Q-switched solid-state laser

Description

[0001] The invention relates to a Q-switched solid-state laser with a resonator having both a passive Q-switch and an active Q-switch.

[0002] Q-switches ("O-switches") are used to generate pulsed laser radiation. Compared to purely pulsed excitation, Q-switches generate shorter pulses. Peak powers can also be higher. Q-switches are known as active and passive Q-switches. Active Q-switches have the disadvantage that they require dedicated control electronics. Furthermore, the minimum resonator length is limited by the size of these Q-switches. Very short resonators are required, for example, if the laser is to be operated with a single longitudinal mode. Passive Q-switches eliminate the need for electronic control. Passive Q-switches can also be made so small that single-mode operation is possible. A disadvantage of passive Q-switches for certain applications, however, is that they exhibit greater jitter, i.e.the exact time of triggering varies from one pulse to the next.

[0003] A passively Q-switched solid-state laser with high pulse energy is described, for example, in Bhandari R. and Taira T., "Palm-top size megawatt peak power ultraviolet microlaser," Optical Engineering, July 2013 / Vol. 52(7), 076102-1 - 076102-6. Nd:YAG is used as the active laser material, and an absorber formed from a doped solid-state material in the form of Cr:YAG serves as the passive Q-switch. With a resonator length of 10 mm, a "free spectral range" is achieved—that is, a separation between the wavelengths of possible neighboring longitudinal modes—that is large enough that only one of the longitudinal modes is at the gain maximum, causing it to oscillate. This laser achieves a pulse energy of 3 mJ with a pulse width of 230 ps and a repetition rate of 100 Hz. The wavelength of the laser is 1064 nm. The resonator is designed to be optically stable ("stable resonator").The design of the resonator can be monolithic, with both end mirrors being planar and a thermal lens with a collecting effect being formed by the active laser material during operation.

[0004] Passively Q-switched solid-state lasers with single longitudinal mode operation and, in contrast, much smaller pulse energies in the range of a few tens of nJ are known in the form of very small monolithic microchip lasers with Nd:YVO4 as the active laser material, see, for example, WO 2016 / 004446 A1. The resonator lengths can be less than 150 µm, and pulse lengths of less than 100 ps can be achieved. These resonators are also stable with flat end mirrors and the formation of a thermal lens during operation.

[0005] Lee HC et al., "High energy, sub-nanosecond linearly polarized passively Q-switched MOPA laser system," Optics and Laser Technology 95 (2017) 81-85, describes a passively Q-switched Nd:YAG / Cr:YAG laser with a stable resonator whose output pulses, which have a pulse energy of approximately 2.5 mJ and a pulse duration of approximately 550 ps, ​​are amplified. The active laser material and the absorber have opposing surfaces arranged at the Brewster angle on their facing sides to effect polarization selection of the laser beam.

[0006] Known passively Q-switched solid-state lasers with optically unstable resonators feature significantly longer resonator lengths with a multitude of longitudinal modes. Such known lasers can achieve high pulse energies with relatively low tilt sensitivity of the optical elements of the resonator and relatively low sensitivity to changes in the thermal lens. However, the intensity curve of the emitted pulses exhibits oscillations and spikes due to the superposition of the various modes. Such lasers with an unstable resonator are disclosed, for example, in WO 2014 / 019003 A1, EP 3 117 494 A1, or WO 2015 / 021488 A1.

[0007] Active Q-switches are known in the form of acousto-optic modulators (AOMs) and electro-optic modulators (EOMs), particularly Pockels cells. Pockels cells, for example, use beta-barium borate (BBO). Acousto-optic modulators achieve short switching times with small beam radii, while electro-optic modulators achieve similarly short switching times even with larger beam radii. The use of an electro-optic modulator also requires a polarizing element in the resonator.

[0008] Stafsudd OM et al., "CO Laser with Simultaneous Active and Passive Q-Switching," APPLIED OPTICS, Vol. 10, No. 1 (1971), pp. 141-143, describes a CO2 laser that uses a slowly rotating mirror for active Q-switching. The occurrence of multiple pulses is observed, which can be eliminated by additionally introducing a saturably absorbing gas, which forms a passive Q-switch. At slow repetition rates, this also allows for a shortening of the pulse duration and an increase in the maximum intensity.

