Open Access
Add to BibSonomy
Abstract
Highly efficient operation of an Yb:YLuGdCOB compact laser was demonstrated, producing an output power of 14.30 W around 1054 nm, with an optical-to-optical efficiency of 70.0% and a slope efficiency of 81%. In dual-polarization oscillation in an emission range of 1080 − 1085 nm, 11.37 W of output power was generated with a slope efficiency of 76%. Efficient passively Q-switched operation was also realized by incorporating a 2D MoTe2 saturable absorber into the compact resonator, producing a maximum pulsed output power of 1.56 W at 1030 nm at a repetition rate of 481 kHz; while the largest attainable pulse energy, minimum pulse duration, and highest peak power were, respectively, 3.87 μJ, 26.2 ns, and 147.7 W. The pulse width proves to be the shortest, while the peak power is the highest ever reported for solid-state lasers passively Q-switched with 2D saturable absorbers.
© 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
It has been known for two decades that the monoclinic rare-earth calcium oxyborates, with a formula of ReCa4O(BO3)3 (ReCOB, Re = Y, Gd, or La), can serve as desirable host crystals for the trivalent Yb active ion [1,2]. These oxyborates crystals can be easily grown from melt with large sizes and good optical quality; have high optical damage thresholds; and are able to accommodate Yb ions in high concentrations. More importantly, the Yb ion doped in ReCOB crystals is of great Stark splitting in its ground state (> 1000 cm−1 [1,2]), which proves to be rather rare in Yb-ion laser materials. Large ground-state splitting can effectively reduce the thermal population in the lower laser level, making it easier to obtain low-threshold high-efficiency laser operation at room temperature. Highly efficient continuous-wave (cw) laser operation of Yb:YCOB or Yb:GdCOB has been demonstrated in a compact plano-concave resonator under simple end pumping conditions, producing output power in the 15 − 20 W level [3–5]. Yb:YCOB or Yb:GdCOB has also been used in mode-locking [6,7], self-frequency doubling [8–10], generating three-level laser oscillation at 976 nm [11], and in high-power thin disk lasers [12,13].
Apart from the ordered crystals, there also exist different Yb-ion doped mixed oxyborates consisting of two or more inert rare-earth ions, such as Yb:YGdCOB and Yb:LuGdCOB [14–17], which have shown comparable or even greater capability of output power scaling in compact cw lasers [15,17,18]. In our recent work, a new (Yb, Y, Lu, Gd) mixed oxyborate crystal, having a specific composition expressed as Yb0.19Y0.34Lu0.12Gd0.35Ca4O(BO3)3, which is abbreviated as Yb:YLuGdCOB in the text, has been developed [19]. Studies of its spectroscopic properties indicated that the absorption cross-section of the zero-phonon transition at about 976 nm is considerably larger than that for Yb:YCOB or Yb:GdCOB [19]; the maximum absorption coefficient of Yb:YLuGdCOB amounts to 18.0 cm−1 for E//Y, making it particularly suitable for applications in compact or microchip lasers.
In this paper we report on the output characteristics of compact Yb:YLuGdCOB lasers operating in cw as well as passive Q-switching modes. Highly efficient cw laser operation was achieved with a Y-cut crystal, producing an output power of 14.30 W at wavelengths around 1054 nm; the slope efficiency could reach 81%. Compared to the previously reported results of a similar Yb:YCOB laser, the laser performance demonstrated in the present work represents a significant improvement. Efficient passively Q-switched laser action was also realized with an X-cut crystal, when a 2D MoTe2 saturable absorber was incorporated into the compact resonator. Pulsed output power of 1.1–1.6 W could be generated at repetition rates of 300–480 kHz, with the minimum pulse duration and maximum peak power being 26.2 ns and 147.7 W, which are, to our knowledge, the best records up to now for solid-state lasers passively Q-switched by 2D saturable absorbers.
2. Description of experiment
The Yb:YLuGdCOB mixed crystal was grown by use of the Czochralski method. The Yb ion concentration in the crystal was 19 at. %, corresponding to an ion density of 8.85 × 1020 cm−3. The crystal samples utilized was X- or Y-cut, 3.2 mm thick, with an aperture of 3.0 mm × 3.0 mm. No antireflection coatings were made on its end faces.
