- Biomedical Optics & Medical Imaging
- Defense & Security
- Electronic Imaging & Signal Processing
- Illumination & Displays
- Lasers & Sources
- Micro/Nano Lithography
- Optical Design & Engineering
- Optoelectronics & Communications
- Remote Sensing
- Sensing & Measurement
- Solar & Alternative Energy
- Sign up for Newsroom E-Alerts
- Information for:
Lasers & Sources
The effects of waveguiding on ultrafast-laser drilling rates
A new technique capable of single-shot spatial measurements across a laser beam captures the impact of optical coherence on penetration rates.
15 November 2006, SPIE Newsroom. DOI: 10.1117/2.1200611.0469
When a laser is used to drill holes into metals, the etch rate decreases with depth: as the hole deepens, penetration slows, as shown in Figure 1, until it stops at some ultimate depth.1 This phenomenon cannot be explained entirely as due to sidewall absorption, which ceases once wall ablation reaches some threshold level.
Figure 1. The graph plots the rate of deepening for a laser-drilled hole in aluminum, in microns per pulse. These estimates are based on the time it takes to drill through aluminum foil of various thicknesses.
We have found that there are optical reasons for this decreased drill rate, such as light scattering and dispersion. These arise because the hole acts as a cylindrical multimode waveguide, which typically disperses the modes. Irregularities in the hole wall introduce phase noise from the modal boundary condition, which progressively scatters the beam. This scattering leads to transverse coherence loss, which affects laser intensity in the propagated beam.
Transverse coherence describes the consistency of phase evolution at different points across the beam. It often depends in a simple way on point separation in a plane.2,3 The degree of coherence can be measured from the visibility (contrast) of fringes that result when intensity-equalized beams, or parts of the same beam, interfere. This may be usefully characterized by a unitless number, ‘global coherence,’ which is the transverse coherence width normalized to the beam's transverse intensity width. To estimate these widths, we map the coherence across the beam after it has exited from the drill hole. In this article, we discuss our method for measuring this map.
While Young's double-slit setup it often used to measure transverse coherence, it is ill-suited for our purposes. Estimating the global coherence with this approach would be slow because it provides only one measurement at a time, and it can be unsuitable for nonrepeatable measurements.
To address these shortcomings, we devised a method based on a folded Michelson interferometer, depicted in Figure 2, that can measure spatial coherence for a single pulse at all transverse separations. In this approach, the interference fringes between the original beam and a left-to-right flipped copy are captured on a digital camera (CCD-COH in Figure 2), and then corrected for spatial beam intensity variations using images of the two beams captured on another digital camera (CCD COH-REF). Each point of the flipped beam interferes with a corresponding portion of the original beam on the other side of the center line. Thus, taken as a whole, the intensity-corrected fringe image can be used to make a coherence map plotted against vertical distance and the transverse separation of the points from the center line (see Figure 3). This is not possible using conventional transverse-coherence measurement techniques.
Figure 2. This schematic illustrates the folded Michelson interferometer used to measure transverse coherence. The images were produced while shooting through 150μm of aluminum. BS1, BS2: Beamsplitters. CCD-COH, CCD COH-REF: Digital cameras used to measure transverse coherence and reference images, respectively. M1, M2, M3: Mirrors. ND: Near-field camera. P: Linear polarizer.
Figure 3. Shown is a map of spatial coherence as a function of position on CCD-COH. The black dots correspond to fringe locations where the coherence was measured. The colors indicate the calculated degree of coherence.
An ultrafast pulsetrain-burst (133MHz repetition rate, 1–10ps pulse length) laser was used in this experiment.4 This kind of system can machine very smooth holes with very little cracking or flaking.5 Before-target and after-target photodiodes were used to measure the time taken to drill the channel. The transverse coherence, coherence, and reference images were captured on each shot as discussed above, together with measurements from a Young's double-slit setup for comparison. We used aluminum foil targets with thicknesses ranging from 25 to 150μm.
While a coherence map may be produced for each shot, it was easier for comparison purposes to use horizontal cross-sections of the map, such as those shown in Figure 4, which can also be used to estimate the global coherence. It is clear from this figure that transverse coherence drops as separation increases, illustrating that the spatial coherence does, indeed, decrease as the beam drills through the channel.
Figure 4. Coherence variations with transverse separation are shown for drilling through 150μm of aluminum. The black lines are cross-sections from a coherence map, and the blue dashed line corresponds to no target. The red dots represent measurements from a Young's double-slit setup.
Applying this method over a range of metal target thicknesses shows that the transverse coherence drop-off increases with hole depth as a channel is drilled, which is consistent with the observed decrease in etch rates for deeper holes. This transverse-coherence information can be correlated with the divergence angle of the modified beam after it exits the channel.
The impact of drilling-related optical waveguiding in other materials, such as glass, may reveal a specific dependence on the native material in which the waveguide boundary conditions are set up.
The authors acknowledge support from the Canadian Institute for Photonic Innovations, the Centre for Photonics of Ontario Centres of Excellence Inc., and the Natural Sciences and Engineering Research Council of Canada.
Jesse Dean, Martin Bercx, Marc Nantel, Robin Marjoribanks
Department of Physics, University of Toronto
Toronto, Ontario, Canada
Jesse Dean is a first-year PhD student at the University of Toronto. He wrote the paper ‘Optical coherence and beamspread in ultrafast-laser pulsetrain-burst hole drilling,’ and presented that poster for Photonics North 2006.
Centre for Photonics, OCE Inc.
Mississauga, Ontario, Canada
1. D. Breitling, C. Fohl, F. Dausinger, T. Kononenko, V. Konov, Drilling of metals,
Femtosecond Technology for Technical and Medical Applications, Topics Appl. Phys. 96,
pp. 131-156, Springer, Berlin, 2004.
5. L. McKinney, F. Frank, D. Graper, J. Dean, P. Forrester, M. Rioblanc, M. Nantel, R. Marjoribanks, Mitigating intrinsic defects and laser damage using pulsetrain-burst (>100MHZ) ultrafast laser processing,
Proc. SPIE 5970,
pp. 436-449, 2005.