Measuring single-shot, picosecond optical damage threshold in Ge, Si, and sapphire with a 5.1-mu (m) laser

Optical photonic structures driven by picosecond, GW-class lasers are emerging as promising novel sources of electron beams and high quality X-rays. Due to quadratic dependence on wavelength of the laser ponderomotive potential, the performance of such sources scales very favorably towards longer drive laser wavelengths. However, to take full advantage of photonic structures at mid-IR spectral region, it is important to determine optical breakdown limits of common optical materials. To this end, an experimental study was carried out at a wavelength of 5 μm, using a frequency-doubled CO2 laser source, with 5 ps pulse length. Single-shot optical breakdowns were detected and characterized at different laser intensities, and damage threshold values of 0.2, 0.3, and 7.0 J/cm, were established for Ge, Si, and sapphire, respectively. The measured damage threshold values were stable and repeatable within individual data sets, and across varying experimental conditions. References and links 1. R. L. Byer, T. Plettner, "Building a pocket-size electron accelerator," Photonics Spectra 42 (7), 64-68 (2008). 2. Y. Takeda and I. Matsui, “Laser linac with grating,” Nucl. Instr. Meth. 62(3), 306-310 (1968). 3. J. Rosenzweig, A. Murokh and C. Pellegrini, “A proposed dielectric loaded resonant laser accelerator,” Phys. Rev. Lett. 74, 2467-2470 (1995). 4. X. E. Lin, “Photonic band gap fiber accelerator,” Phys. Rev. ST Accel. Beams 4, 051301 (2001). 5. B. M. Cowan, “Three-dimensional dielectric photonic crystal structures for laser-driven acceleration,” Phys. Rev. ST Accel. Beams 11, 011301 (2008). 6. I. Staude, C. McGuinness, A. Frölich, R. L. Byer, E. Colby and M. Wegener "Waveguides in three-dimensional photonic bandgap materials for particle-accelerator on a chip architectures," Opt. Express 20, 5607-5612 (2012) 7. E. A. Peralta, K. Soong, R. J. England, E. R. Colby, E. A. Peralta, Z. Wu, B. Montazeri, C. McGuinness, J. McNeur, K. J. Leedle, D. Walz, E. B. Sozer, B. Cowan, B. Schwartz, G. Travish and R. L. Byer “Demonstration of electron acceleration in a laser-driven dielectric microstructure,” Nature 503, 91-94 (2013). 8. J. Power and G. Travish, “Summary of the working group 3,” presented at 16 Advanced Accelerator Concepts Workshop, San Jose, CA, USA, 13-18 Jul. 2014. 9. R. J. England et al., “Dielectric laser accelerators,” Rev. Mod. Phys. 86, 1337 (2014). 10. B. M. Cowan, “Two-dimensional photonic crystal accelerator structures,” Phys. Rev. ST AB 6, 101301 (2003). 11. R. J. England, “Experiment to Demonstrate Acceleration in Optical Photonic Bandgap Structures,” presented at 2011 Particle Accelerator Conference, New York, NY, USA, 28 Mar. 01 Apr. 2011. 12. T. Plettner, P. Lu, R. L. Byer, “Proposed few-optical cycle laser-driven particle accelerator structure,” Phys. Rev. ST Accel. Beams 9, 111301 (2006). 13. R. B. Yoder and J. B. Rosenzweig, “Side-coupled slab-symmetric structure for high-gradient acceleration using terahertz power,” Phys. Rev. ST Accel. Beams 8, 111301 (2005). 14. B. Naranjo, A. Valloni, S. Putterman, and J. B. Rosenzweig, “Stable Charged-Particle Acceleration and Focusing in a Laser Accelerator Using Spatial Harmonics,” Phys. Rev. Lett. 109, 164803 (2012). 15. I. Jovanovic, G. Xu, S.Wandel, “Mid-infrared laser system development for dielectric laser accelerators,” Physics Procedia 52, 68-74 (2014). 16. T. Popmintchev, M.-C. Chen, D. Popmintchev, P. Arpin, S. Brown, S. Ališauskas, G. Andriukaitis, T. Balčiunas, O. D. Mücke, A. Pugzlys, A. Baltuška, B. Shim, S. E. Schrauth, A. Gaeta, C. Hernández-García, L. Plaja, A. Becker, A. Jaron-Becker, M. M. Murnane, and H. C. Kapteyn “Bright coherent ultrahigh harmonics in the kev X-ray regime from mid-infrared femtosecond lasers,” Science 336, 1287–1291 (2012). 17. J. Weisshaupt, V. Juvé, M. Holtz, S. Ku, M. Woerner, T. Elsaesser, S. Ališauskas, A. Pugžlys and A. Baltuška “High-brightness table-top hard X-ray source driven by sub-100-femtosecond mid-infrared pulses”, Nature Photonics 8, 927–930 (2014). 