As shown schematically in Fig. 1, the compact laser cavity was formed with a plane reflector (M1) and a plane output coupler (M2). The M1 mirror was coated for high reflectance (> 99.8%) at 1020–1200 nm and for high transmittance (> 95%) at 976 nm. The transmittance of the output coupler could be chosen in a range from T = 1% to T = 80%. The crystal sample (LC) was fitted into a copper heat-sink that was maintained at a temperature of 15°C during laser operation. A 2D MoTe2 saturable absorber was employed to induce passive Q-switching. It was made on a 0.35 mm thick sapphire substrate by CVD method. The nonlinear absorption (modulation depth) of the 2D MoTe2 was ΔT = 0.9% at wavelengths around 1030 nm, which, along with its other saturation parameters, were measured in our previous work [20]. The cavity length of the laser was determined by the crystal thickness. It was 3.3 mm for cw laser operation, and had to be lengthened to 5.0 mm in the case of passive Q-switching to accommodate the 2D saturable absorber. As the pump source for the laser, a fiber-coupled diode laser emitting at 975.8 nm (bandwidth < 0.5 nm) was utilized, whose fiber core diameter was 105 μm and NA was 0.22. Through a re-imaging unit (ORU) the pump radiation was focused into the Yb:YLuGdCOB crystal sample, with a pump spot radius of about 90 μm. The pump radiation provided by the fiber-coupled diode laser was nominally un-polarized. Our measurement indicated that the power ratio between the horizontal and vertical polarization components varied slightly with the output level, over a range of 1.21–1.24. Given the absorption coefficients of the Yb:YLuGdCOB crystal which could be calculated from the absorption cross-sections at 975.8 nm [19], α = 9.5 cm−1 (E//X), 14.4 cm−1 (E//Y), 7.6 cm−1 (E//Z), one can obtain the unsaturated absorption of the 3.2 mm thick crystal sample: ηa0 = 0.91–0.95 (Y-cut), 0.91–0.99 (X-cut), for any polarization degree of the pump radiation. So the influence of the polarization degree of the pump radiation on the laser action would be fairly limited, in particular for the Y-cut crystal.
Fig. 1. Schematic diagram of the experimental setup. ORU: optical re-imaging unit; LC: laser crystal; SA: saturable absorber.
3. Results and discussion
3.1 Continuous-wave operation
For cw laser operation we chose to use a Y-cut crystal, since this orientation was found to be superior to X- or Z-cut crystal, in terms of output power as well as laser efficiency. A crystal thickness of 3.2 mm proved to be sufficient for pump absorption; the small-signal or unsaturated absorption fraction was measured to be ηa0 = 0.91, which was larger than ∼ 0.89 for a 4 mm long Yb:YCOB (15 at. %) and ∼ 0.85 for a 6 mm long Yb:GdCOB (10 at. %) crystal [3,4]. Efficient cw laser action was realized with the 3.2 mm thick Y-cut crystal in the compact resonator, under various output couplings ranging from T = 1% to T = 50%.
Figure 2 illustrates the output characteristics of the laser for T = 1%, 5%, 20%, 40% and 50%. The absorbed pump power (Pabs) is calculated from the incident pump power (Pin) by Pabs = ηa0Pin. As indicated explicitly in Fig. 2, the laser radiation produced in the case of T = 1% was linearly polarized with E//X; for higher output couplings of T ≥ 5%, however, the polarization state would switch to E//Z. Such polarization changing behavior was also encountered in laser operation of the ordered oxyborates [5,21,22]. It results from the fact that at a very low excitation level, the gain for E//X predominates in the long-wavelength sideband; but for high excitation levels the gain for E//Z in the main emission band is the dominant. The lasing threshold was measured to be Pabs,th = 0.17 W for T = 1%; an output power of 8.04 W could be generated at Pabs = 15.30 W, giving an optical-to-optical efficiency of 52.5%. Above this pumping level, the laser output would get saturated. As can be noted from Fig. 2, increasing the resonator output coupling could effectively mitigate this output saturation behavior. For T = 5%, which proved to be the optimal, a somewhat higher threshold of Pabs,th = 0.31 W was measured; above which the output power was found to increase with Pabs, with a slope efficiency amounting to 81%. At the highest available pumping level of Pabs = 20.43 W, the output power reached 14.30 W, leading to an optical-to-optical efficiency of 70.0%. One sees that despite the lowering of slope efficiency in the high pump region, no saturation behavior was observed up to the maximum pump power, suggesting the presence of more room for further power scaling. With the output coupling increased to T = 20%, the laser action became less efficient compared to the case of T = 5%; the slope efficiency decreased to 70%, while the output power attainable was reduced to 12.71 W. One also sees that an output power of 10.0 W could be generated in the case of T = 40%; and even under a rather high output coupling of T = 50%, efficient laser operation with a slope efficiency of 53% was still achievable, producing an output power of 8.56 W at the highest available pump level. These results also predict the potential of the Yb:YLuGdCOB compact laser in Q-switched pulsed operation, where a sufficiently high resonator output coupling is essential to prevent optical damage to internal elements.
Fig. 2. Output power versus absorbed pump power, measured for the cw Yb:YLuGdCOB laser operating under different output couplings.
