I. Introduction
In recent years, UV light sources have been rapidly developed because they are useful in many applications such as rapid prototyping, laser printing, laser processing, spectroscopy, optical data storage, medical treatment and so on. Compared with other UV lasers, diode-pumped all-solid-state lasers with extracavity harmonic generations intrinsically possess advantages of smaller focused size, higher efficiency, longer life time, higher stability, easier implement and smaller system size etc [10-14].
Although utilizing third-order nonlinearity to directly perform THG exists in principle, typical values of third-order nonlinearity are generally smaller than that of second-order nonlinearity in nonlinear crystals. Practically, the THG of the Nd-doped crystal IR laser consists of two cascaded harmonic generations. The first stage is converting the fundamental IR beam to the visible green light by frequency doubling of the Nd-doped crystal laser at the 4F3/2 → 4I11/2 transition. The generated second harmonic radiation is subsequently mixed with the residual fundamental wave via sum frequency process to produce the UV laser near 0.35 μm.
During the early research on the extracavity THG (ETHG) of the all-solid-state laser, the KTP crystal was employed as the frequency doubler because of its large second-order nonlinearity for the SHG. However, the problems of gray tracking as well as low damage threshold make the KTP crystal unfavorable to be used in high-power and high-repetition-rate extracavity harmonic generation. Moreover, the type-II phase matching condition for the KTP crystal leads the polarizations for the fundamental and the second harmonic waves to be neither perpendicular nor parallel before performing sum frequency process. This means the beam manipulations for the fundamental and second harmonic waves before frequency tripler is needed. As a result, the conversion efficiency from 1064 to 355 nm and the UV output power obtained with the KTP crystal as the frequency doubler were quite low [15,16].
Currently, the most efficient way in obtaining the UV lasers near 0.35 μm is by frequency doubling with the type-I phase matching of a portion of the fundamental Nd-doped crystal laser near 1 μm and subsequently sum frequency generation with
133
type-II phase matching of the residual fundamental beam and the frequency doubled radiation. The easy implementation of the described scheme is due to the fact that the fundamental and frequency doubled beams after the first nonlinear crystal have mutually orthogonal polarizations, which are exactly what is required for type-II phase matching in the second nonlinear crystal. There is no need for beam manipulation between the nonlinear crystals except for possible requirement of focusing optics.
Usually, LBO crystals are utilized for both processes thanks to high damage threshold, relatively large acceptance angle, low absorption in visible and UV ranges, and small walk-off angle etc [17-20].
In this section, the optimized Q-switched Nd:YVO4 and Nd:YLF IR lasers developed in Chap. 2 are employed to generate green and UV radiations in the processes of ESHG and ETHG. For the PQS Nd:YVO4 UV laser, the maximum output power at 355 nm is up to 1.62 W, corresponding to the optical conversion efficiency from 1064 to 355 nm of 26 %. For the AQS Nd:YVO4 UV laser, the maximum output power of 6.65 W at 355 nm with the optical conversion efficiency from 1064 to 355 nm as high as 38.2 % is achieved. For the PQS Nd:YLF UV laser, the largest pulse energy of 360 μJ at 351 nm is efficiently obtained.
II. Passively Q-switched Nd:YVO4 UV laser at 355 nm
Here LBO crystals are exploited as nonlinear frequency converters for SHG and THG of the PQS Nd:YVO4 laser at 1064 nm developed in Sec. 2.2. One LBO crystal with dimensions of 3 × 3 × 15 mm3 was cut at θ = 90°, ϕ = 10.4° for type-I phase-matched SHG at a temperature of 46.6 oC. Both facets of the SHG crystal were AR coated at 1064 and 532 nm. Another LBO crystal with dimensions of 3 × 3 × 10 mm3 was cut at θ = 44°, ϕ = 90° for type-II phase-matched THG at a temperature of 48
oC. Both facets of the THG crystal were AR coated at 1064, 532, and 355 nm. The temperatures of the SHG and THG nonlinear crystals were monitored by thermoelectric controllers with the precision of 0.1 oC. Two convex lenses were used to focus the laser beams into the SHG and THG nonlinear crystals for achieving efficient harmonic generations. The former one with focal length of 38 mm was AR coated at 1064 nm on both sides, the latter one with focal length of 19 mm was AR coated at 1064 and 532 nm on both sides. The optimized geometrical distances of L1, L2, L3 and L4 indicated in Fig.
