stuart_1996#

B. C. Stuart, M. D. Feit, S. Herman, A. M. Rubenchik, B. W. Shore and M. D. Perry, “Nanosecond-to-femtosecond laser-induced breakdown in dielectrics”, Phys. Rev. B 53, 1749 (1996)

Source record · Catalogue · Review register

DOI: 10.1103/physrevb.53.1749

Source kind: measurement

Recorded technique: Front-surface laser damage of polished dielectrics with a 1053-nm Ti:sapphire chirped-pulse amplifier (10 Hz), compressed to 0.4 ps - 1 ns (intensity FWHM), and its second harmonic at 526 nm (275 fs and longer, KD*P doubler); one point at 825 nm, 140 fs (Cr:LiSAF). Focusing by a 1-m lens to 0.3-1.0 mm diameter (1/e^2 intensity; typically 0.5 mm), the spot measured on a CCD from a 4:1 image of a 4 % reflection at the sample plane, >= 98 % Gaussian. Damage criterion: ‘any visible permanent modification to the surface observable with the Nomarski microscope’ after typically 600 shots at 10 Hz per site, 15-30 fluence levels per pulse width (pp. 1750-1751). Plus a kinetic (Fokker-Planck) and rate-equation model of multiphoton-seeded avalanche ionization in fused silica (Sec. IV).

Materials in structured entries: BaF2, CaF2, LiF, MgF2, NaCl, SiO2

Dataset quantities: avalanche_coefficient, band_gap, damage_threshold_fluence, effective_mass, multiphoton_ionization_cross_section

Contents: 9 datasets; 7 formulas.

Curation notes#

Journal pages 1749-1761 = PDF pages 1-13. The PDF prints no DOI; identity from the library’s bibliographic search, and title, authors and journal were checked against page 1.

UNCERTAINTY (p. 1750): ‘Our estimated absolute uncertainty in fluence was 15%, but relative values when changing the laser pulse width or spot size are accurate to 5%.’ Level not stated. The fluence is ‘the effective diameter, as measured on the camera system, ... combined with the measured energy’; whether peak or average fluence is not said. ‘With short (0.4-ps) pulses, damage always occurred at the location corresponding to the peak of the Gaussian spatial profile’ (p. 1751). Short-pulse thresholds are sharp: 2 % below threshold, 10 000 pulses at 0.4 ps produced no damage (p. 1752).

SAMPLES. Fused silica: 1-cm-thick ‘superpolished’ Corning 7940, < 1 nm rms roughness; a thin fused-silica etalon gave the same thresholds (p. 1751). Fluorides: BaF2, CaF2, MgF2, LiF (grades, orientation and polishing not stated beyond ‘scratches left from the polishing process’, p. 1752).

TEXT VALUES TAKEN (p. 1752): ‘Our measured short-pulse (0.4 ps) damage thresholds of BaF2 (1.6 J/cm2), CaF2 (2.0 J/cm2), MgF2 (2.1 J/cm2), and LiF (2.6 J/cm2) scale with band-gap energy’ (the OCR of the library text drops the ‘2’ of J/cm2; the pdf text layer confirms the values). The wavelength is not named in the sentence: 0.4 ps is the shortest 1053-nm pulse (the 1053-nm compressor gives 0.4 ps - 1 ns, p. 1750; the 526-nm pulses start at 275 fs), and Fig. 7 (CaF2) covers 0.4-1000 ps, the 1053-nm range; so 1053 nm is recorded.

FIGURES ONLY (not digitised). Fig. 3: fused-silica damage fluence (1-50 J/cm2, log) against pulse width (0.1-1000 ps, log) at 1053 nm (about 2 J/cm2 at 0.4 ps rising to about 40 J/cm2 at 900 ps), a single 825-nm point at 140 fs with a large error bar, and the tau^1/2 line; above 20 ps the data ‘fit well to a tau^1/2 dependence (actual fit: tau^0.504)’ (p. 1751). Fig. 6: fused silica at 1053 nm (0.4-60 ps) and 526 nm (0.275-7 ps, about 1-2.5 J/cm2), measured points and the calculated curves of the model (0.01-100 ps, 0.3-10 J/cm2), and the calculated multiphoton-ionization-only limit at 1053 nm. Fig. 7: CaF2 at 1053 nm, 0.4-1000 ps, 2-25 J/cm2. Figs. 9-18: the model’s conductivity, avalanche rate against intensity (Fig. 11), electron densities, reflection and energy deposition.

MODEL INPUTS (fused silica; recorded as assumed datasets or in the formulas): ‘We assumed m* = m_e, U_I = 9 eV, and the characteristic phonon energy, U_phon was taken as 0.033 eV’ (pp. 1754-1755), with the electron scattering rates of Arnold, Cartier and DiMaria (ref. 28, Phys. Rev. A 45, 1477 (1992)). The phonon energy has no quantity in the vocabulary. The impact-ionization rate ‘was taken in the Keldysh impact ionization form (ref. 37, Ridley) as 1.5 (epsilon/U_I - 1)^2 fs^-1’ (p. 1753). Four-photon ionization at 526 nm: ‘We used the cross section sigma_4 = 2 x 10^-114 cm^8 sec^3. This was measured for NaCl, but other insulators have nearly the same value (ref. 13)’ (p. 1756); N_s, ‘the solid atom density’, is given no value. Eight-photon ionization at 1053 nm is the fit Eq. (20) to the Keldysh rate. Damage criterion of the model: the critical density ~1e21 cm^-3 at 1053 nm; with 1e19 cm^-3 instead the threshold ‘would decrease by approximately 20%’ (p. 1752).

MODEL RESULTS. The avalanche coefficient alpha (beta = alpha I): flux-doubling estimate ‘between 0.0065 and 0.013 cm2/ps GW’ at 1053 nm, ‘1/alpha lies between 0.08 and 0.16 J/cm2’ (p. 1755); the numerical kinetic solution gives 0.011 cm2/ps GW at 1053 nm (p. 1756, Fig. 11) and 0.013 at 526 nm (p. 1757; the library text reads ‘cm/ps GW’, the pdf text layer ‘cm2/ps GW’); the scale fluence 2/alpha is 0.2 J/cm2 (1053 nm) and 0.15 J/cm2 (526 nm). Other model numbers (notes only): I_b ~ 80 GW/cm2, tau_b ~ 60 ps; p = 0.85 in Eq. (A6); sigma_max E^2 = U_I nu_i(1.5 U_I) near 10 TW/cm2; seed concentrations below ~1e17 cm^-3 unimportant at 10 TW/cm2; the 1e8 cm^-3 ‘lucky electrons’ matter only above ~5 ps; the plasma threshold of the 166-fs calculation is ~1.5 J/cm2 (Figs. 16-18).

REPRODUCED BY: schaffer_2001 (Stuart’s 526/1053 nm factor-two statement), jing_2012 (alpha ~ 0.01 cm2 ps^-1/GW), mao_2004 (Fig. 11 threshold marker, from Stuart ref. 19). Not in the library: Stuart et al., Phys. Rev. Lett. 74, 2248 (1995) (ref. 24) and J. Opt. Soc. Am. B 13, 459 (1996) (cited by the others); Jones et al., Opt. Eng. 28, 1039 (1989) (ref. 13). Du et al., Appl. Phys. Lett. 64, 3071 (1994) (ref. 23) is du_1994 (its thresholds are figures only).

Recorded bibliographic identity status: VALIDATED_WITH_WARNINGS. This concerns identification of the source; it does not establish the scientific correctness of its contents.