Figure 3 shows the laser emission spectra measured at Pabs = 14.43 W for various output couplings. In the case of the lowest output coupling (T = 1%), the lasing spectrum falls within the long-wavelength emission sideband of the mixed crystal [19], with the central lasing wavelength located at 1084.0 nm. Under the optimal output coupling of T = 5%, the laser emission spectrum shifted to the short-wavelength side; the center wavelength was measured as 1054.0 nm. With the resonator output coupling being further increased, the lasing spectrum would shift continually towards the short-wavelength side. For T = 50% the lasing spectrum covering a range of about 1024–1032 nm, has already shifted to within the strongest central part of the main emission band for E//Z [19].
Fig. 3. Lasing spectra measured at Pabs = 14.43 W for the cw Yb:YLuGdCOB laser operating under different output couplings of T = 1%, 5%, 20%, 40% and 50%.
Such a shifting behavior of lasing spectrum upon the increase of resonator output coupling, proves to be quite typical for quasi-three-level laser materials like Yb-ion crystals which possess a wide emission band. As the overall resonator losses are increased, higher gain and thus higher excitation level is required for laser oscillation; with the excitation level being raised continually, the wavelength at which the gain reaches its maximum, tends to shift progressively towards short-wavelength side, as is clearly shown by the gain cross-section curves for Yb:YCOB, Yb:GdCOB, or Yb:LuGdCOB [1,17,21].
As indicated in Figs. 2 and 3, polarization state switching from E//X to E//Z would occur in the laser oscillation, if the resonator output coupling was increased from T = 1% to T = 5%. For an intermediate output coupling, e.g., T = 2%, the two orthogonal polarization states would exist simultaneously in the laser oscillation. Such an output-coupling-dependent polarization variation turns out to be a common feature for laser action of Y-cut oxyborates doped with Yb ion [5,21,22].
It needs to be pointed out that despite the qualitative similarity in the laser polarization properties of these oxyborates, distinct polarization varying behavior is exhibited by different crystals, which also depends upon the crystal length used, the Yb-doping level, and the resonator configuration employed. To reveal the variation character of the polarization state of the Yb:YLuGdCOB laser, we measured, in the case of T = 2%, the pumping level dependence of the output power for the total, the E//X, and the E//Z components. The results are depicted in Fig. 4(a); while the evolution of the lasing spectrum upon pump power increasing is shown in Fig. 4(b).
Fig. 4. Output power versus absorbed pump power for the total, the E//X, and the E//Z components (a) and the lasing spectra at various pumping levels (b), measured for the cw Yb:YLuGdCOB laser operating under an output coupling of T = 2%.
As shown in Fig. 4(a), the whole operational range in this case could be divided roughly into three distinct regions. In the low-pumping region (I), the E//Z component oscillating at wavelengths around 1057 nm [Fig. 4(b)], proved to be the dominant; while the E//X oscillation in the long-wavelength emission sideband was much weak. However, as the pumping level was increased to near the boundary at Pabs ≈ 1.90 W, the E//Z component would also begin to oscillate in the long-wavelength sideband, which is indicated in Fig. 4(b). In contrast to the situation of region I, the laser operation in region II was dominated by the E//X component, which increased much more quickly with the rising pump power than did the E//Z component. It was also found in this operational region that with the pumping level being increased approaching the upper boundary at Pabs ≈ 6.20 W, the E//Z oscillation in the short-wavelength spectral region would get diminished and eventually ceased, leaving only the oscillation in the long-wavelength sideband, as illustrated in Fig. 4(b). One notices from Fig. 4(a) that over the high-pumping region (III), the growth of the output power for E//X became slow compared to the case of region II; whereas the increase of the E//Z output became fast; and in excess of a certain pumping level (Pabs ≈ 12.0 W), the slope efficiencies for the two components became approximately equal, i.e., half of the total slope efficiency of 76%. At an absorbed pump power of 17.1 W, the total output power reached 11.37 W, of which the E//X and E//Z components were measured respectively to be 6.30 and 5.07 W; the total optical-to-optical efficiency being 66.5%. As can be seen from Fig. 4(b), at very high pumping levels the laser emission spectra for the two polarized components overlapped largely, covering a range of about 1080–1085 nm, which remained nearly unchanged.
A Findlay-Clay analysis [23] of the thresholds of laser emission concluded that the residual losses of the resonator, Li, were around 0.03. The output beam quality was also investigated for the cw Yb:YLuGdCOB compact laser. The beam quality factor (M2) was measured as Mx2 = 1.29 and My2 = 1.25 at a very low output power of 0.2 W in the case of T = 5%. Here the subscripts x and y refer to the horizontal and vertical directions. The magnitude of M2 was found to increase gradually with the output level. It was measured to be Mx2 = 3.64 and My2 = 3.69 at an output power of 8.0 W. Figure 5 depicts the measured beam spot radii at various propagation distances. The values for M2 are determined by fitting the measured data to the standard Gaussian beam propagation law. A beam pattern recorded at 8.0 W of output power is also shown in Fig. 5 as an inset.