[0009] From WANG, Xuejun et al., "Single-longitudinal-mode Operation of a 1 W combined actively and passively Q-switched Cr,Nd:YAG Laser," September 5, 2005 (05.09.2005) OPTICS EXPRESS 6697 Vol. 13, No. 18, pp. 6693-6698, a solid-state laser of the type mentioned above with an active and a passive Q-switch emerges. The resonator length was minimized to a value of 66 mm in order to maximize the "free spectral range," i.e., the frequency separation of adjacent longitudinal modes. In conjunction with the use of the passive Q-switch, which serves in a known manner to select a defined longitudinal mode, operation with a single longitudinal mode can be achieved. The active Q-switch modulates the pulses, which thus exhibit relatively low jitter.The passive Q-switch is used to solve the problem of providing single-mode operation, while the active Q-switch solves the problem of generating pulses with low jitter. Similar solid-state lasers are also described in CN 1645691 A and CN 104158082 A. The former document mentions both lasers with stable resonators and those with unstable resonators.

[0010] A microchip laser with combined active and passive Q-switching is known from Arvidsson M et al., "Combined actively and passively Q-switched microchip laser," SPIE PROCEEDINGS, Vol. 3265, May 27, 1998, page 106.

[0011] The object of the invention is to provide an improved Q-switched solid-state laser of the type mentioned above, which, while operating stably with only a single longitudinal mode, produces very short pulses with low jitter. According to the invention, this is achieved by the features of claim 1.

[0012] According to the invention, the loss caused by the active Q-switch in the activated state, based on a forward and return path through the linear, folded or unfolded, resonator, is less than 50%, preferably less than 30%. Furthermore, the resonator has a resonator length of less than 50 mm, preferably less than 25 mm. According to the invention, a solid-state laser is thus provided with such a short resonator that it has only a single longitudinal mode, generating very short pulses and triggering the pulses at precisely defined times.

[0013] A basic idea of ​​the invention is that the loss caused by the active Q-switch can be kept relatively low through interaction with the passive Q-switch, whereby the active Q-switch can be designed with a small overall size. The overall length occupied by the active and passive Q-switches together in the resonator can be significantly shorter than the overall length of conventional active Q-switches, which are designed with a high loss in the activated state, for example, more than 99%. This allows the resonator length and consequently also the pulse length to be further shortened. By using the active Q-switch, even if it only has a relatively low loss, in addition to the passive Q-switch, the time of triggering the laser pulse can still be precisely determined, whereas using a passive Q-switch alone would result in significantly greater jitter.

[0014] According to the invention, the maximum loss caused by the passive Q-switch (when the absorber is completely unsaturated) is more than 50% and is thus advantageously greater, for example, more than twice as great as the loss caused by the active Q-switch in the activated state, which according to the invention is less than 50%. For example, the maximum loss caused by the passive Q-switch can be 90%, while the loss caused by the active Q-switch in the activated state can be 10%.

[0015] When talking about the loss of the active or passive Q-switch, this loss refers to the outward and return paths combined in the case of a linear resonator (i.e., a double pass), while in the case of a ring resonator, it refers to one revolution around the resonator (i.e., a single pass).

[0016] According to the invention, the pulse energy is more than 1 mJ, preferably more than 10 mJ. According to the invention, the resonator is designed as an unstable resonator, wherein the resonator has a mirror in the form of a gradient mirror.

[0017] According to the invention, the pulse duration is less than 500 ps.

[0018] Further advantages and details of the invention are explained below with reference to the accompanying drawings, in which: Fig. 1 a first embodiment of the invention; Fig. 2a ) - e) and Fig. 3 a) - c ) Diagrams explaining the operation of the laser; Fig. 4 a second embodiment of the invention; Fig. 5 a third embodiment of the invention; Fig. 6 a fourth alternative example not falling within the scope of the present invention; Fig. 7 a fifth alternative example not falling within the scope of the present invention.

[0019] A first embodiment of the invention is shown schematically in Fig. 1 It is a solid-state laser with an active laser material 1, which is formed, for example, from Nd:YAG. The doping can be, for example, 1.1%.

[0020] The active laser material 1 is arranged between a first and a second mirror 4, 5, which form the end mirrors of the resonator 3. A passive Q-switch 2 and an active Q-switch 8 are arranged between the mirrors 4, 5.

