) 1275–1278. [40] C. Hartmann, A. Gillner, Ü. Aydin, R. Noll, T. Fehr, C. Gehlen, R. Poprawe, Investigation on laser micro ablation of metals using ns multi-pulses, J. Phys. Conf. Ser. 59 (2007) 440–444. [41] A. Löbe, J. Vrenegor, R. Fleige, V. Sturm, R. Noll, Laser-induced ablation of a steel sample in different ambient gases by use of collinear multiple laser pulses, Anal. Bioanal. Chem. 385 (2006) 326–332. [42] G. Galbács, N. Jedlinszki, G. Cseh, Z. Galbács, L. Túri, Accurate quantitative analysis of gold alloys using multi-pulse laser-induced breakdown spectroscopy and a correlation-based calibration method, Spectrochim. Acta 63B (2008) 591–597. [43] G. Galbács, N. Jedlinszki, K. Herreira, N. Omenetto, B.W. Smith, J.D. Winefordner, A study of ablation, spatial and temporal characteristics of laser-induced plasmas generated by multiple collinear pulses, Appl. Spectrosc. 64 (2010) 161–172. [44] N. Jedlinszki, G. Galbács, An evaluation of the analytical performance of collinear multi pulse laser induced breakdown spectroscopy, Microchem. J. 97 (2011) 255–263. [45] T.A. Labutin, V.N. Lednev, A.A. Ilyin, A.M. Popov, Femtosecond laser-induced

Acknowledgments This project has received funding from the European Union's Horizon 2020 research and innovation programme under grant agreement no 654148 Laserlab-Europe. Appendix A. Supplementary material Supplementary data to this article can be found online at https:// doi.org/10.1016/j.optlastec.2018.10.005. References [1] K.C. Phillips, H.H. Gandhi, E. Mazur, S.K. Sundaram, Ultrafast laser processing of materials: a review, Adv. Opt. Photon. 7 (4) (2015) 684–712. [2] S. Nolte, C. Momma, H. Jacobs, A. Tünnermann, B.N. Chichkov, B. Wellegehausen, H. Welling, Ablation of metals by ultrashort laser pulses, J. Opt. Soc. Am. B 14 (1997) 2716–2722. [3] X. Liu, D. Du, G. Mourou, Laser ablation and micromachining with ultrashort laser pulses, IEEE J. Quant. Electron. 33 (10) (1997) 1706–1716. [4] N. Bloembergen, Laser-induced electric breakdown in solids, IEEE J. Quant. Electron. 10 (1974) 375–386. [5] B.C. Stuart, M.D. Feit, A.M. Rubenchik, B.W. Shore, M.D. Perry, Laser-induced damage in dielectrics with nanosecond to subpicosecond pulses, Phys. Rev. Lett. 74 (1995) 2248. [6] John C. Ion, Laser Processing of Engineering Materials, Elsevier Ltd., 2005. [7] S. Küper, M. Stuke, Ablation of UV-transparent materials with femtosecond UV excimer laser pulses, Microelectron. Eng. 9 (1989) 475–480. [8] S. Küper, M. Stuke, Ablation of polytetrafluoroethylene (Teflon) with femtosecond UV excimer laser pulses, Appl. Phys. Lett. 54 (1989) 4–6. [9] L. Shah, J. Tawney, M. Richardson, K. Richardson, Femtosecond laser deep hole drilling of silicate glasses in air, Appl. Surf. Sci. 183 (2001) 151–164. [10] M. Beresna, M. Gecevičius, P.G. Kazansky, Ultrafast laser direct writing and nanostructuring in transparent materials, Adv. Opt. Photon. 6 (2014) 293. [11] R. Osellame, G. Cerullo, R. Ramponi, Femtosecond Laser Micromachining, Springer,

301

Optics and Laser Technology 111 (2019) 295–302

J. Skruibis et al.

breakdown spectroscopy, J. Anal. At. Spectrom. 31 (2016) 90. [46] A. Baskevicius, O. Balachninaite, M. Karpavicius, S. Butkus, D. Paipulas, V. Sirutkaitis, Monitoring of the Femtosecond Laser Micromachining Process of Materials Immersed in Water by Use of Laser-Induced Breakdown Spectroscopy, J. Laser Micro/Nanoeng. 11 (3) (2016) 381. [47] S. Butkus, D. Paipulas, R. Sirutkaitis, E. Gaižauskas, V. Sirutkaitis, Rapid cutting and drilling of transparent materials via femtosecond laser filamentation, J. Laser Micro/Nanoeng. 9 (3) (2014) 213. [48] E. Tognoni, V. Palleschi, M. Corsi, G. Cristoforetti, Quantitative micro-analysis by laser-induced breakdown spectroscopy: a review of the experimental approaches, Spectrochim. Acta. 57B (2002) 1115–1130. [49] T.Y. Choi, D.J. Hwang, C.P. Grigoropoulos, Femtosecond laser-induced ablation of

[50]

[51]

[52] [53]

302

crystalline silicon upon double beam irradiation, App. Surf. Sci. 197 (2002) 720–725. I. Mirza, N.M. Bulgakova, J. Tomáštík, V. Michálek, O. Haderka, L. Fekete, T. Mocek, Ultrashort pulse laser ablation of dielectrics: thresholds, mechanisms, role of breakdown, Sci. Rep. 6 (1) (2016) 39133. Z. Zhi-Yuan, Z. Yi, Z. Wei-Gong, L. Xin, L. Yu-Tong, Z. Jie, High coupling efficiency generation in water confined laser plasma propulsion, Chin. Phys. Lett. 24 (2) (2007) 501. S. Zhu, Y.F. Lu, M.H. Hong, Laser ablation of solid substrates in a water-confined environment, Appl. Phys. Lett. 79 (2001) 1396. H. Hu, T. Liu, H. Zhai, Comparison of femtosecond laser ablation of aluminum in water and in air by time-resolved optical diagnosis, Opt. Exp. 23 (2015) 628–635.