Fig. 5. Beam spot radius as a function of the propagation distance for the horizontal (x) and vertical (y) directions, measured at an output power of 8.0 W for the cw Yb:YLuGdCOB laser operating under an output coupling of T = 5%. Inset: the beam pattern.
The increase of the M2 value with output power might be attributed to two main physical reasons. Firstly, as the output power and hence the absorbed pump power was increased, the thermal lensing occurring in the laser crystal would get strengthened, which led to reduction in the fundamental mode size. A simple calculation indicates that as the thermal focal length is reduced from fT = ∞ (no thermal lensing) to fT = 50 mm, the fundamental mode radius will decrease from 0.170 to 0.060 mm. So the number of oscillating transverse modes would grow with rising pump power, if the pump spot radius remained unchanged. Secondly, in the situation where the pump spot radius is considerably greater than the sizes of oscillating modes, the gain will be able to accumulate increasingly with the rising pump power in the wings of the pumped area, enabling more higher-order transverse modes to get oscillating.
In the calculation of laser mode size, we supposed an upper limit for the thermal lensing to be fT = 50 mm, which was estimated on the basis of the measured thermal lensing strengthening rate, or the so-called “sensitivity factor”, D = d(1/fT)/dPabs, for the similar Yb:YCOB laser [21]. According to the measurement, D = 1.0 m−1/W in the YX plane of the 3 mm thick Y-cut crystal that was maintained at 12°C (heat-sink temperature) [21]. Using this value for the Yb:YLuGdOB crystal of the current laser, one obtains fT = 50 mm at a pumping level of Pabs = 20 W, very near the maximum pump power absorbed in the crystal.
One notes that the values of M2 measured at high output levels appear to be considerably larger than unity, suggesting the existence of many higher-order transverse modes in the laser beam due to the physical reasons mentioned above. Decreasing the pump beam spot size can reduce the number of oscillating transverse modes and thus improve the output beam quality. This requires a diode laser with higher brightness to be used as the pump source. Additionally, by cooling the laser crystal more effectively, the thermal lensing effect would be mitigated, which proves also to be beneficial to the improvement of the laser beam quality.
To give a brief summary, we list, in Table 1, the primary parameters characterizing the performance of the cw Y-cut Yb:YLuGdCOB compact laser operating under various output couplings ranging from 1% to 50%. These include: the absorbed pump power required for reaching the laser threshold, Pabs,th; the attainable maximum output power, Pout,max; the polarization state, Pol.; optical-to-optical efficiency, ηopt; slope efficiency, ηs; and lasing wavelengths, λl.
Table 1. Primary parameters of the CW Yb:YLuGdCOB laser operating under various output couplings.
It seems to be instructive to make a comparison between the performance of the current Yb:YLuGdCOB and the previously reported Yb:YCOB microchip laser, the latter was made using a 3 mm thick crystal (Yb-ion concentration of 15 at. %) [21]. According to the previous investigation, the most efficient laser action was achieved with Y-cut Yb:YCOB crystal; while Z-cut crystal was found to be more suitable for power scaling [21]. Table 2 lists the main parameters for the Yb:YLuGdCOB and Yb:YCOB lasers, which were measured in the case of T = 5%, the optimal for these compact oxyborate lasers [21]. The maximum absorbed pump power, Pabs,max, at which the highest output power was produced, is also presented for each laser.
Table 2. Comparison of the main parameters between the compact Yb:YLuGdCOB and Yb:YCOB lasers operating under their optimal output coupling of T = 5%.
From Table 2 one can note the significant improvement in laser performance of the mixed Yb:YLuGdCOB compared to the ordered Yb:YCOB: the threshold was much lower (only one third of that for Yb:YCOB); laser efficiencies were higher; while the achievable output power was much higher than produced with Yb:YCOB in nearly the same resonator [21].
To demonstrate more reliably the superior laser properties of the Yb:YLuGdCOB to other similar oxyborates, we conducted a comparative investigation of the cw laser performance of Yb:YCOB, Yb:GdCOB, Yb:YGdCOB (Yb0.14Y0.77Gd0.09Ca4O(BO3)3), and Yb:LuGdCOB (Yb0.09Y0.13Gd0.78Ca4O(BO3)3), in the same compact plane-parallel resonator under the same diode pumping conditions as in the case of the Yb:YLuGdCOB laser. The specific crystal parameters for these oxyborates used are given in Table 3, while the main parameters characterizing the cw laser performance as in Table 2, which were measured under the optimal output coupling (Topt), are listed in Table 4. The unsaturated absorption fraction, ηa0, is also presented for each of the oxyborate crystals.
Table 3. Parameters of the Yb-doped oxyborate crystal samples used in the comparative investigation.