[0021] In the exemplary embodiment, the resonator 3 comprises the mirrors 4, 5, the active laser material 1, the passive Q-switch 2 and the active Q-switch 8.

[0022] The passive Q-switch 2 is formed by a saturable absorber, which is designed in particular as a doped solid-state material, in the exemplary embodiment Cr:YAG (more precisely Cr 4+< :YAG).

[0023] In the exemplary embodiment, the active Q-switch 8 is formed by an electro-optical modulator, in particular a Pockels cell which comprises, for example, beta-barium borate.

[0024] In the exemplary embodiment, the first mirror 4 also serves to couple the Fig. 1 The pump radiation 6, indicated only schematically by an arrow, is directed so that the active laser material 1 is pumped longitudinally (= pump radiation propagates along the same axis as the laser radiation). For this purpose, the first mirror 4 is dichroically coated, making it at least largely transparent to the pump radiation 6 (T > 95%), while being as highly reflective as possible for the wavelength of the laser beam (R > 99.9%) and therefore less transparent. For example, the wavelength of the pump radiation is 808 nm.

[0025] The wavelength of the laser light in the embodiment is 1064 nm.

[0026] The pumping device itself is in Fig. 1 Not shown. It can be a conventional fiber-coupled laser diode with QCW ("quasi-continuous wave") power of, for example, up to 150 W (for a 250 µs pulse duration), operating at a pulse repetition rate in the range of 0-100 Hz. The beam diameter of the pump radiation can be approximately 2.5 mm, for example.

[0027] The resonator 3 is a linear resonator or standing wave resonator. The term "linear resonator" also includes those in which the beam path in the resonator is bent by at least one reflection within the cavity of the resonator, thus also folded resonators. A different functional principle, however, is the ring resonator, in which the radiation circulates. The length of the resonator from mirror 4 to mirror 5 in this embodiment corresponds to the length measured along the axis 7 of the laser beam in the resonator (= resonator length a) from one mirror 4 to the other mirror 5. This resonator length a is less than 50 mm, preferably less than 25 mm. If the laser beam were reflected in the resonator (as, for example, in the embodiments according to the Fig. 4 and 5), the "unfolded" length of the resonator would have to be used as the resonator length a, i.e. again the length of the resonator measured along the axis of the laser beam from one mirror to the other mirror.

[0028] The first mirror 4 is designed as an optical element with negative refractive power. The focal length thus has a negative value, with the focal length being less than 500 mm, preferably less than 250 mm, and particularly preferably less than 150 mm. A beam incident on the first mirror 4 is thus reflected with a high divergence.

[0029] The second mirror 5, through which the laser radiation is coupled out, is a gradient reflectivity mirror (also called a variable reflectivity mirror). Such gradient mirrors are known. The central reflectivity of the gradient mirror can advantageously be in the range of 30% to 60%, e.g., 40%. The radius at which the reflectivity of the gradient mirror has dropped to e-2 can advantageously be in the range of 0.5 mm to 2 mm, e.g., 0.75 mm.

[0030] The reflectivity curve of the gradient mirror (=reflectivity as a function of radius) can in particular have the shape of a Gaussian curve or a super-Gaussian curve.

[0031] Resonator 3 is therefore optically unstable. The magnification of the resonator is >1.2; in the exemplary embodiment, it is approximately 2. The magnification is a measure of the geometric spread of the circulating rays in the unstable resonator, i.e., at M=2, the geometric rays of the eigenmode of the unstable resonator are further away from the axis by a factor of 2 during one orbit.

[0032] The total refractive power of the resonator (= the combined refractive powers of the optical elements for a forward and return path through the resonator 3, with each of the mirrors 4, 5 being included only once) in this exemplary embodiment essentially corresponds to the refractive power of the first mirror 4, since the refractive power of the thermal lens formed by the active laser material 1 during operation is significantly lower. The total refractive power of the resonator is thus negative, with its absolute value being greater than 2 dpt, preferably greater than 4 dpt. For example, the total refractive power can be in the range from -5 dpt to -15 dpt.

