Dual chipsan oscillator and an amplifiercombine to form a low-cost passively Q-switched microlaser that operates at high power and repetition rate.
Daniel Guillot
Passively Q-switched diode-pumped microchip lasers are monolithic, small, simple, and low in cost. The lasers can be manufactured in high volume and produce a high-quality linearly polarized beam useful in various fields such as medical, semiconductor equipment, analytical instrumentation, ranging, and biotechnology; higher harmonics can be generated from such a beam for green or ultraviolet (UV) output. However, obtaining short pulses, high energy per pulse, and high repetition rate simultaneously in a microchip laser is outside the capability of a device based on a single chip.
A microchip laser architecture has been developed to solve this problem. Based on a dual-chip design, the laser can produce 400-ps pulses with tens of kilowatts peak power at a repetition rate of 50 kHza twentyfold increase over single-chip designs.
High power, high repetition rate
Production of short pulses with high energy per pulse is usually achieved using a combination of one oscillator and one amplifier. The oscillator is traditionally a modelocked laser producing very short pulses (typically less than 100 ps) at high frequency (typically a few tens of megahertz) and with a low energy of a few nanojoules per pulse. To increase the pulse energy to several microjoules, an amplifier is used that operates at a lower repetition rate; depending on the pumping configuration, the rate can range from a few kilohertz to a few hundreds of kilohertz. These systems are intricate and complicated to use, requiring active modulation with high-speed electronics to enable short-pulse production from the oscillator, as well as injection and synchronization of the pulses inside the amplifier.
Microchip lasers can be engineered to favor one of the following performance factors: energy per pulse, pulsewidth, or repetition rate. The parameters available to design a laser for a desired output are the cavity length, the saturable-absorber thickness and doping level, the mirror transmission, and the pumping power and density. Most parameters are set at the time of the chip wafer fabrication, but some flexibility remains for optimization through the various means of pumping. Laboratory experiments demonstrate that, for a given microchip, there is a quasilinear dependency of the repetition rate with the pump source brightness (see Fig. 2).
In contrast, the oscillator in a dual-chip amplified microlaser directly produces microjoule pulses at the required repetition rate. These pulses are amplified in only a few passes in a totally passive amplifier, resulting in a great reduction in complexity (see Fig. 1).
Double-pass high-gain amplifier
In the dual-chip design, the output from a microchip injection laser having the desired high frequency is subsequently amplified to increase its energy per pulse. The amplifier crystal is continuously pumped by the diode; there is no electronic synchronization to the injected beam frequency, contrary to what is required with an amplified modelocked laser.
The amplifier crystal is Nd:YVO4 (vanadate), chosen for its large stimulated-emission cross section, its high absorption at 808 nm, and its fluorescence lifetime well suited for amplifying beams having a repetition rate of tens of kilohertz. The pumping of the amplifier crystal is optimized to ensure a good overlap of the pump beam and the microchip laser beam in an end-pump configuration. The selected pump power allows enough energy storage in the amplifier while avoiding thermal effects (due to the focusing of the pump beam) that would degrade the performance.
The quality of the coupling of the pump beam into the amplifier is very important, as it defines the output-beam quality. Best results are obtained with a combination of an aspheric lens, anamorphic prisms, and an objective. The microchip laser beam makes two passes through the amplifier crystal, and two parameters are optimized for best performance: parallelism between the microchip-laser beam polarization and the direction in which the stimulated-emission cross section is the largest and parallelism between the pump-diode beam polarization and the direction of the largest absorption into the microchip.
With the microchip laser providing the high repetition rate and the amplifier supplying the energy, the architecture is simple, resulting in a compact package that does not require any electronic synchronization. The excellent beam properties of the microchip injection laser are unchanged after two passes through the amplifier. The resulting infrared (IR) output is made of TEM00 subnanosecond pulses having a peak power in excess of 10 kW.
Green and UV output can be efficiently generated from the IR beam. Such high efficiency permits the placement of the nonlinear crystals outside the cavity in a single-pass configuration, providing good stability and beam quality with a simple, efficient and compact design. Sliding the appropriate nonlinear-crystal combination directly in front of the focused beam can generate second, third, fourth, and fifth harmonics simply.
Choice of performance
The amplifier works in a saturated mode. For a given pumping configuration, the average power output of the amplifier stays almost the same (500-600 mW) for repetition rates in the 25-50-kHz range. By properly selecting the injected microchip laser properties (pulsewidth, frequency, energy per pulse), it is possible to tailor the laser to a desired set of parameters.
The microchip technology permits customization of the injection laser performance in two ways. First, the microchip pumping configuration allows some flexibility over a small range to favor either maximum energy per pulse or higher frequency output. Second, the microchip cavity design allows for a larger range of adjustment, as well as for the possibility of producing different pulse widths.
With a 50-kHz microchip, for instance, 12-15 µJ per pulse can be obtained at the output of the amplifier. With a 30-kHz higher-energy injection microchip, the output can reach 20 µJ. Pulses as short as 300 ps and pulse-repetition rates as high as 80 kHz have been demonstrated in the laboratory in various microchip-laser configurations.
Applications abound
The uses for a short-pulse, high-peak-power, high-repetition-rate laser with high average output power are many. The inclusion of output wavelengths ranging from IR to deep-UV further expands the application of this sort of laser.
In materials processing, good beam quality combined with high-peak-power repetition rate is ideal for micromarking at high speed and resolution. Due to the short pulsewidth of the dual-chip design, the heat-affected zone created during machining is much smaller, reducing collateral damage on the work piece. The amplified microchip laser has been shown to photoablate most absorbing materialsincluding metals, semiconductors, glasses, polymers, plastics, and biological tissuesat a 25-µm or smaller feature size. With greater than 50 mW of UV output, average power is high enough to replace older helium-cadmium laser technology in photopolymerization applications with a laser that is one-third the size.
In medical and biological applications, the short pulses and multiple wavelengths are well suited to laser-induced fluorescence measurements either in a quasicontinuous-wave mode or in a time-decay mode. With 300-ps UV pulses, very short fluorescence decays can be measured at a 50-kHz repetition rate, enabling high-speed flows of biological samples to be examined. Moreover, because the beam can be focused down to the diffraction limit and injected into a fiber, measurements from remote locations are possible.
Passively Q-switched microlasers are already used in ranging applications in construction, in architecture for mapping the facades of buildings, and in robotics for position control of spinning disks. By boosting the repetition rate while maintaining the same pulsewidth and beam quality, scanning measurements useful in boat navigation-guidance systems and airborne altimeters can now be developed. Remote sensing and environmental monitoring can draw on these same benefits but in the UV, which is commonly the optimum measurement wavelength for these applications.
In semiconductor instrumentation, the high available UV power from an amplified microchip laser combined with small spot sizes resulting from tight focusing improves high-resolution measurements; an added future fifth-harmonic capability will push the resolution limit further. The laser is also an excellent source for generating time-resolved and UV Raman spectra for on-line process control, because the fluorescence that plagues most Raman measurements is far-removed from the signal. The amplified microchip laser can replace air-cooled argon-ion, helium-cadmium, pulsed-nitrogen, and actively Q-switched YAG lasers in many applications and is a simple answer to the need for short pulses at high repetition rate.
DANIEL GUILLOT is president of Nanolase, 31 Chemin du Vieux Chene, Zirst 4101, 38941 Meylan, France; e-mail: [email protected].