Table 4. Main parameters characterizing the cw laser performance of the Yb-doped oxyborates.
As given in Table 4, the optimal output coupling also proved to be Topt = 5%, except for the case of Yb:LuGdCOB where Topt = 3%. In comparison with Yb:YLuGdCOB, the lasing thresholds of these oxyborate crystals were higher; whereas the attainable maximum output power as well as the laser efficiencies were lower. These results confirm the superiority of the mixed Yb:YLuGdCOB over the ordered Yb:YCOB, Yb:GdCOB, and other mixed oxyborates known at present.
Table 5 lists, for the mixed Yb:YLuGdCOB and other Yb-doped oxyborates crystals, the primary spectroscopic parameters that are closely related to laser action. These parameters include: the strongest absorption wavelength, λabs; maximum absorption cross-section, σabs,max; absorption bandwidth, Δλabs; the peak wavelength of the main emission band, λem; maximum emission cross-section, σem,max; and the fluorescence lifetime, τf.
Table 5. Spectroscopic parameters relevant to laser action for the various Yb-doped oxyborates.
One sees from Table 5 that the spectroscopic properties of these oxyborates appear to be very similar, which is expected in view of their identical crystal structure as well as similar chemical compositions. Nonetheless, the magnitude of σabs,max of Yb:YLuGdCOB proves to be considerably larger than those for the ordered Yb:YCOB and Yb:GdCOB. One can also note that the peak emission cross-sections for the mixed crystals turn out to be smaller compared to the ordered crystals.
In comparison with traditional ordered crystals, mixed laser crystals usually possess wide absorption and emission bands arising from the substitution disorder of the active ion in the crystal lattice, which are desirable for pumping and for some laser operations such as mode-locking or wavelength tuning. However, from Table 5 one notes little or no absorption band broadening for the mixed crystals; similarly, no emission band broadening was observed in the emission spectra of the mixed Yb:YLuGdCOB crystal, compared to that of Yb:YCOB or Yb:GdCOB [19]. In such ordered oxyborates, Yb ions could occupy three distinct lattice sites, leading to significant line broadening [1]. As a consequence, the spectral broadening resulting from the substitution disorder in the mixed oxyborates might be masked to a great extent.
3.2 Passive Q-switching induced by 2D MoTe2
For passive Q-switching operation, an X-cut Yb:YLuGdCOB crystal was found to be more appropriate than Y- or Z-cut crystal. The laser cavity length in this case was increased to 5.0 mm, to accommodate the 2D MoTe2 saturable absorber, which was inserted between the laser crystal and the plane output coupler.
Efficient stable passively Q-switched operation was achieved under sufficiently high output couplings in the range of T = 50%−80%. Employing a high enough output coupling proved to be essential for realizing stable pulsed operation induced by 2D saturable absorbers like MoTe2; otherwise, the thermal effects arising mainly from non-saturable absorption in the 2D saturable absorber would get strengthened, at a moderate laser output level, to such an extent that stable passive Q-switching was no longer possible.
Figure 6 shows the pulsed output power versus Pabs for T = 50%, 60%, 70%, and 80%, produced with the X-cut Yb:YLuGdCOB laser passively Q-switched by 2D MoTe2 saturable absorber. The pulsed laser radiation was linearly polarized with E//Z. The most efficient pulsed operation was obtained under the output coupling of T = 50%; a pulsed output power of 1.36 W was generated at Pabs = 3.87 W, giving an optical-to-optical efficiency of 35.1%. Above this output level, however, the Q-switched laser action would become unstable. For a higher output coupling of T = 60%, the laser operation became less efficient; nevertheless, more pulsed output power, 1.56 W, could be produced at Pabs = 5.14 W, before the onset of passive Q-switching instability, the optical-to-optical efficiency being 30.4%. In the cases of T = 70% and 80%, the maximum pulsed output power achievable in stable Q-switched operation, 1.16 W, was the same.
Fig. 6. Pulsed output power versus absorbed pump power, measured for the passively Q-switched Yb:YLuGdCOB/MoTe2 compact laser operating under high output couplings of T = 50%−80%. Continuous lines are linear fit of experimental data.
As the resonator output coupling was increased, the pulse repetition rate (PRR) as well as the pulse duration could be reduced to some extent. For T = 50%, 60%, and 70%, the variation ranges, over which the PRR increased with rising pump power, were respectively 250–521, 240–481, and 210–461 kHz; while the pulse duration decreased with increasing pump power, from 185 to 45, 167 to 33, and 132 to 29 ns. In the case of T = 80%, the repetition rate and pulse duration, measured at the highest applicable pump power of Pabs = 4.74 W, were 300 kHz and 26.2 ns. Figure 7 shows the corresponding laser pulse train (a) and the temporal profile of an individual pulse (b). The pulse-to-pulse amplitude fluctuations were estimated to be about 6% (rms); while the pulsed output power remained much more stable, with a variation of less than 1% at the highest output level. The pulse duration was also found to remain nearly unchanged; the variation was at most 0.2 ns from pulse to pulse at this highest pump power. Timing jitters were found to exist in the laser pulse train, and were determined to be 11%. No Q-switched mode-locking (QML) phenomenon was observed in the pulsed laser operation.