[0033] The length a of the resonator 3 is tuned such that a longitudinal mode of the laser radiation lies precisely at the gain maximum of the active laser material 1, in the exemplary embodiment at 1064 nm. For this purpose, the laser is expediently also temperature-stabilized. Due to the short length of the resonator 3, the "free spectral range" is so large that the neighboring longitudinal modes lie so far outside the gain maximum that they at least essentially do not oscillate (i.e., their energy is less than 10% of that of the fundamental mode), and in particular, do not oscillate at all (i.e., the laser threshold is not exceeded). The width of the gain maximum of the active laser material is also correspondingly small for this purpose. Thus, during operation, the laser essentially has only a single longitudinal mode. Furthermore, during operation, the laser essentially has only a single transverse mode.The design as an unstable resonator favors the fact that higher-order transverse modes, at least for the most part, do not oscillate. Thus, the laser is a single-mode laser with respect to both transverse and longitudinal modes.

[0034] By designing the laser with a high negative refractive power, a large mode size can be achieved despite the short laser length. Advantageously, the beam radius of the laser beam in the active laser material 1 is greater than 500 µm. This allows for high pulse energies to be achieved without causing optical destruction. The pulse energy is preferably greater than 10 mJ.

[0035] The use of Cr:YAG as a saturable absorber can have a stabilizing effect. The absorber initially saturates at the antinodes of the main mode (= "hole burning"). This causes higher absorption for secondary modes because they increasingly have their nodes in a non-saturated region.

[0036] The inlet and outlet surfaces of the active laser material 1 and absorber 2 are advantageously anti-reflection coated for the wavelength of the laser beam.

[0037] In this embodiment, the active laser material 1 and the passive Q-switch 2 are each cut and polished in a so-called "flat-brewster" configuration. The active laser material 1 and the passive Q-switch 2 thus each have a Brewster angle on their mutually facing sides, with these sides being at least substantially parallel to each other, and the opposite sides being perpendicular to the axis 7 of the laser beam. The flat side is advantageously coated with an anti-reflection coating for the laser wavelength (optionally also for the wavelength of the pump radiation). The Brewster surfaces are not necessarily coated. The two Brewster surfaces allow the p-polarization to pass through unhindered, but result in a loss for the s-polarization, so that this at least substantially does not oscillate (i.e.The energy is less than 10% of that of p-polarization), and in particular, it does not oscillate at all (i.e., the laser threshold is not exceeded). This forces the laser to operate at least essentially in p-polarization.

[0038] To avoid an etalon effect for the s-polarization, which could reduce or eliminate the losses for the s-polarization, the distance between the two Brewster surfaces must be precisely chosen so that there is no high transmission of the s-polarization for the wavelengths of those s-polarized modes that are close to the gain maximum of the active laser material 1. Instead or in addition, a slight mutual tilt of the Brewster surfaces can also be provided.

[0039] The active laser material 1 and the passive Q-switch 2 can be monolithic, in which case a dielectric coating is provided between the Brewster surfaces (similar to the well-known "polarizing cubes").

[0040] For example, the transmission of the absorber forming the passive Q-switch 2 can be approximately 30% in the unsaturated state. This results in a transmission of approximately 10% over the forward and return path through the resonator, so that the loss caused by the absorber in the unsaturated state is approximately 90%.

[0041] As already described above, the Brewster surfaces of the passive Q-switch 2 and the active laser material 1 have a polarizing effect at least so strongly that at least essentially only one of the polarizations can oscillate in the resonator, in the exemplary embodiment the p-polarization. If the active Q-switch 8 is activated by applying a high voltage, the polarization of the laser radiation is changed so significantly as it passes through the active Q-switch that, in conjunction with the Brewster surfaces, a loss also occurs with respect to the polarization direction, which can pass unhindered through the Brewster surfaces (in the exemplary embodiment the p-polarization). This loss introduced in the activated state of the active Q-switch 8 (relative to the outward and return paths taken together) can, for example, be in the range of 10%.

[0042] The active Q-switch 8, formed in the exemplary embodiment by a BBO Pockels cell, can be, for example, 3 mm to 7 mm long and have a cross-section of 3 x 3 mm. The surfaces can be gold-plated to apply the high voltage across a large area. The switching voltage can, for example, be in the range of 1000 V to 3000 V.

[0043] Since the active Q-switch 8 only needs to cause a relatively slight change in polarization to achieve the desired, relatively small loss of, for example, 10%, a small size of the active Q-switch 8 is possible.

