Open Access
Add to BibSonomyAbstract
We report on phase measurements and adaptive phase distortion compensation of femtosecond pulses using multiphoton intrapulse interference phase scan (MIIPS) based on second harmonic generation in the plasma generated on the surface of silicon and metals.
©2007 Optical Society of America
1. Introduction
Phase characterization and automated pulse compression through adaptive phase distortion compensation of femtosecond laser pulses is important for reproducible nonlinear optical applications using femtosecond lasers [1]. Currently the characterization of femtosecond pulses is usually carried out by autocorrelation, frequency resolved optical gating (FROG) [2], spectral interferometry for direct electric field reconstruction (SPIDER) [3] or a relatively new method called shaper assisted collinear SPIDER (SAC-SPIDER) [4]. These methods depend on the mode quality of the beam and typically are relatively difficult to set up because they require the overlap of two or more beams in space and in time. Here we use multiphoton intrapulse interference phase scan (MIIPS) [5–7], which is a highly accurate single beam characterization and dispersion compensation method that is not dependent on beam quality. [8]. This system, which uses an adaptive pulse shaper, allows us to explore the effect of pulse shaping (spectral phase) and reproducibility in advanced ultrafast laser material processing [9, 10].
During MIIPS, a series of calibrated phase functions are introduced and the spectra of a nonlinear optical process, typically second harmonic generation (SHG), obtained for the reference phases are used to analytically measure the spectral phase distortions of the pulse. The MIIPS program calculates the phase distortions analytically and uses the adaptive pulse shaper to compensate them. Here we scan a well calibrated function f(ω)=αsin{γ(ω-ω0)-δ(ω)} where α=π, γ is bandwidth of the pulse, ω0 is the carrier frequency and δ(ω) determines the position of the phase mask with respect to the spectrum of the pulse. Phase measurement is based on the fact that the second harmonic spectrum has a maximum, ω max(δ), where local phase distortions for the phase modulated pulse are minimal, i.e. when f′(ω)+ϕ′(ω)→0 where ϕ is the spectral phase of the pulse. By scanning δ in the calibrated phase function f(δ,ω) and measuring the position of the corresponding second harmonic spectrum maximum ωmax(δ), we can directly calculate the unknown phase distortion ϕ using the experimentally measured function δmax(ω), the formula ϕ′(ω)=αγ2×sin[γ(ω-ω0)-δmax(ω)] and double integration in the frequency domain. Details of the MIIPS method can be found in a comprehensive paper [6].
Figure 1 presents SHG-FROG and MIIPS traces for pulses (through a 40x, 0.6 NA objective) that are both compressed to transform limited (top) and uncompressed (bottom). For transform limited pulses, δ max(ω) is a linear function that produces straight parallel features separated by π. Dispersions introduced by the objective, and other optics, result in changes in the spacing and slope of the features. Pulse compression is accomplished by eliminating the phase distortions (including higher order terms) by applying the negative of the phase measured using MIIPS. When the shaper introduces -ϕ(ω), transform limited pulses are obtained, as shown in Fig. 1 (top).
Fig. 1. Comparison of SHG-FROG and MIIPS traces for compressed (TL) and uncompressed pulses. The two sets of measurements were made using the same laser pulses, a 40x, 0.6 NA microscope objective and a SHG crystal. The dashed lines coincide with the MIIPS features observed for TL pulses and are used as a guide to the eye.
Here we demonstrate in-situ characterization and adaptive dispersion compensation of femtosecond laser pulses using the surface second harmonic generation (SSHG) from three different substrates, silicon, copper and aluminum. This study is motivated by analysis of the microscopic morphology of ablated holes using shaped laser pulses in our laboratory, where we found that TL pulses create the sharpest borders[11].
SSHG was first reported by Terhune [12] in 1962 in calcite and further investigated theoretically by Pershan [13], Adler [14] and later, Bloembergen [15, 16] in silicon and germanium. Although some of these early experiments are focused on SSHG generation below the ablation threshold, some reports deal with the SSHG generation in plasma produced by laser matter interaction above the ablation threshold [17–20]. Von der Linde and coworkers have carried out experimental and theoretical studies on SHG generated in plasma produced by femtosecond laser pulses[18, 19]. In our experiments we used normal incidence and collected the SSHG beam confocally in order to both mimic the micromachining environment, where this technique has great potential, and to avoid signal fluctuation due to surface roughness.