4.2.1 were experimentally determined to be approximately 100, 50, 40, and 20 mm, respectively.
After optimization in the extracavity harmonic generations, the dependences of the output powers at 532 and 355 nm on the incident pump power at 1064 nm are shown in Fig. 4.2.2. At the maximum incident pump power of 6.3 W at 1064 nm, the highest output powers at 532 and 355 nm reach 2.2 and 1.62 W with a pulse width as short as 5 ns and a pulse repetition rate of 56 kHz. Accordingly, the highest pulse energies at 532 and 355 nm are found to be 39 and 29 μJ. More importantly, the largest peak powers at 532 and 355 nm as high as 7.8 and 5.8 kW are achieved. The optical conversion efficiencies from 1064 to 355 nm and 808 to 355 nm are up to 26 and 10 %, respectively. With the knife-edge method, the beam quality factors at 355 nm for orthogonal directions were measured to be Mx2 < 1.2 and My2 < 1.3, respectively.
Finally, it is worthwhile to mention that although the intracavity focusing obtained from the three-element resonator can effectively enlarge the ratio of the laser mode area in the gain medium to that in the saturable absorber to meet the second threshold condition, it will not only add complexities to the overall laser cavity but also reduce the peak power that is detrimental for efficient extracavity harmonic generations.
Employing a c-cut Nd:YVO4 that has smaller stimulated emission cross section is
135
another suitable way to satisfy the second threshold condition; however, the non-polarized laser output is problematic in the processes of extracavity harmonic generations, in which the linearly polarized fundamental beam is usually required.
Comparative speaking, using a simple concave-plano resonator to construct a compact high-power PQS Nd:YVO4/Cr4+:YAG laser with constantly linear polarization is a practical method to simultaneously satisfy the second threshold condition and provide adequate peak power for efficient extracavity harmonic generations.
Fig. 4.2.1. Schematic of the experimental setup for the ESHG and ETHG.
Convex lens Convex lens
SHG module THG module Diode-pumped
Q-switched Nd-doped crystal IR laser
L1 L2 L3 L4
LBO LBO
137
Fig. 4.2.2. Dependences of the output power at 532 nm (green curve) and 355 nm (blue curve) on the incident pump power at 1064 nm.
Incident pump power at 1064 nm (W)
0 1 2 3 4 5 6 7
Output power (W)
0.0 0.5 1.0 1.5 2.0 2.5
output power at 532 nm output power at 355 nm
III. Actively Q-switched Nd:YVO4 UV laser at 355 nm
The experimental setup for the SHG and THG of the AQS Nd:YVO4 laser at 1064 nm developed in Sec. 2.3 is the same as that described in Fig. 4.2.1, except that the optimized geometrical distances of L1, L2, L3 and L4 were experimentally determined to be approximately 70, 43, 34, and 21 mm, respectively.
After optimization in the extracavity harmonic generations, the output power, the pulse energy, and the peak power at 532 and 355 nm versus the pulse repetition rate under an incident pump power of 44 W are shown in Figs. 4.2.3(a)-(c), respectively.