18. E. R. Colby, R. J. England, R. J. Noble, “A laser-driven linear collider: sample machine parameters and configuration,” Proc. 24th Particle Accelerator Conf. C110328, 262–264 (2011). 19. F. Wang, C. Adolphsen, and C. Nantista, “Performance limiting effects in X-band accelerators,” Phys. Rev. ST AB 14, 010401 (2011). 20. M. C. Thompson, H. Badakov, A. M. Cook, J. B. Rosenzweig, R. Tikhoplav, G. Travish, I. Blumenfeld, M. J. Hogan, R. Ischebeck, N. Kirby, R. Siemann, D. Walz, P. Muggli, A. Scott, and R. B. Yoder, "Breakdown Limits on Gigavolt-per-Meter Electron-Beam-Driven Wakefields in Dielectric Structures,” Phys. Rev. Lett. 100, 214801 (2008). 21. B. C. Stuart, M. D. Feit, S. Herman, A. M. Rubenchik, B. W. Shore, and M. D. Perry, “Nanosecond-tofemtosecond laser-induced breakdown in dielectrics,” Phys. Rev. B 53(4), 1749–1761 (1996). 22. C. B. Schaffer, A. Brodeur, and E. Mazur, "Laser-induced breakdown and damage in bulk transparent materials induced by tightly focused femtosecond laser pulses." Meas. Science and Tech. 12(11), 1784-1794 (2001). 23. M. Mero, J. Liu, W. Rudolph, D. Ristau, and K. Starke, “Scaling laws of femtosecond laser pulse induced breakdown in oxide films,” Phys. Rev. B 71(11), 115109 (2005). 24. H. O. Jeschke, M. E. Garcia, M. Lenzner, J. Bonse, J. Krüger, W. Kautek, “Laser ablation thresholds of silicon for different pulse durations: theory and experiment,” Appl. Surface Science 197, 839-844 (2002). 25. B. Mangote, L. Gallais, M. Commandré, M. Mende, L. Jensen, H. Ehlers, M. Jupé, D. Ristau, A. Melninkaitis, J. Mirauskas, V. Sirutkaitis, S. Kičas, T. Tolenis, and R. Drazdys, “Femtosecond laser damage resistance of oxide and mixture oxide optical coatings,” Opt. Lett. 37(9), 1478–1480 (2012). 26. C. B. Schaffer, A. O. Jamison, and E. Mazur, “Morphology of femtosecond laser-induced structural changes in bulk transparent materials,” Appl. Phys. Lett. 84, 1441-1443 (2004). 27. K. Soong, R. L. Byer, E. R. Colby, E. A. Peralta, “Laser damage threshold measurements of optical materials for direct laser accelerators,” AIP Conf. Proc. 1507, 511–515 (2012). 28. M. Polyanskiy and M. Babzien, “Ultrashort Pulses,” CO2 Laser-Optimisation and Application, (InTech Open Science, 2012), pp. 139-162 29. “Infrared Nonlinear Crystals”, http://www.oplanchina.com/data/upload/product/ekspla/crystals/nonlinear/IR.pdf 30. Y. C. Lam, D. V. Tran, H. Y. Zheng, V. M. Murukeshan, J. C. Chai and D. E. Hardt, “Surface damage of crystalline silicon by low fluence femtosecond second laser pulses,” Surface Rev. Lett. 11(2), 217-221 (2004). 31. M. Bass, V. N. Mahajan, Handbook of Optics, 3rd ed., New York (McGraw-Hill, 2010), v. <1, 4-5>. 32. R. H. French, “Electronic Band Structure of Al2O3, with Comparison to Alon and AIN,” J. of the American Ceramic Society 73(3), 477-489, (1990). 33. L. V. Keldysh, "Ionization In Field Of A Strong EM Wave," Sov. Phys. Jetp-Ussr, 20(5), 1307 (1965). 34. B. M. Cowan, “Optical damage threshold of silicon for ultrafast infrared pulses”, Proc. SPIE 6720, (2007).


Introduction
The experimental determination of damage thresholds of optical wafers, reported herein, was motivated by the development of the dielectric laser accelerator (DLA), a promising laserpowered technology to reduce the characteristic feature sizes of electron beam sources down to the optical scale [1]. DLAs can reach accelerating gradients exceeding 1 GeV/m, which is more than an order of magnitude improvement over the state-of-the-art radio frequency linear accelerators (RF linacs) presently dominating research and industrial applications. The DLA has been considered for decades [2][3][4], but only recently has the concept gained momentum [5,6] with advances in solid-state laser technology and nanoengineering methods. A significant milestone-the first experimental demonstration of DLA energy gain-has been achieved at the Stanford Linear Accelerator Center (SLAC) in 2013 [7].