[0044] The operation of the laser is explained below using the diagrams of Fig. 2 and 3 explained.

[0045] Fig. 2 a) shows a pulse of the applied pump power (the power P is plotted against time t). For example, the pump power is applied for 250 µs (since the storage time for Nd:YAG is around 240 µs).

[0046] In Fig. 2 Only one pulse of the pump power is shown. Such pulses are repeated periodically at the desired laser pulse frequency, e.g., at a frequency in the range of 10 to 100 Hz.

[0047] Fig. 2 b) shows the structure of the inversion (I). This follows a charging curve, as the inversion is continuously reduced by spontaneous emission. The gain increases accordingly.

[0048] In Fig. 2 c) The laser power is shown (plot of power P versus time t) as it would result without an active Q-switch. The laser pulses of three pump cycles are shown. Due to noise in the pump power, vibrations, pump wavelength inaccuracies from pulse to pulse, instabilities in the laser diode current, etc., these laser pulses are not fired at exactly the same time within the pump pulse (whose time window is Fig. 2 c) schematically represented by a dashed line). Rather, a jitter typically occurs in the range of, for example, 1-2 µs from pulse to pulse.

[0049] Fig. 2 d) shows the situation as it would occur with an active Q-switch that remains activated throughout the entire duration of the pump pulse and beyond. This means that the high voltage is applied to the electro-optical modulator before the start of the pump pulse and is maintained until after the end of the pump pulse. Fig. 2 d) It can be seen that due to the additional loss introduced by the active Q-switch in its activated state, the triggering of the laser pulse is shifted toward the end of the pump pulse, although the jitter still occurs. Depending on the design, the triggering of the laser pulse could also be completely suppressed in this case.

[0050] Fig. 2 e) now shows the operation according to the invention. The active Q-switch is activated at activation time t A . In the exemplary embodiment, activation time t A occurs before the start of the pump pulse. The active Q-switch could also be activated after the start of the pump pulse. However, activation time t A is in any case before the earliest time at which a laser pulse would occur without activation of the active Q-switch. The active Q-switch remains activated beyond the time at which the laser pulse would be triggered at the latest without activation of the active Q-switch, taking jitter into account. Subsequently, but before the end of the pump pulse, the active Q-switch is deactivated (at deactivation time t D ). This triggers the laser pulse. The deactivation naturally occurs before the time at which a laser pulse would be triggered at the earliest if the active Q-switch were activated.

[0051] The short time span between the deactivation of the active Q-switch and the triggering of the laser pulse varies only minimally from pulse to pulse. The moment of triggering the laser pulse is thus precisely defined. The remaining jitter is more than an order of magnitude lower than the jitter that would occur if a passive Q-switch were used alone.

[0052] In the exemplary embodiment, the active Q-switch is activated by applying a high voltage and deactivated by switching off the high voltage.

[0053] Fig. 3 a) shows the loss due to the active Q-switch. This is 10% in the activated state of the active Q-switch in the example. When the active Q-switch is deactivated, the loss caused by the active Q-switch drops to preferably less than 1% (in Fig. 3 a) the remaining loss was neglected).

[0054] Fig. 3 b) shows the loss caused by the passive Q-switch. The maximum loss, i.e., in the fully unsaturated state of the absorber, is approximately 90% in the example. When the absorber is saturated, the loss caused by the passive Q-switch drops to preferably less than 1% (in Fig. 3b ) the remaining loss was neglected).

[0055] Fig. 3 c) shows the onset of the laser pulse in relation to the temporal progression of the losses caused by the active and passive Q-switches. Between the deactivation of the active Q-switch and the maximum intensity of the laser pulse, there is a delay time tv, which, as mentioned, is largely constant from pulse to pulse.

[0056] Out of Fig. 3 It is also evident that immediately before deactivation of the active Q-switch, the loss caused by the passive Q-switch is still greater than the loss caused by the active Q-switch.

[0057] Advantageously, the pulse energy can be greater than 1 mJ. The pulse duration can be less than 500 ps.

[0058] The Brewster surfaces of the active laser material 1 and passive Q-switch 2 could also be omitted and instead another polarizing element could be arranged in the resonator.

[0059] A second embodiment of the invention is shown in Fig. 4 Apart from the differences described below, the design and operation of the laser correspond to the first embodiment, and the description of the first embodiment can be applied analogously.