Fig. 7. The oscilloscope trace of a laser pulse train (a) and the temporal profile of an individual laser pulse (b), measured at the highest applicable pump power (Pabs = 4.74 W) for the Yb:YLuGdCOB/MoTe2 passively Q-switched laser operating under T = 80%.
The pulsed laser emission spectra measured at Pabs = 3.47 W for T = 50%, 60%, 70%, and 80%, are shown in Fig. 8. In contrast to the numerous wavelengths existing in the cw lasing spectrum (Fig. 3), the pulsed laser emission spectrum consisted of only a single oscillation wavelength, despite the presence of a weak emission background. This was due to the longitudinal modes discrimination accompanying the pulse formation process in passively Q-switched lasers. One can see that with the resonator output coupling increased from T = 50% to T = 80%, the pulsed laser wavelength shifted progressively towards the short-wavelength side, from 1033.7 to 1029.3 nm.
Fig. 8. Lasing spectra measured at Pabs = 3.47 W for the Yb:YLuGdCOB/MoTe2 passively Q-switched laser operating under output couplings of T = 50%, 60%, 70% and 80%.
A summary of the passive Q-switching performance of the Yb:YLuGdCOB/MoTe2 laser operating under the various output couplings is presented in Table 6, which lists the primary parameters including the maximum pulsed output power, Pout,max; the PRR range; the largest pulse energy, Ep; the shortest pulsed duration, tp; the highest peak power, Pp; and the laser wavelength, λl. One notes that the highest pulsed output power was obtained in the case of T = 60%; whereas the laser pulses produced under output coupling of T = 80% were of the maximum energy and the minimum duration.
Table 6. Summary of the passive Q-switching performance of the compact Yb:YLuGdCOB/MoTe2 laser operating under different output couplings of T = 50%, 60%, 70%, and 80%.
Experimentally, an X-cut crystal was found to be more appropriate for achieving stable passive Q-switching laser operation, which seems to be common for these Yb-doped ordered or mixed oxyborates. The reasons for this are not quite clear at present. It may be related to the anisotropic thermal properties of such low-symmetry crystals. A study of Yb:YCOB indicated a minimum difference of thermal lensing in the two orthogonal principal planes for an X-cut crystal [21]. It was also shown that for a Y-cut crystal, the thermal lensing appear to be considerably weaker compared to an X- or Z-cut crystal [21], which could facilitate high-power cw laser operation. It might be for this that a Y-cut crystal proved to be more suitable for cw laser action.
These oxyborates are also nonlinear crystals. Green light arising from second harmonic generation (SHG) was observed in our experiment. Due to the lack of phase matching, the green light generated was very weak. In fact, even in a mode-locked Yb:YCOB thin disk laser where the internal laser intensity was much more stronger, the green light was still very weak, with its power limited to 1 mW [13].
Table 7 gives a comparison of the current Yb:YLuGdCOB laser with the previously reported Yb:YCOB, Yb:LaCOB, and Yb:KLuW lasers which were also passively Q-switched with 2D MoTe2 saturable absorber [20,24,25], in the maximum attainable pulsed output power (Pout,max), the corresponding repetition rate (PRR), and lasing wavelength (λl); the largest pulse energy, minimum pulse duration, and highest peak power that were achievable from these lasers.
Table 7. Comparison of the main parameters between the Yb:YLuGdCOB and the previous Yb:YCOB, Yb:LaCOB, Yb:KLuW lasers passively Q-switched with 2D MoTe2 saturable absorber.
One notes from Table 7 the much higher repetition rate reached in the Yb:KLuW laser, which was resulted from the greater emission cross-section, and hence the higher gain, of the laser crystal. While the pulsed output power, repetition rate, and lasing wavelength proved to be comparable to more or less extents for the three oxyborates lasers, the minimum pulse duration of the Yb:YLuGdCOB compact laser is seen to be 2–4 times shorter than obtained in Yb:YCOB and Yb:LaCOB lasers. In fact, the 26.2 ns of pulse width achieved here turns out to be the shortest ever reported for solid-state lasers passively Q-switched with 2D saturable absorbers. So far, sub-30 ns pulse duration has only been obtained in passive Q-switching operation of an Yb:LuPO4 microchip laser induced by 2D WS2 or WSe2 saturable absorber [26,27]. The crystal of Yb:LuPO4, however, is very difficult to grow; up to now, it can only be grown from spontaneous nucleation in high-temperature solution, with crystal thickness limited to ≤ 1 mm.