Most of the DLA work to date has been done at 800 nm due to the commercial availability of Ti:Sa lasers. A number of DLA topographies are being considered [8,9], including photonic band gap fiber and crystal structures [10,11] and transmission-mode grating structures [12]. To relax the spatial acceptance limitations in at least one dimension, planar standing-wave (SW) DLA structures have been developed [13]. Nevertheless, at least one dimension of the electron beam's transverse size has to be confined to a fraction of the laser wavelength, generating strong interest in scaling DLA structures to mid-IR. Recently a new traveling-wave photonic structure DLA was proposed at UCLA [14], operating at 5 µm to take advantage of the emerging availability of high-power, mid-IR, OPA sources [15]. While the 5-µm laser technology itself is more challenging, it is a rapidly evolving field that is also driven by the demands of high-harmonic generation (HHG) [16,17]. Thus, it is expected that mid-IR laser sources suitable for driving DLAs will be commercially available within a few years.
The use of longer wavelength radiation offers several advantages that enable practical DLA devices. Expanding the gap size permits a relaxation of all structure dimensions and fabrication tolerances. Larger structure apertures and stored energies allow higher charges to be accelerated, mitigating both longitudinal and transverse wake-field effects and permitting larger source emittances. Finally, the larger achievable beam loading plays an important role in improving the overall energy efficiency of the acceleration system-a critical consideration for DLAs in the context of future linear collider technology [18].
One important factor, however, is the picosecond optical breakdown limit at mid-IR wavelengths. In general, for any resonant accelerating structure, the maximum achievable gradient is always limited by a breakdown effect. A typical industrially available S-band (3-GHz) microwave linac usually operates at a gradient of about 20 MV/m, and some high performance linacs designed for collider applications can reach up to 100 MV/m at 11 GHz [19]. Beyond these values, however, RF breakdown events are unavoidable and constitute a fundamental limitation in the practical use of microwave technology at high gradients. Since the breakdown threshold increases with operating frequency and is higher for dielectrics [20], it is believed that a DLA will be able to sustain field gradients exceeding 1 GV/m with picosecond pulses. However, before such a DLA photonic structure can be implemented in practice, a picosecond laser damage threshold needs to be established for common photonic building materials.
The general principles for the damage mechanisms near threshold from short-pulse (picoseconds or shorter) lasers are well understood [21][22], and the theory is well supported by experimental data in the visible and near infrared (NIR) spectral ranges [23][24][25][26] including some recent experimental studies which were conducted at SLAC [27] in the context of the DLA development. There is, however, no prior data on the short-pulse breakdown in the mid-IR range. Thus, to facilitate development of mid-IR DLA structures, we investigated the breakdown threshold of three materials: Ge and Si and sapphire (Al 2 O 3 ). We have chosen the popular semiconductors Ge and Si for manufacturability. The fabrication processes for sapphire is less established, however the material has a much larger band gap than either germanium or silicon, and all the experimental data in the optical spectral range [23][24][25][26][27] indicated an anticipated breakdown threshold dependence on the band gap. Thus, sapphire was chosen as an important reference point for comparing mid-IR to NIR optical breakdown properties.

Experiment
The experiments were performed at Brookhaven National Laboratories' Accelerator Test Facility (ATF). The short-pulse CO 2 laser design and operation have been described in detail elsewhere [28]. At ATF, the 5-ps CO 2 pulses of variable peak power-up to TW levels-are available. The 10.2-µm pulse is frequency-doubled to 5.1 µm using ZnGeP 2 (ZGP) non-linear crystal, which at longer wavelengths demonstrated superior performance to other doubling crystals [29] in having the highest nonlinear coefficient and the highest damage threshold.
Initial experiments were conducted in air, with the system set on an optical table, as shown in the photograph in Fig. 1. The CO 2 beam initially used was directly from the regenerative amplifier, generating 5 ps pulses at 10.2 µm with the pulsed energy up to 10 mJ. However, at the location of ZGP crystal (in a different room) the transport resulted in losses of about 55% of the initial CO 2 beam power, resulting in up to 4.5 mJ delivered for second harmonic generation (SHG). Shot-to-shot energy fluctuated over a wide range (2.0-4.5 mJ), but the spatial profile of the laser was very stable and clean (M 2 ~ 1.6) with the waist size on the order of 200 µm. SHG was optimized in two steps: (1) the crystal was rotated in the transverse and horizontal planes to determine the angle where the process was most efficient to accommodate for the small difference between the crystal design wavelength (10.6 µm) and CO 2 laser operating wavelength (10.2 µm), (2) the crystal longitudinal position with respect to the CO 2 laser waist was scanned and optimize to avoid crystal damage and saturation. The optimized position was determined at about 4 cm upstream of the waist, where the CO 2 spot size was about 450 µm, and at 4.5 mJ the conversion efficiency of 14% was achieved, yielding 600-µJ pulses at mid-IR. In the later round of measurements (for the sapphire samples), the CO 2 pulse was passed through an additional amplification stage to improve stability and achieve higher pulsed energy levels at the SHG crystal, but the optimization procedure was the same.