It has been shown before that SSHG signal can be utilized in autocorrelation to measure femtosecond pulses [21, 22]. The main advantage of SSHG-MIIPS is that pulse characterization and optimization are done in-situ, after a high numerical aperture microscope objective. This ensures that phase distortions introduced by the optics are corrected and that the optimum phase is introduced for achieving optimum micromachining at the sample without modifications to the machining setup. This approach does not need a non-linear SHG crystal with phase matching requirements (carrier frequency and bandwidth dependence), which tend to be expensive and prone to laser damage.
2. Experimental section
The experimental setup, shown in Fig. 2, includes a regeneratively amplified Ti:sapphire laser (Spitfire-Spectra Physics) seeded with 100 MHz oscillator (KM Labs, 45 nm FWHM) after compensating for phase distortions using a MIIPS-enabled pulse shaper [5] that is also used to apply arbitrary phase functions. The output laser pulses from the amplifier are centered at 800 nm with 750 µJ/pulse at 1 kHz with pulse duration of 35 fs. A fraction of this output was spatially filtered with a 100 µm pinhole located within an up-collimating Keplerian telescope with 300 mm and 600 mm lenses. The output of the telescope was directed to a 0.6 numerical aperture 40X objectives (Plan Fluor ELWD, Nikon) which focused the laser beam at normal incidence onto the sample, which was mounted on motorized X-Y stages. The stages moved after every laser pulse in order to supply fresh surface for each laser shot. Here we are used a confocal arrangement to mitigate the effects that are arising from surface imperfections. The SSHG signal was collected and focused to a high sensitivity fiber coupled spectrometer (QE65000, Ocean Optics).
Fig. 2. Schematic of the experimental setup. L1, L2 and L3 are 300, 600 and 50 mm focal length lenses. P pinhole, BS beams splitter that transmits 800 nm and reflects 300–600 nm.
3. Results and discussion
The emission spectra from a Si substrate are plotted for three laser intensities in Fig. 3. At low fluence (<0.02 J/cm2) only broad plasma emission is observed. At a higher fluence (0.04 J/cm2) an atomic emission line appears. At a fluence of 0.4 J/cm2 the SSHG signal becomes dominant. This power dependence shows that SSHG, in our case, appears once both plasma and atomic emissions are well developed. We found (data not shown) that at 450 incidence the SSHG signal is primarily emitted in the specular direction and p-polarized light generates more SSHG than s-polarized light. These observations support our conclusion that the SSHG observed is due to an ac electronic current in the plasma [18, 23]. The absence of a blue shift and broadening in the SHG spectrum indicate the source for SHG is weak plasma and not a turbulent or dense plasma [18, 23]. This model holds true for relatively low energies near ablation threshold where perturbations are minimal in the plasma. This also implies that the approach presented here for pulse characterization and adaptive compression is generally applicable to any substrate being processed by femtosecond laser pulses.
Fig. 3. Emission signal is plotted for three different intensities showing the plasma emission (PE), atomic emission (LIBS) and SSHG signal. Note the absence of LIBS and SSHG signal in the spectrum taken with the intensity of <0.1 µJ/pulse but dominant PE signal.
To test the accuracy of SSHG-MIIPS, the trace from the surface of the silicon wafer was measured and compared to SHG-MIIPS measurements obtained using a 50 µm β-barium borate (BBO) SHG crystal. Figure 4 shows the resulting MIIPS traces obtained for transform limited (TL) pulses. The dashed lines coincide with the maximum intensity of the spectrum at each δ value. In both cases we see the parallel lines separated by π, indicating that the pulses are indeed TL. In the case of SSHG-MIIPS, there is a difference between alternating features. The difference results from the different temporal symmetry of the pulses during the MIIPS scan, a subject that is discussed in greater detail elsewhere [11]. However, this difference affects only overall intensity but not the phase measurements in the MIIPS algorithm.
Fig. 4. MIIPS traces obtained after adaptive phase compensation using a BBO SHG crystal (left) and SSHG on the surface of silicon wafer (right).
The effect of phase distortions on the MIIPS traces is illustrated in Fig. 5. In the top panel we show MIIPS traces without compensation for laser pulses with either a positive or a negative 2,000 fs2 chirp. Notice that there is a clear change in the spacing between the MIIPS features for positively and negatively chirped pulses. The dashed lines shown correspond to the TL pulses shown in Fig. 4.