4. Summary
A highly efficient cw Yb:YLuGdCOB compact laser was demonstrated at room temperature. With 20.43 W of pump power absorbed, an output power of 14.30 W was produced at wavelengths around 1054 nm, leading to an optical-to-optical efficiency of 70.0%; while the slope efficiency could reach 81%. Dual-polarization laser operation was realized in an emission range of 1080–1085 nm, generating a total output power of 11.37 W with a slope efficiency of 76%; the two orthogonal polarization components were comparable in the output. By incorporating a 2D MoTe2 saturable absorber into the compact resonator, efficient passively Q-switched operation was achieved under high output couplings, producing a maximum pulsed output power of 1.56 W at 1033 nm at a repetition rate of 481 kHz; whereas the largest attainable pulse energy, minimum pulse duration, and highest peak power were, respectively, 3.87 μJ, 26.2 ns, and 147.7 W.
Funding
National Natural Science Foundation of China (11574170).
Disclosures
The authors declare no conflicts of interest.
References
1. F. Mougel, K. Dardenne, G. Aka, A. Kahn-Harari, and D. Vivien, “Ytterbium-doped Ca4GdO(BO3)3: an efficient infrared laser and self-frequency doubling crystal,” J. Opt. Soc. Am. B 16(1), 164–172 (1999). [CrossRef]
2. A. Aron, G. Aka, B. Viana, A. Kahn-Harari, D. Vivien, F. Druon, F. Balembois, P. Georges, A. Brun, N. Lenain, and M. Jacquet, “Spectroscopic properties and laser performances of Yb:YCOB and potential of the Yb:LaCOB material,” Opt. Mater. 16(1-2), 181–188 (2001). [CrossRef]
3. J. Liu, W. Han, X. Chen, Q. Dai, H. Yu, and H. Zhang, “Continuous-wave and passive Q-switching laser performance of Yb:YCa4O(BO3)3 crystal,” IEEE J. Sel. Top. Quantum Electron. 21(1), 348–355 (2015). [CrossRef]
4. X. Chen, L. Wang, J. Liu, Y. Guo, W. Han, H. Xu, H. Yu, and H. Zhang, “High-power CW and Passively Q-switched laser operation of Yb:GdCa4O(BO3)3 crystal,” Opt. Laser Technol. 79, 74–78 (2016). [CrossRef]
5. Y. Ma, K. Tian, Y. Li, X. Dou, J. Yang, W. Han, H. Xu, and J. Liu, “Anisotropic lasing properties in the 1059–1086 nm rang of Yb:YCa4O(BO3)3 crystal,” Opt. Mater. Express 8(4), 727–735 (2018). [CrossRef]
6. A. Yoshida, A. Schmidt, V. Petrov, C. Fiebig, G. Erbert, J. Liu, H. Zhang, J. Wang, and U. Griebner, “Diode-pumped mode-locked Yb:YCOB laser generating 35 fs pulses,” Opt. Lett. 36(22), 4425–4427 (2011). [CrossRef]
7. Z. Gao, J. Zhu, W. Tian, J. Wang, Z. Zhang, Z. Wei, H. Yu, H. Zhang, and J. Wang, “Generation of 73 fs pulses from a diode pumped Kerr-lens mode-locked Yb:YCa4O(BO3)3 laser,” Opt. Lett. 39(20), 5870–5872 (2014). [CrossRef]
8. F. Khaled, P. Loiseau, G. Aka, and L. Gheorghe, “Rise in power of Yb:YCOB for green light generation by self-frequency doubling,” Opt. Lett. 41(15), 3607–3610 (2016). [CrossRef]
9. Q. Fang, D. Lu, H. Yu, H. Zhang, and J. Wang, “Self-frequency-doubled vibronic yellow Yb:YCOB laser at the wavelength of 570 nm,” Opt. Lett. 41(5), 1002–1005 (2016). [CrossRef]
10. D. Lu, Q. Fang, X. Yu, X. Han, J. Wang, H. Yu, and H. Zhang, “Power scaling of the self-frequency-doubled quasi-two-level Yb:YCOB laser with a 30% slope efficiency,” Opt. Lett. 44(21), 5157–5160 (2019). [CrossRef]
11. J. Xia, H. Liu, Z. Hu, R. Yang, J. Zhang, Z. Jiang, and Y. Lü, “Pure-three-level Yb:GdCOB CW laser at 976 nm,” Opt. Lett. 43(16), 3981–3984 (2018). [CrossRef]
