WAVEFRONT CONTROL FOCUS
Excessive pulse energy may decrease ablation efficiency, induce unwanted surface modifications or generate cracks and nonlinear propagation effects in transparent materials. Similarly, increasing the repetition rate can lead to uncontrolled heat accumulation when the material does not have sufficient time to cool down between consecutive pulses.
A straightforward solution is to attenuate the laser beam, but simply discarding the available energy is hardly attractive for an industrial process. The challenge has therefore progressively shifted from producing more laser power to using the available photons more efficiently. This is where spatial beam shaping provides an answer: instead of removing excess energy, it redistributes it spatially. A single Gaussian spot can be transformed into several processing spots, a top-hat profile, a line, a ring or more complex structured beams. The laser energy can thus be adapted to the process while preserving micrometric resolution and increasing throughput.
HOW TO SHAPE THE WAVEFRONT? Spatial beam shaping relies on a simple principle: modifying the phase of a laser beam in one plane changes its intensity distribution after propagation, particularly in the focal plane of a lens. Contrary to amplitude shaping, where part of the incident energy is deliberately blocked, phase-only modulation can theoretically redistribute nearly all the available energy into the desired pattern. This makes wavefront shaping particularly attractive for high-power ultrafast laser processing.
Several technologies can dynamically control the wavefront( see Figure 1 a-c). Liquid-crystal spatial light modulators( LC SLMs) provide the highest flexibility, with millions of individually controlled pixels and typically a 2-π phase modulation range. They can generate almost arbitrary phase masks, but their response time and laser damage threshold remain limited, particularly at high average power and repetition rate. Deformable mirrors( DMs) offer much better power handling, but their smaller number of actuators limits the spatial complexity of the achievable wavefront. Acousto-optic deflectors( AODs) provide much faster modulation, at the expense of lower shaping flexibility and additional chromatic effects. Several other devices like digital micro mirror devices( DMD) or the tunable acoustic gradient TAG-lens with various advantages have been employed as well with various performances [ 2 ].
The second challenge is to determine the appropriate phase mask. One of the most widely used approaches is the Iterative Fourier Transform Algorithm( IFTA), also known as the Gerchberg – Saxton algorithm [ 3 ]. By successive numerical propagation of the optical field between the modulation and target planes( see Figure 1 d), imposing their respective constraints, the retrieved phase can be calculated and then displayed on the modulator to produce the target intensity distribution.
However, the calculated and experimental distributions never perfectly match. Finite pixel size, limited phase range, fill factor, optical aberrations and imperfect calibration of the modulator introduce errors. In SLMs, the unmodulated fraction of the incident light may also generate an unwanted zero order, while phase discontinuities associated with continuous beam shaping can produce speckle. Both effects reduce shaping fidelity and, importantly for laser processing, waste part of the available energy.
Several strategies can mitigate these limitations. Aberrations can be measured and directly compensated within the phase mask, while iterative algorithms can introduce additional constraints or unconstrained regions to improve convergence. Most importantly, experimental feedback can be incorporated into the phase mask calculation step, allowing the calculated mask to compensate for imperfections of the complete optical system. Yet improving uniformity or suppressing unwanted light generally redistributes energy outside the useful region. Dynamic wavefront shaping therefore always involves a compromise between high spatial fidelity and high energy efficiency.
Photoniques 140 I www. photoniques. com 45