Patent Application
20070280305
The present invention will be more fully appreciated with reference to the detailed description and appended figures, in which:
A microchip, Q-switched, cavity-dumped laser array is disclosed. Each laser in the array is end-pumped by VCSEL and comprises an electro-optic Q-switch mechanism actively controlled by photoconductive switches. The fast response time of the system and its small dimension produce short pulses (ten pico-second range), with high energy (uJ range). The microchip structure may be built using only planar, wafer-like components such that a high-density array of lasers may be manufactured without tight alignment tolerances, providing efficient power or energy scaling.
An illustrative laser structure according to an embodiment of the present invention is shown in
According to one embodiment of the invention, the laser cavity may be implemented in a microchip in stacked layers as shown in
The optical pumping section 110 includes a laser that CW pumps the laser cavity section 105 with a beam. According to one embodiment, the laser 145 may be implemented as a high power VCSEL that is integrated with the microchip structure as shown. The optical pumping section may further include a microlens 150 between the VCSEL and the laser cavity 105 that collimates the beam from the VCSEL and passes it through to the laser cavity 105. A microchannel cooler 155 may be implemented within the pumping section 110 to pass the beam and keep the laser 100 from overheating during operation.
To achieve pico-second pulses, a short cavity is needed, which limits the number and size of the various components. According to one exemplary embodiment of the invention, the cavity size may be approximately 1.25 mm, which corresponds to a cavity round-trip time of 18 pico-seconds, and the total size of the structure including optical pumping is 2.2 mm. Such a short cavity and total size for actively Q-switched microchip laser may be obtained through the use of dual function components: for instance, the use of Nd:YVO4 crystal may be used as the gain medium of the laser to realize both the gain medium and a polarizing beamsplitter due to its birefringent properties; the silicon layer may provides the function of a photoconductive switch, as described further below, and output window for laser; the microchannel cooler, located outside the laser cavity 105, may cool both the gain medium and the pumping VCSELs. However, different and even less efficient arrangements may be implemented when longer cavity lengths and pulse times are tolerable.
When implementing the laser 100 in an array, the array structure may be defined by the VCSEL or laser array 145 used to pump and the electrode patterns on the Pockels cell 120. All the other components are wafers, which don't require precise alignment procedures.
The main components of an embodiment of the invention are the polarizing beamsplitter incorporated into the gain medium 115 in some embodiments, the Pockels cell 120 and the photoconductive switch.
The polarizing beam splitter is used to tune the loss of the cavity. Its function is to change the beam path in the cavity depending on the polarization of the light in the cavity 105. In most lasers, the polarizing beam splitter is component is a prism, or a cube beamsplitter. The main disadvantages are that it is a bulk component, usually not planar and thus difficult to integrate into a microchip and a microchip laser array. To overcome these problems, a birefringent crystal may be used as the gain medium. For example, Nd:YVO4, which is a birefringent crystal, may be used as the gain medium of the laser cavity 105. In addition to the lasing properties, the birefringent properties of the crystal may be used to form a beam displacer that act as a beamsplitter. This is shown in
In the cases shown, illustrated as a vertical laser, depending on the initial polarization of the light, the output beam is laterally displaced compared to the input beam. This allows significant simplification in the overall design by allowing a planar component to be used while keeping the cavity size small. Moreover, by combining gain medium and beamsplitter in the same element, element 115, a larger gain medium piece may be used for the same cavity length, which allows the microchip laser to reach higher energy than conventional microchip lasers where the cavity size is limited.
The thickness needed for the medium 115 may be determined based on the displacement required to shift the laser beam by its own size or another desired amount. For example, assuming the cavity produces a Gaussian beam, and using a cavity length of 1.25 mm, the beam size is approximately 75 um. Based on the birefringence properties of YVO4, the thickness may be approximately 750 um to shift the beam by its own size.
The electro-optic Pockels cell 120 is used for Q-switching by controlling the polarization of the laser beam. The properties of the Pockels cell may be summarized as follows. It generally may need to hold high voltage in some embodiments, to provide a uniform electric field through the crystal and maintain a small size. The configuration for the electro-optic material 125 of the pockels cell 120 may be longitudinal, as shown in
The addition of a waveplate 130 in the Pockels cell 120 is optional. When implemented, it may help reducing the driving voltage for the Pockels cell by introducing a fixed amount of polarization loss. It also may tend to prevent the laser from lasing by itself in absence of Q-switch controls. Without a waveplate 130, this may be obtained by using partial reflectivity mirrors 135 for the cavity.