[0060] The fiber-coupled laser diode modules used as pump sources in the first and second embodiments have limited power and are relatively expensive for the available power. To increase power at lower cost and with less complexity, QCW laser diode stacks are used as the pump radiation source in this embodiment. Designs are known in which several, for example, 1 cm wide, laser diode bars are combined to form a high-power laser diode module that operates in QCW ("quasi-continuous wave"). It is also known to bond the bars directly to one another, so that there is a small distance between the bars, e.g., 140 µm. A stack with 10 bars can thus emit over an area of ​​10 mm x 1.4 mm, with the radiation angle in the x-direction being approximately + / - 5° and that in the y-direction approximately + / - 25°, provided no optics are used.

[0061] A significant difference to the previously described embodiments is that in this embodiment, the active laser material is side-pumped (i.e., by a surface that is at an angle to the axis of the laser beam). For this purpose, a previously described stack 11, which is shown in Fig. 4 is indicated schematically, with which the pump radiation 6 is radiated directly into the active laser material 1.

[0062] The laser beam in the active laser material 1 is reflected by total internal reflection at this side surface 9, through which the pump radiation 6 is irradiated.

[0063] A laser beam incidence angle of 45° is particularly preferred, so that the laser beam axis 7 is bent by 90°. With such an angle of incidence and p-polarization, the incident and outgoing beams do not interfere with each other, as their E-field vectors are perpendicular to each other. As a result, the incident and outgoing beams, considered individually, do not generate any spatial hole burning in the excited active laser material 1, which increases mode stability and efficiency. Otherwise, so-called SHB ("spatial hole burning") could significantly increase the gain of the secondary modes.

[0064] In order to enforce the desired p-polarization, the active laser material 1 and the passive Q-switch 2 can each have a Brewster surface as shown, analogous to the previously described embodiment.

[0065] The length (=resonator length a) of the resonator 3 measured along the axis 7 of the laser beam from the first mirror 4 to the second mirror 5 (=unfolded length of the resonator) can be, for example, 20 mm in this embodiment. This resonator length a is in Fig. 3 indicated by dimensioning lines and arrows at an angle to each other.

[0066] For example, a previously described laser diode stack can emit 2000 W of power for 250 µs, which corresponds to a pump energy of 500 mJ. This allows a laser pulse energy in the range of 100 mJ to be achieved.

[0067] In order to achieve a sufficient expansion of the pump radiation in the y-direction, the air gap between the laser diode stack and the entrance surface into the active laser material 1 can be, for example, in the range of 2-3 mm.

[0068] The total negative refractive power of the resonator 3 is designed to result in a sufficiently large mode radius of the laser radiation. For example, the focal length of the first mirror 4 is in the range of -50 mm. This results in a magnification of approximately 2. The mode radius of the laser radiation when impinging on the second mirror 5 can be in the range of 3.5 mm. The gradient mirror is designed accordingly. The radius at which the reflectivity of the gradient mirror has dropped to e -2< can be in the range of 1.75 mm.

[0069] Polarization selection can additionally be supported by a phase shift between the s- and p-polarization during total internal reflection at the surface of the active laser material 1. The unfolded length of the resonator 3 can be adjusted by fine adjustment such that only for one of the two polarizations a longitudinal mode lies at the gain maximum of the active laser material 1, while the longitudinal modes for the other polarization are shifted to such an extent that they at least essentially do not oscillate (i.e., the energy is less than 10%, preferably less than 1%, of that of the fundamental mode with the polarization that results in a position at the gain maximum), in particular do not oscillate at all (i.e., the laser threshold is not exceeded). Such fine adjustment of the resonator length can be achieved, for example, by adjusting the temperature of the base plate on which the laser is mounted.

[0070] The phase difference between the s- and p-polarizations caused by total internal reflection is approximately 116° at an angle of incidence of 45°. It would also be conceivable to coat this side surface 8 to create a targeted phase difference, for example, of 90°. This coating would have to be transparent to the wavelength of the pump radiation.

[0071] The Brewster surfaces of the active laser material 1 and the passive Q-switch 2 could also be omitted, and polarization selection could be achieved solely by the phase difference between the s- and p-polarization of the laser radiation caused by total internal reflection. An additional or different polarizing element could also be provided.