A picosecond, 5.1-µm, second-harmonic pulse generated by SHG is then separated from the background 10-µm light via a set of dichroic CaF 2 mirrors and focused on the surface of a sample. A portion of the pulse is directed to the pyro detector, which provides shot-to-shot pulsed energy information. After the sample chamber there is also a pyro camera, which provides shot-to-shot images of the mid-IR beam. The IR camera is used to monitor the beam profile throughout the measurements as well as for initial alignment of the 5-µm beam to the reference pinholes positioned at the sample plane. An absolute measurement of mid-IR transmission through calibrated reference set of pinholes is used to derive the peak intensity at the laser plane. The pinholes also aid in the alignment of the in-situ damage detection camera.
A sample holder enabled transverse translation of the sample, so a new area of the sample is exposed after each consecutive shot. For each shot, the pulsed energy was imaged and the location was recorded and stored for comparison with the forensic analysis upon removal of the sample. Alignment and focal properties of the laser pulse were periodically sampled and remained consistent throughout the entire procedure. After the initial round of experiments, the vacuum chamber was introduced around the sample area, to make sure that an exposure to air does not affect the measured breakdown threshold.

Results
The samples were Si and Ge (001) single crystal wafers. The sapphire crystal was oriented along the <0001> direction. Both Ge and Si wafers were N-type with 53.3 to 57.9 Ω-cm resistivity. The samples received exposure from 5.1-µm laser pulses with 0.1-1-mJ energy, 60-µm diameter, and about 5-ps pulse length. After each consecutive shot the samples under test were shifted transversely by about 2 mm in a matrix pattern, and for each shot the matrix position on the wafer and the laser pulse intensity were recorded. The surface damage pattern was first detected in real time with the high magnification CCD in the measurement setup, and later examined forensically by scanning electron microscopy (SEM) using a Hitachi 4800 SEM in the secondary electron imaging mode. Figure 2 shows SEM images of Ge (top) and sapphire (bottom) damage sites. The damage sites in all tested materials show both melting and ablation. In Ge, the large circular transition from dark to light indicates the limit of the melt zone, and the light ring is the frozen surface wave produced by the ablation event. In sapphire, more complex damage morphology is observed where the melt zone starts further away from the center, but following the same concentric surface wave pattern, often with multiple rings (very similar to the model introduced in [30]). In all samples (Ge, Si, and sapphire) there is a visible accumulation of larger ablation debris at the center of the laser damage trace. The size of the damaged area was determined digitally measuring the radius of the ablation area. The peak fluence leading to the damage was determined by approximating the ablation area diameter to zero, whereas diameter of the ablation area was correlated with the laser 1 µm 2 µm 10 µm 20 µm power, as shown in Fig. 3. The shot-to-shot laser intensity variations necessary for such measurement were naturally present, due to the CO 2 laser pulsed energy fluctuations (originating in the regenerative amplifier). For each set of measurements, the peak fluence threshold was obtained as a fit parameter to the curve in Fig. 3 (right). The beam width shotto-shot variations were insignificant. The measurements results are summarized in Table 1. For each set of data, the peak fluence resulting in damage to the optical surface was determined using the method shown in Fig. 3.
The laser damage threshold is defined as a total pulsed energy in 1/e 2 laser beam (as per guidelines of ISO 11254). Thus, in a Gaussian beam approximation, the laser damage threshold value is simply ½ of the measured peak fluence value. The measurements were first carried out in air, and later repeated in vacuum, using a specially fabricated measurement chamber to enable UHV level better than 10 -5 torr. All samples were cleaned prior to shipment and handled in sealed enclosures until installation to remove particulate contamination from the surface. However, the in-air measurements were conducted without any clean room environment provisions, thus some reasonable degree of surface contamination was anticipated. Nevertheless, no quantitative difference in the damage threshold was observed for in-vacuum and out-of-vacuum measurements. Qualitatively, there were also no observable differences in the shape or pattern of the damaged area.

Discussion
Ge and Si are covalently bounded semiconductors [31] with similar electronic structures. They have indirect band gaps of 0.67 eV and 1.1 eV oriented along the <100> direction. Both materials have excitons bound just below the conduction band (at 0.004 eV and 0.015 eV), which is considered to be too close to significantly impact electron seeding. By comparison, sapphire has a strongly ionic character [32] and a direct band gap of 9.9 eV. Its excitons are