In order to dump the laser cavity, the voltage on the Pockels cell 120 is turned off (or reduced considerably) much faster than the cavity transit time. Thus, the external switch used for turning on the Q-switch mechanism cannot be used for this purpose. Instead we use a photoconductive switch 170, embedded in a semiconductor layer in the path of the output beam.
In a preferred embodiment, the photoconductive switch 170 is made within the silicon layer 165. The thickness of the silicon layer and the depth of the doping is determined by the absorption of the silicon at the lasing wavelength and the desired triggering level. However, it is understood that other materials than silicon can also be used to provide the same function such as GaAs or other semiconductor materials. The material choice is dictated by the desired operating wavelength.
The new voltage on the Pockels cell is governed by the size of the second capacitance compared to the first one. The ratio can be adjusted by changing the area of the electrodes, the dielectric constant of materials used or any other factors affecting capacitance.
The cavity dumping process is based on the charge transfer from one capacitance to another. This process may be further enhanced by charging the first capacitance with a voltage V and the second capacitance with a voltage −V using the external Q-switch. Thus, when the photoconductive switch is closed, the net charge on the first capacitance becomes zero, providing a better modulation in the Q factor of the cavity. Also, this method uses two capacitances of the same size, which may reduce the cross section of a laser. Denser arrays may be built with this method. In the case described here, the cross-section of a single laser is 100×600 um. With such as design, more than 30,000 lasers can be fitted on a two-inch diameter structure.
The operation of the laser array is described as follows: an external trigger puts voltage on the Pockels cells and Q-switches all lasers in the array at the same time. Buildup of the laser beams start. For each laser, a small part of the lasing beam is directed to the photoconductive switch via polarization loss, which is controlled by the applied voltage and the retardation of the waveplate. When amplification of the beam reaches intensity sufficient to trigger the photoconductive switch, the voltage on the Pockels cell drops very fast. Thus, polarization losses become predominant which dumps the cavity along the displaced output beam path and through the output beam opening.
The above description constitutes a preferred embodiment of the invention. Many variations are within the scope of the invention. The following are illustrative examples. The Pockels cell shown and described may be made using an electro-optic material in a transverse configuration. The advantage of such a Pockels cell, despite the need for a more complicated electrode design, is that it requires a lower operating voltage than that of a longitudinal configuration.
In another embodiment, an equivalent of the electro-optic Pockels cell may be realized with magneto-optic or Faraday effect material such as bismuth substituted rare earth iron garnet. Due to their strong Faraday effect, these materials are well suited to build a compact structure acting on the polarization state of the light in the cavity. It will be further understood that the dual purpose gain medium may be implemented with any convenient birefringent gain medium, such as Nd:YVO4, Nd:YAG or any other convenient material depending on design considerations. Additionally, the microlens array located inside the cavity 105 may not be implemented and one may instead use the thermal lensing due to the heating of the gain medium to produce a stable cavity. Reference is made to U.S. Pat. No. 5,386,427 to Zayhowski for this purpose. Additionally, the pumping VCSEL array may be implemented as a diode array or any other convenient photon generating source.
Additional components may also be added to the structure. Other components that may be added to the system are unlimited, but include nonlinear crystals for frequency conversion, such as harmonics generation or optical parametric oscillation. Reference is made in this regard to J. J. Zayhowski, “Microchip optical parametric oscillators”, IEEE Photonics Technology Letters, Vol 9, pp 925-927, 1997.
In another embodiment, the laser structure may be modified to ensure coherence of all the lasers in the architecture. This may be accomplished by seeding from CW lasers 410 located on the periphery of an array of lasers 100. The seed is distributed to the lasers through a waveguide structure 400 located on top of the cavity, and enters the lasers through partial reflectivity mirrors. This is shown in the top and side views of
It should be noted that the photoconductive switch in each laser structure allows dumping the cavity when the lasing beam reaches a given level. However, all the Pockels cells electrodes may be linked together in some array configurations to realize a synchronized design. In this scenario, when a photoconductive switch discharges a cell it also helps to trigger its neighbors. The result is that all the lasers tend to be synchronized. This synchronization, when used with the seeding described, contributes to the coherence of all lasers. While the synchronization may be realized in the manner just described, lasers also may be implemented with electrodes that are interconnected or not interconnected, such that each laser is driven and discharged independently from one another.
While particular embodiments of the present invention have been described herein, it will be understood that changes may be made to those embodiments without departing from the spirit and scope of the present invention.