[0072] Fig. 5 shows a third embodiment of the invention. The design corresponds to that of the second embodiment, except for the differences described below. Otherwise, the description of the second embodiment can be applied analogously.

[0073] In this exemplary embodiment, the active laser material 1 has a different shape than in the second exemplary embodiment. It is longer in such a way that two total internal reflections occur. When the laser beam enters the active laser material 1 from the first mirror 4 (after passing through the passive Q-switch 2), total reflection initially occurs at the side surface 9, through which the pump radiation is radiated. Total reflection then occurs at the opposite side surface 10. The laser beam then exits an end surface of the active laser material 1, preferably at a Brewster angle, and from there passes through the active Q-switch 8 to the second mirror 5. The bends of the axis 7 of the laser beam during the two total internal reflections are preferably 90° each. However, other angles are also conceivable and possible in principle.

[0074] The opposing Brewster surfaces of the passive Q-switch 2 and the active laser material 1 have been omitted here. Instead, opposing Brewster surfaces are provided between the active laser material 1 and the active Q-switch 8. Instead or additionally, opposing Brewster surfaces could also be provided between the passive Q-switch 2 and the active laser material 1.

[0075] The active laser material formed in this way can be easily mounted on a base plate. This design may also have advantages with regard to "parasitic lasing." A certain disadvantage is the greater length of the active laser material.

[0076] If the total unfolded length of the resonator can be kept short enough, further total internal reflection of the laser beam in the active laser material would also be conceivable and possible.

[0077] Fig. 6 shows a fourth alternative example not falling within the scope of the present invention. The differences from the first embodiment are explained below. Otherwise, the description of the first embodiment can be applied analogously.

[0078] In this alternative example, the resonator is designed as a stable resonator.

[0079] The mirrors 4, 5, which in turn form the end mirrors of the resonator, can be flat in this alternative example, with the thermal lens forming a focusing element during operation. Another focusing element (= one with a positive refractive power) can also be arranged in the resonator. At least one of the two mirrors 4, 5 can also be designed to be focusing. In the alternative example, this is indicated for mirror 5.

[0080] The single longitudinal mode of the laser could have a relatively small beam radius compared to the first embodiment, e.g., a few tens of µm for the plane mirror 4. The pulse rates can be relatively high, especially above 8 kHz. The pulse energies can be in the range of 1 µJ to 100 µJ, for example.

[0081] The Brewster surfaces of the active laser material 1 and the passive Q-switch 2 are not present in this alternative example. Since no polarizing element is present, the active Q-switch 8 is designed here in the form of an acousto-optical modulator. This is driven by a piezoelectric element (indicated by the horizontal lines).

[0082] In a modification of this alternative example, a polarizing element could be arranged in the resonator; for example, this could be implemented using a "flat-brewster" configuration analogous to the first embodiment. In this case, the active Q-switch 8 could also be formed by an electro-optical modulator.

[0083] In this alternative example, mirror 5 is designed to be partially transparent to couple out the laser beam. For example, mirror 5 can have a transmission of 30% at 1064 nm.

[0084] The passive Q-switch 2 and the active laser material 1 are in Fig. 6 shown separated by a gap. These two elements are preferably bonded together as a single unit, e.g., by diffusion bonding at the respective polished surface. If a "flat-brewster" configuration is not present, a dielectric layer between the two elements can be omitted.

[0085] Fig. 7 shows a fifth alternative example, not falling within the scope of the present invention. In this example, the resonator 30 is designed as a ring resonator. In ring resonators, possible modes circulate within the resonator. For example, as shown, a substantially "triangular" structure can be formed, with reflections at mirrors 4, 5 occurring at two corners of the triangle and total internal reflections in the active laser material 1 occurring at the third corner of the triangle. Further reflections at reflective surfaces are conceivable and possible, e.g., to form a substantially "square" ring structure. The laser radiation can be coupled out by one of the mirrors 4, 5. In the case of the design as an unstable resonator, as is the case in Fig. 7 As shown, this mirror 5 is designed as a gradient mirror. Pumping can, for example, be carried out in a manner analogous to that described in the second embodiment, ie, the active laser material 1 is side-pumped.