12. C. Kränkel, R. Peters, K. Petermann, P. Loiseau, G. Aka, and G. Huber, “Efficient continuous-wave thin disk laser operation of Yb:Ca4YO(BO3)3 in E//Z and E//X orientations with 26 W output power,” J. Opt. Soc. Am. B 26(7), 1310–1314 (2009). [CrossRef]
13. O. H. Heckl, C. Kränkel, C. R. E. Baer, C. J. Saraceno, T. Südmeyer, K. Petermann, G. Huber, and U. Keller, “Continuous-wave and modelocked Yb:YCOB thin disk laser: first demonstration and future prospects,” Opt. Express 18(18), 19201–19208 (2010). [CrossRef]
14. Y. Zhang, Z. Lin, Z. Hu, and G. Wang, “Growth and spectroscopic properties of Yb3+:Gd0.5Y0.5Ca4O(BO3)3 crystal,” J. Alloys Compd. 390(1-2), 194–196 (2005). [CrossRef]
15. X. Chen, H. Xu, W. Han, L. Wang, J. Liu, H. Yu, and H. Zhang, “Spectroscopic properties and high-power laser operation of Yb0.14:Y0.77Gd0.09Ca4O(BO3)3 mixed crystal,” Opt. Mater. 55, 33–37 (2016). [CrossRef]
16. D. Zhong, B. Teng, W. Kong, S. Ji, S. Zhang, J. Li, L. Cao, H. Jing, and L. He, “Growth, structure, spectroscopic and continuous-wave laser properties of a new Yb:GdYCOB crystal,” J. Alloys Compd. 692, 413–419 (2017). [CrossRef]
17. F. Liu, L. Dong, J. Chen, S. Cao, H. Xu, and J. Liu, “Growth, spectroscopic properties and efficient laser action of an Yb0.09Lu0.13Gd0.78Ca4O(BO3)3 crystal,” CrystEngComm 22(36), 6026–6033 (2020). [CrossRef]
18. Y. Ma, Y. Li, X. Dou, H. Yi, H. Xu, W. Han, and J. Liu, “Free-running performance of Yb0.14:Y0.77Gd0.09Ca4O(BO3)3 mixed crystal laser operating around 1084 nm,” Opt. Commun. 427, 244–249 (2018). [CrossRef]
19. F. Liu, L. Dong, J. Chen, and J. Liu, “Spectroscopic and lasing properties of a mixed (Yb, Y, Lu, Gd) calcium oxyborate crystal: Yb0.19Y0.34Lu0.12Gd0.35Ca4O(BO3)3,” J. Lumin. 232, 117789 (2021). [CrossRef]
20. Y. Ma, K. Tian, X. Dou, J. Yang, Y. Li, W. Han, H. Xu, and J. Liu, “Passive Q-switching induced by few-layer MoTe2 in an Yb:YCOB microchip laser,” Opt. Express 26(19), 25147–25155 (2018). [CrossRef]
21. P. Loiko, J. M. Serres, X. Mateos, H. Yu, H. Zhang, J. Liu, K. Yumashev, U. Griebner, V. Petrov, M. Aguiló, and F. Díaz, “Thermal lensing and multiwatt microchip laser operation of Yb:YCOB crystals,” IEEE Photonics J. 8(3), 1–12 (2016). [CrossRef]
22. Y. Li, M. Liu, W. Han, J. Chen, and J. Liu, “High-power dual-polarization laser operation of Yb:LaCa4O(BO3)3 crystal,” Opt. Commun. 451, 192–196 (2019). [CrossRef]
23. D. Findlay and R. A. Clay, “The measurement of internal losses in 4-level lasers,” Phys. Lett. 20(3), 277–278 (1966). [CrossRef]
24. Y. Li, Y. Xu, G. Xu, L. Dong, X. Dou, and J. Liu, “Performance of an Yb:LaCa4O(BO3)3 crystal laser at 1.03–1.04 μm passively Q-switched with 2D MoTe2 saturable absorber,” Infrared Phys. Technol. 99, 167–171 (2019). [CrossRef]
25. K. Tian, Y. Li, J. Yang, X. Dou, H. Xu, W. Han, and J. Liu, “Passively Q-switched Yb:KLu(WO4)2 laser with 2D MoTe2 acting as saturable absorber,” Appl. Phys. B 125(2), 24 (2019). [CrossRef]
26. X. Dou, Y. Ma, M. Zhu, H. Xu, D. Zhong, B. Teng, and J. Liu, “Multi-watt sub-30 ns passively Q-switched Yb:LuPO4/WS2 miniature laser operating under high output couplings,” Opt. Lett. 43(15), 3666–3669 (2018). [CrossRef]
27. X. Dou, Y. Ma, M. Zhu, H. Yi, H. Xu, D. Zhong, B. Teng, H. Yu, and J. Liu, “Passively Q-switched Yb:LuPO4 laser based on a coupled cavity with few-layer WSe2 acting as saturable absorber,” Laser Phys. Lett. 15(9), 095804 (2018). [CrossRef]