Although the conversion efficiency of harmonic generations increases with decreasing the pulse repetition rate, the output power at 1064 nm is proportional to the pulse repetition rate in the range of 20-50 kHz. As a result, the highest output powers for ESHG and ETHG are found to be approximately at a pulse repetition rate of 40 kHz and their values at 532 and 355 nm are 8.38 and 6.65 W, respectively. The corresponding optical conversion efficiencies from 808 to 355 nm and 1064 to 355 nm are up to 15.1
% and 38.2 %, respectively. On the other hand, at a pulse repetition rate of 20 kHz, the largest pulse energy and the highest peak power at 532 nm are found to be 270 μJ and 30 kW, respectively. Similarly, at a pulse repetition rate of 20 kHz, the largest pulse energy and the highest peak power at 355 nm are found to be 200 μJ and 22 kW, respectively. With the knife-edge method, the beam quality factors at 355 nm for orthogonal directions were measured to be Mx2 < 1.2 and My2 < 1.3.
To manifest the influence of parasitic lasing on the extracavity harmonic generations discussed in Sec. 2.3, the Q-switched laser with Lcav = 16 cm was also employed to perform the process of SHG and THG. Experimental results reveal that even though the average output power at 1064 nm obtained with Lcav = 16 cm is nearly the same as that obtained with Lcav = 20 cm, the average output powers at SHG and THG obtained with Lcav = 16 cm were found to be 15-25 % lower than the results obtained with Lcav = 20 cm. The lower conversion efficiencies obtained with Lcav = 16 cm confirm that the parasitic lasing effect leads to the peak-power reduction in Q-switched lasers.
139
Fig. 4.2.3. Dependences of (a) the output powers, (b) the pulse energies, and (c) the peak powers at 532 and 355 nm on the pulse repetition rate at an incident pump power of 44 W.
IV. Passively Q-switched Nd:YLF UV laser at 351 nm
The experimental setup for the SHG and THG of the PQS Nd:YLF laser at 1053 nm developed in Sec. 2.7 is similar to that described in Fig. 4.2.1. The first LBO crystal with dimensions of 3 × 3 × 10 mm3 was cut at = 90˚, = 11.1˚ for type-I phase-matched SHG at a temperature of 47˚C. Both facets of the SHG crystal were AR coated at 1053 and 527 nm. Another LBO crystal with dimensions of 3 × 3 × 10 mm3 was cut at = 46.4˚, = 90˚ for type-II phase-matched THG at a temperature of 48˚C.
Both surfaces of the THG crystal were AR coated at 1053, 527, and 351 nm. The thermoelectric controllers with a precision of 0.1 ˚C were utilized to monitor the temperatures of the SHG and THG nonlinear crystals. The laser beams were focused into the SHG and THG nonlinear crystals with two individual convex lenses for achieving efficient harmonic generations. The former one with focal length of 38 mm was AR coated at 1053 nm on both sides, the latter one with focal length of 19 mm was AR coated at 1053 and 527 nm on both facets. The optimized geometrical distances of L1, L2, L3 and L4 were experimentally determined to be approximately 76, 62, 31, and 29 mm, respectively.
Figures 4.2.4(a) and (b) show the pulse energies at 527 and 351 nm as a function of the pulse repetition rate. The pulse energy for the extracavity harmonic generation is determined by the combined effects of the spatial and temporal properties of the fundamental laser; that is, the beam quality factors and the peak power. Based on the present experimental circumstance, the largest pulse energies at 527 and 351 nm are achieved to be 490 and 360 μJ under a pulse repetition rate of 100 Hz, respectively. On the other hand, the pulse durations at 527 and 351 nm are experimentally found to be in the range 4-9 ns, depending on the ROC of the input mirror. In comparison with the studies in Refs. [21,22], we believe that the end-pumped scheme is a more feasible way to obtain a nearly diffraction-limited pulsed laser, and the efficient extracavity harmonic generations validate the applicability of our cavity design.
141
Fig. 4.2.4. Pulse energies as a function of the pulse repetition rate at (a) 527 nm and (b) 351 nm.
Pulse repetition rate f (Hz)
0 200 400 600 800 1000 1200
Pulse energy at 527 nm ( J)
100
0 200 400 600 800 1000 1200
Pulse energy at 351 nm ( J)
50