Fig. 5. MIIPS traces obtained after the first MIIPS scan without compensation using a BBO SHG crystal (left) and SSHG on the surface of silicon wafer (right). The top panels show MIIPS traces after the introduction of 2,000 fs2 positive chirp. The bottom panels show the MIIPS traces after the introduction of -2,000 fs2 negative chirp. The dashed lines correspond to the position of MIIPS features for TL pulses.
Clearly, SSHG-MIIPS faithfully reproduces the MIIPS traces obtained using an SHG crystal. The retrieved phases using SSHG-MIIPS on a silicon wafer are shown in Fig. 6. The top panel shows the residual phase distortions after pulse compression through adaptive phase distortion compensation. Notice that phase deviations are less than 0.2 radians across the bandwidth of the pulses. The bottom panel shows the spectrum of the pulse and the positive and negative chirp measured using MIIPS. Notice the excellent agreement between the measured phases (solid line) and the phases introduced (dashed line).
Fig. 6. Phases retrieved using SSHG-MIIIPS from the surface of a silicon wafer. The top panel shows the residual phase distortions after adaptive phase compensation using MIIPS. The deviations correspond to ±1 standard deviations following 5 repetitions of the measurements. The bottom panel shows the spectrum of the incident laser pulses together with the measured positive and negative chirp phases (±2,000 fs2) introduced. Notice the excellent agreement between the phases introduce (dashed lines) and the phases measured (blue lines) using SSHG-MIIPS.
As demonstrated in Fig. 6, SSHG-MIIPS from the surface of a silicon wafer provides high fidelity information that can be used for the accurate measurement of arbitrary phase modulation and can also be used for adaptive pulse compression, as the MIIPS algorithm compensates phase distortions. Further, we tested copper and aluminum samples in order to determine the usefulness of this method for metal micromachining. We were able to obtain MIIPS traces with similar precision of phase measurements and compensation to those that are obtained with an SHG crystal and are given in Fig. 7. This indicates that SSHG-MIIPS is versatile enough to use in the micromachining industry to compensate pulses at the point where machining occurs for various substrates in order to achieve efficient and reproducible machining features.
Fig. 7. MIIPS traces obtained using surface generated second harmonic on Cu and Al samples indicating usefulness of MIIPS method to measure phase of the pulse interacting with the sample during metal micromachining.
4. Conclusion
We have demonstrated a phase measurement and compensation method using surface second harmonic generation emitted by the surface plasma. This method is particularly useful for material processing with femtosecond laser pulses. Phase measurement of the femtosecond pulse on the working surface is a critical condition for many applications of ultrafast laser pulses, for example: laser ablation, micromachining, laser surgery, two photon microscopy etc. SSHG-MIIPS can provide highly accurate, fast, robust and reliable phase measurements and control in-situ.
Acknowledgment
We gratefully acknowledge funding for this research from the National Science Foundation Major Research Instrument grant CHE-0421047 and from the Chemical Sciences, Geosciences and Biosciences Division, Office of Science, US Department of Energy.
References and links
1. V. V. Lozovoy and M. Dantus, “Coherent control in femtochemistry,” Chemphyschem 6, 1970–2000 (2005). [CrossRef] [PubMed]
2. R. Trebino and D. J. Kane, “Using Phase Retrieval to measure the intensity and phase of ultrashort pulses - frequency-resolved optical gating,” J. Opt. Soc. Am. A-Opt. Image Sci. Vis. 10, 1101–1111 (1993). [CrossRef]
3. C. Iaconis and I. A. Walmsley, “Spectral phase interferometry for direct electric-field reconstruction of ultrashort optical pulses,” Opt. Lett. 23, 792–794 (1998). [CrossRef]
4. B. von Vacano, T. Buckup, and M. Motzkus, “Shaper-assisted collinear SPIDER: fast and simple broadband pulse compression in nonlinear microscopy,” J. Opt. Soc. Am. B-Opt. Phys. 24, 1091–1100 (2007). [CrossRef]
5. I. Pastirk, B. Resan, A. Fry, J. MacKay, and M. Dantus, “No loss spectral phase correction and arbitrary phase shaping of regeneratively amplified femtosecond pulses using MIIPS,” Opt. Express 14, 9537–9543 (2006). [CrossRef] [PubMed]
6. B. W. Xu, J. M. Gunn, J. M. Dela Cruz, V. V. Lozovoy, and M. Dantus, “Quantitative investigation of the multiphoton intrapulse interference phase scan method for simultaneous phase measurement and compensation of femtosecond laser pulses,” J. Opt. Soc. Am. B-Opt. Phys. 23, 750–759 (2006). [CrossRef]
7. M. Dantus, V. V. Lozovoy, and I. Pastirk, “MIIPS characterize and corrects femtosecond pulses,” Laser Focus World 43, 101 (2007).