[0086] In the case of a ring resonator, the resonator length is the length measured along the axis of the laser beam in the resonator 30 over one revolution through the resonator 30. So that, in an advantageous embodiment, at least essentially only a single longitudinal mode oscillates in the resonator, the resonator length is preferably less than 100 mm. Advantageously, the laser also has at least essentially only a single transverse mode during operation. With regard to the design and configuration of the passive and active Q-switches 2, 8 in the resonator, the statements made in this context regarding the embodiments with a linear resonator apply. The loss of the active or passive Q-switch here refers to a single revolution in the ring resonator (i.e., to only a single passage through the respective Q-switch).

[0087] A ring laser (=laser with a ring resonator) has the advantage that spatial hole burning can be completely eliminated, which increases mode stability with respect to secondary modes. Second harmonic generation optical crystals could be incorporated into the ring resonator, thus generating, for example, green 532 nm pulses from 1064 nm.

[0088] Various further modifications are conceivable and possible without departing from the scope of the invention. For example, in the embodiments other than the one shown in Fig. 6 In the alternative example shown, an acousto-optical modulator can be used as an active Q-switch instead of an electro-optical modulator.

[0089] The order of the elements in the resonator can be varied. The pump radiation could be coupled in a different way, e.g., by an additional dichroic mirror (for longitudinal pumping).

[0090] To improve performance and / or to avoid parasitic lasing, appropriate coatings such as anti-reflection coatings can be provided, as is known in the art.

[0091] For polarization selection of the laser light, other angles could be used instead of the Brewster angle, e.g., a 45° angle. Different parts could also be bonded with a dielectric layer in between.

[0092] Polarization selection could also be performed in a different way than shown, for example, by applying a grating structure to or into a coating of an optical element of the resonator. Polarization selection of laser light is known in various ways.

[0093] A different active laser material 1 than Nd:YAG could be provided, e.g. Yb:YAG or Er:YAG.

[0094] For the passive Q-switch 2, other saturable absorber materials can be used, e.g. Co:MALO (for wavelengths > 1.2 µm), optionally in combination with Nd:YAG, which is operated at the secondary line of 1440 nm.

[0095] For an electro-optical modulator, a material other than BBO could be used, e.g. RTP, KTP, KD*P, etc. legend

[0096] to the reference numbers: 1active laser material 2passive Q-switch 3resonator 4mirror 5mirror 6pump radiation 7axis 8active Q-switch 9side surface 10side surface 11stack

Claims

1. Q-switched solid-state laser comprising an active lasing medium in the form of Nd:YAG, Yb:YAB or Er:YAG and a linear resonator (3, 30) that is non-convoluted or convoluted and has a resonator length (a) of less than 50 mm measured along the axis of the laser beam in the resonator, with either just one longitudinal mode oscillating in the resonator (3) or one longitudinal fundamental mode and further modes oscillating in the resonator, the energy of said further modes being less than 10% of the energy of the fundamental mode, and said resonator also having both a passive Q-switch (2) and an active Q-switch (8), the loss induced by the active Q-switch (8) when in the activated state being less than 50% in relation to a round trip by the resonator (3), characterised in that the resonator (3, 30) is formed as an unstable resonator, one of the mirrors (4, 5) being a gradient mirror, and in that the maximum loss induced by the passive Q-switch (2) is more than 50% in relation to a round trip by the resonator (3), the pulse duration of one pulse of the solid-state laser being less than 500 ps, and the pulse energy of one pulse being more than 1 mJ.

2. Solid-state laser according to claim 1, characterised in that the active Q-switch (8) is an electro-optic modulator, preferably a Pockels cell, or an acousto-optic modulator.

3. Solid-state laser according to claim 1 or claim 2, characterised in that the loss induced by the active Q-switch (8) when in the activated state is less than 30% in relation to a round trip by the resonator (3).

4. Solid-state laser according to any of claims 1 to 3, characterised in that the maximum loss induced by the passive Q-switch (2) is more than 70% in relation to a round trip by the resonator (3).

5. Q-switched solid-state laser according to any of claims 1 to 4, characterised in that the resonator (3, 30) has a resonator length (a) of less than 25 mm.

6. Solid-state laser according to any of claims 1 to 5, characterised in that the active lasing medium (1) is side-pumped.

7. Solid-state laser according to any of claims 1 to 5, characterised in that the active lasing medium (1) is longitudinally pumped.