8. A. M. Weiner, D. E. Leaird, J. S. Patel, and J. R. Wullert, “Programmable Shaping of Femtosecond Optical Pulses by Use of 128-Element Liquid-Crystal Phase Modulator,” IEEE J. Quantum Electron. 28, 908–920 (1992). [CrossRef]
9. V. Hommes, M. Miclea, and R. Hergenroder, “Silicon surface morphology study after exposure to tailored femtosecond pulses,” App. Surf. Sci. 252, 7449–7460 (2006). [CrossRef]
10. R. Stoian, M. Boyle, A. Thoss, A. Rosenfeld, G. Korn, and I. V. Hertel, “Dynamic temporal pulse shaping in advanced ultrafast laser material processing,” App. Phys. A-Materials Science & Processing 77, 265–269 (2003).
11. T. C. Gunaratne, X. Zhu, R. Amin, V. V. Lozovoy, and M. Dantus, “Influence of femtosecond pulse shaping on silicon micromachining monitored by laser induced breakdown spectroscopy and surface second harmonic generation,” Phys. Rev. B., (in preparation) (2007).
12. R. W. Terhune, P. D. Maker, and C. M. Savage, “Optical Harmonic Generation in Calcite,” Phys. Rev. Lett. 8, 404-& (1962). [CrossRef]
13. P. S. Pershan, “Nonlinear Optical Properties of Solids - Energy Considerations,” Phys. Rev. 130, 919-& (1963). [CrossRef]
14. E. Adler, “Nonlinear Optical Frequency Polarization in Dielectric,” Physical Review a-General Physics 134, A728-& (1964).
15. N. Bloembergen, “Wave Propagation in Nonlinear Electromagnetic Media,” Proc. IEEE 51, 124-& (1963). [CrossRef]
16. N. Bloembergen and Y. R. Shen, “Optical Nonlinearities of a plasma,” Phys. Rev. 141, 298–305 (1966). [CrossRef]
17. N. G. Basov, V. Y. Bychenkov, O. N. Krokhin, M. V. Osipov, A. A. Rupasov, V. P. Silin, G. V. Sklizkov, A. N. Starodub, V. T. Tikhoncchuk, and A. S. Shikanov, “Second harmonic generation in a laser plasma,” Sov. J. Quantum Electron 9, 1081–1102 (1979). [CrossRef]
18. D. von der Linde, H. Schulz, T. Engers, and H. Schuler, “Second harmonic generation in plasmas produced by intense femtosecond laser pulses,” IEEE J. Quantum Electron. 28, 2388–2397 (1992). [CrossRef]
19. T. Engers, W. Fendel, H. Schuler, H. Schulz, and D. von der Linde, “second harmonic generation in plasmas prodused by femtosecond laser pulses,” Phys. Rev. A 43, 4564–4567 (1991). [CrossRef] [PubMed]
20. A. Terasevitch, C. Dietrich, K. Sokolowski-Tinten, and D. von der Linde, “3/2 harmonic generation by femtosecond laser pulses in steep-gradient plasmas,” Phys. Rev. E 68, 026410 (2003). [CrossRef]
21. N. D. Whitbread, J. A. R. Williams, J. S. Roberts, I. Bennion, and P. N. Robson, “Optical autocorrelator that used a surface-emitting second-harmonic generator on (211)B GaAs,” Opt. Lett. 19, 2089–2091 (1994). [CrossRef] [PubMed]
22. E. J. Canto-Said, P. Simon, C. Jordan, and G. Marowsky, “Surface second-harmonic generation in Si(111) for autocorrelation measurments of 248 nm femtosecond pulses,” Opt. Lett. 18, 2038–2040 (1993). [CrossRef] [PubMed]
23. W. L. Kruer, The physics of laser plasma interactions (Addison-Wesley Publishing Co., 1988).
